refract encoding method

dsl_details.cc

@@ -895,7 +895,7 @@ DslScanner::DslScanner( string const & fileName ) THROW_SPEC( Ex, Iconv::Ex ):
 
   //iconv.reinit( encoding );
   codec = QTextCodec::codecForName(getEncodingNameFor(encoding));
-  initLineFeed(encoding);
+  lineFeed=Utf8::initLineFeed(encoding);
   // We now can use our own readNextLine() function
 
   wstring str;
@@ -1025,7 +1025,7 @@ bool DslScanner::readNextLine( wstring & out, size_t & offset, bool only_head_wo
     if(readBufferLeft<=0)
         return false;
 
-    int pos = Utf8::findFirstLinePosition(readBufferPtr,readBufferLeft, lineFeed,lineFeedLength);
+    int pos = Utf8::findFirstLinePosition(readBufferPtr,readBufferLeft, lineFeed.lineFeed,lineFeed.length);
     if(pos==-1)
         return false;
     QString line = codec->toUnicode(readBufferPtr, pos);
@@ -1083,31 +1083,6 @@ bool DslScanner::readNextLineWithoutComments( wstring & out, size_t & offset , b
 
 /////////////// DslScanner
 
-void DslScanner::initLineFeed(Utf8::Encoding e)
-{
-	switch (e)
-	{
-    case Utf8::Utf16LE:
-        lineFeed= new char[2] {0x0A,0};
-        lineFeedLength = 2;
-        break;
-    case Utf8::Utf16BE:
-        lineFeed = new char[2] { 0,0x0A};
-        lineFeedLength = 2;
-        break;
-    case Utf8::Windows1252:
-		
-    case Utf8::Windows1251:
-		
-    case Utf8::Utf8:
-		
-    case Utf8::Windows1250:
-	default:
-        lineFeedLength = 1;
-        lineFeed = new char[1] {0x0A};
-	}
-}
-
 void processUnsortedParts( wstring & str, bool strip )
 {
   int refCount = 0;

dsl_details.hh

@@ -24,6 +24,7 @@ using gd::wchar;
 using std::list;
 using std::vector;
 using Utf8::Encoding;
+using Utf8::LineFeed;
 
 
 
@@ -110,8 +111,7 @@ class DslScanner
   char readBuffer[ 65536 ];
   QTextStream* fragStream;
   char * readBufferPtr;
-  const char* lineFeed;
-  int lineFeedLength;
+  LineFeed lineFeed;
   size_t readBufferLeft;
   //qint64 pos;
   unsigned linesRead;
@@ -131,7 +131,6 @@ public:
   /// Returns the detected encoding of this file.
   Encoding getEncoding() const
   { return encoding; }
-  void initLineFeed(Encoding e);
 
   /// Returns the dictionary's name, as was read from file's headers.
   wstring const & getDictionaryName() const

gls.cc

@@ -59,6 +59,7 @@ using BtreeIndexing::WordArticleLink;
 using BtreeIndexing::IndexedWords;
 using BtreeIndexing::IndexInfo;
 using Utf8::Encoding;
+using Utf8::LineFeed;
 
 /////////////// GlsScanner
 
@@ -73,8 +74,7 @@ class GlsScanner
   char readBuffer[ 10000 ];
   char * readBufferPtr;
   size_t readBufferLeft;
-  const char* lineFeed;
-  int lineFeedLength;
+  LineFeed lineFeed;
   unsigned linesRead;
 
 public:
@@ -119,7 +119,6 @@ public:
   /// Reading begins from the first line after the headers (ones which end
   /// by the "### Glossary section:" line).
   bool readNextLine( wstring &, size_t & offset ) THROW_SPEC( Ex, Iconv::Ex );
-  void initLineFeed(Utf8::Encoding e);
   /// Returns the number of lines read so far from the file.
   unsigned getLinesRead() const
   { return linesRead; }
@@ -178,6 +177,7 @@ GlsScanner::GlsScanner( string const & fileName ) THROW_SPEC( Ex, Iconv::Ex ):
 
   codec = QTextCodec::codecForName(Utf8::getEncodingNameFor(encoding));
   // We now can use our own readNextLine() function
+  lineFeed = Utf8::initLineFeed(encoding);
 
   wstring str;
   wstring *currentField = 0;
@@ -243,30 +243,7 @@ GlsScanner::GlsScanner( string const & fileName ) THROW_SPEC( Ex, Iconv::Ex ):
     }
   }
 }
-void GlsScanner::initLineFeed(Utf8::Encoding e)
-{
-    switch (e)
-    {
-    case Utf8::Utf16LE:
-        lineFeed= new char[2] {0x0A,0};
-        lineFeedLength = 2;
-        break;
-    case Utf8::Utf16BE:
-        lineFeed = new char[2] { 0,0x0A};
-        lineFeedLength = 2;
-        break;
-    case Utf8::Windows1252:
 
-    case Utf8::Windows1251:
-
-    case Utf8::Utf8:
-
-    case Utf8::Windows1250:
-    default:
-        lineFeedLength = 1;
-        lineFeed = new char[1] {0x0A};
-    }
-}
 bool GlsScanner::readNextLine( wstring & out, size_t & offset ) THROW_SPEC( Ex,
                                                                        Iconv::Ex )
 {
@@ -286,7 +263,7 @@ bool GlsScanner::readNextLine( wstring & out, size_t & offset ) THROW_SPEC( Ex,
           int result = gzread( f, readBuffer + readBufferLeft,
                                sizeof( readBuffer ) - readBufferLeft );
 
-		  if (result == -1)
+          if (result == -1)
             throw exCantReadGlsFile();
 
           readBufferPtr = readBuffer;
@@ -296,7 +273,7 @@ bool GlsScanner::readNextLine( wstring & out, size_t & offset ) THROW_SPEC( Ex,
       if(readBufferLeft<=0)
           return false;
 
-      int pos = Utf8::findFirstLinePosition(readBufferPtr,readBufferLeft, lineFeed,lineFeedLength);
+      int pos = Utf8::findFirstLinePosition(readBufferPtr,readBufferLeft, lineFeed.lineFeed,lineFeed.length);
       if(pos==-1)
           return false;
       QString line = codec->toUnicode(readBufferPtr, pos);

utf8.cc

@@ -208,4 +208,31 @@ char const* getEncodingNameFor(Encoding e)
     }
 }
 
+LineFeed initLineFeed(Encoding e)
+{
+    LineFeed lf;
+	switch (e)
+	{
+	case Utf8::Utf16LE:
+        lf.lineFeed= new char[2]{ 0x0A,0 };
+        lf.length = 2;
+		break;
+	case Utf8::Utf16BE:
+        lf.lineFeed = new char[2]{ 0,0x0A };
+        lf.length = 2;
+		break;
+	case Utf8::Windows1252:
+
+	case Utf8::Windows1251:
+
+	case Utf8::Utf8:
+
+	case Utf8::Windows1250:
+	default:
+        lf.length = 1;
+        lf.lineFeed = new char[1]{ 0x0A };
+	}
+    return lf;
+}
+
 }

utf8.hh

@@ -2,6 +2,7 @@
  * Part of GoldenDict. Licensed under GPLv3 or later, see the LICENSE file */
 #ifndef __UTF8_HH_INCLUDED__
 #define __UTF8_HH_INCLUDED__
+
 #include <cstdio>
 #include <string>
 #include "cpp_features.hh"
@@ -55,5 +56,15 @@ bool isspace( int c );
 //get the first line in string s1. -1 if not found
 int findFirstLinePosition( char* s1,int s1length, const char* s2,int s2length);
 char const* getEncodingNameFor(Encoding e);
+
+struct LineFeed
+{
+	int length;
+	char* lineFeed;
+
+};
+
+LineFeed initLineFeed(Encoding e);
 }
+
 #endif