Annotation of parser3/src/main/pa_string.C, revision 1.194
1.45 paf 1: /** @file
1.174 paf 2: Parser: string class. @see untalength_t.C.
1.46 paf 3:
1.194 ! paf 4: Copyright (c) 2001-2004 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:
1.194 ! paf 8: static const char * const IDENT_STRING_C="$Date: 2004/01/30 09:56:49 $";
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:
1.185 paf 18: const String String::Empty;
19:
1.193 paf 20: int pa_atoi(const char* str, const String* problem_source) {
21: if(!str)
22: return 0;
23:
24: while(*str && isspace(*str))
25: str++;
26: if(!*str)
27: return 0;
28:
29: int result;
30: char *error_pos;
31: // 0xABC
32: if(str[0]=='0')
33: if(str[1]=='x' || str[1]=='X')
34: result=(int)(unsigned long)strtol(str, &error_pos, 0);
35: else
36: result=(int)strtol(str+1/*skip leading 0*/, &error_pos, 0);
37: else
38: result=(int)strtol(str, &error_pos, 0);
39:
40: while(char c=*error_pos++)
41: if(!isspace(c))
42: throw Exception("number.format",
43: problem_source,
44: problem_source?"invalid number (int)": "'%s' is invalid number (int)", str);
45:
46: return result;
47: }
48:
49: double pa_atod(const char* str, const String* problem_source) {
50: if(!str)
51: return 0;
52:
53: while(*str && isspace(*str))
54: str++;
55: if(!*str)
56: return 0;
57:
58: double result;
59: char *error_pos;
60: // 0xABC
61: if(str[0]=='0')
62: if(str[1]=='x' || str[1]=='X')
63: result=(double)(unsigned long)strtol(str, &error_pos, 0);
64: else
65: result=(double)strtod(str+1/*skip leading 0*/, &error_pos);
66: else
67: result=(double)strtod(str, &error_pos);
68:
69: while(char c=*error_pos++)
70: if(!isspace(c))
71: throw Exception("number.format",
72: problem_source,
73: problem_source?"invalid number (double)": "'%s' is invalid number (double)", str);
74:
75: return result;
76: }
77:
1.176 paf 78: // cord lib extension
79:
80: #ifndef DOXYGEN
81: typedef struct {
82: ssize_t countdown;
83: char target; /* Character we're looking for */
84: } chr_data;
85: #endif
86: static int CORD_range_contains_chr_greater_then_proc(char c, size_t size, void* client_data)
87: {
88: register chr_data * d = (chr_data *)client_data;
89:
90: if (d -> countdown<=0) return(2);
91: d -> countdown -= size;
92: if (c > d -> target) return(1);
93: return(0);
94: }
95: int CORD_range_contains_chr_greater_then(CORD x, size_t i, size_t n, int c)
96: {
97: chr_data d;
98:
99: d.countdown = n;
100: d.target = c;
101: return(CORD_block_iter(x, i, CORD_range_contains_chr_greater_then_proc, &d) == 1/*alternatives: 0 normally ended, 2=struck 'n'*/);
102: }
103:
1.187 paf 104: static int CORD_block_count_proc(char /*c*/, size_t /*size*/, void* client_data)
1.178 paf 105: {
106: int* result=(int*)client_data;
107: (*result)++;
108: return(0); // 0=continue
109: }
110: size_t CORD_block_count(CORD x)
111: {
112: size_t result=0;
113: CORD_block_iter(x, 0, CORD_block_count_proc, &result);
114: return result;
115: }
116:
1.174 paf 117: // helpers
1.139 paf 118:
1.174 paf 119: /// String::match uses this as replace & global search table columns
1.139 paf 120:
1.174 paf 121: const int MAX_MATCH_GROUPS=100;
1.139 paf 122:
1.174 paf 123: class String_match_table_template_columns: public ArrayString {
124: public:
125: String_match_table_template_columns() {
126: *this+=new String("prematch");
127: *this+=new String("match");
128: *this+=new String("postmatch");
129: for(int i=0; i<MAX_MATCH_GROUPS; i++) {
1.176 paf 130: *this+=new String(String::Body::Format(1+i), String::L_CLEAN);
1.174 paf 131: }
132: }
133: };
134:
135: Table string_match_table_template(new String_match_table_template_columns);
136:
1.176 paf 137: // String::Body methods
1.140 paf 138:
1.176 paf 139: String::Body String::Body::Format(int value) {
1.174 paf 140: char local[MAX_NUMBER];
141: size_t length=snprintf(local, MAX_NUMBER, "%d", value);
1.176 paf 142: return String::Body(pa_strdup(local, length), length);
1.120 paf 143: }
144:
1.174 paf 145: static int CORD_batched_iter_fn_generic_hash_code(char c, void * client_data) {
146: uint& result=*static_cast<uint*>(client_data);
147: generic_hash_code(result, c);
148: return 0;
149: }
150: static int CORD_batched_iter_fn_generic_hash_code(const char* s, void * client_data) {
151: uint& result=*static_cast<uint*>(client_data);
152: generic_hash_code(result, s);
153: return 0;
154: };
1.176 paf 155: uint String::Body::hash_code() const {
1.174 paf 156: uint result=0;
157: CORD_iter5(body, 0,
158: CORD_batched_iter_fn_generic_hash_code,
159: CORD_batched_iter_fn_generic_hash_code, &result);
1.120 paf 160: return result;
1.94 parser 161: }
162:
1.174 paf 163: // String methods
164:
165: String::String(const char* cstr, size_t helper_length, bool tainted): body(CORD_EMPTY) {
166: append_help_length(cstr, helper_length, tainted?L_TAINTED:L_CLEAN);
1.115 paf 167: }
1.174 paf 168: String::String(const String::C cstr, bool tainted): body(CORD_EMPTY) {
169: append_know_length(cstr.str, cstr.length, tainted?L_TAINTED:L_CLEAN);
1.5 paf 170: }
1.28 paf 171:
1.174 paf 172: String& String::append_know_length(const char* str, size_t known_length, Language lang) {
173: if(!known_length)
1.9 paf 174: return *this;
1.122 paf 175:
1.176 paf 176: // first: langs
177: langs.append(body, lang, known_length);
178: // next: letters themselves
1.174 paf 179: body.append_know_length(str, known_length);
1.1 paf 180:
1.174 paf 181: ASSERT_STRING_INVARIANT(*this);
1.1 paf 182: return *this;
183: }
1.174 paf 184: String& String::append_help_length(const char* str, size_t helper_length, Language lang) {
185: if(!str)
186: return *this;
187: size_t known_length=helper_length?helper_length:strlen(str);
188: if(!known_length)
189: return *this;
1.1 paf 190:
1.174 paf 191: return append_know_length(str, known_length, lang);
1.5 paf 192: }
1.174 paf 193: String& String::append_strdup(const char* str, size_t helper_length, Language lang) {
194: size_t known_length=helper_length?helper_length:strlen(str);
195: if(!known_length)
196: return *this;
1.5 paf 197:
1.176 paf 198: // first: langs
199: langs.append(body, lang, known_length);
200: // next: letters themselves
1.174 paf 201: body.append_strdup_know_length(str, known_length);
1.33 paf 202:
1.174 paf 203: ASSERT_STRING_INVARIANT(*this);
204: return *this;
1.5 paf 205: }
1.46 paf 206:
1.174 paf 207: /// @todo check in doc: whether it documents NOW bad situation "abc".mid(-1, 3) =were?="ab"
208: String& String::mid(size_t substr_begin, size_t substr_end) const {
209: String& result=*new String;
210:
211: size_t self_length=length();
212: substr_begin=min(substr_begin, self_length);
213: substr_end=min(max(substr_end, substr_begin), self_length);
1.176 paf 214: size_t substr_length=substr_end-substr_begin;
215: if(!substr_length)
1.107 parser 216: return result;
1.53 paf 217:
1.176 paf 218: // first: their langs
219: result.langs.append(result.body, langs, substr_begin, substr_length);
220: // next: letters themselves
221: result.body=body.mid(substr_begin, substr_length);
1.174 paf 222:
223: // SAPI::log("piece of '%s' from %d to %d is '%s'",
224: //cstr(), substr_begin, substr_end, result.cstr());
225: ASSERT_STRING_INVARIANT(result);
1.53 paf 226: return result;
1.54 paf 227: }
228:
1.176 paf 229: size_t String::pos(const String::Body substr, size_t this_offset, Language lang) const {
1.183 paf 230: size_t substr_length=substr.length();
231: while(true) {
232: size_t substr_begin=body.pos(substr, this_offset);
233:
234: if(substr_begin==CORD_NOT_FOUND)
235: return STRING_NOT_FOUND;
1.174 paf 236:
1.183 paf 237: if(langs.check_lang(lang, substr_begin, substr_length))
238: return substr_begin;
239:
240: this_offset=substr_begin+substr_length;
241: }
1.58 paf 242: }
243:
1.174 paf 244: size_t String::pos(const String& substr,
245: size_t this_offset, Language lang) const {
246: return pos(substr.body, this_offset, lang);
1.60 paf 247: }
248:
1.174 paf 249: void String::split(ArrayString& result,
250: size_t& pos_after,
251: const char* delim,
252: Language lang, int limit) const {
253: size_t self_length=length();
254: if(size_t delim_length=strlen(delim)) {
1.186 paf 255: size_t pos_before;
1.60 paf 256: // while we have 'delim'...
1.174 paf 257: for(; (pos_before=pos(delim, pos_after, lang))!=STRING_NOT_FOUND && limit; limit--) {
1.69 paf 258: result+=&mid(pos_after, pos_before);
1.174 paf 259: pos_after=pos_before+delim_length;
1.60 paf 260: }
261: // last piece
1.174 paf 262: if(pos_after<self_length && limit) {
263: result+=&mid(pos_after, self_length);
264: pos_after=self_length;
1.60 paf 265: }
266: } else { // empty delim
267: result+=this;
1.174 paf 268: pos_after+=self_length;
1.60 paf 269: }
270: }
271:
1.174 paf 272: void String::split(ArrayString& result,
273: size_t& pos_after,
274: const String& delim, Language lang,
275: int limit) const {
1.140 paf 276: if(!delim.is_empty()) {
1.186 paf 277: size_t pos_before;
1.60 paf 278: // while we have 'delim'...
1.174 paf 279: for(; (pos_before=pos(delim, pos_after, lang))!=STRING_NOT_FOUND && limit; limit--) {
1.69 paf 280: result+=&mid(pos_after, pos_before);
1.174 paf 281: pos_after=pos_before+delim.length();
1.60 paf 282: }
283: // last piece
1.174 paf 284: if(pos_after<length() && limit) {
285: result+=&mid(pos_after, length());
286: pos_after=length();
1.60 paf 287: }
288: } else { // empty delim
289: result+=this;
1.174 paf 290: pos_after+=length();
1.60 paf 291: }
1.61 paf 292: }
293:
1.174 paf 294: static void regex_options(const String* options, int *result, bool& need_pre_post_match){
1.63 paf 295: struct Regex_option {
1.174 paf 296: const char* keyL;
297: const char* keyU;
1.63 paf 298: int clear, set;
299: int *result;
1.154 paf 300: bool *flag;
1.63 paf 301: } regex_option[]={
1.189 paf 302: {"i", "I", 0, PCRE_CASELESS, result, 0}, // a=A
303: {"s", "S", 0, PCRE_DOTALL, result, 0}, // \n\n$ [default]
304: {"x", "U", 0, PCRE_EXTENDED, result, 0}, // whitespace in regex ignored
305: {"m", "M", PCRE_DOTALL, PCRE_MULTILINE, result, 0}, // ^aaa\n$^bbb\n$
306: {"g", "G", 0, 1, result+1, 0}, // many rows
1.154 paf 307: {"'", 0, 0, 0, 0, &need_pre_post_match},
1.189 paf 308: {0, 0, 0, 0, 0, 0}
1.63 paf 309: };
1.171 paf 310: result[0]=PCRE_EXTRA | PCRE_DOTALL | PCRE_DOLLAR_ENDONLY;
1.63 paf 311: result[1]=0;
312:
1.174 paf 313: if(options && !options->is_empty())
1.153 paf 314: for(Regex_option *o=regex_option; o->keyL; o++)
1.174 paf 315: if(options->pos(o->keyL)!=STRING_NOT_FOUND
316: || (o->keyU && options->pos(o->keyU)!=STRING_NOT_FOUND)) {
1.154 paf 317: if(o->flag)
318: *o->flag=true;
319: else { // result
320: *o->result &= ~o->clear;
321: *o->result |= o->set;
322: }
1.63 paf 323: }
324: }
325:
1.174 paf 326: Table* String::match(Charset& source_charset,
327: const String& regexp,
328: const String* options,
329: Row_action row_action, void *info,
330: bool& just_matched) const {
1.140 paf 331: if(regexp.is_empty())
1.149 paf 332: throw Exception(0,
1.174 paf 333: 0,
1.73 paf 334: "regexp is empty");
1.154 paf 335:
1.174 paf 336: const char* pattern=regexp.cstr();
337: const char* errptr;
1.62 paf 338: int erroffset;
1.173 paf 339: bool need_pre_post_match=false;
1.174 paf 340: int option_bits[2]={0}; regex_options(options, option_bits, need_pre_post_match);
341: bool global=option_bits[1]!=0;
1.63 paf 342: pcre *code=pcre_compile(pattern, option_bits[0],
1.62 paf 343: &errptr, &erroffset,
1.174 paf 344: source_charset.pcre_tables);
1.62 paf 345:
1.67 paf 346: if(!code)
1.149 paf 347: throw Exception(0,
1.174 paf 348: ®exp.mid(erroffset, regexp.length()),
1.74 paf 349: "regular expression syntax error - %s", errptr);
1.62 paf 350:
1.174 paf 351: int subpatterns=pcre_info(code, 0, 0);
352: if(subpatterns<0) {
1.100 parser 353: pcre_free(code);
1.149 paf 354: throw Exception(0,
1.174 paf 355: ®exp,
1.76 paf 356: "pcre_info error (%d)",
1.174 paf 357: subpatterns);
1.63 paf 358: }
359:
1.174 paf 360: const char* subject=cstr();
361: size_t subject_length=strlen(subject);
362: const int oveclength=(1/*match*/+MAX_MATCH_GROUPS)*3;
363: int ovector[oveclength];
1.155 paf 364:
365: // create table
1.173 paf 366: Table::Action_options table_options;
1.174 paf 367: Table& table=*new Table(string_match_table_template, table_options);
1.63 paf 368:
1.64 paf 369: int exec_option_bits=0;
1.154 paf 370: int prestart=0;
371: int poststart=0;
1.174 paf 372: int postfinish=length();
1.63 paf 373: while(true) {
374: int exec_substrings=pcre_exec(code, 0,
1.174 paf 375: subject, subject_length, prestart,
376: exec_option_bits, ovector, oveclength);
1.63 paf 377:
378: if(exec_substrings==PCRE_ERROR_NOMATCH) {
1.100 parser 379: pcre_free(code);
1.174 paf 380: row_action(table, 0/*last time, no raw*/, 0, 0, poststart, postfinish, info);
381: if(global || subpatterns)
382: return &table; // global or with subpatterns=true+result
383: else {
384: just_matched=false; return 0; // not global=no result
385: }
1.63 paf 386: }
387:
388: if(exec_substrings<0) {
1.100 parser 389: pcre_free(code);
1.149 paf 390: throw Exception(0,
1.174 paf 391: ®exp,
1.76 paf 392: "regular expression execute error (%d)",
1.63 paf 393: exec_substrings);
394: }
395:
1.154 paf 396: int prefinish=ovector[0];
397: poststart=ovector[1];
1.174 paf 398: ArrayString* row=new ArrayString;
399: if(need_pre_post_match) {
400: *row+=&mid(0, prefinish); // .prematch column value
401: *row+=&mid(prefinish, poststart); // .match
402: *row+=&mid(poststart, postfinish); // .postmatch
403: } else {
1.185 paf 404: *row+=&Empty; // .prematch column value
405: *row+=&Empty; // .match
406: *row+=&Empty; // .postmatch
1.174 paf 407: }
1.63 paf 408:
409: for(int i=1; i<exec_substrings; i++) {
1.69 paf 410: // -1:-1 case handled peacefully by mid() itself
1.174 paf 411: *row+=&mid(ovector[i*2+0], ovector[i*2+1]); // .i column value
1.63 paf 412: }
413:
1.174 paf 414: row_action(table, row, prestart, prefinish, poststart, postfinish, info);
1.63 paf 415:
1.174 paf 416: if(!global || prestart==poststart) { // not global | going to hang
1.100 parser 417: pcre_free(code);
1.174 paf 418: row_action(table, 0/*last time, no row*/, 0, 0, poststart, postfinish, info);
419: return &table;
1.63 paf 420: }
1.154 paf 421: prestart=poststart;
1.63 paf 422:
423: /*
424: if(option_bits[0] & PCRE_MULTILINE)
1.64 paf 425: exec_option_bits|=PCRE_NOTBOL; // start of subject+startoffset not BOL
1.63 paf 426: */
427: }
1.82 parser 428: }
429:
1.174 paf 430: String& String::change_case(Charset& source_charset, Change_case_kind kind) const {
431: String& result=*new String();
432: if(is_empty())
433: return result;
434:
435: char* new_cstr=cstrm();
1.192 paf 436: size_t new_cstr_len=length();
1.181 paf 437: if(source_charset.isUTF8()) {
438: switch(kind) {
439: case CC_UPPER:
1.192 paf 440: change_case_UTF8((const XMLByte*)new_cstr, new_cstr_len, (XMLByte*)new_cstr, new_cstr_len, UTF8CaseToUpper);
1.181 paf 441: break;
442: case CC_LOWER:
1.192 paf 443: change_case_UTF8((const XMLByte*)new_cstr, new_cstr_len, (XMLByte*)new_cstr, new_cstr_len, UTF8CaseToLower);
1.181 paf 444: break;
445: default:
446: assert(!"unknown change case kind");
447: break; // never
448: }
449:
450: } else {
451: const unsigned char *tables=source_charset.pcre_tables;
1.82 parser 452:
1.181 paf 453: const unsigned char *a;
454: const unsigned char *b;
455: switch(kind) {
456: case CC_UPPER:
457: a=tables+lcc_offset;
458: b=tables+fcc_offset;
459: break;
460: case CC_LOWER:
461: a=tables+lcc_offset;
462: b=0;
463: break;
464: default:
465: assert(!"unknown change case kind");
466: a=b=0; // calm, compiler
467: break; // never
468: }
469:
1.192 paf 470: char *dest=new_cstr;
1.181 paf 471: unsigned char index;
1.190 paf 472: for(const char* current=new_cstr; (index=(unsigned char)*current); current++) {
1.181 paf 473: unsigned char c=a[index];
474: if(b)
475: c=b[c];
476:
477: *dest++=(char)c;
478: }
1.174 paf 479: }
1.176 paf 480: result.langs=langs;
1.174 paf 481: result.body=new_cstr;
1.89 parser 482:
1.101 parser 483: return result;
484: }
485:
1.174 paf 486: const String& String::replace(const Dictionary& dict) const {
487: String& result=*new String();
488: const char* old_cstr=cstr();
489: const char* prematch_begin=old_cstr;
490:
491: const char* current=old_cstr;
492: while(*current) {
1.184 paf 493: if(Dictionary::Subst subst=dict.first_that_begins(current)) {
1.174 paf 494: // prematch
495: if(size_t prematch_length=current-prematch_begin) {
1.179 paf 496: result.langs.append(result.body, langs, prematch_begin-old_cstr, prematch_length);
1.174 paf 497: result.body.append_strdup_know_length(prematch_begin, prematch_length);
1.101 parser 498: }
499:
1.174 paf 500: // match
501: // skip 'a' in 'current'; move prematch_begin
1.184 paf 502: current+=subst.from_length; prematch_begin=current;
1.174 paf 503:
1.184 paf 504: if(const String* b=subst.to) // are there any b?
1.174 paf 505: result<<*b;
506: } else // simply advance
507: current++;
508: }
1.156 paf 509:
1.174 paf 510: // postmatch
511: if(size_t postmatch_length=current-prematch_begin) {
1.179 paf 512: result.langs.append(result.body, langs, prematch_begin-old_cstr, postmatch_length);
1.174 paf 513: result.body.append_strdup_know_length(prematch_begin, postmatch_length);
514: }
1.156 paf 515:
1.174 paf 516: ASSERT_STRING_INVARIANT(result);
1.82 parser 517: return result;
1.61 paf 518: }
1.113 parser 519:
1.180 paf 520: static int serialize_body_char(char c, char** cur) {
521: *((*cur)++)=c;
522: return 0; // 0=continue
523: };
1.174 paf 524: static int serialize_body_piece(const char* s, char** cur) {
525: size_t length=strlen(s);
526: memcpy(*cur, s, length); *cur+=length;
1.178 paf 527: return 0; // 0=continue
1.174 paf 528: };
1.178 paf 529: static int serialize_lang_piece(char alang, size_t asize, char** cur) {
530: // lang
1.191 paf 531: **cur=alang; (*cur)++;
532: // length [WARNING: not cast, addresses must be %4=0 on sparc]
1.178 paf 533: memcpy(*cur, &asize, sizeof(asize)); *cur+=sizeof(asize);
534:
535: return 0; // 0=continue
536: }
1.174 paf 537: String::Cm String::serialize(size_t prolog_length) const {
1.178 paf 538: size_t fragments_count=langs.count();
1.174 paf 539: size_t buf_length=
1.178 paf 540: prolog_length //1
541: +sizeof(size_t) //2
542: +fragments_count*(sizeof(char)+sizeof(size_t)) //3
543: +body.length() //4
544: +1; // for zero terminator used in deserialize
1.174 paf 545: String::Cm result(new(PointerFreeGC) char[buf_length], buf_length);
546:
547: // 1: prolog
548: char *cur=result.str+prolog_length;
1.191 paf 549: // 2: langs.count [WARNING: not cast, addresses must be %4=0 on sparc]
1.174 paf 550: memcpy(cur, &fragments_count, sizeof(fragments_count)); cur+=sizeof(fragments_count);
551: // 3: lang info
1.178 paf 552: langs.for_each(body, serialize_lang_piece, &cur);
1.174 paf 553: // 4: letters
1.180 paf 554: body.for_each(serialize_body_char, serialize_body_piece, &cur);
1.182 paf 555: // 5: zero terminator
556: *cur=0;
1.113 parser 557:
1.174 paf 558: return result;
1.113 parser 559: }
1.174 paf 560: bool String::deserialize(size_t prolog_length, void *buf, size_t buf_length) {
561: if(buf_length<=prolog_length)
1.148 paf 562: return false;
1.174 paf 563: buf_length-=prolog_length;
1.178 paf 564: buf_length-=1; // 5: zero terminator
1.135 paf 565:
1.174 paf 566: // 1: prolog
567: const char* cur=(const char* )buf+prolog_length;
1.113 parser 568:
1.176 paf 569: // 2: langs.count
1.191 paf 570: size_t fragments_count;
571: if(buf_length<sizeof(fragments_count)) // langs.count don't fit?
1.174 paf 572: return false;
1.191 paf 573: // [WARNING: not cast, addresses must be %4=0 on sparc]
574: memcpy(&fragments_count, cur, sizeof(fragments_count)); cur+=sizeof(fragments_count);
575: buf_length-=sizeof(fragments_count);
1.174 paf 576:
577: if(fragments_count) {
578: // 3: lang info
579: size_t total_length=0;
580: for(size_t f=0; f<fragments_count; f++) {
1.191 paf 581: char lang;
582: size_t fragment_length;
583: size_t piece_length=sizeof(lang)+sizeof(fragment_length);
1.174 paf 584: if(buf_length<piece_length) // lang+length
585: return false;
586:
1.191 paf 587: // lang
588: lang=*cur++;
589: // length [WARNING: not cast, addresses must be %4=0 on sparc]
590: memcpy(&fragment_length, cur, sizeof(fragment_length)); cur+=sizeof(fragment_length);
591:
592: // uchar needed to prevent propagating 0x80 bit to upper bytes
593: langs.append(total_length, (String::Language)(uchar)lang, fragment_length);
1.174 paf 594: total_length+=fragment_length;
1.148 paf 595:
1.174 paf 596: buf_length-=piece_length;
597: }
1.128 paf 598:
1.174 paf 599: // 4: letters
600: if(buf_length!=total_length)
1.148 paf 601: return false;
602:
1.178 paf 603: // serialize wrote extra zero byte there, we can rely on that
1.176 paf 604: body=String::Body(cur, buf_length);
1.174 paf 605: }
1.113 parser 606:
1.174 paf 607: ASSERT_STRING_INVARIANT(*this);
1.148 paf 608: return true;
1.176 paf 609: }
610:
611: const char* String::Body::v() const {
612: return CORD_to_const_char_star(body);
613: }
614: const char* String::Languages::v() const {
1.177 paf 615: if(opt.is_not_just_lang)
1.176 paf 616: return CORD_to_const_char_star(langs);
617: else
618: return (const char*)&langs;
619: }
620: const char* String::v() const {
621: #define LIMIT_VIEW 20
622: char* buf=(char*)malloc(MAX_STRING);
623: const char*body_view=body.v();
624: const char*langs_view=langs.v();
625: snprintf(buf, MAX_STRING,
1.178 paf 626: "%d:%.*s%s} "
1.176 paf 627: "{%d:%s",
1.178 paf 628: langs.count(), LIMIT_VIEW, langs_view, strlen(langs_view)>LIMIT_VIEW?"...":"",
1.176 paf 629: strlen(body_view), body_view
630: );
631:
632: return buf;
633: #undef LIMIT_VIEW
1.113 parser 634: }
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