Merge pull request #136 from kiwix/2.0.0

2.0.0

.travis.yml

@@ -1,13 +1,43 @@
 language: cpp
 dist: trusty
-sudo: required
+sudo: false
 cache: ccache
+before_install:
+- if [[ "$TRAVIS_OS_NAME" == "linux" ]]; then eval "${MATRIX_EVAL}"; fi
+- PATH=$PATH:$HOME/bin
+- ${CXX} --version
 install: travis/install_deps.sh
 script: travis/compile.sh
 env:
- - PLATFORM="native_static"
- - PLATFORM="native_dyn"
- - PLATFORM="win32_static"
- - PLATFORM="win32_dyn"
- - PLATFORM="android_arm"
- - PLATFORM="android_arm64"
+  global:
+    - MATRIX_EVAL="CC=gcc-5 && CXX=g++-5"
+  matrix:
+    - PLATFORM="native_static"
+    - PLATFORM="native_dyn"
+    - PLATFORM="win32_static"
+    - PLATFORM="win32_dyn"
+    - PLATFORM="android_arm"
+    - PLATFORM="android_arm64"
+addons:
+  apt:
+    sources:
+      - ubuntu-toolchain-r-test
+    packages:
+      - g++-5
+      - cmake
+      - python3-pip
+      - libbz2-dev
+      - ccache
+      - zlib1g-dev
+      - uuid-dev
+      - libctpp2-dev
+      - ctpp2-utils
+      - libmicrohttpd-dev
+      - g++-mingw-w64-i686
+      - gcc-mingw-w64-i686
+      - gcc-mingw-w64-base
+      - mingw-w64-tools
+matrix:
+  include:
+    - env: PLATFORM="native_dyn"
+      os: osx

ChangeLog

@@ -1,3 +1,47 @@
+kiwix-lib 2.0.0
+===============
+
+ * Introduce a new API to retrive content from a reader.
+   * Introduce the `Entry` class.
+   * Reader's methods return an `Entry`.
+   * Content and other information can be retrieved from the `Entry`.
+   * Older Reader's methods are depreciated.
+ * Add an `OPDSDumper` class to dump a whole `Library` as an OPDS feed.
+ * Add a tool function to get the content of a file.
+ * Add a tool function to create a tempory directory.
+ * Add a `Downloader` class to download a file.
+ * Allow the manager to populate a `Library` from an OPDS feed.
+ * Try to locate libctpp2 in default system libdir and then fallback in 'lib'
+   directory.
+ * Build kiwix-lib setting RPATH.
+ * Build kiwix-lib without warning (werror=true)
+ * Build kiwix-lib on macos.
+
+kiwix-lib 1.1.1
+===============
+
+ * Correct the name of kiwix-lib (from `kiwixlib`) in meson.build to generate
+   dist archive with the correct name.
+ * Libzim version need to be at least 3.2.0
+
+kiwix-lib 1.1.0
+===============
+
+ * Allow for more than 70 search result per page in html results rendering
+   (kiwix/kiwix-tools#92)
+ * Add a small api to do geo queries.
+ * Add multi-search support in the JNI (#67)
+ * Add an API to get only one part of an article.
+ * Add an API to get direct location of an article content in the zim file.
+ * Improve urlencoding
+ * Fix pagination in html results rendering.
+ * Compile using gcc-5 on Travis.
+ * Allow JNI to access search snippets.
+ * JNI throw an exception instead of returning an invalid object if something
+   goes wrong.
+ * Add doctext documentation. (#116)
+ * Various bug fixes.
+
 kiwix-lib 1.0.0
 ===============
 

README.md

@@ -37,6 +37,7 @@ libraries need to be available:
 (package libctpp2-dev on Ubuntu)
 * Xapian ......................................... https://xapian.org/
 (package libxapian-dev on Ubuntu)
+* libaria2 .................................. https://aria2.github.io/
 
 These dependencies may or may not be packaged by your operating
 system. They may also be packaged but only in an older version. The
@@ -68,7 +69,7 @@ Then install Meson itself:
 ```
 virtualenv -p python3 ./ # Create virtualenv
 source bin/activate      # Activate the virtualenv
-pip install meson        # Install Meson
+pip3 install meson        # Install Meson
 hash -r                  # Refresh bash paths
 ```
 

include/common.h

@@ -0,0 +1,24 @@
+
+#ifndef _KIWIX_COMMON_H_
+#define _KIWIX_COMMON_H_
+
+#include <zim/zim.h>
+
+#ifdef __GNUC__
+#define DEPRECATED __attribute__((deprecated))
+#elif defined(_MSC_VER)
+#define DEPRECATED __declspec(deprecated)
+#else
+#praga message("WARNING: You need to implement DEPRECATED for this compiler")
+#define DEPRECATED
+#endif
+
+
+namespace kiwix {
+
+typedef zim::size_type size_type;
+typedef zim::offset_type offset_type;
+
+}
+
+#endif //_KIWIX_COMMON_H_

include/common/pathTools.h

@@ -51,8 +51,10 @@ string appendToDirectory(const string& directoryPath, const string& filename);
 
 unsigned int getFileSize(const string& path);
 string getFileSizeAsString(const string& path);
+string getFileContent(const string& path);
 bool fileExists(const string& path);
 bool makeDirectory(const string& path);
+string makeTmpDirectory();
 bool copyFile(const string& sourcePath, const string& destPath);
 string getLastPathElement(const string& path);
 string getExecutablePath();

include/common/stringTools.h

@@ -23,6 +23,7 @@
 #include <unicode/unistr.h>
 
 #include <fstream>
+#include <iomanip>
 #include <iostream>
 #include <sstream>
 #include <string>
@@ -36,7 +37,6 @@ namespace kiwix
 
 std::string beautifyInteger(const unsigned int number);
 std::string beautifyFileSize(const unsigned int number);
-std::string urlEncode(const std::string& c);
 void printStringInHexadecimal(const char* s);
 void printStringInHexadecimal(UnicodeString s);
 void stringReplacement(std::string& str,
@@ -48,7 +48,9 @@ std::string encodeDiples(const std::string& str);
 
 std::string removeAccents(const std::string& text);
 void loadICUExternalTables();
-std::string urlDecode(const std::string& c);
+
+std::string urlEncode(const std::string& value, bool encodeReserved = false);
+std::string urlDecode(const std::string& value, bool component = false);
 
 std::vector<std::string> split(const std::string&, const std::string&);
 std::vector<std::string> split(const char*, const char*);

include/downloader.h

@@ -0,0 +1,70 @@
+/*
+ * Copyright 2018 Matthieu Gautier <mgautier@kymeria.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#ifndef KIWIX_DOWNLOADER_H
+#define KIWIX_DOWNLOADER_H
+
+#include <string>
+#include <aria2/aria2.h>
+#include <pthread.h>
+
+namespace kiwix
+{
+
+
+struct DownloadedFile {
+  DownloadedFile()
+   : success(false) {}
+  bool success;
+  std::string path;
+};
+
+/**
+ * A tool to download things.
+ *
+ */
+class Downloader
+{
+ public:
+  Downloader();
+  ~Downloader();
+
+  /**
+   * Download a content.
+   *
+   * @param url the url to download
+   * @return the content downloaded.
+   */
+  DownloadedFile download(const std::string& url);
+
+ private:
+  static pthread_mutex_t globalLock;
+
+  aria2::Session* session;
+  DownloadedFile* fileHandle;
+  std::string tmpDir;
+
+  static int downloadEventCallback(aria2::Session* session,
+                                   aria2::DownloadEvent event,
+                                   aria2::A2Gid gid,
+                                   void* userData);
+};
+}
+
+#endif

include/entry.h

@@ -0,0 +1,191 @@
+/*
+ * Copyright 2018 Matthieu Gautier <mgautier@kymeria.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#ifndef KIWIX_ENTRY_H
+#define KIWIX_ENTRY_H
+
+#include <stdio.h>
+#include <zim/article.h>
+#include <exception>
+#include <string>
+#include "common.h"
+
+using namespace std;
+
+namespace kiwix
+{
+
+
+class NoEntry : public std::exception {};
+
+/**
+ * A entry represent an.. entry in a zim file.
+ */
+class Entry
+{
+  public:
+    /**
+     * Default constructor.
+     *
+     * Construct an invalid entry.
+     */
+    Entry() = default;
+
+    /**
+     * Construct an entry making reference to an zim article.
+     *
+     * @param article a zim::Article object
+     */
+    Entry(zim::Article article);
+    virtual ~Entry() = default;
+
+    /**
+     * Get the path of the entry.
+     *
+     * The path is the "key" of an entry.
+     *
+     * @return the path of the entry.
+     */
+    std::string getPath() const;
+    
+    /**
+     * Get the title of the entry.
+     *
+     * @return the title of the entry.
+     */
+    std::string getTitle() const;
+        
+    /**
+     * Get the content of the entry.
+     *
+     * The string is a copy of the content.
+     * If you don't want to do a copy, use get_blob.
+     *
+     * @return the content of the entry.
+     */
+    std::string getContent() const;
+        
+    /**
+     * Get the blob of the entry.
+     *
+     * A blob make reference to the content without copying it.
+     *
+     * @param offset The starting offset of the blob.
+     * @return the blob of the entry.
+     */
+    zim::Blob   getBlob(offset_type offset = 0) const;
+        
+    /**
+     * Get the blob of the entry.
+     *
+     * A blob make reference to the content without copying it.
+     *
+     * @param offset The starting offset of the blob.
+     * @param size The size of the blob.
+     * @return the blob of the entry.
+     */
+    zim::Blob   getBlob(offset_type offset, size_type size) const;
+        
+    /**
+     * Get the info for direct access to the content of the entry.
+     *
+     * Some entry (ie binary ones) have their content plain stored
+     * in the zim file. Knowing the offset where the content is stored
+     * an user can directly read the content in the zim file bypassing the
+     * kiwix-lib/libzim.
+     *
+     * @return A pair specifying where to read the content.
+     *         The string is the real file to read (may be different that .zim
+     *         file if zim is cut).
+     *         The offset is the offset to read in the file.
+     *         Return <"",0> if is not possible to read directly.
+     */
+    std::pair<std::string, offset_type> getDirectAccessInfo() const;
+        
+    /**
+     * Get the size of the entry.
+     *
+     * @return the size of the entry.
+     */
+    size_type   getSize() const;
+
+    /**
+     * Get the mime_type of the entry.
+     *
+     * @return the mime_type of the entry.
+     */
+    std::string getMimetype() const;
+    
+    
+    /**
+     * Get if the entry is a redirect entry.
+     *
+     * @return True if the entry is a redirect.
+     */
+    bool isRedirect() const;
+
+    /**
+     * Get if the entry is a link target entry.
+     *
+     * @return True if the entry is a link target.
+     */
+    bool isLinkTarget() const;
+
+    /**
+     * Get if the entry is a deleted entry.
+     *
+     * @return True if the entry is a deleted entry.
+     */
+    bool isDeleted() const;
+
+    /**
+     * Get the entry pointed by this entry.
+     *
+     * @return the entry pointed.
+     * @throw NoEntry if the entry is not a redirected entry.
+     */
+    Entry getRedirectEntry() const;
+
+    /**
+     * Get the final entry pointed by this entry.
+     *
+     * Follow the redirection until a "not redirecting" entry is found.
+     * If the entry is not a redirected entry, return the entry itself.
+     *
+     * @return the final entry.
+     */
+    Entry getFinalEntry() const;
+
+    /**
+     * Convert the entry to a boolean value.
+     *
+     * @return True if the entry is valid.
+     */
+    explicit operator bool() const { return good(); }
+
+  private:
+    zim::Article article;
+    mutable zim::Article final_article;
+
+    bool good() const { return article.good(); }
+};
+
+}
+
+#endif // KIWIX_ENTRY_H

include/library.h

@@ -38,6 +38,10 @@ namespace kiwix
 {
 enum supportedIndexType { UNKNOWN, XAPIAN };
 
+
+/**
+ * A class to store information about a book (a zim file)
+ */
 class Book
 {
  public:
@@ -78,6 +82,9 @@ class Book
   string faviconMimeType;
 };
 
+/**
+ * A Library store several books.
+ */
 class Library
 {
  public:
@@ -85,7 +92,24 @@ class Library
   ~Library();
 
   string version;
+  /**
+   * Add a book to the library.
+   *
+   * If a book already exist in the library with the same id, update
+   * the existing book instead of adding a new one.
+   *
+   * @param book The book to add.
+   * @return True if the book has been added.
+   *         False if a book has been updated.
+   */
   bool addBook(const Book& book);
+
+  /**
+   * Remove a book from the library.
+   *
+   * @param bookIndex the index of the book to remove.
+   * @return True
+   */
   bool removeBookByIndex(const unsigned int bookIndex);
   vector<kiwix::Book> books;
 

include/manager.h

@@ -39,43 +39,225 @@ namespace kiwix
 enum supportedListMode { LASTOPEN, REMOTE, LOCAL };
 enum supportedListSortBy { TITLE, SIZE, DATE, CREATOR, PUBLISHER };
 
+/**
+ * A tool to manage a `Library`.
+ *
+ * A `Manager` handle a internal `Library`.
+ * This `Library` can be retrived with `cloneLibrary` method.
+ */
 class Manager
 {
  public:
   Manager();
   ~Manager();
 
+  /**
+   * Read a `library.xml` and add book in the file to the library.
+   *
+   * @param path The path to the `library.xml`.
+   * @param readOnly Set if the libray path could be overwritten latter with
+   *                 updated content.
+   * @return True if file has been properly parsed.
+   */
   bool readFile(const string path, const bool readOnly = true);
+
+  /**
+   * Read a `library.xml` and add book in the file to the library.
+   *
+   * @param nativePath The path of the `library.xml`
+   * @param UTF8Path The utf8 version (?) of the path. Also the path where the
+   *                 library will be writen i readOnly is False.
+   * @param readOnly Set if the libray path could be overwritten latter with
+   *                 updated content.
+   * @return True if file has been properly parsed.
+   */
   bool readFile(const string nativePath,
                 const string UTF8Path,
                 const bool readOnly = true);
-  bool readXml(const string xml,
+
+  /**
+   * Load a library content store in the string.
+   *
+   * @param xml The content corresponding of the library xml
+   * @param readOnly Set if the libray path could be overwritten latter with
+   *                 updated content.
+   * @param libraryPath The library path (used to resolve relative path)
+   * @return True if the content has been properly parsed.
+   */
+  bool readXml(const string& xml,
                const bool readOnly = true,
                const string libraryPath = "");
+
+  /**
+   * Load a library content stored in a OPDS stream.
+   *
+   * @param content The content of the OPDS stream.
+   * @param readOnly Set if the library path could be overwritten later with
+   *                 updated content.
+   * @param libraryPath The library path (used to resolve relative path)
+   * @return True if the content has been properly parsed.
+   */
+  bool readOpds(const string& content, const std::string& urlHost);
+
+  /**
+   * Write the library to a file.
+   *
+   * @param path the path of the file to write.
+   * @return True.
+   */
   bool writeFile(const string path);
+
+
+  /**
+   * Remove a book from the library.
+   *
+   * @param bookIndex the index of the book to remove
+   * @return True
+   */
   bool removeBookByIndex(const unsigned int bookIndex);
+
+  /**
+   * Remove a book from the library.
+   *
+   * @param id the id of the book to remove.
+   * @return True if the book were in the library.
+   */
   bool removeBookById(const string id);
+
+  /**
+   * Set the current book.
+   *
+   * @param id The id to add to the stack of current books.
+   *           If id is empty, remove the current book from the stack.
+   * @return True
+   */
   bool setCurrentBookId(const string id);
-  string getCurrentBookId();
+
+  /**
+   * Get the current book id.
+   *
+   * @return The id of the current book (or empty string if no current book).
+   */
+  string getCurrentBookId() const;
+
+  /**
+   * Set the path of the external fulltext index associated to a book.
+   *
+   * @param id The id of the book to set.
+   * @param path The path of the external fullext index.
+   * @param supportedIndexType The type of the fulltext index.
+   * @return True if the book is in the library.
+   */
   bool setBookIndex(const string id,
                     const string path,
-                    const supportedIndexType type);
-  bool setBookIndex(const string id, const string path);
+                    const supportedIndexType type = XAPIAN);
+
+  /**
+   * Set the path of the zim file associated to a book.
+   *
+   * @param id The id of the book to set.
+   * @param path The path of the zim file.
+   * @return True if the book is in the library.
+   */
   bool setBookPath(const string id, const string path);
+
+  /**
+   * Add a book to the library.
+   *
+   * @param pathToOpen The path to the zim file to add.
+   * @param pathToSave The path to store in the library in place of pathToOpen.
+   * @param url        The url of the book to store in the library.
+   * @param checMetaData Tell if we check metadata before adding book to the
+   *                     library.
+   * @return The id of the book if the book has been added to the library.
+   *         Else, an empty string.
+   */
   string addBookFromPathAndGetId(const string pathToOpen,
                                  const string pathToSave = "",
                                  const string url = "",
                                  const bool checkMetaData = false);
+
+  /**
+   * Add a book to the library.
+   *
+   * @param pathToOpen The path to the zim file to add.
+   * @param pathToSave The path to store in the library in place of pathToOpen.
+   * @param url        The url of the book to store in the library.
+   * @param checMetaData Tell if we check metadata before adding book to the
+   *                     library.
+   * @return True if the book has been added to the library.
+   */
+
   bool addBookFromPath(const string pathToOpen,
                        const string pathToSave = "",
                        const string url = "",
                        const bool checkMetaData = false);
+
+  /**
+   * Clone and return the internal library.
+   *
+   * @return A clone of the library.
+   */
   Library cloneLibrary();
+
+  /**
+   * Get the book corresponding to an id.
+   *
+   * @param[in] id The id of the book
+   * @param[out] book The book corresponding to the id.
+   * @return True if the book has been found.
+   */
   bool getBookById(const string id, Book& book);
+
+  /**
+   * Get the current book.
+   *
+   * @param[out] The current book.
+   * @return True if there is a current book.
+   */
   bool getCurrentBook(Book& book);
+
+  /**
+   * Get the number of book in the library.
+   *
+   * @param localBooks If we must count local books (books with a path).
+   * @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);
+
+  /**
+   * Update the "last open date" of a book
+   *
+   * @param id the id of the book.
+   * @return True if the book is in the library.
+   */
   bool updateBookLastOpenDateById(const string id);
+
+  /**
+   * Remove (set to empty) paths of all books in the library.
+   */
   void removeBookPaths();
+
+  /**
+   * List books in the library.
+   *
+   * The books list will be available in public vector member `bookIdList`.
+   *
+   * @param mode The mode of listing :
+   *             - LASTOPEN sort by last opened book.
+   *             - LOCAL list only local file.
+   *             - REMOTE list only remote file.
+   * @param sortBy Attribute to sort by the book list.
+   * @param maxSize Do not list book bigger than maxSize MiB.
+   *                Set to 0 to cancel this filter.
+   * @param language List only books in this language.
+   * @param creator List only books of this creator.
+   * @param publisher List only books of this publisher.
+   * @param search List only books with search in the title, description or
+   *               language.
+   * @return True
+   */
   bool listBooks(const supportedListMode mode,
                  const supportedListSortBy sortBy,
                  const unsigned int maxSize,
@@ -83,9 +265,42 @@ class Manager
                  const string creator,
                  const string publisher,
                  const string search);
+
+  /**
+   * Filter the library and generate a new one with the keep elements.
+   *
+   * @param search List only books with search in the title or description.
+   * @return A `Library`.
+   */
+  Library filter(const string& search);
+
+
+  /**
+   * Get all langagues of the books in the library.
+   *
+   * @return A list of languages.
+   */
   vector<string> getBooksLanguages();
+
+  /**
+   * Get all book creators of the books in the library.
+   *
+   * @return A list of book creators.
+   */
   vector<string> getBooksCreators();
+
+  /**
+   * Get all book publishers of the books in the library.
+   *
+   * @return A list of book publishers.
+   */
   vector<string> getBooksPublishers();
+
+  /**
+   * Get all book ids of the books in the library.
+   *
+   * @return A list of book ids.
+   */
   vector<string> getBooksIds();
 
   string writableLibraryPath;
@@ -99,6 +314,8 @@ class Manager
   bool parseXmlDom(const pugi::xml_document& doc,
                    const bool readOnly,
                    const string libraryPath);
+  bool parseOpdsDom(const pugi::xml_document& doc,
+                    const std::string& urlHost);
 
  private:
   void checkAndCleanBookPaths(Book& book, const string& libraryPath);

include/meson.build

@@ -1,7 +1,11 @@
 headers = [
+  'common.h',
   'library.h',
   'manager.h',
+  'opds_dumper.h',
+  'downloader.h',
   'reader.h',
+  'entry.h',
   'searcher.h'
 ]
 

include/opds_dumper.h

@@ -0,0 +1,107 @@
+/*
+ * Copyright 2017 Matthieu Gautier <mgautier@kymeria.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#ifndef KIWIX_OPDS_DUMPER_H
+#define KIWIX_OPDS_DUMPER_H
+
+#include <time.h>
+#include <sstream>
+#include <string>
+
+#include <pugixml.hpp>
+
+#include "common/base64.h"
+#include "common/pathTools.h"
+#include "common/regexTools.h"
+#include "library.h"
+#include "reader.h"
+
+using namespace std;
+
+namespace kiwix
+{
+
+/**
+ * A tool to dump a `Library` into a opds stream.
+ *
+ */
+class OPDSDumper
+{
+ public:
+  OPDSDumper() = default;
+  OPDSDumper(Library library);
+  ~OPDSDumper();
+
+  /**
+   * Dump the OPDS feed.
+   *
+   * @param id The id of the library.
+   * @return The OPDS feed.
+   */
+  std::string dumpOPDSFeed();
+
+  /**
+   * Set the id of the opds stream.
+   *
+   * @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; }
+
+  /**
+   * Set the root location used when generating url.
+   *
+   * @param rootLocation the root location to use.
+   */
+  void setRootLocation(const std::string& rootLocation) { this->rootLocation = rootLocation; }
+
+  /**
+   * Set the search url.
+   *
+   * @param searchUrl the search url to use.
+   */
+  void setSearchDescriptionUrl(const std::string& searchDescriptionUrl) { this->searchDescriptionUrl = searchDescriptionUrl; }
+
+  /**
+   * 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 rootLocation;
+  std::string searchDescriptionUrl;
+
+ private:
+  pugi::xml_node handleBook(Book book, pugi::xml_node root_node);
+};
+}
+
+#endif // KIWIX_OPDS_DUMPER_H

include/reader.h

@@ -29,6 +29,8 @@
 #include <map>
 #include <sstream>
 #include <string>
+#include "common.h"
+#include "entry.h"
 #include "common/pathTools.h"
 #include "common/stringTools.h"
 
@@ -36,84 +38,458 @@ using namespace std;
 
 namespace kiwix
 {
+
+/**
+ * The Reader class is the class who allow to get an entry content from a zim
+ * file.
+ */
 class Reader
 {
  public:
+  /**
+   * Create a Reader to read a zim file specified by zimFilePath.
+   *
+   * @param zimFilePath The path to the zim file to read.
+   *                    The zim file can be splitted (.zimaa, .zimab, ...).
+   *                    In this case, the file path must still point to the
+   *                    unsplitted path as if the file were not splitted
+   *                    (.zim extesion).
+   */
   Reader(const string zimFilePath);
   ~Reader();
 
-  void reset();
+  /**
+   * Get the number of "displayable" entries in the zim file.
+   *
+   * @return If the zim file has a /M/Counter metadata, return the number of
+   *         entries with the 'text/html' MIMEtype specified in the metadata.
+   *         Else return the number of entries in the 'A' namespace.
+   */
   unsigned int getArticleCount() const;
+
+  /**
+   * Get the number of media in the zim file.
+   *
+   * @return If the zim file has a /M/Counter metadata, return the number of
+   *         entries with the 'image/jpeg', 'image/gif' and 'image/png' in
+   *         the metadata.
+   *         Else return the number of entries in the 'I' namespace.
+   */
   unsigned int getMediaCount() const;
+
+  /**
+   * Get the number of all entries in the zim file.
+   *
+   * @return Return the number of all the entries, whatever their MIMEtype or
+   *         their namespace.
+   */
   unsigned int getGlobalCount() const;
+
+  /**
+   * Get the path of the zim file.
+   *
+   * @return the path of the zim file as given in the constructor.
+   */
   string getZimFilePath() const;
+
+  /**
+   * Get the Id of the zim file.
+   *
+   * @return The uuid stored in the zim file.
+   */
   string getId() const;
-  string getRandomPageUrl() const;
-  string getFirstPageUrl() const;
-  string getMainPageUrl() const;
-  bool getMetatag(const string& url, string& content) const;
+
+  /**
+   * Get the url of a random page.
+   *
+   * Deprecated : Use `getRandomPage` instead.
+   *
+   * @return Url of a random page. The page is picked from all entries in
+   *         the 'A' namespace.
+   *         The main page is excluded from the potential results.
+   */
+  DEPRECATED string getRandomPageUrl() const;
+
+  /**
+   * Get a random page.
+   *
+   * @return A random Entry. The entry is picked from all entries in
+   *         the 'A' namespace.
+   *         The main entry is excluded from the potential results.
+   */
+  Entry getRandomPage() const;
+
+  /**
+   * Get the url of the first page.
+   *
+   * Deprecated : Use `getFirstPage` instead.
+   *
+   * @return Url of the first entry in the 'A' namespace.
+   */
+  DEPRECATED string getFirstPageUrl() const;
+
+  /**
+   * Get the entry of the first page.
+   *
+   * @return The first entry in the 'A' namespace.
+   */
+  Entry getFirstPage() const;
+
+  /**
+   * Get the url of the main page.
+   *
+   * Deprecated : Use `getMainPage` instead.
+   *
+   * @return Url of the main page as specified in the zim file.
+   */
+  DEPRECATED string getMainPageUrl() const;
+
+  /**
+   * Get the entry of the main page.
+   *
+   * @return Entry of the main page as specified in the zim file.
+   */
+  Entry getMainPage() const;
+
+  /**
+   * Get the content of a metadata.
+   *
+   * @param[in]  name The name of the metadata.
+   * @param[out] value The value will be set to the content of the metadata.
+   * @return True if it was possible to get the content of the metadata.
+   */
+  bool getMetatag(const string& name, string& value) const;
+
+  /**
+   * Get the title of the zim file.
+   *
+   * @return The title of zim file as specified in the zim metadata.
+   *         If no title has been set, return a title computed from the
+   *         file path.
+   */
   string getTitle() const;
+
+  /**
+   * Get the description of the zim file.
+   *
+   * @return The description of the zim file as specified in the zim metadata.
+   *         If no description has been set, return the subtitle.
+   */
   string getDescription() const;
+
+  /**
+   * Get the language of the zim file.
+   *
+   * @return The language of the zim file as specified in the zim metadata.
+   */
   string getLanguage() const;
+
+  /**
+   * Get the name of the zim file.
+   *
+   * @return The name of the zim file as specified in the zim metadata.
+   */
   string getName() const;
+
+  /**
+   * Get the tags of the zim file.
+   *
+   * @return The tags of the zim file as specified in the zim metadata.
+   */
   string getTags() const;
+
+  /**
+   * Get the date of the zim file.
+   *
+   * @return The date of the zim file as specified in the zim metadata.
+   */
   string getDate() const;
+
+  /**
+   * Get the creator of the zim file.
+   *
+   * @return The creator of the zim file as specified in the zim metadata.
+   */
   string getCreator() const;
+
+  /**
+   * Get the publisher of the zim file.
+   *
+   * @return The publisher of the zim file as specified in the zim metadata.
+   */
   string getPublisher() const;
+
+  /**
+   * Get the origId of the zim file.
+   *
+   * The origId is only used in the case of patch zim file and is the Id
+   * of the original zim file.
+   *
+   * @return The origId of the zim file as specified in the zim metadata.
+   */
   string getOrigId() const;
+
+  /**
+   * Get the favicon of the zim file.
+   *
+   * @param[out] content The content of the favicon.
+   * @param[out] mimeType The mimeType of the favicon.
+   * @return True if a favicon has been found.
+   */
   bool getFavicon(string& content, string& mimeType) const;
-  bool getPageUrlFromTitle(const string& title, string& url) const;
-  bool getMimeTypeByUrl(const string& url, string& mimeType) const;
-  bool getContentByUrl(const string& url,
+
+  /**
+   * Get an entry associated to an path.
+   *
+   * @param path The path of the entry.
+   * @return The entry.
+   * @throw NoEntry If no entry correspond to the path.
+   */
+  Entry getEntryFromPath(const std::string& path) const;
+
+  /**
+   * Get an entry associated to an url encoded path.
+   *
+   * Equivalent to `getEntryFromPath(urlDecode(path));`
+   *
+   * @param path The url encoded path.
+   * @return The entry.
+   * @throw NoEntry If no entry correspond to the path.
+   */
+  Entry getEntryFromEncodedPath(const std::string& path) const;
+
+  /**
+   * Get un entry associated to a title.
+   *
+   * @param title The title.
+   * @return The entry
+   * throw NoEntry If no entry correspond to the url.
+   */
+  Entry getEntryFromTitle(const std::string& title) const;
+
+  /**
+   * Get the url of a page specified by a title.
+   *
+   * @param[in] title the title of the page.
+   * @param[out] url the url of the page.
+   * @return True if the page can be found.
+   */
+  DEPRECATED bool getPageUrlFromTitle(const string& title, string& url) const;
+
+  /**
+   * Get the mimetype of a entry specified by a url.
+   *
+   * @param[in] url the url of the entry.
+   * @param[out] mimeType the mimeType of the entry.
+   * @return True if the mimeType has been found.
+   */
+  DEPRECATED bool getMimeTypeByUrl(const string& url, string& mimeType) const;
+
+  /**
+   * Get the content of an entry specifed by a url.
+   *
+   * Alias to `getContentByEncodedUrl`
+   */
+  DEPRECATED bool getContentByUrl(const string& url,
                        string& content,
                        string& title,
                        unsigned int& contentLength,
                        string& contentType) const;
-  bool getContentByEncodedUrl(const string& url,
+
+  /**
+   * Get the content of an entry specified by a url encoded url.
+   *
+   * Equivalent to getContentByDecodedUrl(urlDecode(url), ...).
+   */
+  DEPRECATED bool getContentByEncodedUrl(const string& url,
                               string& content,
                               string& title,
                               unsigned int& contentLength,
                               string& contentType,
                               string& baseUrl) const;
-  bool getContentByEncodedUrl(const string& url,
+
+  /**
+   * Get the content of an entry specified by an url encoded url.
+   *
+   * Equivalent to getContentByEncodedUrl but without baseUrl.
+   */
+  DEPRECATED bool getContentByEncodedUrl(const string& url,
                               string& content,
                               string& title,
                               unsigned int& contentLength,
                               string& contentType) const;
-  bool getContentByDecodedUrl(const string& url,
+
+  /**
+   * Get the content of an entry specified by a url.
+   *
+   * @param[in] url The url of the entry.
+   * @param[out] content The content of the entry.
+   * @param[out] title the title of the entry.
+   * @param[out] contentLength The size of the entry (size of content).
+   * @param[out] contentType The mimeType of the entry.
+   * @param[out] baseUrl Return the true url of the entry.
+   *                     If the specified entry is a redirection, contains
+   *                     the url of the targeted entry.
+   * @return True if the entry has been found.
+   */
+  DEPRECATED bool getContentByDecodedUrl(const string& url,
                               string& content,
                               string& title,
                               unsigned int& contentLength,
                               string& contentType,
                               string& baseUrl) const;
-  bool getContentByDecodedUrl(const string& url,
+  /**
+   * Get the content of an entry specified by a url.
+   *
+   * Equivalent to getContentByDecodedUrl but withou the baseUrl.
+   */
+  DEPRECATED bool getContentByDecodedUrl(const string& url,
                               string& content,
                               string& title,
                               unsigned int& contentLength,
                               string& contentType) const;
+
+  /**
+   * Search for entries with title starting with prefix (case sensitive).
+   *
+   * Suggestions are stored in an internal vector and can be retrieved using
+   * `getNextSuggestion` method.
+   *
+   * @param prefix The prefix to search.
+   * @param suggestionsCount How many suggestions to search for.
+   * @param reset If true, remove previous suggestions in the internal vector.
+   *              If false, add suggestions to the internal vector
+   *              (until internal vector size is suggestionCount (or no more
+   *               suggestion))
+   * @return True if some suggestions where added to the internal vector.
+   */
   bool searchSuggestions(const string& prefix,
                          unsigned int suggestionsCount,
                          const bool reset = true);
+
+  /**
+   * Search for entries for the given prefix.
+   *
+   * If the zim file has a internal fulltext index, the suggestions will be
+   * searched using it.
+   * Else the suggestions will be search using `searchSuggestions` while trying
+   * to be smart about case sensitivity (using `getTitleVariants`).
+   *
+   * In any case, suggestions are stored in an internal vector and can be
+   * retrieved using `getNextSuggestion` method.
+   * The internal vector will be reset.
+   *
+   * @param prefix The prefix to search for.
+   * @param suggestionsCount How many suggestions to search for.
+   */
   bool searchSuggestionsSmart(const string& prefix,
                               unsigned int suggestionsCount);
-  bool urlExists(const string& url) const;
+
+  /**
+   * Check if the url exists in the zim file.
+   *
+   * Deprecated : Use `pathExists` instead.
+   *
+   * @param url the url to check.
+   * @return True if the url exits in the zim file.
+   */
+  DEPRECATED bool urlExists(const string& url) const;
+
+  /**
+   * Check if the path exists in the zim file.
+   *
+   * @param path the path to check.
+   * @return True if the path exists in the zim file.
+   */
+  bool pathExists(const string& path) const;
+
+  /**
+   * Check if the zim file has a embedded fulltext index.
+   *
+   * @return True if the zim file has a embedded fulltext index
+   *         and is not split (else the fulltext is not accessible).
+   */
   bool hasFulltextIndex() const;
+
+  /**
+   * Get potential case title variations for a title.
+   *
+   * @param title a title.
+   * @return the list of variantions.
+   */
   std::vector<std::string> getTitleVariants(const std::string& title) const;
+
+  /**
+   * Get the next suggestion title.
+   *
+   * @param[out] title the title of the suggestion.
+   * @return True if title has been set.
+   */
   bool getNextSuggestion(string& title);
+
+  /**
+   * Get the next suggestion title and url.
+   *
+   * @param[out] title the title of the suggestion.
+   * @param[out] url the url of the suggestion.
+   * @return True if title and url have been set.
+   */
   bool getNextSuggestion(string& title, string& url);
+
+  /**
+   * Get if we can check zim file integrity (has a checksum).
+   *
+   * @return True if zim file have a checksum.
+   */
   bool canCheckIntegrity() const;
+
+  /**
+   * Check is zim file is corrupted.
+   *
+   * @return True if zim file is corrupted.
+   */
   bool isCorrupted() const;
-  bool parseUrl(const string& url, char* ns, string& title) const;
+
+  /**
+   * Parse a full url into a namespace and url.
+   *
+   * @param[in] url The full url ("/N/url").
+   * @param[out] ns The namespace (N).
+   * @param[out] title The url (url).
+   * @return True
+   */
+  DEPRECATED bool parseUrl(const string& url, char* ns, string& title) const;
+
+  /**
+   * Return the total size of the zim file.
+   *
+   * If zim file is split, return the sum of all parts' size.
+   *
+   * @return Size of the size file is KiB.
+   */
   unsigned int getFileSize() const;
+
+  /**
+   * Get the zim file handler.
+   *
+   * @return The libzim file handler.
+   */
   zim::File* getZimFileHandler() const;
-  bool getArticleObjectByDecodedUrl(const string& url,
+
+  /**
+   * Get the zim article object associated to a url.
+   *
+   * @param[in] url The url of the article.
+   * @param[out] article The libzim article object.
+   * @return True if the url is good (article.good()).
+   */
+  DEPRECATED bool getArticleObjectByDecodedUrl(const string& url,
                                     zim::Article& article) const;
 
  protected:
   zim::File* zimFileHandler;
   zim::size_type firstArticleOffset;
   zim::size_type lastArticleOffset;
-  zim::size_type currentArticleOffset;
   zim::size_type nsACount;
   zim::size_type nsICount;
   std::string zimFilePath;

include/searcher.h

@@ -53,29 +53,149 @@ class Result
 };
 
 struct SearcherInternal;
+/**
+ * The Searcher class is reponsible to do different kind of search using the
+ * fulltext index.
+ *
+ * Historically, there are two kind of fulltext index :
+ *  - The legacy one, is the external fulltext index. A directory stored outside
+ *    of the zim file.
+ *  - The new one, a embedded fulltext index in the zim file.
+ *
+ * Legacy external fulltext index has to be considered as obsolet format with
+ * less functionnalities:
+ *  - No multi zim search ;
+ *  - No geo_search ;
+ *  - No suggestions search ;
+ *
+ * To reflect this, there is two Search creation "API":
+ *  - One for the external fulltext index, using the constructor taking a
+ *    xapianDirectoryPath) ;
+ *  - One for the embedded fulltext index, using a "empty" constructor and the
+ *  `add_reader` method".
+ *
+ *  On top of that, the Searcher may (if compiled with ctpp2) be used to
+ *  generate a html page for the search result. This use a template that need a
+ *  humanReaderName. This feature is only used by kiwix-serve and this should be
+ *  move outside of Searcher (and with a better API). If you don't use the html
+ *  rendering (getHtml method), you better should simply ignore the different
+ *  humanReadeableName attributes (or give an empty string).
+ */
 class Searcher
 {
  public:
-  Searcher();
+  /**
+   * The default constructor.
+   *
+   * @param humanReadableName The global zim's humanReadableName.
+   *                          Used to generate pagination links.
+   */
+  Searcher(const string& humanReadableName = "");
+
+  /**
+   * The constructor for legacy external fulltext index.
+   *
+   * @param xapianDirectoryPath The path to the external index directory.
+   * @param reader The reader associated to the external index.
+   *               It will be used retrive the article content or generate
+   *               the snippet.
+   * @param humanReadableName The humanReadableName for the zim.
+   */
   Searcher(const string& xapianDirectoryPath,
            Reader* reader,
            const string& humanReadableName);
   ~Searcher();
 
-  void add_reader(Reader* reader, const std::string& humanReaderName);
+  /**
+   * Add a reader (containing embedded fulltext index) to the search.
+   *
+   * @param reader The Reader for the zim containing the fulltext index.
+   * @param humanReaderName The human readable name of the reader.
+   * @return true if the reader has been added.
+   *         false if the reader cannot be added (no embedded fulltext index present)
+   */
+  bool add_reader(Reader* reader, const std::string& humanReaderName);
+
+  /**
+   * Start a search on the zim associated to the Searcher.
+   *
+   * Search results should be retrived using the getNextResult method.
+   *
+   * @param search The search query.
+   * @param resultStart the start offset of the search results (used for pagination).
+   * @param resultEnd the end offset of the search results (used for pagination).
+   * @param verbose print some info on stdout if true.
+   */
   void search(std::string& search,
               unsigned int resultStart,
               unsigned int resultEnd,
               const bool verbose = false);
+
+  /**
+   * Start a geographique search.
+   * The search return result for entry in a disc of center latitude/longitude
+   * and radius distance.
+   *
+   * Search results should be retrived using the getNextResult method.
+   *
+   * @param latitude The latitude of the center point.
+   * @param longitude The longitude of the center point.
+   * @param distance The radius of the disc.
+   * @param resultStart the start offset of the search results (used for pagination).
+   * @param resultEnd the end offset of the search results (used for pagination).
+   * @param verbose print some info on stdout if true.
+   */
+  void geo_search(float latitude, float longitude, float distance,
+                  unsigned int resultStart,
+                  unsigned int resultEnd,
+                  const bool verbose = false);
+
+  /**
+   * Start a suggestion search.
+   * The search made depend of the "version" of the embedded index.
+   *  - If the index is newer enough and have a title namespace, the search is
+   *    made in the titles only.
+   *  - Else the search is made on the whole article content.
+   * In any case, the search is made "partial" (as adding '*' at the end of the query)
+   *
+   * @param search The search query.
+   * @param verbose print some info on stdout if true.
+   */
   void suggestions(std::string& search, const bool verbose = false);
+
+  /**
+   * Get the next result of a started search.
+   * This is the method to use to loop hover the search results.
+   */
   Result* getNextResult();
+
+  /**
+   * Restart the previous search.
+   * Next call to getNextResult will return the first result.
+   */
   void restart_search();
+
+  /**
+   * Get a estimation of the result count.
+   */
   unsigned int getEstimatedResultCount();
+
+  /**
+   * Set protocol prefix.
+   * Only used by getHtml.
+   */
   bool setProtocolPrefix(const std::string prefix);
+
+  /**
+   * Set search protocol prefix.
+   * Only used by getHtml.
+   */
   bool setSearchProtocolPrefix(const std::string prefix);
-  void reset();
 
 #ifdef ENABLE_CTPP2
+  /**
+   * Generate the html page with the resutls of the search.
+   */
   string getHtml();
 #endif
 
@@ -98,7 +218,13 @@ class Searcher
   unsigned int resultStart;
   unsigned int resultEnd;
   std::string contentHumanReadableId;
+
+ private:
+  void reset();
+
 };
+
+
 }
 
 #endif

meson.build

@@ -1,7 +1,7 @@
-project('kiwixlib', 'cpp',
-  version : '1.0.1',
+project('kiwix-lib', 'cpp',
+  version : '2.0.0',
   license : 'GPL',
-  default_options : ['c_std=c11', 'cpp_std=c++11'])
+  default_options : ['c_std=c11', 'cpp_std=c++11', 'werror=true'])
 
 compiler = meson.get_compiler('cpp')
 find_library_in_compiler = meson.version().version_compare('>=0.31.0')
@@ -10,8 +10,9 @@ static_deps = get_option('android') or get_option('default_library') == 'static'
 
 thread_dep = dependency('threads')
 libicu_dep = dependency('icu-i18n', static:static_deps)
-libzim_dep = dependency('libzim', version : '>=3.0.0', static:static_deps)
+libzim_dep = dependency('libzim', version : '>=3.3.0', static:static_deps)
 pugixml_dep = dependency('pugixml', static:static_deps)
+libaria2_dep = dependency('libaria2', static:static_deps)
 
 ctpp2_include_path = ''
 has_ctpp2_dep = false
@@ -48,8 +49,14 @@ else
   ctpp2_include_args = ['-I'+ctpp2_include_path]
   if compiler.has_header('ctpp2/CTPP2Logger.hpp', args:ctpp2_include_args)
     ctpp2_include_dir = include_directories(ctpp2_include_path, is_system:true)
-    ctpp2_lib_path = ctpp2_prefix_install+'/lib'
-    ctpp2_lib = compiler.find_library('ctpp2', dirs:ctpp2_lib_path)
+    ctpp2_lib_path = join_paths(ctpp2_prefix_install, get_option('libdir'))
+    message(ctpp2_lib_path)
+    ctpp2_lib = compiler.find_library('ctpp2', dirs:ctpp2_lib_path, required:false)
+    if not ctpp2_lib.found()
+      ctpp2_lib_path = join_paths(ctpp2_prefix_install, 'lib')
+      message(ctpp2_lib_path)
+      ctpp2_lib = compiler.find_library('ctpp2', dirs:ctpp2_lib_path)
+    endif
     ctpp2_link_args = ['-L'+ctpp2_lib_path, '-lctpp2']
     if meson.is_cross_build() and host_machine.system() == 'windows'
       iconv_lib = compiler.find_library('iconv', required:false)
@@ -66,7 +73,7 @@ endif
 
 xapian_dep = dependency('xapian-core', required:false, static:static_deps)
 
-all_deps = [thread_dep, libicu_dep, libzim_dep, xapian_dep, pugixml_dep]
+all_deps = [thread_dep, libicu_dep, libzim_dep, xapian_dep, pugixml_dep, libaria2_dep]
 if has_ctpp2_dep
   all_deps += [ctpp2_dep]
 endif
@@ -82,7 +89,7 @@ subdir('scripts')
 subdir('static')
 subdir('src')
 
-pkg_requires = ['libzim', 'icu-i18n', 'pugixml']
+pkg_requires = ['libzim', 'icu-i18n', 'pugixml', 'libaria2']
 if xapian_dep.found()
     pkg_requires += ['xapian-core']
 endif

src/android/gen_kiwix.sh

@@ -4,8 +4,13 @@ set -e
 
 BUILD_PATH=$(pwd)
 
-javac -d $BUILD_PATH/src/android $1 $2 $3 $4
+echo "javac -d $BUILD_PATH/src/android $@"
+javac -d $BUILD_PATH/src/android/ "$@"
+
 
 cd $BUILD_PATH/src/android
+echo "javah -jni org.kiwix.kiwixlib"
 javah -jni org.kiwix.kiwixlib.JNIKiwix
+javah -jni org.kiwix.kiwixlib.JNIKiwixReader
+javah -jni org.kiwix.kiwixlib.JNIKiwixSearcher
 cd $BUILD_PATH

src/android/kiwix.cpp

@@ -1,5 +1,6 @@
 /*
  * Copyright (C) 2013 Emmanuel Engelhart <kelson@kiwix.org>
+ * Copyright (C) 2017 Matthieu Gautier <mgautier@kymeria.fr>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU  General Public License as published by
@@ -20,540 +21,24 @@
 #include <jni.h>
 #include "org_kiwix_kiwixlib_JNIKiwix.h"
 
-#include <stdio.h>
-#include <string.h>
-
 #include <iostream>
 #include <string>
 
-#include "common/base64.h"
-#include "reader.h"
-#include "searcher.h"
 #include "unicode/putil.h"
 
-#include <android/log.h>
-#define LOGI(...) __android_log_print(ANDROID_LOG_INFO, "kiwix", __VA_ARGS__)
+#include "utils.h"
 
-#include <xapian.h>
-#include <zim/article.h>
-#include <zim/error.h>
-#include <zim/file.h>
-#include <zim/zim.h>
-
-/* global variables */
-kiwix::Reader* reader = NULL;
-kiwix::Searcher* searcher = NULL;
-
-static pthread_mutex_t readerLock = PTHREAD_MUTEX_INITIALIZER;
-static pthread_mutex_t searcherLock = PTHREAD_MUTEX_INITIALIZER;
-
-/* c2jni type conversion functions */
-jboolean c2jni(const bool& val)
-{
-  return val ? JNI_TRUE : JNI_FALSE;
-}
-jstring c2jni(const std::string& val, JNIEnv* env)
-{
-  return env->NewStringUTF(val.c_str());
-}
-
-jint c2jni(const int val)
-{
-  return (jint)val;
-}
-jint c2jni(const unsigned val)
-{
-  return (unsigned)val;
-}
-/* jni2c type conversion functions */
-bool jni2c(const jboolean& val)
-{
-  return val == JNI_TRUE;
-}
-std::string jni2c(const jstring& val, JNIEnv* env)
-{
-  return std::string(env->GetStringUTFChars(val, 0));
-}
-
-int jni2c(const jint val)
-{
-  return (int)val;
-}
-/* Method to deal with variable passed by reference */
-void setStringObjValue(const std::string& value, const jobject obj, JNIEnv* env)
-{
-  jclass objClass = env->GetObjectClass(obj);
-  jfieldID objFid = env->GetFieldID(objClass, "value", "Ljava/lang/String;");
-  env->SetObjectField(obj, objFid, c2jni(value, env));
-}
-
-void setIntObjValue(const int value, const jobject obj, JNIEnv* env)
-{
-  jclass objClass = env->GetObjectClass(obj);
-  jfieldID objFid = env->GetFieldID(objClass, "value", "I");
-  env->SetIntField(obj, objFid, value);
-}
-
-void setBoolObjValue(const bool value, const jobject obj, JNIEnv* env)
-{
-  jclass objClass = env->GetObjectClass(obj);
-  jfieldID objFid = env->GetFieldID(objClass, "value", "Z");
-  env->SetIntField(obj, objFid, c2jni(value));
-}
-
-/* Kiwix library functions */
-JNIEXPORT jstring JNICALL
-Java_org_kiwix_kiwixlib_JNIKiwix_getMainPage(JNIEnv* env, jobject obj)
-{
-  jstring url;
-
-  pthread_mutex_lock(&readerLock);
-  if (reader != NULL) {
-    try {
-      std::string cUrl = reader->getMainPageUrl();
-      url = c2jni(cUrl, env);
-    } catch (...) {
-      std::cerr << "Unable to get ZIM main page" << std::endl;
-    }
-  }
-  pthread_mutex_unlock(&readerLock);
-
-  return url;
-}
-
-JNIEXPORT jstring JNICALL Java_org_kiwix_kiwixlib_JNIKiwix_getId(JNIEnv* env,
-                                                                 jobject obj)
-{
-  jstring id;
-
-  pthread_mutex_lock(&readerLock);
-  if (reader != NULL) {
-    try {
-      std::string cId = reader->getId();
-      id = c2jni(cId, env);
-    } catch (...) {
-      std::cerr << "Unable to get ZIM id" << std::endl;
-    }
-  }
-  pthread_mutex_unlock(&readerLock);
-
-  return id;
-}
-
-JNIEXPORT jint JNICALL Java_org_kiwix_kiwixlib_JNIKiwix_getFileSize(JNIEnv* env,
-                                                                    jobject obj)
-{
-  jint size;
-
-  pthread_mutex_lock(&readerLock);
-  if (reader != NULL) {
-    try {
-      int cSize = reader->getFileSize();
-      size = c2jni(cSize);
-    } catch (...) {
-      std::cerr << "Unable to get ZIM file size" << std::endl;
-    }
-  }
-  pthread_mutex_unlock(&readerLock);
-
-  return size;
-}
-
-JNIEXPORT jstring JNICALL
-Java_org_kiwix_kiwixlib_JNIKiwix_getCreator(JNIEnv* env, jobject obj)
-{
-  jstring creator;
-
-  pthread_mutex_lock(&readerLock);
-  if (reader != NULL) {
-    try {
-      std::string cCreator = reader->getCreator();
-      creator = c2jni(cCreator, env);
-    } catch (...) {
-      std::cerr << "Unable to get ZIM creator" << std::endl;
-    }
-  }
-  pthread_mutex_unlock(&readerLock);
-
-  return creator;
-}
-
-JNIEXPORT jstring JNICALL
-Java_org_kiwix_kiwixlib_JNIKiwix_getPublisher(JNIEnv* env, jobject obj)
-{
-  jstring publisher;
-
-  pthread_mutex_lock(&readerLock);
-  if (reader != NULL) {
-    try {
-      std::string cPublisher = reader->getPublisher();
-      publisher = c2jni(cPublisher, env);
-    } catch (...) {
-      std::cerr << "Unable to get ZIM creator" << std::endl;
-    }
-  }
-  pthread_mutex_unlock(&readerLock);
-
-  return publisher;
-}
-
-JNIEXPORT jstring JNICALL Java_org_kiwix_kiwixlib_JNIKiwix_getName(JNIEnv* env,
-                                                                   jobject obj)
-{
-  jstring name;
-
-  pthread_mutex_lock(&readerLock);
-  if (reader != NULL) {
-    try {
-      std::string cName = reader->getName();
-      name = c2jni(cName, env);
-    } catch (...) {
-      std::cerr << "Unable to get ZIM name" << std::endl;
-    }
-  }
-  pthread_mutex_unlock(&readerLock);
-
-  return name;
-}
-
-JNIEXPORT jstring JNICALL
-Java_org_kiwix_kiwixlib_JNIKiwix_getFavicon(JNIEnv* env, jobject obj)
-{
-  jstring favicon;
-
-  pthread_mutex_lock(&readerLock);
-  if (reader != NULL) {
-    try {
-      std::string cContent;
-      std::string cMime;
-      reader->getFavicon(cContent, cMime);
-      favicon
-          = c2jni(base64_encode(
-                      reinterpret_cast<const unsigned char*>(cContent.c_str()),
-                      cContent.length()),
-                  env);
-    } catch (...) {
-      std::cerr << "Unable to get ZIM favicon" << std::endl;
-    }
-  }
-  pthread_mutex_unlock(&readerLock);
-
-  return favicon;
-}
-
-JNIEXPORT jstring JNICALL Java_org_kiwix_kiwixlib_JNIKiwix_getDate(JNIEnv* env,
-                                                                   jobject obj)
-{
-  jstring date;
-
-  pthread_mutex_lock(&readerLock);
-  if (reader != NULL) {
-    try {
-      std::string cDate = reader->getDate();
-      date = c2jni(cDate, env);
-    } catch (...) {
-      std::cerr << "Unable to get ZIM date" << std::endl;
-    }
-  }
-  pthread_mutex_unlock(&readerLock);
-
-  return date;
-}
-
-JNIEXPORT jstring JNICALL
-Java_org_kiwix_kiwixlib_JNIKiwix_getLanguage(JNIEnv* env, jobject obj)
-{
-  jstring language;
-
-  pthread_mutex_lock(&readerLock);
-  if (reader != NULL) {
-    try {
-      std::string cLanguage = reader->getLanguage();
-      language = c2jni(cLanguage, env);
-    } catch (...) {
-      std::cerr << "Unable to get ZIM language" << std::endl;
-    }
-  }
-  pthread_mutex_unlock(&readerLock);
-
-  return language;
-}
-
-JNIEXPORT jstring JNICALL Java_org_kiwix_kiwixlib_JNIKiwix_getMimeType(
-    JNIEnv* env, jobject obj, jstring url)
-{
-  jstring mimeType;
-
-  pthread_mutex_lock(&readerLock);
-  if (reader != NULL) {
-    std::string cUrl = jni2c(url, env);
-    try {
-      std::string cMimeType;
-      reader->getMimeTypeByUrl(cUrl, cMimeType);
-      mimeType = c2jni(cMimeType, env);
-    } catch (...) {
-      std::cerr << "Unable to get mime-type for url " << cUrl << std::endl;
-    }
-  }
-  pthread_mutex_unlock(&readerLock);
-
-  return mimeType;
-}
-
-JNIEXPORT jboolean JNICALL
-Java_org_kiwix_kiwixlib_JNIKiwix_loadZIM(JNIEnv* env, jobject obj, jstring path)
-{
-  jboolean retVal = JNI_TRUE;
-  std::string cPath = jni2c(path, env);
-
-  pthread_mutex_lock(&readerLock);
-  try {
-    if (reader != NULL) {
-      delete reader;
-    }
-    reader = new kiwix::Reader(cPath);
-  } catch (...) {
-    std::cerr << "Unable to load ZIM " << cPath << std::endl;
-    reader = NULL;
-    retVal = JNI_FALSE;
-  }
-  pthread_mutex_unlock(&readerLock);
-
-  return retVal;
-}
-
-JNIEXPORT jbyteArray JNICALL Java_org_kiwix_kiwixlib_JNIKiwix_getContent(
-    JNIEnv* env, jobject obj, jstring url, jobject titleObj, jobject mimeTypeObj, jobject sizeObj)
-{
-  /* Default values */
-  setStringObjValue("", titleObj, env);
-  setStringObjValue("", mimeTypeObj, env);
-  setIntObjValue(0, sizeObj, env);
-  jbyteArray data = env->NewByteArray(0);
-
-  /* Retrieve the content */
-  if (reader != NULL) {
-    std::string cUrl = jni2c(url, env);
-    std::string cData;
-    std::string cTitle;
-    std::string cMimeType;
-    unsigned int cSize = 0;
-
-    pthread_mutex_lock(&readerLock);
-    try {
-      if (reader->getContentByUrl(cUrl, cData, cTitle, cSize, cMimeType)) {
-        data = env->NewByteArray(cSize);
-        env->SetByteArrayRegion(
-            data, 0, cSize, reinterpret_cast<const jbyte*>(cData.c_str()));
-        setStringObjValue(cMimeType, mimeTypeObj, env);
-        setStringObjValue(cTitle, titleObj, env);
-        setIntObjValue(cSize, sizeObj, env);
-      }
-    } catch (...) {
-      std::cerr << "Unable to get content for url " << cUrl << std::endl;
-    }
-    pthread_mutex_unlock(&readerLock);
-  }
-
-  return data;
-}
-
-JNIEXPORT jboolean JNICALL Java_org_kiwix_kiwixlib_JNIKiwix_searchSuggestions(
-    JNIEnv* env, jobject obj, jstring prefix, jint count)
-{
-  jboolean retVal = JNI_FALSE;
-  std::string cPrefix = jni2c(prefix, env);
-  unsigned int cCount = jni2c(count);
-
-  pthread_mutex_lock(&readerLock);
-  try {
-    if (reader != NULL) {
-      if (reader->searchSuggestionsSmart(cPrefix, cCount)) {
-        retVal = JNI_TRUE;
-      }
-    }
-  } catch (...) {
-    std::cerr << "Unable to search suggestions for pattern " << cPrefix
-              << std::endl;
-  }
-  pthread_mutex_unlock(&readerLock);
-
-  return retVal;
-}
-
-JNIEXPORT jboolean JNICALL Java_org_kiwix_kiwixlib_JNIKiwix_getNextSuggestion(
-    JNIEnv* env, jobject obj, jobject titleObj)
-{
-  jboolean retVal = JNI_FALSE;
-  std::string cTitle;
-
-  pthread_mutex_lock(&readerLock);
-  try {
-    if (reader != NULL) {
-      if (reader->getNextSuggestion(cTitle)) {
-        setStringObjValue(cTitle, titleObj, env);
-        retVal = JNI_TRUE;
-      }
-    }
-  } catch (...) {
-    std::cerr << "Unable to get next suggestion" << std::endl;
-  }
-  pthread_mutex_unlock(&readerLock);
-
-  return retVal;
-}
-
-JNIEXPORT jboolean JNICALL Java_org_kiwix_kiwixlib_JNIKiwix_getPageUrlFromTitle(
-    JNIEnv* env, jobject obj, jstring title, jobject urlObj)
-{
-  jboolean retVal = JNI_FALSE;
-  std::string cTitle = jni2c(title, env);
-  std::string cUrl;
-
-  pthread_mutex_lock(&readerLock);
-  try {
-    if (reader != NULL) {
-      if (reader->getPageUrlFromTitle(cTitle, cUrl)) {
-        setStringObjValue(cUrl, urlObj, env);
-        retVal = JNI_TRUE;
-      }
-    }
-  } catch (...) {
-    std::cerr << "Unable to get URL for title " << cTitle << std::endl;
-  }
-  pthread_mutex_unlock(&readerLock);
-
-  return retVal;
-}
-
-JNIEXPORT jboolean JNICALL Java_org_kiwix_kiwixlib_JNIKiwix_getTitle(
-    JNIEnv* env, jobject obj, jobject titleObj)
-{
-  jboolean retVal = JNI_FALSE;
-  std::string cTitle;
-
-  pthread_mutex_lock(&readerLock);
-  try {
-    if (reader != NULL) {
-      std::string cTitle = reader->getTitle();
-      setStringObjValue(cTitle, titleObj, env);
-      retVal = JNI_TRUE;
-    }
-  } catch (...) {
-    std::cerr << "Unable to get ZIM title" << std::endl;
-  }
-  pthread_mutex_unlock(&readerLock);
-
-  return retVal;
-}
-
-JNIEXPORT jstring JNICALL
-Java_org_kiwix_kiwixlib_JNIKiwix_getDescription(JNIEnv* env, jobject obj)
-{
-  jstring description;
-
-  pthread_mutex_lock(&readerLock);
-  if (reader != NULL) {
-    try {
-      std::string cDescription = reader->getDescription();
-      description = c2jni(cDescription, env);
-    } catch (...) {
-      std::cerr << "Unable to get ZIM description" << std::endl;
-    }
-  }
-  pthread_mutex_unlock(&readerLock);
-
-  return description;
-}
-
-JNIEXPORT jboolean JNICALL Java_org_kiwix_kiwixlib_JNIKiwix_getRandomPage(
-    JNIEnv* env, jobject obj, jobject urlObj)
-{
-  jboolean retVal = JNI_FALSE;
-  std::string cUrl;
-
-  pthread_mutex_lock(&readerLock);
-  try {
-    if (reader != NULL) {
-      std::string cUrl = reader->getRandomPageUrl();
-      setStringObjValue(cUrl, urlObj, env);
-      retVal = JNI_TRUE;
-    }
-  } catch (...) {
-    std::cerr << "Unable to get random page" << std::endl;
-  }
-  pthread_mutex_unlock(&readerLock);
-
-  return retVal;
-}
+pthread_mutex_t globalLock = PTHREAD_RECURSIVE_MUTEX_INITIALIZER;
 
 JNIEXPORT void JNICALL Java_org_kiwix_kiwixlib_JNIKiwix_setDataDirectory(
     JNIEnv* env, jobject obj, jstring dirStr)
 {
   std::string cPath = jni2c(dirStr, env);
 
-  pthread_mutex_lock(&readerLock);
+  Lock l;
   try {
     u_setDataDirectory(cPath.c_str());
   } catch (...) {
     std::cerr << "Unable to set data directory " << cPath << std::endl;
   }
-  pthread_mutex_unlock(&readerLock);
-}
-
-JNIEXPORT jboolean JNICALL Java_org_kiwix_kiwixlib_JNIKiwix_loadFulltextIndex(
-    JNIEnv* env, jobject obj, jstring path)
-{
-  jboolean retVal = JNI_TRUE;
-  std::string cPath = jni2c(path, env);
-
-  pthread_mutex_lock(&searcherLock);
-  try {
-    if (searcher != NULL) {
-      delete searcher;
-    }
-    if (!reader || !reader->hasFulltextIndex()) {
-      // Use old API (no embedded full text index).
-      searcher = new kiwix::Searcher(cPath, reader, "");
-    } else {
-      // Use the new API. We don't care about the human readable name as
-      // we don't use it (in android).
-      searcher = new kiwix::Searcher();
-      searcher->add_reader(reader, "");
-    }
-  } catch (...) {
-    searcher = NULL;
-    retVal = JNI_FALSE;
-    std::cerr << "Unable to load full text index " << cPath << std::endl;
-  }
-  pthread_mutex_unlock(&searcherLock);
-
-  return retVal;
-}
-
-JNIEXPORT jstring JNICALL Java_org_kiwix_kiwixlib_JNIKiwix_indexedQuery(
-    JNIEnv* env, jclass obj, jstring query, jint count)
-{
-  std::string cQuery = jni2c(query, env);
-  unsigned int cCount = jni2c(count);
-  kiwix::Result* p_result;
-  std::string result;
-
-  pthread_mutex_lock(&searcherLock);
-  try {
-    if (searcher != NULL) {
-      searcher->search(cQuery, 0, count);
-      while ((p_result = searcher->getNextResult())
-             && !(p_result->get_title().empty())
-             && !(p_result->get_url().empty())) {
-        result += p_result->get_title() + "\n";
-        delete p_result;
-      }
-    }
-  } catch (...) {
-    std::cerr << "Unable to make indexed query " << cQuery << std::endl;
-  }
-  pthread_mutex_unlock(&searcherLock);
-
-  return env->NewStringUTF(result.c_str());
 }

src/android/kiwixreader.cpp

@@ -0,0 +1,399 @@
+/*
+ * Copyright (C) 2013 Emmanuel Engelhart <kelson@kiwix.org>
+ * Copyright (C) 2017 Matthieu Gautier <mgautier@kymeria.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+
+#include <jni.h>
+#include <zim/file.h>
+#include "org_kiwix_kiwixlib_JNIKiwixReader.h"
+
+#include "common/base64.h"
+#include "reader.h"
+#include "utils.h"
+
+/* Kiwix Reader JNI functions */
+JNIEXPORT jlong JNICALL Java_org_kiwix_kiwixlib_JNIKiwixReader_getNativeReader(
+    JNIEnv* env, jobject obj, jstring filename)
+{
+  std::string cPath = jni2c(filename, env);
+
+  Lock l;
+  kiwix::Reader* reader = nullptr;
+  try {
+    reader = new kiwix::Reader(cPath);
+  } catch (...) {
+    std::cerr << "Unable to load ZIM " << cPath << std::endl;
+    reader = NULL;
+  }
+
+  return reinterpret_cast<jlong>(new Handle<kiwix::Reader>(reader));
+}
+
+JNIEXPORT void JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixReader_dispose(JNIEnv* env, jobject obj)
+{
+  Handle<kiwix::Reader>::dispose(env, obj);
+}
+
+#define READER (Handle<kiwix::Reader>::getHandle(env, obj))
+
+/* Kiwix library functions */
+JNIEXPORT jstring JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixReader_getMainPage(JNIEnv* env, jobject obj)
+{
+  jstring url;
+
+  try {
+    std::string cUrl = READER->getMainPage().getPath();
+    url = c2jni(cUrl, env);
+  } catch (...) {
+    std::cerr << "Unable to get ZIM main page" << std::endl;
+  }
+  return url;
+}
+
+JNIEXPORT jstring JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixReader_getId(JNIEnv* env, jobject obj)
+{
+  jstring id;
+
+  try {
+    std::string cId = READER->getId();
+    id = c2jni(cId, env);
+  } catch (...) {
+    std::cerr << "Unable to get ZIM id" << std::endl;
+  }
+
+  return id;
+}
+
+JNIEXPORT jint JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixReader_getFileSize(JNIEnv* env, jobject obj)
+{
+  jint size;
+
+  try {
+    int cSize = READER->getFileSize();
+    size = c2jni(cSize);
+  } catch (...) {
+    std::cerr << "Unable to get ZIM file size" << std::endl;
+  }
+
+  return size;
+}
+
+JNIEXPORT jstring JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixReader_getCreator(JNIEnv* env, jobject obj)
+{
+  jstring creator;
+
+  try {
+    std::string cCreator = READER->getCreator();
+    creator = c2jni(cCreator, env);
+  } catch (...) {
+    std::cerr << "Unable to get ZIM creator" << std::endl;
+  }
+
+  return creator;
+}
+
+JNIEXPORT jstring JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixReader_getPublisher(JNIEnv* env, jobject obj)
+{
+  jstring publisher;
+
+  try {
+    std::string cPublisher = READER->getPublisher();
+    publisher = c2jni(cPublisher, env);
+  } catch (...) {
+    std::cerr << "Unable to get ZIM creator" << std::endl;
+  }
+  return publisher;
+}
+
+JNIEXPORT jstring JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixReader_getName(JNIEnv* env, jobject obj)
+{
+  jstring name;
+
+  try {
+    std::string cName = READER->getName();
+    name = c2jni(cName, env);
+  } catch (...) {
+    std::cerr << "Unable to get ZIM name" << std::endl;
+  }
+  return name;
+}
+
+JNIEXPORT jstring JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixReader_getFavicon(JNIEnv* env, jobject obj)
+{
+  jstring favicon;
+
+  try {
+    std::string cContent;
+    std::string cMime;
+    READER->getFavicon(cContent, cMime);
+    favicon = c2jni(
+        base64_encode(reinterpret_cast<const unsigned char*>(cContent.c_str()),
+                      cContent.length()),
+        env);
+  } catch (...) {
+    std::cerr << "Unable to get ZIM favicon" << std::endl;
+  }
+  return favicon;
+}
+
+JNIEXPORT jstring JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixReader_getDate(JNIEnv* env, jobject obj)
+{
+  jstring date;
+
+  try {
+    std::string cDate = READER->getDate();
+    date = c2jni(cDate, env);
+  } catch (...) {
+    std::cerr << "Unable to get ZIM date" << std::endl;
+  }
+  return date;
+}
+
+JNIEXPORT jstring JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixReader_getLanguage(JNIEnv* env, jobject obj)
+{
+  jstring language;
+
+  try {
+    std::string cLanguage = READER->getLanguage();
+    language = c2jni(cLanguage, env);
+  } catch (...) {
+    std::cerr << "Unable to get ZIM language" << std::endl;
+  }
+
+  return language;
+}
+
+JNIEXPORT jstring JNICALL Java_org_kiwix_kiwixlib_JNIKiwixReader_getMimeType(
+    JNIEnv* env, jobject obj, jstring url)
+{
+  jstring mimeType;
+
+  std::string cUrl = jni2c(url, env);
+  try {
+    auto entry = READER->getEntryFromEncodedPath(cUrl);
+    auto cMimeType = entry.getMimetype();
+    mimeType = c2jni(cMimeType, env);
+  } catch (...) {
+    std::cerr << "Unable to get mime-type for url " << cUrl << std::endl;
+  }
+  return mimeType;
+}
+
+JNIEXPORT jbyteArray JNICALL Java_org_kiwix_kiwixlib_JNIKiwixReader_getContent(
+    JNIEnv* env, jobject obj, jstring url, jobject titleObj, jobject mimeTypeObj, jobject sizeObj)
+{
+  /* Default values */
+  setStringObjValue("", titleObj, env);
+  setStringObjValue("", mimeTypeObj, env);
+  setIntObjValue(0, sizeObj, env);
+  jbyteArray data = env->NewByteArray(0);
+
+  /* Retrieve the content */
+  std::string cUrl = jni2c(url, env);
+  unsigned int cSize = 0;
+
+  try {
+    auto entry = READER->getEntryFromEncodedPath(cUrl);
+    entry = entry.getFinalEntry();
+    cSize = entry.getSize();
+    setIntObjValue(cSize, sizeObj, env);
+
+    data = env->NewByteArray(cSize);
+    env->SetByteArrayRegion(
+        data, 0, cSize, reinterpret_cast<const jbyte*>(entry.getBlob().data()));
+
+    setStringObjValue(entry.getMimetype(), mimeTypeObj, env);
+    setStringObjValue(entry.getTitle(), titleObj, env);
+  } catch (...) {
+    std::cerr << "Unable to get content for url " << cUrl << std::endl;
+  }
+
+  return data;
+}
+
+JNIEXPORT jbyteArray JNICALL Java_org_kiwix_kiwixlib_JNIKiwixReader_getContentPart(
+    JNIEnv* env, jobject obj, jstring url, jint offset, jint len, jobject sizeObj)
+{
+  jbyteArray data = env->NewByteArray(0);
+  setIntObjValue(0, sizeObj, env);
+
+  /* Default values */
+  /* Retrieve the content */
+  std::string cUrl = jni2c(url, env);
+  unsigned int cOffset = jni2c(offset);
+  unsigned int cLen = jni2c(len);
+  try {
+    auto entry = READER->getEntryFromEncodedPath(cUrl);
+    entry = entry.getFinalEntry();
+
+    if (cLen == 0) {
+      setIntObjValue(entry.getSize(), sizeObj, env);
+    } else if (cOffset+cLen < entry.getSize()) {
+      auto blob = entry.getBlob(cOffset, cLen);
+      data = env->NewByteArray(cLen);
+      env->SetByteArrayRegion(
+          data, 0, cLen, reinterpret_cast<const jbyte*>(blob.data()));
+      setIntObjValue(cLen, sizeObj, env);
+    }
+  } catch (...) {
+     std::cerr << "Unable to get partial content for url " << cUrl
+               << "(" << cOffset << ":" << cLen << ")" << std::endl;
+  }
+  return data;
+}
+
+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);
+
+   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);
+  } catch (...) {
+    std::cerr << "Unable to locate direct access information for url " << cUrl
+              << std::endl;
+
+  }
+  return pair;
+}
+
+JNIEXPORT jboolean JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixReader_searchSuggestions(JNIEnv* env,
+                                                         jobject obj,
+                                                         jstring prefix,
+                                                         jint count)
+{
+  jboolean retVal = JNI_FALSE;
+  std::string cPrefix = jni2c(prefix, env);
+  unsigned int cCount = jni2c(count);
+
+  try {
+    if (READER->searchSuggestionsSmart(cPrefix, cCount)) {
+      retVal = JNI_TRUE;
+    }
+  } catch (...) {
+    std::cerr << "Unable to search suggestions for pattern " << cPrefix
+              << std::endl;
+  }
+
+  return retVal;
+}
+
+JNIEXPORT jboolean JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixReader_getNextSuggestion(JNIEnv* env,
+                                                         jobject obj,
+                                                         jobject titleObj)
+{
+  jboolean retVal = JNI_FALSE;
+  std::string cTitle;
+
+  try {
+    if (READER->getNextSuggestion(cTitle)) {
+      setStringObjValue(cTitle, titleObj, env);
+      retVal = JNI_TRUE;
+    }
+  } catch (...) {
+    std::cerr << "Unable to get next suggestion" << std::endl;
+  }
+
+  return retVal;
+}
+
+JNIEXPORT jboolean JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixReader_getPageUrlFromTitle(JNIEnv* env,
+                                                           jobject obj,
+                                                           jstring title,
+                                                           jobject urlObj)
+{
+  std::string cTitle = jni2c(title, env);
+
+  try {
+    auto entry = READER->getEntryFromTitle(cTitle);
+    entry = entry.getFinalEntry();
+    setStringObjValue(entry.getPath(), urlObj, env);
+    return JNI_TRUE;
+  } catch (...) {
+    std::cerr << "Unable to get URL for title " << cTitle << std::endl;
+  }
+
+  return JNI_FALSE;
+}
+
+JNIEXPORT jstring JNICALL Java_org_kiwix_kiwixlib_JNIKiwixReader_getTitle(
+    JNIEnv* env, jobject obj)
+{
+  jstring title;
+
+  try {
+    std::string cTitle = READER->getTitle();
+    title = c2jni(cTitle, env);
+  } catch (...) {
+    std::cerr << "Unable to get ZIM title" << std::endl;
+  }
+  return title;
+}
+
+JNIEXPORT jstring JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixReader_getDescription(JNIEnv* env, jobject obj)
+{
+  jstring description;
+
+  try {
+    std::string cDescription = READER->getDescription();
+    description = c2jni(cDescription, env);
+  } catch (...) {
+    std::cerr << "Unable to get ZIM description" << std::endl;
+  }
+  return description;
+}
+
+JNIEXPORT jboolean JNICALL Java_org_kiwix_kiwixlib_JNIKiwixReader_getRandomPage(
+    JNIEnv* env, jobject obj, jobject urlObj)
+{
+  jboolean retVal = JNI_FALSE;
+  std::string cUrl;
+
+  try {
+    std::string cUrl = READER->getRandomPage().getPath();
+    setStringObjValue(cUrl, urlObj, env);
+    retVal = JNI_TRUE;
+  } catch (...) {
+    std::cerr << "Unable to get random page" << std::endl;
+  }
+
+  return retVal;
+}

src/android/kiwixsearcher.cpp

@@ -0,0 +1,124 @@
+/*
+ * Copyright (C) 2013 Emmanuel Engelhart <kelson@kiwix.org>
+ * Copyright (C) 2017 Matthieu Gautier <mgautier@kymeria.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+
+#include <zim/file.h>
+#include "org_kiwix_kiwixlib_JNIKiwixSearcher.h"
+#include "org_kiwix_kiwixlib_JNIKiwixSearcher_Result.h"
+
+#include "reader.h"
+#include "searcher.h"
+#include "utils.h"
+
+#define SEARCHER (Handle<kiwix::Searcher>::getHandle(env, obj))
+#define RESULT (Handle<kiwix::Result>::getHandle(env, obj))
+
+
+JNIEXPORT void JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixSearcher_dispose(JNIEnv* env, jobject obj)
+{
+  Handle<kiwix::Searcher>::dispose(env, obj);
+}
+
+/* Kiwix Reader JNI functions */
+JNIEXPORT jlong JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixSearcher_getNativeHandle(JNIEnv* env,
+                                                          jobject obj)
+{
+  kiwix::Searcher* searcher = new kiwix::Searcher();
+  return reinterpret_cast<jlong>(new Handle<kiwix::Searcher>(searcher));
+}
+
+/* Kiwix library functions */
+JNIEXPORT void JNICALL Java_org_kiwix_kiwixlib_JNIKiwixSearcher_addReader(
+    JNIEnv* env, jobject obj, jobject reader)
+{
+  auto searcher = SEARCHER;
+
+  searcher->add_reader(*(Handle<kiwix::Reader>::getHandle(env, reader)), "");
+}
+
+JNIEXPORT void JNICALL Java_org_kiwix_kiwixlib_JNIKiwixSearcher_search(
+    JNIEnv* env, jobject obj, jstring query, jint count)
+{
+  std::string cquery = jni2c(query, env);
+  unsigned int ccount = jni2c(count);
+
+  SEARCHER->search(cquery, 0, ccount);
+}
+
+JNIEXPORT jobject JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixSearcher_getNextResult(JNIEnv* env,
+                                                         jobject obj)
+{
+  jobject result = nullptr;
+
+  kiwix::Result* cresult = SEARCHER->getNextResult();
+  if (cresult != nullptr) {
+    jclass resultclass
+        = env->FindClass("org/kiwix/kiwixlib/JNIKiwixSearcher$Result");
+    jmethodID ctor = env->GetMethodID(
+        resultclass, "<init>", "(Lorg/kiwix/kiwixlib/JNIKiwixSearcher;JLorg/kiwix/kiwixlib/JNIKiwixSearcher;)V");
+    result = env->NewObject(resultclass, ctor, obj, reinterpret_cast<jlong>(new Handle<kiwix::Result>(cresult)), obj);
+  }
+  return result;
+}
+
+JNIEXPORT void JNICALL Java_org_kiwix_kiwixlib_JNIKiwixSearcher_00024Result_dispose(
+    JNIEnv* env, jobject obj)
+{
+  Handle<kiwix::Result>::dispose(env, obj);
+}
+
+JNIEXPORT jstring JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixSearcher_00024Result_getUrl(JNIEnv* env,
+                                                        jobject obj)
+{
+  try {
+    return c2jni(RESULT->get_url(), env);
+  } catch (...) {
+    return nullptr;
+  }
+}
+
+JNIEXPORT jstring JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixSearcher_00024Result_getTitle(JNIEnv* env,
+                                                          jobject obj)
+{
+  try {
+    return c2jni(RESULT->get_title(), env);
+  } catch (...) {
+    return nullptr;
+  }
+}
+
+JNIEXPORT jstring JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixSearcher_00024Result_getSnippet(JNIEnv* env,
+                                                            jobject obj)
+{
+  return c2jni(RESULT->get_snippet(), env);
+}
+
+JNIEXPORT jstring JNICALL
+Java_org_kiwix_kiwixlib_JNIKiwixSearcher_00024Result_getContent(JNIEnv* env,
+                                                            jobject obj)
+{
+  return c2jni(RESULT->get_content(), env);
+}

src/android/meson.build

@@ -3,14 +3,25 @@ jni_generator = find_program('gen_kiwix.sh')
 
 kiwix_jni = custom_target('jni',
   input: ['org/kiwix/kiwixlib/JNIKiwix.java',
+          'org/kiwix/kiwixlib/JNIKiwixReader.java',
+          'org/kiwix/kiwixlib/JNIKiwixSearcher.java',
           'org/kiwix/kiwixlib/JNIKiwixInt.java',
           'org/kiwix/kiwixlib/JNIKiwixString.java',
-          'org/kiwix/kiwixlib/JNIKiwixBool.java'],
-  output: ['org_kiwix_kiwixlib_JNIKiwix.h'],
+          'org/kiwix/kiwixlib/JNIKiwixBool.java',
+          'org/kiwix/kiwixlib/JNIKiwixException.java',
+          'org/kiwix/kiwixlib/Pair.java'],
+  output: ['org_kiwix_kiwixlib_JNIKiwix.h',
+           'org_kiwix_kiwixlib_JNIKiwixReader.h',
+           'org_kiwix_kiwixlib_JNIKiwixSearcher.h',
+           'org_kiwix_kiwixlib_JNIKiwixSearcher_Result.h'],
   command:[jni_generator, '@INPUT@']
 )
 
-kiwix_sources += ['android/kiwix.cpp', kiwix_jni]
+kiwix_sources += [
+    'android/kiwix.cpp',
+    'android/kiwixreader.cpp',
+    'android/kiwixsearcher.cpp',
+    kiwix_jni]
 
 install_subdir('org', install_dir: 'kiwix-lib/java')
 install_subdir('res', install_dir: 'kiwix-lib')

src/android/org/kiwix/kiwixlib/JNIKiwix.java

@@ -1,5 +1,6 @@
 /*
  * Copyright (C) 2013 Emmanuel Engelhart <kelson@kiwix.org>
+ * Copyright (C) 2017 Matthieu Gautier <mgautier@kymeria.fr>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU  General Public License as published by
@@ -19,56 +20,12 @@
 
 package org.kiwix.kiwixlib;
 
+import org.kiwix.kiwixlib.JNIKiwixReader;
 import org.kiwix.kiwixlib.JNIKiwixString;
-import org.kiwix.kiwixlib.JNIKiwixBool;
-import org.kiwix.kiwixlib.JNIKiwixInt;
 
 public class JNIKiwix
 {
   static { System.loadLibrary("kiwix"); }
-  public native String getMainPage();
-
-  public native String getId();
-
-  public native String getLanguage();
-
-  public native String getMimeType(String url);
-
-  public native boolean loadZIM(String path);
-
-  public native boolean loadFulltextIndex(String path);
-
-  public native byte[] getContent(String url, JNIKiwixString title, JNIKiwixString mimeType, JNIKiwixInt size);
-
-  public native boolean searchSuggestions(String prefix, int count);
-
-  public native boolean getNextSuggestion(JNIKiwixString title);
-
-  public native boolean getPageUrlFromTitle(String title, JNIKiwixString url);
-
-  public native boolean getTitle(JNIKiwixString title);
-
-  public native String getDescription();
-
-  public native String getDate();
-
-  public native String getFavicon();
-
-  public native String getCreator();
-
-  public native String getPublisher();
-
-  public native String getName();
-
-  public native int getFileSize();
-
-  public native int getArticleCount();
-
-  public native int getMediaCount();
-
-  public native boolean getRandomPage(JNIKiwixString url);
 
   public native void setDataDirectory(String icuDataDir);
-
-  public static native String indexedQuery(String db, int count);
 }

src/android/org/kiwix/kiwixlib/JNIKiwixException.java

@@ -0,0 +1,27 @@
+/*
+ * Copyright (C) 2017 Matthieu Gautier <mgautier@kymeria.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+package org.kiwix.kiwixlib;
+
+public class JNIKiwixException extends Exception
+{
+  public JNIKiwixException(String message) {
+    super(message);
+  }
+}

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

@@ -0,0 +1,127 @@
+/*
+ * Copyright (C) 2013 Emmanuel Engelhart <kelson@kiwix.org>
+ * Copyright (C) 2017 Matthieu Gautier <mgautier@kymeria.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+package org.kiwix.kiwixlib;
+
+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;
+
+public class JNIKiwixReader
+{
+  public native String getMainPage();
+
+  public native String getTitle();
+
+  public native String getId();
+
+  public native String getLanguage();
+
+  public native String getMimeType(String url);
+
+  public native byte[] getContent(String url,
+                                  JNIKiwixString title,
+                                  JNIKiwixString mimeType,
+                                  JNIKiwixInt size);
+
+  /**
+   * getContentPart.
+   *
+   * Get only a part of the content of the article.
+   * Return a byte array of `len` size starting from offset `offset`.
+   * Set `size` to the number of bytes read
+   * (`len` if everything is ok, 0 in case of error).
+   * If `len` == 0, no bytes are read but `size` is set to the total size of the
+   * article.
+   */
+  public native byte[] getContentPart(String url,
+                                      int offest,
+                                      int len,
+                                      JNIKiwixInt size);
+
+  /**
+   * getDirectAccessInformation.
+   *
+   * Return information giving where the content is located in the zim file.
+   *
+   * Some contents (binary content) are stored uncompressed in the zim file.
+   * Knowing this information, it could be interesting to directly open
+   * 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.
+   *
+   * 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 boolean searchSuggestions(String prefix, int count);
+
+  public native boolean getNextSuggestion(JNIKiwixString title);
+
+  public native boolean getPageUrlFromTitle(String title, JNIKiwixString url);
+
+  public native String getDescription();
+
+  public native String getDate();
+
+  public native String getFavicon();
+
+  public native String getCreator();
+
+  public native String getPublisher();
+
+  public native String getName();
+
+  public native int getFileSize();
+
+  public native int getArticleCount();
+
+  public native int getMediaCount();
+
+  public native boolean getRandomPage(JNIKiwixString url);
+
+  public JNIKiwixSearcher search(String query, int count)
+  {
+    JNIKiwixSearcher searcher = new JNIKiwixSearcher();
+    searcher.addKiwixReader(this);
+    searcher.search(query, count);
+    return searcher;
+  }
+
+  public JNIKiwixReader(String filename) throws JNIKiwixException
+  {
+    nativeHandle = getNativeReader(filename);
+    if (nativeHandle == 0) {
+        throw new JNIKiwixException("Cannot open zimfile "+filename);
+    }
+  }
+  public JNIKiwixReader() {
+
+  }
+  public native void dispose();
+
+  private native long getNativeReader(String filename);
+  private long nativeHandle;
+}

src/android/org/kiwix/kiwixlib/JNIKiwixSearcher.java

@@ -0,0 +1,67 @@
+/*
+ * Copyright (C) 2013 Emmanuel Engelhart <kelson@kiwix.org>
+ * Copyright (C) 2017 Matthieu Gautier <mgautier@kymeria.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+
+package org.kiwix.kiwixlib;
+
+import org.kiwix.kiwixlib.JNIKiwixReader;
+import java.util.Vector;
+
+public class JNIKiwixSearcher
+{
+  public class Result
+  {
+    private long nativeHandle;
+    private JNIKiwixSearcher searcher;
+    public Result(long handle, JNIKiwixSearcher _searcher)
+    {
+      nativeHandle = handle;
+      searcher = _searcher;
+    }
+    public native String getUrl();
+    public native String getTitle();
+    public native String getContent();
+    public native String getSnippet();
+    public native void dispose();
+  }
+
+  public JNIKiwixSearcher()
+  {
+    nativeHandle = getNativeHandle();
+    usedReaders = new Vector();
+  }
+  public native void dispose();
+
+  private native long getNativeHandle();
+  private long nativeHandle;
+  private Vector usedReaders;
+
+  public native void addReader(JNIKiwixReader reader);
+  public void addKiwixReader(JNIKiwixReader reader)
+  {
+    addReader(reader);
+    usedReaders.addElement(reader);
+  };
+
+  public native void search(String query, int count);
+
+  public native Result getNextResult();
+  public native boolean hasMoreResult();
+}

src/android/org/kiwix/kiwixlib/Pair.java

@@ -0,0 +1,26 @@
+/*
+ * Copyright (C) 2017 Matthieu Gautier <mgautier@kymeria.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+package org.kiwix.kiwixlib;
+
+public class Pair
+{
+  public String filename;
+  public int offset;
+}

src/android/utils.h

@@ -0,0 +1,150 @@
+/*
+ * Copyright (C) 2013 Emmanuel Engelhart <kelson@kiwix.org>
+ * Copyright (C) 2017 Matthieu Gautier <mgautier@kymeria.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+
+#ifndef _ANDROID_JNI_UTILS_H
+#define _ANDROID_JNI_UTILS_H
+
+#include <jni.h>
+
+#include <pthread.h>
+#include <string>
+
+extern pthread_mutex_t globalLock;
+
+inline jfieldID getHandleField(JNIEnv* env, jobject obj)
+{
+  jclass c = env->GetObjectClass(obj);
+  // J is the type signature for long:
+  return env->GetFieldID(c, "nativeHandle", "J");
+}
+
+class Lock
+{
+ protected:
+  pthread_mutex_t* lock;
+
+ public:
+  Lock() : lock(&globalLock) { pthread_mutex_lock(lock); }
+  Lock(const Lock&) = delete;
+  Lock& operator=(const Lock&) = delete;
+  Lock(Lock&& other) : lock(&globalLock) { other.lock = nullptr; }
+  virtual ~Lock()
+  {
+    if (lock) {
+      pthread_mutex_unlock(lock);
+    }
+  }
+};
+
+template <class T>
+class LockedHandle;
+
+template <class T>
+class Handle
+{
+ protected:
+  T* h;
+
+ public:
+  Handle(T* h) : h(h){};
+
+  // No destructor. This must and will be handled by dispose method.
+
+  static LockedHandle<T> getHandle(JNIEnv* env, jobject obj)
+  {
+    jlong handle = env->GetLongField(obj, getHandleField(env, obj));
+    return LockedHandle<T>(reinterpret_cast<Handle<T>*>(handle));
+  }
+
+  static void dispose(JNIEnv* env, jobject obj)
+  {
+    auto lHandle = getHandle(env, obj);
+    auto handle = lHandle.h;
+    delete handle->h;
+    delete handle;
+  }
+  friend class LockedHandle<T>;
+};
+
+template <class T>
+struct LockedHandle : public Lock {
+  Handle<T>* h;
+  LockedHandle(Handle<T>* h) : h(h) {}
+  T* operator->() { return h->h; }
+  T* operator*() { return h->h; }
+  operator bool() const { return (h->h != nullptr); }
+};
+
+/* c2jni type conversion functions */
+inline jboolean c2jni(const bool& val) { return val ? JNI_TRUE : JNI_FALSE; }
+inline jstring c2jni(const std::string& val, JNIEnv* env)
+{
+  return env->NewStringUTF(val.c_str());
+}
+
+inline jint c2jni(const int val) { return (jint)val; }
+inline jint c2jni(const unsigned val) { return (unsigned)val; }
+/* jni2c type conversion functions */
+inline bool jni2c(const jboolean& val) { return val == JNI_TRUE; }
+inline std::string jni2c(const jstring& val, JNIEnv* env)
+{
+  const char* chars = env->GetStringUTFChars(val, 0);
+  std::string ret(chars);
+  env->ReleaseStringUTFChars(val, chars);
+  return ret;
+}
+
+inline int jni2c(const jint val) { return (int)val; }
+/* Method to deal with variable passed by reference */
+inline void setStringObjValue(const std::string& value,
+                              const jobject obj,
+                              JNIEnv* env)
+{
+  jclass objClass = env->GetObjectClass(obj);
+  jfieldID objFid = env->GetFieldID(objClass, "value", "Ljava/lang/String;");
+  env->SetObjectField(obj, objFid, c2jni(value, env));
+}
+
+inline void setIntObjValue(const int value, const jobject obj, JNIEnv* env)
+{
+  jclass objClass = env->GetObjectClass(obj);
+  jfieldID objFid = env->GetFieldID(objClass, "value", "I");
+  env->SetIntField(obj, objFid, value);
+}
+
+inline void setBoolObjValue(const bool value, const jobject obj, JNIEnv* env)
+{
+  jclass objClass = env->GetObjectClass(obj);
+  jfieldID objFid = env->GetFieldID(objClass, "value", "Z");
+  env->SetIntField(obj, objFid, c2jni(value));
+}
+
+inline void setPairObjValue(const std::string& filename, const int offset,
+                            const jobject obj, JNIEnv* env)
+{
+  jclass objClass = env->GetObjectClass(obj);
+  jfieldID filenameFid = env->GetFieldID(objClass, "filename", "Ljava/lang/String;");
+  env->SetObjectField(obj, filenameFid, c2jni(filename, env));
+  jfieldID offsetFid = env->GetFieldID(objClass, "offset", "I");
+  env->SetIntField(obj, offsetFid, offset);
+}
+
+#endif // _ANDROID_JNI_UTILS_H

src/common/networkTools.cpp

@@ -19,13 +19,15 @@
 
 #include <common/networkTools.h>
 
+
+
 std::map<std::string, std::string> kiwix::getNetworkInterfaces()
 {
   std::map<std::string, std::string> interfaces;
 
 #ifdef _WIN32
   SOCKET sd = WSASocket(AF_INET, SOCK_DGRAM, 0, 0, 0, 0);
-  if (sd == SOCKET_ERROR) {
+  if (sd == (SOCKET)SOCKET_ERROR) {
     std::cerr << "Failed to get a socket. Error " << WSAGetLastError()
               << std::endl;
     return interfaces;

src/common/pathTools.cpp

@@ -188,6 +188,20 @@ string getFileSizeAsString(const string& path)
   return convert.str();
 }
 
+string getFileContent(const string& path)
+{
+  std::ifstream f(path, std::ios::in|std::ios::ate);
+  std::string content;
+  if (f.is_open()) {
+    auto size = f.tellg();
+    content.reserve(size);
+    f.seekg(0, std::ios::beg);
+    content.assign((std::istreambuf_iterator<char>(f)),
+                    std::istreambuf_iterator<char>());
+  }
+  return content;
+}
+
 bool fileExists(const string& path)
 {
 #ifdef _WIN32
@@ -214,6 +228,30 @@ bool makeDirectory(const string& path)
   return status == 0;
 }
 
+string makeTmpDirectory()
+{
+#ifdef _WIN32
+  char cbase[MAX_PATH+1];
+  int base_len = GetTempPath(MAX_PATH+1, cbase);
+  UUID uuid;
+  UuidCreate(&uuid);
+  char* dir_name;
+  UuidToString(&uuid, reinterpret_cast<unsigned char**>(&dir_name));
+  string dir(cbase, base_len);
+  dir += dir_name;
+  _mkdir(dir.c_str());
+  RpcStringFree(reinterpret_cast<unsigned char**>(&dir_name));
+#else
+  string base = "/tmp";
+  auto _template = base + "/kiwix-lib_XXXXXX";
+  char* _template_array = new char[_template.size()+1];
+  memcpy(_template_array, _template.c_str(), _template.size());
+  string dir = mkdtemp(_template_array);
+  delete[] _template_array;
+#endif
+  return dir;
+}
+
 /* Try to create a link and if does not work then make a copy */
 bool copyFile(const string& sourcePath, const string& destPath)
 {

src/common/stringTools.cpp

@@ -133,73 +133,130 @@ std::string kiwix::encodeDiples(const std::string& str)
   return result;
 }
 
-// Urlencode
-// based on javascript encodeURIComponent()
-
-std::string char2hex(char dec)
-{
-  char dig1 = (dec & 0xF0) >> 4;
-  char dig2 = (dec & 0x0F);
-  if (0 <= dig1 && dig1 <= 9) {
-    dig1 += 48;  // 0,48inascii
-  }
-  if (10 <= dig1 && dig1 <= 15) {
-    dig1 += 97 - 10;  // a,97inascii
-  }
-  if (0 <= dig2 && dig2 <= 9) {
-    dig2 += 48;
-  }
-  if (10 <= dig2 && dig2 <= 15) {
-    dig2 += 97 - 10;
-  }
-
-  std::string r;
-  r.append(&dig1, 1);
-  r.append(&dig2, 1);
-  return r;
-}
-
-std::string kiwix::urlEncode(const std::string& c)
-{
-  std::string escaped = "";
-  int max = c.length();
-  for (int i = 0; i < max; i++) {
-    if ((48 <= c[i] && c[i] <= 57) ||  // 0-9
-        (65 <= c[i] && c[i] <= 90)
-        ||  // abc...xyz
-        (97 <= c[i] && c[i] <= 122)
-        ||  // ABC...XYZ
-        (c[i] == '~' || c[i] == '!' || c[i] == '*' || c[i] == '(' || c[i] == ')'
-         || c[i] == '\'')) {
-      escaped.append(&c[i], 1);
-    } else {
-      escaped.append("%");
-      escaped.append(char2hex(c[i]));  // converts char 255 to string "ff"
-    }
-  }
-  return escaped;
-}
-
 #endif
 
-static char charFromHex(std::string a)
+/* urlEncode() based on javascript encodeURI() &
+   encodeURIComponent(). Mostly code from rstudio/httpuv (GPLv3) */
+
+bool isReservedUrlChar(char c)
 {
-  std::istringstream Blat(a);
-  int Z;
-  Blat >> std::hex >> Z;
-  return char(Z);
+  switch (c) {
+  case ';':
+  case ',':
+  case '/':
+  case '?':
+  case ':':
+  case '@':
+  case '&':
+  case '=':
+  case '+':
+  case '$':
+    return true;
+  default:
+    return false;
+  }
 }
 
-std::string kiwix::urlDecode(const std::string& originalUrl)
+bool needsEscape(char c, bool encodeReserved)
 {
-  std::string url = originalUrl;
-  std::string::size_type pos = 0;
-  while ((pos = url.find('%', pos)) != std::string::npos
-         && pos + 2 < url.length()) {
-    url.replace(pos, 3, 1, charFromHex(url.substr(pos + 1, 2)));
-    ++pos;
+  if (c >= 'a' && c <= 'z')
+    return false;
+  if (c >= 'A' && c <= 'Z')
+    return false;
+  if (c >= '0' && c <= '9')
+    return false;
+  if (isReservedUrlChar(c))
+    return encodeReserved;
+  switch (c) {
+  case '-':
+  case '_':
+  case '.':
+  case '!':
+  case '~':
+  case '*':
+  case '\'':
+  case '(':
+  case ')':
+    return false;
   }
-  return url;
+  return true;
+}
+
+int hexToInt(char c) {
+  switch (c) {
+  case '0': return 0;
+  case '1': return 1;
+  case '2': return 2;
+  case '3': return 3;
+  case '4': return 4;
+  case '5': return 5;
+  case '6': return 6;
+  case '7': return 7;
+  case '8': return 8;
+  case '9': return 9;
+  case 'A': case 'a': return 10;
+  case 'B': case 'b': return 11;
+  case 'C': case 'c': return 12;
+  case 'D': case 'd': return 13;
+  case 'E': case 'e': return 14;
+  case 'F': case 'f': return 15;
+  default: return -1;
+  }
+}
+
+std::string kiwix::urlEncode(const std::string& value, bool encodeReserved)
+{
+  std::ostringstream os;
+  os << std::hex << std::uppercase;
+  for (std::string::const_iterator it = value.begin();
+       it != value.end();
+       it++) {
+
+    if (!needsEscape(*it, encodeReserved)) {
+      os << *it;
+    } else {
+      os << '%' << std::setw(2) << static_cast<unsigned int>(static_cast<unsigned char>(*it));
+    }
+  }
+  return os.str();
+}
+
+std::string kiwix::urlDecode(const std::string& value, bool component)
+{
+  std::ostringstream os;
+  for (std::string::const_iterator it = value.begin();
+       it != value.end();
+       it++) {
+
+    // If there aren't enough characters left for this to be a
+    // valid escape code, just use the character and move on
+    if (it > value.end() - 3) {
+      os << *it;
+      continue;
+    }
+
+    if (*it == '%') {
+      char hi = *(++it);
+      char lo = *(++it);
+      int iHi = hexToInt(hi);
+      int iLo = hexToInt(lo);
+      if (iHi < 0 || iLo < 0) {
+	// Invalid escape sequence
+	os << '%' << hi << lo;
+	continue;
+      }
+      char c = (char)(iHi << 4 | iLo);
+      if (!component && isReservedUrlChar(c)) {
+	os << '%' << hi << lo;
+      } else {
+	os << c;
+      }
+    } else {
+      os << *it;
+    }
+  }
+
+  return os.str();
 }
 
 /* Split string in a token array */

src/downloader.cpp

@@ -0,0 +1,112 @@
+/*
+ * Copyright 2018 Matthieu Gautier <mgautier@kymeria.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#include "downloader.h"
+#include "common/pathTools.h"
+
+#include <unistd.h>
+#include <iostream>
+
+namespace kiwix
+{
+
+pthread_mutex_t Downloader::globalLock = PTHREAD_MUTEX_INITIALIZER;
+
+
+/* Constructor */
+Downloader::Downloader()
+{
+  aria2::SessionConfig config;
+  config.downloadEventCallback = Downloader::downloadEventCallback;
+  config.userData = this;
+  tmpDir = makeTmpDirectory();
+  aria2::KeyVals options;
+  options.push_back(std::pair<std::string, std::string>("dir", tmpDir));
+  session = aria2::sessionNew(options, config);
+}
+
+
+/* Destructor */
+Downloader::~Downloader()
+{
+  aria2::sessionFinal(session);
+  rmdir(tmpDir.c_str());
+}
+
+
+int Downloader::downloadEventCallback(aria2::Session* session,
+                                      aria2::DownloadEvent event,
+                                      aria2::A2Gid gid,
+                                      void* userData)
+{
+  Downloader* downloader = static_cast<Downloader*>(userData);
+
+  auto fileHandle = downloader->fileHandle;
+  auto dh = aria2::getDownloadHandle(session, gid);
+
+  if (!dh) {
+    return 0;
+  }
+
+  switch (event) {
+    case aria2::EVENT_ON_DOWNLOAD_COMPLETE:
+      {
+        if (dh->getNumFiles() > 0) {
+          auto f = dh->getFile(1);
+          fileHandle->path = f.path;
+          fileHandle->success = true;
+        }
+      }
+      break;
+    case aria2::EVENT_ON_DOWNLOAD_ERROR:
+      {
+        fileHandle->success = false;
+      }
+      break;
+    default:
+      break;
+  }
+  aria2::deleteDownloadHandle(dh);
+  return 0;
+}
+
+DownloadedFile Downloader::download(const std::string& url) {
+  pthread_mutex_lock(&globalLock);
+  DownloadedFile fileHandle;
+  try {
+    std::vector<std::string> uris = {url};
+    aria2::KeyVals options;
+    aria2::A2Gid gid;
+    int ret;
+    DownloadedFile fileHandle;
+    
+    ret = aria2::addUri(session, &gid, uris, options);
+    if (ret < 0) {
+      std::cerr << "Failed to download" << std::endl;
+    } else {
+      this->fileHandle = &fileHandle;
+      aria2::run(session, aria2::RUN_DEFAULT);
+    }
+  } catch (...) {};
+  this->fileHandle = nullptr;
+  pthread_mutex_unlock(&globalLock);
+  return fileHandle;
+}
+
+}

src/entry.cpp

@@ -0,0 +1,138 @@
+/*
+ * Copyright 2011 Emmanuel Engelhart <kelson@kiwix.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#include "reader.h"
+#include <time.h>
+
+#include <zim/search.h>
+
+namespace kiwix
+{
+
+Entry::Entry(zim::Article article)
+  : article(article)
+{
+}
+
+#define RETURN_IF_INVALID(WHAT) if(!good()) { return (WHAT); }
+
+std::string Entry::getPath() const
+{
+  RETURN_IF_INVALID("");
+  return article.getLongUrl();
+}
+
+std::string Entry::getTitle() const
+{
+  RETURN_IF_INVALID("");
+  return article.getTitle();
+}
+
+std::string Entry::getContent() const
+{
+  RETURN_IF_INVALID("");
+  return article.getData();
+}
+
+zim::Blob Entry::getBlob(offset_type offset) const
+{
+  RETURN_IF_INVALID(zim::Blob());
+  return article.getData(offset);
+}
+
+zim::Blob Entry::getBlob(offset_type offset, size_type size) const
+{
+  RETURN_IF_INVALID(zim::Blob());
+  return article.getData(offset, size);
+}
+
+std::pair<std::string, offset_type> Entry::getDirectAccessInfo() const
+{
+  RETURN_IF_INVALID(std::make_pair("", 0));
+  return article.getDirectAccessInformation();
+}
+
+size_type Entry::getSize() const
+{
+  RETURN_IF_INVALID(0);
+  return article.getArticleSize();
+}
+
+std::string Entry::getMimetype() const
+{
+  RETURN_IF_INVALID("");
+  try {
+    return article.getMimeType();
+  } catch (exception& e) {
+    return "application/octet-stream";
+  }
+}
+
+bool Entry::isRedirect() const
+{
+  RETURN_IF_INVALID(false);
+  return article.isRedirect();
+}
+
+bool Entry::isLinkTarget() const
+{
+  RETURN_IF_INVALID(false);
+  return article.isLinktarget();
+}
+
+bool Entry::isDeleted() const
+{
+  RETURN_IF_INVALID(false);
+  return article.isDeleted();
+}
+
+Entry Entry::getRedirectEntry() const
+{
+  RETURN_IF_INVALID(Entry());
+  if ( !article.isRedirect() ) {
+    throw NoEntry();
+  }
+
+  auto targeted_article = article.getRedirectArticle();
+  if ( !targeted_article.good()) {
+    throw NoEntry();
+  }
+  return targeted_article;
+}
+
+Entry Entry::getFinalEntry() const
+{
+  RETURN_IF_INVALID(Entry());
+  if (final_article.good()) {
+    return final_article;
+  }
+
+  int loopCounter = 42;
+  final_article = article;
+  while (final_article.isRedirect() && loopCounter--) {
+    final_article = final_article.getRedirectArticle();
+    if ( !final_article.good()) {
+      throw NoEntry();
+    }
+  }
+
+  return final_article;
+}
+
+}

src/manager.cpp

@@ -18,6 +18,7 @@
  */
 
 #include "manager.h"
+#include "downloader.h"
 
 namespace kiwix
 {
@@ -88,7 +89,7 @@ bool Manager::parseXmlDom(const pugi::xml_document& doc,
   return true;
 }
 
-bool Manager::readXml(const string xml,
+bool Manager::readXml(const string& xml,
                       const bool readOnly,
                       const string libraryPath)
 {
@@ -103,6 +104,67 @@ bool Manager::readXml(const string xml,
   return true;
 }
 
+
+
+bool Manager::parseOpdsDom(const pugi::xml_document& doc, const std::string& urlHost)
+{
+  pugi::xml_node libraryNode = doc.child("feed");
+
+  for (pugi::xml_node entryNode = libraryNode.child("entry"); entryNode;
+       entryNode = entryNode.next_sibling("entry")) {
+    kiwix::Book book;
+
+    book.readOnly = false;
+    book.id = entryNode.child("id").child_value();
+    book.title = entryNode.child("title").child_value();
+    book.description = entryNode.child("summary").child_value();
+    book.language = entryNode.child("language").child_value();
+    book.date = entryNode.child("updated").child_value();
+    book.creator = entryNode.child("author").child("name").child_value();
+    for(pugi::xml_node linkNode = entryNode.child("link"); linkNode;
+        linkNode = linkNode.next_sibling("link")) {
+       std::string rel = linkNode.attribute("rel").value();
+
+       if (rel == "http://opds-spec.org/image/thumbnail") {
+         auto faviconUrl = urlHost + linkNode.attribute("href").value();
+         auto downloader = Downloader();
+         auto fileHandle = downloader.download(faviconUrl);
+         if (fileHandle.success) {
+           auto content = getFileContent(fileHandle.path);
+           book.favicon = base64_encode((const unsigned char*)content.data(), content.size());
+           book.faviconMimeType = linkNode.attribute("type").value();
+         } else {
+           std::cerr << "Cannot get favicon content from " << faviconUrl << std::endl;
+         }
+
+       } else if (rel == "http://opds-spec.org/acquisition/open-access") {
+         book.url = linkNode.attribute("href").value();
+       }
+    }
+
+    /* Update the book properties with the new importer */
+    library.addBook(book);
+  }
+
+  return true;
+}
+
+
+
+bool Manager::readOpds(const string& content, const std::string& urlHost)
+{
+  pugi::xml_document doc;
+  pugi::xml_parse_result result
+      = doc.load_buffer_inplace((void*)content.data(), content.size());
+
+  if (result) {
+    this->parseOpdsDom(doc, urlHost);
+    return true;
+  }
+
+  return false;
+}
+
 bool Manager::readFile(const string path, const bool readOnly)
 {
   return this->readFile(path, path, readOnly);
@@ -231,6 +293,7 @@ bool Manager::writeFile(const string path)
   return true;
 }
 
+
 bool Manager::setCurrentBookId(const string id)
 {
   if (library.current.empty() || library.current.top() != id) {
@@ -243,7 +306,7 @@ bool Manager::setCurrentBookId(const string id)
   return true;
 }
 
-string Manager::getCurrentBookId()
+string Manager::getCurrentBookId() const
 {
   return library.current.empty() ? "" : library.current.top();
 }
@@ -495,11 +558,6 @@ bool Manager::setBookIndex(const string id,
   return false;
 }
 
-bool Manager::setBookIndex(const string id, const string path)
-{
-  return this->setBookIndex(id, path, XAPIAN);
-}
-
 bool Manager::setBookPath(const string id, const string path)
 {
   std::vector<kiwix::Book>::iterator itr;
@@ -630,6 +688,24 @@ bool Manager::listBooks(const supportedListMode mode,
   return true;
 }
 
+
+Library Manager::filter(const std::string& search) {
+  Library library;
+
+  if (search.empty()) {
+    return library;
+  }
+
+  for(auto book:this->library.books) {
+     if (matchRegex(book.title, "\\Q" + search + "\\E")
+         || matchRegex(book.description, "\\Q" + search + "\\E")) {
+       library.addBook(book);
+     }
+  }
+
+  return library;
+}
+
 void Manager::checkAndCleanBookPaths(Book& book, const string& libraryPath)
 {
   if (!book.path.empty()) {

src/meson.build

@@ -1,7 +1,10 @@
 kiwix_sources = [
   'library.cpp',
   'manager.cpp',
+  'opds_dumper.cpp',
+  'downloader.cpp',
   'reader.cpp',
+  'entry.cpp',
   'searcher.cpp',
   'common/base64.cpp',
   'common/pathTools.cpp',
@@ -41,4 +44,5 @@ kiwixlib = library('kiwix',
                    dependencies : all_deps,
                    version: meson.project_version(),
                    install: true,
-                   install_dir: install_dir)
+                   install_dir: install_dir,
+                   install_rpath: '$ORIGIN')

src/opds_dumper.cpp

@@ -0,0 +1,135 @@
+/*
+ * Copyright 2017 Matthieu Gautier <mgautier@kymeria.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU  General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#include "opds_dumper.h"
+
+namespace kiwix
+{
+/* Constructor */
+OPDSDumper::OPDSDumper(Library library)
+  : library(library)
+{
+}
+/* Destructor */
+OPDSDumper::~OPDSDumper()
+{
+}
+
+struct xml_string_writer: pugi::xml_writer
+{
+    std::string result;
+
+    virtual void write(const void* data, size_t size)
+    {
+        result.append(static_cast<const char*>(data), size);
+    }
+};
+
+std::string node_to_string(pugi::xml_node node)
+{
+    xml_string_writer writer;
+    node.print(writer, "  ");
+
+    return writer.result;
+}
+
+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();
+}
+
+#define ADD_TEXT_ENTRY(node, child, value) (node).append_child((child)).append_child(pugi::node_pcdata).set_value((value).c_str())
+
+pugi::xml_node OPDSDumper::handleBook(Book book, pugi::xml_node root_node) {
+  auto entry_node = root_node.append_child("entry");
+  ADD_TEXT_ENTRY(entry_node, "title", book.title);
+  ADD_TEXT_ENTRY(entry_node, "id", "urn:uuid:"+book.id);
+  ADD_TEXT_ENTRY(entry_node, "icon", rootLocation + "/meta?name=favicon&content=" + book.getHumanReadableIdFromPath());
+  ADD_TEXT_ENTRY(entry_node, "updated", date);
+  ADD_TEXT_ENTRY(entry_node, "summary", book.description);
+
+  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.creator);
+
+  if (! book.url.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.url.c_str();
+  }
+
+  if (! book.faviconMimeType.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.faviconMimeType.c_str();
+    image_link.append_attribute("href") = (rootLocation + "/meta?name=favicon&content=" + book.getHumanReadableIdFromPath()).c_str();
+  }
+  return entry_node;
+}
+
+string OPDSDumper::dumpOPDSFeed()
+{
+  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";
+
+  ADD_TEXT_ENTRY(root_node, "id", id);
+
+  ADD_TEXT_ENTRY(root_node, "title", title);
+  ADD_TEXT_ENTRY(root_node, "updated", date);
+
+  auto self_link_node = root_node.append_child("link");
+  self_link_node.append_attribute("rel") = "self";
+  self_link_node.append_attribute("href") = "";
+  self_link_node.append_attribute("type") = "application/atom+xml";
+
+
+  if (!searchDescriptionUrl.empty() ) {
+    auto search_link = root_node.append_child("link");
+    search_link.append_attribute("rel") = "search";
+    search_link.append_attribute("type") = "application/opensearchdescription+xml";
+    search_link.append_attribute("href") = searchDescriptionUrl.c_str();
+  }
+
+  for (auto book: library.books) {
+    handleBook(book, root_node);
+  }
+
+  return node_to_string(root_node);
+}
+
+}

src/reader.cpp

@@ -80,7 +80,6 @@ Reader::Reader(const string zimFilePath) : zimFileHandler(NULL)
     this->firstArticleOffset
         = this->zimFileHandler->getNamespaceBeginOffset('A');
     this->lastArticleOffset = this->zimFileHandler->getNamespaceEndOffset('A');
-    this->currentArticleOffset = this->firstArticleOffset;
     this->nsACount = this->zimFileHandler->getNamespaceCount('A');
     this->nsICount = this->zimFileHandler->getNamespaceCount('I');
     this->zimFilePath = zimFilePath;
@@ -102,11 +101,6 @@ zim::File* Reader::getZimFileHandler() const
 {
   return this->zimFileHandler;
 }
-/* Reset the cursor for GetNextArticle() */
-void Reader::reset()
-{
-  this->currentArticleOffset = this->firstArticleOffset;
-}
 std::map<const std::string, unsigned int> Reader::parseCounterMetadata() const
 {
   std::map<const std::string, unsigned int> counters;
@@ -196,79 +190,88 @@ string Reader::getId() const
 /* Return a page url from a title */
 bool Reader::getPageUrlFromTitle(const string& title, string& url) const
 {
-  /* Extract the content from the zim file */
-  zim::Article article = this->zimFileHandler->getArticleByTitle('A', title);
-
-  if (!article.good()) {
+  try {
+    auto entry = getEntryFromTitle(title);
+    entry = entry.getFinalEntry();
+    url = entry.getPath();
+    return true;
+  } catch (NoEntry& e) {
     return false;
   }
-
-  unsigned int loopCounter = 0;
-  while (article.isRedirect() && loopCounter++ < 42) {
-    article = article.getRedirectArticle();
-  }
-
-  url = article.getLongUrl();
-  return true;
 }
 
 /* Return an URL from a title */
 string Reader::getRandomPageUrl() const
 {
+  return getRandomPage().getPath();
+}
+
+Entry Reader::getRandomPage() const
+{
+  if (!this->zimFileHandler) {
+    throw NoEntry();
+  }
+
   zim::Article article;
-  zim::size_type idx;
-  std::string mainPageUrl = this->getMainPageUrl();
+  std::string mainPagePath = this->getMainPage().getPath();
+  int watchdog = 42;
 
   do {
-    idx = this->firstArticleOffset
+    auto idx = this->firstArticleOffset
           + (zim::size_type)((double)rand() / ((double)RAND_MAX + 1)
                              * this->nsACount);
     article = zimFileHandler->getArticle(idx);
-  } while (article.getLongUrl() == mainPageUrl);
+    if (!watchdog--) {
+      throw NoEntry();
+    }
+  } while (!article.good() && article.getLongUrl() == mainPagePath);
 
-  return article.getLongUrl();
+  return article;
 }
 
 /* Return the welcome page URL */
 string Reader::getMainPageUrl() const
 {
-  string url = "";
+  return getMainPage().getPath();
+}
 
-  if (this->zimFileHandler->getFileheader().hasMainPage()) {
-    zim::Article article = zimFileHandler->getArticle(
-        this->zimFileHandler->getFileheader().getMainPage());
-    url = article.getLongUrl();
-
-    if (url.empty()) {
-      url = getFirstPageUrl();
-    }
-  } else {
-    url = getFirstPageUrl();
+Entry Reader::getMainPage() const
+{
+  if (!this->zimFileHandler) {
+    throw NoEntry();
   }
 
-  return url;
+  string url = "";
+
+  zim::Article article;
+  if (this->zimFileHandler->getFileheader().hasMainPage())
+  {
+    article = zimFileHandler->getArticle(
+        this->zimFileHandler->getFileheader().getMainPage());
+  }
+
+  if (!article.good())
+  {
+    return getFirstPage();
+  }
+
+  return article;
 }
 
 bool Reader::getFavicon(string& content, string& mimeType) const
 {
-  unsigned int contentLength = 0;
-  string title;
+  static const char* const paths[] = {"-/favicon.png", "I/favicon.png", "I/favicon", "-/favicon"};
 
-  this->getContentByUrl("/-/favicon.png", content, title, contentLength, mimeType);
-
-  if (content.empty()) {
-    this->getContentByUrl("/I/favicon.png", content, title, contentLength, mimeType);
-
-    if (content.empty()) {
-      this->getContentByUrl("/I/favicon", content, title, contentLength, mimeType);
-
-      if (content.empty()) {
-        this->getContentByUrl("/-/favicon", content, title, contentLength, mimeType);
-      }
-    }
+  for (auto &path: paths) {
+    try {
+      auto entry = getEntryFromPath(path);
+      content = entry.getContent();
+      mimeType = entry.getMimetype();
+      return true;
+    } catch(NoEntry& e) {};
   }
 
-  return content.empty() ? false : true;
+  return false;
 }
 
 string Reader::getZimFilePath() const
@@ -278,11 +281,13 @@ string Reader::getZimFilePath() const
 /* Return a metatag value */
 bool Reader::getMetatag(const string& name, string& value) const
 {
-  unsigned int contentLength = 0;
-  string contentType = "";
-  string title;
-
-  return this->getContentByUrl("/M/" + name, value, title, contentLength, contentType);
+  try {
+    auto entry = getEntryFromPath("M/"+name);
+    value = entry.getContent();
+    return true;
+  } catch(NoEntry& e) {
+    return false;
+  }
 }
 
 string Reader::getTitle() const
@@ -381,12 +386,26 @@ string Reader::getOrigId() const
 /* Return the first page URL */
 string Reader::getFirstPageUrl() const
 {
-  zim::size_type firstPageOffset = zimFileHandler->getNamespaceBeginOffset('A');
-  zim::Article article = zimFileHandler->getArticle(firstPageOffset);
-  return article.getLongUrl();
+  return getFirstPage().getPath();
 }
 
-bool Reader::parseUrl(const string& url, char* ns, string& title) const
+Entry Reader::getFirstPage() const
+{
+  if (!this->zimFileHandler) {
+    throw NoEntry();
+  }
+
+  auto firstPageOffset = zimFileHandler->getNamespaceBeginOffset('A');
+  auto article = zimFileHandler->getArticle(firstPageOffset);
+
+  if (! article.good()) {
+    throw NoEntry();
+  }
+
+  return article;
+}
+
+bool _parseUrl(const string& url, char* ns, string& title)
 {
   /* Offset to visit the url */
   unsigned int urlLength = url.size();
@@ -420,6 +439,52 @@ bool Reader::parseUrl(const string& url, char* ns, string& title) const
   return true;
 }
 
+bool Reader::parseUrl(const string& url, char* ns, string& title) const
+{
+  return _parseUrl(url, ns, title);
+}
+
+Entry Reader::getEntryFromPath(const std::string& path) const
+{
+  char ns = 0;
+  std::string short_url;
+
+  if (!this->zimFileHandler) {
+    throw NoEntry();
+  }
+  _parseUrl(path, &ns, short_url);
+
+  if (short_url.empty() && ns == 0) {
+    return getMainPage();
+  }
+
+  auto article = zimFileHandler->getArticle(ns, short_url);
+  if (!article.good()) {
+    throw NoEntry();
+  }
+
+  return article;
+}
+
+Entry Reader::getEntryFromEncodedPath(const std::string& path) const
+{
+  return getEntryFromPath(urlDecode(path));
+}
+
+Entry Reader::getEntryFromTitle(const std::string& title) const
+{
+  if (!this->zimFileHandler) {
+    throw NoEntry();
+  }
+
+  auto article = this->zimFileHandler->getArticleByTitle('A', title);
+  if (!article.good()) {
+    throw NoEntry();
+  }
+
+  return article;
+}
+
 /* Return article by url */
 bool Reader::getArticleObjectByDecodedUrl(const string& url,
                                           zim::Article& article) const
@@ -431,11 +496,11 @@ bool Reader::getArticleObjectByDecodedUrl(const string& url,
   /* Parse the url */
   char ns = 0;
   string urlStr;
-  this->parseUrl(url, &ns, urlStr);
+  _parseUrl(url, &ns, urlStr);
 
   /* Main page */
   if (urlStr.empty() && ns == 0) {
-    this->parseUrl(this->getMainPageUrl(), &ns, urlStr);
+    _parseUrl(this->getMainPage().getPath(), &ns, urlStr);
   }
 
   /* Extract the content from the zim file */
@@ -446,26 +511,53 @@ bool Reader::getArticleObjectByDecodedUrl(const string& url,
 /* Return the mimeType without the content */
 bool Reader::getMimeTypeByUrl(const string& url, string& mimeType) const
 {
-  if (this->zimFileHandler == NULL) {
-    return false;
-  }
-
-  zim::Article article;
-  if (this->getArticleObjectByDecodedUrl(url, article)) {
-    try {
-      mimeType = article.getMimeType();
-    } catch (exception& e) {
-      cerr << "Unable to get the mimetype for " << url << ":" << e.what()
-           << endl;
-      mimeType = "application/octet-stream";
-    }
+  try {
+    auto entry = getEntryFromPath(url);
+    mimeType = entry.getMimetype();
     return true;
-  } else {
+  } catch (NoEntry& e) {
     mimeType = "";
     return false;
   }
 }
 
+bool get_content_by_decoded_url(const Reader& reader,
+                                const string& url,
+                                string& content,
+                                string& title,
+                                unsigned int& contentLength,
+                                string& contentType,
+                                string& baseUrl)
+{
+  content = "";
+  contentType = "";
+  contentLength = 0;
+
+  try {
+    auto entry = reader.getEntryFromPath(url);
+    entry = entry.getFinalEntry();
+    baseUrl = entry.getPath();
+    contentType = entry.getMimetype();
+    content = entry.getContent();
+    contentLength = entry.getSize();
+    title = entry.getTitle();
+
+    /* Try to set a stub HTML header/footer if necesssary */
+    if (contentType.find("text/html") != string::npos
+      && content.find("<body") == std::string::npos
+      && content.find("<BODY") == std::string::npos) {
+      content = "<html><head><title>" + title +
+              "</title><meta http-equiv=\"Content-Type\" content=\"text/html; "
+              "charset=utf-8\" /></head><body>" +
+              content + "</body></html>";
+    }
+    return true;
+  } catch (NoEntry& e) {
+    return false;
+  }
+}
+
+
 /* Get a content from a zim file */
 bool Reader::getContentByUrl(const string& url,
                              string& content,
@@ -473,7 +565,14 @@ bool Reader::getContentByUrl(const string& url,
                              unsigned int& contentLength,
                              string& contentType) const
 {
-  return this->getContentByEncodedUrl(url, content, title, contentLength, contentType);
+  std::string stubRedirectUrl;
+  return get_content_by_decoded_url(*this,
+                                kiwix::urlDecode(url),
+                                content,
+                                title,
+                                contentLength,
+                                contentType,
+                                stubRedirectUrl);
 }
 
 bool Reader::getContentByEncodedUrl(const string& url,
@@ -483,8 +582,13 @@ bool Reader::getContentByEncodedUrl(const string& url,
                                     string& contentType,
                                     string& baseUrl) const
 {
-  return this->getContentByDecodedUrl(
-      kiwix::urlDecode(url), content, title, contentLength, contentType, baseUrl);
+  return get_content_by_decoded_url(*this,
+                                kiwix::urlDecode(url),
+                                content,
+                                title,
+                                contentLength,
+                                contentType,
+                                baseUrl);
 }
 
 bool Reader::getContentByEncodedUrl(const string& url,
@@ -494,12 +598,13 @@ bool Reader::getContentByEncodedUrl(const string& url,
                                     string& contentType) const
 {
   std::string stubRedirectUrl;
-  return this->getContentByEncodedUrl(kiwix::urlDecode(url),
-                                      content,
-                                      title,
-                                      contentLength,
-                                      contentType,
-                                      stubRedirectUrl);
+  return get_content_by_decoded_url(*this,
+                                kiwix::urlDecode(url),
+                                content,
+                                title,
+                                contentLength,
+                                contentType,
+                                stubRedirectUrl);
 }
 
 bool Reader::getContentByDecodedUrl(const string& url,
@@ -509,12 +614,13 @@ bool Reader::getContentByDecodedUrl(const string& url,
                                     string& contentType) const
 {
   std::string stubRedirectUrl;
-  return this->getContentByDecodedUrl(kiwix::urlDecode(url),
-                                      content,
-                                      title,
-                                      contentLength,
-                                      contentType,
-                                      stubRedirectUrl);
+  return get_content_by_decoded_url(*this,
+                                url,
+                                content,
+                                title,
+                                contentLength,
+                                contentType,
+                                stubRedirectUrl);
 }
 
 bool Reader::getContentByDecodedUrl(const string& url,
@@ -524,64 +630,31 @@ bool Reader::getContentByDecodedUrl(const string& url,
                                     string& contentType,
                                     string& baseUrl) const
 {
-  content = "";
-  contentType = "";
-  contentLength = 0;
-
-  zim::Article article;
-  if (!this->getArticleObjectByDecodedUrl(url, article)) {
-    return false;
-  }
-
-  /* If redirect */
-  unsigned int loopCounter = 0;
-  while (article.isRedirect() && loopCounter++ < 42) {
-    article = article.getRedirectArticle();
-  }
-
-  if (loopCounter < 42) {
-    /* Compute base url (might be different from the url if redirects */
-    baseUrl
-        = "/" + std::string(1, article.getNamespace()) + "/" + article.getUrl();
-
-    /* Get the content mime-type */
-    try {
-      contentType
-          = string(article.getMimeType().data(), article.getMimeType().size());
-    } catch (exception& e) {
-      cerr << "Unable to get the mimetype for " << baseUrl << ":" << e.what()
-           << endl;
-      contentType = "application/octet-stream";
-    }
-
-    /* Get the data */
-    content = string(article.getData().data(), article.getArticleSize());
-    title = article.getTitle();
-  }
-
-  /* Try to set a stub HTML header/footer if necesssary */
-  if (contentType.find("text/html") != string::npos
-      && content.find("<body") == std::string::npos
-      && content.find("<BODY") == std::string::npos) {
-    content = "<html><head><title>" + article.getTitle() +
-              "</title><meta http-equiv=\"Content-Type\" content=\"text/html; "
-              "charset=utf-8\" /></head><body>" +
-              content + "</body></html>";
-  }
-
-  /* Get the data length */
-  contentLength = article.getArticleSize();
-
-  return true;
+  return get_content_by_decoded_url(*this,
+                                url,
+                                content,
+                                title,
+                                contentLength,
+                                contentType,
+                                baseUrl);
 }
 
 /* Check if an article exists */
 bool Reader::urlExists(const string& url) const
 {
+  return pathExists(url);
+}
+
+bool Reader::pathExists(const string& path) const
+{
+  if (!zimFileHandler)
+  {
+    return false;
+  }
+
   char ns = 0;
   string titleStr;
-  this->parseUrl(url, &ns, titleStr);
-  titleStr = "/" + titleStr;
+  _parseUrl(path, &ns, titleStr);
   zim::File::const_iterator findItr = zimFileHandler->find(ns, titleStr);
   return findItr != zimFileHandler->end() && findItr->getUrl() == titleStr;
 }
@@ -589,8 +662,13 @@ bool Reader::urlExists(const string& url) const
 /* Does the ZIM file has a fulltext index */
 bool Reader::hasFulltextIndex() const
 {
-  return ( this->urlExists("/Z/fulltextIndex/xapian")
-        && !zimFileHandler->is_multiPart() );
+  if (!zimFileHandler || zimFileHandler->is_multiPart() )
+  {
+    return false;
+  }
+
+  return ( pathExists("Z//fulltextIndex/xapian")
+        || pathExists("X/fulltext/xapian"));
 }
 
 /* Search titles by prefix */

src/searcher.cpp

@@ -17,6 +17,9 @@
  * MA 02110-1301, USA.
  */
 
+
+#include <cmath>
+
 #include "searcher.h"
 #include "kiwixlib-resources.h"
 #include "reader.h"
@@ -33,12 +36,14 @@
 using namespace CTPP;
 #endif
 
+#define MAX_SEARCH_LEN 140
+
 namespace kiwix
 {
 class _Result : public Result
 {
  public:
-  _Result(Searcher* searcher, zim::Search::iterator& iterator);
+  _Result(zim::Search::iterator& iterator);
   virtual ~_Result(){};
 
   virtual std::string get_url();
@@ -51,7 +56,6 @@ class _Result : public Result
   virtual int get_readerIndex();
 
  private:
-  Searcher* searcher;
   zim::Search::iterator iterator;
 };
 
@@ -83,17 +87,17 @@ Searcher::Searcher(const string& xapianDirectoryPath,
       resultCountPerPage(0),
       estimatedResultCount(0),
       resultStart(0),
-      resultEnd(0)
+      resultEnd(0),
+      contentHumanReadableId(humanReadableName)
 {
   loadICUExternalTables();
   if (!reader || !reader->hasFulltextIndex()) {
     internal->_xapianSearcher = new XapianSearcher(xapianDirectoryPath, reader);
   }
-  this->contentHumanReadableId = humanReadableName;
   this->humanReaderNames.push_back(humanReadableName);
 }
 
-Searcher::Searcher()
+Searcher::Searcher(const std::string& humanReadableName)
     : internal(new SearcherInternal()),
       searchPattern(""),
       protocolPrefix("zim://"),
@@ -101,7 +105,8 @@ Searcher::Searcher()
       resultCountPerPage(0),
       estimatedResultCount(0),
       resultStart(0),
-      resultEnd(0)
+      resultEnd(0),
+      contentHumanReadableId(humanReadableName)
 {
   loadICUExternalTables();
 }
@@ -112,10 +117,14 @@ Searcher::~Searcher()
   delete internal;
 }
 
-void Searcher::add_reader(Reader* reader, const std::string& humanReadableName)
+bool Searcher::add_reader(Reader* reader, const std::string& humanReadableName)
 {
+  if (!reader->hasFulltextIndex()) {
+      return false;
+  }
   this->readers.push_back(reader);
   this->humanReaderNames.push_back(humanReadableName);
+  return true;
 }
 
 /* Search strings in the database */
@@ -141,9 +150,9 @@ void Searcher::search(std::string& search,
   if (resultStart != resultEnd) {
     /* Avoid big researches */
     this->resultCountPerPage = resultEnd - resultStart;
-    if (this->resultCountPerPage > 70) {
-      resultEnd = resultStart + 70;
-      this->resultCountPerPage = 70;
+    if (this->resultCountPerPage > MAX_SEARCH_LEN) {
+      resultEnd = resultStart + MAX_SEARCH_LEN;
+      this->resultCountPerPage = MAX_SEARCH_LEN;
     }
 
     /* Perform the search */
@@ -160,7 +169,9 @@ void Searcher::search(std::string& search,
       std::vector<const zim::File*> zims;
       for (auto current = this->readers.begin(); current != this->readers.end();
            current++) {
-        zims.push_back((*current)->getZimFileHandler());
+        if ( (*current)->hasFulltextIndex() ) {
+            zims.push_back((*current)->getZimFileHandler());
+        }
       }
       zim::Search* search = new zim::Search(zims);
       search->set_query(unaccentedSearch);
@@ -174,11 +185,68 @@ void Searcher::search(std::string& search,
   return;
 }
 
+
+void Searcher::geo_search(float latitude, float longitude, float distance,
+                          unsigned int resultStart,
+                          unsigned int resultEnd,
+                          const bool verbose)
+{
+  this->reset();
+
+  if (verbose == true) {
+    cout << "Performing geo query `" << distance << "&(" << latitude << ";" << longitude << ")'" << endl;
+  }
+
+  /* If resultEnd & resultStart inverted */
+  if (resultStart > resultEnd) {
+    resultEnd += resultStart;
+    resultStart = resultEnd - resultStart;
+    resultEnd -= resultStart;
+  }
+
+  /* Try to find results */
+  if (resultStart == resultEnd) {
+    return;
+  }
+
+  if (internal->_xapianSearcher) {
+    return;
+  }
+
+  /* Avoid big researches */
+  this->resultCountPerPage = resultEnd - resultStart;
+  if (this->resultCountPerPage > MAX_SEARCH_LEN) {
+    resultEnd = resultStart + MAX_SEARCH_LEN;
+    this->resultCountPerPage = MAX_SEARCH_LEN;
+  }
+
+  /* Perform the search */
+  std::ostringstream oss;
+  oss << "Articles located less than " << distance << " meters of " << latitude << ";" << longitude;
+  this->searchPattern = oss.str();
+  this->resultStart = resultStart;
+  this->resultEnd = resultEnd;
+
+  std::vector<const zim::File*> zims;
+  for (auto current = this->readers.begin(); current != this->readers.end();
+       current++) {
+    zims.push_back((*current)->getZimFileHandler());
+  }
+  zim::Search* search = new zim::Search(zims);
+  search->set_query("");
+  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();
+}
+
+
 void Searcher::restart_search()
 {
   if (internal->_xapianSearcher) {
     internal->_xapianSearcher->restart_search();
-  } else {
+  } else if (internal->_search) {
     internal->current_iterator = internal->_search->begin();
   }
 }
@@ -187,8 +255,9 @@ Result* Searcher::getNextResult()
 {
   if (internal->_xapianSearcher) {
     return internal->_xapianSearcher->getNextResult();
-  } else if (internal->current_iterator != internal->_search->end()) {
-    Result* result = new _Result(this, internal->current_iterator);
+  } else if (internal->_search &&
+             internal->current_iterator != internal->_search->end()) {
+    Result* result = new _Result(internal->current_iterator);
     internal->current_iterator++;
     return result;
   }
@@ -254,8 +323,8 @@ bool Searcher::setSearchProtocolPrefix(const std::string prefix)
   return true;
 }
 
-_Result::_Result(Searcher* searcher, zim::Search::iterator& iterator)
-    : searcher(searcher), iterator(iterator)
+_Result::_Result(zim::Search::iterator& iterator)
+    : iterator(iterator)
 {
 }
 
@@ -298,7 +367,12 @@ int _Result::get_readerIndex()
 
 string Searcher::getHtml()
 {
-  SimpleVM oSimpleVM;
+  SimpleVM oSimpleVM(
+      1024, //iIMaxFunctions (default value)
+      4096, //iIMaxArgStackSize (default value)
+      4096, //iIMaxCodeStackSize (default value)
+      10240 * 2 //iIMaxSteps (default*2)
+  );
 
   // Fill data
   CDT oData;
@@ -367,7 +441,7 @@ string Searcher::getHtml()
   oData["resultRange"] = this->resultCountPerPage;
   oData["resultLastPageStart"]
       = this->estimatedResultCount > this->resultCountPerPage
-            ? this->estimatedResultCount - this->resultCountPerPage
+            ? std::round(this->estimatedResultCount / this->resultCountPerPage) * this->resultCountPerPage
             : 0;
   oData["protocolPrefix"] = this->protocolPrefix;
   oData["searchProtocolPrefix"] = this->searchProtocolPrefix;

src/xapian/myhtmlparse.h

@@ -40,6 +40,7 @@ class MyHtmlParser : public HtmlParser {
 	void process_text(const string &text);
 	void opening_tag(const string &tag);
 	void closing_tag(const string &tag);
+        using HtmlParser::parse_html;
 	void parse_html(const string &text, const string &charset_,
 			bool charset_from_meta_);
 	MyHtmlParser() :

src/xapianSearcher.cpp

@@ -193,13 +193,8 @@ std::string XapianResult::get_content()
   if (!searcher->reader) {
     return "";
   }
-  std::string content;
-  std::string title;
-  unsigned int contentLength;
-  std::string contentType;
-  searcher->reader->getContentByUrl(
-      get_url(), content, title, contentLength, contentType);
-  return content;
+  auto entry = searcher->reader->getEntryFromEncodedPath(get_url());
+  return entry.getContent();
 }
 
 int XapianResult::get_size()

travis/compile.sh

@@ -29,7 +29,12 @@ case ${PLATFORM} in
 esac
 
 cd ${TRAVIS_BUILD_DIR}
-export PKG_CONFIG_PATH=${INSTALL_DIR}/lib/x86_64-linux-gnu/pkgconfig
+if [[ "${TRAVIS_OS_NAME}" == "osx" ]]
+then
+  export PKG_CONFIG_PATH=${INSTALL_DIR}/lib/pkgconfig
+else
+  export PKG_CONFIG_PATH=${INSTALL_DIR}/lib/x86_64-linux-gnu/pkgconfig
+fi
 meson . build -Dctpp2-install-prefix=${INSTALL_DIR} ${MESON_OPTION}
 cd build
 ninja

travis/install_deps.sh

@@ -3,41 +3,27 @@
 set -e
 
 REPO_NAME=${TRAVIS_REPO_SLUG#*/}
-ARCHIVE_NAME=deps_${PLATFORM}_${REPO_NAME}.tar.gz
-
-# Packages.
-case ${PLATFORM} in
-     "native_static")
-         PACKAGES="gcc cmake libbz2-dev ccache zlib1g-dev uuid-dev libctpp2-dev ctpp2-utils"
-         ;;
-     "native_dyn")
-         PACKAGES="gcc cmake libbz2-dev ccache zlib1g-dev uuid-dev libctpp2-dev ctpp2-utils libmicrohttpd-dev"
-         ;;
-     "win32_static")
-         PACKAGES="g++-mingw-w64-i686 gcc-mingw-w64-i686 gcc-mingw-w64-base mingw-w64-tools ccache ctpp2-utils"
-         ;;
-     "win32_dyn")
-         PACKAGES="g++-mingw-w64-i686 gcc-mingw-w64-i686 gcc-mingw-w64-base mingw-w64-tools ccache ctpp2-utils"
-         ;;
-     "android_arm")
-         PACKAGES="gcc cmake ccache ctpp2-utils"
-         ;;
-     "android_arm64")
-         PACKAGES="gcc cmake ccache ctpp2-utils"
-         ;;
-esac
-
-sudo apt-get update -qq
-sudo apt-get install -qq python3-pip ${PACKAGES}
-sudo pip3 install meson
+ARCHIVE_NAME=deps_${TRAVIS_OS_NAME}_${PLATFORM}_${REPO_NAME}.tar.gz
 
 # Ninja
 cd $HOME
-git clone git://github.com/ninja-build/ninja.git
-cd ninja
-git checkout release
-./configure.py --bootstrap
-sudo cp ninja /bin
+if [[ "$TRAVIS_OS_NAME" == "osx" ]]
+then
+  brew update
+  brew upgrade python3
+  pip3 install meson==0.43.0
+
+  wget https://github.com/ninja-build/ninja/releases/download/v1.8.2/ninja-mac.zip
+  unzip ninja-mac.zip ninja
+else
+  pip3 install --user meson==0.43.0
+
+  wget https://github.com/ninja-build/ninja/releases/download/v1.8.2/ninja-linux.zip
+  unzip ninja-linux.zip ninja
+fi
+
+mkdir -p $HOME/bin
+cp ninja $HOME/bin
 
 # Dependencies comming from kiwix-build.
 cd ${HOME}