Patch httplib.h for gcc maybe-uninitialized warning.

Recent version of gcc warn about the use `buf` where it may be
uninitialized.
But here, buf is used as a output buffer to initialize it.

I'm not sure we can fix this differently than ignoring the warning.

.codecov.yml

@@ -0,0 +1,12 @@
+codecov:
+  notify:
+    require_ci_to_pass: yes
+
+coverage:
+  status:
+    project:
+      default:
+        threshold: 1%
+
+ignore:
+  - "test"

.github/FUNDING.yml

@@ -0,0 +1,12 @@
+# These are supported funding model platforms
+
+github: kiwix
+patreon: # Replace with a single Patreon username
+open_collective: # Replace with a single Open Collective username
+ko_fi: # Replace with a single Ko-fi username
+tidelift: # Replace with a single Tidelift platform-name/package-name e.g., npm/babel
+community_bridge: # Replace with a single Community Bridge project-name e.g., cloud-foundry
+liberapay: # Replace with a single Liberapay username
+issuehunt: # Replace with a single IssueHunt username
+otechie: # Replace with a single Otechie username
+custom: # https://kiwix.org/support-us/

.github/move.yml

@@ -0,0 +1,27 @@
+# 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/stale.yml

@@ -0,0 +1,15 @@
+daysUntilClose: false
+staleLabel: stale
+
+issues:
+  daysUntilStale: 60
+  markComment: >
+    This issue has been automatically marked as stale because it has not had
+    recent activity. It will be now be reviewed manually. Thank you
+    for your contributions.
+pulls:
+   daysUntilStale: 7
+   markComment: >
+     This pull request has been automatically marked as stale because it has not had
+     recent activity. It will be now be reviewed manually. Thank you
+     for your contributions.

.github/workflows/ci.yml

@@ -0,0 +1,163 @@
+name: CI
+
+on: [push]
+
+jobs:
+  Macos:
+    runs-on: macos-latest
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v1
+      - name: Setup python 3.5
+        uses: actions/setup-python@v1
+        with:
+          python-version: '3.5'
+      - 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 $HOME
+      - name: Compile
+        shell: bash
+        run: |
+          export PKG_CONFIG_PATH=$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
+        env:
+          SKIP_BIG_MEMORY_TEST: 1
+      - name: Publish coverage
+        shell: bash
+        run: |
+          curl https://codecov.io/bash -o codecov.sh
+          bash codecov.sh -n osx_native_dyn -Z
+          rm codecov.sh
+        env:
+          CODECOV_TOKEN: ${{ secrets.CODECOV_TOKEN }}
+
+  Linux:
+    strategy:
+      fail-fast: false
+      matrix:
+        name:
+          - native_static
+          - native_dyn
+          - native_dyn_bionic
+          - android_arm
+          - android_arm64
+          - win32_static
+          - win32_dyn
+        include:
+          - name: native_static
+            target: native_static
+            image_variant: xenial
+            lib_postfix: '/x86_64-linux-gnu'
+          - name: native_dyn
+            target: native_dyn
+            image_variant: xenial
+            lib_postfix: '/x86_64-linux-gnu'
+          - name: native_dyn_bionic
+            target: native_dyn
+            image_variant: bionic
+            lib_postfix: '/x86_64-linux-gnu'
+          - name: android_arm
+            target: android_arm
+            image_variant: xenial
+            lib_postfix: '/x86_64-linux-gnu'
+          - name: android_arm64
+            target: android_arm64
+            image_variant: xenial
+            lib_postfix: '/x86_64-linux-gnu'
+          - name: win32_static
+            target: win32_static
+            image_variant: f31
+            lib_postfix: '64'
+          - name: win32_dyn
+            target: win32_dyn
+            image_variant: f31
+            lib_postfix: '64'
+    env:
+      HOME: /home/runner
+    runs-on: ubuntu-latest
+    container:
+      image: "kiwix/kiwix-build_ci:${{matrix.image_variant}}-26"
+    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
+    - name: Compile
+      shell: bash
+      run: |
+        meson --version
+        if [[ "${{matrix.target}}" =~ .*_dyn ]]; then
+          MESON_OPTION="--default-library=shared"
+        else
+          MESON_OPTION="--default-library=static"
+        fi
+        if [[ "${{matrix.target}}" =~ native_.* ]]; then
+          MESON_OPTION="$MESON_OPTION -Db_coverage=true"
+        else
+          MESON_OPTION="$MESON_OPTION --cross-file $HOME/BUILD_${{matrix.target}}/meson_cross_file.txt"
+        fi
+        if [[ "${{matrix.target}}" =~ android_.* ]]; then
+          MESON_OPTION="$MESON_OPTION -Dandroid=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"
+    - name: Test
+      if: startsWith(matrix.target, 'native_')
+      shell: bash
+      run: |
+        cd $HOME/libkiwix/build
+        meson test --verbose
+        ninja coverage
+      env:
+        LD_LIBRARY_PATH: "/home/runner/BUILD_${{matrix.target}}/INSTALL/lib:/home/runner/BUILD_${{matrix.target}}/INSTALL/lib${{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 }}

.github/workflows/package.yml

@@ -0,0 +1,87 @@
+name: Packages
+on: [push, pull_request]
+
+jobs:
+  build-deb:
+    runs-on: ubuntu-latest
+    strategy:
+      fail-fast: false
+      matrix:
+        distro: [ubuntu-hirsute, ubuntu-groovy, ubuntu-focal, ubuntu-bionic]
+    steps:
+      - uses: actions/checkout@v2
+
+      # Determine which PPA we should upload to
+      - name: PPA
+        id: ppa
+        run: |
+          if [[ $REF == refs/tags* ]]
+          then
+            echo "::set-output name=ppa::kiwixteam/release"
+          else
+            echo "::set-output name=ppa::kiwixteam/dev"
+          fi
+        env:
+          REF: ${{ github.ref }}
+
+      - uses: legoktm/gh-action-auto-dch@master
+        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
+        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
+        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
+        with:
+          name: Packages for ${{ matrix.distro }}
+          path: output
+
+      - uses: legoktm/gh-action-dput@master
+        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-')
+        with:
+          gpg_key: ${{ secrets.LAUNCHPAD_GPG }}
+          repository: ppa:kiwixteam/dev
+          packages: output/*_source.changes
+
+      - uses: legoktm/gh-action-dput@master
+        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-')
+        with:
+          gpg_key: ${{ secrets.LAUNCHPAD_GPG }}
+          repository: ppa:kiwixteam/release
+          packages: output/*_source.changes
+

.gitignore

@@ -0,0 +1,7 @@
+.idea/
+*.swp
+subprojects/googletest-release*
+*.class
+build/
+.vscode/
+builddir/

.travis.yml

@@ -1,36 +0,0 @@
-language: cpp
-dist: trusty
-sudo: false
-cache: ccache
-before_install:
-- PATH=$PATH:$HOME/bin
-install: travis/install_deps.sh
-script: travis/compile.sh
-env:
-  matrix:
-    - PLATFORM="native_static"
-    - PLATFORM="native_dyn"
-    - PLATFORM="win32_static"
-    - PLATFORM="win32_dyn"
-    - PLATFORM="android_arm"
-    - PLATFORM="android_arm64"
-addons:
-  apt:
-    packages:
-      - cmake
-      - python3-pip
-      - libbz2-dev
-      - ccache
-      - zlib1g-dev
-      - uuid-dev
-      - libctpp2-dev
-      - ctpp2-utils
-      - libmicrohttpd-dev
-      - g++-mingw-w64-i686
-      - gcc-mingw-w64-i686
-      - gcc-mingw-w64-base
-      - mingw-w64-tools
-matrix:
-  include:
-    - env: PLATFORM="native_dyn"
-      os: osx

COPYING

@@ -77,7 +77,7 @@ modification follow.
 
   "Copyright" also means copyright-like laws that apply to other kinds of
 works, such as semiconductor masks.
- 
+
   "The Program" refers to any copyrightable work licensed under this
 License.  Each licensee is addressed as "you".  "Licensees" and
 "recipients" may be individuals or organizations.
@@ -510,7 +510,7 @@ actual knowledge that, but for the patent license, your conveying the
 covered work in a country, or your recipient's use of the covered work
 in a country, would infringe one or more identifiable patents in that
 country that you have reason to believe are valid.
-  
+
   If, pursuant to or in connection with a single transaction or
 arrangement, you convey, or propagate by procuring conveyance of, a
 covered work, and grant a patent license to some of the parties

ChangeLog

@@ -1,3 +1,283 @@
+libkiwix 10.0.0
+===============
+
+ * ...
+
+kiwix-lib 9.4.1
+===============
+
+ * Fix `M/Counter` parsing.
+ * [SERVER] Adjust body padding-top for taskbar
+ * Fix potential crash when stoping a server not started.
+ * Various fix in build system and the CI.
+
+kiwix-lib 9.4.0
+===============
+
+ * [SERVER] Make the headers handling case insensitive.
+ * [SERVER] Make server answer 204 http status code for empty search
+ * [PACKAGING] Made CI build deb packages.
+ * [SERVER] Add a way to prevent taskbar and external link bloquer at article
+   level.
+ * Fix meson file to be compatible with meson 0.45
+ * [SERVER] Update search requests to use pageStart/pageLength instead of
+   pageStart/pageEnd arguments.
+ * [SERVER] Set a fixed favicon size in the main page.
+ * [SERVER] Refactor the response system code to better handling future new
+   libzim api.
+ * Fix segmentation fault around exchange with aria2 process making
+   kiwix-desktop crash at exit.
+
+kiwix-lib 9.3.1
+===============
+
+ * Fix handling of samba path on windows.
+ * Do not include `kiwix_config.h` in public header.
+ * Fix compilation with libmicrohttpd v0.97.1
+ * Increase default test timeout to 160seconds/test.
+ * Add automatic debian packaging.
+ * Use non-minified version of jquery-ui.js
+ * Pass `-latomic` compile option for sh4 architecture.
+ * Make mesion install `kiwix-compile-resources` man page.
+
+kiwix-lib 9.3.0
+===============
+
+ * Add a thread safe method to search suggestions.
+   Old methods are now deprecated.
+
+kiwix-lib 9.2.3
+===============
+
+ * Add test on byte-range
+ * Fix compilation on bionic and windows.
+ * Allow building using debian packaged kainjow-mustache
+ * Pass `-latomic` compile option for architectures that need it
+
+kiwix-lib 9.2.2
+===============
+
+ * Fix handling on empty content in byte range management (wrong assert)
+
+kiwix-lib 9.2.1
+===============
+
+ * Fix support of byte range request.
+
+kiwix-lib 9.2
+=============
+
+ * Add tests
+ * Refactoring server code.
+ * [SERVER] Add HEAD, Etag and If-None-Match support.
+ * [SERVER] Compress opds catalog answers.
+
+kiwix-lib 9.1.2
+===============
+
+ * Do not use the pathToSave if it is empty.
+
+kiwix-lib 9.1.1
+===============
+
+ * Fix the detection of the dataDirectory on windows.
+
+kiwix-lib 9.1.0
+===============
+
+ * [JAVA] Add a method to get the size of an article.
+ * Add a new method to the libray to get the book by path.
+ * Add a option to make the server blocks external link.
+   Links are intercepted by js an redirected to a "portail" page.
+ * [ODPS] Correctly handle book's articleCount and mediaCount.
+
+kiwix-lib 9.0.1
+===============
+
+ * [JAVA] Use a long to store the offset of an article in the zim file instead
+    of an int.
+
+kiwix-lib 9.0.0
+===============
+
+ * [OPDS] Correctly set the id of the OPDS stream.
+ * [OPDS] Do not try to filter the catalog if no filter field is given in the
+   request.
+ * [WINDOWS] Correctly convert path to wide chars when opening the library.xml
+ * [LIBRARY] Remove the function the read file using a native path.
+   All path must be utf8, no need to pass a native path along the utf8 path.
+ * [TEST] Fix tests using the main function of gtest instead of custom one.
+ * [CI] Move to github CI instead of Travis.
+ * The `Book::update` method always update the book's fields. Even if they are
+   not empty.
+ * [JAVA] Add wrapping around the library manager (opds parsing)
+ * [ARIA2] Add api option to start download with option (destination folder)
+ * [OPDS] Fixes about opds parsing, generation (missing attributes)
+   and requesting (server)
+ * Add methods on `Book` to get specific tag values (was on `Reader` only)
+ * Add flavour attribute to `Book`
+ * Fix opensearch description.
+ * Trust the given library.xml (by default) instead of reading the value from
+   the zim files.
+ * [OPDS] Be able to filter the content by name or size.
+ * [WINDOWS] Fix launching subcommand when there is spaces in the path.
+
+kiwix-lib 8.2.2
+===============
+
+ * Improve a few compilation scripts
+
+kiwix-lib 8.2.1
+===============
+
+ * Reintroduce kiwix-serve taskbar
+
+kiwix-lib 8.2.0
+===============
+
+ * More debug information if aria2c command fails
+ * Allow to set kiwix-serve port
+ * Better (dead) bookmarks mgmt
+
+kiwix-lib 8.1.0
+===============
+
+ * Fix pathTools manipulation.
+ * Add missing implementation of getArticleCount and getMediaCount on android.
+ * Correctly convert windows path to utf8.
+ * Add code coverage in the CI
+
+kiwix-lib 8.0.1
+===============
+
+ * Fix join function
+
+kiwix-lib 8.0.0
+===============
+
+ * Add new methods to get all (and new) metadata from the zim file.
+ * Add methods to get the value of a specific tag.
+ * [API Change] Convert tags value to the new convention.
+ * [API Change] Rename `getMatatag` method to `getMetadata`
+ * [ABI Change] Correctly detect executable path in appimage.
+
+kiwix-lib 7.0.0
+===============
+
+ * [API break] Add a argument to kiwix-serve to specify the library to use.
+
+kiwix-lib 6.0.4
+===============
+
+ * Fix HTML rendering of the search result if there is no result.
+ * Do not crash at html rendering if request ask for 0 results (start == end)
+ * Correctly find the executable path if we are using AppImage
+
+kiwix-lib 6.0.3
+===============
+
+ * force one column suggestion in kiwix-serve suggestions
+ * fix fulltext search link in suggestions
+ * UI fixes in kiwix-serve rendering
+
+kiwix-lib 6.0.2
+===============
+
+ * Correctly set the groupId in the pom file.
+
+kiwix-lib 6.0.1
+===============
+
+ * Generate the pom file for android/maven
+
+kiwix-lib 6.0.0
+===============
+
+ * Move the server code in kiwix-lib (from kiwix-serve).
+ * Add unit test on regex functions.
+ * Fix computerAbsolutePath (thread safe, memory leak).
+ * Correctly set the book's path as valid if we construct the book from a
+   reader.
+ * [JNI] Add a method to know if a article is a redirection.
+ * Do not embed the gtest dependency.
+ * [JNI] Add a constructor to JNIKiwixString.
+ * Change order of search of the favicon urls.
+ * Clean a lot of unecessary includes in headers. (potential "API break")
+
+kiwix-lib 5.2.0
+===============
+
+ * kiwix-serve integration (as a seperated process).
+ * Fix crash in the suggestion search.
+ * Better API to filter the library books.
+ * New kiwix-lib application for android. (.aar)
+ * Use ReLinker to link with libkiwix.so in android.
+ * Correctly set the verbosity of zim search.
+
+kiwix-lib 5.1.0
+===============
+
+ * Add function to pause, resume and stop downloads.
+ * Add zim's tags in the opds stream.
+ * Addapt to new libzim 5.0.0 API.
+
+kiwix-lib 5.0.0
+===============
+
+ * Remove error message when trying to open an wrong zim file.
+ * Rewrite `makeTmpDirectory` to not use uuid functions on windows.
+ * [API break] Remove `getNetworkInterfaces` and `getBestPublicIp`.
+ * Remove rpath
+ * Detect infinite (and too long) redirect loops.
+
+kiwix-lib 4.1.0
+===============
+
+ * Allow the library to be filtered by tags.
+ * Fix language mapping.
+ * Update README about mustache dependency.
+
+kiwix-lib 4.0.1
+===============
+
+ * Fix "maybe uninitialize variable" issue.
+ * Ensure path are stored correctly (absolute path) in the library.
+ * [CI] Use the new deps archive xz
+
+kiwix-lib 4.0.0
+===============
+
+ * [API break] Remove support for external index.
+ * Move to the mustache templating system instead of ctpp2.
+ * Make meson.build works for meson>=0.43.0
+ * [API break] Move the basic tools from the `common` directory to `tools`.
+
+kiwix-lib 3.1.1
+===============
+
+ * The OPDS feed book's date must be the date of the book, not the date of the
+ feed generation.
+ * Convert the standard opds date to our format (YYYY-MM-DD)
+ * Remove duplicate language attribute in the libxml dumper.
+ * Create the datadirectory to not fail to write a file in a non-existent
+   directory
+
+kiwix-lib 3.1.0
+===============
+
+ * Add a method to get the favicon url of book (if available).
+ * Move dump code of library.xml in a specific class.
+ * Add a first support to bookmarks
+
+kiwix-lib 3.0.3
+===============
+
+ * Add the 'en' language to the mapping alpha2-code ('en') to alpha3-code
+ ('eng').
+ * Correctly write the 'ArticleCount' and 'MediaCount' in the library.xml.
+ * Correctly fill the book size for the zim file size.
+ * Fix launch of aria2c.
+
 kiwix-lib 3.0.2
 ===============
 

README.md

@@ -1,44 +1,53 @@
-Kiwix library
-=============
+Libkiwix
+========
 
-The Kiwix library provides the Kiwix software core. It contains the
-code shared by all Kiwix ports (Windows, Linux, OSX, Android, ...).
+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, ...).
+
+[![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)
+[![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)
+[![License: GPL v3](https://img.shields.io/badge/License-GPLv3-blue.svg)](https://www.gnu.org/licenses/gpl-3.0)
 
 Disclaimer
 ----------
 
 This document assumes you have a little knowledge about software
 compilation. If you experience difficulties with the dependencies or
-with the Kiwix libary compilation itself, we recommend to have a look
-to [kiwix-build](https://github.com/kiwix/kiwix-build).
+with the Libkiwix compilation itself, we recommend to have a look to
+[kiwix-build](https://github.com/kiwix/kiwix-build).
 
 Preamble
 --------
 
-Although the Kiwix library can be (cross-)compiled on/for many
-sytems, 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 releases of Ubuntu and Fedora.
+Although the Libkiwix can be (cross-)compiled on/for many sytems, 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
+releases of Ubuntu and Fedora.
 
 Dependencies
 ------------
 
-The Kiwix library relies on many third parts software libraries. They
-are prerequisites to the Kiwix library compilation. Following
-libraries need to be available:
+The Libkiwix relies on many third party software libraries. They are
+prerequisites to the Libkiwix compilation. Following libraries need to
+be available:
+* [ICU](https://site.icu-project.org/) (package `libicu-dev` on Ubuntu)
+* [ZIM](https://openzim.org/) (package `libzim-dev` on Ubuntu)
+* [Pugixml](https://pugixml.org/) (package `libpugixml-dev` on Ubuntu)
+* [Mustache](https://github.com/kainjow/Mustache) (Just copy the
+header `mustache.hpp` somewhere it can be found by the compiler and/or
+set CPPFLAGS with correct `-I` option). Use Mustache version 4.1 or above.
+* [Libcurl](https://curl.se/libcurl) (`libcurl4-gnutls-dev`, `libcurl4-nss-dev` or `libcurl4-openssl-dev` on Ubuntu)
+* [Microhttpd](https://www.gnu.org/software/libmicrohttpd) (package `libmicrohttpd-dev` on Ubuntu)
+* [Zlib](https://zlib.net/) (package `zlib1g-dev` on Ubuntu)
 
-* ICU ................................... http://site.icu-project.org/
-(package libicu-dev on Ubuntu)
-* ZIM ........................................ http://www.openzim.org/
-(package libzim-dev on Ubuntu)
-* Pugixml ........................................ http://pugixml.org/
-(package libpugixml-dev on Ubuntu)
-* ctpp2 ........................................ http://ctpp.havoc.ru/
-(package libctpp2-dev on Ubuntu)
-* Xapian ......................................... https://xapian.org/
-(package libxapian-dev on Ubuntu)
-* libaria2 .................................. https://aria2.github.io/
-(no package on Ubuntu)
+To test the code:
+* [Google Test](https://github.com/google/googletest) (package `googletest` on Ubuntu)
+
+The following dependency needs to be available at runtime:
+* [Aria2](https://aria2.github.io/) (package `aria2` on Ubuntu)
 
 These dependencies may or may not be packaged by your operating
 system. They may also be packaged but only in an older version. The
@@ -47,75 +56,79 @@ In the worse case, you will have to download and compile bleeding edge
 version by hand.
 
 If you want to install these dependencies locally, then use the
-kiwix-lib directory as install prefix.
-
-If you compile ctpp2 from source and want to compile the Kiwix library
-statically then you will probably need to rename ctpp2 static library
-from ctpp2-st.a to ctpp2.a.
+`libkiwix` directory as install prefix.
 
 Environment
 -------------
 
-The Kiwix library builds using [Meson](http://mesonbuild.com/) version
-0.39 or higher. Meson relies itself on Ninja, pkg-config and few other
+The Libkiwix builds using [Meson](https://mesonbuild.com/) version
+0.45 or higher. Meson relies itself on Ninja, pkg-config and few other
 compilation tools.
 
 Install first the few common compilation tools:
-* Meson
-* Ninja
-* Pkg-config
+* [Meson](https://mesonbuild.com/)
+* [Ninja](https://ninja-build.org/)
+* [pkg-config](https://www.freedesktop.org/wiki/Software/pkg-config/)
 
 These tools should be packaged if you use a cutting edge operating
-system. If not, have a look to the "Troubleshooting" section.
+system. If not, have a look to the [Troubleshooting](#Troubleshooting)
+section.
 
 Compilation
 -----------
 
-Once all dependencies are installed, you can compile the Kiwix library
+Once all dependencies are installed, you can compile the Libkiwix
 with:
-```
+```bash
 meson . build
 ninja -C build
 ```
 
 By default, it will compile dynamic linked libraries. All binary files
-will be created in the "build" directory created automatically by
+will be created in the `build` directory created automatically by
 Meson. If you want statically linked libraries, you can add
 `--default-library=static` option to the Meson command.
 
 Depending of you system, `ninja` may be called `ninja-build`.
 
+Testing
+-------
+
+To run the automated tests:
+```bash
+cd build
+meson test
+```
+
 Installation
 ------------
 
-If you want to install the Kiwix library and the headers you just have
+If you want to install the Libkiwix and the headers you just have
 compiled on your system, here we go:
-
-```
+```bash
 ninja -C build install
 ```
 
-You might need to run the command as root (or using 'sudo'), depending
+You might need to run the command as root (or using `sudo`), depending
 where you want to install the libraries. After the installation
-succeeded, you may need to run ldconfig (as root).
+succeeded, you may need to run `ldconfig` (as `root`).
 
 Uninstallation
 ------------
 
 If you want to uninstall the Kiwix library:
-
-```
+```bash
 ninja -C build uninstall
 ```
 
-Like for the installation, you might need to run the command as root
-(or using 'sudo').
+Like for the installation, you might need to run the command as `root`
+(or using `sudo`).
 
 Troubleshooting
 ---------------
 
 If you need to install Meson "manually":
-```
+```bash
 virtualenv -p python3 ./ # Create virtualenv
 source bin/activate      # Activate the virtualenv
 pip3 install meson       # Install Meson
@@ -123,7 +136,7 @@ hash -r                  # Refresh bash paths
 ```
 
 If you need to install Ninja "manually":
-```
+```bash
 git clone git://github.com/ninja-build/ninja.git
 cd ninja
 git checkout release
@@ -133,6 +146,10 @@ 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
@@ -142,4 +159,5 @@ repository.
 License
 -------
 
-GPLv3 or later, see COPYING for more details.
+[GPLv3](https://www.gnu.org/licenses/gpl-3.0) or later, see
+[COPYING](COPYING) for more details.

android-kiwix-lib-publisher/.gitignore

@@ -0,0 +1,13 @@
+*.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

@@ -0,0 +1,25 @@
+// 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

@@ -0,0 +1,15 @@
+# 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

@@ -0,0 +1,6 @@
+#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

@@ -0,0 +1,172 @@
+#!/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

@@ -0,0 +1,84 @@
+@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

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

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

@@ -0,0 +1,64 @@
+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

@@ -0,0 +1,21 @@
+# 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

@@ -0,0 +1,10 @@
+<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

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

debian/changelog

@@ -0,0 +1,5 @@
+libkiwix (0.0.0) unstable; urgency=medium
+
+  * Initial release
+
+ -- Kunal Mehta <legoktm@debian.org>  Wed, 08 Jul 2020 18:12:32 -0700

debian/control

@@ -0,0 +1,46 @@
+Source: libkiwix
+Priority: optional
+Maintainer: Kiwix team <kiwix@kiwix.org>
+Build-Depends: debhelper-compat (= 13),
+ meson,
+ pkg-config,
+ libzim-dev (>= 6.1.8),
+ libcurl4-gnutls-dev,
+ libicu-dev,
+ libgtest-dev,
+ libkainjow-mustache-dev,
+ liblzma-dev,
+ libmicrohttpd-dev,
+ libpugixml-dev,
+ zlib1g-dev
+Standards-Version: 4.5.0
+Section: libs
+Homepage: https://github.com/kiwix/libkiwix
+Rules-Requires-Root: no
+
+Package: libkiwix-dev
+Section: libdevel
+Architecture: any
+Multi-Arch: same
+Depends: libkiwix10 (= ${binary:Version}), ${misc:Depends}, python3,
+ libzim-dev (>= 6.0.0),
+ libicu-dev,
+ libpugixml-dev,
+ libcurl4-gnutls-dev,
+ libmicrohttpd-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
+ Kiwix ports (Windows, Linux, OSX, Android, etc.).
+ .
+ This package contains development files.
+
+Package: libkiwix10
+Architecture: any
+Multi-Arch: same
+Depends: ${shlibs:Depends}, ${misc:Depends}, aria2
+Conflicts: libkiwix0, libkiwix3, libkiwix9
+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
+ Kiwix ports (Windows, Linux, OSX, Android, etc.).

debian/copyright

@@ -0,0 +1 @@
+See COPYING in the repository root.

debian/libkiwix-dev.install

@@ -0,0 +1,4 @@
+usr/include
+usr/lib/*/libkiwix.so
+usr/lib/*/pkgconfig
+usr/bin

debian/libkiwix-dev.manpages

@@ -0,0 +1 @@
+usr/share/man/man1/kiwix-compile-resources.1*

debian/libkiwix10.install

@@ -0,0 +1 @@
+usr/lib/*/libkiwix.so.*

debian/rules

@@ -0,0 +1,8 @@
+#!/usr/bin/make -f
+export DEB_BUILD_MAINT_OPTIONS = hardening=+all
+
+%:
+	dh $@ --buildsystem=meson
+
+override_dh_auto_test:
+	dh_auto_test -- -t 3

debian/source/format

@@ -0,0 +1 @@
+3.0 (native)

include/book.h

@@ -28,7 +28,6 @@ class xml_node;
 
 namespace kiwix
 {
-enum supportedIndexType { UNKNOWN, XAPIAN };
 
 class OPDSDumper;
 class Reader;
@@ -46,14 +45,12 @@ class Book
   void update(const Reader& reader);
   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();
+  std::string getHumanReadableIdFromPath() const;
 
   bool readOnly() const { return m_readOnly; }
   const std::string& getId() const { return m_id; }
   const std::string& getPath() const { return m_path; }
   bool isPathValid() const { return m_pathValid; }
-  const std::string& getIndexPath() const { return m_indexPath; }
-  const supportedIndexType& getIndexType() const { return m_indexType; }
   const std::string& getTitle() const { return m_title; }
   const std::string& getDescription() const { return m_description; }
   const std::string& getLanguage() const { return m_language; }
@@ -62,12 +59,17 @@ class Book
   const std::string& getDate() const { return m_date; }
   const std::string& getUrl() const { return m_url; }
   const std::string& getName() const { return m_name; }
+  std::string getCategory() const;
   const std::string& getTags() const { return m_tags; }
+  std::string getTagStr(const std::string& tagName) const;
+  bool getTagBool(const std::string& tagName) const;
+  const std::string& getFlavour() const { return m_flavour; }
   const std::string& getOrigId() const { return m_origId; }
   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; }
   const std::string& getDownloadId() const { return m_downloadId; }
 
@@ -75,8 +77,6 @@ class Book
   void setId(const std::string& id) { m_id = id; }
   void setPath(const std::string& path);
   void setPathValid(bool valid) { m_pathValid = valid; }
-  void setIndexPath(const std::string& indexPath);
-  void setIndexType(supportedIndexType indexType) { m_indexType = indexType;}
   void setTitle(const std::string& title) { m_title = title; }
   void setDescription(const std::string& description) { m_description = description; }
   void setLanguage(const std::string& language) { m_language = language; }
@@ -85,6 +85,7 @@ class Book
   void setDate(const std::string& date) { m_date = date; }
   void setUrl(const std::string& url) { m_url = url; }
   void setName(const std::string& name) { m_name = name; }
+  void setFlavour(const std::string& flavour) { m_flavour = flavour; }
   void setTags(const std::string& tags) { m_tags = tags; }
   void setOrigId(const std::string& origId) { m_origId = origId; }
   void setArticleCount(uint64_t articleCount) { m_articleCount = articleCount; }
@@ -94,27 +95,30 @@ class Book
   void setFaviconMimeType(const std::string& faviconMimeType) { m_faviconMimeType = faviconMimeType; }
   void setDownloadId(const std::string& downloadId) { m_downloadId = downloadId; }
 
+ private:
+  std::string getCategoryFromTags() const;
+
  protected:
   std::string m_id;
   std::string m_downloadId;
   std::string m_path;
-  bool m_pathValid;
-  std::string m_indexPath;
-  supportedIndexType m_indexType;
+  bool m_pathValid = false;
   std::string m_title;
   std::string m_description;
+  std::string m_category;
   std::string m_language;
   std::string m_creator;
   std::string m_publisher;
   std::string m_date;
   std::string m_url;
   std::string m_name;
+  std::string m_flavour;
   std::string m_tags;
   std::string m_origId;
-  uint64_t m_articleCount;
-  uint64_t m_mediaCount;
-  bool m_readOnly;
-  uint64_t m_size;
+  uint64_t m_articleCount = 0;
+  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;

include/bookmark.h

@@ -0,0 +1,68 @@
+/*
+ * Copyright 2018 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_BOOKMARK_H
+#define KIWIX_BOOKMARK_H
+
+#include <string>
+
+namespace pugi {
+class xml_node;
+}
+
+namespace kiwix
+{
+
+/**
+ * A class to store information about a bookmark (an article in a book)
+ */
+class Bookmark
+{
+ public:
+  Bookmark();
+  ~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& getUrl() const { return m_url; }
+  const std::string& getTitle() const { return m_title; }
+  const std::string& getLanguage() const { return m_language; }
+  const std::string& getDate() const { return m_date; }
+
+  void setBookId(const std::string& bookId) { m_bookId = bookId; }
+  void setBookTitle(const std::string& bookTitle) { m_bookTitle = bookTitle; }
+  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; }
+  void setDate(const std::string& date) { m_date = date; }
+
+ protected:
+  std::string m_bookId;
+  std::string m_bookTitle;
+  std::string m_url;
+  std::string m_title;
+  std::string m_language;
+  std::string m_date;
+};
+
+}
+
+#endif

include/common/pathTools.h

@@ -1,64 +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_PATHTOOLS_H
-#define KIWIX_PATHTOOLS_H
-
-#include <fcntl.h>
-#include <limits.h>
-#include <stdio.h>
-#include <string.h>
-#include <sys/stat.h>
-#include <sys/types.h>
-#include <fstream>
-#include <ios>
-#include <iostream>
-#include <sstream>
-#include <string>
-#include <vector>
-
-#ifdef _WIN32
-#include <direct.h>
-#endif
-
-#include "stringTools.h"
-
-using namespace std;
-
-bool isRelativePath(const string& path);
-string computeAbsolutePath(const string path, const string relativePath);
-string computeRelativePath(const string path, const string absolutePath);
-string removeLastPathElement(const string path,
-                             const bool removePreSeparator = false,
-                             const bool removePostSeparator = false);
-string appendToDirectory(const string& directoryPath, const string& filename);
-
-unsigned int getFileSize(const string& path);
-string getFileSizeAsString(const string& path);
-string getFileContent(const string& path);
-bool fileExists(const string& path);
-bool makeDirectory(const string& path);
-string makeTmpDirectory();
-bool copyFile(const string& sourcePath, const string& destPath);
-string getLastPathElement(const string& path);
-string getExecutablePath();
-string getCurrentDirectory();
-string getDataDirectory();
-bool writeTextFile(const string& path, const string& content);
-#endif

include/ctpp2/CTPP2VMStringLoader.hpp

@@ -1,79 +0,0 @@
-/*
- * Copyright 2013 Renaud Gaudin <reg@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 _CTPP2_VM_STRING_LOADER_HPP__
-#define _CTPP2_VM_STRING_LOADER_HPP__ 1
-
-#include <ctpp2/CTPP2VMLoader.hpp>
-#include <ctpp2/CTPP2Util.hpp>
-#include <ctpp2/CTPP2Exception.hpp>
-#include <ctpp2/CTPP2VMExecutable.hpp>
-#include <ctpp2/CTPP2VMInstruction.hpp>
-#include <ctpp2/CTPP2VMMemoryCore.hpp>
-
-#include <sys/types.h>
-#include <sys/stat.h>
-
-#include <errno.h>
-#include <stdio.h>
-#include <stdlib.h>
-
-#include <iostream>
-#include <string>
-
-/**
-  @file VMStringLoader.hpp
-  @brief Load program core from file
-*/
-
-namespace CTPP // C++ Template Engine
-{
-// FWD
-struct VMExecutable;
-
-/**
-  @class VMStringLoader CTPP2VMStringLoader.hpp <CTPP2VMStringLoader.hpp>
-  @brief Load program core from file
-*/
-class CTPP2DECL VMStringLoader:
-  public VMLoader
-{
-public:
-	/**
-	*/
-	VMStringLoader(CCHAR_P rawContent, size_t rawContentSize);
-	/**
-	  @brief Get ready-to-run program
-	*/
-	const VMMemoryCore * GetCore() const;
-
-	/**
-	  @brief A destructor
-	*/
-	~VMStringLoader() throw();
-private:
-	/** Program core             */
-	VMExecutable  * oCore;
-	/** Ready-to-run program     */
-	VMMemoryCore  * pVMMemoryCore;
-};
-
-} // namespace CTPP
-#endif // _CTPP2_VM_STRING_LOADER_HPP__
-// End.

include/downloader.h

@@ -23,8 +23,8 @@
 #include <string>
 #include <vector>
 #include <map>
-#include <pthread.h>
 #include <memory>
+#include <stdexcept>
 
 namespace kiwix
 {
@@ -54,6 +54,9 @@ class Download {
       m_status(K_UNKNOWN),
       m_did(did) {};
   void updateStatus(bool follow=false);
+  void pauseDownload();
+  void resumeDownload();
+  void cancelDownload();
   StatusResult getStatus()          { return m_status; }
   std::string  getDid()             { return m_did; }
   std::string  getFollowedBy()      { return m_followedBy; }
@@ -89,7 +92,7 @@ class Downloader
 
   void close();
 
-  Download* startDownload(const std::string& uri);
+  Download* startDownload(const std::string& uri, const std::vector<std::pair<std::string, std::string>>& options = {});
   Download* getDownload(const std::string& did);
 
   size_t getNbDownload() { return m_knownDownloads.size(); }

include/entry.h

@@ -1,5 +1,5 @@
 /*
- * Copyright 2018 Matthieu Gautier <mgautier@kymeria.fr>
+ * 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
@@ -21,9 +21,11 @@
 #define KIWIX_ENTRY_H
 
 #include <stdio.h>
-#include <zim/article.h>
+#include <zim/entry.h>
+#include <zim/item.h>
 #include <exception>
 #include <string>
+
 #include "common.h"
 
 using namespace std;
@@ -40,19 +42,12 @@ class NoEntry : public std::exception {};
 class Entry
 {
   public:
-    /**
-     * Default constructor.
-     *
-     * Construct an invalid entry.
-     */
-    Entry() = default;
-
     /**
      * Construct an entry making reference to an zim article.
      *
      * @param article a zim::Article object
      */
-    Entry(zim::Article article);
+    Entry(zim::Entry entry);
     virtual ~Entry() = default;
 
     /**
@@ -62,15 +57,15 @@ class Entry
      *
      * @return the path of the entry.
      */
-    std::string getPath() const;
-    
+    std::string getPath() const { return entry.getPath(); }
+
     /**
      * Get the title of the entry.
      *
      * @return the title of the entry.
      */
-    std::string getTitle() const;
-        
+    std::string getTitle() const { return entry.getTitle(); }
+
     /**
      * Get the content of the entry.
      *
@@ -79,8 +74,8 @@ class Entry
      *
      * @return the content of the entry.
      */
-    std::string getContent() const;
-        
+    std::string getContent() const { return entry.getItem().getData(); }
+
     /**
      * Get the blob of the entry.
      *
@@ -89,8 +84,8 @@ class Entry
      * @param offset The starting offset of the blob.
      * @return the blob of the entry.
      */
-    zim::Blob   getBlob(offset_type offset = 0) const;
-        
+    zim::Blob   getBlob(offset_type offset = 0) const { return entry.getItem().getData(offset); }
+
     /**
      * Get the blob of the entry.
      *
@@ -100,15 +95,15 @@ class Entry
      * @param size The size of the blob.
      * @return the blob of the entry.
      */
-    zim::Blob   getBlob(offset_type offset, size_type size) const;
-        
+    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
-     * kiwix-lib/libzim.
+     * libkiwix/libzim.
      *
      * @return A pair specifying where to read the content.
      *         The string is the real file to read (may be different that .zim
@@ -116,8 +111,8 @@ class Entry
      *         The offset is the offset to read in the file.
      *         Return <"",0> if is not possible to read directly.
      */
-    std::pair<std::string, offset_type> getDirectAccessInfo() const;
-        
+    zim::Item::DirectAccessInfo getDirectAccessInfo() const { return entry.getItem().getDirectAccessInformation(); }
+
     /**
      * Get the size of the entry.
      *
@@ -131,8 +126,8 @@ class Entry
      * @return the mime_type of the entry.
      */
     std::string getMimetype() const;
-    
-    
+
+
     /**
      * Get if the entry is a redirect entry.
      *
@@ -173,17 +168,14 @@ class Entry
     Entry getFinalEntry() const;
 
     /**
-     * Convert the entry to a boolean value.
+     * Get the zim entry wrapped by this (kiwix) entry
      *
-     * @return True if the entry is valid.
+     * @return the zim entry
      */
-    explicit operator bool() const { return good(); }
+    const zim::Entry& getZimEntry() const { return entry; }
 
   private:
-    zim::Article article;
-    mutable zim::Article final_article;
-
-    bool good() const { return article.good(); }
+    zim::Entry entry;
 };
 
 }

include/kiwixserve.h

@@ -0,0 +1,30 @@
+#ifndef KIWIXLIB_KIWIX_SERVE_H_
+#define KIWIXLIB_KIWIX_SERVE_H_
+
+#include <memory>
+#include <string>
+
+class Subprocess;
+namespace kiwix {
+
+class KiwixServe
+{
+  public:
+    KiwixServe(const std::string& libraryPath, int port = 8181);
+    ~KiwixServe();
+
+    void run();
+    void shutDown();
+    bool isRunning();
+    int getPort() { return m_port; }
+    int setPort(int port);
+
+  private:
+  std::unique_ptr<Subprocess> mp_kiwixServe;
+  int m_port;
+  std::string m_libraryPath;
+};
+
+}; //end namespace kiwix
+
+#endif // KIWIXLIB_KIWIX_SERVE_H_

include/library.h

@@ -23,14 +23,19 @@
 #include <string>
 #include <vector>
 #include <map>
+#include <memory>
+
+#include "book.h"
+#include "bookmark.h"
+#include "common.h"
 
 #define KIWIX_LIBRARY_VERSION "20110515"
 
 namespace kiwix
 {
 
-class Book;
 class OPDSDumper;
+class Library;
 
 enum supportedListSortBy { UNSORTED, TITLE, SIZE, DATE, CREATOR, PUBLISHER };
 enum supportedListMode {
@@ -42,17 +47,124 @@ enum supportedListMode {
   VALID = 1 << 4,
   NOVALID = 1 << 5
 };
+
+class Filter {
+  public: // types
+    using Tags = std::vector<std::string>;
+
+  private: // data
+    uint64_t activeFilters;
+    Tags _acceptTags;
+    Tags _rejectTags;
+    std::string _category;
+    std::string _lang;
+    std::string _publisher;
+    std::string _creator;
+    size_t _maxSize;
+    std::string _query;
+    bool _queryIsPartial;
+    std::string _name;
+
+  public: // functions
+    Filter();
+    ~Filter() = default;
+
+    /**
+     *  Set the filter to check local.
+     *
+     *  A local book is a book with a path.
+     *  If accept is true, only local book are accepted.
+     *  If accept is false, only non local book are accepted.
+     */
+    Filter& local(bool accept);
+
+    /**
+     *  Set the filter to check remote.
+     *
+     *  A remote book is a book with a url.
+     *  If accept is true, only remote book are accepted.
+     *  If accept is false, only non remote book are accepted.
+     */
+    Filter& remote(bool accept);
+
+    /**
+     *  Set the filter to check validity.
+     *
+     *  A valid book is a book with a path pointing to a existing zim file.
+     *  If accept is true, only valid book are accepted.
+     *  If accept is false, only non valid book are accepted.
+     */
+    Filter& valid(bool accept);
+
+    /**
+     * Set the filter to only accept book with corresponding tag.
+     */
+    Filter& acceptTags(const Tags& tags);
+    Filter& rejectTags(const Tags& tags);
+
+    Filter& category(std::string category);
+    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);
+
+    bool hasQuery() const;
+    const std::string& getQuery() const { return _query; }
+    bool queryIsPartial() const { return _queryIsPartial; }
+
+    bool hasName() const;
+    const std::string& getName() const { return _name; }
+
+    bool hasCategory() const;
+    const std::string& getCategory() const { return _category; }
+
+    bool hasLang() const;
+    const std::string& getLang() const { return _lang; }
+
+    bool hasPublisher() const;
+    const std::string& getPublisher() const { return _publisher; }
+
+    bool hasCreator() const;
+    const std::string& getCreator() const { return _creator; }
+
+    const Tags& getAcceptTags() const { return _acceptTags; }
+    const Tags& getRejectTags() const { return _rejectTags; }
+
+private: // functions
+    friend class Library;
+
+    bool accept(const Book& book) const;
+};
+
+
 /**
  * A Library store several books.
  */
 class Library
 {
-  std::map<std::string, kiwix::Book> books;
+  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 std::vector<std::string> BookIdCollection;
+
  public:
   Library();
   ~Library();
 
-  std::string version;
+  /**
+   * Library is not a copiable object. However it can be moved.
+   */
+  Library(const Library& ) = delete;
+  Library(Library&& );
+  void operator=(const Library& ) = delete;
+  Library& operator=(Library&& );
+
   /**
    * Add a book to the library.
    *
@@ -65,7 +177,27 @@ class Library
    */
   bool addBook(const Book& book);
 
+  /**
+   * Add a bookmark to the library.
+   *
+   * @param bookmark the book to add.
+   */
+  void addBookmark(const Bookmark& bookmark);
+
+  /**
+   * Remove a bookmarkk
+   *
+   * @param zimId The zimId of the bookmark.
+   * @param url The url of the bookmark.
+   * @return True if the bookmark has been removed.
+   */
+  bool removeBookmark(const std::string& zimId, const std::string& url);
+
+  const Book& getBookById(const std::string& id) const;
   Book& getBookById(const std::string& id);
+  const Book& getBookByPath(const std::string& path) const;
+  Book& getBookByPath(const std::string& path);
+  std::shared_ptr<Reader> getReaderById(const std::string& id);
 
   /**
    * Remove a book from the library.
@@ -79,9 +211,17 @@ class Library
    * Write the library to a file.
    *
    * @param path the path of the file to write to.
-   * @return True if the library has been correctly save.
+   * @return True if the library has been correctly saved.
    */
-  bool writeToFile(const std::string& path);
+  bool writeToFile(const std::string& path) const;
+
+  /**
+   * Write the library bookmarks to a file.
+   *
+   * @param path the path of the file to write to.
+   * @return True if the library has been correctly saved.
+   */
+  bool writeBookmarksToFile(const std::string& path) const;
 
   /**
    * Get the number of book in the library.
@@ -90,35 +230,49 @@ class Library
    * @param remoteBooks If we must count remote books (books with an url)
    * @return The number of books.
    */
-  unsigned int getBookCount(const bool localBooks, const bool remoteBooks);
+  unsigned int getBookCount(const bool localBooks, const bool remoteBooks) const;
 
   /**
-   * Get all langagues of the books in the library.
+   * Get all languagues of the books in the library.
    *
    * @return A list of languages.
    */
-  std::vector<std::string> getBooksLanguages();
+  std::vector<std::string> getBooksLanguages() const;
+
+  /**
+   * Get all categories of the books in the library.
+   *
+   * @return A list of categories.
+   */
+  std::vector<std::string> getBooksCategories() const;
 
   /**
    * Get all book creators of the books in the library.
    *
    * @return A list of book creators.
    */
-  std::vector<std::string> getBooksCreators();
+  std::vector<std::string> getBooksCreators() const;
 
   /**
    * Get all book publishers of the books in the library.
    *
    * @return A list of book publishers.
    */
-  std::vector<std::string> getBooksPublishers();
+  std::vector<std::string> getBooksPublishers() const;
+
+  /**
+   * Get all bookmarks.
+   *
+   * @return A list of bookmarks
+   */
+  const std::vector<kiwix::Bookmark> getBookmarks(bool onlyValidBookmarks = true) const;
 
   /**
    * Get all book ids of the books in the library.
    *
    * @return A list of book ids.
    */
-  std::vector<std::string> getBooksIds();
+  BookIdCollection getBooksIds() const;
 
   /**
    * Filter the library and generate a new one with the keep elements.
@@ -128,9 +282,27 @@ class Library
    * @param search List only books with search in the title or description.
    * @return The list of bookIds corresponding to the query.
    */
-  std::vector<std::string> filter(const std::string& search);
+  DEPRECATED BookIdCollection filter(const std::string& search) const;
 
 
+  /**
+   * Filter the library and return the id of the keep elements.
+   *
+   * @param filter The filter to use.
+   * @return The list of bookIds corresponding to the filter.
+   */
+  BookIdCollection filter(const Filter& filter) const;
+
+
+  /**
+   * Sort (in place) bookIds using the given comparator.
+   *
+   * @param bookIds the list of book Ids to sort
+   * @param comparator how to sort the books
+   * @return The sorted list of books
+   */
+  void sort(BookIdCollection& bookIds, supportedListSortBy sortBy, bool ascending) const;
+
   /**
    * List books in the library.
    *
@@ -153,17 +325,24 @@ class Library
    *                Set to 0 to cancel this filter.
    * @return The list of bookIds corresponding to the query.
    */
-  std::vector<std::string> listBooksIds(
+  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 = "",
-    size_t maxSize = 0);
+    const std::vector<std::string>& tags = {},
+    size_t maxSize = 0) const;
 
   friend class OPDSDumper;
+  friend class libXMLDumper;
+
+private: // functions
+  BookIdCollection filterViaBookDB(const Filter& filter) const;
+  void updateBookDB(const Book& book);
 };
+
 }
 
 #endif

include/libxml_dumper.h

@@ -0,0 +1,83 @@
+/*
+ * Copyright 2018 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_LIBXML_DUMPER_H
+#define KIWIX_LIBXML_DUMPER_H
+
+#include <string>
+#include <vector>
+
+#include <pugixml.hpp>
+
+#include "library.h"
+
+namespace kiwix
+{
+
+/**
+ * A tool to dump a `Library` into a basic library.xml
+ *
+ */
+class LibXMLDumper
+{
+ public:
+  LibXMLDumper() = default;
+  LibXMLDumper(const Library* library);
+  ~LibXMLDumper();
+
+  /**
+   * Dump the library.xml
+   *
+   * @param id The id of the library.
+   * @return The library.xml content.
+   */
+  std::string dumpLibXMLContent(const std::vector<std::string>& bookIds);
+
+
+  /**
+   * Dump the bookmark of the library.
+   *
+   * @return The bookmark.xml content.
+   */
+  std::string dumpLibXMLBookmark();
+
+  /**
+   * Set the base directory used.
+   *
+   * @param baseDir the base directory to use.
+   */
+  void setBaseDir(const std::string& baseDir) { this->baseDir = baseDir; }
+
+  /**
+   * Set the library to dump.
+   *
+   * @param library The library to dump.
+   */
+  void setLibrary(const Library* library) { this->library = library; }
+
+ protected:
+  const kiwix::Library* library;
+  std::string baseDir;
+ private:
+  void handleBook(Book book, pugi::xml_node root_node);
+  void handleBookmark(Bookmark bookmark, pugi::xml_node root_node);
+};
+}
+
+#endif // KIWIX_OPDS_DUMPER_H

include/manager.h

@@ -38,6 +38,7 @@ class LibraryManipulator {
  public:
   virtual ~LibraryManipulator() {}
   virtual bool addBookToLibrary(Book book) = 0;
+  virtual void addBookmarkToLibrary(Bookmark bookmark) = 0;
 };
 
 class DefaultLibraryManipulator : public LibraryManipulator {
@@ -48,15 +49,15 @@ class DefaultLibraryManipulator : public LibraryManipulator {
   bool addBookToLibrary(Book book) {
     return library->addBook(book);
   }
+  void addBookmarkToLibrary(Bookmark bookmark) {
+    library->addBookmark(bookmark);
+  }
  private:
    kiwix::Library* library;
 };
 
 /**
  * A tool to manage a `Library`.
- *
- * A `Manager` handle a internal `Library`.
- * This `Library` can be retrived with `cloneLibrary` method.
  */
 class Manager
 {
@@ -68,26 +69,12 @@ class Manager
   /**
    * Read a `library.xml` and add book in the file to the library.
    *
-   * @param path The path to the `library.xml`.
+   * @param path The (utf8) path to the `library.xml`.
    * @param readOnly Set if the libray path could be overwritten latter with
    *                 updated content.
    * @return True if file has been properly parsed.
    */
-  bool readFile(const std::string& path, const bool readOnly = true);
-
-  /**
-   * Read a `library.xml` and add book in the file to the library.
-   *
-   * @param nativePath The path of the `library.xml`
-   * @param UTF8Path The utf8 version (?) of the path. Also the path where the
-   *                 library will be writen i readOnly is False.
-   * @param readOnly Set if the libray path could be overwritten latter with
-   *                 updated content.
-   * @return True if file has been properly parsed.
-   */
-  bool readFile(const std::string& nativePath,
-                const std::string& UTF8Path,
-                const bool readOnly = true);
+  bool readFile(const std::string& path, bool readOnly = true, bool trustLibrary = true);
 
   /**
    * Load a library content store in the string.
@@ -100,7 +87,8 @@ class Manager
    */
   bool readXml(const std::string& xml,
                const bool readOnly = true,
-               const std::string& libraryPath = "");
+               const std::string& libraryPath = "",
+               bool trustLibrary = true);
 
   /**
    * Load a library content stored in a OPDS stream.
@@ -113,6 +101,15 @@ class Manager
    */
   bool readOpds(const std::string& content, const std::string& urlHost);
 
+
+  /**
+   * Load a bookmark file.
+   *
+   * @param path The path of the file to read.
+   * @return True if the content has been properly parsed.
+   */
+  bool readBookmarkFile(const std::string& path);
+
   /**
    * Add a book to the library.
    *
@@ -145,55 +142,6 @@ class Manager
                        const std::string& url = "",
                        const bool checkMetaData = false);
 
-  /**
-   * Get the book corresponding to an id.
-   *
-   * @param[in] id The id of the book
-   * @param[out] book The book corresponding to the id.
-   * @return True if the book has been found.
-   */
-  bool getBookById(const std::string& id, Book& book);
-
-  /**
-   * Update the "last open date" of a book
-   *
-   * @param id the id of the book.
-   * @return True if the book is in the library.
-   */
-  bool updateBookLastOpenDateById(const std::string& id);
-
-  /**
-   * Remove (set to empty) paths of all books in the library.
-   */
-  void removeBookPaths();
-
-  /**
-   * List books in the library.
-   *
-   * The books list will be available in public vector member `bookIdList`.
-   *
-   * @param mode The mode of listing :
-   *             - LASTOPEN sort by last opened book.
-   *             - LOCAL list only local file.
-   *             - REMOTE list only remote file.
-   * @param sortBy Attribute to sort by the book list.
-   * @param maxSize Do not list book bigger than maxSize MiB.
-   *                Set to 0 to cancel this filter.
-   * @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 search List only books with search in the title, description or
-   *               language.
-   * @return True
-   */
-  bool listBooks(const supportedListMode mode,
-                 const supportedListSortBy sortBy,
-                 const unsigned int maxSize,
-                 const std::string& language,
-                 const std::string& creator,
-                 const std::string& publisher,
-                 const std::string& search);
-
   std::string writableLibraryPath;
 
   bool m_hasSearchResult = false;
@@ -207,13 +155,12 @@ class Manager
 
   bool readBookFromPath(const std::string& path, Book* book);
   bool parseXmlDom(const pugi::xml_document& doc,
-                   const bool readOnly,
-                   const std::string& libraryPath);
+                   bool readOnly,
+                   const std::string& libraryPath,
+                   bool trustLibrary);
   bool parseOpdsDom(const pugi::xml_document& doc,
                     const std::string& urlHost);
 
- private:
-  void checkAndCleanBookPaths(Book& book, const std::string& libraryPath);
 };
 }
 

include/meson.build

@@ -1,35 +1,30 @@
 headers = [
   'book.h',
+  'bookmark.h',
   'common.h',
   'library.h',
   'manager.h',
+  'libxml_dumper.h',
   'opds_dumper.h',
   'downloader.h',
   'reader.h',
   'entry.h',
-  'searcher.h'
+  'searcher.h',
+  'search_renderer.h',
+  'server.h',
+  'kiwixserve.h',
+  'name_mapper.h'
 ]
 
-if xapian_dep.found()
-  headers += ['xapianSearcher.h']
-endif
-
 install_headers(headers, subdir:'kiwix')
 
 install_headers(
-  'common/base64.h',
-  'common/networkTools.h',
-  'common/otherTools.h',
-  'common/pathTools.h',
-  'common/regexTools.h',
-  'common/stringTools.h',
-  subdir:'kiwix/common'
+  'tools/base64.h',
+  'tools/networkTools.h',
+  'tools/otherTools.h',
+  'tools/pathTools.h',
+  'tools/regexTools.h',
+  'tools/stringTools.h',
+  subdir:'kiwix/tools'
 )
 
-if has_ctpp2_dep
-  install_headers(
-    'ctpp2/CTPP2VMStringLoader.hpp',
-    subdir:'kiwix/ctpp2'
-  )
-endif
-

include/name_mapper.h

@@ -0,0 +1,61 @@
+/*
+ * Copyright 2019 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_NAMEMAPPER_H
+#define KIWIX_NAMEMAPPER_H
+
+#include <string>
+#include <map>
+
+namespace kiwix
+{
+
+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;
+};
+
+
+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; };
+};
+
+class HumanReadableNameMapper : public NameMapper {
+  private:
+    std::map<std::string, std::string> m_idToName;
+    std::map<std::string, std::string> m_nameToId;
+
+  public:
+    HumanReadableNameMapper(kiwix::Library& library, bool withAlias);
+    virtual ~HumanReadableNameMapper() = default;
+    virtual std::string getNameForId(const std::string& id);
+    virtual std::string getIdForName(const std::string& name);
+};
+
+
+
+}
+
+#endif

include/opds_dumper.h

@@ -26,9 +26,9 @@
 
 #include <pugixml.hpp>
 
-#include "common/base64.h"
-#include "common/pathTools.h"
-#include "common/regexTools.h"
+#include "tools/base64.h"
+#include "tools/pathTools.h"
+#include "tools/regexTools.h"
 #include "library.h"
 #include "reader.h"
 
@@ -51,24 +51,35 @@ class OPDSDumper
   /**
    * Dump the OPDS feed.
    *
-   * @param id The id of the library.
+   * @param bookIds the ids of the books to include in the feed
+   * @param query the query used to obtain the list of book ids
    * @return The OPDS feed.
    */
-  std::string dumpOPDSFeed(const std::vector<std::string>& bookIds);
+  std::string dumpOPDSFeed(const std::vector<std::string>& bookIds, const std::string& query) const;
 
   /**
-   * Set the id of the opds stream.
+   * Dump the OPDS feed.
+   *
+   * @param bookIds the ids of the books to include in the feed
+   * @param query the query used to obtain the list of book ids
+   * @return The OPDS feed.
+   */
+  std::string dumpOPDSFeedV2(const std::vector<std::string>& bookIds, const std::string& query) 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;
+
+  /**
+   * Set the id of the library.
    *
    * @param id the id to use.
    */
-  void setId(const std::string& id) { this->id = id;}
-
-  /**
-   * Set the title oft the opds stream.
-   *
-   * @param title the title to use.
-   */
-  void setTitle(const std::string& title) { this->title = title; }
+  void setLibraryId(const std::string& id) { this->libraryId = id;}
 
   /**
    * Set the root location used when generating url.
@@ -77,13 +88,6 @@ class OPDSDumper
    */
   void setRootLocation(const std::string& rootLocation) { this->rootLocation = rootLocation; }
 
-  /**
-   * Set the search url.
-   *
-   * @param searchUrl the search url to use.
-   */
-  void setSearchDescriptionUrl(const std::string& searchDescriptionUrl) { this->searchDescriptionUrl = searchDescriptionUrl; }
-
   /**
    * Set some informations about the search results.
    *
@@ -93,27 +97,13 @@ class OPDSDumper
    */
   void setOpenSearchInfo(int totalResult, int startIndex, int count);
 
-  /**
-   * Set the library to dump.
-   *
-   * @param library The library to dump.
-   */
-  void setLibrary(Library* library) { this->library = library; }
-
  protected:
   kiwix::Library* library;
-  std::string id;
-  std::string title;
-  std::string date;
+  std::string libraryId;
   std::string rootLocation;
-  std::string searchDescriptionUrl;
   int m_totalResults;
   int m_startIndex;
   int m_count;
-  bool m_isSearchResult = false;
-
- private:
-  pugi::xml_node handleBook(Book book, pugi::xml_node root_node);
 };
 }
 

include/reader.h

@@ -21,28 +21,63 @@
 #define KIWIX_READER_H
 
 #include <stdio.h>
-#include <zim/article.h>
-#include <zim/file.h>
-#include <zim/fileiterator.h>
 #include <zim/zim.h>
+#include <zim/archive.h>
 #include <exception>
 #include <map>
 #include <sstream>
 #include <string>
 #include "common.h"
 #include "entry.h"
-#include "common/pathTools.h"
-#include "common/stringTools.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:
@@ -55,8 +90,12 @@ class Reader
    *                    unsplitted path as if the file were not splitted
    *                    (.zim extesion).
    */
-  Reader(const string zimFilePath);
-  ~Reader();
+  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.
@@ -99,17 +138,6 @@ class Reader
    */
   string getId() const;
 
-  /**
-   * Get the url of a random page.
-   *
-   * Deprecated : Use `getRandomPage` instead.
-   *
-   * @return Url of a random page. The page is picked from all entries in
-   *         the 'A' namespace.
-   *         The main page is excluded from the potential results.
-   */
-  DEPRECATED string getRandomPageUrl() const;
-
   /**
    * Get a random page.
    *
@@ -119,31 +147,6 @@ class Reader
    */
   Entry getRandomPage() const;
 
-  /**
-   * Get the url of the first page.
-   *
-   * Deprecated : Use `getFirstPage` instead.
-   *
-   * @return Url of the first entry in the 'A' namespace.
-   */
-  DEPRECATED string getFirstPageUrl() const;
-
-  /**
-   * Get the entry of the first page.
-   *
-   * @return The first entry in the 'A' namespace.
-   */
-  Entry getFirstPage() const;
-
-  /**
-   * Get the url of the main page.
-   *
-   * Deprecated : Use `getMainPage` instead.
-   *
-   * @return Url of the main page as specified in the zim file.
-   */
-  DEPRECATED string getMainPageUrl() const;
-
   /**
    * Get the entry of the main page.
    *
@@ -158,31 +161,7 @@ class Reader
    * @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 getMetatag(const string& name, string& value) 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 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 language of the zim file.
-   *
-   * @return The language of the zim file as specified in the zim metadata.
-   */
-  string getLanguage() const;
+  bool getMetadata(const string& name, string& value) const;
 
   /**
    * Get the name of the zim file.
@@ -192,18 +171,13 @@ class Reader
   string getName() const;
 
   /**
-   * Get the tags of the zim file.
+   * Get the title of the zim file.
    *
-   * @return The tags of the zim file as specified in the zim metadata.
+   * @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 getTags() const;
-
-  /**
-   * Get the date of the zim file.
-   *
-   * @return The date of the zim file as specified in the zim metadata.
-   */
-  string getDate() const;
+  string getTitle() const;
 
   /**
    * Get the creator of the zim file.
@@ -219,6 +193,100 @@ class Reader
    */
   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.
    *
@@ -267,93 +335,15 @@ class Reader
    */
   Entry getEntryFromTitle(const std::string& title) const;
 
-  /**
-   * Get the url of a page specified by a title.
-   *
-   * @param[in] title the title of the page.
-   * @param[out] url the url of the page.
-   * @return True if the page can be found.
-   */
-  DEPRECATED bool getPageUrlFromTitle(const string& title, string& url) const;
-
-  /**
-   * Get the mimetype of a entry specified by a url.
-   *
-   * @param[in] url the url of the entry.
-   * @param[out] mimeType the mimeType of the entry.
-   * @return True if the mimeType has been found.
-   */
-  DEPRECATED bool getMimeTypeByUrl(const string& url, string& mimeType) const;
-
-  /**
-   * Get the content of an entry specifed by a url.
-   *
-   * Alias to `getContentByEncodedUrl`
-   */
-  DEPRECATED bool getContentByUrl(const string& url,
-                       string& content,
-                       string& title,
-                       unsigned int& contentLength,
-                       string& contentType) const;
-
-  /**
-   * Get the content of an entry specified by a url encoded url.
-   *
-   * Equivalent to getContentByDecodedUrl(urlDecode(url), ...).
-   */
-  DEPRECATED bool getContentByEncodedUrl(const string& url,
-                              string& content,
-                              string& title,
-                              unsigned int& contentLength,
-                              string& contentType,
-                              string& baseUrl) const;
-
-  /**
-   * Get the content of an entry specified by an url encoded url.
-   *
-   * Equivalent to getContentByEncodedUrl but without baseUrl.
-   */
-  DEPRECATED bool getContentByEncodedUrl(const string& url,
-                              string& content,
-                              string& title,
-                              unsigned int& contentLength,
-                              string& contentType) const;
-
-  /**
-   * Get the content of an entry specified by a url.
-   *
-   * @param[in] url The url of the entry.
-   * @param[out] content The content of the entry.
-   * @param[out] title the title of the entry.
-   * @param[out] contentLength The size of the entry (size of content).
-   * @param[out] contentType The mimeType of the entry.
-   * @param[out] baseUrl Return the true url of the entry.
-   *                     If the specified entry is a redirection, contains
-   *                     the url of the targeted entry.
-   * @return True if the entry has been found.
-   */
-  DEPRECATED bool getContentByDecodedUrl(const string& url,
-                              string& content,
-                              string& title,
-                              unsigned int& contentLength,
-                              string& contentType,
-                              string& baseUrl) const;
-  /**
-   * Get the content of an entry specified by a url.
-   *
-   * Equivalent to getContentByDecodedUrl but withou the baseUrl.
-   */
-  DEPRECATED bool getContentByDecodedUrl(const string& url,
-                              string& content,
-                              string& title,
-                              unsigned int& contentLength,
-                              string& contentType) 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.
@@ -361,12 +351,49 @@ class Reader
    *              If false, add suggestions to the internal vector
    *              (until internal vector size is suggestionCount (or no more
    *               suggestion))
-   * @return True if some suggestions where added to the internal vector.
+   * @return True if some suggestions have been added to the internal vector.
    */
-  bool searchSuggestions(const string& prefix,
+  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.
    *
@@ -381,19 +408,13 @@ class Reader
    *
    * @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);
+   bool searchSuggestionsSmart(const string& prefix,
+                              unsigned int suggestionsCount,
+                              SuggestionsList_t& results);
 
-  /**
-   * Check if the url exists in the zim file.
-   *
-   * Deprecated : Use `pathExists` instead.
-   *
-   * @param url the url to check.
-   * @return True if the url exits in the zim file.
-   */
-  DEPRECATED bool urlExists(const string& url) const;
 
   /**
    * Check if the path exists in the zim file.
@@ -425,7 +446,7 @@ class Reader
    * @param[out] title the title of the suggestion.
    * @return True if title has been set.
    */
-  bool getNextSuggestion(string& title);
+  DEPRECATED bool getNextSuggestion(string& title);
 
   /**
    * Get the next suggestion title and url.
@@ -434,7 +455,7 @@ class Reader
    * @param[out] url the url of the suggestion.
    * @return True if title and url have been set.
    */
-  bool getNextSuggestion(string& title, string& url);
+  DEPRECATED bool getNextSuggestion(string& title, string& url);
 
   /**
    * Get if we can check zim file integrity (has a checksum).
@@ -450,16 +471,6 @@ class Reader
    */
   bool isCorrupted() const;
 
-  /**
-   * Parse a full url into a namespace and url.
-   *
-   * @param[in] url The full url ("/N/url").
-   * @param[out] ns The namespace (N).
-   * @param[out] title The url (url).
-   * @return True
-   */
-  DEPRECATED bool parseUrl(const string& url, char* ns, string& title) const;
-
   /**
    * Return the total size of the zim file.
    *
@@ -474,28 +485,14 @@ class Reader
    *
    * @return The libzim file handler.
    */
-  zim::File* getZimFileHandler() const;
-
-  /**
-   * Get the zim article object associated to a url.
-   *
-   * @param[in] url The url of the article.
-   * @param[out] article The libzim article object.
-   * @return True if the url is good (article.good()).
-   */
-  DEPRECATED bool getArticleObjectByDecodedUrl(const string& url,
-                                    zim::Article& article) const;
+  zim::Archive* getZimArchive() const;
 
  protected:
-  zim::File* zimFileHandler;
-  zim::size_type firstArticleOffset;
-  zim::size_type lastArticleOffset;
-  zim::size_type nsACount;
-  zim::size_type nsICount;
+  std::unique_ptr<zim::Archive> zimArchive;
   std::string zimFilePath;
 
-  std::vector<std::vector<std::string>> suggestions;
-  std::vector<std::vector<std::string>>::iterator suggestionsOffset;
+  SuggestionsList_t suggestions;
+  SuggestionsList_t::iterator suggestionsOffset;
 
  private:
   std::map<const std::string, unsigned int> parseCounterMetadata() const;

include/search_renderer.h

@@ -0,0 +1,91 @@
+/*
+ * 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_SEARCH_RENDERER_H
+#define KIWIX_SEARCH_RENDERER_H
+
+#include <string>
+
+namespace kiwix
+{
+
+class Searcher;
+class NameMapper;
+/**
+ * The SearcherRenderer class is used to render a search result to a html page.
+ */
+class SearchRenderer
+{
+ public:
+  /**
+   * The default constructor.
+   *
+   * @param humanReadableName The global zim's humanReadableName.
+   *                          Used to generate pagination links.
+   */
+  SearchRenderer(Searcher* searcher, NameMapper* mapper);
+
+  ~SearchRenderer();
+
+  void setSearchPattern(const std::string& pattern);
+
+  /**
+   * Set the search content id.
+   */
+  void setSearchContent(const std::string& name);
+
+  /**
+   * Set protocol prefix.
+   */
+  void setProtocolPrefix(const std::string& prefix);
+
+  /**
+   * Set search protocol prefix.
+   */
+  void setSearchProtocolPrefix(const std::string& prefix);
+
+  /**
+   * set result count per page
+   */
+  void setPageLength(unsigned int pageLength){
+    this->pageLength  = pageLength;
+  }
+
+  /**
+   * Generate the html page with the resutls of the search.
+   */
+  std::string getHtml();
+
+ protected:
+  std::string beautifyInteger(const unsigned int number);
+  Searcher* mp_searcher;
+  NameMapper* mp_nameMapper;
+  std::string searchContent;
+  std::string searchPattern;
+  std::string protocolPrefix;
+  std::string searchProtocolPrefix;
+  unsigned int pageLength;
+  unsigned int estimatedResultCount;
+  unsigned int resultStart;
+};
+
+
+}
+
+#endif

include/searcher.h

@@ -27,11 +27,10 @@
 #include <cctype>
 #include <locale>
 #include <string>
+#include <memory>
 #include <vector>
-#include <vector>
-#include "common/pathTools.h"
-#include "common/stringTools.h"
-#include "kiwix_config.h"
+#include "tools/pathTools.h"
+#include "tools/stringTools.h"
 
 using namespace std;
 
@@ -49,72 +48,35 @@ class Result
   virtual std::string get_content() = 0;
   virtual int get_wordCount() = 0;
   virtual int get_size() = 0;
-  virtual int get_readerIndex() = 0;
+  virtual std::string get_zimId() = 0;
 };
 
 struct SearcherInternal;
 /**
  * The Searcher class is reponsible to do different kind of search using the
  * fulltext index.
- *
- * Historically, there are two kind of fulltext index :
- *  - The legacy one, is the external fulltext index. A directory stored outside
- *    of the zim file.
- *  - The new one, a embedded fulltext index in the zim file.
- *
- * Legacy external fulltext index has to be considered as obsolet format with
- * less functionnalities:
- *  - No multi zim search ;
- *  - No geo_search ;
- *  - No suggestions search ;
- *
- * To reflect this, there is two Search creation "API":
- *  - One for the external fulltext index, using the constructor taking a
- *    xapianDirectoryPath) ;
- *  - One for the embedded fulltext index, using a "empty" constructor and the
- *  `add_reader` method".
- *
- *  On top of that, the Searcher may (if compiled with ctpp2) be used to
- *  generate a html page for the search result. This use a template that need a
- *  humanReaderName. This feature is only used by kiwix-serve and this should be
- *  move outside of Searcher (and with a better API). If you don't use the html
- *  rendering (getHtml method), you better should simply ignore the different
- *  humanReadeableName attributes (or give an empty string).
  */
 class Searcher
 {
  public:
   /**
    * The default constructor.
-   *
-   * @param humanReadableName The global zim's humanReadableName.
-   *                          Used to generate pagination links.
    */
-  Searcher(const string& humanReadableName = "");
+  Searcher();
 
-  /**
-   * The constructor for legacy external fulltext index.
-   *
-   * @param xapianDirectoryPath The path to the external index directory.
-   * @param reader The reader associated to the external index.
-   *               It will be used retrive the article content or generate
-   *               the snippet.
-   * @param humanReadableName The humanReadableName for the zim.
-   */
-  Searcher(const string& xapianDirectoryPath,
-           Reader* reader,
-           const string& humanReadableName);
   ~Searcher();
 
   /**
    * Add a reader (containing embedded fulltext index) to the search.
    *
    * @param reader The Reader for the zim containing the fulltext index.
-   * @param humanReaderName The human readable name of the reader.
    * @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, const std::string& humanReaderName);
+  bool add_reader(Reader* reader);
+
+
+  Reader* get_reader(int index);
 
   /**
    * Start a search on the zim associated to the Searcher.
@@ -126,7 +88,7 @@ class Searcher
    * @param resultEnd the end offset of the search results (used for pagination).
    * @param verbose print some info on stdout if true.
    */
-  void search(std::string& search,
+  void search(const std::string& search,
               unsigned int resultStart,
               unsigned int resultEnd,
               const bool verbose = false);
@@ -180,24 +142,8 @@ class Searcher
    */
   unsigned int getEstimatedResultCount();
 
-  /**
-   * Set protocol prefix.
-   * Only used by getHtml.
-   */
-  bool setProtocolPrefix(const std::string prefix);
-
-  /**
-   * Set search protocol prefix.
-   * Only used by getHtml.
-   */
-  bool setSearchProtocolPrefix(const std::string prefix);
-
-#ifdef ENABLE_CTPP2
-  /**
-   * Generate the html page with the resutls of the search.
-   */
-  string getHtml();
-#endif
+  unsigned int getResultStart() { return resultStart; }
+  unsigned int getResultEnd() { return resultEnd; }
 
  protected:
   std::string beautifyInteger(const unsigned int number);
@@ -208,16 +154,11 @@ class Searcher
                      const bool verbose = false);
 
   std::vector<Reader*> readers;
-  std::vector<std::string> humanReaderNames;
-  SearcherInternal* internal;
+  std::unique_ptr<SearcherInternal> internal;
   std::string searchPattern;
-  std::string protocolPrefix;
-  std::string searchProtocolPrefix;
-  unsigned int resultCountPerPage;
   unsigned int estimatedResultCount;
   unsigned int resultStart;
   unsigned int resultEnd;
-  std::string contentHumanReadableId;
 
  private:
   void reset();

include/server.h

@@ -0,0 +1,78 @@
+/*
+ * Copyright 2019 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_SERVER_H
+#define KIWIX_SERVER_H
+
+#include <string>
+#include <memory>
+
+namespace kiwix
+{
+  class Library;
+  class NameMapper;
+  class InternalServer;
+
+  class Server {
+     public:
+       /**
+        * The default constructor.
+        *
+        * @param library The library to serve.
+        */
+       Server(Library* library, NameMapper* nameMapper=nullptr);
+
+       virtual ~Server();
+
+       /**
+        * Serve the content.
+        */
+       bool start();
+
+       /**
+        * Stop the daemon.
+        */
+       void stop();
+
+       void setRoot(const std::string& root);
+       void setAddress(const std::string& addr) { m_addr = addr; }
+       void setPort(int port) { m_port = port; }
+       void setNbThreads(int threads) { m_nbThreads = threads; }
+       void setVerbose(bool verbose) { m_verbose = verbose; }
+       void setTaskbar(bool withTaskbar, bool withLibraryButton)
+        { m_withTaskbar = withTaskbar; m_withLibraryButton = withLibraryButton; }
+       void setBlockExternalLinks(bool blockExternalLinks)
+        { m_blockExternalLinks = blockExternalLinks; }
+
+     protected:
+       Library* mp_library;
+       NameMapper* mp_nameMapper;
+       std::string m_root = "";
+       std::string m_addr = "";
+       int m_port = 80;
+       int m_nbThreads = 1;
+       bool m_verbose = false;
+       bool m_withTaskbar = true;
+       bool m_withLibraryButton = true;
+       bool m_blockExternalLinks = false;
+       std::unique_ptr<InternalServer> mp_server;
+  };
+}
+
+#endif

include/tools/networkTools.h

@@ -20,13 +20,10 @@
 #ifndef KIWIX_NETWORKTOOLS_H
 #define KIWIX_NETWORKTOOLS_H
 
-#include <map>
 #include <string>
 
 namespace kiwix
 {
-std::map<std::string, std::string> getNetworkInterfaces();
-std::string getBestPublicIp();
 std::string download(const std::string& url);
 }
 

include/tools/otherTools.h

@@ -0,0 +1,56 @@
+/*
+ * Copyright 2014 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_OTHERTOOLS_H
+#define KIWIX_OTHERTOOLS_H
+
+#include <string>
+#include <vector>
+#include <map>
+#include <zim/zim.h>
+#include <mustache.hpp>
+
+namespace pugi {
+  class xml_node;
+}
+
+namespace kiwix
+{
+  void sleep(unsigned int milliseconds);
+  std::string nodeToString(const pugi::xml_node& node);
+  std::string converta2toa3(const std::string& a2code);
+
+  /*
+   * Convert all format tag string to new format
+   */
+  std::vector<std::string> convertTags(const std::string& tags_str);
+  std::string getTagValueFromTagList(const std::vector<std::string>& tagList,
+                                     const std::string& tagName);
+  bool convertStrToBool(const std::string& value);
+
+  using MimeCounterType = std::map<const std::string, zim::entry_index_type>;
+  MimeCounterType parseMimetypeCounter(const std::string& counterData);
+
+  std::string gen_date_str();
+  std::string gen_uuid(const std::string& s);
+
+  std::string render_template(const std::string& template_str, kainjow::mustache::data data);
+}
+
+#endif

include/tools/pathTools.h

@@ -0,0 +1,48 @@
+/*
+ * 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_PATHTOOLS_H
+#define KIWIX_PATHTOOLS_H
+
+#include <string>
+
+#ifdef _WIN32
+std::string WideToUtf8(const std::wstring& wstr);
+std::wstring Utf8ToWide(const std::string& str);
+#endif
+bool isRelativePath(const std::string& path);
+std::string computeAbsolutePath(const std::string& path, const std::string& relativePath);
+std::string computeRelativePath(const std::string& path, const std::string& absolutePath);
+std::string removeLastPathElement(const std::string& path);
+std::string appendToDirectory(const std::string& directoryPath, const std::string& filename);
+
+unsigned int getFileSize(const std::string& path);
+std::string getFileSizeAsString(const std::string& path);
+std::string getFileContent(const std::string& path);
+bool fileExists(const std::string& path);
+bool makeDirectory(const std::string& path);
+std::string makeTmpDirectory();
+bool copyFile(const std::string& sourcePath, const std::string& destPath);
+std::string getLastPathElement(const std::string& path);
+std::string getExecutablePath(bool realPathOnly = false);
+std::string getCurrentDirectory();
+std::string getDataDirectory();
+bool writeTextFile(const std::string& path, const std::string& content);
+std::string getMimeTypeForFile(const std::string& filename);
+#endif

include/tools/regexTools.h

@@ -20,9 +20,6 @@
 #ifndef KIWIX_REGEXTOOLS_H
 #define KIWIX_REGEXTOOLS_H
 
-#include <unicode/regex.h>
-#include <unicode/ucnv.h>
-#include <map>
 #include <string>
 
 bool matchRegex(const std::string& content, const std::string& regex);
@@ -30,7 +27,10 @@ std::string replaceRegex(const std::string& content,
                          const std::string& replacement,
                          const std::string& regex);
 std::string appendToFirstOccurence(const std::string& content,
-                                   const std::string regex,
+                                   const std::string& regex,
                                    const std::string& replacement);
+std::string prependToFirstOccurence(const std::string& content,
+                                    const std::string& regex,
+                                    const std::string& replacement);
 
 #endif

include/tools/stringTools.h

@@ -22,19 +22,12 @@
 
 #include <unicode/unistr.h>
 
-#include <fstream>
-#include <iomanip>
-#include <iostream>
-#include <sstream>
 #include <string>
 #include <vector>
-
-#include "pathTools.h"
+#include <sstream>
 
 namespace kiwix
 {
-#ifndef __ANDROID__
-
 std::string beautifyInteger(uint64_t number);
 std::string beautifyFileSize(uint64_t number);
 void printStringInHexadecimal(const char* s);
@@ -44,18 +37,14 @@ void stringReplacement(std::string& str,
                        const std::string& newStr);
 std::string encodeDiples(const std::string& str);
 
-#endif
-
 std::string removeAccents(const std::string& text);
 void loadICUExternalTables();
 
 std::string urlEncode(const std::string& value, bool encodeReserved = false);
 std::string urlDecode(const std::string& value, bool component = false);
 
-std::vector<std::string> split(const std::string&, const std::string&);
-std::vector<std::string> split(const char*, const char*);
-std::vector<std::string> split(const std::string&, const char*);
-std::vector<std::string> split(const char*, const std::string&);
+std::vector<std::string> split(const std::string& str, const std::string& delims, bool trimEmpty = true, bool keepDelim = false);
+std::string join(const std::vector<std::string>& list, const std::string& sep);
 
 std::string ucAll(const std::string& word);
 std::string lcAll(const std::string& word);
@@ -79,5 +68,7 @@ T extractFromString(const std::string& str) {
     iss >> ret;
     return ret;
 }
+
+bool startsWith(const std::string& base, const std::string& start);
 } //namespace kiwix
 #endif

include/xapianSearcher.h

@@ -1,98 +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_XAPIAN_SEARCHER_H
-#define KIWIX_XAPIAN_SEARCHER_H
-
-#include <xapian.h>
-#include "reader.h"
-#include "searcher.h"
-
-#include <map>
-#include <string>
-
-using namespace std;
-
-namespace kiwix
-{
-class XapianSearcher;
-
-class XapianResult : public Result
-{
- public:
-  XapianResult(XapianSearcher* searcher, Xapian::MSetIterator& iterator);
-  virtual ~XapianResult(){};
-
-  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 int get_readerIndex() { return 0; };
-
- private:
-  XapianSearcher* searcher;
-  Xapian::MSetIterator iterator;
-  Xapian::Document document;
-};
-
-class NoXapianIndexInZim : public exception
-{
-  virtual const char* what() const throw()
-  {
-    return "There is no fulltext index in the zim file";
-  }
-};
-
-class XapianSearcher
-{
-  friend class XapianResult;
-
- public:
-  XapianSearcher(const string& xapianDirectoryPath, Reader* reader);
-  virtual ~XapianSearcher(){};
-  void searchInIndex(string& search,
-                     const unsigned int resultStart,
-                     const unsigned int resultEnd,
-                     const bool verbose = false);
-  virtual Result* getNextResult();
-  void restart_search();
-
-  Xapian::MSet results;
-
- protected:
-  void closeIndex();
-  void openIndex(const string& xapianDirectoryPath);
-  void setup_queryParser();
-
-  Reader* reader;
-  Xapian::Database readableDatabase;
-  std::string language;
-  std::string stopwords;
-  Xapian::QueryParser queryParser;
-  Xapian::Stem stemmer;
-  Xapian::SimpleStopper stopper;
-  Xapian::MSetIterator current_result;
-  std::map<std::string, int> valuesmap;
-};
-}
-
-#endif

meson.build

@@ -1,23 +1,54 @@
-project('kiwix-lib', 'cpp',
-  version : '3.0.2',
-  license : 'GPL',
+project('libkiwix', 'cpp',
+  version : '10.0.0', # Also change this in android-kiwix-lib-publisher/kiwixLibAndroid/build.gradle
+  license : 'GPLv3+',
   default_options : ['c_std=c11', 'cpp_std=c++11', 'werror=true'])
 
 compiler = meson.get_compiler('cpp')
-find_library_in_compiler = meson.version().version_compare('>=0.31.0')
 
-static_deps = get_option('android') or get_option('default_library') == 'static'
-if get_option('android')
+wrapper = get_option('wrapper')
+
+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 = []
 endif
 
-thread_dep = dependency('threads')
+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
+  thread_dep = dependency('threads')
+else
+  thread_dep = dependency('', required:false)
+endif
 libicu_dep = dependency('icu-i18n', static:static_deps)
-libzim_dep = dependency('libzim', version : '>=4.0.0', static:static_deps)
 pugixml_dep = dependency('pugixml', static:static_deps)
 libcurl_dep = dependency('libcurl', static:static_deps)
+microhttpd_dep = dependency('libmicrohttpd', static:static_deps)
+zlib_dep = dependency('zlib', static:static_deps)
+xapian_dep = dependency('xapian-core', static:static_deps)
+
+if compiler.has_header('mustache.hpp')
+  extra_include = []
+elif compiler.has_header('mustache.hpp', args: '-I/usr/include/kainjow')
+  extra_include = ['/usr/include/kainjow']
+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.')
+endif
+
 
 extra_cflags = ''
 if target_machine.system() == 'windows' and static_deps
@@ -25,77 +56,16 @@ if target_machine.system() == 'windows' and static_deps
   extra_cflags += '-DCURL_STATICLIB'
 endif
 
-
-
-ctpp2_include_path = ''
-has_ctpp2_dep = false
-ctpp2_prefix_install = get_option('ctpp2-install-prefix')
-ctpp2_link_args = []
-if ctpp2_prefix_install == ''
-  if compiler.has_header('ctpp2/CTPP2Logger.hpp')
-    if find_library_in_compiler
-      ctpp2_lib = compiler.find_library('ctpp2')
-    else
-      ctpp2_lib = find_library('ctpp2')
-    endif
-    ctpp2_link_args = ['-lctpp2']
-    if meson.is_cross_build() and host_machine.system() == 'windows'
-      if find_library_in_compiler
-        iconv_lib = compiler.find_library('iconv', required:false)
-      else
-        iconv_lib = find_library('iconv', required:false)
-      endif
-      if iconv_lib.found()
-        ctpp2_link_args += ['-liconv']
-      endif
-    endif
-    has_ctpp2_dep = true
-    ctpp2_dep = declare_dependency(link_args:ctpp2_link_args)
-  else
-    message('ctpp2/CTPP2Logger.hpp not found. Compiling without CTPP2 support')
-  endif
-else
-  if not find_library_in_compiler
-    error('For custom ctpp2_prefix_install you need a meson version >=0.31.0')
-  endif
-  ctpp2_include_path = ctpp2_prefix_install + '/include'
-  ctpp2_include_args = ['-I'+ctpp2_include_path]
-  if compiler.has_header('ctpp2/CTPP2Logger.hpp', args:ctpp2_include_args)
-    ctpp2_include_dir = include_directories(ctpp2_include_path, is_system:true)
-    ctpp2_lib_path = join_paths(ctpp2_prefix_install, get_option('libdir'))
-    message(ctpp2_lib_path)
-    ctpp2_lib = compiler.find_library('ctpp2', dirs:ctpp2_lib_path, required:false)
-    if not ctpp2_lib.found()
-      ctpp2_lib_path = join_paths(ctpp2_prefix_install, 'lib')
-      message(ctpp2_lib_path)
-      ctpp2_lib = compiler.find_library('ctpp2', dirs:ctpp2_lib_path)
-    endif
-    ctpp2_link_args = ['-L'+ctpp2_lib_path, '-lctpp2']
-    if meson.is_cross_build() and host_machine.system() == 'windows'
-      iconv_lib = compiler.find_library('iconv', required:false)
-      if iconv_lib.found()
-        ctpp2_link_args += ['-liconv']
-      endif
-    endif
-    has_ctpp2_dep = true
-    ctpp2_dep = declare_dependency(include_directories:ctpp2_include_dir, link_args:ctpp2_link_args)
-  else
-    message('ctpp2/CTPP2Logger.hpp not found. Compiling without CTPP2 support')
-  endif
+if target_machine.system() == 'windows'
+  add_project_arguments('-DNOMINMAX', language: 'cpp')
 endif
 
-xapian_dep = dependency('xapian-core', required:false, static:static_deps)
+all_deps = [thread_dep, libicu_dep, libzim_dep, pugixml_dep, libcurl_dep, microhttpd_dep, zlib_dep, xapian_dep]
 
-all_deps = [thread_dep, libicu_dep, libzim_dep, xapian_dep, pugixml_dep, libcurl_dep]
-if has_ctpp2_dep
-  all_deps += [ctpp2_dep]
-endif
-
-inc = include_directories('include')
+inc = include_directories('include', extra_include)
 
 conf = configuration_data()
 conf.set('VERSION', '"@0@"'.format(meson.project_version()))
-conf.set('ENABLE_CTPP2', has_ctpp2_dep)
 
 if build_machine.system() == 'windows'
      extra_link_args = ['-lshlwapi', '-lwinmm']
@@ -109,17 +79,7 @@ subdir('static')
 subdir('src')
 subdir('test')
 
-pkg_requires = ['libzim', 'icu-i18n', 'pugixml', 'libcurl']
-if xapian_dep.found()
-    pkg_requires += ['xapian-core']
-endif
-
-if has_ctpp2_dep
-  extra_libs += ctpp2_link_args
-  if ctpp2_include_path != ''
-    extra_cflags = ' -I'+ctpp2_include_path
-  endif
-endif
+pkg_requires = ['libzim', 'icu-i18n', 'pugixml', 'libcurl', 'libmicrohttpd', 'xapian-core']
 
 pkg_conf = configuration_data()
 pkg_conf.set('prefix', get_option('prefix'))

meson_options.txt

@@ -1,4 +1,2 @@
-option('ctpp2-install-prefix', type : 'string', value : '',
-  description : 'Prefix where ctpp libs has been installed')
-option('android', type : 'boolean', value : false,
-  description : 'Do we make a kiwix-lib for android')
+option('wrapper', type:'array', choices:['java', 'android'], value:[],
+  description: 'The wrapper to generate.')

scripts/ctpp2c.sh

@@ -1,8 +0,0 @@
-#!/usr/bin/env bash
-
-
-ctpp2c=$1
-SOURCE=$(pwd)/$2
-DEST=$3
-
-$ctpp2c $SOURCE $DEST

scripts/kiwix-compile-resources

@@ -81,12 +81,12 @@ class Resource:
             data_identifier="_".join([""]+self.identifier),
             resource_content=",\n    ".join(", ".join("{:#04x}".format(i) for i in r) for r in sliced),
             resource_len=len(self.data),
-            namespaces_open=" ".join("namespace {} {{".format(id) for id in self.identifier[:-1]), 
+            namespaces_open=" ".join("namespace {} {{".format(id) for id in self.identifier[:-1]),
             namespaces_close=" ".join(["}"]*(len(self.identifier)-1)),
             identifier=self.identifier[-1],
             env_identifier="RES_"+"_".join(self.identifier)+"_PATH"
         )
-    
+
     def dump_getter(self):
         return resource_getter_template.format(
             common_name=self.filename,
@@ -95,11 +95,11 @@ class Resource:
 
     def dump_decl(self):
         return resource_decl_template.format(
-            namespaces_open=" ".join("namespace {} {{".format(id) for id in self.identifier[:-1]), 
+            namespaces_open=" ".join("namespace {} {{".format(id) for id in self.identifier[:-1]),
             namespaces_close=" ".join(["}"]*(len(self.identifier)-1)),
             identifier=self.identifier[-1]
         )
-    
+
 
 
 master_c_template = """//This file is automaically generated. Do not modify it.
@@ -113,7 +113,7 @@ static std::string init_resource(const char* name, const unsigned char* content,
     char * resPath = getenv(name);
     if (NULL == resPath)
         return std::string(reinterpret_cast<const char*>(content), len);
-    
+
     std::ifstream ifs(resPath);
     if (!ifs.good())
         return std::string(reinterpret_cast<const char*>(content), len);
@@ -137,7 +137,7 @@ def gen_c_file(resources, basename):
        include_file=basename,
        basename=to_identifier(basename)
     )
- 
+
 
 
 master_h_template = """//This file is automaically generated. Do not modify it.

scripts/kiwix-compile-resources.1

@@ -0,0 +1,20 @@
+.TH KIWIX-COMPILE-RESOURCES "1" "August 2017" "Kiwix" "User Commands"
+.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
+.SH DESCRIPTION
+.TP
+resource_file
+The list of 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
+\fB\-\-hfile\fR HFILE
+The h file name to generate
+.SH AUTHOR
+Matthieu Gautier <mgautier@kymeria.fr>

scripts/meson.build

@@ -1,5 +1,6 @@
 
 res_compiler = find_program('kiwix-compile-resources')
-intermediate_ctpp2c = find_program('ctpp2c.sh')
 
 install_data(res_compiler.path(), install_dir:get_option('bindir'))
+
+install_man('kiwix-compile-resources.1')

src/android/meson.build

@@ -1,26 +0,0 @@
-
-kiwix_jni = custom_target('jni',
-  input: ['org/kiwix/kiwixlib/JNIKiwix.java',
-          'org/kiwix/kiwixlib/JNIKiwixReader.java',
-          'org/kiwix/kiwixlib/JNIKiwixSearcher.java',
-          'org/kiwix/kiwixlib/JNIKiwixInt.java',
-          'org/kiwix/kiwixlib/JNIKiwixString.java',
-          'org/kiwix/kiwixlib/JNIKiwixBool.java',
-          'org/kiwix/kiwixlib/JNIKiwixException.java',
-          'org/kiwix/kiwixlib/Pair.java'],
-  output: ['org_kiwix_kiwixlib_JNIKiwix.h',
-           'org_kiwix_kiwixlib_JNIKiwixReader.h',
-           'org_kiwix_kiwixlib_JNIKiwixSearcher.h',
-           'org_kiwix_kiwixlib_JNIKiwixSearcher_Result.h'],
-  command:['javac', '-d', '@OUTDIR@', '-h', '@OUTDIR@', '@INPUT@']
-)
-
-kiwix_sources += [
-    'android/kiwix.cpp',
-    'android/kiwixreader.cpp',
-    'android/kiwixsearcher.cpp',
-    kiwix_jni]
-
-install_subdir('org', install_dir: 'kiwix-lib/java')
-install_subdir('res', install_dir: 'kiwix-lib')
-install_data('AndroidManifest.xml', install_dir: 'kiwix-lib')

src/android/utils.h

@@ -1,150 +0,0 @@
-/*
- * Copyright (C) 2013 Emmanuel Engelhart <kelson@kiwix.org>
- * Copyright (C) 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 _ANDROID_JNI_UTILS_H
-#define _ANDROID_JNI_UTILS_H
-
-#include <jni.h>
-
-#include <pthread.h>
-#include <string>
-
-extern pthread_mutex_t globalLock;
-
-inline jfieldID getHandleField(JNIEnv* env, jobject obj)
-{
-  jclass c = env->GetObjectClass(obj);
-  // J is the type signature for long:
-  return env->GetFieldID(c, "nativeHandle", "J");
-}
-
-class Lock
-{
- protected:
-  pthread_mutex_t* lock;
-
- public:
-  Lock() : lock(&globalLock) { pthread_mutex_lock(lock); }
-  Lock(const Lock&) = delete;
-  Lock& operator=(const Lock&) = delete;
-  Lock(Lock&& other) : lock(&globalLock) { other.lock = nullptr; }
-  virtual ~Lock()
-  {
-    if (lock) {
-      pthread_mutex_unlock(lock);
-    }
-  }
-};
-
-template <class T>
-class LockedHandle;
-
-template <class T>
-class Handle
-{
- protected:
-  T* h;
-
- public:
-  Handle(T* h) : h(h){};
-
-  // No destructor. This must and will be handled by dispose method.
-
-  static LockedHandle<T> getHandle(JNIEnv* env, jobject obj)
-  {
-    jlong handle = env->GetLongField(obj, getHandleField(env, obj));
-    return LockedHandle<T>(reinterpret_cast<Handle<T>*>(handle));
-  }
-
-  static void dispose(JNIEnv* env, jobject obj)
-  {
-    auto lHandle = getHandle(env, obj);
-    auto handle = lHandle.h;
-    delete handle->h;
-    delete handle;
-  }
-  friend class LockedHandle<T>;
-};
-
-template <class T>
-struct LockedHandle : public Lock {
-  Handle<T>* h;
-  LockedHandle(Handle<T>* h) : h(h) {}
-  T* operator->() { return h->h; }
-  T* operator*() { return h->h; }
-  operator bool() const { return (h->h != nullptr); }
-};
-
-/* c2jni type conversion functions */
-inline jboolean c2jni(const bool& val) { return val ? JNI_TRUE : JNI_FALSE; }
-inline jstring c2jni(const std::string& val, JNIEnv* env)
-{
-  return env->NewStringUTF(val.c_str());
-}
-
-inline jint c2jni(const int val) { return (jint)val; }
-inline jint c2jni(const unsigned val) { return (unsigned)val; }
-/* jni2c type conversion functions */
-inline bool jni2c(const jboolean& val) { return val == JNI_TRUE; }
-inline std::string jni2c(const jstring& val, JNIEnv* env)
-{
-  const char* chars = env->GetStringUTFChars(val, 0);
-  std::string ret(chars);
-  env->ReleaseStringUTFChars(val, chars);
-  return ret;
-}
-
-inline int jni2c(const jint val) { return (int)val; }
-/* Method to deal with variable passed by reference */
-inline void setStringObjValue(const std::string& value,
-                              const jobject obj,
-                              JNIEnv* env)
-{
-  jclass objClass = env->GetObjectClass(obj);
-  jfieldID objFid = env->GetFieldID(objClass, "value", "Ljava/lang/String;");
-  env->SetObjectField(obj, objFid, c2jni(value, env));
-}
-
-inline void setIntObjValue(const int value, const jobject obj, JNIEnv* env)
-{
-  jclass objClass = env->GetObjectClass(obj);
-  jfieldID objFid = env->GetFieldID(objClass, "value", "I");
-  env->SetIntField(obj, objFid, value);
-}
-
-inline void setBoolObjValue(const bool value, const jobject obj, JNIEnv* env)
-{
-  jclass objClass = env->GetObjectClass(obj);
-  jfieldID objFid = env->GetFieldID(objClass, "value", "Z");
-  env->SetIntField(obj, objFid, c2jni(value));
-}
-
-inline void setPairObjValue(const std::string& filename, const int offset,
-                            const jobject obj, JNIEnv* env)
-{
-  jclass objClass = env->GetObjectClass(obj);
-  jfieldID filenameFid = env->GetFieldID(objClass, "filename", "Ljava/lang/String;");
-  env->SetObjectField(obj, filenameFid, c2jni(filename, env));
-  jfieldID offsetFid = env->GetFieldID(objClass, "offset", "I");
-  env->SetIntField(obj, offsetFid, offset);
-}
-
-#endif // _ANDROID_JNI_UTILS_H

src/aria2.cpp

@@ -2,23 +2,40 @@
 
 #include "aria2.h"
 #include "xmlrpc.h"
+#include <iostream>
 #include <sstream>
 #include <thread>
 #include <chrono>
-#include <common/otherTools.h>
-#include <common/pathTools.h>
+#include <tools/otherTools.h>
+#include <tools/pathTools.h>
+#include <tools/stringTools.h>
 #include <downloader.h> // For AriaError
 
+#ifdef _WIN32
+# define ARIA2_CMD "aria2c.exe"
+#else
+# define ARIA2_CMD "aria2c"
+# include <unistd.h>
+#endif
+
+
+#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; \
+  }
+
 namespace kiwix {
 
 Aria2::Aria2():
   mp_aria(nullptr),
   m_port(42042),
   m_secret("kiwixariarpc"),
-  mp_curl(nullptr),
-  m_lock(PTHREAD_MUTEX_INITIALIZER)
+  m_curlErrorBuffer(new char[CURL_ERROR_SIZE]),
+  mp_curl(nullptr)
 {
   m_downloadDir = getDataDirectory();
+  makeDirectory(m_downloadDir);
   std::vector<const char*> callCmd;
 
   std::string rpc_port = "--rpc-listen-port=" + to_string(m_port);
@@ -36,11 +53,17 @@ Aria2::Aria2():
   std::string rpc_secret = "--rpc-secret=" + m_secret;
   m_secret = "token:"+m_secret;
 
-#ifdef _WIN32
-  callCmd.push_back("aria2c.exe");
-#else
-  callCmd.push_back("aria2c");
-#endif
+  std::string aria2cmd = appendToDirectory(
+    removeLastPathElement(getExecutablePath(true)),
+    ARIA2_CMD);
+  if (fileExists(aria2cmd)) {
+    // A local aria2c exe exists (packaged with kiwix-desktop), use it.
+    callCmd.push_back(aria2cmd.c_str());
+  } else {
+    // Try to use a potential installed aria2c.
+    callCmd.push_back(ARIA2_CMD);
+  }
+  callCmd.push_back("--follow-metalink=mem");
   callCmd.push_back("--enable-rpc");
   callCmd.push_back(rpc_secret.c_str());
   callCmd.push_back(rpc_port.c_str());
@@ -61,28 +84,38 @@ Aria2::Aria2():
   callCmd.push_back("--max-concurrent-downloads=42");
   callCmd.push_back("--rpc-max-request-size=6M");
   callCmd.push_back("--file-allocation=none");
+  std::string launchCmd;
+  for (auto &cmd : callCmd) {
+      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());
 
   int watchdog = 50;
   while(--watchdog) {
-    std::this_thread::sleep_for(std::chrono::microseconds(100));
+    sleep(10);
+    m_curlErrorBuffer[0] = 0;
     auto res = curl_easy_perform(mp_curl);
     if (res == CURLE_OK) {
       break;
+    } else if (watchdog == 1) {
+      LOG_ARIA_ERROR();
     }
   }
   if (!watchdog) {
     curl_easy_cleanup(mp_curl);
-    throw std::runtime_error("Cannot connect to aria2c rpc");
+    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);
 }
 
@@ -94,47 +127,56 @@ void Aria2::close()
 
 size_t write_callback_to_iss(char* ptr, size_t size, size_t nmemb, void* userdata)
 {
-  auto str = static_cast<std::stringstream*>(userdata);
-  str->write(ptr, nmemb);
+  auto outStream = static_cast<std::stringstream*>(userdata);
+  outStream->write(ptr, nmemb);
   return nmemb;
 }
 
 std::string Aria2::doRequest(const MethodCall& methodCall)
 {
-  pthread_mutex_lock(&m_lock);
   auto requestContent = methodCall.toString();
-  std::stringstream stringstream;
+  std::stringstream outStream;
   CURLcode res;
-  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, &stringstream);
-  res = curl_easy_perform(mp_curl);
-  if (res == CURLE_OK) {
-    long response_code;
+  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);
-    pthread_mutex_unlock(&m_lock);
-    if (response_code != 200) {
-      throw std::runtime_error("Invalid return code from aria");
-    }
-    auto responseContent = stringstream.str();
-    MethodResponse response(responseContent);
-    if (response.isFault()) {
-      throw AriaError(response.getFault().getFaultString());
-    }
-    return responseContent;
   }
-  pthread_mutex_unlock(&m_lock);
-  throw std::runtime_error("Cannot perform request");
+
+  auto responseContent = outStream.str();
+  if (response_code != 200) {
+    std::cerr << "ERROR: Invalid return code (" << response_code << ") from aria :" << std::endl;
+    std::cerr << responseContent << std::endl;
+    throw std::runtime_error("Invalid return code from aria");
+  }
+
+  MethodResponse response(responseContent);
+  if (response.isFault()) {
+    throw AriaError(response.getFault().getFaultString());
+  }
+  return responseContent;
 }
 
-std::string Aria2::addUri(const std::vector<std::string>& uris)
+std::string Aria2::addUri(const std::vector<std::string>& uris, const std::vector<std::pair<std::string, std::string>>& options)
 {
   MethodCall methodCall("aria2.addUri", m_secret);
   auto uriParams = methodCall.newParamValue().getArray();
   for (auto& uri : uris) {
     uriParams.addValue().set(uri);
   }
+  for (auto& option : options) {
+    methodCall.newParamValue().getStruct().addMember(option.first).getValue().set(option.second);
+  }
   auto ret = doRequest(methodCall);
   MethodResponse response(ret);
   return response.getParamValue(0).getAsS();
@@ -158,7 +200,6 @@ std::vector<std::string> Aria2::tellActive()
   MethodCall methodCall("aria2.tellActive", m_secret);
   auto statusArray = methodCall.newParamValue().getArray();
   statusArray.addValue().set(std::string("gid"));
-  statusArray.addValue().set(std::string("following"));
   auto responseContent = doRequest(methodCall);
   MethodResponse response(responseContent);
   std::vector<std::string> activeGID;
@@ -173,6 +214,27 @@ std::vector<std::string> Aria2::tellActive()
   return activeGID;
 }
 
+std::vector<std::string> Aria2::tellWaiting()
+{
+  MethodCall methodCall("aria2.tellWaiting", m_secret);
+  methodCall.newParamValue().set(0);
+  methodCall.newParamValue().set(99); // max number of downloads to be returned, don't know how to set this properly assumed that there will not be more than 99 paused downloads.
+  auto statusArray = methodCall.newParamValue().getArray();
+  statusArray.addValue().set(std::string("gid"));
+  auto responseContent = doRequest(methodCall);
+  MethodResponse response(responseContent);
+  std::vector<std::string> waitingGID;
+  int index = 0;
+  while(true) {
+    try {
+      auto structNode = response.getParamValue(0).getArray().getValue(index++).getStruct();
+      auto gidNode = structNode.getMember("gid");
+      waitingGID.push_back(gidNode.getValue().getAsS());
+    } catch (InvalidRPCNode& e) { break; }
+  }
+  return waitingGID;
+}
+
 void Aria2::saveSession()
 {
   MethodCall methodCall("aria2.saveSession", m_secret);
@@ -186,5 +248,25 @@ void Aria2::shutdown()
   doRequest(methodCall);
 }
 
+void Aria2::pause(const std::string& gid)
+{
+    MethodCall methodCall("aria2.pause", m_secret);
+    methodCall.newParamValue().set(gid);
+    doRequest(methodCall);
+}
+
+void Aria2::unpause(const std::string& gid)
+{
+    MethodCall methodCall("aria2.unpause", m_secret);
+    methodCall.newParamValue().set(gid);
+    doRequest(methodCall);
+}
+
+void Aria2::remove(const std::string& gid)
+{
+    MethodCall methodCall("aria2.remove", m_secret);
+    methodCall.newParamValue().set(gid);
+    doRequest(methodCall);
+}
 
 } // end namespace kiwix

src/aria2.h

@@ -12,8 +12,8 @@
 #include "xmlrpc.h"
 
 #include <memory>
+#include <mutex>
 #include <curl/curl.h>
-#include <pthread.h>
 
 namespace kiwix {
 
@@ -24,8 +24,9 @@ class Aria2
     int m_port;
     std::string m_secret;
     std::string m_downloadDir;
+    std::unique_ptr<char[]> m_curlErrorBuffer;
     CURL* mp_curl;
-    pthread_mutex_t m_lock;
+    std::mutex m_lock;
 
     std::string doRequest(const MethodCall& methodCall);
 
@@ -34,11 +35,15 @@ class Aria2
     virtual ~Aria2();
     void close();
 
-    std::string addUri(const std::vector<std::string>& uri);
+    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);
     std::vector<std::string> tellActive();
+    std::vector<std::string> tellWaiting();
     void saveSession();
     void shutdown();
+    void pause(const std::string& gid);
+    void unpause(const std::string& gid);
+    void remove(const std::string& gid);
 };
 
 }; //end namespace kiwix

src/book.cpp

@@ -20,16 +20,19 @@
 #include "book.h"
 #include "reader.h"
 
-#include "common/base64.h"
-#include "common/regexTools.h"
-#include "common/networkTools.h"
+#include "tools/base64.h"
+#include "tools/regexTools.h"
+#include "tools/networkTools.h"
+#include "tools/otherTools.h"
 
 #include <pugixml.hpp>
 
 namespace kiwix
 {
 /* Constructor */
-Book::Book() : m_readOnly(false)
+Book::Book() :
+  m_pathValid(false),
+  m_readOnly(false)
 {
 }
 /* Destructor */
@@ -42,52 +45,56 @@ bool Book::update(const kiwix::Book& other)
   if (m_readOnly)
     return false;
 
+  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;
 
-  if (m_path.empty()) {
-    m_path = other.m_path;
-  }
+  m_downloadId = other.m_downloadId;
 
-  if (m_url.empty()) {
-    m_url = other.m_url;
-  }
-
-  if (m_tags.empty()) {
-    m_tags = other.m_tags;
-  }
-
-  if (m_name.empty()) {
-    m_name = other.m_name;
-  }
-
-  if (m_indexPath.empty()) {
-    m_indexPath = other.m_indexPath;
-    m_indexType = other.m_indexType;
-  }
-
-  if (m_faviconMimeType.empty()) {
-    m_favicon = other.m_favicon;
-    m_faviconMimeType = other.m_faviconMimeType;
-  }
   return true;
 }
 
 void Book::update(const kiwix::Reader& reader)
 {
   m_path = reader.getZimFilePath();
+  m_pathValid = true;
   m_id = reader.getId();
+  m_title = reader.getTitle();
   m_description = reader.getDescription();
   m_language = reader.getLanguage();
-  m_date = reader.getDate();
   m_creator = reader.getCreator();
   m_publisher = reader.getPublisher();
-  m_title = reader.getTitle();
+  m_date = reader.getDate();
   m_name = reader.getName();
+  m_flavour = reader.getFlavour();
   m_tags = reader.getTags();
+  m_category = getCategoryFromTags();
   m_origId = reader.getOrigId();
   m_articleCount = reader.getArticleCount();
   m_mediaCount = reader.getMediaCount();
-  m_size = reader.getFileSize() << 10;
+  m_size = static_cast<uint64_t>(reader.getFileSize()) << 10;
+  m_pathValid = true;
 
   reader.getFavicon(m_favicon, m_faviconMimeType);
 }
@@ -101,36 +108,41 @@ void Book::updateFromXml(const pugi::xml_node& node, const std::string& baseDir)
     path = computeAbsolutePath(baseDir, path);
   }
   m_path = path;
-  path = ATTR("indexPath");
-  if (!path.empty()) {
-    if (isRelativePath(path)) {
-      path = computeAbsolutePath(baseDir, path);
-    }
-    m_indexPath = path;
-    m_indexType = XAPIAN;
-  }
+  m_pathValid = fileExists(path);
   m_title = ATTR("title");
-  m_name = ATTR("name");
-  m_tags = ATTR("tags");
   m_description = ATTR("description");
   m_language = ATTR("language");
-  m_date = ATTR("date");
   m_creator = ATTR("creator");
   m_publisher = ATTR("publisher");
+  m_date = ATTR("date");
   m_url = ATTR("url");
+  m_name = ATTR("name");
+  m_flavour = ATTR("flavour");
+  m_tags = ATTR("tags");
   m_origId = ATTR("origId");
   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");
   try {
     m_downloadId = ATTR("downloadId");
   } catch(...) {}
+  const auto catattr = node.attribute("category");
+  m_category = catattr.empty() ? getCategoryFromTags() : catattr.value();
 }
 #undef ATTR
 
 
+static std::string fromOpdsDate(const std::string& date)
+{
+  //The opds date use the standard <YYYY>-<MM>-<DD>T<HH>:<mm>:<SS>Z
+  //and we want <YYYY>-<MM>-<DD>. That's easy, let's take the first 10 char
+  return date.substr(0, 10);
+}
+
+
 #define VALUE(name) node.child(name).child_value()
 void Book::updateFromOpds(const pugi::xml_node& node, const std::string& urlHost)
 {
@@ -138,11 +150,20 @@ void Book::updateFromOpds(const pugi::xml_node& node, const std::string& urlHost
   if (!m_id.compare(0, 9, "urn:uuid:")) {
     m_id.erase(0, 9);
   }
+  // No path on opds.
   m_title = VALUE("title");
-  m_description = VALUE("description");
+  m_description = VALUE("summary");
   m_language = VALUE("language");
-  m_date = VALUE("updated");
   m_creator = node.child("author").child("name").child_value();
+  m_publisher = node.child("publisher").child("name").child_value();
+  m_date = fromOpdsDate(VALUE("updated"));
+  m_name = VALUE("name");
+  m_flavour = VALUE("flavour");
+  m_tags = VALUE("tags");
+  const auto catnode = node.child("category");
+  m_category = catnode.empty() ? getCategoryFromTags() : catnode.child_value();
+  m_articleCount = strtoull(VALUE("articleCount"), 0, 0);
+  m_mediaCount = strtoull(VALUE("mediaCount"), 0, 0);
   for(auto linkNode = node.child("link"); linkNode;
            linkNode = linkNode.next_sibling("link")) {
     std::string rel = linkNode.attribute("rel").value();
@@ -160,7 +181,7 @@ void Book::updateFromOpds(const pugi::xml_node& node, const std::string& urlHost
 }
 #undef VALUE
 
-std::string Book::getHumanReadableIdFromPath()
+std::string Book::getHumanReadableIdFromPath() const
 {
   std::string id = m_path;
   if (!id.empty()) {
@@ -186,13 +207,6 @@ void Book::setPath(const std::string& path)
    : path;
 }
 
-void Book::setIndexPath(const std::string& indexPath)
-{
-  m_indexPath = isRelativePath(indexPath)
-    ? computeAbsolutePath(getCurrentDirectory(), indexPath)
-    : indexPath;
-}
-
 const std::string& Book::getFavicon() const {
   if (m_favicon.empty() && !m_faviconUrl.empty()) {
     try {
@@ -204,4 +218,29 @@ const std::string& Book::getFavicon() const {
   return m_favicon;
 }
 
+std::string Book::getTagStr(const std::string& tagName) const {
+  return getTagValueFromTagList(convertTags(m_tags), tagName);
+}
+
+bool Book::getTagBool(const std::string& tagName) const {
+  return convertStrToBool(getTagStr(tagName));
+}
+
+std::string Book::getCategory() const
+{
+  return m_category;
+}
+
+std::string Book::getCategoryFromTags() const
+{
+  try
+  {
+    return getTagStr("category");
+  }
+  catch ( const std::out_of_range& )
+  {
+    return "";
+  }
+}
+
 }

src/bookmark.cpp

@@ -0,0 +1,47 @@
+/*
+ * Copyright 2018 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 "bookmark.h"
+
+#include <pugixml.hpp>
+
+namespace kiwix
+{
+/* Constructor */
+Bookmark::Bookmark()
+{
+}
+
+/* Destructor */
+Bookmark::~Bookmark()
+{
+}
+
+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_language = bookNode.child("language").child_value();
+  m_date = bookNode.child("date").child_value();
+  m_title = node.child("title").child_value();
+  m_url = node.child("url").child_value();
+}
+
+}

src/common/networkTools.cpp

@@ -1,209 +0,0 @@
-/*
- * Copyright 2012 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 <common/networkTools.h>
-
-#ifdef _WIN32
-#include <winsock2.h>
-#include <ws2tcpip.h>
-#else
-#include <net/if.h>
-#include <netdb.h>
-#include <stdio.h>
-#include <stdlib.h>
-#include <string.h>
-#include <sys/ioctl.h>
-#include <sys/socket.h>
-#include <sys/types.h>
-#include <unistd.h>
-#endif
-
-#include <curl/curl.h>
-
-#include <sstream>
-#include <iostream>
-
-
-std::map<std::string, std::string> kiwix::getNetworkInterfaces()
-{
-  std::map<std::string, std::string> interfaces;
-
-#ifdef _WIN32
-  SOCKET sd = WSASocket(AF_INET, SOCK_DGRAM, 0, 0, 0, 0);
-  if (sd == (SOCKET)SOCKET_ERROR) {
-    std::cerr << "Failed to get a socket. Error " << WSAGetLastError()
-              << std::endl;
-    return interfaces;
-  }
-
-  INTERFACE_INFO InterfaceList[20];
-  unsigned long nBytesReturned;
-  if (WSAIoctl(sd,
-               SIO_GET_INTERFACE_LIST,
-               0,
-               0,
-               &InterfaceList,
-               sizeof(InterfaceList),
-               &nBytesReturned,
-               0,
-               0)
-      == SOCKET_ERROR) {
-    std::cerr << "Failed calling WSAIoctl: error " << WSAGetLastError()
-              << std::endl;
-    return interfaces;
-  }
-
-  int nNumInterfaces = nBytesReturned / sizeof(INTERFACE_INFO);
-  for (int i = 0; i < nNumInterfaces; ++i) {
-    sockaddr_in* pAddress;
-    pAddress = (sockaddr_in*)&(InterfaceList[i].iiAddress);
-
-    /* Add to the map */
-    std::string interfaceName = std::string(inet_ntoa(pAddress->sin_addr));
-    std::string interfaceIp = std::string(inet_ntoa(pAddress->sin_addr));
-    interfaces.insert(
-        std::pair<std::string, std::string>(interfaceName, interfaceIp));
-  }
-#else
-  /* Get Network interfaces information */
-  char buf[16384];
-  struct ifconf ifconf;
-  int fd = socket(PF_INET, SOCK_DGRAM, 0); /* Only IPV4 */
-  ifconf.ifc_len = sizeof buf;
-  ifconf.ifc_buf = buf;
-  if (ioctl(fd, SIOCGIFCONF, &ifconf) != 0) {
-    perror("ioctl(SIOCGIFCONF)");
-    exit(EXIT_FAILURE);
-  }
-
-  /* Go through each interface */
-  int i;
-  size_t len;
-  struct ifreq* ifreq;
-  ifreq = ifconf.ifc_req;
-  for (i = 0; i < ifconf.ifc_len;) {
-    if (ifreq->ifr_addr.sa_family == AF_INET) {
-      /* Get the network interface ip */
-      char host[128] = {0};
-      const int error = getnameinfo(&(ifreq->ifr_addr),
-                                    sizeof ifreq->ifr_addr,
-                                    host,
-                                    sizeof host,
-                                    0,
-                                    0,
-                                    NI_NUMERICHOST);
-      if (!error) {
-        std::string interfaceName = std::string(ifreq->ifr_name);
-        std::string interfaceIp = std::string(host);
-        /* Add to the map */
-        interfaces.insert(
-            std::pair<std::string, std::string>(interfaceName, interfaceIp));
-      } else {
-        perror("getnameinfo()");
-      }
-    }
-
-/* some systems have ifr_addr.sa_len and adjust the length that
- * way, but not mine. weird */
-#ifndef __linux__
-    len = IFNAMSIZ + ifreq->ifr_addr.sa_len;
-#else
-    len = sizeof *ifreq;
-#endif
-    ifreq = (struct ifreq*)((char*)ifreq + len);
-    i += len;
-  }
-#endif
-  return interfaces;
-}
-
-std::string kiwix::getBestPublicIp()
-{
-  std::map<std::string, std::string> interfaces = kiwix::getNetworkInterfaces();
-
-#ifndef _WIN32
-  const char* const prioritizedNames[]
-      = {"eth0", "eth1", "wlan0", "wlan1", "en0", "en1"};
-  const int count = (sizeof prioritizedNames) / (sizeof prioritizedNames[0]);
-  for (int i = 0; i < count; ++i) {
-    std::map<std::string, std::string>::const_iterator it
-        = interfaces.find(prioritizedNames[i]);
-    if (it != interfaces.end()) {
-      return it->second;
-    }
-  }
-#endif
-
-  for (std::map<std::string, std::string>::iterator iter = interfaces.begin();
-       iter != interfaces.end();
-       ++iter) {
-    std::string interfaceIp = iter->second;
-    if (interfaceIp.length() >= 7 && interfaceIp.substr(0, 7) == "192.168") {
-      return interfaceIp;
-    }
-  }
-
-  for (std::map<std::string, std::string>::iterator iter = interfaces.begin();
-       iter != interfaces.end();
-       ++iter) {
-    std::string interfaceIp = iter->second;
-    if (interfaceIp.length() >= 7 && interfaceIp.substr(0, 7) == "172.16.") {
-      return interfaceIp;
-    }
-  }
-
-  for (std::map<std::string, std::string>::iterator iter = interfaces.begin();
-       iter != interfaces.end();
-       ++iter) {
-    std::string interfaceIp = iter->second;
-    if (interfaceIp.length() >= 3 && interfaceIp.substr(0, 3) == "10.") {
-      return interfaceIp;
-    }
-  }
-
-  return "127.0.0.1";
-}
-
-size_t write_callback_to_iss(char* ptr, size_t size, size_t nmemb, void* userdata)
-{
-  auto str = static_cast<std::stringstream*>(userdata);
-  str->write(ptr, nmemb);
-  return nmemb;
-}
-
-std::string kiwix::download(const std::string& url) {
-  auto curl = curl_easy_init();
-  std::stringstream ss;
-  curl_easy_setopt(curl, CURLOPT_URL, url.c_str());
-  curl_easy_setopt(curl, CURLOPT_HTTPGET, 1L);
-  curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, &write_callback_to_iss);
-  curl_easy_setopt(curl, CURLOPT_WRITEDATA, &ss);
-  auto res = curl_easy_perform(curl);
-  if (res != CURLE_OK) {
-    curl_easy_cleanup(curl);
-    throw std::runtime_error("Cannot perform request");
-  }
-  long response_code;
-  curl_easy_getinfo(curl, CURLINFO_RESPONSE_CODE, &response_code);
-  curl_easy_cleanup(curl);
-  if (response_code != 200) {
-    throw std::runtime_error("Invalid return code from server");
-  }
-  return ss.str();
-}

src/common/otherTools.cpp

@@ -1,300 +0,0 @@
-/*
- * Copyright 2014 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 <common/otherTools.h>
-#include <map>
-
-std::map<std::string, std::string> codeisomapping {
-  { "af", "afg" },
-  { "ax", "ala" },
-  { "al", "alb" },
-  { "dz", "dza" },
-  { "as", "asm" },
-  { "ad", "and" },
-  { "ao", "ago" },
-  { "ai", "aia" },
-  { "aq", "ata" },
-  { "ag", "atg" },
-  { "ar", "arg" },
-  { "am", "arm" },
-  { "aw", "abw" },
-  { "au", "aus" },
-  { "at", "aut" },
-  { "az", "aze" },
-  { "bs", "bhs" },
-  { "bh", "bhr" },
-  { "bd", "bgd" },
-  { "bb", "brb" },
-  { "by", "blr" },
-  { "be", "bel" },
-  { "bz", "blz" },
-  { "bj", "ben" },
-  { "bm", "bmu" },
-  { "bt", "btn" },
-  { "bo", "bol" },
-  { "ba", "bih" },
-  { "bw", "bwa" },
-  { "bv", "bvt" },
-  { "br", "bra" },
-  { "vg", "vgb" },
-  { "io", "iot" },
-  { "bn", "brn" },
-  { "bg", "bgr" },
-  { "bf", "bfa" },
-  { "bi", "bdi" },
-  { "kh", "khm" },
-  { "cm", "cmr" },
-  { "ca", "can" },
-  { "cv", "cpv" },
-  { "ky", "cym" },
-  { "cf", "caf" },
-  { "td", "tcd" },
-  { "cl", "chl" },
-  { "cn", "chn" },
-  { "hk", "hkg" },
-  { "mo", "mac" },
-  { "cx", "cxr" },
-  { "cc", "cck" },
-  { "co", "col" },
-  { "km", "com" },
-  { "cg", "cog" },
-  { "cd", "cod" },
-  { "ck", "cok" },
-  { "cr", "cri" },
-  { "ci", "civ" },
-  { "hr", "hrv" },
-  { "cu", "cub" },
-  { "cy", "cyp" },
-  { "cz", "cze" },
-  { "dk", "dnk" },
-  { "dj", "dji" },
-  { "dm", "dma" },
-  { "do", "dom" },
-  { "ec", "ecu" },
-  { "eg", "egy" },
-  { "sv", "slv" },
-  { "gq", "gnq" },
-  { "er", "eri" },
-  { "ee", "est" },
-  { "et", "eth" },
-  { "fk", "flk" },
-  { "fo", "fro" },
-  { "fj", "fji" },
-  { "fi", "fin" },
-  { "fr", "fra" },
-  { "gf", "guf" },
-  { "pf", "pyf" },
-  { "tf", "atf" },
-  { "ga", "gab" },
-  { "gm", "gmb" },
-  { "ge", "geo" },
-  { "de", "deu" },
-  { "gh", "gha" },
-  { "gi", "gib" },
-  { "gr", "grc" },
-  { "gl", "grl" },
-  { "gd", "grd" },
-  { "gp", "glp" },
-  { "gu", "gum" },
-  { "gt", "gtm" },
-  { "gg", "ggy" },
-  { "gn", "gin" },
-  { "gw", "gnb" },
-  { "gy", "guy" },
-  { "ht", "hti" },
-  { "hm", "hmd" },
-  { "va", "vat" },
-  { "hn", "hnd" },
-  { "hu", "hun" },
-  { "is", "isl" },
-  { "in", "ind" },
-  { "id", "idn" },
-  { "ir", "irn" },
-  { "iq", "irq" },
-  { "ie", "irl" },
-  { "im", "imn" },
-  { "il", "isr" },
-  { "it", "ita" },
-  { "jm", "jam" },
-  { "jp", "jpn" },
-  { "je", "jey" },
-  { "jo", "jor" },
-  { "kz", "kaz" },
-  { "ke", "ken" },
-  { "ki", "kir" },
-  { "kp", "prk" },
-  { "kr", "kor" },
-  { "kw", "kwt" },
-  { "kg", "kgz" },
-  { "la", "lao" },
-  { "lv", "lva" },
-  { "lb", "lbn" },
-  { "ls", "lso" },
-  { "lr", "lbr" },
-  { "ly", "lby" },
-  { "li", "lie" },
-  { "lt", "ltu" },
-  { "lu", "lux" },
-  { "mk", "mkd" },
-  { "mg", "mdg" },
-  { "mw", "mwi" },
-  { "my", "mys" },
-  { "mv", "mdv" },
-  { "ml", "mli" },
-  { "mt", "mlt" },
-  { "mh", "mhl" },
-  { "mq", "mtq" },
-  { "mr", "mrt" },
-  { "mu", "mus" },
-  { "yt", "myt" },
-  { "mx", "mex" },
-  { "fm", "fsm" },
-  { "md", "mda" },
-  { "mc", "mco" },
-  { "mn", "mng" },
-  { "me", "mne" },
-  { "ms", "msr" },
-  { "ma", "mar" },
-  { "mz", "moz" },
-  { "mm", "mmr" },
-  { "na", "nam" },
-  { "nr", "nru" },
-  { "np", "npl" },
-  { "nl", "nld" },
-  { "an", "ant" },
-  { "nc", "ncl" },
-  { "nz", "nzl" },
-  { "ni", "nic" },
-  { "ne", "ner" },
-  { "ng", "nga" },
-  { "nu", "niu" },
-  { "nf", "nfk" },
-  { "mp", "mnp" },
-  { "no", "nor" },
-  { "om", "omn" },
-  { "pk", "pak" },
-  { "pw", "plw" },
-  { "ps", "pse" },
-  { "pa", "pan" },
-  { "pg", "png" },
-  { "py", "pry" },
-  { "pe", "per" },
-  { "ph", "phl" },
-  { "pn", "pcn" },
-  { "pl", "pol" },
-  { "pt", "prt" },
-  { "pr", "pri" },
-  { "qa", "qat" },
-  { "re", "reu" },
-  { "ro", "rou" },
-  { "ru", "rus" },
-  { "rw", "rwa" },
-  { "bl", "blm" },
-  { "sh", "shn" },
-  { "kn", "kna" },
-  { "lc", "lca" },
-  { "mf", "maf" },
-  { "pm", "spm" },
-  { "vc", "vct" },
-  { "ws", "wsm" },
-  { "sm", "smr" },
-  { "st", "stp" },
-  { "sa", "sau" },
-  { "sn", "sen" },
-  { "rs", "srb" },
-  { "sc", "syc" },
-  { "sl", "sle" },
-  { "sg", "sgp" },
-  { "sk", "svk" },
-  { "si", "svn" },
-  { "sb", "slb" },
-  { "so", "som" },
-  { "za", "zaf" },
-  { "gs", "sgs" },
-  { "ss", "ssd" },
-  { "es", "esp" },
-  { "lk", "lka" },
-  { "sd", "sdn" },
-  { "sr", "sur" },
-  { "sj", "sjm" },
-  { "sz", "swz" },
-  { "se", "swe" },
-  { "ch", "che" },
-  { "sy", "syr" },
-  { "tw", "twn" },
-  { "tj", "tjk" },
-  { "tz", "tza" },
-  { "th", "tha" },
-  { "tl", "tls" },
-  { "tg", "tgo" },
-  { "tk", "tkl" },
-  { "to", "ton" },
-  { "tt", "tto" },
-  { "tn", "tun" },
-  { "tr", "tur" },
-  { "tm", "tkm" },
-  { "tc", "tca" },
-  { "tv", "tuv" },
-  { "ug", "uga" },
-  { "ua", "ukr" },
-  { "ae", "are" },
-  { "gb", "gbr" },
-  { "us", "usa" },
-  { "um", "umi" },
-  { "uy", "ury" },
-  { "uz", "uzb" },
-  { "vu", "vut" },
-  { "ve", "ven" },
-  { "vn", "vnm" },
-  { "vi", "vir" },
-  { "wf", "wlf" },
-  { "eh", "esh" },
-  { "ye", "yem" },
-  { "zm", "zmb" },
-  { "zw", "zwe" }
-};
-
-void kiwix::sleep(unsigned int milliseconds)
-{
-#ifdef _WIN32
-  Sleep(milliseconds);
-#else
-  usleep(1000 * milliseconds);
-#endif
-}
-
-
-struct XmlStringWriter: pugi::xml_writer
-{
-  std::string result;
-  virtual void write(const void* data, size_t size){
-    result.append(static_cast<const char*>(data), size);
-  }
-};
-
-std::string kiwix::nodeToString(pugi::xml_node node)
-{
-  XmlStringWriter writer;
-  node.print(writer, "  ");
-  return writer.result;
-}
-
-std::string kiwix::converta2toa3(const std::string& a2code){
-  return codeisomapping.at(a2code);
-}

src/common/pathTools.cpp

@@ -1,335 +0,0 @@
-/*
- * Copyright 2011-2014 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 <common/pathTools.h>
-
-#ifdef __APPLE__
-#include <limits.h>
-#include <mach-o/dyld.h>
-#elif _WIN32
-#include <direct.h>
-#include <windows.h>
-#include "shlwapi.h"
-#define getcwd _getcwd  // stupid MSFT "deprecation" warning
-#endif
-
-#ifdef _WIN32
-#else
-#include <unistd.h>
-#endif
-
-#ifdef _WIN32
-const std::string SEPARATOR("\\");
-#else
-const std::string SEPARATOR("/");
-#include <unistd.h>
-#endif
-
-#include <stdlib.h>
-
-#ifndef PATH_MAX
-#define PATH_MAX 1024
-#endif
-
-bool isRelativePath(const string& path)
-{
-#ifdef _WIN32
-  return path.empty() || path.substr(1, 2) == ":\\" ? false : true;
-#else
-  return path.empty() || path.substr(0, 1) == "/" ? false : true;
-#endif
-}
-
-string computeRelativePath(const string path, const string absolutePath)
-{
-  std::vector<std::string> pathParts = kiwix::split(path, SEPARATOR);
-  std::vector<std::string> absolutePathParts
-      = kiwix::split(absolutePath, SEPARATOR);
-
-  unsigned int commonCount = 0;
-  while (commonCount < pathParts.size()
-         && commonCount < absolutePathParts.size()
-         && pathParts[commonCount] == absolutePathParts[commonCount]) {
-      commonCount++;
-  }
-
-  string relativePath;
-#ifdef _WIN32
-  /* On Windows you have a token more because the root is represented
-     by a letter */
-  if (commonCount == 0) {
-    relativePath = ".." + SEPARATOR;
-  }
-#endif
-
-  for (unsigned int i = commonCount; i < pathParts.size(); i++) {
-    relativePath += ".." + SEPARATOR;
-  }
-  for (unsigned int i = commonCount; i < absolutePathParts.size(); i++) {
-    relativePath += absolutePathParts[i];
-    relativePath += i + 1 < absolutePathParts.size() ? SEPARATOR : "";
-  }
-  return relativePath;
-}
-
-/* Warning: the relative path must be with slashes */
-string computeAbsolutePath(const string path, const string relativePath)
-{
-  string absolutePath;
-
-  if (path.empty()) {
-    char* path = NULL;
-    size_t size = 0;
-
-#ifdef _WIN32
-    path = _getcwd(path, size);
-#else
-    path = getcwd(path, size);
-#endif
-
-    absolutePath = string(path) + SEPARATOR;
-  } else {
-    absolutePath = path.substr(path.length() - 1, 1) == SEPARATOR
-                       ? path
-                       : path + SEPARATOR;
-  }
-
-#if _WIN32
-  char* cRelativePath = _strdup(relativePath.c_str());
-#else
-  char* cRelativePath = strdup(relativePath.c_str());
-#endif
-  char* token = strtok(cRelativePath, "/");
-
-  while (token != NULL) {
-    if (string(token) == "..") {
-      absolutePath = removeLastPathElement(absolutePath, true, false);
-      token = strtok(NULL, "/");
-    } else if (strcmp(token, ".") && strcmp(token, "")) {
-      absolutePath += string(token);
-      token = strtok(NULL, "/");
-      if (token != NULL) {
-        absolutePath += SEPARATOR;
-      }
-    } else {
-      token = strtok(NULL, "/");
-    }
-  }
-
-  return absolutePath;
-}
-
-string removeLastPathElement(const string path,
-                             const bool removePreSeparator,
-                             const bool removePostSeparator)
-{
-  string newPath = path;
-  size_t offset = newPath.find_last_of(SEPARATOR);
-  if (removePreSeparator &&
-#ifndef _WIN32
-      offset != newPath.find_first_of(SEPARATOR) &&
-#endif
-      offset == newPath.length() - 1) {
-    newPath = newPath.substr(0, offset);
-    offset = newPath.find_last_of(SEPARATOR);
-  }
-  newPath = removePostSeparator ? newPath.substr(0, offset)
-                                : newPath.substr(0, offset + 1);
-  return newPath;
-}
-
-string appendToDirectory(const string& directoryPath, const string& filename)
-{
-  string newPath = directoryPath + SEPARATOR + filename;
-  return newPath;
-}
-
-string getLastPathElement(const string& path)
-{
-  return path.substr(path.find_last_of(SEPARATOR) + 1);
-}
-
-unsigned int getFileSize(const string& path)
-{
-#ifdef _WIN32
-  struct _stat filestatus;
-  _stat(path.c_str(), &filestatus);
-#else
-  struct stat filestatus;
-  stat(path.c_str(), &filestatus);
-#endif
-
-  return filestatus.st_size / 1024;
-}
-
-string getFileSizeAsString(const string& path)
-{
-  ostringstream convert;
-  convert << getFileSize(path);
-  return convert.str();
-}
-
-string getFileContent(const string& path)
-{
-  std::ifstream f(path, std::ios::in|std::ios::ate);
-  std::string content;
-  if (f.is_open()) {
-    auto size = f.tellg();
-    content.reserve(size);
-    f.seekg(0, std::ios::beg);
-    content.assign((std::istreambuf_iterator<char>(f)),
-                    std::istreambuf_iterator<char>());
-  }
-  return content;
-}
-
-bool fileExists(const string& path)
-{
-#ifdef _WIN32
-  return PathFileExists(path.c_str());
-#else
-  bool flag = false;
-  fstream fin;
-  fin.open(path.c_str(), ios::in);
-  if (fin.is_open()) {
-    flag = true;
-  }
-  fin.close();
-  return flag;
-#endif
-}
-
-bool makeDirectory(const string& path)
-{
-#ifdef _WIN32
-  int status = _mkdir(path.c_str());
-#else
-  int status = mkdir(path.c_str(), S_IRWXU | S_IRWXG | S_IROTH | S_IXOTH);
-#endif
-  return status == 0;
-}
-
-string makeTmpDirectory()
-{
-#ifdef _WIN32
-  char cbase[MAX_PATH+1];
-  int base_len = GetTempPath(MAX_PATH+1, cbase);
-  UUID uuid;
-  UuidCreate(&uuid);
-  char* dir_name;
-  UuidToString(&uuid, reinterpret_cast<unsigned char**>(&dir_name));
-  string dir(cbase, base_len);
-  dir += dir_name;
-  _mkdir(dir.c_str());
-  RpcStringFree(reinterpret_cast<unsigned char**>(&dir_name));
-#else
-  string base = "/tmp";
-  auto _template = base + "/kiwix-lib_XXXXXX";
-  char* _template_array = new char[_template.size()+1];
-  memcpy(_template_array, _template.c_str(), _template.size());
-  string dir = mkdtemp(_template_array);
-  delete[] _template_array;
-#endif
-  return dir;
-}
-
-/* Try to create a link and if does not work then make a copy */
-bool copyFile(const string& sourcePath, const string& destPath)
-{
-  try {
-#ifndef _WIN32
-    if (link(sourcePath.c_str(), destPath.c_str()) != 0) {
-#endif
-      std::ifstream infile(sourcePath.c_str(), std::ios_base::binary);
-      std::ofstream outfile(destPath.c_str(), std::ios_base::binary);
-      outfile << infile.rdbuf();
-#ifndef _WIN32
-    }
-#endif
-  } catch (exception& e) {
-    cerr << e.what() << endl;
-    return false;
-  }
-
-  return true;
-}
-
-string getExecutablePath()
-{
-  char binRootPath[PATH_MAX];
-
-#ifdef _WIN32
-  GetModuleFileName(NULL, binRootPath, PATH_MAX);
-  return std::string(binRootPath);
-#elif __APPLE__
-  uint32_t max = (uint32_t)PATH_MAX;
-  _NSGetExecutablePath(binRootPath, &max);
-  return std::string(binRootPath);
-#else
-  ssize_t size = readlink("/proc/self/exe", binRootPath, PATH_MAX);
-  if (size != -1) {
-    return std::string(binRootPath, size);
-  }
-#endif
-
-  return "";
-}
-
-bool writeTextFile(const string& path, const string& content)
-{
-  std::ofstream file;
-  file.open(path.c_str());
-  file << content;
-  file.close();
-  return true;
-}
-
-string getCurrentDirectory()
-{
-  char* a_cwd = getcwd(NULL, 0);
-  string s_cwd(a_cwd);
-  free(a_cwd);
-  return s_cwd;
-}
-
-string getDataDirectory()
-{
-#ifdef _WIN32
-  char* cDataDir = ::getenv("APPDATA");
-#else
-  char* cDataDir = ::getenv("KIWIX_DATA_DIR");
-#endif
-  std::string dataDir = cDataDir==nullptr ? "" : cDataDir;
-  if (!dataDir.empty())
-    return dataDir;
-#ifdef _WIN32
-  cDataDir = ::getenv("USERPROFILE");
-  dataDir = cDataDir==nullptr ? getCurrentDirectory() : cDataDir;
-#else
-  cDataDir = ::getenv("XDG_DATA_HOME");
-  dataDir = cDataDir==nullptr ? "" : cDataDir;
-  if (dataDir.empty()) {
-    cDataDir = ::getenv("HOME");
-    dataDir = cDataDir==nullptr ? getCurrentDirectory() : cDataDir;
-    dataDir = appendToDirectory(dataDir, ".local");
-    dataDir = appendToDirectory(dataDir, "share");
-  }
-#endif
-  return appendToDirectory(dataDir, "kiwix");
-}

src/config.h.in

@@ -1,4 +1,3 @@
 
 #mesondefine VERSION
 
-#mesondefine ENABLE_CTPP2

src/ctpp2/CTPP2VMStringLoader.cpp

@@ -1,210 +0,0 @@
-/*
- * Copyright 2013 Renaud Gaudin <reg@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 <ctpp2/CTPP2VMStringLoader.hpp>
-
-namespace CTPP // C++ Template Engine
-{
-
-//
-// Convert byte order
-//
-static void ConvertExecutable(VMExecutable * oCore)
-{
-	// Code entry point
-	oCore -> entry_point = Swap32(oCore -> entry_point);
-	// Offset of code segment
-	oCore -> code_offset = Swap32(oCore -> code_offset);
-
-	// Code segment size
-	oCore -> code_size   = Swap32(oCore -> code_size);
-
-	// Offset of static text segment
-	oCore -> syscalls_offset    = Swap32(oCore -> syscalls_offset);
-	// Static text segment size
-	oCore -> syscalls_data_size = Swap32(oCore -> syscalls_data_size);
-
-	// Offset of static text index segment
-	oCore -> syscalls_index_offset = Swap32(oCore -> syscalls_index_offset);
-	// Static text index segment size
-	oCore -> syscalls_index_size   = Swap32(oCore -> syscalls_index_size);
-
-	// Offset of static data segment
-	oCore -> static_data_offset    = Swap32(oCore -> static_data_offset);
-
-	// Static data segment size
-	oCore -> static_data_data_size = Swap32(oCore -> static_data_data_size);
-
-	// Offset of static text segment
-	oCore -> static_text_offset    = Swap32(oCore -> static_text_offset);
-	// Static text segment size
-	oCore -> static_text_data_size = Swap32(oCore -> static_text_data_size);
-
-	// Offset of static text index segment
-	oCore -> static_text_index_offset = Swap32(oCore -> static_text_index_offset);
-	// Static text index segment size
-	oCore -> static_text_index_size   = Swap32(oCore -> static_text_index_size);
-
-	// Version 2.2+
-	// Offset of static data bit index
-	oCore -> static_data_bit_index_offset = Swap32(oCore -> static_data_bit_index_offset);
-	/// Offset of static data bit index
-	oCore -> static_data_bit_index_size = Swap32(oCore -> static_data_bit_index_size);
-
-	// Platform
-	oCore -> platform      = Swap64(oCore -> platform);
-
-	// Ugly-jolly hack!
-	// ... dereferencing type-punned pointer will break strict-aliasing rules ...
-	UINT_64 iTMP;
-	memcpy(&iTMP, &(oCore -> ieee754double), sizeof(UINT_64));
-	iTMP = Swap64(iTMP);
-	memcpy(&(oCore -> ieee754double), &iTMP, sizeof(UINT_64));
-
-	// Cyclic Redundancy Check
-	oCore -> crc = 0;
-
-	// Convert data structures
-
-	// Convert code segment
-	VMInstruction * pInstructions = const_cast<VMInstruction *>(VMExecutable::GetCodeSeg(oCore));
-	UINT_32 iI = 0;
-	UINT_32 iSteps = oCore -> code_size / sizeof(VMInstruction);
-	for(iI = 0; iI < iSteps; ++iI)
-	{
-		pInstructions -> instruction = Swap32(pInstructions -> instruction);
-		pInstructions -> argument    = Swap32(pInstructions -> argument);
-		pInstructions -> reserved    = Swap64(pInstructions -> reserved);
-		++pInstructions;
-	}
-
-	// Convert syscalls index
-	TextDataIndex * pTextIndex = const_cast<TextDataIndex *>(VMExecutable::GetSyscallsIndexSeg(oCore));
-	iSteps = oCore -> syscalls_index_size / sizeof(TextDataIndex);
-	for(iI = 0; iI < iSteps; ++iI)
-	{
-		pTextIndex -> offset = Swap32(pTextIndex -> offset);
-		pTextIndex -> length = Swap32(pTextIndex -> length);
-		++pTextIndex;
-	}
-
-	// Convert static text index
-	pTextIndex = const_cast<TextDataIndex *>(VMExecutable::GetStaticTextIndexSeg(oCore));
-	iSteps = oCore -> static_text_index_size / sizeof(TextDataIndex);
-	for(iI = 0; iI < iSteps; ++iI)
-	{
-		pTextIndex -> offset = Swap32(pTextIndex -> offset);
-		pTextIndex -> length = Swap32(pTextIndex -> length);
-		++pTextIndex;
-	}
-
-	// Convert static data
-	StaticDataVar * pStaticDataVar = const_cast<StaticDataVar *>(VMExecutable::GetStaticDataSeg(oCore));
-	iSteps = oCore -> static_data_data_size / sizeof(StaticDataVar);
-	for(iI = 0; iI < iSteps; ++iI)
-	{
-		(*pStaticDataVar).i_data = Swap64((*pStaticDataVar).i_data);
-		++pStaticDataVar;
-	}
-}
-
-//
-// Constructor
-//
-VMStringLoader::VMStringLoader(CCHAR_P rawContent, size_t rawContentSize)
-{
-    oCore = (VMExecutable *)malloc(rawContentSize + 1);
-    memcpy(oCore, rawContent, rawContentSize);
-
-	if (oCore -> magic[0] == 'C' &&
-	    oCore -> magic[1] == 'T' &&
-	    oCore -> magic[2] == 'P' &&
-	    oCore -> magic[3] == 'P')
-	{
-		// Check version
-		if (oCore -> version[0] >= 1)
-		{
-			// Platform-dependent data (byte order)
-			if (oCore -> platform == 0x4142434445464748ull)
-			{
-#ifdef _DEBUG
-				fprintf(stderr, "Big/Little Endian conversion: Nothing to do\n");
-#endif
-
-				// Nothing to do, only check crc
-				UINT_32 iCRC = oCore -> crc;
-				oCore -> crc = 0;
-
-				// Calculate CRC of file
-                // KELSON: next line used to refer to oStat.st_size
-                // changed it to rawContentSize
-				if (iCRC != crc32((UCCHAR_P)oCore, rawContentSize))
-				{
-					free(oCore);
-					throw CTPPLogicError("CRC checksum invalid");
-				}
-			}
-			// Platform-dependent data (byte order)
-			else if (oCore -> platform == 0x4847464544434241ull)
-			{
-				// Need to reconvert data
-#ifdef _DEBUG
-				fprintf(stderr, "Big/Little Endian conversion: Need to reconvert core\n");
-#endif
-				ConvertExecutable(oCore);
-			}
-			else
-			{
-				free(oCore);
-				throw CTPPLogicError("Conversion of middle-end architecture does not supported.");
-			}
-
-			// Check IEEE 754 format
-			if (oCore -> ieee754double != 15839800103804824402926068484019465486336.0)
-			{
-				free(oCore);
-				throw CTPPLogicError("IEEE 754 format is broken, cannot convert file");
-			}
-		}
-
-		pVMMemoryCore = new VMMemoryCore(oCore);
-	}
-	else
-	{
-		free(oCore);
-		throw CTPPLogicError("Not an CTPP bytecode file.");
-	}
-}
-
-//
-// Get ready-to-run program
-//
-const VMMemoryCore * VMStringLoader::GetCore() const { return pVMMemoryCore; }
-
-//
-// A destructor
-//
-VMStringLoader::~VMStringLoader() throw()
-{
-	delete pVMMemoryCore;
-	free(oCore);
-}
-
-} // namespace CTPP
-// End.

src/downloader.cpp

@@ -18,7 +18,8 @@
  */
 
 #include "downloader.h"
-#include "common/pathTools.h"
+#include "tools/pathTools.h"
+#include "tools/stringTools.h"
 
 #include <algorithm>
 #include <thread>
@@ -28,7 +29,7 @@
 
 #include "aria2.h"
 #include "xmlrpc.h"
-#include "common/otherTools.h"
+#include "tools/otherTools.h"
 #include <pugixml.hpp>
 
 namespace kiwix
@@ -36,6 +37,8 @@ namespace kiwix
 
 void Download::updateStatus(bool follow)
 {
+  if (m_status == Download::K_REMOVED)
+    return;
   static std::vector<std::string> statusKey = {"status", "files", "totalLength",
                                                "completedLength", "followedBy",
                                                "downloadSpeed", "verifiedLength"};
@@ -93,17 +96,55 @@ void Download::updateStatus(bool follow)
   }
 }
 
+void Download::resumeDownload()
+{
+    if (!m_followedBy.empty())
+        mp_aria->unpause(m_followedBy);
+    else
+        mp_aria->unpause(m_did);
+    updateStatus(true);
+}
+
+void Download::pauseDownload()
+{
+    if (!m_followedBy.empty())
+        mp_aria->pause(m_followedBy);
+    else
+        mp_aria->pause(m_did);
+    updateStatus(true);
+}
+
+void Download::cancelDownload()
+{
+    if (!m_followedBy.empty())
+        mp_aria->remove(m_followedBy);
+    else
+        mp_aria->remove(m_did);
+    m_status = Download::K_REMOVED;
+}
+
 /* Constructor */
 Downloader::Downloader() :
   mp_aria(new Aria2())
 {
-  for (auto gid : mp_aria->tellActive()) {
-    m_knownDownloads[gid] = std::unique_ptr<Download>(new Download(mp_aria, gid));
-    m_knownDownloads[gid]->updateStatus();
+  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();
+    }
+  } catch (std::exception& e) {
+    std::cerr << "aria2 tellWaiting failed : " << e.what() << std::endl;
   }
 }
 
-
 /* Destructor */
 Downloader::~Downloader()
 {
@@ -122,7 +163,7 @@ std::vector<std::string> Downloader::getDownloadIds() {
   return ret;
 }
 
-Download* Downloader::startDownload(const std::string& uri)
+Download* Downloader::startDownload(const std::string& uri, const std::vector<std::pair<std::string, std::string>>& options)
 {
   for (auto& p: m_knownDownloads) {
     auto& d = p.second;
@@ -131,7 +172,7 @@ Download* Downloader::startDownload(const std::string& uri)
       return d.get();
   }
   std::vector<std::string> uris = {uri};
-  auto gid = mp_aria->addUri(uris);
+  auto gid = mp_aria->addUri(uris, options);
   m_knownDownloads[gid] = std::unique_ptr<Download>(new Download(mp_aria, gid));
   return m_knownDownloads[gid].get();
 }
@@ -139,14 +180,23 @@ Download* Downloader::startDownload(const std::string& uri)
 Download* Downloader::getDownload(const std::string& did)
 {
   try {
+    m_knownDownloads.at(did).get()->updateStatus(true);
     return m_knownDownloads.at(did).get();
-  } catch(exception& e) {
+  } 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();
+      }
+  }
     throw e;
   }
 }

src/entry.cpp

@@ -1,5 +1,5 @@
 /*
- * Copyright 2011 Emmanuel Engelhart <kelson@kiwix.org>
+ * 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
@@ -20,119 +20,54 @@
 #include "reader.h"
 #include <time.h>
 
-#include <zim/search.h>
-
 namespace kiwix
 {
 
-Entry::Entry(zim::Article article)
-  : article(article)
+Entry::Entry(zim::Entry entry)
+  : entry(entry)
 {
 }
 
-#define RETURN_IF_INVALID(WHAT) if(!good()) { return (WHAT); }
-
-std::string Entry::getPath() const
-{
-  RETURN_IF_INVALID("");
-  return article.getLongUrl();
-}
-
-std::string Entry::getTitle() const
-{
-  RETURN_IF_INVALID("");
-  return article.getTitle();
-}
-
-std::string Entry::getContent() const
-{
-  RETURN_IF_INVALID("");
-  return article.getData();
-}
-
-zim::Blob Entry::getBlob(offset_type offset) const
-{
-  RETURN_IF_INVALID(zim::Blob());
-  return article.getData(offset);
-}
-
-zim::Blob Entry::getBlob(offset_type offset, size_type size) const
-{
-  RETURN_IF_INVALID(zim::Blob());
-  return article.getData(offset, size);
-}
-
-std::pair<std::string, offset_type> Entry::getDirectAccessInfo() const
-{
-  RETURN_IF_INVALID(std::make_pair("", 0));
-  return article.getDirectAccessInformation();
-}
-
 size_type Entry::getSize() const
 {
-  RETURN_IF_INVALID(0);
-  return article.getArticleSize();
+  if (entry.isRedirect()) {
+    return 0;
+  } else {
+    return entry.getItem().getSize();
+  }
 }
 
 std::string Entry::getMimetype() const
 {
-  RETURN_IF_INVALID("");
-  try {
-    return article.getMimeType();
-  } catch (exception& e) {
-    return "application/octet-stream";
-  }
+  return entry.getItem(true).getMimetype();
 }
 
 bool Entry::isRedirect() const
 {
-  RETURN_IF_INVALID(false);
-  return article.isRedirect();
-}
-
-bool Entry::isLinkTarget() const
-{
-  RETURN_IF_INVALID(false);
-  return article.isLinktarget();
-}
-
-bool Entry::isDeleted() const
-{
-  RETURN_IF_INVALID(false);
-  return article.isDeleted();
+  return entry.isRedirect();
 }
 
 Entry Entry::getRedirectEntry() const
 {
-  RETURN_IF_INVALID(Entry());
-  if ( !article.isRedirect() ) {
+  if ( !entry.isRedirect() ) {
     throw NoEntry();
   }
 
-  auto targeted_article = article.getRedirectArticle();
-  if ( !targeted_article.good()) {
-    throw NoEntry();
-  }
-  return targeted_article;
+  return entry.getRedirectEntry();
 }
 
 Entry Entry::getFinalEntry() const
 {
-  RETURN_IF_INVALID(Entry());
-  if (final_article.good()) {
-    return final_article;
-  }
-
   int loopCounter = 42;
-  final_article = article;
-  while (final_article.isRedirect() && loopCounter--) {
-    final_article = final_article.getRedirectArticle();
-    if ( !final_article.good()) {
-      throw NoEntry();
-    }
+  auto final_entry = entry;
+  while (final_entry.isRedirect() && loopCounter--) {
+    final_entry = final_entry.getRedirectEntry();
   }
-
-  return final_article;
+  // Prevent infinite loops.
+  if (final_entry.isRedirect()) {
+    throw NoEntry();
+  }
+  return final_entry;
 }
 
 }

src/kiwixserve.cpp

@@ -0,0 +1,81 @@
+#include "kiwixserve.h"
+#include "subprocess.h"
+
+#ifdef _WIN32
+    # define KIWIXSERVE_CMD "kiwix-serve.exe"
+    # include <windows.h>
+#else
+    # define KIWIXSERVE_CMD "kiwix-serve"
+    # include <unistd.h>
+#endif
+
+#include "tools/pathTools.h"
+#include "tools/stringTools.h"
+
+namespace kiwix {
+
+KiwixServe::KiwixServe(const std::string& libraryPath, int port)
+  : m_port(port),
+    m_libraryPath(libraryPath)
+{
+}
+
+KiwixServe::~KiwixServe()
+{
+    shutDown();
+}
+
+void KiwixServe::run()
+{
+    #ifdef _WIN32
+        int pid = GetCurrentProcessId();
+    #else
+        pid_t pid = getpid();
+
+    #endif
+
+    std::vector<const char*> callCmd;
+    std::string kiwixServeCmd = appendToDirectory(
+        removeLastPathElement(getExecutablePath(true)),
+        KIWIXSERVE_CMD);
+    if (fileExists(kiwixServeCmd)) {
+        // A local kiwix-serve exe exists (packaged with kiwix-desktop), use it.
+        callCmd.push_back(kiwixServeCmd.c_str());
+    } else {
+        // Try to use a potential installed kiwix-serve.
+        callCmd.push_back(KIWIXSERVE_CMD);
+    }
+    std::string attachProcessOpt = "-a" + to_string(pid);
+    std::string portOpt = "-p" + to_string(m_port);
+    callCmd.push_back(attachProcessOpt.c_str());
+    callCmd.push_back(portOpt.c_str());
+    callCmd.push_back("-l");
+    callCmd.push_back(m_libraryPath.c_str());
+    mp_kiwixServe = Subprocess::run(callCmd);
+}
+
+void KiwixServe::shutDown()
+{
+    if (mp_kiwixServe)
+        mp_kiwixServe->kill();
+}
+
+bool KiwixServe::isRunning()
+{
+    if (mp_kiwixServe) {
+        return (mp_kiwixServe->isRunning());
+    }
+    return false;
+}
+
+int KiwixServe::setPort(int port)
+{
+    if (port >= 1 && port <= 65535) {
+        m_port = port;
+    } else {
+        return -1;
+    }
+    return m_port;
+}
+
+}

src/library.cpp

@@ -19,20 +19,53 @@
 
 #include "library.h"
 #include "book.h"
+#include "reader.h"
+#include "libxml_dumper.h"
 
-#include "common/base64.h"
-#include "common/regexTools.h"
-#include "common/pathTools.h"
+#include "tools/base64.h"
+#include "tools/regexTools.h"
+#include "tools/pathTools.h"
+#include "tools/stringTools.h"
 
 #include <pugixml.hpp>
 #include <algorithm>
+#include <set>
+#include <unicode/locid.h>
+#include <xapian.h>
 
 namespace kiwix
 {
+
+namespace
+{
+
+std::string iso639_3ToXapian(const std::string& lang) {
+  return icu::Locale(lang.c_str()).getLanguage();
+};
+
+std::string normalizeText(const std::string& text)
+{
+  return removeAccents(text);
+}
+
+} // unnamed namespace
+
+class Library::BookDB : public Xapian::WritableDatabase
+{
+public:
+  BookDB() : Xapian::WritableDatabase("", Xapian::DB_BACKEND_INMEMORY) {}
+};
+
 /* Constructor */
-Library::Library() : version(KIWIX_LIBRARY_VERSION)
+Library::Library()
+  : m_bookDB(new BookDB)
 {
 }
+
+Library::Library(Library&& ) = default;
+
+Library& Library::operator=(Library&& ) = default;
+
 /* Destructor */
 Library::~Library()
 {
@@ -42,32 +75,90 @@ Library::~Library()
 bool Library::addBook(const Book& book)
 {
   /* Try to find it */
+  updateBookDB(book);
   try {
-    auto& oldbook = books.at(book.getId());
+    auto& oldbook = m_books.at(book.getId());
     oldbook.update(book);
     return false;
   } catch (std::out_of_range&) {
-    books[book.getId()] = book;
+    m_books[book.getId()] = book;
     return true;
   }
 }
 
+void Library::addBookmark(const Bookmark& bookmark)
+{
+  m_bookmarks.push_back(bookmark);
+}
+
+bool Library::removeBookmark(const std::string& zimId, const std::string& url)
+{
+  for(auto it=m_bookmarks.begin(); it!=m_bookmarks.end(); it++) {
+    if (it->getBookId() == zimId && it->getUrl() == url) {
+      m_bookmarks.erase(it);
+      return true;
+    }
+  }
+  return false;
+}
+
+
 
 bool Library::removeBookById(const std::string& id)
 {
-  return books.erase(id) == 1;
+  m_bookDB->delete_document("Q" + id);
+  m_readers.erase(id);
+  return m_books.erase(id) == 1;
+}
+
+const Book& Library::getBookById(const std::string& id) const
+{
+  return m_books.at(id);
 }
 
 Book& Library::getBookById(const std::string& id)
 {
-  return books.at(id);
+  const Library& const_self = *this;
+  return const_cast<Book&>(const_self.getBookById(id));
+}
+
+const Book& Library::getBookByPath(const std::string& path) const
+{
+  for(auto& it: m_books) {
+    auto& book = it.second;
+    if (book.getPath() == path)
+      return book;
+  }
+  std::ostringstream ss;
+  ss << "No book with path " << path << " in the library." << std::endl;
+  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)
+{
+  try {
+    return m_readers.at(id);
+  } catch (std::out_of_range& e) {}
+
+  auto book = getBookById(id);
+  if (!book.isPathValid())
+    return nullptr;
+  auto sptr = make_shared<Reader>(book.getPath());
+  m_readers[id] = sptr;
+  return sptr;
 }
 
 unsigned int Library::getBookCount(const bool localBooks,
-                                   const bool remoteBooks)
+                                   const bool remoteBooks) const
 {
   unsigned int result = 0;
-  for (auto& pair: books) {
+  for (auto& pair: m_books) {
     auto& book = pair.second;
     if ((!book.getPath().empty() && localBooks)
         || (book.getPath().empty() && remoteBooks)) {
@@ -77,100 +168,26 @@ unsigned int Library::getBookCount(const bool localBooks,
   return result;
 }
 
-bool Library::writeToFile(const std::string& path) {
-  pugi::xml_document doc;
-  auto baseDir = removeLastPathElement(path, true, false);
-
-  /* Add the library node */
-  pugi::xml_node libraryNode = doc.append_child("library");
-
-  if (!version.empty())
-    libraryNode.append_attribute("version") = version.c_str();
-
-  /* Add each book */
-  for (auto& pair: books) {
-    auto& book = pair.second;
-    if (!book.readOnly()) {
-      pugi::xml_node bookNode = libraryNode.append_child("book");
-      bookNode.append_attribute("id") = book.getId().c_str();
-
-      if (!book.getPath().empty()) {
-        bookNode.append_attribute("path") = computeRelativePath(
-            baseDir, book.getPath()).c_str();
-      }
-
-      if (!book.getIndexPath().empty()) {
-        bookNode.append_attribute("indexPath") = computeRelativePath(
-            baseDir, book.getIndexPath()).c_str();
-        bookNode.append_attribute("indexType") = "xapian";
-      }
-
-      if (book.getOrigId().empty()) {
-        if (!book.getTitle().empty())
-          bookNode.append_attribute("title") = book.getTitle().c_str();
-
-        if (!book.getName().empty())
-          bookNode.append_attribute("name") = book.getName().c_str();
-
-        if (!book.getTags().empty())
-          bookNode.append_attribute("tags") = book.getTags().c_str();
-
-        if (!book.getDescription().empty())
-          bookNode.append_attribute("description") = book.getDescription().c_str();
-
-        if (!book.getLanguage().empty())
-          bookNode.append_attribute("language") = book.getLanguage().c_str();
-
-        if (!book.getCreator().empty())
-          bookNode.append_attribute("creator") = book.getCreator().c_str();
-
-        if (!book.getPublisher().empty())
-          bookNode.append_attribute("publisher") = book.getPublisher().c_str();
-
-        if (!book.getFavicon().empty())
-          bookNode.append_attribute("favicon") = base64_encode(book.getFavicon()).c_str();
-
-        if (!book.getFaviconMimeType().empty())
-          bookNode.append_attribute("faviconMimeType")
-              = book.getFaviconMimeType().c_str();
-      } else {
-        bookNode.append_attribute("origId") = book.getOrigId().c_str();
-      }
-
-      if (!book.getDate().empty()) {
-        bookNode.append_attribute("date") = book.getDate().c_str();
-      }
-
-      if (!book.getUrl().empty()) {
-        bookNode.append_attribute("url") = book.getUrl().c_str();
-      }
-
-      if (!book.getArticleCount())
-        bookNode.append_attribute("articleCount") = to_string(book.getArticleCount()).c_str();
-
-      if (!book.getMediaCount())
-        bookNode.append_attribute("mediaCount") = to_string(book.getMediaCount()).c_str();
-
-      if (book.getSize()) {
-        bookNode.append_attribute("size") = to_string(book.getSize()>>10).c_str();
-      }
-
-      if (!book.getDownloadId().empty()) {
-        bookNode.append_attribute("downloadId") = book.getDownloadId().c_str();
-      }
-    }
-  }
-
-  /* saving file */
-  return doc.save_file(path.c_str());
+bool Library::writeToFile(const std::string& path) const
+{
+  auto baseDir = removeLastPathElement(path);
+  LibXMLDumper dumper(this);
+  dumper.setBaseDir(baseDir);
+  return writeTextFile(path, dumper.dumpLibXMLContent(getBooksIds()));
 }
 
-std::vector<std::string> Library::getBooksLanguages()
+bool Library::writeBookmarksToFile(const std::string& path) const
+{
+  LibXMLDumper dumper(this);
+  return writeTextFile(path, dumper.dumpLibXMLBookmark());
+}
+
+std::vector<std::string> Library::getBooksLanguages() const
 {
   std::vector<std::string> booksLanguages;
   std::map<std::string, bool> booksLanguagesMap;
 
-  for (auto& pair: books) {
+  for (auto& pair: m_books) {
     auto& book = pair.second;
     auto& language = book.getLanguage();
     if (booksLanguagesMap.find(language) == booksLanguagesMap.end()) {
@@ -184,12 +201,27 @@ std::vector<std::string> Library::getBooksLanguages()
   return booksLanguages;
 }
 
-std::vector<std::string> Library::getBooksCreators()
+std::vector<std::string> Library::getBooksCategories() const
+{
+  std::set<std::string> categories;
+
+  for (const auto& pair: m_books) {
+    const auto& book = pair.second;
+    const auto& c = book.getCategory();
+    if ( !c.empty() ) {
+      categories.insert(c);
+    }
+  }
+
+  return std::vector<std::string>(categories.begin(), categories.end());
+}
+
+std::vector<std::string> Library::getBooksCreators() const
 {
   std::vector<std::string> booksCreators;
   std::map<std::string, bool> booksCreatorsMap;
 
-  for (auto& pair: books) {
+  for (auto& pair: m_books) {
     auto& book = pair.second;
     auto& creator = book.getCreator();
     if (booksCreatorsMap.find(creator) == booksCreatorsMap.end()) {
@@ -203,12 +235,12 @@ std::vector<std::string> Library::getBooksCreators()
   return booksCreators;
 }
 
-std::vector<std::string> Library::getBooksPublishers()
+std::vector<std::string> Library::getBooksPublishers() const
 {
   std::vector<std::string> booksPublishers;
   std::map<std::string, bool> booksPublishersMap;
 
-  for (auto& pair:books) {
+  for (auto& pair:m_books) {
     auto& book = pair.second;
     auto& publisher = book.getPublisher();
     if (booksPublishersMap.find(publisher) == booksPublishersMap.end()) {
@@ -222,145 +254,532 @@ std::vector<std::string> Library::getBooksPublishers()
   return booksPublishers;
 }
 
-std::vector<std::string> Library::getBooksIds()
+const std::vector<kiwix::Bookmark> Library::getBookmarks(bool onlyValidBookmarks) const
 {
-  std::vector<std::string> bookIds;
+  if (!onlyValidBookmarks) {
+    return m_bookmarks;
+  }
+  std::vector<kiwix::Bookmark> validBookmarks;
+  auto booksId = getBooksIds();
+  for(auto& bookmark:m_bookmarks) {
+    if (std::find(booksId.begin(), booksId.end(), bookmark.getBookId()) != booksId.end()) {
+      validBookmarks.push_back(bookmark);
+    }
+  }
+  return validBookmarks;
+}
 
-  for (auto& pair: books) {
+Library::BookIdCollection Library::getBooksIds() const
+{
+  BookIdCollection bookIds;
+
+  for (auto& pair: m_books) {
     bookIds.push_back(pair.first);
   }
 
   return bookIds;
 }
 
-std::vector<std::string> Library::filter(const std::string& search)
+Library::BookIdCollection Library::filter(const std::string& search) const
 {
   if (search.empty()) {
     return getBooksIds();
   }
 
-  std::vector<std::string> bookIds;
-  for(auto& pair:books) {
-     auto& book = pair.second;
-     if (matchRegex(book.getTitle(), "\\Q" + search + "\\E")
-         || matchRegex(book.getDescription(), "\\Q" + search + "\\E")) {
-       bookIds.push_back(pair.first);
-     }
+  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 (...) {}
+  Xapian::Document doc;
+  indexer.set_document(doc);
+
+  const std::string title = normalizeText(book.getTitle());
+  const std::string desc = normalizeText(book.getDescription());
+
+  // Index title and description without prefixes for general search
+  indexer.index_text(title);
+  indexer.increase_termpos();
+  indexer.index_text(desc);
+
+  // 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");
+  indexer.index_text(normalizeText(book.getCreator()),   1, "A");
+  indexer.index_text(normalizeText(book.getPublisher()), 1, "XP");
+  indexer.index_text(normalizeText(book.getName()),      1, "XN");
+  indexer.index_text(normalizeText(book.getCategory()),  1, "XC");
+
+  for ( const auto& tag : split(normalizeText(book.getTags()), ";") )
+    doc.add_boolean_term("XT" + tag);
+
+  const std::string idterm = "Q" + book.getId();
+  doc.add_boolean_term(idterm);
+
+  doc.set_data(book.getId());
+
+  m_bookDB->replace_document(idterm, doc);
+}
+
+namespace
+{
+
+bool willSelectEverything(const Xapian::Query& query)
+{
+  return query.get_type() == Xapian::Query::LEAF_MATCH_ALL;
+}
+
+
+Xapian::Query buildXapianQueryFromFilterQuery(const Filter& filter)
+{
+  if ( !filter.hasQuery() || filter.getQuery().empty() ) {
+    // This is a thread-safe way to construct an equivalent of
+    // a Xapian::Query::MatchAll query
+    return Xapian::Query(std::string());
+  }
+
+  Xapian::QueryParser queryParser;
+  queryParser.set_default_op(Xapian::Query::OP_AND);
+  queryParser.add_prefix("title", "S");
+  queryParser.add_prefix("description", "XD");
+  queryParser.add_prefix("name", "XN");
+  queryParser.add_prefix("category", "XC");
+  queryParser.add_prefix("lang", "L");
+  queryParser.add_prefix("publisher", "XP");
+  queryParser.add_prefix("creator", "A");
+  queryParser.add_prefix("tag", "XT");
+  const auto partialQueryFlag = filter.queryIsPartial()
+                              ? Xapian::QueryParser::FLAG_PARTIAL
+                              : 0;
+  // Language assumed for the query is not known for sure so stemming
+  // is not applied
+  //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;
+  return queryParser.parse_query(normalizeText(filter.getQuery()), flags);
+}
+
+Xapian::Query nameQuery(const std::string& name)
+{
+  return Xapian::Query("XN" + normalizeText(name));
+}
+
+Xapian::Query categoryQuery(const std::string& category)
+{
+  return Xapian::Query("XC" + normalizeText(category));
+}
+
+Xapian::Query langQuery(const std::string& lang)
+{
+  return Xapian::Query("L" + normalizeText(lang));
+}
+
+Xapian::Query publisherQuery(const std::string& publisher)
+{
+  Xapian::QueryParser queryParser;
+  queryParser.set_default_op(Xapian::Query::OP_OR);
+  queryParser.set_stemming_strategy(Xapian::QueryParser::STEM_NONE);
+  const auto flags = 0;
+  const auto q = queryParser.parse_query(normalizeText(publisher), flags, "XP");
+  return Xapian::Query(Xapian::Query::OP_PHRASE, q.get_terms_begin(), q.get_terms_end(), q.get_length());
+}
+
+Xapian::Query creatorQuery(const std::string& creator)
+{
+  Xapian::QueryParser queryParser;
+  queryParser.set_default_op(Xapian::Query::OP_OR);
+  queryParser.set_stemming_strategy(Xapian::QueryParser::STEM_NONE);
+  const auto flags = 0;
+  const auto q = queryParser.parse_query(normalizeText(creator), flags, "A");
+  return Xapian::Query(Xapian::Query::OP_PHRASE, q.get_terms_begin(), q.get_terms_end(), q.get_length());
+}
+
+Xapian::Query tagsQuery(const Filter::Tags& acceptTags, const Filter::Tags& rejectTags)
+{
+  Xapian::Query q = Xapian::Query(std::string());
+  if (!acceptTags.empty()) {
+    for ( const auto& tag : acceptTags )
+      q &= Xapian::Query("XT" + normalizeText(tag));
+  }
+
+  if (!rejectTags.empty()) {
+    for ( const auto& tag : rejectTags )
+      q = Xapian::Query(Xapian::Query::OP_AND_NOT, q, "XT" + normalizeText(tag));
+  }
+  return q;
+}
+
+Xapian::Query buildXapianQuery(const Filter& filter)
+{
+  auto q = buildXapianQueryFromFilterQuery(filter);
+  if ( filter.hasName() ) {
+    q = Xapian::Query(Xapian::Query::OP_AND, q, nameQuery(filter.getName()));
+  }
+  if ( filter.hasCategory() ) {
+    q = Xapian::Query(Xapian::Query::OP_AND, q, categoryQuery(filter.getCategory()));
+  }
+  if ( filter.hasLang() ) {
+    q = Xapian::Query(Xapian::Query::OP_AND, q, langQuery(filter.getLang()));
+  }
+  if ( filter.hasPublisher() ) {
+    q = Xapian::Query(Xapian::Query::OP_AND, q, publisherQuery(filter.getPublisher()));
+  }
+  if ( filter.hasCreator() ) {
+    q = Xapian::Query(Xapian::Query::OP_AND, q, creatorQuery(filter.getCreator()));
+  }
+  if ( !filter.getAcceptTags().empty() || !filter.getRejectTags().empty() ) {
+    const auto tq = tagsQuery(filter.getAcceptTags(), filter.getRejectTags());
+    q = Xapian::Query(Xapian::Query::OP_AND, q, tq);;
+  }
+  return q;
+}
+
+} // unnamed namespace
+
+Library::BookIdCollection Library::filterViaBookDB(const Filter& filter) const
+{
+  const auto query = buildXapianQuery(filter);
+
+  if ( willSelectEverything(query) )
+    return getBooksIds();
+
+  BookIdCollection bookIds;
+
+  Xapian::Enquire enquire(*m_bookDB);
+  enquire.set_query(query);
+  const auto results = enquire.get_mset(0, m_books.size());
+  for ( auto it = results.begin(); it != results.end(); ++it  ) {
+    bookIds.push_back(it.get_document().get_data());
   }
 
   return bookIds;
 }
 
-template<supportedListSortBy sort>
-struct Comparator {
-  Library* lib;
-  Comparator(Library* lib) : lib(lib) {}
-
-  bool operator() (const std::string& id1, const std::string& id2) {
-    return get_keys(id1) < get_keys(id2);
+Library::BookIdCollection Library::filter(const Filter& filter) const
+{
+  BookIdCollection result;
+  for(auto id : filterViaBookDB(filter)) {
+    if(filter.accept(m_books.at(id))) {
+      result.push_back(id);
+    }
   }
+  return result;
+}
 
-  std::string get_keys(const std::string& id);
-  unsigned int get_keyi(const std::string& id);
+template<supportedListSortBy SORT>
+struct KEY_TYPE {
+  typedef std::string TYPE;
 };
 
 template<>
-std::string Comparator<TITLE>::get_keys(const std::string& id)
+struct KEY_TYPE<SIZE> {
+  typedef size_t TYPE;
+};
+
+template<supportedListSortBy sort>
+class Comparator {
+  private:
+    const Library* const lib;
+    const bool     ascending;
+
+    inline typename KEY_TYPE<sort>::TYPE get_key(const std::string& id);
+
+  public:
+    Comparator(const Library* lib, bool ascending) : lib(lib), ascending(ascending) {}
+    inline bool operator() (const std::string& id1, const std::string& id2) {
+      if (ascending) {
+        return get_key(id1) < get_key(id2);
+      } else {
+        return get_key(id2) < get_key(id1);
+      }
+    }
+};
+
+template<>
+std::string Comparator<TITLE>::get_key(const std::string& id)
 {
   return lib->getBookById(id).getTitle();
 }
 
 template<>
-unsigned int Comparator<SIZE>::get_keyi(const std::string& id)
+size_t Comparator<SIZE>::get_key(const std::string& id)
 {
   return lib->getBookById(id).getSize();
 }
 
 template<>
-bool Comparator<SIZE>::operator() (const std::string& id1, const std::string& id2)
-{
-  return get_keyi(id1) < get_keyi(id2);
-}
-
-template<>
-std::string Comparator<DATE>::get_keys(const std::string& id)
+std::string Comparator<DATE>::get_key(const std::string& id)
 {
   return lib->getBookById(id).getDate();
 }
 
 template<>
-std::string Comparator<CREATOR>::get_keys(const std::string& id)
+std::string Comparator<CREATOR>::get_key(const std::string& id)
 {
   return lib->getBookById(id).getCreator();
 }
 
 template<>
-std::string Comparator<PUBLISHER>::get_keys(const std::string& id)
+std::string Comparator<PUBLISHER>::get_key(const std::string& id)
 {
   return lib->getBookById(id).getPublisher();
 }
 
+void Library::sort(BookIdCollection& bookIds, supportedListSortBy sort, bool ascending) const
+{
+  switch(sort) {
+    case TITLE:
+      std::sort(bookIds.begin(), bookIds.end(), Comparator<TITLE>(this, ascending));
+      break;
+    case SIZE:
+      std::sort(bookIds.begin(), bookIds.end(), Comparator<SIZE>(this, ascending));
+      break;
+    case DATE:
+      std::sort(bookIds.begin(), bookIds.end(), Comparator<DATE>(this, ascending));
+      break;
+    case CREATOR:
+      std::sort(bookIds.begin(), bookIds.end(), Comparator<CREATOR>(this, ascending));
+      break;
+    case PUBLISHER:
+      std::sort(bookIds.begin(), bookIds.end(), Comparator<PUBLISHER>(this, ascending));
+      break;
+    default:
+      break;
+  }
+}
 
-std::vector<std::string> Library::listBooksIds(
+
+Library::BookIdCollection Library::listBooksIds(
     int mode,
     supportedListSortBy sortBy,
     const std::string& search,
     const std::string& language,
     const std::string& creator,
     const std::string& publisher,
-    size_t maxSize) {
+    const std::vector<std::string>& tags,
+    size_t maxSize) const {
 
-  std::vector<std::string> bookIds;
-  for(auto& pair:books) {
-    auto& book = pair.second;
-    auto local = !book.getPath().empty();
-    if (mode & LOCAL && !local)
-      continue;
-    if (mode & NOLOCAL && local)
-      continue;
-    auto valid = book.isPathValid();
-    if (mode & VALID && !valid)
-      continue;
-    if (mode & NOVALID && valid)
-      continue;
-    auto remote = !book.getUrl().empty();
-    if (mode & REMOTE && !remote)
-      continue;
-    if (mode & NOREMOTE && remote)
-      continue;
-    if (maxSize != 0 && book.getSize() > maxSize)
-      continue;
-    if (!language.empty() && book.getLanguage() != language)
-      continue;
-    if (!publisher.empty() && book.getPublisher() != publisher)
-      continue;
-    if (!creator.empty() && book.getCreator() != creator)
-      continue;
-    if (!search.empty() && !(matchRegex(book.getTitle(), "\\Q" + search + "\\E")
-                          || matchRegex(book.getDescription(), "\\Q" + search + "\\E")))
-      continue;
-    bookIds.push_back(pair.first);
-  }
+  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);
 
-  switch(sortBy) {
-    case TITLE:
-      std::sort(bookIds.begin(), bookIds.end(), Comparator<TITLE>(this));
-      break;
-    case SIZE:
-      std::sort(bookIds.begin(), bookIds.end(), Comparator<SIZE>(this));
-      break;
-    case DATE:
-      std::sort(bookIds.begin(), bookIds.end(), Comparator<DATE>(this));
-      break;
-    case CREATOR:
-      std::sort(bookIds.begin(), bookIds.end(), Comparator<CREATOR>(this));
-      break;
-    case PUBLISHER:
-      std::sort(bookIds.begin(), bookIds.end(), Comparator<PUBLISHER>(this));
-      break;
-    default:
-      break;
-  }
+  auto bookIds = filter(_filter);
+
+  sort(bookIds, sortBy, true);
   return bookIds;
 }
+
+Filter::Filter()
+  : activeFilters(0),
+    _maxSize(0)
+{};
+
+#define FLAG(x) (1 << x)
+enum filterTypes {
+  NONE = 0,
+  _LOCAL = FLAG(0),
+  _REMOTE = FLAG(1),
+  _NOLOCAL = FLAG(2),
+  _NOREMOTE = FLAG(3),
+  _VALID = FLAG(4),
+  _NOVALID = FLAG(5),
+  ACCEPTTAGS = FLAG(6),
+  REJECTTAGS = FLAG(7),
+  LANG = FLAG(8),
+  _PUBLISHER = FLAG(9),
+  _CREATOR = FLAG(10),
+  MAXSIZE = FLAG(11),
+  QUERY = FLAG(12),
+  NAME = FLAG(13),
+  CATEGORY = FLAG(14),
+};
+
+Filter& Filter::local(bool accept)
+{
+  if (accept) {
+    activeFilters |= _LOCAL;
+    activeFilters &= ~_NOLOCAL;
+  } else {
+    activeFilters |= _NOLOCAL;
+    activeFilters &= ~_LOCAL;
+  }
+  return *this;
+}
+
+Filter& Filter::remote(bool accept)
+{
+  if (accept) {
+    activeFilters |= _REMOTE;
+    activeFilters &= ~_NOREMOTE;
+  } else {
+    activeFilters |= _NOREMOTE;
+    activeFilters &= ~_REMOTE;
+  }
+  return *this;
+}
+
+Filter& Filter::valid(bool accept)
+{
+  if (accept) {
+    activeFilters |= _VALID;
+    activeFilters &= ~_NOVALID;
+  } else {
+    activeFilters |= _NOVALID;
+    activeFilters &= ~_VALID;
+  }
+  return *this;
+}
+
+Filter& Filter::acceptTags(const Tags& tags)
+{
+  _acceptTags = tags;
+  activeFilters |= ACCEPTTAGS;
+  return *this;
+}
+
+Filter& Filter::rejectTags(const Tags& tags)
+{
+  _rejectTags = tags;
+  activeFilters |= REJECTTAGS;
+  return *this;
+}
+
+Filter& Filter::category(std::string category)
+{
+  _category = category;
+  activeFilters |= CATEGORY;
+  return *this;
+}
+
+Filter& Filter::lang(std::string lang)
+{
+  _lang = lang;
+  activeFilters |= LANG;
+  return *this;
+}
+
+Filter& Filter::publisher(std::string publisher)
+{
+  _publisher = publisher;
+  activeFilters |= _PUBLISHER;
+  return *this;
+}
+
+Filter& Filter::creator(std::string creator)
+{
+  _creator = creator;
+  activeFilters |= _CREATOR;
+  return *this;
+}
+
+Filter& Filter::maxSize(size_t maxSize)
+{
+  _maxSize = maxSize;
+  activeFilters |= MAXSIZE;
+  return *this;
+}
+
+Filter& Filter::query(std::string query, bool partial)
+{
+  _query = query;
+  _queryIsPartial = partial;
+  activeFilters |= QUERY;
+  return *this;
+}
+
+Filter& Filter::name(std::string name)
+{
+  _name = name;
+  activeFilters |= NAME;
+  return *this;
+}
+
+#define ACTIVE(X) (activeFilters & (X))
+#define FILTER(TAG, TEST) if (ACTIVE(TAG) && !(TEST)) { return false; }
+bool Filter::hasQuery() const
+{
+  return ACTIVE(QUERY);
+}
+
+bool Filter::hasName() const
+{
+  return ACTIVE(NAME);
+}
+
+bool Filter::hasCategory() const
+{
+  return ACTIVE(CATEGORY);
+}
+
+bool Filter::hasLang() const
+{
+  return ACTIVE(LANG);
+}
+
+bool Filter::hasPublisher() const
+{
+  return ACTIVE(_PUBLISHER);
+}
+
+bool Filter::hasCreator() const
+{
+  return ACTIVE(_CREATOR);
+}
+
+bool Filter::accept(const Book& book) const
+{
+  auto local = !book.getPath().empty();
+  FILTER(_LOCAL, local)
+  FILTER(_NOLOCAL, !local)
+
+  auto valid = book.isPathValid();
+  FILTER(_VALID, valid)
+  FILTER(_NOVALID, !valid)
+
+  auto remote = !book.getUrl().empty();
+  FILTER(_REMOTE, remote)
+  FILTER(_NOREMOTE, !remote)
+
+  FILTER(MAXSIZE, book.getSize() <= _maxSize)
+
+  return true;
+}
+
 }

src/libxml_dumper.cpp

@@ -0,0 +1,142 @@
+/*
+ * 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.
+ */
+
+#include "libxml_dumper.h"
+#include "book.h"
+
+#include "tools/base64.h"
+#include "tools/stringTools.h"
+#include "tools/otherTools.h"
+#include "tools/pathTools.h"
+
+namespace kiwix
+{
+/* Constructor */
+LibXMLDumper::LibXMLDumper(const Library* library)
+  : library(library)
+{
+}
+/* Destructor */
+LibXMLDumper::~LibXMLDumper()
+{
+}
+
+#define ADD_ATTRIBUTE(node, name, value) { (node).append_attribute((name)) = (value).c_str(); }
+#define ADD_ATTR_NOT_EMPTY(node, name, value) { if (!(value).empty()) ADD_ATTRIBUTE(node, name, value); }
+
+void LibXMLDumper::handleBook(Book book, pugi::xml_node root_node) {
+  if (book.readOnly())
+    return;
+
+  auto entry_node = root_node.append_child("book");
+  ADD_ATTRIBUTE(entry_node, "id", book.getId());
+
+  if (!book.getPath().empty()) {
+    ADD_ATTRIBUTE(entry_node, "path", computeRelativePath(baseDir, book.getPath()));
+  }
+
+  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, "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()));
+  } else {
+    ADD_ATTRIBUTE(entry_node, "origId", book.getOrigId());
+  }
+
+  ADD_ATTR_NOT_EMPTY(entry_node, "date", book.getDate());
+  ADD_ATTR_NOT_EMPTY(entry_node, "url", book.getUrl());
+
+  if (book.getArticleCount())
+    ADD_ATTRIBUTE(entry_node, "articleCount", to_string(book.getArticleCount()));
+
+  if (book.getMediaCount())
+    ADD_ATTRIBUTE(entry_node, "mediaCount", to_string(book.getMediaCount()));
+
+  if (book.getSize())
+    ADD_ATTRIBUTE(entry_node, "size", to_string(book.getSize()>>10));
+
+  ADD_ATTR_NOT_EMPTY(entry_node, "downloadId", book.getDownloadId());
+}
+
+#define ADD_TEXT_ENTRY(node, child, value) (node).append_child((child)).append_child(pugi::node_pcdata).set_value((value).c_str())
+
+void LibXMLDumper::handleBookmark(Bookmark bookmark, pugi::xml_node root_node) {
+
+  auto entry_node = root_node.append_child("bookmark");
+  auto book_node = entry_node.append_child("book");
+
+  try {
+    auto book = library->getBookById(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, "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, "language", bookmark.getLanguage());
+    ADD_TEXT_ENTRY(book_node, "date", bookmark.getDate());
+  }
+  ADD_TEXT_ENTRY(entry_node, "title", bookmark.getTitle());
+  ADD_TEXT_ENTRY(entry_node, "url", bookmark.getUrl());
+}
+
+
+std::string LibXMLDumper::dumpLibXMLContent(const std::vector<std::string>& bookIds)
+{
+  pugi::xml_document doc;
+
+  /* Add the library node */
+  pugi::xml_node libraryNode = doc.append_child("library");
+
+  libraryNode.append_attribute("version") = KIWIX_LIBRARY_VERSION;
+
+  if (library) {
+    for (auto& bookId: bookIds) {
+      handleBook(library->getBookById(bookId), libraryNode);
+    }
+  }
+  return nodeToString(libraryNode);
+}
+
+std::string LibXMLDumper::dumpLibXMLBookmark()
+{
+  pugi::xml_document doc;
+
+  /* Add the library node */
+  pugi::xml_node bookmarksNode = doc.append_child("bookmarks");
+
+  if (library) {
+    for (auto& bookmark: library->getBookmarks()) {
+      handleBookmark(bookmark, bookmarksNode);
+    }
+  }
+  return nodeToString(bookmarksNode);
+}
+
+}

src/manager.cpp

@@ -19,6 +19,8 @@
 
 #include "manager.h"
 
+#include "tools/pathTools.h"
+
 #include <pugixml.hpp>
 
 namespace kiwix
@@ -46,8 +48,9 @@ Manager::~Manager()
   }
 }
 bool Manager::parseXmlDom(const pugi::xml_document& doc,
-                          const bool readOnly,
-                          const std::string& libraryPath)
+                          bool readOnly,
+                          const std::string& libraryPath,
+                          bool trustLibrary)
 {
   pugi::xml_node libraryNode = doc.child("library");
 
@@ -59,14 +62,10 @@ bool Manager::parseXmlDom(const pugi::xml_document& doc,
 
     book.setReadOnly(readOnly);
     book.updateFromXml(bookNode,
-                       removeLastPathElement(libraryPath, true, false));
+                       removeLastPathElement(libraryPath));
 
-    /* Update the book properties with the new importer */
-    if (libraryVersion.empty()
-        || atoi(libraryVersion.c_str()) <= atoi(KIWIX_LIBRARY_VERSION)) {
-      if (!book.getPath().empty()) {
-        this->readBookFromPath(book.getPath(), &book);
-      }
+    if (!trustLibrary && !book.getPath().empty()) {
+      this->readBookFromPath(book.getPath(), &book);
     }
     manipulator->addBookToLibrary(book);
   }
@@ -75,15 +74,16 @@ bool Manager::parseXmlDom(const pugi::xml_document& doc,
 }
 
 bool Manager::readXml(const std::string& xml,
-                      const bool readOnly,
-                      const std::string& libraryPath)
+                      bool readOnly,
+                      const std::string& libraryPath,
+                      bool trustLibrary)
 {
   pugi::xml_document doc;
   pugi::xml_parse_result result
-      = doc.load_buffer_inplace((void*)xml.data(), xml.size());
+      = doc.load_buffer((void*)xml.data(), xml.size());
 
   if (result) {
-    this->parseXmlDom(doc, readOnly, libraryPath);
+    this->parseXmlDom(doc, readOnly, libraryPath, trustLibrary);
   }
 
   return true;
@@ -124,7 +124,7 @@ bool Manager::readOpds(const std::string& content, const std::string& urlHost)
 {
   pugi::xml_document doc;
   pugi::xml_parse_result result
-      = doc.load_buffer_inplace((void*)content.data(), content.size());
+      = doc.load_buffer((void*)content.data(), content.size());
 
   if (result) {
     this->parseOpdsDom(doc, urlHost);
@@ -134,21 +134,22 @@ bool Manager::readOpds(const std::string& content, const std::string& urlHost)
   return false;
 }
 
-bool Manager::readFile(const std::string& path, const bool readOnly)
-{
-  return this->readFile(path, path, readOnly);
-}
-
-bool Manager::readFile(const std::string& nativePath,
-                       const std::string& UTF8Path,
-                       const bool readOnly)
+bool Manager::readFile(
+  const std::string& path,
+  bool readOnly,
+  bool trustLibrary)
 {
   bool retVal = true;
   pugi::xml_document doc;
-  pugi::xml_parse_result result = doc.load_file(nativePath.c_str());
+
+#ifdef _WIN32
+  pugi::xml_parse_result result = doc.load_file(Utf8ToWide(path).c_str());
+#else
+  pugi::xml_parse_result result = doc.load_file(path.c_str());
+#endif
 
   if (result) {
-    this->parseXmlDom(doc, readOnly, UTF8Path);
+    this->parseXmlDom(doc, readOnly, path, trustLibrary);
   } else {
     retVal = false;
   }
@@ -157,7 +158,7 @@ bool Manager::readFile(const std::string& nativePath,
    * able to know where to save the library if new content are
    * available */
   if (!readOnly) {
-    this->writableLibraryPath = UTF8Path;
+    this->writableLibraryPath = path;
   }
 
   return retVal;
@@ -174,10 +175,10 @@ std::string Manager::addBookFromPathAndGetId(const std::string& pathToOpen,
   kiwix::Book book;
 
   if (this->readBookFromPath(pathToOpen, &book)) {
-    if (pathToSave != pathToOpen) {
+    if (!pathToSave.empty() && pathToSave != pathToOpen) {
       book.setPath(isRelativePath(pathToSave)
                 ? computeAbsolutePath(
-                      removeLastPathElement(writableLibraryPath, true, false),
+                      removeLastPathElement(writableLibraryPath),
                       pathToSave)
                 : pathToSave);
     }
@@ -208,12 +209,15 @@ bool Manager::addBookFromPath(const std::string& pathToOpen,
 
 bool Manager::readBookFromPath(const std::string& path, kiwix::Book* book)
 {
+  std::string tmp_path = path;
+  if (isRelativePath(path)) {
+    tmp_path = computeAbsolutePath(getCurrentDirectory(), path);
+  }
   try {
-    kiwix::Reader reader(path);
+    kiwix::Reader reader(tmp_path);
     book->update(reader);
     book->setPathValid(true);
   } catch (const std::exception& e) {
-    std::cerr << "Invalid " << path << " : " << e.what() << std::endl;
     book->setPathValid(false);
     return false;
   }
@@ -221,4 +225,27 @@ bool Manager::readBookFromPath(const std::string& path, kiwix::Book* book)
   return true;
 }
 
+bool Manager::readBookmarkFile(const std::string& path)
+{
+  pugi::xml_document doc;
+  pugi::xml_parse_result result = doc.load_file(path.c_str());
+
+  if (!result) {
+    return false;
+  }
+
+  pugi::xml_node libraryNode = doc.child("bookmarks");
+
+  for (pugi::xml_node node = libraryNode.child("bookmark"); node;
+       node = node.next_sibling("bookmark")) {
+    kiwix::Bookmark bookmark;
+
+    bookmark.updateFromXml(node);
+
+    manipulator->addBookmarkToLibrary(bookmark);
+  }
+
+  return true;
+}
+
 }

src/meson.build

@@ -1,22 +1,32 @@
 kiwix_sources = [
   'book.cpp',
+  'bookmark.cpp',
   'library.cpp',
   'manager.cpp',
+  'libxml_dumper.cpp',
   'opds_dumper.cpp',
   'downloader.cpp',
   'reader.cpp',
   'entry.cpp',
+  'server.cpp',
   'searcher.cpp',
+  'search_renderer.cpp',
   'subprocess.cpp',
   'aria2.cpp',
-  'common/base64.cpp',
-  'common/pathTools.cpp',
-  'common/regexTools.cpp',
-  'common/stringTools.cpp',
-  'common/networkTools.cpp',
-  'common/otherTools.cpp',
-  'xapian/htmlparse.cc',
-  'xapian/myhtmlparse.cc'
+  'tools/base64.cpp',
+  'tools/pathTools.cpp',
+  'tools/regexTools.cpp',
+  'tools/stringTools.cpp',
+  'tools/networkTools.cpp',
+  'tools/otherTools.cpp',
+  'kiwixserve.cpp',
+  'name_mapper.cpp',
+  'server/byte_range.cpp',
+  'server/etag.cpp',
+  'server/request_context.cpp',
+  'server/response.cpp',
+  'server/internalServer.cpp',
+  'server/internalServer_catalog_v2.cpp'
 ]
 kiwix_sources += lib_resources
 
@@ -26,20 +36,14 @@ else
   kiwix_sources += 'subprocess_unix.cpp'
 endif
 
-if xapian_dep.found()
-  kiwix_sources += ['xapianSearcher.cpp']
-endif
-
-if get_option('android')
-  subdir('android')
+if wrapper.contains('android')
   install_dir = 'kiwix-lib/jniLibs/' + meson.get_cross_property('android_abi')
 else
   install_dir = get_option('libdir')
 endif
 
-
-if has_ctpp2_dep
-  kiwix_sources += ['ctpp2/CTPP2VMStringLoader.cpp']
+if wrapper.contains('android') or wrapper.contains('java')
+  subdir('wrapper/java')
 endif
 
 config_h = configure_file(output : 'kiwix_config.h',
@@ -51,7 +55,7 @@ kiwixlib = library('kiwix',
                    kiwix_sources,
                    include_directories : inc,
                    dependencies : all_deps,
+                   link_args: extra_libs,
                    version: meson.project_version(),
                    install: true,
-                   install_dir: install_dir,
-                   install_rpath: '$ORIGIN')
+                   install_dir: install_dir)

src/microhttpd_wrapper.h

@@ -1,5 +1,5 @@
 /*
- * Copyright 2014 Emmanuel Engelhart <kelson@kiwix.org>
+ * Copyright 2020 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
@@ -17,22 +17,8 @@
  * MA 02110-1301, USA.
  */
 
-#ifndef KIWIX_OTHERTOOLS_H
-#define KIWIX_OTHERTOOLS_H
-
-#ifdef _WIN32
-#include <windows.h>
-#else
-#include <unistd.h>
-#endif
-
-#include <pugixml.hpp>
-
-namespace kiwix
-{
-void sleep(unsigned int milliseconds);
-std::string nodeToString(pugi::xml_node node);
-std::string converta2toa3(const std::string& a2code);
-}
+#include <microhttpd.h>
 
+#if MHD_VERSION < 0x00097002
+typedef int MHD_Result;
 #endif

src/name_mapper.cpp

@@ -0,0 +1,62 @@
+/*
+ * Copyright 2019 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 "name_mapper.h"
+#include "library.h"
+#include "tools/regexTools.h"
+#include <iostream>
+
+namespace kiwix {
+
+HumanReadableNameMapper::HumanReadableNameMapper(kiwix::Library& library, bool withAlias) {
+  for (auto& bookId: library.filter(kiwix::Filter().local(true).valid(true))) {
+    auto& currentBook = library.getBookById(bookId);
+    auto bookName = currentBook.getHumanReadableIdFromPath();
+    m_idToName[bookId] = bookName;
+    m_nameToId[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;
+    }
+  }
+}
+
+std::string HumanReadableNameMapper::getNameForId(const std::string& id) {
+  return m_idToName.at(id);
+}
+
+std::string HumanReadableNameMapper::getIdForName(const std::string& name) {
+  return m_nameToId.at(name);
+}
+
+}

src/opds_dumper.cpp

@@ -20,10 +20,14 @@
 #include "opds_dumper.h"
 #include "book.h"
 
-#include <common/otherTools.h>
+#include "tools/otherTools.h"
+
+#include "kiwixlib-resources.h"
+#include <mustache.hpp>
 
 namespace kiwix
 {
+
 /* Constructor */
 OPDSDumper::OPDSDumper(Library* library)
   : library(library)
@@ -34,104 +38,111 @@ OPDSDumper::~OPDSDumper()
 {
 }
 
-std::string gen_date_str()
-{
-  auto now = time(0);
-  auto tm = localtime(&now);
-
-  std::stringstream is;
-  is << std::setw(2) << std::setfill('0')
-     << 1900+tm->tm_year << "-"
-     << std::setw(2) << std::setfill('0') << tm->tm_mon << "-"
-     << std::setw(2) << std::setfill('0') << tm->tm_mday << "T"
-     << std::setw(2) << std::setfill('0') << tm->tm_hour << ":"
-     << std::setw(2) << std::setfill('0') << tm->tm_min << ":"
-     << std::setw(2) << std::setfill('0') << tm->tm_sec << "Z";
-  return is.str();
-}
-
 void OPDSDumper::setOpenSearchInfo(int totalResults, int startIndex, int count)
 {
   m_totalResults = totalResults;
   m_startIndex = startIndex,
   m_count = count;
-  m_isSearchResult = true;
 }
 
-#define ADD_TEXT_ENTRY(node, child, value) (node).append_child((child)).append_child(pugi::node_pcdata).set_value((value).c_str())
-
-pugi::xml_node OPDSDumper::handleBook(Book book, pugi::xml_node root_node) {
-  auto entry_node = root_node.append_child("entry");
-  ADD_TEXT_ENTRY(entry_node, "title", book.getTitle());
-  ADD_TEXT_ENTRY(entry_node, "id", "urn:uuid:"+book.getId());
-  ADD_TEXT_ENTRY(entry_node, "icon", rootLocation + "/meta?name=favicon&content=" + book.getHumanReadableIdFromPath());
-  ADD_TEXT_ENTRY(entry_node, "updated", date);
-  ADD_TEXT_ENTRY(entry_node, "summary", book.getDescription());
-
-  auto content_node = entry_node.append_child("link");
-  content_node.append_attribute("type") = "text/html";
-  content_node.append_attribute("href") = (rootLocation + "/" + book.getHumanReadableIdFromPath()).c_str();
-
-  auto author_node = entry_node.append_child("author");
-  ADD_TEXT_ENTRY(author_node, "name", book.getCreator());
-
-  if (! book.getUrl().empty()) {
-    auto acquisition_link = entry_node.append_child("link");
-    acquisition_link.append_attribute("rel") = "http://opds-spec.org/acquisition/open-access";
-    acquisition_link.append_attribute("type") = "application/x-zim";
-    acquisition_link.append_attribute("href") = book.getUrl().c_str();
-    acquisition_link.append_attribute("length") = to_string(book.getSize()).c_str();
-  }
-
-  if (! book.getFaviconMimeType().empty() ) {
-    auto image_link = entry_node.append_child("link");
-    image_link.append_attribute("rel") = "http://opds-spec.org/image/thumbnail";
-    image_link.append_attribute("type") = book.getFaviconMimeType().c_str();
-    image_link.append_attribute("href") = (rootLocation + "/meta?name=favicon&content=" + book.getHumanReadableIdFromPath()).c_str();
-  }
-  return entry_node;
-}
-
-string OPDSDumper::dumpOPDSFeed(const std::vector<std::string>& bookIds)
+namespace
 {
-  date = gen_date_str();
-  pugi::xml_document doc;
 
-  auto root_node = doc.append_child("feed");
-  root_node.append_attribute("xmlns") = "http://www.w3.org/2005/Atom";
-  root_node.append_attribute("xmlns:opds") = "http://opds-spec.org/2010/catalog";
+typedef kainjow::mustache::data MustacheData;
+typedef kainjow::mustache::list BookData;
 
-  ADD_TEXT_ENTRY(root_node, "id", id);
-
-  ADD_TEXT_ENTRY(root_node, "title", title);
-  ADD_TEXT_ENTRY(root_node, "updated", date);
-
-  if (m_isSearchResult) {
-    ADD_TEXT_ENTRY(root_node, "totalResults", to_string(m_totalResults));
-    ADD_TEXT_ENTRY(root_node, "startIndex", to_string(m_startIndex));
-    ADD_TEXT_ENTRY(root_node, "itemsPerPage", to_string(m_count));
+BookData getBookData(const Library* library, const std::vector<std::string>& bookIds)
+{
+  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()},
+      {"name", book.getName()},
+      {"title", book.getTitle()},
+      {"description", book.getDescription()},
+      {"language", book.getLanguage()},
+      {"content_id",  book.getHumanReadableIdFromPath()},
+      {"updated", book.getDate() + "T00:00:00Z"},
+      {"category", book.getCategory()},
+      {"flavour", book.getFlavour()},
+      {"tags", book.getTags()},
+      {"article_count", to_string(book.getArticleCount())},
+      {"media_count", to_string(book.getMediaCount())},
+      {"author_name", book.getCreator()},
+      {"publisher_name", book.getPublisher()},
+      {"url", bookUrl},
+      {"size", to_string(book.getSize())},
+    });
   }
 
-  auto self_link_node = root_node.append_child("link");
-  self_link_node.append_attribute("rel") = "self";
-  self_link_node.append_attribute("href") = "";
-  self_link_node.append_attribute("type") = "application/atom+xml";
+  return bookData;
+}
 
+} // unnamed namespace
 
-  if (!searchDescriptionUrl.empty() ) {
-    auto search_link = root_node.append_child("link");
-    search_link.append_attribute("rel") = "search";
-    search_link.append_attribute("type") = "application/opensearchdescription+xml";
-    search_link.append_attribute("href") = searchDescriptionUrl.c_str();
+string OPDSDumper::dumpOPDSFeed(const std::vector<std::string>& bookIds, const std::string& query) const
+{
+  const auto bookData = getBookData(library, bookIds);
+  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)},
+     {"totalResults", to_string(m_totalResults)},
+     {"startIndex", to_string(m_startIndex)},
+     {"itemsPerPage", to_string(m_count)},
+     {"books", bookData }
+  };
+
+  return render_template(RESOURCE::templates::catalog_entries_xml, template_data);
+}
+
+string OPDSDumper::dumpOPDSFeedV2(const std::vector<std::string>& bookIds, const std::string& query) const
+{
+  const auto bookData = getBookData(library, bookIds);
+
+  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)},
+     {"totalResults", to_string(m_totalResults)},
+     {"startIndex", to_string(m_startIndex)},
+     {"itemsPerPage", to_string(m_count)},
+     {"books", bookData }
+  };
+
+  return render_template(RESOURCE::templates::catalog_v2_entries_xml, template_data);
+}
+
+std::string OPDSDumper::categoriesOPDSFeed(const std::vector<std::string>& categories) 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)}
+    });
   }
 
-  if (library) {
-    for (auto& bookId: bookIds) {
-      handleBook(library->getBookById(bookId), root_node);
-    }
-  }
-
-  return nodeToString(root_node);
+  return render_template(
+             RESOURCE::templates::catalog_v2_categories_xml,
+             kainjow::mustache::object{
+               {"date", now},
+               {"endpoint_root", rootLocation + "/catalog/v2"},
+               {"feed_id", gen_uuid(libraryId + "/categories")},
+               {"categories", categoryData }
+             }
+  );
 }
 
 }

src/reader.cpp

@@ -21,6 +21,10 @@
 #include <time.h>
 
 #include <zim/search.h>
+#include <zim/item.h>
+#include <zim/error.h>
+
+#include "tools/otherTools.h"
 
 inline char hi(char v)
 {
@@ -63,7 +67,9 @@ std::string hexUUID(std::string in)
 namespace kiwix
 {
 /* Constructor */
-Reader::Reader(const string zimFilePath) : zimFileHandler(NULL)
+Reader::Reader(const string zimFilePath)
+  :  zimArchive(nullptr),
+     zimFilePath(zimFilePath)
 {
   string tmpZimFilePath = zimFilePath;
 
@@ -74,58 +80,45 @@ Reader::Reader(const string zimFilePath) : zimFileHandler(NULL)
     tmpZimFilePath.resize(tmpZimFilePath.size() - 2);
   }
 
-  this->zimFileHandler = new zim::File(tmpZimFilePath);
-
-  if (this->zimFileHandler != NULL) {
-    this->firstArticleOffset
-        = this->zimFileHandler->getNamespaceBeginOffset('A');
-    this->lastArticleOffset = this->zimFileHandler->getNamespaceEndOffset('A');
-    this->nsACount = this->zimFileHandler->getNamespaceCount('A');
-    this->nsICount = this->zimFileHandler->getNamespaceCount('I');
-    this->zimFilePath = zimFilePath;
-  }
+  zimArchive.reset(new zim::Archive(tmpZimFilePath));
 
   /* initialize random seed: */
-  srand(time(NULL));
+  srand(time(nullptr));
 }
 
-/* Destructor */
-Reader::~Reader()
+#ifndef _WIN32
+Reader::Reader(int fd)
+  :  zimArchive(new zim::Archive(fd)),
+     zimFilePath("")
 {
-  if (this->zimFileHandler != NULL) {
-    delete this->zimFileHandler;
+  /* 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 {};
   }
 }
 
-zim::File* Reader::getZimFileHandler() const
-{
-  return this->zimFileHandler;
-}
-std::map<const std::string, unsigned int> Reader::parseCounterMetadata() const
-{
-  std::map<const std::string, unsigned int> counters;
-  string mimeType, item, counterString;
-  unsigned int counter;
-
-  zim::Article article = this->zimFileHandler->getArticle('M', "Counter");
-
-  if (article.good()) {
-    stringstream ssContent(article.getData());
-
-    while (getline(ssContent, item, ';')) {
-      stringstream ssItem(item);
-      getline(ssItem, mimeType, '=');
-      getline(ssItem, counterString, '=');
-      if (!counterString.empty() && !mimeType.empty()) {
-        sscanf(counterString.c_str(), "%u", &counter);
-        counters.insert(pair<string, int>(mimeType, counter));
-      }
-    }
-  }
-
-  return counters;
-}
-
 /* Get the count of articles which can be indexed/displayed */
 unsigned int Reader::getArticleCount() const
 {
@@ -133,12 +126,9 @@ unsigned int Reader::getArticleCount() const
       = this->parseCounterMetadata();
   unsigned int counter = 0;
 
-  if (counterMap.empty()) {
-    counter = this->nsACount;
-  } else {
-    auto it = counterMap.find("text/html");
-    if (it != counterMap.end()) {
-      counter = it->second;
+  for(auto &pair:counterMap) {
+    if (startsWith(pair.first, "text/html")) {
+      counter += pair.second;
     }
   }
 
@@ -152,151 +142,84 @@ unsigned int Reader::getMediaCount() const
       = this->parseCounterMetadata();
   unsigned int counter = 0;
 
-  if (counterMap.empty()) {
-    counter = this->nsICount;
-  } else {
-    auto it = counterMap.find("image/jpeg");
-    if (it != counterMap.end()) {
-      counter += it->second;
-    }
-
-    it = counterMap.find("image/gif");
-    if (it != counterMap.end()) {
-      counter += it->second;
-    }
-
-    it = counterMap.find("image/png");
-    if (it != counterMap.end()) {
-      counter += it->second;
+  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 this->zimFileHandler->getCountArticles();
+  return zimArchive->getEntryCount();
 }
 
 /* Return the UID of the ZIM file */
 string Reader::getId() const
 {
   std::ostringstream s;
-  s << this->zimFileHandler->getFileheader().getUuid();
+  s << zimArchive->getUuid();
   return s.str();
 }
 
-/* Return a page url from a title */
-bool Reader::getPageUrlFromTitle(const string& title, string& url) const
-{
-  try {
-    auto entry = getEntryFromTitle(title);
-    entry = entry.getFinalEntry();
-    url = entry.getPath();
-    return true;
-  } catch (NoEntry& e) {
-    return false;
-  }
-}
-
-/* Return an URL from a title */
-string Reader::getRandomPageUrl() const
-{
-  return getRandomPage().getPath();
-}
-
 Entry Reader::getRandomPage() const
 {
-  if (!this->zimFileHandler) {
+  try {
+    return zimArchive->getRandomEntry();
+  } catch(...) {
     throw NoEntry();
   }
-
-  zim::Article article;
-  std::string mainPagePath = this->getMainPage().getPath();
-  int watchdog = 42;
-
-  do {
-    auto idx = this->firstArticleOffset
-          + (zim::size_type)((double)rand() / ((double)RAND_MAX + 1)
-                             * this->nsACount);
-    article = zimFileHandler->getArticle(idx);
-    if (!watchdog--) {
-      throw NoEntry();
-    }
-  } while (!article.good() && article.getLongUrl() == mainPagePath);
-
-  return article;
-}
-
-/* Return the welcome page URL */
-string Reader::getMainPageUrl() const
-{
-  return getMainPage().getPath();
 }
 
 Entry Reader::getMainPage() const
 {
-  if (!this->zimFileHandler) {
-    throw NoEntry();
-  }
-
-  string url = "";
-
-  zim::Article article;
-  if (this->zimFileHandler->getFileheader().hasMainPage())
-  {
-    article = zimFileHandler->getArticle(
-        this->zimFileHandler->getFileheader().getMainPage());
-  }
-
-  if (!article.good())
-  {
-    return getFirstPage();
-  }
-
-  return article;
+  return zimArchive->getMainEntry();
 }
 
 bool Reader::getFavicon(string& content, string& mimeType) const
 {
-  static const char* const paths[] = {"-/favicon.png", "I/favicon.png", "I/favicon", "-/favicon"};
-
-  for (auto &path: paths) {
-    try {
-      auto entry = getEntryFromPath(path);
-      entry = entry.getFinalEntry();
-      content = entry.getContent();
-      mimeType = entry.getMimetype();
-      return true;
-    } catch(NoEntry& e) {};
-  }
+  try {
+    auto item = zimArchive->getIllustrationItem();
+    content = item.getData();
+    mimeType = item.getMimetype();
+    return true;
+  } catch(zim::EntryNotFound& e) {};
 
   return false;
 }
 
 string Reader::getZimFilePath() const
 {
-  return this->zimFilePath;
+  return zimFilePath;
 }
 /* Return a metatag value */
-bool Reader::getMetatag(const string& name, string& value) const
+bool Reader::getMetadata(const string& name, string& value) const
 {
   try {
-    auto entry = getEntryFromPath("M/"+name);
-    value = entry.getContent();
+    value = zimArchive->getMetadata(name);
     return true;
-  } catch(NoEntry& e) {
+  } 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;
-  this->getMetatag("Title", value);
+  string value = zimArchive->getMetadata("Title");
   if (value.empty()) {
-    value = getLastPathElement(zimFileHandler->getFilename());
+    value = getLastPathElement(zimFilePath);
     std::replace(value.begin(), value.end(), '_', ' ');
     size_t pos = value.find(".zim");
     value = value.substr(0, pos);
@@ -304,65 +227,98 @@ string Reader::getTitle() const
   return value;
 }
 
-string Reader::getName() const
+string Reader::getCreator() const
 {
-  string value;
-  this->getMetatag("Name", value);
-  return value;
+  METADATA("Creator")
 }
 
-string Reader::getTags() const
+string Reader::getPublisher() const
 {
-  string value;
-  this->getMetatag("Tags", value);
-  return value;
+  METADATA("Publisher")
+}
+
+string Reader::getDate() const
+{
+  METADATA("Date")
 }
 
 string Reader::getDescription() const
 {
   string value;
-  this->getMetatag("Description", value);
+  this->getMetadata("Description", value);
 
   /* Mediawiki Collection tends to use the "Subtitle" name */
   if (value.empty()) {
-    this->getMetatag("Subtitle", value);
+    this->getMetadata("Subtitle", value);
   }
 
   return value;
 }
 
+string Reader::getLongDescription() const
+{
+  METADATA("LongDescription")
+}
+
 string Reader::getLanguage() const
 {
-  string value;
-  this->getMetatag("Language", value);
-  return value;
+  METADATA("Language")
 }
 
-string Reader::getDate() const
+string Reader::getLicense() const
 {
-  string value;
-  this->getMetatag("Date", value);
-  return value;
+  METADATA("License")
 }
 
-string Reader::getCreator() const
+string Reader::getTags(bool original) const
 {
-  string value;
-  this->getMetatag("Creator", value);
-  return value;
+  string tags_str;
+  getMetadata("Tags", tags_str);
+  if (original) {
+    return tags_str;
+  }
+  auto tags = convertTags(tags_str);
+  return join(tags, ";");
 }
 
-string Reader::getPublisher() const
+
+string Reader::getTagStr(const std::string& tagName) const
 {
-  string value;
-  this->getMetatag("Publisher", value);
-  return value;
+  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->getMetatag("startfileuid", value);
+  this->getMetadata("startfileuid", value);
   if (value.empty()) {
     return "";
   }
@@ -384,83 +340,17 @@ string Reader::getOrigId() const
   return origID;
 }
 
-/* Return the first page URL */
-string Reader::getFirstPageUrl() const
-{
-  return getFirstPage().getPath();
-}
-
-Entry Reader::getFirstPage() const
-{
-  if (!this->zimFileHandler) {
-    throw NoEntry();
-  }
-
-  auto firstPageOffset = zimFileHandler->getNamespaceBeginOffset('A');
-  auto article = zimFileHandler->getArticle(firstPageOffset);
-
-  if (! article.good()) {
-    throw NoEntry();
-  }
-
-  return article;
-}
-
-bool _parseUrl(const string& url, char* ns, string& title)
-{
-  /* Offset to visit the url */
-  unsigned int urlLength = url.size();
-  unsigned int offset = 0;
-
-  /* Ignore the first '/' */
-  if (url[offset] == '/')
-    offset++;
-
-  if (url[offset] == '/' || offset >= urlLength)
-    return false;
-
-  /* Get namespace */
-  *ns = url[offset++];
-
-  if (url[offset] != '/' || offset >= urlLength)
-    return false;
-
-  offset++;
-
-  if ( offset >= urlLength)
-    return false;
-
-  /* Get content title */
-  title = url.substr(offset, urlLength - offset);
-
-  return true;
-}
-
-bool Reader::parseUrl(const string& url, char* ns, string& title) const
-{
-  return _parseUrl(url, ns, title);
-}
-
 Entry Reader::getEntryFromPath(const std::string& path) const
 {
-  char ns = 0;
-  std::string short_url;
-
-  if (!this->zimFileHandler) {
-    throw NoEntry();
-  }
-  _parseUrl(path, &ns, short_url);
-
-  if (short_url.empty() && ns == 0) {
+  if (path.empty() || path == "/") {
     return getMainPage();
   }
 
-  auto article = zimFileHandler->getArticle(ns, short_url);
-  if (!article.good()) {
+  try {
+    return zimArchive->getEntryByPath(path);
+  } catch (zim::EntryNotFound& e) {
     throw NoEntry();
   }
-
-  return article;
 }
 
 Entry Reader::getEntryFromEncodedPath(const std::string& path) const
@@ -470,212 +360,30 @@ Entry Reader::getEntryFromEncodedPath(const std::string& path) const
 
 Entry Reader::getEntryFromTitle(const std::string& title) const
 {
-  if (!this->zimFileHandler) {
+  try {
+    return zimArchive->getEntryByTitle(title);
+  } catch(zim::EntryNotFound& e) {
     throw NoEntry();
   }
-
-  auto article = this->zimFileHandler->getArticleByTitle('A', title);
-  if (!article.good()) {
-    throw NoEntry();
-  }
-
-  return article;
-}
-
-/* Return article by url */
-bool Reader::getArticleObjectByDecodedUrl(const string& url,
-                                          zim::Article& article) const
-{
-  if (this->zimFileHandler == NULL) {
-    return false;
-  }
-
-  /* Parse the url */
-  char ns = 0;
-  string urlStr;
-  _parseUrl(url, &ns, urlStr);
-
-  /* Main page */
-  if (urlStr.empty() && ns == 0) {
-    _parseUrl(this->getMainPage().getPath(), &ns, urlStr);
-  }
-
-  /* Extract the content from the zim file */
-  article = zimFileHandler->getArticle(ns, urlStr);
-  return article.good();
-}
-
-/* Return the mimeType without the content */
-bool Reader::getMimeTypeByUrl(const string& url, string& mimeType) const
-{
-  try {
-    auto entry = getEntryFromPath(url);
-    mimeType = entry.getMimetype();
-    return true;
-  } catch (NoEntry& e) {
-    mimeType = "";
-    return false;
-  }
-}
-
-bool get_content_by_decoded_url(const Reader& reader,
-                                const string& url,
-                                string& content,
-                                string& title,
-                                unsigned int& contentLength,
-                                string& contentType,
-                                string& baseUrl)
-{
-  content = "";
-  contentType = "";
-  contentLength = 0;
-
-  try {
-    auto entry = reader.getEntryFromPath(url);
-    entry = entry.getFinalEntry();
-    baseUrl = entry.getPath();
-    contentType = entry.getMimetype();
-    content = entry.getContent();
-    contentLength = entry.getSize();
-    title = entry.getTitle();
-
-    /* Try to set a stub HTML header/footer if necesssary */
-    if (contentType.find("text/html") != string::npos
-      && content.find("<body") == std::string::npos
-      && content.find("<BODY") == std::string::npos) {
-      content = "<html><head><title>" + title +
-              "</title><meta http-equiv=\"Content-Type\" content=\"text/html; "
-              "charset=utf-8\" /></head><body>" +
-              content + "</body></html>";
-    }
-    return true;
-  } catch (NoEntry& e) {
-    return false;
-  }
-}
-
-
-/* Get a content from a zim file */
-bool Reader::getContentByUrl(const string& url,
-                             string& content,
-                             string& title,
-                             unsigned int& contentLength,
-                             string& contentType) const
-{
-  std::string stubRedirectUrl;
-  return get_content_by_decoded_url(*this,
-                                kiwix::urlDecode(url),
-                                content,
-                                title,
-                                contentLength,
-                                contentType,
-                                stubRedirectUrl);
-}
-
-bool Reader::getContentByEncodedUrl(const string& url,
-                                    string& content,
-                                    string& title,
-                                    unsigned int& contentLength,
-                                    string& contentType,
-                                    string& baseUrl) const
-{
-  return get_content_by_decoded_url(*this,
-                                kiwix::urlDecode(url),
-                                content,
-                                title,
-                                contentLength,
-                                contentType,
-                                baseUrl);
-}
-
-bool Reader::getContentByEncodedUrl(const string& url,
-                                    string& content,
-                                    string& title,
-                                    unsigned int& contentLength,
-                                    string& contentType) const
-{
-  std::string stubRedirectUrl;
-  return get_content_by_decoded_url(*this,
-                                kiwix::urlDecode(url),
-                                content,
-                                title,
-                                contentLength,
-                                contentType,
-                                stubRedirectUrl);
-}
-
-bool Reader::getContentByDecodedUrl(const string& url,
-                                    string& content,
-                                    string& title,
-                                    unsigned int& contentLength,
-                                    string& contentType) const
-{
-  std::string stubRedirectUrl;
-  return get_content_by_decoded_url(*this,
-                                url,
-                                content,
-                                title,
-                                contentLength,
-                                contentType,
-                                stubRedirectUrl);
-}
-
-bool Reader::getContentByDecodedUrl(const string& url,
-                                    string& content,
-                                    string& title,
-                                    unsigned int& contentLength,
-                                    string& contentType,
-                                    string& baseUrl) const
-{
-  return get_content_by_decoded_url(*this,
-                                url,
-                                content,
-                                title,
-                                contentLength,
-                                contentType,
-                                baseUrl);
-}
-
-/* Check if an article exists */
-bool Reader::urlExists(const string& url) const
-{
-  return pathExists(url);
 }
 
 bool Reader::pathExists(const string& path) const
 {
-  if (!zimFileHandler)
-  {
-    return false;
-  }
-
-  char ns = 0;
-  string titleStr;
-  _parseUrl(path, &ns, titleStr);
-  zim::File::const_iterator findItr = zimFileHandler->find(ns, titleStr);
-  return findItr != zimFileHandler->end() && findItr->getUrl() == titleStr;
+  return zimArchive->hasEntryByPath(path);
 }
 
 /* Does the ZIM file has a fulltext index */
 bool Reader::hasFulltextIndex() const
 {
-  if (!zimFileHandler || zimFileHandler->is_multiPart() )
-  {
-    return false;
-  }
-
-  return ( pathExists("Z//fulltextIndex/xapian")
-        || pathExists("X/fulltext/xapian"));
+  return zimArchive->hasFulltextIndex();
 }
 
 /* Search titles by prefix */
+
 bool Reader::searchSuggestions(const string& prefix,
                                unsigned int suggestionsCount,
                                const bool reset)
 {
-  bool retVal = false;
-  zim::File::const_iterator articleItr;
-
   /* Reset the suggestions otherwise check if the suggestions number is less
    * than the suggestionsCount */
   if (reset) {
@@ -687,39 +395,48 @@ bool Reader::searchSuggestions(const string& prefix,
     }
   }
 
+  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 (articleItr = zimFileHandler->findByTitle('A', prefix);
-       articleItr != zimFileHandler->end()
-       && articleItr->getTitle().compare(0, prefix.size(), prefix) == 0
-       && this->suggestions.size() < suggestionsCount;
-       ++articleItr) {
+  for (auto& entry: zimArchive->findByTitle(prefix)) {
+    if (results.size() >= suggestionsCount) {
+      break;
+    }
     /* Extract the interesting part of article title & url */
     std::string normalizedArticleTitle
-        = kiwix::normalize(articleItr->getTitle());
-    std::string articleFinalUrl = "/A/" + articleItr->getUrl();
-    if (articleItr->isRedirect()) {
-      zim::Article article = *articleItr;
-      unsigned int loopCounter = 0;
-      while (article.isRedirect() && loopCounter++ < 42) {
-        article = article.getRedirectArticle();
-      }
-      articleFinalUrl = "/A/" + article.getUrl();
-    }
+        = 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<std::vector<std::string>>::iterator suggestionItr;
-    for (suggestionItr = this->suggestions.begin();
-         suggestionItr != this->suggestions.end();
+    std::vector<SuggestionItem>::iterator suggestionItr;
+    for (suggestionItr = results.begin();
+         suggestionItr != results.end();
          suggestionItr++) {
-      int result = normalizedArticleTitle.compare((*suggestionItr)[2]);
-      if (result == 0 && articleFinalUrl.compare((*suggestionItr)[1]) == 0) {
+      int result = normalizedArticleTitle.compare((*suggestionItr).getNormalizedTitle());
+      if (result == 0 && articleFinalUrl.compare((*suggestionItr).getPath()) == 0) {
         insert = false;
         break;
       } else if (result < 0) {
@@ -729,20 +446,14 @@ bool Reader::searchSuggestions(const string& prefix,
 
     /* Insert if possible */
     if (insert) {
-      std::vector<std::string> suggestion;
-      suggestion.push_back(articleItr->getTitle());
-      suggestion.push_back(articleFinalUrl);
-      suggestion.push_back(normalizedArticleTitle);
-      this->suggestions.insert(suggestionItr, suggestion);
+      SuggestionItem suggestion(entry.getTitle(), normalizedArticleTitle, articleFinalUrl);
+      results.insert(suggestionItr, suggestion);
     }
 
     /* Suggestions where found */
     retVal = true;
   }
 
-  /* Set the cursor to the begining */
-  this->suggestionsOffset = this->suggestions.begin();
-
   return retVal;
 }
 
@@ -757,36 +468,50 @@ std::vector<std::string> Reader::getTitleVariants(
   return variants;
 }
 
-/* Try also a few variations of the prefix to have better results */
+
 bool Reader::searchSuggestionsSmart(const string& prefix,
                                     unsigned int suggestionsCount)
 {
-  std::vector<std::string> variants = this->getTitleVariants(prefix);
-  bool retVal;
-
   this->suggestions.clear();
   this->suggestionsOffset = this->suggestions.begin();
-  /* Try to search in the title using fulltext search database */
-  const zim::Search* suggestionSearch
-      = this->getZimFileHandler()->suggestions(prefix, 0, suggestionsCount);
 
-  if (suggestionSearch->get_matches_estimated()) {
-    for (auto current = suggestionSearch->begin();
-         current != suggestionSearch->end();
+  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++) {
-      std::vector<std::string> suggestion;
-      suggestion.push_back(current->getTitle());
-      suggestion.push_back("/A/" + current->getUrl());
-      suggestion.push_back(kiwix::normalize(current->getTitle()));
-      this->suggestions.push_back(suggestion);
+      SuggestionItem suggestion(current.getTitle(), kiwix::normalize(current.getTitle()),
+                                current.getPath(), current.getSnippet());
+      results.push_back(suggestion);
     }
-    this->suggestionsOffset = this->suggestions.begin();
     retVal = true;
   } else {
     for (std::vector<std::string>::iterator variantsItr = variants.begin();
          variantsItr != variants.end();
          variantsItr++) {
-      retVal = this->searchSuggestions(*variantsItr, suggestionsCount, false)
+      retVal = this->searchSuggestions(*variantsItr, suggestionsCount, results)
                || retVal;
     }
   }
@@ -799,7 +524,7 @@ bool Reader::getNextSuggestion(string& title)
 {
   if (this->suggestionsOffset != this->suggestions.end()) {
     /* title */
-    title = (*(this->suggestionsOffset))[0];
+    title = (*(this->suggestionsOffset)).getTitle();
 
     /* increment the cursor for the next call */
     this->suggestionsOffset++;
@@ -814,8 +539,8 @@ bool Reader::getNextSuggestion(string& title, string& url)
 {
   if (this->suggestionsOffset != this->suggestions.end()) {
     /* title */
-    title = (*(this->suggestionsOffset))[0];
-    url = (*(this->suggestionsOffset))[1];
+    title = (*(this->suggestionsOffset)).getTitle();
+    url = (*(this->suggestionsOffset)).getPath();
 
     /* increment the cursor for the next call */
     this->suggestionsOffset++;
@@ -829,14 +554,14 @@ bool Reader::getNextSuggestion(string& title, string& url)
 /* Check if the file has as checksum */
 bool Reader::canCheckIntegrity() const
 {
-  return this->zimFileHandler->getChecksum() != "";
+  return zimArchive->hasChecksum();
 }
 
 /* Return true if corrupted, false otherwise */
 bool Reader::isCorrupted() const
 {
   try {
-    if (this->zimFileHandler->verify() == true) {
+    if (zimArchive->check() == true) {
       return false;
     }
   } catch (exception& e) {
@@ -850,13 +575,7 @@ bool Reader::isCorrupted() const
 /* Return the file size, works also for splitted files */
 unsigned int Reader::getFileSize() const
 {
-  zim::File* file = this->getZimFileHandler();
-  zim::offset_type size = 0;
-
-  if (file != NULL) {
-    size = file->getFilesize();
-  }
-
-  return (size / 1024);
+  return zimArchive->getFilesize() / 1024;
 }
+
 }

src/search_renderer.cpp

@@ -0,0 +1,149 @@
+/*
+ * 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 "search_renderer.h"
+#include "searcher.h"
+#include "reader.h"
+#include "library.h"
+#include "name_mapper.h"
+
+#include <zim/search.h>
+
+#include <mustache.hpp>
+#include "kiwixlib-resources.h"
+
+
+namespace kiwix
+{
+
+/* Constructor */
+SearchRenderer::SearchRenderer(Searcher* searcher, NameMapper* mapper)
+    : mp_searcher(searcher),
+      mp_nameMapper(mapper),
+      protocolPrefix("zim://"),
+      searchProtocolPrefix("search://?")
+{}
+
+/* Destructor */
+SearchRenderer::~SearchRenderer() = default;
+
+void SearchRenderer::setSearchPattern(const std::string& pattern)
+{
+  this->searchPattern = pattern;
+}
+
+void SearchRenderer::setSearchContent(const std::string& name)
+{
+  this->searchContent = name;
+}
+
+void SearchRenderer::setProtocolPrefix(const std::string& prefix)
+{
+  this->protocolPrefix = prefix;
+}
+
+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;
+  }
+
+  // pages
+  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 (estimatedResultCount > pageLength) {
+      lastPageStart = ((estimatedResultCount-1)/pageLength)*pageLength;
+    }
+  }
+
+  resultEnd = resultStart+pageLength; //setting result end
+
+  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);
+    }
+    pages.push_back(page);
+  }
+
+  std::string template_str = RESOURCE::templates::search_result_html;
+  kainjow::mustache::mustache tmpl(template_str);
+
+  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);
+
+  std::stringstream ss;
+  tmpl.render(allData, [&ss](const std::string& str) { ss << str; });
+  return ss.str();
+}
+
+}

src/searcher.cpp

@@ -22,20 +22,12 @@
 
 #include "searcher.h"
 #include "reader.h"
-#include "xapianSearcher.h"
 
 #include <zim/search.h>
 
-#ifdef ENABLE_CTPP2
-#include <ctpp2/CDT.hpp>
-#include <ctpp2/CTPP2FileLogger.hpp>
-#include <ctpp2/CTPP2SimpleVM.hpp>
-#include "ctpp2/CTPP2VMStringLoader.hpp"
+#include <mustache.hpp>
 #include "kiwixlib-resources.h"
 
-using namespace CTPP;
-#endif
-
 #define MAX_SEARCH_LEN 140
 
 namespace kiwix
@@ -43,7 +35,7 @@ namespace kiwix
 class _Result : public Result
 {
  public:
-  _Result(zim::Search::iterator& iterator);
+  _Result(zim::SearchResultSet::iterator iterator);
   virtual ~_Result(){};
 
   virtual std::string get_url();
@@ -53,60 +45,28 @@ class _Result : public Result
   virtual std::string get_content();
   virtual int get_wordCount();
   virtual int get_size();
-  virtual int get_readerIndex();
+  virtual std::string get_zimId();
 
  private:
-  zim::Search::iterator iterator;
+  zim::SearchResultSet::iterator iterator;
 };
 
-struct SearcherInternal {
-  const zim::Search* _search;
-  XapianSearcher* _xapianSearcher;
-  zim::Search::iterator current_iterator;
-
-  SearcherInternal() : _search(NULL), _xapianSearcher(NULL) {}
-  ~SearcherInternal()
+struct SearcherInternal : zim::SearchResultSet {
+  explicit SearcherInternal(const zim::SearchResultSet& srs)
+    : zim::SearchResultSet(srs)
+    , current_iterator(srs.begin())
   {
-    if (_search != NULL) {
-      delete _search;
-    }
-    if (_xapianSearcher != NULL) {
-      delete _xapianSearcher;
-    }
   }
+
+  zim::SearchResultSet::iterator current_iterator;
 };
 
 /* Constructor */
-Searcher::Searcher(const string& xapianDirectoryPath,
-                   Reader* reader,
-                   const string& humanReadableName)
-    : internal(new SearcherInternal()),
-      searchPattern(""),
-      protocolPrefix("zim://"),
-      searchProtocolPrefix("search://?"),
-      resultCountPerPage(0),
+Searcher::Searcher()
+    : searchPattern(""),
       estimatedResultCount(0),
       resultStart(0),
-      resultEnd(0),
-      contentHumanReadableId(humanReadableName)
-{
-  loadICUExternalTables();
-  if (!reader || !reader->hasFulltextIndex()) {
-    internal->_xapianSearcher = new XapianSearcher(xapianDirectoryPath, reader);
-  }
-  this->humanReaderNames.push_back(humanReadableName);
-}
-
-Searcher::Searcher(const std::string& humanReadableName)
-    : internal(new SearcherInternal()),
-      searchPattern(""),
-      protocolPrefix("zim://"),
-      searchProtocolPrefix("search://?"),
-      resultCountPerPage(0),
-      estimatedResultCount(0),
-      resultStart(0),
-      resultEnd(0),
-      contentHumanReadableId(humanReadableName)
+      resultEnd(0)
 {
   loadICUExternalTables();
 }
@@ -114,21 +74,25 @@ Searcher::Searcher(const std::string& humanReadableName)
 /* Destructor */
 Searcher::~Searcher()
 {
-  delete internal;
 }
 
-bool Searcher::add_reader(Reader* reader, const std::string& humanReadableName)
+bool Searcher::add_reader(Reader* reader)
 {
   if (!reader->hasFulltextIndex()) {
       return false;
   }
   this->readers.push_back(reader);
-  this->humanReaderNames.push_back(humanReadableName);
   return true;
 }
 
+
+Reader* Searcher::get_reader(int readerIndex)
+{
+  return readers.at(readerIndex);
+}
+
 /* Search strings in the database */
-void Searcher::search(std::string& search,
+void Searcher::search(const std::string& search,
                       unsigned int resultStart,
                       unsigned int resultEnd,
                       const bool verbose)
@@ -139,47 +103,27 @@ void Searcher::search(std::string& search,
     cout << "Performing query `" << search << "'" << endl;
   }
 
-  /* If resultEnd & resultStart inverted */
-  if (resultStart > resultEnd) {
-    resultEnd += resultStart;
-    resultStart = resultEnd - resultStart;
-    resultEnd -= resultStart;
-  }
-
+  this->searchPattern = search;
+  this->resultStart = resultStart;
+  this->resultEnd = resultEnd;
   /* Try to find results */
   if (resultStart != resultEnd) {
-    /* Avoid big researches */
-    this->resultCountPerPage = resultEnd - resultStart;
-    if (this->resultCountPerPage > MAX_SEARCH_LEN) {
-      resultEnd = resultStart + MAX_SEARCH_LEN;
-      this->resultCountPerPage = MAX_SEARCH_LEN;
-    }
-
     /* Perform the search */
-    this->searchPattern = search;
-    this->resultStart = resultStart;
-    this->resultEnd = resultEnd;
     string unaccentedSearch = removeAccents(search);
-    if (internal->_xapianSearcher) {
-      internal->_xapianSearcher->searchInIndex(
-          unaccentedSearch, resultStart, resultEnd, verbose);
-      this->estimatedResultCount
-          = internal->_xapianSearcher->results.get_matches_estimated();
-    } else {
-      std::vector<const zim::File*> zims;
-      for (auto current = this->readers.begin(); current != this->readers.end();
-           current++) {
-        if ( (*current)->hasFulltextIndex() ) {
-            zims.push_back((*current)->getZimFileHandler());
-        }
+    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::Search* search = new zim::Search(zims);
-      search->set_query(unaccentedSearch);
-      search->set_range(resultStart, resultEnd);
-      internal->_search = search;
-      internal->current_iterator = internal->_search->begin();
-      this->estimatedResultCount = internal->_search->get_matches_estimated();
     }
+    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;
@@ -197,29 +141,6 @@ void Searcher::geo_search(float latitude, float longitude, float distance,
     cout << "Performing geo query `" << distance << "&(" << latitude << ";" << longitude << ")'" << endl;
   }
 
-  /* If resultEnd & resultStart inverted */
-  if (resultStart > resultEnd) {
-    resultEnd += resultStart;
-    resultStart = resultEnd - resultStart;
-    resultEnd -= resultStart;
-  }
-
-  /* Try to find results */
-  if (resultStart == resultEnd) {
-    return;
-  }
-
-  if (internal->_xapianSearcher) {
-    return;
-  }
-
-  /* Avoid big researches */
-  this->resultCountPerPage = resultEnd - resultStart;
-  if (this->resultCountPerPage > MAX_SEARCH_LEN) {
-    resultEnd = resultStart + MAX_SEARCH_LEN;
-    this->resultCountPerPage = MAX_SEARCH_LEN;
-  }
-
   /* Perform the search */
   std::ostringstream oss;
   oss << "Articles located less than " << distance << " meters of " << latitude << ";" << longitude;
@@ -227,36 +148,38 @@ void Searcher::geo_search(float latitude, float longitude, float distance,
   this->resultStart = resultStart;
   this->resultEnd = resultEnd;
 
-  std::vector<const zim::File*> zims;
+  /* Try to find results */
+  if (resultStart == resultEnd) {
+    return;
+  }
+
+  std::vector<zim::Archive> archives;
   for (auto current = this->readers.begin(); current != this->readers.end();
        current++) {
-    zims.push_back((*current)->getZimFileHandler());
+    archives.push_back(*(*current)->getZimArchive());
   }
-  zim::Search* search = new zim::Search(zims);
-  search->set_query("");
-  search->set_georange(latitude, longitude, distance);
-  search->set_range(resultStart, resultEnd);
-  internal->_search = search;
-  internal->current_iterator = internal->_search->begin();
-  this->estimatedResultCount = internal->_search->get_matches_estimated();
+  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->_xapianSearcher) {
-    internal->_xapianSearcher->restart_search();
-  } else if (internal->_search) {
-    internal->current_iterator = internal->_search->begin();
+  if (internal.get()) {
+    internal->current_iterator = internal->begin();
   }
 }
 
 Result* Searcher::getNextResult()
 {
-  if (internal->_xapianSearcher) {
-    return internal->_xapianSearcher->getNextResult();
-  } else if (internal->_search &&
-             internal->current_iterator != internal->_search->end()) {
+  if (internal.get() &&
+             internal->current_iterator != internal->end()) {
     Result* result = new _Result(internal->current_iterator);
     internal->current_iterator++;
     return result;
@@ -272,37 +195,31 @@ void Searcher::reset()
   return;
 }
 
-void Searcher::suggestions(std::string& search, const bool verbose)
+void Searcher::suggestions(std::string& searchPattern, const bool verbose)
 {
   this->reset();
 
   if (verbose == true) {
-    cout << "Performing suggestion query `" << search << "`" << endl;
+    cout << "Performing suggestion query `" << searchPattern << "`" << endl;
   }
 
-  this->searchPattern = search;
+  this->searchPattern = searchPattern;
   this->resultStart = 0;
   this->resultEnd = 10;
-  string unaccentedSearch = removeAccents(search);
+  string unaccentedSearch = removeAccents(searchPattern);
 
-  if (internal->_xapianSearcher) {
-    /* [TODO] Suggestion on a external database ?
-     * We do not support that. */
-    this->estimatedResultCount = 0;
-  } else {
-    std::vector<const zim::File*> zims;
-    for (auto current = this->readers.begin(); current != this->readers.end();
-         current++) {
-      zims.push_back((*current)->getZimFileHandler());
-    }
-    zim::Search* search = new zim::Search(zims);
-    search->set_query(unaccentedSearch);
-    search->set_range(resultStart, resultEnd);
-    search->set_suggestion_mode(true);
-    internal->_search = search;
-    internal->current_iterator = internal->_search->begin();
-    this->estimatedResultCount = internal->_search->get_matches_estimated();
+  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 */
@@ -311,155 +228,45 @@ unsigned int Searcher::getEstimatedResultCount()
   return this->estimatedResultCount;
 }
 
-bool Searcher::setProtocolPrefix(const std::string prefix)
-{
-  this->protocolPrefix = prefix;
-  return true;
-}
-
-bool Searcher::setSearchProtocolPrefix(const std::string prefix)
-{
-  this->searchProtocolPrefix = prefix;
-  return true;
-}
-
-_Result::_Result(zim::Search::iterator& iterator)
+_Result::_Result(zim::SearchResultSet::iterator iterator)
     : iterator(iterator)
 {
 }
 
 std::string _Result::get_url()
 {
-  return iterator.get_url();
+  return iterator.getPath();
 }
 std::string _Result::get_title()
 {
-  return iterator.get_title();
+  return iterator.getTitle();
 }
 int _Result::get_score()
 {
-  return iterator.get_score();
+  return iterator.getScore();
 }
 std::string _Result::get_snippet()
 {
-  return iterator.get_snippet();
+  return iterator.getSnippet();
 }
 std::string _Result::get_content()
 {
-  if (iterator->good()) {
-    return iterator->getData();
-  }
-  return "";
+  return iterator->getItem(true).getData();
 }
 int _Result::get_size()
 {
-  return iterator.get_size();
+  return iterator.getSize();
 }
 int _Result::get_wordCount()
 {
-  return iterator.get_wordCount();
+  return iterator.getWordCount();
 }
-int _Result::get_readerIndex()
+std::string _Result::get_zimId()
 {
-  return iterator.get_fileIndex();
-}
-#ifdef ENABLE_CTPP2
-
-string Searcher::getHtml()
-{
-  SimpleVM oSimpleVM(
-      1024, //iIMaxFunctions (default value)
-      4096, //iIMaxArgStackSize (default value)
-      4096, //iIMaxCodeStackSize (default value)
-      10240 * 2 //iIMaxSteps (default*2)
-  );
-
-  // Fill data
-  CDT oData;
-  CDT resultsCDT(CDT::ARRAY_VAL);
-
-  this->restart_search();
-  Result* p_result = NULL;
-  while ((p_result = this->getNextResult())) {
-    CDT result;
-    result["title"] = p_result->get_title();
-    result["url"] = p_result->get_url();
-    result["snippet"] = p_result->get_snippet();
-    result["contentId"] = humanReaderNames[p_result->get_readerIndex()];
-
-    if (p_result->get_size() >= 0) {
-      result["size"] = kiwix::beautifyInteger(p_result->get_size());
-    }
-
-    if (p_result->get_wordCount() >= 0) {
-      result["wordCount"] = kiwix::beautifyInteger(p_result->get_wordCount());
-    }
-
-    resultsCDT.PushBack(result);
-    delete p_result;
-  }
-  this->restart_search();
-  oData["results"] = resultsCDT;
-
-  // pages
-  CDT pagesCDT(CDT::ARRAY_VAL);
-
-  unsigned int pageStart
-      = this->resultStart / this->resultCountPerPage >= 5
-            ? this->resultStart / this->resultCountPerPage - 4
-            : 0;
-  unsigned int pageCount
-      = this->estimatedResultCount / this->resultCountPerPage + 1 - pageStart;
-
-  if (pageCount > 10) {
-    pageCount = 10;
-  } else if (pageCount == 1) {
-    pageCount = 0;
-  }
-
-  for (unsigned int i = pageStart; i < pageStart + pageCount; i++) {
-    CDT page;
-    page["label"] = i + 1;
-    page["start"] = i * this->resultCountPerPage;
-    page["end"] = (i + 1) * this->resultCountPerPage;
-
-    if (i * this->resultCountPerPage == this->resultStart) {
-      page["selected"] = true;
-    }
-
-    pagesCDT.PushBack(page);
-  }
-  oData["pages"] = pagesCDT;
-
-  oData["count"] = kiwix::beautifyInteger(this->estimatedResultCount);
-  oData["searchPattern"] = kiwix::encodeDiples(this->searchPattern);
-  oData["searchPatternEncoded"] = urlEncode(this->searchPattern);
-  oData["resultStart"] = this->resultStart + 1;
-  oData["resultEnd"] = (this->resultEnd > this->estimatedResultCount
-                            ? this->estimatedResultCount
-                            : this->resultEnd);
-  oData["resultRange"] = this->resultCountPerPage;
-  oData["resultLastPageStart"]
-      = this->estimatedResultCount > this->resultCountPerPage
-            ? std::round(this->estimatedResultCount / this->resultCountPerPage) * this->resultCountPerPage
-            : 0;
-  oData["protocolPrefix"] = this->protocolPrefix;
-  oData["searchProtocolPrefix"] = this->searchProtocolPrefix;
-  oData["contentId"] = this->contentHumanReadableId;
-
-  std::string template_ct2 = RESOURCE::results_ct2;
-  VMStringLoader oLoader(template_ct2.c_str(), template_ct2.size());
-
-  FileLogger oLogger(stderr);
-
-  // DEBUG only (write output to stdout)
-  // oSimpleVM.Run(oData, oLoader, stdout, oLogger);
-
-  std::string sResult;
-  oSimpleVM.Run(oData, oLoader, sResult, oLogger);
-
-  return sResult;
+  std::ostringstream s;
+  s << iterator.getZimId();
+  return s.str();
 }
 
-#endif
+
 }

src/server.cpp

@@ -0,0 +1,73 @@
+/*
+ * Copyright 2019 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 "server.h"
+
+#include "library.h"
+#include "name_mapper.h"
+
+#include <string>
+
+#include "server/internalServer.h"
+
+namespace kiwix {
+
+Server::Server(Library* library, NameMapper* nameMapper) :
+  mp_library(library),
+  mp_nameMapper(nameMapper),
+  mp_server(nullptr)
+{
+}
+
+Server::~Server() = default;
+
+bool Server::start() {
+  mp_server.reset(new InternalServer(
+    mp_library,
+    mp_nameMapper,
+    m_addr,
+    m_port,
+    m_root,
+    m_nbThreads,
+    m_verbose,
+    m_withTaskbar,
+    m_withLibraryButton,
+    m_blockExternalLinks));
+  return mp_server->start();
+}
+
+void Server::stop() {
+  if (mp_server) {
+    mp_server->stop();
+    mp_server.reset(nullptr);
+  }
+}
+
+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);
+  }
+}
+
+}

src/server/byte_range.cpp

@@ -0,0 +1,127 @@
+/*
+ * Copyright 2020 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 "byte_range.h"
+
+#include "tools/stringTools.h"
+
+#include <cassert>
+#include <algorithm>
+
+namespace kiwix {
+
+namespace {
+
+ByteRange parseByteRange(const std::string& rangeStr)
+{
+  std::istringstream iss(rangeStr);
+
+  int64_t start, end = INT64_MAX;
+  if (iss >> start) {
+    if ( start < 0 ) {
+      if ( iss.eof() )
+        return ByteRange(-start);
+    } else {
+      char c;
+      if (iss >> c && c=='-') {
+        iss >> end; // if this fails, end is not modified, which is OK
+        if (iss.eof() && start <= end)
+          return ByteRange(ByteRange::PARSED, start, end);
+      }
+    }
+  }
+
+  return ByteRange(ByteRange::INVALID, 0, INT64_MAX);
+}
+
+} // unnamed namespace
+
+ByteRange::ByteRange()
+  : kind_(NONE)
+  , first_(0)
+  , last_(INT64_MAX)
+{}
+
+ByteRange::ByteRange(Kind kind, int64_t first, int64_t last)
+  : kind_(kind)
+  , first_(first)
+  , last_(last)
+{
+  assert(kind != NONE);
+  assert(first >= 0);
+  assert(last >= first || (first == 0 && last == -1));
+}
+
+ByteRange::ByteRange(int64_t suffix_length)
+  : kind_(PARSED)
+  , first_(-suffix_length)
+  , last_(INT64_MAX)
+{
+  assert(suffix_length > 0);
+}
+
+int64_t ByteRange::first() const
+{
+  assert(kind_ > PARSED);
+  return first_;
+}
+
+int64_t ByteRange::last() const
+{
+  assert(kind_ > PARSED);
+  return last_;
+}
+
+int64_t ByteRange::length() const
+{
+  assert(kind_ > PARSED);
+  return last_ + 1 - first_;
+}
+
+ByteRange ByteRange::parse(const std::string& rangeStr)
+{
+  const std::string byteUnitSpec("bytes=");
+  if ( ! kiwix::startsWith(rangeStr, byteUnitSpec) )
+    return ByteRange(INVALID, 0, INT64_MAX);
+
+  return parseByteRange(rangeStr.substr(byteUnitSpec.size()));
+}
+
+ByteRange ByteRange::resolve(int64_t contentSize) const
+{
+  if ( kind() == NONE )
+    return ByteRange(RESOLVED_FULL_CONTENT, 0, contentSize-1);
+
+  if ( kind() == INVALID )
+    return ByteRange(RESOLVED_UNSATISFIABLE, 0, contentSize-1);
+
+  const int64_t resolved_first = first_ < 0
+                               ? std::max(int64_t(0), contentSize + first_)
+                               : first_;
+
+  const int64_t resolved_last = std::min(contentSize-1, last_);
+
+  if ( resolved_first > resolved_last )
+    return ByteRange(RESOLVED_UNSATISFIABLE, 0, contentSize-1);
+
+  return ByteRange(RESOLVED_PARTIAL_CONTENT, resolved_first, resolved_last);
+}
+
+} // namespace kiwix

src/server/byte_range.h

@@ -0,0 +1,86 @@
+/*
+ * Copyright 2020 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 KIWIXLIB_SERVER_BYTE_RANGE_H
+#define KIWIXLIB_SERVER_BYTE_RANGE_H
+
+#include <cstdint>
+#include <string>
+
+namespace kiwix {
+
+class ByteRange
+{
+  public: // types
+    // ByteRange is parsed in a request, then it must be resolved (taking
+    // into account the actual size of the requested resource) before
+    // being applied in the response.
+    // The Kind enum represents possible states in such a lifecycle.
+    enum Kind {
+      // The request is not a range request (no Range header)
+      NONE,
+
+      // The value of the Range header is not a valid continuous
+      // range. Note that a valid (according to RFC7233) sequence of multiple
+      // byte ranges is considered invalid in the current implementation
+      // (i.e. only single-range partial requests are supported).
+      INVALID,
+
+      // This byte-range has been successfully parsed from the request
+      PARSED,
+
+      // This is a response to a regular (non-range) request
+      RESOLVED_FULL_CONTENT,
+
+      // The range request is invalid or unsatisfiable
+      RESOLVED_UNSATISFIABLE,
+
+      // This is a response to a (satisfiable) range request
+      RESOLVED_PARTIAL_CONTENT,
+    };
+
+  public: // functions
+    // Constructs a ByteRange object of NONE kind
+    ByteRange();
+
+    // Constructs a ByteRange object of the given kind (except NONE)
+    ByteRange(Kind kind, int64_t first, int64_t last);
+
+    // Constructs a ByteRange object of PARSED kind corresponding to a
+    // range request of the form "Range: bytes=-suffix_length"
+    explicit ByteRange(int64_t suffix_length);
+
+    Kind kind() const { return kind_; }
+    int64_t first() const;
+    int64_t last() const;
+    int64_t length() const;
+
+    static ByteRange parse(const std::string& rangeStr);
+    ByteRange resolve(int64_t contentSize) const;
+
+  private: // data
+    Kind kind_;
+    int64_t first_;
+    int64_t last_;
+};
+
+} // namespace kiwix
+
+#endif //KIWIXLIB_SERVER_BYTE_RANGE_H

src/server/etag.cpp

@@ -0,0 +1,135 @@
+/*
+ * Copyright 2020 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 "etag.h"
+
+#include "tools/stringTools.h"
+
+#include <algorithm>
+#include <sstream>
+
+namespace kiwix {
+
+namespace {
+
+// Characters in the options part of the ETag could in principle be picked up
+// from the latin alphabet in natural order (the character corresponding to
+// ETag::Option opt would be 'a'+opt; that would somewhat simplify the code in
+// this file). However it is better to have some mnemonics in the option names,
+// hence below variable: all_options[opt] corresponds to the character going
+// 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";
+
+static_assert(ETag::OPTION_COUNT == sizeof(all_options) - 1, "");
+
+bool isValidServerId(const std::string& s)
+{
+  return !s.empty() && s.find_first_of("\"/") == std::string::npos;
+}
+
+bool isSubsequenceOf(const std::string& s, const std::string& sortedString)
+{
+  std::string::size_type i = 0;
+  for ( const char c : s )
+  {
+    const std::string::size_type j = sortedString.find(c, i);
+    if ( j == std::string::npos )
+      return false;
+    i = j+1;
+  }
+  return true;
+}
+
+bool isValidOptionsString(const std::string& s)
+{
+  return isSubsequenceOf(s, all_options);
+}
+
+} // namespace
+
+
+void ETag::set_option(Option opt)
+{
+  if ( ! get_option(opt) )
+  {
+    m_options.push_back(all_options[opt]);
+    std::sort(m_options.begin(), m_options.end());
+  }
+}
+
+bool ETag::get_option(Option opt) const
+{
+  return m_options.find(all_options[opt]) != std::string::npos;
+}
+
+std::string ETag::get_etag() const
+{
+  if ( m_serverId.empty() )
+    return std::string();
+
+  return "\"" + m_serverId + "/" + m_options + "\"";
+}
+
+ETag::ETag(const std::string& serverId, const std::string& options)
+{
+  if ( isValidServerId(serverId) && isValidOptionsString(options) )
+  {
+    m_serverId = serverId;
+    m_options = options;
+  }
+}
+
+ETag ETag::parse(std::string s)
+{
+  if ( kiwix::startsWith("W/", s) )
+    s = s.substr(2);
+
+  if ( s.front() != '"' || s.back() != '"' )
+    return ETag();
+
+  s = s.substr(1, s.size()-2);
+
+  const std::string::size_type i = s.find('/');
+  if ( i == std::string::npos )
+    return ETag();
+
+  return ETag(s.substr(0, i), s.substr(i+1));
+}
+
+ETag ETag::match(const std::string& etags, const std::string& server_id)
+{
+  std::istringstream ss(etags);
+  std::string etag_str;
+  while ( ss >> etag_str )
+  {
+    if ( etag_str.back() == ',' )
+      etag_str.pop_back();
+
+    const ETag etag = parse(etag_str);
+    if ( etag && etag.m_serverId == server_id )
+      return etag;
+  }
+
+  return ETag();
+}
+
+} // namespace kiwix

src/server/etag.h

@@ -0,0 +1,85 @@
+/*
+ * Copyright 2020 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 KIWIXLIB_SERVER_ETAG_H
+#define KIWIXLIB_SERVER_ETAG_H
+
+#include <string>
+
+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
+//
+// 2. Options -  Zero or more characters encoding 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
+// may contain the slash symbol.
+// Examples of valid Kiwix server ETags (including the double quotes):
+//
+//   "abcdefghijklmn/"
+//   "1234567890/z"
+//   "1234567890/cz"
+//
+// 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
+// (Not Modified) response without following the full code path that would
+// discover the necessary options.
+
+class ETag
+{
+  public: // types
+    enum Option {
+      CACHEABLE_ENTITY,
+      COMPRESSED_CONTENT,
+      OPTION_COUNT
+    };
+
+  public: // functions
+    ETag() {}
+
+    void set_server_id(const std::string& id) { m_serverId = id; }
+    void set_option(Option opt);
+
+    explicit operator bool() const { return !m_serverId.empty(); }
+
+    bool get_option(Option opt) const;
+    std::string get_etag() const;
+
+
+    static ETag match(const std::string& etags, const std::string& server_id);
+
+  private: // functions
+    ETag(const std::string& serverId, const std::string& options);
+
+    static ETag parse(std::string s);
+
+  private: // data
+    std::string m_serverId;
+    std::string m_options;
+};
+
+} // namespace kiwix
+
+#endif // KIWIXLIB_SERVER_ETAG_H

src/server/internalServer.cpp

@@ -0,0 +1,810 @@
+/*
+ * Copyright 2019 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 "internalServer.h"
+
+#ifdef __FreeBSD__
+#include <netinet/in.h>
+#endif
+
+#ifdef _WIN32
+# if !defined(__MINGW32__) && (_MSC_VER < 1600)
+#   include "stdint4win.h"
+# endif
+# include <winsock2.h>
+# include <ws2tcpip.h>
+# ifdef __GNUC__
+  // inet_pton is not declared in mingw, even if the function exists.
+  extern "C" {
+    WINSOCK_API_LINKAGE  INT WSAAPI inet_pton( INT Family, PCSTR pszAddrString, PVOID pAddrBuf);
+  }
+# endif
+  typedef UINT64 uint64_t;
+  typedef UINT16 uint16_t;
+#endif
+
+extern "C" {
+#include "microhttpd_wrapper.h"
+}
+
+#include "tools/otherTools.h"
+#include "tools/pathTools.h"
+#include "tools/regexTools.h"
+#include "tools/stringTools.h"
+#include "library.h"
+#include "name_mapper.h"
+#include "entry.h"
+#include "searcher.h"
+#include "search_renderer.h"
+#include "opds_dumper.h"
+
+#include <zim/uuid.h>
+
+#include <mustache.hpp>
+
+#include <atomic>
+#include <string>
+#include <vector>
+#include <chrono>
+#include "kiwixlib-resources.h"
+
+#ifndef _WIN32
+# include <arpa/inet.h>
+#endif
+
+#include "request_context.h"
+#include "response.h"
+
+#define MAX_SEARCH_LEN 140
+#define KIWIX_MIN_CONTENT_SIZE_TO_DEFLATE 100
+
+namespace kiwix {
+
+namespace
+{
+
+inline std::string normalizeRootUrl(std::string rootUrl)
+{
+  while ( !rootUrl.empty() && rootUrl.back() == '/' )
+    rootUrl.pop_back();
+
+  while ( !rootUrl.empty() && rootUrl.front() == '/' )
+    rootUrl = rootUrl.substr(1);
+  return rootUrl.empty() ? rootUrl : "/" + rootUrl;
+}
+
+} // unnamed namespace
+
+static IdNameMapper defaultNameMapper;
+
+static MHD_Result staticHandlerCallback(void* cls,
+                                        struct MHD_Connection* connection,
+                                        const char* url,
+                                        const char* method,
+                                        const char* version,
+                                        const char* upload_data,
+                                        size_t* upload_data_size,
+                                        void** cont_cls);
+
+
+InternalServer::InternalServer(Library* library,
+                               NameMapper* nameMapper,
+                               std::string addr,
+                               int port,
+                               std::string root,
+                               int nbThreads,
+                               bool verbose,
+                               bool withTaskbar,
+                               bool withLibraryButton,
+                               bool blockExternalLinks) :
+  m_addr(addr),
+  m_port(port),
+  m_root(normalizeRootUrl(root)),
+  m_nbThreads(nbThreads),
+  m_verbose(verbose),
+  m_withTaskbar(withTaskbar),
+  m_withLibraryButton(withLibraryButton),
+  m_blockExternalLinks(blockExternalLinks),
+  mp_daemon(nullptr),
+  mp_library(library),
+  mp_nameMapper(nameMapper ? nameMapper : &defaultNameMapper)
+{}
+
+bool InternalServer::start() {
+#ifdef _WIN32
+  int flags = MHD_USE_SELECT_INTERNALLY;
+#else
+  int flags = MHD_USE_POLL_INTERNALLY;
+#endif
+  if (m_verbose.load())
+    flags |= MHD_USE_DEBUG;
+
+
+  struct sockaddr_in sockAddr;
+  memset(&sockAddr, 0, sizeof(sockAddr));
+  sockAddr.sin_family = AF_INET;
+  sockAddr.sin_port = htons(m_port);
+  if (m_addr.empty()) {
+    if (0 != INADDR_ANY)
+      sockAddr.sin_addr.s_addr = htonl(INADDR_ANY);
+  } else {
+    if (inet_pton(AF_INET, m_addr.c_str(), &(sockAddr.sin_addr.s_addr)) == 0) {
+      std::cerr << "Ip address " << m_addr << "  is not a valid ip address" << std::endl;
+      return false;
+    }
+  }
+
+  mp_daemon = MHD_start_daemon(flags,
+                            m_port,
+                            NULL,
+                            NULL,
+                            &staticHandlerCallback,
+                            this,
+                            MHD_OPTION_SOCK_ADDR, &sockAddr,
+                            MHD_OPTION_THREAD_POOL_SIZE, m_nbThreads,
+                            MHD_OPTION_END);
+  if (mp_daemon == nullptr) {
+    std::cerr << "Unable to instantiate the HTTP daemon. The port " << m_port
+              << " is maybe already occupied or need more permissions to be open. "
+                 "Please try as root or with a port number higher or equal to 1024."
+              << std::endl;
+    return false;
+  }
+  auto server_start_time = std::chrono::system_clock::now().time_since_epoch();
+  m_server_id = kiwix::to_string(server_start_time.count());
+  m_library_id = m_server_id;
+  return true;
+}
+
+void InternalServer::stop()
+{
+  MHD_stop_daemon(mp_daemon);
+}
+
+static MHD_Result staticHandlerCallback(void* cls,
+                                        struct MHD_Connection* connection,
+                                        const char* url,
+                                        const char* method,
+                                        const char* version,
+                                        const char* upload_data,
+                                        size_t* upload_data_size,
+                                        void** cont_cls)
+{
+  InternalServer* _this = static_cast<InternalServer*>(cls);
+
+  return _this->handlerCallback(connection,
+                                url,
+                                method,
+                                version,
+                                upload_data,
+                                upload_data_size,
+                                cont_cls);
+}
+
+MHD_Result InternalServer::handlerCallback(struct MHD_Connection* connection,
+                                           const char* url,
+                                           const char* method,
+                                           const char* version,
+                                           const char* upload_data,
+                                           size_t* upload_data_size,
+                                           void** cont_cls)
+{
+  auto start_time = std::chrono::steady_clock::now();
+  if (m_verbose.load() ) {
+    printf("======================\n");
+    printf("Requesting : \n");
+    printf("full_url  : %s\n", url);
+  }
+  RequestContext request(connection, m_root, url, method, version);
+
+  if (m_verbose.load() ) {
+    request.print_debug_info();
+  }
+  /* Unexpected method */
+  if (request.get_method() != RequestMethod::GET
+   && request.get_method() != RequestMethod::POST
+   && request.get_method() != RequestMethod::HEAD) {
+    printf("Reject request because of unhandled request method.\n");
+    printf("----------------------\n");
+    return MHD_NO;
+  }
+
+  auto response = handle_request(request);
+
+  if (response->getReturnCode() == MHD_HTTP_INTERNAL_SERVER_ERROR) {
+    printf("========== INTERNAL ERROR !! ============\n");
+    if (!m_verbose.load()) {
+      printf("Requesting : \n");
+      printf("full_url : %s\n", url);
+      request.print_debug_info();
+    }
+  }
+
+  if (response->getReturnCode() == MHD_HTTP_OK && !etag_not_needed(request))
+    response->set_server_id(m_server_id);
+
+  auto ret = response->send(request, connection);
+  auto end_time = std::chrono::steady_clock::now();
+  auto time_span = std::chrono::duration_cast<std::chrono::duration<double>>(end_time - start_time);
+  if (m_verbose.load()) {
+    printf("Request time : %fs\n", time_span.count());
+    printf("----------------------\n");
+  }
+  return ret;
+}
+
+std::unique_ptr<Response> InternalServer::handle_request(const RequestContext& request)
+{
+  try {
+    if (! request.is_valid_url())
+      return Response::build_404(*this, request, "", "");
+
+    const ETag etag = get_matching_if_none_match_etag(request);
+    if ( etag )
+      return Response::build_304(*this, etag);
+
+    if (startsWith(request.get_url(), "/skin/"))
+      return handle_skin(request);
+
+    if (startsWith(request.get_url(), "/catalog/"))
+      return handle_catalog(request);
+
+    if (request.get_url() == "/meta")
+      return handle_meta(request);
+
+    if (request.get_url() == "/search")
+      return handle_search(request);
+
+    if (request.get_url() == "/suggest")
+     return handle_suggest(request);
+
+    if (request.get_url() == "/random")
+      return handle_random(request);
+
+    if (request.get_url() == "/catch/external")
+      return handle_captured_external(request);
+
+    return handle_content(request);
+  } catch (std::exception& e) {
+    fprintf(stderr, "===== Unhandled error : %s\n", e.what());
+    return Response::build_500(*this, e.what());
+  } catch (...) {
+    fprintf(stderr, "===== Unhandled unknown error\n");
+    return Response::build_500(*this, "Unknown error");
+  }
+}
+
+MustacheData InternalServer::get_default_data() const
+{
+  MustacheData data;
+  data.set("root", m_root);
+  return data;
+}
+
+bool InternalServer::etag_not_needed(const RequestContext& request) const
+{
+  const std::string url = request.get_url();
+  return kiwix::startsWith(url, "/catalog")
+      || url == "/search"
+      || url == "/suggest"
+      || url == "/random"
+      || url == "/catch/external";
+}
+
+ETag
+InternalServer::get_matching_if_none_match_etag(const RequestContext& r) const
+{
+  try {
+    const std::string etag_list = r.get_header(MHD_HTTP_HEADER_IF_NONE_MATCH);
+    return ETag::match(etag_list, m_server_id);
+  } catch (const std::out_of_range&) {
+    return ETag();
+  }
+}
+
+std::unique_ptr<Response> InternalServer::build_homepage(const RequestContext& request)
+{
+  return ContentResponse::build(*this, RESOURCE::templates::index_html, get_default_data(), "text/html; charset=utf-8", true);
+}
+
+std::unique_ptr<Response> InternalServer::handle_meta(const RequestContext& request)
+{
+  std::string bookName;
+  std::string bookId;
+  std::string meta_name;
+  std::shared_ptr<Reader> reader;
+  try {
+    bookName = request.get_argument("content");
+    bookId = mp_nameMapper->getIdForName(bookName);
+    meta_name = request.get_argument("name");
+    reader = mp_library->getReaderById(bookId);
+  } catch (const std::out_of_range& e) {
+    return Response::build_404(*this, request, bookName, "");
+  }
+
+  if (reader == nullptr) {
+    return Response::build_404(*this, request, bookName, "");
+  }
+
+  std::string content;
+  std::string mimeType = "text";
+
+  if (meta_name == "title") {
+    content = reader->getTitle();
+  } else if (meta_name == "description") {
+    content = reader->getDescription();
+  } else if (meta_name == "language") {
+    content = reader->getLanguage();
+  } else if (meta_name == "name") {
+    content = reader->getName();
+  } else if (meta_name == "tags") {
+    content = reader->getTags();
+  } else if (meta_name == "date") {
+    content = reader->getDate();
+  } else if (meta_name == "creator") {
+    content = reader->getCreator();
+  } else if (meta_name == "publisher") {
+    content = reader->getPublisher();
+  } else if (meta_name == "favicon") {
+    reader->getFavicon(content, mimeType);
+  } else {
+    return Response::build_404(*this, request, bookName, "");
+  }
+
+  auto response = ContentResponse::build(*this, content, mimeType);
+  response->set_cacheable();
+  return std::move(response);
+}
+
+std::unique_ptr<Response> InternalServer::handle_suggest(const RequestContext& request)
+{
+  if (m_verbose.load()) {
+    printf("** running handle_suggest\n");
+  }
+
+  std::string content;
+  std::string mimeType;
+  unsigned int maxSuggestionCount = 10;
+  unsigned int suggestionCount = 0;
+
+  std::string bookName;
+  std::string bookId;
+  std::string term;
+  std::shared_ptr<Reader> reader;
+  try {
+    bookName = request.get_argument("content");
+    bookId = mp_nameMapper->getIdForName(bookName);
+    term = request.get_argument("term");
+    reader = mp_library->getReaderById(bookId);
+  } catch (const std::out_of_range&) {
+    return Response::build_404(*this, request, bookName, "");
+  }
+
+  if (m_verbose.load()) {
+    printf("Searching suggestions for: \"%s\"\n", term.c_str());
+  }
+
+  MustacheData results{MustacheData::type::list};
+
+  bool first = true;
+  if (reader != nullptr) {
+    /* Get the suggestions */
+    SuggestionsList_t suggestions;
+    reader->searchSuggestionsSmart(term, maxSuggestionCount, suggestions);
+    for(auto& suggestion:suggestions) {
+      MustacheData result;
+      result.set("label", suggestion.getTitle());
+
+      if (suggestion.hasSnippet()) {
+        result.set("label", suggestion.getSnippet());
+      }
+
+      result.set("value", suggestion.getTitle());
+      result.set("kind", "path");
+      result.set("path", suggestion.getPath());
+      result.set("first", first);
+      first = false;
+      results.push_back(result);
+      suggestionCount++;
+    }
+  }
+
+  /* Propose the fulltext search if possible */
+  if (reader->hasFulltextIndex()) {
+    MustacheData result;
+    result.set("label", "containing '" + term + "'...");
+    result.set("value", term + " ");
+    result.set("kind", "pattern");
+    result.set("first", first);
+    results.push_back(result);
+  }
+
+  auto data = get_default_data();
+  data.set("suggestions", results);
+
+  auto response = ContentResponse::build(*this, RESOURCE::templates::suggestion_json, data, "application/json; charset=utf-8");
+  return std::move(response);
+}
+
+std::unique_ptr<Response> InternalServer::handle_skin(const RequestContext& request)
+{
+  if (m_verbose.load()) {
+    printf("** running handle_skin\n");
+  }
+
+  auto resourceName = request.get_url().substr(1);
+  try {
+    auto response = ContentResponse::build(
+        *this,
+        getResource(resourceName),
+        getMimeTypeForFile(resourceName));
+    response->set_cacheable();
+    return std::move(response);
+  } catch (const ResourceNotFound& e) {
+    return Response::build_404(*this, request, "", "");
+  }
+}
+
+std::unique_ptr<Response> InternalServer::handle_search(const RequestContext& request)
+{
+  if (m_verbose.load()) {
+    printf("** running handle_search\n");
+  }
+
+  std::string bookName;
+  std::string bookId;
+  try {
+    bookName = request.get_argument("content");
+    bookId = mp_nameMapper->getIdForName(bookName);
+  } catch (const std::out_of_range&) {}
+
+  std::string patternString;
+  try {
+    patternString = request.get_argument("pattern");
+  } catch (const std::out_of_range&) {}
+
+  /* Retrive geo search */
+  bool has_geo_query = false;
+  float latitude = 0;
+  float longitude = 0;
+  float distance = 0;
+  try {
+    latitude = request.get_argument<float>("latitude");
+    longitude = request.get_argument<float>("longitude");
+    distance = request.get_argument<float>("distance");
+    has_geo_query = true;
+  } catch(const std::out_of_range&) {}
+    catch(const std::invalid_argument&) {}
+
+  std::shared_ptr<Reader> reader(nullptr);
+  try {
+    reader = mp_library->getReaderById(bookId);
+  } catch (const std::out_of_range&) {}
+
+  /* Make the search */
+  if ( (!reader && !bookName.empty())
+    || (patternString.empty() && ! has_geo_query) ) {
+    auto data = get_default_data();
+    data.set("pattern", encodeDiples(patternString));
+    auto response = ContentResponse::build(*this, RESOURCE::templates::no_search_result_html, data, "text/html; charset=utf-8");
+    response->set_taskbar(bookName, reader ? reader->getTitle() : "");
+    response->set_code(MHD_HTTP_NOT_FOUND);
+    return std::move(response);
+  }
+
+  Searcher searcher;
+  if (reader) {
+    searcher.add_reader(reader.get());
+  } else {
+    for (auto& bookId: mp_library->filter(kiwix::Filter().local(true).valid(true))) {
+      auto currentReader = mp_library->getReaderById(bookId);
+      if (currentReader) {
+        searcher.add_reader(currentReader.get());
+      }
+    }
+  }
+
+  auto start = 0;
+  try {
+    start = request.get_argument<unsigned int>("start");
+  } catch (const std::exception&) {}
+
+  auto pageLength = 25;
+  try {
+    pageLength = request.get_argument<unsigned int>("pageLength");
+  } catch (const std::exception&) {}
+  if (pageLength > MAX_SEARCH_LEN) {
+    pageLength = MAX_SEARCH_LEN;
+  }
+  if (pageLength == 0) {
+    pageLength = 25;
+  }
+
+  auto end = start + pageLength;
+
+  /* Get the results */
+  try {
+    if (patternString.empty()) {
+      searcher.geo_search(latitude, longitude, distance,
+                           start, end, m_verbose.load());
+    } else {
+      searcher.search(patternString,
+                       start, end, m_verbose.load());
+    }
+    SearchRenderer renderer(&searcher, mp_nameMapper);
+    renderer.setSearchPattern(patternString);
+    renderer.setSearchContent(bookName);
+    renderer.setProtocolPrefix(m_root + "/");
+    renderer.setSearchProtocolPrefix(m_root + "/search?");
+    renderer.setPageLength(pageLength);
+    auto response = ContentResponse::build(*this, renderer.getHtml(), "text/html; charset=utf-8");
+    response->set_taskbar(bookName, reader ? reader->getTitle() : "");
+
+    return std::move(response);
+  } catch (const std::exception& e) {
+    std::cerr << e.what() << std::endl;
+    return Response::build_500(*this, e.what());
+  }
+}
+
+std::unique_ptr<Response> InternalServer::handle_random(const RequestContext& request)
+{
+  if (m_verbose.load()) {
+    printf("** running handle_random\n");
+  }
+
+  std::string bookName;
+  std::string bookId;
+  std::shared_ptr<Reader> reader;
+  try {
+    bookName = request.get_argument("content");
+    bookId = mp_nameMapper->getIdForName(bookName);
+    reader = mp_library->getReaderById(bookId);
+  } catch (const std::out_of_range&) {
+    return Response::build_404(*this, request, bookName, "");
+  }
+
+  if (reader == nullptr) {
+    return Response::build_404(*this, request, bookName, "");
+  }
+
+  try {
+    auto entry = reader->getRandomPage();
+    return build_redirect(bookName, entry.getFinalEntry());
+  } catch(kiwix::NoEntry& e) {
+    return Response::build_404(*this, request, bookName, "");
+  }
+}
+
+std::unique_ptr<Response> InternalServer::handle_captured_external(const RequestContext& request)
+{
+  std::string source = "";
+  try {
+    source = kiwix::urlDecode(request.get_argument("source"));
+  } catch (const std::out_of_range& e) {}
+
+  if (source.empty())
+    return Response::build_404(*this, request, "", "");
+
+  auto data = get_default_data();
+  data.set("source", source);
+  return ContentResponse::build(*this, RESOURCE::templates::captured_external_html, data, "text/html; charset=utf-8");
+}
+
+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 Response::build_404(*this, request, "", "");
+  }
+
+  if (url == "v2") {
+    return handle_catalog_v2(request);
+  }
+
+  if (url != "searchdescription.xml" && url != "root.xml" && url != "search") {
+    return Response::build_404(*this, request, "", "");
+  }
+
+  if (url == "searchdescription.xml") {
+    auto response = ContentResponse::build(*this, RESOURCE::opensearchdescription_xml, get_default_data(), "application/opensearchdescription+xml");
+    return std::move(response);
+  }
+
+  zim::Uuid uuid;
+  kiwix::OPDSDumper opdsDumper(mp_library);
+  opdsDumper.setRootLocation(m_root);
+  opdsDumper.setLibraryId(m_library_id);
+  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(
+      *this,
+      opdsDumper.dumpOPDSFeed(bookIdsToDump, request.get_query()),
+      "application/atom+xml; profile=opds-catalog; kind=acquisition; charset=utf-8");
+  return std::move(response);
+}
+
+namespace
+{
+
+Filter get_search_filter(const RequestContext& request)
+{
+    auto filter = kiwix::Filter().valid(true).local(true);
+    try {
+      filter.query(request.get_argument("q"));
+    } catch (const std::out_of_range&) {}
+    try {
+      filter.maxSize(request.get_argument<unsigned long>("maxsize"));
+    } catch (...) {}
+    try {
+      filter.name(request.get_argument("name"));
+    } catch (const std::out_of_range&) {}
+    try {
+      filter.category(request.get_argument("category"));
+    } catch (const std::out_of_range&) {}
+    try {
+      filter.lang(request.get_argument("lang"));
+    } catch (const std::out_of_range&) {}
+    try {
+      filter.acceptTags(kiwix::split(request.get_argument("tag"), ";"));
+    } catch (...) {}
+    try {
+      filter.rejectTags(kiwix::split(request.get_argument("notag"), ";"));
+    } catch (...) {}
+    return filter;
+}
+
+template<class T>
+std::vector<T> subrange(const std::vector<T>& v, size_t s, size_t n)
+{
+  const size_t e = std::min(v.size(), s+n);
+  return std::vector<T>(v.begin()+std::min(v.size(), s), v.begin()+e);
+}
+
+} // unnamed namespace
+
+std::vector<std::string>
+InternalServer::search_catalog(const RequestContext& request,
+                               kiwix::OPDSDumper& opdsDumper)
+{
+    const auto filter = get_search_filter(request);
+    const std::string q = filter.hasQuery()
+                        ? filter.getQuery()
+                        : "<Empty query>";
+    std::vector<std::string> bookIdsToDump = mp_library->filter(filter);
+    const auto totalResults = bookIdsToDump.size();
+    const size_t count = request.get_optional_param("count", 10UL);
+    const size_t startIndex = request.get_optional_param("start", 0UL);
+    bookIdsToDump = subrange(bookIdsToDump, startIndex, count);
+    opdsDumper.setOpenSearchInfo(totalResults, startIndex, bookIdsToDump.size());
+    return bookIdsToDump;
+}
+
+namespace
+{
+
+std::string get_book_name(const RequestContext& request)
+{
+  try {
+    return request.get_url_part(0);
+  } catch (const std::out_of_range& e) {
+    return std::string();
+  }
+}
+
+std::string searchSuggestionHTML(const std::string& searchURL, const std::string& pattern)
+{
+  kainjow::mustache::mustache tmpl("Make a full text search for <a href=\"{{{searchURL}}}\">{{pattern}}</a>");
+  MustacheData data;
+  data.set("pattern", pattern);
+  data.set("searchURL", searchURL);
+  return (tmpl.render(data));
+}
+
+} // unnamed namespace
+
+std::shared_ptr<Reader>
+InternalServer::get_reader(const std::string& bookName) const
+{
+  std::shared_ptr<Reader> reader;
+  try {
+    const std::string bookId = mp_nameMapper->getIdForName(bookName);
+    reader = mp_library->getReaderById(bookId);
+  } catch (const std::out_of_range& e) {
+  }
+  return reader;
+}
+
+std::unique_ptr<Response>
+InternalServer::build_redirect(const std::string& bookName, const kiwix::Entry& entry) const
+{
+  auto redirectUrl = m_root + "/" + bookName + "/" + kiwix::urlEncode(entry.getPath());
+  return Response::build_redirect(*this, redirectUrl);
+}
+
+std::unique_ptr<Response> InternalServer::handle_content(const RequestContext& request)
+{
+  const std::string url = request.get_url();
+  const std::string pattern = url.substr((url.find_last_of('/'))+1);
+  if (m_verbose.load()) {
+    printf("** running handle_content\n");
+  }
+
+  const std::string bookName = get_book_name(request);
+  if (bookName.empty())
+    return build_homepage(request);
+
+  const std::shared_ptr<Reader> reader = get_reader(bookName);
+  if (reader == nullptr) {
+    std::string searchURL = m_root+"/search?pattern="+pattern; // Make a full search on the entire library.
+    const std::string details = searchSuggestionHTML(searchURL, kiwix::urlDecode(pattern));
+
+    return Response::build_404(*this, request, bookName, "", details);
+  }
+
+  auto urlStr = request.get_url().substr(bookName.size()+1);
+  if (urlStr[0] == '/') {
+    urlStr = urlStr.substr(1);
+  }
+
+  try {
+    auto entry = reader->getEntryFromPath(urlStr);
+    if (entry.isRedirect() || urlStr.empty()) {
+      // If urlStr is empty, we want to mainPage.
+      // We must do a redirection to the real page.
+      return build_redirect(bookName, entry.getFinalEntry());
+    }
+    auto response = ItemResponse::build(*this, request, entry.getZimEntry().getItem());
+    try {
+      dynamic_cast<ContentResponse&>(*response).set_taskbar(bookName, reader->getTitle());
+    } catch (std::bad_cast& e) {}
+
+    if (m_verbose.load()) {
+      printf("Found %s\n", entry.getPath().c_str());
+      printf("mimeType: %s\n", entry.getMimetype().c_str());
+    }
+
+    return response;
+  } catch(kiwix::NoEntry& e) {
+    if (m_verbose.load())
+      printf("Failed to find %s\n", urlStr.c_str());
+
+    std::string searchURL = m_root+"/search?content="+bookName+"&pattern="+pattern; // Make a search on this specific book only.
+    const std::string details = searchSuggestionHTML(searchURL, kiwix::urlDecode(pattern));
+
+    return Response::build_404(*this, request, bookName, reader->getTitle(), details);
+  }
+}
+
+}

src/server/internalServer.h

@@ -0,0 +1,122 @@
+/*
+ * Copyright 2019 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 KIWIXLIB_SERVER_INTERNALSERVER_H
+#define KIWIXLIB_SERVER_INTERNALSERVER_H
+
+
+extern "C" {
+#include "microhttpd_wrapper.h"
+}
+
+#include "library.h"
+#include "name_mapper.h"
+
+#include <mustache.hpp>
+
+#include <atomic>
+#include <string>
+
+#include "server/request_context.h"
+#include "server/response.h"
+
+namespace kiwix {
+
+typedef kainjow::mustache::data MustacheData;
+
+class Entry;
+class OPDSDumper;
+
+class InternalServer {
+  public:
+    InternalServer(Library* library,
+                   NameMapper* nameMapper,
+                   std::string addr,
+                   int port,
+                   std::string root,
+                   int nbThreads,
+                   bool verbose,
+                   bool withTaskbar,
+                   bool withLibraryButton,
+                   bool blockExternalLinks);
+    virtual ~InternalServer() = default;
+
+    MHD_Result handlerCallback(struct MHD_Connection* connection,
+                               const char* url,
+                               const char* method,
+                               const char* version,
+                               const char* upload_data,
+                               size_t* upload_data_size,
+                               void** cont_cls);
+    bool start();
+    void stop();
+
+  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_homepage(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_categories(const RequestContext& request);
+    std::unique_ptr<Response> handle_meta(const RequestContext& request);
+    std::unique_ptr<Response> handle_search(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_captured_external(const RequestContext& request);
+    std::unique_ptr<Response> handle_content(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;
+
+  private: // data
+    std::string m_addr;
+    int m_port;
+    std::string m_root;
+    int m_nbThreads;
+    std::atomic_bool m_verbose;
+    bool m_withTaskbar;
+    bool m_withLibraryButton;
+    bool m_blockExternalLinks;
+    struct MHD_Daemon* mp_daemon;
+
+    Library* mp_library;
+    NameMapper* mp_nameMapper;
+
+    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);
+
+};
+
+}
+
+#endif //KIWIXLIB_SERVER_INTERNALSERVER_H

src/server/internalServer_catalog_v2.cpp

@@ -0,0 +1,109 @@
+/*
+ * 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

@@ -0,0 +1,196 @@
+/*
+ * Copyright 2009-2016 Emmanuel Engelhart <kelson@kiwix.org>
+ * 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.
+ */
+
+
+#include "request_context.h"
+#include <string.h>
+#include <stdexcept>
+#include <sstream>
+#include <cstdio>
+#include <atomic>
+
+#include "tools/stringTools.h"
+
+namespace kiwix {
+
+static std::atomic_ullong s_requestIndex(0);
+
+namespace {
+
+RequestMethod str2RequestMethod(const std::string& method) {
+  if      (method == "GET")     return RequestMethod::GET;
+  else if (method == "HEAD")    return RequestMethod::HEAD;
+  else if (method == "POST")    return RequestMethod::POST;
+  else if (method == "PUT")     return RequestMethod::PUT;
+  else if (method == "DELETE")  return RequestMethod::DELETE_;
+  else if (method == "CONNECT") return RequestMethod::CONNECT;
+  else if (method == "OPTIONS") return RequestMethod::OPTIONS;
+  else if (method == "TRACE")   return RequestMethod::TRACE;
+  else if (method == "PATCH")   return RequestMethod::PATCH;
+  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,
+                               const std::string& _method,
+                               const std::string& version) :
+  full_url(_url),
+  url(fullURL2LocalURL(_url, rootLocation)),
+  method(str2RequestMethod(_method)),
+  version(version),
+  requestIndex(s_requestIndex++),
+  acceptEncodingDeflate(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);
+
+  try {
+    acceptEncodingDeflate =
+        (get_header(MHD_HTTP_HEADER_ACCEPT_ENCODING).find("deflate") != 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&) {}
+}
+
+RequestContext::~RequestContext()
+{}
+
+MHD_Result RequestContext::fill_header(void *__this, enum MHD_ValueKind kind,
+                                       const char *key, const char *value)
+{
+  RequestContext *_this = static_cast<RequestContext*>(__this);
+  _this->headers[lcAll(key)] = value;
+  return MHD_YES;
+}
+
+MHD_Result RequestContext::fill_argument(void *__this, enum MHD_ValueKind kind,
+                                         const char *key, const char* value)
+{
+  RequestContext *_this = static_cast<RequestContext*>(__this);
+  _this->arguments[key] = value == nullptr ? "" : value;
+  return MHD_YES;
+}
+
+void RequestContext::print_debug_info() const {
+  printf("method    : %s (%d)\n", method==RequestMethod::GET ? "GET" :
+                                  method==RequestMethod::POST ? "POST" :
+                                  "OTHER", (int)method);
+  printf("version   : %s\n", version.c_str());
+  printf("request#  : %lld\n", requestIndex);
+  printf("headers   :\n");
+  for (auto it=headers.begin(); it!=headers.end(); it++) {
+    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());
+  }
+  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("has_range : %d\n", byteRange_.kind() != ByteRange::NONE);
+  printf("is_valid_url : %d\n", is_valid_url());
+  printf(".............\n");
+}
+
+
+RequestMethod RequestContext::get_method() const {
+  return method;
+}
+
+std::string RequestContext::get_url() const {
+  return url;
+}
+
+std::string RequestContext::get_url_part(int number) const {
+  size_t start = 1;
+  while(true) {
+    auto found = url.find('/', start);
+    if (number == 0) {
+      if (found == std::string::npos) {
+        return url.substr(start);
+      } else {
+        return url.substr(start, found-start);
+      }
+    } else {
+      if (found == std::string::npos) {
+        throw std::out_of_range("No parts");
+      }
+      start = found + 1;
+      number -= 1;
+    }
+  }
+}
+
+std::string RequestContext::get_full_url() const {
+  return full_url;
+}
+
+bool RequestContext::is_valid_url() const {
+  return !url.empty();
+}
+
+ByteRange RequestContext::get_range() const {
+  return byteRange_;
+}
+
+template<>
+std::string RequestContext::get_argument(const std::string& name) const {
+  return arguments.at(name);
+}
+
+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;
+}
+
+}

src/server/request_context.h

@@ -0,0 +1,119 @@
+/*
+ * Copyright 2009-2016 Emmanuel Engelhart <kelson@kiwix.org>
+ * 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 REQUEST_CONTEXT_H
+#define REQUEST_CONTEXT_H
+
+#include <string>
+#include <sstream>
+#include <map>
+#include <stdexcept>
+
+#include "byte_range.h"
+
+extern "C" {
+#include "microhttpd_wrapper.h"
+}
+
+namespace kiwix {
+
+enum class RequestMethod {
+    GET,
+    HEAD,
+    POST,
+    PUT,
+    DELETE_,
+    CONNECT,
+    OPTIONS,
+    TRACE,
+    PATCH,
+    OTHER
+};
+
+class KeyError : public std::runtime_error {};
+class IndexError: public std::runtime_error {};
+
+
+class RequestContext {
+  public: // functions
+    RequestContext(struct MHD_Connection* connection,
+                   std::string rootLocation,
+                   const std::string& url,
+                   const std::string& method,
+                   const std::string& version);
+    ~RequestContext();
+
+    void print_debug_info() const;
+
+    bool is_valid_url() const;
+
+    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;
+    }
+
+    template<class T>
+    T get_optional_param(const std::string& name, T default_value) const
+    {
+      try {
+        return get_argument<T>(name);
+      } catch (...) {}
+      return default_value;
+    }
+
+
+    RequestMethod get_method() const;
+    std::string get_url() const;
+    std::string get_url_part(int part) const;
+    std::string get_full_url() const;
+    std::string get_query() const;
+
+    ByteRange get_range() const;
+
+    bool can_compress() const { return acceptEncodingDeflate; }
+
+  private: // data
+    std::string full_url;
+    std::string url;
+    RequestMethod method;
+    std::string version;
+    unsigned long long requestIndex;
+
+    bool acceptEncodingDeflate;
+
+    ByteRange byteRange_;
+    std::map<std::string, std::string> headers;
+    std::map<std::string, std::string> arguments;
+
+  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*);
+};
+
+template<> std::string RequestContext::get_argument(const std::string& name) const;
+
+}
+
+#endif //REQUEST_CONTEXT_H

src/server/response.cpp

@@ -0,0 +1,420 @@
+/*
+ * Copyright 2019 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 "response.h"
+#include "request_context.h"
+#include "internalServer.h"
+#include "kiwixlib-resources.h"
+
+#include "tools/regexTools.h"
+#include "tools/stringTools.h"
+#include "tools/otherTools.h"
+
+#include "string.h"
+#include <mustache.hpp>
+#include <zlib.h>
+
+
+#define KIWIX_MIN_CONTENT_SIZE_TO_DEFLATE 100
+
+namespace kiwix {
+
+namespace
+{
+// some utilities
+
+std::string get_mime_type(const zim::Item& item)
+{
+  try {
+    return item.getMimetype();
+  } catch (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;
+}
+
+
+} // unnamed namespace
+
+Response::Response(bool verbose)
+  : m_verbose(verbose),
+    m_returnCode(MHD_HTTP_OK)
+{
+  add_header(MHD_HTTP_HEADER_ACCESS_CONTROL_ALLOW_ORIGIN, "*");
+}
+
+std::unique_ptr<Response> Response::build(const InternalServer& server)
+{
+  return std::unique_ptr<Response>(new Response(server.m_verbose.load()));
+}
+
+std::unique_ptr<Response> Response::build_304(const InternalServer& server, const ETag& etag)
+{
+  auto response = Response::build(server);
+  response->set_code(MHD_HTTP_NOT_MODIFIED);
+  response->m_etag = etag;
+  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)
+{
+  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);
+
+  return std::move(response);
+}
+
+std::unique_ptr<Response> Response::build_416(const InternalServer& server, size_t resourceLength)
+{
+  auto response = Response::build(server);
+// [FIXME] (compile with recent enough version of libmicrohttpd)
+//  response->set_code(MHD_HTTP_RANGE_NOT_SATISFIABLE);
+  response->set_code(416);
+  std::ostringstream oss;
+  oss << "bytes */" << resourceLength;
+  response->add_header(MHD_HTTP_HEADER_CONTENT_RANGE, oss.str());
+
+  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)
+{
+  auto response = Response::build(server);
+  response->m_returnCode = MHD_HTTP_FOUND;
+  response->add_header(MHD_HTTP_HEADER_LOCATION, redirectUrl);
+  return response;
+}
+
+static MHD_Result print_key_value (void *cls, enum MHD_ValueKind kind,
+                                   const char *key, const char *value)
+{
+  printf (" - %s: '%s'\n", key, value);
+  return MHD_YES;
+}
+
+
+struct RunningResponse {
+   zim::Item item;
+   int range_start;
+
+   RunningResponse(zim::Item item,
+                   int range_start) :
+     item(item),
+     range_start(range_start)
+   {}
+};
+
+static ssize_t callback_reader_from_item(void* cls,
+                                  uint64_t pos,
+                                  char* buf,
+                                  size_t max)
+{
+  RunningResponse* response = static_cast<RunningResponse*>(cls);
+
+  size_t max_size_to_set = min<size_t>(
+    max,
+    response->item.getSize() - pos - response->range_start);
+
+  if (max_size_to_set <= 0) {
+    return MHD_CONTENT_READER_END_WITH_ERROR;
+  }
+
+  zim::Blob blob = response->item.getData(response->range_start+pos, max_size_to_set);
+  memcpy(buf, blob.data(), max_size_to_set);
+  return max_size_to_set;
+}
+
+static void callback_free_response(void* cls)
+{
+  RunningResponse* response = static_cast<RunningResponse*>(cls);
+  delete response;
+}
+
+
+
+void print_response_info(int retCode, MHD_Response* response)
+{
+  printf("Response :\n");
+  printf("httpResponseCode : %d\n", retCode);
+  printf("headers :\n");
+  MHD_get_response_headers(response, print_key_value, nullptr);
+}
+
+
+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);
+}
+
+MHD_Response*
+Response::create_mhd_response(const RequestContext& request)
+{
+  MHD_Response* response = MHD_create_response_from_buffer(0, nullptr, MHD_RESPMEM_PERSISTENT);
+  return response;
+}
+
+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;
+    }
+  }
+
+  MHD_Response* response = MHD_create_response_from_buffer(
+    m_content.size(), const_cast<char*>(m_content.data()), MHD_RESPMEM_MUST_COPY);
+
+  if (shouldCompress) {
+    MHD_add_response_header(
+        response, MHD_HTTP_HEADER_VARY, "Accept-Encoding");
+    MHD_add_response_header(
+        response, MHD_HTTP_HEADER_CONTENT_ENCODING, "deflate");
+  }
+  return response;
+}
+
+MHD_Result Response::send(const RequestContext& request, 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");
+  const std::string etag = m_etag.get_etag();
+  if ( ! etag.empty() )
+    MHD_add_response_header(response, MHD_HTTP_HEADER_ETAG, etag.c_str());
+  for(auto& p: m_customHeaders) {
+    MHD_add_response_header(response, p.first.c_str(), p.second.c_str());
+  }
+
+  if (m_returnCode == MHD_HTTP_OK && m_byteRange.kind() == ByteRange::RESOLVED_PARTIAL_CONTENT)
+    m_returnCode = MHD_HTTP_PARTIAL_CONTENT;
+
+  if (m_verbose)
+    print_response_info(m_returnCode, response);
+
+  auto ret = MHD_queue_response(connection, m_returnCode, response);
+  MHD_destroy_response(response);
+  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),
+  m_content(content),
+  m_mimeType(mimetype),
+  m_withTaskbar(withTaskbar),
+  m_withLibraryButton(withLibraryButton),
+  m_blockExternalLinks(blockExternalLinks),
+  m_bookName(""),
+  m_bookTitle("")
+{
+  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)
+{
+   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));
+}
+
+std::unique_ptr<ContentResponse> ContentResponse::build(const InternalServer& server, const std::string& template_str, kainjow::mustache::data data, const std::string& mimetype, bool isHomePage) {
+  auto content = render_template(template_str, data);
+  return ContentResponse::build(server, content, mimetype, isHomePage);
+}
+
+ItemResponse::ItemResponse(bool verbose, const zim::Item& item, const std::string& mimetype, const ByteRange& byterange) :
+  Response(verbose),
+  m_item(item),
+  m_mimeType(mimetype)
+{
+  m_byteRange = byterange;
+  set_cacheable();
+  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)
+{
+  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();
+    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();
+    return response;
+  }
+
+  return std::unique_ptr<Response>(new ItemResponse(
+        server.m_verbose.load(),
+        item,
+        mimetype,
+        byteRange));
+}
+
+MHD_Response*
+ItemResponse::create_mhd_response(const RequestContext& request)
+{
+  const auto content_length = m_byteRange.length();
+  MHD_Response* response = MHD_create_response_from_callback(content_length,
+                                               16384,
+                                               callback_reader_from_item,
+                                               new RunningResponse(m_item, m_byteRange.first()),
+                                               callback_free_response);
+  MHD_add_response_header(response, MHD_HTTP_HEADER_ACCEPT_RANGES, "bytes");
+  if ( m_byteRange.kind() == ByteRange::RESOLVED_PARTIAL_CONTENT ) {
+    std::ostringstream oss;
+    oss << "bytes " << m_byteRange.first() << "-" << m_byteRange.last()
+        << "/" << m_item.getSize();
+
+    MHD_add_response_header(response,
+      MHD_HTTP_HEADER_CONTENT_RANGE, oss.str().c_str());
+  }
+
+  MHD_add_response_header(response,
+    MHD_HTTP_HEADER_CONTENT_LENGTH, kiwix::to_string(content_length).c_str());
+  return response;
+}
+
+
+}

src/server/response.h

@@ -0,0 +1,122 @@
+/*
+ * Copyright 2019 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 KIWIXLIB_SERVER_RESPONSE_H
+#define KIWIXLIB_SERVER_RESPONSE_H
+
+#include <string>
+#include <map>
+
+#include <mustache.hpp>
+#include "byte_range.h"
+#include "entry.h"
+#include "etag.h"
+
+extern "C" {
+#include "microhttpd_wrapper.h"
+}
+
+namespace kiwix {
+
+class InternalServer;
+class RequestContext;
+
+class EntryResponse;
+
+class Response {
+  public:
+    Response(bool verbose);
+    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);
+
+    MHD_Result send(const RequestContext& request, 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 add_header(const std::string& name, const std::string& value) { m_customHeaders[name] = value; }
+
+    int getReturnCode() const { return m_returnCode; }
+
+  private: // functions
+    virtual MHD_Response* create_mhd_response(const RequestContext& request);
+    MHD_Response* create_error_response(const RequestContext& request) const;
+
+  protected: // data
+    bool m_verbose;
+    int m_returnCode;
+    ByteRange m_byteRange;
+    ETag m_etag;
+    std::map<std::string, std::string> m_customHeaders;
+
+    friend class ItemResponse;
+};
+
+
+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);
+
+    void set_taskbar(const std::string& bookName, const std::string& bookTitle);
+
+  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 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);
+
+  private:
+    MHD_Response* create_mhd_response(const RequestContext& request);
+
+    zim::Item m_item;
+    std::string m_mimeType;
+};
+
+}
+
+#endif //KIWIXLIB_SERVER_RESPONSE_H

src/subprocess_unix.cpp

@@ -12,45 +12,31 @@
 UnixImpl::UnixImpl():
   m_pid(0),
   m_running(false),
-  m_mutex(PTHREAD_MUTEX_INITIALIZER),
-  m_waitingThread()
+  m_shouldQuit(false)
 {
 }
 
 UnixImpl::~UnixImpl()
 {
   kill();
-// Android has no pthread_cancel :(
-#ifdef __ANDROID__
-  pthread_kill(m_waitingThread, SIGUSR1);
-#else
-  pthread_cancel(m_waitingThread);
-#endif
+  m_shouldQuit = true;
+  m_waitingThread.join();
 }
 
-#ifdef __ANDROID__
-void thread_exit_handler(int sig) {
-  pthread_exit(0);
-}
-#endif
-
 void* UnixImpl::waitForPID(void* _self)
 {
-#ifdef __ANDROID__
-  struct sigaction actions;
-  memset(&actions, 0, sizeof(actions));
-  sigemptyset(&actions.sa_mask);
-  actions.sa_flags = 0;
-  actions.sa_handler = thread_exit_handler;
-  sigaction(SIGUSR1, &actions, NULL);
-#endif
-
   UnixImpl* self = static_cast<UnixImpl*>(_self);
-  waitpid(self->m_pid, NULL, WEXITED);
+  while (true) {
+    if (!waitpid(self->m_pid, NULL, WNOHANG)) {
+      break;
+    }
+    if (self->m_shouldQuit) {
+      return nullptr;
+    }
+    std::this_thread::sleep_for(std::chrono::milliseconds(100));
+  }
 
-  pthread_mutex_lock(&self->m_mutex);
   self->m_running = false;
-  pthread_mutex_unlock(&self->m_mutex);
 
   return self;
 }
@@ -67,14 +53,14 @@ void UnixImpl::run(commandLine_t& commandLine)
       commandLine.push_back(NULL);
       if (execvp(binary, const_cast<char* const*>(commandLine.data()))) {
         perror("Cannot launch\n");
-        exit(-1);
+        _exit(-1);
       }
 
       break;
     default:
       m_pid = pid;
       m_running = true;
-      pthread_create(&m_waitingThread, NULL, waitForPID, this);
+      m_waitingThread = std::thread(waitForPID, this);
       break;
   }
 }
@@ -86,8 +72,5 @@ bool UnixImpl::kill()
 
 bool UnixImpl::isRunning()
 {
-  pthread_mutex_lock(&m_mutex);
-  bool ret = m_running;
-  pthread_mutex_unlock(&m_mutex);
-  return ret;
+  return m_running;
 }

src/subprocess_unix.h

@@ -3,16 +3,16 @@
 
 #include "subprocess.h"
 
-#include <pthread.h>
-
+#include <atomic>
+#include <thread>
 
 class UnixImpl : public SubprocessImpl
 {
   private:
     int m_pid;
-    bool m_running;
-    pthread_mutex_t m_mutex;
-    pthread_t m_waitingThread;
+    std::atomic<bool> m_running;
+    std::atomic<bool> m_shouldQuit;
+    std::thread m_waitingThread;
 
   public:
     UnixImpl();

src/subprocess_windows.cpp

@@ -4,13 +4,15 @@
 
 #include <windows.h>
 #include <winbase.h>
+#include <shlwapi.h>
 #include <iostream>
 #include <sstream>
 
 WinImpl::WinImpl():
   m_pid(0),
   m_running(false),
-  m_handle(INVALID_HANDLE_VALUE)
+  m_subprocessHandle(INVALID_HANDLE_VALUE),
+  m_waitingThreadHandle(INVALID_HANDLE_VALUE)
 {
   InitializeCriticalSection(&m_criticalSection);
 }
@@ -18,14 +20,15 @@ WinImpl::WinImpl():
 WinImpl::~WinImpl()
 {
   kill();
-  CloseHandle(m_handle);
+  WaitForSingleObject(m_waitingThreadHandle, INFINITE);
+  CloseHandle(m_subprocessHandle);
   DeleteCriticalSection(&m_criticalSection);
 }
 
 DWORD WINAPI WinImpl::waitForPID(void* _self)
 {
   WinImpl* self = static_cast<WinImpl*>(_self);
-  WaitForSingleObject(self->m_handle, INFINITE);
+  WaitForSingleObject(self->m_subprocessHandle, INFINITE);
 
   EnterCriticalSection(&self->m_criticalSection);
   self->m_running = false;
@@ -34,11 +37,11 @@ DWORD WINAPI WinImpl::waitForPID(void* _self)
   return 0;
 }
 
-std::unique_ptr<wchar_t[]> toWideChar(const std::string& value)
+std::unique_ptr<wchar_t[]> toWideChar(const std::string& value, size_t min_size = 0)
 {
-  auto size = MultiByteToWideChar(CP_UTF8, 0,
+  size_t size = MultiByteToWideChar(CP_UTF8, 0,
                 value.c_str(), -1, nullptr, 0);
-  auto wdata = std::unique_ptr<wchar_t[]>(new wchar_t[size]);
+  auto wdata = std::unique_ptr<wchar_t[]>(new wchar_t[size>min_size?size:min_size]);
   auto ret = MultiByteToWideChar(CP_UTF8, 0,
                 value.c_str(), -1, wdata.get(), size);
   if (0 == ret) {
@@ -46,6 +49,9 @@ std::unique_ptr<wchar_t[]> toWideChar(const std::string& value)
     oss << "Cannot convert to wchar : " << GetLastError();
     throw std::runtime_error(oss.str());
   }
+  if (size < min_size) {
+    memset(wdata.get() + size, 0, min_size-size);
+  }
   return wdata;
 }
 
@@ -55,14 +61,16 @@ void WinImpl::run(commandLine_t& commandLine)
   STARTUPINFOW startInfo = {0};
   PROCESS_INFORMATION procInfo;
   startInfo.cb = sizeof(startInfo);
-  std::ostringstream oss;
+  std::wostringstream oss;
   for(auto& item: commandLine) {
-    oss << item << " ";
+    auto witem = toWideChar(item, MAX_PATH);
+    PathQuoteSpacesW(witem.get());
+    oss << witem.get() << " ";
   }
-  auto wCommandLine = toWideChar(oss.str());
+  auto wCommandLine = oss.str();
   if (CreateProcessW(
     NULL,
-    wCommandLine.get(),
+    const_cast<wchar_t*>(wCommandLine.c_str()),
     NULL,
     NULL,
     false,
@@ -70,18 +78,19 @@ void WinImpl::run(commandLine_t& commandLine)
     NULL,
     NULL,
     &startInfo,
-    &procInfo)) {
+    &procInfo))
+  {
     m_pid = procInfo.dwProcessId;
-    m_handle = procInfo.hProcess;
+    m_subprocessHandle = procInfo.hProcess;
     CloseHandle(procInfo.hThread);
     m_running = true;
-    CreateThread(NULL, 0, &waitForPID, this, 0, NULL	);
+    m_waitingThreadHandle = CreateThread(NULL, 0, &waitForPID, this, 0, NULL);
   }
 }
 
 bool WinImpl::kill()
 {
-  return TerminateProcess(m_handle, 0);
+  return TerminateProcess(m_subprocessHandle, 0);
 }
 
 bool WinImpl::isRunning()