Merge pull request #734 from kiwix/br_10.1.0

Add release badge

.github/workflows/ci.yml

@@ -8,19 +8,20 @@ jobs:
     steps:
       - name: Checkout code
         uses: actions/checkout@v1
-      - name: Setup python 3.5
-        uses: actions/setup-python@v1
+      - name: Setup python 3.10
+        uses: actions/setup-python@v2
         with:
-          python-version: '3.5'
+          python-version: '3.10'
       - name: Install packages
-        run:
+        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
+          ARCHIVE_NAME=deps2_osx_native_dyn_libkiwix.tar.xz
           wget -O- http://tmp.kiwix.org/ci/${ARCHIVE_NAME} | tar -xJ -C $HOME
       - name: Compile
         shell: bash
@@ -36,17 +37,8 @@ jobs:
           export LD_LIBRARY_PATH=$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:
@@ -93,7 +85,7 @@ jobs:
       HOME: /home/runner
     runs-on: ubuntu-latest
     container:
-      image: "kiwix/kiwix-build_ci:${{matrix.image_variant}}-26"
+      image: "kiwix/kiwix-build_ci:${{matrix.image_variant}}-31"
     steps:
     - name: Extract branch name
       shell: bash
@@ -114,7 +106,7 @@ jobs:
     - name: Install deps
       shell: bash
       run: |
-        ARCHIVE_NAME=deps2_${OS_NAME}_${{matrix.target}}_kiwix-lib.tar.xz
+        ARCHIVE_NAME=deps2_${OS_NAME}_${{matrix.target}}_libkiwix.tar.xz
         wget -O- http://tmp.kiwix.org/ci/${ARCHIVE_NAME} | tar -xJ -C /home/runner
     - name: Compile
       shell: bash
@@ -133,7 +125,7 @@ jobs:
         if [[ "${{matrix.target}}" =~ android_.* ]]; then
           MESON_OPTION="$MESON_OPTION -Dandroid=true"
         fi
-        cd $HOME/kiwix-lib
+        cd $HOME/libkiwix
         meson . build ${MESON_OPTION}
         cd build
         ninja
@@ -144,7 +136,7 @@ jobs:
       if: startsWith(matrix.target, 'native_')
       shell: bash
       run: |
-        cd $HOME/kiwix-lib/build
+        cd $HOME/libkiwix/build
         meson test --verbose
         ninja coverage
       env:
@@ -153,7 +145,7 @@ jobs:
     - name: Publish coverage
       shell: bash
       run: |
-        cd $HOME/kiwix-lib
+        cd $HOME/libkiwix
         curl https://codecov.io/bash -o codecov.sh
         bash codecov.sh -n "${OS_NAME}_${{matrix.target}}" -Z
         rm codecov.sh

.github/workflows/package.yml

@@ -7,7 +7,11 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        distro: [ubuntu-hirsute, ubuntu-groovy, ubuntu-focal, ubuntu-bionic]
+        distro:
+          - ubuntu-jammy
+          - ubuntu-impish
+          - ubuntu-focal
+          - ubuntu-bionic
     steps:
       - uses: actions/checkout@v2
 
@@ -30,18 +34,18 @@ jobs:
           email: release+launchpad@kiwix.org
           distro: ${{ matrix.distro }}
 
-      - uses: legoktm/gh-action-build-deb@ubuntu-hirsute
-        if: matrix.distro == 'ubuntu-hirsute'
-        name: Build package for ubuntu-hirsute
-        id: build-ubuntu-hirsute
+      - uses: legoktm/gh-action-build-deb@ubuntu-jammy
+        if: matrix.distro == 'ubuntu-jammy'
+        name: Build package for ubuntu-jammy
+        id: build-ubuntu-jammy
         with:
           args: --no-sign
           ppa: ${{ steps.ppa.outputs.ppa }}
 
-      - uses: legoktm/gh-action-build-deb@ubuntu-groovy
-        if: matrix.distro == 'ubuntu-groovy'
-        name: Build package for ubuntu-groovy
-        id: build-ubuntu-groovy
+      - uses: legoktm/gh-action-build-deb@ubuntu-impish
+        if: matrix.distro == 'ubuntu-impish'
+        name: Build package for ubuntu-impish
+        id: build-ubuntu-impish
         with:
           args: --no-sign
           ppa: ${{ steps.ppa.outputs.ppa }}

.gitignore

@@ -4,3 +4,6 @@ subprojects/googletest-release*
 *.class
 build/
 .vscode/
+builddir/
+.cache/
+.clangd/

ChangeLog

@@ -1,3 +1,102 @@
+libkiwix 10.1.0
+===============
+
+ This release is an important one as it fixes a Xss vulnerability introduced
+ in libkiwix 10.0.0
+
+ * [SECURITY] Fix a Xss attack vulnerability (introduced in 10.0.0) (@juuz0 #721)
+ * [server] Add a option to set a limit on the number of connexion per IP (@kelson42 #700)
+ * [server] Do not display a lang tag in the UI if the book has no language (@juuz0 #706)
+ * [server] Add the book title associated to a search results (@thavelick #705, @mgautierfr #718)
+ * Add `dc:issued` to opds output stream (@veloman-yunkan #715)
+ * Add handling of several languages not provided by ICU (@juuz0 #701)
+ * [server] Add a caching system for search and suggestion (@maneeshpm #620)
+ * Fix cross-compilation (@kelson42 #703)
+ * Add unit-testing of suggestions and error pages (@veloman-yunkan #709 #710 #727)
+ * Better testing system of html response (@veloman-yunkan #725)
+
+libkiwix 10.0.1
+===============
+
+ * [server] The catalog search interpret `count=0` as no limit.
+   This was the case for a long time. This was changed unintentionally
+   (@veloman-yunkan #686)
+ * [server] Correctly generere a human friendly title in the server frontend.
+   (@juuz0 #687, @kelson42 #689)
+ * [server] Fix download button if there is no url do download from.
+   (@juuz0 #691)
+ * Add non-minified isotope.pkdg.js
+   Needed for debian packaging as we need the source and minified version is
+   not the source (@legoktm #693)
+ * [server] Add a tooltip with the full language for the lang tag.
+ * CI fixes (@kelson42 @legoktm)
+
+libkiwix 10.0.0
+===============
+
+ This release is huge release.
+ The project has been renamed to libkiwix, it is more coherent with the library name.
+
+ * Server front page :
+  - Use js in the front page to display the available book,
+     using the OPDS stream as source. The front page is now populated only with
+     the visible books and user can search for books. (@MananJethwany #530, #541, #534)
+     (@kelson42 #628)
+  - Revamp css (@MananJethwany #559)
+  - Correctly Convert 3iso language code to 2iso (@juuz0 #672)
+
+ * Server suggestions search :
+  - Add pagination for suggestion search (@maneeshpm #591)
+  - Fix suggestion system (@MananJethwany #498)
+  - Provide the kind and path (when adapted) to the suggestion answer (@MananJethwany #464)
+  - The displayed suggestion have now highligth on the searched terms (@maneeshpm #505)
+  - Properly handle html encoding of suggestions (@veloman-yunkan #458)
+
+ * Server improvements :
+  - Remove meta endpoints (@mgautier #669)
+  - Add raw endpoints to get the raw content of a zim (@mgautierfr #646)
+  - Add details on 404 error pages (@soumyankar #490)
+  - Fix headbar insertion when `<head>` tag has attributes (@kelson42 #440)
+  - Better headbar insertion (after charset definition) (@kelson42 #442)
+
+ * New OPDS Stream v2 :
+  - Add a list of categories (@veloman-yunkan)
+  - Support for partial entries (@veloman-yunkan #602)
+  - Support multiple icons size in the OPDS stream (@veloman-yunkan #577 #630)
+  - Add language endpoint to catalog (@veloman-yunkan #553)
+  - Add illustration API to get the illustration of a book (@mgautierfr #645)
+  - OPDS search can now filter books by category (@veloman-yunkan #459)
+
+ * Library improvements :
+  - Allow the libray to be live reloaded when the library.xml changes (@veloman-yunkan #636)
+  - Properly handle removing of book from the library (@veloman-yunkan #485)
+  - Use xapian to search for books in the library (@veloman-yunkan #460, #488)
+
+ * Added methods/functions :
+  - Fix `fileExist` and introduce `fileReadable` (@juuz0 #668)
+  - Add `getVersions` and `printVersions` functions (@kelson42 #665)
+  - Add `getNetworkInterfaces()` and `getBestPublicIP()` functions (@juuz0 #622)
+  - Add `get_zimid()` method to the search result (@maneeshpm #510)
+
+ * Various improvements :
+  - Better secret value for aria2c rpc (@juuz0 #666)
+  - Avoid duplicated Archive/Reader in the Searcher (@veloman-yunkan #648)
+  - Add basic documentation (@mgautierfr #640)
+  - Do not use Reader internally (@maneeshpm #536 #576)
+  - Remove dependency headers from our public headers (@mgautierfr #574)
+  - Downloader now don't write metalink on the filesystem (@kelson42 #502)
+  - Support opening a zim file using a fd (@veloman-yukan #429)
+  - Use C++11 std::thread instead of pthread (@mgautierfr #445)
+  - [READER] Do not crash if zim file has no `Counter` metadata (@mgautierfr #449)
+  - Ensure libzim dependency is compiled with xapian (@mgautierfr #434)
+  - Support video and audio mimetype in `getMediaCount` (@kelson42 #439)
+  - Better parsing of the counterMap (@kelson42 #437)
+  - Adapt libkiwix to libzim 7.0.0 (@mgautierfr #428)
+  - Remove deprecated methods (@mgautierfr)
+  - CI: Build package for Ubuntu Hirsute, Impish and Jammy (@legoktm #431 #568) and remove Groovy
+  - Fix compilation for FreeBSD (@swills g#432)
+  - Many fixes and improvement (@MananJethwany, @maneeshpm, @veloman-yunkan, @mgautierfr)
+
 kiwix-lib 9.4.1
 ===============
 

README.md

@@ -1,50 +1,51 @@
-Kiwix library
-=============
+Libkiwix
+========
 
-The Kiwix library provides the [Kiwix](https://kiwix.org) software
-suite core. It contains the code shared by all Kiwix ports (Windows,
+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, ...).
 
-[![Download](https://api.bintray.com/packages/kiwix/kiwix/kiwixlib/images/download.svg)](https://bintray.com/kiwix/kiwix/kiwixlib/_latestVersion)
-[![Build Status](https://github.com/kiwix/kiwix-lib/workflows/CI/badge.svg?query=branch%3Amaster)](https://github.com/kiwix/kiwix-lib/actions?query=branch%3Amaster)
-[![CodeFactor](https://www.codefactor.io/repository/github/kiwix/kiwix-lib/badge)](https://www.codefactor.io/repository/github/kiwix/kiwix-lib)
-[![Codecov](https://codecov.io/gh/kiwix/kiwix-lib/branch/master/graph/badge.svg)](https://codecov.io/gh/kiwix/kiwix-lib)
+[![Release](https://img.shields.io/github/v/tag/kiwix/libkiwix?label=release&sort=semver)](https://download.kiwix.org/release/libkiwix/)
+[![Repositories](https://img.shields.io/repology/repositories/libkiwix?label=repositories)](https://github.com/kiwix/libkiwix/wiki/Repology)
+[![Build Status](https://github.com/kiwix/libkiwix/workflows/CI/badge.svg?query=branch%3Amaster)](https://github.com/kiwix/libkiwix/actions?query=branch%3Amaster)
+[![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)
 
-[![Packaging status](https://repology.org/badge/vertical-allrepos/kiwix-lib.svg)](https://repology.org/project/kiwix-lib/versions)
-
 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 3 only.
-* [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)
+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)
+
+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)
@@ -56,12 +57,12 @@ 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.
+`libkiwix` directory as install prefix.
 
 Environment
 -------------
 
-The Kiwix library builds using [Meson](https://mesonbuild.com/) version
+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.
 
@@ -77,7 +78,7 @@ 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
@@ -85,12 +86,20 @@ 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`.
 
+The android wrapper uses deprecated methods of libkiwix so it cannot be compiled
+with `werror=true` (the default). So you must pass `-Dwerror=false` to meson:
+
+```bash
+meson . build -Dwrapper=android -Dwerror=false
+ninja -C build
+```
+
 Testing
 -------
 
@@ -103,7 +112,7 @@ 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
@@ -146,6 +155,57 @@ cp ninja ../bin
 cd ..
 ```
 
+Custom Index Page
+-----------------
+
+to use custom welcome page mention `customIndexPage` argument in `kiwix::internalServer()` or use `kiwix::server->setCustomIndexTemplate()`.
+(note - while using custom html file please mention all external links as absolute path.)
+
+to create a HTML template with custom JS you need to have a look at various OPDS based endpoints as mentioned [here](https://wiki.kiwix.org/wiki/OPDS) to load books.
+
+To use JS provided by kiwix-serve you can use the following template to start with ->
+
+```
+<!DOCTYPE html>
+<html xmlns="http://www.w3.org/1999/xhtml">
+  <head>
+    <meta charset="UTF-8" />
+    <meta name="viewport" content="width=device-width,initial-scale=1" />
+    <title><-- Custom Tittle --></title>
+    <script
+      type="text/javascript"
+      src="{{root}}/skin/jquery-ui/external/jquery/jquery.js"
+    ></script>
+    <script
+      type="text/javascript"
+      src="{{root}}/skin/jquery-ui/jquery-ui.min.js"
+    ></script>
+    <script src="{{root}}/skin/isotope.pkgd.min.js" defer></script>
+    <script src="{{root}}/skin/iso6391To3.js"></script>
+    <script type="text/javascript" src="{{root}}/skin/index.js" defer></script>
+  </head>
+  <body>
+  </body>
+</html>
+```
+
+- To get books listed using `index.js` add - `<div class="book__list"></div>` under body tag.
+- To get number of books listed add - `<h3 class="kiwixHomeBody__results"></h3>` under body tag.
+- To add language select box add - `<select id="languageFilter"></select>` under body tag.
+- To add language select box add - `<select id="categoryFilter"></select>` under body tag.
+- To add search box for books use following form -
+    ```
+        <form id='kiwixSearchForm'>
+        <input type="text" name="q" placeholder="Search" id="searchFilter" class='kiwixSearch filter'>
+        <input type="submit" class="searchButton" value="Search"/>
+        </form>
+    ```
+
+
+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

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

@@ -26,10 +26,10 @@ task writePom {
     project {
         groupId 'org.kiwix.kiwixlib'
         artifactId 'kiwixlib'
-        version '10.0.0' + (System.env.KIWIXLIB_BUILDVERSION == null ? '' : '-'+System.env.KIWIXLIB_BUILDVERSION)
+        version '10.1.0' + (System.env.KIWIXLIB_BUILDVERSION == null ? '' : '-'+System.env.KIWIXLIB_BUILDVERSION)
         packaging 'aar'
         name 'kiwixlib'
-        url 'https://github.com/kiwix/kiwix-lib'
+        url 'https://github.com/kiwix/libkiwix'
         licenses {
             license {
                 name 'GPLv3'
@@ -44,9 +44,9 @@ task writePom {
           }
       }
       scm {
-          connection 'https://github.com/kiwix/kiwix-lib.git'
-          developerConnection 'https://github.com/kiwix/kiwix-lib.git'
-          url 'https://github.com/kiwix/kiwix-lib'
+          connection 'https://github.com/kiwix/libkiwix.git'
+          developerConnection 'https://github.com/kiwix/libkiwix.git'
+          url 'https://github.com/kiwix/libkiwix'
       }
     }
   }.withXml {

debian/control

@@ -4,7 +4,7 @@ Maintainer: Kiwix team <kiwix@kiwix.org>
 Build-Depends: debhelper-compat (= 13),
  meson,
  pkg-config,
- libzim-dev (>= 6.1.8),
+ libzim-dev (>= 7.2.0~),
  libcurl4-gnutls-dev,
  libicu-dev,
  libgtest-dev,
@@ -15,7 +15,7 @@ Build-Depends: debhelper-compat (= 13),
  zlib1g-dev
 Standards-Version: 4.5.0
 Section: libs
-Homepage: https://github.com/kiwix/kiwix-lib
+Homepage: https://github.com/kiwix/libkiwix
 Rules-Requires-Root: no
 
 Package: libkiwix-dev
@@ -23,7 +23,7 @@ Section: libdevel
 Architecture: any
 Multi-Arch: same
 Depends: libkiwix10 (= ${binary:Version}), ${misc:Depends}, python3,
- libzim-dev (>= 6.0.0),
+ libzim-dev (>= 7.2.0~),
  libicu-dev,
  libpugixml-dev,
  libcurl4-gnutls-dev,

docs/.gitignore

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

docs/conf.py

@@ -0,0 +1,72 @@
+# Configuration file for the Sphinx documentation builder.
+#
+# This file only contains a selection of the most common options. For a full
+# list see the documentation:
+# https://www.sphinx-doc.org/en/master/usage/configuration.html
+
+# -- Path setup --------------------------------------------------------------
+
+# If extensions (or modules to document with autodoc) are in another directory,
+# add these directories to sys.path here. If the directory is relative to the
+# documentation root, use os.path.abspath to make it absolute, like shown here.
+#
+import os
+# import sys
+# sys.path.insert(0, os.path.abspath('.'))
+
+
+# -- Project information -----------------------------------------------------
+
+project = 'libkiwix'
+copyright = '2022, libkiwix-team'
+author = 'libzim-team'
+
+
+# -- General configuration ---------------------------------------------------
+
+on_rtd = os.environ.get('READTHEDOCS', None) == 'True'
+
+# Add any Sphinx extension module names here, as strings. They can be
+# extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
+# ones.
+extensions = [
+    'breathe',
+    'exhale'
+]
+
+# Add any paths that contain templates here, relative to this directory.
+templates_path = ['_templates']
+
+# List of patterns, relative to source directory, that match files and
+# directories to ignore when looking for source files.
+# This pattern also affects html_static_path and html_extra_path.
+exclude_patterns = ['_build', 'Thumbs.db', '.DS_Store']
+
+
+if not on_rtd:
+    html_theme = 'sphinx_rtd_theme'
+
+# Add any paths that contain custom static files (such as style sheets) here,
+# relative to this directory. They are copied after the builtin static files,
+# so a file named "default.css" will overwrite the builtin "default.css".
+html_static_path = ['_static']
+
+breathe_projects = {
+    "libkiwix": "./xml"
+}
+breathe_default_project = 'libkiwix'
+
+exhale_args = {
+    "containmentFolder":   "./api",
+    "rootFileName":        "ref_api.rst",
+    "rootFileTitle":       "Reference API",
+    "doxygenStripFromPath":"..",
+    "treeViewIsBootstrap": True,
+    "createTreeView" : True,
+    "exhaleExecutesDoxygen": True,
+    "exhaleDoxygenStdin":    "INPUT = ../include"
+}
+
+primary_domain = 'cpp'
+
+highlight_language = 'cpp'

docs/index.rst

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

docs/meson.build

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

docs/requirements.txt

@@ -0,0 +1,3 @@
+breathe
+exhale
+sphinx<4

docs/usage.rst

@@ -0,0 +1,17 @@
+Libkiwix programming
+====================
+
+Introduction
+------------
+
+libkiwix is written in C++. To use the library, you need the include files of libkiwix have
+to link against libzim.
+
+Errors are handled with exceptions. When something goes wrong, libzim throws an error,
+which is always derived from std::exception.
+
+All classes are defined in the namespace kiwix.
+
+libkiwix is a set of tools to manage zim files and provide some common functionnality.
+While libkiwix has some wrappers around libzim classes, they are deprecated and will be removed
+in the future.

format_code.sh

@@ -7,6 +7,7 @@ files=(
 "include/common/otherTools.h"
 "include/common/regexTools.h"
 "include/common/networkTools.h"
+"include/common/archiveTools.h"
 "include/manager.h"
 "include/reader.h"
 "include/kiwix.h"
@@ -22,6 +23,7 @@ files=(
 "src/common/pathTools.cpp"
 "src/common/regexTools.cpp"
 "src/common/otherTools.cpp"
+"src/common/archiveTools.cpp"
 "src/common/networkTools.cpp"
 "src/common/stringTools.cpp"
 "src/xapianSearcher.cpp"

include/book.h

@@ -21,11 +21,19 @@
 #define KIWIX_BOOK_H
 
 #include <string>
+#include <vector>
+#include <memory>
+#include <mutex>
+#include "common.h"
 
 namespace pugi {
 class xml_node;
 }
 
+namespace zim {
+class Archive;
+}
+
 namespace kiwix
 {
 
@@ -37,12 +45,32 @@ class Reader;
  */
 class Book
 {
- public:
+ public: // types
+  class Illustration
+  {
+    friend class Book;
+   public:
+    uint16_t width = 48;
+    uint16_t height = 48;
+    std::string mimeType;
+    std::string url;
+
+    const std::string& getData() const;
+
+   private:
+    mutable std::string data;
+    mutable std::mutex mutex;
+  };
+
+  typedef std::vector<std::shared_ptr<const Illustration>> Illustrations;
+
+ public: // functions
   Book();
   ~Book();
 
   bool update(const Book& other);
-  void update(const Reader& reader);
+  DEPRECATED void update(const Reader& reader);
+  void update(const zim::Archive& archive);
   void updateFromXml(const pugi::xml_node& node, const std::string& baseDir);
   void updateFromOpds(const pugi::xml_node& node, const std::string& urlHost);
   std::string getHumanReadableIdFromPath() const;
@@ -59,6 +87,7 @@ 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;
@@ -67,9 +96,13 @@ class Book
   const uint64_t& getArticleCount() const { return m_articleCount; }
   const uint64_t& getMediaCount() const { return m_mediaCount; }
   const uint64_t& getSize() const { return m_size; }
-  const std::string& getFavicon() const;
-  const std::string& getFaviconUrl() const { return m_faviconUrl; }
-  const std::string& getFaviconMimeType() const { return m_faviconMimeType; }
+  DEPRECATED const std::string& getFavicon() const;
+  DEPRECATED const std::string& getFaviconUrl() const;
+  DEPRECATED const std::string& getFaviconMimeType() const;
+
+  Illustrations getIllustrations() const;
+  std::shared_ptr<const Illustration> getIllustration(unsigned int size) const;
+
   const std::string& getDownloadId() const { return m_downloadId; }
 
   void setReadOnly(bool readOnly) { m_readOnly = readOnly; }
@@ -90,17 +123,20 @@ class Book
   void setArticleCount(uint64_t articleCount) { m_articleCount = articleCount; }
   void setMediaCount(uint64_t mediaCount) { m_mediaCount = mediaCount; }
   void setSize(uint64_t size) { m_size = size; }
-  void setFavicon(const std::string& favicon) { m_favicon = favicon; }
-  void setFaviconMimeType(const std::string& faviconMimeType) { m_faviconMimeType = faviconMimeType; }
   void setDownloadId(const std::string& downloadId) { m_downloadId = downloadId; }
 
- protected:
+ private: // functions
+  std::string getCategoryFromTags() const;
+  const Illustration& getDefaultIllustration() const;
+
+ protected: // data
   std::string m_id;
   std::string m_downloadId;
   std::string m_path;
   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;
@@ -114,9 +150,11 @@ class Book
   uint64_t m_mediaCount = 0;
   bool m_readOnly = false;
   uint64_t m_size = 0;
-  mutable std::string m_favicon;
-  std::string m_faviconUrl;
-  std::string m_faviconMimeType;
+  Illustrations m_illustrations;
+
+  // Used as the return value of getDefaultIllustration() when no default
+  // illustration is found in the book
+  static const Illustration missingDefaultIllustration;
 };
 
 }

include/entry.h

@@ -47,7 +47,7 @@ class Entry
      *
      * @param article a zim::Article object
      */
-    Entry(zim::Entry entry);
+    DEPRECATED Entry(zim::Entry entry) : Entry(entry, true) {};
     virtual ~Entry() = default;
 
     /**
@@ -103,7 +103,7 @@ class 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
@@ -111,7 +111,7 @@ 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 { return entry.getItem().getDirectAccessInformation(); }
+    zim::Item::DirectAccessInfo getDirectAccessInfo() const { return entry.getItem().getDirectAccessInformation(); }
 
     /**
      * Get the size of the entry.
@@ -176,6 +176,16 @@ class Entry
 
   private:
     zim::Entry entry;
+
+  private:
+    // Entry is deprecated, so we've marked the constructor as deprecated.
+    // But we still need to construct the entry (in our deprecated code)
+    // To avoid warning because we use deprecated function, we create a second
+    // constructor not deprecated. The `bool marker` is unused, it sole purpose
+    // is to change the signature to have two different constructor.
+    // This one is not deprecated and we must use it in our private code.
+    Entry(zim::Entry entry, bool marker);
+    friend class Reader;
 };
 
 }

include/kiwix.h

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

include/library.h

@@ -24,6 +24,8 @@
 #include <vector>
 #include <map>
 #include <memory>
+#include <mutex>
+#include <zim/archive.h>
 
 #include "book.h"
 #include "bookmark.h"
@@ -35,6 +37,7 @@ namespace kiwix
 {
 
 class OPDSDumper;
+class Library;
 
 enum supportedListSortBy { UNSORTED, TITLE, SIZE, DATE, CREATOR, PUBLISHER };
 enum supportedListMode {
@@ -48,18 +51,23 @@ enum supportedListMode {
 };
 
 class Filter {
-  private:
+  public: // types
+    using Tags = std::vector<std::string>;
+
+  private: // data
     uint64_t activeFilters;
-    std::vector<std::string> _acceptTags;
-    std::vector<std::string> _rejectTags;
+    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:
+  public: // functions
     Filter();
     ~Filter() = default;
 
@@ -93,33 +101,105 @@ class Filter {
     /**
      * Set the filter to only accept book with corresponding tag.
      */
-    Filter& acceptTags(std::vector<std::string> tags);
-    Filter& rejectTags(std::vector<std::string> tags);
+    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);
+    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;
 };
 
+/**
+ * This class is not part of the libkiwix API. Its only purpose is
+ * to simplify the implementation of the Library's move operations
+ * and avoid bugs should new data members be added to Library.
+ */
+class LibraryBase
+{
+protected: // types
+  typedef uint64_t LibraryRevision;
+
+  struct Entry : Book
+  {
+    LibraryRevision lastUpdatedRevision = 0;
+
+    // May also keep the Archive and Reader pointers here and get
+    // rid of the m_readers and m_archives data members in Library
+  };
+
+protected: // data
+  LibraryRevision m_revision;
+  std::map<std::string, Entry> m_books;
+  std::map<std::string, std::shared_ptr<Reader>> m_readers;
+  std::map<std::string, std::shared_ptr<zim::Archive>> m_archives;
+  std::vector<kiwix::Bookmark> m_bookmarks;
+  class BookDB;
+  std::unique_ptr<BookDB> m_bookDB;
+
+protected: // functions
+  LibraryBase();
+  ~LibraryBase();
+
+  LibraryBase(LibraryBase&& );
+  LibraryBase& operator=(LibraryBase&& );
+};
 
 /**
  * A Library store several books.
  */
-class Library
+class Library : private LibraryBase
 {
-  std::map<std::string, kiwix::Book> m_books;
-  std::map<std::string, std::shared_ptr<Reader>> m_readers;
-  std::vector<kiwix::Bookmark> m_bookmarks;
+  // all data fields must be added in LibraryBase
+  mutable std::mutex m_mutex;
+
+ public:
+  typedef LibraryRevision Revision;
+  typedef std::vector<std::string> BookIdCollection;
+  typedef std::map<std::string, int> AttributeCounts;
 
  public:
   Library();
   ~Library();
 
+  /**
+   * 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.
    *
@@ -132,6 +212,11 @@ class Library
    */
   bool addBook(const Book& book);
 
+  /**
+   * A self-explanatory alias for addBook()
+   */
+  bool addOrUpdateBook(const Book& book) { return addBook(book); }
+
   /**
    * Add a bookmark to the library.
    *
@@ -148,9 +233,15 @@ class Library
    */
   bool removeBookmark(const std::string& zimId, const std::string& url);
 
-  Book& getBookById(const std::string& id);
-  Book& getBookByPath(const std::string& path);
-  std::shared_ptr<Reader> getReaderById(const std::string& id);
+  // XXX: This is a non-thread-safe operation
+  const Book& getBookById(const std::string& id) const;
+  // XXX: This is a non-thread-safe operation
+  const Book& getBookByPath(const std::string& path) const;
+
+  Book getBookByIdThreadSafe(const std::string& id) const;
+
+  DEPRECATED std::shared_ptr<Reader> getReaderById(const std::string& id);
+  std::shared_ptr<zim::Archive> getArchiveById(const std::string& id);
 
   /**
    * Remove a book from the library.
@@ -166,7 +257,7 @@ class Library
    * @param path the path of the file to write to.
    * @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.
@@ -174,7 +265,7 @@ class Library
    * @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);
+  bool writeBookmarksToFile(const std::string& path) const;
 
   /**
    * Get the number of book in the library.
@@ -183,53 +274,56 @@ 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 languagues of the books in the library with counts.
+   *
+   * @return A list of languages with the count of books in each language.
+   */
+  AttributeCounts getBooksLanguagesWithCounts() const;
+
+  /**
+   * Get all categories of the books in the library.
+   *
+   * @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 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();
-
-  /**
-   * Filter the library and generate a new one with the keep elements.
-   *
-   * This is equivalent to `listBookIds(ALL, UNSORTED, search)`.
-   *
-   * @param search List only books with search in the title or description.
-   * @return The list of bookIds corresponding to the query.
-   */
-  DEPRECATED std::vector<std::string> filter(const std::string& search);
-
+  BookIdCollection getBooksIds() const;
 
   /**
    * Filter the library and return the id of the keep elements.
@@ -237,7 +331,7 @@ class Library
    * @param filter The filter to use.
    * @return The list of bookIds corresponding to the filter.
    */
-  std::vector<std::string> filter(const Filter& filter);
+  BookIdCollection filter(const Filter& filter) const;
 
 
   /**
@@ -247,43 +341,40 @@ class Library
    * @param comparator how to sort the books
    * @return The sorted list of books
    */
-  void sort(std::vector<std::string>& bookIds, supportedListSortBy sortBy, bool ascending);
+  void sort(BookIdCollection& bookIds, supportedListSortBy sortBy, bool ascending) const;
 
   /**
-   * List books in the library.
+   * Return the current revision of the library.
    *
-   * @param mode The mode of listing :
-   *             - LOCAL    : list only local books (with a path).
-   *             - REMOTE   : list only remote books (with an url).
-   *             - VALID    : list only valid books (without a path or with a
-   *                          path pointing to a valid zim file).
-   *             - NOLOCAL  : list only books without valid path.
-   *             - NOREMOTE : list only books without url.
-   *             - NOVALID  : list only books not valid.
-   *             - ALL : Do not do any filter (LOCAL or REMOTE)
-   *             - Flags can be combined.
-   * @param sortBy Attribute to sort by the book list.
-   * @param search List only books with search in the title, description.
-   * @param language List only books in this language.
-   * @param creator List only books of this creator.
-   * @param publisher List only books of this publisher.
-   * @param maxSize Do not list book bigger than maxSize.
-   *                Set to 0 to cancel this filter.
-   * @return The list of bookIds corresponding to the query.
+   * The revision of the library is updated (incremented by one) only by
+   * the addBook() operation.
+   *
+   * @return Current revision of the library.
    */
-  DEPRECATED std::vector<std::string> listBooksIds(
-    int supportedListMode = ALL,
-    supportedListSortBy sortBy = UNSORTED,
-    const std::string& search = "",
-    const std::string& language = "",
-    const std::string& creator = "",
-    const std::string& publisher = "",
-    const std::vector<std::string>& tags = {},
-    size_t maxSize = 0);
+  LibraryRevision getRevision() const;
+
+  /**
+   * Remove books that have not been updated since the specified revision.
+   *
+   * @param  rev the library revision to use
+   * @return Count of books that were removed by this operation.
+   */
+  uint32_t removeBooksNotUpdatedSince(LibraryRevision rev);
 
   friend class OPDSDumper;
   friend class libXMLDumper;
+
+private: // types
+  typedef const std::string& (Book::*BookStrPropMemFn)() const;
+
+private: // functions
+  AttributeCounts getBookAttributeCounts(BookStrPropMemFn p) const;
+  std::vector<std::string> getBookPropValueSet(BookStrPropMemFn p) const;
+  BookIdCollection filterViaBookDB(const Filter& filter) const;
+  void updateBookDB(const Book& book);
+  void dropReader(const std::string& bookId);
 };
+
 }
 
 #endif

include/libxml_dumper.h

@@ -38,7 +38,7 @@ class LibXMLDumper
 {
  public:
   LibXMLDumper() = default;
-  LibXMLDumper(Library* library);
+  LibXMLDumper(const Library* library);
   ~LibXMLDumper();
 
   /**
@@ -69,10 +69,10 @@ class LibXMLDumper
    *
    * @param library The library to dump.
    */
-  void setLibrary(Library* library) { this->library = library; }
+  void setLibrary(const Library* library) { this->library = library; }
 
  protected:
-  kiwix::Library* library;
+  const kiwix::Library* library;
   std::string baseDir;
  private:
   void handleBook(Book book, pugi::xml_node root_node);

include/manager.h

@@ -26,6 +26,7 @@
 
 #include <string>
 #include <vector>
+#include <memory>
 
 namespace pugi {
 class xml_document;
@@ -34,26 +35,25 @@ class xml_document;
 namespace kiwix
 {
 
-class LibraryManipulator {
- public:
-  virtual ~LibraryManipulator() {}
-  virtual bool addBookToLibrary(Book book) = 0;
-  virtual void addBookmarkToLibrary(Bookmark bookmark) = 0;
-};
+class LibraryManipulator
+{
+ public: // functions
+  explicit LibraryManipulator(Library* library);
+  virtual ~LibraryManipulator();
 
-class DefaultLibraryManipulator : public LibraryManipulator {
- public:
-  DefaultLibraryManipulator(Library* library) :
-    library(library) {}
-  virtual ~DefaultLibraryManipulator() {}
-  bool addBookToLibrary(Book book) {
-    return library->addBook(book);
-  }
-  void addBookmarkToLibrary(Bookmark bookmark) {
-    library->addBookmark(bookmark);
-  }
- private:
-   kiwix::Library* library;
+  Library& getLibrary() const { return library; }
+
+  bool addBookToLibrary(const Book& book);
+  void addBookmarkToLibrary(const Bookmark& bookmark);
+  uint32_t removeBooksNotUpdatedSince(Library::Revision rev);
+
+ protected: // overrides
+  virtual void bookWasAddedToLibrary(const Book& book);
+  virtual void bookmarkWasAddedToLibrary(const Bookmark& bookmark);
+  virtual void booksWereRemovedFromLibrary();
+
+ private: // data
+  kiwix::Library& library;
 };
 
 /**
@@ -61,10 +61,12 @@ class DefaultLibraryManipulator : public LibraryManipulator {
  */
 class Manager
 {
- public:
-  Manager(LibraryManipulator* manipulator);
-  Manager(Library* library);
-  ~Manager();
+ public: // types
+  typedef std::vector<std::string> Paths;
+
+ public: // functions
+  explicit Manager(LibraryManipulator* manipulator);
+  explicit Manager(Library* library);
 
   /**
    * Read a `library.xml` and add book in the file to the library.
@@ -72,10 +74,22 @@ class Manager
    * @param path The (utf8) path to the `library.xml`.
    * @param readOnly Set if the libray path could be overwritten latter with
    *                 updated content.
+   * @param trustLibrary use book metadata coming from XML.
    * @return True if file has been properly parsed.
    */
   bool readFile(const std::string& path, bool readOnly = true, bool trustLibrary = true);
 
+  /**
+   * Sync the contents of the library with one or more `library.xml` files.
+   *
+   * The metadata of the library files is trusted unconditionally.
+   * Any books not present in the input library.xml files are removed
+   * from the library.
+   *
+   * @param paths The (utf8) paths to the `library.xml` files.
+   */
+  void reload(const Paths& paths);
+
   /**
    * Load a library content store in the string.
    *
@@ -150,8 +164,7 @@ class Manager
   uint64_t m_itemsPerPage = 0;
 
  protected:
-  kiwix::LibraryManipulator* manipulator;
-  bool mustDeleteManipulator;
+  std::shared_ptr<kiwix::LibraryManipulator> manipulator;
 
   bool readBookFromPath(const std::string& path, Book* book);
   bool parseXmlDom(const pugi::xml_document& doc,

include/meson.build

@@ -13,18 +13,9 @@ headers = [
   'search_renderer.h',
   'server.h',
   'kiwixserve.h',
-  'name_mapper.h'
+  'name_mapper.h',
+  'tools.h',
+  'version.h'
 ]
 
 install_headers(headers, subdir:'kiwix')
-
-install_headers(
-  'tools/base64.h',
-  'tools/networkTools.h',
-  'tools/otherTools.h',
-  'tools/pathTools.h',
-  'tools/regexTools.h',
-  'tools/stringTools.h',
-  subdir:'kiwix/tools'
-)
-

include/name_mapper.h

@@ -22,6 +22,8 @@
 
 #include <string>
 #include <map>
+#include <memory>
+#include <mutex>
 
 namespace kiwix
 {
@@ -31,15 +33,15 @@ class Library;
 class NameMapper {
   public:
     virtual ~NameMapper() = default;
-    virtual std::string getNameForId(const std::string& id) = 0;
-    virtual std::string getIdForName(const std::string& name) = 0;
+    virtual std::string getNameForId(const std::string& id) const = 0;
+    virtual std::string getIdForName(const std::string& name) const = 0;
 };
 
 
 class IdNameMapper : public NameMapper {
   public:
-    virtual std::string getNameForId(const std::string& id) { return id; };
-    virtual std::string getIdForName(const std::string& name) { return name; };
+    virtual std::string getNameForId(const std::string& id) const { return id; };
+    virtual std::string getIdForName(const std::string& name) const { return name; };
 };
 
 class HumanReadableNameMapper : public NameMapper {
@@ -50,11 +52,29 @@ class HumanReadableNameMapper : public NameMapper {
   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);
+    virtual std::string getNameForId(const std::string& id) const;
+    virtual std::string getIdForName(const std::string& name) const;
 };
 
+class UpdatableNameMapper : public NameMapper {
+    typedef std::shared_ptr<NameMapper> NameMapperHandle;
+  public:
+    UpdatableNameMapper(Library& library, bool withAlias);
 
+    virtual std::string getNameForId(const std::string& id) const;
+    virtual std::string getIdForName(const std::string& name) const;
+
+    void update();
+
+  private:
+    NameMapperHandle currentNameMapper() const;
+
+  private:
+    mutable std::mutex mutex;
+    Library& library;
+    NameMapperHandle nameMapper;
+    const bool withAlias;
+};
 
 }
 

include/opds_catalog.h

@@ -0,0 +1,33 @@
+/*
+ * 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.
+ */
+
+#ifndef KIWIX_OPDS_CATALOG_H
+#define KIWIX_OPDS_CATALOG_H
+
+
+#include "library.h"
+
+namespace kiwix
+{
+
+std::string getSearchUrl(const Filter& f);
+
+} // namespace kiwix
+
+#endif // KIWIX_OPDS_CATALOG_H

include/opds_dumper.h

@@ -26,9 +26,6 @@
 
 #include <pugixml.hpp>
 
-#include "tools/base64.h"
-#include "tools/pathTools.h"
-#include "tools/regexTools.h"
 #include "library.h"
 #include "reader.h"
 
@@ -51,24 +48,50 @@ 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
+   * @param partial whether the feed should include partial or complete entries
+   * @return The OPDS feed.
+   */
+  std::string dumpOPDSFeedV2(const std::vector<std::string>& bookIds, const std::string& query, bool partial) const;
+
+  /**
+   * Dump the OPDS complete entry document.
+   *
+   * @param bookId the id of the book
+   * @return The OPDS complete entry document.
+   */
+  std::string dumpOPDSCompleteEntry(const std::string& bookId) const;
+
+  /**
+   * Dump the categories OPDS feed.
+   *
+   * @return The OPDS feed.
+   */
+  std::string categoriesOPDSFeed() const;
+
+  /**
+   * Dump the languages OPDS feed.
+   *
+   * @return The OPDS feed.
+   */
+  std::string languagesOPDSFeed() 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 +100,6 @@ 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 +109,13 @@ 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

@@ -29,20 +29,52 @@
 #include <string>
 #include "common.h"
 #include "entry.h"
-#include "tools/pathTools.h"
-#include "tools/stringTools.h"
 
 using namespace std;
 
 namespace kiwix
 {
 
+/**
+ * The SuggestionItem is a helper class that contains the info about a single
+ * suggestion item.
+ */
+class SuggestionItem
+{
+  // Functions
+  public:
+    // Create a sugggestion item.
+    explicit SuggestionItem(const std::string& title, const std::string& normalizedTitle,
+                            const std::string& path, const std::string& snippet = "") :
+      title(title),
+      normalizedTitle(normalizedTitle),
+      path(path),
+      snippet(snippet) {}
+
+  public:
+    const std::string& getTitle() const           { return title;}
+    const std::string& getNormalizedTitle() const { return normalizedTitle;}
+    const std::string& getPath() const            { return path;}
+    const std::string& getSnippet() const         { return snippet;}
+
+    bool hasSnippet() const                { return !snippet.empty();}
+
+  // Data
+  private:
+    std::string title;
+    std::string normalizedTitle;
+    std::string path;
+    std::string snippet;
+};
+
 /**
  * The Reader class is the class who allow to get an entry content from a zim
  * file.
+ *
+ * Reader is now deprecated. Directly use `zim::Archive`.
  */
 
-using SuggestionsList_t = std::vector<std::vector<std::string>>;
+using SuggestionsList_t = std::vector<SuggestionItem>;
 class Reader
 {
  public:
@@ -55,7 +87,19 @@ class Reader
    *                    unsplitted path as if the file were not splitted
    *                    (.zim extesion).
    */
-  Reader(const string zimFilePath);
+  explicit DEPRECATED Reader(const string zimFilePath);
+
+   /**
+   * Create a Reader to read a zim file given by the Archive.
+   *
+   * @param archive The shared pointer to the Archive object.
+   */
+  explicit DEPRECATED Reader(const std::shared_ptr<zim::Archive> archive)
+   : Reader(archive, true) {};
+#ifndef _WIN32
+  explicit DEPRECATED Reader(int fd);
+  DEPRECATED Reader(int fd, zim::offset_type offset, zim::size_type size);
+#endif
   ~Reader() = default;
 
   /**
@@ -248,16 +292,6 @@ class Reader
    */
   string getScraper() const;
 
-  /**
-   * Get the origId of the zim file.
-   *
-   * The origId is only used in the case of patch zim file and is the Id
-   * of the original zim file.
-   *
-   * @return The origId of the zim file as specified in the zim metadata.
-   */
-  string getOrigId() const;
-
   /**
    * Get the favicon of the zim file.
    *
@@ -449,7 +483,7 @@ class Reader
   zim::Archive* getZimArchive() const;
 
  protected:
-  std::unique_ptr<zim::Archive> zimArchive;
+  std::shared_ptr<zim::Archive> zimArchive;
   std::string zimFilePath;
 
   SuggestionsList_t suggestions;
@@ -457,6 +491,15 @@ class Reader
 
  private:
   std::map<const std::string, unsigned int> parseCounterMetadata() const;
+
+  // Reader is deprecated, so we've marked the constructor as deprecated.
+  // But we still need to construct the reader (in our deprecated code)
+  // To avoid warning because we use deprecated function, we create a
+  // constructor not deprecated. The `bool marker` is unused, it sole purpose
+  // is to change the signature to have a different constructor.
+  // This one is not deprecated and we must use it in our private code.
+  Reader(const std::shared_ptr<zim::Archive> archive, bool marker);
+  friend class Library;
 };
 }
 

include/search_renderer.h

@@ -21,6 +21,8 @@
 #define KIWIX_SEARCH_RENDERER_H
 
 #include <string>
+#include <zim/search.h>
+#include "library.h"
 
 namespace kiwix
 {
@@ -34,12 +36,42 @@ class SearchRenderer
 {
  public:
   /**
-   * The default constructor.
+   * Construct a SearchRenderer from a Searcher.
    *
-   * @param humanReadableName The global zim's humanReadableName.
-   *                          Used to generate pagination links.
+   * This method is now deprecated. Construct the renderer from a
+   * `zim::SearchResultSet`
+   *
+   * @param searcher The `Searcher` to render.
+   * @param mapper The `NameMapper` to use to do the rendering.
    */
-  SearchRenderer(Searcher* searcher, NameMapper* mapper);
+  DEPRECATED SearchRenderer(Searcher* searcher, NameMapper* mapper);
+
+  /**
+   * Construct a SearchRenderer from a SearchResultSet.
+   *
+   * The constructed version of the SearchRenderer will not introduce
+   * the book name for each result. It is better to use the other constructor
+   * with a Library pointer to have a better html page.
+   *
+   * @param srs The `SearchResultSet` to render.
+   * @param mapper The `NameMapper` to use to do the rendering.
+   * @param start The start offset used for the srs.
+   * @param estimatedResultCount The estimatedResultCount of the whole search
+   */
+  SearchRenderer(zim::SearchResultSet srs, NameMapper* mapper,
+                 unsigned int start, unsigned int estimatedResultCount);
+
+  /**
+   * Construct a SearchRenderer from a SearchResultSet.
+   *
+   * @param srs The `SearchResultSet` to render.
+   * @param mapper The `NameMapper` to use to do the rendering.
+   * @param library The `Library` to use to look up book details for search results.
+   * @param start The start offset used for the srs.
+   * @param estimatedResultCount The estimatedResultCount of the whole search
+   */
+  SearchRenderer(zim::SearchResultSet srs, NameMapper* mapper, Library* library,
+                 unsigned int start, unsigned int estimatedResultCount);
 
   ~SearchRenderer();
 
@@ -74,8 +106,9 @@ class SearchRenderer
 
  protected:
   std::string beautifyInteger(const unsigned int number);
-  Searcher* mp_searcher;
+  zim::SearchResultSet m_srs;
   NameMapper* mp_nameMapper;
+  Library* mp_library;
   std::string searchContent;
   std::string searchPattern;
   std::string protocolPrefix;

include/searcher.h

@@ -27,10 +27,10 @@
 #include <cctype>
 #include <locale>
 #include <string>
+#include <memory>
 #include <vector>
-#include <vector>
-#include "tools/pathTools.h"
-#include "tools/stringTools.h"
+
+#include <zim/search.h>
 
 using namespace std;
 
@@ -48,13 +48,16 @@ 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;
+struct SuggestionInternal;
 /**
  * The Searcher class is reponsible to do different kind of search using the
  * fulltext index.
+ *
+ * The Searcher is now deprecated. Use libzim search feature.
  */
 class Searcher
 {
@@ -62,7 +65,7 @@ class Searcher
   /**
    * The default constructor.
    */
-  Searcher();
+  DEPRECATED Searcher();
 
   ~Searcher();
 
@@ -85,12 +88,12 @@ class Searcher
    *
    * @param search The search query.
    * @param resultStart the start offset of the search results (used for pagination).
-   * @param resultEnd the end offset of the search results (used for pagination).
+   * @param maxResultCount Maximum results to get from start (used for pagination).
    * @param verbose print some info on stdout if true.
    */
   void search(const std::string& search,
               unsigned int resultStart,
-              unsigned int resultEnd,
+              unsigned int maxResultCount,
               const bool verbose = false);
 
   /**
@@ -104,12 +107,12 @@ class Searcher
    * @param longitude The longitude of the center point.
    * @param distance The radius of the disc.
    * @param resultStart the start offset of the search results (used for pagination).
-   * @param resultEnd the end offset of the search results (used for pagination).
+   * @param maxResultCount Maximum number of results to get from start (used for pagination).
    * @param verbose print some info on stdout if true.
    */
   void geo_search(float latitude, float longitude, float distance,
                   unsigned int resultStart,
-                  unsigned int resultEnd,
+                  unsigned int maxResultCount,
                   const bool verbose = false);
 
   /**
@@ -142,23 +145,29 @@ class Searcher
    */
   unsigned int getEstimatedResultCount();
 
+  /**
+   * Get a SearchResultSet object for current search
+   */
+  zim::SearchResultSet getSearchResultSet();
+
   unsigned int getResultStart() { return resultStart; }
-  unsigned int getResultEnd() { return resultEnd; }
+  unsigned int getMaxResultCount() { return maxResultCount; }
 
  protected:
   std::string beautifyInteger(const unsigned int number);
   void closeIndex();
   void searchInIndex(string& search,
                      const unsigned int resultStart,
-                     const unsigned int resultEnd,
+                     const unsigned int maxResultCount,
                      const bool verbose = false);
 
   std::vector<Reader*> readers;
-  SearcherInternal* internal;
+  std::unique_ptr<SearcherInternal> internal;
+  std::unique_ptr<SuggestionInternal> suggestionInternal;
   std::string searchPattern;
   unsigned int estimatedResultCount;
   unsigned int resultStart;
-  unsigned int resultEnd;
+  unsigned int maxResultCount;
 
  private:
   void reset();

include/server.h

@@ -54,23 +54,29 @@ namespace kiwix
        void setAddress(const std::string& addr) { m_addr = addr; }
        void setPort(int port) { m_port = port; }
        void setNbThreads(int threads) { m_nbThreads = threads; }
+       void setIpConnectionLimit(int limit) { m_ipConnectionLimit = limit; }
        void setVerbose(bool verbose) { m_verbose = verbose; }
+       void setIndexTemplateString(const std::string& indexTemplateString) { m_indexTemplateString = indexTemplateString; }
        void setTaskbar(bool withTaskbar, bool withLibraryButton)
         { m_withTaskbar = withTaskbar; m_withLibraryButton = withLibraryButton; }
        void setBlockExternalLinks(bool blockExternalLinks)
         { m_blockExternalLinks = blockExternalLinks; }
-
+       int getPort();
+       std::string getAddress();
+       
      protected:
        Library* mp_library;
        NameMapper* mp_nameMapper;
        std::string m_root = "";
        std::string m_addr = "";
+       std::string m_indexTemplateString = "";
        int m_port = 80;
        int m_nbThreads = 1;
        bool m_verbose = false;
        bool m_withTaskbar = true;
        bool m_withLibraryButton = true;
        bool m_blockExternalLinks = false;
+       int m_ipConnectionLimit = 0;
        std::unique_ptr<InternalServer> mp_server;
   };
 }

include/tools.h

@@ -0,0 +1,220 @@
+/*
+ * Copyright 2021 Matthieu Gautier <mgautier@kymeria.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#ifndef KIWIX_TOOLS_H
+#define KIWIX_TOOLS_H
+
+#include <string>
+#include <vector>
+#include <map>
+
+namespace kiwix {
+
+/**
+ * Return the current directory.
+ *
+ * @return the current directory (utf8 encoded)
+ */
+std::string getCurrentDirectory();
+
+/**
+ * Return the data directory.
+ *
+ * The data directory is a directory where to put data (zim files, ...)
+ * It depends of the platform and it may be changed by user using environment variable.
+ *
+ * The resolution order is :
+ * - `KIWIX_DATA_DIR` env variable (if set).
+ * - On Windows :
+ *    . `$APPDATA/kiwix` if $APPDATA is set
+ *    . `$USERPROFILE/kiwix` if $USERPROFILE is set
+ * - Else :
+ *    . `$XDG_DATA_HOME/kiwix`if $XDG_DATA_HOME is set
+ *    . `$HOME/.local/share/kiwx` if $HOWE is set
+ * - current directory
+ *
+ * @return the path of the data directory (utf8 encoded)
+ */
+std::string getDataDirectory();
+
+/** Return the path of the executable
+ *
+ * Some application may be packaged in auto extractible archive (Appimage) and the
+ * real executable is different of the path of the archive.
+ * If `realPathOnly` is true, return the path of the real executable instead of the
+ * archive launched by the user.
+ *
+ * @param realPathOnly If we must return the real path of the executable.
+ * @return the path of the executable (utf8 encoded)
+ */
+std::string getExecutablePath(bool realPathOnly = false);
+
+/** Tell if the path is a relative path.
+ *
+ * This function is provided as a small helper. It is probably better to use native tools
+ * to manipulate paths.
+ *
+ * @param path A utf8 encoded path.
+ * @return true if the path is relative.
+ */
+bool isRelativePath(const std::string& path);
+
+/** Append a path to another one.
+ *
+ * This function is provided as a small helper. It is probably better to use native tools
+ * to manipulate paths.
+ *
+ * @param basePath the base path.
+ * @param relativePath a path to add to the base path, must be a relative path.
+ * @return The concatenation of the paths, using the right separator.
+ */
+std::string appendToDirectory(const std::string& basePath, const std::string& relativePath);
+
+/** Remove the last element of a path.
+ *
+ * This function is provided as a small helper. It is probably better to use native tools
+ * to manipulate paths.
+ *
+ * @param path a path.
+ * @return The parent directory (or empty string if none).
+ */
+std::string removeLastPathElement(const std::string& path);
+
+/** Get the last element of a path.
+ *
+ * This function is provided as a small helper. It is probably better to use native tools
+ * to manipulate paths.
+ *
+ * @param path a path.
+ * @return The base name of the path or empty string if none (ending with a separator).
+ */
+std::string getLastPathElement(const std::string& path);
+
+/** Compute the absolute path of a relative path based on another one
+ *
+ * Equivalent to appendToDirectory followed by a normalization of the path.
+ *
+ * This function is provided as a small helper. It is probably better to use native tools
+ * to manipulate paths.
+ *
+ * @param path the base path (if empty, current directory is taken).
+ * @param relativePath the relative path.
+ * @return a absolute path.
+ */
+std::string computeAbsolutePath(const std::string& path, const std::string& relativePath);
+
+/** Compute the relative path of a path relative to another one
+ *
+ * This function is provided as a small helper. It is probably better to use native tools
+ * to manipulate paths.
+ *
+ * @param path the base path.
+ * @param absolutePath the absolute path to find the relative path for.
+ * @return a relative path (pointing to absolutePath, relative to path).
+ */
+std::string computeRelativePath(const std::string& path, const std::string& absolutePath);
+
+/** Sleep the current thread.
+ *
+ * This function is provided as a small helper. It is probably better to use native tools.
+ *
+ * @param milliseconds The number of milliseconds to wait for.
+ */
+void sleep(unsigned int milliseconds);
+
+/** Split a string
+ *
+ * This function is provided as a small helper. It is probably better to use native tools.
+ *
+ * Assuming text = "foo:;bar;baz,oups;"
+ *
+ * split(text, ":;", true, true) => ["foo", ":", ";", "bar", ";", "baz,oups", ";"]
+ * split(text, ":;", true, false) => ["foo", "bar", "baz,oups"] (default)
+ * split(text, ":;", false, true) => ["foo", ":", "", ";", "bar", ";", "baz,oups", ";", ""]
+ * split(text, ":;", false, false) => ["foo", "", "bar", "baz,oups", ""]
+ *
+ * @param str The string to split.
+ * @param delims A string of potential delimiters.
+ *               Each charater in the string can be a individual delimiters.
+ * @param dropEmpty true if empty part must be dropped from the result.
+ * @param keepDelim true if delimiter must be included from the result.
+ * @return a list of part (potentially containing delimiters)
+ */
+std::vector<std::string> split(const std::string& str, const std::string& delims, bool dropEmpty=true, bool keepDelim = false);
+
+/** Convert language code from iso2 code to iso3
+ *
+ * This function is provided as a small helper. It is probably better to use native tools
+ * to manipulate locales.
+ *
+ * @param a2code a iso2 code string.
+ * @return the corresponding iso3 code.
+ * @throw std::out_of_range if iso2 code is not known.
+ */
+std::string converta2toa3(const std::string& a2code);
+
+/** Extracts content from given file.
+ *
+ * This function provides content of a file provided it's path.
+ *
+ * @param path The absolute path provided in string format.
+ * @return Content of corresponding file in string format.
+ */
+std::string getFileContent(const std::string& path);
+
+/** Checks if file exists.
+ *
+ * This function returns boolean stating if file exists.
+ *
+ * @param path The absolute path provided in string format.
+ * @return Boolean representing if file exists or not.
+ */
+bool fileExists(const std::string& path);
+
+/** Checks if file is readable.
+ *
+ * This function returns boolean stating if file is readable.
+ *
+ * @param path The absolute path provided in string format.
+ * @return Boolean representing if file is readale or not.
+ */
+bool fileReadable(const std::string& path);
+
+/** Provides mimetype from filename.
+ *
+ * This function provides mimetype from file-name.
+ *
+ * @param filename string containing filename.
+ * @return mimetype from filename in string format.
+ */
+std::string getMimeTypeForFile(const std::string& filename);
+
+/** Provides all available network interfaces
+ *
+ * This function provides the available IPv4 network interfaces
+ */
+std::map<std::string, std::string> getNetworkInterfaces();
+
+/** Provides the best IP address
+ * This function provides the best IP address from the list given by getNetworkInterfaces
+ */
+std::string getBestPublicIp();
+
+}
+#endif // KIWIX_TOOLS_H

include/version.h

@@ -0,0 +1,33 @@
+/*
+ * Copyright 2021 Emmanuel Engelhart <kelson@kiwix.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * 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_VERSION_H
+#define KIWIX_VERSION_H
+
+#include <string>
+#include <vector>
+
+namespace kiwix
+{
+  typedef std::vector<std::pair<std::string, std::string>> LibVersions;
+  LibVersions getVersions();
+  void printVersions(std::ostream& out = std::cout);
+}
+
+#endif // KIWIX_VERSION_H

meson.build

@@ -1,5 +1,5 @@
-project('kiwix-lib', 'cpp',
-  version : '10.0.0', # Also change this in android-kiwix-lib-publisher/kiwixLibAndroid/build.gradle
+project('libkiwix', 'cpp',
+  version : '10.1.0', # Also change this in android-kiwix-lib-publisher/kiwixLibAndroid/build.gradle
   license : 'GPLv3+',
   default_options : ['c_std=c11', 'cpp_std=c++11', 'werror=true'])
 
@@ -18,12 +18,12 @@ if wrapper.contains('java')
   add_languages('java')
 endif
 
-# See https://github.com/kiwix/kiwix-lib/issues/371
-if ['arm', 'mips', 'm68k', 'ppc', 'sh4'].contains(target_machine.cpu_family())
+# See https://github.com/kiwix/libkiwix/issues/371
+if ['arm', 'mips', 'm68k', 'ppc', 'sh4'].contains(host_machine.cpu_family())
   extra_libs += '-latomic'
 endif
 
-if (compiler.get_id() == 'gcc' and build_machine.system() == 'linux') or target_machine.system() == 'freebsd'
+if (compiler.get_id() == 'gcc' and build_machine.system() == 'linux') or host_machine.system() == 'freebsd'
   # C++ std::thread is implemented using pthread on linux by gcc
   thread_dep = dependency('threads')
 else
@@ -34,6 +34,7 @@ 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 = []
@@ -43,24 +44,28 @@ else
   error('Cannot found header mustache.hpp')
 endif
 
-libzim_dep = dependency('libzim', version : '>=7.0.0', static:static_deps)
+libzim_dep = dependency('libzim', version : '>=7.2.0', 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
+if host_machine.system() == 'windows' and static_deps
   add_project_arguments('-DCURL_STATICLIB', language : 'cpp')
   extra_cflags += '-DCURL_STATICLIB'
 endif
 
-all_deps = [thread_dep, libicu_dep, libzim_dep, pugixml_dep, libcurl_dep, microhttpd_dep, zlib_dep]
+if host_machine.system() == 'windows'
+  add_project_arguments('-DNOMINMAX', language: 'cpp')
+endif
+
+all_deps = [thread_dep, libicu_dep, libzim_dep, pugixml_dep, libcurl_dep, microhttpd_dep, zlib_dep, xapian_dep]
 
 inc = include_directories('include', extra_include)
 
 conf = configuration_data()
-conf.set('VERSION', '"@0@"'.format(meson.project_version()))
+conf.set('LIBKIWIX_VERSION', '"@0@"'.format(meson.project_version()))
 
 if build_machine.system() == 'windows'
      extra_link_args = ['-lshlwapi', '-lwinmm']
@@ -73,8 +78,11 @@ subdir('scripts')
 subdir('static')
 subdir('src')
 subdir('test')
+if get_option('doc')
+  subdir('docs')
+endif
 
-pkg_requires = ['libzim', 'icu-i18n', 'pugixml', 'libcurl', 'libmicrohttpd']
+pkg_requires = ['libzim', 'icu-i18n', 'pugixml', 'libcurl', 'libmicrohttpd', 'xapian-core']
 
 pkg_conf = configuration_data()
 pkg_conf.set('prefix', get_option('prefix'))

meson_options.txt

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

src/aria2.cpp

@@ -3,13 +3,15 @@
 #include "aria2.h"
 #include "xmlrpc.h"
 #include <iostream>
+#include <algorithm>
 #include <sstream>
 #include <thread>
 #include <chrono>
-#include <tools/otherTools.h>
-#include <tools/pathTools.h>
-#include <tools/stringTools.h>
-#include <downloader.h> // For AriaError
+#include "tools.h"
+#include "tools/pathTools.h"
+#include "tools/stringTools.h"
+#include "tools/otherTools.h"
+#include "downloader.h" // For AriaError
 
 #ifdef _WIN32
 # define ARIA2_CMD "aria2c.exe"
@@ -30,7 +32,7 @@ namespace kiwix {
 Aria2::Aria2():
   mp_aria(nullptr),
   m_port(42042),
-  m_secret("kiwixariarpc"),
+  m_secret(getNewRpcSecret()),
   m_curlErrorBuffer(new char[CURL_ERROR_SIZE]),
   mp_curl(nullptr)
 {
@@ -63,11 +65,12 @@ Aria2::Aria2():
     // 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());
   callCmd.push_back(download_dir.c_str());
-  if (fileExists(session_file)) {
+  if (fileReadable(session_file)) {
     callCmd.push_back(inputFile.c_str());
   }
   callCmd.push_back(session.c_str());
@@ -194,6 +197,13 @@ std::string Aria2::tellStatus(const std::string& gid, const std::vector<std::str
   return doRequest(methodCall);
 }
 
+std::string Aria2::getNewRpcSecret()
+{
+  std::string uuid = gen_uuid("");
+  uuid.erase(std::remove(uuid.begin(), uuid.end(), '-'));
+  return uuid.substr(0, 9);
+}
+
 std::vector<std::string> Aria2::tellActive()
 {
   MethodCall methodCall("aria2.tellActive", m_secret);

src/aria2.h

@@ -37,6 +37,7 @@ class Aria2
 
     std::string addUri(const std::vector<std::string>& uri, const std::vector<std::pair<std::string, std::string>>& options = {});
     std::string tellStatus(const std::string& gid, const std::vector<std::string>& statusKey);
+    static std::string getNewRpcSecret();
     std::vector<std::string> tellActive();
     std::vector<std::string> tellWaiting();
     void saveSession();

src/book.cpp

@@ -20,10 +20,16 @@
 #include "book.h"
 #include "reader.h"
 
+#include "tools.h"
 #include "tools/base64.h"
 #include "tools/regexTools.h"
 #include "tools/networkTools.h"
 #include "tools/otherTools.h"
+#include "tools/stringTools.h"
+#include "tools/pathTools.h"
+#include "tools/archiveTools.h"
+
+#include <zim/archive.h>
 
 #include <pugixml.hpp>
 
@@ -35,11 +41,17 @@ Book::Book() :
   m_readOnly(false)
 {
 }
+
 /* Destructor */
 Book::~Book()
 {
 }
 
+Book::Illustrations Book::getIllustrations() const
+{
+  return m_illustrations;
+}
+
 bool Book::update(const kiwix::Book& other)
 {
   if (m_readOnly)
@@ -48,53 +60,43 @@ bool Book::update(const kiwix::Book& other)
   if (m_id != other.m_id)
     return false;
 
-  m_readOnly = other.m_readOnly;
-  m_path = other.m_path;
-  m_pathValid = other.m_pathValid;
-  m_title = other.m_title;
-  m_description = other.m_description;
-  m_language = other.m_language;
-  m_creator = other.m_creator;
-  m_publisher = other.m_publisher;
-  m_date = other.m_date;
-  m_url = other.m_url;
-  m_name = other.m_name;
-  m_flavour = other.m_flavour;
-  m_tags = other.m_tags;
-  m_origId = other.m_origId;
-  m_articleCount = other.m_articleCount;
-  m_mediaCount = other.m_mediaCount;
-  m_size = other.m_size;
-  m_favicon = other.m_favicon;
-  m_faviconMimeType = other.m_faviconMimeType;
-  m_faviconUrl = other.m_faviconUrl;
-
-  m_downloadId = other.m_downloadId;
-
+  *this = other;
   return true;
 }
 
 void Book::update(const kiwix::Reader& reader)
 {
-  m_path = reader.getZimFilePath();
-  m_pathValid = true;
-  m_id = reader.getId();
-  m_title = reader.getTitle();
-  m_description = reader.getDescription();
-  m_language = reader.getLanguage();
-  m_creator = reader.getCreator();
-  m_publisher = reader.getPublisher();
-  m_date = reader.getDate();
-  m_name = reader.getName();
-  m_flavour = reader.getFlavour();
-  m_tags = reader.getTags();
-  m_origId = reader.getOrigId();
-  m_articleCount = reader.getArticleCount();
-  m_mediaCount = reader.getMediaCount();
-  m_size = static_cast<uint64_t>(reader.getFileSize()) << 10;
-  m_pathValid = true;
+  update(*reader.getZimArchive());
+}
 
-  reader.getFavicon(m_favicon, m_faviconMimeType);
+void Book::update(const zim::Archive& archive) {
+  m_path = archive.getFilename();
+  m_pathValid = true;
+  m_id = getArchiveId(archive);
+  m_title = getArchiveTitle(archive);
+  m_description = getMetaDescription(archive);
+  m_language = getMetaLanguage(archive);
+  m_creator = getMetaCreator(archive);
+  m_publisher = getMetaPublisher(archive);
+  m_date = getMetaDate(archive);
+  m_name = getMetaName(archive);
+  m_flavour = getMetaFlavour(archive);
+  m_tags = getMetaTags(archive);
+  m_category = getCategoryFromTags();
+  m_articleCount = archive.getArticleCount();
+  m_mediaCount = getArchiveMediaCount(archive);
+  m_size = static_cast<uint64_t>(getArchiveFileSize(archive)) << 10;
+
+  m_illustrations.clear();
+  for ( const auto illustrationSize : archive.getIllustrationSizes() ) {
+    const auto illustration = std::make_shared<Illustration>();
+    const zim::Item illustrationItem = archive.getIllustrationItem(illustrationSize);
+    illustration->width = illustration->height = illustrationSize;
+    illustration->mimeType = illustrationItem.getMimetype();
+    illustration->data = illustrationItem.getData();
+    // NOTE: illustration->url is left uninitialized
+    m_illustrations.push_back(illustration);
+  }
 }
 
 #define ATTR(name) node.attribute(name).value()
@@ -106,7 +108,7 @@ void Book::updateFromXml(const pugi::xml_node& node, const std::string& baseDir)
     path = computeAbsolutePath(baseDir, path);
   }
   m_path = path;
-  m_pathValid = fileExists(path);
+  m_pathValid = fileReadable(path);
   m_title = ATTR("title");
   m_description = ATTR("description");
   m_language = ATTR("language");
@@ -121,12 +123,19 @@ void Book::updateFromXml(const pugi::xml_node& node, const std::string& baseDir)
   m_articleCount = strtoull(ATTR("articleCount"), 0, 0);
   m_mediaCount = strtoull(ATTR("mediaCount"), 0, 0);
   m_size = strtoull(ATTR("size"), 0, 0) << 10;
-  m_favicon = base64_decode(ATTR("favicon"));
-  m_faviconMimeType = ATTR("faviconMimeType");
-  m_faviconUrl = ATTR("faviconUrl");
+  std::string favicon_mimetype = ATTR("faviconMimeType");
+  if (! favicon_mimetype.empty()) {
+    const auto favicon = std::make_shared<Illustration>();
+    favicon->data = base64_decode(ATTR("favicon"));
+    favicon->mimeType = favicon_mimetype;
+    favicon->url = ATTR("faviconUrl");
+    m_illustrations.assign(1, favicon);
+  }
   try {
     m_downloadId = ATTR("downloadId");
   } catch(...) {}
+  const auto catattr = node.attribute("category");
+  m_category = catattr.empty() ? getCategoryFromTags() : catattr.value();
 }
 #undef ATTR
 
@@ -152,10 +161,14 @@ void Book::updateFromOpds(const pugi::xml_node& node, const std::string& urlHost
   m_language = VALUE("language");
   m_creator = node.child("author").child("name").child_value();
   m_publisher = node.child("publisher").child("name").child_value();
-  m_date = fromOpdsDate(VALUE("updated"));
+  const std::string dcIssuedDate = VALUE("dc:issued");
+  m_date = dcIssuedDate.empty() ? VALUE("updated") : dcIssuedDate;
+  m_date = fromOpdsDate(m_date);
   m_name = VALUE("name");
   m_flavour = VALUE("flavour");
   m_tags = VALUE("tags");
+  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;
@@ -167,8 +180,11 @@ void Book::updateFromOpds(const pugi::xml_node& node, const std::string& urlHost
       m_size = strtoull(linkNode.attribute("length").value(), 0, 0);
     }
     if (rel == "http://opds-spec.org/image/thumbnail") {
-      m_faviconUrl = urlHost + linkNode.attribute("href").value();
-      m_faviconMimeType = linkNode.attribute("type").value();
+      const auto favicon = std::make_shared<Illustration>();
+      favicon->data.clear();
+      favicon->url = urlHost + linkNode.attribute("href").value();
+      favicon->mimeType = linkNode.attribute("type").value();
+      m_illustrations.assign(1, favicon);
     }
  }
 
@@ -179,7 +195,7 @@ std::string Book::getHumanReadableIdFromPath() const
 {
   std::string id = m_path;
   if (!id.empty()) {
-    kiwix::removeAccents(id);
+    id = kiwix::removeAccents(id);
 
 #ifdef _WIN32
     id = replaceRegex(id, "", "^.*\\\\");
@@ -201,15 +217,54 @@ void Book::setPath(const std::string& path)
    : path;
 }
 
-const std::string& Book::getFavicon() const {
-  if (m_favicon.empty() && !m_faviconUrl.empty()) {
-    try {
-      m_favicon = download(m_faviconUrl);
-    } catch(...) {
-      std::cerr << "Cannot download favicon from " << m_faviconUrl;
+const Book::Illustration Book::missingDefaultIllustration;
+
+std::shared_ptr<const Book::Illustration> Book::getIllustration(unsigned int size) const
+{
+  for ( const auto& ilPtr : m_illustrations ) {
+    if (ilPtr->width == size && ilPtr->height == size) {
+      return ilPtr;
     }
   }
-  return m_favicon;
+  throw std::runtime_error("Cannot find illustration");
+}
+
+const Book::Illustration& Book::getDefaultIllustration() const
+{
+  try {
+    return *getIllustration(48);
+  } catch (...) {
+    return missingDefaultIllustration;
+  }
+}
+
+const std::string& Book::Illustration::getData() const
+{
+  if (data.empty() && !url.empty()) {
+    const std::lock_guard<std::mutex> l(mutex);
+    if ( data.empty() ) {
+      try {
+        data = download(url);
+      } catch(...) {
+        std::cerr << "Cannot download favicon from " << url;
+      }
+    }
+  }
+  return data;
+}
+
+const std::string& Book::getFavicon() const {
+  return getDefaultIllustration().getData();
+}
+
+const std::string& Book::getFaviconUrl() const
+{
+  return getDefaultIllustration().url;
+}
+
+const std::string& Book::getFaviconMimeType() const
+{
+  return getDefaultIllustration().mimeType;
 }
 
 std::string Book::getTagStr(const std::string& tagName) const {
@@ -220,4 +275,21 @@ 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/config.h.in

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

src/entry.cpp

@@ -23,7 +23,7 @@
 namespace kiwix
 {
 
-Entry::Entry(zim::Entry entry)
+Entry::Entry(zim::Entry entry, bool _marker)
   : entry(entry)
 {
 }
@@ -53,7 +53,7 @@ Entry Entry::getRedirectEntry() const
     throw NoEntry();
   }
 
-  return entry.getRedirectEntry();
+  return Entry(entry.getRedirectEntry(), true);
 }
 
 Entry Entry::getFinalEntry() const
@@ -67,7 +67,7 @@ Entry Entry::getFinalEntry() const
   if (final_entry.isRedirect()) {
     throw NoEntry();
   }
-  return final_entry;
+  return Entry(final_entry, true);
 }
 
 }

src/kiwixserve.cpp

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

src/library.cpp

@@ -22,6 +22,7 @@
 #include "reader.h"
 #include "libxml_dumper.h"
 
+#include "tools.h"
 #include "tools/base64.h"
 #include "tools/regexTools.h"
 #include "tools/pathTools.h"
@@ -30,14 +31,67 @@
 #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);
+}
+
+bool booksReferToTheSameArchive(const Book& book1, const Book& book2)
+{
+  return book1.isPathValid()
+      && book2.isPathValid()
+      && book1.getPath() == book2.getPath();
+}
+
+} // unnamed namespace
+
+class LibraryBase::BookDB : public Xapian::WritableDatabase
+{
+public:
+  BookDB() : Xapian::WritableDatabase("", Xapian::DB_BACKEND_INMEMORY) {}
+};
+
+LibraryBase::LibraryBase()
+  : m_bookDB(new BookDB)
+{
+}
+
+LibraryBase::~LibraryBase()
+{
+}
+
+LibraryBase::LibraryBase(LibraryBase&& ) = default;
+LibraryBase& LibraryBase::operator=(LibraryBase&& ) = default;
+
 /* Constructor */
 Library::Library()
 {
 }
+
+Library::Library(Library&& other)
+  : LibraryBase(std::move(other))
+{
+}
+
+Library& Library::operator=(Library&& other)
+{
+  LibraryBase::operator=(std::move(other));
+  return *this;
+}
+
 /* Destructor */
 Library::~Library()
 {
@@ -46,24 +100,37 @@ Library::~Library()
 
 bool Library::addBook(const Book& book)
 {
+  std::lock_guard<std::mutex> lock(m_mutex);
+  ++m_revision;
   /* Try to find it */
+  updateBookDB(book);
   try {
     auto& oldbook = m_books.at(book.getId());
-    oldbook.update(book);
+    if ( ! booksReferToTheSameArchive(oldbook, book) ) {
+      dropReader(book.getId());
+    }
+    oldbook.update(book); // XXX: This may have no effect if oldbook is readonly
+                          // XXX: Then m_bookDB will become out-of-sync with
+                          // XXX: the real contents of the library.
+    oldbook.lastUpdatedRevision = m_revision;
     return false;
   } catch (std::out_of_range&) {
-    m_books[book.getId()] = book;
+    Entry& newEntry = m_books[book.getId()];
+    static_cast<Book&>(newEntry) = book;
+    newEntry.lastUpdatedRevision = m_revision;
     return true;
   }
 }
 
 void Library::addBookmark(const Bookmark& bookmark)
 {
+  std::lock_guard<std::mutex> lock(m_mutex);
   m_bookmarks.push_back(bookmark);
 }
 
 bool Library::removeBookmark(const std::string& zimId, const std::string& url)
 {
+  std::lock_guard<std::mutex> lock(m_mutex);
   for(auto it=m_bookmarks.begin(); it!=m_bookmarks.end(); it++) {
     if (it->getBookId() == zimId && it->getUrl() == url) {
       m_bookmarks.erase(it);
@@ -74,19 +141,63 @@ bool Library::removeBookmark(const std::string& zimId, const std::string& url)
 }
 
 
+void Library::dropReader(const std::string& id)
+{
+  m_readers.erase(id);
+  m_archives.erase(id);
+}
 
 bool Library::removeBookById(const std::string& id)
 {
+  std::lock_guard<std::mutex> lock(m_mutex);
+  m_bookDB->delete_document("Q" + id);
+  dropReader(id);
   return m_books.erase(id) == 1;
 }
 
-Book& Library::getBookById(const std::string& id)
+Library::Revision Library::getRevision() const
 {
+  std::lock_guard<std::mutex> lock(m_mutex);
+  return m_revision;
+}
+
+uint32_t Library::removeBooksNotUpdatedSince(LibraryRevision libraryRevision)
+{
+  BookIdCollection booksToRemove;
+  {
+    std::lock_guard<std::mutex> lock(m_mutex);
+    for ( const auto& entry : m_books) {
+      if ( entry.second.lastUpdatedRevision <= libraryRevision ) {
+        booksToRemove.push_back(entry.first);
+      }
+    }
+  }
+
+  uint32_t countOfRemovedBooks = 0;
+  for ( const auto& id : booksToRemove ) {
+    if ( removeBookById(id) )
+      ++countOfRemovedBooks;
+  }
+  return countOfRemovedBooks;
+}
+
+const Book& Library::getBookById(const std::string& id) const
+{
+  // XXX: Doesn't make sense to lock this operation since it cannot
+  // XXX: guarantee thread-safety because of its return type
   return m_books.at(id);
 }
 
-Book& Library::getBookByPath(const std::string& path)
+Book Library::getBookByIdThreadSafe(const std::string& id) const
 {
+  std::lock_guard<std::mutex> lock(m_mutex);
+  return getBookById(id);
+}
+
+const Book& Library::getBookByPath(const std::string& path) const
+{
+  // XXX: Doesn't make sense to lock this operation since it cannot
+  // XXX: guarantee thread-safety because of its return type
   for(auto& it: m_books) {
     auto& book = it.second;
     if (book.getPath() == path)
@@ -100,20 +211,40 @@ Book& Library::getBookByPath(const std::string& path)
 std::shared_ptr<Reader> Library::getReaderById(const std::string& id)
 {
   try {
+    std::lock_guard<std::mutex> lock(m_mutex);
     return m_readers.at(id);
   } catch (std::out_of_range& e) {}
 
+  const auto archive = getArchiveById(id);
+  if ( !archive )
+    return nullptr;
+
+  const shared_ptr<Reader> reader(new Reader(archive, true));
+  std::lock_guard<std::mutex> lock(m_mutex);
+  m_readers[id] = reader;
+  return reader;
+}
+
+std::shared_ptr<zim::Archive> Library::getArchiveById(const std::string& id)
+{
+  std::lock_guard<std::mutex> lock(m_mutex);
+  try {
+    return m_archives.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;
+
+  auto sptr = make_shared<zim::Archive>(book.getPath());
+  m_archives[id] = sptr;
   return sptr;
 }
 
 unsigned int Library::getBookCount(const bool localBooks,
-                                   const bool remoteBooks)
+                                   const bool remoteBooks) const
 {
+  std::lock_guard<std::mutex> lock(m_mutex);
   unsigned int result = 0;
   for (auto& pair: m_books) {
     auto& book = pair.second;
@@ -125,84 +256,96 @@ unsigned int Library::getBookCount(const bool localBooks,
   return result;
 }
 
-bool Library::writeToFile(const std::string& path)
+bool Library::writeToFile(const std::string& path) const
 {
+  const auto allBookIds = getBooksIds();
+
   auto baseDir = removeLastPathElement(path);
   LibXMLDumper dumper(this);
   dumper.setBaseDir(baseDir);
-  return writeTextFile(path, dumper.dumpLibXMLContent(getBooksIds()));
+  std::string xml;
+  {
+    std::lock_guard<std::mutex> lock(m_mutex);
+    xml = dumper.dumpLibXMLContent(allBookIds);
+  };
+  return writeTextFile(path, xml);
 }
 
-bool Library::writeBookmarksToFile(const std::string& path)
+bool Library::writeBookmarksToFile(const std::string& path) const
 {
   LibXMLDumper dumper(this);
-  return writeTextFile(path, dumper.dumpLibXMLBookmark());
+  // NOTE: LibXMLDumper::dumpLibXMLBookmark uses Library in a thread-safe way
+  const std::string xml = dumper.dumpLibXMLBookmark();
+  return writeTextFile(path, xml);
 }
 
-std::vector<std::string> Library::getBooksLanguages()
+Library::AttributeCounts Library::getBookAttributeCounts(BookStrPropMemFn p) const
 {
-  std::vector<std::string> booksLanguages;
-  std::map<std::string, bool> booksLanguagesMap;
+  std::lock_guard<std::mutex> lock(m_mutex);
+  AttributeCounts propValueCounts;
 
-  for (auto& pair: m_books) {
-    auto& book = pair.second;
-    auto& language = book.getLanguage();
-    if (booksLanguagesMap.find(language) == booksLanguagesMap.end()) {
-      if (book.getOrigId().empty()) {
-        booksLanguagesMap[language] = true;
-        booksLanguages.push_back(language);
-      }
+  for (const auto& pair: m_books) {
+    const auto& book = pair.second;
+    if (book.getOrigId().empty()) {
+      propValueCounts[(book.*p)()] += 1;
+    }
+  }
+  return propValueCounts;
+}
+
+std::vector<std::string> Library::getBookPropValueSet(BookStrPropMemFn p) const
+{
+  std::vector<std::string> result;
+  for ( const auto& kv : getBookAttributeCounts(p) ) {
+    result.push_back(kv.first);
+  }
+  return result;
+}
+
+std::vector<std::string> Library::getBooksLanguages() const
+{
+  return getBookPropValueSet(&Book::getLanguage);
+}
+
+Library::AttributeCounts Library::getBooksLanguagesWithCounts() const
+{
+  return getBookAttributeCounts(&Book::getLanguage);
+}
+
+std::vector<std::string> Library::getBooksCategories() const
+{
+  std::lock_guard<std::mutex> lock(m_mutex);
+  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 booksLanguages;
+  return std::vector<std::string>(categories.begin(), categories.end());
 }
 
-std::vector<std::string> Library::getBooksCreators()
+std::vector<std::string> Library::getBooksCreators() const
 {
-  std::vector<std::string> booksCreators;
-  std::map<std::string, bool> booksCreatorsMap;
-
-  for (auto& pair: m_books) {
-    auto& book = pair.second;
-    auto& creator = book.getCreator();
-    if (booksCreatorsMap.find(creator) == booksCreatorsMap.end()) {
-      if (book.getOrigId().empty()) {
-        booksCreatorsMap[creator] = true;
-        booksCreators.push_back(creator);
-      }
-    }
-  }
-
-  return booksCreators;
+  return getBookPropValueSet(&Book::getCreator);
 }
 
-std::vector<std::string> Library::getBooksPublishers()
+std::vector<std::string> Library::getBooksPublishers() const
 {
-  std::vector<std::string> booksPublishers;
-  std::map<std::string, bool> booksPublishersMap;
-
-  for (auto& pair:m_books) {
-    auto& book = pair.second;
-    auto& publisher = book.getPublisher();
-    if (booksPublishersMap.find(publisher) == booksPublishersMap.end()) {
-      if (book.getOrigId().empty()) {
-        booksPublishersMap[publisher] = true;
-        booksPublishers.push_back(publisher);
-      }
-    }
-  }
-
-  return booksPublishers;
+  return getBookPropValueSet(&Book::getPublisher);
 }
 
-const std::vector<kiwix::Bookmark> Library::getBookmarks(bool onlyValidBookmarks)
+const std::vector<kiwix::Bookmark> Library::getBookmarks(bool onlyValidBookmarks) const
 {
   if (!onlyValidBookmarks) {
     return m_bookmarks;
   }
   std::vector<kiwix::Bookmark> validBookmarks;
   auto booksId = getBooksIds();
+  std::lock_guard<std::mutex> lock(m_mutex);
   for(auto& bookmark:m_bookmarks) {
     if (std::find(booksId.begin(), booksId.end(), bookmark.getBookId()) != booksId.end()) {
       validBookmarks.push_back(bookmark);
@@ -211,9 +354,10 @@ const std::vector<kiwix::Bookmark> Library::getBookmarks(bool onlyValidBookmarks
   return validBookmarks;
 }
 
-std::vector<std::string> Library::getBooksIds()
+Library::BookIdCollection Library::getBooksIds() const
 {
-  std::vector<std::string> bookIds;
+  std::lock_guard<std::mutex> lock(m_mutex);
+  BookIdCollection bookIds;
 
   for (auto& pair: m_books) {
     bookIds.push_back(pair.first);
@@ -222,26 +366,199 @@ std::vector<std::string> Library::getBooksIds()
   return bookIds;
 }
 
-std::vector<std::string> Library::filter(const std::string& search)
-{
-  if (search.empty()) {
-    return getBooksIds();
-  }
 
-  return filter(Filter().query(search));
+void Library::updateBookDB(const Book& book)
+{
+  Xapian::Stem stemmer;
+  Xapian::TermGenerator indexer;
+  const std::string lang = book.getLanguage();
+  try {
+    stemmer = Xapian::Stem(iso639_3ToXapian(lang));
+    indexer.set_stemmer(stemmer);
+    indexer.set_stemming_strategy(Xapian::TermGenerator::STEM_SOME);
+  } catch (...) {}
+  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;
 }
 
 
-std::vector<std::string> Library::filter(const Filter& filter)
+Xapian::Query buildXapianQueryFromFilterQuery(const Filter& filter)
 {
-  std::vector<std::string> bookIds;
-  for(auto& pair:m_books) {
-    auto book = pair.second;
-    if(filter.accept(book)) {
-      bookIds.push_back(pair.first);
+  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;
+
+  std::lock_guard<std::mutex> lock(m_mutex);
+  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;
+}
+
+Library::BookIdCollection Library::filter(const Filter& filter) const
+{
+  BookIdCollection result;
+  const auto preliminaryResult = filterViaBookDB(filter);
+  std::lock_guard<std::mutex> lock(m_mutex);
+  for(auto id : preliminaryResult) {
+    if(filter.accept(m_books.at(id))) {
+      result.push_back(id);
     }
   }
-  return bookIds;
+  return result;
 }
 
 template<supportedListSortBy SORT>
@@ -257,13 +574,13 @@ struct KEY_TYPE<SIZE> {
 template<supportedListSortBy sort>
 class Comparator {
   private:
-    Library* lib;
-    bool     ascending;
+    const Library* const lib;
+    const bool     ascending;
 
     inline typename KEY_TYPE<sort>::TYPE get_key(const std::string& id);
 
   public:
-    Comparator(Library* lib, bool ascending) : lib(lib), ascending(ascending) {}
+    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);
@@ -303,8 +620,13 @@ std::string Comparator<PUBLISHER>::get_key(const std::string& id)
   return lib->getBookById(id).getPublisher();
 }
 
-void Library::sort(std::vector<std::string>& bookIds, supportedListSortBy sort, bool ascending)
+void Library::sort(BookIdCollection& bookIds, supportedListSortBy sort, bool ascending) const
 {
+  // NOTE: Can reimplement this method in a way that doesn't require locking
+  // NOTE: for the entire duration of the sort. Will need to obtain (under a
+  // NOTE: lock) the required atributes from the books once, and then the
+  // NOTE: sorting will run on a copy of data without locking.
+  std::lock_guard<std::mutex> lock(m_mutex);
   switch(sort) {
     case TITLE:
       std::sort(bookIds.begin(), bookIds.end(), Comparator<TITLE>(this, ascending));
@@ -327,48 +649,6 @@ void Library::sort(std::vector<std::string>& bookIds, supportedListSortBy sort,
 }
 
 
-std::vector<std::string> Library::listBooksIds(
-    int mode,
-    supportedListSortBy sortBy,
-    const std::string& search,
-    const std::string& language,
-    const std::string& creator,
-    const std::string& publisher,
-    const std::vector<std::string>& tags,
-    size_t maxSize) {
-
-  Filter _filter;
-  if (mode & LOCAL)
-    _filter.local(true);
-  if (mode & NOLOCAL)
-    _filter.local(false);
-  if (mode & VALID)
-    _filter.valid(true);
-  if (mode & NOVALID)
-    _filter.valid(false);
-  if (mode & REMOTE)
-    _filter.remote(true);
-  if (mode & NOREMOTE)
-    _filter.remote(false);
-  if (!tags.empty())
-    _filter.acceptTags(tags);
-  if (maxSize != 0)
-    _filter.maxSize(maxSize);
-  if (!language.empty())
-    _filter.lang(language);
-  if (!publisher.empty())
-    _filter.publisher(publisher);
-  if (!creator.empty())
-    _filter.creator(creator);
-  if (!search.empty())
-    _filter.query(search);
-
-  auto bookIds = filter(_filter);
-
-  sort(bookIds, sortBy, true);
-  return bookIds;
-}
-
 Filter::Filter()
   : activeFilters(0),
     _maxSize(0)
@@ -391,6 +671,7 @@ enum filterTypes {
   MAXSIZE = FLAG(11),
   QUERY = FLAG(12),
   NAME = FLAG(13),
+  CATEGORY = FLAG(14),
 };
 
 Filter& Filter::local(bool accept)
@@ -429,20 +710,27 @@ Filter& Filter::valid(bool accept)
   return *this;
 }
 
-Filter& Filter::acceptTags(std::vector<std::string> tags)
+Filter& Filter::acceptTags(const Tags& tags)
 {
   _acceptTags = tags;
   activeFilters |= ACCEPTTAGS;
   return *this;
 }
 
-Filter& Filter::rejectTags(std::vector<std::string> tags)
+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;
@@ -471,9 +759,10 @@ Filter& Filter::maxSize(size_t maxSize)
   return *this;
 }
 
-Filter& Filter::query(std::string query)
+Filter& Filter::query(std::string query, bool partial)
 {
   _query = query;
+  _queryIsPartial = partial;
   activeFilters |= QUERY;
   return *this;
 }
@@ -487,6 +776,36 @@ Filter& Filter::name(std::string name)
 
 #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();
@@ -502,40 +821,8 @@ bool Filter::accept(const Book& book) const
   FILTER(_NOREMOTE, !remote)
 
   FILTER(MAXSIZE, book.getSize() <= _maxSize)
-  FILTER(LANG, book.getLanguage() == _lang)
-  FILTER(_PUBLISHER, book.getPublisher() == _publisher)
-  FILTER(_CREATOR, book.getCreator() == _creator)
-  FILTER(NAME, book.getName() == _name)
-
-  if (ACTIVE(ACCEPTTAGS)) {
-    if (!_acceptTags.empty()) {
-      auto vBookTags = split(book.getTags(), ";");
-      std::set<std::string> sBookTags(vBookTags.begin(), vBookTags.end());
-      for (auto& t: _acceptTags) {
-        if (sBookTags.find(t) == sBookTags.end()) {
-          return false;
-        }
-      }
-    }
-  }
-  if (ACTIVE(REJECTTAGS)) {
-    if (!_rejectTags.empty()) {
-      auto vBookTags = split(book.getTags(), ";");
-      std::set<std::string> sBookTags(vBookTags.begin(), vBookTags.end());
-      for (auto& t: _rejectTags) {
-        if (sBookTags.find(t) != sBookTags.end()) {
-          return false;
-        }
-      }
-    }
-  }
-  if ( ACTIVE(QUERY)
-    && !(matchRegex(book.getTitle(), "\\Q" + _query + "\\E")
-        || matchRegex(book.getDescription(), "\\Q" + _query + "\\E")))
-    return false;
 
   return true;
-
 }
 
 }

src/libxml_dumper.cpp

@@ -20,15 +20,15 @@
 #include "libxml_dumper.h"
 #include "book.h"
 
+#include "tools.h"
 #include "tools/base64.h"
 #include "tools/stringTools.h"
 #include "tools/otherTools.h"
-#include "tools/pathTools.h"
 
 namespace kiwix
 {
 /* Constructor */
-LibXMLDumper::LibXMLDumper(Library* library)
+LibXMLDumper::LibXMLDumper(const Library* library)
   : library(library)
 {
 }
@@ -60,10 +60,13 @@ void LibXMLDumper::handleBook(Book book, pugi::xml_node root_node) {
     ADD_ATTR_NOT_EMPTY(entry_node, "name", book.getName());
     ADD_ATTR_NOT_EMPTY(entry_node, "flavour", book.getFlavour());
     ADD_ATTR_NOT_EMPTY(entry_node, "tags", book.getTags());
-    ADD_ATTR_NOT_EMPTY(entry_node, "faviconMimeType", book.getFaviconMimeType());
-    ADD_ATTR_NOT_EMPTY(entry_node, "faviconUrl", book.getFaviconUrl());
-    if (!book.getFavicon().empty())
-      ADD_ATTRIBUTE(entry_node, "favicon", base64_encode(book.getFavicon()));
+    try {
+      auto defaultIllustration = book.getIllustration(48);
+      ADD_ATTR_NOT_EMPTY(entry_node, "faviconMimeType", defaultIllustration->mimeType);
+      ADD_ATTR_NOT_EMPTY(entry_node, "faviconUrl", defaultIllustration->url);
+      if (!defaultIllustration->getData().empty())
+        ADD_ATTRIBUTE(entry_node, "favicon", base64_encode(defaultIllustration->getData()));
+    } catch(...) {}
   } else {
     ADD_ATTRIBUTE(entry_node, "origId", book.getOrigId());
   }
@@ -91,7 +94,7 @@ void LibXMLDumper::handleBookmark(Bookmark bookmark, pugi::xml_node root_node) {
   auto book_node = entry_node.append_child("book");
 
   try {
-    auto book = library->getBookById(bookmark.getBookId());
+    auto book = library->getBookByIdThreadSafe(bookmark.getBookId());
     ADD_TEXT_ENTRY(book_node, "id", book.getId());
     ADD_TEXT_ENTRY(book_node, "title", book.getTitle());
     ADD_TEXT_ENTRY(book_node, "language", book.getLanguage());

src/manager.cpp

@@ -19,34 +19,88 @@
 
 #include "manager.h"
 
+#include "tools.h"
 #include "tools/pathTools.h"
 
 #include <pugixml.hpp>
 
 namespace kiwix
 {
+
+namespace
+{
+
+struct NoDelete
+{
+  template<class T> void operator()(T*) {}
+};
+
+} // unnamed namespace
+
+////////////////////////////////////////////////////////////////////////////////
+// LibraryManipulator
+////////////////////////////////////////////////////////////////////////////////
+
+LibraryManipulator::LibraryManipulator(Library* library)
+  : library(*library)
+{}
+
+LibraryManipulator::~LibraryManipulator()
+{}
+
+bool LibraryManipulator::addBookToLibrary(const Book& book)
+{
+  const auto ret = library.addBook(book);
+  if ( ret ) {
+    bookWasAddedToLibrary(book);
+  }
+  return ret;
+}
+
+void LibraryManipulator::addBookmarkToLibrary(const Bookmark& bookmark)
+{
+  library.addBookmark(bookmark);
+  bookmarkWasAddedToLibrary(bookmark);
+}
+
+uint32_t LibraryManipulator::removeBooksNotUpdatedSince(Library::Revision rev)
+{
+  const auto n = library.removeBooksNotUpdatedSince(rev);
+  if ( n != 0 ) {
+    booksWereRemovedFromLibrary();
+  }
+  return n;
+}
+
+void LibraryManipulator::bookWasAddedToLibrary(const Book& book)
+{
+}
+
+void LibraryManipulator::bookmarkWasAddedToLibrary(const Bookmark& bookmark)
+{
+}
+
+void LibraryManipulator::booksWereRemovedFromLibrary()
+{
+}
+
+////////////////////////////////////////////////////////////////////////////////
+// Manager
+////////////////////////////////////////////////////////////////////////////////
+
 /* Constructor */
 Manager::Manager(LibraryManipulator* manipulator):
   writableLibraryPath(""),
-  manipulator(manipulator),
-  mustDeleteManipulator(false)
+  manipulator(manipulator, NoDelete())
 {
 }
 
 Manager::Manager(Library* library) :
   writableLibraryPath(""),
-  manipulator(new DefaultLibraryManipulator(library)),
-  mustDeleteManipulator(true)
+  manipulator(new LibraryManipulator(library))
 {
 }
 
-/* Destructor */
-Manager::~Manager()
-{
-  if (mustDeleteManipulator) {
-    delete manipulator;
-  }
-}
 bool Manager::parseXmlDom(const pugi::xml_document& doc,
                           bool readOnly,
                           const std::string& libraryPath,
@@ -80,7 +134,7 @@ bool Manager::readXml(const std::string& xml,
 {
   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, trustLibrary);
@@ -124,7 +178,7 @@ 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);
@@ -214,8 +268,8 @@ bool Manager::readBookFromPath(const std::string& path, kiwix::Book* book)
     tmp_path = computeAbsolutePath(getCurrentDirectory(), path);
   }
   try {
-    kiwix::Reader reader(tmp_path);
-    book->update(reader);
+    zim::Archive archive(tmp_path);
+    book->update(archive);
     book->setPathValid(true);
   } catch (const std::exception& e) {
     book->setPathValid(false);
@@ -248,4 +302,21 @@ bool Manager::readBookmarkFile(const std::string& path)
   return true;
 }
 
+void Manager::reload(const Paths& paths)
+{
+  const auto libRevision = manipulator->getLibrary().getRevision();
+  for (std::string path : paths) {
+    if (!path.empty()) {
+      if ( kiwix::isRelativePath(path) )
+        path = kiwix::computeAbsolutePath(kiwix::getCurrentDirectory(), path);
+
+      if (!readFile(path, false, true)) {
+        throw std::runtime_error("Failed to load the XML library file '" + path + "'.");
+      }
+    }
+  }
+
+  manipulator->removeBooksNotUpdatedSince(libRevision);
+}
+
 }

src/meson.build

@@ -19,13 +19,17 @@ kiwix_sources = [
   'tools/stringTools.cpp',
   'tools/networkTools.cpp',
   'tools/otherTools.cpp',
+  'tools/archiveTools.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.cpp',
+  'server/internalServer_catalog_v2.cpp',
+  'opds_catalog.cpp',
+  'version.cpp'
 ]
 kiwix_sources += lib_resources
 

src/name_mapper.cpp

@@ -51,12 +51,54 @@ HumanReadableNameMapper::HumanReadableNameMapper(kiwix::Library& library, bool w
   }
 }
 
-std::string HumanReadableNameMapper::getNameForId(const std::string& id) {
+std::string HumanReadableNameMapper::getNameForId(const std::string& id) const {
   return m_idToName.at(id);
 }
 
-std::string HumanReadableNameMapper::getIdForName(const std::string& name) {
+std::string HumanReadableNameMapper::getIdForName(const std::string& name) const {
   return m_nameToId.at(name);
 }
 
+////////////////////////////////////////////////////////////////////////////////
+// UpdatableNameMapper
+////////////////////////////////////////////////////////////////////////////////
+
+UpdatableNameMapper::UpdatableNameMapper(Library& lib, bool withAlias)
+  : library(lib)
+  , withAlias(withAlias)
+{
+  update();
+}
+
+void UpdatableNameMapper::update()
+{
+  const auto newNameMapper = new HumanReadableNameMapper(library, withAlias);
+  std::lock_guard<std::mutex> lock(mutex);
+  nameMapper.reset(newNameMapper);
+}
+
+UpdatableNameMapper::NameMapperHandle
+UpdatableNameMapper::currentNameMapper() const
+{
+  // Return a copy of the handle to the current NameMapper object. It will
+  // ensure that the object survives any call to UpdatableNameMapper::update()
+  // made before the completion of any pending operation on that object.
+  std::lock_guard<std::mutex> lock(mutex);
+  return nameMapper;
+}
+
+std::string UpdatableNameMapper::getNameForId(const std::string& id) const
+{
+  // Ensure that the current nameMapper object survives a concurrent call
+  // to UpdatableNameMapper::update()
+  return currentNameMapper()->getNameForId(id);
+}
+
+std::string UpdatableNameMapper::getIdForName(const std::string& name) const
+{
+  // Ensure that the current nameMapper object survives a concurrent call
+  // to UpdatableNameMapper::update()
+  return currentNameMapper()->getIdForName(name);
+}
+
 }

src/opds_catalog.cpp

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

src/opds_dumper.cpp

@@ -20,11 +20,16 @@
 #include "opds_dumper.h"
 #include "book.h"
 
+#include "kiwixlib-resources.h"
+#include <mustache.hpp>
+#include <unicode/locid.h>
+
+#include "tools/stringTools.h"
 #include "tools/otherTools.h"
-#include <iomanip>
 
 namespace kiwix
 {
+
 /* Constructor */
 OPDSDumper::OPDSDumper(Library* library)
   : library(library)
@@ -35,120 +40,249 @@ 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();
-}
-
-static std::string gen_date_from_yyyy_mm_dd(const std::string& date)
-{
-  std::stringstream is;
-  is << date << "T00:00::00: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, "id", "urn:uuid:"+book.getId());
-  ADD_TEXT_ENTRY(entry_node, "title", book.getTitle());
-  ADD_TEXT_ENTRY(entry_node, "summary", book.getDescription());
-  ADD_TEXT_ENTRY(entry_node, "language", book.getLanguage());
-  ADD_TEXT_ENTRY(entry_node, "updated", gen_date_from_yyyy_mm_dd(book.getDate()));
-  ADD_TEXT_ENTRY(entry_node, "name", book.getName());
-  ADD_TEXT_ENTRY(entry_node, "flavour", book.getFlavour());
-  ADD_TEXT_ENTRY(entry_node, "tags", book.getTags());
-  ADD_TEXT_ENTRY(entry_node, "articleCount", to_string(book.getArticleCount()));
-  ADD_TEXT_ENTRY(entry_node, "mediaCount", to_string(book.getMediaCount()));
-  ADD_TEXT_ENTRY(entry_node, "icon", rootLocation + "/meta?name=favicon&content=" + book.getHumanReadableIdFromPath());
-
-  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());
-
-  auto publisher_node = entry_node.append_child("publisher");
-  ADD_TEXT_ENTRY(publisher_node, "name", book.getPublisher());
-
-  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 BooksData;
+typedef kainjow::mustache::list IllustrationInfo;
 
-  ADD_TEXT_ENTRY(root_node, "id", id);
+IllustrationInfo getBookIllustrationInfo(const Book& book)
+{
+    kainjow::mustache::list illustrations;
+    if ( book.isPathValid() ) {
+      for ( const auto& illustration : book.getIllustrations() ) {
+        // For now, we are handling only sizexsize@1 illustration.
+        // So we can simply pass one size to mustache.
+        illustrations.push_back(kainjow::mustache::object{
+          {"icon_size", to_string(illustration->width)},
+          {"icon_mimetype", illustration->mimeType}
+        });
+      }
+    }
+    return illustrations;
+}
 
-  ADD_TEXT_ENTRY(root_node, "title", title);
-  ADD_TEXT_ENTRY(root_node, "updated", date);
+kainjow::mustache::object getSingleBookData(const Book& book)
+{
+    const auto bookDate = book.getDate() + "T00:00:00Z";
+    const MustacheData bookUrl = book.getUrl().empty()
+                               ? MustacheData(false)
+                               : MustacheData(book.getUrl());
+    return kainjow::mustache::object{
+      {"id", book.getId()},
+      {"name", book.getName()},
+      {"title", book.getTitle()},
+      {"description", book.getDescription()},
+      {"language", book.getLanguage()},
+      {"content_id",  urlEncode(book.getHumanReadableIdFromPath(), true)},
+      {"updated", bookDate}, // XXX: this should be the entry update datetime
+      {"book_date", bookDate},
+      {"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())},
+      {"icons", getBookIllustrationInfo(book)},
+    };
+}
 
-  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));
-  }
+std::string getSingleBookEntryXML(const Book& book, bool withXMLHeader, const std::string& rootLocation, const std::string& endpointRoot, bool partial)
+{
+  auto data = getSingleBookData(book);
+  data["with_xml_header"] = MustacheData(withXMLHeader);
+  data["dump_partial_entries"] = MustacheData(partial);
+  data["endpoint_root"] = endpointRoot;
+  data["root"] = rootLocation;
+  return render_template(RESOURCE::templates::catalog_v2_entry_xml, data);
+}
 
-  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";
-
-
-  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();
-  }
-
-  if (library) {
-    for (auto& bookId: bookIds) {
-      handleBook(library->getBookById(bookId), root_node);
+BooksData getBooksData(const Library* library, const std::vector<std::string>& bookIds, const std::string& rootLocation, const std::string& endpointRoot, bool partial)
+{
+  BooksData booksData;
+  for ( const auto& bookId : bookIds ) {
+    try {
+      const Book book = library->getBookByIdThreadSafe(bookId);
+      booksData.push_back(kainjow::mustache::object{
+          {"entry", getSingleBookEntryXML(book, false, rootLocation, endpointRoot, partial)}
+      });
+    } catch ( const std::out_of_range& ) {
+      // the book was removed from the library since its id was obtained
+      // ignore it
     }
   }
 
-  return nodeToString(root_node);
+  return booksData;
+}
+
+std::map<std::string, std::string> iso639_3 = {
+  {"atj", "atikamekw"},
+  {"azb", "آذربایجان دیلی"},
+  {"bcl", "central bikol"},
+  {"bgs", "tagabawa"},
+  {"bxr", "буряад хэлэн"},
+  {"cbk", "chavacano"},
+  {"cdo", "閩東語"},
+  {"dag", "Dagbani"},
+  {"diq", "dimli"},
+  {"dty", "डोटेली"},
+  {"eml", "emiliân-rumagnōl"},
+  {"fbs", "српскохрватски"},
+  {"ido", "ido"},
+  {"kbp", "kabɩyɛ"},
+  {"kld", "Gamilaraay"},
+  {"lbe", "лакку маз"},
+  {"lbj", "ལ་དྭགས་སྐད་"},
+  {"map", "Austronesian"},
+  {"mhr", "марий йылме"},
+  {"mnw", "ဘာသာမန်"},
+  {"myn", "mayan"},
+  {"nah", "nahuatl"},
+  {"nai", "north American Indian"},
+  {"nds", "plattdütsch"},
+  {"nrm", "bhasa narom"},
+  {"olo", "livvi"},
+  {"pih", "Pitcairn-Norfolk"},
+  {"pnb", "Western Panjabi"},
+  {"rmr", "Caló"},
+  {"rmy", "romani shib"},
+  {"roa", "romance languages"},
+  {"twi", "twi"}
+};
+
+std::once_flag fillLanguagesFlag;
+
+void fillLanguagesMap()
+{
+  for (auto icuLangPtr = icu::Locale::getISOLanguages(); *icuLangPtr != NULL; ++icuLangPtr) {
+    auto lang = *icuLangPtr;
+    const icu::Locale locale(lang);
+    icu::UnicodeString ustring;
+    locale.getDisplayLanguage(locale, ustring);
+    std::string displayLanguage;
+    ustring.toUTF8String(displayLanguage);
+    std::string iso3LangCode = locale.getISO3Language();
+    iso639_3.insert({iso3LangCode, displayLanguage});
+  }
+}
+
+std::string getLanguageSelfName(const std::string& lang) {
+  const auto itr = iso639_3.find(lang);
+  if (itr != iso639_3.end()) {
+    return itr->second;
+  }
+  return lang;
+};
+
+} // unnamed namespace
+
+string OPDSDumper::dumpOPDSFeed(const std::vector<std::string>& bookIds, const std::string& query) const
+{
+  const auto booksData = getBooksData(library, bookIds, rootLocation, "", false);
+  const kainjow::mustache::object template_data{
+     {"date", gen_date_str()},
+     {"root", rootLocation},
+     {"feed_id", gen_uuid(libraryId + "/catalog/search?"+query)},
+     {"filter", query.empty() ? MustacheData(false) : MustacheData(query)},
+     {"totalResults", to_string(m_totalResults)},
+     {"startIndex", to_string(m_startIndex)},
+     {"itemsPerPage", to_string(m_count)},
+     {"books", booksData }
+  };
+
+  return render_template(RESOURCE::templates::catalog_entries_xml, template_data);
+}
+
+string OPDSDumper::dumpOPDSFeedV2(const std::vector<std::string>& bookIds, const std::string& query, bool partial) const
+{
+  const auto endpointRoot = rootLocation + "/catalog/v2";
+  const auto booksData = getBooksData(library, bookIds, rootLocation, endpointRoot, partial);
+
+  const char* const endpoint = partial ? "/partial_entries" : "/entries";
+  const kainjow::mustache::object template_data{
+     {"date", gen_date_str()},
+     {"endpoint_root", endpointRoot},
+     {"feed_id", gen_uuid(libraryId + endpoint + "?" + 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", booksData },
+     {"dump_partial_entries", MustacheData(partial)}
+  };
+
+  return render_template(RESOURCE::templates::catalog_v2_entries_xml, template_data);
+}
+
+std::string OPDSDumper::dumpOPDSCompleteEntry(const std::string& bookId) const
+{
+  return getSingleBookEntryXML(library->getBookById(bookId), true, rootLocation, "", false);
+}
+
+std::string OPDSDumper::categoriesOPDSFeed() const
+{
+  const auto now = gen_date_str();
+  kainjow::mustache::list categoryData;
+  for ( const auto& category : library->getBooksCategories() ) {
+    const auto urlencodedCategoryName = urlEncode(category);
+    categoryData.push_back(kainjow::mustache::object{
+      {"name", category},
+      {"urlencoded_name",  urlencodedCategoryName},
+      {"updated", now},
+      {"id", gen_uuid(libraryId + "/categories/" + urlencodedCategoryName)}
+    });
+  }
+
+  return 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 }
+             }
+  );
+}
+
+std::string OPDSDumper::languagesOPDSFeed() const
+{
+  const auto now = gen_date_str();
+  kainjow::mustache::list languageData;
+  std::call_once(fillLanguagesFlag, fillLanguagesMap);
+  for ( const auto& langAndBookCount : library->getBooksLanguagesWithCounts() ) {
+    const std::string languageCode = langAndBookCount.first;
+    const int bookCount = langAndBookCount.second;
+    const auto languageSelfName = getLanguageSelfName(languageCode);
+    languageData.push_back(kainjow::mustache::object{
+      {"lang_code",  languageCode},
+      {"lang_self_name", languageSelfName},
+      {"book_count", to_string(bookCount)},
+      {"updated", now},
+      {"id", gen_uuid(libraryId + "/languages/" + languageCode)}
+    });
+  }
+
+  return render_template(
+             RESOURCE::templates::catalog_v2_languages_xml,
+             kainjow::mustache::object{
+               {"date", now},
+               {"endpoint_root", rootLocation + "/catalog/v2"},
+               {"feed_id", gen_uuid(libraryId + "/languages")},
+               {"languages", languageData }
+             }
+  );
 }
 
 }

src/reader.cpp

@@ -21,48 +21,14 @@
 #include <time.h>
 
 #include <zim/search.h>
+#include <zim/suggestion.h>
 #include <zim/item.h>
 #include <zim/error.h>
 
+#include "tools.h"
+#include "tools/stringTools.h"
 #include "tools/otherTools.h"
-
-inline char hi(char v)
-{
-  char hex[] = "0123456789abcdef";
-  return hex[(v >> 4) & 0xf];
-}
-
-inline char lo(char v)
-{
-  char hex[] = "0123456789abcdef";
-  return hex[v & 0xf];
-}
-
-std::string hexUUID(std::string in)
-{
-  std::ostringstream out;
-  for (unsigned n = 0; n < 4; ++n) {
-    out << hi(in[n]) << lo(in[n]);
-  }
-  out << '-';
-  for (unsigned n = 4; n < 6; ++n) {
-    out << hi(in[n]) << lo(in[n]);
-  }
-  out << '-';
-  for (unsigned n = 6; n < 8; ++n) {
-    out << hi(in[n]) << lo(in[n]);
-  }
-  out << '-';
-  for (unsigned n = 8; n < 10; ++n) {
-    out << hi(in[n]) << lo(in[n]);
-  }
-  out << '-';
-  for (unsigned n = 10; n < 16; ++n) {
-    out << hi(in[n]) << lo(in[n]);
-  }
-  std::string op = out.str();
-  return op;
-}
+#include "tools/archiveTools.h"
 
 namespace kiwix
 {
@@ -86,6 +52,29 @@ Reader::Reader(const string zimFilePath)
   srand(time(nullptr));
 }
 
+Reader::Reader(const std::shared_ptr<zim::Archive> archive, bool _marker)
+  : zimArchive(archive),
+    zimFilePath(archive->getFilename())
+  {}
+
+#ifndef _WIN32
+Reader::Reader(int fd)
+  :  zimArchive(new zim::Archive(fd)),
+     zimFilePath("")
+{
+  /* initialize random seed: */
+  srand(time(nullptr));
+}
+
+Reader::Reader(int fd, zim::offset_type offset, zim::size_type size)
+  :  zimArchive(new zim::Archive(fd, offset, size)),
+     zimFilePath("")
+{
+  /* initialize random seed: */
+  srand(time(nullptr));
+}
+#endif // #ifndef _WIN32
+
 zim::Archive* Reader::getZimArchive() const
 {
   return zimArchive.get();
@@ -93,12 +82,7 @@ zim::Archive* Reader::getZimArchive() const
 
 MimeCounterType Reader::parseCounterMetadata() const
 {
-  try {
-    auto counterContent = zimArchive->getMetadata("Counter");
-    return parseMimetypeCounter(counterContent);
-  } catch (zim::EntryNotFound& e) {
-    return {};
-  }
+  return kiwix::parseArchiveCounter(*zimArchive);
 }
 
 /* Get the count of articles which can be indexed/displayed */
@@ -120,19 +104,7 @@ unsigned int Reader::getArticleCount() const
 /* Get the count of medias content in the ZIM file */
 unsigned int Reader::getMediaCount() const
 {
-  std::map<const std::string, unsigned int> counterMap
-      = this->parseCounterMetadata();
-  unsigned int counter = 0;
-
-  for (auto &pair:counterMap) {
-    if (startsWith(pair.first, "image/") ||
-        startsWith(pair.first, "video/") ||
-        startsWith(pair.first, "audio/")) {
-      counter += pair.second;
-    }
-  }
-
-  return counter;
+  return kiwix::getArchiveMediaCount(*zimArchive);
 }
 
 /* Get the total of all items of a ZIM file, redirects included */
@@ -144,48 +116,26 @@ unsigned int Reader::getGlobalCount() const
 /* Return the UID of the ZIM file */
 string Reader::getId() const
 {
-  std::ostringstream s;
-  s << zimArchive->getUuid();
-  return s.str();
+  return kiwix::getArchiveId(*zimArchive);
 }
 
 Entry Reader::getRandomPage() const
 {
-  auto mainPagePath = zimArchive->getMainEntry().getPath();
-  int watchdog = 42;
-
-  while (--watchdog){
-    auto idx = (zim::size_type)((double)rand() / ((double)RAND_MAX + 1)
-                              * zimArchive->getEntryCount());
-    auto entry = zimArchive->getEntryByPath(idx);
-
-    if (entry.getPath()==mainPagePath) {
-      continue;
-    }
-    auto item = entry.getItem(true);
-    if (item.getMimetype() == "text/html") {
-      return entry;
-    }
+  try {
+    return Entry(zimArchive->getRandomEntry(), true);
+  } catch(...) {
+    throw NoEntry();
   }
-  throw NoEntry();
 }
 
 Entry Reader::getMainPage() const
 {
-  return zimArchive->getMainEntry();
+  return Entry(zimArchive->getMainEntry(), true);
 }
 
 bool Reader::getFavicon(string& content, string& mimeType) const
 {
-  try {
-    auto entry = zimArchive->getFaviconEntry();
-    auto item = entry.getItem(true);
-    content = item.getData();
-    mimeType = item.getMimetype();
-    return true;
-  } catch(zim::EntryNotFound& e) {};
-
-  return false;
+  return kiwix::getArchiveFavicon(*zimArchive, 48, content, mimeType);
 }
 
 string Reader::getZimFilePath() const
@@ -207,47 +157,32 @@ bool Reader::getMetadata(const string& name, string& value) const
 
 string Reader::getName() const
 {
-  METADATA("Name")
+  return kiwix::getMetaName(*zimArchive);
 }
 
 string Reader::getTitle() const
 {
-  string value = zimArchive->getMetadata("Title");
-  if (value.empty()) {
-    value = getLastPathElement(zimFilePath);
-    std::replace(value.begin(), value.end(), '_', ' ');
-    size_t pos = value.find(".zim");
-    value = value.substr(0, pos);
-  }
-  return value;
+  return kiwix::getArchiveTitle(*zimArchive);
 }
 
 string Reader::getCreator() const
 {
-  METADATA("Creator")
+  return kiwix::getMetaCreator(*zimArchive);
 }
 
 string Reader::getPublisher() const
 {
-  METADATA("Publisher")
+  return kiwix::getMetaPublisher(*zimArchive);
 }
 
 string Reader::getDate() const
 {
-  METADATA("Date")
+  return kiwix::getMetaDate(*zimArchive);
 }
 
 string Reader::getDescription() const
 {
-  string value;
-  this->getMetadata("Description", value);
-
-  /* Mediawiki Collection tends to use the "Subtitle" name */
-  if (value.empty()) {
-    this->getMetadata("Subtitle", value);
-  }
-
-  return value;
+  return kiwix::getMetaDescription(*zimArchive);
 }
 
 string Reader::getLongDescription() const
@@ -257,7 +192,7 @@ string Reader::getLongDescription() const
 
 string Reader::getLanguage() const
 {
-  METADATA("Language")
+  return kiwix::getMetaLanguage(*zimArchive);
 }
 
 string Reader::getLicense() const
@@ -267,13 +202,7 @@ string Reader::getLicense() const
 
 string Reader::getTags(bool original) const
 {
-  string tags_str;
-  getMetadata("Tags", tags_str);
-  if (original) {
-    return tags_str;
-  }
-  auto tags = convertTags(tags_str);
-  return join(tags, ";");
+  return kiwix::getMetaTags(*zimArchive, original);
 }
 
 
@@ -296,7 +225,7 @@ string Reader::getRelation() const
 
 string Reader::getFlavour() const
 {
-  METADATA("Flavour")
+  return kiwix::getMetaFlavour(*zimArchive);
 }
 
 string Reader::getSource() const
@@ -310,39 +239,10 @@ string Reader::getScraper() const
 }
 #undef METADATA
 
-string Reader::getOrigId() const
-{
-  string value;
-  this->getMetadata("startfileuid", value);
-  if (value.empty()) {
-    return "";
-  }
-  std::string id = value;
-  std::string origID;
-  std::string temp = "";
-  unsigned int k = 0;
-  char tempArray[16] = "";
-  for (unsigned int i = 0; i < id.size(); i++) {
-    if (id[i] == '\n') {
-      tempArray[k] = atoi(temp.c_str());
-      temp = "";
-      k++;
-    } else {
-      temp += id[i];
-    }
-  }
-  origID = hexUUID(tempArray);
-  return origID;
-}
-
 Entry Reader::getEntryFromPath(const std::string& path) const
 {
-  if (path.empty() || path == "/") {
-    return getMainPage();
-  }
-
   try {
-    return zimArchive->getEntryByPath(path);
+    return Entry(kiwix::getEntryFromPath(*zimArchive, path), true);
   } catch (zim::EntryNotFound& e) {
     throw NoEntry();
   }
@@ -356,7 +256,7 @@ Entry Reader::getEntryFromEncodedPath(const std::string& path) const
 Entry Reader::getEntryFromTitle(const std::string& title) const
 {
   try {
-    return zimArchive->getEntryByTitle(title);
+    return Entry(zimArchive->getEntryByTitle(title), true);
   } catch(zim::EntryNotFound& e) {
     throw NoEntry();
   }
@@ -426,12 +326,12 @@ bool Reader::searchSuggestions(const string& prefix,
        article is already in the suggestions list (with an other
        title) */
     bool insert = true;
-    std::vector<std::vector<std::string>>::iterator suggestionItr;
+    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) {
@@ -441,10 +341,7 @@ bool Reader::searchSuggestions(const string& prefix,
 
     /* Insert if possible */
     if (insert) {
-      std::vector<std::string> suggestion;
-      suggestion.push_back(entry.getTitle());
-      suggestion.push_back(articleFinalUrl);
-      suggestion.push_back(normalizedArticleTitle);
+      SuggestionItem suggestion(entry.getTitle(), normalizedArticleTitle, articleFinalUrl);
       results.insert(suggestionItr, suggestion);
     }
 
@@ -458,12 +355,7 @@ bool Reader::searchSuggestions(const string& prefix,
 std::vector<std::string> Reader::getTitleVariants(
     const std::string& title) const
 {
-  std::vector<std::string> variants;
-  variants.push_back(title);
-  variants.push_back(kiwix::ucFirst(title));
-  variants.push_back(kiwix::lcFirst(title));
-  variants.push_back(kiwix::toTitle(title));
-  return variants;
+  return kiwix::getTitleVariants(title);
 }
 
 
@@ -486,35 +378,36 @@ bool Reader::searchSuggestionsSmart(const string& prefix,
                                     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 suggestionSearch = zim::Search(*zimArchive);
-  suggestionSearch.set_query(prefix);
-  suggestionSearch.set_range(0, suggestionsCount);
-  suggestionSearch.set_suggestion_mode(true);
+  auto suggestionSearcher = zim::SuggestionSearcher(*zimArchive);
 
-  if (suggestionSearch.get_matches_estimated()) {
-    for (auto current = suggestionSearch.begin();
-         current != suggestionSearch.end();
-         current++) {
-      std::vector<std::string> suggestion;
-      suggestion.push_back(current->getTitle());
-      suggestion.push_back(current->getPath());
-      suggestion.push_back(kiwix::normalize(current->getTitle()));
+  if (zimArchive->hasTitleIndex()) {
+    auto suggestionSearch = suggestionSearcher.suggest(prefix);
+    const auto suggestions = suggestionSearch.getResults(0, suggestionsCount);
+    for (auto current : suggestions) {
+      SuggestionItem suggestion(current.getTitle(), kiwix::normalize(current.getTitle()),
+                                current.getPath(), current.getSnippet());
       results.push_back(suggestion);
     }
-    retVal = true;
   } else {
+    // Check some of the variants of the prefix
     for (std::vector<std::string>::iterator variantsItr = variants.begin();
          variantsItr != variants.end();
          variantsItr++) {
-      retVal = this->searchSuggestions(*variantsItr, suggestionsCount, results)
-               || retVal;
+      auto suggestionSearch = suggestionSearcher.suggest(*variantsItr);
+      for (auto current : suggestionSearch.getResults(0, suggestionsCount)) {
+        if (results.size() >= suggestionsCount) {
+          break;
+        }
+
+        SuggestionItem suggestion(current.getTitle(), kiwix::normalize(current.getTitle()),
+                                current.getPath(), current.getSnippet());
+        results.push_back(suggestion);
+      }
     }
   }
 
-  return retVal;
+  return results.size() > 0;
 }
 
 /* Get next suggestion */
@@ -522,7 +415,7 @@ 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++;
@@ -537,8 +430,8 @@ 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++;
@@ -573,7 +466,7 @@ bool Reader::isCorrupted() const
 /* Return the file size, works also for splitted files */
 unsigned int Reader::getFileSize() const
 {
-  return zimArchive->getFilesize() / 1024;
+  return kiwix::getArchiveFileSize(*zimArchive);
 }
 
 }

src/search_renderer.cpp

@@ -26,21 +26,41 @@
 #include "library.h"
 #include "name_mapper.h"
 
+#include "tools/archiveTools.h"
+
 #include <zim/search.h>
 
 #include <mustache.hpp>
 #include "kiwixlib-resources.h"
-
+#include "tools/stringTools.h"
 
 namespace kiwix
 {
 
 /* Constructor */
 SearchRenderer::SearchRenderer(Searcher* searcher, NameMapper* mapper)
-    : mp_searcher(searcher),
+    : SearchRenderer(
+        searcher->getSearchResultSet(),
+        mapper,
+        nullptr,
+        searcher->getEstimatedResultCount(),
+        searcher->getResultStart())
+{}
+
+SearchRenderer::SearchRenderer(zim::SearchResultSet srs, NameMapper* mapper,
+                      unsigned int start, unsigned int estimatedResultCount)
+    : SearchRenderer(srs, mapper, nullptr, start, estimatedResultCount)
+{}
+
+SearchRenderer::SearchRenderer(zim::SearchResultSet srs, NameMapper* mapper, Library* library,
+                      unsigned int start, unsigned int estimatedResultCount)
+    : m_srs(srs),
       mp_nameMapper(mapper),
+      mp_library(library),
       protocolPrefix("zim://"),
-      searchProtocolPrefix("search://?")
+      searchProtocolPrefix("search://?"),
+      estimatedResultCount(estimatedResultCount),
+      resultStart(start)
 {}
 
 /* Destructor */
@@ -70,31 +90,30 @@ 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())) {
+  for (auto it = m_srs.begin(); it != m_srs.end(); it++) {
     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());
-    auto readerIndex = p_result->get_readerIndex();
-    auto reader = mp_searcher->get_reader(readerIndex);
-    result.set("resultContentId", mp_nameMapper->getNameForId(reader->getId()));
+    result.set("title", it.getTitle());
+    result.set("url", it.getPath());
+    result.set("snippet", it.getSnippet());
+    std::string zim_id(it.getZimId());
+    result.set("resultContentId", mp_nameMapper->getNameForId(zim_id));
+    if (!mp_library) {
+      result.set("bookTitle", kainjow::mustache::data(false));
+    } else {
+      result.set("bookTitle", mp_library->getBookById(zim_id).getTitle());
+    }
 
-    if (p_result->get_wordCount() >= 0) {
-      result.set("wordCount", kiwix::beautifyInteger(p_result->get_wordCount()));
+    if (it.getWordCount() >= 0) {
+      result.set("wordCount", kiwix::beautifyInteger(it.getWordCount()));
     }
 
     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;
@@ -148,4 +167,4 @@ std::string SearchRenderer::getHtml()
   return ss.str();
 }
 
-}
+}

src/searcher.cpp

@@ -18,14 +18,16 @@
  */
 
 
-#include <cmath>
 
 #include "searcher.h"
 #include "reader.h"
 
 #include <zim/search.h>
+#include <zim/suggestion.h>
 
 #include <mustache.hpp>
+#include <cmath>
+#include "tools/stringTools.h"
 #include "kiwixlib-resources.h"
 
 #define MAX_SEARCH_LEN 140
@@ -35,7 +37,8 @@ namespace kiwix
 class _Result : public Result
 {
  public:
-  _Result(zim::Search::iterator& iterator);
+  _Result(zim::SearchResultSet::iterator iterator);
+  _Result(SuggestionItem suggestionItem);
   virtual ~_Result(){};
 
   virtual std::string get_url();
@@ -45,32 +48,38 @@ 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;
+  SuggestionItem suggestionItem;
+  bool isSuggestion;
 };
 
-struct SearcherInternal {
-  const zim::Search* _search;
-  zim::Search::iterator current_iterator;
-
-  SearcherInternal() : _search(NULL) {}
-  ~SearcherInternal()
+struct SearcherInternal : zim::SearchResultSet {
+  explicit SearcherInternal(const zim::SearchResultSet& srs)
+    : zim::SearchResultSet(srs)
+    , current_iterator(srs.begin())
   {
-    if (_search != NULL) {
-      delete _search;
-    }
   }
+
+  zim::SearchResultSet::iterator current_iterator;
+};
+
+struct SuggestionInternal : zim::SuggestionResultSet {
+  explicit SuggestionInternal(const zim::SuggestionResultSet& srs)
+    : zim::SuggestionResultSet(srs),
+      currentIterator(srs.begin()) {}
+
+  zim::SuggestionResultSet::iterator currentIterator;
 };
 
 /* Constructor */
 Searcher::Searcher()
-    : internal(new SearcherInternal()),
-      searchPattern(""),
+    : searchPattern(""),
       estimatedResultCount(0),
       resultStart(0),
-      resultEnd(0)
+      maxResultCount(0)
 {
   loadICUExternalTables();
 }
@@ -78,7 +87,6 @@ Searcher::Searcher()
 /* Destructor */
 Searcher::~Searcher()
 {
-  delete internal;
 }
 
 bool Searcher::add_reader(Reader* reader)
@@ -86,6 +94,12 @@ bool Searcher::add_reader(Reader* reader)
   if (!reader->hasFulltextIndex()) {
       return false;
   }
+
+  for ( const Reader* const existing_reader : readers ) {
+    if ( existing_reader->getZimArchive()->getUuid() == reader->getZimArchive()->getUuid() )
+      return false;
+  }
+
   this->readers.push_back(reader);
   return true;
 }
@@ -99,7 +113,7 @@ Reader* Searcher::get_reader(int readerIndex)
 /* Search strings in the database */
 void Searcher::search(const std::string& search,
                       unsigned int resultStart,
-                      unsigned int resultEnd,
+                      unsigned int maxResultCount,
                       const bool verbose)
 {
   this->reset();
@@ -110,9 +124,9 @@ void Searcher::search(const std::string& search,
 
   this->searchPattern = search;
   this->resultStart = resultStart;
-  this->resultEnd = resultEnd;
+  this->maxResultCount = maxResultCount;
   /* Try to find results */
-  if (resultStart != resultEnd) {
+  if (maxResultCount != 0) {
     /* Perform the search */
     string unaccentedSearch = removeAccents(search);
     std::vector<zim::Archive> archives;
@@ -122,13 +136,13 @@ void Searcher::search(const std::string& search,
           archives.push_back(*(*current)->getZimArchive());
       }
     }
-    zim::Search* search = new zim::Search(archives);
-    search->set_verbose(verbose);
-    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);
+    searcher.setVerbose(verbose);
+    zim::Query query;
+    query.setQuery(unaccentedSearch);
+    zim::Search search = searcher.search(query);
+    internal.reset(new SearcherInternal(search.getResults(resultStart, maxResultCount)));
+    this->estimatedResultCount = search.getEstimatedMatches();
   }
 
   return;
@@ -137,7 +151,7 @@ void Searcher::search(const std::string& search,
 
 void Searcher::geo_search(float latitude, float longitude, float distance,
                           unsigned int resultStart,
-                          unsigned int resultEnd,
+                          unsigned int maxResultCount,
                           const bool verbose)
 {
   this->reset();
@@ -151,10 +165,10 @@ void Searcher::geo_search(float latitude, float longitude, float distance,
   oss << "Articles located less than " << distance << " meters of " << latitude << ";" << longitude;
   this->searchPattern = oss.str();
   this->resultStart = resultStart;
-  this->resultEnd = resultEnd;
+  this->maxResultCount = maxResultCount;
 
   /* Try to find results */
-  if (resultStart == resultEnd) {
+  if (maxResultCount == 0) {
     return;
   }
 
@@ -163,31 +177,41 @@ void Searcher::geo_search(float latitude, float longitude, float distance,
        current++) {
     archives.push_back(*(*current)->getZimArchive());
   }
-  zim::Search* search = new zim::Search(archives);
-  search->set_verbose(verbose);
-  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);
+  searcher.setVerbose(verbose);
+  zim::Query query;
+  query.setQuery("");
+  query.setGeorange(latitude, longitude, distance);
+  zim::Search search = searcher.search(query);
+  internal.reset(new SearcherInternal(search.getResults(resultStart, maxResultCount)));
+  this->estimatedResultCount = search.getEstimatedMatches();
 }
 
 
 void Searcher::restart_search()
 {
-  if (internal->_search) {
-    internal->current_iterator = internal->_search->begin();
+  if (internal.get()) {
+    internal->current_iterator = internal->begin();
   }
 }
 
 Result* Searcher::getNextResult()
 {
-  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;
+  } else if (suggestionInternal.get() &&
+             suggestionInternal->currentIterator != suggestionInternal->end()) {
+    SuggestionItem item(
+                        suggestionInternal->currentIterator->getTitle(),
+                        normalize(suggestionInternal->currentIterator->getTitle()),
+                        suggestionInternal->currentIterator->getPath(),
+                        suggestionInternal->currentIterator->getSnippet()
+                      );
+    Result* result = new _Result(item);
+    suggestionInternal->currentIterator++;
+    return result;
   }
   return NULL;
 }
@@ -210,22 +234,16 @@ void Searcher::suggestions(std::string& searchPattern, const bool verbose)
 
   this->searchPattern = searchPattern;
   this->resultStart = 0;
-  this->resultEnd = 10;
+  this->maxResultCount = 10;
   string unaccentedSearch = removeAccents(searchPattern);
 
-  std::vector<zim::Archive> archives;
-  for (auto current = this->readers.begin(); current != this->readers.end();
-       current++) {
-    archives.push_back(*(*current)->getZimArchive());
-  }
-  zim::Search* search = new zim::Search(archives);
-  search->set_verbose(verbose);
-  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();
+  // Multizim suggestion is not supported as of now! taking only one archive
+  zim::Archive archive = *(*this->readers.begin())->getZimArchive();
+  zim::SuggestionSearcher searcher(archive);
+  searcher.setVerbose(verbose);
+  zim::SuggestionSearch search = searcher.suggest(searchPattern);
+  suggestionInternal.reset(new SuggestionInternal(search.getResults(resultStart, maxResultCount)));
+  this->estimatedResultCount = search.getEstimatedMatches();
 }
 
 /* Return the result count estimation */
@@ -234,42 +252,78 @@ unsigned int Searcher::getEstimatedResultCount()
   return this->estimatedResultCount;
 }
 
-_Result::_Result(zim::Search::iterator& iterator)
-    : iterator(iterator)
+zim::SearchResultSet Searcher::getSearchResultSet()
 {
+  return *(this->internal);
 }
 
+_Result::_Result(zim::SearchResultSet::iterator iterator)
+    : iterator(iterator),
+      suggestionItem("", "", ""),
+      isSuggestion(false)
+{}
+
+_Result::_Result(SuggestionItem item)
+    : iterator(),
+      suggestionItem(item.getTitle(), item.getNormalizedTitle(), item.getPath(), item.getSnippet()),
+      isSuggestion(true)
+{}
+
 std::string _Result::get_url()
 {
-  return iterator.get_url();
+  if (isSuggestion) {
+    return suggestionItem.getPath();
+  }
+  return iterator.getPath();
 }
 std::string _Result::get_title()
 {
-  return iterator.get_title();
+  if (isSuggestion) {
+    return suggestionItem.getTitle();
+  }
+  return iterator.getTitle();
 }
 int _Result::get_score()
 {
-  return iterator.get_score();
+  if (isSuggestion) {
+    return 0;
+  }
+  return iterator.getScore();
 }
 std::string _Result::get_snippet()
 {
-  return iterator.get_snippet();
+  if (isSuggestion) {
+    return suggestionItem.getSnippet();
+  }
+  return iterator.getSnippet();
 }
 std::string _Result::get_content()
 {
+  if (isSuggestion) return "";
   return iterator->getItem(true).getData();
 }
 int _Result::get_size()
 {
-  return iterator.get_size();
+  if (isSuggestion) {
+    return 0;
+  }
+  return iterator.getSize();
 }
 int _Result::get_wordCount()
 {
-  return iterator.get_wordCount();
+  if (isSuggestion) {
+    return 0;
+  }
+  return iterator.getWordCount();
 }
-int _Result::get_readerIndex()
+std::string _Result::get_zimId()
 {
-  return iterator.get_fileIndex();
+  if (isSuggestion) {
+    return "";
+  }
+  std::ostringstream s;
+  s << iterator.getZimId();
+  return s.str();
 }
 
 

src/server.cpp

@@ -48,7 +48,9 @@ bool Server::start() {
     m_verbose,
     m_withTaskbar,
     m_withLibraryButton,
-    m_blockExternalLinks));
+    m_blockExternalLinks,
+    m_indexTemplateString,
+    m_ipConnectionLimit));
   return mp_server->start();
 }
 
@@ -70,4 +72,14 @@ void Server::setRoot(const std::string& root)
   }
 }
 
+int Server::getPort()
+{
+  return mp_server->getPort();
+}
+
+std::string Server::getAddress()
+{
+  return mp_server->getAddress();
+}
+
 }

src/server/internalServer.cpp

@@ -43,10 +43,12 @@ extern "C" {
 #include "microhttpd_wrapper.h"
 }
 
-#include "tools/otherTools.h"
+#include "tools.h"
 #include "tools/pathTools.h"
 #include "tools/regexTools.h"
 #include "tools/stringTools.h"
+#include "tools/archiveTools.h"
+#include "tools/networkTools.h"
 #include "library.h"
 #include "name_mapper.h"
 #include "entry.h"
@@ -55,6 +57,9 @@ extern "C" {
 #include "opds_dumper.h"
 
 #include <zim/uuid.h>
+#include <zim/error.h>
+#include <zim/entry.h>
+#include <zim/item.h>
 
 #include <mustache.hpp>
 
@@ -73,9 +78,37 @@ extern "C" {
 
 #define MAX_SEARCH_LEN 140
 #define KIWIX_MIN_CONTENT_SIZE_TO_DEFLATE 100
+#define DEFAULT_CACHE_SIZE 2
 
 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;
+}
+
+// Returns the value of env var `name` if found, otherwise returns defaultVal
+unsigned int getCacheLength(const char* name, unsigned int defaultVal) {
+  try {
+    const char* envString = std::getenv(name);
+    if (envString == nullptr) {
+      throw std::runtime_error("Environment variable not set");
+    }
+    return extractFromString<unsigned int>(envString);
+  } catch (...) {}
+
+  return defaultVal;
+}
+} // unnamed namespace
+
 static IdNameMapper defaultNameMapper;
 
 static MHD_Result staticHandlerCallback(void* cls,
@@ -97,18 +130,25 @@ InternalServer::InternalServer(Library* library,
                                bool verbose,
                                bool withTaskbar,
                                bool withLibraryButton,
-                               bool blockExternalLinks) :
+                               bool blockExternalLinks,
+                               std::string indexTemplateString,
+                               int ipConnectionLimit) :
   m_addr(addr),
   m_port(port),
-  m_root(root),
+  m_root(normalizeRootUrl(root)),
   m_nbThreads(nbThreads),
   m_verbose(verbose),
   m_withTaskbar(withTaskbar),
   m_withLibraryButton(withLibraryButton),
   m_blockExternalLinks(blockExternalLinks),
+  m_indexTemplateString(indexTemplateString.empty() ? RESOURCE::templates::index_html : indexTemplateString),
+  m_ipConnectionLimit(ipConnectionLimit),
   mp_daemon(nullptr),
   mp_library(library),
-  mp_nameMapper(nameMapper ? nameMapper : &defaultNameMapper)
+  mp_nameMapper(nameMapper ? nameMapper : &defaultNameMapper),
+  searcherCache(getCacheLength("SEARCHER_CACHE_SIZE", std::max((unsigned int) (mp_library->getBookCount(true, true)*0.1), 1U))),
+  searchCache(getCacheLength("SEARCH_CACHE_SIZE", DEFAULT_CACHE_SIZE)),
+  suggestionSearcherCache(getCacheLength("SUGGESTION_SEARCHER_CACHE_SIZE", std::max((unsigned int) (mp_library->getBookCount(true, true)*0.1), 1U)))
 {}
 
 bool InternalServer::start() {
@@ -120,21 +160,21 @@ bool InternalServer::start() {
   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)
+    if (0 != INADDR_ANY) {
       sockAddr.sin_addr.s_addr = htonl(INADDR_ANY);
+    }
+    m_addr = kiwix::getBestPublicIp();
   } 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,
@@ -143,6 +183,7 @@ bool InternalServer::start() {
                             this,
                             MHD_OPTION_SOCK_ADDR, &sockAddr,
                             MHD_OPTION_THREAD_POOL_SIZE, m_nbThreads,
+                            MHD_OPTION_PER_IP_CONNECTION_LIMIT, m_ipConnectionLimit,
                             MHD_OPTION_END);
   if (mp_daemon == nullptr) {
     std::cerr << "Unable to instantiate the HTTP daemon. The port " << m_port
@@ -153,6 +194,7 @@ bool InternalServer::start() {
   }
   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;
 }
 
@@ -237,20 +279,20 @@ std::unique_ptr<Response> InternalServer::handle_request(const RequestContext& r
 {
   try {
     if (! request.is_valid_url())
-      return Response::build_404(*this, request, "");
+      return Response::build_404(*this, request.get_full_url(), "", "");
 
     const ETag etag = get_matching_if_none_match_etag(request);
     if ( etag )
       return Response::build_304(*this, etag);
 
-    if (kiwix::startsWith(request.get_url(), "/skin/"))
+    if (startsWith(request.get_url(), "/skin/"))
       return handle_skin(request);
 
-    if (startsWith(request.get_url(), "/catalog"))
+    if (startsWith(request.get_url(), "/catalog/"))
       return handle_catalog(request);
 
-    if (request.get_url() == "/meta")
-      return handle_meta(request);
+    if (startsWith(request.get_url(), "/raw/"))
+      return handle_raw(request);
 
     if (request.get_url() == "/search")
       return handle_search(request);
@@ -281,27 +323,6 @@ MustacheData InternalServer::get_default_data() const
   return data;
 }
 
-MustacheData InternalServer::homepage_data() const
-{
-  auto data = get_default_data();
-
-  MustacheData books{MustacheData::type::list};
-  for (auto& bookId: mp_library->filter(kiwix::Filter().local(true).valid(true))) {
-    auto& currentBook = mp_library->getBookById(bookId);
-
-    MustacheData book;
-    book.set("name", mp_nameMapper->getNameForId(bookId));
-    book.set("title", currentBook.getTitle());
-    book.set("description", currentBook.getDescription());
-    book.set("articleCount", beautifyInteger(currentBook.getArticleCount()));
-    book.set("mediaCount", beautifyInteger(currentBook.getMediaCount()));
-    books.push_back(book);
-  }
-
-  data.set("books", books);
-  return data;
-}
-
 bool InternalServer::etag_not_needed(const RequestContext& request) const
 {
   const std::string url = request.get_url();
@@ -325,109 +346,110 @@ InternalServer::get_matching_if_none_match_etag(const RequestContext& r) const
 
 std::unique_ptr<Response> InternalServer::build_homepage(const RequestContext& request)
 {
-  return ContentResponse::build(*this, RESOURCE::templates::index_html, homepage_data(), "text/html; charset=utf-8");
+  return ContentResponse::build(*this, m_indexTemplateString, get_default_data(), "text/html; charset=utf-8", true);
 }
 
-std::unique_ptr<Response> InternalServer::handle_meta(const RequestContext& request)
+/**
+ * Archive and Zim handlers begin
+ **/
+
+SuggestionsList_t getSuggestions(SuggestionSearcherCache& cache, const zim::Archive* const archive,
+                  const std::string& bookId, const std::string& queryString, int start, int suggestionCount)
 {
-  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);
-  }
+  SuggestionsList_t suggestions;
+  std::shared_ptr<zim::SuggestionSearcher> searcher;
+  searcher = cache.getOrPut(bookId, [=](){ return make_shared<zim::SuggestionSearcher>(*archive); });
 
-  if (reader == nullptr) {
-    return Response::build_404(*this, request, bookName);
-  }
+  if (archive->hasTitleIndex()) {
+    auto search = searcher->suggest(queryString);
+    auto srs = search.getResults(start, suggestionCount);
 
-  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);
+    for (auto it : srs) {
+      SuggestionItem suggestion(it.getTitle(), kiwix::normalize(it.getTitle()),
+                                it.getPath(), it.getSnippet());
+      suggestions.push_back(suggestion);
+    }
   } else {
-    return Response::build_404(*this, request, bookName);
+    // TODO: This case should be handled by libzim
+    std::vector<std::string> variants = getTitleVariants(queryString);
+    int currCount = 0;
+    for (auto it = variants.begin(); it != variants.end() && currCount < suggestionCount; it++) {
+      auto search = searcher->suggest(queryString);
+      auto srs = search.getResults(0, suggestionCount);
+      for (auto it : srs) {
+        SuggestionItem suggestion(it.getTitle(), kiwix::normalize(it.getTitle()),
+                                  it.getPath());
+        suggestions.push_back(suggestion);
+        currCount++;
+      }
+    }
   }
-
-  auto response = ContentResponse::build(*this, content, mimeType);
-  response->set_cacheable();
-  return std::move(response);
+  return suggestions;
 }
 
+
 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;
+  std::string bookName, bookId;
+  std::shared_ptr<zim::Archive> archive;
   try {
     bookName = request.get_argument("content");
     bookId = mp_nameMapper->getIdForName(bookName);
-    term = request.get_argument("term");
-    reader = mp_library->getReaderById(bookId);
+    archive = mp_library->getArchiveById(bookId);
   } catch (const std::out_of_range&) {
-    return Response::build_404(*this, request, bookName);
+    // error handled by the archive == nullptr check below
+  }
+
+  if (archive == nullptr) {
+    const std::string error_details = "No such book: " + bookName;
+    return Response::build_404(*this, "", bookName, "", error_details);
+  }
+
+  const auto queryString = request.get_optional_param("term", std::string());
+  const auto start = request.get_optional_param<unsigned int>("start", 0);
+  unsigned int count = request.get_optional_param<unsigned int>("count", 10);
+  if (count == 0) {
+    count = 10;
   }
 
   if (m_verbose.load()) {
-    printf("Searching suggestions for: \"%s\"\n", term.c_str());
+    printf("Searching suggestions for: \"%s\"\n", queryString.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[0]);
-      result.set("value", suggestion[0]);
-      result.set("first", first);
-      first = false;
-      results.push_back(result);
-      suggestionCount++;
+
+  /* Get the suggestions */
+  SuggestionsList_t suggestions = getSuggestions(suggestionSearcherCache, archive.get(),
+                                                  bookId, queryString, start, count);
+  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);
   }
 
+
   /* Propose the fulltext search if possible */
-  if (reader->hasFulltextIndex()) {
+  if (archive->hasFulltextIndex()) {
     MustacheData result;
-    result.set("label", "containing '" + term + "'...");
-    result.set("value", term + " ");
+    result.set("label", "containing '" + queryString + "'...");
+    result.set("value", queryString + " ");
+    result.set("kind", "pattern");
     result.set("first", first);
     results.push_back(result);
   }
@@ -454,7 +476,7 @@ std::unique_ptr<Response> InternalServer::handle_skin(const RequestContext& requ
     response->set_cacheable();
     return std::move(response);
   } catch (const ResourceNotFound& e) {
-    return Response::build_404(*this, request, "");
+    return Response::build_404(*this, request.get_full_url(), "", "");
   }
 }
 
@@ -464,13 +486,6 @@ std::unique_ptr<Response> InternalServer::handle_search(const RequestContext& re
     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");
@@ -489,54 +504,38 @@ std::unique_ptr<Response> InternalServer::handle_search(const RequestContext& re
   } catch(const std::out_of_range&) {}
     catch(const std::invalid_argument&) {}
 
-  std::shared_ptr<Reader> reader(nullptr);
+  std::string bookName, bookId;
+  std::shared_ptr<zim::Archive> archive;
   try {
-    reader = mp_library->getReaderById(bookId);
+    bookName = request.get_argument("content");
+    bookId = mp_nameMapper->getIdForName(bookName);
+    archive = mp_library->getArchiveById(bookId);
   } catch (const std::out_of_range&) {}
 
-  /* Try first to load directly the article */
-  if (reader != nullptr && !patternString.empty()) {
-    std::string patternCorrespondingUrl;
-    auto variants = reader->getTitleVariants(patternString);
-    auto variantsItr = variants.begin();
-
-    while (patternCorrespondingUrl.empty() && variantsItr != variants.end()) {
-      try {
-        auto entry = reader->getEntryFromTitle(*variantsItr);
-        entry = entry.getFinalEntry();
-        patternCorrespondingUrl = entry.getPath();
-        break;
-      } catch(kiwix::NoEntry& e) {
-        variantsItr++;
-      }
-    }
-
-    /* If article found then redirect directly to it */
-    if (!patternCorrespondingUrl.empty()) {
-      auto redirectUrl = m_root + "/" + bookName + "/" + patternCorrespondingUrl;
-      return Response::build_redirect(*this, redirectUrl);
-    }
-  }
-
   /* Make the search */
-  if ( (!reader && !bookName.empty())
+  if ( (!archive && !bookName.empty())
     || (patternString.empty() && ! has_geo_query) ) {
     auto data = get_default_data();
     data.set("pattern", encodeDiples(patternString));
+    data.set("root", m_root);
     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_taskbar(bookName, archive ? getArchiveTitle(*archive) : "");
     response->set_code(MHD_HTTP_NOT_FOUND);
     return std::move(response);
   }
 
-  Searcher searcher;
-  if (reader) {
-    searcher.add_reader(reader.get());
+  std::shared_ptr<zim::Searcher> searcher;
+  if (archive) {
+    searcher = searcherCache.getOrPut(bookId, [=](){ return std::make_shared<zim::Searcher>(*archive);});
   } 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 currentArchive = mp_library->getArchiveById(bookId);
+      if (currentArchive) {
+        if (! searcher) {
+          searcher = std::make_shared<zim::Searcher>(*currentArchive);
+        } else {
+          searcher->addArchive(*currentArchive);
+        }
       }
     }
   }
@@ -557,30 +556,42 @@ std::unique_ptr<Response> InternalServer::handle_search(const RequestContext& re
     pageLength = 25;
   }
 
-  auto end = start + pageLength;
-
   /* Get the results */
+  std::string queryString;
   try {
+    zim::Query query;
     if (patternString.empty()) {
-      searcher.geo_search(latitude, longitude, distance,
-                           start, end, m_verbose.load());
+      // Execute geo-search
+      if (m_verbose.load()) {
+        cout << "Performing geo query `" << distance << "&(" << latitude << ";" << longitude << ")'" << endl;
+      }
+
+      query.setQuery("");
+      queryString = "GEO:" + to_string(latitude) + to_string(longitude) + to_string(distance);
+      query.setGeorange(latitude, longitude, distance);
     } else {
-      searcher.search(patternString,
-                       start, end, m_verbose.load());
+      // Execute Ft search
+      if (m_verbose.load()) {
+          cout << "Performing query `" << patternString << "'" << endl;
+      }
+
+      queryString = "FT:" + removeAccents(patternString);
+      query.setQuery(queryString);
     }
-    SearchRenderer renderer(&searcher, mp_nameMapper);
+    queryString = bookId + queryString;
+
+    std::shared_ptr<zim::Search> search;
+    search = searchCache.getOrPut(queryString, [=](){ return make_shared<zim::Search>(searcher->search(query));});
+
+    SearchRenderer renderer(search->getResults(start, pageLength), mp_nameMapper, mp_library, start,
+                            search->getEstimatedMatches());
     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() : "");
-    //changing status code if no result obtained
-    if(searcher.getEstimatedResultCount() == 0)
-    {
-      response->set_code(MHD_HTTP_NO_CONTENT);
-    }
+    response->set_taskbar(bookName, archive ? getArchiveTitle(*archive) : "");
 
     return std::move(response);
   } catch (const std::exception& e) {
@@ -596,25 +607,26 @@ std::unique_ptr<Response> InternalServer::handle_random(const RequestContext& re
   }
 
   std::string bookName;
-  std::string bookId;
-  std::shared_ptr<Reader> reader;
+  std::shared_ptr<zim::Archive> archive;
   try {
     bookName = request.get_argument("content");
-    bookId = mp_nameMapper->getIdForName(bookName);
-    reader = mp_library->getReaderById(bookId);
+    const std::string bookId = mp_nameMapper->getIdForName(bookName);
+    archive = mp_library->getArchiveById(bookId);
   } catch (const std::out_of_range&) {
-    return Response::build_404(*this, request, bookName);
+    // error handled by the archive == nullptr check below
   }
 
-  if (reader == nullptr) {
-    return Response::build_404(*this, request, bookName);
+  if (archive == nullptr) {
+    const std::string error_details = "No such book: " + bookName;
+    return Response::build_404(*this, "", bookName, "", error_details);
   }
 
   try {
-    auto entry = reader->getRandomPage();
-    return build_redirect(bookName, entry.getFinalEntry());
-  } catch(kiwix::NoEntry& e) {
-    return Response::build_404(*this, request, bookName);
+    auto entry = archive->getRandomEntry();
+    return build_redirect(bookName, getFinalItem(*archive, entry));
+  } catch(zim::EntryNotFound& e) {
+    const std::string error_details = "Oops! Failed to pick a random article :(";
+    return Response::build_404(*this, "", bookName, getArchiveTitle(*archive), error_details);
   }
 }
 
@@ -626,7 +638,7 @@ std::unique_ptr<Response> InternalServer::handle_captured_external(const Request
   } catch (const std::out_of_range& e) {}
 
   if (source.empty())
-    return Response::build_404(*this, request, "");
+    return Response::build_404(*this, request.get_full_url(), "", "");
 
   auto data = get_default_data();
   data.set("source", source);
@@ -645,11 +657,15 @@ std::unique_ptr<Response> InternalServer::handle_catalog(const RequestContext& r
     host = request.get_header("Host");
     url  = request.get_url_part(1);
   } catch (const std::out_of_range&) {
-    return Response::build_404(*this, request, "");
+    return Response::build_404(*this, request.get_full_url(), "", "");
+  }
+
+  if (url == "v2") {
+    return handle_catalog_v2(request);
   }
 
   if (url != "searchdescription.xml" && url != "root.xml" && url != "search") {
-    return Response::build_404(*this, request, "");
+    return Response::build_404(*this, request.get_full_url(), "", "");
   }
 
   if (url == "searchdescription.xml") {
@@ -658,62 +674,80 @@ std::unique_ptr<Response> InternalServer::handle_catalog(const RequestContext& r
   }
 
   zim::Uuid uuid;
-  kiwix::OPDSDumper opdsDumper;
+  kiwix::OPDSDumper opdsDumper(mp_library);
   opdsDumper.setRootLocation(m_root);
-  opdsDumper.setSearchDescriptionUrl("catalog/searchdescription.xml");
-  opdsDumper.setLibrary(mp_library);
+  opdsDumper.setLibraryId(m_library_id);
   std::vector<std::string> bookIdsToDump;
   if (url == "root.xml") {
-    opdsDumper.setTitle("All zims");
     uuid = zim::Uuid::generate(host);
     bookIdsToDump = mp_library->filter(kiwix::Filter().valid(true).local(true).remote(true));
   } else if (url == "search") {
-    auto filter = kiwix::Filter().valid(true).local(true).remote(true);
-    string query("<Empty query>");
-    size_t count(10);
-    size_t startIndex(0);
+    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 {
-      query = request.get_argument("q");
-      filter.query(query);
+      filter.query(request.get_argument("q"));
     } catch (const std::out_of_range&) {}
     try {
-      filter.maxSize(extractFromString<unsigned long>(request.get_argument("maxsize")));
+      filter.maxSize(request.get_argument<unsigned long>("maxsize"));
     } catch (...) {}
     try {
       filter.name(request.get_argument("name"));
     } catch (const std::out_of_range&) {}
     try {
-      filter.lang(request.get_argument("lang"));
+      filter.category(request.get_argument("category"));
     } catch (const std::out_of_range&) {}
     try {
-      count = extractFromString<unsigned long>(request.get_argument("count"));
-    } catch (...) {}
-    try {
-      startIndex = extractFromString<unsigned long>(request.get_argument("start"));
-    } catch (...) {}
+      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 (...) {}
-    opdsDumper.setTitle("Search result for " + query);
-    uuid = zim::Uuid::generate();
-    bookIdsToDump = mp_library->filter(filter);
-    auto totalResults = bookIdsToDump.size();
-    bookIdsToDump.erase(bookIdsToDump.begin(), bookIdsToDump.begin()+startIndex);
-    if (count>0 && bookIdsToDump.size() > count) {
-      bookIdsToDump.resize(count);
-    }
-    opdsDumper.setOpenSearchInfo(totalResults, startIndex, bookIdsToDump.size());
-  }
+    return filter;
+}
 
-  opdsDumper.setId(kiwix::to_string(uuid));
-  auto response = ContentResponse::build(
-      *this,
-      opdsDumper.dumpOPDSFeed(bookIdsToDump),
-      "application/atom+xml; profile=opds-catalog; kind=acquisition; charset=utf-8");
-  return std::move(response);
+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);
+    const size_t intendedCount = count > 0 ? count : bookIdsToDump.size();
+    bookIdsToDump = subrange(bookIdsToDump, startIndex, intendedCount);
+    opdsDumper.setOpenSearchInfo(totalResults, startIndex, bookIdsToDump.size());
+    return bookIdsToDump;
 }
 
 namespace
@@ -728,29 +762,28 @@ std::string get_book_name(const RequestContext& request)
   }
 }
 
-} // unnamed namespace
-
-std::shared_ptr<Reader>
-InternalServer::get_reader(const std::string& bookName) const
+std::string searchSuggestionHTML(const std::string& searchURL, const std::string& pattern)
 {
-  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;
+  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::unique_ptr<Response>
-InternalServer::build_redirect(const std::string& bookName, const kiwix::Entry& entry) const
+InternalServer::build_redirect(const std::string& bookName, const zim::Item& item) const
 {
-  auto redirectUrl = m_root + "/" + bookName + "/" + kiwix::urlEncode(entry.getPath());
+  auto redirectUrl = m_root + "/" + bookName + "/" + kiwix::urlEncode(item.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");
   }
@@ -759,9 +792,17 @@ std::unique_ptr<Response> InternalServer::handle_content(const RequestContext& r
   if (bookName.empty())
     return build_homepage(request);
 
-  const std::shared_ptr<Reader> reader = get_reader(bookName);
-  if (reader == nullptr) {
-    return Response::build_404(*this, request, bookName);
+  std::shared_ptr<zim::Archive> archive;
+  try {
+    const std::string bookId = mp_nameMapper->getIdForName(bookName);
+    archive = mp_library->getArchiveById(bookId);
+  } catch (const std::out_of_range& e) {}
+
+  if (archive == nullptr) {
+    std::string searchURL = m_root + "/search?pattern=" + kiwix::urlEncode(pattern, true); // Make a full search on the entire library.
+    const std::string details = searchSuggestionHTML(searchURL, kiwix::urlDecode(pattern));
+
+    return Response::build_404(*this, request.get_full_url(), bookName, "", details);
   }
 
   auto urlStr = request.get_url().substr(bookName.size()+1);
@@ -770,28 +811,89 @@ std::unique_ptr<Response> InternalServer::handle_content(const RequestContext& r
   }
 
   try {
-    auto entry = reader->getEntryFromPath(urlStr);
+    auto entry = getEntryFromPath(*archive, 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());
+      return build_redirect(bookName, getFinalItem(*archive, entry));
     }
-    auto response = ItemResponse::build(*this, request, entry.getZimEntry().getItem());
+    auto response = ItemResponse::build(*this, request, entry.getItem());
     try {
-      dynamic_cast<ContentResponse&>(*response).set_taskbar(bookName, reader->getTitle());
+      dynamic_cast<ContentResponse&>(*response).set_taskbar(bookName, getArchiveTitle(*archive));
     } 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());
+      printf("mimeType: %s\n", entry.getItem(true).getMimetype().c_str());
     }
 
     return response;
-  } catch(kiwix::NoEntry& e) {
+  } catch(zim::EntryNotFound& e) {
     if (m_verbose.load())
       printf("Failed to find %s\n", urlStr.c_str());
 
-    return Response::build_404(*this, request, bookName);
+    std::string searchURL = m_root + "/search?content=" + bookName + "&pattern=" + kiwix::urlEncode(pattern, true); // Make a search on this specific book only.
+    const std::string details = searchSuggestionHTML(searchURL, kiwix::urlDecode(pattern));
+
+    return Response::build_404(*this, request.get_full_url(), bookName, getArchiveTitle(*archive), details);
+  }
+}
+
+
+std::unique_ptr<Response> InternalServer::handle_raw(const RequestContext& request)
+{
+  if (m_verbose.load()) {
+    printf("** running handle_raw\n");
+  }
+
+  std::string bookName;
+  std::string kind;
+  try {
+     bookName = request.get_url_part(1);
+     kind = request.get_url_part(2);
+  } catch (const std::out_of_range& e) {
+     return Response::build_404(*this, request.get_full_url(), bookName, "", "");
+  }
+
+  if (kind != "meta" && kind!= "content") {
+    const std::string error_details = kind + " is not a valid request for raw content.";
+    return Response::build_404(*this, request.get_full_url(), bookName, "", error_details);
+  }
+
+  std::shared_ptr<zim::Archive> archive;
+  try {
+    const std::string bookId = mp_nameMapper->getIdForName(bookName);
+    archive = mp_library->getArchiveById(bookId);
+  } catch (const std::out_of_range& e) {}
+
+  if (archive == nullptr) {
+    const std::string error_details = "No such book: " + bookName;
+    return Response::build_404(*this, request.get_full_url(), bookName, "", error_details);
+  }
+
+  // Remove the beggining of the path:
+  // /raw/<bookName>/<kind>/foo
+  // ^^^^^          ^      ^
+  //   5      +     1  +   1   = 7
+  auto itemPath = request.get_url().substr(bookName.size()+kind.size()+7);
+
+  try {
+    if (kind == "meta") {
+      auto item = archive->getMetadataItem(itemPath);
+      return ItemResponse::build(*this, request, item, /*raw=*/true);
+    } else {
+      auto entry = archive->getEntryByPath(itemPath);
+      if (entry.isRedirect()) {
+        return build_redirect(bookName, entry.getItem(true));
+      }
+      return ItemResponse::build(*this, request, entry.getItem(), /*raw=*/true);
+    }
+  } catch (zim::EntryNotFound& e ) {
+    if (m_verbose.load()) {
+      printf("Failed to find %s\n", itemPath.c_str());
+    }
+    const std::string error_details = "Cannot find " + kind + " entry " + itemPath;
+    return Response::build_404(*this, request.get_full_url(), bookName, getArchiveTitle(*archive), error_details);
   }
 }
 

src/server/internalServer.h

@@ -28,6 +28,9 @@ extern "C" {
 #include "library.h"
 #include "name_mapper.h"
 
+#include <zim/search.h>
+#include <zim/suggestion.h>
+
 #include <mustache.hpp>
 
 #include <atomic>
@@ -36,11 +39,17 @@ extern "C" {
 #include "server/request_context.h"
 #include "server/response.h"
 
+#include "tools/concurrent_cache.h"
+
 namespace kiwix {
 
 typedef kainjow::mustache::data MustacheData;
+typedef ConcurrentCache<string, std::shared_ptr<zim::Searcher>> SearcherCache;
+typedef ConcurrentCache<string, std::shared_ptr<zim::Search>> SearchCache;
+typedef ConcurrentCache<string, std::shared_ptr<zim::SuggestionSearcher>> SuggestionSearcherCache;
 
 class Entry;
+class OPDSDumper;
 
 class InternalServer {
   public:
@@ -53,7 +62,9 @@ class InternalServer {
                    bool verbose,
                    bool withTaskbar,
                    bool withLibraryButton,
-                   bool blockExternalLinks);
+                   bool blockExternalLinks,
+                   std::string indexTemplateString,
+                   int ipConnectionLimit);
     virtual ~InternalServer() = default;
 
     MHD_Result handlerCallback(struct MHD_Connection* connection,
@@ -65,24 +76,34 @@ class InternalServer {
                                void** cont_cls);
     bool start();
     void stop();
+    std::string getAddress() { return m_addr; }
+    int getPort() { return m_port; } 
 
   private: // functions
     std::unique_ptr<Response> handle_request(const RequestContext& request);
-    std::unique_ptr<Response> build_redirect(const std::string& bookName, const kiwix::Entry& entry) const;
+    std::unique_ptr<Response> build_redirect(const std::string& bookName, const zim::Item& item) const;
     std::unique_ptr<Response> build_homepage(const RequestContext& request);
     std::unique_ptr<Response> handle_skin(const RequestContext& request);
     std::unique_ptr<Response> handle_catalog(const RequestContext& request);
-    std::unique_ptr<Response> handle_meta(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, bool partial);
+    std::unique_ptr<Response> handle_catalog_v2_complete_entry(const RequestContext& request, const std::string& entryId);
+    std::unique_ptr<Response> handle_catalog_v2_categories(const RequestContext& request);
+    std::unique_ptr<Response> handle_catalog_v2_languages(const RequestContext& request);
+    std::unique_ptr<Response> handle_catalog_v2_illustration(const RequestContext& request);
     std::unique_ptr<Response> handle_search(const RequestContext& request);
     std::unique_ptr<Response> handle_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::unique_ptr<Response> handle_raw(const RequestContext& request);
+
+    std::vector<std::string> search_catalog(const RequestContext& request,
+                                            kiwix::OPDSDumper& opdsDumper);
 
     MustacheData get_default_data() const;
-    MustacheData homepage_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;
 
@@ -95,16 +116,23 @@ class InternalServer {
     bool m_withTaskbar;
     bool m_withLibraryButton;
     bool m_blockExternalLinks;
+    std::string m_indexTemplateString;
+    int m_ipConnectionLimit;
     struct MHD_Daemon* mp_daemon;
 
     Library* mp_library;
     NameMapper* mp_nameMapper;
 
+    SearcherCache searcherCache;
+    SearchCache searchCache;
+    SuggestionSearcherCache suggestionSearcherCache;
+
     std::string m_server_id;
+    std::string m_library_id;
 
     friend std::unique_ptr<Response> Response::build(const InternalServer& server);
-    friend std::unique_ptr<ContentResponse> ContentResponse::build(const InternalServer& server, const std::string& content, const std::string& mimetype);
-    friend std::unique_ptr<Response> ItemResponse::build(const InternalServer& server, const RequestContext& request, const zim::Item& item);
+    friend std::unique_ptr<ContentResponse> ContentResponse::build(const InternalServer& server, const std::string& content, const std::string& mimetype, bool isHomePage, bool raw);
+    friend std::unique_ptr<Response> ItemResponse::build(const InternalServer& server, const RequestContext& request, const zim::Item& item, bool raw);
     friend std::unique_ptr<Response> Response::build_500(const InternalServer& server, const std::string& msg);
 
 };

src/server/internalServer_catalog_v2.cpp

@@ -0,0 +1,165 @@
+/*
+ * 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.get_full_url(), "", "");
+  }
+
+  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 == "entry") {
+    const std::string entryId  = request.get_url_part(3);
+    return handle_catalog_v2_complete_entry(request, entryId);
+  } else if (url == "entries") {
+    return handle_catalog_v2_entries(request, /*partial=*/false);
+  } else if (url == "partial_entries") {
+    return handle_catalog_v2_entries(request, /*partial=*/true);
+  } else if (url == "categories") {
+    return handle_catalog_v2_categories(request);
+  } else if (url == "languages") {
+    return handle_catalog_v2_languages(request);
+  } else if (url == "illustration") {
+    return handle_catalog_v2_illustration(request);
+  } else {
+    return Response::build_404(*this, request.get_full_url(), "", "");
+  }
+}
+
+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")},
+               {"partial_entries_feed_id", gen_uuid(m_library_id + "/partial_entries")},
+               {"category_list_feed_id", gen_uuid(m_library_id + "/categories")},
+               {"language_list_feed_id", gen_uuid(m_library_id + "/languages")}
+             },
+             "application/atom+xml;profile=opds-catalog;kind=navigation"
+  );
+}
+
+std::unique_ptr<Response> InternalServer::handle_catalog_v2_entries(const RequestContext& request, bool partial)
+{
+  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(), partial);
+  return ContentResponse::build(
+             *this,
+             opdsFeed,
+             "application/atom+xml;profile=opds-catalog;kind=acquisition"
+  );
+}
+
+std::unique_ptr<Response> InternalServer::handle_catalog_v2_complete_entry(const RequestContext& request, const std::string& entryId)
+{
+  try {
+    mp_library->getBookById(entryId);
+  } catch (const std::out_of_range&) {
+    return Response::build_404(*this, request.get_full_url(), "", "");
+  }
+
+  OPDSDumper opdsDumper(mp_library);
+  opdsDumper.setRootLocation(m_root);
+  opdsDumper.setLibraryId(m_library_id);
+  const auto opdsFeed = opdsDumper.dumpOPDSCompleteEntry(entryId);
+  return ContentResponse::build(
+             *this,
+             opdsFeed,
+             "application/atom+xml;type=entry;profile=opds-catalog"
+  );
+}
+
+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(),
+             "application/atom+xml;profile=opds-catalog;kind=navigation"
+  );
+}
+
+std::unique_ptr<Response> InternalServer::handle_catalog_v2_languages(const RequestContext& request)
+{
+  OPDSDumper opdsDumper(mp_library);
+  opdsDumper.setRootLocation(m_root);
+  opdsDumper.setLibraryId(m_library_id);
+  return ContentResponse::build(
+             *this,
+             opdsDumper.languagesOPDSFeed(),
+             "application/atom+xml;profile=opds-catalog;kind=navigation"
+  );
+}
+
+std::unique_ptr<Response> InternalServer::handle_catalog_v2_illustration(const RequestContext& request)
+{
+  try {
+    const auto bookName  = request.get_url_part(3);
+    const auto bookId = mp_nameMapper->getIdForName(bookName);
+    auto book = mp_library->getBookByIdThreadSafe(bookId);
+    auto size = request.get_argument<unsigned int>("size");
+    auto illustration = book.getIllustration(size);
+    return ContentResponse::build(*this, illustration->getData(), illustration->mimeType);
+  } catch(...) {
+    return Response::build_404(*this, request.get_full_url(), "", "");
+  }
+}
+
+} // namespace kiwix

src/server/request_context.cpp

@@ -183,4 +183,14 @@ 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

@@ -74,11 +74,21 @@ class RequestContext {
         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;
 

src/server/response.cpp

@@ -24,6 +24,7 @@
 
 #include "tools/regexTools.h"
 #include "tools/stringTools.h"
+#include "tools/otherTools.h"
 
 #include "string.h"
 #include <mustache.hpp>
@@ -38,17 +39,6 @@ namespace
 {
 // some utilities
 
-std::string render_template(const std::string& template_str, kainjow::mustache::data data)
-{
-  kainjow::mustache::mustache tmpl(template_str);
-  kainjow::mustache::data urlencode{kainjow::mustache::lambda2{
-                               [](const std::string& str,const kainjow::mustache::renderer& r) { return urlEncode(r(str), true); }}};
-  data.set("urlencoded", urlencode);
-  std::stringstream ss;
-  tmpl.render(data, [&ss](const std::string& str) { ss << str; });
-  return ss.str();
-}
-
 std::string get_mime_type(const zim::Item& item)
 {
   try {
@@ -93,14 +83,17 @@ std::unique_ptr<Response> Response::build_304(const InternalServer& server, cons
   return response;
 }
 
-std::unique_ptr<Response> Response::build_404(const InternalServer& server, const RequestContext& request, const std::string& bookName)
+std::unique_ptr<Response> Response::build_404(const InternalServer& server, const std::string& url, const std::string& bookName, const std::string& bookTitle, const std::string& details)
 {
   MustacheData results;
-  results.set("url", request.get_full_url());
+  if ( !url.empty() ) {
+    results.set("url", 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, "");
+  response->set_taskbar(bookName, bookTitle);
 
   return std::move(response);
 }
@@ -124,7 +117,16 @@ std::unique_ptr<Response> Response::build_500(const InternalServer& server, cons
   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"));
+      new ContentResponse(
+        server.m_root, //root
+        true, //verbose
+        true, //raw
+        false, //withTaskbar
+        false, //withLibraryButton
+        false, //blockExternalLinks
+        content, //content
+        "text/html" //mimetype
+  ));
   response->set_code(MHD_HTTP_INTERNAL_SERVER_ERROR);
   return response;
 }
@@ -199,10 +201,10 @@ void ContentResponse::introduce_taskbar()
   kainjow::mustache::data data;
   data.set("root", m_root);
   data.set("content", m_bookName);
-  data.set("hascontent", !m_bookName.empty());
+  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_part_html, data);
+  auto head_content = render_template(RESOURCE::templates::head_taskbar_html, data);
   m_content = prependToFirstOccurence(
     m_content,
     "</head[ \\t]*>",
@@ -221,12 +223,19 @@ 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 = appendToFirstOccurence(
+  m_content = prependToFirstOccurence(
     m_content,
-    "<head>",
+    "</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
 {
@@ -238,8 +247,11 @@ ContentResponse::can_compress(const RequestContext& request) const
 bool
 ContentResponse::contentDecorationAllowed() const
 {
-    return (startsWith(m_mimeType, "text/html")
-         && m_mimeType.find(";raw=true") == std::string::npos);
+  if (m_raw) {
+    return false;
+  }
+  return (startsWith(m_mimeType, "text/html")
+    && m_mimeType.find(";raw=true") == std::string::npos);
 }
 
 MHD_Response*
@@ -253,6 +265,8 @@ MHD_Response*
 ContentResponse::create_mhd_response(const RequestContext& request)
 {
   if (contentDecorationAllowed()) {
+    inject_root_link();
+
     if (m_withTaskbar) {
       introduce_taskbar();
     }
@@ -325,11 +339,12 @@ void ContentResponse::set_taskbar(const std::string& bookName, const std::string
 }
 
 
-ContentResponse::ContentResponse(const std::string& root, bool verbose, bool withTaskbar, bool withLibraryButton, bool blockExternalLinks, const std::string& content, const std::string& mimetype) :
+ContentResponse::ContentResponse(const std::string& root, bool verbose, bool raw, 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_raw(raw),
   m_withTaskbar(withTaskbar),
   m_withLibraryButton(withLibraryButton),
   m_blockExternalLinks(blockExternalLinks),
@@ -339,21 +354,33 @@ ContentResponse::ContentResponse(const std::string& root, bool verbose, bool wit
   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)
+std::unique_ptr<ContentResponse> ContentResponse::build(
+  const InternalServer& server,
+  const std::string& content,
+  const std::string& mimetype,
+  bool isHomePage,
+  bool raw)
 {
    return std::unique_ptr<ContentResponse>(new ContentResponse(
         server.m_root,
         server.m_verbose.load(),
-        server.m_withTaskbar,
+        raw,
+        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) {
+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);
+  return ContentResponse::build(server, content, mimetype, isHomePage);
 }
 
 ItemResponse::ItemResponse(bool verbose, const zim::Item& item, const std::string& mimetype, const ByteRange& byterange) :
@@ -366,14 +393,14 @@ ItemResponse::ItemResponse(bool verbose, const zim::Item& item, const std::strin
   add_header(MHD_HTTP_HEADER_CONTENT_TYPE, m_mimeType);
 }
 
-std::unique_ptr<Response> ItemResponse::build(const InternalServer& server, const RequestContext& request, const zim::Item& item)
+std::unique_ptr<Response> ItemResponse::build(const InternalServer& server, const RequestContext& request, const zim::Item& item, bool raw)
 {
   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);
+    auto response = ContentResponse::build(server, item.getData(), mimetype, /*isHomePage=*/false, raw);
     response->set_cacheable();
     response->m_byteRange = byteRange;
     return std::move(response);

src/server/response.h

@@ -47,7 +47,7 @@ class Response {
 
     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);
+    static std::unique_ptr<Response> build_404(const InternalServer& server, const std::string& url, 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);
@@ -78,9 +78,27 @@ class Response {
 
 class ContentResponse : public Response {
   public:
-    ContentResponse(const std::string& root, bool verbose, bool withTaskbar, bool withLibraryButton, bool blockExternalLinks, const std::string& content, const std::string& mimetype);
-    static std::unique_ptr<ContentResponse> build(const InternalServer& server, const std::string& content, const std::string& mimetype);
-    static std::unique_ptr<ContentResponse> build(const InternalServer& server, const std::string& template_str, kainjow::mustache::data data, const std::string& mimetype);
+    ContentResponse(
+      const std::string& root,
+      bool verbose,
+      bool raw,
+      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,
+      bool raw = 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);
 
@@ -89,6 +107,7 @@ class ContentResponse : public Response {
 
     void introduce_taskbar();
     void inject_externallinks_blocker();
+    void inject_root_link();
     bool can_compress(const RequestContext& request) const;
     bool contentDecorationAllowed() const;
 
@@ -97,6 +116,7 @@ class ContentResponse : public Response {
     std::string m_root;
     std::string m_content;
     std::string m_mimeType;
+    bool m_raw;
     bool m_withTaskbar;
     bool m_withLibraryButton;
     bool m_blockExternalLinks;
@@ -107,7 +127,7 @@ class ContentResponse : public Response {
 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);
+    static std::unique_ptr<Response> build(const InternalServer& server, const RequestContext& request, const zim::Item& item, bool raw = false);
 
   private:
     MHD_Response* create_mhd_response(const RequestContext& request);

src/tools/archiveTools.cpp

@@ -0,0 +1,158 @@
+/*
+ * Copyright 2021 Maneesh P M <manu.pm55@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#include "archiveTools.h"
+
+#include "tools.h"
+#include "pathTools.h"
+#include "otherTools.h"
+#include "stringTools.h"
+
+#include <zim/error.h>
+#include <zim/item.h>
+
+namespace kiwix
+{
+std::string getMetadata(const zim::Archive& archive, const std::string& name) {
+    try {
+        return archive.getMetadata(name);
+    } catch (zim::EntryNotFound& e) {
+        return "";
+    }
+}
+
+std::string getArchiveTitle(const zim::Archive& archive) {
+  std::string value = getMetadata(archive, "Title");
+  if (value.empty()) {
+    value = getLastPathElement(archive.getFilename());
+    std::replace(value.begin(), value.end(), '_', ' ');
+    size_t pos = value.find(".zim");
+    value = value.substr(0, pos);
+  }
+  return value;
+}
+
+std::string getMetaDescription(const zim::Archive& archive) {
+  std::string value;
+  value = getMetadata(archive, "Description");
+
+  /* Mediawiki Collection tends to use the "Subtitle" name */
+  if (value.empty()) {
+    value = getMetadata(archive, "Subtitle");
+  }
+
+  return value;
+}
+
+std::string getMetaTags(const zim::Archive& archive, bool original) {
+  std::string tags_str = getMetadata(archive, "Tags");
+  if (original) {
+    return tags_str;
+  }
+  auto tags = convertTags(tags_str);
+  return join(tags, ";");
+}
+
+std::string getMetaLanguage(const zim::Archive& archive) {
+  return getMetadata(archive, "Language");
+}
+
+std::string getMetaName(const zim::Archive& archive) {
+  return getMetadata(archive, "Name");
+}
+
+std::string getMetaDate(const zim::Archive& archive) {
+  return getMetadata(archive, "Date");
+}
+
+std::string getMetaCreator(const zim::Archive& archive) {
+  return getMetadata(archive, "Creator");
+}
+
+std::string getMetaPublisher(const zim::Archive& archive) {
+  return getMetadata(archive, "Publisher");
+}
+
+std::string getMetaFlavour(const zim::Archive& archive) {
+  return getMetadata(archive, "Flavour");
+}
+
+std::string getArchiveId(const zim::Archive& archive) {
+  return (std::string) archive.getUuid();
+}
+
+bool getArchiveFavicon(const zim::Archive& archive, unsigned size,
+                           std::string& content, std::string& mimeType){
+  try {
+    auto item = archive.getIllustrationItem(size);
+    content = item.getData();
+    mimeType = item.getMimetype();
+    return true;
+  } catch(zim::EntryNotFound& e) {};
+
+  return false;
+}
+
+// should this be in libzim
+unsigned int getArchiveMediaCount(const zim::Archive& archive) {
+  std::map<const std::string, unsigned int> counterMap = parseArchiveCounter(archive);
+  unsigned int counter = 0;
+
+  for (auto &pair:counterMap) {
+    if (startsWith(pair.first, "image/") ||
+        startsWith(pair.first, "video/") ||
+        startsWith(pair.first, "audio/")) {
+      counter += pair.second;
+    }
+  }
+
+  return counter;
+}
+
+unsigned int getArchiveFileSize(const zim::Archive& archive) {
+  return archive.getFilesize() / 1024;
+}
+
+zim::Item getFinalItem(const zim::Archive& archive, const zim::Entry& entry)
+{
+  return entry.getItem(true);
+}
+
+zim::Entry getEntryFromPath(const zim::Archive& archive, const std::string& path)
+{
+  try {
+    return archive.getEntryByPath(path);
+  } catch (zim::EntryNotFound& e) {
+    if (path.empty() || path == "/") {
+      return archive.getMainEntry();
+    }
+  }
+  throw zim::EntryNotFound("Cannot find entry for non empty path");
+}
+
+MimeCounterType parseArchiveCounter(const zim::Archive& archive) {
+  try {
+    auto counterContent = archive.getMetadata("Counter");
+    return parseMimetypeCounter(counterContent);
+  } catch (zim::EntryNotFound& e) {
+    return {};
+  }
+}
+
+} // kiwix

src/tools/archiveTools.h

@@ -0,0 +1,59 @@
+/*
+ * Copyright 2021 Maneesh P M <manu.pm55@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#ifndef KIWIX_ARCHIVETOOLS_H
+#define KIWIX_ARCHIVETOOLS_H
+
+#include <zim/archive.h>
+#include <tools/otherTools.h>
+
+/**
+ * This file contains all the functions that would make handling data related to
+ * an archive easier.
+ **/
+
+namespace kiwix
+{
+    std::string getMetadata(const zim::Archive& archive, const std::string& name);
+    std::string getArchiveTitle(const zim::Archive& archive);
+    std::string getMetaDescription(const zim::Archive& archive);
+    std::string getMetaTags(const zim::Archive& archive, bool original = false);
+    std::string getMetaLanguage(const zim::Archive& archive);
+    std::string getMetaName(const zim::Archive& archive);
+    std::string getMetaDate(const zim::Archive& archive);
+    std::string getMetaCreator(const zim::Archive& archive);
+    std::string getMetaPublisher(const zim::Archive& archive);
+    std::string getMetaFlavour(const zim::Archive& archive);
+    std::string getArchiveId(const zim::Archive& archive);
+
+    bool getArchiveFavicon(const zim::Archive& archive, unsigned size,
+                           std::string& content, std::string& mimeType);
+
+    unsigned int getArchiveMediaCount(const zim::Archive& archive);
+    unsigned int getArchiveFileSize(const zim::Archive& archive);
+
+    zim::Item getFinalItem(const zim::Archive& archive, const zim::Entry& entry);
+
+    zim::Entry getEntryFromPath(const zim::Archive& archive, const std::string& path);
+
+    MimeCounterType parseArchiveCounter(const zim::Archive& archive);
+
+}
+
+#endif

src/tools/concurrent_cache.h

@@ -0,0 +1,95 @@
+/*
+ * Copyright (C) 2021 Matthieu Gautier <mgautier@kymeria.fr>
+ * Copyright (C) 2020 Veloman Yunkan
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * is provided AS IS, WITHOUT ANY WARRANTY; without even the implied
+ * warranty of MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, and
+ * NON-INFRINGEMENT.  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 St, Fifth Floor, Boston, MA  02110-1301 USA
+ *
+ */
+
+#ifndef ZIM_CONCURRENT_CACHE_H
+#define ZIM_CONCURRENT_CACHE_H
+
+#include "lrucache.h"
+
+#include <future>
+#include <mutex>
+
+namespace kiwix
+{
+
+/**
+   ConcurrentCache implements a concurrent thread-safe cache
+
+   Compared to kiwix::lru_cache, each access operation is slightly more expensive.
+   However, different slots of the cache can be safely accessed concurrently
+   with minimal blocking. Concurrent access to the same element is also
+   safe, and, in case of a cache miss, will block until that element becomes
+   available.
+ */
+template <typename Key, typename Value>
+class ConcurrentCache
+{
+private: // types
+  typedef std::shared_future<Value> ValuePlaceholder;
+  typedef lru_cache<Key, ValuePlaceholder> Impl;
+
+public: // types
+  explicit ConcurrentCache(size_t maxEntries)
+    : impl_(maxEntries)
+  {}
+
+  // Gets the entry corresponding to the given key. If the entry is not in the
+  // cache, it is obtained by calling f() (without any arguments) and the
+  // result is put into the cache.
+  //
+  // The cache as a whole is locked only for the duration of accessing
+  // the respective slot. If, in the case of the a cache miss, the generation
+  // of the missing element takes a long time, only attempts to access that
+  // element will block - the rest of the cache remains open to concurrent
+  // access.
+  template<class F>
+  Value getOrPut(const Key& key, F f)
+  {
+    std::promise<Value> valuePromise;
+    std::unique_lock<std::mutex> l(lock_);
+    const auto x = impl_.getOrPut(key, valuePromise.get_future().share());
+    l.unlock();
+    if ( x.miss() ) {
+      try {
+        valuePromise.set_value(f());
+      } catch (std::exception& e) {
+        drop(key);
+        throw;
+      }
+    }
+
+    return x.value().get();
+  }
+
+  bool drop(const Key& key)
+  {
+    std::unique_lock<std::mutex> l(lock_);
+    return impl_.drop(key);
+  }
+
+private: // data
+  Impl impl_;
+  std::mutex lock_;
+};
+
+} // namespace kiwix
+
+#endif // ZIM_CONCURRENT_CACHE_H
+

src/tools/lrucache.h

@@ -0,0 +1,160 @@
+/*
+ * Copyrigth (c) 2021, Matthieu Gautier <mgautier@kymeria.fr>
+ * Copyright (c) 2020, Veloman Yunkan
+ * Copyright (c) 2014, lamerman
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *
+ * * Redistributions of source code must retain the above copyright notice, this
+ *   list of conditions and the following disclaimer.
+ *
+ * * Redistributions in binary form must reproduce the above copyright notice,
+ *   this list of conditions and the following disclaimer in the documentation
+ *   and/or other materials provided with the distribution.
+ *
+ * * Neither the name of lamerman nor the names of its
+ *   contributors may be used to endorse or promote products derived from
+ *   this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+ * DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+ * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+ * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+ * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * File:   lrucache.hpp
+ * Author: Alexander Ponomarev
+ *
+ * Created on June 20, 2013, 5:09 PM
+ */
+
+#ifndef _LRUCACHE_HPP_INCLUDED_
+#define _LRUCACHE_HPP_INCLUDED_
+
+#include <map>
+#include <list>
+#include <cstddef>
+#include <stdexcept>
+#include <cassert>
+
+namespace kiwix {
+
+template<typename key_t, typename value_t>
+class lru_cache {
+public: // types
+  typedef typename std::pair<key_t, value_t> key_value_pair_t;
+  typedef typename std::list<key_value_pair_t>::iterator list_iterator_t;
+
+  enum AccessStatus {
+    HIT, // key was found in the cache
+    PUT, // key was not in the cache but was created by the getOrPut() access
+    MISS // key was not in the cache; get() access failed
+  };
+
+  class AccessResult
+  {
+    const AccessStatus status_;
+    const value_t val_;
+  public:
+    AccessResult(const value_t& val, AccessStatus status)
+      : status_(status), val_(val)
+    {}
+    AccessResult() : status_(MISS), val_() {}
+
+    bool hit() const { return status_ == HIT; }
+    bool miss() const { return !hit(); }
+    const value_t& value() const
+    {
+      if ( status_ == MISS )
+        throw std::range_error("There is no such key in cache");
+      return val_;
+    }
+
+    operator const value_t& () const { return value(); }
+  };
+
+public: // functions
+  explicit lru_cache(size_t max_size) :
+    _max_size(max_size) {
+  }
+
+  // If 'key' is present in the cache, returns the associated value,
+  // otherwise puts the given value into the cache (and returns it with
+  // a status of a cache miss).
+  AccessResult getOrPut(const key_t& key, const value_t& value) {
+    auto it = _cache_items_map.find(key);
+    if (it != _cache_items_map.end()) {
+      _cache_items_list.splice(_cache_items_list.begin(), _cache_items_list, it->second);
+      return AccessResult(it->second->second, HIT);
+    } else {
+      putMissing(key, value);
+      return AccessResult(value, PUT);
+    }
+  }
+
+  void put(const key_t& key, const value_t& value) {
+    auto it = _cache_items_map.find(key);
+    if (it != _cache_items_map.end()) {
+      _cache_items_list.splice(_cache_items_list.begin(), _cache_items_list, it->second);
+      it->second->second = value;
+    } else {
+      putMissing(key, value);
+    }
+  }
+
+  AccessResult get(const key_t& key) {
+    auto it = _cache_items_map.find(key);
+    if (it == _cache_items_map.end()) {
+      return AccessResult();
+    } else {
+      _cache_items_list.splice(_cache_items_list.begin(), _cache_items_list, it->second);
+      return AccessResult(it->second->second, HIT);
+    }
+  }
+
+  bool drop(const key_t& key) {
+    try {
+      auto list_it = _cache_items_map.at(key);
+      _cache_items_list.erase(list_it);
+      _cache_items_map.erase(key);
+      return true;
+    } catch (std::out_of_range& e) {
+      return false;
+    }
+  }
+
+  bool exists(const key_t& key) const {
+    return _cache_items_map.find(key) != _cache_items_map.end();
+  }
+
+  size_t size() const {
+    return _cache_items_map.size();
+  }
+
+private: // functions
+  void putMissing(const key_t& key, const value_t& value) {
+    assert(_cache_items_map.find(key) == _cache_items_map.end());
+    _cache_items_list.push_front(key_value_pair_t(key, value));
+    _cache_items_map[key] = _cache_items_list.begin();
+    if (_cache_items_map.size() > _max_size) {
+      _cache_items_map.erase(_cache_items_list.back().first);
+      _cache_items_list.pop_back();
+    }
+  }
+
+private: // data
+  std::list<key_value_pair_t> _cache_items_list;
+  std::map<key_t, list_iterator_t> _cache_items_map;
+  size_t _max_size;
+};
+
+} // namespace kiwix
+
+#endif  /* _LRUCACHE_HPP_INCLUDED_ */

src/tools/networkTools.cpp

@@ -1,5 +1,6 @@
 /*
  * Copyright 2012 Emmanuel Engelhart <kelson@kiwix.org>
+ * Copyright 2021 Nikhil Tanwar <2002nikhiltanwar@gmail.com>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU  General Public License as published by
@@ -17,6 +18,7 @@
  * MA 02110-1301, USA.
  */
 
+#include "tools.h"
 #include <tools/networkTools.h>
 
 #include <stdio.h>
@@ -29,6 +31,17 @@
 #include <iostream>
 #include <stdexcept>
 
+#ifdef _WIN32
+#include <winsock2.h>
+#include <ws2tcpip.h>
+#include <iostream>
+#else
+#include <unistd.h>
+#include <sys/ioctl.h>
+#include <sys/socket.h>
+#include <net/if.h>
+#include <netdb.h>
+#endif
 
 size_t write_callback_to_iss(char* ptr, size_t size, size_t nmemb, void* userdata)
 {
@@ -57,3 +70,104 @@ std::string kiwix::download(const std::string& url) {
   }
   return ss.str();
 }
+
+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 == INVALID_SOCKET) {
+    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.AddressIn);
+    if(pAddress->sin_family == AF_INET) {
+    /* Add to the map */
+    std::string interfaceName = std::string(inet_ntoa(pAddress->sin_addr));
+    interfaces[interfaceName] = interfaceName;
+    }
+  }
+#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)");
+  }
+
+  /* Go through each interface */
+  struct ifreq *ifreq;
+  ifreq = ifconf.ifc_req;
+  for (int 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[interfaceName] = interfaceIp;
+      } else {
+        perror("getnameinfo()");
+      }
+    }
+
+    /* some systems have ifr_addr.sa_len and adjust the length that
+     * way, but not mine. weird */
+    size_t len;
+#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() {
+  auto interfaces = getNetworkInterfaces();
+
+#ifndef _WIN32
+  const char* const prioritizedNames[] =
+      { "eth0", "eth1", "wlan0", "wlan1", "en0", "en1" };
+  for(auto name: prioritizedNames) {
+    auto it = interfaces.find(name);
+    if(it != interfaces.end()) {
+      return it->second;
+    }
+  }
+#endif
+
+  const char* const prefixes[] = { "192.168", "172.16.", "10.0" };
+  for(auto prefix : prefixes){
+    for(auto& itr : interfaces) {
+      auto interfaceIp = itr.second;
+      if (interfaceIp.find(prefix) == 0) {
+        return interfaceIp;
+      }
+    }
+  }
+
+  return "127.0.0.1";
+}

src/tools/otherTools.cpp

@@ -17,8 +17,13 @@
  * MA 02110-1301, USA.
  */
 
+// Implement function declared in tools.h and tools/otherTools.h
+#include "tools.h"
 #include "tools/otherTools.h"
+
+
 #include <algorithm>
+#include <iomanip>
 
 #ifdef _WIN32
 #include <windows.h>
@@ -32,6 +37,8 @@
 #include <sstream>
 #include <pugixml.hpp>
 
+#include <zim/uuid.h>
+
 
 static std::map<std::string, std::string> codeisomapping {
 { "aa", "aar" },
@@ -341,3 +348,35 @@ kiwix::MimeCounterType kiwix::parseMimetypeCounter(const std::string& counterDat
 
   return counters;
 }
+
+std::string kiwix::gen_date_str()
+{
+  auto now = std::time(0);
+  auto tm = std::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+1 << "-"
+     << 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();
+}
+
+std::string kiwix::gen_uuid(const std::string& s)
+{
+  return kiwix::to_string(zim::Uuid::generate(s));
+}
+
+std::string kiwix::render_template(const std::string& template_str, kainjow::mustache::data data)
+{
+  kainjow::mustache::mustache tmpl(template_str);
+  kainjow::mustache::data urlencode{kainjow::mustache::lambda2{
+                               [](const std::string& str,const kainjow::mustache::renderer& r) { return urlEncode(r(str), true); }}};
+  data.set("urlencoded", urlencode);
+  std::stringstream ss;
+  tmpl.render(data, [&ss](const std::string& str) { ss << str; });
+  return ss.str();
+}

src/tools/otherTools.h

@@ -24,6 +24,7 @@
 #include <vector>
 #include <map>
 #include <zim/zim.h>
+#include <mustache.hpp>
 
 namespace pugi {
   class xml_node;
@@ -31,9 +32,7 @@ namespace pugi {
 
 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
@@ -45,6 +44,11 @@ namespace kiwix
 
   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

src/tools/pathTools.cpp

@@ -17,7 +17,10 @@
  * MA 02110-1301, USA.
  */
 
+// Implement method defined in <kiwix/tools.h> and "tools/pathTools.h"
+#include "tools.h"
 #include "tools/pathTools.h"
+
 #include <stdexcept>
 
 #ifdef __APPLE__
@@ -59,7 +62,6 @@
 #define PATH_MAX 1024
 #endif
 
-
 #ifdef _WIN32
 std::string WideToUtf8(const std::wstring& wstr)
 {
@@ -78,7 +80,7 @@ std::wstring Utf8ToWide(const std::string& str)
 }
 #endif
 
-bool isRelativePath(const std::string& path)
+bool kiwix::isRelativePath(const std::string& path)
 {
 #ifdef _WIN32
   if (path.size() < 3 ) {
@@ -173,7 +175,7 @@ std::vector<std::string> normalizeParts(std::vector<std::string>& parts, bool ab
   return ret;
 }
 
-std::string computeRelativePath(const std::string& path, const std::string& absolutePath)
+std::string kiwix::computeRelativePath(const std::string& path, const std::string& absolutePath)
 {
   auto parts = kiwix::split(path, SEPARATOR, false);
   auto pathParts = normalizeParts(parts, false);
@@ -198,11 +200,11 @@ std::string computeRelativePath(const std::string& path, const std::string& abso
   return ret;
 }
 
-std::string computeAbsolutePath(const std::string& path, const std::string& relativePath)
+std::string kiwix::computeAbsolutePath(const std::string& path, const std::string& relativePath)
 {
   std::string absolutePath = path;
   if (path.empty()) {
-    absolutePath = getCurrentDirectory();
+    absolutePath = kiwix::getCurrentDirectory();
   }
 
   auto parts = kiwix::split(absolutePath, SEPARATOR, false);
@@ -215,7 +217,7 @@ std::string computeAbsolutePath(const std::string& path, const std::string& rela
   return ret;
 }
 
-std::string removeLastPathElement(const std::string& path)
+std::string kiwix::removeLastPathElement(const std::string& path)
 {
   auto parts_ = kiwix::split(path, SEPARATOR, false);
   auto parts = normalizeParts(parts_, false);
@@ -226,7 +228,7 @@ std::string removeLastPathElement(const std::string& path)
   return ret;
 }
 
-std::string appendToDirectory(const std::string& directoryPath, const std::string& filename)
+std::string kiwix::appendToDirectory(const std::string& directoryPath, const std::string& filename)
 {
   std::string newPath = directoryPath;
   if (!directoryPath.empty() && directoryPath.back() != SEPARATOR[0]) {
@@ -236,7 +238,7 @@ std::string appendToDirectory(const std::string& directoryPath, const std::strin
   return newPath;
 }
 
-std::string getLastPathElement(const std::string& path)
+std::string kiwix::getLastPathElement(const std::string& path)
 {
   auto parts_ = kiwix::split(path, SEPARATOR);
   auto parts = normalizeParts(parts_, false);
@@ -267,7 +269,7 @@ std::string getFileSizeAsString(const std::string& path)
   return convert.str();
 }
 
-std::string getFileContent(const std::string& path)
+std::string kiwix::getFileContent(const std::string& path)
 {
 #ifdef _WIN32
   auto wpath = Utf8ToWide(path);
@@ -300,19 +302,21 @@ std::string getFileContent(const std::string& path)
   return content;
 }
 
-bool fileExists(const std::string& path)
+bool kiwix::fileExists(const std::string& path)
 {
 #ifdef _WIN32
-  return PathFileExistsW(Utf8ToWide(path).c_str());
+  return (_waccess_s(Utf8ToWide(path).c_str(), 0) == 0);
 #else
-  bool flag = false;
-  std::fstream fin;
-  fin.open(path.c_str(), std::ios::in);
-  if (fin.is_open()) {
-    flag = true;
-  }
-  fin.close();
-  return flag;
+  return (access(path.c_str(), F_OK) == 0);
+#endif
+}
+
+bool kiwix::fileReadable(const std::string& path)
+{
+#ifdef _WIN32
+  return (_waccess_s(Utf8ToWide(path).c_str(), 4) == 0);
+#else
+  return (access(path.c_str(), R_OK) == 0);
 #endif
 }
 
@@ -339,7 +343,7 @@ std::string makeTmpDirectory()
   _wmkdir(ctmp);
   return WideToUtf8(ctmp);
 #else
-  char _template_array[] = {"/tmp/kiwix-lib_XXXXXX"};
+  char _template_array[] = {"/tmp/libkiwix_XXXXXX"};
   std::string dir = mkdtemp(_template_array);
   return dir;
 #endif
@@ -366,7 +370,7 @@ bool copyFile(const std::string& sourcePath, const std::string& destPath)
 #endif
 }
 
-std::string getExecutablePath(bool realPathOnly)
+std::string kiwix::getExecutablePath(bool realPathOnly)
 {
   if (!realPathOnly) {
     char* cAppImage = ::getenv("APPIMAGE");
@@ -420,7 +424,7 @@ bool writeTextFile(const std::string& path, const std::string& content)
   return true;
 }
 
-std::string getCurrentDirectory()
+std::string kiwix::getCurrentDirectory()
 {
 #ifdef _WIN32
   wchar_t* a_cwd = _wgetcwd(NULL, 0);
@@ -434,7 +438,7 @@ std::string getCurrentDirectory()
   return ret;
 }
 
-std::string getDataDirectory()
+std::string kiwix::getDataDirectory()
 {
 // Try to get the dataDir from the `KIWIX_DATA_DIR` env var
 #ifdef _WIN32
@@ -503,7 +507,7 @@ static std::map<std::string, std::string> extMimeTypes = {
 };
 
 /* Try to get the mimeType from the file extension */
-std::string getMimeTypeForFile(const std::string& filename)
+std::string kiwix::getMimeTypeForFile(const std::string& filename)
 {
   std::string mimeType = "text/plain";
   auto pos = filename.find_last_of(".");
@@ -524,4 +528,3 @@ std::string getMimeTypeForFile(const std::string& filename)
 
   return mimeType;
 }
-

src/tools/pathTools.h

@@ -26,23 +26,13 @@
 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
+

src/tools/stringTools.cpp

@@ -17,9 +17,11 @@
  * MA 02110-1301, USA.
  */
 
-#include <tools/stringTools.h>
+// Implement function declared in tools.h and tools/stringTools.h
+#include "tools.h"
+#include "tools/stringTools.h"
 
-#include <tools/pathTools.h>
+#include "tools/pathTools.h"
 #include <unicode/normlzr.h>
 #include <unicode/rep.h>
 #include <unicode/translit.h>
@@ -268,7 +270,7 @@ std::string kiwix::urlDecode(const std::string& value, bool component)
 /* Split string in a token array */
 std::vector<std::string> kiwix::split(const std::string& str,
                                       const std::string& delims,
-                                      bool trimEmpty,
+                                      bool dropEmpty,
                                       bool keepDelim)
 {
   std::string::size_type lastPos = 0;
@@ -277,7 +279,7 @@ std::vector<std::string> kiwix::split(const std::string& str,
   while( (pos = str.find_first_of(delims, lastPos)) < str.length() )
   {
     auto token = str.substr(lastPos, pos - lastPos);
-    if (!trimEmpty || !token.empty()) {
+    if (!dropEmpty || !token.empty()) {
       tokens.push_back(token);
     }
     if (keepDelim) {
@@ -287,7 +289,7 @@ std::vector<std::string> kiwix::split(const std::string& str,
   }
 
   auto token = str.substr(lastPos);
-  if (!trimEmpty || !token.empty()) {
+  if (!dropEmpty || !token.empty()) {
     tokens.push_back(token);
   }
   return tokens;
@@ -395,3 +397,11 @@ bool kiwix::startsWith(const std::string& base, const std::string& start)
         && std::equal(start.begin(), start.end(), base.begin());
 }
 
+std::vector<std::string> kiwix::getTitleVariants(const std::string& title) {
+  std::vector<std::string> variants;
+  variants.push_back(title);
+  variants.push_back(kiwix::ucFirst(title));
+  variants.push_back(kiwix::lcFirst(title));
+  variants.push_back(kiwix::toTitle(title));
+  return variants;
+}

src/tools/stringTools.h

@@ -43,7 +43,6 @@ 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& 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);
@@ -70,5 +69,7 @@ T extractFromString(const std::string& str) {
 }
 
 bool startsWith(const std::string& base, const std::string& start);
+
+std::vector<std::string> getTitleVariants(const std::string& title);
 } //namespace kiwix
 #endif

src/version.cpp

@@ -0,0 +1,77 @@
+/*
+ * Copyright 2021 Emmanuel Engelhart <kelson@kiwix.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * 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 <iostream>
+#include <sstream>
+
+#include <version.h>
+#include <zim/zim.h>
+#include <kiwix_config.h>
+#include <unicode/uversion.h>
+#include <pugixml.hpp>
+#include <curl/curl.h>
+#include <microhttpd.h>
+#include <xapian.h>
+#include <mustache.hpp>
+#include <zlib.h>
+
+namespace kiwix
+{
+  LibVersions getVersions() {
+    LibVersions versions = {
+      { "libkiwix",      LIBKIWIX_VERSION    },
+      { "libzim",        LIBZIM_VERSION      },
+      { "libxapian",     XAPIAN_VERSION      },
+      { "libcurl",       LIBCURL_VERSION     },
+      { "libmicrohttpd", MHD_get_version()   },
+      { "libz",          ZLIB_VERSION        }
+    };
+
+    // U_ICU_VERSION does not include the patch level if 0
+    std::ostringstream libicu_version;
+    libicu_version << U_ICU_VERSION_MAJOR_NUM << "." << U_ICU_VERSION_MINOR_NUM << "." << U_ICU_VERSION_PATCHLEVEL_NUM;
+    versions.push_back({ "libicu", libicu_version.str() });
+
+    // No human readable version string for pugixml
+    const unsigned pugixml_major = (PUGIXML_VERSION - PUGIXML_VERSION % 1000) / 1000;
+    const unsigned pugixml_minor = (PUGIXML_VERSION - pugixml_major * 1000 - PUGIXML_VERSION % 10) / 10;
+    const unsigned pugixml_patch = PUGIXML_VERSION - pugixml_major * 1000 - pugixml_minor * 10;
+    std::ostringstream libpugixml_version;
+    libpugixml_version << pugixml_major << "." << pugixml_minor << "." << pugixml_patch;
+    versions.push_back({ "libpugixml", libpugixml_version.str() });
+
+    // Needs version 5.0 of Mustache
+#if defined(KAINJOW_MUSTACHE_VERSION_MAJOR)
+    std::ostringstream libmustache_version;
+    libmustache_version << KAINJOW_MUSTACHE_VERSION_MAJOR << "." <<
+      KAINJOW_MUSTACHE_VERSION_MINOR << "." << KAINJOW_MUSTACHE_VERSION_PATCH;
+    versions.push_back({ "libmustache", libmustache_version.str() });
+#endif
+
+    return versions;
+  }
+
+  void printVersions(std::ostream& out) {
+    LibVersions versions = getVersions();
+    for (const auto& iter : versions) {
+      out << (iter != versions.front() ? "+ " : "")
+                << iter.first << " " << iter.second << std::endl;
+    }
+  }
+} //namespace kiwix

src/wrapper/java/book.cpp

@@ -69,6 +69,7 @@ GETTER(jstring, getDate)
 GETTER(jstring, getUrl)
 GETTER(jstring, getName)
 GETTER(jstring, getFlavour)
+GETTER(jstring, getCategory)
 GETTER(jstring, getTags)
 GETTER(jlong, getArticleCount)
 GETTER(jlong, getMediaCount)

src/wrapper/java/kiwixreader.cpp

@@ -45,6 +45,72 @@ JNIEXPORT jlong JNICALL Java_org_kiwix_kiwixlib_JNIKiwixReader_getNativeReader(
   }
 }
 
+namespace
+{
+
+int jni2fd(const jobject& fdObj, JNIEnv* env)
+{
+  jclass class_fdesc = env->FindClass("java/io/FileDescriptor");
+  jfieldID field_fd = env->GetFieldID(class_fdesc, "fd", "I");
+  if ( field_fd == NULL )
+  {
+    env->ExceptionClear();
+    // Under Android the (private) 'fd' field of java.io.FileDescriptor has been
+    // renamed to 'descriptor'. See, for example,
+    // https://android.googlesource.com/platform/libcore/+/refs/tags/android-8.1.0_r1/ojluni/src/main/java/java/io/FileDescriptor.java#55
+    field_fd = env->GetFieldID(class_fdesc, "descriptor", "I");
+  }
+  return env->GetIntField(fdObj, field_fd);
+}
+
+} // unnamed namespace
+
+JNIEXPORT jlong JNICALL Java_org_kiwix_kiwixlib_JNIKiwixReader_getNativeReaderByFD(
+    JNIEnv* env, jobject obj, jobject fdObj)
+{
+#ifndef _WIN32
+  int fd = jni2fd(fdObj, env);
+
+  LOG("Attempting to create reader with fd: %d", fd);
+  Lock l;
+  try {
+    kiwix::Reader* reader = new kiwix::Reader(fd);
+    return reinterpret_cast<jlong>(new Handle<kiwix::Reader>(reader));
+  } catch (std::exception& e) {
+    LOG("Error opening ZIM file");
+    LOG(e.what());
+    return 0;
+  }
+#else
+  jclass exception = env->FindClass("java/lang/UnsupportedOperationException");
+  env->ThrowNew(exception, "org.kiwix.kiwixlib.JNIKiwixReader.getNativeReaderByFD() is not supported under Windows");
+  return 0;
+#endif
+}
+
+JNIEXPORT jlong JNICALL Java_org_kiwix_kiwixlib_JNIKiwixReader_getNativeReaderEmbedded(
+    JNIEnv* env, jobject obj, jobject fdObj, jlong offset, jlong size)
+{
+#ifndef _WIN32
+  int fd = jni2fd(fdObj, env);
+
+  LOG("Attempting to create reader with fd: %d", fd);
+  Lock l;
+  try {
+    kiwix::Reader* reader = new kiwix::Reader(fd, offset, size);
+    return reinterpret_cast<jlong>(new Handle<kiwix::Reader>(reader));
+  } catch (std::exception& e) {
+    LOG("Error opening ZIM file");
+    LOG(e.what());
+    return 0;
+  }
+#else
+  jclass exception = env->FindClass("java/lang/UnsupportedOperationException");
+  env->ThrowNew(exception, "org.kiwix.kiwixlib.JNIKiwixReader.getNativeReaderEmbedded() is not supported under Windows");
+  return 0;
+#endif
+}
+
 JNIEXPORT void JNICALL
 Java_org_kiwix_kiwixlib_JNIKiwixReader_dispose(JNIEnv* env, jobject obj)
 {
@@ -325,22 +391,22 @@ JNIEXPORT jobject JNICALL
 Java_org_kiwix_kiwixlib_JNIKiwixReader_getDirectAccessInformation(
     JNIEnv* env, jobject obj, jstring url)
 {
-   jclass classPair = env->FindClass("org/kiwix/kiwixlib/Pair");
-   jmethodID midPairinit = env->GetMethodID(classPair, "<init>", "()V");
-   jobject pair = env->NewObject(classPair, midPairinit);
-   setPairObjValue("", 0, pair, env);
+   jclass daiClass = env->FindClass("org/kiwix/kiwixlib/DirectAccessInfo");
+   jmethodID daiInitMethod = env->GetMethodID(daiClass, "<init>", "()V");
+   jobject dai = env->NewObject(daiClass, daiInitMethod);
+   setDaiObjValue("", 0, dai, env);
 
    std::string cUrl = jni2c(url, env);
    try {
     auto entry = READER->getEntryFromEncodedPath(cUrl);
     entry = entry.getFinalEntry();
     auto part_info = entry.getDirectAccessInfo();
-    setPairObjValue(part_info.first, part_info.second, pair, env);
+    setDaiObjValue(part_info.first, part_info.second, dai, env);
   } catch (std::exception& e) {
     LOG("Unable to get direct access info for url: %s", cUrl.c_str());
     LOG(e.what());
   }
-  return pair;
+  return dai;
 }
 
 JNIEXPORT jboolean JNICALL

src/wrapper/java/meson.build

@@ -12,7 +12,7 @@ java_sources = files([
   'org/kiwix/kiwixlib/JNIKiwixString.java',
   'org/kiwix/kiwixlib/JNIKiwixBool.java',
   'org/kiwix/kiwixlib/JNIKiwixException.java',
-  'org/kiwix/kiwixlib/Pair.java'
+  'org/kiwix/kiwixlib/DirectAccessInfo.java'
 ])
 
 kiwix_jni = custom_target('jni',

src/wrapper/java/org/kiwix/kiwixlib/Book.java

@@ -24,6 +24,7 @@ public class Book
   public native String getUrl();
   public native String getName();
   public native String getFlavour();
+  public native String getCategory();
   public native String getTags();
   /**
    * Return the value associated to the tag tagName

src/wrapper/java/org/kiwix/kiwixlib/DirectAccessInfo.java

@@ -19,7 +19,7 @@
 
 package org.kiwix.kiwixlib;
 
-public class Pair
+public class DirectAccessInfo
 {
   public String filename;
   public long offset;

src/wrapper/java/org/kiwix/kiwixlib/JNIKiwixReader.java

@@ -24,7 +24,8 @@ import org.kiwix.kiwixlib.JNIKiwixException;
 import org.kiwix.kiwixlib.JNIKiwixString;
 import org.kiwix.kiwixlib.JNIKiwixInt;
 import org.kiwix.kiwixlib.JNIKiwixSearcher;
-import org.kiwix.kiwixlib.Pair;
+import org.kiwix.kiwixlib.DirectAccessInfo;
+import java.io.FileDescriptor;
 
 public class JNIKiwixReader
 {
@@ -102,13 +103,13 @@ public class JNIKiwixReader
    * the zim file (or zim part) and directly read the content from it (and so
    * bypassing the libzim).
    *
-   * Return a `Pair` (filename, offset) where the content is located.
+   * Return a `DirectAccessInfo` (filename, offset) where the content is located.
    *
    * If the content cannot be directly accessed (content is compressed or zim
    * file is cut in the middle of the content), the filename is an empty string
    * and offset is zero.
    */
-  public native Pair getDirectAccessInformation(String url);
+  public native DirectAccessInfo getDirectAccessInformation(String url);
 
   public native boolean searchSuggestions(String prefix, int count);
 
@@ -151,11 +152,31 @@ public class JNIKiwixReader
         throw new JNIKiwixException("Cannot open zimfile "+filename);
     }
   }
+
+  public JNIKiwixReader(FileDescriptor fd) throws JNIKiwixException
+  {
+    nativeHandle = getNativeReaderByFD(fd);
+    if (nativeHandle == 0) {
+        throw new JNIKiwixException("Cannot open zimfile by fd "+fd.toString());
+    }
+  }
+
+  public JNIKiwixReader(FileDescriptor fd, long offset, long size)
+      throws JNIKiwixException
+  {
+    nativeHandle = getNativeReaderEmbedded(fd, offset, size);
+    if (nativeHandle == 0) {
+        throw new JNIKiwixException(String.format("Cannot open embedded zimfile (fd=%s, offset=%d, size=%d)", fd, offset, size));
+    }
+  }
+
   public JNIKiwixReader() {
 
   }
   public native void dispose();
 
   private native long getNativeReader(String filename);
+  private native long getNativeReaderByFD(FileDescriptor fd);
+  private native long getNativeReaderEmbedded(FileDescriptor fd, long offset, long size);
   private long nativeHandle;
 }

src/wrapper/java/org/kiwix/testing/catalog.xml

@@ -0,0 +1,19 @@
+<feed xmlns="http://www.w3.org/2005/Atom" xmlns:opds="http://opds-spec.org/2010/catalog">
+  <id>00000000-0000-0000-0000-000000000000</id>
+  <entry>
+    <title>Test ZIM file</title>
+    <id>urn:uuid:86c91e51-55bf-8882-464e-072aca37a3e8</id>
+    <icon>/meta?name=favicon&amp;content=small</icon>
+    <updated>2020-11-27:00::00:Z</updated>
+    <language>en</language>
+    <summary>This is a ZIM file used in libzim unit-tests</summary>
+    <tags>unit;test</tags>
+    <link type="text/html" href="/small" />
+    <author>
+      <name>Kiwix</name>
+    </author>
+    <link rel="http://opds-spec.org/acquisition/open-access" type="application/x-zim" href="http://localhost/small.zim" length="78982" />
+    <link rel="http://opds-spec.org/image/thumbnail" type="image/png" href="/meta?name=favicon&amp;content=small" />
+  </entry>
+</feed>
+

src/wrapper/java/org/kiwix/testing/compile_and_run_test.sh

@@ -0,0 +1,37 @@
+#!/usr/bin/bash
+
+# This script compiles and runs the unit test to test the java wrapper.
+# This is not integrated in meson because ... this is not so easy.
+
+die()
+{
+  echo >&2 "!!! ERROR: $*"
+  exit 1
+}
+
+
+KIWIX_LIB_JAR=$1
+if [ -z $KIWIX_LIB_JAR ]
+then
+  die "You must give the path to the kiwixlib.jar as first argument"
+fi
+
+KIWIX_LIB_DIR=$2
+if [ -z $KIWIX_LIB_DIR ]
+then
+  die "You must give the path to directory containing libkiwix.so as second argument"
+fi
+
+KIWIX_LIB_JAR=$(readlink -f "$KIWIX_LIB_JAR")
+KIWIX_LIB_DIR=$(readlink -f "$KIWIX_LIB_DIR")
+TEST_SOURCE_DIR=$(dirname "$(readlink -f $0)")
+
+cd "$TEST_SOURCE_DIR"
+
+javac -g -d . -s . -cp "junit-4.13.jar:$KIWIX_LIB_JAR" test.java \
+  || die "Compilation failed"
+
+java -Djava.library.path="$KIWIX_LIB_DIR" \
+     -cp "junit-4.13.jar:hamcrest-core-1.3.jar:$KIWIX_LIB_JAR:." \
+     org.junit.runner.JUnitCore test \
+  || die "Unit test failed"

src/wrapper/java/org/kiwix/testing/compile_test.sh

@@ -1,26 +0,0 @@
-#!/usr/bin/bash
-
-# This script compile the unit test to test the java wrapper.
-# This is not integrated in meson because ... this is not so easy.
-
-
-KIWIX_LIB_JAR=$1
-if [ -z $KIWIX_LIB_JAR ]
-then
-  echo "You must give the path to the kiwixlib.jar as first argument"
-  exit 1
-fi
-
-KIWIX_LIB_DIR=$2
-if [ -z $KIWIX_LIB_DIR ]
-then
-  echo "You must give the path to directory containing libkiwix.so as second argument"
-  exit 1
-fi
-TEST_SOURCE_DIR=$(dirname $(readlink -f $0))
-
-
-javac -g -d . -s . -cp $TEST_SOURCE_DIR/junit-4.13.jar:$KIWIX_LIB_JAR $TEST_SOURCE_DIR/test.java
-
-java -Djava.library.path=$KIWIX_LIB_DIR -cp $TEST_SOURCE_DIR/junit-4.13.jar:$TEST_SOURCE_DIR/hamcrest-core-1.3.jar:$KIWIX_LIB_JAR:. org.junit.runner.JUnitCore test
-

src/wrapper/java/org/kiwix/testing/create_test_zimfiles

@@ -0,0 +1,28 @@
+#!/usr/bin/env bash
+
+die()
+{
+  echo >&2 "!!! ERROR: $*"
+  exit 1
+}
+
+cd "$(dirname "$0")"
+rm -f small.zim
+zimwriterfs --withoutFTIndex \
+            -w main.html \
+            -f favicon.png \
+            -l en \
+            -t "Test ZIM file" \
+            -d "N/A" \
+            -c "N/A" \
+            -p "N/A" \
+            small_zimfile_data \
+            small.zim \
+&& echo 'small.zim was successfully created' \
+|| die 'Failed to create small.zim'
+
+printf "BEGINZIM" > small.zim.embedded \
+&& cat small.zim >> small.zim.embedded \
+&& printf "ENDZIM" >> small.zim.embedded \
+&& echo 'small.zim.embedded was successfully created' \
+|| die 'Failed to create small.zim.embedded'

src/wrapper/java/org/kiwix/testing/small_zimfile_data/main.html

@@ -0,0 +1,11 @@
+<html>
+  <head>
+    <meta charset="UTF-8">
+    <title>Test ZIM file</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  </head>
+
+  <body>
+    Test ZIM file
+  </body>
+</html>

src/wrapper/java/org/kiwix/testing/test.java

@@ -10,36 +10,143 @@ static {
   System.loadLibrary("kiwix");
 }
 
-private static String getCatalogContent()
+private static byte[] getFileContent(String path)
 throws IOException
 {
-  BufferedReader reader = new BufferedReader(new FileReader("catalog.xml"));
-  String line;
-  StringBuilder sb = new StringBuilder();
-  while ((line = reader.readLine()) != null)
-  {
-    sb.append(line + "\n");
-  }
-  reader.close();
-  return sb.toString();
+  File file = new File(path);
+  DataInputStream in = new DataInputStream(
+                         new BufferedInputStream(
+                           new FileInputStream(file)));
+  byte[] data = new byte[(int)file.length()];
+  in.read(data);
+  return data;
+}
+
+private static byte[] getFileContentPartial(String path, int offset, int size)
+throws IOException
+{
+  File file = new File(path);
+  DataInputStream in = new DataInputStream(
+                         new BufferedInputStream(
+                           new FileInputStream(file)));
+  byte[] data = new byte[size];
+  in.skipBytes(offset);
+  in.read(data, 0, size);
+  return data;
+}
+
+private static String getTextFileContent(String path)
+throws IOException
+{
+  return new String(getFileContent(path));
 }
 
 @Test
-public void testSome()
+public void testReader()
+throws JNIKiwixException, IOException
+{
+  JNIKiwixReader reader = new JNIKiwixReader("small.zim");
+  assertEquals("Test ZIM file", reader.getTitle());
+  assertEquals(45, reader.getFileSize()); // The file size is in KiB
+  assertEquals("A/main.html", reader.getMainPage());
+  String s = getTextFileContent("small_zimfile_data/main.html");
+  byte[] c = reader.getContent(new JNIKiwixString("A/main.html"),
+                                new JNIKiwixString(),
+                                new JNIKiwixString(),
+                                new JNIKiwixInt());
+  assertEquals(s, new String(c));
+
+  byte[] faviconData = getFileContent("small_zimfile_data/favicon.png");
+  assertEquals(faviconData.length, reader.getArticleSize("I/favicon.png"));
+  c = reader.getContent(new JNIKiwixString("I/favicon.png"),
+                                new JNIKiwixString(),
+                                new JNIKiwixString(),
+                                new JNIKiwixInt());
+  assertTrue(Arrays.equals(faviconData, c));
+
+  DirectAccessInfo dai = reader.getDirectAccessInformation("I/favicon.png");
+  assertNotEquals("", dai.filename);
+  c = getFileContentPartial(dai.filename, (int)dai.offset, faviconData.length);
+  assertTrue(Arrays.equals(faviconData, c));
+}
+
+@Test
+public void testReaderByFd()
+throws JNIKiwixException, IOException
+{
+  FileInputStream fis = new FileInputStream("small.zim");
+  JNIKiwixReader reader = new JNIKiwixReader(fis.getFD());
+  assertEquals("Test ZIM file", reader.getTitle());
+  assertEquals(45, reader.getFileSize()); // The file size is in KiB
+  assertEquals("A/main.html", reader.getMainPage());
+  String s = getTextFileContent("small_zimfile_data/main.html");
+  byte[] c = reader.getContent(new JNIKiwixString("A/main.html"),
+                                new JNIKiwixString(),
+                                new JNIKiwixString(),
+                                new JNIKiwixInt());
+  assertEquals(s, new String(c));
+
+  byte[] faviconData = getFileContent("small_zimfile_data/favicon.png");
+  assertEquals(faviconData.length, reader.getArticleSize("I/favicon.png"));
+  c = reader.getContent(new JNIKiwixString("I/favicon.png"),
+                                new JNIKiwixString(),
+                                new JNIKiwixString(),
+                                new JNIKiwixInt());
+  assertTrue(Arrays.equals(faviconData, c));
+
+  DirectAccessInfo dai = reader.getDirectAccessInformation("I/favicon.png");
+  assertNotEquals("", dai.filename);
+  c = getFileContentPartial(dai.filename, (int)dai.offset, faviconData.length);
+  assertTrue(Arrays.equals(faviconData, c));
+}
+
+@Test
+public void testReaderWithAnEmbeddedArchive()
+throws JNIKiwixException, IOException
+{
+  File plainArchive = new File("small.zim");
+  FileInputStream fis = new FileInputStream("small.zim.embedded");
+  JNIKiwixReader reader = new JNIKiwixReader(fis.getFD(), 8, plainArchive.length());
+  assertEquals("Test ZIM file", reader.getTitle());
+  assertEquals(45, reader.getFileSize()); // The file size is in KiB
+  assertEquals("A/main.html", reader.getMainPage());
+  String s = getTextFileContent("small_zimfile_data/main.html");
+  byte[] c = reader.getContent(new JNIKiwixString("A/main.html"),
+                                new JNIKiwixString(),
+                                new JNIKiwixString(),
+                                new JNIKiwixInt());
+  assertEquals(s, new String(c));
+
+  byte[] faviconData = getFileContent("small_zimfile_data/favicon.png");
+  assertEquals(faviconData.length, reader.getArticleSize("I/favicon.png"));
+  c = reader.getContent(new JNIKiwixString("I/favicon.png"),
+                                new JNIKiwixString(),
+                                new JNIKiwixString(),
+                                new JNIKiwixInt());
+  assertTrue(Arrays.equals(faviconData, c));
+
+  DirectAccessInfo dai = reader.getDirectAccessInformation("I/favicon.png");
+  assertNotEquals("", dai.filename);
+  c = getFileContentPartial(dai.filename, (int)dai.offset, faviconData.length);
+  assertTrue(Arrays.equals(faviconData, c));
+}
+
+@Test
+public void testLibrary()
 throws IOException
 {
   Library lib = new Library();
   Manager manager = new Manager(lib);
-  String content = getCatalogContent();
-  manager.readOpds(content, "https://library.kiwix.org");
-  assertEquals(lib.getBookCount(true, true), 10);
+  String content = getTextFileContent("catalog.xml");
+  manager.readOpds(content, "http://localhost");
+  assertEquals(lib.getBookCount(true, true), 1);
   String[] bookIds = lib.getBooksIds();
-  assertEquals(bookIds.length, 10);
+  assertEquals(bookIds.length, 1);
   Book book = lib.getBookById(bookIds[0]);
-  assertEquals(book.getTitle(), "Wikisource");
-  assertEquals(book.getTags(), "wikisource;_category:wikisource;_pictures:no;_videos:no;_details:yes;_ftindex:yes");
-  assertEquals(book.getFaviconUrl(), "https://library.kiwix.org/meta?name=favicon&content=wikisource_fr_all_nopic_2020-01");
-  assertEquals(book.getUrl(), "http://download.kiwix.org/zim/wikisource/wikisource_fr_all_nopic_2020-01.zim.meta4");
+  assertEquals(book.getTitle(), "Test ZIM file");
+  assertEquals(book.getTags(), "unit;test");
+  assertEquals(book.getFaviconUrl(), "http://localhost/meta?name=favicon&content=small");
+  assertEquals(book.getUrl(), "http://localhost/small.zim");
 }
 
 static

src/wrapper/java/utils.h

@@ -246,7 +246,7 @@ inline void setBoolObjValue(const bool value, const jobject obj, JNIEnv* env)
   env->SetIntField(obj, objFid, c2jni(value, env));
 }
 
-inline void setPairObjValue(const std::string& filename, const long offset,
+inline void setDaiObjValue(const std::string& filename, const long offset,
                             const jobject obj, JNIEnv* env)
 {
   jclass objClass = env->GetObjectClass(obj);

static/catalog_v2_searchdescription.xml

@@ -0,0 +1,10 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<OpenSearchDescription xmlns="http://a9.com/-/spec/opensearch/1.1/">
+  <ShortName>Zim catalog search</ShortName>
+  <Description>Search zim files in the catalog.</Description>
+  <Url type="application/atom+xml;profile=opds-catalog;kind=acquisition"
+       xmlns:atom="http://www.w3.org/2005/Atom"
+       xmlns:k="http://kiwix.org/opensearchextension/1.0"
+       indexOffset="0"
+       template="{{endpoint_root}}/entries?q={searchTerms?}&lang={language?}&name={k:name?}&tag={k:tag?}&maxsize={k:maxsize?}&count={count?}&start={startIndex?}"/>
+</OpenSearchDescription>

static/opensearchdescription.xml

@@ -6,5 +6,5 @@
        xmlns:atom="http://www.w3.org/2005/Atom"
        xmlns:k="http://kiwix.org/opensearchextension/1.0"
        indexOffset="0"
-       template="/{{root}}/catalog/search?q={searchTerms?}&lang={language?}&name={k:name?}&tag={k:tag?}&notag={k:notag?}&maxsize={k:maxsize?}&count={count?}&start={startIndex?}"/>
+       template="{{root}}/catalog/search?q={searchTerms?}&lang={language?}&name={k:name?}&tag={k:tag?}&notag={k:notag?}&maxsize={k:maxsize?}&count={count?}&start={startIndex?}"/>
 </OpenSearchDescription>

static/resources_list.txt

@@ -18,17 +18,36 @@ skin/jquery-ui/images/ui-bg_glass_95_fef1ec_1x400.png
 skin/jquery-ui/jquery-ui.theme.min.css
 skin/jquery-ui/jquery-ui.min.css
 skin/caret.png
+skin/bittorrent.png
+skin/magnet.png
+skin/download.png
+skin/hash.png
+skin/search-icon.svg
 skin/taskbar.js
+skin/iso6391To3.js
+skin/isotope.pkgd.min.js
+skin/index.js
 skin/taskbar.css
+skin/index.css
+skin/fonts/Poppins.ttf
+skin/fonts/Roboto.ttf
 skin/block_external.js
+skin/search_results.css
 templates/search_result.html
 templates/no_search_result.html
 templates/404.html
 templates/500.html
 templates/index.html
 templates/suggestion.json
-templates/head_part.html
+templates/head_taskbar.html
 templates/taskbar_part.html
 templates/external_blocker_part.html
 templates/captured_external.html
+templates/catalog_entries.xml
+templates/catalog_v2_root.xml
+templates/catalog_v2_entries.xml
+templates/catalog_v2_entry.xml
+templates/catalog_v2_categories.xml
+templates/catalog_v2_languages.xml
 opensearchdescription.xml
+catalog_v2_searchdescription.xml

static/skin/block_external.js

@@ -1,5 +1,6 @@
+const root = document.querySelector( `link[type='root']` ).getAttribute("href");
 // `block_path` variable used by openzim/warc2zim to detect whether URL blocking is enabled or not
-var block_path = "/catch/external";
+var block_path = `${root}/catch/external`;
 // called only on external links
 function capture_event(e, target) { target.setAttribute("href", encodeURI(block_path + "?source=" + target.href)); }
 

static/skin/index.js

@@ -0,0 +1,438 @@
+(function() {
+    const root = $(`link[type='root']`).attr('href');
+    const incrementalLoadingParams = {
+        start: 0,
+        count: viewPortToCount()
+    };
+    const bookOrderMap = new Map();
+    const filterCookieName = 'filters';
+    const oneDayDelta = 86400000;
+    let loader;
+    let footer;
+    let fadeOutDiv;
+    let iso;
+    let isFetching = false;
+    let noResultInjected = false;
+    let filters = getCookie(filterCookieName);
+    let params = new URLSearchParams(window.location.search || filters || '');
+    let timer;
+    let languages = {};
+
+    function queryUrlBuilder() {
+        let url = `${root}/catalog/search?`;
+        url += Object.keys(incrementalLoadingParams).map(key => `${key}=${incrementalLoadingParams[key]}`).join("&");
+        params.forEach((value, key) => {url+= value ? `&${key}=${value}` : ''});
+        return (url);
+    }
+
+    function setCookie(cookieName, cookieValue) {
+        const date = new Date();
+        date.setTime(date.getTime() + oneDayDelta);
+        document.cookie = `${cookieName}=${cookieValue};expires=${date.toUTCString()};sameSite=Strict`;
+    }
+
+    function getCookie(cookieName) {
+        const name = cookieName + "=";
+        let result;
+        decodeURIComponent(document.cookie).split('; ').forEach(val => {
+            if (val.indexOf(name) === 0) {
+                result = val.substring(name.length);
+            }
+        });
+        return result;
+    }
+
+    const humanFriendlySize = (fileSize) => {
+        if (fileSize === 0) {
+            return '';
+        }
+        const units = ['bytes', 'kB', 'MB', 'GB', 'TB'];
+        let quotient = Math.floor(Math.log10(fileSize) / 3);
+        quotient = quotient < units.length ? quotient : units.length - 1;
+        fileSize /= (1000 ** quotient);
+        return `${+fileSize.toFixed(2)} ${units[quotient]}`;
+    };
+
+    const humanFriendlyTitle = (title) => {
+        if (typeof title === 'string' && title.length > 0) {
+            title = title.replace(/_/g, ' ');
+            if (title.length > 0) {
+                return htmlEncode(title[0].toUpperCase() + title.slice(1));
+            }
+        }
+        return '';
+    }
+
+    function htmlEncode(str) {
+        return str.replace(/[\u00A0-\u9999<>\&]/gim, (i) => `&#${i.charCodeAt(0)};`);
+    }
+
+    function viewPortToCount(){
+        const zoom = Math.floor((( window.outerWidth - 10 ) / window.innerWidth) * 100);
+        return Math.floor(window.innerHeight/(3*zoom) + 1)*(window.innerWidth/(2.5*zoom) + 1);
+    }
+
+    function getInnerHtml(node, query) {
+        const queryNode = node.querySelector(query);
+        return queryNode != null ? queryNode.innerHTML : "";
+    }
+
+    function generateBookHtml(book, sort = false) {
+        const link = book.querySelector('link[type="text/html"]').getAttribute('href');
+        let iconUrl;
+        book.querySelectorAll('link[rel="http://opds-spec.org/image/thumbnail"]').forEach(link => {
+            if (link.getAttribute('type').split(';')[1] == 'width=48' && !iconUrl) {
+                iconUrl = link.getAttribute('href');
+            }
+        });
+        const title =  getInnerHtml(book, 'title');
+        const description = getInnerHtml(book, 'summary');
+        const id = getInnerHtml(book, 'id');
+        const langCode = getInnerHtml(book, 'language');
+        const language = languages[langCode];
+        const tags = getInnerHtml(book, 'tags');
+        let tagHtml = tags.split(';').filter(tag => {return !(tag.split(':')[0].startsWith('_'))})
+            .map((tag) => {return tag.charAt(0).toUpperCase() + tag.slice(1)})
+            .join(' | ').replace(/_/g, ' ');
+        let downloadLink;
+        let zimSize = 0;
+        try {
+            const downloadBookLink = book.querySelector('link[type="application/x-zim"]')
+            zimSize = parseInt(downloadBookLink.getAttribute('length'));
+            downloadLink = downloadBookLink.getAttribute('href').split('.meta4')[0];
+        } catch {
+            downloadLink = '';
+        }
+        const humanFriendlyZimSize = humanFriendlySize(zimSize);
+
+        const divTag = document.createElement('div');
+        divTag.setAttribute('class', 'book');
+        divTag.setAttribute('data-id', id);
+        if (sort) {
+            divTag.setAttribute('data-idx', bookOrderMap.get(id));
+        }
+        const faviconAttr = iconUrl != undefined ? `style="background-image: url('${iconUrl}')"` : '';
+        const languageAttr = langCode != '' ? `title="${language}" aria-label="${language}"` : 'style="background-color: transparent"';
+        divTag.innerHTML = `<a class="book__link" href="${link}" data-hover="Preview">
+            <div class="book__wrapper">
+            <div class="book__icon" ${faviconAttr}></div>
+            <div class="book__header">
+                <div id="book__title">${title}</div>
+                ${downloadLink ? `<div class="book__download"><span data-link="${downloadLink}">Download ${humanFriendlyZimSize ? ` - ${humanFriendlyZimSize}</span></div>`: ''}` : ''}
+            </div>
+            <div class="book__description" title="${description}">${description}</div>
+            <div class="book__languageTag" ${languageAttr}>${getLanguageCodeToDisplay(langCode)}</div>
+            <div class="book__tags"><div class="book__tags--wrapper">${tagHtml}</div></div>
+            </div></div></a>`;
+        return divTag;
+    }
+
+    function getLanguageCodeToDisplay(langCode3Letter) {
+        const langCode2Letter = (Object.keys(iso6391To3).find(key => iso6391To3[key] === langCode3Letter));
+        const res = (langCode2Letter != undefined) ? langCode2Letter : langCode3Letter;
+        return res.toUpperCase();
+    }
+
+    function toggleFooter(show=false) {
+        if (show) {
+            footer.style.display = 'block';
+        } else {
+            footer.style.display = 'none';
+            fadeOutDiv.style.display = 'block';
+        }
+    }
+
+    function insertModal(button) {
+        const downloadLink = button.getAttribute('data-link');
+        button.addEventListener('click', (event) => {
+            event.preventDefault();
+            document.body.insertAdjacentHTML('beforeend', `<div class="modal-wrapper">
+                <div class="modal">
+                    <div class="modal-heading">
+                        <div class="modal-title">
+                            <div>
+                                Download
+                            </div>
+                        </div>
+                        <div onclick="closeModal()" class="modal-close-button">
+                            <div>
+                                <svg xmlns="http://www.w3.org/2000/svg" width="14" height="14" viewBox="0 0 14 14" fill="none">
+                                    <path fill-rule="evenodd" clip-rule="evenodd" d="M13.7071 1.70711C14.0976 1.31658 14.0976 
+                                    0.683417 13.7071 0.292893C13.3166 -0.0976311 12.6834 -0.0976311 12.2929 0.292893L7 5.58579L1.70711
+                                    0.292893C1.31658 -0.0976311 0.683417 -0.0976311 0.292893 0.292893C-0.0976311 0.683417
+                                    -0.0976311 1.31658 0.292893 1.70711L5.58579 7L0.292893 12.2929C-0.0976311 12.6834
+                                    -0.0976311 13.3166 0.292893 13.7071C0.683417 14.0976 1.31658 14.0976 1.70711 13.7071L7
+                                    8.41421L12.2929 13.7071C12.6834 14.0976 13.3166 14.0976 13.7071 13.7071C14.0976 13.3166 
+                                    14.0976 12.6834 13.7071 12.2929L8.41421 7L13.7071 1.70711Z" fill="black" />
+                                </svg>
+                            </div>
+                        </div>
+                    </div>
+                    <div class="modal-content">
+                        <div class="modal-regular-download">
+                            <a href="${downloadLink}" download>
+                                <img src="../skin/download.png" alt="direct download" />
+                                <div>Direct</div>
+                            </a>
+                        </div>
+                        <div class="modal-regular-download">
+                            <a href="${downloadLink}.sha256" download>
+                                <img src="../skin/hash.png" alt="download hash" />
+                                <div>Sha256 hash</div>
+                            </a>
+                        </div>
+                        <div class="modal-regular-download">
+                            <a href="${downloadLink}.magnet" target="_blank">
+                                <img src="../skin/magnet.png" alt="download magnet" />
+                                <div>Magnet link</div>
+                            </a>
+                        </div>
+                        <div class="modal-regular-download">
+                            <a href="${downloadLink}.torrent" download>
+                                <img src="../skin/bittorrent.png" alt="download torrent" />
+                                <div>Torrent file</div>
+                            </a>
+                        </div>
+                    </div>
+                </div>
+            </div>`);
+        })
+    }
+
+    async function getBookCount(query) {
+        const url = `${root}/catalog/search?${query}`;
+        return await fetch(url).then(async (resp) => {
+            const data = new window.DOMParser().parseFromString(await resp.text(), 'application/xml');
+            return parseInt(data.querySelector('totalResults').innerHTML);
+        });
+    }
+
+    async function loadBooks() {
+        loader.style.display = 'block';
+        return await fetch(queryUrlBuilder()).then(async (resp) => {
+            const data = new window.DOMParser().parseFromString(await resp.text(), 'application/xml');
+            const books = data.querySelectorAll('entry');
+            books.forEach((book, idx) => {
+                bookOrderMap.set(getInnerHtml(book, 'id'), idx);
+            });
+            incrementalLoadingParams.start += books.length;
+            const results = parseInt(data.querySelector('totalResults').innerHTML)
+            if (results === bookOrderMap.size) {
+                incrementalLoadingParams.count = 0;
+                toggleFooter(true);
+            } else {
+                toggleFooter();
+            }
+            const kiwixResultText = document.querySelector('.kiwixHomeBody__results')
+            if (results) {
+                let resultText = `${results} books`;
+                if (results === 1) {
+                    resultText = `${results} book`;
+                }
+                kiwixResultText.innerHTML = resultText;
+            } else {
+                kiwixResultText.innerHTML = ``;
+            }
+            loader.style.display = 'none';
+            return books;
+        });
+    }
+
+    async function loadAndDisplayOptions(nodeQuery, query, valueEntryNode) {
+        await fetch(query).then(async (resp) => {
+            const data = new window.DOMParser().parseFromString(await resp.text(), 'application/xml');
+            let optionStr = '';
+            data.querySelectorAll('entry').forEach(entry => {
+                const title = getInnerHtml(entry, 'title');
+                const value = getInnerHtml(entry, valueEntryNode);
+                const hfTitle = humanFriendlyTitle(title);
+                if (valueEntryNode == 'language') {
+                    languages[value] = hfTitle;
+                }
+                optionStr += (hfTitle != '') ? `<option value="${value}">${hfTitle}</option>` : '';
+            });
+            document.querySelector(nodeQuery).innerHTML += optionStr;
+        });
+    }
+
+    function checkAndInjectEmptyMessage() {
+        const kiwixHomeBody = document.querySelector('.kiwixHomeBody');
+        if (!bookOrderMap.size) {
+            if (!noResultInjected) {
+                noResultInjected = true;
+                iso.remove(document.getElementsByClassName('book__list')[0].getElementsByTagName('div'));
+                iso.layout();
+                setTimeout(() => {
+                    const divTag = document.createElement('div');
+                    divTag.setAttribute('class', 'noResults');
+                    divTag.innerHTML = `No result. Would you like to <a href="/?lang=">reset filter</a>?`;
+                    kiwixHomeBody.append(divTag);
+                    kiwixHomeBody.setAttribute('style', 'display: flex; justify-content: center; align-items: center');
+                    divTag.getElementsByTagName('a')[0].onclick = (event) => {
+                        event.preventDefault();
+                        window.history.pushState({}, null, `${window.location.href.split('?')[0]}?lang=`);
+                        setCookie(filterCookieName, 'lang=');
+                        resetAndFilter();
+                        document.querySelectorAll('.filter').forEach(filter => {
+                            filter.value = params.get(filter.name) || '';
+                            if (filter.value) {
+                                filter.style = 'background-color: #858585; color: #fff';
+                            } else {
+                                filter.style = 'background-color: #ffffff; color: black';
+                            }
+                        })
+                    };
+                    loader.setAttribute('style', 'position: absolute; top: 50%');
+                }, 300);
+            }
+            return true;
+        } else if (noResultInjected) {
+            noResultInjected = false;
+            document.getElementsByClassName('noResults')[0].remove();
+            kiwixHomeBody.removeAttribute('style');
+        }
+        loader.removeAttribute('style');
+        return false;
+    }
+
+    async function loadAndDisplayBooks(sort = false) {
+        if (isFetching) return;
+        isFetching = true;
+        await loadAndDisplayBooksUnguarded(sort);
+        isFetching = false;
+    }
+
+    async function loadAndDisplayBooksUnguarded(sort) {
+        let books = await loadBooks();
+        if (checkAndInjectEmptyMessage()) {return}
+        const booksToFilter = new Set();
+        const booksToDelete = new Set();
+        iso.arrange({
+            filter: function (idx, elem) {
+                const id = elem.getAttribute('data-id');
+                const retVal = bookOrderMap.has(id);
+                if (retVal) {
+                    booksToFilter.add(id);
+                    if (sort) {
+                        elem.setAttribute('data-idx', bookOrderMap.get(id));
+                        iso.updateSortData(elem);
+                    }
+                } else {
+                    booksToDelete.add(elem);
+                }
+                return retVal;
+            }
+        });
+        books = [...books].filter((book) => {return !booksToFilter.has(getInnerHtml(book, 'id'))});
+        booksToDelete.forEach(book => {iso.remove(book);});
+        books.forEach((book) => {
+            iso.insert(generateBookHtml(book, sort))
+            const downloadButton = document.querySelector(`[data-id="${getInnerHtml(book, 'id')}"] .book__download span`);
+            if (downloadButton) {
+                insertModal(downloadButton);
+            }
+        });
+    }
+
+    async function resetAndFilter(filterType = '', filterValue = '') {
+        isFetching = false;
+        incrementalLoadingParams.start = 0;
+        incrementalLoadingParams.count = viewPortToCount();
+        fadeOutDiv.style.display = 'none';
+        bookOrderMap.clear();
+        params = new URLSearchParams(window.location.search);
+        if (filterType) {
+            params.set(filterType, filterValue);
+            window.history.pushState({}, null, `${window.location.href.split('?')[0]}?${params.toString()}`);
+            setCookie(filterCookieName, params.toString());
+        }
+        document.querySelectorAll('.filter').forEach(filter => {
+            if (filter.value) {
+                filter.style = 'background-color: #858585; color: #fff';
+            } else {
+                filter.style = 'background-color: #ffffff; color: black';
+            }
+        });
+        await loadAndDisplayBooks(true);
+    }
+
+    window.addEventListener('popstate', async () => {
+        await resetAndFilter();
+        document.querySelectorAll('.filter').forEach(filter => {filter.value = params.get(filter.name) || ''});
+    });
+
+    async function loadSubset() {
+        if (window.innerHeight + window.scrollY >= (document.body.offsetHeight * 0.98)) {
+            if (incrementalLoadingParams.count) {
+                loadAndDisplayBooks();
+            }
+            else {
+                fadeOutDiv.style.display = 'none';
+            }
+        }
+    }
+
+    window.addEventListener('resize', (event) => {
+        if (timer) {clearTimeout(timer)}
+        timer = setTimeout(() => {
+            incrementalLoadingParams.count = incrementalLoadingParams.count && viewPortToCount();
+            loadSubset();
+        }, 100, event);
+    });
+
+    window.addEventListener('scroll', loadSubset);
+
+    window.onload = async () => {
+        iso = new Isotope( '.book__list', {
+            itemSelector: '.book',
+            getSortData:{
+                weight: function( itemElem ) {
+                    const index = itemElem.getAttribute('data-idx');
+                    return index ? parseInt(index) : Infinity;
+                }
+            },
+            sortBy: 'weight',
+            layoutMode: 'cellsByRow',
+            cellsByRow: {
+                columnWidth: '.book',
+                rowHeight: '.book'
+            }
+        });
+        footer = document.getElementById('kiwixfooter');
+        fadeOutDiv = document.getElementById('fadeOut');
+        loader = document.querySelector('.loader');
+        await loadAndDisplayOptions('#languageFilter', `${root}/catalog/v2/languages`, 'language');
+        await loadAndDisplayOptions('#categoryFilter', `${root}/catalog/v2/categories`, 'title');
+        await loadAndDisplayBooks();
+        document.querySelectorAll('.filter').forEach(filter => {
+            filter.addEventListener('change', () => {resetAndFilter(filter.name, filter.value)});
+        });
+        if (filters) {
+            window.history.pushState({}, null, `${window.location.href.split('?')[0]}?${params.toString()}`);
+        }
+        params.forEach((value, key) => {
+            const selectBox = document.getElementsByName(key)[0];
+            if (selectBox) {
+                selectBox.value = value
+            }
+        });
+        document.getElementById('kiwixSearchForm').onsubmit = (event) => {event.preventDefault()};
+        if (!window.location.search) {
+            const browserLang = navigator.language.split('-')[0];
+            const langFilter = document.getElementById('languageFilter');
+            const lang = browserLang.length === 3 ? browserLang : iso6391To3[browserLang];
+            if (await getBookCount(`lang=${lang}`)) {
+                langFilter.value = lang;
+                langFilter.dispatchEvent(new Event('change'));
+            }
+        }
+        document.querySelectorAll('.filter').forEach(filter => {
+            if (filter.value) {
+                filter.style = 'background-color: #858585; color: #fff';
+            } else {
+                filter.style = 'background-color: #ffffff; color: black';
+            }
+        });
+        setCookie(filterCookieName, params.toString());
+    }
+})();

static/skin/iso6391To3.js

@@ -0,0 +1,187 @@
+// Source: https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes
+// eslint-disable-next-line no-unused-vars
+const iso6391To3 = {
+    "aa": "aar",
+    "ab": "abk",
+    "ae": "ave",
+    "af": "afr",
+    "ak": "aka",
+    "am": "amh",
+    "an": "arg",
+    "ar": "ara",
+    "as": "asm",
+    "av": "ava",
+    "ay": "aym",
+    "az": "aze",
+    "ba": "bak",
+    "be": "bel",
+    "bg": "bul",
+    "bi": "bis",
+    "bm": "bam",
+    "bn": "ben",
+    "bo": "bod",
+    "br": "bre",
+    "bs": "bos",
+    "ca": "cat",
+    "ce": "che",
+    "ch": "cha",
+    "co": "cos",
+    "cr": "cre",
+    "cs": "ces",
+    "cu": "chu",
+    "cv": "chv",
+    "cy": "cym",
+    "da": "dan",
+    "de": "deu",
+    "dv": "div",
+    "dz": "dzo",
+    "el": "ell",
+    "ee": "ewe",
+    "en": "eng",
+    "eo": "epo",
+    "es": "spa",
+    "et": "est",
+    "eu": "eus",
+    "fa": "fas",
+    "ff": "ful",
+    "fi": "fin",
+    "fj": "fij",
+    "fo": "fao",
+    "fr": "fra",
+    "fy": "fry",
+    "ga": "gle",
+    "gd": "gla",
+    "gl": "glg",
+    "gn": "grn",
+    "gu": "guj",
+    "gv": "glv",
+    "ha": "hau",
+    "he": "heb",
+    "hi": "hin",
+    "ho": "hmo",
+    "hr": "hrv",
+    "ht": "hat",
+    "hu": "hun",
+    "hy": "hye",
+    "hz": "her",
+    "ia": "ina",
+    "id": "ind",
+    "ie": "ile",
+    "ig": "ibo",
+    "ii": "iii",
+    "ik": "ipk",
+    "io": "ido",
+    "is": "isl",
+    "it": "ita",
+    "iu": "iku",
+    "ja": "jpn",
+    "jv": "jav",
+    "ka": "kat",
+    "kg": "kon",
+    "ki": "kik",
+    "kj": "kua",
+    "kl": "kal",
+    "kk": "kaz",
+    "km": "khm",
+    "kn": "kan",
+    "ko": "kor",
+    "kr": "kau",
+    "ks": "kas",
+    "ku": "kur",
+    "kv": "kom",
+    "kw": "cor",
+    "ky": "kir",
+    "la": "lat",
+    "lb": "ltz",
+    "lg": "lug",
+    "li": "lim",
+    "ln": "lin",
+    "lo": "lao",
+    "lt": "lit",
+    "lu": "lub",
+    "lv": "lav",
+    "mg": "mlg",
+    "mh": "mah",
+    "mi": "mri",
+    "mk": "mkd",
+    "ml": "mal",
+    "mn": "mon",
+    "mr": "mar",
+    "ms": "msa",
+    "mt": "mlt",
+    "my": "mya",
+    "na": "nau",
+    "nb": "nob",
+    "nd": "nde",
+    "ne": "nep",
+    "ng": "ndo",
+    "nl": "nld",
+    "nn": "nno",
+    "no": "nor",
+    "nr": "nbl",
+    "nv": "nav",
+    "ny": "nya",
+    "oc": "oci",
+    "oj": "oji",
+    "om": "orm",
+    "or": "ori",
+    "os": "oss",
+    "pa": "pan",
+    "pi": "pli",
+    "pl": "pol",
+    "ps": "pus",
+    "pt": "por",
+    "qu": "que",
+    "rm": "roh",
+    "rn": "run",
+    "ro": "ron",
+    "ru": "rus",
+    "rw": "kin",
+    "sa": "san",
+    "sc": "srd",
+    "sd": "snd",
+    "se": "sme",
+    "sm": "smo",
+    "sg": "sag",
+    "si": "sin",
+    "sk": "slk",
+    "sl": "slv",
+    "sn": "sna",
+    "so": "som",
+    "sq": "sqi",
+    "sr": "srp",
+    "ss": "ssw",
+    "st": "sot",
+    "su": "sun",
+    "sv": "swe",
+    "sw": "swa",
+    "ta": "tam",
+    "te": "tel",
+    "tg": "tgk",
+    "th": "tha",
+    "ti": "tir",
+    "tk": "tuk",
+    "tl": "tgl",
+    "tn": "tsn",
+    "to": "ton",
+    "tr": "tur",
+    "ts": "tso",
+    "tt": "tat",
+    "tw": "twi",
+    "ty": "tah",
+    "ug": "uig",
+    "uk": "ukr",
+    "ur": "urd",
+    "uz": "uzb",
+    "ve": "ven",
+    "vi": "vie",
+    "vo": "vol",
+    "wa": "wln",
+    "wo": "wol",
+    "xh": "xho",
+    "yi": "yid",
+    "yo": "yor",
+    "za": "zha",
+    "zh": "zho",
+    "zu": "zul"
+}