|
|
| version 1.33.2.19.2.2, 2003/03/19 16:28:08 | version 1.33.4.1, 2003/03/17 14:48:05 |
|---|---|
| Line 1 | Line 1 |
| /** @file | /** @file |
| Parser: Charset connection implementation. | Parser: Charset connection implementation. |
| Copyright(c) 2001-2003 ArtLebedev Group (http://www.artlebedev.com) | Copyright(c) 2001, 2003 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) |
| */ | */ |
| static const char* IDENT_CHARSET_C="$Date$"; | static const char* IDENT_CHARSET_C="$Date$"; |
| #include "pa_charset.h" | #include "pa_charset.h" |
| #include "pa_charsets.h" | |
| #ifdef XML | #ifdef XML |
| #include "libxml/encoding.h" | #include "libxml/encoding.h" |
| #endif | #endif |
| // globals | |
| // consts | |
| #define MAX_CHARSET_UNI_CODES 500 | |
| // helpers | // helpers |
| inline void prepare_case_tables(unsigned char *tables) { | inline void prepare_case_tables(unsigned char *tables) { |
| Line 31 inline void cstr2ctypes(unsigned char *t | Line 37 inline void cstr2ctypes(unsigned char *t |
| ctypes_table[c]|=bit; | ctypes_table[c]|=bit; |
| } | } |
| } | } |
| inline unsigned int to_wchar_code(const char* cstr) { | inline unsigned int to_wchar_code(const char *cstr) { |
| if(!cstr || !*cstr) | if(!cstr || !*cstr) |
| return 0; | return 0; |
| if(cstr[1]==0) | if(cstr[1]==0) |
| Line 40 inline unsigned int to_wchar_code(const | Line 46 inline unsigned int to_wchar_code(const |
| char *error_pos; | char *error_pos; |
| return(unsigned int)strtol(cstr, &error_pos, 0); | return(unsigned int)strtol(cstr, &error_pos, 0); |
| } | } |
| inline bool to_bool(const char* cstr) { | inline bool to_bool(const char *cstr) { |
| return cstr && *cstr!=0; | return cstr && *cstr!=0; |
| } | } |
| static void element2ctypes(unsigned char c, bool belongs, | static void element2ctypes(unsigned char c, bool belongs, |
| Line 68 static void element2case(unsigned char f | Line 74 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(Pool& apool, const String& aname, const String *request_file_spec): Pooled(apool), |
| fname(aname), | fname(aname) { |
| fname_cstr(aname->cstr()) { | |
| for(char *c=fname_cstr; *c; c++) | char *name_cstr=fname.cstr(); |
| for(char *c=name_cstr; *c; c++) | |
| *c = toupper(*c); | *c = toupper(*c); |
| if(afile_spec) { | if(request_file_spec) { |
| fisUTF8=false; | fisUTF8=false; |
| load_definition(afile_spec); | loadDefinition(*request_file_spec); |
| #ifdef XML | #ifdef XML |
| addEncoding(fname_cstr); | addEncoding(name_cstr); |
| #endif | #endif |
| } else { | } else { |
| fisUTF8=true; | fisUTF8=true; |
| Line 88 Charset::Charset(const String& aname, co | Line 94 Charset::Charset(const String& aname, co |
| } | } |
| #ifdef XML | #ifdef XML |
| initTranscoder(fname, fname_cstr); | initTranscoder(&aname, name_cstr); |
| #endif | #endif |
| } | } |
| Charset::~Charset() { | Charset::~Charset() { |
| // @todonow unregister encodings | |
| #ifdef XML | #ifdef XML |
| // not deleting transcoder, that's not our business | // not deleting transcoder, that's not our business |
| #endif | #endif |
| } | } |
| void Charset::load_definition(const String& afile_spec) { | void Charset::loadDefinition(const String& request_file_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 115 void Charset::load_definition(const Stri |
| cstr2ctypes(pcre_tables,(const unsigned char *)"*+?{^.$|()[", ctype_meta); | cstr2ctypes(pcre_tables,(const unsigned char *)"*+?{^.$|()[", ctype_meta); |
| // charset | // charset |
| memset(&tables, 0, sizeof(tables)); | memset(tables.fromTable, 0, sizeof(tables.fromTable)); |
| tables.toTable=(Charset_TransRec *)calloc(sizeof(Charset_TransRec)*MAX_CHARSET_UNI_CODES); | |
| tables.toTableSize=0; | |
| // strangly vital | // strangly vital |
| tables.toTable[tables.toTableSize].intCh=0; | tables.toTable[tables.toTableSize].intCh=0; |
| tables.toTable[tables.toTableSize].extCh=(XMLByte)0; | tables.toTable[tables.toTableSize].extCh=(XMLByte)0; |
| tables.toTableSize++; | tables.toTableSize++; |
| // loading text | // loading text |
| Pool pool_for_load_only; | char *data=file_read_text(pool(), request_file_spec); |
| char *data=file_read_text(pool_for_load_only, *UTF8_charset, afile_spec); | |
| // ignore header | // ignore header |
| getrow(&data); | getrow(&data); |
| Line 148 void Charset::load_definition(const Stri | Line 154 void Charset::load_definition(const Stri |
| // 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, | &request_file_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); |
| XMLCh unicode=(XMLCh)to_wchar_code(cell); | XMLCh unicode=(XMLCh)to_wchar_code(cell); |
| Line 208 static XMLByte xlatOneTo(const XMLCh toX | Line 214 static XMLByte xlatOneTo(const XMLCh toX |
| return not_found; | return not_found; |
| } | } |
| void Charset::transcode( | void Charset::transcode(Pool& pool, |
| const Charset& source_charset, const void* source_body, size_t source_content_length, | const Charset& source_charset, const void *source_body, size_t source_content_length, |
| const Charset& dest_charset, const void *& dest_body, size_t& dest_content_length | const Charset& dest_charset, const void *& dest_body, size_t& dest_content_length |
| ) { | ) { |
| if(!source_content_length) { | if(!source_content_length) { |
| Line 235 void Charset::transcode( | Line 241 void Charset::transcode( |
| dest_body, dest_content_length); | dest_body, dest_content_length); |
| break; | break; |
| case 0x11: | case 0x11: |
| dest_body=pool.copy((char*)source_body, dest_content_length=source_content_length); | dest_body=source_body; |
| dest_content_length=source_content_length; | |
| break; | break; |
| } | } |
| } | } |
| Line 366 of ocetes consumed. | Line 373 of ocetes consumed. |
| } | } |
| /// @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; |
| Line 377 static int transcodeFromUTF8( | Line 384 static int transcodeFromUTF8( |
| // 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) { |
| Line 445 of ocetes consumed. | Line 452 of ocetes consumed. |
| } | } |
| /// @todo not so memory-hungry with prescan | /// @todo not so memory-hungry with prescan |
| void Charset::transcodeToUTF8( | void Charset::transcodeToUTF8(Pool& pool, |
| const void* source_body, size_t source_content_length, | const void *source_body, size_t source_content_length, |
| const void *& adest_body, size_t& dest_content_length) const { | const void *& adest_body, size_t& dest_content_length) const { |
| dest_content_length=source_content_length*6/*so that surly enough, max utf8 seq len=6*/; | dest_content_length=source_content_length*6/*so that surly enough, max utf8 seq len=6*/; |
| XMLByte *dest_body=new XMLByte[dest_content_length]; | XMLByte *dest_body=(XMLByte*)pool.malloc_atomic(dest_content_length); |
| if(::transcodeToUTF8( | if(::transcodeToUTF8( |
| (XMLByte *)source_body, source_content_length, | (XMLByte *)source_body, source_content_length, |
| Line 462 void Charset::transcodeToUTF8( | Line 469 void Charset::transcodeToUTF8( |
| // return | // return |
| adest_body=dest_body; | adest_body=dest_body; |
| } | } |
| void Charset::transcodeFromUTF8( | void Charset::transcodeFromUTF8(Pool& pool, |
| const void* source_body, size_t source_content_length, | const void *source_body, size_t source_content_length, |
| const void *& adest_body, size_t& dest_content_length) const { | const void *& adest_body, size_t& dest_content_length) const { |
| dest_content_length=source_content_length*6/*so that surly enough, "ÿ" has max ratio */; | dest_content_length=source_content_length*6/*so that surly enough, "ÿ" has max ratio */; |
| XMLByte *dest_body=new XMLByte[dest_content_length]; | XMLByte *dest_body=(XMLByte*)pool.malloc_atomic(dest_content_length); |
| if(::transcodeFromUTF8( | if(::transcodeFromUTF8( |
| (XMLByte *)source_body, source_content_length, | (XMLByte *)source_body, source_content_length, |
| Line 481 void Charset::transcodeFromUTF8( | Line 488 void Charset::transcodeFromUTF8( |
| } | } |
| /// transcode using both charsets | /// transcode using both charsets |
| void Charset::transcodeToCharset( | void Charset::transcodeToCharset(Pool& pool, |
| const Charset& dest_charset, | const Charset& dest_charset, |
| const void* source_body, size_t source_content_length, | const void *source_body, size_t source_content_length, |
| const void *& adest_body, size_t& adest_content_length) const { | const void *& adest_body, size_t& adest_content_length) const { |
| if(&dest_charset==this) { | if(&dest_charset==this) { |
| adest_body=pool.copy((char*)source_body, adest_content_length=source_content_length); | adest_body=source_body; |
| adest_content_length=source_content_length; | |
| } else { | } else { |
| size_t dest_content_length=source_content_length; | size_t dest_content_length=source_content_length; |
| unsigned char *dest_body=new unsigned char[dest_content_length]; | unsigned char *dest_body=(unsigned char *)pool.malloc_atomic(dest_content_length); |
| const XMLByte* srcPtr=(XMLByte *)source_body; | const XMLByte* srcPtr=(XMLByte *)source_body; |
| const XMLByte* srcEnd=(XMLByte *)source_body+source_content_length; | const XMLByte* srcEnd=(XMLByte *)source_body+source_content_length; |
| Line 536 static int xml256CharEncodingOut | Line 544 static int xml256CharEncodingOut |
| void Charset::addEncoding(char *name_cstr) { | void Charset::addEncoding(char *name_cstr) { |
| xmlCharEncodingHandler* handler=new xmlCharEncodingHandler; | xmlCharEncodingHandler *handler= |
| fcreated_handler=xmlCharEncodingHandlerPtr(handler); | (xmlCharEncodingHandler *)malloc(sizeof(xmlCharEncodingHandler)); |
| handler->name=name_cstr; | handler->name=name_cstr; |
| handler->input=xml256CharEncodingInputFunc; handler->inputInfo=&tables; | handler->input=xml256CharEncodingInputFunc; handler->inputInfo=&tables; |
| handler->output=xml256CharEncodingOutputFunc; handler->outputInfo=&tables; | handler->output=xml256CharEncodingOutputFunc; handler->outputInfo=&tables; |
| Line 546 void Charset::addEncoding(char *name_cst | Line 553 void Charset::addEncoding(char *name_cst |
| xmlRegisterCharEncodingHandler(handler); | xmlRegisterCharEncodingHandler(handler); |
| } | } |
| void Charset::initTranscoder(const String* source, const char* name_cstr) { | void Charset::initTranscoder(const String *source, const char *name_cstr) { |
| ftranscoder=xmlFindCharEncodingHandler(name_cstr); | ftranscoder=xmlFindCharEncodingHandler(name_cstr); |
| transcoder(source); // check right way | transcoder(source); // check right way |
| } | } |
| xmlCharEncodingHandler& Charset::transcoder(const String* source) { | xmlCharEncodingHandler *Charset::transcoder(const String *source) { |
| if(!ftranscoder) | if(!ftranscoder) |
| throw Exception("parser.runtime", | throw Exception("parser.runtime", |
| source, | source, |
| "unsupported encoding"); | "unsupported encoding"); |
| return *ftranscoder; | return ftranscoder; |
| } | } |
| const char* Charset::transcode_cstrxmlChar* s) { | const char *Charset::transcode_cstr(xmlChar *s) { |
| if(!s) | if(!s) |
| return ""; | return ""; |
| int inlen=strlen((const char* )s); | int inlen=strlen((const char *)s); |
| int outlen=inlen+1; // max | int outlen=inlen+1; // max |
| char *out=new char[outlen]; | char *out=(char *)malloc_atomic(outlen*sizeof(char)); |
| int error; | int error; |
| if(xmlCharEncodingOutputFunc output=transcoder(0).output) { | if(xmlCharEncodingOutputFunc output=transcoder(0)->output) { |
| error=output( | error=output( |
| (unsigned char*)out, &outlen, | (unsigned char*)out, &outlen, |
| (const unsigned char*)s, &inlen, | (const unsigned char*)s, &inlen, |
| transcoder(0).outputInfo); | transcoder(0)->outputInfo); |
| } else { | } else { |
| memcpy(out, s, outlen=inlen); | memcpy(out, s, outlen=inlen); |
| error=0; | error=0; |
| Line 585 const char* Charset::transcode_cstrxmlCh | Line 592 const char* Charset::transcode_cstrxmlCh |
| out[outlen/*surely would be less then on input*/]=0; | out[outlen/*surely would be less then on input*/]=0; |
| return out; | return out; |
| } | } |
| const String& Charset::transcodexmlChar* s | String& Charset::transcode(xmlChar *s |
| #ifndef NO_STRING_ORIGIN | #ifndef NO_STRING_ORIGIN |
| , const String& origin | , const String *origin |
| #endif | #endif |
| ) { | ) { |
| const String& result(new String()); | String& result=*NEW String(pool()); |
| result->APPEND_CLEAN( | result.APPEND_CLEAN(transcode_cstr(s), 0/*auto-size*/, origin->origin().file, origin->origin().line); |
| transcode_cstr(pool, s), 0/*auto-size*/, | |
| origin->origin().file, origin->origin().line); | |
| return result; | return result; |
| } | } |
| const char* Charset::transcode_cstrGdomeDOMString* s) { | const char *Charset::transcode_cstr(GdomeDOMString *s) { |
| return s?transcode_cstr(pool, BAD_CAST s->str):""; | return s?transcode_cstr(BAD_CAST s->str):""; |
| } | } |
| const String& Charset::transcodeGdomeDOMString* s | String& Charset::transcode(GdomeDOMString *s |
| #ifndef NO_STRING_ORIGIN | #ifndef NO_STRING_ORIGIN |
| , const String& origin | , const String *origin |
| #endif | #endif |
| ) { | ) { |
| const String& result(new String()); | String& result=*NEW String(pool()); |
| result->APPEND_CLEAN(transcode_cstr(pool, s), 0/*auto-size*/, origin->origin().file, origin->origin().line); | result.APPEND_CLEAN(transcode_cstr(s), 0/*auto-size*/, origin->origin().file, origin->origin().line); |
| return result; | return result; |
| } | } |
| /// @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) { | unsigned char *out; |
| unsigned char* out; | |
| int outlen; | int outlen; |
| int error; | int error; |
| if(xmlCharEncodingInputFunc input=transcoder(0).input) { | if(xmlCharEncodingInputFunc input=transcoder(0)->input) { |
| outlen=buf_size*6/*max*/; | outlen=buf_size*6/*max*/; |
| out=(unsigned char*)malloc_func(outlen+1); | out=(unsigned char*)malloc_atomic((outlen+1)*sizeof(unsigned char)); |
| error=input( | error=input( |
| out, &outlen, | out, &outlen, |
| (const unsigned char *)buf, (int *)&buf_size, | (const unsigned char *)buf, (int *)&buf_size, |
| transcoder(0).inputInfo); | transcoder(0)->inputInfo); |
| } else { | } else { |
| outlen=buf_size; | outlen=buf_size; |
| out=(unsigned char*)malloc_func(outlen+1); | out=(unsigned char*)malloc_atomic((outlen+1)*sizeof(unsigned char)); |
| memcpy(out, buf, outlen); | memcpy(out, buf, outlen); |
| error=0; | error=0; |
| } | } |
| Line 635 void* Charset::transcode_buf2mchar(trans | Line 639 void* Charset::transcode_buf2mchar(trans |
| "transcode_buf failed (%d)", error); | "transcode_buf failed (%d)", error); |
| out[outlen/*surely would be less then on input*/]=0; | out[outlen/*surely would be less then on input*/]=0; |
| return out; | return (xmlChar *)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((gchar*)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::UL_UNSPECIFIED); | const char *cstr=s.cstr(String::UL_UNSPECIFIED); |
| return transcode_buf2dom(cstr, strlen(cstr)); | return transcode_buf2dom(cstr, strlen(cstr)); |
| } | } |