1 /* xgettext Smalltalk backend.
2 Copyright (C) 2002-2003, 2005 Free Software Foundation, Inc.
3
4 This file was written by Bruno Haible <haible@clisp.cons.org>, 2002.
5
6 This program is free software; you can redistribute it and/or modify
7 it under the terms of the GNU General Public License as published by
8 the Free Software Foundation; either version 2, or (at your option)
9 any later version.
10
11 This program is distributed in the hope that it will be useful,
12 but WITHOUT ANY WARRANTY; without even the implied warranty of
13 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 GNU General Public License for more details.
15
16 You should have received a copy of the GNU General Public License
17 along with this program; if not, write to the Free Software Foundation,
18 Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */
19
20 #ifdef HAVE_CONFIG_H
21 # include "config.h"
22 #endif
23
24 #include <errno.h>
25 #include <stdio.h>
26 #include <stdlib.h>
27
28 #include "message.h"
29 #include "xgettext.h"
30 #include "x-smalltalk.h"
31 #include "error.h"
32 #include "xalloc.h"
33 #include "exit.h"
34 #include "gettext.h"
35
36 #define _(s) gettext(s)
37
38 #define SIZEOF(a) (sizeof(a) / sizeof(a[0]))
39
40
41 /* The relevant parts of the Smalltalk syntax are:
42
43 stringliteral ::= string | stringconst | symconst
44 stringconst ::= "#"string
45 string ::= "'"[char]*"'"
46 symconst ::= "#"symbol
47 symbol ::= id | binsel | keysel[keysel]*
48 keysel ::= id":"
49 id ::= letter[letter|digit]*
50 letter ::= "A".."Z" | "a".."z"
51 digit ::= "0".."9"
52 binsel ::= selchar[selchar]
53 selchar ::= "+" | "-" | "*" | "/" | "~" | "|" | "," | "<" | ">"
54 | "=" | "&" | "@" | "?" | "%" | "\"
55
56 Strings can contain any characters; to include the string delimiter itself,
57 it must be duplicated.
58
59 Character constants are written "$"char
60
61 Comments are enclosed within double quotes.
62
63 In well-formed expressions, {} and [] and () are balanced.
64 */
65
66
67 /* ======================== Reading of characters. ======================== */
68
69
70 /* Real filename, used in error messages about the input file. */
71 static const char *real_file_name;
72
73 /* Logical filename and line number, used to label the extracted messages. */
74 static char *logical_file_name;
75 static int line_number;
76
77 /* The input file stream. */
78 static FILE *fp;
79
80
81 /* 1. line_number handling. */
82
83 static int
phase1_getc()84 phase1_getc ()
85 {
86 int c = getc (fp);
87
88 if (c == EOF)
89 {
90 if (ferror (fp))
91 error (EXIT_FAILURE, errno, _("error while reading \"%s\""),
92 real_file_name);
93 return EOF;
94 }
95
96 if (c == '\n')
97 line_number++;
98
99 return c;
100 }
101
102 /* Supports only one pushback character. */
103 static void
phase1_ungetc(int c)104 phase1_ungetc (int c)
105 {
106 if (c != EOF)
107 {
108 if (c == '\n')
109 --line_number;
110
111 ungetc (c, fp);
112 }
113 }
114
115
116 /* Accumulating comments. */
117
118 static char *buffer;
119 static size_t bufmax;
120 static size_t buflen;
121
122 static inline void
comment_start()123 comment_start ()
124 {
125 buflen = 0;
126 }
127
128 static inline void
comment_add(int c)129 comment_add (int c)
130 {
131 if (buflen >= bufmax)
132 {
133 bufmax = 2 * bufmax + 10;
134 buffer = xrealloc (buffer, bufmax);
135 }
136 buffer[buflen++] = c;
137 }
138
139 static inline void
comment_line_end()140 comment_line_end ()
141 {
142 while (buflen >= 1
143 && (buffer[buflen - 1] == ' ' || buffer[buflen - 1] == '\t'))
144 --buflen;
145 if (buflen >= bufmax)
146 {
147 bufmax = 2 * bufmax + 10;
148 buffer = xrealloc (buffer, bufmax);
149 }
150 buffer[buflen] = '\0';
151 savable_comment_add (buffer);
152 }
153
154
155 /* These are for tracking whether comments count as immediately before
156 keyword. */
157 static int last_comment_line;
158 static int last_non_comment_line;
159
160
161 /* ========================== Reading of tokens. ========================== */
162
163
164 enum token_type_ty
165 {
166 token_type_eof,
167 token_type_uniq, /* # */
168 token_type_symbol, /* symbol */
169 token_type_string_literal, /* string, stringconst, symbolconst */
170 token_type_other /* misc. operator */
171 };
172 typedef enum token_type_ty token_type_ty;
173
174 typedef struct token_ty token_ty;
175 struct token_ty
176 {
177 token_type_ty type;
178 char *string; /* for token_type_string_literal, token_type_symbol */
179 int line_number;
180 };
181
182
183 /* 2. Combine characters into tokens. Discard comments and whitespace. */
184
185 static token_ty phase2_pushback[1];
186 static int phase2_pushback_length;
187
188 static void
phase2_get(token_ty * tp)189 phase2_get (token_ty *tp)
190 {
191 static char *buffer;
192 static int bufmax;
193 int bufpos;
194 int c;
195
196 if (phase2_pushback_length)
197 {
198 *tp = phase2_pushback[--phase2_pushback_length];
199 return;
200 }
201
202 tp->string = NULL;
203
204 for (;;)
205 {
206 tp->line_number = line_number;
207 c = phase1_getc ();
208 switch (c)
209 {
210 case EOF:
211 tp->type = token_type_eof;
212 return;
213
214 case '"':
215 {
216 /* Comment. */
217 int lineno;
218
219 comment_start ();
220 lineno = line_number;
221 for (;;)
222 {
223 c = phase1_getc ();
224 if (c == '"' || c == EOF)
225 break;
226 if (c == '\n')
227 {
228 comment_line_end ();
229 comment_start ();
230 }
231 else
232 {
233 /* We skip all leading white space, but not EOLs. */
234 if (!(buflen == 0 && (c == ' ' || c == '\t')))
235 comment_add (c);
236 }
237 }
238 comment_line_end ();
239 last_comment_line = lineno;
240 continue;
241 }
242
243 case '\n':
244 if (last_non_comment_line > last_comment_line)
245 savable_comment_reset ();
246 /* FALLTHROUGH */
247 case ' ':
248 case '\t':
249 case '\r':
250 /* Ignore whitespace. */
251 continue;
252 }
253
254 last_non_comment_line = tp->line_number;
255
256 switch (c)
257 {
258 case '\'':
259 /* String literal. */
260 bufpos = 0;
261 for (;;)
262 {
263 c = phase1_getc ();
264 if (c == EOF)
265 break;
266 if (c == '\'')
267 {
268 c = phase1_getc ();
269 if (c != '\'')
270 {
271 phase1_ungetc (c);
272 break;
273 }
274 }
275 if (bufpos >= bufmax)
276 {
277 bufmax = 2 * bufmax + 10;
278 buffer = xrealloc (buffer, bufmax);
279 }
280 buffer[bufpos++] = c;
281 }
282 if (bufpos >= bufmax)
283 {
284 bufmax = 2 * bufmax + 10;
285 buffer = xrealloc (buffer, bufmax);
286 }
287 buffer[bufpos] = 0;
288 tp->type = token_type_string_literal;
289 tp->string = xstrdup (buffer);
290 return;
291
292 case '+':
293 case '-':
294 case '*':
295 case '/':
296 case '~':
297 case '|':
298 case ',':
299 case '<':
300 case '>':
301 case '=':
302 case '&':
303 case '@':
304 case '?':
305 case '%':
306 case '\\':
307 {
308 char *name;
309 int c2 = phase1_getc ();
310 switch (c2)
311 {
312 case '+':
313 case '-':
314 case '*':
315 case '/':
316 case '~':
317 case '|':
318 case ',':
319 case '<':
320 case '>':
321 case '=':
322 case '&':
323 case '@':
324 case '?':
325 case '%':
326 name = xmalloc (3);
327 name[0] = c;
328 name[1] = c2;
329 name[2] = '\0';
330 tp->type = token_type_symbol;
331 tp->string = name;
332 return;
333 default:
334 phase1_ungetc (c2);
335 break;
336 }
337 name = xmalloc (2);
338 name[0] = c;
339 name[1] = '\0';
340 tp->type = token_type_symbol;
341 tp->string = name;
342 return;
343 }
344
345 case 'A': case 'B': case 'C': case 'D': case 'E': case 'F':
346 case 'G': case 'H': case 'I': case 'J': case 'K': case 'L':
347 case 'M': case 'N': case 'O': case 'P': case 'Q': case 'R':
348 case 'S': case 'T': case 'U': case 'V': case 'W': case 'X':
349 case 'Y': case 'Z':
350 case 'a': case 'b': case 'c': case 'd': case 'e': case 'f':
351 case 'g': case 'h': case 'i': case 'j': case 'k': case 'l':
352 case 'm': case 'n': case 'o': case 'p': case 'q': case 'r':
353 case 's': case 't': case 'u': case 'v': case 'w': case 'x':
354 case 'y': case 'z':
355 /* Recognize id or id":"[id":"]* or id":"[id":"]*id. */
356 bufpos = 0;
357 for (;;)
358 {
359 if (bufpos >= bufmax)
360 {
361 bufmax = 2 * bufmax + 10;
362 buffer = xrealloc (buffer, bufmax);
363 }
364 buffer[bufpos++] = c;
365 c = phase1_getc ();
366 switch (c)
367 {
368 case 'A': case 'B': case 'C': case 'D': case 'E': case 'F':
369 case 'G': case 'H': case 'I': case 'J': case 'K': case 'L':
370 case 'M': case 'N': case 'O': case 'P': case 'Q': case 'R':
371 case 'S': case 'T': case 'U': case 'V': case 'W': case 'X':
372 case 'Y': case 'Z':
373 case 'a': case 'b': case 'c': case 'd': case 'e': case 'f':
374 case 'g': case 'h': case 'i': case 'j': case 'k': case 'l':
375 case 'm': case 'n': case 'o': case 'p': case 'q': case 'r':
376 case 's': case 't': case 'u': case 'v': case 'w': case 'x':
377 case 'y': case 'z':
378 case '0': case '1': case '2': case '3': case '4':
379 case '5': case '6': case '7': case '8': case '9':
380 continue;
381 case ':':
382 if (bufpos >= bufmax)
383 {
384 bufmax = 2 * bufmax + 10;
385 buffer = xrealloc (buffer, bufmax);
386 }
387 buffer[bufpos++] = c;
388 c = phase1_getc ();
389 switch (c)
390 {
391 case 'A': case 'B': case 'C': case 'D': case 'E': case 'F':
392 case 'G': case 'H': case 'I': case 'J': case 'K': case 'L':
393 case 'M': case 'N': case 'O': case 'P': case 'Q': case 'R':
394 case 'S': case 'T': case 'U': case 'V': case 'W': case 'X':
395 case 'Y': case 'Z':
396 case 'a': case 'b': case 'c': case 'd': case 'e': case 'f':
397 case 'g': case 'h': case 'i': case 'j': case 'k': case 'l':
398 case 'm': case 'n': case 'o': case 'p': case 'q': case 'r':
399 case 's': case 't': case 'u': case 'v': case 'w': case 'x':
400 case 'y': case 'z':
401 continue;
402 default:
403 phase1_ungetc (c);
404 break;
405 }
406 break;
407 default:
408 phase1_ungetc (c);
409 break;
410 }
411 break;
412 }
413 if (bufpos >= bufmax)
414 {
415 bufmax = 2 * bufmax + 10;
416 buffer = xrealloc (buffer, bufmax);
417 }
418 buffer[bufpos] = '\0';
419 tp->string = xstrdup (buffer);
420 tp->type = token_type_symbol;
421 return;
422
423 case '#':
424 /* Uniquification operator. */
425 tp->type = token_type_uniq;
426 return;
427
428 case '$':
429 c = phase1_getc ();
430 tp->type = token_type_other;
431 return;
432
433 default:
434 tp->type = token_type_other;
435 return;
436 }
437 }
438 }
439
440 /* Supports only one pushback token. */
441 static void
phase2_unget(token_ty * tp)442 phase2_unget (token_ty *tp)
443 {
444 if (tp->type != token_type_eof)
445 {
446 if (phase2_pushback_length == SIZEOF (phase2_pushback))
447 abort ();
448 phase2_pushback[phase2_pushback_length++] = *tp;
449 }
450 }
451
452
453 /* 3. Combine "# string_literal" and "# symbol" to a single token. */
454
455 static void
x_smalltalk_lex(token_ty * tp)456 x_smalltalk_lex (token_ty *tp)
457 {
458 phase2_get (tp);
459 if (tp->type == token_type_uniq)
460 {
461 token_ty token2;
462
463 phase2_get (&token2);
464 if (token2.type == token_type_symbol
465 || token2.type == token_type_string_literal)
466 {
467 tp->type = token_type_string_literal;
468 tp->string = token2.string;
469 }
470 else
471 phase2_unget (&token2);
472 }
473 }
474
475
476 /* ========================= Extracting strings. ========================== */
477
478 /* The file is broken into tokens. Scan the token stream, looking for the
479 following patterns
480 NLS ? <string>
481 NLS at: <string>
482 NLS at: <string> plural: <string>
483 where <string> is one of
484 string_literal
485 # string_literal
486 # symbol
487 */
488
489 void
extract_smalltalk(FILE * f,const char * real_filename,const char * logical_filename,flag_context_list_table_ty * flag_table,msgdomain_list_ty * mdlp)490 extract_smalltalk (FILE *f,
491 const char *real_filename, const char *logical_filename,
492 flag_context_list_table_ty *flag_table,
493 msgdomain_list_ty *mdlp)
494 {
495 message_list_ty *mlp = mdlp->item[0]->messages;
496
497 fp = f;
498 real_file_name = real_filename;
499 logical_file_name = xstrdup (logical_filename);
500 line_number = 1;
501
502 last_comment_line = -1;
503 last_non_comment_line = -1;
504
505 /* Eat tokens until eof is seen. */
506 {
507 /* 0 when no "NLS" has been seen.
508 1 after "NLS".
509 2 after "NLS ?".
510 3 after "NLS at:".
511 4 after "NLS at: <string>".
512 5 after "NLS at: <string> plural:". */
513 int state;
514 /* Remember the message containing the msgid, for msgid_plural.
515 Non-NULL in states 4, 5. */
516 message_ty *plural_mp = NULL;
517
518 /* Start state is 0. */
519 state = 0;
520
521 for (;;)
522 {
523 token_ty token;
524
525 x_smalltalk_lex (&token);
526
527 switch (token.type)
528 {
529 case token_type_symbol:
530 state = (strcmp (token.string, "NLS") == 0 ? 1 :
531 strcmp (token.string, "?") == 0 && state == 1 ? 2 :
532 strcmp (token.string, "at:") == 0 && state == 1 ? 3 :
533 strcmp (token.string, "plural:") == 0 && state == 4 ? 5 :
534 0);
535 free (token.string);
536 break;
537
538 case token_type_string_literal:
539 if (state == 2)
540 {
541 lex_pos_ty pos;
542 pos.file_name = logical_file_name;
543 pos.line_number = token.line_number;
544 remember_a_message (mlp, NULL, token.string, null_context,
545 &pos, savable_comment);
546 state = 0;
547 break;
548 }
549 if (state == 3)
550 {
551 lex_pos_ty pos;
552 pos.file_name = logical_file_name;
553 pos.line_number = token.line_number;
554 plural_mp = remember_a_message (mlp, NULL, token.string,
555 null_context, &pos,
556 savable_comment);
557 state = 4;
558 break;
559 }
560 if (state == 5)
561 {
562 lex_pos_ty pos;
563 pos.file_name = logical_file_name;
564 pos.line_number = token.line_number;
565 remember_a_message_plural (plural_mp, token.string,
566 null_context, &pos,
567 savable_comment);
568 state = 0;
569 break;
570 }
571 state = 0;
572 free (token.string);
573 break;
574
575 case token_type_uniq:
576 case token_type_other:
577 state = 0;
578 break;
579
580 case token_type_eof:
581 break;
582
583 default:
584 abort ();
585 }
586
587 if (token.type == token_type_eof)
588 break;
589 }
590 }
591
592 /* Close scanner. */
593 fp = NULL;
594 real_file_name = NULL;
595 logical_file_name = NULL;
596 line_number = 0;
597 }
598