Annotation of parser3/src/main/pa_string.C, revision 1.172.2.21.2.9
1.45 paf 1: /** @file
1.172.2.21.2. (paf 2:): Parser: string class. @see untalength_t.C.
1.46 paf 3:
1.172.2.11 paf 4: Copyright (c) 2001-2003 ArtLebedev Group (http://www.artlebedev.com)
1.138 paf 5: Author: Alexandr Petrosian <paf@design.ru> (http://paf.design.ru)
1.164 paf 6: */
1.46 paf 7:
1.172.2.21.2. (paf 8:): static const char* IDENT_STRING_C="$Date: 2003/03/20 13:09:54 $";
1.4 paf 9:
1.70 paf 10: #include "pcre.h"
11:
1.12 paf 12: #include "pa_string.h"
1.22 paf 13: #include "pa_exception.h"
1.61 paf 14: #include "pa_table.h"
1.101 parser 15: #include "pa_dictionary.h"
1.132 paf 16: #include "pa_charset.h"
1.60 paf 17:
1.172.2.2 paf 18: // helpers
1.139 paf 19:
1.172.2.2 paf 20: /// String::match uses this as replace & global search table columns
1.139 paf 21:
1.172.2.4 paf 22: const int MAX_MATCH_GROUPS=100;
23:
1.172.2.14 paf 24: class String_match_table_template_columns: public ArrayString {
1.172.2.2 paf 25: public:
1.172.2.4 paf 26: String_match_table_template_columns() {
1.172.2.21.2. (paf 27:): *this+=new String("prematch");
28:): *this+=new String("match");
29:): *this+=new String("postmatch");
1.172.2.4 paf 30: for(int i=0; i<MAX_MATCH_GROUPS; i++) {
1.172.2.21.2. (paf 31:): char *cname=new(PointerFreeGC) char[3/*strlen("100")*/+1/*terminating 0*/];
32:): *this+=new String(cname, sprintf(cname, "%d", 1+i));
1.172.2.2 paf 33: }
1.172.2.21 paf 34: }
1.172.2.4 paf 35: };
36:
1.172.2.21.2. (paf 37:): Table string_match_table_template(new String_match_table_template_columns);
1.172.2.2 paf 38:
1.172.2.21.2. (paf 39:): // String::ArrayFragment methods
40:):
41:): String::ArrayFragment& String::ArrayFragment::append(const ArrayFragment& src,
42:): size_t substr_start, size_t substr_end) {
43:): if(substr_start==substr_end)
44:): return *this;
45:):
46:): size_t fragment_start=0;
47:): size_t fragment_end;
48:): for(Array_iterator<element_type> i(src); ; fragment_start=fragment_end) {
49:): const element_type& fragment=i.next();
50:): fragment_end=fragment_start+fragment.length;
51:):
52:): if(substr_start<fragment_start) // not reached fragments which may include 'substr'?
53:): continue;
54:):
55:): // found first fragment including piece of 'substr'
56:): size_t piece_end=min(substr_end, fragment_end);
57:): *this+=Fragment(fragment.lang, piece_end-substr_start);
58:):
59:): while(substr_end>fragment_end) { // are there more fragments including pieces of 'substr'?
60:): const element_type& fragment=i.next();
61:): fragment_end=fragment_start+fragment.length;
62:):
63:): if(substr_end>fragment_end) // are there still more?
64:): simple_append(Fragment(fragment.lang, fragment.length)); // appending whole fragment
65:): else { // no, it was last
66:): simple_append(Fragment(fragment.lang, substr_end-fragment_start));
67:): goto break2;
68:): }
69:): }
70:):
71:): break;
72:): }
73:): break2:
74:):
75:): return *this;
76:): }
77:): /*
78:): void String::ArrayFragment::mid(ArrayFragment& result, size_t substr_start, size_t substr_end) {
79:): }
80:): */
81:):
82:): // String methods
1.172.2.2 paf 83:
1.172.2.21.2. (paf 84:): String::String(const char* cstr, size_t helper_length, bool tainted): body(CORD_EMPTY) {
85:): append_cstr(cstr, helper_length, tainted?UL_TAINTED:UL_CLEAN);
1.1 paf 86: }
1.140 paf 87:
1.172.2.21.2. (paf 88:): String::String(const String& src): body(src.body), fragments(src.fragments) {}
1.28 paf 89:
1.172.2.21.2. (paf 90:): String& String::append_cstr(const char* heap_variable, size_t helper_length, Untaint_lang lang) {
91:): if(!heap_variable)
92:): return *this;
93:): size_t length=helper_length?helper_length:strlen(heap_variable);
94:): if(!length)
95:): return *this;
96:):
97:): body=CORD_cat_char_star(body, heap_variable, length);
98:): fragments+=Fragment(lang, length);
99:):
100:): return *this;
101:): }
102:):
103:): String& String::append_copy(const char* auto_variable, size_t helper_length, Untaint_lang lang) {
104:): if(!auto_variable)
1.9 paf 105: return *this;
1.172.2.21.2. (paf 106:): size_t length=helper_length?helper_length:strlen(auto_variable);
107:): if(!length)
1.9 paf 108: return *this;
1.122 paf 109:
1.172.2.21.2. (paf 110:): body=pa_strdup(auto_variable, length);
111:): fragments+=Fragment(lang, length);
1.1 paf 112:
113: return *this;
114: }
115:
1.172.2.21.2. (paf 116:): static int CORD_batched_iter_fn_generic_hash_code(const char* s, void * client_data) {
117:): uint& result=*static_cast<uint*>(client_data);
118:): generic_hash_code(result, s);
119:): return 0;
120:): };
1.16 paf 121: uint String::hash_code() const {
1.7 paf 122: uint result=0;
1.172.2.21.2. (paf 123:): CORD_iter5(body, 0, 0, CORD_batched_iter_fn_generic_hash_code, &result);
1.5 paf 124: return result;
125: }
126:
1.172.2.21.2. (paf 127:): const String& String::mid(size_t substr_start, size_t substr_end) const {
128:): String& result=*new String;
1.33 paf 129:
1.172.2.21.2. (paf 130:): substr_start=min(substr_start, length());
131:): substr_end=max(substr_start, substr_end);
132:): if(substr_start==substr_end)
133:): return result;
1.52 paf 134:
1.172.2.21.2. (paf 135:): // first: letters themselves
136:): result.body=CORD_substr(body, substr_start, substr_end-substr_start);
1.46 paf 137:
1.172.2.21.2. (paf 138:): // next: their langs
139:): result.fragments.append(fragments, substr_start, substr_end);
1.53 paf 140:
1.60 paf 141: // SAPI::log(pool(), "piece of '%s' from %d to %d is '%s'",
1.172.2.21.2. (paf 142:): //cstr(), substr_start, substr_end, result.cstr());
1.53 paf 143: return result;
1.54 paf 144: }
145:
1.172.2.21.2. (paf 146:): size_t String::pos(CORD substr,
147:): size_t this_offset, Untaint_lang lang) const {
148:): // first: letters themselves
149:): size_t substr_start=CORD_str(body, this_offset, substr);
150:): if(substr_start==CORD_NOT_FOUND)
151:): return STRING_NOT_FOUND;
152:):
153:): // next: check the lang when specified
154:):
155:): if(lang==UL_UNSPECIFIED) // ignore lang?
156:): return substr_start;
157:):
158:): // substr must be in one fragment, and fragments' lang must = lang
159:): size_t substr_end=substr_start+CORD_len(substr);
160:): size_t fragment_start=0;
161:): size_t fragment_end;
162:): for(Array_iterator<ArrayFragment::element_type> i(fragments); i.has_next(); fragment_start=fragment_end) {
163:): const Fragment& fragment=i.next();
164:): fragment_end=fragment_start+fragment.length;
165:):
166:): if(substr_start<fragment_start) // not reached fragments which may include 'result'?
167:): continue;
168:):
169:): if(substr_end>fragment_end) // end of substr OUT of current fragment?
170:): break;
171:):
172:): return fragment.lang==lang?substr_end:STRING_NOT_FOUND;
173:): }
174:):
175:): return STRING_NOT_FOUND;
1.58 paf 176: }
177:
1.172.2.21.2. (paf 178:): size_t String::pos(const String& substr,
179:): size_t this_offset, Untaint_lang lang) const {
180:): return pos(substr.body, this_offset, lang);
1.60 paf 181: }
182:
1.172.2.14 paf 183: void String::split(ArrayString& result,
1.172.2.21.2. (paf 184:): size_t& pos_after,
185:): const char* delim,
1.172.2.10 paf 186: Untaint_lang lang, int limit) {
1.172.2.21.2. (paf 187:): size_t self_length=length();
188:): if(size_t delim_length=strlen(delim)) {
1.60 paf 189: int pos_before;
190: // while we have 'delim'...
1.172.2.21.2. (paf 191:): for(; (pos_before=pos(delim, pos_after, lang))>=0 && limit; limit--) {
192:): result+=&mid(pos_after, pos_before);
193:): pos_after=pos_before+delim_length;
1.60 paf 194: }
195: // last piece
1.172.2.21.2. (paf 196:): if(pos_after<self_length && limit) {
197:): result+=&mid(pos_after, self_length);
198:): pos_after=self_length;
1.60 paf 199: }
200: } else { // empty delim
1.172.2.21.2. (paf 201:): result+=this;
202:): pos_after+=self_length;
1.60 paf 203: }
204: }
205:
1.172.2.14 paf 206: void String::split(ArrayString& result,
1.172.2.21.2. (paf 207:): size_t& pos_after,
1.60 paf 208: const String& delim, Untaint_lang lang,
1.172.2.21.2. (paf 209:): int limit) const {
210:): if(delim) {
1.60 paf 211: int pos_before;
212: // while we have 'delim'...
213: for(; (pos_before=pos(delim, pos_after, lang))>=0 && limit; limit--) {
1.172.2.21.2. (paf 214:): result+=&mid(pos_after, pos_before);
215:): pos_after=pos_before+delim.length();
1.60 paf 216: }
217: // last piece
1.172.2.21.2. (paf 218:): if(pos_after<length() && limit) {
219:): result+=&mid(pos_after, length());
220:): pos_after=length();
1.60 paf 221: }
222: } else { // empty delim
1.172.2.21.2. (paf 223:): result+=this;
224:): pos_after+=length();
1.60 paf 225: }
1.61 paf 226: }
227:
1.172.2.21.2. (paf 228:): static void regex_options(const String& options, int *result, bool& need_pre_post_match){
1.63 paf 229: struct Regex_option {
1.172.2.11 paf 230: const char* keyL;
231: const char* keyU;
1.63 paf 232: int clear, set;
233: int *result;
1.154 paf 234: bool *flag;
1.63 paf 235: } regex_option[]={
1.153 paf 236: {"i", "I", 0, PCRE_CASELESS, result}, // a=A
237: {"s", "S", 0, PCRE_DOTALL, result}, // \n\n$ [default]
238: {"x", "U", 0, PCRE_EXTENDED, result}, // whitespace in regex ignored
239: {"m", "M", PCRE_DOTALL, PCRE_MULTILINE, result}, // ^aaa\n$^bbb\n$
240: {"g", "G", 0, true, result+1}, // many rows
1.154 paf 241: {"'", 0, 0, 0, 0, &need_pre_post_match},
242: {0}
1.63 paf 243: };
1.171 paf 244: result[0]=PCRE_EXTRA | PCRE_DOTALL | PCRE_DOLLAR_ENDONLY;
1.63 paf 245: result[1]=0;
246:
247: if(options)
1.153 paf 248: for(Regex_option *o=regex_option; o->keyL; o++)
1.172.2.21.2. (paf 249:): if(options.pos(o->keyL)>=0
250:): || (o->keyU && options.pos(o->keyU)>=0)) {
1.154 paf 251: if(o->flag)
252: *o->flag=true;
253: else { // result
254: *o->result &= ~o->clear;
255: *o->result |= o->set;
256: }
1.63 paf 257: }
258: }
259:
1.172.2.21.2. (paf 260:): Table* String::match(Charset& source_charset,
1.172.2.4 paf 261: const String& regexp,
1.172.2.21.2. (paf 262:): const String& options,
1.172.2.4 paf 263: Row_action row_action, void *info,
1.172.2.20 paf 264: bool& just_matched) const {
1.172.2.21.2. (paf 265:): if(!regexp)
266:): throw Exception(0,
267:): 0,
1.73 paf 268: "regexp is empty");
1.154 paf 269:
1.172.2.21.2. (paf 270:): const char* pattern=regexp.cstr();
1.172.2.11 paf 271: const char* errptr;
1.62 paf 272: int erroffset;
1.172.2.21.2. (paf 273:): bool need_pre_post_match=false;
1.154 paf 274: int option_bits[2]; regex_options(options, option_bits, need_pre_post_match);
1.172.2.20 paf 275: bool global=option_bits[1]!=0;
1.172.2.21.2. (paf 276:): pcre *code=pcre_compile(pattern, option_bits[0],
1.62 paf 277: &errptr, &erroffset,
1.172.2.1 paf 278: source_charset.pcre_tables);
1.62 paf 279:
1.67 paf 280: if(!code)
1.172.2.21.2. (paf 281:): throw Exception(0,
282:): ®exp.mid(erroffset, regexp.length()),
1.74 paf 283: "regular expression syntax error - %s", errptr);
1.62 paf 284:
1.172.2.20 paf 285: int subpatterns=pcre_info(code, 0, 0);
286: if(subpatterns<0) {
1.100 parser 287: pcre_free(code);
1.149 paf 288: throw Exception(0,
1.172.2.21.2. (paf 289:): ®exp,
1.76 paf 290: "pcre_info error (%d)",
1.172.2.20 paf 291: subpatterns);
1.63 paf 292: }
293:
1.172.2.21.2. (paf 294:): const char* subject=cstr();
295:): size_t subject_length=strlen(subject);
296:): const int oveclength=(1/*match*/+MAX_MATCH_GROUPS)*3;
297:): int ovector[oveclength];
1.155 paf 298:
299: // create table
1.172.2.21.2. (paf 300:): Table* table=new Table(string_match_table_template);
1.63 paf 301:
1.64 paf 302: int exec_option_bits=0;
1.154 paf 303: int prestart=0;
304: int poststart=0;
1.172.2.21.2. (paf 305:): int postfinish=length();
1.63 paf 306: while(true) {
307: int exec_substrings=pcre_exec(code, 0,
1.172.2.21.2. (paf 308:): subject, subject_length, prestart,
309:): exec_option_bits, ovector, oveclength);
1.63 paf 310:
311: if(exec_substrings==PCRE_ERROR_NOMATCH) {
1.100 parser 312: pcre_free(code);
1.172.2.21.2. (paf 313:): row_action(table, 0/*last time, no raw*/, 0, 0, poststart, postfinish, info);
1.172.2.20 paf 314: if(global || subpatterns)
315: return table; // global or with subpatterns=true+result
316: else {
1.172.2.21.2. (paf 317:): just_matched=false; return 0; // not global=no result
1.172.2.20 paf 318: }
1.63 paf 319: }
320:
321: if(exec_substrings<0) {
1.100 parser 322: pcre_free(code);
1.172.2.21.2. (paf 323:): throw Exception(0,
324:): ®exp,
1.76 paf 325: "regular expression execute error (%d)",
1.63 paf 326: exec_substrings);
327: }
328:
1.154 paf 329: int prefinish=ovector[0];
330: poststart=ovector[1];
1.172.2.21.2. (paf 331:): ArrayString* row=new ArrayString;
1.172.2.4 paf 332: if(need_pre_post_match) {
1.172.2.21.2. (paf 333:): *row+=&mid(0, prefinish); // .prematch column value
334:): *row+=&mid(prefinish, poststart); // .match
335:): *row+=&mid(poststart, postfinish); // .postmatch
1.172.2.4 paf 336: } else {
1.172.2.21.2. (paf 337:): *row+=0; // .prematch column value
338:): *row+=0; // .match
339:): *row+=0; // .postmatch
1.172.2.4 paf 340: }
1.63 paf 341:
342: for(int i=1; i<exec_substrings; i++) {
1.69 paf 343: // -1:-1 case handled peacefully by mid() itself
1.172.2.21.2. (paf 344:): *row+=&mid(ovector[i*2+0], ovector[i*2+1]); // .i column value
1.63 paf 345: }
346:
1.172.2.20 paf 347: row_action(table, row, prestart, prefinish, poststart, postfinish, info);
1.63 paf 348:
1.172.2.20 paf 349: if(!global || prestart==poststart) { // not global | going to hang
1.100 parser 350: pcre_free(code);
1.172.2.21.2. (paf 351:): row_action(table, 0/*last time, no row*/, 0, 0, poststart, postfinish, info);
1.172.2.20 paf 352: return table;
1.63 paf 353: }
1.154 paf 354: prestart=poststart;
1.63 paf 355:
356: /*
357: if(option_bits[0] & PCRE_MULTILINE)
1.64 paf 358: exec_option_bits|=PCRE_NOTBOL; // start of subject+startoffset not BOL
1.63 paf 359: */
360: }
1.82 parser 361: }
362:
1.172.2.21.2. (paf 363:): String& String::change_case(Charset& source_charset, Change_case_kind kind) {
364:): String& result=*new String();
1.172.2.4 paf 365:
1.172.2.1 paf 366: const unsigned char *tables=source_charset.pcre_tables;
1.82 parser 367:
368: const unsigned char *a;
369: const unsigned char *b;
370: switch(kind) {
371: case CC_UPPER:
372: a=tables+lcc_offset;
373: b=tables+fcc_offset;
374: break;
375: case CC_LOWER:
376: a=tables+lcc_offset;
377: b=0;
378: break;
379: default:
1.172.2.21.2. (paf 380:): throw Exception(0,
381:): this,
1.82 parser 382: "unknown change case kind #%d",
383: static_cast<int>(kind)); // never
384: a=b=0; // calm, compiler
385: break; // never
386: }
387:
1.172.2.21.2. (paf 388:): char* new_cstr=cstrm();
389:): char *dest=new_cstr;
390:): unsigned char index;
391:): for(const char* current=new_cstr; index=(unsigned char)*current; current++) {
392:): unsigned char c=a[index];
393:): if(b)
394:): c=b[c];
1.82 parser 395:
1.172.2.21.2. (paf 396:): *dest++=(char)c;
397:): }
398:): result.body=new_cstr;
399:): result.fragments.append(fragments, 0, fragments.count());
1.89 parser 400:
1.101 parser 401: return result;
402: }
403:
1.172.2.21.2. (paf 404:): const String& String::replace(const Dictionary& dict) const {
405:): String& result=*new String();
406:): const char* old_cstr=cstr();
407:): const char* prematch_start=old_cstr;
408:):
409:): for(const char* current=old_cstr; *current; ) {
410:): if(Table::element_type row=dict.first_that_starts(current)) {
411:): // prematch
412:): if(size_t prematch_length=current-prematch_start) {
413:): result.body=CORD_cat_char_star(result.body, prematch_start, prematch_length);
414:): result.fragments.append(fragments, prematch_start-old_cstr, prematch_length);
1.101 parser 415: }
416:
1.172.2.21.2. (paf 417:): // match
418:):
419:): const String* a=row->get(0);
420:): // skip 'a' in 'current'; move prematch_start
421:): current+=a->length(); prematch_start=current;
1.170 paf 422:
1.172.2.21.2. (paf 423:): if(row->count()>1) { // are there any b?
424:): const String* b=row->get(1);
425:): if(size_t b_length=b->length()) {
426:): result.body=CORD_cat(result.body, b->cstr_to_cord());
427:): result.fragments.append(b->fragments, 0, b->fragments.count());
428:): }
429:): }
430:): } else // simply advance
431:): current++;
432:): }
1.156 paf 433:
1.172.2.21.2. (paf 434:): // postmatch
435:): if(size_t prematch_length=current-prematch_start) {
436:): result.body=CORD_cat_char_star(result.body, prematch_start, prematch_length);
437:): result.fragments.append(fragments, prematch_start-old_cstr, prematch_length);
438:): }
1.156 paf 439:
1.89 parser 440: return result;
441: }
442:
1.172.2.21.2. (paf 443:): double String::as_double() const {
1.89 parser 444: double result;
1.172.2.21.2. (paf 445:): const char *str=cstr();
446:):
447:): while(*str && isspace(*str))
448:): str++;
449:): if(!*str)
1.162 paf 450: return 0;
1.161 paf 451:
1.102 parser 452: char *error_pos;
1.89 parser 453: // 0xABC
1.172.2.21.2. (paf 454:): if(str[0]=='0')
455:): if(str[1]=='x' || str[1]=='X')
456:): result=(double)(unsigned long)strtol(str, &error_pos, 0);
1.99 parser 457: else
1.172.2.21.2. (paf 458:): result=(double)strtod(str+1/*skip leading 0*/, &error_pos);
1.89 parser 459: else
1.172.2.21.2. (paf 460:): result=(double)strtod(str, &error_pos);
1.89 parser 461:
1.159 paf 462: while(char c=*error_pos++)
463: if(!isspace(c))
464: throw Exception("number.format",
1.172.2.21.2. (paf 465:): this,
1.159 paf 466: "invalid number (double)");
1.89 parser 467:
468: return result;
469: }
1.172.2.21.2. (paf 470:): int String::as_int() const {
1.89 parser 471: int result;
1.172.2.21.2. (paf 472:): const char *str=cstr();
473:):
474:): while(*str && isspace(*str))
475:): str++;
476:): if(!*str)
1.162 paf 477: return 0;
1.161 paf 478:
1.102 parser 479: char *error_pos;
1.89 parser 480: // 0xABC
1.172.2.21.2. (paf 481:): if(str[0]=='0')
482:): if(str[1]=='x' || str[1]=='X')
483:): result=(int)(unsigned long)strtol(str, &error_pos, 0);
1.99 parser 484: else
1.172.2.21.2. (paf 485:): result=(int)strtol(str+1/*skip leading 0*/, &error_pos, 0);
1.89 parser 486: else
1.172.2.21.2. (paf 487:): result=(int)strtol(str, &error_pos, 0);
1.89 parser 488:
1.159 paf 489: while(char c=*error_pos++)
490: if(!isspace(c))
491: throw Exception("number.format",
1.172.2.21.2. (paf 492:): this,
1.159 paf 493: "invalid number (int)");
1.82 parser 494:
495: return result;
1.61 paf 496: }
1.113 parser 497:
1.172.2.4 paf 498: inline void uint2uchars(uint word, uchar *bytes) {
499: bytes[0]=word&0xFF;
500: bytes[1]=(word>>8)&0xFF;
501: bytes[2]=(word>>16)&0xFF;
502: bytes[3]=(word>>24)&0xFF;
503: }
504: inline uint uchars2uint(uchar *bytes) {
505: return bytes[3]<<24
506: | bytes[2]<<16
507: | bytes[1]<<8
508: | bytes[0];
509: }
510:
1.172.2.21.2. (paf 511:): static int CORD_batched_iter_fn_append(const char* s, void* client_data) {
512:): char*& cur=*static_cast<char**>(client_data);
513:):
514:): size_t length=strlen(s);
515:): memcpy(cur, s, length); cur+=length;
516:): return 0;
517:): };
518:): String::C String::serialize(size_t prolog_length) const {
519:): size_t buf_length=
520:): prolog_length
521:): +fragments.count()*(sizeof(Untaint_lang)+sizeof(size_t))
522:): +length();
523:): C result(new(PointerFreeGC) char[buf_length], buf_length);
524:):
525:): // 1: prolog
526:): char *cur=result.str+prolog_length;
527:):
528:):
529:): // 2: fragments.count
530:): size_t fragments_count=fragments.count();
531:): memcpy(cur, &fragments_count, sizeof(fragments_count)); cur+=sizeof(fragments_count);
1.113 parser 532:
1.172.2.21.2. (paf 533:): // 3: lang info
534:): for(Array_iterator<ArrayFragment::element_type> i(fragments); i.has_next(); ) {
535:): const Fragment& fragment=i.next();
1.123 paf 536: // lang
1.172.2.21.2. (paf 537:): memcpy(cur, &fragment.lang, sizeof(fragment.lang)); cur+=sizeof(fragment.lang);
538:): // length
539:): memcpy(cur, &fragment.length, sizeof(fragment.length)); cur+=sizeof(fragment.length);
540:): }
541:):
542:): // 4: letters
543:): CORD_iter5(body, 0, 0, CORD_batched_iter_fn_append, &cur);
544:):
545:): return result;
1.113 parser 546: }
1.172.2.21.2. (paf 547:): bool String::deserialize(size_t prolog_length, void *buf, size_t buf_length, const char* file) {
548:): if(buf_length<=prolog_length)
1.148 paf 549: return false;
1.172.2.21.2. (paf 550:): buf_length-=prolog_length;
1.113 parser 551:
1.172.2.21.2. (paf 552:): // 1: prolog
553:): const char* cur=(const char* )buf+prolog_length;
1.148 paf 554:
1.172.2.21.2. (paf 555:): // 2: fragments.count
556:): if(buf_length<sizeof(size_t)) // fragments.count don't fit?
557:): return false;
558:): size_t fragments_count=*reinterpret_cast<const size_t*>(cur); cur+=sizeof(size_t);
559:): buf_length-=sizeof(size_t);
1.128 paf 560:
1.172.2.21.2. (paf 561:): // 3: lang info
562:): size_t total_length=0;
563:): for(size_t f=0; f<fragments_count; f++) {
564:): size_t piece_length=sizeof(Untaint_lang)+sizeof(size_t);
565:): if(buf_length<piece_length) // lang+length
1.148 paf 566: return false;
567:
1.172.2.21.2. (paf 568:): Untaint_lang lang=*reinterpret_cast<const Untaint_lang *>(cur); cur+=sizeof(Untaint_lang);
569:): size_t fragment_length=*reinterpret_cast<const size_t*>(cur); cur+=sizeof(size_t);
570:): fragments+=Fragment(lang, fragment_length);
571:): total_length+=fragment_length;
1.113 parser 572:
1.172.2.21.2. (paf 573:): buf_length-=piece_length;
1.113 parser 574: }
1.172.2.21.2. (paf 575:):
576:): // 4: letters
577:): if(buf_length!=total_length)
578:): return false;
579:):
580:): body=CORD_cat_char_star(CORD_EMPTY, cur, buf_length);
581:):
1.148 paf 582: return true;
1.113 parser 583: }
E-mail:
![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to pa_string.C CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [parser3project] / parser3 / src / main