fixup! fixup! fixup! fixup! Introduce Manager::addBooksFromDirectory()

skipInvalid=false makes little sense, if an invalid book is found, we can simply choose to ignore it rather stopping the whole operation midway.

.github/move.yml

@@ -1,27 +0,0 @@
-# Configuration for Move Issues - https://github.com/dessant/move-issues
-
-# Delete the command comment when it contains no other content
-deleteCommand: true
-
-# Close the source issue after moving
-closeSourceIssue: true
-
-# Lock the source issue after moving
-lockSourceIssue: false
-
-# Mention issue and comment authors
-mentionAuthors: true
-
-# Preserve mentions in the issue content
-keepContentMentions: true
-
-# Move labels that also exist on the target repository
-moveLabels: true
-
-# Set custom aliases for targets
-# aliases:
-#   r: repo
-#   or: owner/repo
-
-# Repository to extend settings from
-# _extends: repo

.github/workflows/ci.yml

@@ -1,163 +1,200 @@
 name: CI
 
-on: [push]
+on:
+  push:
+    branches:
+      - main
+  pull_request:
 
 jobs:
-  Macos:
-    runs-on: macos-latest
+  macOS:
+    strategy:
+      fail-fast: false
+      matrix:
+        target:
+          - macos-aarch64-dyn
+          - macos-x86_64-dyn
+          - ios-arm64-dyn
+          - ios-x86_64-dyn
+        include:
+          - target: macos-aarch64-dyn
+            arch_name: arm64-apple-macos
+            run_test: true
+          - target: macos-x86_64-dyn
+            arch_name: x86_64-apple-darwin
+            run_test: true
+          - target: ios-arm64-dyn
+            arch_name: aarch64-apple-ios
+            run_test: false
+          - target: ios-x86_64-dyn
+            arch_name: x86-apple-ios-simulator
+            run_test: false
+    runs-on: macos-15
+
+    env:
+      HOME: /Users/runner
     steps:
-      - name: Checkout code
-        uses: actions/checkout@v1
-      - name: Setup python 3.5
-        uses: actions/setup-python@v1
-        with:
-          python-version: '3.5'
+      - name: Retrieve source code
+        uses: actions/checkout@v4
+
       - name: Install packages
         run: |
           brew update
-          brew install gcovr pkg-config ninja
-      - name: Install python modules
-        run: pip3 install meson==0.49.2 pytest
-      - name: Install deps
-        shell: bash
-        run: |
-          ARCHIVE_NAME=deps2_osx_native_dyn_kiwix-lib.tar.xz
-          wget -O- http://tmp.kiwix.org/ci/${ARCHIVE_NAME} | tar -xJ -C $HOME
+          brew install ninja meson
+
+      - name: Install dependencies
+        uses: kiwix/kiwix-build/actions/dl_deps_archive@main
+        with:
+          target_platform: ${{ matrix.target }}
+
       - name: Compile
+        env:
+          PKG_CONFIG_PATH: ${{env.HOME}}/BUILD_${{matrix.arch_name}}/INSTALL/lib/pkgconfig
+          CPPFLAGS: -I${{env.HOME}}/BUILD_${{matrix.arch_name}}/INSTALL/include
+          MESON_OPTION: --default-library=shared -Db_coverage=true
+          MESON_CROSSFILE: ${{env.HOME}}/BUILD_${{matrix.arch_name}}/meson_cross_file.txt
         shell: bash
         run: |
-          export PKG_CONFIG_PATH=$HOME/BUILD_native_dyn/INSTALL/lib/pkgconfig
-          export CPPFLAGS="-I$HOME/BUILD_native_dyn/INSTALL/include"
-          meson . build --default-library=shared -Db_coverage=true
-          cd build
-          ninja
-      - name: Test
-        shell: bash
-        run: |
-          export LD_LIBRARY_PATH=$HOME/BUILD_native_dyn/INSTALL/lib:$HOME/BUILD_native_dyn/INSTALL/lib64
-          cd build
-          meson test --verbose
-          ninja coverage
+          if [ -e $MESON_CROSSFILE ]; then
+            MESON_OPTION="$MESON_OPTION --cross-file $MESON_CROSSFILE -Dstatic-linkage=true"
+          fi
+          meson . build ${MESON_OPTION}
+          ninja -C build
+
+      - name: Test libkiwix
+        if: matrix.run_test
         env:
           SKIP_BIG_MEMORY_TEST: 1
-      - name: Publish coverage
-        shell: bash
+          LD_LIBRARY_PATH: ${{env.HOME}}/BUILD_${{matrix.arch_name}}/INSTALL/lib:${{env.HOME}}/BUILD_${{matrix.arch_name}}/INSTALL/lib64
+        run: meson test -C build --verbose
+
+  Windows:
+    runs-on: windows-2022
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Setup python 3.10
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.10'
+
+      - name: Install packages
+        run:
+          choco install pkgconfiglite ninja
+
+      - name: Install python modules
+        run: pip3 install meson
+
+      - name: Setup MSVC compiler
+        uses: bus1/cabuild/action/msdevshell@v1
+        with:
+          architecture: x64
+
+      - name: Install dependencies
+        uses: kiwix/kiwix-build/actions/dl_deps_archive@main
+        with:
+          target_platform: win-x86_64-static
+
+      - name: Compile
+        shell: cmd
         run: |
-          curl https://codecov.io/bash -o codecov.sh
-          bash codecov.sh -n osx_native_dyn -Z
-          rm codecov.sh
+          set PKG_CONFIG_PATH=%cd%\BUILD_win-amd64\INSTALL\lib\pkgconfig
+          set CPPFLAGS=-I%cd%\BUILD_win-amd64\INSTALL\include
+          meson.exe setup . build -Dwerror=false --default-library=static --buildtype=release
+          cd build
+          ninja.exe
+
+      - name: Test
+        shell: cmd
+        run: |
+          cd build
+          meson.exe test --verbose
         env:
-          CODECOV_TOKEN: ${{ secrets.CODECOV_TOKEN }}
+          WAIT_TIME_FACTOR_TEST: 10
 
   Linux:
     strategy:
       fail-fast: false
       matrix:
-        name:
-          - native_static
-          - native_dyn
-          - native_dyn_bionic
-          - android_arm
-          - android_arm64
-          - win32_static
-          - win32_dyn
+        target:
+          - linux-x86_64-static
+          - linux-x86_64-dyn
+          - android-arm
+          - android-arm64
+        image_variant: ['jammy']
         include:
-          - name: native_static
-            target: native_static
-            image_variant: xenial
+          - target: linux-x86_64-static
             lib_postfix: '/x86_64-linux-gnu'
-          - name: native_dyn
-            target: native_dyn
-            image_variant: xenial
+            arch_name: linux-x86_64
+            run_test: true
+            coverage: true
+          - target: linux-x86_64-dyn
             lib_postfix: '/x86_64-linux-gnu'
-          - name: native_dyn_bionic
-            target: native_dyn
-            image_variant: bionic
-            lib_postfix: '/x86_64-linux-gnu'
-          - name: android_arm
-            target: android_arm
-            image_variant: xenial
-            lib_postfix: '/x86_64-linux-gnu'
-          - name: android_arm64
-            target: android_arm64
-            image_variant: xenial
-            lib_postfix: '/x86_64-linux-gnu'
-          - name: win32_static
-            target: win32_static
-            image_variant: f31
-            lib_postfix: '64'
-          - name: win32_dyn
-            target: win32_dyn
-            image_variant: f31
-            lib_postfix: '64'
+            arch_name: linux-x86_64
+            run_test: true
+            coverage: true
+          - target: android-arm
+            lib_postfix: '/arm-linux-androideabi'
+            arch_name: arm-linux-androideabi
+            run_test: false
+            coverage: false
+          - target: android-arm64
+            lib_postfix: '/aarch64-linux-android'
+            arch_name: aarch64-linux-android
+            run_test: false
+            coverage: false
     env:
       HOME: /home/runner
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-22.04
     container:
-      image: "kiwix/kiwix-build_ci:${{matrix.image_variant}}-26"
+      image: "ghcr.io/kiwix/kiwix-build_ci_${{matrix.image_variant}}:2025-06-07"
     steps:
-    - name: Extract branch name
-      shell: bash
-      run: echo "##[set-output name=branch;]$(echo ${GITHUB_REF#refs/heads/})"
-      id: extract_branch
     - name: Checkout code
-      shell: python
-      run: |
-        from subprocess import check_call
-        from os import environ
-        command = [
-          'git', 'clone',
-          'https://github.com/${{github.repository}}',
-          '--depth=1',
-          '--branch', '${{steps.extract_branch.outputs.branch}}'
-        ]
-        check_call(command, cwd=environ['HOME'])
-    - name: Install deps
-      shell: bash
-      run: |
-        ARCHIVE_NAME=deps2_${OS_NAME}_${{matrix.target}}_kiwix-lib.tar.xz
-        wget -O- http://tmp.kiwix.org/ci/${ARCHIVE_NAME} | tar -xJ -C /home/runner
+      uses: actions/checkout@v4
+    - name: Install dependencies
+      uses: kiwix/kiwix-build/actions/dl_deps_archive@main
+      with:
+        target_platform: ${{ matrix.target }}
     - name: Compile
       shell: bash
       run: |
         meson --version
-        if [[ "${{matrix.target}}" =~ .*_dyn ]]; then
+        if [[ "${{matrix.target}}" =~ .*-dyn ]]; then
           MESON_OPTION="--default-library=shared"
         else
           MESON_OPTION="--default-library=static"
         fi
-        if [[ "${{matrix.target}}" =~ native_.* ]]; then
-          MESON_OPTION="$MESON_OPTION -Db_coverage=true"
+        if [ -e "${{env.HOME}}/BUILD_${{matrix.arch_name}}/meson_cross_file.txt" ]; then
+          MESON_OPTION="$MESON_OPTION --cross-file ${{env.HOME}}/BUILD_${{matrix.arch_name}}/meson_cross_file.txt"
         else
-          MESON_OPTION="$MESON_OPTION --cross-file $HOME/BUILD_${{matrix.target}}/meson_cross_file.txt"
+          MESON_OPTION="$MESON_OPTION -Db_coverage=true"
         fi
-        if [[ "${{matrix.target}}" =~ android_.* ]]; then
-          MESON_OPTION="$MESON_OPTION -Dandroid=true"
+        if [[ "${{matrix.target}}" =~ android-.* ]]; then
+          MESON_OPTION="$MESON_OPTION -Dstatic-linkage=true"
         fi
-        cd $HOME/libkiwix
         meson . build ${MESON_OPTION}
         cd build
         ninja
       env:
-        PKG_CONFIG_PATH: "/home/runner/BUILD_${{matrix.target}}/INSTALL/lib/pkgconfig:/home/runner/BUILD_${{matrix.target}}/INSTALL/lib${{matrix.lib_postfix}}/pkgconfig"
-        CPPFLAGS: "-I/home/runner/BUILD_${{matrix.target}}/INSTALL/include"
+        PKG_CONFIG_PATH: "/home/runner/BUILD_${{matrix.arch_name}}/INSTALL/lib/pkgconfig:/home/runner/BUILD_${{matrix.arch_name}}/INSTALL/lib${{matrix.lib_postfix}}/pkgconfig"
+        CPPFLAGS: "-I/home/runner/BUILD_${{matrix.arch_name}}/INSTALL/include"
     - name: Test
-      if: startsWith(matrix.target, 'native_')
+      if: matrix.run_test
       shell: bash
       run: |
-        cd $HOME/libkiwix/build
+        cd build
         meson test --verbose
-        ninja coverage
+        if [[ "${{matrix.coverage}}" = "true" ]]; then
+          ninja coverage
+        fi
       env:
-        LD_LIBRARY_PATH: "/home/runner/BUILD_${{matrix.target}}/INSTALL/lib:/home/runner/BUILD_${{matrix.target}}/INSTALL/lib${{matrix.lib_postfix}}"
+        LD_LIBRARY_PATH: "/home/runner/BUILD_${{matrix.arch_name}}/INSTALL/lib:/home/runner/BUILD_${{matrix.arch_name}}/INSTALL/lib${{matrix.lib_postfix}}"
         SKIP_BIG_MEMORY_TEST: 1
+
     - name: Publish coverage
-      shell: bash
-      run: |
-        cd $HOME/libkiwix
-        curl https://codecov.io/bash -o codecov.sh
-        bash codecov.sh -n "${OS_NAME}_${{matrix.target}}" -Z
-        rm codecov.sh
-      if: startsWith(matrix.target, 'native_') && matrix.image_variant == 'xenial'
-      env:
-        CODECOV_TOKEN: ${{ secrets.CODECOV_TOKEN }}
+      if: matrix.coverage
+      uses: codecov/codecov-action@v3
+      with:
+        token: ${{ secrets.CODECOV_TOKEN }}

.github/workflows/package.yml

@@ -1,15 +1,29 @@
 name: Packages
-on: [push, pull_request]
+
+on:
+  pull_request:
+  push:
+    branches:
+      - main
+  release:
+    types: [published]
 
 jobs:
   build-deb:
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-22.04
     strategy:
       fail-fast: false
       matrix:
-        distro: [ubuntu-hirsute, ubuntu-groovy, ubuntu-focal, ubuntu-bionic]
+        distro:
+#          - debian-unstable
+#          - debian-trixie
+#          - debian-bookworm
+#          - debian-bullseye
+          - ubuntu-noble
+          - ubuntu-jammy
+
     steps:
-      - uses: actions/checkout@v2
+      - uses: actions/checkout@v4
 
       # Determine which PPA we should upload to
       - name: PPA
@@ -17,71 +31,81 @@ jobs:
         run: |
           if [[ $REF == refs/tags* ]]
           then
-            echo "::set-output name=ppa::kiwixteam/release"
+            echo "ppa=kiwixteam/release" >> $GITHUB_OUTPUT
           else
-            echo "::set-output name=ppa::kiwixteam/dev"
+            echo "ppa=kiwixteam/dev" >> $GITHUB_OUTPUT
           fi
         env:
           REF: ${{ github.ref }}
 
-      - uses: legoktm/gh-action-auto-dch@master
+      - uses: legoktm/gh-action-auto-dch@main
         with:
           fullname: Kiwix builder
           email: release+launchpad@kiwix.org
           distro: ${{ matrix.distro }}
 
-      - uses: legoktm/gh-action-build-deb@ubuntu-hirsute
-        if: matrix.distro == 'ubuntu-hirsute'
-        name: Build package for ubuntu-hirsute
-        id: build-ubuntu-hirsute
+#      - uses: legoktm/gh-action-build-deb@debian-unstable
+#        if: matrix.distro == 'debian-unstable'
+#        name: Build package for debian-unstable
+#        id: build-debian-unstable
+#        with:
+#          args: --no-sign
+#
+#      - uses: legoktm/gh-action-build-deb@b47978ba8498dc8b8153cc3b5f99a5fc1afa5de1 # pin@debian-trixie
+#        if: matrix.distro == 'debian-trixie'
+#        name: Build package for debian-trixie
+#        id: build-debian-trixie
+#        with:
+#          args: --no-sign
+#
+#      - uses: legoktm/gh-action-build-deb@1f4e86a6bb34aaad388167eaf5eb85d553935336 # pin@debian-bookworm
+#        if: matrix.distro == 'debian-bookworm'
+#        name: Build package for debian-bookworm
+#        id: build-debian-bookworm
+#        with:
+#          args: --no-sign
+#
+#      - uses: legoktm/gh-action-build-deb@084b4263209252ec80a75d2c78a586192c17f18d # pin@debian-bullseye
+#        if: matrix.distro == 'debian-bullseye'
+#        name: Build package for debian-bullseye
+#        id: build-debian-bullseye
+#        with:
+#          args: --no-sign
+
+      - uses: legoktm/gh-action-build-deb@9114a536498b65c40b932209b9833aa942bf108d # pin@ubuntu-noble
+        if: matrix.distro == 'ubuntu-noble'
+        name: Build package for ubuntu-noble
+        id: build-ubuntu-noble
         with:
           args: --no-sign
           ppa: ${{ steps.ppa.outputs.ppa }}
 
-      - uses: legoktm/gh-action-build-deb@ubuntu-groovy
-        if: matrix.distro == 'ubuntu-groovy'
-        name: Build package for ubuntu-groovy
-        id: build-ubuntu-groovy
+      - uses: legoktm/gh-action-build-deb@ubuntu-jammy
+        if: matrix.distro == 'ubuntu-jammy'
+        name: Build package for ubuntu-jammy
+        id: build-ubuntu-jammy
         with:
           args: --no-sign
           ppa: ${{ steps.ppa.outputs.ppa }}
 
-      - uses: legoktm/gh-action-build-deb@ubuntu-focal
-        if: matrix.distro == 'ubuntu-focal'
-        name: Build package for ubuntu-focal
-        id: build-ubuntu-focal
-        with:
-          args: --no-sign
-          ppa: ${{ steps.ppa.outputs.ppa }}
-
-      - uses: legoktm/gh-action-build-deb@ubuntu-bionic
-        if: matrix.distro == 'ubuntu-bionic'
-        name: Build package for ubuntu-bionic
-        id: build-ubuntu-bionic
-        with:
-          args: --no-sign
-          ppa: ${{ steps.ppa.outputs.ppa }}
-
-      - uses: actions/upload-artifact@v2
+      - uses: actions/upload-artifact@v4
         with:
           name: Packages for ${{ matrix.distro }}
           path: output
 
-      - uses: legoktm/gh-action-dput@master
+      - uses: legoktm/gh-action-dput@main
         name: Upload dev package
-        # Only upload on pushes to master
-        if: github.event_name == 'push' && github.event.ref == 'refs/heads/master' && startswith(matrix.distro, 'ubuntu-')
+        # Only upload on pushes to main
+        if: github.event_name == 'push' && github.event.ref == 'refs/heads/main' && startswith(matrix.distro, 'ubuntu-')
         with:
           gpg_key: ${{ secrets.LAUNCHPAD_GPG }}
           repository: ppa:kiwixteam/dev
           packages: output/*_source.changes
 
-      - uses: legoktm/gh-action-dput@master
+      - uses: legoktm/gh-action-dput@main
         name: Upload release package
-        # Only upload on pushes to master or tag
-        if: github.event_name == 'push' && startsWith(github.event.ref, 'refs/tags') && startswith(matrix.distro, 'ubuntu-')
+        if: github.event_name == 'release' && startswith(matrix.distro, 'ubuntu-')
         with:
           gpg_key: ${{ secrets.LAUNCHPAD_GPG }}
           repository: ppa:kiwixteam/release
           packages: output/*_source.changes
-

.gitignore

@@ -5,3 +5,5 @@ subprojects/googletest-release*
 build/
 .vscode/
 builddir/
+.cache/
+.clangd/

.readthedocs.yaml

@@ -0,0 +1,21 @@
+# Read the Docs configuration file
+# See https://docs.readthedocs.io/en/stable/config-file/v2.html for details
+
+# Required
+version: 2
+
+# Set the version of Python and other tools you might need
+build:
+  os: ubuntu-22.04
+  tools:
+    python: "3.11"
+
+# Build documentation in the docs/ directory with Sphinx
+sphinx:
+  configuration: docs/conf.py
+
+# We recommend specifying your dependencies to enable reproducible builds:
+# https://docs.readthedocs.io/en/stable/guides/reproducible-builds.html
+python:
+  install:
+  - requirements: docs/requirements.txt

ChangeLog

@@ -1,7 +1,288 @@
+libkiwix 14.1.1
+===============
+
+* Server:
+  - Fix regression for kiwix-serve --nosearchbar (@veloman-yunkan #1250)
+  - Avoid results content interpretation... crash in fulltext search (@vighnesh-sawant #1241)
+  - Fix for intermittent /content/blank.html errors (@veloman-yunkan #1249)
+
+libkiwix 14.1.0
+===============
+
+ * Server:
+  - Viewer detects & tracks intrapage navigation anchors too (@veloman-yunkan #1213)
+  - Add support for catalog only mode (@veloman-yunkan #1219)
+  - Add API which returns server access url (@vighnesh-sawant #1234)
+  - Fix chrome searchbar placeholder text overflow (@aditii2712 #1185)
+  - Fix magnet link queryStyring (@rgaudin #1160)
+  - Improve chrome printing stylesheet (@kelson42 #1202)
+  - Default white background (@kelson42 #1205)
+
+ * Other:
+  - Switched to the new libzim illustrations API (@veloman-yunkan #1226)
+  - Stop building Windows with DEBUG symbols in CI (@kelson42 #1165)
+  - Update many things in the CI/CD (@kelson42 #1203 #1194 #1209 #1207 #1235)
+  - Requires now libzim 9.4.0 (@kelson42 #1231)
+  - Fix compilation for FreeBSD (@OICe2 #1173 #1174)
+  - Wait up to 1s to let aria2c to start before complaining (@kelson42 #1169)
+
+libkiwix 14.0.0
+===============
+
+ * Server:
+  - Support of IPv6 (@veloman-yunkan @aryanA101a #1074 #1093)
+  - Better public IP configuration/detection (@sgourdas #1132)
+  - Fix API errors in catalog searches if Xapian keyword in used (@veloman-yunkan #1137)
+  - Clearly define which Web browsers are supported (@kelson42 @rgaudin @jaifroid @benoit74 #1132)
+  - Improve welcome page download buttons (@veloman-yunkan #1094)
+  - Better handling of external (non-HTTP) links (@veloman-yunkan #1123)
+  - Fix book illustration size on welcome page to 48x48 pixels (@veloman-yunkan #1127)
+  - Remove "Multiple Languages" in language filter (@veloman-yunkan #1098)
+  - Stop transforming tags casing (@kelson42 @veloman-yunkan #1079 #1121)
+  - ZIM file size consistently advertised in MiB (@harsha-mangena #1132)
+  - Few new supported languages in the filter (@kelson42 #1080)
+  - Improve accesskeys (@kelson42 #1075)
+  - Add OpenSearch <link> to head of pages (@kelson42 #1070)
+ * Compilation/Packaging:
+  - Multiple fixes around deb packaging (@kelson42 #1108 #1114 #1135)
+  - Generating of libkiwix.pc via Meson (@veloman-yunkan #1133)
+  - Native Windows CI/CD (@mgautierfr @kelson42 #1113 #1125)
+  - Better check (maximum) libzim version (@kelson42 #1124)
+  - Multiple automated tests improvements (@veloman-yunkan #1068 #1067)
+ * Other:
+  - Deleted supported env. variable `$KIWIX_DATA_DIR` and `kiwix::getDataDirectory()` (@sgourdas #1107)
+  - New string slugification for filenames (@shaopenglin #1105)
+  - Multiple improvements around aria2c download mgmt. (@veloman-yunkan #1097)
+
+libkiwix 13.1.0
+===============
+
+ * Server:
+  - Properly translated error pages (@veloman-yunkan #1032)
+  - Properly translated search result page (@veloman-yunkan #1046)
+  - Default UI language is resolved in frontend (@veloman-yunkan #1044)
+  - Better support of older Web browsers by polyfilling replaceAll() (@veloman-yunkan #1054)
+ * New API to migrate bookmarks between books (@mgautierfr #1043)
+ * Fixed compilation on Haiku OS (@Begasus #1048)
+
+libkiwix 13.0.0
+===============
+
+ * Server:
+  - Improved look & feel of kiwix-serve UI (@veloman-yunkan #917 #1021)
+  - Increase tolerance to malformed (control characters) ZIM entry titles (@veloman-yunkan #1023)
+  - API allowing to filter many categories at once (@juuz0 #974)
+  - Cookie-less user language control (@veloman-yumkan #997)
+  - Hack to fix Mirrorbrain based broken magnet URLs (@rgaudin #1001)
+ * Fix handling of books with 'Name' metadata with dots (@mgautier #1016)
+ * New method beautifyFileSize() to provide nice-looking book sizes (@vuuz0 #971)
+ * Fix a few missing includes (@mgautierfr #978)
+ * New functions to read - kiwix-serve - languages and categories streams (@juuz0 #967)
+ * Add support of Fon language (@kelson42 #1013)
+ * C++17 code base compliancy (@mgautierfr #996)
+ * Use everywhere std::shared_ptr in place of raw pointer (@mgautierfr #991)
+ * Do not use [[nodiscard]] attribute on compiler not supporting it (@mgautierfr #1003)
+ * Add a non minified version of autoComplete.js (@mgautierfr #1008)
+ * Multiple CI/CD improvements (@kelson42 #982)
+
+libkiwix 12.1.0
+===============
+
+ * Server:
+  - Introduce a `/nojs` endpoint to browse catalog and zim files with a browser without js (@juuz0 #897)
+  - Translate the viewer (@veloman-yunkan #871 #846)
+  - Display `mul` on tile when zim is multi-languages (@juuz0 #934)
+  - Suggestion links point to the `/content` endpoint (@veloman-yunkan #862)
+  - Correctly compress web fonts in http answers (@kelson42 #856)
+  - Correctly encode link in suggestions (@veloman-yunkan #859 #860 #963)
+  - Correctly encode url redirection (@veloman-yunkan #866 #890)
+  - Properly handle user language, through cookies and http headers (@veloman-yunkan #849 #869)
+  - Fix url encoding (@veloman-yunkan #870)
+  - Fix viewer for viewer for SeaMonkey (@veloman-yunkan #887)
+  - Make the downloader threadsafe (@mgautierfr #886)
+  - Add RSS feed in the main page (pointing to the catalog) (@juuz0 #882 #920)
+  - Correctly set the mimetype for json and ico (@veloman-yunkan #892)
+  - `count=-1` correspond to unlimited count (instead of 0) (@veloman-yunkan #894)
+  - Keep the navigation bar on top (@juuz0 #896)
+  - Make the viewer's iframe "safe" (@veloman-yunkan #906 #930)
+  - Correctly escape search link in XML Opds output (@veloman-yunkan #936)
+  - Store values needed for the viewer js in the url fragment instead of the query string (@juuz0 #907)
+  - Get rid of legacy OPDS API usage in the viewer (@veloman-yunkan #939)
+  - Fix charset encoding declaration in OPDS response MIME types (@veloman-yunkan #942)
+  - Fix PDF in the viewer (@veloman-yunkan #940)
+  - Fix external links handling in the viewer (@veloman-yunkan #959)
+  - Add tests of searching with accents (@mgautierfs #954)
+ * Fix handling of missing illustration in the book (@veloman-yunkan #961)
+ * Add support for multi languages zim files (@veloman-yunkan #904)
+ * Fix includes for openbsd (@bentley #949)
+ * Fix pathes in git to allow git clone on Windows (@adamlamar #868)
+ * Switch to `main` as principal branch (instead of `master`) (@kelson42)
+ * Remove libkiwix android publisher from the repository (@kelson42 #884)
+ * Various fixes of meson and CI. (@mgautierfr @kelson42)
+
+libkiwix 12.0.0
+===============
+
+ * [API Break] Remove wrapper around libzim (@mgautierfr #789)
+ * Allow kiwix-serve to use custom resource files (@veloman-yunkan #779)
+ * Properly handle searchProtocolPrefix when rendering search result (@veloman-yunkan #823)
+ * Prevent search on multi language content (@veloman-yunkan #838)
+ * Use new `zim::Archive::getMediaCount` from libzim (@mgautierfr #836)
+ * Catalog:
+  - Include tags in free text catalog search (@veloman-yunkan #802)
+  - Illustration's url is based on book's uuid (@veloman-yunkan #804)
+  - Cleanup of the opds-dumper (@veloman-yunkan #829)
+  - Allow filtering of catalog content using multiple languages (@veloman-yunkan #841)
+  - Make opds-dumper respect the namemapper (@mgautierfr #837)
+ * Server:
+  - Correctly handle `\` in suggestion json generation (@veloman-yunkan #843)
+  - Better http caching (@veloman-yunkan #833)
+  - Make `/suggest` endpoint thread-safe (@veloman-yunkan #834)
+  - Better redirection of main page (@veloman-yunkan #827)
+  - Remove jquery (@mgautierfr @juuz0 #796)
+  - Better Viewer of zim content :
+    . Introduce `/content` endpoints (@veloman-yunkan #806)
+    . Switch to iframe based content viewer (@veloman-yunkan #716)
+  - Optimised design of the welcome page:
+    . Alignement (@juuz0 @kelson42 #786)
+    . Exit download modal on pressing escape key (@juzz0 #800)
+    . Add favicon for different devices (@juzz0 #805)
+    . Fix auto hidding of the toolbar (@veloman-yunkan #821)
+    . Allow user to filter books by tags in the front page (@juuz0 #711)
+ * CI :
+   - Trigger CI on pull_request (@kelson42 #791)
+   - Drop Ubuntu Impish packaging (@legoktm #825)
+   - Add Ubuntu Kinetic packaging (@legoktm #801)
+ * Testing:
+   - Test ICULanguageInfo (@veloman-yunkan #795)
+   - Introduce fake `test` language to test i18n (@veloman-yunkan #848)
+ * Fix documentation (@kelson42 #816)
+ * Udpate translation (#787 #839 #847)
+
+libkiwix 11.0.0
+===============
+
+ * [server] Add support for internationalization (@veloman-yunkan #679)
+ * [server] Use gzip compression instead of deflat (mgautierfr #757)
+ * [server] Version the static resources. This allow better invalidating
+   browser cache when resources are changed (@veloman-yunkan #712)
+ * [server|front] Use integer to query the host for page length (@juuz0 #772)
+ * [server] Improve multizim search API:
+   - Improvement of the cache system
+   - Better API to select on which books to search in.
+   - SysAdmin is now able to limit the number of book we search in for a multizim search
+ * [server] Introduce a opensearch API for multizim fulltext search
+ * [wrapper] Remove java wrapper
+ * Testing:
+   - Testing of search result pages content (@veloman-yunkan #765)
+   - Better testing structure of xml search result (@veloman-yunkan #780)
+
+libkiwix 10.1.1
+===============
+
+ * Correctly detect the number of article for older zims (<=6) (@mgautier #743)
+ * [server] Fix fulltext search (@mgautierfr #724)
+ * [server][internal] New way to build Error message (@veloman-yunkan #732 #738 #744)
+ * Fix CI (@mgautierfr #736)
+
+libkiwix 10.1.0
+===============
+
+ This release is an important one as it fixes a Xss vulnerability introduced
+ in libkiwix 10.0.0
+
+ * [SECURITY] Fix a Xss attack vulnerability (introduced in 10.0.0) (@juuz0 #721)
+ * [server] Add a option to set a limit on the number of connexion per IP (@kelson42 #700)
+ * [server] Do not display a lang tag in the UI if the book has no language (@juuz0 #706)
+ * [server] Add the book title associated to a search results (@thavelick #705, @mgautierfr #718)
+ * Add `dc:issued` to opds output stream (@veloman-yunkan #715)
+ * Add handling of several languages not provided by ICU (@juuz0 #701)
+ * [server] Add a caching system for search and suggestion (@maneeshpm #620)
+ * Fix cross-compilation (@kelson42 #703)
+ * Add unit-testing of suggestions and error pages (@veloman-yunkan #709 #710 #727)
+ * Better testing system of html response (@veloman-yunkan #725)
+
+libkiwix 10.0.1
+===============
+
+ * [server] The catalog search interpret `count=0` as no limit.
+   This was the case for a long time. This was changed unintentionally
+   (@veloman-yunkan #686)
+ * [server] Correctly generere a human friendly title in the server frontend.
+   (@juuz0 #687, @kelson42 #689)
+ * [server] Fix download button if there is no url do download from.
+   (@juuz0 #691)
+ * Add non-minified isotope.pkdg.js
+   Needed for debian packaging as we need the source and minified version is
+   not the source (@legoktm #693)
+ * [server] Add a tooltip with the full language for the lang tag.
+ * CI fixes (@kelson42 @legoktm)
+
 libkiwix 10.0.0
 ===============
 
- * ...
+ This release is huge release.
+ The project has been renamed to libkiwix, it is more coherent with the library name.
+
+ * Server front page :
+  - Use js in the front page to display the available book,
+     using the OPDS stream as source. The front page is now populated only with
+     the visible books and user can search for books. (@MananJethwany #530, #541, #534)
+     (@kelson42 #628)
+  - Revamp css (@MananJethwany #559)
+  - Correctly Convert 3iso language code to 2iso (@juuz0 #672)
+
+ * Server suggestions search :
+  - Add pagination for suggestion search (@maneeshpm #591)
+  - Fix suggestion system (@MananJethwany #498)
+  - Provide the kind and path (when adapted) to the suggestion answer (@MananJethwany #464)
+  - The displayed suggestion have now highligth on the searched terms (@maneeshpm #505)
+  - Properly handle html encoding of suggestions (@veloman-yunkan #458)
+
+ * Server improvements :
+  - Remove meta endpoints (@mgautier #669)
+  - Add raw endpoints to get the raw content of a zim (@mgautierfr #646)
+  - Add details on 404 error pages (@soumyankar #490)
+  - Fix headbar insertion when `<head>` tag has attributes (@kelson42 #440)
+  - Better headbar insertion (after charset definition) (@kelson42 #442)
+
+ * New OPDS Stream v2 :
+  - Add a list of categories (@veloman-yunkan)
+  - Support for partial entries (@veloman-yunkan #602)
+  - Support multiple icons size in the OPDS stream (@veloman-yunkan #577 #630)
+  - Add language endpoint to catalog (@veloman-yunkan #553)
+  - Add illustration API to get the illustration of a book (@mgautierfr #645)
+  - OPDS search can now filter books by category (@veloman-yunkan #459)
+
+ * Library improvements :
+  - Allow the libray to be live reloaded when the library.xml changes (@veloman-yunkan #636)
+  - Properly handle removing of book from the library (@veloman-yunkan #485)
+  - Use xapian to search for books in the library (@veloman-yunkan #460, #488)
+
+ * Added methods/functions :
+  - Fix `fileExist` and introduce `fileReadable` (@juuz0 #668)
+  - Add `getVersions` and `printVersions` functions (@kelson42 #665)
+  - Add `getNetworkInterfaces()` and `getBestPublicIP()` functions (@juuz0 #622)
+  - Add `get_zimid()` method to the search result (@maneeshpm #510)
+
+ * Various improvements :
+  - Better secret value for aria2c rpc (@juuz0 #666)
+  - Avoid duplicated Archive/Reader in the Searcher (@veloman-yunkan #648)
+  - Add basic documentation (@mgautierfr #640)
+  - Do not use Reader internally (@maneeshpm #536 #576)
+  - Remove dependency headers from our public headers (@mgautierfr #574)
+  - Downloader now don't write metalink on the filesystem (@kelson42 #502)
+  - Support opening a zim file using a fd (@veloman-yukan #429)
+  - Use C++11 std::thread instead of pthread (@mgautierfr #445)
+  - [READER] Do not crash if zim file has no `Counter` metadata (@mgautierfr #449)
+  - Ensure libzim dependency is compiled with xapian (@mgautierfr #434)
+  - Support video and audio mimetype in `getMediaCount` (@kelson42 #439)
+  - Better parsing of the counterMap (@kelson42 #437)
+  - Adapt libkiwix to libzim 7.0.0 (@mgautierfr #428)
+  - Remove deprecated methods (@mgautierfr)
+  - CI: Build package for Ubuntu Hirsute, Impish and Jammy (@legoktm #431 #568) and remove Groovy
+  - Fix compilation for FreeBSD (@swills g#432)
+  - Many fixes and improvement (@MananJethwany, @maneeshpm, @veloman-yunkan, @mgautierfr)
 
 kiwix-lib 9.4.1
 ===============

README.md

@@ -5,10 +5,12 @@ The Libkiwix provides the [Kiwix](https://kiwix.org) software suite
 core. It contains the code shared by all Kiwix ports (Windows,
 GNU/Linux, macOS, Android, iOS, ...).
 
+[![Release](https://img.shields.io/github/v/tag/kiwix/libkiwix?label=release&sort=semver)](https://download.kiwix.org/release/libkiwix/)
 [![Repositories](https://img.shields.io/repology/repositories/libkiwix?label=repositories)](https://github.com/kiwix/libkiwix/wiki/Repology)
-[![Build Status](https://github.com/kiwix/libkiwix/workflows/CI/badge.svg?query=branch%3Amaster)](https://github.com/kiwix/libkiwix/actions?query=branch%3Amaster)
+[![Build Status](https://github.com/kiwix/libkiwix/actions/workflows/ci.yml/badge.svg?branch=main)](https://github.com/kiwix/libkiwix/actions?query=branch%3Amain)
+[![Doc](https://readthedocs.org/projects/libkiwix/badge/?style=flat)](https://libkiwix.readthedocs.org/en/latest/?badge=latest)
 [![CodeFactor](https://www.codefactor.io/repository/github/kiwix/libkiwix/badge)](https://www.codefactor.io/repository/github/kiwix/libkiwix)
-[![Codecov](https://codecov.io/gh/kiwix/libkiwix/branch/master/graph/badge.svg)](https://codecov.io/gh/kiwix/libkiwix)
+[![Codecov](https://codecov.io/gh/kiwix/libkiwix/branch/main/graph/badge.svg)](https://codecov.io/gh/kiwix/libkiwix)
 [![License: GPL v3](https://img.shields.io/badge/License-GPLv3-blue.svg)](https://www.gnu.org/licenses/gpl-3.0)
 
 Disclaimer
@@ -22,9 +24,9 @@ with the Libkiwix compilation itself, we recommend to have a look to
 Preamble
 --------
 
-Although the Libkiwix can be (cross-)compiled on/for many sytems, the
+Although the Libkiwix can be (cross-)compiled on/for many systems, the
 following documentation explains how to do it on POSIX ones. It is
-primarly thought for GNU/Linux systems and has been tested on recent
+primarily thought for GNU/Linux systems and has been tested on recent
 releases of Ubuntu and Fedora.
 
 Dependencies
@@ -52,7 +54,7 @@ The following dependency needs to be available at runtime:
 These dependencies may or may not be packaged by your operating
 system. They may also be packaged but only in an older version. The
 compilation script will tell you if one of them is missing or too old.
-In the worse case, you will have to download and compile bleeding edge
+In the worst case, you will have to download and compile bleeding edge
 version by hand.
 
 If you want to install these dependencies locally, then use the
@@ -91,6 +93,41 @@ Meson. If you want statically linked libraries, you can add
 
 Depending of you system, `ninja` may be called `ninja-build`.
 
+The android wrapper uses deprecated methods of libkiwix so it cannot be compiled
+with `werror=true` (the default). So you must pass `-Dwerror=false` to meson:
+
+```bash
+meson . build -Dwrapper=android -Dwerror=false
+ninja -C build
+```
+
+Static files compilation
+------------------------
+
+Libkiwix has a few static files 'compiled' within the binary
+code. This is mostly Javascript/HTML/pictures necessary for the HTTP
+daemon.
+
+These static files are available in the `static` directory and are
+compiled by custom Python code available in this repository `scripts`
+directory. This happens automatically at compilation time without any
+additional command to run.
+
+To avoid HTTP caching issues, the URLs (to the static content) are
+appended with a `cacheid` parameter (this is called "cache
+busting"). This `cacheid` value derived from the
+[sha1sum](https://en.wikipedia.org/wiki/Sha1sum) of each targeted
+static file. As a consequence, each time you change a static file, the
+corresponding `cacheid` value will change.
+
+To properly test this feature, this `cacheid` needs to be added
+manually to the automated tests and has to be commited. After
+modifying the needed static file, [run the automated
+tests](#Testing). They will fail, but the inspection of the testing
+log will give you the new `cacheid` value(s). Finally update
+`test/server.cpp` with the appropriate `cacheid` value(s) which have
+changed.
+
 Testing
 -------
 
@@ -114,7 +151,7 @@ where you want to install the libraries. After the installation
 succeeded, you may need to run `ldconfig` (as `root`).
 
 Uninstallation
-------------
+--------------
 
 If you want to uninstall the Kiwix library:
 ```bash
@@ -124,6 +161,55 @@ ninja -C build uninstall
 Like for the installation, you might need to run the command as `root`
 (or using `sudo`).
 
+Custom Index Page
+-----------------
+
+to use custom welcome page mention `customIndexPage` argument in `kiwix::internalServer()` or use `kiwix::server->setCustomIndexTemplate()`.
+(note - while using custom html file please mention all external links as absolute path.)
+
+to create a HTML template with custom JS you need to have a look at various OPDS based endpoints as mentioned [here](https://wiki.kiwix.org/wiki/OPDS) to load books.
+
+To use JS provided by kiwix-serve you can use the following template to start with ->
+
+```
+<!DOCTYPE html>
+<html xmlns="http://www.w3.org/1999/xhtml">
+  <head>
+    <meta charset="UTF-8" />
+    <meta name="viewport" content="width=device-width,initial-scale=1" />
+    <title><-- Custom Tittle --></title>
+    <script src="{{root}}/skin/isotope.pkgd.min.js" defer></script>
+    <script src="{{root}}/skin/iso6391To3.js"></script>
+    <script type="text/javascript" src="{{root}}/skin/index.js" defer></script>
+  </head>
+  <body>
+  </body>
+</html>
+```
+
+- To get books listed using `index.js` add - `<div class="book__list"></div>` under body tag.
+- To get number of books listed add - `<h3 class="kiwixHomeBody__results"></h3>` under body tag.
+- To add language select box add - `<select id="languageFilter"></select>` under body tag.
+- To add category select box add - `<select id="categoryFilter"></select>` under body tag.
+- To add search box for books use following form -
+    ```
+        <form id='kiwixSearchForm'>
+        <input type="text" name="q" placeholder="Search" id="searchFilter" class='kiwixSearch filter'>
+        <input type="submit" class="kiwixButton" value="Search"/>
+        </form>
+    ```
+
+
+If you compile manually Libmicrohttpd, you might need to compile it
+without GNU TLS, a bug here will impeach further compilation
+otherwise.
+
+If the compilation still fails, you might need to get a more recent
+version of a dependency than the one packaged by your Linux
+distribution. Try then with a source tarball distributed by the
+problematic upstream project or even directly from the source code
+repository.
+
 Troubleshooting
 ---------------
 
@@ -146,16 +232,6 @@ cp ninja ../bin
 cd ..
 ```
 
-If you compile manually Libmicrohttpd, you might need to compile it
-without GNU TLS, a bug here will empeach further compilation
-otherwise.
-
-If the compilation still fails, you might need to get a more recent
-version of a dependency than the one packaged by your Linux
-distribution. Try then with a source tarball distributed by the
-problematic upstream project or even directly from the source code
-repository.
-
 License
 -------
 

android-kiwix-lib-publisher/.gitignore

@@ -1,13 +0,0 @@
-*.iml
-.gradle
-/local.properties
-/.idea/caches
-/.idea/libraries
-/.idea/modules.xml
-/.idea/workspace.xml
-/.idea/navEditor.xml
-/.idea/assetWizardSettings.xml
-.DS_Store
-/build
-/captures
-.externalNativeBuild

android-kiwix-lib-publisher/build.gradle

@@ -1,25 +0,0 @@
-// Top-level build file where you can add configuration options common to all sub-projects/modules.
-
-buildscript {
-    repositories {
-        google()
-        jcenter()
-
-    }
-    dependencies {
-        classpath 'com.android.tools.build:gradle:3.4.1'
-        // NOTE: Do not place your application dependencies here; they belong
-        // in the individual module build.gradle files
-    }
-}
-
-allprojects {
-    repositories {
-        google()
-        jcenter()
-    }
-}
-
-task clean(type: Delete) {
-    delete rootProject.buildDir
-}

android-kiwix-lib-publisher/gradle.properties

@@ -1,15 +0,0 @@
-# Project-wide Gradle settings.
-# IDE (e.g. Android Studio) users:
-# Gradle settings configured through the IDE *will override*
-# any settings specified in this file.
-# For more details on how to configure your build environment visit
-# http://www.gradle.org/docs/current/userguide/build_environment.html
-# Specifies the JVM arguments used for the daemon process.
-# The setting is particularly useful for tweaking memory settings.
-org.gradle.jvmargs=-Xmx1536m
-# When configured, Gradle will run in incubating parallel mode.
-# This option should only be used with decoupled projects. More details, visit
-# http://www.gradle.org/docs/current/userguide/multi_project_builds.html#sec:decoupled_projects
-# org.gradle.parallel=true
-# Kotlin code style for this project: "official" or "obsolete":
-kotlin.code.style=official

android-kiwix-lib-publisher/gradle/wrapper/gradle-wrapper.properties

@@ -1,6 +0,0 @@
-#Wed Jun 19 15:28:39 BST 2019
-distributionBase=GRADLE_USER_HOME
-distributionPath=wrapper/dists
-zipStoreBase=GRADLE_USER_HOME
-zipStorePath=wrapper/dists
-distributionUrl=https\://services.gradle.org/distributions/gradle-5.1.1-all.zip

android-kiwix-lib-publisher/gradlew

@@ -1,172 +0,0 @@
-#!/usr/bin/env sh
-
-##############################################################################
-##
-##  Gradle start up script for UN*X
-##
-##############################################################################
-
-# Attempt to set APP_HOME
-# Resolve links: $0 may be a link
-PRG="$0"
-# Need this for relative symlinks.
-while [ -h "$PRG" ] ; do
-    ls=`ls -ld "$PRG"`
-    link=`expr "$ls" : '.*-> \(.*\)$'`
-    if expr "$link" : '/.*' > /dev/null; then
-        PRG="$link"
-    else
-        PRG=`dirname "$PRG"`"/$link"
-    fi
-done
-SAVED="`pwd`"
-cd "`dirname \"$PRG\"`/" >/dev/null
-APP_HOME="`pwd -P`"
-cd "$SAVED" >/dev/null
-
-APP_NAME="Gradle"
-APP_BASE_NAME=`basename "$0"`
-
-# Add default JVM options here. You can also use JAVA_OPTS and GRADLE_OPTS to pass JVM options to this script.
-DEFAULT_JVM_OPTS=""
-
-# Use the maximum available, or set MAX_FD != -1 to use that value.
-MAX_FD="maximum"
-
-warn () {
-    echo "$*"
-}
-
-die () {
-    echo
-    echo "$*"
-    echo
-    exit 1
-}
-
-# OS specific support (must be 'true' or 'false').
-cygwin=false
-msys=false
-darwin=false
-nonstop=false
-case "`uname`" in
-  CYGWIN* )
-    cygwin=true
-    ;;
-  Darwin* )
-    darwin=true
-    ;;
-  MINGW* )
-    msys=true
-    ;;
-  NONSTOP* )
-    nonstop=true
-    ;;
-esac
-
-CLASSPATH=$APP_HOME/gradle/wrapper/gradle-wrapper.jar
-
-# Determine the Java command to use to start the JVM.
-if [ -n "$JAVA_HOME" ] ; then
-    if [ -x "$JAVA_HOME/jre/sh/java" ] ; then
-        # IBM's JDK on AIX uses strange locations for the executables
-        JAVACMD="$JAVA_HOME/jre/sh/java"
-    else
-        JAVACMD="$JAVA_HOME/bin/java"
-    fi
-    if [ ! -x "$JAVACMD" ] ; then
-        die "ERROR: JAVA_HOME is set to an invalid directory: $JAVA_HOME
-
-Please set the JAVA_HOME variable in your environment to match the
-location of your Java installation."
-    fi
-else
-    JAVACMD="java"
-    which java >/dev/null 2>&1 || die "ERROR: JAVA_HOME is not set and no 'java' command could be found in your PATH.
-
-Please set the JAVA_HOME variable in your environment to match the
-location of your Java installation."
-fi
-
-# Increase the maximum file descriptors if we can.
-if [ "$cygwin" = "false" -a "$darwin" = "false" -a "$nonstop" = "false" ] ; then
-    MAX_FD_LIMIT=`ulimit -H -n`
-    if [ $? -eq 0 ] ; then
-        if [ "$MAX_FD" = "maximum" -o "$MAX_FD" = "max" ] ; then
-            MAX_FD="$MAX_FD_LIMIT"
-        fi
-        ulimit -n $MAX_FD
-        if [ $? -ne 0 ] ; then
-            warn "Could not set maximum file descriptor limit: $MAX_FD"
-        fi
-    else
-        warn "Could not query maximum file descriptor limit: $MAX_FD_LIMIT"
-    fi
-fi
-
-# For Darwin, add options to specify how the application appears in the dock
-if $darwin; then
-    GRADLE_OPTS="$GRADLE_OPTS \"-Xdock:name=$APP_NAME\" \"-Xdock:icon=$APP_HOME/media/gradle.icns\""
-fi
-
-# For Cygwin, switch paths to Windows format before running java
-if $cygwin ; then
-    APP_HOME=`cygpath --path --mixed "$APP_HOME"`
-    CLASSPATH=`cygpath --path --mixed "$CLASSPATH"`
-    JAVACMD=`cygpath --unix "$JAVACMD"`
-
-    # We build the pattern for arguments to be converted via cygpath
-    ROOTDIRSRAW=`find -L / -maxdepth 1 -mindepth 1 -type d 2>/dev/null`
-    SEP=""
-    for dir in $ROOTDIRSRAW ; do
-        ROOTDIRS="$ROOTDIRS$SEP$dir"
-        SEP="|"
-    done
-    OURCYGPATTERN="(^($ROOTDIRS))"
-    # Add a user-defined pattern to the cygpath arguments
-    if [ "$GRADLE_CYGPATTERN" != "" ] ; then
-        OURCYGPATTERN="$OURCYGPATTERN|($GRADLE_CYGPATTERN)"
-    fi
-    # Now convert the arguments - kludge to limit ourselves to /bin/sh
-    i=0
-    for arg in "$@" ; do
-        CHECK=`echo "$arg"|egrep -c "$OURCYGPATTERN" -`
-        CHECK2=`echo "$arg"|egrep -c "^-"`                                 ### Determine if an option
-
-        if [ $CHECK -ne 0 ] && [ $CHECK2 -eq 0 ] ; then                    ### Added a condition
-            eval `echo args$i`=`cygpath --path --ignore --mixed "$arg"`
-        else
-            eval `echo args$i`="\"$arg\""
-        fi
-        i=$((i+1))
-    done
-    case $i in
-        (0) set -- ;;
-        (1) set -- "$args0" ;;
-        (2) set -- "$args0" "$args1" ;;
-        (3) set -- "$args0" "$args1" "$args2" ;;
-        (4) set -- "$args0" "$args1" "$args2" "$args3" ;;
-        (5) set -- "$args0" "$args1" "$args2" "$args3" "$args4" ;;
-        (6) set -- "$args0" "$args1" "$args2" "$args3" "$args4" "$args5" ;;
-        (7) set -- "$args0" "$args1" "$args2" "$args3" "$args4" "$args5" "$args6" ;;
-        (8) set -- "$args0" "$args1" "$args2" "$args3" "$args4" "$args5" "$args6" "$args7" ;;
-        (9) set -- "$args0" "$args1" "$args2" "$args3" "$args4" "$args5" "$args6" "$args7" "$args8" ;;
-    esac
-fi
-
-# Escape application args
-save () {
-    for i do printf %s\\n "$i" | sed "s/'/'\\\\''/g;1s/^/'/;\$s/\$/' \\\\/" ; done
-    echo " "
-}
-APP_ARGS=$(save "$@")
-
-# Collect all arguments for the java command, following the shell quoting and substitution rules
-eval set -- $DEFAULT_JVM_OPTS $JAVA_OPTS $GRADLE_OPTS "\"-Dorg.gradle.appname=$APP_BASE_NAME\"" -classpath "\"$CLASSPATH\"" org.gradle.wrapper.GradleWrapperMain "$APP_ARGS"
-
-# by default we should be in the correct project dir, but when run from Finder on Mac, the cwd is wrong
-if [ "$(uname)" = "Darwin" ] && [ "$HOME" = "$PWD" ]; then
-  cd "$(dirname "$0")"
-fi
-
-exec "$JAVACMD" "$@"

android-kiwix-lib-publisher/gradlew.bat

@@ -1,84 +0,0 @@
-@if "%DEBUG%" == "" @echo off
-@rem ##########################################################################
-@rem
-@rem  Gradle startup script for Windows
-@rem
-@rem ##########################################################################
-
-@rem Set local scope for the variables with windows NT shell
-if "%OS%"=="Windows_NT" setlocal
-
-set DIRNAME=%~dp0
-if "%DIRNAME%" == "" set DIRNAME=.
-set APP_BASE_NAME=%~n0
-set APP_HOME=%DIRNAME%
-
-@rem Add default JVM options here. You can also use JAVA_OPTS and GRADLE_OPTS to pass JVM options to this script.
-set DEFAULT_JVM_OPTS=
-
-@rem Find java.exe
-if defined JAVA_HOME goto findJavaFromJavaHome
-
-set JAVA_EXE=java.exe
-%JAVA_EXE% -version >NUL 2>&1
-if "%ERRORLEVEL%" == "0" goto init
-
-echo.
-echo ERROR: JAVA_HOME is not set and no 'java' command could be found in your PATH.
-echo.
-echo Please set the JAVA_HOME variable in your environment to match the
-echo location of your Java installation.
-
-goto fail
-
-:findJavaFromJavaHome
-set JAVA_HOME=%JAVA_HOME:"=%
-set JAVA_EXE=%JAVA_HOME%/bin/java.exe
-
-if exist "%JAVA_EXE%" goto init
-
-echo.
-echo ERROR: JAVA_HOME is set to an invalid directory: %JAVA_HOME%
-echo.
-echo Please set the JAVA_HOME variable in your environment to match the
-echo location of your Java installation.
-
-goto fail
-
-:init
-@rem Get command-line arguments, handling Windows variants
-
-if not "%OS%" == "Windows_NT" goto win9xME_args
-
-:win9xME_args
-@rem Slurp the command line arguments.
-set CMD_LINE_ARGS=
-set _SKIP=2
-
-:win9xME_args_slurp
-if "x%~1" == "x" goto execute
-
-set CMD_LINE_ARGS=%*
-
-:execute
-@rem Setup the command line
-
-set CLASSPATH=%APP_HOME%\gradle\wrapper\gradle-wrapper.jar
-
-@rem Execute Gradle
-"%JAVA_EXE%" %DEFAULT_JVM_OPTS% %JAVA_OPTS% %GRADLE_OPTS% "-Dorg.gradle.appname=%APP_BASE_NAME%" -classpath "%CLASSPATH%" org.gradle.wrapper.GradleWrapperMain %CMD_LINE_ARGS%
-
-:end
-@rem End local scope for the variables with windows NT shell
-if "%ERRORLEVEL%"=="0" goto mainEnd
-
-:fail
-rem Set variable GRADLE_EXIT_CONSOLE if you need the _script_ return code instead of
-rem the _cmd.exe /c_ return code!
-if  not "" == "%GRADLE_EXIT_CONSOLE%" exit 1
-exit /b 1
-
-:mainEnd
-if "%OS%"=="Windows_NT" endlocal
-
-:omega

android-kiwix-lib-publisher/kiwixLibAndroid/.gitignore

@@ -1 +0,0 @@
-/build

android-kiwix-lib-publisher/kiwixLibAndroid/build.gradle

@@ -1,64 +0,0 @@
-apply plugin: 'com.android.library'
-apply plugin: 'maven'
-
-android {
-    compileSdkVersion 28
-    defaultConfig {
-        minSdkVersion 15
-        targetSdkVersion 28
-        versionCode 1
-        versionName "1.0"
-    }
-    buildTypes {
-        release {
-            minifyEnabled false
-            proguardFiles getDefaultProguardFile('proguard-android-optimize.txt'), 'proguard-rules.pro'
-        }
-    }
-}
-
-dependencies {
-    implementation 'com.getkeepsafe.relinker:relinker:1.3.1'
-}
-
-task writePom {
-  pom {
-    project {
-        groupId 'org.kiwix.kiwixlib'
-        artifactId 'kiwixlib'
-        version '10.0.0' + (System.env.KIWIXLIB_BUILDVERSION == null ? '' : '-'+System.env.KIWIXLIB_BUILDVERSION)
-        packaging 'aar'
-        name 'kiwixlib'
-        url 'https://github.com/kiwix/libkiwix'
-        licenses {
-            license {
-                name 'GPLv3'
-                url 'https://www.gnu.org/licenses/gpl-3.0.en.html'
-            }
-        }
-      developers {
-          developer {
-              id 'kiwix'
-              name 'kiwix'
-              email 'contact@kiwix.org'
-          }
-      }
-      scm {
-          connection 'https://github.com/kiwix/libkiwix.git'
-          developerConnection 'https://github.com/kiwix/libkiwix.git'
-          url 'https://github.com/kiwix/libkiwix'
-      }
-    }
-  }.withXml {
-    def dependenciesNode = asNode().appendNode('dependencies')
-
-    //Iterate over the implementation dependencies, adding a <dependency> node for each
-    configurations.implementation.allDependencies.each {
-      def dependencyNode = dependenciesNode.appendNode('dependency')
-      dependencyNode.appendNode('groupId', it.group)
-      dependencyNode.appendNode('artifactId', it.name)
-      dependencyNode.appendNode('version', it.version)
-    }
-  }.writeTo("$buildDir/pom.xml")
-}
-

android-kiwix-lib-publisher/kiwixLibAndroid/proguard-rules.pro

@@ -1,21 +0,0 @@
-# Add project specific ProGuard rules here.
-# You can control the set of applied configuration files using the
-# proguardFiles setting in build.gradle.
-#
-# For more details, see
-#   http://developer.android.com/guide/developing/tools/proguard.html
-
-# If your project uses WebView with JS, uncomment the following
-# and specify the fully qualified class name to the JavaScript interface
-# class:
-#-keepclassmembers class fqcn.of.javascript.interface.for.webview {
-#   public *;
-#}
-
-# Uncomment this to preserve the line number information for
-# debugging stack traces.
-#-keepattributes SourceFile,LineNumberTable
-
-# If you keep the line number information, uncomment this to
-# hide the original source file name.
-#-renamesourcefileattribute SourceFile

android-kiwix-lib-publisher/kiwixLibAndroid/src/main/AndroidManifest.xml

@@ -1,10 +0,0 @@
-<manifest xmlns:android="http://schemas.android.com/apk/res/android"
-    package="org.kiwix.kiwixlib">
-
-    <application
-        android:allowBackup="true"
-        android:supportsRtl="true">
-
-    </application>
-
-</manifest>

android-kiwix-lib-publisher/settings.gradle

@@ -1 +0,0 @@
-include ':kiwixLibAndroid'

debian/control

@@ -3,13 +3,12 @@ Priority: optional
 Maintainer: Kiwix team <kiwix@kiwix.org>
 Build-Depends: debhelper-compat (= 13),
  meson,
- pkg-config,
- libzim-dev (>= 6.1.8),
+ pkgconf,
+ libzim-dev (>= 9.0), libzim-dev (<< 10.0),
  libcurl4-gnutls-dev,
  libicu-dev,
  libgtest-dev,
  libkainjow-mustache-dev,
- liblzma-dev,
  libmicrohttpd-dev,
  libpugixml-dev,
  zlib1g-dev
@@ -22,12 +21,13 @@ Package: libkiwix-dev
 Section: libdevel
 Architecture: any
 Multi-Arch: same
-Depends: libkiwix10 (= ${binary:Version}), ${misc:Depends}, python3,
- libzim-dev (>= 6.0.0),
+Depends: libkiwix14 (= ${binary:Version}), ${misc:Depends}, python3,
+ libzim-dev (>= 9.0), libzim-dev (<< 10.0),
  libicu-dev,
  libpugixml-dev,
  libcurl4-gnutls-dev,
- libmicrohttpd-dev
+ libmicrohttpd-dev,
+ zlib1g-dev
 Description: library of common code for Kiwix (development)
  Kiwix is an offline Wikipedia reader. libkiwix provides the
  software core for Kiwix, and contains the code shared by all
@@ -35,11 +35,12 @@ Description: library of common code for Kiwix (development)
  .
  This package contains development files.
 
-Package: libkiwix10
+Package: libkiwix14
 Architecture: any
 Multi-Arch: same
 Depends: ${shlibs:Depends}, ${misc:Depends}, aria2
-Conflicts: libkiwix0, libkiwix3, libkiwix9
+Conflicts: libkiwix0, libkiwix3, libkiwix9, libkiwix10, libkiwix11, libkiwix12, libkiwix13
+Replaces: libkiwix0, libkiwix3, libkiwix9, libkiwix10, libkiwix11, libkiwix12, libkiwix13
 Description: library of common code for Kiwix
  Kiwix is an offline Wikipedia reader. libkiwix provides the
  software core for Kiwix, and contains the code shared by all

debian/libkiwix-dev.manpages

@@ -1 +1,2 @@
 usr/share/man/man1/kiwix-compile-resources.1*
+usr/share/man/man1/kiwix-compile-i18n.1*

docs/.gitignore

@@ -0,0 +1,2 @@
+api
+xml

docs/conf.py

@@ -0,0 +1,68 @@
+# Configuration file for the Sphinx documentation builder.
+#
+# This file only contains a selection of the most common options. For a full
+# list see the documentation:
+# https://www.sphinx-doc.org/en/master/usage/configuration.html
+
+# -- Path setup --------------------------------------------------------------
+
+# If extensions (or modules to document with autodoc) are in another directory,
+# add these directories to sys.path here. If the directory is relative to the
+# documentation root, use os.path.abspath to make it absolute, like shown here.
+#
+import os
+# import sys
+# sys.path.insert(0, os.path.abspath('.'))
+
+
+# -- Project information -----------------------------------------------------
+
+project = 'libkiwix'
+copyright = '2022, libkiwix-team'
+author = 'libkiwix-team'
+
+
+# -- General configuration ---------------------------------------------------
+
+# Add any Sphinx extension module names here, as strings. They can be
+# extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
+# ones.
+extensions = [
+    'breathe',
+    'exhale'
+]
+
+# Add any paths that contain templates here, relative to this directory.
+templates_path = ['_templates']
+
+# List of patterns, relative to source directory, that match files and
+# directories to ignore when looking for source files.
+# This pattern also affects html_static_path and html_extra_path.
+exclude_patterns = ['_build', 'Thumbs.db', '.DS_Store']
+
+html_theme = 'sphinx_rtd_theme'
+
+# Add any paths that contain custom static files (such as style sheets) here,
+# relative to this directory. They are copied after the builtin static files,
+# so a file named "default.css" will overwrite the builtin "default.css".
+html_static_path = ['_static']
+
+breathe_projects = {
+    "libkiwix": "./xml"
+}
+breathe_default_project = 'libkiwix'
+
+exhale_args = {
+    "containmentFolder":   "./api",
+    "rootFileName":        "ref_api.rst",
+    "rootFileTitle":       "Reference API",
+    "doxygenStripFromPath":"..",
+    "treeViewIsBootstrap": True,
+    "createTreeView" : True,
+    "exhaleExecutesDoxygen": True,
+    "exhaleDoxygenStdin":    "INPUT = ../include"
+}
+
+primary_domain = 'cpp'
+
+highlight_language = 'cpp'

docs/index.rst

@@ -0,0 +1,14 @@
+.. libkiwix documentation master file, created by
+   sphinx-quickstart on Fri Jul 24 15:40:50 2020.
+   You can adapt this file completely to your liking, but it should at least
+   contain the root `toctree` directive.
+
+Welcome to libkiwix's documentation!
+==================================
+
+.. toctree::
+   :maxdepth: 2
+   :caption: Contents:
+
+   usage
+   api/ref_api

docs/meson.build

@@ -0,0 +1,7 @@
+
+sphinx = find_program('sphinx-build', native:true)
+
+sphinx_target = run_target('doc',
+    command: [sphinx, '-bhtml',
+              meson.current_source_dir(),
+              meson.current_build_dir()])

docs/requirements.txt

@@ -0,0 +1,3 @@
+breathe
+exhale
+sphinx_rtd_theme

docs/usage.rst

@@ -0,0 +1,15 @@
+Libkiwix programming
+====================
+
+Introduction
+------------
+
+libkiwix is written in C++. To use the library, you need the include files of libkiwix have
+to link against libzim.
+
+Errors are handled with exceptions. When something goes wrong, libkiwix throws an error,
+which is always derived from std::exception.
+
+All classes are defined in the namespace kiwix.
+
+libkiwix is a set of tools to manage zim files and provide some common functionnality.

format_code.sh

@@ -1,36 +0,0 @@
-#!/usr/bin/bash
-
-files=(
-"include/library.h"
-"include/common/stringTools.h"
-"include/common/pathTools.h"
-"include/common/otherTools.h"
-"include/common/regexTools.h"
-"include/common/networkTools.h"
-"include/manager.h"
-"include/reader.h"
-"include/kiwix.h"
-"include/xapianSearcher.h"
-"include/searcher.h"
-"src/library.cpp"
-"src/android/kiwix.cpp"
-"src/android/org/kiwix/kiwixlib/JNIKiwixBool.java"
-"src/android/org/kiwix/kiwixlib/JNIKiwix.java"
-"src/android/org/kiwix/kiwixlib/JNIKiwixString.java"
-"src/android/org/kiwix/kiwixlib/JNIKiwixInt.java"
-"src/searcher.cpp"
-"src/common/pathTools.cpp"
-"src/common/regexTools.cpp"
-"src/common/otherTools.cpp"
-"src/common/networkTools.cpp"
-"src/common/stringTools.cpp"
-"src/xapianSearcher.cpp"
-"src/manager.cpp"
-"src/reader.cpp"
-)
-
-for i in "${files[@]}"
-do
-  echo $i
-  clang-format -i -style=file $i
-done

include/book.h

@@ -21,28 +21,54 @@
 #define KIWIX_BOOK_H
 
 #include <string>
+#include <vector>
+#include <memory>
+#include <mutex>
+#include "common.h"
 
 namespace pugi {
 class xml_node;
 }
 
+namespace zim {
+class Archive;
+}
+
 namespace kiwix
 {
 
 class OPDSDumper;
-class Reader;
 
 /**
  * A class to store information about a book (a zim file)
  */
 class Book
 {
- public:
+ public: // types
+  class Illustration
+  {
+    friend class Book;
+   public:
+    uint16_t width = 48;
+    uint16_t height = 48;
+    std::string mimeType;
+    std::string url;
+
+    const std::string& getData() const;
+
+   private:
+    mutable std::string data;
+    mutable std::mutex mutex;
+  };
+
+  typedef std::vector<std::shared_ptr<const Illustration>> Illustrations;
+
+ public: // functions
   Book();
   ~Book();
 
   bool update(const Book& other);
-  void update(const Reader& reader);
+  void update(const zim::Archive& archive);
   void updateFromXml(const pugi::xml_node& node, const std::string& baseDir);
   void updateFromOpds(const pugi::xml_node& node, const std::string& urlHost);
   std::string getHumanReadableIdFromPath() const;
@@ -53,7 +79,9 @@ class Book
   bool isPathValid() const { return m_pathValid; }
   const std::string& getTitle() const { return m_title; }
   const std::string& getDescription() const { return m_description; }
-  const std::string& getLanguage() const { return m_language; }
+  DEPRECATED const std::string& getLanguage() const { return m_language; }
+  const std::string& getCommaSeparatedLanguages() const { return m_language; }
+  const std::vector<std::string> getLanguages() const;
   const std::string& getCreator() const { return m_creator; }
   const std::string& getPublisher() const { return m_publisher; }
   const std::string& getDate() const { return m_date; }
@@ -68,9 +96,13 @@ class Book
   const uint64_t& getArticleCount() const { return m_articleCount; }
   const uint64_t& getMediaCount() const { return m_mediaCount; }
   const uint64_t& getSize() const { return m_size; }
-  const std::string& getFavicon() const;
-  const std::string& getFaviconUrl() const { return m_faviconUrl; }
-  const std::string& getFaviconMimeType() const { return m_faviconMimeType; }
+  DEPRECATED const std::string& getFavicon() const;
+  DEPRECATED const std::string& getFaviconUrl() const;
+  DEPRECATED const std::string& getFaviconMimeType() const;
+
+  Illustrations getIllustrations() const;
+  std::shared_ptr<const Illustration> getIllustration(unsigned int size) const;
+
   const std::string& getDownloadId() const { return m_downloadId; }
 
   void setReadOnly(bool readOnly) { m_readOnly = readOnly; }
@@ -91,14 +123,13 @@ class Book
   void setArticleCount(uint64_t articleCount) { m_articleCount = articleCount; }
   void setMediaCount(uint64_t mediaCount) { m_mediaCount = mediaCount; }
   void setSize(uint64_t size) { m_size = size; }
-  void setFavicon(const std::string& favicon) { m_favicon = favicon; }
-  void setFaviconMimeType(const std::string& faviconMimeType) { m_faviconMimeType = faviconMimeType; }
   void setDownloadId(const std::string& downloadId) { m_downloadId = downloadId; }
 
- private:
+ private: // functions
   std::string getCategoryFromTags() const;
+  const Illustration& getDefaultIllustration() const;
 
- protected:
+ protected: // data
   std::string m_id;
   std::string m_downloadId;
   std::string m_path;
@@ -119,9 +150,11 @@ class Book
   uint64_t m_mediaCount = 0;
   bool m_readOnly = false;
   uint64_t m_size = 0;
-  mutable std::string m_favicon;
-  std::string m_faviconUrl;
-  std::string m_faviconMimeType;
+  Illustrations m_illustrations;
+
+  // Used as the return value of getDefaultIllustration() when no default
+  // illustration is found in the book
+  static const Illustration missingDefaultIllustration;
 };
 
 }

include/bookmark.h

@@ -29,19 +29,33 @@ class xml_node;
 namespace kiwix
 {
 
+class Book;
 /**
  * A class to store information about a bookmark (an article in a book)
  */
 class Bookmark
 {
  public:
+  /**
+   *  Create an empty bookmark.
+   *
+   * Bookmark must be populated with `set*` methods
+   */
   Bookmark();
+
+  /**
+   * Create a bookmark given a Book, a path and a title.
+   */
+  Bookmark(const Book& book, const std::string& path, const std::string& title);
+
   ~Bookmark();
 
   void updateFromXml(const pugi::xml_node& node);
 
   const std::string& getBookId() const { return m_bookId; }
   const std::string& getBookTitle()  const { return m_bookTitle; }
+  const std::string& getBookName()  const { return m_bookName; }
+  const std::string& getBookFlavour()  const { return m_bookFlavour; }
   const std::string& getUrl() const { return m_url; }
   const std::string& getTitle() const { return m_title; }
   const std::string& getLanguage() const { return m_language; }
@@ -49,6 +63,8 @@ class Bookmark
 
   void setBookId(const std::string& bookId) { m_bookId = bookId; }
   void setBookTitle(const std::string& bookTitle) { m_bookTitle = bookTitle; }
+  void setBookName(const std::string& bookName) { m_bookName = bookName; }
+  void setBookFlavour(const std::string& bookFlavour) { m_bookFlavour = bookFlavour; }
   void setUrl(const std::string& url) { m_url = url; }
   void setTitle(const std::string& title) { m_title = title; }
   void setLanguage(const std::string& language) { m_language = language; }
@@ -57,6 +73,8 @@ class Bookmark
  protected:
   std::string m_bookId;
   std::string m_bookTitle;
+  std::string m_bookName;
+  std::string m_bookFlavour;
   std::string m_url;
   std::string m_title;
   std::string m_language;

include/common.h

@@ -16,6 +16,7 @@
 
 namespace kiwix {
 
+enum class IpMode { IPV4, IPV6, ALL, AUTO }; // AUTO: Server decides the protocol
 typedef zim::size_type size_type;
 typedef zim::offset_type offset_type;
 

include/downloader.h

@@ -25,6 +25,7 @@
 #include <map>
 #include <memory>
 #include <stdexcept>
+#include <mutex>
 
 namespace kiwix
 {
@@ -43,6 +44,14 @@ class AriaError : public std::runtime_error {
 };
 
 
+/**
+ * A representation of a current download.
+ *
+ * `Download` is not thread safe. User must care to not call method on a
+ * same download from different threads.
+ * However, it is safe to use different `Download`s from different threads.
+ */
+
 class Download {
  public:
   typedef enum { K_ACTIVE, K_WAITING, K_PAUSED, K_ERROR, K_COMPLETE, K_REMOVED, K_UNKNOWN } StatusResult;
@@ -53,19 +62,89 @@ class Download {
     : mp_aria(p_aria),
       m_status(K_UNKNOWN),
       m_did(did) {};
-  void updateStatus(bool follow=false);
+
+  /**
+   * Update the status of the download.
+   *
+   * This call make an aria rpc call and is blocking.
+   * Some download (started with a metalink) are in fact several downloads.
+   * - A first one to download the metadlink.
+   * - A second one to download the real file.
+   *
+   * If `follow` is true, updateStatus tries to detect that and tracks
+   * the second download when the first one is finished.
+   * By passing false to `follow`, `Download` will only track the first download.
+   *
+   * `getFoo` methods are based on the last statusUpdate.
+   *
+   * @param follow: Do we have to follow following downloads.
+   */
+  void updateStatus(bool follow);
+
+  /**
+   * Pause the download (and call updateStatus)
+   */
   void pauseDownload();
+
+  /**
+   * Resume the download (and call updateStatus)
+   */
   void resumeDownload();
+
+  /**
+   * Cancel the download.
+   *
+   * A canceled downlod cannot be resume and updateStatus does nothing.
+   * However, you can still get information based on the last known information.
+   */
   void cancelDownload();
-  StatusResult getStatus()          { return m_status; }
-  std::string  getDid()             { return m_did; }
-  std::string  getFollowedBy()      { return m_followedBy; }
-  uint64_t     getTotalLength()     { return m_totalLength; }
-  uint64_t     getCompletedLength() { return m_completedLength; }
-  uint64_t     getDownloadSpeed()   { return m_downloadSpeed; }
-  uint64_t     getVerifiedLength()  { return m_verifiedLength; }
-  std::string  getPath()            { return m_path; }
-  std::vector<std::string>&  getUris()             { return m_uris; }
+
+  /*
+   * Get the status of the download.
+   */
+  StatusResult getStatus() const             { return m_status; }
+
+  /*
+   * Get the id of the download.
+   */
+  const std::string&  getDid() const         { return m_did; }
+
+  /*
+   * Get the id of the "second" download.
+   *
+   * Set only if the "first" download is a metalink and is complete.
+   */
+  const std::string&  getFollowedBy() const  { return m_followedBy; }
+
+  /*
+   * Get the total length of the download.
+   */
+  uint64_t     getTotalLength() const        { return m_totalLength; }
+
+  /*
+   * Get the completed length of the download.
+   */
+  uint64_t     getCompletedLength() const    { return m_completedLength; }
+
+  /*
+   * Get the download speed of the download.
+   */
+  uint64_t     getDownloadSpeed() const      { return m_downloadSpeed; }
+
+  /*
+   * Get the verified length of the download.
+   */
+  uint64_t     getVerifiedLength() const     { return m_verifiedLength; }
+
+  /*
+   * Get the path (local file) of the download.
+   */
+  const std::string&  getPath() const        { return m_path; }
+
+  /*
+   * Get the download uris of the download.
+   */
+  const std::vector<std::string>&  getUris() const { return m_uris; }
 
  protected:
   std::shared_ptr<Aria2> mp_aria;
@@ -83,23 +162,69 @@ class Download {
 /**
  * A tool to download things.
  *
+ * A Downloader manages `Download` using aria2 in the background.
+ * `Downloader` is threadsafe.
+ * However, the returned `Download`s are NOT threadsafe.
  */
 class Downloader
 {
- public:
-  Downloader();
+ public: // types
+  typedef std::vector<std::pair<std::string, std::string>> Options;
+
+ public: // functions
+  /*
+  * Create a new Downloader object.
+  *
+  * @param sessionFileDir: The directory where aria2 will store its session file.
+  */
+  explicit Downloader(std::string sessionFileDir);
   virtual ~Downloader();
 
   void close();
 
-  Download* startDownload(const std::string& uri, const std::vector<std::pair<std::string, std::string>>& options = {});
-  Download* getDownload(const std::string& did);
+  /**
+   * Start a new download.
+   *
+   * This method is thread safe and returns a pointer to a newly created
+   * `Download` or an existing one with a matching URI. In the latter case
+   * the options parameter is ignored, which can lead to surprising results.
+   * For example, if the old and new download requests (sharing the same URI)
+   * have different values for the download directory or output file name
+   * options, after the download is reported to be complete the downloaded file
+   * will be present only at the location specified for the first request.
+   *
+   * User should call `update` on the returned `Download` to have an accurate status.
+   *
+   * @param uri: The uri of the thing to download.
+   * @param downloadDir: The download directory where the thing should be stored (takes precedence over any "dir" in `options`).
+   * @param options: A series of pair <option_name, option_value> to pass to aria.
+   * @return: The newly created Download.
+   */
+  std::shared_ptr<Download> startDownload(const std::string& uri, const std::string& downloadDir, Options options = {});
 
-  size_t getNbDownload() { return m_knownDownloads.size(); }
-  std::vector<std::string> getDownloadIds();
+  /**
+   * Get a download corrsponding to a download id (did)
+   * User should call `update` on the returned `Download` to have an accurate status.
+   *
+   * @param did: The download id to search for.
+   * @return: The Download corresponding to did.
+   * @throw: Throw std::out_of_range if did is not found.
+   */
+  std::shared_ptr<Download> getDownload(const std::string& did);
 
- private:
-  std::map<std::string, std::unique_ptr<Download>> m_knownDownloads;
+  /**
+   * Get the number of downloads currently managed.
+   */
+  size_t getNbDownload() const;
+
+  /**
+   * Get the ids of the managed downloads.
+   */
+  std::vector<std::string> getDownloadIds() const;
+
+ private: // data
+  mutable std::mutex m_lock;
+  std::map<std::string, std::shared_ptr<Download>> m_knownDownloads;
   std::shared_ptr<Aria2> mp_aria;
 };
 }

include/entry.h

@@ -1,183 +0,0 @@
-/*
- * Copyright 2018-2020 Matthieu Gautier <mgautier@kymeria.fr>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU  General Public License as published by
- * the Free Software Foundation; either version 3 of the License, or
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
- * MA 02110-1301, USA.
- */
-
-#ifndef KIWIX_ENTRY_H
-#define KIWIX_ENTRY_H
-
-#include <stdio.h>
-#include <zim/entry.h>
-#include <zim/item.h>
-#include <exception>
-#include <string>
-
-#include "common.h"
-
-using namespace std;
-
-namespace kiwix
-{
-
-
-class NoEntry : public std::exception {};
-
-/**
- * A entry represent an.. entry in a zim file.
- */
-class Entry
-{
-  public:
-    /**
-     * Construct an entry making reference to an zim article.
-     *
-     * @param article a zim::Article object
-     */
-    Entry(zim::Entry entry);
-    virtual ~Entry() = default;
-
-    /**
-     * Get the path of the entry.
-     *
-     * The path is the "key" of an entry.
-     *
-     * @return the path of the entry.
-     */
-    std::string getPath() const { return entry.getPath(); }
-
-    /**
-     * Get the title of the entry.
-     *
-     * @return the title of the entry.
-     */
-    std::string getTitle() const { return entry.getTitle(); }
-
-    /**
-     * Get the content of the entry.
-     *
-     * The string is a copy of the content.
-     * If you don't want to do a copy, use get_blob.
-     *
-     * @return the content of the entry.
-     */
-    std::string getContent() const { return entry.getItem().getData(); }
-
-    /**
-     * Get the blob of the entry.
-     *
-     * A blob make reference to the content without copying it.
-     *
-     * @param offset The starting offset of the blob.
-     * @return the blob of the entry.
-     */
-    zim::Blob   getBlob(offset_type offset = 0) const { return entry.getItem().getData(offset); }
-
-    /**
-     * Get the blob of the entry.
-     *
-     * A blob make reference to the content without copying it.
-     *
-     * @param offset The starting offset of the blob.
-     * @param size The size of the blob.
-     * @return the blob of the entry.
-     */
-    zim::Blob   getBlob(offset_type offset, size_type size) const { return entry.getItem().getData(offset, size); }
-
-    /**
-     * Get the info for direct access to the content of the entry.
-     *
-     * Some entry (ie binary ones) have their content plain stored
-     * in the zim file. Knowing the offset where the content is stored
-     * an user can directly read the content in the zim file bypassing the
-     * libkiwix/libzim.
-     *
-     * @return A pair specifying where to read the content.
-     *         The string is the real file to read (may be different that .zim
-     *         file if zim is cut).
-     *         The offset is the offset to read in the file.
-     *         Return <"",0> if is not possible to read directly.
-     */
-    zim::Item::DirectAccessInfo getDirectAccessInfo() const { return entry.getItem().getDirectAccessInformation(); }
-
-    /**
-     * Get the size of the entry.
-     *
-     * @return the size of the entry.
-     */
-    size_type   getSize() const;
-
-    /**
-     * Get the mime_type of the entry.
-     *
-     * @return the mime_type of the entry.
-     */
-    std::string getMimetype() const;
-
-
-    /**
-     * Get if the entry is a redirect entry.
-     *
-     * @return True if the entry is a redirect.
-     */
-    bool isRedirect() const;
-
-    /**
-     * Get if the entry is a link target entry.
-     *
-     * @return True if the entry is a link target.
-     */
-    bool isLinkTarget() const;
-
-    /**
-     * Get if the entry is a deleted entry.
-     *
-     * @return True if the entry is a deleted entry.
-     */
-    bool isDeleted() const;
-
-    /**
-     * Get the entry pointed by this entry.
-     *
-     * @return the entry pointed.
-     * @throw NoEntry if the entry is not a redirected entry.
-     */
-    Entry getRedirectEntry() const;
-
-    /**
-     * Get the final entry pointed by this entry.
-     *
-     * Follow the redirection until a "not redirecting" entry is found.
-     * If the entry is not a redirected entry, return the entry itself.
-     *
-     * @return the final entry.
-     */
-    Entry getFinalEntry() const;
-
-    /**
-     * Get the zim entry wrapped by this (kiwix) entry
-     *
-     * @return the zim entry
-     */
-    const zim::Entry& getZimEntry() const { return entry; }
-
-  private:
-    zim::Entry entry;
-};
-
-}
-
-#endif // KIWIX_ENTRY_H

include/i18n.h

@@ -0,0 +1,92 @@
+/*
+ * Copyright 2024 Veloman Yunkan <veloman.yunkan@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#ifndef KIWIX_I18N
+#define KIWIX_I18N
+
+#include <map>
+#include <string>
+
+namespace kiwix
+{
+
+std::string getTranslatedString(const std::string& lang, const std::string& key);
+
+namespace i18n
+{
+
+typedef std::map<std::string, std::string> Parameters;
+
+std::string expandParameterizedString(const std::string& lang,
+                                      const std::string& key,
+                                      const Parameters& params);
+
+class GetTranslatedString
+{
+public:
+  explicit GetTranslatedString(const std::string& lang) : m_lang(lang) {}
+
+  std::string operator()(const std::string& key) const
+  {
+    return getTranslatedString(m_lang, key);
+  }
+
+  std::string operator()(const std::string& key, const Parameters& params) const
+  {
+    return expandParameterizedString(m_lang, key, params);
+  }
+
+private:
+  const std::string m_lang;
+};
+
+} // namespace i18n
+
+class ParameterizedMessage
+{
+public: // types
+  typedef i18n::Parameters Parameters;
+
+public: // functions
+  ParameterizedMessage(const std::string& msgId, const Parameters& params)
+    : msgId(msgId)
+    , params(params)
+  {}
+
+  std::string getText(const std::string& lang) const;
+
+  const std::string& getMsgId()  const { return msgId; }
+  const Parameters&  getParams() const { return params; }
+
+private: // data
+  const std::string msgId;
+  const Parameters  params;
+};
+
+inline ParameterizedMessage nonParameterizedMessage(const std::string& msgId)
+{
+  const ParameterizedMessage::Parameters noParams;
+  return ParameterizedMessage(msgId, noParams);
+}
+
+std::string translateBookCategory(const std::string& lang, const std::string& category);
+
+} // namespace kiwix
+
+#endif // KIWIX_I18N

include/kiwix.h

@@ -22,4 +22,4 @@
 
 #include "library.h"
 
-#endif
+#endif

include/library.h

@@ -24,6 +24,9 @@
 #include <vector>
 #include <map>
 #include <memory>
+#include <mutex>
+#include <zim/archive.h>
+#include <zim/search.h>
 
 #include "book.h"
 #include "bookmark.h"
@@ -31,6 +34,10 @@
 
 #define KIWIX_LIBRARY_VERSION "20110515"
 
+namespace Xapian {
+class WritableDatabase;
+};
+
 namespace kiwix
 {
 
@@ -48,6 +55,22 @@ enum supportedListMode {
   NOVALID = 1 << 5
 };
 
+enum MigrationMode {
+  /** When migrating bookmarks, do not allow to migrate to an older book than the currently pointed one
+   * (or date stored in the bookmark if book is invalid)
+   *
+   * If no newer books are found, no upgrade is made.
+   */
+  UPGRADE_ONLY = 0,
+
+  /** Try hard to do a migration. This mostly does:
+   * - Try to find a newer book.
+   * - If book is invalid: find a best book, potentially older.
+   * Older book will never be returned if current book is a valid one.
+   */
+  ALLOW_DOWNGRADE = 1,
+};
+
 class Filter {
   public: // types
     using Tags = std::vector<std::string>;
@@ -64,6 +87,7 @@ class Filter {
     std::string _query;
     bool _queryIsPartial;
     std::string _name;
+    std::string _flavour;
 
   public: // functions
     Filter();
@@ -102,13 +126,30 @@ class Filter {
     Filter& acceptTags(const Tags& tags);
     Filter& rejectTags(const Tags& tags);
 
+    /**
+     *  Set the filter to only accept books in the specified category.
+     *
+     *  Multiple categories can be specified as a comma-separated list (in
+     *  which case a book in any of those categories will match).
+     */
     Filter& category(std::string category);
+
+    /**
+     *  Set the filter to only accept books in the specified language.
+     *
+     *  Multiple languages can be specified as a comma-separated list (in
+     *  which case a book in any of those languages will match).
+     */
     Filter& lang(std::string lang);
+
     Filter& publisher(std::string publisher);
     Filter& creator(std::string creator);
     Filter& maxSize(size_t size);
     Filter& query(std::string query, bool partial=true);
     Filter& name(std::string name);
+    Filter& flavour(std::string flavour);
+    Filter& clearLang();
+    Filter& clearCategory();
 
     bool hasQuery() const;
     const std::string& getQuery() const { return _query; }
@@ -129,6 +170,9 @@ class Filter {
     bool hasCreator() const;
     const std::string& getCreator() const { return _creator; }
 
+    bool hasFlavour() const;
+    const std::string& getFlavour() const { return _flavour; }
+
     const Tags& getAcceptTags() const { return _acceptTags; }
     const Tags& getRejectTags() const { return _rejectTags; }
 
@@ -139,31 +183,68 @@ private: // functions
 };
 
 
+class ZimSearcher : public zim::Searcher
+{
+  public:
+    explicit ZimSearcher(zim::Searcher&& searcher)
+      : zim::Searcher(searcher)
+    {}
+
+    std::unique_lock<std::mutex> getLock() {
+      return std::unique_lock<std::mutex>(m_mutex);
+    }
+    virtual ~ZimSearcher() = default;
+  private:
+    std::mutex m_mutex;
+};
+
+template<typename, typename>
+class ConcurrentCache;
+
+template<typename, typename>
+class MultiKeyCache;
+
+using LibraryPtr = std::shared_ptr<Library>;
+using ConstLibraryPtr = std::shared_ptr<const Library>;
+
+
+// Some compiler we use don't have [[nodiscard]] attribute.
+// We don't want to declare `create` with it in this case.
+#define LIBKIWIX_NODISCARD
+#if defined __has_cpp_attribute
+  #if __has_cpp_attribute (nodiscard)
+    #undef LIBKIWIX_NODISCARD
+    #define LIBKIWIX_NODISCARD [[nodiscard]]
+  #endif
+#endif
+
 /**
  * A Library store several books.
  */
-class Library
+class Library: public std::enable_shared_from_this<Library>
 {
-  std::map<std::string, kiwix::Book> m_books;
-  std::map<std::string, std::shared_ptr<Reader>> m_readers;
-  std::vector<kiwix::Bookmark> m_bookmarks;
-  class BookDB;
-  std::unique_ptr<BookDB> m_bookDB;
-
  public:
+  typedef uint64_t Revision;
   typedef std::vector<std::string> BookIdCollection;
+  typedef std::map<std::string, int> AttributeCounts;
+  typedef std::set<std::string> BookIdSet;
+
+ private:
+  Library();
 
  public:
-  Library();
+  LIBKIWIX_NODISCARD static LibraryPtr create() {
+    return LibraryPtr(new Library());
+  }
   ~Library();
 
   /**
    * Library is not a copiable object. However it can be moved.
    */
   Library(const Library& ) = delete;
-  Library(Library&& );
+  Library(Library&& ) = delete;
   void operator=(const Library& ) = delete;
-  Library& operator=(Library&& );
+  Library& operator=(Library&& ) = delete;
 
   /**
    * Add a book to the library.
@@ -177,6 +258,11 @@ class Library
    */
   bool addBook(const Book& book);
 
+  /**
+   * A self-explanatory alias for addBook()
+   */
+  bool addOrUpdateBook(const Book& book) { return addBook(book); }
+
   /**
    * Add a bookmark to the library.
    *
@@ -185,7 +271,7 @@ class Library
   void addBookmark(const Bookmark& bookmark);
 
   /**
-   * Remove a bookmarkk
+   * Remove a bookmark
    *
    * @param zimId The zimId of the bookmark.
    * @param url The url of the bookmark.
@@ -193,11 +279,78 @@ class Library
    */
   bool removeBookmark(const std::string& zimId, const std::string& url);
 
+  /**
+   * Migrate all invalid bookmarks.
+   *
+   * All invalid bookmarks (ie pointing to unknown books, no check is made on bookmark pointing to
+   * invalid articles of valid book) will be migrated (if possible) to a better book.
+   * "Better book", will be determined using method `getBestTargetBookId`.
+   *
+   * @return A tuple<int, int>: <The number of bookmarks updated>, <Number of invalid bookmarks before migration was performed>.
+   */
+  std::tuple<int, int> migrateBookmarks(MigrationMode migrationMode = ALLOW_DOWNGRADE);
+
+  /**
+   * Migrate all bookmarks associated to a specific book.
+   *
+   * All bookmarks associated to `sourceBookId` book will be migrated to a better book.
+   * "Better book", will be determined using method `getBestTargetBookId`.
+   *
+   * @param sourceBookId the source bookId of the bookmarks to migrate.
+   * @param migrationMode how we will find the best book.
+   * @return The number of bookmarks updated.
+   */
+  int migrateBookmarks(const std::string& sourceBookId, MigrationMode migrationMode = UPGRADE_ONLY);
+
+  /**
+   * Migrate bookmarks
+   *
+   * Migrate all bookmarks pointing to `source` to `destination`.
+   *
+   * @param sourceBookId the source bookId of the bookmarks to migrate.
+   * @param targetBookId the destination bookId to migrate the bookmarks to.
+   * @return The number of bookmarks updated.
+   */
+  int migrateBookmarks(const std::string& sourceBookId, const std::string& targetBookId);
+ 
+  /**
+   * Get the best available bookId for a bookmark.
+   *
+   * Given a bookmark, return the best available bookId.
+   * "best available bookId" is determined using heuristitcs based on book name, flavour and date.
+   *
+   * @param bookmark The bookmark to search the bookId for.
+   * @param migrationMode The migration mode to use.
+   * @return A bookId. Potentially empty string if no suitable book found.
+   */
+  std::string getBestTargetBookId(const Bookmark& bookmark, MigrationMode migrationMode) const;
+
+  /**
+   * Get the best bookId for a combination of book's name, flavour and date.
+   *
+   * Given a bookName (mandatory), try to find the best book.
+   * If preferedFlavour is given, will try to find a book with the same flavour. If not found, return a book with a different flavour.
+   * If minDate is given, return a book newer than minDate. If not found, return a empty bookId.
+   *
+   * @param bookName The name of the book
+   * @param preferedFlavour The prefered flavour.
+   * @param minDate the minimal book date acceptable. Must be a string in the format "YYYY-MM-DD".
+   * @return A bookId corresponding to the query, or empty string if not found.
+   */
+  std::string getBestTargetBookId(const std::string& bookName, const std::string& preferedFlavour="", const std::string& minDate="") const;
+
+  // XXX: This is a non-thread-safe operation
   const Book& getBookById(const std::string& id) const;
-  Book& getBookById(const std::string& id);
+  // XXX: This is a non-thread-safe operation
   const Book& getBookByPath(const std::string& path) const;
-  Book& getBookByPath(const std::string& path);
-  std::shared_ptr<Reader> getReaderById(const std::string& id);
+
+  Book getBookByIdThreadSafe(const std::string& id) const;
+
+  std::shared_ptr<zim::Archive> getArchiveById(const std::string& id);
+  std::shared_ptr<ZimSearcher> getSearcherById(const std::string& id) {
+    return getSearcherByIds(BookIdSet{id});
+  }
+  std::shared_ptr<ZimSearcher> getSearcherByIds(const BookIdSet& ids);
 
   /**
    * Remove a book from the library.
@@ -239,6 +392,13 @@ class Library
    */
   std::vector<std::string> getBooksLanguages() const;
 
+  /**
+   * Get all languagues of the books in the library with counts.
+   *
+   * @return A list of languages with the count of books in each language.
+   */
+  AttributeCounts getBooksLanguagesWithCounts() const;
+
   /**
    * Get all categories of the books in the library.
    *
@@ -274,17 +434,6 @@ class Library
    */
   BookIdCollection getBooksIds() const;
 
-  /**
-   * Filter the library and generate a new one with the keep elements.
-   *
-   * This is equivalent to `listBookIds(ALL, UNSORTED, search)`.
-   *
-   * @param search List only books with search in the title or description.
-   * @return The list of bookIds corresponding to the query.
-   */
-  DEPRECATED BookIdCollection filter(const std::string& search) const;
-
-
   /**
    * Filter the library and return the id of the keep elements.
    *
@@ -304,45 +453,59 @@ class Library
   void sort(BookIdCollection& bookIds, supportedListSortBy sortBy, bool ascending) const;
 
   /**
-   * List books in the library.
+   * Return the current revision of the library.
    *
-   * @param mode The mode of listing :
-   *             - LOCAL    : list only local books (with a path).
-   *             - REMOTE   : list only remote books (with an url).
-   *             - VALID    : list only valid books (without a path or with a
-   *                          path pointing to a valid zim file).
-   *             - NOLOCAL  : list only books without valid path.
-   *             - NOREMOTE : list only books without url.
-   *             - NOVALID  : list only books not valid.
-   *             - ALL : Do not do any filter (LOCAL or REMOTE)
-   *             - Flags can be combined.
-   * @param sortBy Attribute to sort by the book list.
-   * @param search List only books with search in the title, description.
-   * @param language List only books in this language.
-   * @param creator List only books of this creator.
-   * @param publisher List only books of this publisher.
-   * @param maxSize Do not list book bigger than maxSize.
-   *                Set to 0 to cancel this filter.
-   * @return The list of bookIds corresponding to the query.
+   * The revision of the library is updated (incremented by one) by
+   * the addBook() and removeBookById() operations.
+   *
+   * @return Current revision of the library.
    */
-  DEPRECATED BookIdCollection listBooksIds(
-    int supportedListMode = ALL,
-    supportedListSortBy sortBy = UNSORTED,
-    const std::string& search = "",
-    const std::string& language = "",
-    const std::string& creator = "",
-    const std::string& publisher = "",
-    const std::vector<std::string>& tags = {},
-    size_t maxSize = 0) const;
+  Revision getRevision() const;
+
+  /**
+   * Remove books that have not been updated since the specified revision.
+   *
+   * @param  rev the library revision to use
+   * @return Count of books that were removed by this operation.
+   */
+  uint32_t removeBooksNotUpdatedSince(Revision rev);
 
   friend class OPDSDumper;
   friend class libXMLDumper;
 
+private: // types
+  typedef const std::string& (Book::*BookStrPropMemFn)() const;
+  struct Entry : Book
+  {
+    Library::Revision lastUpdatedRevision = 0;
+  };
+
 private: // functions
+  AttributeCounts getBookAttributeCounts(BookStrPropMemFn p) const;
+  std::vector<std::string> getBookPropValueSet(BookStrPropMemFn p) const;
   BookIdCollection filterViaBookDB(const Filter& filter) const;
+  std::string getBestFromBookCollection(BookIdCollection books, const Bookmark& bookmark, MigrationMode migrationMode) const;
+  unsigned int getBookCount_not_protected(const bool localBooks, const bool remoteBooks) const;
   void updateBookDB(const Book& book);
+  void dropCache(const std::string& bookId);
+
+private: //data
+  mutable std::recursive_mutex m_mutex;
+  Library::Revision m_revision;
+  std::map<std::string, Entry> m_books;
+  using ArchiveCache = ConcurrentCache<std::string, std::shared_ptr<zim::Archive>>;
+  std::unique_ptr<ArchiveCache> mp_archiveCache;
+  using SearcherCache = MultiKeyCache<std::string, std::shared_ptr<ZimSearcher>>;
+  std::unique_ptr<SearcherCache> mp_searcherCache;
+  std::vector<kiwix::Bookmark> m_bookmarks;
+  std::unique_ptr<Xapian::WritableDatabase> m_bookDB;
 };
 
+// We don't need it anymore and we don't want to polute any other potential usage
+// of `LIBKIWIX_NODISCARD` token.
+#undef LIBKIWIX_NODISCARD
+
+
 }
 
 #endif

include/manager.h

@@ -22,10 +22,10 @@
 
 #include "book.h"
 #include "library.h"
-#include "reader.h"
 
 #include <string>
 #include <vector>
+#include <memory>
 
 namespace pugi {
 class xml_document;
@@ -34,26 +34,25 @@ class xml_document;
 namespace kiwix
 {
 
-class LibraryManipulator {
- public:
-  virtual ~LibraryManipulator() {}
-  virtual bool addBookToLibrary(Book book) = 0;
-  virtual void addBookmarkToLibrary(Bookmark bookmark) = 0;
-};
+class LibraryManipulator
+{
+ public: // functions
+  explicit LibraryManipulator(LibraryPtr library);
+  virtual ~LibraryManipulator();
 
-class DefaultLibraryManipulator : public LibraryManipulator {
- public:
-  DefaultLibraryManipulator(Library* library) :
-    library(library) {}
-  virtual ~DefaultLibraryManipulator() {}
-  bool addBookToLibrary(Book book) {
-    return library->addBook(book);
-  }
-  void addBookmarkToLibrary(Bookmark bookmark) {
-    library->addBookmark(bookmark);
-  }
- private:
-   kiwix::Library* library;
+  LibraryPtr getLibrary() const { return library; }
+
+  bool addBookToLibrary(const Book& book);
+  void addBookmarkToLibrary(const Bookmark& bookmark);
+  uint32_t removeBooksNotUpdatedSince(Library::Revision rev);
+
+ protected: // overrides
+  virtual void bookWasAddedToLibrary(const Book& book);
+  virtual void bookmarkWasAddedToLibrary(const Bookmark& bookmark);
+  virtual void booksWereRemovedFromLibrary();
+
+ private: // data
+  LibraryPtr library;
 };
 
 /**
@@ -61,10 +60,12 @@ class DefaultLibraryManipulator : public LibraryManipulator {
  */
 class Manager
 {
- public:
-  Manager(LibraryManipulator* manipulator);
-  Manager(Library* library);
-  ~Manager();
+ public: // types
+  typedef std::vector<std::string> Paths;
+
+ public: // functions
+  explicit Manager(LibraryManipulator manipulator);
+  explicit Manager(LibraryPtr library);
 
   /**
    * Read a `library.xml` and add book in the file to the library.
@@ -72,10 +73,22 @@ class Manager
    * @param path The (utf8) path to the `library.xml`.
    * @param readOnly Set if the libray path could be overwritten latter with
    *                 updated content.
+   * @param trustLibrary use book metadata coming from XML.
    * @return True if file has been properly parsed.
    */
   bool readFile(const std::string& path, bool readOnly = true, bool trustLibrary = true);
 
+  /**
+   * Sync the contents of the library with one or more `library.xml` files.
+   *
+   * The metadata of the library files is trusted unconditionally.
+   * Any books not present in the input library.xml files are removed
+   * from the library.
+   *
+   * @param paths The (utf8) paths to the `library.xml` files.
+   */
+  void reload(const Paths& paths);
+
   /**
    * Load a library content store in the string.
    *
@@ -142,6 +155,15 @@ class Manager
                        const std::string& url = "",
                        const bool checkMetaData = false);
 
+  /**
+   * Add all books from the directory tree into the library.
+   * 
+   * @param path          The path of the directory to scan.
+   * @param verboseFlag   Verbose logs flag.
+   */
+  void addBooksFromDirectory(const std::string& path,
+                             const bool verboseFlag = false);
+
   std::string writableLibraryPath;
 
   bool m_hasSearchResult = false;
@@ -150,8 +172,7 @@ class Manager
   uint64_t m_itemsPerPage = 0;
 
  protected:
-  kiwix::LibraryManipulator* manipulator;
-  bool mustDeleteManipulator;
+  kiwix::LibraryManipulator manipulator;
 
   bool readBookFromPath(const std::string& path, Book* book);
   bool parseXmlDom(const pugi::xml_document& doc,

include/meson.build

@@ -4,27 +4,15 @@ headers = [
   'common.h',
   'library.h',
   'manager.h',
-  'libxml_dumper.h',
-  'opds_dumper.h',
   'downloader.h',
-  'reader.h',
-  'entry.h',
-  'searcher.h',
   'search_renderer.h',
   'server.h',
+  'spelling_correction.h',
   'kiwixserve.h',
-  'name_mapper.h'
+  'name_mapper.h',
+  'tools.h',
+  'version.h',
+  'i18n.h'
 ]
 
 install_headers(headers, subdir:'kiwix')
-
-install_headers(
-  'tools/base64.h',
-  'tools/networkTools.h',
-  'tools/otherTools.h',
-  'tools/pathTools.h',
-  'tools/regexTools.h',
-  'tools/stringTools.h',
-  subdir:'kiwix/tools'
-)
-

include/name_mapper.h

@@ -22,6 +22,8 @@
 
 #include <string>
 #include <map>
+#include <memory>
+#include <mutex>
 
 namespace kiwix
 {
@@ -31,15 +33,15 @@ class Library;
 class NameMapper {
   public:
     virtual ~NameMapper() = default;
-    virtual std::string getNameForId(const std::string& id) = 0;
-    virtual std::string getIdForName(const std::string& name) = 0;
+    virtual std::string getNameForId(const std::string& id) const = 0;
+    virtual std::string getIdForName(const std::string& name) const = 0;
 };
 
 
 class IdNameMapper : public NameMapper {
   public:
-    virtual std::string getNameForId(const std::string& id) { return id; };
-    virtual std::string getIdForName(const std::string& name) { return name; };
+    virtual std::string getNameForId(const std::string& id) const { return id; };
+    virtual std::string getIdForName(const std::string& name) const { return name; };
 };
 
 class HumanReadableNameMapper : public NameMapper {
@@ -48,13 +50,34 @@ class HumanReadableNameMapper : public NameMapper {
     std::map<std::string, std::string> m_nameToId;
 
   public:
-    HumanReadableNameMapper(kiwix::Library& library, bool withAlias);
+    HumanReadableNameMapper(const kiwix::Library& library, bool withAlias);
     virtual ~HumanReadableNameMapper() = default;
-    virtual std::string getNameForId(const std::string& id);
-    virtual std::string getIdForName(const std::string& name);
+    virtual std::string getNameForId(const std::string& id) const;
+    virtual std::string getIdForName(const std::string& name) const;
+
+  private:
+    void mapName(const kiwix::Library& lib, std::string name, std::string id);
 };
 
+class UpdatableNameMapper : public NameMapper {
+    typedef std::shared_ptr<NameMapper> NameMapperHandle;
+  public:
+    UpdatableNameMapper(std::shared_ptr<Library> library, bool withAlias);
 
+    virtual std::string getNameForId(const std::string& id) const;
+    virtual std::string getIdForName(const std::string& name) const;
+
+    void update();
+
+  private:
+    NameMapperHandle currentNameMapper() const;
+
+  private:
+    mutable std::mutex mutex;
+    std::shared_ptr<Library> library;
+    NameMapperHandle nameMapper;
+    const bool withAlias;
+};
 
 }
 

include/opds_catalog.h

@@ -1,6 +1,5 @@
 /*
- * Copyright (C) 2013 Emmanuel Engelhart <kelson@kiwix.org>
- * Copyright (C) 2017 Matthieu Gautier <mgautier@kymeria.fr>
+ * Copyright 2021 Veloman Yunkan <veloman.yunkan@gmail.com>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU  General Public License as published by
@@ -18,9 +17,17 @@
  * MA 02110-1301, USA.
  */
 
-package org.kiwix.kiwixlib;
+#ifndef KIWIX_OPDS_CATALOG_H
+#define KIWIX_OPDS_CATALOG_H
 
-public class JNIICU
+
+#include "library.h"
+
+namespace kiwix
 {
-  static public native void setDataDirectory(String icuDataDir);
-}
+
+std::string getSearchUrl(const Filter& f);
+
+} // namespace kiwix
+
+#endif // KIWIX_OPDS_CATALOG_H

include/reader.h

@@ -1,502 +0,0 @@
-/*
- * Copyright 2011 Emmanuel Engelhart <kelson@kiwix.org>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU  General Public License as published by
- * the Free Software Foundation; either version 3 of the License, or
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
- * MA 02110-1301, USA.
- */
-
-#ifndef KIWIX_READER_H
-#define KIWIX_READER_H
-
-#include <stdio.h>
-#include <zim/zim.h>
-#include <zim/archive.h>
-#include <exception>
-#include <map>
-#include <sstream>
-#include <string>
-#include "common.h"
-#include "entry.h"
-#include "tools/pathTools.h"
-#include "tools/stringTools.h"
-
-using namespace std;
-
-namespace kiwix
-{
-
-/**
- * The SuggestionItem is a helper class that contains the info about a single
- * suggestion item.
- */
-
-class SuggestionItem
-{
-  // Functions
-  private:
-    // Create a sugggestion item.
-    explicit SuggestionItem(std::string title, std::string normalizedTitle,
-                            std::string path, std::string snippet = "") :
-      title(title),
-      normalizedTitle(normalizedTitle),
-      path(path),
-      snippet(snippet) {}
-
-  public:
-    const std::string getTitle()           {return title;}
-    const std::string getNormalizedTitle() {return normalizedTitle;}
-    const std::string getPath()            {return path;}
-    const std::string getSnippet()         {return snippet;}
-
-    const bool hasSnippet()                {return !snippet.empty();}
-
-  // Data
-  private:
-    std::string title;
-    std::string normalizedTitle;
-    std::string path;
-    std::string snippet;
-
-  friend class Reader;
-};
-
-/**
- * The Reader class is the class who allow to get an entry content from a zim
- * file.
- */
-
-using SuggestionsList_t = std::vector<SuggestionItem>;
-class Reader
-{
- public:
-  /**
-   * Create a Reader to read a zim file specified by zimFilePath.
-   *
-   * @param zimFilePath The path to the zim file to read.
-   *                    The zim file can be splitted (.zimaa, .zimab, ...).
-   *                    In this case, the file path must still point to the
-   *                    unsplitted path as if the file were not splitted
-   *                    (.zim extesion).
-   */
-  explicit Reader(const string zimFilePath);
-#ifndef _WIN32
-  explicit Reader(int fd);
-  Reader(int fd, zim::offset_type offset, zim::size_type size);
-#endif
-  ~Reader() = default;
-
-  /**
-   * Get the number of "displayable" entries in the zim file.
-   *
-   * @return If the zim file has a /M/Counter metadata, return the number of
-   *         entries with the 'text/html' MIMEtype specified in the metadata.
-   *         Else return the number of entries in the 'A' namespace.
-   */
-  unsigned int getArticleCount() const;
-
-  /**
-   * Get the number of media in the zim file.
-   *
-   * @return If the zim file has a /M/Counter metadata, return the number of
-   *         entries with the 'image/jpeg', 'image/gif' and 'image/png' in
-   *         the metadata.
-   *         Else return the number of entries in the 'I' namespace.
-   */
-  unsigned int getMediaCount() const;
-
-  /**
-   * Get the number of all entries in the zim file.
-   *
-   * @return Return the number of all the entries, whatever their MIMEtype or
-   *         their namespace.
-   */
-  unsigned int getGlobalCount() const;
-
-  /**
-   * Get the path of the zim file.
-   *
-   * @return the path of the zim file as given in the constructor.
-   */
-  string getZimFilePath() const;
-
-  /**
-   * Get the Id of the zim file.
-   *
-   * @return The uuid stored in the zim file.
-   */
-  string getId() const;
-
-  /**
-   * Get a random page.
-   *
-   * @return A random Entry. The entry is picked from all entries in
-   *         the 'A' namespace.
-   *         The main entry is excluded from the potential results.
-   */
-  Entry getRandomPage() const;
-
-  /**
-   * Get the entry of the main page.
-   *
-   * @return Entry of the main page as specified in the zim file.
-   */
-  Entry getMainPage() const;
-
-  /**
-   * Get the content of a metadata.
-   *
-   * @param[in]  name The name of the metadata.
-   * @param[out] value The value will be set to the content of the metadata.
-   * @return True if it was possible to get the content of the metadata.
-   */
-  bool getMetadata(const string& name, string& value) const;
-
-  /**
-   * Get the name of the zim file.
-   *
-   * @return The name of the zim file as specified in the zim metadata.
-   */
-  string getName() const;
-
-  /**
-   * Get the title of the zim file.
-   *
-   * @return The title of zim file as specified in the zim metadata.
-   *         If no title has been set, return a title computed from the
-   *         file path.
-   */
-  string getTitle() const;
-
-  /**
-   * Get the creator of the zim file.
-   *
-   * @return The creator of the zim file as specified in the zim metadata.
-   */
-  string getCreator() const;
-
-  /**
-   * Get the publisher of the zim file.
-   *
-   * @return The publisher of the zim file as specified in the zim metadata.
-   */
-  string getPublisher() const;
-
-  /**
-   * Get the date of the zim file.
-   *
-   * @return The date of the zim file as specified in the zim metadata.
-   */
-  string getDate() const;
-
-  /**
-   * Get the description of the zim file.
-   *
-   * @return The description of the zim file as specified in the zim metadata.
-   *         If no description has been set, return the subtitle.
-   */
-  string getDescription() const;
-
-  /**
-   * Get the long description of the zim file.
-   *
-   * @return The long description of the zim file as specifed in the zim metadata.
-   */
-  string getLongDescription() const;
-
-  /**
-   * Get the language of the zim file.
-   *
-   * @return The language of the zim file as specified in the zim metadata.
-   */
-  string getLanguage() const;
-
-  /**
-   * Get the license of the zim file.
-   *
-   * @return The license of the zim file as specified in the zim metadata.
-   */
-  string getLicense() const;
-
-  /**
-   * Get the tags of the zim file.
-   *
-   * @param original If true, return the original tags as specified in the zim metadata.
-   *                 Else, try to convert it to the new 'normalized' format.
-   * @return The tags of the zim file.
-   */
-  string getTags(bool original=false) const;
-
-  /**
-   * Get the value (as a string) of a specific tag.
-   *
-   * According to https://wiki.openzim.org/wiki/Tags
-   *
-   * @return The value of the specified tag.
-   * @throw  std::out_of_range if the specified tag is not found.
-   */
-  string getTagStr(const std::string& tagName) const;
-
-  /**
-   * Get the boolean value of a specific tag.
-   *
-   * According to https://wiki.openzim.org/wiki/Tags
-   *
-   * @return The boolean value of the specified tag.
-   * @throw  std::out_of_range if the specified tag is not found.
-   *         std::domain_error if the value of the tag cannot be convert to bool.
-   */
-  bool getTagBool(const std::string& tagName) const;
-
-  /**
-   * Get the relations of the zim file.
-   *
-   * @return The relation of the zim file as specified in the zim metadata.
-   */
-  string getRelation() const;
-
-  /**
-   * Get the flavour of the zim file.
-   *
-   * @return The flavour of the zim file as specified in the zim metadata.
-   */
-  string getFlavour() const;
-
-  /**
-   * Get the source of the zim file.
-   *
-   * @return The source of the zim file as specified in the zim metadata.
-   */
-  string getSource() const;
-
-  /**
-   * Get the scraper of the zim file.
-   *
-   * @return The scraper of the zim file as specified in the zim metadata.
-   */
-  string getScraper() const;
-
-  /**
-   * Get the origId of the zim file.
-   *
-   * The origId is only used in the case of patch zim file and is the Id
-   * of the original zim file.
-   *
-   * @return The origId of the zim file as specified in the zim metadata.
-   */
-  string getOrigId() const;
-
-  /**
-   * Get the favicon of the zim file.
-   *
-   * @param[out] content The content of the favicon.
-   * @param[out] mimeType The mimeType of the favicon.
-   * @return True if a favicon has been found.
-   */
-  bool getFavicon(string& content, string& mimeType) const;
-
-  /**
-   * Get an entry associated to an path.
-   *
-   * @param path The path of the entry.
-   * @return The entry.
-   * @throw NoEntry If no entry correspond to the path.
-   */
-  Entry getEntryFromPath(const std::string& path) const;
-
-  /**
-   * Get an entry associated to an url encoded path.
-   *
-   * Equivalent to `getEntryFromPath(urlDecode(path));`
-   *
-   * @param path The url encoded path.
-   * @return The entry.
-   * @throw NoEntry If no entry correspond to the path.
-   */
-  Entry getEntryFromEncodedPath(const std::string& path) const;
-
-  /**
-   * Get un entry associated to a title.
-   *
-   * @param title The title.
-   * @return The entry
-   * throw NoEntry If no entry correspond to the url.
-   */
-  Entry getEntryFromTitle(const std::string& title) const;
-
-  /**
-   * Search for entries with title starting with prefix (case sensitive).
-   *
-   * Suggestions are stored in an internal vector and can be retrieved using
-   * `getNextSuggestion` method.
-   * This method is not thread safe and is deprecated. Use :
-   * bool searchSuggestions(const string& prefix,
-   *                        unsigned int suggestionsCount,
-   *                        SuggestionsList_t& results);
-   *
-   * @param prefix The prefix to search.
-   * @param suggestionsCount How many suggestions to search for.
-   * @param reset If true, remove previous suggestions in the internal vector.
-   *              If false, add suggestions to the internal vector
-   *              (until internal vector size is suggestionCount (or no more
-   *               suggestion))
-   * @return True if some suggestions have been added to the internal vector.
-   */
-  DEPRECATED bool searchSuggestions(const string& prefix,
-                         unsigned int suggestionsCount,
-                         const bool reset = true);
-
-  /**
-   * Search for entries with title starting with prefix (case sensitive).
-   *
-   * Suggestions are added to the `result` vector.
-   *
-   * @param prefix The prefix to search.
-   * @param suggestionsCount How many suggestions to search for.
-   * @param result The vector where to store the suggestions.
-   * @return True if some suggestions have been added to the vector.
-   */
-
-  bool searchSuggestions(const string& prefix,
-                         unsigned int suggestionsCount,
-                         SuggestionsList_t& resuls);
-
-  /**
-   * Search for entries for the given prefix.
-   *
-   * If the zim file has a internal fulltext index, the suggestions will be
-   * searched using it.
-   * Else the suggestions will be search using `searchSuggestions` while trying
-   * to be smart about case sensitivity (using `getTitleVariants`).
-   *
-   * In any case, suggestions are stored in an internal vector and can be
-   * retrieved using `getNextSuggestion` method.
-   * The internal vector will be reset.
-   * This method is not thread safe and is deprecated. Use :
-   * bool searchSuggestionsSmart(const string& prefix,
-   *                             unsigned int suggestionsCount,
-   *                             SuggestionsList_t& results);
-   *
-   * @param prefix The prefix to search for.
-   * @param suggestionsCount How many suggestions to search for.
-   */
-  DEPRECATED bool searchSuggestionsSmart(const string& prefix,
-                              unsigned int suggestionsCount);
-
-  /**
-   * Search for entries for the given prefix.
-   *
-   * If the zim file has a internal fulltext index, the suggestions will be
-   * searched using it.
-   * Else the suggestions will be search using `searchSuggestions` while trying
-   * to be smart about case sensitivity (using `getTitleVariants`).
-   *
-   * In any case, suggestions are stored in an internal vector and can be
-   * retrieved using `getNextSuggestion` method.
-   * The internal vector will be reset.
-   *
-   * @param prefix The prefix to search for.
-   * @param suggestionsCount How many suggestions to search for.
-   * @param results The vector where to store the suggestions
-   * @return True if some suggestions have been added to the results.
-   */
-   bool searchSuggestionsSmart(const string& prefix,
-                              unsigned int suggestionsCount,
-                              SuggestionsList_t& results);
-
-
-  /**
-   * Check if the path exists in the zim file.
-   *
-   * @param path the path to check.
-   * @return True if the path exists in the zim file.
-   */
-  bool pathExists(const string& path) const;
-
-  /**
-   * Check if the zim file has a embedded fulltext index.
-   *
-   * @return True if the zim file has a embedded fulltext index
-   *         and is not split (else the fulltext is not accessible).
-   */
-  bool hasFulltextIndex() const;
-
-  /**
-   * Get potential case title variations for a title.
-   *
-   * @param title a title.
-   * @return the list of variantions.
-   */
-  std::vector<std::string> getTitleVariants(const std::string& title) const;
-
-  /**
-   * Get the next suggestion title.
-   *
-   * @param[out] title the title of the suggestion.
-   * @return True if title has been set.
-   */
-  DEPRECATED bool getNextSuggestion(string& title);
-
-  /**
-   * Get the next suggestion title and url.
-   *
-   * @param[out] title the title of the suggestion.
-   * @param[out] url the url of the suggestion.
-   * @return True if title and url have been set.
-   */
-  DEPRECATED bool getNextSuggestion(string& title, string& url);
-
-  /**
-   * Get if we can check zim file integrity (has a checksum).
-   *
-   * @return True if zim file have a checksum.
-   */
-  bool canCheckIntegrity() const;
-
-  /**
-   * Check is zim file is corrupted.
-   *
-   * @return True if zim file is corrupted.
-   */
-  bool isCorrupted() const;
-
-  /**
-   * Return the total size of the zim file.
-   *
-   * If zim file is split, return the sum of all parts' size.
-   *
-   * @return Size of the size file is KiB.
-   */
-  unsigned int getFileSize() const;
-
-  /**
-   * Get the zim file handler.
-   *
-   * @return The libzim file handler.
-   */
-  zim::Archive* getZimArchive() const;
-
- protected:
-  std::unique_ptr<zim::Archive> zimArchive;
-  std::string zimFilePath;
-
-  SuggestionsList_t suggestions;
-  SuggestionsList_t::iterator suggestionsOffset;
-
- private:
-  std::map<const std::string, unsigned int> parseCounterMetadata() const;
-};
-}
-
-#endif

include/search_renderer.h

@@ -21,11 +21,12 @@
 #define KIWIX_SEARCH_RENDERER_H
 
 #include <string>
+#include <zim/search.h>
+#include "library.h"
 
 namespace kiwix
 {
 
-class Searcher;
 class NameMapper;
 /**
  * The SearcherRenderer class is used to render a search result to a html page.
@@ -34,21 +35,25 @@ class SearchRenderer
 {
  public:
   /**
-   * The default constructor.
+   * Construct a SearchRenderer from a SearchResultSet.
    *
-   * @param humanReadableName The global zim's humanReadableName.
-   *                          Used to generate pagination links.
+   * @param srs The `SearchResultSet` to render.
+   * @param start The start offset used for the srs.
+   * @param estimatedResultCount The estimatedResultCount of the whole search
    */
-  SearchRenderer(Searcher* searcher, NameMapper* mapper);
+  SearchRenderer(zim::SearchResultSet srs, unsigned int start, unsigned int estimatedResultCount);
 
   ~SearchRenderer();
 
+  /**
+   * Set the search pattern used to do the search
+   */
   void setSearchPattern(const std::string& pattern);
 
   /**
-   * Set the search content id.
+   * Set the querystring used to select books
    */
-  void setSearchContent(const std::string& name);
+  void setSearchBookQuery(const std::string& bookQuery);
 
   /**
    * Set protocol prefix.
@@ -68,21 +73,46 @@ class SearchRenderer
   }
 
   /**
-   * Generate the html page with the resutls of the search.
+   * set user language
    */
-  std::string getHtml();
+  void setUserLang(const std::string& lang){
+    this->userlang  = lang;
+  }
+
+  /**
+   * Generate the html page with the resutls of the search.
+   *
+   * @param mapper The `NameMapper` to use to do the rendering.
+   * @param library The `Library` to use to look up book details for search results.
+                    May be nullptr. In this case, bookName is not set in the rendered string.
+   * @return The html string
+   */
+  std::string getHtml(const NameMapper& mapper, const Library* library);
+
+  /**
+   * Generate the xml page with the resutls of the search.
+   *
+   * @param mapper The `NameMapper` to use to do the rendering.
+   * @param library The `Library` to use to look up book details for search results.
+                    May be nullptr. In this case, bookName is not set in the rendered string.
+   * @return The xml string
+   */
+  std::string getXml(const NameMapper& mapper, const Library* library);
+
+ protected: // function
+  std::string renderTemplate(const std::string& tmpl_str, const NameMapper& mapper, const Library *library);
 
  protected:
   std::string beautifyInteger(const unsigned int number);
-  Searcher* mp_searcher;
-  NameMapper* mp_nameMapper;
-  std::string searchContent;
+  zim::SearchResultSet m_srs;
+  std::string searchBookQuery;
   std::string searchPattern;
   std::string protocolPrefix;
   std::string searchProtocolPrefix;
   unsigned int pageLength;
   unsigned int estimatedResultCount;
   unsigned int resultStart;
+  std::string userlang = "en";
 };
 
 

include/searcher.h

@@ -1,171 +0,0 @@
-/*
- * Copyright 2011 Emmanuel Engelhart <kelson@kiwix.org>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU  General Public License as published by
- * the Free Software Foundation; either version 3 of the License, or
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
- * MA 02110-1301, USA.
- */
-
-#ifndef KIWIX_SEARCHER_H
-#define KIWIX_SEARCHER_H
-
-#include <stdio.h>
-#include <stdlib.h>
-#include <unicode/putil.h>
-#include <algorithm>
-#include <cctype>
-#include <locale>
-#include <string>
-#include <memory>
-#include <vector>
-#include "tools/pathTools.h"
-#include "tools/stringTools.h"
-
-using namespace std;
-
-namespace kiwix
-{
-class Reader;
-class Result
-{
- public:
-  virtual ~Result(){};
-  virtual std::string get_url() = 0;
-  virtual std::string get_title() = 0;
-  virtual int get_score() = 0;
-  virtual std::string get_snippet() = 0;
-  virtual std::string get_content() = 0;
-  virtual int get_wordCount() = 0;
-  virtual int get_size() = 0;
-  virtual std::string get_zimId() = 0;
-};
-
-struct SearcherInternal;
-/**
- * The Searcher class is reponsible to do different kind of search using the
- * fulltext index.
- */
-class Searcher
-{
- public:
-  /**
-   * The default constructor.
-   */
-  Searcher();
-
-  ~Searcher();
-
-  /**
-   * Add a reader (containing embedded fulltext index) to the search.
-   *
-   * @param reader The Reader for the zim containing the fulltext index.
-   * @return true if the reader has been added.
-   *         false if the reader cannot be added (no embedded fulltext index present)
-   */
-  bool add_reader(Reader* reader);
-
-
-  Reader* get_reader(int index);
-
-  /**
-   * Start a search on the zim associated to the Searcher.
-   *
-   * Search results should be retrived using the getNextResult method.
-   *
-   * @param search The search query.
-   * @param resultStart the start offset of the search results (used for pagination).
-   * @param resultEnd the end offset of the search results (used for pagination).
-   * @param verbose print some info on stdout if true.
-   */
-  void search(const std::string& search,
-              unsigned int resultStart,
-              unsigned int resultEnd,
-              const bool verbose = false);
-
-  /**
-   * Start a geographique search.
-   * The search return result for entry in a disc of center latitude/longitude
-   * and radius distance.
-   *
-   * Search results should be retrived using the getNextResult method.
-   *
-   * @param latitude The latitude of the center point.
-   * @param longitude The longitude of the center point.
-   * @param distance The radius of the disc.
-   * @param resultStart the start offset of the search results (used for pagination).
-   * @param resultEnd the end offset of the search results (used for pagination).
-   * @param verbose print some info on stdout if true.
-   */
-  void geo_search(float latitude, float longitude, float distance,
-                  unsigned int resultStart,
-                  unsigned int resultEnd,
-                  const bool verbose = false);
-
-  /**
-   * Start a suggestion search.
-   * The search made depend of the "version" of the embedded index.
-   *  - If the index is newer enough and have a title namespace, the search is
-   *    made in the titles only.
-   *  - Else the search is made on the whole article content.
-   * In any case, the search is made "partial" (as adding '*' at the end of the query)
-   *
-   * @param search The search query.
-   * @param verbose print some info on stdout if true.
-   */
-  void suggestions(std::string& search, const bool verbose = false);
-
-  /**
-   * Get the next result of a started search.
-   * This is the method to use to loop hover the search results.
-   */
-  Result* getNextResult();
-
-  /**
-   * Restart the previous search.
-   * Next call to getNextResult will return the first result.
-   */
-  void restart_search();
-
-  /**
-   * Get a estimation of the result count.
-   */
-  unsigned int getEstimatedResultCount();
-
-  unsigned int getResultStart() { return resultStart; }
-  unsigned int getResultEnd() { return resultEnd; }
-
- protected:
-  std::string beautifyInteger(const unsigned int number);
-  void closeIndex();
-  void searchInIndex(string& search,
-                     const unsigned int resultStart,
-                     const unsigned int resultEnd,
-                     const bool verbose = false);
-
-  std::vector<Reader*> readers;
-  std::unique_ptr<SearcherInternal> internal;
-  std::string searchPattern;
-  unsigned int estimatedResultCount;
-  unsigned int resultStart;
-  unsigned int resultEnd;
-
- private:
-  void reset();
-
-};
-
-
-}
-
-#endif

include/server.h

@@ -22,6 +22,7 @@
 
 #include <string>
 #include <memory>
+#include "tools.h"
 
 namespace kiwix
 {
@@ -36,7 +37,7 @@ namespace kiwix
         *
         * @param library The library to serve.
         */
-       Server(Library* library, NameMapper* nameMapper=nullptr);
+       Server(std::shared_ptr<Library> library, std::shared_ptr<NameMapper> nameMapper=nullptr);
 
        virtual ~Server();
 
@@ -51,26 +52,42 @@ namespace kiwix
        void stop();
 
        void setRoot(const std::string& root);
-       void setAddress(const std::string& addr) { m_addr = addr; }
+       void setAddress(const std::string& addr);
        void setPort(int port) { m_port = port; }
        void setNbThreads(int threads) { m_nbThreads = threads; }
+       void setMultiZimSearchLimit(unsigned int limit) { m_multizimSearchLimit = limit; }
+       void setIpConnectionLimit(int limit) { m_ipConnectionLimit = limit; }
        void setVerbose(bool verbose) { m_verbose = verbose; }
+       void setIndexTemplateString(const std::string& indexTemplateString) { m_indexTemplateString = indexTemplateString; }
        void setTaskbar(bool withTaskbar, bool withLibraryButton)
         { m_withTaskbar = withTaskbar; m_withLibraryButton = withLibraryButton; }
        void setBlockExternalLinks(bool blockExternalLinks)
         { m_blockExternalLinks = blockExternalLinks; }
+       void setCatalogOnlyMode(bool enable) { m_catalogOnlyMode = enable; }
+       void setContentServerUrl(std::string url) { m_contentServerUrl = url; }
+       void setIpMode(IpMode mode) { m_ipMode = mode; }
+       int getPort() const;
+       IpAddress getAddress() const;
+       IpMode getIpMode() const;
+       std::vector<std::string> getServerAccessUrls() const;
 
      protected:
-       Library* mp_library;
-       NameMapper* mp_nameMapper;
+       std::shared_ptr<Library> mp_library;
+       std::shared_ptr<NameMapper> mp_nameMapper;
        std::string m_root = "";
-       std::string m_addr = "";
+       IpAddress m_addr;
+       std::string m_indexTemplateString = "";
        int m_port = 80;
        int m_nbThreads = 1;
+       unsigned int m_multizimSearchLimit = 0;
        bool m_verbose = false;
        bool m_withTaskbar = true;
        bool m_withLibraryButton = true;
        bool m_blockExternalLinks = false;
+       IpMode m_ipMode = IpMode::AUTO;
+       int m_ipConnectionLimit = 0;
+       bool m_catalogOnlyMode = false;
+       std::string m_contentServerUrl;
        std::unique_ptr<InternalServer> mp_server;
   };
 }

include/spelling_correction.h

@@ -1,6 +1,5 @@
 /*
- * Copyright (C) 2013 Emmanuel Engelhart <kelson@kiwix.org>
- * Copyright (C) 2017 Matthieu Gautier <mgautier@kymeria.fr>
+ * Copyright (C) 2025 Veloman Yunkan
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU  General Public License as published by
@@ -18,27 +17,42 @@
  * MA 02110-1301, USA.
  */
 
-#include <jni.h>
-#include "org_kiwix_kiwixlib_JNIICU.h"
+#ifndef KIWIX_SPELLING_CORRECTION_H
+#define KIWIX_SPELLING_CORRECTION_H
 
-#include <iostream>
+#include <filesystem>
+#include <memory>
 #include <string>
+#include <vector>
 
-#include "unicode/putil.h"
-
-#include "utils.h"
-
-std::mutex globalLock;
-
-JNIEXPORT void JNICALL Java_org_kiwix_kiwixlib_JNIICU_setDataDirectory(
-    JNIEnv* env, jclass kclass, jstring dirStr)
+namespace zim
 {
-  std::string cPath = jni2c(dirStr, env);
-
-  Lock l;
-  try {
-    u_setDataDirectory(cPath.c_str());
-  } catch (...) {
-    std::cerr << "Unable to set data directory " << cPath << std::endl;
-  }
+class Archive;
 }
+
+namespace Xapian
+{
+class Database;
+}
+
+namespace kiwix
+{
+
+class SpellingsDB
+{
+public: // functions
+  SpellingsDB(const zim::Archive& archive, std::filesystem::path cacheDirPath);
+  ~SpellingsDB();
+
+  SpellingsDB(const SpellingsDB& ) = delete;
+  void operator=(const SpellingsDB& ) = delete;
+
+  std::vector<std::string> getSpellingCorrections(const std::string& word, uint32_t maxCount) const;
+
+private: // data
+  std::unique_ptr<Xapian::Database> impl_;
+};
+
+} // namespace kiwix
+
+#endif // KIWIX_SPELLING_CORRECTION_H

include/tools.h

@@ -0,0 +1,271 @@
+/*
+ * Copyright 2021 Matthieu Gautier <mgautier@kymeria.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#ifndef KIWIX_TOOLS_H
+#define KIWIX_TOOLS_H
+
+#include <string>
+#include <vector>
+#include <map>
+#include <cstdint>
+#include "common.h"
+
+namespace kiwix
+{
+
+struct IpAddress
+{
+    std::string addr;  // IPv4 address
+    std::string addr6; // IPv6 address
+};
+
+typedef std::pair<std::string, std::string> LangNameCodePair;
+typedef std::vector<LangNameCodePair> FeedLanguages;
+typedef std::vector<std::string> FeedCategories;
+
+/**
+ * Return the current directory.
+ *
+ * @return the current directory (utf8 encoded)
+ */
+std::string getCurrentDirectory();
+
+/** Return the path of the executable
+ *
+ * Some application may be packaged in auto extractible archive (Appimage) and the
+ * real executable is different of the path of the archive.
+ * If `realPathOnly` is true, return the path of the real executable instead of the
+ * archive launched by the user.
+ *
+ * @param realPathOnly If we must return the real path of the executable.
+ * @return the path of the executable (utf8 encoded)
+ */
+std::string getExecutablePath(bool realPathOnly = false);
+
+/** Tell if the path is a relative path.
+ *
+ * This function is provided as a small helper. It is probably better to use native tools
+ * to manipulate paths.
+ *
+ * @param path A utf8 encoded path.
+ * @return true if the path is relative.
+ */
+bool isRelativePath(const std::string& path);
+
+/** Append a path to another one.
+ *
+ * This function is provided as a small helper. It is probably better to use native tools
+ * to manipulate paths.
+ *
+ * @param basePath the base path.
+ * @param relativePath a path to add to the base path, must be a relative path.
+ * @return The concatenation of the paths, using the right separator.
+ */
+std::string appendToDirectory(const std::string& basePath, const std::string& relativePath);
+
+/** Remove the last element of a path.
+ *
+ * This function is provided as a small helper. It is probably better to use native tools
+ * to manipulate paths.
+ *
+ * @param path a path.
+ * @return The parent directory (or empty string if none).
+ */
+std::string removeLastPathElement(const std::string& path);
+
+/** Get the last element of a path.
+ *
+ * This function is provided as a small helper. It is probably better to use native tools
+ * to manipulate paths.
+ *
+ * @param path a path.
+ * @return The base name of the path or empty string if none (ending with a separator).
+ */
+std::string getLastPathElement(const std::string& path);
+
+/** Compute the absolute path of a relative path based on another one
+ *
+ * Equivalent to appendToDirectory followed by a normalization of the path.
+ *
+ * This function is provided as a small helper. It is probably better to use native tools
+ * to manipulate paths.
+ *
+ * @param path the base path (if empty, current directory is taken).
+ * @param relativePath the relative path.
+ * @return a absolute path.
+ */
+std::string computeAbsolutePath(const std::string& path, const std::string& relativePath);
+
+/** Compute the relative path of a path relative to another one
+ *
+ * This function is provided as a small helper. It is probably better to use native tools
+ * to manipulate paths.
+ *
+ * @param path the base path.
+ * @param absolutePath the absolute path to find the relative path for.
+ * @return a relative path (pointing to absolutePath, relative to path).
+ */
+std::string computeRelativePath(const std::string& path, const std::string& absolutePath);
+
+/** Sleep the current thread.
+ *
+ * This function is provided as a small helper. It is probably better to use native tools.
+ *
+ * @param milliseconds The number of milliseconds to wait for.
+ */
+void sleep(unsigned int milliseconds);
+
+/** Split a string
+ *
+ * This function is provided as a small helper. It is probably better to use native tools.
+ *
+ * Assuming text = "foo:;bar;baz,oups;"
+ *
+ * split(text, ":;", true, true) => ["foo", ":", ";", "bar", ";", "baz,oups", ";"]
+ * split(text, ":;", true, false) => ["foo", "bar", "baz,oups"] (default)
+ * split(text, ":;", false, true) => ["foo", ":", "", ";", "bar", ";", "baz,oups", ";", ""]
+ * split(text, ":;", false, false) => ["foo", "", "bar", "baz,oups", ""]
+ *
+ * @param str The string to split.
+ * @param delims A string of potential delimiters.
+ *               Each charater in the string can be a individual delimiters.
+ * @param dropEmpty true if empty part must be dropped from the result.
+ * @param keepDelim true if delimiter must be included from the result.
+ * @return a list of part (potentially containing delimiters)
+ */
+std::vector<std::string> split(const std::string& str, const std::string& delims, bool dropEmpty=true, bool keepDelim = false);
+
+/** Convert language code from iso2 code to iso3
+ *
+ * This function is provided as a small helper. It is probably better to use native tools
+ * to manipulate locales.
+ *
+ * @param a2code a iso2 code string.
+ * @return the corresponding iso3 code.
+ * @throw std::out_of_range if iso2 code is not known.
+ */
+std::string converta2toa3(const std::string& a2code);
+
+/** Extracts content from given file.
+ *
+ * This function provides content of a file provided it's path.
+ *
+ * @param path The absolute path provided in string format.
+ * @return Content of corresponding file in string format.
+ */
+std::string getFileContent(const std::string& path);
+
+/** Checks if file exists.
+ *
+ * This function returns boolean stating if file exists.
+ *
+ * @param path The absolute path provided in string format.
+ * @return Boolean representing if file exists or not.
+ */
+bool fileExists(const std::string& path);
+
+/** Checks if file is readable.
+ *
+ * This function returns boolean stating if file is readable.
+ *
+ * @param path The absolute path provided in string format.
+ * @return Boolean representing if file is readale or not.
+ */
+bool fileReadable(const std::string& path);
+
+/** Provides mimetype from filename.
+ *
+ * This function provides mimetype from file-name.
+ *
+ * @param filename string containing filename.
+ * @return mimetype from filename in string format.
+ */
+std::string getMimeTypeForFile(const std::string& filename);
+
+/** Provides all available network interfaces
+ *
+ * This function provides the available IPv4 and IPv6 network interfaces
+ * as a map from the interface name to its IPv4 and/or IPv6 address(es).
+ */
+std::map<std::string, IpAddress> getNetworkInterfacesIPv4Or6();
+
+/** Provides all available IPv4 network interfaces
+ *
+ * This function provides the available IPv4 network interfaces
+ * as a map from the interface name to its IPv4 address.
+ *
+ * Provided for backward compatibility with libkiwix v13.1.0.
+ */
+std::map<std::string, std::string> getNetworkInterfaces();
+
+/** Provides the best IP address
+ * This function provides the best IP addresses for both ipv4 and ipv6 protocols,
+ * in an IpAddress struct, based on the list given by getNetworkInterfacesIPv4Or6()
+ */
+IpAddress getBestPublicIps();
+
+/** Provides the best IPv4 adddress
+ * Equivalent to getBestPublicIp(false). Provided for backward compatibility
+ * with libkiwix v13.1.0.
+ */
+std::string getBestPublicIp();
+
+/** Converts file size to human readable format.
+ *
+ * This function will convert a number to its equivalent size using units.
+ *
+ * @param number file size in bytes.
+ * @return a human-readable string representation of the size, e.g., "2.3 KB", "1.8 MB", "5.2 GB".
+ */
+std::string beautifyFileSize(uint64_t number);
+
+/**
+ * Load languages stored in an OPDS stream.
+ *
+ * @param content the OPDS stream.
+ * @return vector containing pairs of language code and their corresponding full language name.
+ */
+FeedLanguages readLanguagesFromFeed(const std::string& content);
+
+/**
+ * Load categories stored in an OPDS stream .
+ *
+ * @param content the OPDS stream.
+ * @return vector containing category strings.
+ */
+FeedCategories readCategoriesFromFeed(const std::string& content);
+
+/**
+ * Retrieve the full language name associated with a given ISO 639-3 language code.
+ *
+ * @param lang ISO 639-3 language code.
+ * @return full language name.
+ */
+std::string getLanguageSelfName(const std::string& lang);
+
+/**
+ * Slugifies the filename by converting any characters reserved by the operating
+ * system to '_'. Note filename is only the file name and not a path.
+ *
+ * @param filename Valid UTF-8 encoded file name string. 
+ * @return slugified string.
+ */
+std::string getSlugifiedFileName(const std::string& filename);
+}
+#endif // KIWIX_TOOLS_H

include/version.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2013 Emmanuel Engelhart <kelson@kiwix.org>
+ * Copyright 2021 Emmanuel Engelhart <kelson@kiwix.org>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU  General Public License as published by
@@ -17,21 +17,18 @@
  * MA 02110-1301, USA.
  */
 
-package org.kiwix.kiwixlib;
+#ifndef KIWIX_VERSION_H
+#define KIWIX_VERSION_H
 
-public class JNIKiwixString
+#include <string>
+#include <vector>
+#include <iostream>
+
+namespace kiwix
 {
-  public String value;
-
-  public JNIKiwixString(String value) {
-    this.value = value;
-  }
-
-  public JNIKiwixString() {
-    this("");
-  }
-
-  public String getValue() {
-    return value;
-  }
+  typedef std::vector<std::pair<std::string, std::string>> LibVersions;
+  LibVersions getVersions();
+  void printVersions(std::ostream& out = std::cout);
 }
+
+#endif // KIWIX_VERSION_H

kiwix.pc.in

@@ -1,10 +0,0 @@
-prefix=@prefix@
-libdir=${prefix}/lib64
-includedir=${prefix}/include
-
-Name: libkiwix
-Description: A library that contains a lot of things used by used by other kiwix programs
-Version: @version@
-Requires: @requires@
-Libs: -L${libdir} -lkiwix @extra_libs@
-Cflags: -I${includedir}/ @extra_cflags@

meson.build

@@ -1,35 +1,61 @@
 project('libkiwix', 'cpp',
-  version : '10.0.0', # Also change this in android-kiwix-lib-publisher/kiwixLibAndroid/build.gradle
+  version : '14.1.1',
   license : 'GPLv3+',
-  default_options : ['c_std=c11', 'cpp_std=c++11', 'werror=true'])
+  default_options : ['c_std=c11', 'cpp_std=c++17', 'werror=true'])
 
 compiler = meson.get_compiler('cpp')
+static_deps = get_option('static-linkage') or get_option('default_library') == 'static'
+extra_libs = []
 
-wrapper = get_option('wrapper')
+# Atomics as compiled by GCC or clang can lead to external references to
+# functions depending on the type size and the platform. LLVM provides them in
+# 'libcompiler_rt', which clang normally automatically links in, while GNU
+# provides them in 'libatomic', which GCC *does not* link in automatically (but
+# this is probably going to change, see
+# https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81358). Regardless of the setup
+# of the compiler driver itself (GCC or clang), we can thus assume that if some
+# atomic references can't be resolved, then 'libatomic' is missing.
+atomics_program = '''
+#include <atomic>
+#include <cstdint>
 
-static_deps = wrapper.contains('android') or wrapper.contains('java') or get_option('default_library') == 'static'
-if wrapper.contains('android')
-  extra_libs = ['-llog']
-else
-  extra_libs = []
+using namespace std;
+
+int main() {
+  volatile atomic_bool a_b(true);
+  volatile atomic_ullong a_ull(-1);
+  // Next two lines are to cover atomic<socket_t> from 'httplib.h'.
+  volatile atomic<uint32_t> a_u32(-1);
+  volatile atomic<uint64_t> a_u64(-1);
+
+  return atomic_load(&a_b) == false && atomic_load(&a_ull) == 0 &&
+    atomic_load(&a_u32) == 0 && atomic_load(&a_u64) == 0;
+}
+'''
+if not compiler.links(atomics_program,
+                      name: 'compiler driver readily supports atomics')
+  libatomic = compiler.find_library('atomic')
+  compiler.links(atomics_program, name: 'atomics work with libatomic',
+                 dependencies: libatomic, required: true)
+  extra_libs += ['-latomic']
 endif
 
-if wrapper.contains('java')
-  add_languages('java')
-endif
-
-# See https://github.com/kiwix/libkiwix/issues/371
-if ['arm', 'mips', 'm68k', 'ppc', 'sh4'].contains(target_machine.cpu_family())
-  extra_libs += '-latomic'
-endif
-
-if (compiler.get_id() == 'gcc' and build_machine.system() == 'linux') or target_machine.system() == 'freebsd'
-  # C++ std::thread is implemented using pthread on linux by gcc
+# C++ std::thread is implemented using pthread on Linux by GCC, and on FreeBSD
+# for both GCC and LLVM.
+if (host_machine.system() == 'linux' and compiler.get_id() == 'gcc') or \
+   host_machine.system() == 'freebsd'
   thread_dep = dependency('threads')
 else
   thread_dep = dependency('', required:false)
 endif
+
 libicu_dep = dependency('icu-i18n', static:static_deps)
+if libicu_dep.version().version_compare('>= 76')
+  libicu_deps = [libicu_dep, dependency('icu-uc', static:static_deps)]
+else
+  libicu_deps = [libicu_dep]
+endif
+
 pugixml_dep = dependency('pugixml', static:static_deps)
 libcurl_dep = dependency('libcurl', static:static_deps)
 microhttpd_dep = dependency('libmicrohttpd', static:static_deps)
@@ -44,52 +70,53 @@ else
   error('Cannot found header mustache.hpp')
 endif
 
-libzim_dep = dependency('libzim', version : '>=7.0.0', static:static_deps)
-if not compiler.has_header_symbol('zim/zim.h', 'LIBZIM_WITH_XAPIAN')
-  error('Libzim seems to be compiled without xapian. Xapian support is mandatory.')
+libzim_dep = dependency('libzim', version:['>=9.4.0', '<10.0.0'], static:static_deps)
+
+if not compiler.has_header_symbol('zim/zim.h', 'LIBZIM_WITH_XAPIAN', dependencies: libzim_dep)
+  error('Libzim seems to be compiled without Xapian. Xapian support is mandatory.')
 endif
 
 
 extra_cflags = ''
-if target_machine.system() == 'windows' and static_deps
+if host_machine.system() == 'windows' and static_deps
   add_project_arguments('-DCURL_STATICLIB', language : 'cpp')
   extra_cflags += '-DCURL_STATICLIB'
 endif
 
-if target_machine.system() == 'windows'
+if host_machine.system() == 'windows'
   add_project_arguments('-DNOMINMAX', language: 'cpp')
+  extra_libs += ['-liphlpapi']
 endif
 
-all_deps = [thread_dep, libicu_dep, libzim_dep, pugixml_dep, libcurl_dep, microhttpd_dep, zlib_dep, xapian_dep]
+if build_machine.system() == 'windows'
+  extra_libs += ['-lshlwapi', '-lwinmm']
+endif
+
+
+# Dependencies as string
+all_deps = [thread_dep, libzim_dep, pugixml_dep, libcurl_dep, microhttpd_dep, zlib_dep, xapian_dep]
+
+# Dependencies as array
+all_deps += libicu_deps
 
 inc = include_directories('include', extra_include)
 
 conf = configuration_data()
-conf.set('VERSION', '"@0@"'.format(meson.project_version()))
-
-if build_machine.system() == 'windows'
-     extra_link_args = ['-lshlwapi', '-lwinmm']
-else
-     extra_link_args = []
-endif
+conf.set('LIBKIWIX_VERSION', '"@0@"'.format(meson.project_version()))
 
 subdir('include')
 subdir('scripts')
 subdir('static')
 subdir('src')
 subdir('test')
+if get_option('doc')
+  subdir('docs')
+endif
 
-pkg_requires = ['libzim', 'icu-i18n', 'pugixml', 'libcurl', 'libmicrohttpd', 'xapian-core']
-
-pkg_conf = configuration_data()
-pkg_conf.set('prefix', get_option('prefix'))
-pkg_conf.set('requires', ' '.join(pkg_requires))
-pkg_conf.set('extra_libs', ' '.join(extra_libs))
-pkg_conf.set('extra_cflags', extra_cflags)
-pkg_conf.set('version', meson.project_version())
-configure_file(output : 'kiwix.pc',
-               configuration : pkg_conf,
-               input : 'kiwix.pc.in',
-               install_dir: get_option('libdir')+'/pkgconfig'
-              )
-
+pkg_mod = import('pkgconfig')
+pkg_mod.generate(libraries : [libkiwix] + extra_libs,
+                 version : meson.project_version(),
+                 name : 'libkiwix',
+                 filebase : 'libkiwix',
+                 description : 'A library that contains useful primitives that Kiwix readers have in common',
+                 extra_cflags: extra_cflags)

meson_options.txt

@@ -1,2 +1,4 @@
-option('wrapper', type:'array', choices:['java', 'android'], value:[],
-  description: 'The wrapper to generate.')
+option('static-linkage', type : 'boolean', value : false,
+  description : 'Link statically with the dependencies.')
+option('doc', type : 'boolean', value : false,
+  description : 'Build the documentations.')

scripts/format_code.sh

@@ -0,0 +1,14 @@
+#!/usr/bin/bash
+
+# Compute 'src' path
+SCRIPT_DIR=$(dirname "$0")
+REPO_DIR=$(readlink -f "$SCRIPT_DIR"/..)
+DIRS="src include"
+
+# Apply formating to all *.cpp and *.h files
+cd "$REPO_DIR"
+for FILE in $(find $DIRS -name '*.h' -o -name '*.cpp')
+do
+  echo $FILE
+  clang-format -i -style=file "$FILE"
+done

scripts/kiwix-compile-i18n

@@ -0,0 +1,148 @@
+#!/usr/bin/env python3
+
+'''
+Copyright 2022 Veloman Yunkan <veloman.yunkan@gmail.com>
+
+This program is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3 of the License, or any
+later version.
+
+This program is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with this program; if not, write to the Free Software
+Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+02110-1301, USA.
+'''
+
+import argparse
+import os.path
+import re
+import json
+
+def to_identifier(name):
+    ident = re.sub(r'[^0-9a-zA-Z]', '_', name)
+    if ident[0].isnumeric():
+        return "_"+ident
+    return ident
+
+def lang_code(filename):
+    filename = os.path.basename(filename)
+    lang = to_identifier(os.path.splitext(filename)[0])
+    print(filename, '->', lang)
+    return lang
+
+from string import Template
+
+def expand_cxx_template(t, **kwargs):
+    return Template(t).substitute(**kwargs)
+
+def cxx_string_literal(s):
+    # Taking advantage of the fact the JSON string escape rules match
+    # those of C++
+    return 'u8' + json.dumps(s)
+
+string_table_cxx_template = '''
+const I18nString $TABLE_NAME[] = {
+  $TABLE_ENTRIES
+};
+'''
+
+lang_table_entry_cxx_template = '''
+    {
+        $LANG_STRING_LITERAL,
+        ARRAY_ELEMENT_COUNT($STRING_TABLE_NAME),
+        $STRING_TABLE_NAME
+    }'''
+
+cxxfile_template = '''// This file is automatically generated. Do not modify it.
+
+#include "server/i18n_utils.h"
+
+namespace kiwix {
+namespace i18n {
+
+namespace
+{
+
+$STRING_DATA
+
+} // unnamed namespace
+
+#define ARRAY_ELEMENT_COUNT(a) (sizeof(a)/sizeof(a[0]))
+
+extern const I18nStringTable stringTables[] = {
+  $LANG_TABLE
+};
+
+extern const size_t langCount = $LANG_COUNT;
+
+} // namespace i18n
+} // namespace kiwix
+'''
+
+class Resource:
+    def __init__(self, filename):
+        filename = filename.strip()
+        self.filename = filename
+        self.lang_code = lang_code(filename)
+        with open(filename, 'r', encoding='utf-8') as f:
+            self.data = f.read()
+
+    def get_string_table_name(self):
+        return "string_table_for_" + self.lang_code
+
+    def get_string_table(self):
+        table_entries = ",\n  ".join(self.get_string_table_entries())
+        return expand_cxx_template(string_table_cxx_template,
+                    TABLE_NAME=self.get_string_table_name(),
+                    TABLE_ENTRIES=table_entries)
+
+    def get_string_table_entries(self):
+        d = json.loads(self.data)
+        for k in sorted(d.keys()):
+            if k != "@metadata":
+                key_string = cxx_string_literal(k)
+                value_string = cxx_string_literal(d[k])
+                yield '{ ' + key_string + ', ' + value_string + ' }'
+
+    def get_lang_table_entry(self):
+        return expand_cxx_template(lang_table_entry_cxx_template,
+                LANG_STRING_LITERAL=cxx_string_literal(self.lang_code),
+                STRING_TABLE_NAME=self.get_string_table_name())
+
+
+
+def gen_c_file(resources):
+    string_data = []
+    lang_table = []
+    for r in resources:
+        string_data.append(r.get_string_table())
+        lang_table.append(r.get_lang_table_entry())
+
+    return expand_cxx_template(cxxfile_template,
+                STRING_DATA="\n".join(string_data),
+                LANG_TABLE=",\n  ".join(lang_table),
+                LANG_COUNT=len(resources)
+           )
+
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument('--cxxfile',
+                        required=True,
+                        help='The Cpp file name to generate')
+    parser.add_argument('i18n_resource_files', nargs='+',
+                        help='The list of resources to compile.')
+    args = parser.parse_args()
+
+    resources = [Resource(filename) for filename in args.i18n_resource_files]
+
+    with open(args.cxxfile, 'w') as f:
+        f.write(gen_c_file(resources))
+

scripts/kiwix-compile-i18n.1

@@ -0,0 +1,18 @@
+.TH KIWIX-COMPILE-I18N "1" "January 2022" "Kiwix" "User Commands"
+.SH NAME
+kiwix-compile-i18n \- helper to compile Kiwix i18n (internationalization) data
+.SH SYNOPSIS
+\fBkiwix\-compile\-i18n\fR [\-h] \-\-cxxfile CXXFILE i18n_resource_files ...\fR
+.SH DESCRIPTION
+.TP
+i18n_resource_files ...
+The list of i18n resources to compile.
+.TP
+\fB\-h\fR, \fB\-\-help\fR
+show a help message and exit
+.TP
+\fB\-\-cxxfile\fR CXXFILE
+The Cpp file name to generate
+.TP
+.SH AUTHOR
+Veloman Yunkan <veloman.yunkan@gmail.com>

scripts/kiwix-compile-resources

@@ -52,26 +52,60 @@ resource_getter_template = """
         return RESOURCE::{identifier};
 """
 
+resource_cacheid_getter_template = """
+    if (name == "{common_name}")
+        return "{cacheid}";
+"""
+
 resource_decl_template = """{namespaces_open}
 extern const std::string {identifier};
 {namespaces_close}"""
 
+BINARY_RESOURCE_EXTENSIONS = {'.ico', '.png', '.ttf'}
+
+TEXT_RESOURCE_EXTENSIONS = {
+    '.css',
+    '.html',
+    '.js',
+    '.json',
+    '.svg',
+    '.tmpl',
+    '.webmanifest',
+    '.xml',
+}
+
+if not BINARY_RESOURCE_EXTENSIONS.isdisjoint(TEXT_RESOURCE_EXTENSIONS):
+    raise RuntimeError(f"The following file type extensions are declared to be both binary and text: {BINARY_RESOURCE_EXTENSIONS.intersection(TEXT_RESOURCE_EXTENSIONS)}")
+
+def is_binary_resource(filename):
+    _, extension = os.path.splitext(filename)
+    is_binary = extension in BINARY_RESOURCE_EXTENSIONS
+    is_text = extension in TEXT_RESOURCE_EXTENSIONS
+    if not is_binary and not is_text:
+        # all file type extensions of static resources must be listed
+        # in either BINARY_RESOURCE_EXTENSIONS or TEXT_RESOURCE_EXTENSIONS
+        raise RuntimeError(f"Unknown file type extension: {extension}")
+    return is_binary
+
 class Resource:
-    def __init__(self, base_dirs, filename):
-        filename = filename.strip()
+    def __init__(self, base_dirs, filename, cacheid=None):
+        filename = filename
         self.filename = filename
         self.identifier = full_identifier(filename)
+        self.cacheid = cacheid
         found = False
         for base_dir in base_dirs:
             try:
                 with open(os.path.join(base_dir, filename), 'rb') as f:
                     self.data = f.read()
+                    if not is_binary_resource(filename):
+                        self.data = self.data.replace(b"\r\n", b"\n")
                 found = True
                 break
             except FileNotFoundError:
                 continue
         if not found:
-            raise Exception("Impossible to found {}".format(filename))
+            raise Exception("Resource not found: {}".format(filename))
 
     def dump_impl(self):
         nb_row = len(self.data)//16 + (1 if len(self.data) % 16 else 0)
@@ -93,6 +127,12 @@ class Resource:
             identifier="::".join(self.identifier)
         )
 
+    def dump_cacheid_getter(self):
+        return resource_cacheid_getter_template.format(
+            common_name=self.filename,
+            cacheid=self.cacheid
+        )
+
     def dump_decl(self):
         return resource_decl_template.format(
             namespaces_open=" ".join("namespace {} {{".format(id) for id in self.identifier[:-1]),
@@ -102,7 +142,7 @@ class Resource:
 
 
 
-master_c_template = """//This file is automaically generated. Do not modify it.
+master_c_template = """//This file is automatically generated. Do not modify it.
 
 #include <stdlib.h>
 #include <fstream>
@@ -123,7 +163,12 @@ static std::string init_resource(const char* name, const unsigned char* content,
 
 const std::string& getResource_{basename}(const std::string& name) {{
 {RESOURCES_GETTER}
-    throw ResourceNotFound("Resource not found.");
+    throw ResourceNotFound("Resource not found: " + name);
+}}
+
+const char* getResourceCacheId_{basename}(const std::string& name) {{
+{RESOURCE_CACHEID_GETTER}
+    return nullptr;
 }}
 
 {RESOURCES}
@@ -134,6 +179,7 @@ def gen_c_file(resources, basename):
     return master_c_template.format(
        RESOURCES="\n\n".join(r.dump_impl() for r in resources),
        RESOURCES_GETTER="\n\n".join(r.dump_getter() for r in resources),
+       RESOURCE_CACHEID_GETTER="\n\n".join(r.dump_cacheid_getter() for r in resources if r.cacheid is not None),
        include_file=basename,
        basename=to_identifier(basename)
     )
@@ -159,8 +205,10 @@ class ResourceNotFound : public std::runtime_error {{
 }};
 
 const std::string& getResource_{basename}(const std::string& name);
+const char* getResourceCacheId_{basename}(const std::string& name);
 
 #define getResource(a) (getResource_{basename}(a))
+#define getResourceCacheId(a) (getResourceCacheId_{basename}(a))
 
 #endif // KIWIX_{BASENAME}
 
@@ -182,15 +230,17 @@ if __name__ == "__main__":
     parser.add_argument('--source_dir',
                         help="Additional directory where to look for resources.",
                         action='append')
-    parser.add_argument('resource_file',
+    parser.add_argument('resource_files', nargs='+',
                         help='The list of resources to compile.')
     args = parser.parse_args()
 
-    base_dir = os.path.dirname(os.path.realpath(args.resource_file))
     source_dir = args.source_dir or []
-    with open(args.resource_file, 'r') as f:
-        resources = [Resource([base_dir]+source_dir, filename)
-                        for filename in f.readlines()]
+    resources = []
+    for resfile in args.resource_files:
+        base_dir = os.path.dirname(os.path.realpath(resfile))
+        with open(resfile, 'r') as f:
+            resources += [Resource([base_dir]+source_dir, *line.strip().split())
+                            for line in f.readlines()]
 
     h_identifier = to_identifier(os.path.basename(args.hfile))
     with open(args.hfile, 'w') as f:

scripts/kiwix-compile-resources.1

@@ -2,7 +2,7 @@
 .SH NAME
 kiwix-compile-resources \- helper to compile and generate some Kiwix resources
 .SH SYNOPSIS
-\fBkiwix\-compile\-resources\fR [\-h] [\-\-cxxfile CXXFILE] [\-\-hfile HFILE] resource_file\fR
+\fBkiwix\-compile\-resources\fR [\-h] [\-\-cxxfile CXXFILE] [\-\-hfile HFILE] resource_file ...\fR
 .SH DESCRIPTION
 .TP
 resource_file

scripts/kiwix-resources

@@ -0,0 +1,135 @@
+#!/usr/bin/env python3
+
+'''
+Copyright 2022 Veloman Yunkan <veloman.yunkan@gmail.com>
+
+This program is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3 of the License, or any
+later version.
+
+This program is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with this program; if not, write to the Free Software
+Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+02110-1301, USA.
+'''
+
+import argparse
+import hashlib
+import os.path
+import re
+
+def read_resource_file(resource_file_path):
+    with open(resource_file_path, 'r') as f:
+        return [line.strip() for line in f]
+
+def list_resources(resource_file_path):
+    for resource_path in read_resource_file(resource_file_path):
+        print(resource_path)
+
+def compute_resource_revision(resource_path):
+    with open(os.path.join(OUT_DIR, resource_path), 'rb') as f:
+        return hashlib.sha1(f.read()).hexdigest()[:8]
+
+resource_revisions = {}
+
+def get_resource_revision(res):
+    if not res in resource_revisions:
+        preprocess_resource(res)
+        resource_revisions[res] = compute_resource_revision(res)
+    return resource_revisions[res]
+
+RESOURCE_WITH_CACHEID_URL_PATTERN=r'(?P<pre>.*/(?P<resource>skin/[^"?]+)\?)KIWIXCACHEID(?P<post>[^"]*)'
+
+def set_cacheid(resource_matchobj):
+    pre = resource_matchobj.group('pre')
+    resource = resource_matchobj.group('resource')
+    post = resource_matchobj.group('post')
+    cacheid = 'cacheid=' + get_resource_revision(resource)
+    return pre + cacheid + post
+
+def preprocess_text(s):
+    if 'KIWIXCACHEID' in s:
+        s = re.sub(RESOURCE_WITH_CACHEID_URL_PATTERN, set_cacheid, s)
+        assert not 'KIWIXCACHEID' in s
+    return s
+
+def get_preprocessed_resource(srcpath):
+    """Get the transformed content of a resource
+
+    If the resource at srcpath is modified by preprocessing then this function
+    returns the transformed content of the resource. Otherwise it returns None.
+    """
+    try:
+        with open(srcpath, 'r') as resource_file:
+            content = resource_file.read()
+            preprocessed_content = preprocess_text(content)
+        return preprocessed_content if preprocessed_content != content else None
+    except UnicodeDecodeError:
+        # It was a binary resource
+        return None
+
+
+def symlink_resource(src, resource_path):
+    if os.path.exists(resource_path):
+        if os.path.islink(resource_path) and os.readlink(resource_path) == src:
+            return
+        os.remove(resource_path)
+    os.symlink(src, resource_path)
+
+def preprocess_resource(resource_path):
+    print('Preprocessing', resource_path, '...')
+    resource_dir = os.path.dirname(resource_path)
+    if resource_dir != '':
+        os.makedirs(os.path.join(OUT_DIR, resource_dir), exist_ok=True)
+    srcpath = os.path.join(BASE_DIR, resource_path)
+    outpath = os.path.join(OUT_DIR, resource_path)
+    if os.path.exists(outpath):
+        os.remove(outpath)
+    preprocessed_content = get_preprocessed_resource(srcpath)
+    if preprocessed_content is None:
+        symlink_resource(srcpath, outpath)
+    else:
+        with open(outpath, 'w') as target:
+            print(preprocessed_content, end='', file=target)
+
+
+def copy_resource_list_file(src_path, dst_path):
+    with open(src_path, 'r') as src:
+        with open(dst_path, 'w') as dst:
+            for line in src:
+                res = line.strip()
+                if line.startswith("skin/") and res in resource_revisions:
+                    dst.write(res + " " + resource_revisions[res] + "\n")
+                else:
+                    dst.write(line)
+
+def preprocess_resources(resource_file_path):
+    resource_filename = os.path.basename(resource_file_path)
+    for resource in read_resource_file(resource_file_path):
+        if resource.startswith('skin/'):
+            get_resource_revision(resource)
+        else:
+            preprocess_resource(resource)
+    copy_resource_list_file(resource_file_path, os.path.join(OUT_DIR, resource_filename))
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    commands = parser.add_mutually_exclusive_group()
+    commands.add_argument('--list-all', action='store_true')
+    commands.add_argument('--preprocess', action='store_true')
+    parser.add_argument('--outdir')
+    parser.add_argument('resource_file')
+    args = parser.parse_args()
+    BASE_DIR = os.path.dirname(os.path.realpath(args.resource_file))
+    OUT_DIR = args.outdir
+
+    if args.list_all:
+        list_resources(args.resource_file)
+    elif args.preprocess:
+        preprocess_resources(args.resource_file)

scripts/meson.build

@@ -1,6 +1,13 @@
 
+res_manager = find_program('kiwix-resources')
 res_compiler = find_program('kiwix-compile-resources')
 
 install_data(res_compiler.path(), install_dir:get_option('bindir'))
 
 install_man('kiwix-compile-resources.1')
+
+i18n_compiler = find_program('kiwix-compile-i18n')
+
+install_data(i18n_compiler.path(), install_dir:get_option('bindir'))
+
+install_man('kiwix-compile-i18n.1')

src/aria2.cpp

@@ -3,13 +3,16 @@
 #include "aria2.h"
 #include "xmlrpc.h"
 #include <iostream>
+#include <algorithm>
+#include <fstream>
 #include <sstream>
 #include <thread>
 #include <chrono>
-#include <tools/otherTools.h>
-#include <tools/pathTools.h>
-#include <tools/stringTools.h>
-#include <downloader.h> // For AriaError
+#include "tools.h"
+#include "tools/pathTools.h"
+#include "tools/stringTools.h"
+#include "tools/otherTools.h"
+#include "downloader.h" // For AriaError
 
 #ifdef _WIN32
 # define ARIA2_CMD "aria2c.exe"
@@ -22,25 +25,46 @@
 #define LOG_ARIA_ERROR() \
   { \
     std::cerr << "ERROR: aria2 RPC request failed. (" << res << ")." << std::endl; \
-    std::cerr << (m_curlErrorBuffer[0] ? m_curlErrorBuffer.get() : curl_easy_strerror(res)) << std::endl; \
+    std::cerr << (curlErrorBuffer[0] ? curlErrorBuffer : curl_easy_strerror(res)) << std::endl; \
   }
 
 namespace kiwix {
 
-Aria2::Aria2():
+namespace {
+
+void pauseAnyActiveDownloads(const std::string& ariaSessionFilePath)
+{
+    std::ifstream inputFile(ariaSessionFilePath);
+    if ( !inputFile )
+        return;
+
+    std::ostringstream ss;
+    std::string line;
+    while ( std::getline(inputFile, line) ) {
+        if ( !startsWith(line, " pause=") ) {
+            ss << line << "\n";
+        }
+        if ( !line.empty() && line[0] != ' ' && line[0] != '#' ) {
+            ss << " pause=true\n";
+        }
+    }
+
+    std::ofstream outputFile(ariaSessionFilePath);
+    outputFile << ss.str();
+}
+
+} // unnamed namespace
+
+Aria2::Aria2(std::string sessionFileDir):
   mp_aria(nullptr),
   m_port(42042),
-  m_secret("kiwixariarpc"),
-  m_curlErrorBuffer(new char[CURL_ERROR_SIZE]),
-  mp_curl(nullptr)
+  m_secret(getNewRpcSecret())
 {
-  m_downloadDir = getDataDirectory();
-  makeDirectory(m_downloadDir);
   std::vector<const char*> callCmd;
 
   std::string rpc_port = "--rpc-listen-port=" + to_string(m_port);
-  std::string download_dir = "--dir=" + getDataDirectory();
-  std::string session_file = appendToDirectory(getDataDirectory(), "kiwix.session");
+  std::string session_file = appendToDirectory(sessionFileDir, "kiwix.session");
+  pauseAnyActiveDownloads(session_file);
   std::string session = "--save-session=" + session_file;
   std::string inputFile = "--input-file=" + session_file;
 //  std::string log_dir = "--log=\"" + logDir + "\"";
@@ -67,8 +91,7 @@ Aria2::Aria2():
   callCmd.push_back("--enable-rpc");
   callCmd.push_back(rpc_secret.c_str());
   callCmd.push_back(rpc_port.c_str());
-  callCmd.push_back(download_dir.c_str());
-  if (fileExists(session_file)) {
+  if (fileReadable(session_file)) {
     callCmd.push_back(inputFile.c_str());
   }
   callCmd.push_back(session.c_str());
@@ -89,36 +112,42 @@ Aria2::Aria2():
       launchCmd.append(cmd).append(" ");
   }
   mp_aria = Subprocess::run(callCmd);
-  mp_curl = curl_easy_init();
 
-  curl_easy_setopt(mp_curl, CURLOPT_URL, "http://localhost/rpc");
-  curl_easy_setopt(mp_curl, CURLOPT_PORT, m_port);
-  curl_easy_setopt(mp_curl, CURLOPT_POST, 1L);
-  curl_easy_setopt(mp_curl, CURLOPT_ERRORBUFFER, m_curlErrorBuffer.get());
+  CURL* p_curl = curl_easy_init();
+  char curlErrorBuffer[CURL_ERROR_SIZE];
 
-  int watchdog = 50;
-  while(--watchdog) {
+  curl_easy_setopt(p_curl, CURLOPT_URL, "http://localhost/rpc");
+  curl_easy_setopt(p_curl, CURLOPT_PORT, m_port);
+  curl_easy_setopt(p_curl, CURLOPT_POST, 1L);
+  curl_easy_setopt(p_curl, CURLOPT_ERRORBUFFER, curlErrorBuffer);
+  curl_easy_setopt(p_curl, CURLOPT_TIMEOUT_MS, 100);
+
+  typedef std::chrono::duration<double> Seconds;
+
+  const double MAX_WAITING_TIME_SECONDS = 1;
+  const auto t0 = std::chrono::steady_clock::now();
+  bool maxWaitingTimeWasExceeded = false;
+
+  CURLcode res = CURLE_OK;
+  while ( !maxWaitingTimeWasExceeded ) {
     sleep(10);
-    m_curlErrorBuffer[0] = 0;
-    auto res = curl_easy_perform(mp_curl);
+    curlErrorBuffer[0] = 0;
+    res = curl_easy_perform(p_curl);
     if (res == CURLE_OK) {
       break;
-    } else if (watchdog == 1) {
-      LOG_ARIA_ERROR();
     }
+
+    const auto dt = std::chrono::steady_clock::now() - t0;
+    const double elapsedTime = std::chrono::duration_cast<Seconds>(dt).count();
+    maxWaitingTimeWasExceeded = elapsedTime > MAX_WAITING_TIME_SECONDS;
   }
-  if (!watchdog) {
-    curl_easy_cleanup(mp_curl);
+  curl_easy_cleanup(p_curl);
+  if ( maxWaitingTimeWasExceeded ) {
+    LOG_ARIA_ERROR();
     throw std::runtime_error("Cannot connect to aria2c rpc. Aria2c launch cmd : " + launchCmd);
   }
 }
 
-Aria2::~Aria2()
-{
-  std::unique_lock<std::mutex> lock(m_lock);
-  curl_easy_cleanup(mp_curl);
-}
-
 void Aria2::close()
 {
   saveSession();
@@ -138,20 +167,25 @@ std::string Aria2::doRequest(const MethodCall& methodCall)
   std::stringstream outStream;
   CURLcode res;
   long response_code;
-  {
-    std::unique_lock<std::mutex> lock(m_lock);
-    curl_easy_setopt(mp_curl, CURLOPT_POSTFIELDSIZE, requestContent.size());
-    curl_easy_setopt(mp_curl, CURLOPT_POSTFIELDS, requestContent.c_str());
-    curl_easy_setopt(mp_curl, CURLOPT_WRITEFUNCTION, &write_callback_to_iss);
-    curl_easy_setopt(mp_curl, CURLOPT_WRITEDATA, &outStream);
-    m_curlErrorBuffer[0] = 0;
-    res = curl_easy_perform(mp_curl);
-    if (res != CURLE_OK) {
-      LOG_ARIA_ERROR();
-      throw std::runtime_error("Cannot perform request");
-    }
-    curl_easy_getinfo(mp_curl, CURLINFO_RESPONSE_CODE, &response_code);
+  char curlErrorBuffer[CURL_ERROR_SIZE];
+  CURL* p_curl = curl_easy_init();
+  curl_easy_setopt(p_curl, CURLOPT_URL, "http://localhost/rpc");
+  curl_easy_setopt(p_curl, CURLOPT_PORT, m_port);
+  curl_easy_setopt(p_curl, CURLOPT_POST, 1L);
+  curl_easy_setopt(p_curl, CURLOPT_ERRORBUFFER, curlErrorBuffer);
+  curl_easy_setopt(p_curl, CURLOPT_POSTFIELDSIZE, requestContent.size());
+  curl_easy_setopt(p_curl, CURLOPT_POSTFIELDS, requestContent.c_str());
+  curl_easy_setopt(p_curl, CURLOPT_WRITEFUNCTION, &write_callback_to_iss);
+  curl_easy_setopt(p_curl, CURLOPT_WRITEDATA, &outStream);
+  curlErrorBuffer[0] = 0;
+  res = curl_easy_perform(p_curl);
+  if (res != CURLE_OK) {
+    LOG_ARIA_ERROR();
+    curl_easy_cleanup(p_curl);
+    throw std::runtime_error("Cannot perform request");
   }
+  curl_easy_getinfo(p_curl, CURLINFO_RESPONSE_CODE, &response_code);
+  curl_easy_cleanup(p_curl);
 
   auto responseContent = outStream.str();
   if (response_code != 200) {
@@ -195,6 +229,13 @@ std::string Aria2::tellStatus(const std::string& gid, const std::vector<std::str
   return doRequest(methodCall);
 }
 
+std::string Aria2::getNewRpcSecret()
+{
+  std::string uuid = gen_uuid("");
+  uuid.erase(std::remove(uuid.begin(), uuid.end(), '-'));
+  return uuid.substr(0, 9);
+}
+
 std::vector<std::string> Aria2::tellActive()
 {
   MethodCall methodCall("aria2.tellActive", m_secret);

src/aria2.h

@@ -12,7 +12,6 @@
 #include "xmlrpc.h"
 
 #include <memory>
-#include <mutex>
 #include <curl/curl.h>
 
 namespace kiwix {
@@ -23,20 +22,16 @@ class Aria2
     std::unique_ptr<Subprocess> mp_aria;
     int m_port;
     std::string m_secret;
-    std::string m_downloadDir;
-    std::unique_ptr<char[]> m_curlErrorBuffer;
-    CURL* mp_curl;
-    std::mutex m_lock;
-
     std::string doRequest(const MethodCall& methodCall);
 
   public:
-    Aria2();
-    virtual ~Aria2();
+    explicit Aria2(std::string sessionFileDir);
+    virtual ~Aria2() = default;
     void close();
 
     std::string addUri(const std::vector<std::string>& uri, const std::vector<std::pair<std::string, std::string>>& options = {});
     std::string tellStatus(const std::string& gid, const std::vector<std::string>& statusKey);
+    static std::string getNewRpcSecret();
     std::vector<std::string> tellActive();
     std::vector<std::string> tellWaiting();
     void saveSession();

src/book.cpp

@@ -18,13 +18,18 @@
  */
 
 #include "book.h"
-#include "reader.h"
 
+#include "tools.h"
 #include "tools/base64.h"
 #include "tools/regexTools.h"
 #include "tools/networkTools.h"
 #include "tools/otherTools.h"
+#include "tools/stringTools.h"
+#include "tools/pathTools.h"
+#include "tools/archiveTools.h"
 
+#include <zim/archive.h>
+#include <zim/item.h>
 #include <pugixml.hpp>
 
 namespace kiwix
@@ -35,11 +40,17 @@ Book::Book() :
   m_readOnly(false)
 {
 }
+
 /* Destructor */
 Book::~Book()
 {
 }
 
+Book::Illustrations Book::getIllustrations() const
+{
+  return m_illustrations;
+}
+
 bool Book::update(const kiwix::Book& other)
 {
   if (m_readOnly)
@@ -48,55 +59,39 @@ bool Book::update(const kiwix::Book& other)
   if (m_id != other.m_id)
     return false;
 
-  m_readOnly = other.m_readOnly;
-  m_path = other.m_path;
-  m_pathValid = other.m_pathValid;
-  m_title = other.m_title;
-  m_description = other.m_description;
-  m_language = other.m_language;
-  m_creator = other.m_creator;
-  m_publisher = other.m_publisher;
-  m_date = other.m_date;
-  m_url = other.m_url;
-  m_name = other.m_name;
-  m_flavour = other.m_flavour;
-  m_tags = other.m_tags;
-  m_category = other.m_category;
-  m_origId = other.m_origId;
-  m_articleCount = other.m_articleCount;
-  m_mediaCount = other.m_mediaCount;
-  m_size = other.m_size;
-  m_favicon = other.m_favicon;
-  m_faviconMimeType = other.m_faviconMimeType;
-  m_faviconUrl = other.m_faviconUrl;
-
-  m_downloadId = other.m_downloadId;
-
+  *this = other;
   return true;
 }
 
-void Book::update(const kiwix::Reader& reader)
-{
-  m_path = reader.getZimFilePath();
+void Book::update(const zim::Archive& archive) {
+  m_path = archive.getFilename();
   m_pathValid = true;
-  m_id = reader.getId();
-  m_title = reader.getTitle();
-  m_description = reader.getDescription();
-  m_language = reader.getLanguage();
-  m_creator = reader.getCreator();
-  m_publisher = reader.getPublisher();
-  m_date = reader.getDate();
-  m_name = reader.getName();
-  m_flavour = reader.getFlavour();
-  m_tags = reader.getTags();
+  m_id = std::string(archive.getUuid());
+  m_title = getArchiveTitle(archive);
+  m_description = getMetaDescription(archive);
+  m_language = getMetaLanguage(archive);
+  m_creator = getMetaCreator(archive);
+  m_publisher = getMetaPublisher(archive);
+  m_date = getMetaDate(archive);
+  m_name = getMetaName(archive);
+  m_flavour = getMetaFlavour(archive);
+  m_tags = getMetaTags(archive);
   m_category = getCategoryFromTags();
-  m_origId = reader.getOrigId();
-  m_articleCount = reader.getArticleCount();
-  m_mediaCount = reader.getMediaCount();
-  m_size = static_cast<uint64_t>(reader.getFileSize()) << 10;
-  m_pathValid = true;
+  m_articleCount = archive.getArticleCount();
+  m_mediaCount = archive.getMediaCount();
+  m_size = static_cast<uint64_t>(getArchiveFileSize(archive)) << 10;
 
-  reader.getFavicon(m_favicon, m_faviconMimeType);
+  m_illustrations.clear();
+  for ( const auto& illustrationInfo : archive.getIllustrationInfos() ) {
+    const auto illustration = std::make_shared<Illustration>();
+    const zim::Item illustrationItem = archive.getIllustrationItem(illustrationInfo);
+    illustration->width = illustrationInfo.width;
+    illustration->height = illustrationInfo.height;
+    illustration->mimeType = illustrationItem.getMimetype();
+    illustration->data = illustrationItem.getData();
+    // NOTE: illustration->url is left uninitialized
+    m_illustrations.push_back(illustration);
+  }
 }
 
 #define ATTR(name) node.attribute(name).value()
@@ -108,7 +103,7 @@ void Book::updateFromXml(const pugi::xml_node& node, const std::string& baseDir)
     path = computeAbsolutePath(baseDir, path);
   }
   m_path = path;
-  m_pathValid = fileExists(path);
+  m_pathValid = fileReadable(path);
   m_title = ATTR("title");
   m_description = ATTR("description");
   m_language = ATTR("language");
@@ -123,9 +118,15 @@ void Book::updateFromXml(const pugi::xml_node& node, const std::string& baseDir)
   m_articleCount = strtoull(ATTR("articleCount"), 0, 0);
   m_mediaCount = strtoull(ATTR("mediaCount"), 0, 0);
   m_size = strtoull(ATTR("size"), 0, 0) << 10;
-  m_favicon = base64_decode(ATTR("favicon"));
-  m_faviconMimeType = ATTR("faviconMimeType");
-  m_faviconUrl = ATTR("faviconUrl");
+  const std::string faviconMimeType = ATTR("faviconMimeType");
+  const std::string faviconBase64EncodedData = ATTR("favicon");
+  if ( !faviconMimeType.empty() && !faviconBase64EncodedData.empty() ) {
+    const auto favicon = std::make_shared<Illustration>();
+    favicon->data = base64_decode(faviconBase64EncodedData);
+    favicon->mimeType = faviconMimeType;
+    favicon->url = ATTR("faviconUrl");
+    m_illustrations.assign(1, favicon);
+  }
   try {
     m_downloadId = ATTR("downloadId");
   } catch(...) {}
@@ -156,7 +157,9 @@ void Book::updateFromOpds(const pugi::xml_node& node, const std::string& urlHost
   m_language = VALUE("language");
   m_creator = node.child("author").child("name").child_value();
   m_publisher = node.child("publisher").child("name").child_value();
-  m_date = fromOpdsDate(VALUE("updated"));
+  const std::string dcIssuedDate = VALUE("dc:issued");
+  m_date = dcIssuedDate.empty() ? VALUE("updated") : dcIssuedDate;
+  m_date = fromOpdsDate(m_date);
   m_name = VALUE("name");
   m_flavour = VALUE("flavour");
   m_tags = VALUE("tags");
@@ -173,8 +176,11 @@ void Book::updateFromOpds(const pugi::xml_node& node, const std::string& urlHost
       m_size = strtoull(linkNode.attribute("length").value(), 0, 0);
     }
     if (rel == "http://opds-spec.org/image/thumbnail") {
-      m_faviconUrl = urlHost + linkNode.attribute("href").value();
-      m_faviconMimeType = linkNode.attribute("type").value();
+      const auto favicon = std::make_shared<Illustration>();
+      favicon->data.clear();
+      favicon->url = urlHost + linkNode.attribute("href").value();
+      favicon->mimeType = linkNode.attribute("type").value();
+      m_illustrations.assign(1, favicon);
     }
  }
 
@@ -185,7 +191,7 @@ std::string Book::getHumanReadableIdFromPath() const
 {
   std::string id = m_path;
   if (!id.empty()) {
-    kiwix::removeAccents(id);
+    id = kiwix::removeAccents(id);
 
 #ifdef _WIN32
     id = replaceRegex(id, "", "^.*\\\\");
@@ -207,15 +213,54 @@ void Book::setPath(const std::string& path)
    : path;
 }
 
-const std::string& Book::getFavicon() const {
-  if (m_favicon.empty() && !m_faviconUrl.empty()) {
-    try {
-      m_favicon = download(m_faviconUrl);
-    } catch(...) {
-      std::cerr << "Cannot download favicon from " << m_faviconUrl;
+const Book::Illustration Book::missingDefaultIllustration;
+
+std::shared_ptr<const Book::Illustration> Book::getIllustration(unsigned int size) const
+{
+  for ( const auto& ilPtr : m_illustrations ) {
+    if (ilPtr->width == size && ilPtr->height == size) {
+      return ilPtr;
     }
   }
-  return m_favicon;
+  throw std::runtime_error("Cannot find illustration");
+}
+
+const Book::Illustration& Book::getDefaultIllustration() const
+{
+  try {
+    return *getIllustration(48);
+  } catch (...) {
+    return missingDefaultIllustration;
+  }
+}
+
+const std::string& Book::Illustration::getData() const
+{
+  if (data.empty() && !url.empty()) {
+    const std::lock_guard<std::mutex> l(mutex);
+    if ( data.empty() ) {
+      try {
+        data = download(url);
+      } catch(...) {
+        std::cerr << "Cannot download favicon from " << url;
+      }
+    }
+  }
+  return data;
+}
+
+const std::string& Book::getFavicon() const {
+  return getDefaultIllustration().getData();
+}
+
+const std::string& Book::getFaviconUrl() const
+{
+  return getDefaultIllustration().url;
+}
+
+const std::string& Book::getFaviconMimeType() const
+{
+  return getDefaultIllustration().mimeType;
 }
 
 std::string Book::getTagStr(const std::string& tagName) const {
@@ -243,4 +288,9 @@ std::string Book::getCategoryFromTags() const
   }
 }
 
+const std::vector<std::string> Book::getLanguages() const
+{
+  return kiwix::split(m_language, ",");
+}
+
 }

src/bookmark.cpp

@@ -18,6 +18,7 @@
  */
 
 #include "bookmark.h"
+#include "book.h"
 
 #include <pugixml.hpp>
 
@@ -28,6 +29,17 @@ Bookmark::Bookmark()
 {
 }
 
+Bookmark::Bookmark(const Book& book, const std::string& path, const std::string& title):
+  m_bookId(book.getId()),
+  m_bookTitle(book.getTitle()),
+  m_bookName(book.getName()),
+  m_bookFlavour(book.getFlavour()),
+  m_url(path),
+  m_title(title),
+  m_language(book.getCommaSeparatedLanguages()),
+  m_date(book.getDate())
+{}
+
 /* Destructor */
 Bookmark::~Bookmark()
 {
@@ -38,6 +50,8 @@ void Bookmark::updateFromXml(const pugi::xml_node& node)
   auto bookNode = node.child("book");
   m_bookId = bookNode.child("id").child_value();
   m_bookTitle = bookNode.child("title").child_value();
+  m_bookName = bookNode.child("name").child_value();
+  m_bookFlavour = bookNode.child("flavour").child_value();
   m_language = bookNode.child("language").child_value();
   m_date = bookNode.child("date").child_value();
   m_title = node.child("title").child_value();

src/config.h.in

@@ -1,3 +1,3 @@
 
-#mesondefine VERSION
+#mesondefine LIBKIWIX_VERSION
 

src/downloader.cpp

@@ -18,6 +18,7 @@
  */
 
 #include "downloader.h"
+#include "tools.h"
 #include "tools/pathTools.h"
 #include "tools/stringTools.h"
 
@@ -124,38 +125,50 @@ void Download::cancelDownload()
 }
 
 /* Constructor */
-Downloader::Downloader() :
-  mp_aria(new Aria2())
+Downloader::Downloader(std::string sessionFileDir) :
+  mp_aria(new Aria2(sessionFileDir))
 {
-  try {
-    for (auto gid : mp_aria->tellActive()) {
-      m_knownDownloads[gid] = std::unique_ptr<Download>(new Download(mp_aria, gid));
-      m_knownDownloads[gid]->updateStatus();
-    }
-  } catch (std::exception& e) {
-    std::cerr << "aria2 tellActive failed : " << e.what() << std::endl;
-  }
   try {
     for (auto gid : mp_aria->tellWaiting()) {
       m_knownDownloads[gid] = std::unique_ptr<Download>(new Download(mp_aria, gid));
-      m_knownDownloads[gid]->updateStatus();
+      m_knownDownloads[gid]->updateStatus(false);
     }
   } catch (std::exception& e) {
     std::cerr << "aria2 tellWaiting failed : " << e.what() << std::endl;
   }
+  try {
+    for (auto gid : mp_aria->tellActive()) {
+      if( m_knownDownloads.find(gid) == m_knownDownloads.end()) {
+        m_knownDownloads[gid] = std::unique_ptr<Download>(new Download(mp_aria, gid));
+        m_knownDownloads[gid]->updateStatus(false);
+      }
+    }
+  } catch (std::exception& e) {
+    std::cerr << "aria2 tellActive failed : " << e.what() << std::endl;
+  }
 }
 
 /* Destructor */
 Downloader::~Downloader()
 {
+  close();
 }
 
 void Downloader::close()
 {
-  mp_aria->close();
+  if ( mp_aria ) {
+    try {
+      mp_aria->close();
+    } catch (const std::exception& err) {
+      std::cerr << "ERROR: Failed to save the downloader state: "
+                << err.what() << std::endl;
+    }
+    mp_aria.reset();
+  }
 }
 
-std::vector<std::string> Downloader::getDownloadIds() {
+std::vector<std::string> Downloader::getDownloadIds() const {
+  std::unique_lock<std::mutex> lock(m_lock);
   std::vector<std::string> ret;
   for(auto& p:m_knownDownloads) {
     ret.push_back(p.first);
@@ -163,42 +176,82 @@ std::vector<std::string> Downloader::getDownloadIds() {
   return ret;
 }
 
-Download* Downloader::startDownload(const std::string& uri, const std::vector<std::pair<std::string, std::string>>& options)
+namespace
 {
+
+bool downloadCanBeReused(const Download& d,
+                         const std::string& uri,
+                         const Downloader::Options& /*options*/)
+{
+  const auto& uris = d.getUris();
+  const bool sameURI = std::find(uris.begin(), uris.end(), uri) != uris.end();
+
+  if ( !sameURI )
+    return false;
+
+  switch ( d.getStatus() ) {
+  case Download::K_ERROR:
+  case Download::K_UNKNOWN:
+  case Download::K_REMOVED:
+    return false;
+
+  case Download::K_ACTIVE:
+  case Download::K_WAITING:
+  case Download::K_PAUSED:
+    return true; // XXX: what if options are different?
+
+  case Download::K_COMPLETE:
+    return fileExists(d.getPath()); // XXX: what if options are different?
+  }
+
+  return false;
+}
+
+} // unnamed namespace
+
+std::shared_ptr<Download> Downloader::startDownload(const std::string& uri, const std::string& downloadDir, Options options)
+{
+  std::unique_lock<std::mutex> lock(m_lock);
+  options.erase(std::remove_if(options.begin(), options.end(), [](const auto& option) {
+    return option.first == "dir";
+  }), options.end());
+  options.push_back({"dir", downloadDir});
   for (auto& p: m_knownDownloads) {
     auto& d = p.second;
-    auto& uris = d->getUris();
-    if (std::find(uris.begin(), uris.end(), uri) != uris.end())
-      return d.get();
+    if ( downloadCanBeReused(*d, uri, options) )
+      return d;
   }
   std::vector<std::string> uris = {uri};
   auto gid = mp_aria->addUri(uris, options);
-  m_knownDownloads[gid] = std::unique_ptr<Download>(new Download(mp_aria, gid));
-  return m_knownDownloads[gid].get();
+  m_knownDownloads[gid] = std::make_shared<Download>(mp_aria, gid);
+  return m_knownDownloads[gid];
 }
 
-Download* Downloader::getDownload(const std::string& did)
+std::shared_ptr<Download> Downloader::getDownload(const std::string& did)
 {
+  std::unique_lock<std::mutex> lock(m_lock);
   try {
-    m_knownDownloads.at(did).get()->updateStatus(true);
-    return m_knownDownloads.at(did).get();
+    return m_knownDownloads.at(did);
   } catch(std::exception& e) {
-    for (auto gid : mp_aria->tellActive()) {
-      if (gid == did) {
-        m_knownDownloads[gid] = std::unique_ptr<Download>(new Download(mp_aria, gid));
-        m_knownDownloads.at(gid).get()->updateStatus(true);
-        return m_knownDownloads[gid].get();
-      }
-    }
     for (auto gid : mp_aria->tellWaiting()) {
       if (gid == did) {
-        m_knownDownloads[gid] = std::unique_ptr<Download>(new Download(mp_aria, gid));
-        m_knownDownloads.at(gid).get()->updateStatus(true);
-        return m_knownDownloads[gid].get();
+        m_knownDownloads[gid] = std::make_shared<Download>(mp_aria, gid);
+        return m_knownDownloads[gid];
       }
-  }
+    }
+    for (auto gid : mp_aria->tellActive()) {
+      if (gid == did) {
+        m_knownDownloads[gid] = std::make_shared<Download>(mp_aria, gid);
+        return m_knownDownloads[gid];
+      }
+    }
     throw e;
   }
 }
 
+size_t Downloader::getNbDownload() const {
+  std::unique_lock<std::mutex> lock(m_lock);
+  return m_knownDownloads.size();
+}
+
 }

src/entry.cpp

@@ -1,73 +0,0 @@
-/*
- * Copyright 2018-2020 Matthieu Gautier <mgautier@kymeria.fr>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU  General Public License as published by
- * the Free Software Foundation; either version 3 of the License, or
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
- * MA 02110-1301, USA.
- */
-
-#include "reader.h"
-#include <time.h>
-
-namespace kiwix
-{
-
-Entry::Entry(zim::Entry entry)
-  : entry(entry)
-{
-}
-
-size_type Entry::getSize() const
-{
-  if (entry.isRedirect()) {
-    return 0;
-  } else {
-    return entry.getItem().getSize();
-  }
-}
-
-std::string Entry::getMimetype() const
-{
-  return entry.getItem(true).getMimetype();
-}
-
-bool Entry::isRedirect() const
-{
-  return entry.isRedirect();
-}
-
-Entry Entry::getRedirectEntry() const
-{
-  if ( !entry.isRedirect() ) {
-    throw NoEntry();
-  }
-
-  return entry.getRedirectEntry();
-}
-
-Entry Entry::getFinalEntry() const
-{
-  int loopCounter = 42;
-  auto final_entry = entry;
-  while (final_entry.isRedirect() && loopCounter--) {
-    final_entry = final_entry.getRedirectEntry();
-  }
-  // Prevent infinite loops.
-  if (final_entry.isRedirect()) {
-    throw NoEntry();
-  }
-  return final_entry;
-}
-
-}

src/html_dumper.cpp

@@ -0,0 +1,144 @@
+#include "html_dumper.h"
+#include "libkiwix-resources.h"
+#include "tools/otherTools.h"
+#include "tools.h"
+#include "tools/regexTools.h"
+#include "server/i18n_utils.h"
+
+namespace kiwix
+{
+
+/* Constructor */
+HTMLDumper::HTMLDumper(const Library* library, const NameMapper* nameMapper)
+  : LibraryDumper(library, nameMapper)
+{
+}
+/* Destructor */
+HTMLDumper::~HTMLDumper()
+{
+}
+
+namespace {
+
+std::string humanFriendlyTitle(std::string title)
+{
+  std::string humanFriendlyString = replaceRegex(title, "_", " ");
+  humanFriendlyString[0] = toupper(humanFriendlyString[0]);
+  return humanFriendlyString;
+}
+
+kainjow::mustache::object getLangTag(const std::vector<std::string>& bookLanguages) {
+  std::string langShortString = "";
+  std::string langFullString = "???";
+
+  //if more than 1 languages then show "mul" else show the language
+  if(bookLanguages.size() > 1) {
+    std::vector<std::string> mulLanguages;
+    langShortString = "mul";
+    for (const auto& lang : bookLanguages) {
+      const std::string fullLang = getLanguageSelfName(lang);
+      mulLanguages.push_back(fullLang);
+    }
+    langFullString = kiwix::join(mulLanguages, ",");
+  } else if(bookLanguages.size() == 1) {
+    langShortString = bookLanguages[0];
+    langFullString = getLanguageSelfName(langShortString);
+  }
+
+  kainjow::mustache::object langTag;
+    langTag["langShortString"] = langShortString;
+    langTag["langFullString"] = langFullString;
+  return langTag;
+}
+
+kainjow::mustache::list getTagList(std::string tags)
+{
+  const auto tagsList = kiwix::split(tags, ";", true, false);
+  kainjow::mustache::list finalTagList;
+  for (auto tag : tagsList) {
+  if (tag[0] != '_')
+    finalTagList.push_back(kainjow::mustache::object{
+      {"tag", tag}
+    });
+  }
+  return finalTagList;
+}
+
+} // unnamed namespace
+
+std::string HTMLDumper::dumpPlainHTML(kiwix::Filter filter) const
+{
+  kainjow::mustache::list booksData;
+  const auto filteredBooks = library->filter(filter);
+  const auto searchQuery = filter.getQuery();
+  auto languages = getLanguageData();
+  auto categories = getCategoryData();
+
+  for (auto &category : categories) {
+    const auto categoryName = category.get("name")->string_value();
+    if (categoryName == filter.getCategory()) {
+      category["selected"] = true;
+    }
+    category["hf_name"] = humanFriendlyTitle(categoryName);
+  }
+
+  for (auto &language : languages) {
+    if (language.get("lang_code")->string_value() == filter.getLang()) {
+      language["selected"] = true;
+    }
+  }
+
+  for ( const auto& bookId : filteredBooks ) {
+    const auto bookObj = library->getBookById(bookId);
+    const auto bookTitle = bookObj.getTitle();
+    std::string contentId = "";
+    try {
+      contentId = urlEncode(nameMapper->getNameForId(bookId));
+    } catch (...) {}
+    const auto bookDescription = bookObj.getDescription();
+    const auto bookIconUrl = rootLocation + "/catalog/v2/illustration/" + bookId +  "/?size=48";
+    const auto tags = bookObj.getTags();
+    const auto downloadAvailable = (bookObj.getUrl() != "");
+    const auto langTagObj = getLangTag(bookObj.getLanguages());
+    std::string faviconAttr = "style=background-image:url(" + bookIconUrl + ")";
+    booksData.push_back(kainjow::mustache::object{
+      {"id", contentId},
+      {"title", bookTitle},
+      {"description", bookDescription},
+      {"langTag", langTagObj},
+      {"faviconAttr", faviconAttr},
+      {"tagList", getTagList(tags)},
+      {"downloadAvailable", downloadAvailable}
+    });
+  }
+
+  auto getTranslation = i18n::GetTranslatedStringWithMsgId(m_userLang);
+
+  const auto translations = kainjow::mustache::object{
+                    getTranslation("search"),
+                    getTranslation("download"),
+                    getTranslation("count-of-matching-books", {{"COUNT", to_string(filteredBooks.size())}}),
+                    getTranslation("book-filtering-all-categories"),
+                    getTranslation("book-filtering-all-languages"),
+                    getTranslation("powered-by-kiwix-html"),
+                    getTranslation("welcome-to-kiwix-server"),
+                    getTranslation("preview-book"),
+                    getTranslation("welcome-page-overzealous-filter", {{"URL", "?lang="}})
+  };
+
+  return render_template(
+             RESOURCE::templates::no_js_library_page_html,
+             kainjow::mustache::object{
+               {"root", rootLocation},
+               {"contentAccessUrl", onlyAsNonEmptyMustacheValue(contentAccessUrl)},
+               {"books", booksData },
+               {"searchQuery", searchQuery},
+               {"languages", languages},
+               {"categories", categories},
+               {"noResults", filteredBooks.size() == 0},
+               {"translations", translations}
+             }
+  );
+}
+
+} // namespace kiwix

src/html_dumper.h

@@ -1,6 +1,5 @@
 /*
- * Copyright (C) 2013 Emmanuel Engelhart <kelson@kiwix.org>
- * Copyright (C) 2017 Matthieu Gautier <mgautier@kymeria.fr>
+ * Copyright 2023 Nikhil Tanwar <2002nikhiltanwar@gmail.com>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU  General Public License as published by
@@ -18,19 +17,34 @@
  * MA 02110-1301, USA.
  */
 
-package org.kiwix.kiwixlib;
+#ifndef KIWIX_HTML_DUMPER_H
+#define KIWIX_HTML_DUMPER_H
 
-import android.content.Context;
-import com.getkeepsafe.relinker.ReLinker;
-import org.kiwix.kiwixlib.JNIICU;
+#include <string>
 
-public class JNIKiwix
+#include "library_dumper.h"
+
+namespace kiwix
 {
-  public JNIKiwix(final Context context){
-    ReLinker.loadLibrary(context, "kiwix");
-  }
 
-  public void setDataDirectory(String icuDataDir) {
-    JNIICU.setDataDirectory(icuDataDir);
-  }
+/**
+ * A class to dump Library in HTML format.
+ */
+class HTMLDumper : public LibraryDumper
+{
+ public:
+  HTMLDumper(const Library* library, const NameMapper* NameMapper);
+  ~HTMLDumper();
+
+
+  /**
+   * Dump library in HTML
+   *
+   * @return HTML content
+   */
+  std::string dumpPlainHTML(kiwix::Filter filter) const;
+};
+
 }
+
+#endif // KIWIX_HTML_DUMPER_H

src/kiwixserve.cpp

@@ -9,6 +9,7 @@
     # include <unistd.h>
 #endif
 
+#include "tools.h"
 #include "tools/pathTools.h"
 #include "tools/stringTools.h"
 

src/library.cpp

@@ -19,23 +19,28 @@
 
 #include "library.h"
 #include "book.h"
-#include "reader.h"
 #include "libxml_dumper.h"
 
+#include "tools.h"
 #include "tools/base64.h"
 #include "tools/regexTools.h"
 #include "tools/pathTools.h"
 #include "tools/stringTools.h"
+#include "tools/otherTools.h"
+#include "tools/concurrent_cache.h"
 
 #include <pugixml.hpp>
 #include <algorithm>
 #include <set>
+#include <cmath>
 #include <unicode/locid.h>
 #include <xapian.h>
 
 namespace kiwix
 {
 
+
+
 namespace
 {
 
@@ -48,51 +53,101 @@ std::string normalizeText(const std::string& text)
   return removeAccents(text);
 }
 
+bool booksReferToTheSameArchive(const Book& book1, const Book& book2)
+{
+  return book1.isPathValid()
+      && book2.isPathValid()
+      && book1.getPath() == book2.getPath();
+}
+
 } // unnamed namespace
 
-class Library::BookDB : public Xapian::WritableDatabase
+template<typename Key, typename Value>
+class MultiKeyCache: public ConcurrentCache<std::set<Key>, Value>
 {
-public:
-  BookDB() : Xapian::WritableDatabase("", Xapian::DB_BACKEND_INMEMORY) {}
+  public:
+    explicit MultiKeyCache(size_t maxEntries)
+      : ConcurrentCache<std::set<Key>, Value>(maxEntries)
+    {}
+
+    bool drop(const Key& key)
+    {
+      std::unique_lock<std::mutex> l(this->lock_);
+      bool removed = false;
+      for(auto& cache_key: this->impl_.keys()) {
+        if(cache_key.find(key)!=cache_key.end()) {
+          removed |= this->impl_.drop(cache_key);
+        }
+      }
+      return removed;
+    }
 };
 
+unsigned int
+Library::getBookCount_not_protected(const bool localBooks, const bool remoteBooks) const
+{
+  unsigned int result = 0;
+  for (auto& pair: m_books) {
+    auto& book = pair.second;
+    if ((!book.getPath().empty() && localBooks)
+        || (!book.getUrl().empty() && remoteBooks)) {
+      result++;
+    }
+  }
+  return result;
+}
+
 /* Constructor */
 Library::Library()
-  : m_bookDB(new BookDB)
+  : mp_archiveCache(new ArchiveCache(std::max(getEnvVar<int>("KIWIX_ARCHIVE_CACHE_SIZE", 1), 1))),
+    mp_searcherCache(new SearcherCache(std::max(getEnvVar<int>("KIWIX_SEARCHER_CACHE_SIZE", 1), 1))),
+    m_bookDB(new Xapian::WritableDatabase("", Xapian::DB_BACKEND_INMEMORY))
 {
 }
 
-Library::Library(Library&& ) = default;
-
-Library& Library::operator=(Library&& ) = default;
-
 /* Destructor */
-Library::~Library()
-{
-}
-
+Library::~Library() = default;
 
 bool Library::addBook(const Book& book)
 {
+  std::lock_guard<std::recursive_mutex> lock(m_mutex);
+  ++m_revision;
   /* Try to find it */
   updateBookDB(book);
   try {
     auto& oldbook = m_books.at(book.getId());
-    oldbook.update(book);
+    if ( ! booksReferToTheSameArchive(oldbook, book) ) {
+      dropCache(book.getId());
+    }
+    oldbook.update(book); // XXX: This may have no effect if oldbook is readonly
+                          // XXX: Then m_bookDB will become out-of-sync with
+                          // XXX: the real contents of the library.
+    oldbook.lastUpdatedRevision = m_revision;
     return false;
   } catch (std::out_of_range&) {
-    m_books[book.getId()] = book;
+    auto& newEntry = m_books[book.getId()];
+    static_cast<Book&>(newEntry) = book;
+    newEntry.lastUpdatedRevision = m_revision;
+    size_t new_cache_size = static_cast<size_t>(std::ceil(getBookCount_not_protected(true, true)*0.1));
+    if (getEnvVar<int>("KIWIX_ARCHIVE_CACHE_SIZE", -1) <= 0) {
+      mp_archiveCache->setMaxSize(new_cache_size);
+    }
+    if (getEnvVar<int>("KIWIX_SEARCHER_CACHE_SIZE", -1) <= 0) {
+      mp_searcherCache->setMaxSize(new_cache_size);
+    }
     return true;
   }
 }
 
 void Library::addBookmark(const Bookmark& bookmark)
 {
+  std::lock_guard<std::recursive_mutex> lock(m_mutex);
   m_bookmarks.push_back(bookmark);
 }
 
 bool Library::removeBookmark(const std::string& zimId, const std::string& url)
 {
+  std::lock_guard<std::recursive_mutex> lock(m_mutex);
   for(auto it=m_bookmarks.begin(); it!=m_bookmarks.end(); it++) {
     if (it->getBookId() == zimId && it->getUrl() == url) {
       m_bookmarks.erase(it);
@@ -102,28 +157,227 @@ bool Library::removeBookmark(const std::string& zimId, const std::string& url)
   return false;
 }
 
+std::tuple<int, int> Library::migrateBookmarks(MigrationMode migrationMode) {
+  std::set<std::string> sourceBooks;
+  int invalidBookmarks = 0;
+  {
+    std::lock_guard<std::recursive_mutex> lock(m_mutex);
+    for(auto& bookmark:m_bookmarks) {
+      if (m_books.find(bookmark.getBookId()) == m_books.end()) {
+        invalidBookmarks += 1;
+        sourceBooks.insert(bookmark.getBookId());
+      }
+    }
+  }
+  int changed = 0;
+  for(auto& sourceBook:sourceBooks) {
+    changed += migrateBookmarks(sourceBook, migrationMode);
+  }
+  return std::make_tuple(changed, invalidBookmarks);
+}
 
+std::string Library::getBestFromBookCollection(BookIdCollection books, const Bookmark& bookmark, MigrationMode migrationMode) const {
+  // This function try to get the best book for a bookmark from a book collection.
+  // It assumes that all books in the collection are "acceptable".
+  // (this definiton is not clear but for now it is book's name is equal to bookmark's bookName)
+  //
+  // The algorithm first sort the colletion by "flavour equality" and date.
+  // "flavour equality" is if book's flavour is same that bookmark's flavour (let's say "flavourA" here)
+  // So we have the sorted collection:
+  // - flavourA, date 5
+  // - flavourA, date 4
+  // - flavourB, date 6
+  // - flavourC, date 5
+  // - flavourB, date 3
+  //
+  // Then, depending of migrationMode:
+  // - If ALLOW_DOWNGRADE => take the first one
+  // - If UPGRADE_ONLY => loop on books until we find a book newer than bookmark.
+  //   So if bookmark date is 5 => flavourB, date 6
+  //      if bookmark date is 4 => flavourA, date 5
+  //      if bookmark date is 7 => No book
+
+  if (books.empty()) {
+    return "";
+  }
+
+  sort(books, DATE, false);
+  stable_sort(books.begin(), books.end(), [&](const std::string& bookId1, const std::string& bookId2) {
+    const auto& book1 = getBookById(bookId1);
+    const auto& book2 = getBookById(bookId2);
+    bool same_flavour1 = book1.getFlavour() == bookmark.getBookFlavour();
+    bool same_flavour2 = book2.getFlavour() == bookmark.getBookFlavour();
+    // return True if bookId1 is before bookId2, ie if same_flavour1 and not same_flavour2
+    return same_flavour1 > same_flavour2;
+  });
+
+  if (migrationMode == ALLOW_DOWNGRADE) {
+    return books[0];
+  } else {
+    for (const auto& bookId: books) {
+      const auto& book = getBookById(bookId);
+      if (book.getDate() >= bookmark.getDate()) {
+        return bookId;
+      }
+    }
+  }
+
+  return "";
+}
+
+std::string remove_quote(std::string input) {
+  std::replace(input.begin(), input.end(), '"', ' ');
+  return input;
+}
+
+std::string Library::getBestTargetBookId(const std::string& bookName, const std::string& preferedFlavour, const std::string& minDate) const {
+  // Let's reuse our algorithm based on bookmark.
+  MigrationMode migrationMode = UPGRADE_ONLY;
+  auto bookmark = Bookmark();
+  bookmark.setBookName(bookName);
+  bookmark.setBookFlavour(preferedFlavour);
+
+  if (minDate.empty()) {
+    migrationMode = ALLOW_DOWNGRADE;
+  } else {
+    bookmark.setDate(minDate);
+  }
+
+  return getBestTargetBookId(bookmark, migrationMode);
+}
+
+std::string Library::getBestTargetBookId(const Bookmark& bookmark, MigrationMode migrationMode) const {
+  std::lock_guard<std::recursive_mutex> lock(m_mutex);
+  // Search for a existing book with the same name
+  auto book_filter = Filter();
+  if (!bookmark.getBookName().empty()) {
+    book_filter.name(bookmark.getBookName());
+  } else {
+    // We don't have a name stored (older bookmarks)
+    // Fallback on title (All bookmarks should have one, but let's be safe against wrongly filled bookmark)
+    if (bookmark.getBookTitle().empty()) {
+        // No bookName nor bookTitle, no way to find target book.
+        return "";
+    }
+    book_filter.query("title:\"" + remove_quote(bookmark.getBookTitle()) + "\"");
+  }
+  auto targetBooks = filter(book_filter);
+  auto bestBook = getBestFromBookCollection(targetBooks, bookmark, migrationMode);
+  if (bestBook.empty()) {
+    try {
+      getBookById(bookmark.getBookId());
+      return bookmark.getBookId();
+    } catch (std::out_of_range&) {}
+  }
+  return bestBook;
+}
+
+int Library::migrateBookmarks(const std::string& sourceBookId, MigrationMode migrationMode) {
+  std::lock_guard<std::recursive_mutex> lock(m_mutex);
+
+  Bookmark firstBookmarkToChange;
+  for(auto& bookmark:m_bookmarks) {
+    if (bookmark.getBookId() == sourceBookId) {
+      firstBookmarkToChange = bookmark;
+      break;
+    }
+  }
+
+  if (firstBookmarkToChange.getBookId().empty()) {
+    return 0;
+  }
+
+  std::string betterBook = getBestTargetBookId(firstBookmarkToChange, migrationMode);
+
+  if (betterBook.empty()) {
+    return 0;
+  }
+
+  return migrateBookmarks(sourceBookId, betterBook);
+}
+
+int Library::migrateBookmarks(const std::string& sourceBookId, const std::string& targetBookId) {
+  if (sourceBookId == targetBookId) {
+    return 0;
+  }
+  int changed = 0;
+  for (auto& bookmark:m_bookmarks) {
+    if (bookmark.getBookId() == sourceBookId) {
+      bookmark.setBookId(targetBookId);
+      changed +=1;
+    }
+  }
+  return changed;
+}
+
+
+void Library::dropCache(const std::string& id)
+{
+  mp_archiveCache->drop(id);
+  mp_searcherCache->drop(id);
+}
 
 bool Library::removeBookById(const std::string& id)
 {
+  std::lock_guard<std::recursive_mutex> lock(m_mutex);
   m_bookDB->delete_document("Q" + id);
-  m_readers.erase(id);
-  return m_books.erase(id) == 1;
+  dropCache(id);
+  // We do not change the cache size here
+  // Most of the time, the book is remove in case of library refresh, it is
+  // often associated with addBook calls (which will properly set the cache size)
+  // Having a too big cache is not a problem here (or it would have been before)
+  // (And setMaxSize doesn't actually reduce the cache size, extra cached items
+  //  will be removed in put or getOrPut).
+  const bool bookWasRemoved = m_books.erase(id) == 1;
+  if ( bookWasRemoved ) {
+    ++m_revision;
+  }
+  return bookWasRemoved;
+}
+
+Library::Revision Library::getRevision() const
+{
+  std::lock_guard<std::recursive_mutex> lock(m_mutex);
+  return m_revision;
+}
+
+uint32_t Library::removeBooksNotUpdatedSince(Revision libraryRevision)
+{
+  BookIdCollection booksToRemove;
+  {
+    std::lock_guard<std::recursive_mutex> lock(m_mutex);
+    for ( const auto& entry : m_books) {
+      if ( entry.second.lastUpdatedRevision <= libraryRevision ) {
+        booksToRemove.push_back(entry.first);
+      }
+    }
+  }
+
+  uint32_t countOfRemovedBooks = 0;
+  for ( const auto& id : booksToRemove ) {
+    if ( removeBookById(id) )
+      ++countOfRemovedBooks;
+  }
+  return countOfRemovedBooks;
 }
 
 const Book& Library::getBookById(const std::string& id) const
 {
+  // XXX: Doesn't make sense to lock this operation since it cannot
+  // XXX: guarantee thread-safety because of its return type
   return m_books.at(id);
 }
 
-Book& Library::getBookById(const std::string& id)
+Book Library::getBookByIdThreadSafe(const std::string& id) const
 {
-  const Library& const_self = *this;
-  return const_cast<Book&>(const_self.getBookById(id));
+  std::lock_guard<std::recursive_mutex> lock(m_mutex);
+  return getBookById(id);
 }
 
 const Book& Library::getBookByPath(const std::string& path) const
 {
+  // XXX: Doesn't make sense to lock this operation since it cannot
+  // XXX: guarantee thread-safety because of its return type
   for(auto& it: m_books) {
     auto& book = it.second;
     if (book.getPath() == path)
@@ -134,75 +388,124 @@ const Book& Library::getBookByPath(const std::string& path) const
   throw std::out_of_range(ss.str());
 }
 
-Book& Library::getBookByPath(const std::string& path)
-{
-  const Library& const_self = *this;
-  return const_cast<Book&>(const_self.getBookByPath(path));
-}
-
-std::shared_ptr<Reader> Library::getReaderById(const std::string& id)
+std::shared_ptr<zim::Archive> Library::getArchiveById(const std::string& id)
 {
   try {
-    return m_readers.at(id);
-  } catch (std::out_of_range& e) {}
-
-  auto book = getBookById(id);
-  if (!book.isPathValid())
+    return mp_archiveCache->getOrPut(id,
+    [&](){
+      auto book = getBookById(id);
+      if (!book.isPathValid()) {
+        throw std::invalid_argument("");
+      }
+      return std::make_shared<zim::Archive>(book.getPath());
+    });
+  } catch (std::invalid_argument&) {
     return nullptr;
-  auto sptr = make_shared<Reader>(book.getPath());
-  m_readers[id] = sptr;
-  return sptr;
+  }
+}
+
+std::shared_ptr<ZimSearcher> Library::getSearcherByIds(const BookIdSet& ids)
+{
+  assert(!ids.empty());
+  try {
+    return mp_searcherCache->getOrPut(ids,
+    [&](){
+      std::vector<zim::Archive> archives;
+      for(auto& id:ids) {
+        auto archive = getArchiveById(id);
+        if(!archive) {
+          throw std::invalid_argument("");
+        }
+        archives.push_back(*archive);
+      }
+      return std::make_shared<ZimSearcher>(zim::Searcher(archives));
+    });
+  } catch (std::invalid_argument&) {
+    return nullptr;
+  }
 }
 
 unsigned int Library::getBookCount(const bool localBooks,
                                    const bool remoteBooks) const
 {
-  unsigned int result = 0;
-  for (auto& pair: m_books) {
-    auto& book = pair.second;
-    if ((!book.getPath().empty() && localBooks)
-        || (book.getPath().empty() && remoteBooks)) {
-      result++;
-    }
-  }
-  return result;
+  std::lock_guard<std::recursive_mutex> lock(m_mutex);
+  return getBookCount_not_protected(localBooks, remoteBooks);
 }
 
 bool Library::writeToFile(const std::string& path) const
 {
+  const auto allBookIds = getBooksIds();
+
   auto baseDir = removeLastPathElement(path);
   LibXMLDumper dumper(this);
   dumper.setBaseDir(baseDir);
-  return writeTextFile(path, dumper.dumpLibXMLContent(getBooksIds()));
+  std::string xml;
+  {
+    std::lock_guard<std::recursive_mutex> lock(m_mutex);
+    xml = dumper.dumpLibXMLContent(allBookIds);
+  };
+  return writeTextFile(path, xml);
 }
 
 bool Library::writeBookmarksToFile(const std::string& path) const
 {
   LibXMLDumper dumper(this);
-  return writeTextFile(path, dumper.dumpLibXMLBookmark());
+  // NOTE: LibXMLDumper::dumpLibXMLBookmark uses Library in a thread-safe way
+  const std::string xml = dumper.dumpLibXMLBookmark();
+  return writeTextFile(path, xml);
+}
+
+Library::AttributeCounts Library::getBookAttributeCounts(BookStrPropMemFn p) const
+{
+  std::lock_guard<std::recursive_mutex> lock(m_mutex);
+  AttributeCounts propValueCounts;
+
+  for (const auto& pair: m_books) {
+    const auto& book = pair.second;
+    if (book.getOrigId().empty()) {
+      propValueCounts[(book.*p)()] += 1;
+    }
+  }
+  return propValueCounts;
+}
+
+std::vector<std::string> Library::getBookPropValueSet(BookStrPropMemFn p) const
+{
+  std::vector<std::string> result;
+  for ( const auto& kv : getBookAttributeCounts(p) ) {
+    result.push_back(kv.first);
+  }
+  return result;
 }
 
 std::vector<std::string> Library::getBooksLanguages() const
 {
-  std::vector<std::string> booksLanguages;
-  std::map<std::string, bool> booksLanguagesMap;
+  std::vector<std::string> langs;
+  for ( const auto& langAndCount : getBooksLanguagesWithCounts() ) {
+    langs.push_back(langAndCount.first);
+  }
+  return langs;
+}
 
-  for (auto& pair: m_books) {
-    auto& book = pair.second;
-    auto& language = book.getLanguage();
-    if (booksLanguagesMap.find(language) == booksLanguagesMap.end()) {
-      if (book.getOrigId().empty()) {
-        booksLanguagesMap[language] = true;
-        booksLanguages.push_back(language);
+Library::AttributeCounts Library::getBooksLanguagesWithCounts() const
+{
+  std::lock_guard<std::recursive_mutex> lock(m_mutex);
+  AttributeCounts langsWithCounts;
+
+  for (const auto& pair: m_books) {
+    const auto& book = pair.second;
+    if (book.getOrigId().empty()) {
+      for ( const auto& lang : book.getLanguages() ) {
+        ++langsWithCounts[lang];
       }
     }
   }
-
-  return booksLanguages;
+  return langsWithCounts;
 }
 
 std::vector<std::string> Library::getBooksCategories() const
 {
+  std::lock_guard<std::recursive_mutex> lock(m_mutex);
   std::set<std::string> categories;
 
   for (const auto& pair: m_books) {
@@ -218,40 +521,12 @@ std::vector<std::string> Library::getBooksCategories() const
 
 std::vector<std::string> Library::getBooksCreators() const
 {
-  std::vector<std::string> booksCreators;
-  std::map<std::string, bool> booksCreatorsMap;
-
-  for (auto& pair: m_books) {
-    auto& book = pair.second;
-    auto& creator = book.getCreator();
-    if (booksCreatorsMap.find(creator) == booksCreatorsMap.end()) {
-      if (book.getOrigId().empty()) {
-        booksCreatorsMap[creator] = true;
-        booksCreators.push_back(creator);
-      }
-    }
-  }
-
-  return booksCreators;
+  return getBookPropValueSet(&Book::getCreator);
 }
 
 std::vector<std::string> Library::getBooksPublishers() const
 {
-  std::vector<std::string> booksPublishers;
-  std::map<std::string, bool> booksPublishersMap;
-
-  for (auto& pair:m_books) {
-    auto& book = pair.second;
-    auto& publisher = book.getPublisher();
-    if (booksPublishersMap.find(publisher) == booksPublishersMap.end()) {
-      if (book.getOrigId().empty()) {
-        booksPublishersMap[publisher] = true;
-        booksPublishers.push_back(publisher);
-      }
-    }
-  }
-
-  return booksPublishers;
+  return getBookPropValueSet(&Book::getPublisher);
 }
 
 const std::vector<kiwix::Bookmark> Library::getBookmarks(bool onlyValidBookmarks) const
@@ -261,6 +536,7 @@ const std::vector<kiwix::Bookmark> Library::getBookmarks(bool onlyValidBookmarks
   }
   std::vector<kiwix::Bookmark> validBookmarks;
   auto booksId = getBooksIds();
+  std::lock_guard<std::recursive_mutex> lock(m_mutex);
   for(auto& bookmark:m_bookmarks) {
     if (std::find(booksId.begin(), booksId.end(), bookmark.getBookId()) != booksId.end()) {
       validBookmarks.push_back(bookmark);
@@ -271,6 +547,7 @@ const std::vector<kiwix::Bookmark> Library::getBookmarks(bool onlyValidBookmarks
 
 Library::BookIdCollection Library::getBooksIds() const
 {
+  std::lock_guard<std::recursive_mutex> lock(m_mutex);
   BookIdCollection bookIds;
 
   for (auto& pair: m_books) {
@@ -280,26 +557,19 @@ Library::BookIdCollection Library::getBooksIds() const
   return bookIds;
 }
 
-Library::BookIdCollection Library::filter(const std::string& search) const
-{
-  if (search.empty()) {
-    return getBooksIds();
-  }
-
-  return filter(Filter().query(search));
-}
-
 
 void Library::updateBookDB(const Book& book)
 {
   Xapian::Stem stemmer;
   Xapian::TermGenerator indexer;
-  const std::string lang = book.getLanguage();
-  try {
-    stemmer = Xapian::Stem(iso639_3ToXapian(lang));
-    indexer.set_stemmer(stemmer);
-    indexer.set_stemming_strategy(Xapian::TermGenerator::STEM_SOME);
-  } catch (...) {}
+  const auto langs = book.getLanguages();
+  if ( langs.size() == 1 ) {
+    try {
+      stemmer = Xapian::Stem(iso639_3ToXapian(langs[0]));
+      indexer.set_stemmer(stemmer);
+      indexer.set_stemming_strategy(Xapian::TermGenerator::STEM_SOME);
+    } catch (...) {}
+  }
   Xapian::Document doc;
   indexer.set_document(doc);
 
@@ -314,14 +584,22 @@ void Library::updateBookDB(const Book& book)
   // Index all fields for field-based search
   indexer.index_text(title, 1, "S");
   indexer.index_text(desc,  1, "XD");
-  indexer.index_text(lang,  1, "L");
+  for ( const auto& lang : langs ) {
+    indexer.index_text(lang,  1, "L");
+  }
   indexer.index_text(normalizeText(book.getCreator()),   1, "A");
   indexer.index_text(normalizeText(book.getPublisher()), 1, "XP");
-  indexer.index_text(normalizeText(book.getName()),      1, "XN");
+  doc.add_term("XN"+normalizeText(book.getName()));
+  indexer.index_text(normalizeText(book.getFlavour()),  1, "XF");
   indexer.index_text(normalizeText(book.getCategory()),  1, "XC");
 
-  for ( const auto& tag : split(normalizeText(book.getTags()), ";") )
+  for ( const auto& tag : split(normalizeText(book.getTags()), ";") ) {
     doc.add_boolean_term("XT" + tag);
+    if ( tag[0] != '_' ) {
+      indexer.increase_termpos();
+      indexer.index_text(tag);
+    }
+  }
 
   const std::string idterm = "Q" + book.getId();
   doc.add_boolean_term(idterm);
@@ -353,6 +631,7 @@ Xapian::Query buildXapianQueryFromFilterQuery(const Filter& filter)
   queryParser.add_prefix("title", "S");
   queryParser.add_prefix("description", "XD");
   queryParser.add_prefix("name", "XN");
+  queryParser.add_prefix("flavour", "XF");
   queryParser.add_prefix("category", "XC");
   queryParser.add_prefix("lang", "L");
   queryParser.add_prefix("publisher", "XP");
@@ -366,8 +645,6 @@ Xapian::Query buildXapianQueryFromFilterQuery(const Filter& filter)
   //queryParser.set_stemmer(Xapian::Stem(iso639_3ToXapian(???)));
   //queryParser.set_stemming_strategy(Xapian::QueryParser::STEM_SOME);
   const auto flags = Xapian::QueryParser::FLAG_PHRASE
-                   | Xapian::QueryParser::FLAG_BOOLEAN
-                   | Xapian::QueryParser::FLAG_BOOLEAN_ANY_CASE
                    | Xapian::QueryParser::FLAG_LOVEHATE
                    | Xapian::QueryParser::FLAG_WILDCARD
                    | partialQueryFlag;
@@ -379,14 +656,35 @@ Xapian::Query nameQuery(const std::string& name)
   return Xapian::Query("XN" + normalizeText(name));
 }
 
-Xapian::Query categoryQuery(const std::string& category)
+Xapian::Query flavourQuery(const std::string& name)
 {
-  return Xapian::Query("XC" + normalizeText(category));
+  return Xapian::Query("XF" + normalizeText(name));
 }
 
-Xapian::Query langQuery(const std::string& lang)
+Xapian::Query multipleParamQuery(const std::string& commaSeparatedList, const std::string& prefix)
 {
-  return Xapian::Query("L" + normalizeText(lang));
+  Xapian::Query q;
+  bool firstIteration = true;
+  for ( const auto& elem : kiwix::split(commaSeparatedList, ",") ) {
+    const Xapian::Query singleQuery(prefix + normalizeText(elem));
+    if ( firstIteration ) {
+      q = singleQuery;
+      firstIteration = false;
+    } else {
+      q = Xapian::Query(Xapian::Query::OP_OR, q, singleQuery);
+    }
+  }
+  return q;
+}
+
+Xapian::Query categoryQuery(const std::string& commaSeparatedCategoryList)
+{
+  return multipleParamQuery(commaSeparatedCategoryList, "XC");
+}
+
+Xapian::Query langQuery(const std::string& commaSeparatedLanguageList)
+{
+  return multipleParamQuery(commaSeparatedLanguageList, "L");
 }
 
 Xapian::Query publisherQuery(const std::string& publisher)
@@ -430,6 +728,9 @@ Xapian::Query buildXapianQuery(const Filter& filter)
   if ( filter.hasName() ) {
     q = Xapian::Query(Xapian::Query::OP_AND, q, nameQuery(filter.getName()));
   }
+  if ( filter.hasFlavour() )  {
+    q = Xapian::Query(Xapian::Query::OP_AND, q, flavourQuery(filter.getFlavour()));     
+  }
   if ( filter.hasCategory() ) {
     q = Xapian::Query(Xapian::Query::OP_AND, q, categoryQuery(filter.getCategory()));
   }
@@ -460,6 +761,7 @@ Library::BookIdCollection Library::filterViaBookDB(const Filter& filter) const
 
   BookIdCollection bookIds;
 
+  std::lock_guard<std::recursive_mutex> lock(m_mutex);
   Xapian::Enquire enquire(*m_bookDB);
   enquire.set_query(query);
   const auto results = enquire.get_mset(0, m_books.size());
@@ -473,7 +775,9 @@ Library::BookIdCollection Library::filterViaBookDB(const Filter& filter) const
 Library::BookIdCollection Library::filter(const Filter& filter) const
 {
   BookIdCollection result;
-  for(auto id : filterViaBookDB(filter)) {
+  const auto preliminaryResult = filterViaBookDB(filter);
+  std::lock_guard<std::recursive_mutex> lock(m_mutex);
+  for(auto id : preliminaryResult) {
     if(filter.accept(m_books.at(id))) {
       result.push_back(id);
     }
@@ -542,6 +846,11 @@ std::string Comparator<PUBLISHER>::get_key(const std::string& id)
 
 void Library::sort(BookIdCollection& bookIds, supportedListSortBy sort, bool ascending) const
 {
+  // NOTE: Can reimplement this method in a way that doesn't require locking
+  // NOTE: for the entire duration of the sort. Will need to obtain (under a
+  // NOTE: lock) the required atributes from the books once, and then the
+  // NOTE: sorting will run on a copy of data without locking.
+  std::lock_guard<std::recursive_mutex> lock(m_mutex);
   switch(sort) {
     case TITLE:
       std::sort(bookIds.begin(), bookIds.end(), Comparator<TITLE>(this, ascending));
@@ -564,48 +873,6 @@ void Library::sort(BookIdCollection& bookIds, supportedListSortBy sort, bool asc
 }
 
 
-Library::BookIdCollection Library::listBooksIds(
-    int mode,
-    supportedListSortBy sortBy,
-    const std::string& search,
-    const std::string& language,
-    const std::string& creator,
-    const std::string& publisher,
-    const std::vector<std::string>& tags,
-    size_t maxSize) const {
-
-  Filter _filter;
-  if (mode & LOCAL)
-    _filter.local(true);
-  if (mode & NOLOCAL)
-    _filter.local(false);
-  if (mode & VALID)
-    _filter.valid(true);
-  if (mode & NOVALID)
-    _filter.valid(false);
-  if (mode & REMOTE)
-    _filter.remote(true);
-  if (mode & NOREMOTE)
-    _filter.remote(false);
-  if (!tags.empty())
-    _filter.acceptTags(tags);
-  if (maxSize != 0)
-    _filter.maxSize(maxSize);
-  if (!language.empty())
-    _filter.lang(language);
-  if (!publisher.empty())
-    _filter.publisher(publisher);
-  if (!creator.empty())
-    _filter.creator(creator);
-  if (!search.empty())
-    _filter.query(search);
-
-  auto bookIds = filter(_filter);
-
-  sort(bookIds, sortBy, true);
-  return bookIds;
-}
-
 Filter::Filter()
   : activeFilters(0),
     _maxSize(0)
@@ -629,6 +896,7 @@ enum filterTypes {
   QUERY = FLAG(12),
   NAME = FLAG(13),
   CATEGORY = FLAG(14),
+  FLAVOUR = FLAG(15),
 };
 
 Filter& Filter::local(bool accept)
@@ -730,6 +998,25 @@ Filter& Filter::name(std::string name)
   activeFilters |= NAME;
   return *this;
 }
+  
+Filter& Filter::flavour(std::string flavour)
+{
+  _flavour = flavour;
+  activeFilters |= FLAVOUR;
+  return *this;
+}
+
+Filter& Filter::clearLang()
+{
+  activeFilters &= ~LANG;
+  return *this;
+}
+
+Filter& Filter::clearCategory()
+{
+  activeFilters &= ~CATEGORY;
+  return *this;
+}
 
 #define ACTIVE(X) (activeFilters & (X))
 #define FILTER(TAG, TEST) if (ACTIVE(TAG) && !(TEST)) { return false; }
@@ -763,6 +1050,12 @@ bool Filter::hasCreator() const
   return ACTIVE(_CREATOR);
 }
 
+bool Filter::hasFlavour() const
+{
+  return ACTIVE(FLAVOUR);
+}
+
+
 bool Filter::accept(const Book& book) const
 {
   auto local = !book.getPath().empty();

src/library_dumper.cpp

@@ -0,0 +1,61 @@
+#include "library_dumper.h"
+#include "tools/stringTools.h"
+#include "tools/otherTools.h"
+#include "tools.h"
+
+namespace kiwix
+{
+/* Constructor */
+LibraryDumper::LibraryDumper(const Library* library, const NameMapper* nameMapper)
+  : library(library),
+    nameMapper(nameMapper)
+{
+}
+/* Destructor */
+LibraryDumper::~LibraryDumper()
+{
+}
+
+void LibraryDumper::setOpenSearchInfo(int totalResults, int startIndex, int count)
+{
+  m_totalResults = totalResults;
+  m_startIndex = startIndex,
+  m_count = count;
+}
+
+kainjow::mustache::list LibraryDumper::getCategoryData() const
+{
+  const auto now = gen_date_str();
+  kainjow::mustache::list categoryData;
+  for ( const auto& category : library->getBooksCategories() ) {
+    const auto urlencodedCategoryName = urlEncode(category);
+    categoryData.push_back(kainjow::mustache::object{
+      {"name", category},
+      {"urlencoded_name",  urlencodedCategoryName},
+      {"updated", now},
+      {"id", gen_uuid(libraryId + "/categories/" + urlencodedCategoryName)}
+    });
+  }
+  return categoryData;
+}
+
+kainjow::mustache::list LibraryDumper::getLanguageData() const
+{
+  const auto now = gen_date_str();
+  kainjow::mustache::list languageData;
+  for ( const auto& langAndBookCount : library->getBooksLanguagesWithCounts() ) {
+    const std::string languageCode = langAndBookCount.first;
+    const int bookCount = langAndBookCount.second;
+    const auto languageSelfName = getLanguageSelfName(languageCode);
+    languageData.push_back(kainjow::mustache::object{
+      {"lang_code",  languageCode},
+      {"lang_self_name", languageSelfName},
+      {"book_count", to_string(bookCount)},
+      {"updated", now},
+      {"id", gen_uuid(libraryId + "/languages/" + languageCode)}
+    });
+  }
+  return languageData;
+}
+
+} // namespace kiwix

src/library_dumper.h

@@ -0,0 +1,99 @@
+/*
+ * Copyright 2023 Nikhil Tanwar <2002nikhiltanwar@gmail.com>
+ * Copyright 2017 Matthieu Gautier <mgautier@kymeria.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#ifndef KIWIX_LIBRARY_DUMPER_H
+#define KIWIX_LIBRARY_DUMPER_H
+
+#include <string>
+
+
+#include "library.h"
+#include "name_mapper.h"
+#include <mustache.hpp>
+
+namespace kiwix
+{
+
+/**
+ * A base class to dump Library in various formats.
+ *
+ */
+class LibraryDumper
+{
+ public:
+  LibraryDumper(const Library* library, const NameMapper* NameMapper);
+  ~LibraryDumper();
+
+  void setLibraryId(const std::string& id) { this->libraryId = id;}
+
+  /**
+   * Set the root location used when generating url.
+   *
+   * @param rootLocation the root location to use.
+   */
+  void setRootLocation(const std::string& rootLocation) { this->rootLocation = rootLocation; }
+
+  /**
+   * Set the URL for accessing book content
+   *
+   * @param url the URL of the /content endpoint of the content server
+   */
+  void setContentAccessUrl(const std::string& url) { this->contentAccessUrl = url; }
+
+  /**
+   * Set some informations about the search results.
+   *
+   * @param totalResult the total number of results of the search.
+   * @param startIndex the start index of the result.
+   * @param count the number of result of the current set (or page).
+   */
+  void setOpenSearchInfo(int totalResult, int startIndex, int count);
+
+  /**
+   * Sets user default language
+   *
+   * @param userLang the user language to be set
+   */
+  void setUserLanguage(std::string userLang) { this->m_userLang = userLang; }
+
+  /**
+   * Get the data of categories
+   */
+  kainjow::mustache::list getCategoryData() const;
+
+  /**
+   * Get the data of languages
+   */
+  kainjow::mustache::list getLanguageData() const;
+
+ protected:
+  const kiwix::Library* const library;
+  const kiwix::NameMapper* const nameMapper;
+  std::string libraryId;
+  std::string rootLocation;
+  std::string contentAccessUrl;
+  std::string m_userLang;
+  int m_totalResults;
+  int m_startIndex;
+  int m_count;
+};
+}
+
+#endif // KIWIX_LIBRARY_DUMPER_H

src/libxml_dumper.cpp

@@ -20,10 +20,10 @@
 #include "libxml_dumper.h"
 #include "book.h"
 
+#include "tools.h"
 #include "tools/base64.h"
 #include "tools/stringTools.h"
 #include "tools/otherTools.h"
-#include "tools/pathTools.h"
 
 namespace kiwix
 {
@@ -54,16 +54,19 @@ void LibXMLDumper::handleBook(Book book, pugi::xml_node root_node) {
   if (book.getOrigId().empty()) {
     ADD_ATTR_NOT_EMPTY(entry_node, "title", book.getTitle());
     ADD_ATTR_NOT_EMPTY(entry_node, "description", book.getDescription());
-    ADD_ATTR_NOT_EMPTY(entry_node, "language", book.getLanguage());
+    ADD_ATTR_NOT_EMPTY(entry_node, "language", book.getCommaSeparatedLanguages());
     ADD_ATTR_NOT_EMPTY(entry_node, "creator", book.getCreator());
     ADD_ATTR_NOT_EMPTY(entry_node, "publisher", book.getPublisher());
     ADD_ATTR_NOT_EMPTY(entry_node, "name", book.getName());
     ADD_ATTR_NOT_EMPTY(entry_node, "flavour", book.getFlavour());
     ADD_ATTR_NOT_EMPTY(entry_node, "tags", book.getTags());
-    ADD_ATTR_NOT_EMPTY(entry_node, "faviconMimeType", book.getFaviconMimeType());
-    ADD_ATTR_NOT_EMPTY(entry_node, "faviconUrl", book.getFaviconUrl());
-    if (!book.getFavicon().empty())
-      ADD_ATTRIBUTE(entry_node, "favicon", base64_encode(book.getFavicon()));
+    try {
+      auto defaultIllustration = book.getIllustration(48);
+      ADD_ATTR_NOT_EMPTY(entry_node, "faviconMimeType", defaultIllustration->mimeType);
+      ADD_ATTR_NOT_EMPTY(entry_node, "faviconUrl", defaultIllustration->url);
+      if (!defaultIllustration->getData().empty())
+        ADD_ATTRIBUTE(entry_node, "favicon", base64_encode(defaultIllustration->getData()));
+    } catch(...) {}
   } else {
     ADD_ATTRIBUTE(entry_node, "origId", book.getOrigId());
   }
@@ -91,14 +94,18 @@ void LibXMLDumper::handleBookmark(Bookmark bookmark, pugi::xml_node root_node) {
   auto book_node = entry_node.append_child("book");
 
   try {
-    auto book = library->getBookById(bookmark.getBookId());
+    auto book = library->getBookByIdThreadSafe(bookmark.getBookId());
     ADD_TEXT_ENTRY(book_node, "id", book.getId());
     ADD_TEXT_ENTRY(book_node, "title", book.getTitle());
-    ADD_TEXT_ENTRY(book_node, "language", book.getLanguage());
+    ADD_TEXT_ENTRY(book_node, "name", book.getName());
+    ADD_TEXT_ENTRY(book_node, "flavour", book.getFlavour());
+    ADD_TEXT_ENTRY(book_node, "language", book.getCommaSeparatedLanguages());
     ADD_TEXT_ENTRY(book_node, "date", book.getDate());
   } catch (...) {
     ADD_TEXT_ENTRY(book_node, "id", bookmark.getBookId());
     ADD_TEXT_ENTRY(book_node, "title", bookmark.getBookTitle());
+    ADD_TEXT_ENTRY(book_node, "name", bookmark.getBookName());
+    ADD_TEXT_ENTRY(book_node, "flavour", bookmark.getBookFlavour());
     ADD_TEXT_ENTRY(book_node, "language", bookmark.getLanguage());
     ADD_TEXT_ENTRY(book_node, "date", bookmark.getDate());
   }
@@ -132,7 +139,7 @@ std::string LibXMLDumper::dumpLibXMLBookmark()
   pugi::xml_node bookmarksNode = doc.append_child("bookmarks");
 
   if (library) {
-    for (auto& bookmark: library->getBookmarks()) {
+    for (auto& bookmark: library->getBookmarks(false)) {
       handleBookmark(bookmark, bookmarksNode);
     }
   }

src/manager.cpp

@@ -19,34 +19,86 @@
 
 #include "manager.h"
 
+#include "tools.h"
 #include "tools/pathTools.h"
 
 #include <pugixml.hpp>
+#include <filesystem>
+#include <iostream>
+#include <set>
+#include <queue>
+#include <cctype>
+#include <algorithm>
+
+namespace fs = std::filesystem;
 
 namespace kiwix
 {
-/* Constructor */
-Manager::Manager(LibraryManipulator* manipulator):
-  writableLibraryPath(""),
-  manipulator(manipulator),
-  mustDeleteManipulator(false)
-{
-}
 
-Manager::Manager(Library* library) :
-  writableLibraryPath(""),
-  manipulator(new DefaultLibraryManipulator(library)),
-  mustDeleteManipulator(true)
-{
-}
+////////////////////////////////////////////////////////////////////////////////
+// LibraryManipulator
+////////////////////////////////////////////////////////////////////////////////
 
-/* Destructor */
-Manager::~Manager()
+LibraryManipulator::LibraryManipulator(LibraryPtr library)
+  : library(library)
+{}
+
+LibraryManipulator::~LibraryManipulator()
+{}
+
+bool LibraryManipulator::addBookToLibrary(const Book& book)
 {
-  if (mustDeleteManipulator) {
-    delete manipulator;
+  const auto ret = library->addBook(book);
+  if ( ret ) {
+    bookWasAddedToLibrary(book);
   }
+  return ret;
 }
+
+void LibraryManipulator::addBookmarkToLibrary(const Bookmark& bookmark)
+{
+  library->addBookmark(bookmark);
+  bookmarkWasAddedToLibrary(bookmark);
+}
+
+uint32_t LibraryManipulator::removeBooksNotUpdatedSince(Library::Revision rev)
+{
+  const auto n = library->removeBooksNotUpdatedSince(rev);
+  if ( n != 0 ) {
+    booksWereRemovedFromLibrary();
+  }
+  return n;
+}
+
+void LibraryManipulator::bookWasAddedToLibrary(const Book& book)
+{
+}
+
+void LibraryManipulator::bookmarkWasAddedToLibrary(const Bookmark& bookmark)
+{
+}
+
+void LibraryManipulator::booksWereRemovedFromLibrary()
+{
+}
+
+////////////////////////////////////////////////////////////////////////////////
+// Manager
+////////////////////////////////////////////////////////////////////////////////
+
+/* Constructor */
+Manager::Manager(LibraryManipulator manipulator):
+  writableLibraryPath(""),
+  manipulator(manipulator)
+{
+}
+
+Manager::Manager(LibraryPtr library) :
+  writableLibraryPath(""),
+  manipulator(LibraryManipulator(library))
+{
+}
+
 bool Manager::parseXmlDom(const pugi::xml_document& doc,
                           bool readOnly,
                           const std::string& libraryPath,
@@ -67,7 +119,7 @@ bool Manager::parseXmlDom(const pugi::xml_document& doc,
     if (!trustLibrary && !book.getPath().empty()) {
       this->readBookFromPath(book.getPath(), &book);
     }
-    manipulator->addBookToLibrary(book);
+    manipulator.addBookToLibrary(book);
   }
 
   return true;
@@ -112,7 +164,7 @@ bool Manager::parseOpdsDom(const pugi::xml_document& doc, const std::string& url
     book.updateFromOpds(entryNode, urlHost);
 
     /* Update the book properties with the new importer */
-    manipulator->addBookToLibrary(book);
+    manipulator.addBookToLibrary(book);
   }
 
   return true;
@@ -184,10 +236,10 @@ std::string Manager::addBookFromPathAndGetId(const std::string& pathToOpen,
     }
 
     if (!checkMetaData
-        || (checkMetaData && !book.getTitle().empty() && !book.getLanguage().empty()
+        || (!book.getTitle().empty() && !book.getLanguages().empty()
             && !book.getDate().empty())) {
       book.setUrl(url);
-      manipulator->addBookToLibrary(book);
+      manipulator.addBookToLibrary(book);
       return book.getId();
     }
   }
@@ -207,6 +259,58 @@ bool Manager::addBookFromPath(const std::string& pathToOpen,
           .empty());
 }
 
+void Manager::addBooksFromDirectory(const std::string& path,
+                                    const bool verboseFlag)
+{
+  std::set<std::string> iteratedDirs;
+  std::queue<std::string> dirQueue;
+  dirQueue.push(fs::absolute(path).u8string());
+  int totalBooksAdded = 0;
+  if (verboseFlag)
+    std::cout << "Adding books from the directory tree: " << dirQueue.front() << std::endl;
+
+  while (!dirQueue.empty()) {
+    const auto currentPath = dirQueue.front();
+    dirQueue.pop();
+    if (verboseFlag)
+      std::cout << "Visiting directory: " << currentPath << std::endl;
+    for (const auto& dirEntry : fs::directory_iterator(currentPath)) {
+      auto resolvedPath = dirEntry.path();
+      if (fs::is_symlink(dirEntry)) {
+        try {
+          resolvedPath = fs::canonical(dirEntry.path());
+        } catch (const std::exception& e) {
+          std::cerr << "Could not resolve symlink " << resolvedPath.u8string() << " to a valid path. Skipping..." << std::endl;
+          continue;
+        }
+      }
+      const std::string pathString = resolvedPath.u8string();
+      std::string resolvedPathExtension = resolvedPath.extension();
+      std::transform(resolvedPathExtension.begin(), resolvedPathExtension.end(), resolvedPathExtension.begin(),
+          [](unsigned char c){ return std::tolower(c); });
+      if (fs::is_directory(resolvedPath)) {
+        if (iteratedDirs.find(pathString) == iteratedDirs.end())
+          dirQueue.push(pathString);
+        else if (verboseFlag)
+          std::cout << "Already iterated over " << pathString << ". Skipping..." << std::endl;
+      } else if (resolvedPathExtension == ".zim" || resolvedPathExtension == ".zimaa") {
+        if (!this->addBookFromPath(pathString, pathString, "", false)) {
+          std::cerr << "Could not add " << pathString << " into the library." << std::endl;
+        } else if (verboseFlag) {
+          std::cout << "Added " << pathString << " into the library." << std::endl;
+          totalBooksAdded++;
+        }
+      } else if (verboseFlag) {
+        std::cout << "Skipped " << pathString << " - unsupported file type or permission denied." << std::endl;
+      }
+    }
+    iteratedDirs.insert(currentPath);
+  }
+
+  if (verboseFlag)
+    std::cout << "Traversal completed. Total books added: " << totalBooksAdded << std::endl;
+}
+
 bool Manager::readBookFromPath(const std::string& path, kiwix::Book* book)
 {
   std::string tmp_path = path;
@@ -214,8 +318,8 @@ bool Manager::readBookFromPath(const std::string& path, kiwix::Book* book)
     tmp_path = computeAbsolutePath(getCurrentDirectory(), path);
   }
   try {
-    kiwix::Reader reader(tmp_path);
-    book->update(reader);
+    zim::Archive archive(tmp_path);
+    book->update(archive);
     book->setPathValid(true);
   } catch (const std::exception& e) {
     book->setPathValid(false);
@@ -242,10 +346,27 @@ bool Manager::readBookmarkFile(const std::string& path)
 
     bookmark.updateFromXml(node);
 
-    manipulator->addBookmarkToLibrary(bookmark);
+    manipulator.addBookmarkToLibrary(bookmark);
   }
 
   return true;
 }
 
+void Manager::reload(const Paths& paths)
+{
+  const auto libRevision = manipulator.getLibrary()->getRevision();
+  for (std::string path : paths) {
+    if (!path.empty()) {
+      if ( kiwix::isRelativePath(path) )
+        path = kiwix::computeAbsolutePath(kiwix::getCurrentDirectory(), path);
+
+      if (!readFile(path, false, true)) {
+        throw std::runtime_error("Failed to load the XML library file '" + path + "'.");
+      }
+    }
+  }
+
+  manipulator.removeBooksNotUpdatedSince(libRevision);
+}
+
 }

src/meson.build

@@ -5,11 +5,10 @@ kiwix_sources = [
   'manager.cpp',
   'libxml_dumper.cpp',
   'opds_dumper.cpp',
+  'html_dumper.cpp',
+  'library_dumper.cpp',
   'downloader.cpp',
-  'reader.cpp',
-  'entry.cpp',
   'server.cpp',
-  'searcher.cpp',
   'search_renderer.cpp',
   'subprocess.cpp',
   'aria2.cpp',
@@ -18,7 +17,10 @@ kiwix_sources = [
   'tools/regexTools.cpp',
   'tools/stringTools.cpp',
   'tools/networkTools.cpp',
+  'tools/opdsParsingTools.cpp',
+  'tools/languageTools.cpp',
   'tools/otherTools.cpp',
+  'tools/archiveTools.cpp',
   'kiwixserve.cpp',
   'name_mapper.cpp',
   'server/byte_range.cpp',
@@ -26,9 +28,14 @@ kiwix_sources = [
   'server/request_context.cpp',
   'server/response.cpp',
   'server/internalServer.cpp',
-  'server/internalServer_catalog_v2.cpp'
+  'server/internalServer_catalog.cpp',
+  'server/i18n.cpp',
+  'opds_catalog.cpp',
+  'spelling_correction.cpp',
+  'version.cpp'
 ]
 kiwix_sources += lib_resources
+kiwix_sources += i18n_resources
 
 if host_machine.system() == 'windows'
   kiwix_sources += 'subprocess_windows.cpp'
@@ -36,22 +43,14 @@ else
   kiwix_sources += 'subprocess_unix.cpp'
 endif
 
-if wrapper.contains('android')
-  install_dir = 'kiwix-lib/jniLibs/' + meson.get_cross_property('android_abi')
-else
-  install_dir = get_option('libdir')
-endif
-
-if wrapper.contains('android') or wrapper.contains('java')
-  subdir('wrapper/java')
-endif
+install_dir = get_option('libdir')
 
 config_h = configure_file(output : 'kiwix_config.h',
                configuration : conf,
                input : 'config.h.in')
 install_headers(config_h, subdir:'kiwix')
 
-kiwixlib = library('kiwix',
+libkiwix = library('kiwix',
                    kiwix_sources,
                    include_directories : inc,
                    dependencies : all_deps,

src/name_mapper.cpp

@@ -24,39 +24,85 @@
 
 namespace kiwix {
 
-HumanReadableNameMapper::HumanReadableNameMapper(kiwix::Library& library, bool withAlias) {
-  for (auto& bookId: library.filter(kiwix::Filter().local(true).valid(true))) {
+HumanReadableNameMapper::HumanReadableNameMapper(const kiwix::Library& library, bool withAlias) {
+  for (auto& bookId: library.filter(kiwix::Filter())) {
     auto& currentBook = library.getBookById(bookId);
     auto bookName = currentBook.getHumanReadableIdFromPath();
     m_idToName[bookId] = bookName;
-    m_nameToId[bookName] = bookId;
+    mapName(library, bookName, bookId);
 
     if (!withAlias)
       continue;
 
     auto aliasName = replaceRegex(bookName, "", "_[[:digit:]]{4}-[[:digit:]]{2}$");
-    if (aliasName == bookName) {
-      continue;
-    }
-    if (m_nameToId.find(aliasName) == m_nameToId.end()) {
-      m_nameToId[aliasName] = bookId;
-    } else {
-      auto alreadyPresentPath = library.getBookById(m_nameToId[aliasName]).getPath();
-      std::cerr << "Path collision: " << alreadyPresentPath
-                << " and " << currentBook.getPath()
-                << " can't share the same URL path '" << aliasName << "'."
-                << " Therefore, only " << alreadyPresentPath
-                << " will be served." << std::endl;
+    if (aliasName != bookName) {
+      mapName(library, aliasName, bookId);
     }
   }
 }
 
-std::string HumanReadableNameMapper::getNameForId(const std::string& id) {
+void HumanReadableNameMapper::mapName(const Library& library, std::string name, std::string bookId) {
+  if (m_nameToId.find(name) == m_nameToId.end()) {
+    m_nameToId[name] = bookId;
+  } else {
+    const auto& currentBook = library.getBookById(bookId);
+    auto alreadyPresentPath = library.getBookById(m_nameToId[name]).getPath();
+    std::cerr << "Path collision: '" << alreadyPresentPath
+              << "' and '" << currentBook.getPath()
+              << "' can't share the same URL path '" << name << "'."
+              << " Therefore, only '" << alreadyPresentPath
+              << "' will be served." << std::endl;
+  }
+}
+
+std::string HumanReadableNameMapper::getNameForId(const std::string& id) const {
   return m_idToName.at(id);
 }
 
-std::string HumanReadableNameMapper::getIdForName(const std::string& name) {
+std::string HumanReadableNameMapper::getIdForName(const std::string& name) const {
   return m_nameToId.at(name);
 }
 
+////////////////////////////////////////////////////////////////////////////////
+// UpdatableNameMapper
+////////////////////////////////////////////////////////////////////////////////
+
+UpdatableNameMapper::UpdatableNameMapper(LibraryPtr lib, bool withAlias)
+  : library(lib)
+  , withAlias(withAlias)
+{
+  update();
+}
+
+void UpdatableNameMapper::update()
+{
+  const auto newNameMapper = new HumanReadableNameMapper(*library, withAlias);
+  std::lock_guard<std::mutex> lock(mutex);
+  nameMapper.reset(newNameMapper);
+}
+
+UpdatableNameMapper::NameMapperHandle
+UpdatableNameMapper::currentNameMapper() const
+{
+  // Return a copy of the handle to the current NameMapper object. It will
+  // ensure that the object survives any call to UpdatableNameMapper::update()
+  // made before the completion of any pending operation on that object.
+  std::lock_guard<std::mutex> lock(mutex);
+  return nameMapper;
+}
+
+std::string UpdatableNameMapper::getNameForId(const std::string& id) const
+{
+  // Ensure that the current nameMapper object survives a concurrent call
+  // to UpdatableNameMapper::update()
+  return currentNameMapper()->getNameForId(id);
+}
+
+std::string UpdatableNameMapper::getIdForName(const std::string& name) const
+{
+  // Ensure that the current nameMapper object survives a concurrent call
+  // to UpdatableNameMapper::update()
+  return currentNameMapper()->getIdForName(name);
+}
+
 }

src/opds_catalog.cpp

@@ -0,0 +1,74 @@
+/*
+ * Copyright 2021 Veloman Yunkan <veloman.yunkan@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#include "opds_catalog.h"
+#include "tools/stringTools.h"
+
+#include <sstream>
+
+namespace kiwix
+{
+
+namespace
+{
+const char opdsSearchEndpoint[] = "/catalog/v2/entries";
+
+enum Separator { AMP };
+
+std::ostringstream& operator<<(std::ostringstream& oss, Separator sep)
+{
+  if ( oss.tellp() > 0 )
+    oss << "&";
+  return oss;
+}
+
+std::string buildSearchString(const Filter& f)
+{
+  std::ostringstream oss;
+  if ( f.hasQuery() )
+    oss << AMP << "q=" << urlEncode(f.getQuery());
+
+  if ( f.hasCategory() )
+    oss << AMP << "category=" << urlEncode(f.getCategory());
+
+  if ( f.hasLang() )
+    oss << AMP << "lang=" << urlEncode(f.getLang());
+
+  if ( f.hasName() )
+    oss << AMP << "name=" << urlEncode(f.getName());
+
+  if ( !f.getAcceptTags().empty() )
+    oss << AMP << "tag=" << urlEncode(join(f.getAcceptTags(), ";"));
+
+  return oss.str();
+}
+
+} // unnamed namespace
+
+std::string getSearchUrl(const Filter& f)
+{
+  const std::string searchString = buildSearchString(f);
+
+  if ( searchString.empty() )
+    return opdsSearchEndpoint;
+  else
+    return opdsSearchEndpoint + ("?" + searchString);
+}
+
+} // namespace kiwix

src/opds_dumper.cpp

@@ -20,17 +20,18 @@
 #include "opds_dumper.h"
 #include "book.h"
 
-#include "tools/otherTools.h"
-
-#include "kiwixlib-resources.h"
+#include "libkiwix-resources.h"
 #include <mustache.hpp>
 
+#include "tools/stringTools.h"
+#include "tools/otherTools.h"
+
 namespace kiwix
 {
 
 /* Constructor */
-OPDSDumper::OPDSDumper(Library* library)
-  : library(library)
+OPDSDumper::OPDSDumper(const Library* library, const NameMapper* nameMapper)
+  : LibraryDumper(library, nameMapper)
 {
 }
 /* Destructor */
@@ -38,35 +39,46 @@ OPDSDumper::~OPDSDumper()
 {
 }
 
-void OPDSDumper::setOpenSearchInfo(int totalResults, int startIndex, int count)
-{
-  m_totalResults = totalResults;
-  m_startIndex = startIndex,
-  m_count = count;
-}
-
 namespace
 {
 
-typedef kainjow::mustache::data MustacheData;
-typedef kainjow::mustache::list BookData;
+const std::string XML_HEADER(R"(<?xml version="1.0" encoding="UTF-8"?>)");
 
-BookData getBookData(const Library* library, const std::vector<std::string>& bookIds)
+typedef kainjow::mustache::data MustacheData;
+typedef kainjow::mustache::list BooksData;
+typedef kainjow::mustache::list IllustrationInfo;
+
+IllustrationInfo getBookIllustrationInfo(const Book& book)
 {
-  BookData bookData;
-  for ( const auto& bookId : bookIds ) {
-    const Book& book = library->getBookById(bookId);
-    const MustacheData bookUrl = book.getUrl().empty()
-                               ? MustacheData(false)
-                               : MustacheData(book.getUrl());
-    bookData.push_back(kainjow::mustache::object{
-      {"id", "urn:uuid:"+book.getId()},
+    kainjow::mustache::list illustrations;
+    for ( const auto& illustration : book.getIllustrations() ) {
+      // For now, we are handling only sizexsize@1 illustration.
+      // So we can simply pass one size to mustache.
+      illustrations.push_back(kainjow::mustache::object{
+        {"icon_size", to_string(illustration->width)},
+        {"icon_mimetype", illustration->mimeType}
+      });
+    }
+    return illustrations;
+}
+
+std::string fullEntryXML(const Book& book,
+                         const std::string& rootLocation,
+                         const std::string& contentAccessUrl,
+                         const std::string& contentId)
+{
+    const auto bookDate = book.getDate() + "T00:00:00Z";
+    const kainjow::mustache::object data{
+      {"root",  rootLocation},
+      {"contentAccessUrl",  onlyAsNonEmptyMustacheValue(contentAccessUrl)},
+      {"id", book.getId()},
       {"name", book.getName()},
       {"title", book.getTitle()},
       {"description", book.getDescription()},
-      {"language", book.getLanguage()},
-      {"content_id",  book.getHumanReadableIdFromPath()},
-      {"updated", book.getDate() + "T00:00:00Z"},
+      {"language", book.getCommaSeparatedLanguages()},
+      {"content_id",  urlEncode(contentId)},
+      {"updated", bookDate}, // XXX: this should be the entry update datetime
+      {"book_date", bookDate},
       {"category", book.getCategory()},
       {"flavour", book.getFlavour()},
       {"tags", book.getTags()},
@@ -74,66 +86,106 @@ BookData getBookData(const Library* library, const std::vector<std::string>& boo
       {"media_count", to_string(book.getMediaCount())},
       {"author_name", book.getCreator()},
       {"publisher_name", book.getPublisher()},
-      {"url", bookUrl},
+      {"url", onlyAsNonEmptyMustacheValue(book.getUrl())},
       {"size", to_string(book.getSize())},
-    });
+      {"icons", getBookIllustrationInfo(book)},
+    };
+    return render_template(RESOURCE::templates::catalog_v2_entry_xml, data);
+}
+
+std::string partialEntryXML(const Book& book, const std::string& rootLocation)
+{
+    const auto bookDate = book.getDate() + "T00:00:00Z";
+    const kainjow::mustache::object data{
+      {"root",  rootLocation},
+      {"endpoint_root", rootLocation + "/catalog/v2"},
+      {"id", book.getId()},
+      {"title", book.getTitle()},
+      {"updated", bookDate}, // XXX: this should be the entry update datetime
+    };
+    const auto xmlTemplate = RESOURCE::templates::catalog_v2_partial_entry_xml;
+    return render_template(xmlTemplate, data);
+}
+
+BooksData getBooksData(const Library* library,
+                       const NameMapper* nameMapper,
+                       const std::vector<std::string>& bookIds,
+                       const std::string& rootLocation,
+                       const std::string& contentAccessUrl,
+                       bool partial)
+{
+  BooksData booksData;
+  for ( const auto& bookId : bookIds ) {
+    try {
+      const Book book = library->getBookByIdThreadSafe(bookId);
+      const std::string contentId = nameMapper->getNameForId(bookId);
+      const auto entryXML = partial
+                          ? partialEntryXML(book, rootLocation)
+                          : fullEntryXML(book, rootLocation, contentAccessUrl, contentId);
+      booksData.push_back(kainjow::mustache::object{ {"entry", entryXML} });
+    } catch ( const std::out_of_range& ) {
+      // the book was removed from the library since its id was obtained
+      // ignore it
+    }
   }
 
-  return bookData;
+  return booksData;
 }
 
 } // unnamed namespace
 
 string OPDSDumper::dumpOPDSFeed(const std::vector<std::string>& bookIds, const std::string& query) const
 {
-  const auto bookData = getBookData(library, bookIds);
+  const auto booksData = getBooksData(library, nameMapper, bookIds, rootLocation, contentAccessUrl, false);
   const kainjow::mustache::object template_data{
      {"date", gen_date_str()},
      {"root", rootLocation},
      {"feed_id", gen_uuid(libraryId + "/catalog/search?"+query)},
-     {"filter", query.empty() ? MustacheData(false) : MustacheData(query)},
+     {"filter", onlyAsNonEmptyMustacheValue(query)},
      {"totalResults", to_string(m_totalResults)},
      {"startIndex", to_string(m_startIndex)},
      {"itemsPerPage", to_string(m_count)},
-     {"books", bookData }
+     {"books", booksData }
   };
 
   return render_template(RESOURCE::templates::catalog_entries_xml, template_data);
 }
 
-string OPDSDumper::dumpOPDSFeedV2(const std::vector<std::string>& bookIds, const std::string& query) const
+string OPDSDumper::dumpOPDSFeedV2(const std::vector<std::string>& bookIds, const std::string& query, bool partial) const
 {
-  const auto bookData = getBookData(library, bookIds);
+  const auto endpointRoot = rootLocation + "/catalog/v2";
+  const auto booksData = getBooksData(library, nameMapper, bookIds, rootLocation, contentAccessUrl, partial);
 
+  const char* const endpoint = partial ? "/partial_entries" : "/entries";
+  const std::string url = endpoint + (query.empty() ? "" : "?" + query);
   const kainjow::mustache::object template_data{
      {"date", gen_date_str()},
-     {"endpoint_root", rootLocation + "/catalog/v2"},
-     {"feed_id", gen_uuid(libraryId + "/entries?"+query)},
-     {"filter", query.empty() ? MustacheData(false) : MustacheData(query)},
-     {"query", query.empty() ? "" : "?" + urlEncode(query)},
+     {"endpoint_root", endpointRoot},
+     {"feed_id", gen_uuid(libraryId + endpoint + "?" + query)},
+     {"filter", onlyAsNonEmptyMustacheValue(query)},
+     {"self_url", url},
      {"totalResults", to_string(m_totalResults)},
      {"startIndex", to_string(m_startIndex)},
      {"itemsPerPage", to_string(m_count)},
-     {"books", bookData }
+     {"books", booksData }
   };
 
   return render_template(RESOURCE::templates::catalog_v2_entries_xml, template_data);
 }
 
-std::string OPDSDumper::categoriesOPDSFeed(const std::vector<std::string>& categories) const
+std::string OPDSDumper::dumpOPDSCompleteEntry(const std::string& bookId) const
+{
+  const auto book = library->getBookById(bookId);
+  const std::string contentId = nameMapper->getNameForId(bookId);
+  return XML_HEADER
+         + "\n"
+         + fullEntryXML(book, rootLocation, contentAccessUrl, contentId);
+}
+
+std::string OPDSDumper::categoriesOPDSFeed() const
 {
   const auto now = gen_date_str();
-  kainjow::mustache::list categoryData;
-  for ( const auto& category : categories ) {
-    const auto urlencodedCategoryName = urlEncode(category);
-    categoryData.push_back(kainjow::mustache::object{
-      {"name", category},
-      {"urlencoded_name",  urlencodedCategoryName},
-      {"updated", now},
-      {"id", gen_uuid(libraryId + "/categories/" + urlencodedCategoryName)}
-    });
-  }
-
+  kainjow::mustache::list categoryData = getCategoryData();
   return render_template(
              RESOURCE::templates::catalog_v2_categories_xml,
              kainjow::mustache::object{
@@ -145,4 +197,19 @@ std::string OPDSDumper::categoriesOPDSFeed(const std::vector<std::string>& categ
   );
 }
 
+std::string OPDSDumper::languagesOPDSFeed() const
+{
+  const auto now = gen_date_str();
+  kainjow::mustache::list languageData = getLanguageData();
+  return render_template(
+             RESOURCE::templates::catalog_v2_languages_xml,
+             kainjow::mustache::object{
+               {"date", now},
+               {"endpoint_root", rootLocation + "/catalog/v2"},
+               {"feed_id", gen_uuid(libraryId + "/languages")},
+               {"languages", languageData }
+             }
+  );
+}
+
 }

src/opds_dumper.h

@@ -26,11 +26,9 @@
 
 #include <pugixml.hpp>
 
-#include "tools/base64.h"
-#include "tools/pathTools.h"
-#include "tools/regexTools.h"
 #include "library.h"
-#include "reader.h"
+#include "name_mapper.h"
+#include "library_dumper.h"
 
 using namespace std;
 
@@ -41,11 +39,10 @@ namespace kiwix
  * A tool to dump a `Library` into a opds stream.
  *
  */
-class OPDSDumper
+class OPDSDumper : public LibraryDumper
 {
  public:
-  OPDSDumper() = default;
-  OPDSDumper(Library* library);
+  OPDSDumper(const Library* library, const NameMapper* NameMapper);
   ~OPDSDumper();
 
   /**
@@ -62,48 +59,32 @@ class OPDSDumper
    *
    * @param bookIds the ids of the books to include in the feed
    * @param query the query used to obtain the list of book ids
+   * @param partial whether the feed should include partial or complete entries
    * @return The OPDS feed.
    */
-  std::string dumpOPDSFeedV2(const std::vector<std::string>& bookIds, const std::string& query) const;
+  std::string dumpOPDSFeedV2(const std::vector<std::string>& bookIds, const std::string& query, bool partial) const;
+
+  /**
+   * Dump the OPDS complete entry document.
+   *
+   * @param bookId the id of the book
+   * @return The OPDS complete entry document.
+   */
+  std::string dumpOPDSCompleteEntry(const std::string& bookId) const;
 
   /**
    * Dump the categories OPDS feed.
    *
-   * @param categories list of category names
    * @return The OPDS feed.
    */
-  std::string categoriesOPDSFeed(const std::vector<std::string>& categories) const;
+  std::string categoriesOPDSFeed() const;
 
   /**
-   * Set the id of the library.
+   * Dump the languages OPDS feed.
    *
-   * @param id the id to use.
+   * @return The OPDS feed.
    */
-  void setLibraryId(const std::string& id) { this->libraryId = id;}
-
-  /**
-   * Set the root location used when generating url.
-   *
-   * @param rootLocation the root location to use.
-   */
-  void setRootLocation(const std::string& rootLocation) { this->rootLocation = rootLocation; }
-
-  /**
-   * Set some informations about the search results.
-   *
-   * @param totalResult the total number of results of the search.
-   * @param startIndex the start index of the result.
-   * @param count the number of result of the current set (or page).
-   */
-  void setOpenSearchInfo(int totalResult, int startIndex, int count);
-
- protected:
-  kiwix::Library* library;
-  std::string libraryId;
-  std::string rootLocation;
-  int m_totalResults;
-  int m_startIndex;
-  int m_count;
+  std::string languagesOPDSFeed() const;
 };
 }
 

src/reader.cpp

@@ -1,581 +0,0 @@
-/*
- * Copyright 2011 Emmanuel Engelhart <kelson@kiwix.org>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU  General Public License as published by
- * the Free Software Foundation; either version 3 of the License, or
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
- * MA 02110-1301, USA.
- */
-
-#include "reader.h"
-#include <time.h>
-
-#include <zim/search.h>
-#include <zim/item.h>
-#include <zim/error.h>
-
-#include "tools/otherTools.h"
-
-inline char hi(char v)
-{
-  char hex[] = "0123456789abcdef";
-  return hex[(v >> 4) & 0xf];
-}
-
-inline char lo(char v)
-{
-  char hex[] = "0123456789abcdef";
-  return hex[v & 0xf];
-}
-
-std::string hexUUID(std::string in)
-{
-  std::ostringstream out;
-  for (unsigned n = 0; n < 4; ++n) {
-    out << hi(in[n]) << lo(in[n]);
-  }
-  out << '-';
-  for (unsigned n = 4; n < 6; ++n) {
-    out << hi(in[n]) << lo(in[n]);
-  }
-  out << '-';
-  for (unsigned n = 6; n < 8; ++n) {
-    out << hi(in[n]) << lo(in[n]);
-  }
-  out << '-';
-  for (unsigned n = 8; n < 10; ++n) {
-    out << hi(in[n]) << lo(in[n]);
-  }
-  out << '-';
-  for (unsigned n = 10; n < 16; ++n) {
-    out << hi(in[n]) << lo(in[n]);
-  }
-  std::string op = out.str();
-  return op;
-}
-
-namespace kiwix
-{
-/* Constructor */
-Reader::Reader(const string zimFilePath)
-  :  zimArchive(nullptr),
-     zimFilePath(zimFilePath)
-{
-  string tmpZimFilePath = zimFilePath;
-
-  /* Remove potential trailing zimaa */
-  size_t found = tmpZimFilePath.rfind("zimaa");
-  if (found != string::npos && tmpZimFilePath.size() > 5
-      && found == tmpZimFilePath.size() - 5) {
-    tmpZimFilePath.resize(tmpZimFilePath.size() - 2);
-  }
-
-  zimArchive.reset(new zim::Archive(tmpZimFilePath));
-
-  /* initialize random seed: */
-  srand(time(nullptr));
-}
-
-#ifndef _WIN32
-Reader::Reader(int fd)
-  :  zimArchive(new zim::Archive(fd)),
-     zimFilePath("")
-{
-  /* initialize random seed: */
-  srand(time(nullptr));
-}
-
-Reader::Reader(int fd, zim::offset_type offset, zim::size_type size)
-  :  zimArchive(new zim::Archive(fd, offset, size)),
-     zimFilePath("")
-{
-  /* initialize random seed: */
-  srand(time(nullptr));
-}
-#endif // #ifndef _WIN32
-
-zim::Archive* Reader::getZimArchive() const
-{
-  return zimArchive.get();
-}
-
-MimeCounterType Reader::parseCounterMetadata() const
-{
-  try {
-    auto counterContent = zimArchive->getMetadata("Counter");
-    return parseMimetypeCounter(counterContent);
-  } catch (zim::EntryNotFound& e) {
-    return {};
-  }
-}
-
-/* Get the count of articles which can be indexed/displayed */
-unsigned int Reader::getArticleCount() const
-{
-  std::map<const std::string, unsigned int> counterMap
-      = this->parseCounterMetadata();
-  unsigned int counter = 0;
-
-  for(auto &pair:counterMap) {
-    if (startsWith(pair.first, "text/html")) {
-      counter += pair.second;
-    }
-  }
-
-  return counter;
-}
-
-/* Get the count of medias content in the ZIM file */
-unsigned int Reader::getMediaCount() const
-{
-  std::map<const std::string, unsigned int> counterMap
-      = this->parseCounterMetadata();
-  unsigned int counter = 0;
-
-  for (auto &pair:counterMap) {
-    if (startsWith(pair.first, "image/") ||
-        startsWith(pair.first, "video/") ||
-        startsWith(pair.first, "audio/")) {
-      counter += pair.second;
-    }
-  }
-
-  return counter;
-}
-
-/* Get the total of all items of a ZIM file, redirects included */
-unsigned int Reader::getGlobalCount() const
-{
-  return zimArchive->getEntryCount();
-}
-
-/* Return the UID of the ZIM file */
-string Reader::getId() const
-{
-  std::ostringstream s;
-  s << zimArchive->getUuid();
-  return s.str();
-}
-
-Entry Reader::getRandomPage() const
-{
-  try {
-    return zimArchive->getRandomEntry();
-  } catch(...) {
-    throw NoEntry();
-  }
-}
-
-Entry Reader::getMainPage() const
-{
-  return zimArchive->getMainEntry();
-}
-
-bool Reader::getFavicon(string& content, string& mimeType) const
-{
-  try {
-    auto item = zimArchive->getIllustrationItem();
-    content = item.getData();
-    mimeType = item.getMimetype();
-    return true;
-  } catch(zim::EntryNotFound& e) {};
-
-  return false;
-}
-
-string Reader::getZimFilePath() const
-{
-  return zimFilePath;
-}
-/* Return a metatag value */
-bool Reader::getMetadata(const string& name, string& value) const
-{
-  try {
-    value = zimArchive->getMetadata(name);
-    return true;
-  } catch(zim::EntryNotFound& e) {
-    return false;
-  }
-}
-
-#define METADATA(NAME) std::string v; getMetadata(NAME, v); return v;
-
-string Reader::getName() const
-{
-  METADATA("Name")
-}
-
-string Reader::getTitle() const
-{
-  string value = zimArchive->getMetadata("Title");
-  if (value.empty()) {
-    value = getLastPathElement(zimFilePath);
-    std::replace(value.begin(), value.end(), '_', ' ');
-    size_t pos = value.find(".zim");
-    value = value.substr(0, pos);
-  }
-  return value;
-}
-
-string Reader::getCreator() const
-{
-  METADATA("Creator")
-}
-
-string Reader::getPublisher() const
-{
-  METADATA("Publisher")
-}
-
-string Reader::getDate() const
-{
-  METADATA("Date")
-}
-
-string Reader::getDescription() const
-{
-  string value;
-  this->getMetadata("Description", value);
-
-  /* Mediawiki Collection tends to use the "Subtitle" name */
-  if (value.empty()) {
-    this->getMetadata("Subtitle", value);
-  }
-
-  return value;
-}
-
-string Reader::getLongDescription() const
-{
-  METADATA("LongDescription")
-}
-
-string Reader::getLanguage() const
-{
-  METADATA("Language")
-}
-
-string Reader::getLicense() const
-{
-  METADATA("License")
-}
-
-string Reader::getTags(bool original) const
-{
-  string tags_str;
-  getMetadata("Tags", tags_str);
-  if (original) {
-    return tags_str;
-  }
-  auto tags = convertTags(tags_str);
-  return join(tags, ";");
-}
-
-
-string Reader::getTagStr(const std::string& tagName) const
-{
-  string tags_str;
-  getMetadata("Tags", tags_str);
-  return getTagValueFromTagList(convertTags(tags_str), tagName);
-}
-
-bool Reader::getTagBool(const std::string& tagName) const
-{
-  return convertStrToBool(getTagStr(tagName));
-}
-
-string Reader::getRelation() const
-{
-  METADATA("Relation")
-}
-
-string Reader::getFlavour() const
-{
-  METADATA("Flavour")
-}
-
-string Reader::getSource() const
-{
-  METADATA("Source")
-}
-
-string Reader::getScraper() const
-{
-  METADATA("Scraper")
-}
-#undef METADATA
-
-string Reader::getOrigId() const
-{
-  string value;
-  this->getMetadata("startfileuid", value);
-  if (value.empty()) {
-    return "";
-  }
-  std::string id = value;
-  std::string origID;
-  std::string temp = "";
-  unsigned int k = 0;
-  char tempArray[16] = "";
-  for (unsigned int i = 0; i < id.size(); i++) {
-    if (id[i] == '\n') {
-      tempArray[k] = atoi(temp.c_str());
-      temp = "";
-      k++;
-    } else {
-      temp += id[i];
-    }
-  }
-  origID = hexUUID(tempArray);
-  return origID;
-}
-
-Entry Reader::getEntryFromPath(const std::string& path) const
-{
-  if (path.empty() || path == "/") {
-    return getMainPage();
-  }
-
-  try {
-    return zimArchive->getEntryByPath(path);
-  } catch (zim::EntryNotFound& e) {
-    throw NoEntry();
-  }
-}
-
-Entry Reader::getEntryFromEncodedPath(const std::string& path) const
-{
-  return getEntryFromPath(urlDecode(path, true));
-}
-
-Entry Reader::getEntryFromTitle(const std::string& title) const
-{
-  try {
-    return zimArchive->getEntryByTitle(title);
-  } catch(zim::EntryNotFound& e) {
-    throw NoEntry();
-  }
-}
-
-bool Reader::pathExists(const string& path) const
-{
-  return zimArchive->hasEntryByPath(path);
-}
-
-/* Does the ZIM file has a fulltext index */
-bool Reader::hasFulltextIndex() const
-{
-  return zimArchive->hasFulltextIndex();
-}
-
-/* Search titles by prefix */
-
-bool Reader::searchSuggestions(const string& prefix,
-                               unsigned int suggestionsCount,
-                               const bool reset)
-{
-  /* Reset the suggestions otherwise check if the suggestions number is less
-   * than the suggestionsCount */
-  if (reset) {
-    this->suggestions.clear();
-    this->suggestionsOffset = this->suggestions.begin();
-  } else {
-    if (this->suggestions.size() > suggestionsCount) {
-      return false;
-    }
-  }
-
-  auto ret =  searchSuggestions(prefix, suggestionsCount, this->suggestions);
-
-  /* Set the cursor to the begining */
-  this->suggestionsOffset = this->suggestions.begin();
-
-  return ret;
-}
-
-
-bool Reader::searchSuggestions(const string& prefix,
-                               unsigned int suggestionsCount,
-                               SuggestionsList_t& results)
-{
-  bool retVal = false;
-
-  /* Return if no prefix */
-  if (prefix.size() == 0) {
-    return false;
-  }
-
-  for (auto& entry: zimArchive->findByTitle(prefix)) {
-    if (results.size() >= suggestionsCount) {
-      break;
-    }
-    /* Extract the interesting part of article title & url */
-    std::string normalizedArticleTitle
-        = kiwix::normalize(entry.getTitle());
-
-    // Get the final path.
-    auto item = entry.getItem(true);
-    std::string articleFinalUrl = item.getPath();
-
-    /* Go through all already found suggestions and skip if this
-       article is already in the suggestions list (with an other
-       title) */
-    bool insert = true;
-    std::vector<SuggestionItem>::iterator suggestionItr;
-    for (suggestionItr = results.begin();
-         suggestionItr != results.end();
-         suggestionItr++) {
-      int result = normalizedArticleTitle.compare((*suggestionItr).getNormalizedTitle());
-      if (result == 0 && articleFinalUrl.compare((*suggestionItr).getPath()) == 0) {
-        insert = false;
-        break;
-      } else if (result < 0) {
-        break;
-      }
-    }
-
-    /* Insert if possible */
-    if (insert) {
-      SuggestionItem suggestion(entry.getTitle(), normalizedArticleTitle, articleFinalUrl);
-      results.insert(suggestionItr, suggestion);
-    }
-
-    /* Suggestions where found */
-    retVal = true;
-  }
-
-  return retVal;
-}
-
-std::vector<std::string> Reader::getTitleVariants(
-    const std::string& title) const
-{
-  std::vector<std::string> variants;
-  variants.push_back(title);
-  variants.push_back(kiwix::ucFirst(title));
-  variants.push_back(kiwix::lcFirst(title));
-  variants.push_back(kiwix::toTitle(title));
-  return variants;
-}
-
-
-bool Reader::searchSuggestionsSmart(const string& prefix,
-                                    unsigned int suggestionsCount)
-{
-  this->suggestions.clear();
-  this->suggestionsOffset = this->suggestions.begin();
-
-  auto ret = searchSuggestionsSmart(prefix, suggestionsCount, this->suggestions);
-
-  this->suggestionsOffset = this->suggestions.begin();
-
-  return ret;
-}
-
-/* Try also a few variations of the prefix to have better results */
-bool Reader::searchSuggestionsSmart(const string& prefix,
-                                    unsigned int suggestionsCount,
-                                    SuggestionsList_t& results)
-{
-  std::vector<std::string> variants = this->getTitleVariants(prefix);
-  bool retVal = false;
-
-  /* Try to search in the title using fulltext search database */
-
-  auto suggestionSearcher = zim::Searcher(*zimArchive);
-  zim::Query suggestionQuery;
-  suggestionQuery.setQuery(prefix, true);
-  auto suggestionSearch = suggestionSearcher.search(suggestionQuery);
-
-  if (suggestionSearch.getEstimatedMatches()) {
-    const auto suggestions = suggestionSearch.getResults(0, suggestionsCount);
-    for (auto current = suggestions.begin();
-         current != suggestions.end();
-         current++) {
-      SuggestionItem suggestion(current.getTitle(), kiwix::normalize(current.getTitle()),
-                                current.getPath(), current.getSnippet());
-      results.push_back(suggestion);
-    }
-    retVal = true;
-  } else {
-    for (std::vector<std::string>::iterator variantsItr = variants.begin();
-         variantsItr != variants.end();
-         variantsItr++) {
-      retVal = this->searchSuggestions(*variantsItr, suggestionsCount, results)
-               || retVal;
-    }
-  }
-
-  return retVal;
-}
-
-/* Get next suggestion */
-bool Reader::getNextSuggestion(string& title)
-{
-  if (this->suggestionsOffset != this->suggestions.end()) {
-    /* title */
-    title = (*(this->suggestionsOffset)).getTitle();
-
-    /* increment the cursor for the next call */
-    this->suggestionsOffset++;
-
-    return true;
-  }
-
-  return false;
-}
-
-bool Reader::getNextSuggestion(string& title, string& url)
-{
-  if (this->suggestionsOffset != this->suggestions.end()) {
-    /* title */
-    title = (*(this->suggestionsOffset)).getTitle();
-    url = (*(this->suggestionsOffset)).getPath();
-
-    /* increment the cursor for the next call */
-    this->suggestionsOffset++;
-
-    return true;
-  }
-
-  return false;
-}
-
-/* Check if the file has as checksum */
-bool Reader::canCheckIntegrity() const
-{
-  return zimArchive->hasChecksum();
-}
-
-/* Return true if corrupted, false otherwise */
-bool Reader::isCorrupted() const
-{
-  try {
-    if (zimArchive->check() == true) {
-      return false;
-    }
-  } catch (exception& e) {
-    cerr << e.what() << endl;
-    return true;
-  }
-
-  return true;
-}
-
-/* Return the file size, works also for splitted files */
-unsigned int Reader::getFileSize() const
-{
-  return zimArchive->getFilesize() / 1024;
-}
-
-}

src/search_renderer.cpp

@@ -21,26 +21,61 @@
 #include <cmath>
 
 #include "search_renderer.h"
-#include "searcher.h"
-#include "reader.h"
 #include "library.h"
 #include "name_mapper.h"
 
+#include "tools/archiveTools.h"
+
 #include <zim/search.h>
 
 #include <mustache.hpp>
-#include "kiwixlib-resources.h"
+#include "libkiwix-resources.h"
+#include "tools/stringTools.h"
 
+#include "server/i18n_utils.h"
 
 namespace kiwix
 {
 
+namespace
+{
+
+ParameterizedMessage searchResultsPageTitleMsg(const std::string& searchPattern)
+{
+  return ParameterizedMessage("search-results-page-title",
+                              {{"SEARCH_PATTERN", searchPattern}}
+  );
+}
+
+ParameterizedMessage searchResultsPageHeaderMsg(const std::string& searchPattern,
+                                                const kainjow::mustache::data& r)
+{
+  if ( r.get("count")->string_value() == "0" ) {
+    return ParameterizedMessage("empty-search-results-page-header",
+                                {{"SEARCH_PATTERN", searchPattern}}
+    );
+  } else {
+    return ParameterizedMessage("search-results-page-header",
+                                {
+                                  {"SEARCH_PATTERN", searchPattern},
+                                  {"START", r.get("startLabel")->string_value()},
+                                  {"END",   r.get("end")  ->string_value()},
+                                  {"COUNT", r.get("count")->string_value()},
+                                }
+    );
+  }
+}
+
+} // unnamed namespace
+
 /* Constructor */
-SearchRenderer::SearchRenderer(Searcher* searcher, NameMapper* mapper)
-    : mp_searcher(searcher),
-      mp_nameMapper(mapper),
+SearchRenderer::SearchRenderer(zim::SearchResultSet srs,
+                      unsigned int start, unsigned int estimatedResultCount)
+    : m_srs(srs),
       protocolPrefix("zim://"),
-      searchProtocolPrefix("search://?")
+      searchProtocolPrefix("search://"),
+      estimatedResultCount(estimatedResultCount),
+      resultStart(start)
 {}
 
 /* Destructor */
@@ -48,12 +83,12 @@ SearchRenderer::~SearchRenderer() = default;
 
 void SearchRenderer::setSearchPattern(const std::string& pattern)
 {
-  this->searchPattern = pattern;
+  searchPattern = pattern;
 }
 
-void SearchRenderer::setSearchContent(const std::string& name)
+void SearchRenderer::setSearchBookQuery(const std::string& bookQuery)
 {
-  this->searchContent = name;
+  searchBookQuery = bookQuery;
 }
 
 void SearchRenderer::setProtocolPrefix(const std::string& prefix)
@@ -66,84 +101,175 @@ void SearchRenderer::setSearchProtocolPrefix(const std::string& prefix)
   this->searchProtocolPrefix = prefix;
 }
 
-std::string SearchRenderer::getHtml()
-{
-  kainjow::mustache::data results{kainjow::mustache::data::type::list};
-
-  mp_searcher->restart_search();
-  Result* p_result = NULL;
-  while ((p_result = mp_searcher->getNextResult())) {
-    kainjow::mustache::data result;
-    result.set("title", p_result->get_title());
-    result.set("url", p_result->get_url());
-    result.set("snippet", p_result->get_snippet());
-    result.set("resultContentId", mp_nameMapper->getNameForId(p_result->get_zimId()));
-
-    if (p_result->get_wordCount() >= 0) {
-      result.set("wordCount", kiwix::beautifyInteger(p_result->get_wordCount()));
-    }
-
-    results.push_back(result);
-    delete p_result;
+std::string extractValueFromQuery(const std::string& query, const std::string& key) {
+  const std::string p = key + "=";
+  const size_t i = query.find(p);
+  if (i == std::string::npos) {
+    return "";
   }
+  std::string r = query.substr(i + p.size());
+  return r.substr(0, r.find("&"));
+}
 
-  // pages
+kainjow::mustache::data buildQueryData
+(
+  const std::string& searchProtocolPrefix,
+  const std::string& pattern,
+  const std::string& bookQuery
+) {
+  kainjow::mustache::data query;
+  query.set("pattern", kiwix::encodeDiples(pattern));
+  std::ostringstream ss;
+  ss << searchProtocolPrefix << "?pattern=" << urlEncode(pattern);
+  ss << "&" << bookQuery;
+  query.set("unpaginatedQuery", ss.str());
+  auto lang = extractValueFromQuery(bookQuery, "books.filter.lang");
+  if(!lang.empty()) {
+    query.set("lang", lang);
+  }
+  return query;
+}
+
+kainjow::mustache::data buildPagination(
+  unsigned int pageLength,
+  unsigned int resultsCount,
+  unsigned int resultsStart
+)
+{
+  assert(pageLength!=0);
+  kainjow::mustache::data pagination;
   kainjow::mustache::data pages{kainjow::mustache::data::type::list};
 
-  auto resultStart = mp_searcher->getResultStart();
-  auto resultEnd = 0U;
-  auto estimatedResultCount = mp_searcher->getEstimatedResultCount();
-  auto currentPage = 0U;
-  auto pageStart = 0U;
-  auto pageEnd = 0U;
-  auto lastPageStart = 0U;
-  if (pageLength) {
-    currentPage = resultStart/pageLength;
-    pageStart = currentPage > 4 ? currentPage-4 : 0;
-    pageEnd = currentPage + 5;
-    if (pageEnd > estimatedResultCount / pageLength) {
-      pageEnd = (estimatedResultCount + pageLength - 1) / pageLength;
+  if (resultsCount == 0) {
+    // Easy case
+    pagination.set("itemsPerPage", to_string(pageLength));
+    pagination.set("hasPages", false);
+    pagination.set("pages", pages);
+    return pagination;
+  }
+
+  // First we want to display pages starting at a multiple of `pageLength`
+  // so, let's calculate the start index of the current page.
+  auto currentPage = resultsStart/pageLength;
+  auto lastPage = ((resultsCount-1)/pageLength);
+  auto lastPageStart = lastPage*pageLength;
+  auto nbPages = lastPage + 1;
+
+  auto firstPageGenerated = currentPage > 4 ? currentPage-4 : 0;
+  auto lastPageGenerated = std::min(currentPage+4, lastPage);
+
+  if (nbPages != 1) {
+    if (firstPageGenerated!=0) {
+      kainjow::mustache::data page;
+      page.set("label", "◀");
+      page.set("start", to_string(0));
+      page.set("current", false);
+      pages.push_back(page);
     }
-    if (estimatedResultCount > pageLength) {
-      lastPageStart = ((estimatedResultCount-1)/pageLength)*pageLength;
+
+    for (auto i=firstPageGenerated; i<=lastPageGenerated; i++) {
+      kainjow::mustache::data page;
+      page.set("label", to_string(i+1));
+      page.set("start", to_string(i*pageLength));
+      page.set("current", bool(i == currentPage));
+      pages.push_back(page);
+    }
+
+    if (lastPageGenerated!=lastPage) {
+      kainjow::mustache::data page;
+      page.set("label", "▶");
+      page.set("start", to_string(lastPageStart));
+      page.set("current", false);
+      pages.push_back(page);
     }
   }
 
-  resultEnd = resultStart+pageLength; //setting result end
+  pagination.set("itemsPerPage", to_string(pageLength));
+  pagination.set("hasPages", firstPageGenerated < lastPageGenerated);
+  pagination.set("pages", pages);
+  return pagination;
+}
 
-  for (unsigned int i = pageStart; i < pageEnd; i++) {
-    kainjow::mustache::data page;
-    page.set("label", to_string(i + 1));
-    page.set("start", to_string(i * pageLength));
-
-    if (i == currentPage) {
-      page.set("selected", true);
+std::string SearchRenderer::renderTemplate(const std::string& tmpl_str, const NameMapper& nameMapper, const Library* library)
+{
+  const std::string absPathPrefix = protocolPrefix;
+  // Build the results list
+  kainjow::mustache::data items{kainjow::mustache::data::type::list};
+  for (auto it = m_srs.begin(); it != m_srs.end(); it++) {
+    kainjow::mustache::data result;
+    const std::string zim_id(it.getZimId());
+    const auto path = nameMapper.getNameForId(zim_id) + "/" + it.getPath();
+    result.set("title", it.getTitle());
+    result.set("absolutePath", absPathPrefix + urlEncode(path));
+    result.set("snippet", it.getSnippet());
+    if (library) {
+      const std::string bookTitle = library->getBookById(zim_id).getTitle();
+      const ParameterizedMessage bookInfoMsg("search-result-book-info",
+          {{"BOOK_TITLE", bookTitle}}
+      );
+      result.set("bookInfo",  bookInfoMsg.getText(userlang)); // for HTML
+      result.set("bookTitle", bookTitle); // for XML
     }
-    pages.push_back(page);
+    if (it.getWordCount() >= 0) {
+      const auto wordCountStr = kiwix::beautifyInteger(it.getWordCount());
+      const ParameterizedMessage wordCountMsg("word-count",
+          {{"COUNT", wordCountStr}}
+      );
+      result.set("wordCountInfo", wordCountMsg.getText(userlang)); // for HTML
+      result.set("wordCount", wordCountStr); // for XML
+    }
+
+    items.push_back(result);
   }
+  kainjow::mustache::data results;
+  results.set("items", items);
+  results.set("count", kiwix::beautifyInteger(estimatedResultCount));
+  results.set("start", kiwix::beautifyInteger(resultStart));
+  results.set("startLabel", kiwix::beautifyInteger(resultStart+1));
+  results.set("end", kiwix::beautifyInteger(std::min(resultStart+pageLength, estimatedResultCount)));
 
-  std::string template_str = RESOURCE::templates::search_result_html;
-  kainjow::mustache::mustache tmpl(template_str);
+  // pagination
+  auto pagination = buildPagination(
+    pageLength,
+    estimatedResultCount,
+    resultStart
+  );
 
-  kainjow::mustache::data allData;
-  allData.set("results", results);
-  allData.set("pages", pages);
-  allData.set("hasResults", estimatedResultCount != 0);
-  allData.set("hasPages", pageStart != pageEnd);
-  allData.set("count", kiwix::beautifyInteger(estimatedResultCount));
-  allData.set("searchPattern", kiwix::encodeDiples(this->searchPattern));
-  allData.set("searchPatternEncoded", urlEncode(this->searchPattern));
-  allData.set("resultStart", to_string(resultStart + 1));
-  allData.set("resultEnd", to_string(min(resultEnd, estimatedResultCount)));
-  allData.set("pageLength", to_string(pageLength));
-  allData.set("resultLastPageStart", to_string(lastPageStart));
-  allData.set("protocolPrefix", this->protocolPrefix);
-  allData.set("searchProtocolPrefix", this->searchProtocolPrefix);
-  allData.set("contentId", this->searchContent);
+  kainjow::mustache::data query = buildQueryData(
+    searchProtocolPrefix,
+    searchPattern,
+    searchBookQuery
+  );
+
+  const auto pageHeaderMsg = searchResultsPageHeaderMsg(searchPattern, results);
+  const kainjow::mustache::object allData{
+    {"PAGE_TITLE", searchResultsPageTitleMsg(searchPattern).getText(userlang)},
+    {"PAGE_HEADER", pageHeaderMsg.getText(userlang)},
+    {"searchProtocolPrefix", searchProtocolPrefix},
+    {"results", results},
+    {"pagination", pagination},
+    {"query", query},
+  };
+
+  kainjow::mustache::mustache tmpl(tmpl_str);
 
   std::stringstream ss;
   tmpl.render(allData, [&ss](const std::string& str) { ss << str; });
+  if (!tmpl.is_valid()) {
+    throw std::runtime_error("Error while rendering search results: " + tmpl.error_message());
+  }
   return ss.str();
 }
 
-}
+std::string SearchRenderer::getHtml(const NameMapper& mapper, const Library* library)
+{
+  return renderTemplate(RESOURCE::templates::search_result_html, mapper, library);
+}
+
+std::string SearchRenderer::getXml(const NameMapper& mapper, const Library* library)
+{
+  return renderTemplate(RESOURCE::templates::search_result_xml, mapper, library);
+}
+
+
+}

src/searcher.cpp

@@ -1,272 +0,0 @@
-/*
- * Copyright 2011 Emmanuel Engelhart <kelson@kiwix.org>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU  General Public License as published by
- * the Free Software Foundation; either version 3 of the License, or
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
- * MA 02110-1301, USA.
- */
-
-
-#include <cmath>
-
-#include "searcher.h"
-#include "reader.h"
-
-#include <zim/search.h>
-
-#include <mustache.hpp>
-#include "kiwixlib-resources.h"
-
-#define MAX_SEARCH_LEN 140
-
-namespace kiwix
-{
-class _Result : public Result
-{
- public:
-  _Result(zim::SearchResultSet::iterator iterator);
-  virtual ~_Result(){};
-
-  virtual std::string get_url();
-  virtual std::string get_title();
-  virtual int get_score();
-  virtual std::string get_snippet();
-  virtual std::string get_content();
-  virtual int get_wordCount();
-  virtual int get_size();
-  virtual std::string get_zimId();
-
- private:
-  zim::SearchResultSet::iterator iterator;
-};
-
-struct SearcherInternal : zim::SearchResultSet {
-  explicit SearcherInternal(const zim::SearchResultSet& srs)
-    : zim::SearchResultSet(srs)
-    , current_iterator(srs.begin())
-  {
-  }
-
-  zim::SearchResultSet::iterator current_iterator;
-};
-
-/* Constructor */
-Searcher::Searcher()
-    : searchPattern(""),
-      estimatedResultCount(0),
-      resultStart(0),
-      resultEnd(0)
-{
-  loadICUExternalTables();
-}
-
-/* Destructor */
-Searcher::~Searcher()
-{
-}
-
-bool Searcher::add_reader(Reader* reader)
-{
-  if (!reader->hasFulltextIndex()) {
-      return false;
-  }
-  this->readers.push_back(reader);
-  return true;
-}
-
-
-Reader* Searcher::get_reader(int readerIndex)
-{
-  return readers.at(readerIndex);
-}
-
-/* Search strings in the database */
-void Searcher::search(const std::string& search,
-                      unsigned int resultStart,
-                      unsigned int resultEnd,
-                      const bool verbose)
-{
-  this->reset();
-
-  if (verbose == true) {
-    cout << "Performing query `" << search << "'" << endl;
-  }
-
-  this->searchPattern = search;
-  this->resultStart = resultStart;
-  this->resultEnd = resultEnd;
-  /* Try to find results */
-  if (resultStart != resultEnd) {
-    /* Perform the search */
-    string unaccentedSearch = removeAccents(search);
-    std::vector<zim::Archive> archives;
-    for (auto current = this->readers.begin(); current != this->readers.end();
-         current++) {
-      if ( (*current)->hasFulltextIndex() ) {
-          archives.push_back(*(*current)->getZimArchive());
-      }
-    }
-    zim::Searcher searcher(archives);
-    zim::Query query;
-    query.setQuery(unaccentedSearch, false);
-    query.setVerbose(verbose);
-    zim::Search search = searcher.search(query);
-    internal.reset(new SearcherInternal(search.getResults(resultStart, resultEnd)));
-    this->estimatedResultCount = search.getEstimatedMatches();
-  }
-
-  return;
-}
-
-
-void Searcher::geo_search(float latitude, float longitude, float distance,
-                          unsigned int resultStart,
-                          unsigned int resultEnd,
-                          const bool verbose)
-{
-  this->reset();
-
-  if (verbose == true) {
-    cout << "Performing geo query `" << distance << "&(" << latitude << ";" << longitude << ")'" << endl;
-  }
-
-  /* Perform the search */
-  std::ostringstream oss;
-  oss << "Articles located less than " << distance << " meters of " << latitude << ";" << longitude;
-  this->searchPattern = oss.str();
-  this->resultStart = resultStart;
-  this->resultEnd = resultEnd;
-
-  /* Try to find results */
-  if (resultStart == resultEnd) {
-    return;
-  }
-
-  std::vector<zim::Archive> archives;
-  for (auto current = this->readers.begin(); current != this->readers.end();
-       current++) {
-    archives.push_back(*(*current)->getZimArchive());
-  }
-  zim::Searcher searcher(archives);
-  zim::Query query;
-  query.setVerbose(verbose);
-  query.setQuery("", false);
-  query.setGeorange(latitude, longitude, distance);
-  zim::Search search = searcher.search(query);
-  internal.reset(new SearcherInternal(search.getResults(resultStart, resultEnd)));
-  this->estimatedResultCount = search.getEstimatedMatches();
-}
-
-
-void Searcher::restart_search()
-{
-  if (internal.get()) {
-    internal->current_iterator = internal->begin();
-  }
-}
-
-Result* Searcher::getNextResult()
-{
-  if (internal.get() &&
-             internal->current_iterator != internal->end()) {
-    Result* result = new _Result(internal->current_iterator);
-    internal->current_iterator++;
-    return result;
-  }
-  return NULL;
-}
-
-/* Reset the results */
-void Searcher::reset()
-{
-  this->estimatedResultCount = 0;
-  this->searchPattern = "";
-  return;
-}
-
-void Searcher::suggestions(std::string& searchPattern, const bool verbose)
-{
-  this->reset();
-
-  if (verbose == true) {
-    cout << "Performing suggestion query `" << searchPattern << "`" << endl;
-  }
-
-  this->searchPattern = searchPattern;
-  this->resultStart = 0;
-  this->resultEnd = 10;
-  string unaccentedSearch = removeAccents(searchPattern);
-
-  std::vector<zim::Archive> archives;
-  for (auto current = this->readers.begin(); current != this->readers.end();
-       current++) {
-    archives.push_back(*(*current)->getZimArchive());
-  }
-  zim::Searcher searcher(archives);
-  zim::Query query;
-  query.setVerbose(verbose);
-  query.setQuery(unaccentedSearch, true);
-  zim::Search search = searcher.search(query);
-  internal.reset(new SearcherInternal(search.getResults(resultStart, resultEnd)));
-  this->estimatedResultCount = search.getEstimatedMatches();
-}
-
-/* Return the result count estimation */
-unsigned int Searcher::getEstimatedResultCount()
-{
-  return this->estimatedResultCount;
-}
-
-_Result::_Result(zim::SearchResultSet::iterator iterator)
-    : iterator(iterator)
-{
-}
-
-std::string _Result::get_url()
-{
-  return iterator.getPath();
-}
-std::string _Result::get_title()
-{
-  return iterator.getTitle();
-}
-int _Result::get_score()
-{
-  return iterator.getScore();
-}
-std::string _Result::get_snippet()
-{
-  return iterator.getSnippet();
-}
-std::string _Result::get_content()
-{
-  return iterator->getItem(true).getData();
-}
-int _Result::get_size()
-{
-  return iterator.getSize();
-}
-int _Result::get_wordCount()
-{
-  return iterator.getWordCount();
-}
-std::string _Result::get_zimId()
-{
-  std::ostringstream s;
-  s << iterator.getZimId();
-  return s.str();
-}
-
-
-}

src/server.cpp

@@ -24,11 +24,28 @@
 
 #include <string>
 
+#include <zim/item.h>
 #include "server/internalServer.h"
 
 namespace kiwix {
 
-Server::Server(Library* library, NameMapper* nameMapper) :
+namespace
+{
+
+std::string makeServerUrl(std::string host, int port, std::string root)
+{
+  const int httpDefaultPort = 80;
+
+  if (port == httpDefaultPort) {
+    return "http://" + host + root;
+  } else {
+    return "http://" + host + ":" + std::to_string(port) + root;
+  }
+}
+
+}  // unnamed namespace
+
+Server::Server(LibraryPtr library, std::shared_ptr<NameMapper> nameMapper) :
   mp_library(library),
   mp_nameMapper(nameMapper),
   mp_server(nullptr)
@@ -45,11 +62,23 @@ bool Server::start() {
     m_port,
     m_root,
     m_nbThreads,
+    m_multizimSearchLimit,
     m_verbose,
     m_withTaskbar,
     m_withLibraryButton,
-    m_blockExternalLinks));
-  return mp_server->start();
+    m_blockExternalLinks,
+    m_ipMode,
+    m_indexTemplateString,
+    m_ipConnectionLimit,
+    m_catalogOnlyMode,
+    m_contentServerUrl));
+  if (mp_server->start()) {
+    // this syncs m_addr of InternalServer and Server as they may diverge
+    m_addr = mp_server->getAddress();
+    return true;
+  } else {
+    return false;
+  }
 }
 
 void Server::stop() {
@@ -62,12 +91,53 @@ void Server::stop() {
 void Server::setRoot(const std::string& root)
 {
   m_root = root;
-  if (m_root[0] != '/') {
-    m_root = "/" + m_root;
-  }
-  if (m_root.back() == '/') {
-    m_root.erase(m_root.size() - 1);
+  while (!m_root.empty() && m_root.back() == '/')
+    m_root.pop_back();
+
+  while (!m_root.empty() && m_root.front() == '/')
+    m_root = m_root.substr(1);
+  m_root = m_root.empty() ? m_root : "/" + m_root;
+}
+
+void Server::setAddress(const std::string& addr)
+{
+  m_addr.addr.clear();
+  m_addr.addr6.clear();
+
+  if (addr.empty()) return;
+
+  if (addr.find(':') != std::string::npos) { // IPv6
+    m_addr.addr6 = (addr[0] == '[') ? addr.substr(1, addr.length() - 2) : addr; // Remove brackets if any
+  } else {
+    m_addr.addr = addr;
   }
 }
 
+int Server::getPort() const
+{
+  return m_port;
+}
+
+IpAddress Server::getAddress() const
+{
+  return m_addr;
+}
+
+IpMode Server::getIpMode() const
+{
+  return mp_server->getIpMode();
+}
+
+std::vector<std::string> Server::getServerAccessUrls() const
+{
+  std::vector<std::string> result;
+  if (!m_addr.addr.empty()) {
+    result.push_back(makeServerUrl(m_addr.addr, m_port, m_root));
+  }
+  if (!m_addr.addr6.empty()) {
+    result.push_back(makeServerUrl("[" + m_addr.addr6 + "]", m_port, m_root));
+  }
+  return result;
+}
+
 }

src/server/etag.cpp

@@ -37,11 +37,11 @@ namespace {
 // into the ETag for ETag::Option opt.
 // IMPORTANT: The characters in all_options must come in sorted order (so that
 // IMPORTANT: isValidOptionsString() works correctly).
-const char all_options[] = "cz";
+const char all_options[] = "Zz";
 
 static_assert(ETag::OPTION_COUNT == sizeof(all_options) - 1, "");
 
-bool isValidServerId(const std::string& s)
+bool isValidETagBody(const std::string& s)
 {
   return !s.empty() && s.find_first_of("\"/") == std::string::npos;
 }
@@ -83,17 +83,17 @@ bool ETag::get_option(Option opt) const
 
 std::string ETag::get_etag() const
 {
-  if ( m_serverId.empty() )
+  if ( m_body.empty() )
     return std::string();
 
-  return "\"" + m_serverId + "/" + m_options + "\"";
+  return "\"" + m_body + "/" + m_options + "\"";
 }
 
-ETag::ETag(const std::string& serverId, const std::string& options)
+ETag::ETag(const std::string& body, const std::string& options)
 {
-  if ( isValidServerId(serverId) && isValidOptionsString(options) )
+  if ( isValidETagBody(body) && isValidOptionsString(options) )
   {
-    m_serverId = serverId;
+    m_body = body;
     m_options = options;
   }
 }
@@ -115,7 +115,7 @@ ETag ETag::parse(std::string s)
   return ETag(s.substr(0, i), s.substr(i+1));
 }
 
-ETag ETag::match(const std::string& etags, const std::string& server_id)
+ETag ETag::match(const std::string& etags, const std::string& body)
 {
   std::istringstream ss(etags);
   std::string etag_str;
@@ -125,7 +125,7 @@ ETag ETag::match(const std::string& etags, const std::string& server_id)
       etag_str.pop_back();
 
     const ETag etag = parse(etag_str);
-    if ( etag && etag.m_serverId == server_id )
+    if ( etag && etag.m_body == body )
       return etag;
   }
 

src/server/etag.h

@@ -28,10 +28,11 @@ namespace kiwix {
 // The ETag string used by Kiwix server (more precisely, its value inside the
 // double quotes) consists of two parts:
 //
-// 1. ServerId - The string obtained on server start up
+// 1. Body    -  A string uniquely identifying the object or state from which
+//               the resource has been obtained.
 //
-// 2. Options -  Zero or more characters encoding the values of some of the
-//               headers of the response
+// 2. Options -  Zero or more characters encoding the type of the ETag and/or
+//               the values of some of the headers of the response
 //
 // The two parts are separated with a slash (/) symbol (which is always present,
 // even when the the options part is empty). Neither portion of a Kiwix ETag
@@ -40,7 +41,7 @@ namespace kiwix {
 //
 //   "abcdefghijklmn/"
 //   "1234567890/z"
-//   "1234567890/cz"
+//   "6f1d19d0-633f-087b-fb55-7ac324ff9baf/Zz"
 //
 // The options part of the Kiwix ETag allows to correctly set the required
 // headers when responding to a conditional If-None-Match request with a 304
@@ -51,7 +52,7 @@ class ETag
 {
   public: // types
     enum Option {
-      CACHEABLE_ENTITY,
+      ZIM_CONTENT,
       COMPRESSED_CONTENT,
       OPTION_COUNT
     };
@@ -59,10 +60,10 @@ class ETag
   public: // functions
     ETag() {}
 
-    void set_server_id(const std::string& id) { m_serverId = id; }
+    void set_body(const std::string& s) { m_body = s; }
     void set_option(Option opt);
 
-    explicit operator bool() const { return !m_serverId.empty(); }
+    explicit operator bool() const { return !m_body.empty(); }
 
     bool get_option(Option opt) const;
     std::string get_etag() const;
@@ -76,7 +77,7 @@ class ETag
     static ETag parse(std::string s);
 
   private: // data
-    std::string m_serverId;
+    std::string m_body;
     std::string m_options;
 };
 

src/server/i18n.cpp

@@ -0,0 +1,205 @@
+/*
+ * Copyright 2022 Veloman Yunkan <veloman.yunkan@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#include "i18n_utils.h"
+
+#include "tools/otherTools.h"
+
+#include <algorithm>
+#include <map>
+
+namespace kiwix
+{
+
+const char* I18nStringTable::get(const std::string& key) const
+{
+  const I18nString* const begin = entries;
+  const I18nString* const end = begin + entryCount;
+  const I18nString* found = std::lower_bound(begin, end, key,
+      [](const I18nString& a, const std::string& k) {
+        return a.key < k;
+  });
+  return (found == end || found->key != key) ? nullptr : found->value;
+}
+
+namespace i18n
+{
+// this data is generated by the i18n resource compiler
+extern const I18nStringTable stringTables[];
+extern const size_t langCount;
+}
+
+namespace
+{
+
+class I18nStringDB
+{
+public: // functions
+  I18nStringDB() {
+    for ( size_t i = 0; i < kiwix::i18n::langCount; ++i ) {
+      const auto& t = kiwix::i18n::stringTables[i];
+      lang2TableMap[t.lang] = &t;
+    }
+    enStrings = lang2TableMap.at("en");
+  };
+
+  std::string get(const std::string& lang, const std::string& key) const {
+    const char* s = getStringsFor(lang)->get(key);
+    if ( s == nullptr ) {
+      s = enStrings->get(key);
+      if ( s == nullptr ) {
+        throw std::runtime_error("Invalid message id");
+      }
+    }
+    return s;
+  }
+
+  size_t getStringCount(const std::string& lang) const {
+    try {
+      return lang2TableMap.at(lang)->entryCount;
+    } catch(const std::out_of_range&) {
+      return 0;
+    }
+  }
+
+private: // functions
+  const I18nStringTable* getStringsFor(const std::string& lang) const {
+    try {
+      return lang2TableMap.at(lang);
+    } catch(const std::out_of_range&) {
+      return enStrings;
+    }
+  }
+
+private: // data
+  std::map<std::string, const I18nStringTable*> lang2TableMap;
+  const I18nStringTable* enStrings;
+};
+
+const I18nStringDB& getStringDb()
+{
+  static const I18nStringDB stringDb;
+  return stringDb;
+}
+
+} // unnamed namespace
+
+std::string getTranslatedString(const std::string& lang, const std::string& key)
+{
+  return getStringDb().get(lang, key);
+}
+
+namespace i18n
+{
+
+std::string expandParameterizedString(const std::string& lang,
+                                      const std::string& key,
+                                      const Parameters& params)
+{
+  kainjow::mustache::object mustacheParams;
+  for( const auto& kv : params ) {
+    mustacheParams[kv.first] = kv.second;
+  }
+  const std::string tmpl = getTranslatedString(lang, key);
+  return render_template(tmpl, mustacheParams);
+}
+
+} // namespace i18n
+
+std::string ParameterizedMessage::getText(const std::string& lang) const
+{
+  return i18n::expandParameterizedString(lang, msgId, params);
+}
+
+namespace
+{
+
+LangPreference parseSingleLanguagePreference(const std::string& s)
+{
+  const size_t langStart = s.find_first_not_of(" \t\n");
+  if ( langStart == std::string::npos ) {
+    return {"", 0};
+  }
+
+  const size_t langEnd = s.find(';', langStart);
+  if ( langEnd == std::string::npos ) {
+    return {s.substr(langStart), 1};
+  }
+
+  const std::string lang = s.substr(langStart, langEnd - langStart);
+  // We don't care about langEnd == langStart which will result in an empty
+  // language name - it will be dismissed by parseUserLanguagePreferences()
+
+  float q = 1.0;
+  int nCharsScanned;
+  if ( 1 == sscanf(s.c_str() + langEnd + 1, "q=%f%n", &q, &nCharsScanned)
+       && langEnd + 1 + nCharsScanned == s.size() ) {
+    return {lang, q};
+  }
+
+  return {"", 0};
+}
+
+} // unnamed namespace
+
+UserLangPreferences parseUserLanguagePreferences(const std::string& s)
+{
+  UserLangPreferences result;
+  std::istringstream iss(s);
+  std::string singleLangPrefStr;
+  while ( std::getline(iss, singleLangPrefStr, ',') )
+  {
+    const auto langPref = parseSingleLanguagePreference(singleLangPrefStr);
+    if ( !langPref.lang.empty() && langPref.preference > 0 ) {
+      result.push_back(langPref);
+    }
+  }
+
+  return result;
+}
+
+std::string selectMostSuitableLanguage(const UserLangPreferences& prefs)
+{
+  if ( prefs.empty() ) {
+    return "en";
+  }
+
+  std::string bestLangSoFar("en");
+  float bestScoreSoFar = 0;
+  const auto& stringDb = getStringDb();
+  for ( const auto& entry : prefs ) {
+    const float score = entry.preference * stringDb.getStringCount(entry.lang);
+    if ( score > bestScoreSoFar ) {
+      bestScoreSoFar = score;
+      bestLangSoFar = entry.lang;
+    }
+  }
+  return bestLangSoFar;
+}
+
+std::string translateBookCategory(const std::string& lang, const std::string& category)
+{
+    try {
+        return getTranslatedString(lang, "book-category." + category);
+    } catch (...) {
+        return category;
+    }
+}
+
+} // namespace kiwix

src/server/i18n_utils.h

@@ -0,0 +1,83 @@
+/*
+ * Copyright 2022 Veloman Yunkan <veloman.yunkan@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#ifndef KIWIX_SERVER_I18N_UTILS
+#define KIWIX_SERVER_I18N_UTILS
+
+#include "i18n.h"
+#include <mustache.hpp>
+
+namespace kiwix
+{
+
+struct I18nString {
+  const char* const key;
+  const char* const value;
+};
+
+struct I18nStringTable {
+  const char* const lang;
+  const size_t entryCount;
+  const I18nString* const entries;
+
+  const char* get(const std::string& key) const;
+};
+
+namespace i18n
+{
+
+class GetTranslatedStringWithMsgId
+{
+  typedef kainjow::mustache::basic_data<std::string> MustacheString;
+  typedef std::pair<std::string, MustacheString> MsgIdAndTranslation;
+
+public:
+  explicit GetTranslatedStringWithMsgId(const std::string& lang) : m_lang(lang) {}
+
+  MsgIdAndTranslation operator()(const std::string& key) const
+  {
+    return {key, getTranslatedString(m_lang, key)};
+  }
+
+  MsgIdAndTranslation operator()(const std::string& key, const Parameters& params) const
+  {
+    return {key, expandParameterizedString(m_lang, key, params)};
+  }
+
+private:
+  const std::string m_lang;
+};
+
+} // namespace i18n
+
+struct LangPreference
+{
+  const std::string lang;
+  const float preference;
+};
+
+typedef std::vector<LangPreference> UserLangPreferences;
+
+UserLangPreferences parseUserLanguagePreferences(const std::string& s);
+
+std::string selectMostSuitableLanguage(const UserLangPreferences& prefs);
+
+} // namespace kiwix
+
+#endif // KIWIX_SERVER_I18N_UTILS

src/server/internalServer.h

@@ -27,6 +27,10 @@ extern "C" {
 
 #include "library.h"
 #include "name_mapper.h"
+#include "tools.h"
+
+#include <zim/search.h>
+#include <zim/suggestion.h>
 
 #include <mustache.hpp>
 
@@ -36,26 +40,77 @@ extern "C" {
 #include "server/request_context.h"
 #include "server/response.h"
 
+#include "tools/concurrent_cache.h"
+
 namespace kiwix {
 
-typedef kainjow::mustache::data MustacheData;
+struct GeoQuery {
+  GeoQuery()
+    : GeoQuery(0, 0, -1)
+  {}
 
-class Entry;
+  GeoQuery(float latitude, float longitude, float distance)
+    : latitude(latitude), longitude(longitude), distance(distance)
+  {}
+  float latitude;
+  float longitude;
+  float distance;
+
+  explicit operator bool() const {
+    return distance >= 0;
+  }
+
+  friend bool operator<(const GeoQuery& l, const GeoQuery& r)
+  {
+      return std::tie(l.latitude, l.longitude, l.distance)
+           < std::tie(r.latitude, r.longitude, r.distance); // keep the same order
+  }
+};
+
+class SearchInfo {
+  public:
+    SearchInfo(const std::string& pattern, GeoQuery geoQuery, const Library::BookIdSet& bookIds, const std::string& bookFilterString);
+
+    zim::Query getZimQuery(bool verbose) const;
+    const Library::BookIdSet& getBookIds() const { return bookIds; }
+
+    friend bool operator<(const SearchInfo& l, const SearchInfo& r)
+    {
+        return std::tie(l.bookIds, l.pattern, l.geoQuery)
+             < std::tie(r.bookIds, r.pattern, r.geoQuery); // keep the same order
+    }
+
+  public: //data
+    std::string pattern;
+    GeoQuery geoQuery;
+    Library::BookIdSet bookIds;
+    std::string bookFilterQuery;
+};
+
+
+typedef kainjow::mustache::data MustacheData;
 class OPDSDumper;
+class LibraryDumper;
 
 class InternalServer {
   public:
-    InternalServer(Library* library,
-                   NameMapper* nameMapper,
-                   std::string addr,
+    InternalServer(LibraryPtr library,
+                   std::shared_ptr<NameMapper> nameMapper,
+                   IpAddress addr,
                    int port,
                    std::string root,
                    int nbThreads,
+                   unsigned int multizimSearchLimit,
                    bool verbose,
                    bool withTaskbar,
                    bool withLibraryButton,
-                   bool blockExternalLinks);
-    virtual ~InternalServer() = default;
+                   bool blockExternalLinks,
+                   IpMode ipMode,
+                   std::string indexTemplateString,
+                   int ipConnectionLimit,
+                   bool catalogOnlyMode,
+                   std::string zimViewerURL);
+    virtual ~InternalServer();
 
     MHD_Result handlerCallback(struct MHD_Connection* connection,
                                const char* url,
@@ -66,55 +121,85 @@ class InternalServer {
                                void** cont_cls);
     bool start();
     void stop();
+    IpAddress getAddress() const { return m_addr; }
+    int getPort() const { return m_port; }
+    IpMode getIpMode() const { return m_ipMode; }
 
   private: // functions
     std::unique_ptr<Response> handle_request(const RequestContext& request);
-    std::unique_ptr<Response> build_redirect(const std::string& bookName, const kiwix::Entry& entry) const;
+    std::unique_ptr<Response> build_redirect(const std::string& bookName, const zim::Item& item) const;
     std::unique_ptr<Response> build_homepage(const RequestContext& request);
+    std::unique_ptr<Response> handle_viewer_settings(const RequestContext& request);
     std::unique_ptr<Response> handle_skin(const RequestContext& request);
     std::unique_ptr<Response> handle_catalog(const RequestContext& request);
     std::unique_ptr<Response> handle_catalog_v2(const RequestContext& request);
     std::unique_ptr<Response> handle_catalog_v2_root(const RequestContext& request);
-    std::unique_ptr<Response> handle_catalog_v2_entries(const RequestContext& request);
+    std::unique_ptr<Response> handle_catalog_v2_entries(const RequestContext& request, bool partial);
+    std::unique_ptr<Response> handle_catalog_v2_complete_entry(const RequestContext& request, const std::string& entryId);
     std::unique_ptr<Response> handle_catalog_v2_categories(const RequestContext& request);
-    std::unique_ptr<Response> handle_meta(const RequestContext& request);
+    std::unique_ptr<Response> handle_no_js(const RequestContext& request);
+    std::unique_ptr<Response> handle_catalog_v2_languages(const RequestContext& request);
+    std::unique_ptr<Response> handle_catalog_v2_illustration(const RequestContext& request);
     std::unique_ptr<Response> handle_search(const RequestContext& request);
+    std::unique_ptr<Response> handle_search_request(const RequestContext& request);
     std::unique_ptr<Response> handle_suggest(const RequestContext& request);
     std::unique_ptr<Response> handle_random(const RequestContext& request);
+    std::unique_ptr<Response> handle_catch(const RequestContext& request);
     std::unique_ptr<Response> handle_captured_external(const RequestContext& request);
     std::unique_ptr<Response> handle_content(const RequestContext& request);
+    std::unique_ptr<Response> handle_raw(const RequestContext& request);
+    std::unique_ptr<Response> handle_locally_customized_resource(const RequestContext& request);
 
     std::vector<std::string> search_catalog(const RequestContext& request,
                                             kiwix::OPDSDumper& opdsDumper);
 
     MustacheData get_default_data() const;
 
-    std::shared_ptr<Reader> get_reader(const std::string& bookName) const;
-    bool etag_not_needed(const RequestContext& r) const;
-    ETag get_matching_if_none_match_etag(const RequestContext& request) const;
+    std::pair<std::string, Library::BookIdSet> selectBooks(const RequestContext& r) const;
+    SearchInfo getSearchInfo(const RequestContext& r) const;
+
+    bool isLocallyCustomizedResource(const std::string& url) const;
+
+    std::string getLibraryId() const;
+
+    std::string getNoJSDownloadPageHTML(const std::string& bookId, const std::string& userLang) const;
+    OPDSDumper getOPDSDumper() const;
+    void setContentAccessUrl(LibraryDumper& libDumper) const;
+
+  private: // types
+    class LockableSuggestionSearcher;
+    typedef ConcurrentCache<SearchInfo, std::shared_ptr<zim::Search>> SearchCache;
+    typedef ConcurrentCache<std::string, std::shared_ptr<LockableSuggestionSearcher>> SuggestionSearcherCache;
 
   private: // data
-    std::string m_addr;
+    IpAddress m_addr;
     int m_port;
-    std::string m_root;
+    std::string m_root;                   // URI-encoded
+    std::string m_rootPrefixOfDecodedURL; // URI-decoded
     int m_nbThreads;
+    unsigned int m_multizimSearchLimit;
     std::atomic_bool m_verbose;
     bool m_withTaskbar;
     bool m_withLibraryButton;
     bool m_blockExternalLinks;
+    IpMode m_ipMode;
+    std::string m_indexTemplateString;
+    int m_ipConnectionLimit;
     struct MHD_Daemon* mp_daemon;
 
-    Library* mp_library;
-    NameMapper* mp_nameMapper;
+    LibraryPtr mp_library;
+    std::shared_ptr<NameMapper> mp_nameMapper;
+
+    SearchCache searchCache;
+    SuggestionSearcherCache suggestionSearcherCache;
 
     std::string m_server_id;
-    std::string m_library_id;
 
-    friend std::unique_ptr<Response> Response::build(const InternalServer& server);
-    friend std::unique_ptr<ContentResponse> ContentResponse::build(const InternalServer& server, const std::string& content, const std::string& mimetype, bool isHomePage);
-    friend std::unique_ptr<Response> ItemResponse::build(const InternalServer& server, const RequestContext& request, const zim::Item& item);
-    friend std::unique_ptr<Response> Response::build_500(const InternalServer& server, const std::string& msg);
+    class CustomizedResources;
+    std::unique_ptr<CustomizedResources> m_customizedResources;
 
+    const bool m_catalogOnlyMode;
+    const std::string m_contentServerUrl;
 };
 
 }

src/server/internalServer_catalog.cpp

@@ -0,0 +1,227 @@
+/*
+ * Copyright 2021 Veloman Yunkan <veloman.yunkan@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#include "internalServer.h"
+
+#include "library.h"
+#include "opds_dumper.h"
+#include "request_context.h"
+#include "response.h"
+#include "tools/otherTools.h"
+#include "libkiwix-resources.h"
+
+#include <mustache.hpp>
+
+#include <string>
+#include <vector>
+
+namespace kiwix {
+
+namespace
+{
+
+enum OPDSResponseKind
+{
+  OPDS_ENTRY,
+  OPDS_NAVIGATION_FEED,
+  OPDS_ACQUISITION_FEED
+};
+
+const std::string opdsMimeType[] = {
+  "application/atom+xml;type=entry;profile=opds-catalog;charset=utf-8",
+  "application/atom+xml;profile=opds-catalog;kind=navigation;charset=utf-8",
+  "application/atom+xml;profile=opds-catalog;kind=acquisition;charset=utf-8"
+};
+
+} // unnamed namespace
+
+OPDSDumper InternalServer::getOPDSDumper() const
+{
+  kiwix::OPDSDumper opdsDumper(mp_library.get(), mp_nameMapper.get());
+  opdsDumper.setRootLocation(m_root);
+  opdsDumper.setLibraryId(getLibraryId());
+  setContentAccessUrl(opdsDumper);
+  return opdsDumper;
+}
+
+std::unique_ptr<Response> InternalServer::handle_catalog(const RequestContext& request)
+{
+  if (m_verbose.load()) {
+    printf("** running handle_catalog");
+  }
+
+  std::string host;
+  std::string url;
+  try {
+    host = request.get_header("Host");
+    url  = request.get_url_part(1);
+  } catch (const std::out_of_range&) {
+    return UrlNotFoundResponse(request);
+  }
+
+  if (url == "v2") {
+    return handle_catalog_v2(request);
+  }
+
+  if (url != "searchdescription.xml" && url != "root.xml" && url != "search") {
+    return UrlNotFoundResponse(request);
+  }
+
+  if (url == "searchdescription.xml") {
+    auto response = ContentResponse::build(RESOURCE::opensearchdescription_xml, get_default_data(), "application/opensearchdescription+xml");
+    return std::move(response);
+  }
+
+  zim::Uuid uuid;
+  kiwix::OPDSDumper opdsDumper = getOPDSDumper();
+  std::vector<std::string> bookIdsToDump;
+  if (url == "root.xml") {
+    uuid = zim::Uuid::generate(host);
+    bookIdsToDump = mp_library->filter(kiwix::Filter().valid(true).local(true).remote(true));
+  } else if (url == "search") {
+    bookIdsToDump = search_catalog(request, opdsDumper);
+    uuid = zim::Uuid::generate();
+  }
+
+  auto response = ContentResponse::build(
+      opdsDumper.dumpOPDSFeed(bookIdsToDump, request.get_query()),
+      opdsMimeType[OPDS_ACQUISITION_FEED]);
+  return std::move(response);
+}
+
+std::unique_ptr<Response> InternalServer::handle_catalog_v2(const RequestContext& request)
+{
+  if (m_verbose.load()) {
+    printf("** running handle_catalog_v2");
+  }
+
+  std::string url;
+  try {
+    url  = request.get_url_part(2);
+  } catch (const std::out_of_range&) {
+    return UrlNotFoundResponse(request);
+  }
+
+  if (url == "root.xml") {
+    return handle_catalog_v2_root(request);
+  } else if (url == "searchdescription.xml") {
+    const std::string endpoint_root = m_root + "/catalog/v2";
+    return ContentResponse::build(
+        RESOURCE::catalog_v2_searchdescription_xml,
+        kainjow::mustache::object({{"endpoint_root", endpoint_root}}),
+        "application/opensearchdescription+xml"
+    );
+  } else if (url == "entry") {
+    const std::string entryId  = request.get_url_part(3);
+    return handle_catalog_v2_complete_entry(request, entryId);
+  } else if (url == "entries") {
+    return handle_catalog_v2_entries(request, /*partial=*/false);
+  } else if (url == "partial_entries") {
+    return handle_catalog_v2_entries(request, /*partial=*/true);
+  } else if (url == "categories") {
+    return handle_catalog_v2_categories(request);
+  } else if (url == "languages") {
+    return handle_catalog_v2_languages(request);
+  } else if (url == "illustration") {
+    return handle_catalog_v2_illustration(request);
+  } else {
+    return UrlNotFoundResponse(request);
+  }
+}
+
+std::unique_ptr<Response> InternalServer::handle_catalog_v2_root(const RequestContext& request)
+{
+  const std::string libraryId = getLibraryId();
+  return ContentResponse::build(
+             RESOURCE::templates::catalog_v2_root_xml,
+             kainjow::mustache::object{
+               {"date", gen_date_str()},
+               {"endpoint_root", m_root + "/catalog/v2"},
+               {"feed_id", gen_uuid(libraryId)},
+               {"all_entries_feed_id", gen_uuid(libraryId + "/entries")},
+               {"partial_entries_feed_id", gen_uuid(libraryId + "/partial_entries")},
+               {"category_list_feed_id", gen_uuid(libraryId + "/categories")},
+               {"language_list_feed_id", gen_uuid(libraryId + "/languages")}
+             },
+             opdsMimeType[OPDS_NAVIGATION_FEED]
+  );
+}
+
+std::unique_ptr<Response> InternalServer::handle_catalog_v2_entries(const RequestContext& request, bool partial)
+{
+  kiwix::OPDSDumper opdsDumper = getOPDSDumper();
+  const auto bookIds = search_catalog(request, opdsDumper);
+  const auto opdsFeed = opdsDumper.dumpOPDSFeedV2(bookIds, request.get_query(), partial);
+  return ContentResponse::build(
+             opdsFeed,
+             opdsMimeType[OPDS_ACQUISITION_FEED]
+  );
+}
+
+std::unique_ptr<Response> InternalServer::handle_catalog_v2_complete_entry(const RequestContext& request, const std::string& entryId)
+{
+  try {
+    mp_library->getBookById(entryId);
+  } catch (const std::out_of_range&) {
+    return UrlNotFoundResponse(request);
+  }
+
+  kiwix::OPDSDumper opdsDumper = getOPDSDumper();
+  const auto opdsFeed = opdsDumper.dumpOPDSCompleteEntry(entryId);
+  return ContentResponse::build(
+             opdsFeed,
+             opdsMimeType[OPDS_ENTRY]
+  );
+}
+
+std::unique_ptr<Response> InternalServer::handle_catalog_v2_categories(const RequestContext& request)
+{
+  kiwix::OPDSDumper opdsDumper = getOPDSDumper();
+  return ContentResponse::build(
+             opdsDumper.categoriesOPDSFeed(),
+             opdsMimeType[OPDS_NAVIGATION_FEED]
+  );
+}
+
+std::unique_ptr<Response> InternalServer::handle_catalog_v2_languages(const RequestContext& request)
+{
+  kiwix::OPDSDumper opdsDumper = getOPDSDumper();
+  return ContentResponse::build(
+             opdsDumper.languagesOPDSFeed(),
+             opdsMimeType[OPDS_NAVIGATION_FEED]
+  );
+}
+
+std::unique_ptr<Response> InternalServer::handle_catalog_v2_illustration(const RequestContext& request)
+{
+  try {
+    const auto bookId  = request.get_url_part(3);
+    auto book = mp_library->getBookByIdThreadSafe(bookId);
+    auto size = request.get_argument<unsigned int>("size");
+    auto illustration = book.getIllustration(size);
+    return ContentResponse::build(
+               illustration->getData(),
+               illustration->mimeType
+    );
+  } catch(...) {
+    return UrlNotFoundResponse(request);
+  }
+}
+
+} // namespace kiwix

src/server/internalServer_catalog_v2.cpp

@@ -1,109 +0,0 @@
-/*
- * Copyright 2021 Veloman Yunkan <veloman.yunkan@gmail.com>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU  General Public License as published by
- * the Free Software Foundation; either version 3 of the License, or
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
- * MA 02110-1301, USA.
- */
-
-#include "internalServer.h"
-
-#include "library.h"
-#include "opds_dumper.h"
-#include "request_context.h"
-#include "response.h"
-#include "tools/otherTools.h"
-#include "kiwixlib-resources.h"
-
-#include <mustache.hpp>
-
-#include <string>
-#include <vector>
-
-namespace kiwix {
-
-std::unique_ptr<Response> InternalServer::handle_catalog_v2(const RequestContext& request)
-{
-  if (m_verbose.load()) {
-    printf("** running handle_catalog_v2");
-  }
-
-  std::string url;
-  try {
-    url  = request.get_url_part(2);
-  } catch (const std::out_of_range&) {
-    return Response::build_404(*this, request, "", "");
-  }
-
-  if (url == "root.xml") {
-    return handle_catalog_v2_root(request);
-  } else if (url == "searchdescription.xml") {
-    const std::string endpoint_root = m_root + "/catalog/v2";
-    return ContentResponse::build(*this,
-        RESOURCE::catalog_v2_searchdescription_xml,
-        kainjow::mustache::object({{"endpoint_root", endpoint_root}}),
-        "application/opensearchdescription+xml"
-    );
-  } else if (url == "entries") {
-    return handle_catalog_v2_entries(request);
-  } else if (url == "categories") {
-    return handle_catalog_v2_categories(request);
-  } else {
-    return Response::build_404(*this, request, "", "");
-  }
-}
-
-std::unique_ptr<Response> InternalServer::handle_catalog_v2_root(const RequestContext& request)
-{
-  return ContentResponse::build(
-             *this,
-             RESOURCE::templates::catalog_v2_root_xml,
-             kainjow::mustache::object{
-               {"date", gen_date_str()},
-               {"endpoint_root", m_root + "/catalog/v2"},
-               {"feed_id", gen_uuid(m_library_id)},
-               {"all_entries_feed_id", gen_uuid(m_library_id + "/entries")},
-               {"category_list_feed_id", gen_uuid(m_library_id + "/categories")}
-             },
-             "application/atom+xml;profile=opds-catalog;kind=navigation"
-  );
-}
-
-std::unique_ptr<Response> InternalServer::handle_catalog_v2_entries(const RequestContext& request)
-{
-  OPDSDumper opdsDumper(mp_library);
-  opdsDumper.setRootLocation(m_root);
-  opdsDumper.setLibraryId(m_library_id);
-  const auto bookIds = search_catalog(request, opdsDumper);
-  const auto opdsFeed = opdsDumper.dumpOPDSFeedV2(bookIds, request.get_query());
-  return ContentResponse::build(
-             *this,
-             opdsFeed,
-             "application/atom+xml;profile=opds-catalog;kind=acquisition"
-  );
-}
-
-std::unique_ptr<Response> InternalServer::handle_catalog_v2_categories(const RequestContext& request)
-{
-  OPDSDumper opdsDumper(mp_library);
-  opdsDumper.setRootLocation(m_root);
-  opdsDumper.setLibraryId(m_library_id);
-  return ContentResponse::build(
-             *this,
-             opdsDumper.categoriesOPDSFeed(mp_library->getBooksCategories()),
-             "application/atom+xml;profile=opds-catalog;kind=navigation"
-  );
-}
-
-} // namespace kiwix

src/server/request_context.cpp

@@ -25,8 +25,10 @@
 #include <sstream>
 #include <cstdio>
 #include <atomic>
+#include <cctype>
 
 #include "tools/stringTools.h"
+#include "i18n_utils.h"
 
 namespace kiwix {
 
@@ -47,67 +49,62 @@ RequestMethod str2RequestMethod(const std::string& method) {
   else                          return RequestMethod::OTHER;
 }
 
-std::string
-fullURL2LocalURL(const std::string& full_url, const std::string& rootLocation)
-{
-  if (rootLocation.empty()) {
-    // nothing special to handle.
-    return full_url;
-  } else if (full_url == rootLocation) {
-    return "/";
-  } else if (full_url.size() > rootLocation.size() &&
-             full_url.substr(0, rootLocation.size()+1) == rootLocation + "/") {
-    return full_url.substr(rootLocation.size());
-  } else {
-    return "";
-  }
-}
-
 } // unnamed namespace
 
-RequestContext::RequestContext(struct MHD_Connection* connection,
-                               std::string rootLocation,
-                               const std::string& _url,
+RequestContext::RequestContext(const std::string& _rootLocation, // URI-encoded
+                               const std::string& unrootedUrl,   // URI-decoded
                                const std::string& _method,
-                               const std::string& version) :
-  full_url(_url),
-  url(fullURL2LocalURL(_url, rootLocation)),
+                               const std::string& version,
+                               const NameValuePairs& headers,
+                               const NameValuePairs& queryArgs) :
+  rootLocation(_rootLocation),
+  url(unrootedUrl),
   method(str2RequestMethod(_method)),
   version(version),
   requestIndex(s_requestIndex++),
-  acceptEncodingDeflate(false),
+  acceptEncodingGzip(false),
   byteRange_()
 {
-  MHD_get_connection_values(connection, MHD_HEADER_KIND, &RequestContext::fill_header, this);
-  MHD_get_connection_values(connection, MHD_GET_ARGUMENT_KIND, &RequestContext::fill_argument, this);
+  for ( const auto& kv : headers ) {
+    add_header(kv.first, kv.second);
+  }
+
+  for ( const auto& kv : queryArgs ) {
+    add_argument(kv.first, kv.second);
+  }
 
   try {
-    acceptEncodingDeflate =
-        (get_header(MHD_HTTP_HEADER_ACCEPT_ENCODING).find("deflate") != std::string::npos);
+    acceptEncodingGzip =
+        (get_header(MHD_HTTP_HEADER_ACCEPT_ENCODING).find("gzip") != std::string::npos);
   } catch (const std::out_of_range&) {}
 
   try {
     byteRange_ = ByteRange::parse(get_header(MHD_HTTP_HEADER_RANGE));
   } catch (const std::out_of_range&) {}
+
+  userlang = determine_user_language();
 }
 
 RequestContext::~RequestContext()
 {}
 
-MHD_Result RequestContext::fill_header(void *__this, enum MHD_ValueKind kind,
-                                       const char *key, const char *value)
+void RequestContext::add_header(const char *key, const char *value)
 {
-  RequestContext *_this = static_cast<RequestContext*>(__this);
-  _this->headers[lcAll(key)] = value;
-  return MHD_YES;
+  this->headers[lcAll(key)] = value;
 }
 
-MHD_Result RequestContext::fill_argument(void *__this, enum MHD_ValueKind kind,
-                                         const char *key, const char* value)
+void RequestContext::add_argument(const char *key, const char* value)
 {
-  RequestContext *_this = static_cast<RequestContext*>(__this);
-  _this->arguments[key] = value == nullptr ? "" : value;
-  return MHD_YES;
+  RequestContext *_this = this;
+  _this->arguments[key].push_back(value == nullptr ? "" : value);
+  if ( ! _this->queryString.empty() ) {
+    _this->queryString += "&";
+  }
+  _this->queryString += urlEncode(key);
+  if ( value ) {
+    _this->queryString += "=";
+    _this->queryString += urlEncode(value);
+  }
 }
 
 void RequestContext::print_debug_info() const {
@@ -121,13 +118,18 @@ void RequestContext::print_debug_info() const {
     printf(" - %s : '%s'\n", it->first.c_str(), it->second.c_str());
   }
   printf("arguments :\n");
-  for (auto it=arguments.begin(); it!=arguments.end(); it++) {
-    printf(" - %s : '%s'\n", it->first.c_str(), it->second.c_str());
+  for (auto& pair:arguments) {
+    printf(" - %s :", pair.first.c_str());
+    bool first = true;
+    for (auto& v: pair.second) {
+      printf("%s %s", first?"":",", v.c_str());
+      first = false;
+    }
+    printf("\n");
   }
   printf("Parsed : \n");
-  printf("full_url: %s\n", full_url.c_str());
   printf("url   : %s\n", url.c_str());
-  printf("acceptEncodingDeflate : %d\n", acceptEncodingDeflate);
+  printf("acceptEncodingGzip : %d\n", acceptEncodingGzip);
   printf("has_range : %d\n", byteRange_.kind() != ByteRange::NONE);
   printf("is_valid_url : %d\n", is_valid_url());
   printf(".............\n");
@@ -163,11 +165,15 @@ std::string RequestContext::get_url_part(int number) const {
 }
 
 std::string RequestContext::get_full_url() const {
-  return full_url;
+  return rootLocation + urlEncode(url);
+}
+
+std::string RequestContext::get_root_path() const {
+  return rootLocation.empty() ? "/" : rootLocation;
 }
 
 bool RequestContext::is_valid_url() const {
-  return !url.empty();
+  return url.empty() || url[0] == '/';
 }
 
 ByteRange RequestContext::get_range() const {
@@ -176,21 +182,37 @@ ByteRange RequestContext::get_range() const {
 
 template<>
 std::string RequestContext::get_argument(const std::string& name) const {
-  return arguments.at(name);
+  return arguments.at(name)[0];
 }
 
 std::string RequestContext::get_header(const std::string& name) const {
   return headers.at(lcAll(name));
 }
 
-std::string RequestContext::get_query() const {
-  std::string q;
-  const char* sep = "";
-  for ( const auto& a : arguments ) {
-    q += sep + a.first + '=' + a.second;
-    sep = "&";
-  }
-  return q;
+std::string RequestContext::get_user_language() const
+{
+  return userlang.lang;
+}
+
+RequestContext::UserLanguage RequestContext::determine_user_language() const
+{
+  try {
+    return {UserLanguage::SelectorKind::QUERY_PARAM, get_argument("userlang")};
+  } catch(const std::out_of_range&) {}
+
+  try {
+    const std::string acceptLanguage = get_header("Accept-Language");
+    const auto userLangPrefs = parseUserLanguagePreferences(acceptLanguage);
+    const auto lang = selectMostSuitableLanguage(userLangPrefs);
+    return {UserLanguage::SelectorKind::ACCEPT_LANGUAGE_HEADER, lang};
+  } catch(const std::out_of_range&) {}
+
+  return {UserLanguage::SelectorKind::DEFAULT, "en"};
+}
+
+std::string RequestContext::get_requested_format() const
+{
+  return get_optional_param<std::string>("format", "html");
 }
 
 }

src/server/request_context.h

@@ -25,9 +25,11 @@
 #include <string>
 #include <sstream>
 #include <map>
+#include <vector>
 #include <stdexcept>
 
 #include "byte_range.h"
+#include "../tools/stringTools.h"
 
 extern "C" {
 #include "microhttpd_wrapper.h"
@@ -53,12 +55,17 @@ class IndexError: public std::runtime_error {};
 
 
 class RequestContext {
+  public: // types
+    typedef std::vector<std::pair<const char*, const char*>> NameValuePairs;
+
   public: // functions
-    RequestContext(struct MHD_Connection* connection,
-                   std::string rootLocation,
-                   const std::string& url,
+    RequestContext(const std::string& rootLocation, // URI-encoded
+                   const std::string& unrootedUrl,  // URI-decoded
                    const std::string& method,
-                   const std::string& version);
+                   const std::string& version,
+                   const NameValuePairs& headers,
+                   const NameValuePairs& queryArgs);
+
     ~RequestContext();
 
     void print_debug_info() const;
@@ -68,10 +75,11 @@ class RequestContext {
     std::string get_header(const std::string& name) const;
     template<typename T=std::string>
     T get_argument(const std::string& name) const {
-        std::istringstream stream(arguments.at(name));
-        T v;
-        stream >> v;
-        return v;
+        return extractFromString<T>(get_argument(name));
+    }
+
+    std::vector<std::string> get_arguments(const std::string& name) const {
+      return arguments.at(name);
     }
 
     template<class T>
@@ -88,28 +96,68 @@ class RequestContext {
     std::string get_url() const;
     std::string get_url_part(int part) const;
     std::string get_full_url() const;
-    std::string get_query() const;
+    std::string get_root_path() const;
+
+    std::string get_query() const { return queryString; }
+
+    template<class F>
+    std::string get_query(F filter, bool mustEncode) const {
+      std::string q;
+      const char* sep = "";
+      auto encode = [=](const std::string& value) { return mustEncode?urlEncode(value):value; };
+      for ( const auto& a : arguments ) {
+        if (!filter(a.first)) {
+          continue;
+        }
+        for (const auto& v: a.second) {
+          q += sep + encode(a.first) + '=' + encode(v);
+          sep = "&";
+        }
+      }
+      return q;
+    }
 
     ByteRange get_range() const;
 
-    bool can_compress() const { return acceptEncodingDeflate; }
+    bool can_compress() const { return acceptEncodingGzip; }
+
+    std::string get_user_language() const;
+    std::string get_requested_format() const;
+
+  private: // types
+    struct UserLanguage
+    {
+      enum SelectorKind
+      {
+        QUERY_PARAM,
+        ACCEPT_LANGUAGE_HEADER,
+        DEFAULT
+      };
+
+      SelectorKind selectedBy;
+      std::string  lang;
+    };
 
   private: // data
-    std::string full_url;
+    std::string rootLocation;
     std::string url;
     RequestMethod method;
     std::string version;
     unsigned long long requestIndex;
 
-    bool acceptEncodingDeflate;
+    bool acceptEncodingGzip;
 
     ByteRange byteRange_;
     std::map<std::string, std::string> headers;
-    std::map<std::string, std::string> arguments;
+    std::map<std::string, std::vector<std::string>> arguments;
+    std::string queryString;
+    UserLanguage userlang;
 
   private: // functions
-    static MHD_Result fill_header(void *, enum MHD_ValueKind, const char*, const char*);
-    static MHD_Result fill_argument(void *, enum MHD_ValueKind, const char*, const char*);
+    UserLanguage determine_user_language() const;
+
+    void add_header(const char* name, const char* value);
+    void add_argument(const char* name, const char* value);
 };
 
 template<> std::string RequestContext::get_argument(const std::string& name) const;

src/server/response.cpp

@@ -20,85 +20,506 @@
 #include "response.h"
 #include "request_context.h"
 #include "internalServer.h"
-#include "kiwixlib-resources.h"
+#include "libkiwix-resources.h"
 
 #include "tools/regexTools.h"
 #include "tools/stringTools.h"
 #include "tools/otherTools.h"
+#include "tools/archiveTools.h"
 
 #include "string.h"
 #include <mustache.hpp>
 #include <zlib.h>
 
+#include <array>
+#include <list>
+#include <map>
+#include <regex>
 
-#define KIWIX_MIN_CONTENT_SIZE_TO_DEFLATE 100
+// This is somehow a magic value.
+// If this value is too small, we will compress (and lost cpu time) too much
+// content.
+// If this value is too big, we will not compress enough content and send too
+// much data.
+// If we assume that MTU is 1500 Bytes it is useless to compress
+// content smaller as the content will be sent in one packet anyway.
+// 1400 Bytes seems to be a common accepted limit.
+#define KIWIX_MIN_CONTENT_SIZE_TO_COMPRESS 1400
 
 namespace kiwix {
 
 namespace
 {
+typedef kainjow::mustache::data MustacheData;
+
 // some utilities
 
 std::string get_mime_type(const zim::Item& item)
 {
   try {
     return item.getMimetype();
-  } catch (exception& e) {
+  } catch (std::exception& e) {
     return "application/octet-stream";
   }
 }
 
 bool is_compressible_mime_type(const std::string& mimeType)
 {
-  return mimeType.find("text/") != string::npos
-      || mimeType.find("application/javascript") != string::npos
-      || mimeType.find("application/atom") != string::npos
-      || mimeType.find("application/opensearchdescription") != string::npos
-      || mimeType.find("application/json") != string::npos;
+  return mimeType.find("text/") != std::string::npos
+      || mimeType.find("application/javascript") != std::string::npos
+      || mimeType.find("application/atom") != std::string::npos
+      || mimeType.find("application/opensearchdescription") != std::string::npos
+      || mimeType.find("application/json") != std::string::npos
+
+      // Web fonts
+      || mimeType.find("application/font-") != std::string::npos
+      || mimeType.find("application/x-font-") != std::string::npos
+      || mimeType.find("application/vnd.ms-fontobject") != std::string::npos
+      || mimeType.find("font/") != std::string::npos;
+}
+
+bool compress(std::string &content) {
+  z_stream strm;
+  strm.zalloc = Z_NULL;
+  strm.zfree = Z_NULL;
+  strm.opaque = Z_NULL;
+
+  auto ret = deflateInit2(&strm, Z_DEFAULT_COMPRESSION, Z_DEFLATED, 31, 8,
+                          Z_DEFAULT_STRATEGY);
+  if (ret != Z_OK) { return false; }
+
+  strm.avail_in = static_cast<decltype(strm.avail_in)>(content.size());
+  strm.next_in =
+      const_cast<Bytef *>(reinterpret_cast<const Bytef *>(content.data()));
+
+  std::string compressed;
+
+  std::array<char, 16384> buff{};
+  do {
+    strm.avail_out = buff.size();
+    strm.next_out = reinterpret_cast<Bytef *>(buff.data());
+    ret = deflate(&strm, Z_FINISH);
+    assert(ret != Z_STREAM_ERROR);
+    compressed.append(buff.data(), buff.size() - strm.avail_out);
+  } while (strm.avail_out == 0);
+
+  assert(ret == Z_STREAM_END);
+  assert(strm.avail_in == 0);
+
+  content.swap(compressed);
+
+  deflateEnd(&strm);
+  return true;
 }
 
 
+const char* getCacheControlHeader(Response::Kind k)
+{
+  switch(k) {
+    case Response::STATIC_RESOURCE: return "max-age=31536000, immutable";
+    case Response::ZIM_CONTENT:     return "max-age=3600, must-revalidate";
+    default:                        return "max-age=0, must-revalidate";
+  }
+}
+
 } // unnamed namespace
 
-Response::Response(bool verbose)
-  : m_verbose(verbose),
-    m_returnCode(MHD_HTTP_OK)
+Response::Response()
+  : m_returnCode(MHD_HTTP_OK)
 {
   add_header(MHD_HTTP_HEADER_ACCESS_CONTROL_ALLOW_ORIGIN, "*");
 }
 
-std::unique_ptr<Response> Response::build(const InternalServer& server)
+void Response::set_kind(Kind k)
 {
-  return std::unique_ptr<Response>(new Response(server.m_verbose.load()));
+  m_kind = k;
+  if ( k == ZIM_CONTENT )
+    m_etag.set_option(ETag::ZIM_CONTENT);
 }
 
-std::unique_ptr<Response> Response::build_304(const InternalServer& server, const ETag& etag)
+std::unique_ptr<Response> Response::build()
 {
-  auto response = Response::build(server);
+  return std::make_unique<Response>();
+}
+
+std::unique_ptr<Response> Response::build_304(const ETag& etag)
+{
+  auto response = Response::build();
   response->set_code(MHD_HTTP_NOT_MODIFIED);
   response->m_etag = etag;
+  if ( etag.get_option(ETag::ZIM_CONTENT) ) {
+    response->set_kind(Response::ZIM_CONTENT);
+  }
   if ( etag.get_option(ETag::COMPRESSED_CONTENT) ) {
     response->add_header(MHD_HTTP_HEADER_VARY, "Accept-Encoding");
   }
   return response;
 }
 
-std::unique_ptr<Response> Response::build_404(const InternalServer& server, const RequestContext& request, const std::string& bookName, const std::string& bookTitle, const std::string& details)
+
+namespace
 {
-  MustacheData results;
-  results.set("url", request.get_full_url());
-  results.set("details", details);
 
-  auto response = ContentResponse::build(server, RESOURCE::templates::_404_html, results, "text/html");
-  response->set_code(MHD_HTTP_NOT_FOUND);
-  response->set_taskbar(bookName, bookTitle);
+// This class was introduced in order to work around the missing support
+// for std::variant (and std::optional) under some of the current build
+// platforms.
+template<class T>
+class Optional
+{
+public: // functions
+    Optional() {}
+    Optional(const T& t) : ptr(new T(t)) {}
+    Optional(const Optional& o) : ptr(o.has_value() ? new T(*o) : nullptr) {}
+    Optional(Optional&& o) : ptr(std::move(o.ptr)) {}
 
-  return std::move(response);
+    Optional& operator=(const Optional& o)
+    {
+      *this = Optional(o);
+      return *this;
+    }
+
+    Optional& operator=(Optional&& o)
+    {
+      ptr = std::move(o.ptr);
+      return *this;
+    }
+
+    bool has_value() const { return ptr.get() != nullptr; }
+    const T& operator*() const { return *ptr; }
+    T& operator*() { return *ptr; }
+
+private: // data
+    std::unique_ptr<T> ptr;
+};
+
+} // unnamed namespace
+
+class ContentResponseBlueprint::Data
+{
+public:
+  typedef std::list<Data>             List;
+  typedef std::map<std::string, Data> Object;
+
+private:
+  // std::variant<std::string, bool, List, Object> data;
+  // XXX: libkiwix is compiled on platforms where std::variant
+  // XXX: is not yet supported. Hence this hack. Only one
+  // XXX: of the below data members is expected to contain a value.
+  Optional<std::string> m_stringValue;
+  Optional<bool>        m_boolValue;
+  Optional<List>        m_listValue;
+  Optional<Object>      m_objectValue;
+
+public:
+  Data() {}
+  Data(const std::string& s) : m_stringValue(s) {}
+  Data(bool b)               : m_boolValue(b)   {}
+  Data(const List& l)        : m_listValue(l)   {}
+  Data(const Object& o)      : m_objectValue(o) {}
+
+  MustacheData toMustache(const std::string& lang) const;
+
+  Data& operator[](const std::string& key)
+  {
+    return (*m_objectValue)[key];
+  }
+
+  void push_back(const Data& d) { (*m_listValue).push_back(d); }
+
+  static Data onlyAsNonEmptyValue(const std::string& s)
+  {
+    return s.empty() ? Data(false) : Data(s);
+  }
+
+  static Data from(const ParameterizedMessage& pmsg)
+  {
+    Object obj;
+    for(const auto& kv : pmsg.getParams()) {
+      obj[kv.first] = kv.second;
+    }
+    return Object{
+              { "msgid", pmsg.getMsgId() },
+              { "params", Data(obj) }
+    };
+  }
+
+  static Data fromMsgId(const std::string& nonParameterizedMsgId)
+  {
+    return from(nonParameterizedMessage(nonParameterizedMsgId));
+  }
+
+  static Data staticMultiParagraphText(const std::string& msgIdPrefix, size_t n)
+  {
+    Object paragraphs;
+    for ( size_t i = 1; i <= n; ++i ) {
+      std::ostringstream oss;
+      oss << "p" << i;
+      const std::string pId = oss.str();
+      paragraphs[pId] = fromMsgId(msgIdPrefix + "." + pId);
+    }
+    return paragraphs;
+  }
+
+  std::string asJSON() const;
+  void dumpJSON(std::ostream& os) const;
+
+private:
+  bool isString() const { return m_stringValue.has_value(); }
+  bool isList()   const { return m_listValue.has_value();   }
+  bool isObject() const { return m_objectValue.has_value(); }
+
+  const std::string& stringValue() const { return *m_stringValue; }
+  bool               boolValue()   const { return *m_boolValue;   }
+  const List&        listValue()   const { return *m_listValue;   }
+  const Object&      objectValue() const { return *m_objectValue; }
+
+  const Data* get(const std::string& key) const
+  {
+    if ( !isObject() )
+      return nullptr;
+
+    const auto& obj = objectValue();
+    const auto it = obj.find(key);
+    return it != obj.end() ? &it->second : nullptr;
+  }
+};
+
+MustacheData ContentResponseBlueprint::Data::toMustache(const std::string& lang) const
+{
+  if ( this->isList() ) {
+    kainjow::mustache::list l;
+    for ( const auto& x : this->listValue() ) {
+      l.push_back(x.toMustache(lang));
+    }
+    return l;
+  } else if ( this->isObject() ) {
+    const Data* msgId = this->get("msgid");
+    const Data* msgParams = this->get("params");
+    if ( msgId && msgId->isString() && msgParams && msgParams->isObject() ) {
+      std::map<std::string, std::string> params;
+      for(const auto& kv : msgParams->objectValue()) {
+        params[kv.first] = kv.second.stringValue();
+      }
+      const ParameterizedMessage msg(msgId->stringValue(), ParameterizedMessage::Parameters(params));
+      return msg.getText(lang);
+    } else {
+      kainjow::mustache::object o;
+      for ( const auto& kv : this->objectValue() ) {
+        o[kv.first] = kv.second.toMustache(lang);
+      }
+      return o;
+    }
+  } else if ( this->isString() ) {
+    return this->stringValue();
+  } else {
+    return this->boolValue();
+  }
 }
 
-std::unique_ptr<Response> Response::build_416(const InternalServer& server, size_t resourceLength)
+void ContentResponseBlueprint::Data::dumpJSON(std::ostream& os) const
 {
-  auto response = Response::build(server);
+  if ( this->isString() ) {
+    os << '"' << escapeForJSON(this->stringValue()) << '"';
+  } else if ( this->isList() ) {
+    const char * sep = " ";
+    os << "[";
+
+    for ( const auto& x :  this->listValue() ) {
+      os << sep;
+      x.dumpJSON(os);
+      sep = ", ";
+    }
+    os << " ]";
+  } else if ( this->isObject() ) {
+    const char * sep = " ";
+    os << "{";
+    for ( const auto& kv : this->objectValue() ) {
+      os << sep << '"' << kv.first << "\" : ";
+      kv.second.dumpJSON(os);
+      sep = ", ";
+    }
+    os << " }";
+  } else {
+    os << (this->boolValue() ? "true" : "false");
+  }
+}
+
+std::string ContentResponseBlueprint::Data::asJSON() const
+{
+  std::ostringstream oss;
+  this->dumpJSON(oss);
+
+  // This JSON is going to be used in HTML inside a <script></script> tag.
+  // If it contains "</script>" (or "</script >") as a substring, then the HTML
+  // parser will be confused. Since for a valid JSON that may happen only inside
+  // a JSON string, we can safely take advantage of the answers to
+  // https://stackoverflow.com/questions/28259389/how-to-put-script-in-a-javascript-string
+  // and work around the issue by inserting an otherwise harmless backslash.
+  return std::regex_replace(oss.str(), std::regex("</script"), "</scr\\ipt");
+}
+
+ContentResponseBlueprint::ContentResponseBlueprint(const RequestContext* request,
+                         int httpStatusCode,
+                         const std::string& mimeType,
+                         const std::string& templateStr,
+                         bool includeKiwixResponseData)
+  : m_request(*request)
+  , m_httpStatusCode(httpStatusCode)
+  , m_mimeType(mimeType)
+  , m_template(templateStr)
+  , m_includeKiwixResponseData(includeKiwixResponseData)
+  , m_data(new Data)
+{}
+
+ContentResponseBlueprint::~ContentResponseBlueprint() = default;
+
+std::unique_ptr<ContentResponse> ContentResponseBlueprint::generateResponseObject() const
+{
+  kainjow::mustache::data d = m_data->toMustache(m_request.get_user_language());
+  if ( m_includeKiwixResponseData ) {
+    d.set("KIWIX_RESPONSE_TEMPLATE", escapeForJSON(m_template, false));
+    d.set("KIWIX_RESPONSE_DATA", m_data->asJSON());
+  }
+  auto r = ContentResponse::build(m_template, d, m_mimeType);
+  r->set_code(m_httpStatusCode);
+  return r;
+}
+
+NewHTTP404Response::NewHTTP404Response(const RequestContext& request,
+                                       const std::string& root,
+                                       const std::string& urlPath)
+  : ContentResponseBlueprint(&request,
+                             MHD_HTTP_NOT_FOUND,
+                             "text/html; charset=utf-8",
+                             RESOURCE::templates::sexy404_html,
+                             /*includeKiwixResponseData=*/true)
+{
+  *this->m_data = Data(Data::Object{
+                    {"root", root },
+                    {"url_path", urlPath},
+                    {"PAGE_TITLE",   Data::fromMsgId("new-404-page-title")},
+                    {"PAGE_HEADING", Data::fromMsgId("new-404-page-heading")},
+                    {"404_img_text", Data::fromMsgId("404-img-text")},
+                    {"path_was_not_found_msg", Data::fromMsgId("path-was-not-found")},
+                    {"advice", Data::staticMultiParagraphText("404-advice", 5)},
+  });
+}
+
+BlockExternalLinkResponse::BlockExternalLinkResponse(const RequestContext& request,
+                                       const std::string& root,
+                                       const std::string& externalUrl)
+  : ContentResponseBlueprint(&request,
+                             MHD_HTTP_OK,
+                             "text/html; charset=utf-8",
+                             RESOURCE::templates::captured_external_html,
+                             /*includeKiwixResponseData=*/true)
+{
+  *this->m_data = Data(Data::Object{
+                    {"root", root },
+                    {"external_link_detected", Data::fromMsgId("external-link-detected") },
+                    {"url", externalUrl },
+                    {"caution_warning", Data::fromMsgId("caution-warning") },
+                    {"external_link_intro", Data::fromMsgId("external-link-intro") },
+                    {"advice", Data::staticMultiParagraphText("external-link-advice", 3)},
+  });
+}
+
+HTTPErrorResponse::HTTPErrorResponse(const RequestContext& request,
+                                     int httpStatusCode,
+                                     const std::string& pageTitleMsgId,
+                                     const std::string& headingMsgId,
+                                     const std::string& cssUrl,
+                                     bool includeKiwixResponseData)
+  : ContentResponseBlueprint(&request,
+                             httpStatusCode,
+                             request.get_requested_format() == "html" ? "text/html; charset=utf-8" : "application/xml; charset=utf-8",
+                             request.get_requested_format() == "html" ? RESOURCE::templates::error_html : RESOURCE::templates::error_xml,
+                             includeKiwixResponseData)
+{
+  Data::List emptyList;
+  *this->m_data = Data(Data::Object{
+                    {"CSS_URL", Data::onlyAsNonEmptyValue(cssUrl) },
+                    {"PAGE_TITLE",   Data::fromMsgId(pageTitleMsgId)},
+                    {"PAGE_HEADING", Data::fromMsgId(headingMsgId)},
+                    {"details", emptyList}
+  });
+}
+
+HTTP404Response::HTTP404Response(const RequestContext& request)
+  : HTTPErrorResponse(request,
+                      MHD_HTTP_NOT_FOUND,
+                      "404-page-title",
+                      "404-page-heading",
+                      std::string(),
+                      /*includeKiwixResponseData=*/true)
+{
+}
+
+UrlNotFoundResponse::UrlNotFoundResponse(const RequestContext& request)
+    : HTTP404Response(request)
+{
+  const std::string requestUrl = urlDecode(m_request.get_full_url(), false);
+  *this += ParameterizedMessage("url-not-found", {{"url", requestUrl}});
+}
+
+HTTPErrorResponse& HTTPErrorResponse::operator+(const ParameterizedMessage& details)
+{
+  (*m_data)["details"].push_back(Data::Object{{"p", Data::from(details)}});
+  return *this;
+}
+
+HTTPErrorResponse& HTTPErrorResponse::operator+=(const ParameterizedMessage& details)
+{
+  // operator+() is already a state-modifying operator (akin to operator+=)
+  return *this + details;
+}
+
+
+HTTP400Response::HTTP400Response(const RequestContext& request)
+  : HTTPErrorResponse(request,
+                      MHD_HTTP_BAD_REQUEST,
+                      "400-page-title",
+                      "400-page-heading",
+                      std::string(),
+                      /*includeKiwixResponseData=*/true)
+{
+  std::string requestUrl = urlDecode(m_request.get_full_url(), false);
+  const auto query = m_request.get_query();
+  if (!query.empty()) {
+    requestUrl += "?" + encodeDiples(query);
+  }
+  *this += ParameterizedMessage("invalid-request", {{"url", requestUrl}});
+}
+
+HTTP500Response::HTTP500Response(const RequestContext& request,
+                                 const std::string& root,
+                                 const std::string& urlPath,
+                                 const std::string& errorText)
+  : ContentResponseBlueprint(&request,
+                             MHD_HTTP_INTERNAL_SERVER_ERROR,
+                             "text/html; charset=utf-8",
+                             RESOURCE::templates::sexy500_html,
+                             /*includeKiwixResponseData=*/true)
+{
+  auto pageParams = Data::Object{
+    {"root", root },
+    {"url_path", urlPath},
+    {"PAGE_TITLE",   Data::fromMsgId("500-page-title")},
+    {"PAGE_HEADING", Data::fromMsgId("500-page-heading")},
+    {"PAGE_TEXT",    Data::fromMsgId("500-page-text")},
+    {"500_img_text", Data::fromMsgId("500-img-text")},
+  };
+
+  if ( !errorText.empty() ) {
+    pageParams["error"] = errorText;
+  }
+
+  *this->m_data = Data(pageParams);
+}
+
+std::unique_ptr<Response> Response::build_416(size_t resourceLength)
+{
+  auto response = Response::build();
 // [FIXME] (compile with recent enough version of libmicrohttpd)
 //  response->set_code(MHD_HTTP_RANGE_NOT_SATISFIABLE);
   response->set_code(416);
@@ -109,21 +530,10 @@ std::unique_ptr<Response> Response::build_416(const InternalServer& server, size
   return response;
 }
 
-std::unique_ptr<Response> Response::build_500(const InternalServer& server, const std::string& msg)
-{
-  MustacheData data;
-  data.set("error", msg);
-  auto content = render_template(RESOURCE::templates::_500_html, data);
-  std::unique_ptr<Response> response (
-      new ContentResponse(server.m_root, true, false, false, false, content, "text/html"));
-  response->set_code(MHD_HTTP_INTERNAL_SERVER_ERROR);
-  return response;
-}
 
-
-std::unique_ptr<Response> Response::build_redirect(const InternalServer& server, const std::string& redirectUrl)
+std::unique_ptr<Response> Response::build_redirect(const std::string& redirectUrl)
 {
-  auto response = Response::build(server);
+  auto response = Response::build();
   response->m_returnCode = MHD_HTTP_FOUND;
   response->add_header(MHD_HTTP_HEADER_LOCATION, redirectUrl);
   return response;
@@ -155,7 +565,7 @@ static ssize_t callback_reader_from_item(void* cls,
 {
   RunningResponse* response = static_cast<RunningResponse*>(cls);
 
-  size_t max_size_to_set = min<size_t>(
+  size_t max_size_to_set = std::min<size_t>(
     max,
     response->item.getSize() - pos - response->range_start);
 
@@ -185,59 +595,12 @@ void print_response_info(int retCode, MHD_Response* response)
 }
 
 
-void ContentResponse::introduce_taskbar()
-{
-  kainjow::mustache::data data;
-  data.set("root", m_root);
-  data.set("content", m_bookName);
-  data.set("hascontent", (!m_bookName.empty() && !m_bookTitle.empty()));
-  data.set("title", m_bookTitle);
-  data.set("withlibrarybutton", m_withLibraryButton);
-  auto head_content = render_template(RESOURCE::templates::head_taskbar_html, data);
-  m_content = prependToFirstOccurence(
-    m_content,
-    "</head[ \\t]*>",
-    head_content);
-
-  auto taskbar_part = render_template(RESOURCE::templates::taskbar_part_html, data);
-  m_content = appendToFirstOccurence(
-    m_content,
-    "<body[^>]*>",
-    taskbar_part);
-}
-
-
-void ContentResponse::inject_externallinks_blocker()
-{
-  kainjow::mustache::data data;
-  data.set("root", m_root);
-  auto script_tag = render_template(RESOURCE::templates::external_blocker_part_html, data);
-  m_content = prependToFirstOccurence(
-    m_content,
-    "</head[ \\t]*>",
-    script_tag);
-}
-
-void ContentResponse::inject_root_link(){
-  m_content = prependToFirstOccurence(
-    m_content,
-    "</head[ \\t]*>",
-    "<link type=\"root\" href=\"" + m_root + "\">");
-}
-
 bool
 ContentResponse::can_compress(const RequestContext& request) const
 {
   return request.can_compress()
       && is_compressible_mime_type(m_mimeType)
-      && (m_content.size() > KIWIX_MIN_CONTENT_SIZE_TO_DEFLATE);
-}
-
-bool
-ContentResponse::contentDecorationAllowed() const
-{
-    return (startsWith(m_mimeType, "text/html")
-         && m_mimeType.find(";raw=true") == std::string::npos);
+      && (m_content.size() > KIWIX_MIN_CONTENT_SIZE_TO_COMPRESS);
 }
 
 MHD_Response*
@@ -250,56 +613,27 @@ Response::create_mhd_response(const RequestContext& request)
 MHD_Response*
 ContentResponse::create_mhd_response(const RequestContext& request)
 {
-  if (contentDecorationAllowed()) {
-    inject_root_link();
-
-    if (m_withTaskbar) {
-      introduce_taskbar();
-    }
-    if (m_blockExternalLinks) {
-      inject_externallinks_blocker();
-    }
-  }
-
-  bool shouldCompress = can_compress(request);
-  if (shouldCompress) {
-    std::vector<Bytef> compr_buffer(compressBound(m_content.size()));
-    uLongf comprLen = compr_buffer.capacity();
-    int err = compress(&compr_buffer[0],
-                       &comprLen,
-                       (const Bytef*)(m_content.data()),
-                       m_content.size());
-    if (err == Z_OK && comprLen > 2 && comprLen < (m_content.size() + 2)) {
-      /* /!\ Internet Explorer has a bug with deflate compression.
-         It can not handle the first two bytes (compression headers)
-         We need to chunk them off (move the content 2bytes)
-         It has no incidence on other browsers
-         See http://www.subbu.org/blog/2008/03/ie7-deflate-or-not and comments */
-      m_content = string((char*)&compr_buffer[2], comprLen - 2);
-      m_etag.set_option(ETag::COMPRESSED_CONTENT);
-    } else {
-      shouldCompress = false;
-    }
-  }
+  const bool isCompressed = can_compress(request) && compress(m_content);
 
   MHD_Response* response = MHD_create_response_from_buffer(
     m_content.size(), const_cast<char*>(m_content.data()), MHD_RESPMEM_MUST_COPY);
 
-  if (shouldCompress) {
+  if (isCompressed) {
+    m_etag.set_option(ETag::COMPRESSED_CONTENT);
     MHD_add_response_header(
         response, MHD_HTTP_HEADER_VARY, "Accept-Encoding");
     MHD_add_response_header(
-        response, MHD_HTTP_HEADER_CONTENT_ENCODING, "deflate");
+        response, MHD_HTTP_HEADER_CONTENT_ENCODING, "gzip");
   }
   return response;
 }
 
-MHD_Result Response::send(const RequestContext& request, MHD_Connection* connection)
+MHD_Result Response::send(const RequestContext& request, bool verbose, MHD_Connection* connection)
 {
   MHD_Response* response = create_mhd_response(request);
 
   MHD_add_response_header(response, MHD_HTTP_HEADER_CACHE_CONTROL,
-    m_etag.get_option(ETag::CACHEABLE_ENTITY) ? "max-age=2723040, public" : "no-cache, no-store, must-revalidate");
+                          getCacheControlHeader(m_kind));
   const std::string etag = m_etag.get_etag();
   if ( ! etag.empty() )
     MHD_add_response_header(response, MHD_HTTP_HEADER_ETAG, etag.c_str());
@@ -310,7 +644,7 @@ MHD_Result Response::send(const RequestContext& request, MHD_Connection* connect
   if (m_returnCode == MHD_HTTP_OK && m_byteRange.kind() == ByteRange::RESOLVED_PARTIAL_CONTENT)
     m_returnCode = MHD_HTTP_PARTIAL_CONTENT;
 
-  if (m_verbose)
+  if (verbose)
     print_response_info(m_returnCode, response);
 
   auto ret = MHD_queue_response(connection, m_returnCode, response);
@@ -318,78 +652,60 @@ MHD_Result Response::send(const RequestContext& request, MHD_Connection* connect
   return ret;
 }
 
-void ContentResponse::set_taskbar(const std::string& bookName, const std::string& bookTitle)
-{
-  m_bookName = bookName;
-  m_bookTitle = bookTitle;
-}
-
-
-ContentResponse::ContentResponse(const std::string& root, bool verbose, bool withTaskbar, bool withLibraryButton, bool blockExternalLinks, const std::string& content, const std::string& mimetype) :
-  Response(verbose),
-  m_root(root),
+ContentResponse::ContentResponse(const std::string& content, const std::string& mimetype) :
+  Response(),
   m_content(content),
-  m_mimeType(mimetype),
-  m_withTaskbar(withTaskbar),
-  m_withLibraryButton(withLibraryButton),
-  m_blockExternalLinks(blockExternalLinks),
-  m_bookName(""),
-  m_bookTitle("")
+  m_mimeType(mimetype)
 {
   add_header(MHD_HTTP_HEADER_CONTENT_TYPE, m_mimeType);
 }
 
-std::unique_ptr<ContentResponse> ContentResponse::build(const InternalServer& server, const std::string& content, const std::string& mimetype, bool isHomePage)
+std::unique_ptr<ContentResponse> ContentResponse::build(
+  const std::string& content,
+  const std::string& mimetype)
 {
-   return std::unique_ptr<ContentResponse>(new ContentResponse(
-        server.m_root,
-        server.m_verbose.load(),
-        server.m_withTaskbar && !isHomePage,
-        server.m_withLibraryButton,
-        server.m_blockExternalLinks,
-        content,
-        mimetype));
+   return std::make_unique<ContentResponse>(content, mimetype);
 }
 
-std::unique_ptr<ContentResponse> ContentResponse::build(const InternalServer& server, const std::string& template_str, kainjow::mustache::data data, const std::string& mimetype, bool isHomePage) {
+std::unique_ptr<ContentResponse> ContentResponse::build(
+  const std::string& template_str,
+  kainjow::mustache::data data,
+  const std::string& mimetype)
+{
   auto content = render_template(template_str, data);
-  return ContentResponse::build(server, content, mimetype, isHomePage);
+  return ContentResponse::build(content, mimetype);
 }
 
-ItemResponse::ItemResponse(bool verbose, const zim::Item& item, const std::string& mimetype, const ByteRange& byterange) :
-  Response(verbose),
+ItemResponse::ItemResponse(const zim::Item& item, const std::string& mimetype, const ByteRange& byterange) :
+  Response(),
   m_item(item),
   m_mimeType(mimetype)
 {
   m_byteRange = byterange;
-  set_cacheable();
+  set_kind(Response::ZIM_CONTENT);
   add_header(MHD_HTTP_HEADER_CONTENT_TYPE, m_mimeType);
 }
 
-std::unique_ptr<Response> ItemResponse::build(const InternalServer& server, const RequestContext& request, const zim::Item& item)
+std::unique_ptr<Response> ItemResponse::build(const RequestContext& request, const zim::Item& item)
 {
   const std::string mimetype = get_mime_type(item);
   auto byteRange = request.get_range().resolve(item.getSize());
   const bool noRange = byteRange.kind() == ByteRange::RESOLVED_FULL_CONTENT;
   if (noRange && is_compressible_mime_type(mimetype)) {
     // Return a contentResponse
-    auto response = ContentResponse::build(server, item.getData(), mimetype);
-    response->set_cacheable();
+    auto response = ContentResponse::build(item.getData(), mimetype);
+    response->set_kind(Response::ZIM_CONTENT);
     response->m_byteRange = byteRange;
     return std::move(response);
   }
 
   if (byteRange.kind() == ByteRange::RESOLVED_UNSATISFIABLE) {
-    auto response = Response::build_416(server, item.getSize());
-    response->set_cacheable();
+    auto response = Response::build_416(item.getSize());
+    response->set_kind(Response::ZIM_CONTENT);
     return response;
   }
 
-  return std::unique_ptr<Response>(new ItemResponse(
-        server.m_verbose.load(),
-        item,
-        mimetype,
-        byteRange));
+  return std::make_unique<ItemResponse>(item, mimetype, byteRange);
 }
 
 MHD_Response*

src/server/response.h

@@ -26,37 +26,47 @@
 
 #include <mustache.hpp>
 #include "byte_range.h"
-#include "entry.h"
 #include "etag.h"
+#include "i18n_utils.h"
+
+#include <zim/item.h>
 
 extern "C" {
 #include "microhttpd_wrapper.h"
 }
 
+namespace zim {
+class Archive;
+} // namespace zim
+
 namespace kiwix {
 
-class InternalServer;
 class RequestContext;
 
-class EntryResponse;
-
 class Response {
   public:
-    Response(bool verbose);
+    enum Kind
+    {
+      STATIC_RESOURCE,
+      ZIM_CONTENT,
+      DYNAMIC_CONTENT
+    };
+
+  public:
+    Response();
     virtual ~Response() = default;
 
-    static std::unique_ptr<Response> build(const InternalServer& server);
-    static std::unique_ptr<Response> build_304(const InternalServer& server, const ETag& etag);
-    static std::unique_ptr<Response> build_404(const InternalServer& server, const RequestContext& request, const std::string& bookName, const std::string& bookTitle, const std::string& details="");
-    static std::unique_ptr<Response> build_416(const InternalServer& server, size_t resourceLength);
-    static std::unique_ptr<Response> build_500(const InternalServer& server, const std::string& msg);
-    static std::unique_ptr<Response> build_redirect(const InternalServer& server, const std::string& redirectUrl);
+    static std::unique_ptr<Response> build();
+    static std::unique_ptr<Response> build_304(const ETag& etag);
+    static std::unique_ptr<Response> build_416(size_t resourceLength);
+    static std::unique_ptr<Response> build_redirect(const std::string& redirectUrl);
 
-    MHD_Result send(const RequestContext& request, MHD_Connection* connection);
+    MHD_Result send(const RequestContext& request, bool verbose, MHD_Connection* connection);
 
     void set_code(int code) { m_returnCode = code; }
-    void set_cacheable() { m_etag.set_option(ETag::CACHEABLE_ENTITY); }
-    void set_server_id(const std::string& id) { m_etag.set_server_id(id); }
+    void set_kind(Kind k);
+    Kind get_kind() const { return m_kind; }
+    void set_etag_body(const std::string& id) { m_etag.set_body(id); }
     void add_header(const std::string& name, const std::string& value) { m_customHeaders[name] = value; }
 
     int getReturnCode() const { return m_returnCode; }
@@ -66,7 +76,7 @@ class Response {
     MHD_Response* create_error_response(const RequestContext& request) const;
 
   protected: // data
-    bool m_verbose;
+    Kind m_kind = DYNAMIC_CONTENT;
     int m_returnCode;
     ByteRange m_byteRange;
     ETag m_etag;
@@ -78,37 +88,110 @@ class Response {
 
 class ContentResponse : public Response {
   public:
-    ContentResponse(const std::string& root, bool verbose, bool withTaskbar, bool withLibraryButton, bool blockExternalLinks, const std::string& content, const std::string& mimetype);
-    static std::unique_ptr<ContentResponse> build(const InternalServer& server, const std::string& content, const std::string& mimetype, bool isHomePage = false);
-    static std::unique_ptr<ContentResponse> build(const InternalServer& server, const std::string& template_str, kainjow::mustache::data data, const std::string& mimetype, bool isHomePage = false);
+    ContentResponse(
+      const std::string& content,
+      const std::string& mimetype);
 
-    void set_taskbar(const std::string& bookName, const std::string& bookTitle);
+    static std::unique_ptr<ContentResponse> build(
+      const std::string& content,
+      const std::string& mimetype);
+
+    static std::unique_ptr<ContentResponse> build(
+      const std::string& template_str,
+      kainjow::mustache::data data,
+      const std::string& mimetype);
+
+    const std::string& getContent() const { return m_content; }
+    const std::string& getMimeType() const { return m_mimeType; }
 
   private:
     MHD_Response* create_mhd_response(const RequestContext& request);
 
-    void introduce_taskbar();
-    void inject_externallinks_blocker();
-    void inject_root_link();
     bool can_compress(const RequestContext& request) const;
-    bool contentDecorationAllowed() const;
 
 
   private:
-    std::string m_root;
     std::string m_content;
     std::string m_mimeType;
-    bool m_withTaskbar;
-    bool m_withLibraryButton;
-    bool m_blockExternalLinks;
-    std::string m_bookName;
-    std::string m_bookTitle;
  };
 
+class ContentResponseBlueprint
+{
+public: // functions
+  ContentResponseBlueprint(const RequestContext* request,
+                           int httpStatusCode,
+                           const std::string& mimeType,
+                           const std::string& templateStr,
+                           bool includeKiwixResponseData = false);
+
+  ~ContentResponseBlueprint();
+
+  operator std::unique_ptr<Response>() const
+  {
+    return generateResponseObject();
+  }
+
+  std::unique_ptr<ContentResponse> generateResponseObject() const;
+
+protected: // types
+  class Data;
+
+protected: //data
+  const RequestContext& m_request;
+  const int m_httpStatusCode;
+  const std::string m_mimeType;
+  const std::string m_template;
+  const bool m_includeKiwixResponseData;
+  std::unique_ptr<Data> m_data;
+};
+
+struct NewHTTP404Response : ContentResponseBlueprint
+{
+  NewHTTP404Response(const RequestContext& request,
+                     const std::string& root,
+                     const std::string& urlPath);
+};
+
+struct HTTPErrorResponse : ContentResponseBlueprint
+{
+  HTTPErrorResponse(const RequestContext& request,
+                    int httpStatusCode,
+                    const std::string& pageTitleMsgId,
+                    const std::string& headingMsgId,
+                    const std::string& cssUrl = "",
+                    bool includeKiwixResponseData = false);
+
+  HTTPErrorResponse& operator+(const ParameterizedMessage& errorDetails);
+  HTTPErrorResponse& operator+=(const ParameterizedMessage& errorDetails);
+};
+
+struct HTTP404Response : HTTPErrorResponse
+{
+  explicit HTTP404Response(const RequestContext& request);
+};
+
+struct UrlNotFoundResponse : HTTP404Response
+{
+  explicit UrlNotFoundResponse(const RequestContext& request);
+};
+
+struct HTTP400Response : HTTPErrorResponse
+{
+  explicit HTTP400Response(const RequestContext& request);
+};
+
+struct HTTP500Response : ContentResponseBlueprint
+{
+  HTTP500Response(const RequestContext& request,
+                  const std::string& root,
+                  const std::string& urlPath,
+                  const std::string& error = "");
+};
+
 class ItemResponse : public Response {
   public:
-    ItemResponse(bool verbose, const zim::Item& item, const std::string& mimetype, const ByteRange& byterange);
-    static std::unique_ptr<Response> build(const InternalServer& server, const RequestContext& request, const zim::Item& item);
+    ItemResponse(const zim::Item& item, const std::string& mimetype, const ByteRange& byterange);
+    static std::unique_ptr<Response> build(const RequestContext& request, const zim::Item& item);
 
   private:
     MHD_Response* create_mhd_response(const RequestContext& request);
@@ -117,6 +200,13 @@ class ItemResponse : public Response {
     std::string m_mimeType;
 };
 
+struct BlockExternalLinkResponse : ContentResponseBlueprint
+{
+  BlockExternalLinkResponse(const RequestContext& request,
+                            const std::string& root,
+                            const std::string& externalUrl);
+};
+
 }
 
 #endif //KIWIXLIB_SERVER_RESPONSE_H

src/spelling_correction.cpp

@@ -0,0 +1,108 @@
+/*
+ * Copyright (C) 2025 Veloman Yunkan
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#include "spelling_correction.h"
+#include "zim/archive.h"
+
+#include <sstream>
+#include <stdexcept>
+
+#include <xapian.h>
+
+namespace kiwix
+{
+
+namespace
+{
+
+std::vector<std::string> getAllTitles(const zim::Archive& a)
+{
+  std::vector<std::string> result;
+  for (const auto& entry : a.iterByPath() ) {
+     result.push_back(entry.getTitle());
+  }
+  return result;
+}
+
+void createXapianDB(std::string path, const zim::Archive& archive)
+{
+  const int flags = Xapian::DB_BACKEND_GLASS|Xapian::DB_CREATE;
+  const auto tmpDbPath = path + ".tmp";
+  Xapian::WritableDatabase db(tmpDbPath, flags);
+  for (const auto& t : getAllTitles(archive)) {
+    db.add_spelling(t);
+  }
+  db.commit();
+  db.compact(path, Xapian::DBCOMPACT_SINGLE_FILE);
+  db.close();
+  std::filesystem::remove_all(tmpDbPath);
+}
+
+std::string spellingsDBPathForZIMArchive(std::filesystem::path cacheDirPath, const zim::Archive& a)
+{
+  // The version of spellings DB must be updated each time an important change
+  // to the implementation is made that renders using the previous version
+  // impossible or undesirable.
+  const char SPELLINGS_DB_VERSION[] = "0.1";
+
+  std::ostringstream filename;
+  filename << a.getUuid() << ".spellingsdb.v" << SPELLINGS_DB_VERSION;
+  return (cacheDirPath / filename.str()).string();
+}
+
+std::unique_ptr<Xapian::Database> openOrCreateXapianDB(std::filesystem::path cacheDirPath, const zim::Archive& archive)
+{
+  const auto path = spellingsDBPathForZIMArchive(cacheDirPath, archive);
+  try
+  {
+    return std::make_unique<Xapian::Database>(path);
+  }
+  catch (const Xapian::DatabaseOpeningError& )
+  {
+    createXapianDB(path, archive);
+    return std::make_unique<Xapian::Database>(path);
+  }
+}
+
+} // unnamed namespace
+
+SpellingsDB::SpellingsDB(const zim::Archive& archive, std::filesystem::path cacheDirPath)
+  : impl_(openOrCreateXapianDB(cacheDirPath, archive))
+{
+}
+
+SpellingsDB::~SpellingsDB()
+{
+}
+
+std::vector<std::string> SpellingsDB::getSpellingCorrections(const std::string& word, uint32_t maxCount) const
+{
+  if ( maxCount > 1 ) {
+    throw std::runtime_error("More than one spelling correction was requested");
+  }
+
+  std::vector<std::string> result;
+  const auto term = impl_->get_spelling_suggestion(word, 3);
+  if ( !term.empty() ) {
+    result.push_back(term);
+  }
+  return result;
+}
+
+} // namespace kiwix

src/tools/archiveTools.cpp

@@ -0,0 +1,128 @@
+/*
+ * Copyright 2021 Maneesh P M <manu.pm55@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#include "archiveTools.h"
+
+#include "tools.h"
+#include "pathTools.h"
+#include "otherTools.h"
+#include "stringTools.h"
+
+#include <zim/error.h>
+#include <zim/item.h>
+
+namespace kiwix
+{
+std::string getMetadata(const zim::Archive& archive, const std::string& name) {
+    try {
+        return archive.getMetadata(name);
+    } catch (zim::EntryNotFound& e) {
+        return "";
+    }
+}
+
+std::string getArchiveTitle(const zim::Archive& archive) {
+  std::string value = getMetadata(archive, "Title");
+  if (value.empty()) {
+    value = getLastPathElement(archive.getFilename());
+    std::replace(value.begin(), value.end(), '_', ' ');
+    size_t pos = value.find(".zim");
+    value = value.substr(0, pos);
+  }
+  return value;
+}
+
+std::string getMetaDescription(const zim::Archive& archive) {
+  std::string value;
+  value = getMetadata(archive, "Description");
+
+  /* Mediawiki Collection tends to use the "Subtitle" name */
+  if (value.empty()) {
+    value = getMetadata(archive, "Subtitle");
+  }
+
+  return value;
+}
+
+std::string getMetaTags(const zim::Archive& archive, bool original) {
+  std::string tags_str = getMetadata(archive, "Tags");
+  if (original) {
+    return tags_str;
+  }
+  auto tags = convertTags(tags_str);
+  return join(tags, ";");
+}
+
+std::string getMetaLanguage(const zim::Archive& archive) {
+  return getMetadata(archive, "Language");
+}
+
+std::string getMetaName(const zim::Archive& archive) {
+  return getMetadata(archive, "Name");
+}
+
+std::string getMetaDate(const zim::Archive& archive) {
+  return getMetadata(archive, "Date");
+}
+
+std::string getMetaCreator(const zim::Archive& archive) {
+  return getMetadata(archive, "Creator");
+}
+
+std::string getMetaPublisher(const zim::Archive& archive) {
+  return getMetadata(archive, "Publisher");
+}
+
+std::string getMetaFlavour(const zim::Archive& archive) {
+  return getMetadata(archive, "Flavour");
+}
+
+bool getArchiveFavicon(const zim::Archive& archive, unsigned size,
+                           std::string& content, std::string& mimeType){
+  try {
+    auto item = archive.getIllustrationItem(size);
+    content = item.getData();
+    mimeType = item.getMimetype();
+    return true;
+  } catch(zim::EntryNotFound& e) {};
+
+  return false;
+}
+
+unsigned int getArchiveFileSize(const zim::Archive& archive) {
+  return archive.getFilesize() / 1024;
+}
+
+zim::Item getFinalItem(const zim::Archive& archive, const zim::Entry& entry)
+{
+  return entry.getItem(true);
+}
+
+zim::Entry getEntryFromPath(const zim::Archive& archive, const std::string& path)
+{
+  try {
+    return archive.getEntryByPath(path);
+  } catch (zim::EntryNotFound& e) {
+    if (path.empty() || path == "/") {
+      return archive.getMainEntry();
+    }
+  }
+  throw zim::EntryNotFound("Cannot find entry for non empty path");
+}
+} // kiwix

src/tools/archiveTools.h

@@ -0,0 +1,56 @@
+/*
+ * Copyright 2021 Maneesh P M <manu.pm55@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#ifndef KIWIX_ARCHIVETOOLS_H
+#define KIWIX_ARCHIVETOOLS_H
+
+#include <zim/archive.h>
+#include <tools/otherTools.h>
+
+/**
+ * This file contains all the functions that would make handling data related to
+ * an archive easier.
+ **/
+
+namespace kiwix
+{
+    std::string getMetadata(const zim::Archive& archive, const std::string& name);
+    std::string getArchiveTitle(const zim::Archive& archive);
+    std::string getMetaDescription(const zim::Archive& archive);
+    std::string getMetaTags(const zim::Archive& archive, bool original = false);
+    std::string getMetaLanguage(const zim::Archive& archive);
+    std::string getMetaName(const zim::Archive& archive);
+    std::string getMetaDate(const zim::Archive& archive);
+    std::string getMetaCreator(const zim::Archive& archive);
+    std::string getMetaPublisher(const zim::Archive& archive);
+    std::string getMetaFlavour(const zim::Archive& archive);
+
+    bool getArchiveFavicon(const zim::Archive& archive, unsigned size,
+                           std::string& content, std::string& mimeType);
+
+    unsigned int getArchiveMediaCount(const zim::Archive& archive);
+    unsigned int getArchiveArticleCount(const zim::Archive& archive);
+    unsigned int getArchiveFileSize(const zim::Archive& archive);
+
+    zim::Item getFinalItem(const zim::Archive& archive, const zim::Entry& entry);
+
+    zim::Entry getEntryFromPath(const zim::Archive& archive, const std::string& path);
+}
+
+#endif