|
|
| version 1.195, 2004/02/27 15:07:46 | version 1.216, 2008/08/18 12:01:29 |
|---|---|
| Line 1 | Line 1 |
| /** @file | /** @file |
| Parser: string class. @see untalength_t.C. | Parser: string class. @see untalength_t.C. |
| Copyright (c) 2001-2004 ArtLebedev Group (http://www.artlebedev.com) | Copyright (c) 2001-2005 ArtLebedev Group (http://www.artlebedev.com) |
| Author: Alexandr Petrosian <paf@design.ru> (http://paf.design.ru) | Author: Alexandr Petrosian <paf@design.ru> (http://paf.design.ru) |
| */ | */ |
| static const char * const IDENT_STRING_C="$Date$"; | static const char * const IDENT_STRING_C="$Date$"; |
| #include "pcre.h" | |
| #include "pa_string.h" | #include "pa_string.h" |
| #include "pa_exception.h" | #include "pa_exception.h" |
| #include "pa_table.h" | #include "pa_table.h" |
| #include "pa_dictionary.h" | #include "pa_dictionary.h" |
| #include "pa_charset.h" | #include "pa_charset.h" |
| #include "pcre.h" | |
| const String String::Empty; | const String String::Empty; |
| int pa_atoi(const char* str, const String* problem_source) { | int pa_atoi(const char* str, const String* problem_source) { |
| if(!str) | if(!str) |
| return 0; | return 0; |
| while(*str && isspace(*str)) | while(*str && isspace((unsigned char)*str)) |
| str++; | str++; |
| if(!*str) | if(!*str) |
| return 0; | return 0; |
| int result; | int result; |
| char *error_pos; | char *error_pos; |
| bool negative=false; | |
| if(str[0]=='-') { | |
| negative=true; | |
| str++; | |
| } else if(str[0]=='+') { | |
| str++; | |
| } | |
| // 0xABC | // 0xABC |
| if(str[0]=='0') | if(str[0]=='0') |
| if(str[1]=='x' || str[1]=='X') | if(str[1]=='x' || str[1]=='X') |
| result=(int)(unsigned long)strtol(str, &error_pos, 0); | result=(int)(unsigned long)strtol(str, &error_pos, 0); |
| else | else { |
| result=(int)strtol(str+1/*skip leading 0*/, &error_pos, 0); | // skip leading 0000, to disable octal interpretation |
| do str++; while(*str=='0'); | |
| result=(int)strtol(str, &error_pos, 0); | |
| } | |
| else | else |
| result=(int)strtol(str, &error_pos, 0); | result=(int)strtol(str, &error_pos, 0); |
| if(negative) | |
| result=-result; | |
| while(char c=*error_pos++) | while(char c=*error_pos++) |
| if(!isspace(c)) | if(!isspace((unsigned char)c)) |
| throw Exception("number.format", | throw Exception("number.format", |
| problem_source, | problem_source, |
| problem_source?"invalid number (int)": "'%s' is invalid number (int)", str); | problem_source?"invalid number (int)": "'%s' is invalid number (int)", str); |
| Line 50 double pa_atod(const char* str, const St | Line 62 double pa_atod(const char* str, const St |
| if(!str) | if(!str) |
| return 0; | return 0; |
| while(*str && isspace(*str)) | while(*str && isspace((unsigned char)*str)) |
| str++; | str++; |
| if(!*str) | if(!*str) |
| return 0; | return 0; |
| double result; | double result; |
| char *error_pos; | char *error_pos; |
| bool negative=false; | |
| if(str[0]=='-') { | |
| negative=true; | |
| str++; | |
| } else if(str[0]=='+') { | |
| str++; | |
| } | |
| // 0xABC | // 0xABC |
| if(str[0]=='0') | if(str[0]=='0') |
| if(str[1]=='x' || str[1]=='X') | if(str[1]=='x' || str[1]=='X') |
| result=(double)(unsigned long)strtol(str, &error_pos, 0); | result=(double)(unsigned long)strtol(str, &error_pos, 0); |
| else | else { |
| result=(double)strtod(str+1/*skip leading 0*/, &error_pos); | // skip leading 0000, to disable octal interpretation |
| do str++; while(*str=='0'); | |
| result=(double)strtod(str, &error_pos); | |
| } | |
| else | else |
| result=(double)strtod(str, &error_pos); | result=(double)strtod(str, &error_pos); |
| if(negative) | |
| result=-result; | |
| while(char c=*error_pos++) | while(char c=*error_pos++) |
| if(!isspace(c)) | if(!isspace((unsigned char)c)) |
| throw Exception("number.format", | throw Exception("number.format", |
| problem_source, | problem_source, |
| problem_source?"invalid number (double)": "'%s' is invalid number (double)", str); | problem_source?"invalid number (double)": "'%s' is invalid number (double)", str); |
| Line 80 double pa_atod(const char* str, const St | Line 104 double pa_atod(const char* str, const St |
| #ifndef DOXYGEN | #ifndef DOXYGEN |
| typedef struct { | typedef struct { |
| ssize_t countdown; | ssize_t countdown; |
| char target; /* Character we're looking for */ | int target; /* Character we're looking for */ |
| } chr_data; | } chr_data; |
| #endif | #endif |
| static int CORD_range_contains_chr_greater_then_proc(char c, size_t size, void* client_data) | static int CORD_range_contains_chr_greater_then_proc(char c, size_t size, void* client_data) |
| Line 149 String::Body String::Body::trim(String:: | Line 173 String::Body String::Body::trim(String:: |
| return *this; | return *this; |
| if(!chars) | if(!chars) |
| chars=" \t\n"; // white space | chars=" \t\n"; // white space |
| Body result=*this; | |
| size_t start=0; | size_t start=0; |
| size_t end=our_length; | size_t end=our_length; |
| bool trim_start=(kind!=TRIM_END); | // from left... |
| if(trim_start) { | if(kind!=TRIM_END) { |
| CORD_pos pos; set_pos(pos, 0); | CORD_pos pos; set_pos(pos, 0); |
| while(true) { | while(true) { |
| char c=CORD_pos_fetch(pos); | char c=CORD_pos_fetch(pos); |
| Line 167 String::Body String::Body::trim(String:: | Line 190 String::Body String::Body::trim(String:: |
| CORD_next(pos); | CORD_next(pos); |
| } | } |
| } | } |
| // todo справа | // from right.. |
| if(kind!=TRIM_START) { | |
| CORD_pos pos; set_pos(pos, end-1); | |
| while(true) { | |
| char c=CORD_pos_fetch(pos); | |
| if(strchr(chars, c)) { | |
| if(--end==0) // optimization: NO need to check for 'end>=start', that's(<) impossible | |
| return 0; // all chars are empty, just return empty string | |
| } else | |
| break; | |
| CORD_prev(pos); | |
| } | |
| } | |
| if(start==0 && end==our_length) // nobody moved a thing | if(start==0 && end==our_length) // nobody moved a thing |
| return *this; | return *this; |
| Line 243 String& String::append_strdup(const char | Line 279 String& String::append_strdup(const char |
| return *this; | return *this; |
| } | } |
| size_t String::length(Charset& charset) const { | |
| if(charset.isUTF8()){ | |
| const XMLByte* srcPtr=(const XMLByte*)cstrm(); | |
| return lengthUTF8(srcPtr, srcPtr+body.length()); | |
| } else | |
| return body.length(); | |
| } | |
| /// @todo check in doc: whether it documents NOW bad situation "abc".mid(-1, 3) =were?="ab" | /// @todo check in doc: whether it documents NOW bad situation "abc".mid(-1, 3) =were?="ab" |
| String& String::mid(size_t substr_begin, size_t substr_end) const { | String& String::mid(size_t substr_begin, size_t substr_end) const { |
| String& result=*new String; | String& result=*new String; |
| Line 259 String& String::mid(size_t substr_begin, | Line 303 String& String::mid(size_t substr_begin, |
| // next: letters themselves | // next: letters themselves |
| result.body=body.mid(substr_begin, substr_length); | result.body=body.mid(substr_begin, substr_length); |
| // SAPI::log("piece of '%s' from %d to %d is '%s'", | ASSERT_STRING_INVARIANT(result); |
| //cstr(), substr_begin, substr_end, result.cstr()); | return result; |
| } | |
| // from, to and helper_length in characters, not in bytes (it's important for utf-8) | |
| String& String::mid(Charset& charset, size_t from, size_t to, size_t helper_length) const { | |
| String& result=*new String; | |
| size_t self_length=(helper_length)?helper_length:length(charset); | |
| if(!self_length) | |
| return result; | |
| from=min(min(to, from), self_length); | |
| to=min(max(to, from), self_length); | |
| size_t substr_length=to-from; | |
| if(!substr_length) | |
| return result; | |
| if(charset.isUTF8()){ | |
| const XMLByte* srcPtr=(const XMLByte*)cstrm(); | |
| const XMLByte* srcEnd=srcPtr+body.length(); | |
| // convert 'from' and 'substr_length' from 'characters' to 'bytes' | |
| from=getUTF8BytePos(srcPtr, srcEnd, from); | |
| substr_length=getUTF8BytePos(srcPtr+from, srcEnd, substr_length); | |
| if(!substr_length) | |
| return result; | |
| } | |
| // first: their langs | |
| result.langs.append(result.body, langs, from, substr_length); | |
| // next: letters themselves | |
| result.body=body.mid(from, substr_length); | |
| ASSERT_STRING_INVARIANT(result); | ASSERT_STRING_INVARIANT(result); |
| return result; | return result; |
| } | } |
| Line 285 size_t String::pos(const String& substr, | Line 364 size_t String::pos(const String& substr, |
| return pos(substr.body, this_offset, lang); | return pos(substr.body, this_offset, lang); |
| } | } |
| size_t String::pos(Charset& charset, const String& substr, | |
| size_t this_offset, Language lang) const { | |
| size_t result; | |
| if(charset.isUTF8()){ | |
| const XMLByte* srcPtr=(const XMLByte*)cstrm(); | |
| const XMLByte* srcEnd=srcPtr+body.length(); | |
| // convert 'this_offset' from 'characters' to 'bytes' | |
| this_offset=getUTF8BytePos(srcPtr, srcEnd, this_offset); | |
| result=pos(substr.body, this_offset, lang); | |
| if(result==CORD_NOT_FOUND) | |
| return STRING_NOT_FOUND; | |
| // convert 'result' from 'bytes' to 'characters' | |
| result=getUTF8CharPos(srcPtr, srcEnd, result); | |
| } else { | |
| result=pos(substr.body, this_offset, lang); | |
| if(result==CORD_NOT_FOUND) | |
| return STRING_NOT_FOUND; | |
| } | |
| return result; | |
| } | |
| void String::split(ArrayString& result, | void String::split(ArrayString& result, |
| size_t& pos_after, | size_t& pos_after, |
| const char* delim, | const char* delim, |
| Line 330 void String::split(ArrayString& result, | Line 435 void String::split(ArrayString& result, |
| } | } |
| } | } |
| static void regex_options(const String* options, int *result, bool& need_pre_post_match){ | enum Match_feature { |
| MF_NEED_PRE_POST_MATCH = 0x01, | |
| MF_JUST_COUNT_MATCHES = 0x02 | |
| }; | |
| static void regex_options(const String* options, int* result, int* match_features){ | |
| struct Regex_option { | struct Regex_option { |
| const char* keyL; | const char* key; |
| const char* keyU; | const char* keyAlt; |
| int clear, set; | int clear; |
| int set; | |
| int *result; | int *result; |
| bool *flag; | int flag; |
| } regex_option[]={ | } regex_option[]={ |
| {"i", "I", 0, PCRE_CASELESS, result, 0}, // a=A | {"i", "I", 0, PCRE_CASELESS, result, 0}, // a=A |
| {"s", "S", 0, PCRE_DOTALL, result, 0}, // \n\n$ [default] | {"s", "S", 0, PCRE_DOTALL, result, 0}, // ^\n\n$ [default] |
| {"x", "U", 0, PCRE_EXTENDED, result, 0}, // whitespace in regex ignored | |
| {"m", "M", PCRE_DOTALL, PCRE_MULTILINE, result, 0}, // ^aaa\n$^bbb\n$ | {"m", "M", PCRE_DOTALL, PCRE_MULTILINE, result, 0}, // ^aaa\n$^bbb\n$ |
| {"x", 0, 0, PCRE_EXTENDED, result, 0}, // whitespace in regex ignored | |
| {"U", 0, 0, PCRE_UNGREEDY, result, 0}, // ungreedy patterns (greedy by default) | |
| {"g", "G", 0, 1, result+1, 0}, // many rows | {"g", "G", 0, 1, result+1, 0}, // many rows |
| {"'", 0, 0, 0, 0, &need_pre_post_match}, | {"'", 0, 0, 0, 0, MF_NEED_PRE_POST_MATCH}, |
| {"n", 0, 0, 0, 0, MF_JUST_COUNT_MATCHES}, | |
| {0, 0, 0, 0, 0, 0} | {0, 0, 0, 0, 0, 0} |
| }; | }; |
| result[0]=PCRE_EXTRA | PCRE_DOTALL | PCRE_DOLLAR_ENDONLY; | result[0]=PCRE_EXTRA | PCRE_DOTALL | PCRE_DOLLAR_ENDONLY; |
| result[1]=0; | result[1]=0; |
| if(options && !options->is_empty()) | if(options && !options->is_empty()) |
| for(Regex_option *o=regex_option; o->keyL; o++) | for(Regex_option *o=regex_option; o->key; o++) |
| if(options->pos(o->keyL)!=STRING_NOT_FOUND | if( |
| || (o->keyU && options->pos(o->keyU)!=STRING_NOT_FOUND)) { | options->pos(o->key)!=STRING_NOT_FOUND |
| if(o->flag) | || (o->keyAlt && options->pos(o->keyAlt)!=STRING_NOT_FOUND) |
| *o->flag=true; | ){ |
| else { // result | if(o->flag){ |
| (*match_features) |= o->flag; | |
| } else { | |
| *o->result &= ~o->clear; | *o->result &= ~o->clear; |
| *o->result |= o->set; | *o->result |= o->set; |
| } | } |
| Line 366 Table* String::match(Charset& source_cha | Line 481 Table* String::match(Charset& source_cha |
| const String& regexp, | const String& regexp, |
| const String* options, | const String* options, |
| Row_action row_action, void *info, | Row_action row_action, void *info, |
| bool& just_matched) const { | int& matches_count) const { |
| if(regexp.is_empty()) | if(regexp.is_empty()) |
| throw Exception(0, | throw Exception(0, |
| 0, | 0, |
| "regexp is empty"); | "regexp is empty"); |
| const char* pattern=regexp.cstr(); | const char* pattern=regexp.cstr(String::L_UNSPECIFIED); // fix any tainted with L_REGEX |
| const char* errptr; | const char* errptr; |
| int erroffset; | int erroffset; |
| bool need_pre_post_match=false; | int option_bits[2]={0}; |
| int option_bits[2]={0}; regex_options(options, option_bits, need_pre_post_match); | int match_features=0; |
| regex_options(options, option_bits, &match_features); | |
| bool need_pre_post_match=(match_features & MF_NEED_PRE_POST_MATCH) != 0; | |
| bool just_count_matches=(match_features & MF_JUST_COUNT_MATCHES) != 0; | |
| bool global=option_bits[1]!=0; | bool global=option_bits[1]!=0; |
| if(source_charset.isUTF8()) | |
| option_bits[0]=option_bits[0] | PCRE_UTF8; | |
| pcre *code=pcre_compile(pattern, option_bits[0], | pcre *code=pcre_compile(pattern, option_bits[0], |
| &errptr, &erroffset, | &errptr, &erroffset, |
| source_charset.pcre_tables); | source_charset.pcre_tables); |
| Line 417 Table* String::match(Charset& source_cha | Line 539 Table* String::match(Charset& source_cha |
| if(exec_substrings==PCRE_ERROR_NOMATCH) { | if(exec_substrings==PCRE_ERROR_NOMATCH) { |
| pcre_free(code); | pcre_free(code); |
| row_action(table, 0/*last time, no raw*/, 0, 0, poststart, postfinish, info); | row_action(table, 0/*last time, no raw*/, 0, 0, poststart, postfinish, info); |
| if(global || subpatterns) | // if(global || subpatterns) |
| return &table; // global or with subpatterns=true+result | // return &table; // global or with subpatterns=true+result |
| else { | // else { |
| just_matched=false; return 0; // not global=no result | // just_matched=false; return 0; // not global=no result |
| } | // } |
| return just_count_matches ? 0 : &table; | |
| } | } |
| if(exec_substrings<0) { | if(exec_substrings<0) { |
| Line 450 Table* String::match(Charset& source_cha | Line 573 Table* String::match(Charset& source_cha |
| *row+=&mid(ovector[i*2+0], ovector[i*2+1]); // .i column value | *row+=&mid(ovector[i*2+0], ovector[i*2+1]); // .i column value |
| } | } |
| matches_count++; | |
| row_action(table, row, prestart, prefinish, poststart, postfinish, info); | row_action(table, row, prestart, prefinish, poststart, postfinish, info); |
| if(!global || prestart==poststart) { // not global | going to hang | if(!global || prestart==poststart) { // not global | going to hang |
| pcre_free(code); | pcre_free(code); |
| row_action(table, 0/*last time, no row*/, 0, 0, poststart, postfinish, info); | row_action(table, 0/*last time, no row*/, 0, 0, poststart, postfinish, info); |
| return &table; | return just_count_matches ? 0 : &table; |
| // return &table; | |
| } | } |
| prestart=poststart; | prestart=poststart; |
| Line 472 String& String::change_case(Charset& sou | Line 597 String& String::change_case(Charset& sou |
| return result; | return result; |
| char* new_cstr=cstrm(); | char* new_cstr=cstrm(); |
| size_t new_cstr_len=length(); | |
| if(source_charset.isUTF8()) { | if(source_charset.isUTF8()) { |
| size_t new_cstr_len=length(); | |
| switch(kind) { | switch(kind) { |
| case CC_UPPER: | case CC_UPPER: |
| change_case_UTF8((const XMLByte*)new_cstr, new_cstr_len, (XMLByte*)new_cstr, new_cstr_len, UTF8CaseToUpper); | change_case_UTF8((const XMLByte*)new_cstr, new_cstr_len, (XMLByte*)new_cstr, new_cstr_len, UTF8CaseToUpper); |
| Line 522 String& String::change_case(Charset& sou | Line 648 String& String::change_case(Charset& sou |
| return result; | return result; |
| } | } |
| const String& String::escape(Charset& source_charset) const { | |
| if(is_empty()) | |
| return *this; | |
| return Charset::escape(*this, source_charset); | |
| } | |
| const String& String::replace(const Dictionary& dict) const { | const String& String::replace(const Dictionary& dict) const { |
| String& result=*new String(); | String& result=*new String(); |
| const char* old_cstr=cstr(); | const char* old_cstr=cstr(); |
| Line 575 static int serialize_lang_piece(char ala | Line 708 static int serialize_lang_piece(char ala |
| } | } |
| String::Cm String::serialize(size_t prolog_length) const { | String::Cm String::serialize(size_t prolog_length) const { |
| size_t fragments_count=langs.count(); | size_t fragments_count=langs.count(); |
| size_t body_length=body.length(); | |
| size_t buf_length= | size_t buf_length= |
| prolog_length //1 | prolog_length //1 |
| +sizeof(size_t) //2 | +sizeof(size_t) //2 |
| +fragments_count*(sizeof(char)+sizeof(size_t)) //3 | +body_length //3 |
| +body.length() //4 | +1 // 4 for zero terminator used in deserialize |
| +1; // for zero terminator used in deserialize | +sizeof(size_t) //5 |
| +fragments_count*(sizeof(char)+sizeof(size_t)); //6 | |
| String::Cm result(new(PointerFreeGC) char[buf_length], buf_length); | String::Cm result(new(PointerFreeGC) char[buf_length], buf_length); |
| // 1: prolog | // 1: prolog |
| char *cur=result.str+prolog_length; | char *cur=result.str+prolog_length; |
| // 2: langs.count [WARNING: not cast, addresses must be %4=0 on sparc] | // 2: chars.count [WARNING: not cast, addresses must be %4=0 on sparc] |
| memcpy(cur, &body_length, sizeof(body_length)); cur+=sizeof(body_length); | |
| // 3: letters | |
| body.for_each(serialize_body_char, serialize_body_piece, &cur); | |
| // 4: zero terminator | |
| *cur++=0; | |
| // 5: langs.count [WARNING: not cast, addresses must be %4=0 on sparc] | |
| memcpy(cur, &fragments_count, sizeof(fragments_count)); cur+=sizeof(fragments_count); | memcpy(cur, &fragments_count, sizeof(fragments_count)); cur+=sizeof(fragments_count); |
| // 3: lang info | // 6: lang info |
| langs.for_each(body, serialize_lang_piece, &cur); | langs.for_each(body, serialize_lang_piece, &cur); |
| // 4: letters | |
| body.for_each(serialize_body_char, serialize_body_piece, &cur); | |
| // 5: zero terminator | |
| *cur=0; | |
| return result; | return result; |
| } | } |
| bool String::deserialize(size_t prolog_length, void *buf, size_t buf_length) { | bool String::deserialize(size_t prolog_size, void *buf, size_t buf_size) { |
| if(buf_length<=prolog_length) | size_t in_buf=buf_size; |
| if(in_buf<=prolog_size) | |
| return false; | return false; |
| buf_length-=prolog_length; | in_buf-=prolog_size; |
| buf_length-=1; // 5: zero terminator | |
| // 1: prolog | // 1: prolog |
| const char* cur=(const char* )buf+prolog_length; | const char* cur=(const char* )buf+prolog_size; |
| // 2: langs.count | // 2: chars.count |
| size_t body_length; | |
| if(in_buf<sizeof(body_length)) // body.length don't fit? | |
| return false; | |
| // [WARNING: not cast, addresses must be %4=0 on sparc] | |
| memcpy(&body_length, cur, sizeof(body_length)); cur+=sizeof(body_length); | |
| in_buf-=sizeof(body_length); | |
| if(in_buf<body_length+1) // letters+terminator don't fit? | |
| return false; | |
| // 4: zero terminator | |
| if(cur[body_length] != 0) // in place? | |
| return false; | |
| // 3: letters | |
| body=String::Body(cur, body_length); | |
| cur+=body_length+1; | |
| in_buf-=body_length+1; | |
| // 5: langs.count | |
| size_t fragments_count; | size_t fragments_count; |
| if(buf_length<sizeof(fragments_count)) // langs.count don't fit? | if(in_buf<sizeof(fragments_count)) // langs.count don't fit? |
| return false; | return false; |
| // [WARNING: not cast, addresses must be %4=0 on sparc] | // [WARNING: not cast, addresses must be %4=0 on sparc] |
| memcpy(&fragments_count, cur, sizeof(fragments_count)); cur+=sizeof(fragments_count); | memcpy(&fragments_count, cur, sizeof(fragments_count)); cur+=sizeof(fragments_count); |
| buf_length-=sizeof(fragments_count); | in_buf-=sizeof(fragments_count); |
| if(fragments_count) { | if(fragments_count) { |
| // 3: lang info | // 6: lang info |
| size_t total_length=0; | size_t total_length=0; |
| for(size_t f=0; f<fragments_count; f++) { | for(size_t f=0; f<fragments_count; f++) { |
| char lang; | char lang; |
| size_t fragment_length; | size_t fragment_length; |
| size_t piece_length=sizeof(lang)+sizeof(fragment_length); | size_t piece_length=sizeof(lang)+sizeof(fragment_length); |
| if(buf_length<piece_length) // lang+length | if(in_buf<piece_length) // lang+length |
| return false; | return false; |
| // lang | // lang |
| Line 628 bool String::deserialize(size_t prolog_l | Line 784 bool String::deserialize(size_t prolog_l |
| // length [WARNING: not cast, addresses must be %4=0 on sparc] | // length [WARNING: not cast, addresses must be %4=0 on sparc] |
| memcpy(&fragment_length, cur, sizeof(fragment_length)); cur+=sizeof(fragment_length); | memcpy(&fragment_length, cur, sizeof(fragment_length)); cur+=sizeof(fragment_length); |
| size_t combined_length=total_length+fragment_length; | |
| if(combined_length>body_length) | |
| return false; // file curruption | |
| // uchar needed to prevent propagating 0x80 bit to upper bytes | // uchar needed to prevent propagating 0x80 bit to upper bytes |
| langs.append(total_length, (String::Language)(uchar)lang, fragment_length); | langs.append(total_length, (String::Language)(uchar)lang, fragment_length); |
| total_length+=fragment_length; | total_length=combined_length; |
| in_buf-=piece_length; | |
| buf_length-=piece_length; | |
| } | } |
| // 4: letters | if(total_length!=body_length) // length(all language fragments) vs length(letters) |
| if(buf_length!=total_length) | |
| return false; | return false; |
| // serialize wrote extra zero byte there, we can rely on that | |
| body=String::Body(cur, buf_length); | |
| } | } |
| if(in_buf!=0) // some strange extra bytes | |
| return false; | |
| ASSERT_STRING_INVARIANT(*this); | ASSERT_STRING_INVARIANT(*this); |
| return true; | return true; |
| Line 650 bool String::deserialize(size_t prolog_l | Line 806 bool String::deserialize(size_t prolog_l |
| const char* String::Body::v() const { | const char* String::Body::v() const { |
| return CORD_to_const_char_star(body); | return CORD_to_const_char_star(body); |
| } | } |
| void String::Body::dump() const { | |
| CORD_dump(body); | |
| } | |
| const char* String::Languages::v() const { | const char* String::Languages::v() const { |
| if(opt.is_not_just_lang) | if(opt.is_not_just_lang) |
| return CORD_to_const_char_star(langs); | return CORD_to_const_char_star(langs); |
| else | else |
| return (const char*)&langs; | return (const char*)&langs; |
| } | } |
| void String::Languages::dump() const { | |
| if(opt.is_not_just_lang) | |
| CORD_dump(langs); | |
| else | |
| puts((const char*)&langs); | |
| } | |
| const char* String::v() const { | const char* String::v() const { |
| const int LIMIT_VIEW=20; | const uint LIMIT_VIEW=20; |
| char* buf=(char*)malloc(MAX_STRING); | char* buf=(char*)malloc(MAX_STRING); |
| const char*body_view=body.v(); | const char*body_view=body.v(); |
| const char*langs_view=langs.v(); | const char*langs_view=langs.v(); |
| Line 670 const char* String::v() const { | Line 836 const char* String::v() const { |
| return buf; | return buf; |
| } | } |
| void String::dump() const { | |
| body.dump(); | |
| langs.dump(); | |
| } | |
| const String& String::trim(String::Trim_kind kind, const char* chars) const { | const String& String::trim(String::Trim_kind kind, const char* chars) const { |
| if(!length()) | if(!length()) |
| return *this; | return *this; |
| Line 693 const String& String::trim(String::Trim_ | Line 862 const String& String::trim(String::Trim_ |
| ASSERT_STRING_INVARIANT(result); | ASSERT_STRING_INVARIANT(result); |
| return result; | return result; |
| } | |
| } |