--- parser3/src/main/pa_charset.C 2003/01/27 15:35:24 1.33.2.1 +++ parser3/src/main/pa_charset.C 2003/03/26 18:52:50 1.33.2.19.2.12 @@ -1,26 +1,19 @@ /** @file 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(http://paf.design.ru) */ -static const char* IDENT_CHARSET_C="$Date: 2003/01/27 15:35:24 $"; +static const char* IDENT_CHARSET_C="$Date: 2003/03/26 18:52:50 $"; #include "pa_charset.h" -#include "pa_globals.h" +#include "pa_charsets.h" #ifdef XML #include "libxml/encoding.h" #endif -// globals - - -// consts - -#define MAX_CHARSET_UNI_CODES 500 - // helpers inline void prepare_case_tables(unsigned char *tables) { @@ -38,7 +31,7 @@ inline void cstr2ctypes(unsigned char *t ctypes_table[c]|=bit; } } -inline unsigned int to_wchar_code(const char *cstr) { +inline unsigned int to_wchar_code(const char* cstr) { if(!cstr || !*cstr) return 0; if(cstr[1]==0) @@ -47,7 +40,7 @@ inline unsigned int to_wchar_code(const char *error_pos; return(unsigned int)strtol(cstr, &error_pos, 0); } -inline bool to_bool(const char *cstr) { +inline bool to_bool(const char* cstr) { return cstr && *cstr!=0; } static void element2ctypes(unsigned char c, bool belongs, @@ -75,16 +68,16 @@ static void element2case(unsigned char f // methods extern "C" unsigned char pcre_default_tables[]; // pcre/chartables.c -Charset::Charset(Pool& pool, ConstStringPtr aname, ConstStringPtr afile_spec): +Charset::Charset(const String& aname, const String* afile_spec): fname(aname), - fname_cstr(aname->cstr()) { + fname_cstr(aname.cstrm()) { for(char *c=fname_cstr; *c; c++) - *c = toupper(*c); + *c = toupper(*c); if(afile_spec) { fisUTF8=false; - loadDefinition(pool, fname); + load_definition(*afile_spec); #ifdef XML addEncoding(fname_cstr); #endif @@ -99,14 +92,7 @@ Charset::Charset(Pool& pool, ConstString #endif } -Charset::~Charset() { - // @todonow unregister encodings -#ifdef XML - // not deleting transcoder, that's not our business -#endif -} - -void Charset::loadDefinition(Pool& pool, ConstStringPtr afile_spec) { +void Charset::load_definition(const String& afile_spec) { // pcre_tables // lowcase, flipcase, bits digit+word+whitespace, masks @@ -117,16 +103,14 @@ void Charset::loadDefinition(Pool& pool, cstr2ctypes(pcre_tables,(const unsigned char *)"*+?{^.$|()[", ctype_meta); // charset - memset(tables.fromTable, 0, sizeof(tables.fromTable)); - tables.toTable=(Charset_TransRec *)pool.calloc(sizeof(Charset_TransRec)*MAX_CHARSET_UNI_CODES); - tables.toTableSize=0; + memset(&tables, 0, sizeof(tables)); // strangly vital tables.toTable[tables.toTableSize].intCh=0; tables.toTable[tables.toTableSize].extCh=(XMLByte)0; tables.toTableSize++; // loading text - char *data=file_read_text(pool, utf8_charset, afile_spec); + char *data=file_read_text(UTF8_charset, afile_spec); // ignore header getrow(&data); @@ -156,7 +140,7 @@ void Charset::loadDefinition(Pool& pool, // charset if(tables.toTableSize>MAX_CHARSET_UNI_CODES) throw Exception("parser.runtime", - afile_spec, + &afile_spec, "charset must contain not more then %d unicode values", MAX_CHARSET_UNI_CODES); XMLCh unicode=(XMLCh)to_wchar_code(cell); @@ -216,8 +200,8 @@ static XMLByte xlatOneTo(const XMLCh toX return not_found; } -void Charset::transcode(Pool& pool, - const Charset& source_charset, const void *source_body, size_t source_content_length, +void Charset::transcode( + 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 ) { if(!source_content_length) { @@ -228,17 +212,17 @@ void Charset::transcode(Pool& pool, switch((source_charset.isUTF8()?0x10:0x00)|(dest_charset.isUTF8()?0x01:0x00)) { default: // 0x00 - source_charset.transcodeToCharset(pool, dest_charset, + source_charset.transcodeToCharset(dest_charset, source_body, source_content_length, dest_body, dest_content_length); break; case 0x01: - source_charset.transcodeToUTF8(pool, + source_charset.transcodeToUTF8( source_body, source_content_length, dest_body, dest_content_length); break; case 0x10: - dest_charset.transcodeFromUTF8(pool, + dest_charset.transcodeFromUTF8( source_body, source_content_length, dest_body, dest_content_length); break; @@ -375,7 +359,7 @@ of ocetes consumed. } /// @todo digital entites only when xml/html output [at output in html/xml mode, in html part of a letter] static int transcodeFromUTF8( - const XMLByte *srcData, size_t& srcLen, + const XMLByte* srcData, size_t& srcLen, XMLByte* toFill, size_t& toFillLen, const Charset::Tables& tables) { const XMLByte* srcPtr=srcData; @@ -386,7 +370,7 @@ static int transcodeFromUTF8( // We now loop until we either run out of input data, or room to store while ((srcPtr < srcEnd) && (outPtr < outEnd)) { // 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 if(firstByte<= 127) { @@ -417,7 +401,7 @@ static int transcodeFromUTF8( default: throw Exception(0, - Exception::undefined_source, + 0, "transcodeFromUTF8 error: wrong trailingBytes value(%d)", trailingBytes); } tmpVal-=gUTFOffsets[trailingBytes]; @@ -432,7 +416,7 @@ static int transcodeFromUTF8( outPtr+=sprintf((char *)outPtr, "&#%d;", tmpVal); // &#decimal; } else throw Exception(0, - Exception::undefined_source, + 0, "transcodeFromUTF8 error: too big tmpVal(0x%08X)", tmpVal); } @@ -454,11 +438,11 @@ of ocetes consumed. } /// @todo not so memory-hungry with prescan -void Charset::transcodeToUTF8(Pool& pool, - const void *source_body, size_t source_content_length, +void Charset::transcodeToUTF8( + const void* source_body, size_t source_content_length, 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*/; - XMLByte *dest_body=(XMLByte*)pool.malloc(dest_content_length); + XMLByte *dest_body=new(PointerFreeGC) XMLByte[dest_content_length]; if(::transcodeToUTF8( (XMLByte *)source_body, source_content_length, @@ -471,11 +455,11 @@ void Charset::transcodeToUTF8(Pool& pool // return adest_body=dest_body; } -void Charset::transcodeFromUTF8(Pool& pool, - const void *source_body, size_t source_content_length, +void Charset::transcodeFromUTF8( + const void* source_body, size_t source_content_length, 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 */; - XMLByte *dest_body=(XMLByte*)pool.malloc(dest_content_length); + XMLByte *dest_body=new(PointerFreeGC) XMLByte[dest_content_length]; if(::transcodeFromUTF8( (XMLByte *)source_body, source_content_length, @@ -490,16 +474,16 @@ void Charset::transcodeFromUTF8(Pool& po } /// transcode using both charsets -void Charset::transcodeToCharset(Pool& pool, - const Charset& dest_charset, - const void *source_body, size_t source_content_length, - const void *& adest_body, size_t& adest_content_length) const { +void Charset::transcodeToCharset( + const Charset& dest_charset, + const void* source_body, size_t source_content_length, + const void *& adest_body, size_t& adest_content_length) const { if(&dest_charset==this) { adest_body=source_body; adest_content_length=source_content_length; } else { size_t dest_content_length=source_content_length; - unsigned char *dest_body=(unsigned char *)pool.malloc(dest_content_length); + unsigned char *dest_body=new(PointerFreeGC) unsigned char[dest_content_length]; const XMLByte* srcPtr=(XMLByte *)source_body; const XMLByte* srcEnd=(XMLByte *)source_body+source_content_length; @@ -527,9 +511,9 @@ static int xml256CharEncodingInp int *inlen, void *info) { return transcodeToUTF8( - in, *(size_t*)inlen, - out, *(size_t*)outlen, - *(const Charset::Tables *)info); + in, *(size_t*)inlen, + out, *(size_t*)outlen, + *(const Charset::Tables *)info); } static int xml256CharEncodingOutputFunc ( @@ -539,15 +523,14 @@ static int xml256CharEncodingOut int *inlen, void *info) { return transcodeFromUTF8( - in, *(size_t*)inlen, - out, *(size_t*)outlen, - *(const Charset::Tables *)info); + in, *(size_t*)inlen, + out, *(size_t*)outlen, + *(const Charset::Tables *)info); } void Charset::addEncoding(char *name_cstr) { - xmlCharEncodingHandler *handler= - (xmlCharEncodingHandler *)malloc(sizeof(xmlCharEncodingHandler)); + xmlCharEncodingHandler* handler=new(PointerFreeGC) xmlCharEncodingHandler; handler->name=name_cstr; handler->input=xml256CharEncodingInputFunc; handler->inputInfo=&tables; handler->output=xml256CharEncodingOutputFunc; handler->outputInfo=&tables; @@ -555,33 +538,33 @@ void Charset::addEncoding(char *name_cst xmlRegisterCharEncodingHandler(handler); } -void Charset::initTranscoder(const String *source, const char *name_cstr) { +void Charset::initTranscoder(const String& name, const char* name_cstr) { ftranscoder=xmlFindCharEncodingHandler(name_cstr); - transcoder(source); // check right way + transcoder(name); // check right way } -xmlCharEncodingHandler *Charset::transcoder(const String *source) { +xmlCharEncodingHandler& Charset::transcoder(const String& name) { if(!ftranscoder) throw Exception("parser.runtime", - source, + &name, "unsupported encoding"); - return ftranscoder; + return *ftranscoder; } -const char *Charset::transcode_cstr(xmlChar *s) { +const char* Charset::transcode_cstr(xmlChar* s) { if(!s) return ""; - int inlen=strlen((const char *)s); + int inlen=strlen((const char* )s); int outlen=inlen+1; // max - char *out=(char *)malloc(outlen*sizeof(char)); + char *out=new(PointerFreeGC) char[outlen]; int error; - if(xmlCharEncodingOutputFunc output=transcoder(0)->output) { + if(xmlCharEncodingOutputFunc output=transcoder(fname).output) { error=output( (unsigned char*)out, &outlen, (const unsigned char*)s, &inlen, - transcoder(0)->outputInfo); + transcoder(fname).outputInfo); } else { memcpy(out, s, outlen=inlen); error=0; @@ -594,43 +577,32 @@ const char *Charset::transcode_cstr(xmlC out[outlen/*surely would be less then on input*/]=0; return out; } -String& Charset::transcode(xmlChar *s -#ifndef NO_STRING_ORIGIN - , const String *origin -#endif - ) { - String& result=*NEW String(pool()); - result.APPEND_CLEAN(transcode_cstr(s), 0/*auto-size*/, origin->origin().file, origin->origin().line); - return result; +const String& Charset::transcode(xmlChar* s) { + return *new String(transcode_cstr(s), 0/*auto-size*/, true); } -const char *Charset::transcode_cstr(GdomeDOMString *s) { +const char* Charset::transcode_cstr(GdomeDOMString* s) { return s?transcode_cstr(BAD_CAST s->str):""; } -String& Charset::transcode(GdomeDOMString *s -#ifndef NO_STRING_ORIGIN - , const String *origin -#endif - ) { - String& result=*NEW String(pool()); - result.APPEND_CLEAN(transcode_cstr(s), 0/*auto-size*/, origin->origin().file, origin->origin().line); - return result; +const String& Charset::transcode(GdomeDOMString* s) { + return *new String(transcode_cstr(s), 0/*auto-size*/, true); } /// @test less memory using -maybe- xmlParserInputBufferCreateMem -xmlChar *Charset::transcode_buf2xchar(const char *buf, size_t buf_size) { - unsigned char *out; +void* Charset::transcode_buf2mchar(transcode_buf_malloc_func malloc_func, + const char* buf, size_t buf_size) { + unsigned char* out; int outlen; int error; - if(xmlCharEncodingInputFunc input=transcoder(0)->input) { + if(xmlCharEncodingInputFunc input=transcoder(fname).input) { outlen=buf_size*6/*max*/; - out=(unsigned char*)malloc((outlen+1)*sizeof(unsigned char)); + out=(unsigned char*)malloc_func(outlen+1); error=input( out, &outlen, (const unsigned char *)buf, (int *)&buf_size, - transcoder(0)->inputInfo); + transcoder(fname).inputInfo); } else { outlen=buf_size; - out=(unsigned char*)malloc((outlen+1)*sizeof(unsigned char)); + out=(unsigned char*)malloc_func(outlen+1); memcpy(out, buf, outlen); error=0; } @@ -641,14 +613,29 @@ xmlChar *Charset::transcode_buf2xchar(co "transcode_buf failed (%d)", error); out[outlen/*surely would be less then on input*/]=0; - return (xmlChar *)out; + return out; +} + +xmlChar* Charset::transcode_buf2xchar(const char* buf, size_t buf_size) { + return static_cast(transcode_buf2mchar(xmlMalloc, buf, buf_size)); } -GdomeDOMString_auto_ptr Charset::transcode_buf2dom(const char *buf, size_t buf_size) { - return GdomeDOMString_auto_ptr((gchar*)transcode_buf2xchar(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(transcode_buf2mchar(g_malloc_wrapper, buf, buf_size)); +} +GdomeDOMString_auto_ptr Charset::transcode_buf2dom(const char* buf, size_t buf_size) { + return GdomeDOMString_auto_ptr(transcode_buf2gchar(buf, buf_size)); } GdomeDOMString_auto_ptr Charset::transcode(const String& s) { - const char *cstr=s.cstr(String::UL_UNSPECIFIED); + const char* cstr=s.cstr(String::L_UNSPECIFIED); return transcode_buf2dom(cstr, strlen(cstr)); } +GdomeDOMString_auto_ptr Charset::transcode(const StringBody s) { + const char* cstr=s.cstr(); + + return transcode_buf2dom(cstr, s.length()); +} #endif