|
|
| version 1.49, 2004/07/07 09:50:25 | version 1.61, 2008/07/16 17:07:16 |
|---|---|
| Line 1 | Line 1 |
| /** @file | /** @file |
| Parser: Charset connection implementation. | Parser: Charset connection implementation. |
| Copyright(c) 2001-2004 ArtLebedev Group (http://www.artlebedev.com) | Copyright(c) 2001-2005 ArtLebedev Group (http://www.artlebedev.com) |
| Author: Alexander Petrosyan<paf@design.ru>(http://paf.design.ru) | Author: Alexander Petrosyan<paf@design.ru>(http://paf.design.ru) |
| */ | */ |
| Line 15 static const char * const IDENT_CHARSET_ | Line 15 static const char * const IDENT_CHARSET_ |
| #endif | #endif |
| //#define PA_PATCHED_LIBXML_BACKWARD | //#define PA_PATCHED_LIBXML_BACKWARD |
| #define PRECALCULATE_DEST_LENGTH | |
| // globals | // globals |
| Line 38 inline void prepare_case_tables(unsigned | Line 39 inline void prepare_case_tables(unsigned |
| unsigned char *lcc_table=tables+lcc_offset; | unsigned char *lcc_table=tables+lcc_offset; |
| unsigned char *fcc_table=tables+fcc_offset; | unsigned char *fcc_table=tables+fcc_offset; |
| for(int i=0; i<0x100; i++) | for(int i=0; i<0x100; i++) |
| lcc_table[i]=fcc_table[i]=i; | lcc_table[i]=fcc_table[i]=(unsigned char)i; |
| } | } |
| inline void cstr2ctypes(unsigned char *tables, const unsigned char *cstr, | inline void cstr2ctypes(unsigned char *tables, const unsigned char *cstr, |
| unsigned char bit) { | unsigned char bit) { |
| Line 134 void Charset::load_definition(Request_ch | Line 135 void Charset::load_definition(Request_ch |
| continue; | continue; |
| // char white-space digit hex-digit letter word lowercase unicode1 unicode2 | // char white-space digit hex-digit letter word lowercase unicode1 unicode2 |
| unsigned int c=0; | unsigned char c=0; |
| char *cell; | char *cell; |
| for(int column=0; (cell=lsplit(&row, '\t')); column++) { | for(int column=0; (cell=lsplit(&row, '\t')); column++) { |
| switch(column) { | switch(column) { |
| case 0: c=to_wchar_code(cell); break; | case 0: c=(unsigned char)to_wchar_code(cell); break; |
| // pcre_tables | // pcre_tables |
| case 1: element2ctypes(c, to_bool(cell), pcre_tables, ctype_space, cbit_space); break; | case 1: element2ctypes(c, to_bool(cell), pcre_tables, ctype_space, cbit_space); break; |
| case 2: element2ctypes(c, to_bool(cell), pcre_tables, ctype_digit, cbit_digit); break; | case 2: element2ctypes(c, to_bool(cell), pcre_tables, ctype_digit, cbit_digit); break; |
| case 3: element2ctypes(c, to_bool(cell), pcre_tables, ctype_xdigit); break; | case 3: element2ctypes(c, to_bool(cell), pcre_tables, ctype_xdigit); break; |
| case 4: element2ctypes(c, to_bool(cell), pcre_tables, ctype_letter); break; | case 4: element2ctypes(c, to_bool(cell), pcre_tables, ctype_letter); break; |
| case 5: element2ctypes(c, to_bool(cell), pcre_tables, ctype_word, cbit_word); break; | case 5: element2ctypes(c, to_bool(cell), pcre_tables, ctype_word, cbit_word); break; |
| case 6: element2case(c, to_wchar_code(cell), pcre_tables); break; | case 6: element2case(c, (unsigned char)to_wchar_code(cell), pcre_tables); break; |
| case 7: | case 7: |
| case 8: | case 8: |
| // charset | // charset |
| if(tables.toTableSize>MAX_CHARSET_UNI_CODES) | if(tables.toTableSize>MAX_CHARSET_UNI_CODES) |
| throw Exception("parser.runtime", | throw Exception(PARSER_RUNTIME, |
| &afile_spec, | &afile_spec, |
| "charset must contain not more then %d unicode values", MAX_CHARSET_UNI_CODES); | "charset must contain not more then %d unicode values", MAX_CHARSET_UNI_CODES); |
| Line 187 void Charset::sort_ToTable() { | Line 188 void Charset::sort_ToTable() { |
| //fclose(f); | //fclose(f); |
| } | } |
| // @todo: precache for spedup searching | |
| static XMLByte xlatOneTo(const XMLCh toXlat, | static XMLByte xlatOneTo(const XMLCh toXlat, |
| const Charset::Tables& tables, | const Charset::Tables& tables, |
| XMLByte not_found) { | XMLByte not_found) { |
| Line 316 static int transcodeToUTF8(const XMLByte | Line 318 static int transcodeToUTF8(const XMLByte |
| // here, so bump up the output pointer and work down as we go. | // here, so bump up the output pointer and work down as we go. |
| outPtr+= encodedBytes; | outPtr+= encodedBytes; |
| switch(encodedBytes) { | switch(encodedBytes) { |
| case 6: *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL); | case 6: *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL); |
| curVal>>= 6; | curVal>>= 6; |
| case 5: *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL); | case 5: *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL); |
| curVal>>= 6; | curVal>>= 6; |
| case 4: *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL); | case 4: *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL); |
| curVal>>= 6; | curVal>>= 6; |
| case 3: *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL); | case 3: *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL); |
| curVal>>= 6; | curVal>>= 6; |
| case 2: *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL); | case 2: *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL); |
| curVal>>= 6; | curVal>>= 6; |
| case 1: *--outPtr = XMLByte(curVal | gFirstByteMark[encodedBytes]); | case 1: *--outPtr = XMLByte(curVal | gFirstByteMark[encodedBytes]); |
| } | } |
| // Add the encoded bytes back in again to indicate we've eaten them | // Add the encoded bytes back in again to indicate we've eaten them |
| Line 364 static int transcodeFromUTF8(const XMLBy | Line 366 static int transcodeFromUTF8(const XMLBy |
| const XMLByte firstByte =* srcPtr; | const XMLByte firstByte =* srcPtr; |
| // Special-case ASCII, which is a leading byte value of<= 127 | // Special-case ASCII, which is a leading byte value of<= 127 |
| if(firstByte<= 127) { | if(firstByte<=127) { |
| *outPtr++= firstByte; | *outPtr++= firstByte; |
| srcPtr++; | srcPtr++; |
| continue; | continue; |
| Line 404 static int transcodeFromUTF8(const XMLBy | Line 406 static int transcodeFromUTF8(const XMLBy |
| if(XMLByte xlat=xlatOneTo(tmpVal, tables, 0)) | if(XMLByte xlat=xlatOneTo(tmpVal, tables, 0)) |
| *outPtr++=xlat; | *outPtr++=xlat; |
| else { | else { |
| if(tmpVal & 0xFFFFFF00) | outPtr+=sprintf((char *)outPtr, "&#%u;", tmpVal); // &#decimal; |
| goto fail; | |
| outPtr+=sprintf((char *)outPtr, "&#%d;", tmpVal); // &#decimal; | |
| } | } |
| } else { | } else { |
| fail: | |
| const XMLByte* recoverPtr=srcPtr-trailingBytes-1; | const XMLByte* recoverPtr=srcPtr-trailingBytes-1; |
| for(uint i=0; i<=trailingBytes; i++) | for(uint i=0; i<=trailingBytes; i++) |
| outPtr+=sprintf((char*)outPtr, "%%%02X", *recoverPtr++); | outPtr+=sprintf((char*)outPtr, "%%%02X", *recoverPtr++); |
| Line 433 of ocetes consumed. | Line 432 of ocetes consumed. |
| return 0; | return 0; |
| } | } |
| /// @todo not so memory-hungry with prescan | static bool is_escaped(char c){ |
| return | |
| !(c<=127 | |
| && ( | |
| ((c>='0') && (c<='9')) | |
| || ((c>='A') && (c<='Z')) | |
| || ((c>='a') && (c<='z')) | |
| || strchr("*@-_+./", c)!=0 | |
| )); | |
| } | |
| // read one utf8 character, return number of bytes needed for store it | |
| static unsigned int readChar(const XMLByte*& srcPtr, const XMLByte* srcEnd, XMLByte& firstByte, XMLCh& UTF8Char){ | |
| if(!srcPtr || !*srcPtr || srcPtr>=srcEnd) | |
| return 0; | |
| firstByte=*srcPtr; | |
| if(firstByte<=127){ | |
| UTF8Char=firstByte; | |
| srcPtr++; | |
| return 1; | |
| } | |
| unsigned int trailingBytes=gUTFBytes[firstByte]; | |
| if(srcPtr+trailingBytes>=srcEnd){ | |
| return 0; // not enough bytes in source string for reading | |
| } | |
| uint tmpVal=0; | |
| switch(trailingBytes){ | |
| case 5: tmpVal+=*srcPtr++; tmpVal<<=6; | |
| case 4: tmpVal+=*srcPtr++; tmpVal<<=6; | |
| case 3: tmpVal+=*srcPtr++; tmpVal<<=6; | |
| case 2: tmpVal+=*srcPtr++; tmpVal<<=6; | |
| case 1: tmpVal+=*srcPtr++; tmpVal<<=6; | |
| case 0: tmpVal+=*srcPtr++; | |
| } | |
| tmpVal-=gUTFOffsets[trailingBytes]; | |
| UTF8Char=tmpVal; | |
| return trailingBytes+1; | |
| } | |
| static unsigned int readChar(const XMLByte*& srcPtr, const XMLByte* srcEnd){ | |
| XMLByte firstByte; | |
| XMLCh UTF8Char; | |
| return readChar(srcPtr, srcEnd, firstByte, UTF8Char); | |
| } | |
| // read char, return number of bytes needed for store it as UTF8 | |
| static unsigned int readChar(const XMLByte*& srcPtr, const XMLByte* srcEnd, XMLByte& firstByte, XMLCh& UTF8Char, const Charset::Tables& tables){ | |
| if(!srcPtr || !*srcPtr || srcPtr>=srcEnd) | |
| return 0; | |
| firstByte=*srcPtr++; | |
| UTF8Char=tables.fromTable[firstByte]; | |
| if(UTF8Char<0x80) | |
| return 1; | |
| else if(UTF8Char<0x800) | |
| return 2; | |
| else if(UTF8Char<0x10000) | |
| return 3; | |
| else if(UTF8Char<0x200000) | |
| return 4; | |
| else if(UTF8Char<0x4000000) | |
| return 5; | |
| else if(UTF8Char<= 0x7FFFFFFF) | |
| return 6; | |
| // will use the replacement character '?' | |
| firstByte=0; | |
| return 1; | |
| } | |
| static int escape(const XMLByte* srcData, size_t& srcLen, | |
| XMLByte* toFill, size_t& toFillLen) { | |
| const XMLByte* srcPtr=srcData; | |
| const XMLByte* srcEnd=srcData+srcLen; | |
| XMLByte* outPtr=toFill; | |
| XMLByte* outEnd=toFill+toFillLen; | |
| XMLByte firstByte; | |
| XMLCh UTF8Char; | |
| uint charSize; | |
| // loop until we either run out of input data, or room to store | |
| while((outPtr < outEnd) && (charSize=readChar(srcPtr, srcEnd, firstByte, UTF8Char))){ | |
| if(charSize==1){ | |
| if(is_escaped(firstByte)) // %XX | |
| outPtr+=sprintf((char*)outPtr, "%%%02X", firstByte); | |
| else | |
| *outPtr++=firstByte; | |
| } else | |
| outPtr+=sprintf((char*)outPtr, "%%u%04X", UTF8Char); // %uXXXX | |
| } | |
| // Update the bytes eaten | |
| srcLen=srcPtr-srcData; | |
| // Return the characters read | |
| toFillLen=outPtr-toFill; | |
| return 0; | |
| } | |
| static int escape(const XMLByte* srcData, size_t& srcLen, | |
| XMLByte *toFill, size_t& toFillLen, | |
| const Charset::Tables& tables) { | |
| const XMLByte* srcPtr=srcData; | |
| const XMLByte* srcEnd=srcData+srcLen; | |
| XMLByte* outPtr=toFill; | |
| //XMLByte* outEnd=toFill+toFillLen; | |
| XMLByte firstByte; | |
| XMLCh UTF8Char; | |
| uint charSize; | |
| while(charSize=readChar(srcPtr, srcEnd, firstByte, UTF8Char, tables)){ | |
| if(charSize==1){ | |
| if(firstByte){ | |
| if(is_escaped(firstByte)) // %XX | |
| outPtr+=sprintf((char*)outPtr, "%%%02X", firstByte); | |
| else | |
| *outPtr++=firstByte; | |
| } else // add replacement char '?' | |
| *outPtr++='?'; | |
| } else | |
| outPtr+=sprintf((char*)outPtr, "%%u%04X", UTF8Char); // %uXXXX | |
| } | |
| // Update the bytes eaten | |
| srcLen = srcPtr - srcData; | |
| // Return the characters read | |
| toFillLen = outPtr - toFill; | |
| return 0; | |
| } | |
| String::C Charset::escape(const String::C src, const Charset& source_charset){ | |
| size_t src_length=src.length; | |
| if(!src_length) | |
| return String::C("", 0); | |
| #ifdef PRECALCULATE_DEST_LENGTH | |
| size_t dest_length=0; | |
| const XMLByte* srcPtr=(XMLByte*)src.str; | |
| const XMLByte* srcEnd=srcPtr+src_length; | |
| XMLByte firstByte; | |
| XMLCh UTF8Char; | |
| if(source_charset.isUTF8()){ | |
| while(uint charSize=readChar(srcPtr, srcEnd, firstByte, UTF8Char)){ | |
| if(charSize==1) | |
| dest_length+=!is_escaped(firstByte)?1:3/*%XX*/; | |
| else | |
| dest_length+=6; // '%uXXXX' | |
| } | |
| } else { | |
| while(uint charSize=readChar(srcPtr, srcEnd, firstByte, UTF8Char, source_charset.tables)){ | |
| if(charSize==1) | |
| dest_length+=(!firstByte/*replacement char '?'*/ || !is_escaped(firstByte))?1:3/*'%XX'*/; | |
| else | |
| dest_length+=6; // '%uXXXX' | |
| } | |
| } | |
| #else | |
| size_t dest_length=src_length*6; // enough for %uXXXX but too memory-hungry | |
| #endif | |
| //throw Exception(0,0,"%u",dest_length); | |
| #ifndef NDEBUG | |
| size_t saved_dest_length=dest_length; | |
| #endif | |
| XMLByte *dest_body=new(PointerFreeGC) XMLByte[dest_length+1/*for terminator*/]; | |
| int status; | |
| if(source_charset.isUTF8()){ | |
| status=::escape((XMLByte *)src.str, src_length, dest_body, dest_length); | |
| } else { | |
| status=::escape((XMLByte *)src.str, src_length, dest_body, dest_length, source_charset.tables); | |
| } | |
| if(status<0) | |
| throw Exception(0, | |
| 0, | |
| "Charset::escapeString buffer overflow"); | |
| assert(dest_length<=saved_dest_length); | |
| dest_body[dest_length]=0; // terminator | |
| return String::C((char*)dest_body, dest_length); | |
| } | |
| const String::C Charset::transcodeToUTF8(const String::C src) const { | const String::C Charset::transcodeToUTF8(const String::C src) const { |
| size_t src_length=src.length; | size_t src_length=src.length; |
| size_t dest_length=src.length*6/*so that surly enough, max utf8 seq len=6*/; | |
| #ifdef PRECALCULATE_DEST_LENGTH | |
| size_t dest_length=0; | |
| const XMLByte* srcPtr=(XMLByte*)src.str; | |
| const XMLByte* srcEnd=srcPtr+src_length; | |
| XMLByte firstByte; | |
| XMLCh UTF8Char; | |
| while(uint charSize=readChar(srcPtr, srcEnd, firstByte, UTF8Char, tables)) | |
| dest_length+=charSize; | |
| #else | |
| size_t dest_length=src_length*6; // so that surly enough (max utf8 seq len=6) but too memory-hyngry | |
| #endif | |
| //throw Exception(0,0,"%u",dest_length); | |
| #ifndef NDEBUG | #ifndef NDEBUG |
| size_t saved_dest_length=dest_length; | size_t saved_dest_length=dest_length; |
| #endif | #endif |
| Line 450 const String::C Charset::transcodeToUTF8 | Line 660 const String::C Charset::transcodeToUTF8 |
| 0, | 0, |
| "Charset::transcodeToUTF8 buffer overflow"); | "Charset::transcodeToUTF8 buffer overflow"); |
| assert(dest_length<=saved_dest_length); dest_body[dest_length]=0; // terminator | assert(dest_length<=saved_dest_length); |
| dest_body[dest_length]=0; // terminator | |
| return String::C((char*)dest_body, dest_length); | return String::C((char*)dest_body, dest_length); |
| } | } |
| Line 474 static XMLCh change_case_UTF8(const XMLC | Line 685 static XMLCh change_case_UTF8(const XMLC |
| return src; | return src; |
| } | } |
| static void store_UTF8(XMLCh src, XMLByte*& outPtr ) { | static void store_UTF8(XMLCh src, XMLByte*& outPtr){ |
| if(!src) { | if(!src) { |
| // use the replacement character | // use the replacement character |
| *outPtr++= '?'; | *outPtr++= '?'; |
| Line 539 void change_case_UTF8(const XMLByte* src | Line 750 void change_case_UTF8(const XMLByte* src |
| // Get the next leading byte out | // Get the next leading byte out |
| const XMLByte firstByte =* srcPtr; | const XMLByte firstByte =* srcPtr; |
| if(firstByte<= 127) { | if(firstByte<=127) { |
| change_case_UTF8(firstByte, outPtr, table); | change_case_UTF8(firstByte, outPtr, table); |
| srcPtr++; | srcPtr++; |
| continue; | continue; |
| Line 583 void change_case_UTF8(const XMLByte* src | Line 794 void change_case_UTF8(const XMLByte* src |
| "change_case_UTF8 error: end pointers do not match"); | "change_case_UTF8 error: end pointers do not match"); |
| } | } |
| static size_t getDecNumLength(XMLCh UTF8Char){ | |
| return | |
| (UTF8Char < 100) | |
| ?2 | |
| :(UTF8Char < 1000) | |
| ?3 | |
| :(UTF8Char < 10000) | |
| ?4 | |
| :5; | |
| } | |
| const String::C Charset::transcodeFromUTF8(const String::C src) const { | const String::C Charset::transcodeFromUTF8(const String::C src) const { |
| size_t src_length=src.length; | size_t src_length=src.length; |
| size_t dest_length=src.length*6/*so that surly enough, "ÿ" has max ratio */; | |
| #ifdef PRECALCULATE_DEST_LENGTH | |
| size_t dest_length=0; | |
| const XMLByte* srcPtr=(XMLByte*)src.str; | |
| const XMLByte* srcEnd=srcPtr+src_length; | |
| XMLByte firstByte; | |
| XMLCh UTF8Char; | |
| while(uint charSize=readChar(srcPtr, srcEnd, firstByte, UTF8Char)){ | |
| if(charSize==1) | |
| dest_length++; | |
| else | |
| dest_length+=(UTF8Char & 0xFFFF0000) | |
| ?charSize*3 // '%XX' for each byte | |
| :(xlatOneTo(UTF8Char, tables, 0)!=0) | |
| ?1 // can convert it to single char | |
| :getDecNumLength(UTF8Char)+3; // &#XX; - &#XXXXX; | |
| } | |
| #else | |
| // so that surly enough, "&#XXX;" has max ratio (huh? 8 bytes needed for '&#XXXXX;') | |
| size_t dest_length=src_length*6; | |
| #endif | |
| //throw Exception(0,0,"%u",dest_length); | |
| #ifndef NDEBUG | #ifndef NDEBUG |
| size_t saved_dest_length=dest_length; | size_t saved_dest_length=dest_length; |
| #endif | #endif |
| Line 600 const String::C Charset::transcodeFromUT | Line 844 const String::C Charset::transcodeFromUT |
| 0, | 0, |
| "Charset::transcodeFromUTF8 buffer overflow"); | "Charset::transcodeFromUTF8 buffer overflow"); |
| assert(dest_length<=saved_dest_length); dest_body[dest_length]=0; // terminator | assert(dest_length<=saved_dest_length); |
| dest_body[dest_length]=0; // terminator | |
| return String::C((char*)dest_body, dest_length); | return String::C((char*)dest_body, dest_length); |
| } | } |
| Line 627 const String::C Charset::transcodeToChar | Line 872 const String::C Charset::transcodeToChar |
| } | } |
| } | } |
| void Charset::store_Char(XMLByte*& outPtr, XMLCh src, XMLByte not_found){ | |
| if(isUTF8()) | |
| store_UTF8(src, outPtr); | |
| else if(char ch=xlatOneTo(src, tables, not_found)) | |
| *outPtr++=ch; | |
| } | |
| #ifdef XML | #ifdef XML |
| static const Charset::Tables* tables[MAX_CHARSETS]; | static const Charset::Tables* tables[MAX_CHARSETS]; |
| Line 723 void Charset::initTranscoder(const Strin | Line 975 void Charset::initTranscoder(const Strin |
| xmlCharEncodingHandler& Charset::transcoder(const String::Body NAME) { | xmlCharEncodingHandler& Charset::transcoder(const String::Body NAME) { |
| if(!ftranscoder) | if(!ftranscoder) |
| throw Exception("parser.runtime", | throw Exception(PARSER_RUNTIME, |
| new String(NAME, String::L_TAINTED), | new String(NAME, String::L_TAINTED), |
| "unsupported encoding"); | "unsupported encoding"); |
| return *ftranscoder; | return *ftranscoder; |
| } | } |
| String::C Charset::transcode_cstr(xmlChar* s) { | String::C Charset::transcode_cstr(const xmlChar* s) { |
| if(!s) | if(!s) |
| return String::C("", 0); | return String::C("", 0); |
| int inlen=strlen((const char*)s); | int inlen=strlen((const char*)s); |
| int outlen=inlen; // max | int outlen=inlen*6/*strlen("ÿ")*/; // max |
| #ifndef NDEBUG | #ifndef NDEBUG |
| int saved_outlen=outlen; | int saved_outlen=outlen; |
| #endif | #endif |
| Line 761 String::C Charset::transcode_cstr(xmlCha | Line 1013 String::C Charset::transcode_cstr(xmlCha |
| assert(outlen<=saved_outlen); out[outlen]=0; | assert(outlen<=saved_outlen); out[outlen]=0; |
| return String::C(out, outlen); | return String::C(out, outlen); |
| } | } |
| const String& Charset::transcode(xmlChar* s) { | const String& Charset::transcode(const xmlChar* s) { |
| String::C cstr=transcode_cstr(s); | |
| return *new String(cstr.str, cstr.length, true); | |
| } | |
| String::C Charset::transcode_cstr(GdomeDOMString* s) { | |
| return s?transcode_cstr(BAD_CAST s->str) | |
| :String::C("", 0); | |
| } | |
| const String& Charset::transcode(GdomeDOMString* s) { | |
| String::C cstr=transcode_cstr(s); | String::C cstr=transcode_cstr(s); |
| return *new String(cstr.str, cstr.length, true); | return *new String(cstr.str, cstr.length, true); |
| } | } |
| Line 783 xmlChar* Charset::transcode_buf2xchar(co | Line 1027 xmlChar* Charset::transcode_buf2xchar(co |
| int saved_outlen; | int saved_outlen; |
| #endif | #endif |
| if(xmlCharEncodingInputFunc input=transcoder(FNAME).input) { | if(xmlCharEncodingInputFunc input=transcoder(FNAME).input) { |
| outlen=buf_size*6/*max*/; | outlen=buf_size*6/*max UTF8 bytes per char*/; |
| #ifndef NDEBUG | #ifndef NDEBUG |
| saved_outlen=outlen; | saved_outlen=outlen; |
| #endif | #endif |
| Line 813 xmlChar* Charset::transcode_buf2xchar(co | Line 1057 xmlChar* Charset::transcode_buf2xchar(co |
| assert(outlen<=saved_outlen); out[outlen]=0; | assert(outlen<=saved_outlen); out[outlen]=0; |
| return out; | return out; |
| } | } |
| GdomeDOMString_auto_ptr Charset::transcode_buf2dom(const char* buf, size_t buf_size) { | xmlChar* Charset::transcode(const String& s) { |
| return GdomeDOMString_auto_ptr(transcode_buf2xchar(buf, buf_size)); | |
| } | |
| GdomeDOMString_auto_ptr Charset::transcode(const String& s) { | |
| const char* cstr=s.cstr(String::L_UNSPECIFIED); | const char* cstr=s.cstr(String::L_UNSPECIFIED); |
| return transcode_buf2dom(cstr, strlen(cstr)); | return transcode_buf2xchar(cstr, strlen(cstr)); |
| } | } |
| GdomeDOMString_auto_ptr Charset::transcode(const String::Body s) { | xmlChar* Charset::transcode(const String::Body s) { |
| const char* cstr=s.cstr(); | const char* cstr=s.cstr(); |
| return transcode_buf2dom(cstr, s.length()); | return transcode_buf2xchar(cstr, s.length()); |
| } | } |
| #endif | #endif |
| Line 871 static void transcode_pair(const String: | Line 1112 static void transcode_pair(const String: |
| *info->source_transcoder, | *info->source_transcoder, |
| *info->dest_transcoder); | *info->dest_transcoder); |
| } | } |
| void Charset::transcode(HashStringString& src, | void Charset::transcode(HashStringString& src, |
| const Charset& source_transcoder, | const Charset& source_transcoder, |
| const Charset& dest_transcoder) { | const Charset& dest_transcoder) { |
| Transcode_pair_info info={&source_transcoder, &dest_transcoder}; | Transcode_pair_info info={&source_transcoder, &dest_transcoder}; |
| src.for_each_ref(transcode_pair, &info); | src.for_each_ref<Transcode_pair_info*>(transcode_pair, &info); |
| } | |
| size_t getUTF8BytePos(const XMLByte* srcBegin, const XMLByte* srcEnd, size_t charPos){ | |
| const XMLByte* ptr=srcBegin; | |
| while(charPos-- && readChar(ptr, srcEnd)); | |
| return ptr-srcBegin; | |
| } | |
| size_t getUTF8CharPos(const XMLByte* srcBegin, const XMLByte* srcEnd, size_t bytePos){ | |
| size_t charPos=0; | |
| const XMLByte* ptr=srcBegin; | |
| const XMLByte* ptrEnd=srcBegin+bytePos; | |
| while(readChar(ptr, srcEnd)){ | |
| if(ptr>ptrEnd) | |
| return charPos; | |
| charPos++; | |
| } | |
| // scan till end but position in bytes still too low | |
| throw Exception(0, | |
| 0, | |
| "Error convertion byte pos to char pos"); | |
| } | |
| size_t lengthUTF8(const XMLByte* srcBegin, const XMLByte* srcEnd){ | |
| size_t size=0; | |
| while(readChar(srcBegin, srcEnd)) | |
| size++; | |
| return size; | |
| } | } |