1 /*
2 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
3 % %
4 % %
5 % %
6 % TTTTT OOO K K EEEEE N N %
7 % T O O K K E NN N %
8 % T O O KKK EEE N N N %
9 % T O O K K E N NN %
10 % T OOO K K EEEEE N N %
11 % %
12 % %
13 % MagickCore Token Methods %
14 % %
15 % Software Design %
16 % Cristy %
17 % January 1993 %
18 % %
19 % %
20 % Copyright 1999-2021 ImageMagick Studio LLC, a non-profit organization %
21 % dedicated to making software imaging solutions freely available. %
22 % %
23 % You may not use this file except in compliance with the License. You may %
24 % obtain a copy of the License at %
25 % %
26 % https://imagemagick.org/script/license.php %
27 % %
28 % Unless required by applicable law or agreed to in writing, software %
29 % distributed under the License is distributed on an "AS IS" BASIS, %
30 % WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. %
31 % See the License for the specific language governing permissions and %
32 % limitations under the License. %
33 % %
34 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
35 %
36 %
37 %
38 */
39
40 /*
41 Include declarations.
42 */
43 #include "MagickCore/studio.h"
44 #include "MagickCore/exception.h"
45 #include "MagickCore/exception-private.h"
46 #include "MagickCore/image.h"
47 #include "MagickCore/image-private.h"
48 #include "MagickCore/memory_.h"
49 #include "MagickCore/memory-private.h"
50 #include "MagickCore/string_.h"
51 #include "MagickCore/string-private.h"
52 #include "MagickCore/token.h"
53 #include "MagickCore/token-private.h"
54 #include "MagickCore/utility.h"
55 #include "MagickCore/utility-private.h"
56
57 /*
58 Typedef declaractions.
59 */
60 struct _TokenInfo
61 {
62 int
63 state;
64
65 MagickStatusType
66 flag;
67
68 ssize_t
69 offset;
70
71 char
72 quote;
73
74 size_t
75 signature;
76 };
77
78 /*
79 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
80 % %
81 % %
82 % %
83 % A c q u i r e T o k e n I n f o %
84 % %
85 % %
86 % %
87 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
88 %
89 % AcquireTokenInfo() allocates the TokenInfo structure.
90 %
91 % The format of the AcquireTokenInfo method is:
92 %
93 % TokenInfo *AcquireTokenInfo()
94 %
95 */
AcquireTokenInfo(void)96 MagickExport TokenInfo *AcquireTokenInfo(void)
97 {
98 TokenInfo
99 *token_info;
100
101 token_info=(TokenInfo *) AcquireCriticalMemory(sizeof(*token_info));
102 token_info->signature=MagickCoreSignature;
103 return(token_info);
104 }
105
106 /*
107 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
108 % %
109 % %
110 % %
111 % D e s t r o y T o k e n I n f o %
112 % %
113 % %
114 % %
115 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
116 %
117 % DestroyTokenInfo() deallocates memory associated with an TokenInfo
118 % structure.
119 %
120 % The format of the DestroyTokenInfo method is:
121 %
122 % TokenInfo *DestroyTokenInfo(TokenInfo *token_info)
123 %
124 % A description of each parameter follows:
125 %
126 % o token_info: Specifies a pointer to an TokenInfo structure.
127 %
128 */
DestroyTokenInfo(TokenInfo * token_info)129 MagickExport TokenInfo *DestroyTokenInfo(TokenInfo *token_info)
130 {
131 (void) LogMagickEvent(TraceEvent,GetMagickModule(),"...");
132 assert(token_info != (TokenInfo *) NULL);
133 assert(token_info->signature == MagickCoreSignature);
134 token_info->signature=(~MagickCoreSignature);
135 token_info=(TokenInfo *) RelinquishMagickMemory(token_info);
136 return(token_info);
137 }
138
139 /*
140 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
141 % %
142 % %
143 % %
144 + G e t N e x t T o k e n %
145 % %
146 % %
147 % %
148 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
149 %
150 % GetNextToken() gets a token from the token stream. A token is defined as
151 % a sequence of characters delimited by whitespace (e.g. clip-path), a
152 % sequence delimited with quotes (.e.g "Quote me"), or a sequence enclosed in
153 % parenthesis (e.g. rgb(0,0,0)). GetNextToken() also recognizes these
154 % separator characters: ':', '=', ',', and ';'. GetNextToken() returns the
155 % length of the consumed token.
156 %
157 % The format of the GetNextToken method is:
158 %
159 % size_t GetNextToken(const char *magick_restrict start,
160 % const char **magick_restrict end,const size_t extent,
161 % char *magick_restrict token)
162 %
163 % A description of each parameter follows:
164 %
165 % o start: the start of the token sequence.
166 %
167 % o end: point to the end of the token sequence.
168 %
169 % o extent: maximum extent of the token.
170 %
171 % o token: copy the token to this buffer.
172 %
173 */
GetNextToken(const char * magick_restrict start,const char ** magick_restrict end,const size_t extent,char * magick_restrict token)174 MagickExport magick_hot_spot size_t GetNextToken(
175 const char *magick_restrict start,const char **magick_restrict end,
176 const size_t extent,char *magick_restrict token)
177 {
178 double
179 value;
180
181 char
182 *magick_restrict q;
183
184 const char
185 *magick_restrict p;
186
187 ssize_t
188 i;
189
190 assert(start != (const char *) NULL);
191 assert(token != (char *) NULL);
192 i=0;
193 p=start;
194 while ((isspace((int) ((unsigned char) *p)) != 0) && (*p != '\0'))
195 p++;
196 switch (*p)
197 {
198 case '\0':
199 break;
200 case '"':
201 case '\'':
202 case '`':
203 case '{':
204 {
205 char
206 escape;
207
208 switch (*p)
209 {
210 case '"': escape='"'; break;
211 case '\'': escape='\''; break;
212 case '`': escape='\''; break;
213 case '{': escape='}'; break;
214 default: escape=(*p); break;
215 }
216 for (p++; *p != '\0'; p++)
217 {
218 if ((*p == '\\') && ((*(p+1) == escape) || (*(p+1) == '\\')))
219 p++;
220 else
221 if (*p == escape)
222 {
223 p++;
224 break;
225 }
226 if (i < (ssize_t) (extent-1))
227 token[i++]=(*p);
228 if ((size_t) (p-start) >= (extent-1))
229 break;
230 }
231 break;
232 }
233 case '/':
234 {
235 if (i < (ssize_t) (extent-1))
236 token[i++]=(*p);
237 p++;
238 if ((*p == '>') || (*p == '/'))
239 {
240 if (i < (ssize_t) (extent-1))
241 token[i++]=(*p);
242 p++;
243 }
244 break;
245 }
246 default:
247 {
248 value=StringToDouble(p,&q);
249 (void) value;
250 if ((p != q) && (*p != ','))
251 {
252 for ( ; (p < q) && (*p != ','); p++)
253 {
254 if (i < (ssize_t) (extent-1))
255 token[i++]=(*p);
256 if ((size_t) (p-start) >= (extent-1))
257 break;
258 }
259 if (*p == '%')
260 {
261 if (i < (ssize_t) (extent-1))
262 token[i++]=(*p);
263 p++;
264 }
265 break;
266 }
267 if ((*p != '\0') && (isalpha((int) ((unsigned char) *p)) == 0) &&
268 (*p != *DirectorySeparator) && (*p != '#') && (*p != '<'))
269 {
270 if (i < (ssize_t) (extent-1))
271 token[i++]=(*p);
272 p++;
273 break;
274 }
275 for ( ; *p != '\0'; p++)
276 {
277 if (((isspace((int) ((unsigned char) *p)) != 0) || (*p == '=') ||
278 (*p == ',') || (*p == ':') || (*p == ';')) && (*(p-1) != '\\'))
279 break;
280 if ((i > 0) && (*p == '<'))
281 break;
282 if (i < (ssize_t) (extent-1))
283 token[i++]=(*p);
284 if (*p == '>')
285 break;
286 if (*p == '(')
287 {
288 for (p++; *p != '\0'; p++)
289 {
290 if (i < (ssize_t) (extent-1))
291 token[i++]=(*p);
292 if ((*p == ')') && (*(p-1) != '\\'))
293 break;
294 if ((size_t) (p-start) >= (extent-1))
295 break;
296 }
297 if (*p == '\0')
298 break;
299 }
300 if ((size_t) (p-start) >= (extent-1))
301 break;
302 }
303 break;
304 }
305 }
306 token[i]='\0';
307 if (LocaleNCompare(token,"url(#",5) == 0)
308 {
309 q=strrchr(token,')');
310 if (q != (char *) NULL)
311 {
312 *q='\0';
313 (void) memmove(token,token+5,(size_t) (q-token-4));
314 }
315 }
316 while (isspace((int) ((unsigned char) *p)) != 0)
317 p++;
318 if (end != (const char **) NULL)
319 *end=(const char *) p;
320 return(p-start+1);
321 }
322
323 /*
324 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
325 % %
326 % %
327 % %
328 % G l o b E x p r e s s i o n %
329 % %
330 % %
331 % %
332 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
333 %
334 % GlobExpression() returns MagickTrue if the expression matches the pattern.
335 %
336 % The format of the GlobExpression function is:
337 %
338 % MagickBooleanType GlobExpression(const char *magick_restrict expression,
339 % const char *magick_restrict pattern,
340 % const MagickBooleanType case_insensitive)
341 %
342 % A description of each parameter follows:
343 %
344 % o expression: Specifies a pointer to a text string containing a file name.
345 %
346 % o pattern: Specifies a pointer to a text string containing a pattern.
347 %
348 % o case_insensitive: set to MagickTrue to ignore the case when matching
349 % an expression.
350 %
351 */
GlobExpression(const char * magick_restrict expression,const char * magick_restrict pattern,const MagickBooleanType case_insensitive)352 MagickExport MagickBooleanType GlobExpression(
353 const char *magick_restrict expression,const char *magick_restrict pattern,
354 const MagickBooleanType case_insensitive)
355 {
356 MagickBooleanType
357 done,
358 match;
359
360 /*
361 Return on empty pattern or '*'.
362 */
363 if (pattern == (char *) NULL)
364 return(MagickTrue);
365 if (GetUTFCode(pattern) == 0)
366 return(MagickTrue);
367 if (LocaleCompare(pattern,"*") == 0)
368 return(MagickTrue);
369 if ((GetUTFCode(pattern+strlen(pattern)-1) == ']') &&
370 (strchr(pattern,'[') != (char *) NULL))
371 {
372 ExceptionInfo
373 *exception;
374
375 ImageInfo
376 *image_info;
377
378 /*
379 Determine if pattern is a scene, i.e. img0001.pcd[2].
380 */
381 image_info=AcquireImageInfo();
382 (void) CopyMagickString(image_info->filename,pattern,MagickPathExtent);
383 exception=AcquireExceptionInfo();
384 (void) SetImageInfo(image_info,0,exception);
385 exception=DestroyExceptionInfo(exception);
386 if (LocaleCompare(image_info->filename,pattern) != 0)
387 {
388 image_info=DestroyImageInfo(image_info);
389 return(MagickFalse);
390 }
391 image_info=DestroyImageInfo(image_info);
392 }
393 /*
394 Evaluate glob expression.
395 */
396 done=MagickFalse;
397 while ((GetUTFCode(pattern) != 0) && (done == MagickFalse))
398 {
399 if (GetUTFCode(expression) == 0)
400 if ((GetUTFCode(pattern) != '{') && (GetUTFCode(pattern) != '*'))
401 break;
402 switch (GetUTFCode(pattern))
403 {
404 case '*':
405 {
406 MagickBooleanType
407 status;
408
409 status=MagickFalse;
410 while (GetUTFCode(pattern) == '*')
411 pattern+=GetUTFOctets(pattern);
412 while ((GetUTFCode(expression) != 0) && (status == MagickFalse))
413 {
414 status=GlobExpression(expression,pattern,case_insensitive);
415 expression+=GetUTFOctets(expression);
416 }
417 if (status != MagickFalse)
418 {
419 while (GetUTFCode(expression) != 0)
420 expression+=GetUTFOctets(expression);
421 while (GetUTFCode(pattern) != 0)
422 pattern+=GetUTFOctets(pattern);
423 }
424 break;
425 }
426 case '[':
427 {
428 int
429 c;
430
431 pattern+=GetUTFOctets(pattern);
432 for ( ; ; )
433 {
434 if ((GetUTFCode(pattern) == 0) || (GetUTFCode(pattern) == ']'))
435 {
436 done=MagickTrue;
437 break;
438 }
439 if (GetUTFCode(pattern) == '\\')
440 {
441 pattern+=GetUTFOctets(pattern);
442 if (GetUTFCode(pattern) == 0)
443 {
444 done=MagickTrue;
445 break;
446 }
447 }
448 if (GetUTFCode(pattern+GetUTFOctets(pattern)) == '-')
449 {
450 c=GetUTFCode(pattern);
451 pattern+=GetUTFOctets(pattern);
452 pattern+=GetUTFOctets(pattern);
453 if (GetUTFCode(pattern) == ']')
454 {
455 done=MagickTrue;
456 break;
457 }
458 if (GetUTFCode(pattern) == '\\')
459 {
460 pattern+=GetUTFOctets(pattern);
461 if (GetUTFCode(pattern) == 0)
462 {
463 done=MagickTrue;
464 break;
465 }
466 }
467 if ((GetUTFCode(expression) < c) ||
468 (GetUTFCode(expression) > GetUTFCode(pattern)))
469 {
470 pattern+=GetUTFOctets(pattern);
471 continue;
472 }
473 }
474 else
475 if (GetUTFCode(pattern) != GetUTFCode(expression))
476 {
477 pattern+=GetUTFOctets(pattern);
478 continue;
479 }
480 pattern+=GetUTFOctets(pattern);
481 while ((GetUTFCode(pattern) != ']') && (GetUTFCode(pattern) != 0))
482 {
483 if ((GetUTFCode(pattern) == '\\') &&
484 (GetUTFCode(pattern+GetUTFOctets(pattern)) > 0))
485 pattern+=GetUTFOctets(pattern);
486 pattern+=GetUTFOctets(pattern);
487 }
488 if (GetUTFCode(pattern) != 0)
489 {
490 pattern+=GetUTFOctets(pattern);
491 expression+=GetUTFOctets(expression);
492 }
493 break;
494 }
495 break;
496 }
497 case '?':
498 {
499 pattern+=GetUTFOctets(pattern);
500 expression+=GetUTFOctets(expression);
501 break;
502 }
503 case '{':
504 {
505 char
506 *target;
507
508 char
509 *p;
510
511 target=AcquireString(pattern);
512 p=target;
513 pattern++;
514 while ((GetUTFCode(pattern) != '}') && (GetUTFCode(pattern) != 0))
515 {
516 *p++=(*pattern++);
517 if ((GetUTFCode(pattern) == ',') || (GetUTFCode(pattern) == '}'))
518 {
519 *p='\0';
520 match=GlobExpression(expression,target,case_insensitive);
521 if (match != MagickFalse)
522 {
523 expression+=MagickMin(strlen(expression),strlen(target));
524 break;
525 }
526 p=target;
527 pattern+=GetUTFOctets(pattern);
528 }
529 }
530 while ((GetUTFCode(pattern) != '}') && (GetUTFCode(pattern) != 0))
531 pattern+=GetUTFOctets(pattern);
532 if (GetUTFCode(pattern) != 0)
533 pattern+=GetUTFOctets(pattern);
534 target=DestroyString(target);
535 break;
536 }
537 case '\\':
538 {
539 pattern+=GetUTFOctets(pattern);
540 if (GetUTFCode(pattern) == 0)
541 break;
542 }
543 default:
544 {
545 if (case_insensitive != MagickFalse)
546 {
547 if (LocaleLowercase((int) GetUTFCode(expression)) != LocaleLowercase((int) GetUTFCode(pattern)))
548 {
549 done=MagickTrue;
550 break;
551 }
552 }
553 else
554 if (GetUTFCode(expression) != GetUTFCode(pattern))
555 {
556 done=MagickTrue;
557 break;
558 }
559 expression+=GetUTFOctets(expression);
560 pattern+=GetUTFOctets(pattern);
561 }
562 }
563 }
564 while (GetUTFCode(pattern) == '*')
565 pattern+=GetUTFOctets(pattern);
566 match=(GetUTFCode(expression) == 0) && (GetUTFCode(pattern) == 0) ?
567 MagickTrue : MagickFalse;
568 return(match);
569 }
570
571 /*
572 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
573 % %
574 % %
575 % %
576 + I s G l o b %
577 % %
578 % %
579 % %
580 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
581 %
582 % IsGlob() returns MagickTrue if the path specification contains a globbing
583 % pattern.
584 %
585 % The format of the IsGlob method is:
586 %
587 % MagickBooleanType IsGlob(const char *geometry)
588 %
589 % A description of each parameter follows:
590 %
591 % o path: the path.
592 %
593 */
IsGlob(const char * path)594 MagickPrivate MagickBooleanType IsGlob(const char *path)
595 {
596 MagickBooleanType
597 status = MagickFalse;
598
599 const char
600 *p;
601
602 if (IsPathAccessible(path) != MagickFalse)
603 return(MagickFalse);
604 for (p=path; *p != '\0'; p++)
605 {
606 switch (*p)
607 {
608 case '*':
609 case '?':
610 case '{':
611 case '}':
612 case '[':
613 case ']':
614 {
615 status=MagickTrue;
616 break;
617 }
618 default:
619 break;
620 }
621 }
622 return(status);
623 }
624
625 /*
626 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
627 % %
628 % %
629 % %
630 % T o k e n i z e r %
631 % %
632 % %
633 % %
634 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
635 %
636 % Tokenizer() is a generalized, finite state token parser. It extracts tokens
637 % one at a time from a string of characters. The characters used for white
638 % space, for break characters, and for quotes can be specified. Also,
639 % characters in the string can be preceded by a specifiable escape character
640 % which removes any special meaning the character may have.
641 %
642 % Here is some terminology:
643 %
644 % o token: A single unit of information in the form of a group of
645 % characters.
646 %
647 % o white space: Apace that gets ignored (except within quotes or when
648 % escaped), like blanks and tabs. in addition, white space terminates a
649 % non-quoted token.
650 %
651 % o break set: One or more characters that separates non-quoted tokens.
652 % Commas are a common break character. The usage of break characters to
653 % signal the end of a token is the same as that of white space, except
654 % multiple break characters with nothing or only white space between
655 % generate a null token for each two break characters together.
656 %
657 % For example, if blank is set to be the white space and comma is set to
658 % be the break character, the line
659 %
660 % A, B, C , , DEF
661 %
662 % ... consists of 5 tokens:
663 %
664 % 1) "A"
665 % 2) "B"
666 % 3) "C"
667 % 4) "" (the null string)
668 % 5) "DEF"
669 %
670 % o Quote character: A character that, when surrounding a group of other
671 % characters, causes the group of characters to be treated as a single
672 % token, no matter how many white spaces or break characters exist in
673 % the group. Also, a token always terminates after the closing quote.
674 % For example, if ' is the quote character, blank is white space, and
675 % comma is the break character, the following string
676 %
677 % A, ' B, CD'EF GHI
678 %
679 % ... consists of 4 tokens:
680 %
681 % 1) "A"
682 % 2) " B, CD" (note the blanks & comma)
683 % 3) "EF"
684 % 4) "GHI"
685 %
686 % The quote characters themselves do not appear in the resultant
687 % tokens. The double quotes are delimiters i use here for
688 % documentation purposes only.
689 %
690 % o Escape character: A character which itself is ignored but which
691 % causes the next character to be used as is. ^ and \ are often used
692 % as escape characters. An escape in the last position of the string
693 % gets treated as a "normal" (i.e., non-quote, non-white, non-break,
694 % and non-escape) character. For example, assume white space, break
695 % character, and quote are the same as in the above examples, and
696 % further, assume that ^ is the escape character. Then, in the string
697 %
698 % ABC, ' DEF ^' GH' I ^ J K^ L ^
699 %
700 % ... there are 7 tokens:
701 %
702 % 1) "ABC"
703 % 2) " DEF ' GH"
704 % 3) "I"
705 % 4) " " (a lone blank)
706 % 5) "J"
707 % 6) "K L"
708 % 7) "^" (passed as is at end of line)
709 %
710 % The format of the Tokenizer method is:
711 %
712 % int Tokenizer(TokenInfo *token_info,const unsigned flag,char *token,
713 % const size_t max_token_length,const char *line,const char *white,
714 % const char *break_set,const char *quote,const char escape,
715 % char *breaker,int *next,char *quoted)
716 %
717 % A description of each parameter follows:
718 %
719 % o flag: right now, only the low order 3 bits are used.
720 %
721 % 1 => convert non-quoted tokens to upper case
722 % 2 => convert non-quoted tokens to lower case
723 % 0 => do not convert non-quoted tokens
724 %
725 % o token: a character string containing the returned next token
726 %
727 % o max_token_length: the maximum size of "token". Characters beyond
728 % "max_token_length" are truncated.
729 %
730 % o string: the string to be parsed.
731 %
732 % o white: a string of the valid white spaces. example:
733 %
734 % char whitesp[]={" \t"};
735 %
736 % blank and tab will be valid white space.
737 %
738 % o break: a string of the valid break characters. example:
739 %
740 % char breakch[]={";,"};
741 %
742 % semicolon and comma will be valid break characters.
743 %
744 % o quote: a string of the valid quote characters. An example would be
745 %
746 % char whitesp[]={"'\"");
747 %
748 % (this causes single and double quotes to be valid) Note that a
749 % token starting with one of these characters needs the same quote
750 % character to terminate it.
751 %
752 % for example:
753 %
754 % "ABC '
755 %
756 % is unterminated, but
757 %
758 % "DEF" and 'GHI'
759 %
760 % are properly terminated. Note that different quote characters
761 % can appear on the same line; only for a given token do the quote
762 % characters have to be the same.
763 %
764 % o escape: the escape character (NOT a string ... only one
765 % allowed). Use zero if none is desired.
766 %
767 % o breaker: the break character used to terminate the current
768 % token. If the token was quoted, this will be the quote used. If
769 % the token is the last one on the line, this will be zero.
770 %
771 % o next: this variable points to the first character of the
772 % next token. it gets reset by "tokenizer" as it steps through the
773 % string. Set it to 0 upon initialization, and leave it alone
774 % after that. You can change it if you want to jump around in the
775 % string or re-parse from the beginning, but be careful.
776 %
777 % o quoted: set to True if the token was quoted and MagickFalse
778 % if not. You may need this information (for example: in C, a
779 % string with quotes around it is a character string, while one
780 % without is an identifier).
781 %
782 % o result: 0 if we haven't reached EOS (end of string), and 1
783 % if we have.
784 %
785 */
786
787 #define IN_WHITE 0
788 #define IN_TOKEN 1
789 #define IN_QUOTE 2
790 #define IN_OZONE 3
791
sindex(int c,const char * string)792 static ssize_t sindex(int c,const char *string)
793 {
794 const char
795 *p;
796
797 for (p=string; *p != '\0'; p++)
798 if (c == (int) (*p))
799 return((ssize_t) (p-string));
800 return(-1);
801 }
802
StoreToken(TokenInfo * token_info,char * string,size_t max_token_length,int c)803 static void StoreToken(TokenInfo *token_info,char *string,
804 size_t max_token_length,int c)
805 {
806 ssize_t
807 i;
808
809 if ((token_info->offset < 0) ||
810 ((size_t) token_info->offset >= (max_token_length-1)))
811 return;
812 i=token_info->offset++;
813 string[i]=(char) c;
814 if (token_info->state == IN_QUOTE)
815 return;
816 switch (token_info->flag & 0x03)
817 {
818 case 1:
819 {
820 string[i]=(char) LocaleUppercase(c);
821 break;
822 }
823 case 2:
824 {
825 string[i]=(char) LocaleLowercase(c);
826 break;
827 }
828 default:
829 break;
830 }
831 }
832
Tokenizer(TokenInfo * token_info,const unsigned flag,char * token,const size_t max_token_length,const char * line,const char * white,const char * break_set,const char * quote,const char escape,char * breaker,int * next,char * quoted)833 MagickExport int Tokenizer(TokenInfo *token_info,const unsigned flag,
834 char *token,const size_t max_token_length,const char *line,const char *white,
835 const char *break_set,const char *quote,const char escape,char *breaker,
836 int *next,char *quoted)
837 {
838 int
839 c;
840
841 ssize_t
842 i;
843
844 *breaker='\0';
845 *quoted='\0';
846 if (line[*next] == '\0')
847 return(1);
848 token_info->state=IN_WHITE;
849 token_info->quote=(char) MagickFalse;
850 token_info->flag=flag;
851 for (token_info->offset=0; (int) line[*next] != 0; (*next)++)
852 {
853 c=(int) line[*next];
854 i=sindex(c,break_set);
855 if (i >= 0)
856 {
857 switch (token_info->state)
858 {
859 case IN_WHITE:
860 case IN_TOKEN:
861 case IN_OZONE:
862 {
863 (*next)++;
864 *breaker=break_set[i];
865 token[token_info->offset]='\0';
866 return(0);
867 }
868 case IN_QUOTE:
869 {
870 StoreToken(token_info,token,max_token_length,c);
871 break;
872 }
873 }
874 continue;
875 }
876 i=sindex(c,quote);
877 if (i >= 0)
878 {
879 switch (token_info->state)
880 {
881 case IN_WHITE:
882 {
883 token_info->state=IN_QUOTE;
884 token_info->quote=quote[i];
885 *quoted=(char) MagickTrue;
886 break;
887 }
888 case IN_QUOTE:
889 {
890 if (quote[i] != token_info->quote)
891 StoreToken(token_info,token,max_token_length,c);
892 else
893 {
894 token_info->state=IN_OZONE;
895 token_info->quote='\0';
896 }
897 break;
898 }
899 case IN_TOKEN:
900 case IN_OZONE:
901 {
902 *breaker=(char) c;
903 token[token_info->offset]='\0';
904 return(0);
905 }
906 }
907 continue;
908 }
909 i=sindex(c,white);
910 if (i >= 0)
911 {
912 switch (token_info->state)
913 {
914 case IN_WHITE:
915 case IN_OZONE:
916 break;
917 case IN_TOKEN:
918 {
919 token_info->state=IN_OZONE;
920 break;
921 }
922 case IN_QUOTE:
923 {
924 StoreToken(token_info,token,max_token_length,c);
925 break;
926 }
927 }
928 continue;
929 }
930 if (c == (int) escape)
931 {
932 if (line[(*next)+1] == '\0')
933 {
934 *breaker='\0';
935 StoreToken(token_info,token,max_token_length,c);
936 (*next)++;
937 token[token_info->offset]='\0';
938 return(0);
939 }
940 switch (token_info->state)
941 {
942 case IN_WHITE:
943 {
944 (*next)--;
945 token_info->state=IN_TOKEN;
946 break;
947 }
948 case IN_TOKEN:
949 case IN_QUOTE:
950 {
951 (*next)++;
952 c=(int) line[*next];
953 StoreToken(token_info,token,max_token_length,c);
954 break;
955 }
956 case IN_OZONE:
957 {
958 token[token_info->offset]='\0';
959 return(0);
960 }
961 }
962 continue;
963 }
964 switch (token_info->state)
965 {
966 case IN_WHITE:
967 {
968 token_info->state=IN_TOKEN;
969 StoreToken(token_info,token,max_token_length,c);
970 break;
971 }
972 case IN_TOKEN:
973 case IN_QUOTE:
974 {
975 StoreToken(token_info,token,max_token_length,c);
976 break;
977 }
978 case IN_OZONE:
979 {
980 token[token_info->offset]='\0';
981 return(0);
982 }
983 }
984 }
985 token[token_info->offset]='\0';
986 return(0);
987 }
988