Patch httplib.h for gcc maybe-uninitialized warning.

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

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

.github/workflows/ci.yml

@@ -13,7 +13,8 @@ jobs:
         with:
           python-version: '3.5'
       - 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
@@ -133,7 +134,7 @@ 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 +145,7 @@ 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 +154,7 @@ 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

.gitignore

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

ChangeLog

@@ -1,3 +1,8 @@
+libkiwix 10.0.0
+===============
+
+ * ...
+
 kiwix-lib 9.4.1
 ===============
 

README.md

@@ -1,50 +1,50 @@
-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)
+[![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 +56,12 @@ 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 +77,7 @@ 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
@@ -85,7 +85,7 @@ 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.
 
@@ -103,7 +103,7 @@ 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 +146,10 @@ cp ninja ../bin
 cd ..
 ```
 
+If you compile manually Libmicrohttpd, you might need to compile it
+without GNU TLS, a bug here will empeach further compilation
+otherwise.
+
 If the compilation still fails, you might need to get a more recent
 version of a dependency than the one packaged by your Linux
 distribution. Try then with a source tarball distributed by the

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

@@ -29,7 +29,7 @@ task writePom {
         version '10.0.0' + (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

@@ -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

include/book.h

@@ -59,6 +59,7 @@ 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;
@@ -94,6 +95,9 @@ class Book
   void setFaviconMimeType(const std::string& faviconMimeType) { m_faviconMimeType = faviconMimeType; }
   void setDownloadId(const std::string& downloadId) { m_downloadId = downloadId; }
 
+ private:
+  std::string getCategoryFromTags() const;
+
  protected:
   std::string m_id;
   std::string m_downloadId;
@@ -101,6 +105,7 @@ class Book
   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;

include/entry.h

@@ -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.

include/library.h

@@ -35,6 +35,7 @@ namespace kiwix
 {
 
 class OPDSDumper;
+class Library;
 
 enum supportedListSortBy { UNSORTED, TITLE, SIZE, DATE, CREATOR, PUBLISHER };
 enum supportedListMode {
@@ -48,18 +49,23 @@ 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;
 
@@ -93,16 +99,42 @@ 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;
 };
 
@@ -115,11 +147,24 @@ class Library
   std::map<std::string, kiwix::Book> m_books;
   std::map<std::string, std::shared_ptr<Reader>> m_readers;
   std::vector<kiwix::Bookmark> m_bookmarks;
+  class BookDB;
+  std::unique_ptr<BookDB> m_bookDB;
+
+ public:
+  typedef std::vector<std::string> BookIdCollection;
 
  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.
    *
@@ -148,7 +193,9 @@ class Library
    */
   bool removeBookmark(const std::string& zimId, const std::string& url);
 
+  const Book& getBookById(const std::string& id) const;
   Book& getBookById(const std::string& id);
+  const Book& getBookByPath(const std::string& path) const;
   Book& getBookByPath(const std::string& path);
   std::shared_ptr<Reader> getReaderById(const std::string& id);
 
@@ -166,7 +213,7 @@ 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 +221,7 @@ 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.
@@ -183,42 +230,49 @@ class Library
    * @param remoteBooks If we must count remote books (books with an url)
    * @return The number of books.
    */
-  unsigned int getBookCount(const bool localBooks, const bool remoteBooks);
+  unsigned int getBookCount(const bool localBooks, const bool remoteBooks) const;
 
   /**
-   * Get all langagues of the books in the library.
+   * Get all languagues of the books in the library.
    *
    * @return A list of languages.
    */
-  std::vector<std::string> getBooksLanguages();
+  std::vector<std::string> getBooksLanguages() const;
+
+  /**
+   * Get all categories of the books in the library.
+   *
+   * @return A list of categories.
+   */
+  std::vector<std::string> getBooksCategories() const;
 
   /**
    * Get all book creators of the books in the library.
    *
    * @return A list of book creators.
    */
-  std::vector<std::string> getBooksCreators();
+  std::vector<std::string> getBooksCreators() const;
 
   /**
    * Get all book publishers of the books in the library.
    *
    * @return A list of book publishers.
    */
-  std::vector<std::string> getBooksPublishers();
+  std::vector<std::string> getBooksPublishers() const;
 
   /**
    * Get all bookmarks.
    *
    * @return A list of bookmarks
    */
-  const std::vector<kiwix::Bookmark> getBookmarks(bool onlyValidBookmarks = true);
+  const std::vector<kiwix::Bookmark> getBookmarks(bool onlyValidBookmarks = true) const;
 
   /**
    * Get all book ids of the books in the library.
    *
    * @return A list of book ids.
    */
-  std::vector<std::string> getBooksIds();
+  BookIdCollection getBooksIds() const;
 
   /**
    * Filter the library and generate a new one with the keep elements.
@@ -228,7 +282,7 @@ class Library
    * @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);
+  DEPRECATED BookIdCollection filter(const std::string& search) const;
 
 
   /**
@@ -237,7 +291,7 @@ 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;
 
 
   /**
@@ -247,7 +301,7 @@ 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.
@@ -271,7 +325,7 @@ class Library
    *                Set to 0 to cancel this filter.
    * @return The list of bookIds corresponding to the query.
    */
-  DEPRECATED std::vector<std::string> listBooksIds(
+  DEPRECATED BookIdCollection listBooksIds(
     int supportedListMode = ALL,
     supportedListSortBy sortBy = UNSORTED,
     const std::string& search = "",
@@ -279,11 +333,16 @@ class Library
     const std::string& creator = "",
     const std::string& publisher = "",
     const std::vector<std::string>& tags = {},
-    size_t maxSize = 0);
+    size_t maxSize = 0) const;
 
   friend class OPDSDumper;
   friend class libXMLDumper;
+
+private: // functions
+  BookIdCollection filterViaBookDB(const Filter& filter) const;
+  void updateBookDB(const Book& book);
 };
+
 }
 
 #endif

include/libxml_dumper.h

@@ -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/opds_dumper.h

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

include/reader.h

@@ -37,12 +37,47 @@ using namespace std;
 namespace kiwix
 {
 
+/**
+ * The SuggestionItem is a helper class that contains the info about a single
+ * suggestion item.
+ */
+
+class SuggestionItem
+{
+  // Functions
+  private:
+    // Create a sugggestion item.
+    explicit SuggestionItem(std::string title, std::string normalizedTitle,
+                            std::string path, std::string snippet = "") :
+      title(title),
+      normalizedTitle(normalizedTitle),
+      path(path),
+      snippet(snippet) {}
+
+  public:
+    const std::string getTitle()           {return title;}
+    const std::string getNormalizedTitle() {return normalizedTitle;}
+    const std::string getPath()            {return path;}
+    const std::string getSnippet()         {return snippet;}
+
+    const bool hasSnippet()                {return !snippet.empty();}
+
+  // Data
+  private:
+    std::string title;
+    std::string normalizedTitle;
+    std::string path;
+    std::string snippet;
+
+  friend class Reader;
+};
+
 /**
  * The Reader class is the class who allow to get an entry content from a zim
  * file.
  */
 
-using SuggestionsList_t = std::vector<std::vector<std::string>>;
+using SuggestionsList_t = std::vector<SuggestionItem>;
 class Reader
 {
  public:
@@ -55,7 +90,11 @@ class Reader
    *                    unsplitted path as if the file were not splitted
    *                    (.zim extesion).
    */
-  Reader(const string zimFilePath);
+  explicit Reader(const string zimFilePath);
+#ifndef _WIN32
+  explicit Reader(int fd);
+  Reader(int fd, zim::offset_type offset, zim::size_type size);
+#endif
   ~Reader() = default;
 
   /**

include/searcher.h

@@ -27,7 +27,7 @@
 #include <cctype>
 #include <locale>
 #include <string>
-#include <vector>
+#include <memory>
 #include <vector>
 #include "tools/pathTools.h"
 #include "tools/stringTools.h"
@@ -48,7 +48,7 @@ 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;
@@ -154,7 +154,7 @@ class Searcher
                      const bool verbose = false);
 
   std::vector<Reader*> readers;
-  SearcherInternal* internal;
+  std::unique_ptr<SearcherInternal> internal;
   std::string searchPattern;
   unsigned int estimatedResultCount;
   unsigned int resultStart;

include/tools/otherTools.h

@@ -24,6 +24,7 @@
 #include <vector>
 #include <map>
 #include <zim/zim.h>
+#include <mustache.hpp>
 
 namespace pugi {
   class xml_node;
@@ -45,6 +46,11 @@ 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

meson.build

@@ -1,4 +1,4 @@
-project('kiwix-lib', 'cpp',
+project('libkiwix', 'cpp',
   version : '10.0.0', # Also change this in android-kiwix-lib-publisher/kiwixLibAndroid/build.gradle
   license : 'GPLv3+',
   default_options : ['c_std=c11', 'cpp_std=c++11', 'werror=true'])
@@ -18,7 +18,7 @@ if wrapper.contains('java')
   add_languages('java')
 endif
 
-# See https://github.com/kiwix/kiwix-lib/issues/371
+# See https://github.com/kiwix/libkiwix/issues/371
 if ['arm', 'mips', 'm68k', 'ppc', 'sh4'].contains(target_machine.cpu_family())
   extra_libs += '-latomic'
 endif
@@ -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 = []
@@ -55,7 +56,11 @@ if target_machine.system() == 'windows' and static_deps
   extra_cflags += '-DCURL_STATICLIB'
 endif
 
-all_deps = [thread_dep, libicu_dep, libzim_dep, pugixml_dep, libcurl_dep, microhttpd_dep, zlib_dep]
+if target_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)
 
@@ -74,7 +79,7 @@ subdir('static')
 subdir('src')
 subdir('test')
 
-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'))

src/aria2.cpp

@@ -63,6 +63,7 @@ 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());

src/book.cpp

@@ -61,6 +61,7 @@ bool Book::update(const kiwix::Book& other)
   m_name = other.m_name;
   m_flavour = other.m_flavour;
   m_tags = other.m_tags;
+  m_category = other.m_category;
   m_origId = other.m_origId;
   m_articleCount = other.m_articleCount;
   m_mediaCount = other.m_mediaCount;
@@ -88,6 +89,7 @@ void Book::update(const kiwix::Reader& reader)
   m_name = reader.getName();
   m_flavour = reader.getFlavour();
   m_tags = reader.getTags();
+  m_category = getCategoryFromTags();
   m_origId = reader.getOrigId();
   m_articleCount = reader.getArticleCount();
   m_mediaCount = reader.getMediaCount();
@@ -127,6 +129,8 @@ void Book::updateFromXml(const pugi::xml_node& node, const std::string& baseDir)
   try {
     m_downloadId = ATTR("downloadId");
   } catch(...) {}
+  const auto catattr = node.attribute("category");
+  m_category = catattr.empty() ? getCategoryFromTags() : catattr.value();
 }
 #undef ATTR
 
@@ -156,6 +160,8 @@ void Book::updateFromOpds(const pugi::xml_node& node, const std::string& urlHost
   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;
@@ -220,4 +226,21 @@ 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/library.cpp

@@ -30,14 +30,42 @@
 #include <pugixml.hpp>
 #include <algorithm>
 #include <set>
+#include <unicode/locid.h>
+#include <xapian.h>
 
 namespace kiwix
 {
 
+namespace
+{
+
+std::string iso639_3ToXapian(const std::string& lang) {
+  return icu::Locale(lang.c_str()).getLanguage();
+};
+
+std::string normalizeText(const std::string& text)
+{
+  return removeAccents(text);
+}
+
+} // unnamed namespace
+
+class Library::BookDB : public Xapian::WritableDatabase
+{
+public:
+  BookDB() : Xapian::WritableDatabase("", Xapian::DB_BACKEND_INMEMORY) {}
+};
+
 /* Constructor */
 Library::Library()
+  : m_bookDB(new BookDB)
 {
 }
+
+Library::Library(Library&& ) = default;
+
+Library& Library::operator=(Library&& ) = default;
+
 /* Destructor */
 Library::~Library()
 {
@@ -47,6 +75,7 @@ Library::~Library()
 bool Library::addBook(const Book& book)
 {
   /* Try to find it */
+  updateBookDB(book);
   try {
     auto& oldbook = m_books.at(book.getId());
     oldbook.update(book);
@@ -77,15 +106,23 @@ bool Library::removeBookmark(const std::string& zimId, const std::string& url)
 
 bool Library::removeBookById(const std::string& id)
 {
+  m_bookDB->delete_document("Q" + id);
+  m_readers.erase(id);
   return m_books.erase(id) == 1;
 }
 
-Book& Library::getBookById(const std::string& id)
+const Book& Library::getBookById(const std::string& id) const
 {
   return m_books.at(id);
 }
 
-Book& Library::getBookByPath(const std::string& path)
+Book& Library::getBookById(const std::string& id)
+{
+  const Library& const_self = *this;
+  return const_cast<Book&>(const_self.getBookById(id));
+}
+
+const Book& Library::getBookByPath(const std::string& path) const
 {
   for(auto& it: m_books) {
     auto& book = it.second;
@@ -97,6 +134,12 @@ Book& Library::getBookByPath(const std::string& path)
   throw std::out_of_range(ss.str());
 }
 
+Book& Library::getBookByPath(const std::string& path)
+{
+  const Library& const_self = *this;
+  return const_cast<Book&>(const_self.getBookByPath(path));
+}
+
 std::shared_ptr<Reader> Library::getReaderById(const std::string& id)
 {
   try {
@@ -112,7 +155,7 @@ std::shared_ptr<Reader> Library::getReaderById(const std::string& id)
 }
 
 unsigned int Library::getBookCount(const bool localBooks,
-                                   const bool remoteBooks)
+                                   const bool remoteBooks) const
 {
   unsigned int result = 0;
   for (auto& pair: m_books) {
@@ -125,7 +168,7 @@ unsigned int Library::getBookCount(const bool localBooks,
   return result;
 }
 
-bool Library::writeToFile(const std::string& path)
+bool Library::writeToFile(const std::string& path) const
 {
   auto baseDir = removeLastPathElement(path);
   LibXMLDumper dumper(this);
@@ -133,13 +176,13 @@ bool Library::writeToFile(const std::string& path)
   return writeTextFile(path, dumper.dumpLibXMLContent(getBooksIds()));
 }
 
-bool Library::writeBookmarksToFile(const std::string& path)
+bool Library::writeBookmarksToFile(const std::string& path) const
 {
   LibXMLDumper dumper(this);
   return writeTextFile(path, dumper.dumpLibXMLBookmark());
 }
 
-std::vector<std::string> Library::getBooksLanguages()
+std::vector<std::string> Library::getBooksLanguages() const
 {
   std::vector<std::string> booksLanguages;
   std::map<std::string, bool> booksLanguagesMap;
@@ -158,7 +201,22 @@ std::vector<std::string> Library::getBooksLanguages()
   return booksLanguages;
 }
 
-std::vector<std::string> Library::getBooksCreators()
+std::vector<std::string> Library::getBooksCategories() const
+{
+  std::set<std::string> categories;
+
+  for (const auto& pair: m_books) {
+    const auto& book = pair.second;
+    const auto& c = book.getCategory();
+    if ( !c.empty() ) {
+      categories.insert(c);
+    }
+  }
+
+  return std::vector<std::string>(categories.begin(), categories.end());
+}
+
+std::vector<std::string> Library::getBooksCreators() const
 {
   std::vector<std::string> booksCreators;
   std::map<std::string, bool> booksCreatorsMap;
@@ -177,7 +235,7 @@ std::vector<std::string> Library::getBooksCreators()
   return booksCreators;
 }
 
-std::vector<std::string> Library::getBooksPublishers()
+std::vector<std::string> Library::getBooksPublishers() const
 {
   std::vector<std::string> booksPublishers;
   std::map<std::string, bool> booksPublishersMap;
@@ -196,7 +254,7 @@ std::vector<std::string> Library::getBooksPublishers()
   return booksPublishers;
 }
 
-const std::vector<kiwix::Bookmark> Library::getBookmarks(bool onlyValidBookmarks)
+const std::vector<kiwix::Bookmark> Library::getBookmarks(bool onlyValidBookmarks) const
 {
   if (!onlyValidBookmarks) {
     return m_bookmarks;
@@ -211,9 +269,9 @@ 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;
+  BookIdCollection bookIds;
 
   for (auto& pair: m_books) {
     bookIds.push_back(pair.first);
@@ -222,7 +280,7 @@ std::vector<std::string> Library::getBooksIds()
   return bookIds;
 }
 
-std::vector<std::string> Library::filter(const std::string& search)
+Library::BookIdCollection Library::filter(const std::string& search) const
 {
   if (search.empty()) {
     return getBooksIds();
@@ -232,16 +290,195 @@ std::vector<std::string> Library::filter(const std::string& search)
 }
 
 
-std::vector<std::string> Library::filter(const Filter& filter)
+void Library::updateBookDB(const Book& book)
 {
-  std::vector<std::string> bookIds;
-  for(auto& pair:m_books) {
-    auto book = pair.second;
-    if(filter.accept(book)) {
-      bookIds.push_back(pair.first);
+  Xapian::Stem stemmer;
+  Xapian::TermGenerator indexer;
+  const std::string lang = book.getLanguage();
+  try {
+    stemmer = Xapian::Stem(iso639_3ToXapian(lang));
+    indexer.set_stemmer(stemmer);
+    indexer.set_stemming_strategy(Xapian::TermGenerator::STEM_SOME);
+  } catch (...) {}
+  Xapian::Document doc;
+  indexer.set_document(doc);
+
+  const std::string title = normalizeText(book.getTitle());
+  const std::string desc = normalizeText(book.getDescription());
+
+  // Index title and description without prefixes for general search
+  indexer.index_text(title);
+  indexer.increase_termpos();
+  indexer.index_text(desc);
+
+  // Index all fields for field-based search
+  indexer.index_text(title, 1, "S");
+  indexer.index_text(desc,  1, "XD");
+  indexer.index_text(lang,  1, "L");
+  indexer.index_text(normalizeText(book.getCreator()),   1, "A");
+  indexer.index_text(normalizeText(book.getPublisher()), 1, "XP");
+  indexer.index_text(normalizeText(book.getName()),      1, "XN");
+  indexer.index_text(normalizeText(book.getCategory()),  1, "XC");
+
+  for ( const auto& tag : split(normalizeText(book.getTags()), ";") )
+    doc.add_boolean_term("XT" + tag);
+
+  const std::string idterm = "Q" + book.getId();
+  doc.add_boolean_term(idterm);
+
+  doc.set_data(book.getId());
+
+  m_bookDB->replace_document(idterm, doc);
+}
+
+namespace
+{
+
+bool willSelectEverything(const Xapian::Query& query)
+{
+  return query.get_type() == Xapian::Query::LEAF_MATCH_ALL;
+}
+
+
+Xapian::Query buildXapianQueryFromFilterQuery(const Filter& filter)
+{
+  if ( !filter.hasQuery() || filter.getQuery().empty() ) {
+    // This is a thread-safe way to construct an equivalent of
+    // a Xapian::Query::MatchAll query
+    return Xapian::Query(std::string());
+  }
+
+  Xapian::QueryParser queryParser;
+  queryParser.set_default_op(Xapian::Query::OP_AND);
+  queryParser.add_prefix("title", "S");
+  queryParser.add_prefix("description", "XD");
+  queryParser.add_prefix("name", "XN");
+  queryParser.add_prefix("category", "XC");
+  queryParser.add_prefix("lang", "L");
+  queryParser.add_prefix("publisher", "XP");
+  queryParser.add_prefix("creator", "A");
+  queryParser.add_prefix("tag", "XT");
+  const auto partialQueryFlag = filter.queryIsPartial()
+                              ? Xapian::QueryParser::FLAG_PARTIAL
+                              : 0;
+  // Language assumed for the query is not known for sure so stemming
+  // is not applied
+  //queryParser.set_stemmer(Xapian::Stem(iso639_3ToXapian(???)));
+  //queryParser.set_stemming_strategy(Xapian::QueryParser::STEM_SOME);
+  const auto flags = Xapian::QueryParser::FLAG_PHRASE
+                   | Xapian::QueryParser::FLAG_BOOLEAN
+                   | Xapian::QueryParser::FLAG_BOOLEAN_ANY_CASE
+                   | Xapian::QueryParser::FLAG_LOVEHATE
+                   | Xapian::QueryParser::FLAG_WILDCARD
+                   | partialQueryFlag;
+  return queryParser.parse_query(normalizeText(filter.getQuery()), flags);
+}
+
+Xapian::Query nameQuery(const std::string& name)
+{
+  return Xapian::Query("XN" + normalizeText(name));
+}
+
+Xapian::Query categoryQuery(const std::string& category)
+{
+  return Xapian::Query("XC" + normalizeText(category));
+}
+
+Xapian::Query langQuery(const std::string& lang)
+{
+  return Xapian::Query("L" + normalizeText(lang));
+}
+
+Xapian::Query publisherQuery(const std::string& publisher)
+{
+  Xapian::QueryParser queryParser;
+  queryParser.set_default_op(Xapian::Query::OP_OR);
+  queryParser.set_stemming_strategy(Xapian::QueryParser::STEM_NONE);
+  const auto flags = 0;
+  const auto q = queryParser.parse_query(normalizeText(publisher), flags, "XP");
+  return Xapian::Query(Xapian::Query::OP_PHRASE, q.get_terms_begin(), q.get_terms_end(), q.get_length());
+}
+
+Xapian::Query creatorQuery(const std::string& creator)
+{
+  Xapian::QueryParser queryParser;
+  queryParser.set_default_op(Xapian::Query::OP_OR);
+  queryParser.set_stemming_strategy(Xapian::QueryParser::STEM_NONE);
+  const auto flags = 0;
+  const auto q = queryParser.parse_query(normalizeText(creator), flags, "A");
+  return Xapian::Query(Xapian::Query::OP_PHRASE, q.get_terms_begin(), q.get_terms_end(), q.get_length());
+}
+
+Xapian::Query tagsQuery(const Filter::Tags& acceptTags, const Filter::Tags& rejectTags)
+{
+  Xapian::Query q = Xapian::Query(std::string());
+  if (!acceptTags.empty()) {
+    for ( const auto& tag : acceptTags )
+      q &= Xapian::Query("XT" + normalizeText(tag));
+  }
+
+  if (!rejectTags.empty()) {
+    for ( const auto& tag : rejectTags )
+      q = Xapian::Query(Xapian::Query::OP_AND_NOT, q, "XT" + normalizeText(tag));
+  }
+  return q;
+}
+
+Xapian::Query buildXapianQuery(const Filter& filter)
+{
+  auto q = buildXapianQueryFromFilterQuery(filter);
+  if ( filter.hasName() ) {
+    q = Xapian::Query(Xapian::Query::OP_AND, q, nameQuery(filter.getName()));
+  }
+  if ( filter.hasCategory() ) {
+    q = Xapian::Query(Xapian::Query::OP_AND, q, categoryQuery(filter.getCategory()));
+  }
+  if ( filter.hasLang() ) {
+    q = Xapian::Query(Xapian::Query::OP_AND, q, langQuery(filter.getLang()));
+  }
+  if ( filter.hasPublisher() ) {
+    q = Xapian::Query(Xapian::Query::OP_AND, q, publisherQuery(filter.getPublisher()));
+  }
+  if ( filter.hasCreator() ) {
+    q = Xapian::Query(Xapian::Query::OP_AND, q, creatorQuery(filter.getCreator()));
+  }
+  if ( !filter.getAcceptTags().empty() || !filter.getRejectTags().empty() ) {
+    const auto tq = tagsQuery(filter.getAcceptTags(), filter.getRejectTags());
+    q = Xapian::Query(Xapian::Query::OP_AND, q, tq);;
+  }
+  return q;
+}
+
+} // unnamed namespace
+
+Library::BookIdCollection Library::filterViaBookDB(const Filter& filter) const
+{
+  const auto query = buildXapianQuery(filter);
+
+  if ( willSelectEverything(query) )
+    return getBooksIds();
+
+  BookIdCollection bookIds;
+
+  Xapian::Enquire enquire(*m_bookDB);
+  enquire.set_query(query);
+  const auto results = enquire.get_mset(0, m_books.size());
+  for ( auto it = results.begin(); it != results.end(); ++it  ) {
+    bookIds.push_back(it.get_document().get_data());
+  }
+
+  return bookIds;
+}
+
+Library::BookIdCollection Library::filter(const Filter& filter) const
+{
+  BookIdCollection result;
+  for(auto id : filterViaBookDB(filter)) {
+    if(filter.accept(m_books.at(id))) {
+      result.push_back(id);
     }
   }
-  return bookIds;
+  return result;
 }
 
 template<supportedListSortBy SORT>
@@ -257,13 +494,13 @@ 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);
@@ -303,7 +540,7 @@ 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
 {
   switch(sort) {
     case TITLE:
@@ -327,7 +564,7 @@ void Library::sort(std::vector<std::string>& bookIds, supportedListSortBy sort,
 }
 
 
-std::vector<std::string> Library::listBooksIds(
+Library::BookIdCollection Library::listBooksIds(
     int mode,
     supportedListSortBy sortBy,
     const std::string& search,
@@ -335,7 +572,7 @@ std::vector<std::string> Library::listBooksIds(
     const std::string& creator,
     const std::string& publisher,
     const std::vector<std::string>& tags,
-    size_t maxSize) {
+    size_t maxSize) const {
 
   Filter _filter;
   if (mode & LOCAL)
@@ -391,6 +628,7 @@ enum filterTypes {
   MAXSIZE = FLAG(11),
   QUERY = FLAG(12),
   NAME = FLAG(13),
+  CATEGORY = FLAG(14),
 };
 
 Filter& Filter::local(bool accept)
@@ -429,20 +667,27 @@ 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 +716,10 @@ 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 +733,36 @@ 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 +778,8 @@ 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

@@ -28,7 +28,7 @@
 namespace kiwix
 {
 /* Constructor */
-LibXMLDumper::LibXMLDumper(Library* library)
+LibXMLDumper::LibXMLDumper(const Library* library)
   : library(library)
 {
 }

src/manager.cpp

@@ -80,7 +80,7 @@ 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 +124,7 @@ bool Manager::readOpds(const std::string& content, const std::string& urlHost)
 {
   pugi::xml_document doc;
   pugi::xml_parse_result result
-      = doc.load_buffer_inplace((void*)content.data(), content.size());
+      = doc.load_buffer((void*)content.data(), content.size());
 
   if (result) {
     this->parseOpdsDom(doc, urlHost);

src/meson.build

@@ -25,7 +25,8 @@ kiwix_sources = [
   'server/etag.cpp',
   'server/request_context.cpp',
   'server/response.cpp',
-  'server/internalServer.cpp'
+  'server/internalServer.cpp',
+  'server/internalServer_catalog_v2.cpp'
 ]
 kiwix_sources += lib_resources
 

src/opds_dumper.cpp

@@ -21,10 +21,13 @@
 #include "book.h"
 
 #include "tools/otherTools.h"
-#include <iomanip>
+
+#include "kiwixlib-resources.h"
+#include <mustache.hpp>
 
 namespace kiwix
 {
+
 /* Constructor */
 OPDSDumper::OPDSDumper(Library* library)
   : library(library)
@@ -35,120 +38,111 @@ OPDSDumper::~OPDSDumper()
 {
 }
 
-std::string gen_date_str()
-{
-  auto now = time(0);
-  auto tm = localtime(&now);
-
-  std::stringstream is;
-  is << std::setw(2) << std::setfill('0')
-     << 1900+tm->tm_year << "-"
-     << std::setw(2) << std::setfill('0') << tm->tm_mon << "-"
-     << std::setw(2) << std::setfill('0') << tm->tm_mday << "T"
-     << std::setw(2) << std::setfill('0') << tm->tm_hour << ":"
-     << std::setw(2) << std::setfill('0') << tm->tm_min << ":"
-     << std::setw(2) << std::setfill('0') << tm->tm_sec << "Z";
-  return is.str();
-}
-
-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 BookData;
 
-  ADD_TEXT_ENTRY(root_node, "id", id);
-
-  ADD_TEXT_ENTRY(root_node, "title", title);
-  ADD_TEXT_ENTRY(root_node, "updated", date);
-
-  if (m_isSearchResult) {
-    ADD_TEXT_ENTRY(root_node, "totalResults", to_string(m_totalResults));
-    ADD_TEXT_ENTRY(root_node, "startIndex", to_string(m_startIndex));
-    ADD_TEXT_ENTRY(root_node, "itemsPerPage", to_string(m_count));
+BookData getBookData(const Library* library, const std::vector<std::string>& bookIds)
+{
+  BookData bookData;
+  for ( const auto& bookId : bookIds ) {
+    const Book& book = library->getBookById(bookId);
+    const MustacheData bookUrl = book.getUrl().empty()
+                               ? MustacheData(false)
+                               : MustacheData(book.getUrl());
+    bookData.push_back(kainjow::mustache::object{
+      {"id", "urn:uuid:"+book.getId()},
+      {"name", book.getName()},
+      {"title", book.getTitle()},
+      {"description", book.getDescription()},
+      {"language", book.getLanguage()},
+      {"content_id",  book.getHumanReadableIdFromPath()},
+      {"updated", book.getDate() + "T00:00:00Z"},
+      {"category", book.getCategory()},
+      {"flavour", book.getFlavour()},
+      {"tags", book.getTags()},
+      {"article_count", to_string(book.getArticleCount())},
+      {"media_count", to_string(book.getMediaCount())},
+      {"author_name", book.getCreator()},
+      {"publisher_name", book.getPublisher()},
+      {"url", bookUrl},
+      {"size", to_string(book.getSize())},
+    });
   }
 
-  auto self_link_node = root_node.append_child("link");
-  self_link_node.append_attribute("rel") = "self";
-  self_link_node.append_attribute("href") = "";
-  self_link_node.append_attribute("type") = "application/atom+xml";
+  return bookData;
+}
 
+} // unnamed namespace
 
-  if (!searchDescriptionUrl.empty() ) {
-    auto search_link = root_node.append_child("link");
-    search_link.append_attribute("rel") = "search";
-    search_link.append_attribute("type") = "application/opensearchdescription+xml";
-    search_link.append_attribute("href") = searchDescriptionUrl.c_str();
+string OPDSDumper::dumpOPDSFeed(const std::vector<std::string>& bookIds, const std::string& query) const
+{
+  const auto bookData = getBookData(library, bookIds);
+  const kainjow::mustache::object template_data{
+     {"date", gen_date_str()},
+     {"root", rootLocation},
+     {"feed_id", gen_uuid(libraryId + "/catalog/search?"+query)},
+     {"filter", query.empty() ? MustacheData(false) : MustacheData(query)},
+     {"totalResults", to_string(m_totalResults)},
+     {"startIndex", to_string(m_startIndex)},
+     {"itemsPerPage", to_string(m_count)},
+     {"books", bookData }
+  };
+
+  return render_template(RESOURCE::templates::catalog_entries_xml, template_data);
+}
+
+string OPDSDumper::dumpOPDSFeedV2(const std::vector<std::string>& bookIds, const std::string& query) const
+{
+  const auto bookData = getBookData(library, bookIds);
+
+  const kainjow::mustache::object template_data{
+     {"date", gen_date_str()},
+     {"endpoint_root", rootLocation + "/catalog/v2"},
+     {"feed_id", gen_uuid(libraryId + "/entries?"+query)},
+     {"filter", query.empty() ? MustacheData(false) : MustacheData(query)},
+     {"query", query.empty() ? "" : "?" + urlEncode(query)},
+     {"totalResults", to_string(m_totalResults)},
+     {"startIndex", to_string(m_startIndex)},
+     {"itemsPerPage", to_string(m_count)},
+     {"books", bookData }
+  };
+
+  return render_template(RESOURCE::templates::catalog_v2_entries_xml, template_data);
+}
+
+std::string OPDSDumper::categoriesOPDSFeed(const std::vector<std::string>& categories) const
+{
+  const auto now = gen_date_str();
+  kainjow::mustache::list categoryData;
+  for ( const auto& category : categories ) {
+    const auto urlencodedCategoryName = urlEncode(category);
+    categoryData.push_back(kainjow::mustache::object{
+      {"name", category},
+      {"urlencoded_name",  urlencodedCategoryName},
+      {"updated", now},
+      {"id", gen_uuid(libraryId + "/categories/" + urlencodedCategoryName)}
+    });
   }
 
-  if (library) {
-    for (auto& bookId: bookIds) {
-      handleBook(library->getBookById(bookId), root_node);
-    }
-  }
-
-  return nodeToString(root_node);
+  return render_template(
+             RESOURCE::templates::catalog_v2_categories_xml,
+             kainjow::mustache::object{
+               {"date", now},
+               {"endpoint_root", rootLocation + "/catalog/v2"},
+               {"feed_id", gen_uuid(libraryId + "/categories")},
+               {"categories", categoryData }
+             }
+  );
 }
 
 }

src/reader.cpp

@@ -86,6 +86,24 @@ Reader::Reader(const string zimFilePath)
   srand(time(nullptr));
 }
 
+#ifndef _WIN32
+Reader::Reader(int fd)
+  :  zimArchive(new zim::Archive(fd)),
+     zimFilePath("")
+{
+  /* initialize random seed: */
+  srand(time(nullptr));
+}
+
+Reader::Reader(int fd, zim::offset_type offset, zim::size_type size)
+  :  zimArchive(new zim::Archive(fd, offset, size)),
+     zimFilePath("")
+{
+  /* initialize random seed: */
+  srand(time(nullptr));
+}
+#endif // #ifndef _WIN32
+
 zim::Archive* Reader::getZimArchive() const
 {
   return zimArchive.get();
@@ -151,23 +169,11 @@ string Reader::getId() const
 
 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 zimArchive->getRandomEntry();
+  } catch(...) {
+    throw NoEntry();
   }
-  throw NoEntry();
 }
 
 Entry Reader::getMainPage() const
@@ -178,8 +184,7 @@ Entry Reader::getMainPage() const
 bool Reader::getFavicon(string& content, string& mimeType) const
 {
   try {
-    auto entry = zimArchive->getFaviconEntry();
-    auto item = entry.getItem(true);
+    auto item = zimArchive->getIllustrationItem();
     content = item.getData();
     mimeType = item.getMimetype();
     return true;
@@ -426,12 +431,12 @@ 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 +446,7 @@ 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);
     }
 
@@ -489,19 +491,19 @@ bool Reader::searchSuggestionsSmart(const string& 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);
 
-  if (suggestionSearch.get_matches_estimated()) {
-    for (auto current = suggestionSearch.begin();
-         current != suggestionSearch.end();
+  auto suggestionSearcher = zim::Searcher(*zimArchive);
+  zim::Query suggestionQuery;
+  suggestionQuery.setQuery(prefix, true);
+  auto suggestionSearch = suggestionSearcher.search(suggestionQuery);
+
+  if (suggestionSearch.getEstimatedMatches()) {
+    const auto suggestions = suggestionSearch.getResults(0, suggestionsCount);
+    for (auto current = suggestions.begin();
+         current != suggestions.end();
          current++) {
-      std::vector<std::string> suggestion;
-      suggestion.push_back(current->getTitle());
-      suggestion.push_back(current->getPath());
-      suggestion.push_back(kiwix::normalize(current->getTitle()));
+      SuggestionItem suggestion(current.getTitle(), kiwix::normalize(current.getTitle()),
+                                current.getPath(), current.getSnippet());
       results.push_back(suggestion);
     }
     retVal = true;
@@ -522,7 +524,7 @@ bool Reader::getNextSuggestion(string& title)
 {
   if (this->suggestionsOffset != this->suggestions.end()) {
     /* title */
-    title = (*(this->suggestionsOffset))[0];
+    title = (*(this->suggestionsOffset)).getTitle();
 
     /* increment the cursor for the next call */
     this->suggestionsOffset++;
@@ -537,8 +539,8 @@ bool Reader::getNextSuggestion(string& title, string& url)
 {
   if (this->suggestionsOffset != this->suggestions.end()) {
     /* title */
-    title = (*(this->suggestionsOffset))[0];
-    url = (*(this->suggestionsOffset))[1];
+    title = (*(this->suggestionsOffset)).getTitle();
+    url = (*(this->suggestionsOffset)).getPath();
 
     /* increment the cursor for the next call */
     this->suggestionsOffset++;

src/search_renderer.cpp

@@ -77,9 +77,7 @@ std::string SearchRenderer::getHtml()
     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("resultContentId", mp_nameMapper->getNameForId(p_result->get_zimId()));
 
     if (p_result->get_wordCount() >= 0) {
       result.set("wordCount", kiwix::beautifyInteger(p_result->get_wordCount()));

src/searcher.cpp

@@ -35,7 +35,7 @@ namespace kiwix
 class _Result : public Result
 {
  public:
-  _Result(zim::Search::iterator& iterator);
+  _Result(zim::SearchResultSet::iterator iterator);
   virtual ~_Result(){};
 
   virtual std::string get_url();
@@ -45,29 +45,25 @@ class _Result : public Result
   virtual std::string get_content();
   virtual int get_wordCount();
   virtual int get_size();
-  virtual int get_readerIndex();
+  virtual std::string get_zimId();
 
  private:
-  zim::Search::iterator iterator;
+  zim::SearchResultSet::iterator iterator;
 };
 
-struct SearcherInternal {
-  const zim::Search* _search;
-  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;
 };
 
 /* Constructor */
 Searcher::Searcher()
-    : internal(new SearcherInternal()),
-      searchPattern(""),
+    : searchPattern(""),
       estimatedResultCount(0),
       resultStart(0),
       resultEnd(0)
@@ -78,7 +74,6 @@ Searcher::Searcher()
 /* Destructor */
 Searcher::~Searcher()
 {
-  delete internal;
 }
 
 bool Searcher::add_reader(Reader* reader)
@@ -122,13 +117,13 @@ 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);
+    zim::Query query;
+    query.setQuery(unaccentedSearch, false);
+    query.setVerbose(verbose);
+    zim::Search search = searcher.search(query);
+    internal.reset(new SearcherInternal(search.getResults(resultStart, resultEnd)));
+    this->estimatedResultCount = search.getEstimatedMatches();
   }
 
   return;
@@ -163,28 +158,28 @@ 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);
+  zim::Query query;
+  query.setVerbose(verbose);
+  query.setQuery("", false);
+  query.setGeorange(latitude, longitude, distance);
+  zim::Search search = searcher.search(query);
+  internal.reset(new SearcherInternal(search.getResults(resultStart, resultEnd)));
+  this->estimatedResultCount = search.getEstimatedMatches();
 }
 
 
 void Searcher::restart_search()
 {
-  if (internal->_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;
@@ -218,14 +213,13 @@ void Searcher::suggestions(std::string& searchPattern, const bool verbose)
        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();
+  zim::Searcher searcher(archives);
+  zim::Query query;
+  query.setVerbose(verbose);
+  query.setQuery(unaccentedSearch, true);
+  zim::Search search = searcher.search(query);
+  internal.reset(new SearcherInternal(search.getResults(resultStart, resultEnd)));
+  this->estimatedResultCount = search.getEstimatedMatches();
 }
 
 /* Return the result count estimation */
@@ -234,26 +228,26 @@ unsigned int Searcher::getEstimatedResultCount()
   return this->estimatedResultCount;
 }
 
-_Result::_Result(zim::Search::iterator& iterator)
+_Result::_Result(zim::SearchResultSet::iterator iterator)
     : iterator(iterator)
 {
 }
 
 std::string _Result::get_url()
 {
-  return iterator.get_url();
+  return iterator.getPath();
 }
 std::string _Result::get_title()
 {
-  return iterator.get_title();
+  return iterator.getTitle();
 }
 int _Result::get_score()
 {
-  return iterator.get_score();
+  return iterator.getScore();
 }
 std::string _Result::get_snippet()
 {
-  return iterator.get_snippet();
+  return iterator.getSnippet();
 }
 std::string _Result::get_content()
 {
@@ -261,15 +255,17 @@ std::string _Result::get_content()
 }
 int _Result::get_size()
 {
-  return iterator.get_size();
+  return iterator.getSize();
 }
 int _Result::get_wordCount()
 {
-  return iterator.get_wordCount();
+  return iterator.getWordCount();
 }
-int _Result::get_readerIndex()
+std::string _Result::get_zimId()
 {
-  return iterator.get_fileIndex();
+  std::ostringstream s;
+  s << iterator.getZimId();
+  return s.str();
 }
 
 

src/server/internalServer.cpp

@@ -76,6 +76,21 @@ extern "C" {
 
 namespace kiwix {
 
+namespace
+{
+
+inline std::string normalizeRootUrl(std::string rootUrl)
+{
+  while ( !rootUrl.empty() && rootUrl.back() == '/' )
+    rootUrl.pop_back();
+
+  while ( !rootUrl.empty() && rootUrl.front() == '/' )
+    rootUrl = rootUrl.substr(1);
+  return rootUrl.empty() ? rootUrl : "/" + rootUrl;
+}
+
+} // unnamed namespace
+
 static IdNameMapper defaultNameMapper;
 
 static MHD_Result staticHandlerCallback(void* cls,
@@ -100,7 +115,7 @@ InternalServer::InternalServer(Library* library,
                                bool blockExternalLinks) :
   m_addr(addr),
   m_port(port),
-  m_root(root),
+  m_root(normalizeRootUrl(root)),
   m_nbThreads(nbThreads),
   m_verbose(verbose),
   m_withTaskbar(withTaskbar),
@@ -153,6 +168,7 @@ 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;
 }
 
@@ -237,16 +253,16 @@ 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, "", "");
 
     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")
@@ -281,27 +297,6 @@ 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();
@@ -325,7 +320,7 @@ 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, RESOURCE::templates::index_html, get_default_data(), "text/html; charset=utf-8", true);
 }
 
 std::unique_ptr<Response> InternalServer::handle_meta(const RequestContext& request)
@@ -340,11 +335,11 @@ std::unique_ptr<Response> InternalServer::handle_meta(const RequestContext& requ
     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);
+    return Response::build_404(*this, request, bookName, "");
   }
 
   if (reader == nullptr) {
-    return Response::build_404(*this, request, bookName);
+    return Response::build_404(*this, request, bookName, "");
   }
 
   std::string content;
@@ -369,7 +364,7 @@ std::unique_ptr<Response> InternalServer::handle_meta(const RequestContext& requ
   } else if (meta_name == "favicon") {
     reader->getFavicon(content, mimeType);
   } else {
-    return Response::build_404(*this, request, bookName);
+    return Response::build_404(*this, request, bookName, "");
   }
 
   auto response = ContentResponse::build(*this, content, mimeType);
@@ -398,7 +393,7 @@ std::unique_ptr<Response> InternalServer::handle_suggest(const RequestContext& r
     term = request.get_argument("term");
     reader = mp_library->getReaderById(bookId);
   } catch (const std::out_of_range&) {
-    return Response::build_404(*this, request, bookName);
+    return Response::build_404(*this, request, bookName, "");
   }
 
   if (m_verbose.load()) {
@@ -414,8 +409,15 @@ std::unique_ptr<Response> InternalServer::handle_suggest(const RequestContext& r
     reader->searchSuggestionsSmart(term, maxSuggestionCount, suggestions);
     for(auto& suggestion:suggestions) {
       MustacheData result;
-      result.set("label", suggestion[0]);
-      result.set("value", suggestion[0]);
+      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);
@@ -428,6 +430,7 @@ std::unique_ptr<Response> InternalServer::handle_suggest(const RequestContext& r
     MustacheData result;
     result.set("label", "containing '" + term + "'...");
     result.set("value", term + " ");
+    result.set("kind", "pattern");
     result.set("first", first);
     results.push_back(result);
   }
@@ -454,7 +457,7 @@ 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, "", "");
   }
 }
 
@@ -494,30 +497,6 @@ std::unique_ptr<Response> InternalServer::handle_search(const RequestContext& re
     reader = mp_library->getReaderById(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())
     || (patternString.empty() && ! has_geo_query) ) {
@@ -576,11 +555,6 @@ std::unique_ptr<Response> InternalServer::handle_search(const RequestContext& re
     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);
-    }
 
     return std::move(response);
   } catch (const std::exception& e) {
@@ -603,18 +577,18 @@ std::unique_ptr<Response> InternalServer::handle_random(const RequestContext& re
     bookId = mp_nameMapper->getIdForName(bookName);
     reader = mp_library->getReaderById(bookId);
   } catch (const std::out_of_range&) {
-    return Response::build_404(*this, request, bookName);
+    return Response::build_404(*this, request, bookName, "");
   }
 
   if (reader == nullptr) {
-    return Response::build_404(*this, request, bookName);
+    return Response::build_404(*this, request, bookName, "");
   }
 
   try {
     auto entry = reader->getRandomPage();
     return build_redirect(bookName, entry.getFinalEntry());
   } catch(kiwix::NoEntry& e) {
-    return Response::build_404(*this, request, bookName);
+    return Response::build_404(*this, request, bookName, "");
   }
 }
 
@@ -626,7 +600,7 @@ 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, "", "");
 
   auto data = get_default_data();
   data.set("source", source);
@@ -645,11 +619,15 @@ 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, "", "");
+  }
+
+  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, "", "");
   }
 
   if (url == "searchdescription.xml") {
@@ -658,62 +636,79 @@ 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);
+    bookIdsToDump = subrange(bookIdsToDump, startIndex, count);
+    opdsDumper.setOpenSearchInfo(totalResults, startIndex, bookIdsToDump.size());
+    return bookIdsToDump;
 }
 
 namespace
@@ -728,6 +723,15 @@ std::string get_book_name(const RequestContext& request)
   }
 }
 
+std::string searchSuggestionHTML(const std::string& searchURL, const std::string& pattern)
+{
+  kainjow::mustache::mustache tmpl("Make a full text search for <a href=\"{{{searchURL}}}\">{{pattern}}</a>");
+  MustacheData data;
+  data.set("pattern", pattern);
+  data.set("searchURL", searchURL);
+  return (tmpl.render(data));
+}
+
 } // unnamed namespace
 
 std::shared_ptr<Reader>
@@ -751,6 +755,8 @@ InternalServer::build_redirect(const std::string& bookName, const kiwix::Entry&
 
 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");
   }
@@ -761,7 +767,10 @@ std::unique_ptr<Response> InternalServer::handle_content(const RequestContext& r
 
   const std::shared_ptr<Reader> reader = get_reader(bookName);
   if (reader == nullptr) {
-    return Response::build_404(*this, request, bookName);
+    std::string searchURL = m_root+"/search?pattern="+pattern; // Make a full search on the entire library.
+    const std::string details = searchSuggestionHTML(searchURL, kiwix::urlDecode(pattern));
+
+    return Response::build_404(*this, request, bookName, "", details);
   }
 
   auto urlStr = request.get_url().substr(bookName.size()+1);
@@ -791,7 +800,10 @@ std::unique_ptr<Response> InternalServer::handle_content(const RequestContext& r
     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="+pattern; // Make a search on this specific book only.
+    const std::string details = searchSuggestionHTML(searchURL, kiwix::urlDecode(pattern));
+
+    return Response::build_404(*this, request, bookName, reader->getTitle(), details);
   }
 }
 

src/server/internalServer.h

@@ -41,6 +41,7 @@ namespace kiwix {
 typedef kainjow::mustache::data MustacheData;
 
 class Entry;
+class OPDSDumper;
 
 class InternalServer {
   public:
@@ -72,6 +73,10 @@ class InternalServer {
     std::unique_ptr<Response> build_homepage(const RequestContext& request);
     std::unique_ptr<Response> handle_skin(const RequestContext& request);
     std::unique_ptr<Response> handle_catalog(const RequestContext& request);
+    std::unique_ptr<Response> handle_catalog_v2(const RequestContext& request);
+    std::unique_ptr<Response> handle_catalog_v2_root(const RequestContext& request);
+    std::unique_ptr<Response> handle_catalog_v2_entries(const RequestContext& request);
+    std::unique_ptr<Response> handle_catalog_v2_categories(const RequestContext& request);
     std::unique_ptr<Response> handle_meta(const RequestContext& request);
     std::unique_ptr<Response> handle_search(const RequestContext& request);
     std::unique_ptr<Response> handle_suggest(const RequestContext& request);
@@ -79,8 +84,10 @@ class InternalServer {
     std::unique_ptr<Response> handle_captured_external(const RequestContext& request);
     std::unique_ptr<Response> handle_content(const RequestContext& request);
 
+    std::vector<std::string> search_catalog(const RequestContext& request,
+                                            kiwix::OPDSDumper& opdsDumper);
+
     MustacheData get_default_data() const;
-    MustacheData homepage_data() const;
 
     std::shared_ptr<Reader> get_reader(const std::string& bookName) const;
     bool etag_not_needed(const RequestContext& r) const;
@@ -101,9 +108,10 @@ class InternalServer {
     NameMapper* mp_nameMapper;
 
     std::string m_server_id;
+    std::string m_library_id;
 
     friend std::unique_ptr<Response> Response::build(const InternalServer& server);
-    friend std::unique_ptr<ContentResponse> ContentResponse::build(const InternalServer& server, const std::string& content, const std::string& mimetype);
+    friend std::unique_ptr<ContentResponse> ContentResponse::build(const InternalServer& server, const std::string& content, const std::string& mimetype, bool isHomePage);
     friend std::unique_ptr<Response> ItemResponse::build(const InternalServer& server, const RequestContext& request, const zim::Item& item);
     friend std::unique_ptr<Response> Response::build_500(const InternalServer& server, const std::string& msg);
 

src/server/internalServer_catalog_v2.cpp

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

src/server/request_context.cpp

@@ -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 +83,15 @@ 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 RequestContext& request, const std::string& bookName, const std::string& bookTitle, const std::string& details)
 {
   MustacheData results;
   results.set("url", request.get_full_url());
+  results.set("details", details);
 
   auto response = ContentResponse::build(server, RESOURCE::templates::_404_html, results, "text/html");
   response->set_code(MHD_HTTP_NOT_FOUND);
-  response->set_taskbar(bookName, "");
+  response->set_taskbar(bookName, bookTitle);
 
   return std::move(response);
 }
@@ -199,10 +190,10 @@ 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 +212,19 @@ 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
 {
@@ -253,6 +251,8 @@ MHD_Response*
 ContentResponse::create_mhd_response(const RequestContext& request)
 {
   if (contentDecorationAllowed()) {
+    inject_root_link();
+
     if (m_withTaskbar) {
       introduce_taskbar();
     }
@@ -339,21 +339,21 @@ 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)
 {
    return std::unique_ptr<ContentResponse>(new ContentResponse(
         server.m_root,
         server.m_verbose.load(),
-        server.m_withTaskbar,
+        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) :

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 RequestContext& request, const std::string& bookName, const std::string& bookTitle, const std::string& details="");
     static std::unique_ptr<Response> build_416(const InternalServer& server, size_t resourceLength);
     static std::unique_ptr<Response> build_500(const InternalServer& server, const std::string& msg);
     static std::unique_ptr<Response> build_redirect(const InternalServer& server, const std::string& redirectUrl);
@@ -79,8 +79,8 @@ 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);
+    static std::unique_ptr<ContentResponse> build(const InternalServer& server, const std::string& content, const std::string& mimetype, bool isHomePage = false);
+    static std::unique_ptr<ContentResponse> build(const InternalServer& server, const std::string& template_str, kainjow::mustache::data data, const std::string& mimetype, bool isHomePage = false);
 
     void set_taskbar(const std::string& bookName, const std::string& bookTitle);
 
@@ -89,6 +89,7 @@ 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;
 

src/tools/otherTools.cpp

@@ -19,6 +19,7 @@
 
 #include "tools/otherTools.h"
 #include <algorithm>
+#include <iomanip>
 
 #ifdef _WIN32
 #include <windows.h>
@@ -32,6 +33,8 @@
 #include <sstream>
 #include <pugixml.hpp>
 
+#include <zim/uuid.h>
+
 
 static std::map<std::string, std::string> codeisomapping {
 { "aa", "aar" },
@@ -341,3 +344,35 @@ 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/pathTools.cpp

@@ -339,7 +339,7 @@ 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

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

@@ -19,6 +19,11 @@ skin/jquery-ui/jquery-ui.theme.min.css
 skin/jquery-ui/jquery-ui.min.css
 skin/caret.png
 skin/taskbar.js
+skin/langList.js
+skin/categoryList.js
+skin/iso6391To3.js
+skin/isotope.pkgd.min.js
+skin/index.js
 skin/taskbar.css
 skin/block_external.js
 templates/search_result.html
@@ -27,8 +32,13 @@ 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_categories.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/categoryList.js

@@ -0,0 +1,19 @@
+// eslint-disable-next-line no-unused-vars
+const categoryList = {
+  "other": "Other",
+  "gutenberg": "Gutenberg",
+  "mooc": "Mooc",
+  "phet": "Phet",
+  "psiram": "Psiram",
+  "stack_exchange": "Stack Exchange",
+  "ted": "Ted",
+  "vikidia": "Vikidia",
+  "wikibooks": "Wikibooks",
+  "wikinews": "Wikinews",
+  "wikipedia": "Wikipedia",
+  "wikiquote": "Wikiquote",
+  "wikisource": "Wikisource",
+  "wikiversity": "Wikiversity",
+  "wikivoyage": "Wikivoyage",
+  "wiktionary": "Wiktionary"
+}

static/skin/index.js

@@ -0,0 +1,250 @@
+(function() {
+    const root = $(`link[type='root']`).attr('href');
+    const incrementalLoadingParams = {
+        start: 0,
+        count: viewPortToCount()
+    };
+    const filterTypes = ['lang', 'category', 'q'];
+    const bookOrderMap = new Map();
+    const filterCookieName = 'filters';
+    const oneDayDelta = 86400000;
+    let footer;
+    let fadeOutDiv;
+    let iso;
+    let isFetching = false;
+    let noResultInjected = false;
+    let filters = getCookie(filterCookieName);
+    let params = new URLSearchParams(window.location.search || filters || '');
+    let timer;
+
+    function queryUrlBuilder() {
+        let url = `${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=${date.toUTCString()};sameSite=Strict`;
+    }
+
+    function getCookie(cookieName) {
+        const name = cookieName + "=";
+        let result;
+        decodeURIComponent(document.cookie).split('; ').forEach(val => {
+            if (val.indexOf(name) === 0) {
+                result = val.substring(name.length);
+            }
+        });
+        return result;
+    }
+
+    function htmlEncode(str) {
+        return str.replace(/[\u00A0-\u9999<>\&]/gim, (i) => `&#${i.charCodeAt(0)};`);
+    }
+
+    function viewPortToCount(){
+        return Math.floor(window.innerHeight/100 + 1)*(window.innerWidth>1000 ? 3 : 2);
+    }
+
+    function getInnerHtml(node, query) {
+        return node.querySelector(query).innerHTML;
+    }
+
+    function generateBookHtml(book, sort = false) {
+        const link = book.querySelector('link').getAttribute('href');
+        const title =  getInnerHtml(book, 'title');
+        const description = getInnerHtml(book, 'summary');
+        const id = getInnerHtml(book, 'id');
+        const iconUrl = getInnerHtml(book, 'icon');
+        const articleCount = getInnerHtml(book, 'articleCount');
+        const mediaCount = getInnerHtml(book, 'mediaCount');
+
+        const linkTag = document.createElement('a');
+        linkTag.setAttribute('class', 'book');
+        linkTag.setAttribute('data-id', id);
+        linkTag.setAttribute('href', link);
+        if (sort) {
+            linkTag.setAttribute('data-idx', bookOrderMap.get(id));
+        }
+        linkTag.innerHTML = `<div class='book__background' style="background-image: url('${iconUrl}');">
+            <div class='book__title' title='${title}'>${title}</div>
+            <div class='book__description' title='${description}'>${description}</div>
+            <div class='book__info'>${articleCount} articles, ${mediaCount} medias</div>
+        </div>`;
+        return linkTag;
+    }
+
+    function toggleFooter(show=false) {
+        if (show) {
+            footer.style.display = 'block';
+        } else {
+            footer.style.display = 'none';
+            fadeOutDiv.style.display = 'block';
+        }
+    }
+
+    async function loadBooks() {
+        const loader = document.querySelector('.loader');
+        loader.style.display = 'block';
+        return await fetch(queryUrlBuilder()).then(async (resp) => {
+            const data = new window.DOMParser().parseFromString(await resp.text(), 'application/xml');
+            const books = data.querySelectorAll('entry');
+            books.forEach((book, idx) => {
+                bookOrderMap.set(getInnerHtml(book, 'id'), idx);
+            });
+            incrementalLoadingParams.start += books.length;
+            if (parseInt(data.querySelector('totalResults').innerHTML) === bookOrderMap.size) {
+                incrementalLoadingParams.count = 0;
+                toggleFooter(true);
+            } else {
+                toggleFooter();
+            }
+            loader.style.display = 'none';
+            return books;
+        });
+    }
+
+    async function loadAndDisplayOptions(nodeQuery, query) {
+        // currently taking an object in place of query, will replace it with query while fetching data from backend later on.
+        document.querySelector(nodeQuery).innerHTML += Object.keys(query)
+            .map((option) => {return `<option value='${option}'>${htmlEncode(query[option])}</option>`})
+        .join('');
+    }
+
+    function checkAndInjectEmptyMessage() {
+        if (!bookOrderMap.size) {
+            if (!noResultInjected) {
+                noResultInjected = true;
+                iso.remove(document.getElementsByClassName('book__list')[0].getElementsByTagName('a'));
+                iso.layout();
+                const spanTag = document.createElement('span');
+                spanTag.setAttribute('class', 'noResults');
+                spanTag.innerHTML = `No result. Would you like to <a href="/?lang=">reset filter?</a>`;
+                document.querySelector('body').append(spanTag);
+                spanTag.getElementsByTagName('a')[0].onclick = (event) => {
+                    event.preventDefault();
+                    window.history.pushState({}, null, `${window.location.href.split('?')[0]}?lang=`);
+                    setCookie(filterCookieName, 'lang=');
+                    resetAndFilter();
+                    filterTypes.forEach(key => {document.getElementsByName(key)[0].value = params.get(key) || ''});
+                };
+            }
+            return true;
+        } else if (noResultInjected) {
+            noResultInjected = false;
+            document.getElementsByClassName('noResults')[0].remove();
+        }
+        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))});
+    }
+
+    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());
+        }
+        await loadAndDisplayBooks(true);
+    }
+
+    window.addEventListener('popstate', async () => {
+        await resetAndFilter();
+        filterTypes.forEach(key => {document.getElementsByName(key)[0].value = params.get(key) || ''});
+    });
+
+    async function loadSubset() {
+        if (window.innerHeight + window.scrollY >= document.body.offsetHeight) {
+            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'
+        });
+        footer = document.getElementById('kiwixfooter');
+        fadeOutDiv = document.getElementById('fadeOut');
+        await loadAndDisplayBooks();
+        await loadAndDisplayOptions('#languageFilter', langList);
+        await loadAndDisplayOptions('#categoryFilter', categoryList);
+        filterTypes.forEach((filter) => {
+            const filterTag = document.getElementsByName(filter)[0];
+            filterTag.addEventListener('change', () => {resetAndFilter(filterTag.name, filterTag.value)});
+        });
+        if (filters) {
+            window.history.pushState({}, null, `${window.location.href.split('?')[0]}?${params.toString()}`);
+        }
+        params.forEach((value, key) => {document.getElementsByName(key)[0].value = value});
+        document.getElementById('kiwixSearchForm').onsubmit = (event) => {event.preventDefault()};
+        if (!window.location.search) {
+            const browserLang = navigator.language.split('-')[0];
+            const langFilter = document.getElementById('languageFilter');
+            langFilter.value = browserLang.length === 3 ? browserLang : iso6391To3[browserLang];
+            langFilter.dispatchEvent(new Event('change'));
+        }
+        setCookie(filterCookieName, params.toString());
+    }
+})();

static/skin/iso6391To3.js

@@ -0,0 +1,124 @@
+// eslint-disable-next-line no-unused-vars
+const iso6391To3 = {
+    "aa": "aar",
+    "af": "afr",
+    "ak": "aka",
+    "am": "amh",
+    "ar": "ara",
+    "as": "asm",
+    "az": "aze",
+    "ba": "bak",
+    "be": "bel",
+    "bg": "bul",
+    "bm": "bam",
+    "bn": "ben",
+    "bo": "bod",
+    "br": "bre",
+    "bs": "bos",
+    "ca": "cat",
+    "ce": "che",
+    "co": "cos",
+    "cs": "ces",
+    "cv": "chv",
+    "cy": "cym",
+    "da": "dan",
+    "de": "deu",
+    "dz": "dzo",
+    "ee": "ewe",
+    "en": "eng",
+    "es": "spa",
+    "et": "est",
+    "eu": "eus",
+    "fa": "fas",
+    "ff": "ful",
+    "fi": "fin",
+    "fo": "fao",
+    "fr": "fra",
+    "ga": "gle",
+    "gl": "glg",
+    "gn": "grn",
+    "gu": "guj",
+    "gv": "glv",
+    "ha": "hau",
+    "he": "heb",
+    "hi": "hin",
+    "hr": "hrv",
+    "hu": "hun",
+    "hy": "hye",
+    "id": "ind",
+    "ig": "ibo",
+    "is": "isl",
+    "it": "ita",
+    "iu": "iku",
+    "ja": "jpn",
+    "jv": "jav",
+    "ka": "kat",
+    "ki": "kik",
+    "kk": "kaz",
+    "km": "khm",
+    "kn": "kan",
+    "ko": "kor",
+    "ks": "kas",
+    "ku": "kur",
+    "kw": "cor",
+    "ky": "kir",
+    "lb": "ltz",
+    "lg": "lug",
+    "ln": "lin",
+    "lo": "lao",
+    "lt": "lit",
+    "lv": "lav",
+    "mg": "mlg",
+    "mi": "mri",
+    "mk": "mkd",
+    "ml": "mal",
+    "mn": "mon",
+    "mr": "mar",
+    "mt": "mlt",
+    "my": "mya",
+    "nl": "nld",
+    "ny": "nya",
+    "om": "orm",
+    "pl": "pol",
+    "pt": "por",
+    "qu": "que",
+    "rm": "roh",
+    "rn": "run",
+    "ro": "ron",
+    "ru": "rus",
+    "rw": "kin",
+    "sa": "san",
+    "sd": "snd",
+    "sg": "sag",
+    "si": "sin",
+    "sk": "slk",
+    "sl": "slv",
+    "sn": "sna",
+    "so": "som",
+    "sq": "sqi",
+    "sr": "srp",
+    "ss": "ssw",
+    "sv": "swe",
+    "ta": "tam",
+    "te": "tel",
+    "tg": "tgk",
+    "th": "tha",
+    "ti": "tir",
+    "tk": "tuk",
+    "tn": "tsn",
+    "tr": "tur",
+    "ts": "tso",
+    "tt": "tat",
+    "ug": "uig",
+    "uk": "ukr",
+    "ur": "urd",
+    "uz": "uzb",
+    "ve": "ven",
+    "vi": "vie",
+    "wa": "wln",
+    "wo": "wol",
+    "xh": "xho",
+    "yo": "yor",
+    "zh": "zho",
+    "zu": "zul"
+}

static/skin/langList.js

@@ -0,0 +1,124 @@
+// eslint-disable-next-line no-unused-vars
+const langList = {
+  "aar": "Afaraf",
+  "afr": "Afrikaans",
+  "aka": "Akan",
+  "amh": "አማርኛ",
+  "ara": "اللغة العربية",
+  "asm": "অসমীয়া",
+  "aze": "azərbaycan dili",
+  "bak": "башҡорт теле",
+  "bel": "беларуская мова",
+  "bul": "български език",
+  "bam": "bamanankan",
+  "ben": "বাংলা",
+  "bod": "བོད་ཡིག",
+  "bre": "brezhoneg",
+  "bos": "bosanski jezik",
+  "cat": "Català",
+  "che": "нохчийн мотт",
+  "cos": "corsu",
+  "ces": "čeština",
+  "chv": "чӑваш чӗлхи",
+  "cym": "Cymraeg",
+  "dan": "dansk",
+  "deu": "Deutsch",
+  "dzo": "རྫོང་ཁ",
+  "ewe": "Eʋegbe",
+  "eng": "English",
+  "spa": "Español",
+  "est": "eesti",
+  "eus": "euskara",
+  "fas": "فارسی",
+  "ful": "Fulfulde",
+  "fin": "suomi",
+  "fao": "føroyskt",
+  "fra": "Français",
+  "gle": "Gaeilge",
+  "glg": "galego",
+  "grn": "Avañe'ẽ",
+  "guj": "ગુજરાતી",
+  "glv": "Gaelg",
+  "hau": "هَوُسَ",
+  "heb": "עברית",
+  "hin": "हिन्दी",
+  "hrv": "hrvatski jezik",
+  "hun": "magyar",
+  "hye": "Հայերեն",
+  "ind": "Bahasa Indonesia",
+  "ibo": "Asụsụ Igbo",
+  "isl": "Íslenska",
+  "ita": "Italiano",
+  "iku": "ᐃᓄᒃᑎᑐᑦ",
+  "jpn": "日本語",
+  "jav": "basa Jawa",
+  "kat": "ქართული",
+  "kik": "Gĩkũyũ",
+  "kaz": "қазақ тілі",
+  "khm": "ខេមរភាសា",
+  "kan": "ಕನ್ನಡ",
+  "kor": "한국어",
+  "kas": "कश्मीरी",
+  "kur": "Kurdî",
+  "cor": "Kernewek",
+  "kir": "Кыргызча",
+  "ltz": "Lëtzebuergesch",
+  "lug": "Luganda",
+  "lin": "Lingála",
+  "lao": "ພາສາ",
+  "lit": "lietuvių kalba",
+  "lav": "latviešu valoda",
+  "mlg": "fiteny malagasy",
+  "mri": "te reo Māori",
+  "mkd": "македонски јазик",
+  "mal": "മലയാളം",
+  "mon": "Монгол хэл",
+  "mar": "मराठी",
+  "mlt": "Malti",
+  "mya": "ဗမာစာ",
+  "nld": "Nederlands",
+  "nya": "chiCheŵa",
+  "orm": "Afaan Oromoo",
+  "pol": "język polski",
+  "por": "Português",
+  "que": "Runa Simi",
+  "roh": "rumantsch grischun",
+  "run": "Ikirundi",
+  "ron": "Română",
+  "rus": "Русский",
+  "kin": "Ikinyarwanda",
+  "san": "संस्कृतम्",
+  "snd": "सिन्धी",
+  "sag": "yângâ tî sängö",
+  "sin": "සිංහල",
+  "slk": "slovenčina",
+  "slv": "slovenski jezik",
+  "sna": "chiShona",
+  "som": "Soomaaliga",
+  "sqi": "Shqip",
+  "srp": "српски језик",
+  "ssw": "SiSwati",
+  "swe": "svenska",
+  "tam": "தமிழ்",
+  "tel": "తెలుగు",
+  "tgk": "тоҷикӣ",
+  "tha": "ไทย",
+  "tir": "ትግርኛ",
+  "tuk": "Türkmen",
+  "tsn": "Setswana",
+  "tur": "Türkçe",
+  "tso": "Xitsonga",
+  "tat": "татар теле",
+  "uig": "ئۇيغۇرچە‎",
+  "ukr": "Українська",
+  "urd": "اردو",
+  "uzb": "Ўзбек",
+  "ven": "Tshivenḓa",
+  "vie": "Tiếng Việt",
+  "wln": "walon",
+  "wol": "Wollof",
+  "xho": "isiXhosa",
+  "yor": "Yorùbá",
+  "zho": "中文",
+  "zul": "isiZulu"
+}

static/skin/taskbar.js

@@ -1,49 +1,96 @@
+function htmlDecode(input) {
+    var doc = new DOMParser().parseFromString(input, "text/html");
+    return doc.documentElement.textContent;
+}
 
-(function ($) {
-    if ($(window).width() < 520) {
-        var didScroll;
-        var lastScrollTop = 0;
-        var delta = 5;
-        // on scroll, let the interval function know the user has scrolled
-        $(window).scroll(function (event) {
-            didScroll = true;
+const jq = jQuery.noConflict(true);
+jq(document).ready(() => {
+    (function ($) {
+        const root = $( `link[type='root']` ).attr("href");
+
+        const bookName = (window.location.pathname == `${root}/search`)
+               ? (new URLSearchParams(window.location.search)).get('content') 
+               : window.location.pathname.split(`${root}/`)[1].split('/')[0];
+    
+        $( "#kiwixsearchbox" ).autocomplete({
+    
+            source: `${root}/suggest?content=${bookName}`,
+            dataType: "json",
+            cache: false,
+    
+            response: function( event, ui ) {
+                  for(const item of ui.content) {
+                      item.label = htmlDecode(item.label);
+                      item.value = htmlDecode(item.value);
+                      if (item.path) item.path = htmlDecode(item.path);
+                  }
+            },
+    
+            select: function(event, ui) {
+                if (ui.item.kind === 'path') {
+                    window.location.href = `${root}/${bookName}/${encodeURI(ui.item.path)}`;
+                } else {
+                    $( "#kiwixsearchbox" ).val(ui.item.value);
+                    $( "#kiwixsearchform" ).submit();
+                }
+            },
+        }).data( "ui-autocomplete" )._renderItem = function( ul, item ) {
+            return $( "<li>" )
+              .data( "ui-autocomplete-item", item )
+              .append( item.label )
+              .appendTo( ul );
+        };
+    
+        /* cybook hack */
+        if (navigator.userAgent.indexOf("bookeen/cybook") != -1) {
+            $("html").addClass("cybook");
+        }
+    
+        if ($(window).width() < 520) {
+            var didScroll;
+            var lastScrollTop = 0;
+            var delta = 5;
+            // on scroll, let the interval function know the user has scrolled
+            $(window).scroll(function (event) {
+                didScroll = true;
+            });
+            // run hasScrolled() and reset didScroll status
+            setInterval(function () {
+                if (didScroll) {
+                    hasScrolled();
+                    didScroll = false;
+                }
+            }, 250);
+            function hasScrolled() {
+                var st = $(this).scrollTop();
+    
+                // Make sure they scroll more than delta
+                if (Math.abs(lastScrollTop - st) <= delta)
+                    return;
+    
+                // If they scrolled down and are past the navbar, add class .nav-up.
+                // This is necessary so you never see what is "behind" the navbar.
+                if (st > lastScrollTop) {
+                    // Scroll Down
+                    $('#kiwixtoolbar').css({ top: '-100%' });
+                } else {
+                    // Scroll Up
+                    $('#kiwixtoolbar').css({ top: '0' });
+                }
+    
+                lastScrollTop = st;
+            }
+        }
+    
+        $('#kiwixsearchbox').on({
+            focus: function () {
+                $('.kiwix_searchform').addClass('full_width');
+                $('label[for="kiwix_button_show_toggle"], .kiwix_button_cont').addClass('searching');
+            },
+            blur: function () {
+                $('.kiwix_searchform').removeClass('full_width');
+                $('label[for="kiwix_button_show_toggle"], .kiwix_button_cont').removeClass('searching');
+            }
         });
-        // run hasScrolled() and reset didScroll status
-        setInterval(function () {
-            if (didScroll) {
-                hasScrolled();
-                didScroll = false;
-            }
-        }, 250);
-        function hasScrolled() {
-            var st = $(this).scrollTop();
-
-            // Make sure they scroll more than delta
-            if (Math.abs(lastScrollTop - st) <= delta)
-                return;
-
-            // If they scrolled down and are past the navbar, add class .nav-up.
-            // This is necessary so you never see what is "behind" the navbar.
-            if (st > lastScrollTop) {
-                // Scroll Down
-                $('#kiwixtoolbar').css({ top: '-100%' });
-            } else {
-                // Scroll Up
-                $('#kiwixtoolbar').css({ top: '0' });
-            }
-
-            lastScrollTop = st;
-        }
-    }
-
-    $('#kiwixsearchbox').on({
-        focus: function () {
-            $('.kiwix_searchform').addClass('full_width');
-            $('label[for="kiwix_button_show_toggle"], .kiwix_button_cont').addClass('searching');
-        },
-        blur: function () {
-            $('.kiwix_searchform').removeClass('full_width');
-            $('label[for="kiwix_button_show_toggle"], .kiwix_button_cont').removeClass('searching');
-        }
-    });
-})(jQuery);
+    })(jq);
+})

static/templates/404.html

@@ -9,5 +9,10 @@
     <p>
       The requested URL "{{url}}" was not found on this server.
     </p>
+    {{#details}}
+    <p>
+      {{{details}}}
+    </p>
+    {{/details}}
   </body>
 </html>

static/templates/catalog_entries.xml

@@ -0,0 +1,38 @@
+<feed xmlns="http://www.w3.org/2005/Atom" xmlns:opds="http://opds-spec.org/2010/catalog">
+  <id>{{feed_id}}</id>
+  <title>{{^filter}}All zims{{/filter}}{{#filter}}Filtered zims ({{filter}}){{/filter}}</title>
+  <updated>{{date}}</updated>
+{{#filter}}
+  <totalResults>{{totalResults}}</totalResults>
+  <startIndex>{{startIndex}}</startIndex>
+  <itemsPerPage>{{itemsPerPage}}</itemsPerPage>
+{{/filter}}
+  <link rel="self" href="" type="application/atom+xml" />
+  <link rel="search" type="application/opensearchdescription+xml" href="{{root}}/catalog/searchdescription.xml" />
+  {{#books}}
+  <entry>
+    <id>{{id}}</id>
+    <title>{{title}}</title>
+    <summary>{{description}}</summary>
+    <language>{{language}}</language>
+    <updated>{{updated}}</updated>
+    <name>{{name}}</name>
+    <flavour>{{flavour}}</flavour>
+    <category>{{category}}</category>
+    <tags>{{tags}}</tags>
+    <articleCount>{{article_count}}</articleCount>
+    <mediaCount>{{media_count}}</mediaCount>
+    <icon>/meta?name=favicon&amp;content={{{content_id}}}</icon>
+    <link type="text/html" href="/{{{content_id}}}" />
+    <author>
+      <name>{{author_name}}</name>
+    </author>
+    <publisher>
+      <name>{{publisher_name}}</name>
+    </publisher>
+    {{#url}}
+    <link rel="http://opds-spec.org/acquisition/open-access" type="application/x-zim" href="{{{url}}}" length="{{{size}}}" />
+    {{/url}}
+  </entry>
+  {{/books}}
+</feed>

static/templates/catalog_v2_categories.xml

@@ -0,0 +1,25 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<feed xmlns="http://www.w3.org/2005/Atom"
+      xmlns:opds="https://specs.opds.io/opds-1.2">
+  <id>{{feed_id}}</id>
+  <link rel="self"
+        href="{{endpoint_root}}/categories"
+        type="application/atom+xml;profile=opds-catalog;kind=navigation"/>
+  <link rel="start"
+        href="{{endpoint_root}}/root.xml"
+        type="application/atom+xml;profile=opds-catalog;kind=navigation"/>
+  <title>List of categories</title>
+  <updated>{{date}}</updated>
+
+  {{#categories}}
+  <entry>
+    <title>{{name}}</title>
+    <link rel="subsection"
+          href="{{endpoint_root}}/entries?category={{{urlencoded_name}}}"
+          type="application/atom+xml;profile=opds-catalog;kind=acquisition"/>
+    <updated>{{updated}}</updated>
+    <id>{{id}}</id>
+    <content type="text">All entries with category of '{{name}}'.</content>
+  </entry>
+  {{/categories}}
+</feed>

static/templates/catalog_v2_entries.xml

@@ -0,0 +1,50 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<feed xmlns="http://www.w3.org/2005/Atom"
+      xmlns:opds="https://specs.opds.io/opds-1.2"
+      xmlns:opensearch="http://a9.com/-/spec/opensearch/1.1/">
+  <id>{{feed_id}}</id>
+
+  <link rel="self"
+        href="{{endpoint_root}}/entries{{{query}}}"
+        type="application/atom+xml;profile=opds-catalog;kind=acquisition"/>
+  <link rel="start"
+        href="{{endpoint_root}}/root.xml"
+        type="application/atom+xml;profile=opds-catalog;kind=navigation"/>
+  <link rel="up"
+        href="{{endpoint_root}}/root.xml"
+        type="application/atom+xml;profile=opds-catalog;kind=navigation"/>
+
+  <title>{{^filter}}All Entries{{/filter}}{{#filter}}Filtered Entries ({{filter}}){{/filter}}</title>
+  <updated>{{date}}</updated>
+{{#filter}}
+  <totalResults>{{totalResults}}</totalResults>
+  <startIndex>{{startIndex}}</startIndex>
+  <itemsPerPage>{{itemsPerPage}}</itemsPerPage>
+{{/filter}}
+  {{#books}}
+  <entry>
+    <id>{{id}}</id>
+    <title>{{title}}</title>
+    <summary>{{description}}</summary>
+    <language>{{language}}</language>
+    <updated>{{updated}}</updated>
+    <name>{{name}}</name>
+    <flavour>{{flavour}}</flavour>
+    <category>{{category}}</category>
+    <tags>{{tags}}</tags>
+    <articleCount>{{article_count}}</articleCount>
+    <mediaCount>{{media_count}}</mediaCount>
+    <icon>/meta?name=favicon&amp;content={{{content_id}}}</icon>
+    <link type="text/html" href="/{{{content_id}}}" />
+    <author>
+      <name>{{author_name}}</name>
+    </author>
+    <publisher>
+      <name>{{publisher_name}}</name>
+    </publisher>
+    {{#url}}
+    <link rel="http://opds-spec.org/acquisition/open-access" type="application/x-zim" href="{{{url}}}" length="{{{size}}}" />
+    {{/url}}
+  </entry>
+  {{/books}}
+</feed>

static/templates/catalog_v2_root.xml

@@ -0,0 +1,35 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<feed xmlns="http://www.w3.org/2005/Atom"
+      xmlns:opds="https://specs.opds.io/opds-1.2">
+  <id>{{feed_id}}</id>
+  <link rel="self"
+        href="{{endpoint_root}}/root.xml"
+        type="application/atom+xml;profile=opds-catalog;kind=navigation"/>
+  <link rel="start"
+        href="{{endpoint_root}}/root.xml"
+        type="application/atom+xml;profile=opds-catalog;kind=navigation"/>
+  <link rel="search"
+        href="{{endpoint_root}}/searchdescription.xml"
+        type="application/opensearchdescription+xml"/>
+  <title>OPDS Catalog Root</title>
+  <updated>{{date}}</updated>
+
+  <entry>
+    <title>All entries</title>
+    <link rel="subsection"
+          href="{{endpoint_root}}/entries"
+          type="application/atom+xml;profile=opds-catalog;kind=acquisition"/>
+    <updated>{{date}}</updated>
+    <id>{{all_entries_feed_id}}</id>
+    <content type="text">All entries from this catalog.</content>
+  </entry>
+  <entry>
+    <title>List of categories</title>
+    <link rel="subsection"
+          href="{{endpoint_root}}/categories"
+          type="application/atom+xml;profile=opds-catalog;kind=navigation"/>
+    <updated>{{date}}</updated>
+    <id>{{category_list_feed_id}}</id>
+    <content type="text">List of all categories in this catalog.</content>
+  </entry>
+</feed>

static/templates/head_part.html

@@ -1,28 +0,0 @@
-<link type="text/css" href="{{root}}/skin/jquery-ui/jquery-ui.min.css" rel="Stylesheet" />
-<link type="text/css" href="{{root}}/skin/jquery-ui/jquery-ui.theme.min.css" rel="Stylesheet" />
-<link type="text/css" href="{{root}}/skin/taskbar.css" rel="Stylesheet" />
-<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>
-  var jk = jQuery.noConflict();
-  jk(function() {
-  jk( "#kiwixsearchbox" ).autocomplete({
-
-  source: "{{root}}/suggest?content={{#urlencoded}}{{{content}}}{{/urlencoded}}",
-  dataType: "json",
-  cache: false,
-
-  select: function(event, ui) {
-  jk( "#kiwixsearchbox" ).val(ui.item.value);
-  jk( "#kiwixsearchform" ).submit();
-  },
-
-  });
-  });
-
-  /* cybook hack */
-  if (navigator.userAgent.indexOf("bookeen/cybook") != -1) {
-  	jk("html").addClass("cybook");
-  }
-</script>
-<script type="text/javascript" src="{{root}}/skin/taskbar.js" async></script>

static/templates/head_taskbar.html

@@ -0,0 +1,6 @@
+<link type="text/css" href="{{root}}/skin/jquery-ui/jquery-ui.min.css" rel="Stylesheet" />
+<link type="text/css" href="{{root}}/skin/jquery-ui/jquery-ui.theme.min.css" rel="Stylesheet" />
+<link type="text/css" href="{{root}}/skin/taskbar.css" rel="Stylesheet" />
+<script type="text/javascript" src="{{root}}/skin/jquery-ui/external/jquery/jquery.js" defer></script>
+<script type="text/javascript" src="{{root}}/skin/jquery-ui/jquery-ui.min.js" defer></script>
+<script type="text/javascript" src="{{root}}/skin/taskbar.js" defer></script>

static/templates/index.html

@@ -1,66 +1,206 @@
 <!DOCTYPE html>
 <html xmlns="http://www.w3.org/1999/xhtml">
-<head>
-  <meta charset="UTF-8" />
-  <title>Welcome to Kiwix Server</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>
-  <link type="text/css" href="{{root}}/skin/jquery-ui/jquery-ui.min.css" rel="Stylesheet" />
-  <link type="text/css" href="{{root}}/skin/jquery-ui/jquery-ui.theme.min.css" rel="Stylesheet" />
-  <style>
-        body {
-                background:
-                radial-gradient(#EEEEEE 15%, transparent 16%) 0 0,
-                radial-gradient(#EEEEEE 15%, transparent 16%) 8px 8px,
-                radial-gradient(rgba(255,255,255,.1) 15%, transparent 20%) 0 1px,
-                radial-gradient(rgba(255,255,255,.1) 15%, transparent 20%) 8px 9px;
-                background-color:#E8E8E8;
-                background-size:16px 16px;
-                margin-left: auto;
-                margin-right: auto;
-                max-width: 1100px;
+  <head>
+    <meta charset="UTF-8" />
+    <title>Welcome to Kiwix Server</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>
+    <link
+      type="text/css"
+      href="{{root}}/skin/jquery-ui/jquery-ui.min.css"
+      rel="Stylesheet"
+    />
+    <link
+      type="text/css"
+      href="{{root}}/skin/jquery-ui/jquery-ui.theme.min.css"
+      rel="Stylesheet"
+    />
+    <style>
+      html {
+        min-height: 100%;
+        position: relative;
+      }
+      body {
+        background: radial-gradient(#eeeeee 15%, transparent 16%) 0 0,
+          radial-gradient(#eeeeee 15%, transparent 16%) 8px 8px,
+          radial-gradient(rgba(255, 255, 255, 0.1) 15%, transparent 20%) 0 1px,
+          radial-gradient(rgba(255, 255, 255, 0.1) 15%, transparent 20%) 8px 9px;
+        background-color: #e8e8e8;
+        background-size: 16px 16px;
+        margin-left: auto;
+        margin-right: auto;
+        max-width: 1100px;
+        min-height: 100%;
+      }
+      .book__list {
+        text-align: center;
+      }
+      .kiwixHomeBody {
+        position: relative;
+        text-align: center;
+        min-height: 100%;
+        margin: 0 0 15px;
+      }
+      .book {
+        display: inline-block;
+        vertical-align: bottom;
+        margin: 8px;
+        padding: 12px 15px;
+        width: 300px;
+        border: 1px solid #ccc;
+        border-radius: 8px;
+        text-align: left;
+        color: #000;
+        font-family: sans-serif;
+        font-size: 13px;
+        background-color: #f1f1f1;
+        box-shadow: 2px 2px 5px 0 #ccc;
+      }
+      #kiwixfooter {
+        text-align: center;
+        margin: 0.5em;
+        position: absolute;
+        bottom: 0;
+        left: 46%;
+      }
+      .kiwixHomeNavbar {
+        display: flex;
+        justify-content: center;
+      }
+      .kiwixFilter {
+        margin: 8px 10px;
+      }
+      .kiwixSearch, .searchButton {
+        margin: 0 13px 0 0;
+      }
+      .kiwixSearchForm {
+        margin: 8px 10px;
+        float: right;
+      }
+      @media (max-width: 1100px) {
+        .kiwixHomeBody {
+          padding: 0 125px;
         }
-        .book__list { text-align: center; }
-        .book {
-                display: inline-block; vertical-align: bottom; margin: 8px; padding: 12px 15px; width: 300px;
-                border: 1px solid #ccc; border-radius: 8px;
-                text-align: left; color: #000; font-family: sans-serif; font-size: 13px;
-                background-color:#F1F1F1;
-                box-shadow: 2px 2px 5px 0px #ccc;
-        }
-        .book:hover { background-color: #F9F9F9; box-shadow: none;}
-        .book__background { background-repeat: no-repeat; background-size: 48px 48px; background-position: top right; }
-        .book__title {
-                padding: 0 55px 0 0;overflow: hidden; text-overflow: ellipsis; white-space: nowrap;
-                font-size: 18px; color: #0645ad; line-height: 1em;
-        }
-        .book__description {
-                padding: 5px 55px 5px 0px; overflow: hidden; text-overflow: ellipsis; white-space: nowrap;
-                font-size: 15px; line-height: 1em;
-        }
-        .book__info { color: #777; font-weight: bold; font-size: 13px; line-height: 1em; }
-  </style>
-  <script type="text/javascript" src="{{root}}/skin/taskbar.js" async></script>
-</head>
-<body class="kiwix">
-
-<div class="kiwix">
-  <div class='book__list'>
-    {{#books}}
-    <a href="{{root}}/{{name}}"><div class='book'>
-        <div class='book__background' style="background-image: url('{{root}}/meta?content={{#urlencoded}}{{{name}}}{{/urlencoded}}&name=favicon');">
-          <div class='book__title' title='{{title}}'>{{title}}</div>
-          <div class='book__description' title='{{description}}'>{{description}}</div>
-          <div class='book__info'>{{articleCount}} articles, {{mediaCount}} medias</div>
-        </div>
-    </div></a>
-    {{/books}}
-  </div>
-</div>
-
-<div id="kiwixfooter">
-  Powered by <a href="https://kiwix.org">Kiwix</a>
-</div>
-
-</body>
+      }
+      .book:hover {
+        background-color: #f9f9f9;
+        box-shadow: none;
+      }
+      .book__background {
+        background-repeat: no-repeat;
+        background-size: 48px 48px;
+        background-position: top right;
+      }
+      .book__title {
+        padding: 0 55px 0 0;
+        overflow: hidden;
+        text-overflow: ellipsis;
+        white-space: nowrap;
+        font-size: 18px;
+        color: #0645ad;
+        line-height: 1em;
+      }
+      .book__description {
+        padding: 5px 55px 5px 0;
+        overflow: hidden;
+        text-overflow: ellipsis;
+        white-space: nowrap;
+        font-size: 15px;
+        line-height: 1em;
+      }
+      .book__info {
+        color: #777;
+        font-weight: bold;
+        font-size: 13px;
+        line-height: 1em;
+      }
+      a:link {
+        text-decoration: none;
+      }
+      a:visited {
+        text-decoration: none;
+      }
+      .noResults {
+        position: absolute;
+        top: 48%;
+        left: 42%;
+      }
+      .loader-spinner {
+        position: absolute;
+        top: -50%;
+        left: 50%;
+        border: 5px solid #f3f3f3;
+        border-radius: 50%;
+        border-top: 5px solid #3498db;
+        width: 40px;
+        height: 40px;
+        margin: auto;
+        -webkit-animation: spin 1s linear infinite; /* Safari */
+        animation: spin 1s linear infinite;
+        margin-top: 35px;
+        margin-bottom: -35px;
+        z-index: 2;
+      }
+      /* Safari */
+      @-webkit-keyframes spin {
+        0% { -webkit-transform: rotate(0deg); }
+        100% { -webkit-transform: rotate(360deg); }
+      }
+      @keyframes spin {
+        0% { transform: rotate(0deg); }
+        100% { transform: rotate(360deg); }
+      }
+      .loader {
+        position: relative;
+        height: 70px;
+        width: 100%;
+      }
+      .fadeOut {
+        position: fixed;
+        display: none;
+        bottom: 0;
+        left: 0;
+        z-index: 1;
+        background: linear-gradient(180deg, rgba(232, 232, 232, 0) 0%, rgb(232, 232, 232) 100%);
+        height: 80px;
+        width: 100%;
+      }
+      .spacer {
+        height: 20px;
+        background: transparent;
+      }
+    </style>
+    <script src="{{root}}/skin/isotope.pkgd.min.js" defer></script>
+    <script src="{{root}}/skin/categoryList.js"></script>
+    <script src="{{root}}/skin/langList.js"></script>
+    <script src="{{root}}/skin/iso6391To3.js"></script>
+    <script type="text/javascript" src="{{root}}/skin/index.js" defer></script>
+  </head>
+  <body class="kiwix">
+    <div class='kiwixHomeNavbar'>
+      <select name="lang" id="languageFilter" class='kiwixFilter'>
+        <option value="" selected>All languages</option>
+      </select>
+      <select name="category" id="categoryFilter" class='kiwixFilter'>
+        <option value="" selected>All categories</option>
+      </select>
+      <form id='kiwixSearchForm' class='kiwixSearchForm'>
+        <input type="text" name="q" id="searchFilter" class='kiwixSearch'>
+        <input type="submit" class="searchButton" value="Search"/>
+      </form>
+    </div>
+    <div class="kiwixHomeBody">
+      <div class="book__list"></div>
+      <div id="fadeOut" class="fadeOut"></div>
+    </div>
+    <div class="loader"><div class="loader-spinner"></div></div>
+    <div class="spacer"></div>
+    <div id="kiwixfooter">Powered by <a href="https://kiwix.org">Kiwix</a></div>
+  </body>
 </html>

static/templates/suggestion.json

@@ -2,6 +2,10 @@
   {{#suggestions}}{{^first}},{{/first}}
   {
     "value" : "{{value}}",
-    "label" : "{{& label}}"
+    "label" : "{{label}}",
+    "kind" : "{{kind}}"
+    {{#path}}
+      , "path" : "{{path}}"
+    {{/path}}
   }{{/suggestions}}
 ]

static/templates/taskbar_part.html

@@ -8,18 +8,18 @@
           <input autocomplete="off" class="ui-autocomplete-input" id="kiwixsearchbox" name="pattern" type="text">
         </form>
       </div>
-      {{#hascontent}}
         <input type="checkbox" id="kiwix_button_show_toggle">
         <label for="kiwix_button_show_toggle"><img src="{{root}}/skin/caret.png" alt=""></label>
         <div class="kiwix_button_cont">
           {{#withlibrarybutton}}
             <a id="kiwix_serve_taskbar_library_button" href="{{root}}/"><button>&#x1f3e0;</button></a>
           {{/withlibrarybutton}}
+          {{#hascontent}}
           <a id="kiwix_serve_taskbar_home_button" href="{{root}}/{{content}}/"><button>{{title}}</button></a>
           <a id="kiwix_serve_taskbar_random_button"
             href="{{root}}/random?content={{#urlencoded}}{{{content}}}{{/urlencoded}}"><button>&#x1F3B2;</button></a>
+          {{/hascontent}}
         </div>
-       {{/hascontent}}
     </div>
   </span>
 </span>

test/book.cpp

@@ -1,10 +1,12 @@
 #include "gtest/gtest.h"
 #include "../include/book.h"
+#include <pugixml.hpp>
 
 TEST(BookTest, updateTest)
 {
     kiwix::Book book;
 
+    book.setId("xyz");
     book.setReadOnly(false);
     book.setPath("/home/user/Downloads/skin-of-color-society_en_all_2019-11.zim");
     book.setPathValid(true);
@@ -20,6 +22,9 @@ TEST(BookTest, updateTest)
     EXPECT_FALSE(newBook.update(book));
 
     newBook.setReadOnly(false);
+    EXPECT_FALSE(newBook.update(book));
+
+    newBook.setId("xyz");
     EXPECT_TRUE(newBook.update(book));
 
     EXPECT_EQ(newBook.readOnly(), book.readOnly());
@@ -27,7 +32,137 @@ TEST(BookTest, updateTest)
     EXPECT_EQ(newBook.isPathValid(), book.isPathValid());
     EXPECT_EQ(newBook.getUrl(), book.getUrl());
     EXPECT_EQ(newBook.getTags(), book.getTags());
+    EXPECT_EQ(newBook.getCategory(), book.getCategory());
     EXPECT_EQ(newBook.getName(), book.getName());
     EXPECT_EQ(newBook.getFavicon(), book.getFavicon());
     EXPECT_EQ(newBook.getFaviconMimeType(), book.getFaviconMimeType());
 }
+
+namespace
+{
+
+struct XMLDoc : pugi::xml_document
+{
+  explicit XMLDoc(const std::string& xml)
+  {
+    load_buffer(xml.c_str(), xml.size());
+  }
+};
+
+} // unnamed namespace
+
+TEST(BookTest, updateFromXMLTest)
+{
+    const XMLDoc xml(R"(
+      <book id="zara"
+            path="./zara.zim"
+            url="https://who.org/zara.zim"
+            title="Catch an infection in 24 hours"
+            description="Complete guide to contagious diseases"
+            language="eng"
+            creator="World Health Organization"
+            publisher="WHO"
+            date="2020-03-31"
+            name="who_contagious_diseases_en"
+            tags="unittest;_category:medicine;_pictures:yes"
+            articleCount="123456"
+            mediaCount="234567"
+            size="345678"
+          >
+      </book>
+    )");
+
+    kiwix::Book book;
+    book.updateFromXml(xml.child("book"), "/data/zim");
+
+    EXPECT_EQ(book.getPath(), "/data/zim/zara.zim");
+    EXPECT_EQ(book.getUrl(), "https://who.org/zara.zim");
+    EXPECT_EQ(book.getTitle(), "Catch an infection in 24 hours");
+    EXPECT_EQ(book.getDescription(), "Complete guide to contagious diseases");
+    EXPECT_EQ(book.getTags(), "unittest;_category:medicine;_pictures:yes");
+    EXPECT_EQ(book.getName(), "who_contagious_diseases_en");
+    EXPECT_EQ(book.getCategory(), "medicine");
+    EXPECT_EQ(book.getArticleCount(), 123456U);
+    EXPECT_EQ(book.getMediaCount(), 234567U);
+    EXPECT_EQ(book.getSize(), 345678U*1024U);
+}
+
+TEST(BookTest, updateFromXMLCategoryHandlingTest)
+{
+  {
+    const XMLDoc xml(R"(
+      <book id="abcd"
+            tags="_category:category_defined_via_tags_only"
+          >
+      </book>
+    )");
+
+    kiwix::Book book;
+    book.updateFromXml(xml.child("book"), "");
+
+    EXPECT_EQ(book.getCategory(), "category_defined_via_tags_only");
+  }
+  {
+    const XMLDoc xml(R"(
+      <book id="abcd"
+            category="category_defined_via_attribute_only"
+          >
+      </book>
+    )");
+
+    kiwix::Book book;
+    book.updateFromXml(xml.child("book"), "");
+
+    EXPECT_EQ(book.getCategory(), "category_defined_via_attribute_only");
+  }
+  {
+    const XMLDoc xml(R"(
+      <book id="abcd"
+            category="category_attribute_overrides_tags"
+            tags="_category:tags_override_category_attribute"
+          >
+      </book>
+    )");
+
+    kiwix::Book book;
+    book.updateFromXml(xml.child("book"), "");
+
+    EXPECT_EQ(book.getCategory(), "category_attribute_overrides_tags");
+  }
+  {
+    const XMLDoc xml(R"(
+      <book id="abcd"
+            tags="_category:tags_override_category_attribute"
+            category="category_attribute_overrides_tags"
+          >
+      </book>
+    )");
+
+    kiwix::Book book;
+    book.updateFromXml(xml.child("book"), "");
+
+    EXPECT_EQ(book.getCategory(), "category_attribute_overrides_tags");
+  }
+}
+
+TEST(BookTest, setTagsDoesntAffectCategory)
+{
+    kiwix::Book book;
+
+    book.setTags("_category:youtube");
+    ASSERT_EQ("", book.getCategory());
+}
+
+TEST(BookTest, updateCopiesCategory)
+{
+    const XMLDoc xml(R"(<book id="abcd" category="ted"></book>)");
+
+    kiwix::Book book;
+    book.updateFromXml(xml.child("book"), "");
+
+    kiwix::Book newBook;
+    newBook.setId("abcd");
+    EXPECT_EQ(newBook.getCategory(), "");
+    newBook.update(book);
+    EXPECT_EQ(newBook.getCategory(), "ted");
+}

test/data/library.xml

@@ -0,0 +1,50 @@
+<library version="1.0">
+  <book
+        id="raycharles"
+        path="./zimfile.zim"
+        url="https://github.com/kiwix/libkiwix/raw/master/test/data/zimfile.zim"
+        title="Ray Charles"
+        description="Wikipedia articles about Ray Charles"
+        language="eng"
+        creator="Wikipedia"
+        publisher="Kiwix"
+        date="2020-03-31"
+        name="wikipedia_en_ray_charles"
+        tags="unittest;wikipedia;_category:wikipedia;_pictures:no;_videos:no;_details:no;_ftindex:yes"
+        articleCount="284"
+        mediaCount="2"
+        size="556"
+      ></book>
+  <book
+        id="raycharles_uncategorized"
+        path="./zimfile.zim"
+        url="https://github.com/kiwix/libkiwix/raw/master/test/data/zimfile.zim"
+        title="Ray (uncategorized) Charles"
+        description="No category is assigned to this library entry."
+        language="eng"
+        creator="Wikipedia"
+        publisher="Kiwix"
+        date="2020-03-31"
+        name="wikipedia_en_ray_charles"
+        tags="unittest;wikipedia;_pictures:no;_videos:no;_details:no"
+        articleCount="284"
+        mediaCount="2"
+        size="123"
+      ></book>
+  <book
+        id="charlesray"
+        path="./zimfile.zim"
+        url="https://github.com/kiwix/libkiwix/raw/master/test/data/zimfile.zim"
+        title="Charles, Ray"
+        description="Wikipedia articles about Ray Charles"
+        language="eng"
+        creator="Wikipedia"
+        publisher="Kiwix"
+        date="2020-03-31"
+        name="wikipedia_en_ray_charles"
+        tags="unittest;wikipedia;_category:jazz;_pictures:no;_videos:no;_details:no;_ftindex:yes"
+        articleCount="284"
+        mediaCount="2"
+        size="556"
+      ></book>
+</library>

test/httplib.h

@@ -2890,12 +2890,15 @@ inline ssize_t Stream::write(const std::string &s) {
 template <typename... Args>
 inline ssize_t Stream::write_format(const char *fmt, const Args &... args) {
   std::array<char, 2048> buf;
-
+#pragma GCC diagnostic push
+#pragma GCC diagnostic ignored "-Wunknown-warning-option"
+#pragma GCC diagnostic ignored "-Wmaybe-uninitialized"
 #if defined(_MSC_VER) && _MSC_VER < 1900
   auto sn = _snprintf_s(buf, bufsiz, buf.size() - 1, fmt, args...);
 #else
   auto sn = snprintf(buf.data(), buf.size() - 1, fmt, args...);
 #endif
+#pragma GCC diagnostic pop
   if (sn <= 0) { return sn; }
 
   auto n = static_cast<size_t>(sn);
@@ -3651,7 +3654,11 @@ Server::process_request(Stream &strm, bool last_connection,
                         const std::function<void(Request &)> &setup_request) {
   std::array<char, 2048> buf{};
 
-  detail::stream_line_reader line_reader(strm, buf.data(), buf.size());
+#pragma GCC diagnostic push
+#pragma GCC diagnostic ignored "-Wunknown-warning-option"
+#pragma GCC diagnostic ignored "-Wmaybe-uninitialized"
+   detail::stream_line_reader line_reader(strm, buf.data(), buf.size());
+#pragma GCC diagnostic pop
 
   // Connection has been closed on client
   if (!line_reader.getline()) { return false; }
@@ -3757,9 +3764,11 @@ inline socket_t Client::create_client_socket() const {
 
 inline bool Client::read_response_line(Stream &strm, Response &res) {
   std::array<char, 2048> buf;
-
+#pragma GCC diagnostic push
+#pragma GCC diagnostic ignored "-Wunknown-warning-option"
+#pragma GCC diagnostic ignored "-Wmaybe-uninitialized"
   detail::stream_line_reader line_reader(strm, buf.data(), buf.size());
-
+#pragma GCC diagnostic pop
   if (!line_reader.getline()) { return false; }
 
   const static std::regex re("(HTTP/1\\.[01]) (\\d+?) .*\r\n");

test/library.cpp

@@ -46,7 +46,7 @@ const char * sampleOpdsStream = R"(
     <updated>2018-06-23T00:00::00:Z</updated>
     <language>fra</language>
     <summary>Tania Louis videos</summary>
-    <tags>youtube</tags>
+    <tags>youtube;_category:category_defined_via_tags_only</tags>
     <link type="text/html" href="/biologie-tout-compris_fr_all_2018-06" />
     <author>
       <name>Tania Louis</name>
@@ -61,6 +61,7 @@ const char * sampleOpdsStream = R"(
     <updated>2019-06-05T00:00::00:Z</updated>
     <language>fra</language>
     <summary>Une page de Wikiquote, le recueil des citations libres.</summary>
+    <category>category_defined_via_category_element_only</category>
     <tags>wikiquote;nopic</tags>
     <link type="text/html" href="/wikiquote_fr_all_nopic_2019-06" />
     <author>
@@ -76,7 +77,8 @@ const char * sampleOpdsStream = R"(
     <updated>2019-06-02T00:00::00:Z</updated>
     <summary>Une sélection d'articles de Wikipédia sur la géographie</summary>
     <language>fra</language>
-    <tags>wikipedia;nopic</tags>
+    <category>category_element_overrides_tags</category>
+    <tags>wikipedia;nopic;_category:tags_override_category_element</tags>
     <link type="text/html" href="/wikipedia_fr_geography_nopic_2019-06" />
     <author>
       <name>Wikipedia</name>
@@ -91,7 +93,8 @@ const char * sampleOpdsStream = R"(
     <updated>2019-05-13T00:00::00:Z</updated>
     <language>fra</language>
     <summary>Une</summary>
-    <tags>wikipedia;nopic</tags>
+    <tags>wikipedia;nopic;_category:tags_override_category_element</tags>
+    <category>category_element_overrides_tags</category>
     <link type="text/html" href="/wikipedia_fr_mathematics_nopic_2019-05" />
     <author>
       <name>Wikipedia</name>
@@ -178,6 +181,42 @@ const char * sampleOpdsStream = R"(
 
 )";
 
+const char sampleLibraryXML[] = R"(
+<library version="1.0">
+  <book
+        id="raycharles"
+        path="./zimfile.zim"
+        url="https://github.com/kiwix/libkiwix/raw/master/test/data/zimfile.zim"
+        title="Ray Charles"
+        description="Wikipedia articles about Ray Charles"
+        language="eng"
+        creator="Wikipedia"
+        publisher="Kiwix"
+        date="2020-03-31"
+        name="wikipedia_en_ray_charles"
+        tags="wikipedia;_category:wikipedia;_pictures:no"
+        articleCount="284"
+        mediaCount="2"
+        size="556"
+      ></book>
+  <book
+        id="example"
+        path="./example.zim"
+        title="An example ZIM archive"
+        description="An eXaMpLe book added to the catalog via XML"
+        language="deu"
+        creator="Wikibooks"
+        publisher="Kiwix & Some Enthusiasts"
+        date="2021-04-11"
+        name="wikibooks_de"
+        tags="unittest;wikibooks;_category:wikibooks"
+        articleCount="12"
+        mediaCount="0"
+        size="126"
+      ></book>
+</library>
+)";
+
 #include "../include/library.h"
 #include "../include/manager.h"
 #include "../include/bookmark.h"
@@ -187,9 +226,13 @@ namespace
 
 class LibraryTest : public ::testing::Test {
  protected:
+  typedef kiwix::Library::BookIdCollection BookIdCollection;
+  typedef std::vector<std::string> TitleCollection;
+
   void SetUp() override {
      kiwix::Manager manager(&lib);
      manager.readOpds(sampleOpdsStream, "foo.urlHost");
+     manager.readXml(sampleLibraryXML, true, "./test/library.xml", true);
   }
 
     kiwix::Bookmark createBookmark(const std::string &id) {
@@ -198,6 +241,15 @@ class LibraryTest : public ::testing::Test {
         return bookmark;
     };
 
+  TitleCollection ids2Titles(const BookIdCollection& ids) {
+    TitleCollection titles;
+    for ( const auto& bookId : ids ) {
+      titles.push_back(lib.getBookById(bookId).getTitle());
+    }
+    std::sort(titles.begin(), titles.end());
+    return titles;
+  }
+
   kiwix::Library lib;
 };
 
@@ -222,41 +274,372 @@ TEST_F(LibraryTest, getBookMarksTest)
 
 TEST_F(LibraryTest, sanityCheck)
 {
-  EXPECT_EQ(lib.getBookCount(true, true), 10U);
-  EXPECT_EQ(lib.getBooksLanguages().size(), 2U);
-  EXPECT_EQ(lib.getBooksCreators().size(), 8U);
-  EXPECT_EQ(lib.getBooksPublishers().size(), 1U);
+  EXPECT_EQ(lib.getBookCount(true, true), 12U);
+  EXPECT_EQ(lib.getBooksLanguages().size(), 3U);
+  EXPECT_EQ(lib.getBooksCreators().size(), 9U);
+  EXPECT_EQ(lib.getBooksPublishers().size(), 3U);
 }
 
-TEST_F(LibraryTest, filterCheck)
+TEST_F(LibraryTest, categoryHandling)
 {
-  auto bookIds = lib.filter(kiwix::Filter());
+  EXPECT_EQ("", lib.getBookById("0c45160e-f917-760a-9159-dfe3c53cdcdd").getCategory());
+  EXPECT_EQ("category_defined_via_tags_only", lib.getBookById("0d0bcd57-d3f6-cb22-44cc-a723ccb4e1b2").getCategory());
+  EXPECT_EQ("category_defined_via_category_element_only", lib.getBookById("0ea1cde6-441d-6c58-f2c7-21c2838e659f").getCategory());
+  EXPECT_EQ("category_element_overrides_tags", lib.getBookById("1123e574-6eef-6d54-28fc-13e4caeae474").getCategory());
+  EXPECT_EQ("category_element_overrides_tags", lib.getBookById("14829621-c490-c376-0792-9de558b57efa").getCategory());
+}
+
+TEST_F(LibraryTest, emptyFilter)
+{
+  const auto bookIds = lib.filter(kiwix::Filter());
   EXPECT_EQ(bookIds, lib.getBooksIds());
+}
 
-  bookIds = lib.filter(kiwix::Filter().lang("eng"));
-  EXPECT_EQ(bookIds.size(), 5U);
+#define EXPECT_FILTER_RESULTS(f, ...)        \
+        EXPECT_EQ(                           \
+          ids2Titles(lib.filter(f)),         \
+          TitleCollection({ __VA_ARGS__ })   \
+        )
 
-  bookIds = lib.filter(kiwix::Filter().acceptTags({"stackexchange"}));
-  EXPECT_EQ(bookIds.size(), 3U);
+TEST_F(LibraryTest, filterLocal)
+{
+  EXPECT_FILTER_RESULTS(kiwix::Filter().local(true),
+    "An example ZIM archive",
+    "Ray Charles"
+  );
 
-  bookIds = lib.filter(kiwix::Filter().acceptTags({"wikipedia"}));
-  EXPECT_EQ(bookIds.size(), 3U);
+  EXPECT_FILTER_RESULTS(kiwix::Filter().local(false),
+    "Encyclopédie de la Tunisie",
+    "Granblue Fantasy Wiki",
+    "Géographie par Wikipédia",
+    "Islam Stack Exchange",
+    "Mathématiques",
+    "Movies & TV Stack Exchange",
+    "Mythology & Folklore Stack Exchange",
+    "TED talks - Business",
+    "Tania Louis",
+    "Wikiquote"
+  );
+}
 
-  bookIds = lib.filter(kiwix::Filter().acceptTags({"wikipedia", "nopic"}));
-  EXPECT_EQ(bookIds.size(), 2U);
+TEST_F(LibraryTest, filterRemote)
+{
+  EXPECT_FILTER_RESULTS(kiwix::Filter().remote(true),
+    "Encyclopédie de la Tunisie",
+    "Granblue Fantasy Wiki",
+    "Géographie par Wikipédia",
+    "Islam Stack Exchange",
+    "Mathématiques",
+    "Movies & TV Stack Exchange",
+    "Mythology & Folklore Stack Exchange",
+    "Ray Charles",
+    "TED talks - Business",
+    "Tania Louis",
+    "Wikiquote"
+  );
 
-  bookIds = lib.filter(kiwix::Filter().acceptTags({"wikipedia"}).rejectTags({"nopic"}));
-  EXPECT_EQ(bookIds.size(), 1U);
+  EXPECT_FILTER_RESULTS(kiwix::Filter().remote(false),
+    "An example ZIM archive"
+  );
+}
 
-  bookIds = lib.filter(kiwix::Filter().query("folklore"));
-  EXPECT_EQ(bookIds.size(), 1U);
+TEST_F(LibraryTest, filterByLanguage)
+{
+  EXPECT_FILTER_RESULTS(kiwix::Filter().lang("eng"),
+    "Granblue Fantasy Wiki",
+    "Islam Stack Exchange",
+    "Movies & TV Stack Exchange",
+    "Mythology & Folklore Stack Exchange",
+    "Ray Charles",
+    "TED talks - Business"
+  );
 
-  bookIds = lib.filter(kiwix::Filter().query("Wiki"));
-  EXPECT_EQ(bookIds.size(), 4U);
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("lang:eng"),
+    "Granblue Fantasy Wiki",
+    "Islam Stack Exchange",
+    "Movies & TV Stack Exchange",
+    "Mythology & Folklore Stack Exchange",
+    "Ray Charles",
+    "TED talks - Business"
+  );
 
-  bookIds = lib.filter(kiwix::Filter().query("Wiki").creator("Wiki"));
-  EXPECT_EQ(bookIds.size(), 1U);
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("eng"),
+    /* no results */
+  );
+}
 
+TEST_F(LibraryTest, filterByTags)
+{
+  EXPECT_FILTER_RESULTS(kiwix::Filter().acceptTags({"stackexchange"}),
+    "Islam Stack Exchange",
+    "Movies & TV Stack Exchange",
+    "Mythology & Folklore Stack Exchange"
+  );
+
+  // filtering by tags is case and diacritics insensitive
+  EXPECT_FILTER_RESULTS(kiwix::Filter().acceptTags({"ståckEXÇhange"}),
+    "Islam Stack Exchange",
+    "Movies & TV Stack Exchange",
+    "Mythology & Folklore Stack Exchange"
+  );
+
+  // filtering by tags requires full match of the search term
+  EXPECT_FILTER_RESULTS(kiwix::Filter().acceptTags({"stackexch"}),
+    /* no results */
+  );
+
+  // in tags with values (tag:value form) the value is an inseparable
+  // part of the tag
+  EXPECT_FILTER_RESULTS(kiwix::Filter().acceptTags({"_category"}),
+    /* no results */
+  );
+  EXPECT_FILTER_RESULTS(kiwix::Filter().acceptTags({"_category:category_defined_via_tags_only"}),
+    "Tania Louis"
+  );
+
+  EXPECT_FILTER_RESULTS(kiwix::Filter().acceptTags({"wikipedia"}),
+    "Encyclopédie de la Tunisie",
+    "Géographie par Wikipédia",
+    "Mathématiques",
+    "Ray Charles"
+  );
+
+  EXPECT_FILTER_RESULTS(kiwix::Filter().acceptTags({"wikipedia", "nopic"}),
+    "Géographie par Wikipédia",
+    "Mathématiques"
+  );
+
+  EXPECT_FILTER_RESULTS(kiwix::Filter().acceptTags({"wikipedia"}).rejectTags({"nopic"}),
+    "Encyclopédie de la Tunisie",
+    "Ray Charles"
+  );
+}
+
+
+TEST_F(LibraryTest, filterByQuery)
+{
+  // filtering by query checks the title
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("Exchange"),
+    "Islam Stack Exchange",
+    "Movies & TV Stack Exchange",
+    "Mythology & Folklore Stack Exchange"
+  );
+
+  // filtering by query checks the description/summary
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("enthusiasts"),
+    "Movies & TV Stack Exchange",
+    "Mythology & Folklore Stack Exchange"
+  );
+
+  // filtering by query is case insensitive on titles
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("ExcHANge"),
+    "Islam Stack Exchange",
+    "Movies & TV Stack Exchange",
+    "Mythology & Folklore Stack Exchange"
+  );
+
+  // filtering by query is diacritics insensitive on titles
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("mathematiques"),
+    "Mathématiques",
+  );
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("èxchângé"),
+    "Islam Stack Exchange",
+    "Movies & TV Stack Exchange",
+    "Mythology & Folklore Stack Exchange"
+  );
+
+  // filtering by query is case insensitive on description/summary
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("enTHUSiaSTS"),
+    "Movies & TV Stack Exchange",
+    "Mythology & Folklore Stack Exchange"
+  );
+
+  // filtering by query is diacritics insensitive on description/summary
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("selection"),
+    "Géographie par Wikipédia"
+  );
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("enthúsïåsts"),
+    "Movies & TV Stack Exchange",
+    "Mythology & Folklore Stack Exchange"
+  );
+
+  // by default, filtering by query assumes partial query
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("Wiki"),
+    "Encyclopédie de la Tunisie",
+    "Granblue Fantasy Wiki",
+    "Géographie par Wikipédia",
+    "Ray Charles",
+    "Wikiquote"
+  );
+
+  // partial query can be disabled
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("Wiki", false),
+    "Granblue Fantasy Wiki"
+  );
+}
+
+
+TEST_F(LibraryTest, filteringByEmptyQueryReturnsAllEntries)
+{
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query(""),
+    "An example ZIM archive",
+    "Encyclopédie de la Tunisie",
+    "Granblue Fantasy Wiki",
+    "Géographie par Wikipédia",
+    "Islam Stack Exchange",
+    "Mathématiques",
+    "Movies & TV Stack Exchange",
+    "Mythology & Folklore Stack Exchange",
+    "Ray Charles",
+    "TED talks - Business",
+    "Tania Louis",
+    "Wikiquote"
+  );
+}
+
+TEST_F(LibraryTest, filterByCreator)
+{
+  EXPECT_FILTER_RESULTS(kiwix::Filter().creator("Wikipedia"),
+    "Encyclopédie de la Tunisie",
+    "Géographie par Wikipédia",
+    "Mathématiques",
+    "Ray Charles"
+  );
+
+  // filtering by creator requires full match of the search term
+  EXPECT_FILTER_RESULTS(kiwix::Filter().creator("Wiki"),
+    "Granblue Fantasy Wiki"
+  );
+
+  // filtering by creator is case and diacritics insensitive
+  EXPECT_FILTER_RESULTS(kiwix::Filter().creator("wIkï"),
+    "Granblue Fantasy Wiki"
+  );
+
+  // filtering by creator doesn't requires full match of the full creator name
+  EXPECT_FILTER_RESULTS(kiwix::Filter().creator("Stack"),
+    "Islam Stack Exchange",
+    "Movies & TV Stack Exchange",
+    "Mythology & Folklore Stack Exchange"
+  );
+
+  // filtering by creator requires a full phrase match (ignoring some non-word terms)
+  EXPECT_FILTER_RESULTS(kiwix::Filter().creator("Movies & TV Stack Exchange"),
+    "Movies & TV Stack Exchange"
+  );
+  EXPECT_FILTER_RESULTS(kiwix::Filter().creator("Movies & TV"),
+    "Movies & TV Stack Exchange"
+  );
+  EXPECT_FILTER_RESULTS(kiwix::Filter().creator("Movies TV"),
+    "Movies & TV Stack Exchange"
+  );
+  EXPECT_FILTER_RESULTS(kiwix::Filter().creator("TV & Movies"),
+    /* no results */
+  );
+  EXPECT_FILTER_RESULTS(kiwix::Filter().creator("TV Movies"),
+    /* no results */
+  );
+
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("creator:Wikipedia"),
+    "Encyclopédie de la Tunisie",
+    "Géographie par Wikipédia",
+    "Mathématiques",
+    "Ray Charles"
+  );
+
+}
+
+TEST_F(LibraryTest, filterByPublisher)
+{
+  EXPECT_FILTER_RESULTS(kiwix::Filter().publisher("Kiwix"),
+    "An example ZIM archive",
+    "Ray Charles"
+  );
+
+  // filtering by publisher requires full match of the search term
+  EXPECT_FILTER_RESULTS(kiwix::Filter().publisher("Kiwi"),
+    /* no results */
+  );
+
+  // filtering by publisher requires a full phrase match
+  EXPECT_FILTER_RESULTS(kiwix::Filter().publisher("Kiwix & Some Enthusiasts"),
+    "An example ZIM archive"
+  );
+  EXPECT_FILTER_RESULTS(kiwix::Filter().publisher("Some Enthusiasts & Kiwix"),
+    /* no results */
+  );
+
+  // filtering by publisher is case and diacritics insensitive
+  EXPECT_FILTER_RESULTS(kiwix::Filter().publisher("kîWIx"),
+    "An example ZIM archive",
+    "Ray Charles"
+  );
+
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("publisher:kiwix"),
+    "An example ZIM archive",
+    "Ray Charles"
+  );
+
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("kiwix"),
+    /* no results */
+  );
+}
+
+TEST_F(LibraryTest, filterByName)
+{
+  EXPECT_FILTER_RESULTS(kiwix::Filter().name("wikibooks_de"),
+    "An example ZIM archive"
+  );
+
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("name:wikibooks_de"),
+    "An example ZIM archive"
+  );
+
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("wikibooks_de"),
+    /* no results */
+  );
+}
+
+TEST_F(LibraryTest, filterByCategory)
+{
+  EXPECT_FILTER_RESULTS(kiwix::Filter().category("category_element_overrides_tags"),
+    "Géographie par Wikipédia",
+    "Mathématiques"
+  );
+
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("category:category_element_overrides_tags"),
+    "Géographie par Wikipédia",
+    "Mathématiques"
+  );
+
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("category_element_overrides_tags"),
+    /* no results */
+  );
+}
+
+TEST_F(LibraryTest, filterByMaxSize)
+{
+  EXPECT_FILTER_RESULTS(kiwix::Filter().maxSize(200000),
+    "An example ZIM archive"
+  );
+}
+
+TEST_F(LibraryTest, filterByMultipleCriteria)
+{
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("Wiki").creator("Wikipedia"),
+    "Encyclopédie de la Tunisie",
+    "Géographie par Wikipédia",
+    "Ray Charles"
+  );
+
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("Wiki").creator("Wikipedia").maxSize(100000000UL),
+    "Encyclopédie de la Tunisie",
+    "Ray Charles"
+  );
+
+  EXPECT_FILTER_RESULTS(kiwix::Filter().query("Wiki").creator("Wikipedia").maxSize(100000000UL).local(false),
+    "Encyclopédie de la Tunisie"
+  );
 }
 
 TEST_F(LibraryTest, getBookByPath)
@@ -271,4 +654,40 @@ TEST_F(LibraryTest, getBookByPath)
   EXPECT_EQ(lib.getBookByPath(path).getId(), book.getId());
   EXPECT_THROW(lib.getBookByPath("non/existant/path.zim"), std::out_of_range);
 }
+
+TEST_F(LibraryTest, removeBookByIdRemovesTheBook)
+{
+  const auto initialBookCount = lib.getBookCount(true, true);
+  ASSERT_GT(initialBookCount, 0U);
+  EXPECT_NO_THROW(lib.getBookById("raycharles"));
+  lib.removeBookById("raycharles");
+  EXPECT_EQ(initialBookCount - 1, lib.getBookCount(true, true));
+  EXPECT_THROW(lib.getBookById("raycharles"), std::out_of_range);
+};
+
+TEST_F(LibraryTest, removeBookByIdDropsTheReader)
+{
+  EXPECT_NE(nullptr, lib.getReaderById("raycharles"));
+  lib.removeBookById("raycharles");
+  EXPECT_THROW(lib.getReaderById("raycharles"), std::out_of_range);
+};
+
+TEST_F(LibraryTest, removeBookByIdUpdatesTheSearchDB)
+{
+  kiwix::Filter f;
+  f.local(true).valid(true).query(R"(title:"ray charles")", false);
+
+  EXPECT_NO_THROW(lib.getBookById("raycharles"));
+  EXPECT_EQ(1U, lib.filter(f).size());
+
+  lib.removeBookById("raycharles");
+
+  EXPECT_THROW(lib.getBookById("raycharles"), std::out_of_range);
+  EXPECT_EQ(0U, lib.filter(f).size());
+
+  // make sure that Library::filter() doesn't add an empty book with
+  // an id surviving in the search DB
+  EXPECT_THROW(lib.getBookById("raycharles"), std::out_of_range);
+};
+
 };

test/manager.cpp

@@ -14,7 +14,7 @@ TEST(ManagerTest, addBookFromPathAndGetIdTest)
     ASSERT_NE(bookId, "");
     kiwix::Book book = lib.getBookById(bookId);
     EXPECT_EQ(book.getPath(), computeAbsolutePath("", "./test/example.zim"));
-    
+
     const std::string pathToSave = "./pathToSave";
     const std::string url = "url";
     bookId = manager.addBookFromPathAndGetId("./test/example.zim", pathToSave, url, true);
@@ -23,3 +23,46 @@ TEST(ManagerTest, addBookFromPathAndGetIdTest)
     EXPECT_EQ(book.getPath(), savedPath);
     EXPECT_EQ(book.getUrl(), url);
 }
+
+const char sampleLibraryXML[] = R"(
+<library version="1.0">
+  <book
+        id="0d0bcd57-d3f6-cb22-44cc-a723ccb4e1b2"
+        path="zimfiles/unittest.zim"
+        url="https://example.com/zimfiles/unittest.zim"
+        title="Unit Test"
+        description="Wikipedia articles about unit testing"
+        language="eng"
+        creator="Wikipedia"
+        publisher="Kiwix"
+        date="2020-03-31"
+        name="wikipedia_en_unit_testing"
+        tags="unittest;wikipedia"
+        articleCount="123"
+        mediaCount="45"
+        size="678"
+      ></book>
+</library>
+)";
+
+TEST(ManagerTest, readXml)
+{
+    kiwix::Library lib;
+    kiwix::Manager manager = kiwix::Manager(&lib);
+
+    EXPECT_EQ(true, manager.readXml(sampleLibraryXML, true, "/data/lib.xml", true));
+    kiwix::Book book = lib.getBookById("0d0bcd57-d3f6-cb22-44cc-a723ccb4e1b2");
+    EXPECT_EQ("/data/zimfiles/unittest.zim", book.getPath());
+    EXPECT_EQ("https://example.com/zimfiles/unittest.zim", book.getUrl());
+    EXPECT_EQ("Unit Test", book.getTitle());
+    EXPECT_EQ("Wikipedia articles about unit testing", book.getDescription());
+    EXPECT_EQ("eng", book.getLanguage());
+    EXPECT_EQ("Wikipedia", book.getCreator());
+    EXPECT_EQ("Kiwix", book.getPublisher());
+    EXPECT_EQ("2020-03-31", book.getDate());
+    EXPECT_EQ("wikipedia_en_unit_testing", book.getName());
+    EXPECT_EQ("unittest;wikipedia", book.getTags());
+    EXPECT_EQ(123U, book.getArticleCount());
+    EXPECT_EQ(45U, book.getMediaCount());
+    EXPECT_EQ(678U*1024, book.getSize());
+}

test/meson.build

@@ -25,7 +25,8 @@ if gtest_dep.found() and not meson.is_cross_build()
     data_files = [
       'example.zim',
       'zimfile.zim',
-      'corner_cases.zim'
+      'corner_cases.zim',
+      'library.xml'
     ]
     foreach file : data_files
         # configure_file(input : 'data/' + file,

test/server.cpp

@@ -53,6 +53,7 @@ public: // types
   typedef std::vector<std::string> FilePathCollection;
 
 public: // functions
+  ZimFileServer(int serverPort, std::string libraryFilePath);
   ZimFileServer(int serverPort, const FilePathCollection& zimpaths);
   ~ZimFileServer();
 
@@ -66,6 +67,9 @@ public: // functions
     return client->Head(path, headers);
   }
 
+private:
+  void run(int serverPort);
+
 private: // data
   kiwix::Library library;
   kiwix::Manager manager;
@@ -74,6 +78,16 @@ private: // data
   std::unique_ptr<httplib::Client> client;
 };
 
+ZimFileServer::ZimFileServer(int serverPort, std::string libraryFilePath)
+: manager(&this->library)
+{
+  if ( isRelativePath(libraryFilePath) )
+    libraryFilePath = computeAbsolutePath(getCurrentDirectory(), libraryFilePath);
+  manager.readFile(libraryFilePath, true, true);
+
+  run(serverPort);
+}
+
 ZimFileServer::ZimFileServer(int serverPort, const FilePathCollection& zimpaths)
 : manager(&this->library)
 {
@@ -82,6 +96,11 @@ ZimFileServer::ZimFileServer(int serverPort, const FilePathCollection& zimpaths)
       throw std::runtime_error("Unable to add the ZIM file '" + zimpath + "'");
   }
 
+  run(serverPort);
+}
+
+void ZimFileServer::run(int serverPort)
+{
   const std::string address = "127.0.0.1";
   nameMapper.reset(new kiwix::HumanReadableNameMapper(library, false));
   server.reset(new kiwix::Server(&library, nameMapper.get()));
@@ -197,15 +216,6 @@ TEST_F(ServerTest, 200)
     EXPECT_EQ(200, zfs1_->GET(res.url)->status) << "res.url: " << res.url;
 }
 
-// seperate test for 204 code
-
-TEST_F(ServerTest, EmptySearchReturnsA204StatusCode)
-{
-  const char* url="/search?content=zimfile&pattern=abcd";
-  auto res=zfs1_->GET(url);
-  EXPECT_EQ(204, res->status) << "res.url: " << url;
-}
-
 TEST_F(ServerTest, CompressibleContentIsCompressedIfAcceptable)
 {
   for ( const Resource& res : resources200Compressible ) {
@@ -230,6 +240,7 @@ const char* urls404[] = {
   "/skin/non-existent-skin-resource",
   "/catalog",
   "/catalog/non-existent-item",
+  "/catalogBLABLABLA/root.xml",
   "/meta",
   "/meta?content=zimfile",
   "/meta?content=zimfile&name=non-existent-item",
@@ -250,14 +261,6 @@ TEST_F(ServerTest, 404)
     EXPECT_EQ(404, zfs1_->GET(url)->status) << "url: " << url;
 }
 
-TEST_F(ServerTest, SuccessfulSearchForAnArticleTitleRedirectsToTheArticle)
-{
-  auto g = zfs1_->GET("/search?content=zimfile&pattern=ray%20charles" );
-  ASSERT_EQ(302, g->status);
-  ASSERT_TRUE(g->has_header("Location"));
-  ASSERT_EQ("/zimfile/A/Ray_Charles", g->get_header_value("Location"));
-}
-
 TEST_F(ServerTest, RandomPageRedirectsToAnExistingArticle)
 {
   auto g = zfs1_->GET("/random?content=zimfile");
@@ -546,3 +549,560 @@ TEST_F(ServerTest, RangeHeaderIsCaseInsensitive)
     EXPECT_EQ(r0->body, r->body);
   }
 }
+
+////////////////////////////////////////////////////////////////////////////////
+// Testing of the library-related functionality of the server
+////////////////////////////////////////////////////////////////////////////////
+
+class LibraryServerTest : public ::testing::Test
+{
+protected:
+  std::unique_ptr<ZimFileServer>   zfs1_;
+
+  const int PORT = 8002;
+
+protected:
+  void SetUp() override {
+    zfs1_.reset(new ZimFileServer(PORT, "./test/library.xml"));
+  }
+
+  void TearDown() override {
+    zfs1_.reset();
+  }
+};
+
+// Returns a copy of 'text' where every line that fully matches 'pattern'
+// preceded by optional whitespace is replaced with the fixed string
+// 'replacement' preserving the leading whitespace
+std::string replaceLines(const std::string& text,
+                         const std::string& pattern,
+                         const std::string& replacement)
+{
+  std::regex regex("^ *" + pattern + "$");
+  std::ostringstream oss;
+  std::istringstream iss(text);
+  std::string line;
+  while ( std::getline(iss, line) ) {
+    if ( std::regex_match(line, regex) ) {
+      for ( size_t i = 0; i < line.size() && line[i] == ' '; ++i )
+        oss << ' ';
+      oss << replacement << "\n";
+    } else {
+      oss << line << "\n";
+    }
+  }
+  return oss.str();
+}
+
+std::string maskVariableOPDSFeedData(std::string s)
+{
+  s = replaceLines(s, R"(<updated>\d\d\d\d-\d\d-\d\dT\d\d:\d\d:\d\dZ</updated>)",
+                      "<updated>YYYY-MM-DDThh:mm:ssZ</updated>");
+  s = replaceLines(s, "<id>[[:xdigit:]]{8}-[[:xdigit:]]{4}-[[:xdigit:]]{4}-[[:xdigit:]]{4}-[[:xdigit:]]{12}</id>",
+                      "<id>12345678-90ab-cdef-1234-567890abcdef</id>");
+  return s;
+}
+
+#define OPDS_FEED_TAG \
+    "<feed xmlns=\"http://www.w3.org/2005/Atom\"" \
+         " xmlns:opds=\"http://opds-spec.org/2010/catalog\">\n"
+
+#define CATALOG_LINK_TAGS \
+    "  <link rel=\"self\" href=\"\" type=\"application/atom+xml\" />\n" \
+    "  <link rel=\"search\""                                            \
+           " type=\"application/opensearchdescription+xml\""            \
+           " href=\"/catalog/searchdescription.xml\" />\n"
+
+#define CHARLES_RAY_CATALOG_ENTRY \
+    "  <entry>\n"                                                       \
+    "    <id>urn:uuid:charlesray</id>\n"                                \
+    "    <title>Charles, Ray</title>\n"                                 \
+    "    <summary>Wikipedia articles about Ray Charles</summary>\n"    \
+    "    <language>eng</language>\n"                                    \
+    "    <updated>YYYY-MM-DDThh:mm:ssZ</updated>\n"                     \
+    "    <name>wikipedia_en_ray_charles</name>\n"                       \
+    "    <flavour></flavour>\n"                                         \
+    "    <category>jazz</category>\n"                                   \
+    "    <tags>unittest;wikipedia;_category:jazz;_pictures:no;_videos:no;_details:no;_ftindex:yes</tags>\n" \
+    "    <articleCount>284</articleCount>\n"                            \
+    "    <mediaCount>2</mediaCount>\n"                                  \
+    "    <icon>/meta?name=favicon&amp;content=zimfile</icon>\n"         \
+    "    <link type=\"text/html\" href=\"/zimfile\" />\n"               \
+    "    <author>\n"                                                    \
+    "      <name>Wikipedia</name>\n"                                    \
+    "    </author>\n"                                                   \
+    "    <publisher>\n"                                                 \
+    "      <name>Kiwix</name>\n"                                        \
+    "    </publisher>\n"                                                \
+    "    <link rel=\"http://opds-spec.org/acquisition/open-access\" type=\"application/x-zim\" href=\"https://github.com/kiwix/libkiwix/raw/master/test/data/zimfile.zim\" length=\"569344\" />\n" \
+    "  </entry>\n"
+
+#define RAY_CHARLES_CATALOG_ENTRY \
+    "  <entry>\n"                                                       \
+    "    <id>urn:uuid:raycharles</id>\n"                                \
+    "    <title>Ray Charles</title>\n"                                  \
+    "    <summary>Wikipedia articles about Ray Charles</summary>\n"     \
+    "    <language>eng</language>\n"                                    \
+    "    <updated>YYYY-MM-DDThh:mm:ssZ</updated>\n"                     \
+    "    <name>wikipedia_en_ray_charles</name>\n"                       \
+    "    <flavour></flavour>\n"                                         \
+    "    <category>wikipedia</category>\n"                              \
+    "    <tags>unittest;wikipedia;_category:wikipedia;_pictures:no;_videos:no;_details:no;_ftindex:yes</tags>\n" \
+    "    <articleCount>284</articleCount>\n"                            \
+    "    <mediaCount>2</mediaCount>\n"                                  \
+    "    <icon>/meta?name=favicon&amp;content=zimfile</icon>\n"         \
+    "    <link type=\"text/html\" href=\"/zimfile\" />\n"               \
+    "    <author>\n"                                                    \
+    "      <name>Wikipedia</name>\n"                                    \
+    "    </author>\n"                                                   \
+    "    <publisher>\n"                                                 \
+    "      <name>Kiwix</name>\n"                                        \
+    "    </publisher>\n"                                                \
+    "    <link rel=\"http://opds-spec.org/acquisition/open-access\" type=\"application/x-zim\" href=\"https://github.com/kiwix/libkiwix/raw/master/test/data/zimfile.zim\" length=\"569344\" />\n" \
+    "  </entry>\n"
+
+#define UNCATEGORIZED_RAY_CHARLES_CATALOG_ENTRY \
+    "  <entry>\n"                                                       \
+    "    <id>urn:uuid:raycharles_uncategorized</id>\n"                  \
+    "    <title>Ray (uncategorized) Charles</title>\n"                  \
+    "    <summary>No category is assigned to this library entry.</summary>\n" \
+    "    <language>eng</language>\n"                                    \
+    "    <updated>YYYY-MM-DDThh:mm:ssZ</updated>\n"                     \
+    "    <name>wikipedia_en_ray_charles</name>\n"                       \
+    "    <flavour></flavour>\n"                                         \
+    "    <category></category>\n"                                \
+    "    <tags>unittest;wikipedia;_pictures:no;_videos:no;_details:no</tags>\n" \
+    "    <articleCount>284</articleCount>\n"                            \
+    "    <mediaCount>2</mediaCount>\n"                                  \
+    "    <icon>/meta?name=favicon&amp;content=zimfile</icon>\n"         \
+    "    <link type=\"text/html\" href=\"/zimfile\" />\n"               \
+    "    <author>\n"                                                    \
+    "      <name>Wikipedia</name>\n"                                    \
+    "    </author>\n"                                                   \
+    "    <publisher>\n"                                                 \
+    "      <name>Kiwix</name>\n"                                        \
+    "    </publisher>\n"                                                \
+    "    <link rel=\"http://opds-spec.org/acquisition/open-access\" type=\"application/x-zim\" href=\"https://github.com/kiwix/libkiwix/raw/master/test/data/zimfile.zim\" length=\"125952\" />\n" \
+    "  </entry>\n"
+
+TEST_F(LibraryServerTest, catalog_root_xml)
+{
+  const auto r = zfs1_->GET("/catalog/root.xml");
+  EXPECT_EQ(r->status, 200);
+  EXPECT_EQ(maskVariableOPDSFeedData(r->body),
+    OPDS_FEED_TAG
+    "  <id>12345678-90ab-cdef-1234-567890abcdef</id>\n"
+    "  <title>All zims</title>\n"
+    "  <updated>YYYY-MM-DDThh:mm:ssZ</updated>\n"
+    "\n"
+    CATALOG_LINK_TAGS
+    CHARLES_RAY_CATALOG_ENTRY
+    RAY_CHARLES_CATALOG_ENTRY
+    UNCATEGORIZED_RAY_CHARLES_CATALOG_ENTRY
+    "</feed>\n"
+  );
+}
+
+TEST_F(LibraryServerTest, catalog_searchdescription_xml)
+{
+  const auto r = zfs1_->GET("/catalog/searchdescription.xml");
+  EXPECT_EQ(r->status, 200);
+  EXPECT_EQ(r->body,
+    "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n"
+    "<OpenSearchDescription xmlns=\"http://a9.com/-/spec/opensearch/1.1/\">\n"
+    "  <ShortName>Zim catalog search</ShortName>\n"
+    "  <Description>Search zim files in the catalog.</Description>\n"
+    "  <Url type=\"application/atom+xml;profile=opds-catalog\"\n"
+    "       xmlns:atom=\"http://www.w3.org/2005/Atom\"\n"
+    "       xmlns:k=\"http://kiwix.org/opensearchextension/1.0\"\n"
+    "       indexOffset=\"0\"\n"
+    "       template=\"/catalog/search?q={searchTerms?}&lang={language?}&name={k:name?}&tag={k:tag?}&notag={k:notag?}&maxsize={k:maxsize?}&count={count?}&start={startIndex?}\"/>\n"
+    "</OpenSearchDescription>\n"
+  );
+}
+
+TEST_F(LibraryServerTest, catalog_search_by_phrase)
+{
+  const auto r = zfs1_->GET("/catalog/search?q=\"ray%20charles\"");
+  EXPECT_EQ(r->status, 200);
+  EXPECT_EQ(maskVariableOPDSFeedData(r->body),
+    OPDS_FEED_TAG
+    "  <id>12345678-90ab-cdef-1234-567890abcdef</id>\n"
+    "  <title>Filtered zims (q=&quot;ray charles&quot;)</title>\n"
+    "  <updated>YYYY-MM-DDThh:mm:ssZ</updated>\n"
+    "  <totalResults>2</totalResults>\n"
+    "  <startIndex>0</startIndex>\n"
+    "  <itemsPerPage>2</itemsPerPage>\n"
+    CATALOG_LINK_TAGS
+    RAY_CHARLES_CATALOG_ENTRY
+    CHARLES_RAY_CATALOG_ENTRY
+    "</feed>\n"
+  );
+}
+
+TEST_F(LibraryServerTest, catalog_search_by_words)
+{
+  const auto r = zfs1_->GET("/catalog/search?q=ray%20charles");
+  EXPECT_EQ(r->status, 200);
+  EXPECT_EQ(maskVariableOPDSFeedData(r->body),
+    OPDS_FEED_TAG
+    "  <id>12345678-90ab-cdef-1234-567890abcdef</id>\n"
+    "  <title>Filtered zims (q=ray charles)</title>\n"
+    "  <updated>YYYY-MM-DDThh:mm:ssZ</updated>\n"
+    "  <totalResults>3</totalResults>\n"
+    "  <startIndex>0</startIndex>\n"
+    "  <itemsPerPage>3</itemsPerPage>\n"
+    CATALOG_LINK_TAGS
+    RAY_CHARLES_CATALOG_ENTRY
+    CHARLES_RAY_CATALOG_ENTRY
+    UNCATEGORIZED_RAY_CHARLES_CATALOG_ENTRY
+    "</feed>\n"
+  );
+}
+
+TEST_F(LibraryServerTest, catalog_prefix_search)
+{
+  {
+    const auto r = zfs1_->GET("/catalog/search?q=description:ray%20description:charles");
+    EXPECT_EQ(r->status, 200);
+    EXPECT_EQ(maskVariableOPDSFeedData(r->body),
+      OPDS_FEED_TAG
+      "  <id>12345678-90ab-cdef-1234-567890abcdef</id>\n"
+      "  <title>Filtered zims (q=description:ray description:charles)</title>\n"
+      "  <updated>YYYY-MM-DDThh:mm:ssZ</updated>\n"
+      "  <totalResults>2</totalResults>\n"
+      "  <startIndex>0</startIndex>\n"
+      "  <itemsPerPage>2</itemsPerPage>\n"
+      CATALOG_LINK_TAGS
+      RAY_CHARLES_CATALOG_ENTRY
+      CHARLES_RAY_CATALOG_ENTRY
+      "</feed>\n"
+    );
+  }
+  {
+    const auto r = zfs1_->GET("/catalog/search?q=title:\"ray%20charles\"");
+    EXPECT_EQ(r->status, 200);
+    EXPECT_EQ(maskVariableOPDSFeedData(r->body),
+      OPDS_FEED_TAG
+      "  <id>12345678-90ab-cdef-1234-567890abcdef</id>\n"
+      "  <title>Filtered zims (q=title:&quot;ray charles&quot;)</title>\n"
+      "  <updated>YYYY-MM-DDThh:mm:ssZ</updated>\n"
+      "  <totalResults>1</totalResults>\n"
+      "  <startIndex>0</startIndex>\n"
+      "  <itemsPerPage>1</itemsPerPage>\n"
+      CATALOG_LINK_TAGS
+      RAY_CHARLES_CATALOG_ENTRY
+      "</feed>\n"
+    );
+  }
+}
+
+TEST_F(LibraryServerTest, catalog_search_with_word_exclusion)
+{
+  const auto r = zfs1_->GET("/catalog/search?q=ray%20-uncategorized");
+  EXPECT_EQ(r->status, 200);
+  EXPECT_EQ(maskVariableOPDSFeedData(r->body),
+    OPDS_FEED_TAG
+    "  <id>12345678-90ab-cdef-1234-567890abcdef</id>\n"
+    "  <title>Filtered zims (q=ray -uncategorized)</title>\n"
+    "  <updated>YYYY-MM-DDThh:mm:ssZ</updated>\n"
+    "  <totalResults>2</totalResults>\n"
+    "  <startIndex>0</startIndex>\n"
+    "  <itemsPerPage>2</itemsPerPage>\n"
+    CATALOG_LINK_TAGS
+    RAY_CHARLES_CATALOG_ENTRY
+    CHARLES_RAY_CATALOG_ENTRY
+    "</feed>\n"
+  );
+}
+
+TEST_F(LibraryServerTest, catalog_search_by_tag)
+{
+  const auto r = zfs1_->GET("/catalog/search?tag=_category:jazz");
+  EXPECT_EQ(r->status, 200);
+  EXPECT_EQ(maskVariableOPDSFeedData(r->body),
+    OPDS_FEED_TAG
+    "  <id>12345678-90ab-cdef-1234-567890abcdef</id>\n"
+    "  <title>Filtered zims (tag=_category:jazz)</title>\n"
+    "  <updated>YYYY-MM-DDThh:mm:ssZ</updated>\n"
+    "  <totalResults>1</totalResults>\n"
+    "  <startIndex>0</startIndex>\n"
+    "  <itemsPerPage>1</itemsPerPage>\n"
+    CATALOG_LINK_TAGS
+    CHARLES_RAY_CATALOG_ENTRY
+    "</feed>\n"
+  );
+}
+
+TEST_F(LibraryServerTest, catalog_search_by_category)
+{
+  const auto r = zfs1_->GET("/catalog/search?category=jazz");
+  EXPECT_EQ(r->status, 200);
+  EXPECT_EQ(maskVariableOPDSFeedData(r->body),
+    OPDS_FEED_TAG
+    "  <id>12345678-90ab-cdef-1234-567890abcdef</id>\n"
+    "  <title>Filtered zims (category=jazz)</title>\n"
+    "  <updated>YYYY-MM-DDThh:mm:ssZ</updated>\n"
+    "  <totalResults>1</totalResults>\n"
+    "  <startIndex>0</startIndex>\n"
+    "  <itemsPerPage>1</itemsPerPage>\n"
+    CATALOG_LINK_TAGS
+    CHARLES_RAY_CATALOG_ENTRY
+    "</feed>\n"
+  );
+}
+
+TEST_F(LibraryServerTest, catalog_search_results_pagination)
+{
+  {
+    const auto r = zfs1_->GET("/catalog/search?count=1");
+    EXPECT_EQ(r->status, 200);
+    EXPECT_EQ(maskVariableOPDSFeedData(r->body),
+      OPDS_FEED_TAG
+      "  <id>12345678-90ab-cdef-1234-567890abcdef</id>\n"
+      "  <title>Filtered zims (count=1)</title>\n"
+      "  <updated>YYYY-MM-DDThh:mm:ssZ</updated>\n"
+      "  <totalResults>3</totalResults>\n"
+      "  <startIndex>0</startIndex>\n"
+      "  <itemsPerPage>1</itemsPerPage>\n"
+      CATALOG_LINK_TAGS
+      CHARLES_RAY_CATALOG_ENTRY
+      "</feed>\n"
+    );
+  }
+  {
+    const auto r = zfs1_->GET("/catalog/search?start=1&count=1");
+    EXPECT_EQ(r->status, 200);
+    EXPECT_EQ(maskVariableOPDSFeedData(r->body),
+      OPDS_FEED_TAG
+      "  <id>12345678-90ab-cdef-1234-567890abcdef</id>\n"
+      "  <title>Filtered zims (count=1&amp;start=1)</title>\n"
+      "  <updated>YYYY-MM-DDThh:mm:ssZ</updated>\n"
+      "  <totalResults>3</totalResults>\n"
+      "  <startIndex>1</startIndex>\n"
+      "  <itemsPerPage>1</itemsPerPage>\n"
+      CATALOG_LINK_TAGS
+      RAY_CHARLES_CATALOG_ENTRY
+      "</feed>\n"
+    );
+  }
+  {
+    const auto r = zfs1_->GET("/catalog/search?start=100&count=10");
+    EXPECT_EQ(r->status, 200);
+    EXPECT_EQ(maskVariableOPDSFeedData(r->body),
+      OPDS_FEED_TAG
+      "  <id>12345678-90ab-cdef-1234-567890abcdef</id>\n"
+      "  <title>Filtered zims (count=10&amp;start=100)</title>\n"
+      "  <updated>YYYY-MM-DDThh:mm:ssZ</updated>\n"
+      "  <totalResults>3</totalResults>\n"
+      "  <startIndex>100</startIndex>\n"
+      "  <itemsPerPage>0</itemsPerPage>\n"
+      CATALOG_LINK_TAGS
+      "  \n"
+      "</feed>\n"
+    );
+  }
+}
+
+TEST_F(LibraryServerTest, catalog_v2_root)
+{
+  const auto r = zfs1_->GET("/catalog/v2/root.xml");
+  EXPECT_EQ(r->status, 200);
+  const char expected_output[] = R"(<?xml version="1.0" encoding="UTF-8"?>
+<feed xmlns="http://www.w3.org/2005/Atom"
+      xmlns:opds="https://specs.opds.io/opds-1.2">
+  <id>12345678-90ab-cdef-1234-567890abcdef</id>
+  <link rel="self"
+        href="/catalog/v2/root.xml"
+        type="application/atom+xml;profile=opds-catalog;kind=navigation"/>
+  <link rel="start"
+        href="/catalog/v2/root.xml"
+        type="application/atom+xml;profile=opds-catalog;kind=navigation"/>
+  <link rel="search"
+        href="/catalog/v2/searchdescription.xml"
+        type="application/opensearchdescription+xml"/>
+  <title>OPDS Catalog Root</title>
+  <updated>YYYY-MM-DDThh:mm:ssZ</updated>
+
+  <entry>
+    <title>All entries</title>
+    <link rel="subsection"
+          href="/catalog/v2/entries"
+          type="application/atom+xml;profile=opds-catalog;kind=acquisition"/>
+    <updated>YYYY-MM-DDThh:mm:ssZ</updated>
+    <id>12345678-90ab-cdef-1234-567890abcdef</id>
+    <content type="text">All entries from this catalog.</content>
+  </entry>
+  <entry>
+    <title>List of categories</title>
+    <link rel="subsection"
+          href="/catalog/v2/categories"
+          type="application/atom+xml;profile=opds-catalog;kind=navigation"/>
+    <updated>YYYY-MM-DDThh:mm:ssZ</updated>
+    <id>12345678-90ab-cdef-1234-567890abcdef</id>
+    <content type="text">List of all categories in this catalog.</content>
+  </entry>
+</feed>
+)";
+  EXPECT_EQ(maskVariableOPDSFeedData(r->body), expected_output);
+}
+
+TEST_F(LibraryServerTest, catalog_v2_searchdescription_xml)
+{
+  const auto r = zfs1_->GET("/catalog/v2/searchdescription.xml");
+  EXPECT_EQ(r->status, 200);
+  EXPECT_EQ(r->body,
+    "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n"
+    "<OpenSearchDescription xmlns=\"http://a9.com/-/spec/opensearch/1.1/\">\n"
+    "  <ShortName>Zim catalog search</ShortName>\n"
+    "  <Description>Search zim files in the catalog.</Description>\n"
+    "  <Url type=\"application/atom+xml;profile=opds-catalog;kind=acquisition\"\n"
+    "       xmlns:atom=\"http://www.w3.org/2005/Atom\"\n"
+    "       xmlns:k=\"http://kiwix.org/opensearchextension/1.0\"\n"
+    "       indexOffset=\"0\"\n"
+    "       template=\"/catalog/v2/entries?q={searchTerms?}&lang={language?}&name={k:name?}&tag={k:tag?}&maxsize={k:maxsize?}&count={count?}&start={startIndex?}\"/>\n"
+    "</OpenSearchDescription>\n"
+  );
+}
+
+TEST_F(LibraryServerTest, catalog_v2_categories)
+{
+  const auto r = zfs1_->GET("/catalog/v2/categories");
+  EXPECT_EQ(r->status, 200);
+  const char expected_output[] = R"(<?xml version="1.0" encoding="UTF-8"?>
+<feed xmlns="http://www.w3.org/2005/Atom"
+      xmlns:opds="https://specs.opds.io/opds-1.2">
+  <id>12345678-90ab-cdef-1234-567890abcdef</id>
+  <link rel="self"
+        href="/catalog/v2/categories"
+        type="application/atom+xml;profile=opds-catalog;kind=navigation"/>
+  <link rel="start"
+        href="/catalog/v2/root.xml"
+        type="application/atom+xml;profile=opds-catalog;kind=navigation"/>
+  <title>List of categories</title>
+  <updated>YYYY-MM-DDThh:mm:ssZ</updated>
+
+  <entry>
+    <title>jazz</title>
+    <link rel="subsection"
+          href="/catalog/v2/entries?category=jazz"
+          type="application/atom+xml;profile=opds-catalog;kind=acquisition"/>
+    <updated>YYYY-MM-DDThh:mm:ssZ</updated>
+    <id>12345678-90ab-cdef-1234-567890abcdef</id>
+    <content type="text">All entries with category of 'jazz'.</content>
+  </entry>
+  <entry>
+    <title>wikipedia</title>
+    <link rel="subsection"
+          href="/catalog/v2/entries?category=wikipedia"
+          type="application/atom+xml;profile=opds-catalog;kind=acquisition"/>
+    <updated>YYYY-MM-DDThh:mm:ssZ</updated>
+    <id>12345678-90ab-cdef-1234-567890abcdef</id>
+    <content type="text">All entries with category of 'wikipedia'.</content>
+  </entry>
+</feed>
+)";
+  EXPECT_EQ(maskVariableOPDSFeedData(r->body), expected_output);
+}
+
+#define CATALOG_V2_ENTRIES_PREAMBLE(q)                        \
+    "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n"            \
+    "<feed xmlns=\"http://www.w3.org/2005/Atom\"\n"           \
+    "      xmlns:opds=\"https://specs.opds.io/opds-1.2\"\n"   \
+    "      xmlns:opensearch=\"http://a9.com/-/spec/opensearch/1.1/\">\n"  \
+    "  <id>12345678-90ab-cdef-1234-567890abcdef</id>\n"       \
+    "\n"                                                      \
+    "  <link rel=\"self\"\n"                                  \
+    "        href=\"/catalog/v2/entries" q "\"\n"             \
+    "        type=\"application/atom+xml;profile=opds-catalog;kind=acquisition\"/>\n" \
+    "  <link rel=\"start\"\n"                                 \
+    "        href=\"/catalog/v2/root.xml\"\n"              \
+    "        type=\"application/atom+xml;profile=opds-catalog;kind=navigation\"/>\n" \
+    "  <link rel=\"up\"\n"                                    \
+    "        href=\"/catalog/v2/root.xml\"\n"              \
+    "        type=\"application/atom+xml;profile=opds-catalog;kind=navigation\"/>\n" \
+    "\n"                                                      \
+
+
+TEST_F(LibraryServerTest, catalog_v2_entries)
+{
+  const auto r = zfs1_->GET("/catalog/v2/entries");
+  EXPECT_EQ(r->status, 200);
+  EXPECT_EQ(maskVariableOPDSFeedData(r->body),
+    CATALOG_V2_ENTRIES_PREAMBLE("")
+    "  <title>All Entries</title>\n"
+    "  <updated>YYYY-MM-DDThh:mm:ssZ</updated>\n"
+    "\n"
+    CHARLES_RAY_CATALOG_ENTRY
+    RAY_CHARLES_CATALOG_ENTRY
+    UNCATEGORIZED_RAY_CHARLES_CATALOG_ENTRY
+    "</feed>\n"
+  );
+}
+
+TEST_F(LibraryServerTest, catalog_v2_entries_filtered_by_range)
+{
+  {
+    const auto r = zfs1_->GET("/catalog/v2/entries?start=1");
+    EXPECT_EQ(r->status, 200);
+    EXPECT_EQ(maskVariableOPDSFeedData(r->body),
+      CATALOG_V2_ENTRIES_PREAMBLE("?start=1")
+      "  <title>Filtered Entries (start=1)</title>\n"
+      "  <updated>YYYY-MM-DDThh:mm:ssZ</updated>\n"
+      "  <totalResults>3</totalResults>\n"
+      "  <startIndex>1</startIndex>\n"
+      "  <itemsPerPage>2</itemsPerPage>\n"
+      RAY_CHARLES_CATALOG_ENTRY
+      UNCATEGORIZED_RAY_CHARLES_CATALOG_ENTRY
+      "</feed>\n"
+    );
+  }
+
+  {
+    const auto r = zfs1_->GET("/catalog/v2/entries?count=2");
+    EXPECT_EQ(r->status, 200);
+    EXPECT_EQ(maskVariableOPDSFeedData(r->body),
+      CATALOG_V2_ENTRIES_PREAMBLE("?count=2")
+      "  <title>Filtered Entries (count=2)</title>\n"
+      "  <updated>YYYY-MM-DDThh:mm:ssZ</updated>\n"
+      "  <totalResults>3</totalResults>\n"
+      "  <startIndex>0</startIndex>\n"
+      "  <itemsPerPage>2</itemsPerPage>\n"
+      CHARLES_RAY_CATALOG_ENTRY
+      RAY_CHARLES_CATALOG_ENTRY
+      "</feed>\n"
+    );
+  }
+
+  {
+    const auto r = zfs1_->GET("/catalog/v2/entries?start=1&count=1");
+    EXPECT_EQ(r->status, 200);
+    EXPECT_EQ(maskVariableOPDSFeedData(r->body),
+      CATALOG_V2_ENTRIES_PREAMBLE("?count=1&start=1")
+      "  <title>Filtered Entries (count=1&amp;start=1)</title>\n"
+      "  <updated>YYYY-MM-DDThh:mm:ssZ</updated>\n"
+      "  <totalResults>3</totalResults>\n"
+      "  <startIndex>1</startIndex>\n"
+      "  <itemsPerPage>1</itemsPerPage>\n"
+      RAY_CHARLES_CATALOG_ENTRY
+      "</feed>\n"
+    );
+  }
+}
+
+TEST_F(LibraryServerTest, catalog_v2_entries_filtered_by_search_terms)
+{
+  const auto r = zfs1_->GET("/catalog/v2/entries?q=\"ray%20charles\"");
+  EXPECT_EQ(r->status, 200);
+  EXPECT_EQ(maskVariableOPDSFeedData(r->body),
+    CATALOG_V2_ENTRIES_PREAMBLE("?q=%22ray%20charles%22")
+    "  <title>Filtered Entries (q=&quot;ray charles&quot;)</title>\n"
+    "  <updated>YYYY-MM-DDThh:mm:ssZ</updated>\n"
+    "  <totalResults>2</totalResults>\n"
+    "  <startIndex>0</startIndex>\n"
+    "  <itemsPerPage>2</itemsPerPage>\n"
+    RAY_CHARLES_CATALOG_ENTRY
+    CHARLES_RAY_CATALOG_ENTRY
+    "</feed>\n"
+  );
+}