|
|
| version 1.33.2.19.2.16, 2003/04/08 12:47:31 | version 1.37, 2003/09/25 09:15:03 |
|---|---|
| Line 68 static void element2case(unsigned char f | Line 68 static void element2case(unsigned char f |
| // methods | // methods |
| extern "C" unsigned char pcre_default_tables[]; // pcre/chartables.c | extern "C" unsigned char pcre_default_tables[]; // pcre/chartables.c |
| Charset::Charset(const String& aname, const String* afile_spec): | Charset::Charset(Request_charsets* charsets, const String::Body ANAME, const String* afile_spec): |
| fname(aname), | FNAME(ANAME), |
| fname_cstr(aname.cstrm()) { | FNAME_CSTR(ANAME.cstrm()) { |
| for(char *c=fname_cstr; *c; c++) | |
| *c = toupper(*c); | |
| if(afile_spec) { | if(afile_spec) { |
| fisUTF8=false; | fisUTF8=false; |
| load_definition(*afile_spec); | load_definition(*charsets, *afile_spec); |
| #ifdef XML | #ifdef XML |
| addEncoding(fname_cstr); | addEncoding(FNAME_CSTR); |
| #endif | #endif |
| } else { | } else { |
| fisUTF8=true; | fisUTF8=true; |
| Line 88 Charset::Charset(const String& aname, co | Line 85 Charset::Charset(const String& aname, co |
| } | } |
| #ifdef XML | #ifdef XML |
| initTranscoder(fname, fname_cstr); | initTranscoder(FNAME, FNAME_CSTR); |
| #endif | #endif |
| } | } |
| void Charset::load_definition(const String& afile_spec) { | void Charset::load_definition(Request_charsets& charsets, const String& afile_spec) { |
| // pcre_tables | // pcre_tables |
| // lowcase, flipcase, bits digit+word+whitespace, masks | // lowcase, flipcase, bits digit+word+whitespace, masks |
| Line 110 void Charset::load_definition(const Stri | Line 107 void Charset::load_definition(const Stri |
| tables.toTableSize++; | tables.toTableSize++; |
| // loading text | // loading text |
| char *data=file_read_text(UTF8_charset, afile_spec); | char *data=file_read_text(charsets, afile_spec); |
| // ignore header | // ignore header |
| getrow(&data); | getrow(&data); |
| Line 177 void Charset::sort_ToTable() { | Line 174 void Charset::sort_ToTable() { |
| } | } |
| 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) { |
| unsigned int lowOfs = 0; | unsigned int lowOfs = 0; |
| unsigned int hiOfs = tables.toTableSize - 1; | unsigned int hiOfs = tables.toTableSize - 1; |
| XMLByte curByte = 0; | XMLByte curByte = 0; |
| do { | do { |
| // Calc the mid point of the low and high offset. | // Calc the mid point of the low and high offset. |
| const unsigned int midOfs =((hiOfs - lowOfs) / 2)+lowOfs; | const unsigned int midOfs =((hiOfs - lowOfs) / 2)+lowOfs; |
| // If our test char is greater than the mid point char, then | // If our test char is greater than the mid point char, then |
| // we move up to the upper half. Else we move to the lower | // we move up to the upper half. Else we move to the lower |
| // half. If its equal, then its our guy. | // half. If its equal, then its our guy. |
| if(toXlat>tables.toTable[midOfs].intCh) | if(toXlat>tables.toTable[midOfs].intCh) |
| lowOfs = midOfs; | lowOfs = midOfs; |
| else if(toXlat<tables.toTable[midOfs].intCh) | else if(toXlat<tables.toTable[midOfs].intCh) |
| hiOfs = midOfs; | hiOfs = midOfs; |
| else | else |
| return tables.toTable[midOfs].extCh; | return tables.toTable[midOfs].extCh; |
| } while(lowOfs+1<hiOfs); | } while(lowOfs+1<hiOfs); |
| return not_found; | return not_found; |
| } | } |
| String::C Charset::transcode(const String::C src, | String::C Charset::transcode(const String::C src, |
| const Charset& source_charset, | const Charset& source_charset, |
| const Charset& dest_charset) { | const Charset& dest_charset) { |
| if(!src.length) | if(!src.length) |
| return String::C(0, 0); | return String::C("", 0); |
| switch((source_charset.isUTF8()?0x10:0x00)|(dest_charset.isUTF8()?0x01:0x00)) { | switch((source_charset.isUTF8()?0x10:0x00)|(dest_charset.isUTF8()?0x01:0x00)) { |
| default: // 0x00 | default: // 0x00 |
| Line 259 static const XMLByte gFirstByteMark[7] = | Line 256 static const XMLByte gFirstByteMark[7] = |
| 0x00, 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC | 0x00, 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC |
| }; | }; |
| static int transcodeToUTF8( | static int transcodeToUTF8(const XMLByte* srcData, size_t& srcLen, |
| const XMLByte* srcData, size_t& srcLen, | |
| XMLByte *toFill, size_t& toFillLen, | XMLByte *toFill, size_t& toFillLen, |
| const Charset::Tables& tables) { | const Charset::Tables& tables) { |
| const XMLByte* srcPtr=srcData; | const XMLByte* srcPtr=srcData; |
| Line 268 static int transcodeToUTF8( | Line 264 static int transcodeToUTF8( |
| XMLByte* outPtr=toFill; | XMLByte* outPtr=toFill; |
| XMLByte* outEnd=toFill+toFillLen; | XMLByte* outEnd=toFill+toFillLen; |
| while(srcPtr<srcEnd) { | while(srcPtr<srcEnd) { |
| uint curVal = tables.fromTable[*srcPtr]; | uint curVal = tables.fromTable[*srcPtr]; |
| if(!curVal) { | if(!curVal) { |
| // use the replacement character | // use the replacement character |
| *outPtr++= '?'; | *outPtr++= '?'; |
| srcPtr++; | srcPtr++; |
| continue; | continue; |
| } | } |
| // Figure out how many bytes we need | |
| unsigned int encodedBytes; | |
| if(curVal<0x80) | |
| encodedBytes = 1; | |
| else if(curVal<0x800) | |
| encodedBytes = 2; | |
| else if(curVal<0x10000) | |
| encodedBytes = 3; | |
| else if(curVal<0x200000) | |
| encodedBytes = 4; | |
| else if(curVal<0x4000000) | |
| encodedBytes = 5; | |
| else if(curVal<= 0x7FFFFFFF) | |
| encodedBytes = 6; | |
| else { | |
| // use the replacement character | |
| *outPtr++= '?'; | |
| srcPtr++; | |
| continue; | |
| } | |
| // If we cannot fully get this char into the output buffer | |
| if (outPtr + encodedBytes > outEnd) | |
| break; | |
| // We can do it, so update the source index | |
| srcPtr++; | |
| // And spit out the bytes. We spit them out in reverse order | |
| // here, so bump up the output pointer and work down as we go. | |
| outPtr+= encodedBytes; | |
| switch(encodedBytes) { | |
| case 6: *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL); | |
| curVal>>= 6; | |
| case 5: *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL); | |
| curVal>>= 6; | |
| case 4: *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL); | |
| curVal>>= 6; | |
| case 3: *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL); | |
| curVal>>= 6; | |
| case 2: *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL); | |
| curVal>>= 6; | |
| case 1: *--outPtr = XMLByte(curVal | gFirstByteMark[encodedBytes]); | |
| } | |
| // Add the encoded bytes back in again to indicate we've eaten them | |
| outPtr+= encodedBytes; | |
| } | |
| // Update the bytes eaten | |
| srcLen = srcPtr - srcData; | |
| // Return the characters read | // Figure out how many bytes we need |
| toFillLen = outPtr - toFill; | unsigned int encodedBytes; |
| if(curVal<0x80) | |
| encodedBytes = 1; | |
| else if(curVal<0x800) | |
| encodedBytes = 2; | |
| else if(curVal<0x10000) | |
| encodedBytes = 3; | |
| else if(curVal<0x200000) | |
| encodedBytes = 4; | |
| else if(curVal<0x4000000) | |
| encodedBytes = 5; | |
| else if(curVal<= 0x7FFFFFFF) | |
| encodedBytes = 6; | |
| else { | |
| // use the replacement character | |
| *outPtr++= '?'; | |
| srcPtr++; | |
| continue; | |
| } | |
| // If we cannot fully get this char into the output buffer | |
| if (outPtr + encodedBytes > outEnd) | |
| break; | |
| // We can do it, so update the source index | |
| srcPtr++; | |
| // And spit out the bytes. We spit them out in reverse order | |
| // here, so bump up the output pointer and work down as we go. | |
| outPtr+= encodedBytes; | |
| switch(encodedBytes) { | |
| case 6: *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL); | |
| curVal>>= 6; | |
| case 5: *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL); | |
| curVal>>= 6; | |
| case 4: *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL); | |
| curVal>>= 6; | |
| case 3: *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL); | |
| curVal>>= 6; | |
| case 2: *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL); | |
| curVal>>= 6; | |
| case 1: *--outPtr = XMLByte(curVal | gFirstByteMark[encodedBytes]); | |
| } | |
| // Add the encoded bytes back in again to indicate we've eaten them | |
| outPtr+= encodedBytes; | |
| } | |
| // Update the bytes eaten | |
| srcLen = srcPtr - srcData; | |
| // Return the characters read | |
| toFillLen = outPtr - toFill; | |
| //return srcPtr==srcEnd?(int)toFillLen:-1; | //return srcPtr==srcEnd?(int)toFillLen:-1; |
| /* | /* |
| xmlCharEncodingInputFunc | xmlCharEncodingInputFunc |
| Line 343 of ocetes consumed. | Line 339 of ocetes consumed. |
| return 0; | return 0; |
| } | } |
| /// @todo digital entites only when xml/html output [at output in html/xml mode, in html part of a letter] | /// @todo digital entites only when xml/html output [at output in html/xml mode, in html part of a letter] |
| static int transcodeFromUTF8( | static int transcodeFromUTF8(const XMLByte* srcData, size_t& srcLen, |
| const XMLByte* srcData, size_t& srcLen, | XMLByte* toFill, size_t& toFillLen, |
| XMLByte* toFill, size_t& toFillLen, | const Charset::Tables& tables) { |
| const Charset::Tables& tables) { | |
| const XMLByte* srcPtr=srcData; | const XMLByte* srcPtr=srcData; |
| const XMLByte* srcEnd=srcData+srcLen; | const XMLByte* srcEnd=srcData+srcLen; |
| XMLByte* outPtr=toFill; | XMLByte* outPtr=toFill; |
| XMLByte* outEnd=toFill+toFillLen; | XMLByte* outEnd=toFill+toFillLen; |
| // We now loop until we either run out of input data, or room to store | // We now loop until we either run out of input data, or room to store |
| while ((srcPtr < srcEnd) && (outPtr < outEnd)) { | while ((srcPtr < srcEnd) && (outPtr < outEnd)) { |
| // Get the next leading byte out | // Get the next leading byte out |
| 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; |
| } | } |
| // See how many trailing src bytes this sequence is going to require | // See how many trailing src bytes this sequence is going to require |
| const unsigned int trailingBytes = gUTFBytes[firstByte]; | const unsigned int trailingBytes = gUTFBytes[firstByte]; |
| // If there are not enough source bytes to do this one, then we | // If there are not enough source bytes to do this one, then we |
| // are done. Note that we done>= here because we are implicitly | // are done. Note that we done>= here because we are implicitly |
| // counting the 1 byte we get no matter what. | // counting the 1 byte we get no matter what. |
| if(srcPtr+trailingBytes>= srcEnd) | if(srcPtr+trailingBytes>= srcEnd) |
| break; | break; |
| // Looks ok, so lets build up the value | // Looks ok, so lets build up the value |
| uint tmpVal=0; | uint tmpVal=0; |
| switch(trailingBytes) { | switch(trailingBytes) { |
| case 5: tmpVal+=*srcPtr++; tmpVal<<=6; | case 5: tmpVal+=*srcPtr++; tmpVal<<=6; |
| case 4: tmpVal+=*srcPtr++; tmpVal<<=6; | case 4: tmpVal+=*srcPtr++; tmpVal<<=6; |
| case 3: tmpVal+=*srcPtr++; tmpVal<<=6; | case 3: tmpVal+=*srcPtr++; tmpVal<<=6; |
| case 2: tmpVal+=*srcPtr++; tmpVal<<=6; | case 2: tmpVal+=*srcPtr++; tmpVal<<=6; |
| case 1: tmpVal+=*srcPtr++; tmpVal<<=6; | case 1: tmpVal+=*srcPtr++; tmpVal<<=6; |
| case 0: tmpVal+=*srcPtr++; | case 0: tmpVal+=*srcPtr++; |
| break; | break; |
| default: | default: |
| throw Exception(0, | throw Exception(0, |
| 0, | 0, |
| "transcodeFromUTF8 error: wrong trailingBytes value(%d)", trailingBytes); | "transcodeFromUTF8 error: wrong trailingBytes value(%d)", trailingBytes); |
| } | } |
| tmpVal-=gUTFOffsets[trailingBytes]; | tmpVal-=gUTFOffsets[trailingBytes]; |
| // If it will fit into a single char, then put it in. Otherwise | // If it will fit into a single char, then put it in. Otherwise |
| // fail [*encode it as a surrogate pair. If its not valid, use the | // fail [*encode it as a surrogate pair. If its not valid, use the |
| // replacement char.*] | // replacement char.*] |
| if(!(tmpVal & 0xFFFF0000)) { | if(!(tmpVal & 0xFFFF0000)) { |
| if(XMLByte xlat=xlatOneTo(tmpVal, tables, 0)) | if(XMLByte xlat=xlatOneTo(tmpVal, tables, 0)) |
| *outPtr++=xlat; | *outPtr++=xlat; |
| else | else |
| outPtr+=sprintf((char *)outPtr, "&#%d;", tmpVal); // &#decimal; | outPtr+=sprintf((char *)outPtr, "&#%d;", tmpVal); // &#decimal; |
| } else | } else |
| throw Exception(0, | throw Exception(0, |
| 0, | 0, |
| "transcodeFromUTF8 error: too big tmpVal(0x%08X)", tmpVal); | "transcodeFromUTF8 error: too big tmpVal(0x%08X)", tmpVal); |
| } | } |
| // Update the bytes eaten | // Update the bytes eaten |
| srcLen = srcPtr - srcData; | srcLen = srcPtr - srcData; |
| // Return the characters read | // Return the characters read |
| toFillLen = outPtr - toFill; | toFillLen = outPtr - toFill; |
| //return srcPtr==srcEnd?(int)toFillLen:-1; | //return srcPtr==srcEnd?(int)toFillLen:-1; |
| /* | /* |
| Line 426 of ocetes consumed. | Line 421 of ocetes consumed. |
| 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*/; | size_t dest_length=src.length*6/*so that surly enough, max utf8 seq len=6*/; |
| #ifndef NDEBUG | |
| size_t saved_dest_length=dest_length; | |
| #endif | |
| XMLByte *dest_body=new(PointerFreeGC) XMLByte[dest_length+1/*for terminator*/]; | XMLByte *dest_body=new(PointerFreeGC) XMLByte[dest_length+1/*for terminator*/]; |
| if(::transcodeToUTF8( | if(::transcodeToUTF8( |
| Line 436 const String::C Charset::transcodeToUTF8 | Line 434 const String::C Charset::transcodeToUTF8 |
| 0, | 0, |
| "Charset::transcodeToUTF8 buffer overflow"); | "Charset::transcodeToUTF8 buffer overflow"); |
| 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); |
| } | } |
| 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 */; | size_t dest_length=src.length*6/*so that surly enough, "ÿ" has max ratio */; |
| #ifndef NDEBUG | |
| size_t saved_dest_length=dest_length; | |
| #endif | |
| XMLByte *dest_body=new(PointerFreeGC) XMLByte[dest_length+1/*for terminator*/]; | XMLByte *dest_body=new(PointerFreeGC) XMLByte[dest_length+1/*for terminator*/]; |
| if(::transcodeFromUTF8( | if(::transcodeFromUTF8( |
| Line 452 const String::C Charset::transcodeFromUT | Line 453 const String::C Charset::transcodeFromUT |
| 0, | 0, |
| "Charset::transcodeFromUTF8 buffer overflow"); | "Charset::transcodeFromUTF8 buffer overflow"); |
| 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 467 const String::C Charset::transcodeToChar | Line 468 const String::C Charset::transcodeToChar |
| XMLByte* output=dest_body; | XMLByte* output=dest_body; |
| const XMLByte* input=(XMLByte *)src.str; | const XMLByte* input=(XMLByte *)src.str; |
| while(char c=*input++) { | while(XMLCh c=*input++) { |
| XMLCh curVal = tables.fromTable[c]; | XMLCh curVal = tables.fromTable[c]; |
| *output++=curVal? | *output++=curVal? |
| xlatOneTo(curVal, dest_charset.tables, '?') // OK | xlatOneTo(curVal, dest_charset.tables, '?') // OK |
| Line 543 void Charset::addEncoding(char *name_cst | Line 544 void Charset::addEncoding(char *name_cst |
| } | } |
| void Charset::initTranscoder(const String& name, const char* name_cstr) { | void Charset::initTranscoder(const String::Body NAME, const char* name_cstr) { |
| ftranscoder=xmlFindCharEncodingHandler(name_cstr); | ftranscoder=xmlFindCharEncodingHandler(name_cstr); |
| transcoder(name); // check right way | transcoder(NAME); // check right way |
| } | } |
| xmlCharEncodingHandler& Charset::transcoder(const String& name) { | xmlCharEncodingHandler& Charset::transcoder(const String::Body NAME) { |
| if(!ftranscoder) | if(!ftranscoder) |
| throw Exception("parser.runtime", | throw Exception("parser.runtime", |
| &name, | new String(NAME, String::L_TAINTED), |
| "unsupported encoding"); | "unsupported encoding"); |
| return *ftranscoder; | return *ftranscoder; |
| } | } |
| const char* Charset::transcode_cstr(xmlChar* s) { | String::C Charset::transcode_cstr(xmlChar* s) { |
| if(!s) | if(!s) |
| return ""; | return String::C("", 0); |
| int inlen=strlen((const char* )s); | int inlen=strlen((const char*)s); |
| int outlen=inlen+1; // max | int outlen=inlen; // max |
| char *out=new(PointerFreeGC) char[outlen]; | #ifndef NDEBUG |
| int saved_outlen=outlen; | |
| #endif | |
| char *out=new(PointerFreeGC) char[outlen+1]; | |
| int error; | int error; |
| if(xmlCharEncodingOutputFunc output=transcoder(fname).output) { | if(xmlCharEncodingOutputFunc output=transcoder(FNAME).output) { |
| error=output( | error=output( |
| (unsigned char*)out, &outlen, | (unsigned char*)out, &outlen, |
| (const unsigned char*)s, &inlen); | (const unsigned char*)s, &inlen); |
| Line 578 const char* Charset::transcode_cstr(xmlC | Line 582 const char* Charset::transcode_cstr(xmlC |
| 0, | 0, |
| "transcode_cstr failed (%d)", error); | "transcode_cstr failed (%d)", error); |
| out[outlen/*surely would be less then on input*/]=0; | assert(outlen<=saved_outlen); out[outlen]=0; |
| return out; | return String::C(out, outlen); |
| } | } |
| const String& Charset::transcode(xmlChar* s) { | const String& Charset::transcode(xmlChar* s) { |
| return *new String(transcode_cstr(s), 0/*auto-size*/, true); | String::C cstr=transcode_cstr(s); |
| return *new String(cstr.str, cstr.length, true); | |
| } | } |
| const char* Charset::transcode_cstr(GdomeDOMString* s) { | String::C Charset::transcode_cstr(GdomeDOMString* s) { |
| return s?transcode_cstr(BAD_CAST s->str):""; | return s?transcode_cstr(BAD_CAST s->str) |
| :String::C("", 0); | |
| } | } |
| const String& Charset::transcode(GdomeDOMString* s) { | const String& Charset::transcode(GdomeDOMString* s) { |
| return *new String(transcode_cstr(s), 0/*auto-size*/, true); | String::C cstr=transcode_cstr(s); |
| return *new String(cstr.str, cstr.length, true); | |
| } | } |
| /// @test less memory using -maybe- xmlParserInputBufferCreateMem | /// @test less memory using -maybe- xmlParserInputBufferCreateMem |
| void* Charset::transcode_buf2mchar(transcode_buf_malloc_func malloc_func, | xmlChar* Charset::transcode_buf2xchar(const char* buf, size_t buf_size) { |
| const char* buf, size_t buf_size) { | xmlChar* out; |
| unsigned char* out; | |
| int outlen; | int outlen; |
| int error; | int error; |
| if(xmlCharEncodingInputFunc input=transcoder(fname).input) { | #ifndef NDEBUG |
| int saved_outlen; | |
| #endif | |
| if(xmlCharEncodingInputFunc input=transcoder(FNAME).input) { | |
| outlen=buf_size*6/*max*/; | outlen=buf_size*6/*max*/; |
| out=(unsigned char*)malloc_func(outlen+1); | #ifndef NDEBUG |
| saved_outlen=outlen; | |
| #endif | |
| out=(xmlChar*)xmlMalloc(outlen+1); | |
| error=input( | error=input( |
| out, &outlen, | out, &outlen, |
| (const unsigned char *)buf, (int *)&buf_size); | (const unsigned char*)buf, (int*)&buf_size); |
| } else { | } else { |
| outlen=buf_size; | outlen=buf_size; |
| out=(unsigned char*)malloc_func(outlen+1); | #ifndef NDEBUG |
| saved_outlen=outlen; | |
| #endif | |
| out=(xmlChar*)xmlMalloc(outlen+1); | |
| memcpy(out, buf, outlen); | memcpy(out, buf, outlen); |
| error=0; | error=0; |
| } | } |
| Line 615 void* Charset::transcode_buf2mchar(trans | Line 630 void* Charset::transcode_buf2mchar(trans |
| 0, | 0, |
| "transcode_buf failed (%d)", error); | "transcode_buf failed (%d)", error); |
| out[outlen/*surely would be less then on input*/]=0; | assert(outlen<=saved_outlen); out[outlen]=0; |
| return out; | return out; |
| } | } |
| xmlChar* Charset::transcode_buf2xchar(const char* buf, size_t buf_size) { | |
| return static_cast<xmlChar*>(transcode_buf2mchar(xmlMalloc, buf, buf_size)); | |
| } | |
| static void* g_malloc_wrapper(size_t size) { | |
| return g_malloc(size); | |
| } | |
| gchar* Charset::transcode_buf2gchar(const char* buf, size_t buf_size) { | |
| return static_cast<gchar*>(transcode_buf2mchar(g_malloc_wrapper, buf, buf_size)); | |
| } | |
| GdomeDOMString_auto_ptr Charset::transcode_buf2dom(const char* buf, size_t buf_size) { | GdomeDOMString_auto_ptr Charset::transcode_buf2dom(const char* buf, size_t buf_size) { |
| return GdomeDOMString_auto_ptr(transcode_buf2gchar(buf, buf_size)); | return GdomeDOMString_auto_ptr(transcode_buf2xchar(buf, buf_size)); |
| } | } |
| GdomeDOMString_auto_ptr Charset::transcode(const String& s) { | 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_buf2dom(cstr, strlen(cstr)); |
| } | } |
| GdomeDOMString_auto_ptr Charset::transcode(const StringBody s) { | GdomeDOMString_auto_ptr Charset::transcode(const String::Body s) { |
| const char* cstr=s.cstr(); | const char* cstr=s.cstr(); |
| return transcode_buf2dom(cstr, s.length()); | return transcode_buf2dom(cstr, s.length()); |
| } | } |
| #endif | #endif |
| String::Body Charset::transcode(const String::Body src, | |
| const Charset& source_transcoder, | |
| const Charset& dest_transcoder) { | |
| const char *src_ptr=src.cstr(); | |
| size_t src_size=strlen(src_ptr); | |
| String::C dest=Charset::transcode(String::C(src_ptr, src_size), | |
| source_transcoder, | |
| dest_transcoder); | |
| return String::Body(dest.str, dest.length); | |
| } | |
| String& Charset::transcode(const String& src, | |
| const Charset& source_transcoder, | |
| const Charset& dest_transcoder) { | |
| if(!src.length()) | |
| return *new String("", 0, false); | |
| return *new String(transcode((String::Body)src, source_transcoder, dest_transcoder), String::L_CLEAN); | |
| } | |
| void Charset::transcode(ArrayString& src, | |
| const Charset& source_transcoder, | |
| const Charset& dest_transcoder) { | |
| for(size_t i=0; i<src.count(); i++) | |
| src.put(i, &transcode(*src[i], source_transcoder, dest_transcoder)); | |
| } | |
| #ifndef DOXYGEN | |
| struct Transcode_pair_info { | |
| const Charset* source_transcoder; | |
| const Charset* dest_transcoder; | |
| }; | |
| #endif | |
| static void transcode_pair(const String::Body akey, | |
| String::Body& avalue, | |
| Transcode_pair_info* info) { | |
| avalue=Charset::transcode(avalue, | |
| *info->source_transcoder, | |
| *info->dest_transcoder); | |
| } | |
| void Charset::transcode(HashStringString& src, | |
| const Charset& source_transcoder, | |
| const Charset& dest_transcoder) { | |
| Transcode_pair_info info={&source_transcoder, &dest_transcoder}; | |
| src.for_each_ref(transcode_pair, &info); | |
| } |