goldendict-ng/gls.cc

/* This file is (c) 2008-2017 Abs62
 * Part of GoldenDict. Licensed under GPLv3 or later, see the LICENSE file */

#include <zlib.h>
#include "gls.hh"
#include "iconv.hh"
#include "dictionary.hh"
#include "ufile.hh"
#include "btreeidx.hh"
#include "folding.hh"
#include "gddebug.hh"
#include "utf8.hh"
#include "wstring_qt.hh"
#include "chunkedstorage.hh"
#include "langcoder.hh"
#include "dictzip.h"
#include "indexedzip.hh"
#include "ftshelpers.hh"
#include "fsencoding.hh"
#include "htmlescape.hh"
#include "filetype.hh"
#include "tiff.hh"

#include <QString>
#include <QSemaphore>
#include <QThreadPool>
#include <QAtomicInt>
// For TIFF conversion
#include <QImage>
#include <QByteArray>
#include <QBuffer>

#include <string>
#include <list>
#include <map>
#include <set>

#ifdef _MSC_VER
#include <stub_msvc.h>
#endif

namespace Gls {

using std::list;
using std::map;
using std::set;
using std::multimap;
using std::pair;

using gd::wstring;
using gd::wchar;

using BtreeIndexing::WordArticleLink;
using BtreeIndexing::IndexedWords;
using BtreeIndexing::IndexInfo;

enum Encoding
{
  Utf8,
  Utf16LE,
  Utf16BE
};

/////////////// GlsScanner

class GlsScanner
{
  gzFile f;
  Encoding encoding;
  Iconv iconv;
  wstring dictionaryName;
  wstring dictionaryDecription, dictionaryAuthor;
  wstring langFrom, langTo;
  char readBuffer[ 65536 ];
  char * readBufferPtr;
  size_t readBufferLeft;
  vector< wchar > wcharBuffer;
  unsigned linesRead;

public:

  DEF_EX( Ex, "Gls scanner exception", Dictionary::Ex )
  DEF_EX_STR( exCantOpen, "Can't open .gls file", Ex )
  DEF_EX( exCantReadGlsFile, "Can't read .gls file", Ex )
  DEF_EX_STR( exMalformedGlsFile, "The .gls file is malformed:", Ex )
  DEF_EX( exEncodingError, "Encoding error", Ex ) // Should never happen really

  GlsScanner( string const & fileName ) throw( Ex, Iconv::Ex );
  ~GlsScanner() throw();

  /// Returns the detected encoding of this file.
  Encoding getEncoding() const
  { return encoding; }

  /// Returns the dictionary's name, as was read from file's headers.
  wstring const & getDictionaryName() const
  { return dictionaryName; }

  /// Returns the dictionary's author, as was read from file's headers.
  wstring const & getDictionaryAuthor() const
  { return dictionaryAuthor; }

  /// Returns the dictionary's description, as was read from file's headers.
  wstring const & getDictionaryDescription() const
  { return dictionaryDecription; }

  /// Returns the dictionary's source language, as was read from file's headers.
  wstring const & getLangFrom() const
  { return langFrom; }

  /// Returns the dictionary's target language, as was read from file's headers.
  wstring const & getLangTo() const
  { return langTo; }

  /// Reads next line from the file. Returns true if reading succeeded --
  /// the string gets stored in the one passed, along with its physical
  /// file offset in the file (the uncompressed one if the file is compressed).
  /// If end of file is reached, false is returned.
  /// Reading begins from the first line after the headers (ones which end
  /// by the "### Glossary section:" line).
  bool readNextLine( wstring &, size_t & offset ) throw( Ex, Iconv::Ex );

  /// Returns the number of lines read so far from the file.
  unsigned getLinesRead() const
  { return linesRead; }

  /// Returns a name to be passed to iconv for the given encoding.
  static char const * getEncodingNameFor( Encoding e )
  {
    switch( e )
    {
      case Utf16LE:
        return Iconv::Utf16Le;
      case Utf16BE:
        return "UTF-16BE";
      case Utf8:
      default:
        return Iconv::Utf8;
    }
  }
};

GlsScanner::GlsScanner( string const & fileName ) throw( Ex, Iconv::Ex ):
  encoding( Utf8 ), iconv( Iconv::GdWchar, Iconv::Utf8 ), readBufferPtr( readBuffer ),
  readBufferLeft( 0 ), wcharBuffer( 64 ), linesRead( 0 )
{
  // Since .dz is backwards-compatible with .gz, we use gz- functions to
  // read it -- they are much nicer than the dict_data- ones.

  f = gd_gzopen( fileName.c_str() );
  if ( !f )
    throw exCantOpen( fileName );

  // Now try guessing the encoding by reading the first two bytes

  unsigned char firstBytes[ 2 ];

  if ( gzread( f, firstBytes, sizeof( firstBytes ) ) != sizeof( firstBytes ) )
  {
    // Apparently the file's too short
    gzclose( f );
    throw exMalformedGlsFile( fileName );
  }

  // If the file begins with the dedicated Unicode marker, we just consume
  // it. If, on the other hand, it's not, we return the bytes back
  if ( firstBytes[ 0 ] == 0xFF && firstBytes[ 1 ] == 0xFE )
    encoding = Utf16LE;
  else
  if ( firstBytes[ 0 ] == 0xFE && firstBytes[ 1 ] == 0xFF )
    encoding = Utf16BE;
  else
  if ( firstBytes[ 0 ] == 0xEF && firstBytes[ 1 ] == 0xBB )
  {
    // Looks like Utf8, read one more byte
    if ( gzread( f, firstBytes, 1 ) != 1 || firstBytes[ 0 ] != 0xBF )
    {
      // Either the file's too short, or the BOM is weird
      gzclose( f );
      throw exMalformedGlsFile( fileName );
    }
    encoding = Utf8;
  }
  else
  {
    if ( gzrewind( f ) )
    {
      gzclose( f );
      throw exCantOpen( fileName );
    }
    encoding = Utf8;
  }

  if( encoding != Utf8 )
    iconv.reinit( Iconv::GdWchar, getEncodingNameFor( encoding ) );

  // We now can use our own readNextLine() function

  wstring str;
  wstring *currentField = 0;
  wstring mark = GD_NATIVE_TO_WS( L"###" );
  wstring titleMark = GD_NATIVE_TO_WS( L"### Glossary title:" );
  wstring authorMark = GD_NATIVE_TO_WS( L"### Author:" );
  wstring descriptionMark = GD_NATIVE_TO_WS( L"### Description:" );
  wstring langFromMark = GD_NATIVE_TO_WS( L"### Source language:" );
  wstring langToMark = GD_NATIVE_TO_WS( L"### Target language:" );
  wstring endOfHeaderMark = GD_NATIVE_TO_WS( L"### Glossary section:" );
  size_t offset;

  for( ; ; )
  {
    if ( !readNextLine( str, offset ) )
    {
      gzclose( f );
      throw exMalformedGlsFile( fileName );
    }

    if( str.compare( 0, 3, mark.c_str(), 3 ) == 0 )
    {
      currentField = 0;

      if( str.compare( 0, titleMark.size(), titleMark ) == 0 )
      {
        dictionaryName = wstring( str, titleMark.size(), str.size() - titleMark.size() );
        currentField = &dictionaryName;
      }
      else
      if( str.compare( 0, authorMark.size(), authorMark ) == 0 )
      {
        dictionaryAuthor = wstring( str, authorMark.size(), str.size() - authorMark.size() );
        currentField = &dictionaryAuthor;
      }
      else
      if( str.compare( 0, descriptionMark.size(), descriptionMark ) == 0 )
      {
        dictionaryDecription = wstring( str, descriptionMark.size(), str.size() - descriptionMark.size() );
        currentField = &dictionaryDecription;
      }
      else
      if( str.compare( 0, langFromMark.size(), langFromMark ) == 0 )
      {
        langFrom = wstring( str, langFromMark.size(), str.size() - langFromMark.size() );
      }
      else
      if( str.compare( 0, langToMark.size(), langToMark ) == 0 )
      {
        langTo = wstring( str, langToMark.size(), str.size() - langToMark.size() );
      }
      else
      if( str.compare( 0, endOfHeaderMark.size(), endOfHeaderMark ) == 0 )
      {
        break;
      }
    }
    else
    {
      /// Handle multiline headers
      if( currentField )
        *currentField += str;
    }
  }
}

bool GlsScanner::readNextLine( wstring & out, size_t & offset ) throw( Ex,
                                                                       Iconv::Ex )
{
  offset = (size_t)( gztell( f ) - readBufferLeft );

  // For now we just read one char at a time
  size_t readMultiple = ( encoding == Utf16LE || encoding == Utf16BE ) ? 2 : 1;

  size_t leftInOut = wcharBuffer.size();

  wchar * outPtr = &wcharBuffer.front();

  for( ; ; )
  {
    // Check that we have bytes to read
    if ( readBufferLeft < 4 ) // To convert one char, we need at most 4 bytes
    {
      if ( !gzeof( f ) )
      {
        // To avoid having to deal with ring logic, we move the remaining bytes
        // to the beginning
        memmove( readBuffer, readBufferPtr, readBufferLeft );

        // Read some more bytes to readBuffer
        int result = gzread( f, readBuffer + readBufferLeft,
                             sizeof( readBuffer ) - readBufferLeft );

        if ( result == -1 )
          throw exCantReadGlsFile();

        readBufferPtr = readBuffer;
        readBufferLeft += (size_t) result;
      }
    }

    if ( readBufferLeft < readMultiple )
    {
      // No more data. Return what we've got so far, forget the last byte if
      // it was a 16-bit Unicode and a file had an odd number of bytes.
      readBufferLeft = 0;

      if ( outPtr != &wcharBuffer.front() )
      {
        // If there was a stray \r, remove it
        if ( outPtr[ -1 ] == L'\r' )
          --outPtr;

        out = wstring( &wcharBuffer.front(), outPtr - &wcharBuffer.front() );

        ++linesRead;

        return true;
      }
      else
        return false;
    }

    // Check that we have chars to write
    if ( leftInOut < 2 ) // With 16-bit wchars, 2 is needed for a surrogate pair
    {
      wcharBuffer.resize( wcharBuffer.size() + 64 );
      outPtr = &wcharBuffer.front() + wcharBuffer.size() - 64 - leftInOut;
      leftInOut += 64;
    }

    // Ok, now convert one char
    size_t outBytesLeft = sizeof( wchar );

    Iconv::Result r =
      iconv.convert( (void const *&)readBufferPtr, readBufferLeft,
                     (void *&)outPtr, outBytesLeft );

    if ( r == Iconv::NeedMoreOut && outBytesLeft == sizeof( wchar ) )
    {
      // Seems to be a surrogate pair with a 16-bit target wchar

      outBytesLeft *= 2;
      r = iconv.convert( (void const *&)readBufferPtr, readBufferLeft,
                     (void *&)outPtr, outBytesLeft );
      --leftInOut; // Complements the next decremention
    }

    if ( outBytesLeft )
      throw exEncodingError();

    --leftInOut;

    // Have we got \n?
    if ( outPtr[ -1 ] == L'\n' )
    {
      --outPtr;

      // Now kill a \r if there is one, and return the result.
      if ( outPtr != &wcharBuffer.front() && outPtr[ -1 ] == L'\r' )
          --outPtr;

      out = wstring( &wcharBuffer.front(), outPtr - &wcharBuffer.front() );

      ++linesRead;

      return true;
    }
  }
}

GlsScanner::~GlsScanner() throw()
{
  gzclose( f );
}

namespace {

////////////////// GLS Dictionary

DEF_EX_STR( exCantReadFile, "Can't read file", Dictionary::Ex )
DEF_EX( exUserAbort, "User abort", Dictionary::Ex )
DEF_EX_STR( exDictzipError, "DICTZIP error", Dictionary::Ex )

enum
{
  Signature = 0x58534c47, // GLSX on little-endian, XSLG on big-endian
  CurrentFormatVersion = 1 + BtreeIndexing::FormatVersion + Folding::Version,
  CurrentZipSupportVersion = 2,
  CurrentFtsIndexVersion = 1
};

struct IdxHeader
{
  uint32_t signature; // First comes the signature, GLSX
  uint32_t formatVersion; // File format version (CurrentFormatVersion)
  uint32_t zipSupportVersion; // Zip support version -- narrows down reindexing
                              // when it changes only for dictionaries with the
                              // zip files
  int glsEncoding; // Which encoding is used for the file indexed
  uint32_t chunksOffset; // The offset to chunks' storage
  uint32_t indexBtreeMaxElements; // Two fields from IndexInfo
  uint32_t indexRootOffset;
  uint32_t articleCount; // Number of articles this dictionary has
  uint32_t wordCount; // Number of headwords this dictionary has
  uint32_t langFrom;  // Source language
  uint32_t langTo;    // Target language
  uint32_t hasZipFile; // Non-zero means there's a zip file with resources
                       // present
  uint32_t zipIndexBtreeMaxElements; // Two fields from IndexInfo of the zip
                                     // resource index.
  uint32_t zipIndexRootOffset;
}
#ifndef _MSC_VER
__attribute__((packed))
#endif
;

bool indexIsOldOrBad( string const & indexFile, bool hasZipFile )
{
  File::Class idx( indexFile, "rb" );

  IdxHeader header;

  return idx.readRecords( &header, sizeof( header ), 1 ) != 1 ||
         header.signature != Signature ||
         header.formatVersion != CurrentFormatVersion ||
         (bool) header.hasZipFile != hasZipFile ||
         ( hasZipFile && header.zipSupportVersion != CurrentZipSupportVersion );
}

class GlsDictionary: public BtreeIndexing::BtreeDictionary
{
  Mutex idxMutex;
  File::Class idx;
  IdxHeader idxHeader;
  dictData * dz;
  ChunkedStorage::Reader chunks;
  Mutex dzMutex;
  Mutex resourceZipMutex;
  IndexedZip resourceZip;
  string dictionaryName;

public:

  GlsDictionary( string const & id, string const & indexFile,
                      vector< string > const & dictionaryFiles );

  ~GlsDictionary();

  virtual string getName() throw()
  { return dictionaryName; }

  virtual map< Dictionary::Property, string > getProperties() throw()
  { return map< Dictionary::Property, string >(); }

  virtual unsigned long getArticleCount() throw()
  { return idxHeader.articleCount; }

  virtual unsigned long getWordCount() throw()
  { return idxHeader.wordCount; }

  inline virtual quint32 getLangFrom() const
  { return idxHeader.langFrom; }

  inline virtual quint32 getLangTo() const
  { return idxHeader.langTo; }

  virtual sptr< Dictionary::WordSearchRequest > findHeadwordsForSynonym( wstring const & )
    throw( std::exception );

  virtual sptr< Dictionary::DataRequest > getArticle( wstring const &,
                                                      vector< wstring > const & alts,
                                                      wstring const & )
    throw( std::exception );

  virtual sptr< Dictionary::DataRequest > getResource( string const & name )
    throw( std::exception );

  virtual QString const& getDescription();

  virtual QString getMainFilename();

  virtual sptr< Dictionary::DataRequest > getSearchResults( QString const & searchString,
                                                            int searchMode, bool matchCase,
                                                            int distanceBetweenWords,
                                                            int maxResults );

  virtual void getArticleText( uint32_t articleAddress, QString & headword, QString & text );

  virtual void makeFTSIndex(QAtomicInt & isCancelled, bool firstIteration );

  virtual void setFTSParameters( Config::FullTextSearch const & fts )
  {
    can_FTS = fts.enabled
              && !fts.disabledTypes.contains( "GLS", Qt::CaseInsensitive )
              && ( fts.maxDictionarySize == 0 || getArticleCount() <= fts.maxDictionarySize );
  }
protected:

  void loadIcon() throw();

private:

  /// Loads the article, storing its headword and formatting the data it has
  /// into an html.
  void loadArticle(  uint32_t address,
                     string & headword,
                     string & articleText );

  /// Loads the article
  void loadArticleText(  uint32_t address,
                         vector< string > & headwords,
                         string & articleText );

  /// Process resource links (images, audios, etc)
  QString & filterResource( QString & article );

  friend class GlsResourceRequest;
  friend class GlsArticleRequest;
  friend class GlsHeadwordsRequest;
};

GlsDictionary::GlsDictionary( string const & id,
                              string const & indexFile,
                              vector< string > const & dictionaryFiles ):
  BtreeDictionary( id, dictionaryFiles ),
  idx( indexFile, "rb" ),
  idxHeader( idx.read< IdxHeader >() ),
  dz( 0 ),
  chunks( idx, idxHeader.chunksOffset )
{
  // Open the .gls file

  DZ_ERRORS error;
  dz = dict_data_open( getDictionaryFilenames()[ 0 ].c_str(), &error, 0 );

  if ( !dz )
    throw exDictzipError( string( dz_error_str( error ) )
                          + "(" + getDictionaryFilenames()[ 0 ] + ")" );

  // Read the dictionary name

  idx.seek( sizeof( idxHeader ) );

  vector< char > dName( idx.read< uint32_t >() );
  idx.read( &dName.front(), dName.size() );
  dictionaryName = string( &dName.front(), dName.size() );

  // Initialize the index

  openIndex( IndexInfo( idxHeader.indexBtreeMaxElements,
                        idxHeader.indexRootOffset ),
             idx, idxMutex );

  // Open a resource zip file, if there's one

  if ( idxHeader.hasZipFile &&
       ( idxHeader.zipIndexBtreeMaxElements ||
         idxHeader.zipIndexRootOffset ) )
  {
    resourceZip.openIndex( IndexInfo( idxHeader.zipIndexBtreeMaxElements,
                                      idxHeader.zipIndexRootOffset ),
                           idx, idxMutex );

    QString zipName = QDir::fromNativeSeparators(
        FsEncoding::decode( getDictionaryFilenames().back().c_str() ) );

    if ( zipName.endsWith( ".zip", Qt::CaseInsensitive ) ) // Sanity check
      resourceZip.openZipFile( zipName );
  }

  // Full-text search parameters

  can_FTS = true;

  ftsIdxName = indexFile + "_FTS";

  if( !Dictionary::needToRebuildIndex( dictionaryFiles, ftsIdxName )
      && !FtsHelpers::ftsIndexIsOldOrBad( ftsIdxName, this ) )
    FTS_index_completed.ref();
}

GlsDictionary::~GlsDictionary()
{
  if ( dz )
    dict_data_close( dz );
}

void GlsDictionary::loadIcon() throw()
{
  if ( dictionaryIconLoaded )
    return;

  QString fileName =
    QDir::fromNativeSeparators( FsEncoding::decode( getDictionaryFilenames()[ 0 ].c_str() ) );

  // Remove the extension
  if ( fileName.endsWith( ".gls.dz", Qt::CaseInsensitive ) )
    fileName.chop( 6 );
  else
    fileName.chop( 3 );

  if ( !loadIconFromFile( fileName ) )
  {
    // Load failed -- use default icon
    dictionaryNativeIcon = dictionaryIcon = QIcon(":/icons/icon32_gls.png");
  }

  dictionaryIconLoaded = true;
}

QString const& GlsDictionary::getDescription()
{
  if( !dictionaryDescription.isEmpty() )
      return dictionaryDescription;

  try {
    GlsScanner scanner( getDictionaryFilenames()[ 0 ] );
    string str = Utf8::encode( scanner.getDictionaryAuthor() );
    if( !str.empty() )
      dictionaryDescription = QString( QObject::tr( "Author: %1%2" ) )
                              .arg( QString::fromUtf8( str.c_str() ) )
                              .arg( "\n\n" );
    str = Utf8::encode( scanner.getDictionaryDescription() );
    if( !str.empty() )
    {
      QString desc = QString::fromUtf8( str.c_str() );
      desc.replace( "\t", "<br/>" );
      desc.replace( "\\n", "<br/>" );
      desc.replace( "<br>", "<br/>", Qt::CaseInsensitive );
      dictionaryDescription += Html::unescape( desc, true );
    }
  }
  catch( std::exception & e )
  {
    gdWarning( "GLS dictionary description reading failed: %s, error: %s\n",
               getName().c_str(), e.what() );
  }

  if( dictionaryDescription.isEmpty() )
    dictionaryDescription = "NONE";

  return dictionaryDescription;
}

QString GlsDictionary::getMainFilename()
{
  return FsEncoding::decode( getDictionaryFilenames()[ 0 ].c_str() );
}

void GlsDictionary::makeFTSIndex( QAtomicInt & isCancelled, bool firstIteration )
{
  if( !( Dictionary::needToRebuildIndex( getDictionaryFilenames(), ftsIdxName )
         || FtsHelpers::ftsIndexIsOldOrBad( ftsIdxName, this ) ) )
    FTS_index_completed.ref();

  if( haveFTSIndex() )
    return;

  if( ensureInitDone().size() )
    return;

  if( firstIteration && getArticleCount() > FTS::MaxDictionarySizeForFastSearch )
    return;

  gdDebug( "Gls: Building the full-text index for dictionary: %s\n",
           getName().c_str() );

  try
  {
    FtsHelpers::makeFTSIndex( this, isCancelled );
    FTS_index_completed.ref();
  }
  catch( std::exception &ex )
  {
    gdWarning( "Gls: Failed building full-text search index for \"%s\", reason: %s\n", getName().c_str(), ex.what() );
    QFile::remove( FsEncoding::decode( ftsIdxName.c_str() ) );
  }
}

void GlsDictionary::loadArticleText( uint32_t address,
                                     vector< string > & headwords,
                                     string & articleText )
{
  vector< char > chunk;
  char * articleProps;
  {
    Mutex::Lock _( idxMutex );

    articleProps = chunks.getBlock( address, chunk );
  }

  uint32_t articleOffset, articleSize;

  memcpy( &articleOffset, articleProps, sizeof( articleOffset ) );
  memcpy( &articleSize, articleProps + sizeof( articleOffset ),
          sizeof( articleSize ) );

  char * articleBody;

  {
    Mutex::Lock _( dzMutex );

    articleBody = dict_data_read_( dz, articleOffset, articleSize, 0, 0 );
  }

  headwords.clear();
  articleText.clear();
  string headword;

  if ( !articleBody )
  {
    articleText = string( "\n\tDICTZIP error: " ) + dict_error_str( dz );
  }
  else
  {
    string articleData = Iconv::toUtf8( GlsScanner::getEncodingNameFor( Encoding( idxHeader.glsEncoding ) ), articleBody, articleSize );
    string::size_type start_pos = 0, end_pos = 0;

    for( ; ; )
    {
      // Replace all "\r\n" by "\n"
      end_pos = articleData.find( "\r\n", start_pos );
      if( end_pos == string::npos )
      {
        articleText += articleData.substr( start_pos, end_pos );
        break;
      }
      else
      {
        articleText += articleData.substr( start_pos, end_pos - start_pos ) + "\n";
        start_pos = end_pos + 2;
      }
    }

    // Find headword
    start_pos = articleText.find( '\n' );
    if( start_pos != string::npos )
    {
      headword = articleText.substr( 0, start_pos );
      articleText = articleText.substr( start_pos + 1, string::npos );
    }

    // Parse headwords

    start_pos = 0;
    end_pos = 0;
    for( ; ; )
    {
      end_pos = headword.find( '|', start_pos );
      if( end_pos == wstring::npos )
      {
        string hw = headword.substr( start_pos );
        if( !hw.empty() )
          headwords.push_back( hw );
        break;
      }
      headwords.push_back( headword.substr( start_pos, end_pos - start_pos ) );
      start_pos = end_pos + 1;
    }
  }
}

void GlsDictionary::loadArticle( uint32_t address,
                                 string & headword,
                                 string & articleText )
{
  string articleBody;
  vector< string > headwords;
  loadArticleText( address, headwords, articleBody );

  QString article = QString::fromLatin1( "<div class=\"glsdict\">" );
  if( headwords.size() )
  {
    // Headwords
    article += "<div class=\"glsdict_headwords\"";
    if( isFromLanguageRTL() )
      article += " dir=\"rtl\"";
    if( headwords.size() > 1 )
    {
      QString altHeadwords;
      for( vector< string >::size_type i = 1; i < headwords.size(); i++ )
      {
        if( i > 1 )
          altHeadwords += ", ";
        altHeadwords += QString::fromUtf8( headwords[ i ].c_str(), headwords[ i ].size() );
      }
      article += " title=\"" + altHeadwords + "\"";
    }
    article += ">";

    headword = headwords.front();
    article += QString::fromUtf8( headword.c_str(), headword.size() );

    article += "</div>";
  }

  if( isToLanguageRTL() )
    article += "<div style=\"display:inline;\" dir=\"rtl\">";

  QString text = QString::fromUtf8( articleBody.c_str(), articleBody.size() );
  article += filterResource( text );

  if( isToLanguageRTL() )
    article += "</div>";

  article +="</div>";

  articleText = string( article.toUtf8().data() );
}

QString & GlsDictionary::filterResource( QString & article )
{
  QRegExp imgRe( "(<\\s*img\\s+[^>]*src\\s*=\\s*[\"']+)(?!(?:data|https?|ftp):)", Qt::CaseInsensitive );
  imgRe.setMinimal( true );
  QRegExp linkRe( "(<\\s*link\\s+[^>]*href\\s*=\\s*[\"']+)(?!(?:data|https?|ftp):)", Qt::CaseInsensitive );
  linkRe.setMinimal( true );

  article.replace( imgRe , "\\1bres://" + QString::fromStdString( getId() ) + "/" )
         .replace( linkRe, "\\1bres://" + QString::fromStdString( getId() ) + "/" );

  // Handle links to articles

  QRegExp linksReg( "<a(\\s*[^>]*)href\\s*=\\s*['\"](bword://)?([^'\"]+)['\"]" );
  linksReg.setMinimal( true );

  int pos = 0;
  while( pos >= 0 )
  {
    pos = linksReg.indexIn( article, pos );
    if( pos < 0 )
      break;

    QString link = linksReg.cap( 3 );
    if( link.indexOf( ':' ) < 0 )
    {
      QString newLink;
      if( link.indexOf( '#' ) < 0 )
        newLink = QString( "<a" ) + linksReg.cap( 1 ) + "href=\"bword:" + link + "\"";

      // Anchors

      if( link.indexOf( '#' ) > 0 )
      {
        newLink = QString( "<a" ) + linksReg.cap( 1 ) + "href=\"gdlookup://localhost/" + link + "\"";
        newLink.replace( "#", "?gdanchor=" );
      }

      if( !newLink.isEmpty() )
      {
        article.replace( pos, linksReg.cap( 0 ).size(), newLink );
        pos += newLink.size();
      }
      else
        pos += linksReg.cap( 0 ).size();
    }
    else
      pos += linksReg.cap( 0 ).size();
  }
  return article;
}

void GlsDictionary::getArticleText( uint32_t articleAddress, QString & headword, QString & text )
{
  try
  {
    vector< string > headwords;
    string articleStr;
    loadArticleText( articleAddress, headwords, articleStr );

    if( !headwords.empty() )
      headword = QString::fromUtf8( headwords.front().data(), headwords.front().size() );

    wstring wstr = Utf8::decode( articleStr );

    text = Html::unescape( gd::toQString( wstr ) );
  }
  catch( std::exception &ex )
  {
    gdWarning( "Gls: Failed retrieving article from \"%s\", reason: %s\n", getName().c_str(), ex.what() );
  }
}

/// GlsDictionary::findHeadwordsForSynonym()

class GlsHeadwordsRequest;

class GlsHeadwordsRequestRunnable: public QRunnable
{
  GlsHeadwordsRequest & r;
  QSemaphore & hasExited;

public:

  GlsHeadwordsRequestRunnable( GlsHeadwordsRequest & r_,
                               QSemaphore & hasExited_ ): r( r_ ),
                                                          hasExited( hasExited_ )
  {}

  ~GlsHeadwordsRequestRunnable()
  {
    hasExited.release();
  }

  virtual void run();
};

class GlsHeadwordsRequest: public Dictionary::WordSearchRequest
{
  friend class GlsHeadwordsRequestRunnable;

  wstring word;
  GlsDictionary & dict;

  QAtomicInt isCancelled;
  QSemaphore hasExited;

public:

  GlsHeadwordsRequest( wstring const & word_, GlsDictionary & dict_ ):
    word( word_ ), dict( dict_ )
  {
    QThreadPool::globalInstance()->start(
      new GlsHeadwordsRequestRunnable( *this, hasExited ) );
  }

  void run(); // Run from another thread by StardictHeadwordsRequestRunnable

  virtual void cancel()
  {
    isCancelled.ref();
  }

  ~GlsHeadwordsRequest()
  {
    isCancelled.ref();
    hasExited.acquire();
  }
};

void GlsHeadwordsRequestRunnable::run()
{
  r.run();
}

void GlsHeadwordsRequest::run()
{
  if ( Qt4x5::AtomicInt::loadAcquire( isCancelled ) )
  {
    finish();
    return;
  }

  try
  {
    vector< WordArticleLink > chain = dict.findArticles( word );

    wstring caseFolded = Folding::applySimpleCaseOnly( word );

    for( unsigned x = 0; x < chain.size(); ++x )
    {
      if ( Qt4x5::AtomicInt::loadAcquire( isCancelled ) )
      {
        finish();
        return;
      }

      string articleText;
      vector< string > headwords;

      dict.loadArticleText( chain[ x ].articleOffset,
                            headwords, articleText );

      wstring headwordDecoded = Utf8::decode( headwords.front() );

      if ( caseFolded != Folding::applySimpleCaseOnly( headwordDecoded ) )
      {
        // The headword seems to differ from the input word, which makes the
        // input word its synonym.
        Mutex::Lock _( dataMutex );

        matches.push_back( headwordDecoded );
      }
    }
  }
  catch( std::exception & e )
  {
    setErrorString( QString::fromUtf8( e.what() ) );
  }

  finish();
}

sptr< Dictionary::WordSearchRequest >
  GlsDictionary::findHeadwordsForSynonym( wstring const & word )
  throw( std::exception )
{
  return synonymSearchEnabled ? new GlsHeadwordsRequest( word, *this ) :
                                Class::findHeadwordsForSynonym( word );
}


/// GlsDictionary::getArticle()

class GlsArticleRequest;

class GlsArticleRequestRunnable: public QRunnable
{
  GlsArticleRequest & r;
  QSemaphore & hasExited;

public:

  GlsArticleRequestRunnable( GlsArticleRequest & r_,
                             QSemaphore & hasExited_ ): r( r_ ),
                                                        hasExited( hasExited_ )
  {}

  ~GlsArticleRequestRunnable()
  {
    hasExited.release();
  }

  virtual void run();
};

class GlsArticleRequest: public Dictionary::DataRequest
{
  friend class GlsArticleRequestRunnable;

  wstring word;
  vector< wstring > alts;
  GlsDictionary & dict;

  QAtomicInt isCancelled;
  QSemaphore hasExited;

public:

  GlsArticleRequest( wstring const & word_,
                     vector< wstring > const & alts_,
                     GlsDictionary & dict_ ):
    word( word_ ), alts( alts_ ), dict( dict_ )
  {
    QThreadPool::globalInstance()->start(
      new GlsArticleRequestRunnable( *this, hasExited ) );
  }

  void run(); // Run from another thread by GlsArticleRequestRunnable

  virtual void cancel()
  {
    isCancelled.ref();
  }

  ~GlsArticleRequest()
  {
    isCancelled.ref();
    hasExited.acquire();
  }
};

void GlsArticleRequestRunnable::run()
{
  r.run();
}

void GlsArticleRequest::run()
{
  if ( Qt4x5::AtomicInt::loadAcquire( isCancelled ) )
  {
    finish();
    return;
  }
  try
  {
    vector< WordArticleLink > chain = dict.findArticles( word );

    for( unsigned x = 0; x < alts.size(); ++x )
    {
      /// Make an additional query for each alt

      vector< WordArticleLink > altChain = dict.findArticles( alts[ x ] );

      chain.insert( chain.end(), altChain.begin(), altChain.end() );
    }

    multimap< wstring, pair< string, string > > mainArticles, alternateArticles;

    set< uint32_t > articlesIncluded; // Some synonims make it that the articles
                                      // appear several times. We combat this
                                      // by only allowing them to appear once.

    wstring wordCaseFolded = Folding::applySimpleCaseOnly( word );

    for( unsigned x = 0; x < chain.size(); ++x )
    {
      if ( Qt4x5::AtomicInt::loadAcquire( isCancelled ) )
      {
        finish();
        return;
      }

      if ( articlesIncluded.find( chain[ x ].articleOffset ) != articlesIncluded.end() )
        continue; // We already have this article in the body.

      // Now grab that article

      string headword, articleText;

      dict.loadArticle( chain[ x ].articleOffset, headword, articleText );

      // Ok. Now, does it go to main articles, or to alternate ones? We list
      // main ones first, and alternates after.

      // We do the case-folded comparison here.

      wstring headwordStripped =
        Folding::applySimpleCaseOnly( Utf8::decode( headword ) );

      multimap< wstring, pair< string, string > > & mapToUse =
        ( wordCaseFolded == headwordStripped ) ?
          mainArticles : alternateArticles;

      mapToUse.insert( pair< wstring, pair< string, string > >(
        Folding::applySimpleCaseOnly( Utf8::decode( headword ) ),
        pair< string, string >( headword, articleText ) ) );

      articlesIncluded.insert( chain[ x ].articleOffset );
    }

    if ( mainArticles.empty() && alternateArticles.empty() )
    {
      // No such word
      finish();
      return;
    }

    string result;

    multimap< wstring, pair< string, string > >::const_iterator i;

    for( i = mainArticles.begin(); i != mainArticles.end(); ++i )
    {
        result += i->second.second;
    }

    for( i = alternateArticles.begin(); i != alternateArticles.end(); ++i )
    {
        result += i->second.second;
    }

    Mutex::Lock _( dataMutex );

    data.resize( result.size() );

    memcpy( &data.front(), result.data(), result.size() );

    hasAnyData = true;
  }
  catch( std::exception & e )
  {
    setErrorString( QString::fromUtf8( e.what() ) );
  }

  finish();
}

sptr< Dictionary::DataRequest > GlsDictionary::getArticle( wstring const & word,
                                                           vector< wstring > const & alts,
                                                           wstring const & )
  throw( std::exception )
{
  return new GlsArticleRequest( word, alts, *this );
}

//////////////// GlsDictionary::getResource()

class GlsResourceRequest;

class GlsResourceRequestRunnable: public QRunnable
{
  GlsResourceRequest & r;
  QSemaphore & hasExited;

public:

  GlsResourceRequestRunnable( GlsResourceRequest & r_,
                              QSemaphore & hasExited_ ): r( r_ ),
                                                         hasExited( hasExited_ )
  {}

  ~GlsResourceRequestRunnable()
  {
    hasExited.release();
  }

  virtual void run();
};

class GlsResourceRequest: public Dictionary::DataRequest
{
  friend class GlsResourceRequestRunnable;

  GlsDictionary & dict;

  string resourceName;

  QAtomicInt isCancelled;
  QSemaphore hasExited;

public:

  GlsResourceRequest( GlsDictionary & dict_,
                      string const & resourceName_ ):
    dict( dict_ ),
    resourceName( resourceName_ )
  {
    QThreadPool::globalInstance()->start(
      new GlsResourceRequestRunnable( *this, hasExited ) );
  }

  void run(); // Run from another thread by GlsResourceRequestRunnable

  virtual void cancel()
  {
    isCancelled.ref();
  }

  ~GlsResourceRequest()
  {
    isCancelled.ref();
    hasExited.acquire();
  }
};

void GlsResourceRequestRunnable::run()
{
  r.run();
}

void GlsResourceRequest::run()
{
  // Some runnables linger enough that they are cancelled before they start
  if ( Qt4x5::AtomicInt::loadAcquire( isCancelled ) )
  {
    finish();
    return;
  }

  try
  {
    string n =
      FsEncoding::dirname( dict.getDictionaryFilenames()[ 0 ] ) +
      FsEncoding::separator() +
      FsEncoding::encode( resourceName );

    GD_DPRINTF( "n is %s\n", n.c_str() );

    try
    {
      Mutex::Lock _( dataMutex );

      File::loadFromFile( n, data );
    }
    catch( File::exCantOpen & )
    {
      n = dict.getDictionaryFilenames()[ 0 ] + ".files" +
          FsEncoding::separator() +
          FsEncoding::encode( resourceName );

      try
      {
        Mutex::Lock _( dataMutex );

        File::loadFromFile( n, data );
      }
      catch( File::exCantOpen & )
      {
        // Try reading from zip file

        if ( dict.resourceZip.isOpen() )
        {
          Mutex::Lock _( dict.resourceZipMutex );

          Mutex::Lock __( dataMutex );

          if ( !dict.resourceZip.loadFile( Utf8::decode( resourceName ), data ) )
            throw; // Make it fail since we couldn't read the archive
        }
        else
          throw;
      }
    }

    if ( Filetype::isNameOfTiff( resourceName ) )
    {
      // Convert it

      dataMutex.lock();

      QImage img = QImage::fromData( (unsigned char *) &data.front(),
                                     data.size() );

#ifdef MAKE_EXTRA_TIFF_HANDLER
      if( img.isNull() )
        GdTiff::tiffToQImage( &data.front(), data.size(), img );
#endif

      dataMutex.unlock();

      if ( !img.isNull() )
      {
        // Managed to load -- now store it back as BMP

        QByteArray ba;
        QBuffer buffer( &ba );
        buffer.open( QIODevice::WriteOnly );
        img.save( &buffer, "BMP" );

        Mutex::Lock _( dataMutex );

        data.resize( buffer.size() );

        memcpy( &data.front(), buffer.data(), data.size() );
      }
    }

    if( Filetype::isNameOfCSS( resourceName ) )
    {
      Mutex::Lock _( dataMutex );

      QString css = QString::fromUtf8( data.data(), data.size() );

      // Correct some url's

      QRegExp links( "url\\(\\s*(['\"]?)([^'\"]*)(['\"]?)\\s*\\)", Qt::CaseInsensitive, QRegExp::RegExp );
      QString id = QString::fromUtf8( dict.getId().c_str() );
      int pos = 0;
      for( ; ; )
      {
        pos = links.indexIn( css, pos );
        if( pos < 0 )
          break;
        QString url = links.cap( 2 );

        if( url.indexOf( ":/" ) >= 0 || url.indexOf( "data:" ) >= 0)
        {
          // External link
          pos += links.cap().size();
          continue;
        }

        QString newUrl = QString( "url(" ) + links.cap( 1 ) + "bres://"
                                           + id + "/" + url + links.cap( 3 ) + ")";
        css.replace( pos, links.cap().size(), newUrl );
        pos += newUrl.size();
      }

      dict.isolateCSS( css );
      QByteArray bytes = css.toUtf8();
      data.resize( bytes.size() );
      memcpy( &data.front(), bytes.constData(), bytes.size() );
    }

    hasAnyData = true;
  }
  catch( std::exception &ex )
  {
    gdWarning( "GLS: Failed loading resource \"%s\" for \"%s\", reason: %s\n",
               resourceName.c_str(), dict.getName().c_str(), ex.what() );
    // Resource not loaded -- we don't set the hasAnyData flag then
  }

  finish();
}

sptr< Dictionary::DataRequest > GlsDictionary::getResource( string const & name )
  throw( std::exception )
{
  return new GlsResourceRequest( *this, name );
}

sptr< Dictionary::DataRequest > GlsDictionary::getSearchResults( QString const & searchString,
                                                                 int searchMode, bool matchCase,
                                                                 int distanceBetweenWords,
                                                                 int maxResults )
{
  return new FtsHelpers::FTSResultsRequest( *this, searchString,searchMode, matchCase, distanceBetweenWords, maxResults );
}

} // anonymous namespace

/// makeDictionaries

vector< sptr< Dictionary::Class > > makeDictionaries(
                                      vector< string > const & fileNames,
                                      string const & indicesDir,
                                      Dictionary::Initializing & initializing )
  throw( std::exception )
{
  vector< sptr< Dictionary::Class > > dictionaries;

  for( vector< string >::const_iterator i = fileNames.begin(); i != fileNames.end();
       ++i )
  {
    // Try .gls and .gls.dz suffixes

    if( !( i->size() >= 4 && strcasecmp( i->c_str() + ( i->size() - 4 ), ".gls" ) == 0 )
        && !( i->size() >= 7 && strcasecmp( i->c_str() + ( i->size() - 7 ), ".gls.dz" ) == 0 ) )
      continue;

    unsigned atLine = 0; // Indicates current line in .gls, for debug purposes

    try
    {
      vector< string > dictFiles( 1, *i );

      string dictId = Dictionary::makeDictionaryId( dictFiles );

      // See if there's a zip file with resources present. If so, include it.

      string baseName = ( (*i)[ i->size() - 4 ] == '.' ) ?
               string( *i, 0, i->size() - 4 ) : string( *i, 0, i->size() - 7 );

      string zipFileName;

      if ( File::tryPossibleZipName( baseName + ".gls.files.zip", zipFileName ) ||
           File::tryPossibleZipName( baseName + ".gls.dz.files.zip", zipFileName ) ||
           File::tryPossibleZipName( baseName + ".GLS.FILES.ZIP", zipFileName ) ||
           File::tryPossibleZipName( baseName + ".GLS.DZ.FILES.ZIP", zipFileName ) )
        dictFiles.push_back( zipFileName );

      string indexFile = indicesDir + dictId;

      if ( Dictionary::needToRebuildIndex( dictFiles, indexFile ) ||
           indexIsOldOrBad( indexFile, zipFileName.size() ) )
      {
        GlsScanner scanner( *i );

        try { // Here we intercept any errors during the read to save line at
              // which the incident happened. We need alive scanner for that.

          // Building the index
          initializing.indexingDictionary( Utf8::encode( scanner.getDictionaryName() ) );

          gdDebug( "Gls: Building the index for dictionary: %s\n",
                   gd::toQString( scanner.getDictionaryName() ).toUtf8().data() );

          File::Class idx( indexFile, "wb" );

          IdxHeader idxHeader;

          memset( &idxHeader, 0, sizeof( idxHeader ) );

          // We write a dummy header first. At the end of the process the header
          // will be rewritten with the right values.

          idx.write( idxHeader );

          string dictionaryName = Utf8::encode( scanner.getDictionaryName() );

          idx.write( (uint32_t) dictionaryName.size() );
          idx.write( dictionaryName.data(), dictionaryName.size() );

          idxHeader.glsEncoding = scanner.getEncoding();

          IndexedWords indexedWords;

          ChunkedStorage::Writer chunks( idx );

          wstring curString;
          size_t curOffset;

          uint32_t articleCount = 0, wordCount = 0;

          for( ; ; )
          {
            // Find the headwords

            if ( !scanner.readNextLine( curString, curOffset ) )
              break; // Clean end of file

            if( curString.empty() )
              continue;

            uint32_t articleOffset = curOffset;

            // Parse headwords

            list< wstring > allEntryWords;
            wstring::size_type start_pos = 0, end_pos = 0;
            for( ; ; )
            {
              end_pos = curString.find( '|', start_pos );
              if( end_pos == wstring::npos )
              {
                wstring headword = curString.substr( start_pos );
                if( !headword.empty() )
                  allEntryWords.push_back( headword );
                break;
              }
              allEntryWords.push_back( curString.substr( start_pos, end_pos - start_pos ) );
              start_pos = end_pos + 1;
            }

            // Skip article body

            for( ; ; )
            {
              if( !scanner.readNextLine( curString, curOffset ) )
                break;
              if( curString.empty() )
                break;
            }

            // Insert new entry

            uint32_t descOffset = chunks.startNewBlock();
            chunks.addToBlock( &articleOffset, sizeof( articleOffset ) );

            uint32_t articleSize = curOffset - articleOffset;
            chunks.addToBlock( &articleSize, sizeof( articleSize ) );

            for( list< wstring >::iterator j = allEntryWords.begin();
                 j != allEntryWords.end(); ++j )
              indexedWords.addWord( *j, descOffset );

            ++articleCount;
            wordCount += allEntryWords.size();
          }

          // Finish with the chunks

          idxHeader.chunksOffset = chunks.finish();

          // Build index

          IndexInfo idxInfo = BtreeIndexing::buildIndex( indexedWords, idx );

          idxHeader.indexBtreeMaxElements = idxInfo.btreeMaxElements;
          idxHeader.indexRootOffset = idxInfo.rootOffset;

          indexedWords.clear(); // Release memory -- no need for this data

          // If there was a zip file, index it too

          if ( zipFileName.size() )
          {
            GD_DPRINTF( "Indexing zip file\n" );

            idxHeader.hasZipFile = 1;

            IndexedWords zipFileNames;
            IndexedZip zipFile;
            if( zipFile.openZipFile( QDir::fromNativeSeparators(
                                     FsEncoding::decode( zipFileName.c_str() ) ) ) )
                zipFile.indexFile( zipFileNames );

            if( !zipFileNames.empty() )
            {
              // Build the resulting zip file index

              IndexInfo idxInfo = BtreeIndexing::buildIndex( zipFileNames, idx );

              idxHeader.zipIndexBtreeMaxElements = idxInfo.btreeMaxElements;
              idxHeader.zipIndexRootOffset = idxInfo.rootOffset;
            }
            else
            {
              // Bad zip file -- no index (though the mark that we have one
              // remains)
              idxHeader.zipIndexBtreeMaxElements = 0;
              idxHeader.zipIndexRootOffset = 0;
            }
          }
          else
            idxHeader.hasZipFile = 0;

          // That concludes it. Update the header.

          idxHeader.signature = Signature;
          idxHeader.formatVersion = CurrentFormatVersion;
          idxHeader.zipSupportVersion = CurrentZipSupportVersion;

          idxHeader.articleCount = articleCount;
          idxHeader.wordCount = wordCount;

          idxHeader.langFrom = LangCoder::findIdForLanguage( scanner.getLangFrom() );
          idxHeader.langTo = LangCoder::findIdForLanguage( scanner.getLangTo() );
          if( idxHeader.langFrom == 0 && idxHeader.langTo == 0 )
          {
            // if no languages found, try dictionary's file name
            QPair<quint32,quint32> langs =
                LangCoder::findIdsForFilename( QString::fromStdString( dictFiles[ 0 ] ) );

            // if no languages found, try dictionary's name
            if ( langs.first == 0 || langs.second == 0 )
            {
              langs =
                LangCoder::findIdsForFilename( QString::fromStdString( dictionaryName ) );
            }
            idxHeader.langFrom = langs.first;
            idxHeader.langTo = langs.second;
          }

          idx.rewind();

          idx.write( &idxHeader, sizeof( idxHeader ) );
        } // In-place try for saving line count
        catch( ... )
        {
          atLine = scanner.getLinesRead();
          throw;
        }

      } // if need to rebuild
      dictionaries.push_back( new GlsDictionary( dictId,
                                                 indexFile,
                                                 dictFiles ) );
    }
    catch( std::exception & e )
    {
      gdWarning( "GLS dictionary reading failed: %s:%u, error: %s\n",
                 i->c_str(), atLine, e.what() );
    }
  }

  return dictionaries;
}

} // namespace Gls