1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
|
/* Copyright (C) 2001-2019 Artifex Software, Inc.
All Rights Reserved.
This software is provided AS-IS with no warranty, either express or
implied.
This software is distributed under license and may not be copied,
modified or distributed except as expressly authorized under the terms
of the license contained in the file LICENSE in this distribution.
Refer to licensing information at http://www.artifex.com or contact
Artifex Software, Inc., 1305 Grant Avenue - Suite 200, Novato,
CA 94945, U.S.A., +1(415)492-9861, for further information.
*/
/* Command line argument list management */
#include "ctype_.h"
#include "stdio_.h"
#include "string_.h"
#include "gsexit.h"
#include "gsmemory.h"
#include "gsargs.h"
#include "gserrors.h"
#include "gp.h"
int codepoint_to_utf8(char *cstr, int rune)
{
int idx = 0;
if (rune < 0x80) {
cstr[idx++] = rune;
} else {
if (rune < 0x800) {
cstr[idx++] = 0xc0 | (rune>>6);
} else {
if (rune < 0x10000) {
cstr[idx++] = 0xe0 | (rune>>12);
} else {
if (rune < 0x200000) {
cstr[idx++] = 0xf0 | (rune>>18);
} else {
/* Shouldn't ever be required, but included for completeness */
if (rune < 0x4000000) {
cstr[idx++] = 0xf8 | (rune>>24);
} else {
cstr[idx++] = 0xfc | (rune>>30);
cstr[idx++] = 0x80 | ((rune>>24) & 0x3f);
}
cstr[idx++] = 0x80 | ((rune>>18) & 0x3f);
}
cstr[idx++] = 0x80 | ((rune>>12) & 0x3f);
}
cstr[idx++] = 0x80 | ((rune>>6) & 0x3f);
}
cstr[idx++] = 0x80 | (rune & 0x3f);
}
return idx;
}
static int get_codepoint_utf8(gp_file *file, const char **astr)
{
int c;
int rune;
int len;
/* This code spots the BOM for utf8 and ignores it. Strictly speaking
* this may be wrong, as we are only supposed to ignore it at the beginning
* of the string, but if anyone is stupid enough to use ZWNBSP (zero width
* non breaking space) in the middle of their strings, then they deserve
* what they get. */
do {
c = (file ? gp_fgetc(file) : (**astr ? (int)(unsigned char)*(*astr)++ : EOF));
if (c == EOF)
return EOF;
if (c < 0x80)
return c;
lead: /* We've just read a byte >= 0x80, presumably a leading byte */
if (c < 0xc0)
continue; /* Illegal - skip it */
else if (c < 0xe0)
len = 1, rune = c & 0x1f;
else if (c < 0xf0)
len = 2, rune = c & 0xf;
else if (c < 0xf8)
len = 3, rune = c & 7;
else if (c < 0xfc)
len = 4, rune = c & 3;
else if (c < 0xfe)
len = 5, rune = c & 1;
else
continue; /* Illegal - skip it */
do {
c = (file ? gp_fgetc(file) : (**astr ? (int)(unsigned char)*(*astr)++ : EOF));
if (c == EOF)
return EOF;
rune = (rune<<6) | (c & 0x3f);
} while (((c & 0xC0) == 0x80) && --len);
if (len) {
/* The rune we are collecting is improperly formed. */
if (c < 0x80) {
/* Just return the simple char we've ended on. */
return c;
}
/* Start collecting again */
goto lead;
}
if (rune == 0xFEFF)
continue; /* BOM. Skip it */
break;
} while (1);
return rune;
}
/* Initialize an arg list. */
int
arg_init(arg_list * pal,
const char **argv,
int argc,
gp_file *(*arg_fopen)(const char *fname, void *fopen_data),
void *fopen_data,
int (*get_codepoint)(gp_file *file, const char **astr),
gs_memory_t *memory)
{
int code;
const char *arg;
pal->expand_ats = true;
pal->arg_fopen = arg_fopen;
pal->fopen_data = fopen_data;
pal->get_codepoint = (get_codepoint ? get_codepoint : get_codepoint_utf8);
pal->memory = memory;
pal->argp = argv;
pal->argn = argc;
pal->depth = 0;
pal->sources[0].is_file = 0;
pal->sources[0].u.s.memory = NULL;
pal->sources[0].u.s.decoded = 0;
pal->sources[0].u.s.parsed = 0;
/* Stash the 0th one */
code = arg_next(pal, &arg, memory);
if (code < 0)
return code;
return gs_lib_ctx_stash_exe(memory->gs_lib_ctx, arg);
}
/* Push a string onto an arg list. */
int
arg_push_memory_string(arg_list * pal, char *str, bool parsed, gs_memory_t * mem)
{
return arg_push_decoded_memory_string(pal, str, parsed, parsed, mem);
}
int
arg_push_decoded_memory_string(arg_list * pal, char *str, bool parsed, bool decoded, gs_memory_t * mem)
{
arg_source *pas;
if (pal->depth+1 == arg_depth_max) {
lprintf("Too much nesting of @-files.\n");
return 1;
}
pas = &pal->sources[++pal->depth];
pas->is_file = false;
pas->u.s.parsed = parsed;
pas->u.s.decoded = decoded;
pas->u.s.chars = str;
pas->u.s.memory = mem;
pas->u.s.str = str;
return 0;
}
/* Clean up an arg list. */
void
arg_finit(arg_list * pal)
{
/* No cleanup is required for level 0 */
while (pal->depth) {
arg_source *pas = &pal->sources[pal->depth--];
if (pas->is_file)
gp_fclose(pas->u.file);
else if (pas->u.s.memory)
gs_free_object(pas->u.s.memory, pas->u.s.chars, "arg_finit");
}
}
static int get_codepoint(arg_list *pal, arg_source *pas)
{
int (*fn)(gp_file *file, const char **str);
fn = (!pas->is_file && pas->u.s.decoded ? get_codepoint_utf8 : pal->get_codepoint);
return fn(pas->is_file ? pas->u.file : NULL, &pas->u.s.str);
}
/* Get the next arg from a list. */
/* Note that these are not copied to the heap. */
/* returns:
* >0 - valid argument
* 0 - arguments exhausted
* <0 - error condition
* *argstr is *always* set: to the arg string if it is valid,
* or to NULL otherwise
*/
int
arg_next(arg_list * pal, const char **argstr, const gs_memory_t *errmem)
{
arg_source *pas;
char *cstr;
int c;
int i;
bool in_quote, eol;
*argstr = NULL;
/* Loop over arguments, finding one to return. */
do {
pas = &pal->sources[pal->depth];
if (!pas->is_file && pas->u.s.parsed) {
/* This string is a "pushed-back" argument (retrieved
* by a preceding arg_next(), but not processed). No
* decoding is required. */
/* assert(pas->u.s.decoded); */
if (strlen(pas->u.s.str) >= arg_str_max) {
errprintf(errmem, "Command too long: %s\n", pas->u.s.str);
return_error(gs_error_Fatal);
}
strcpy(pal->cstr, pas->u.s.str);
*argstr = pal->cstr;
if (pas->u.s.memory)
gs_free_object(pas->u.s.memory, pas->u.s.chars, "arg_next");
pal->depth--;
} else {
/* We need to decode the next argument */
if (pal->depth == 0) {
if (pal->argn <= 0)
return 0; /* all done */
/* Move onto the next argument from the string. */
pal->argn--;
pas->u.s.str = *(pal->argp++);
}
/* Skip a prefix of whitespace. */
do {
c = get_codepoint(pal, pas);
} while (c > 0 && c < 256 && isspace(c));
if (c == EOF) {
/* EOF before any argument characters. */
if (pas->is_file)
gp_fclose(pas->u.file);
else if (pas->u.s.memory)
gs_free_object(pas->u.s.memory, pas->u.s.chars,
"arg_next");
/* If depth is 0, then we are reading from the simple
* argument list and we just hit an "empty" argument
* (such as -o ""). Return this. */
if (pal->depth == 0)
{
*argstr = pal->cstr;
pal->cstr[0] = 0;
break;
}
/* If depth > 0, then we're reading from a response
* file, and we've hit the end of the response file.
* Pop up one level and continue. */
pal->depth--;
continue; /* Next argument */
}
#define is_eol(c) (c == '\r' || c == '\n')
/* Convert from astr into pal->cstr, and return it as *argstr. */
*argstr = cstr = pal->cstr;
in_quote = false;
/* We keep track of whether we have just read an "eol" or not,
* in order to skip # characters at the start of a line
* (possibly preceeded by whitespace). We do NOT want this to
* apply to the start of arguments in the arg list, so only
* set eol to be true, if we are in a file. */
eol = pal->depth > 0;
for (i = 0;;) {
if (c == EOF) {
if (in_quote) {
cstr[i] = 0;
errprintf(errmem,
"Unterminated quote in @-file: %s\n", cstr);
return_error(gs_error_Fatal);
}
break; /* End of arg */
}
/* c != 0 */
/* If we aren't parsing from the arglist (i.e. depth > 0)
* then we break on whitespace (unless we're in quotes). */
if (pal->depth > 0 && !in_quote && c > 0 && c < 256 && isspace(c))
break; /* End of arg */
/* c isn't leading or terminating whitespace. */
if (c == '#' && eol) {
/* Skip a comment. */
do {
c = get_codepoint(pal, pas);
} while (c != 0 && !is_eol(c) && c != EOF);
if (c == '\r')
c = get_codepoint(pal, pas);
if (c == '\n')
c = get_codepoint(pal, pas);
continue; /* Next char */
}
if (c == '\\') {
/* Check for \ followed by newline. */
c = get_codepoint(pal, pas);
if (is_eol(c)) {
if (c == '\r')
c = get_codepoint(pal, pas);
if (c == '\n')
c = get_codepoint(pal, pas);
eol = true;
continue; /* Next char */
}
/* \ anywhere else is treated as a printing character. */
/* This is different from the Unix shells. */
if (i >= arg_str_max - 1) {
cstr[i] = 0;
errprintf(errmem, "Command too long: %s\n", cstr);
return_error(gs_error_Fatal);
}
cstr[i++] = '\\';
eol = false;
continue; /* Next char */
}
/* c will become part of the argument */
if (i >= arg_str_max - 1) {
cstr[i] = 0;
errprintf(errmem, "Command too long: %s\n", cstr);
return_error(gs_error_Fatal);
}
/* Allow quotes to protect whitespace. */
/* (special cases have already been handled and don't reach this point) */
if (c == '"')
in_quote = !in_quote;
else
i += codepoint_to_utf8(&cstr[i], c);
eol = is_eol(c);
c = get_codepoint(pal, pas);
}
cstr[i] = 0;
}
/* At this point *argstr is full of utf8 encoded argument. */
/* If it's an @filename argument, then deal with it, and never return
* it to the caller. */
if (pal->expand_ats && **argstr == '@') {
char *fname;
gp_file *f;
if (pal->depth+1 == arg_depth_max) {
errprintf(errmem, "Too much nesting of @-files.\n");
return_error(gs_error_Fatal);
}
fname = (char *)*argstr + 1; /* skip @ */
if (gs_add_control_path(pal->memory, gs_permit_file_reading, fname) < 0)
return_error(gs_error_Fatal);
f = (*pal->arg_fopen) (fname, pal->fopen_data);
DISCARD(gs_remove_control_path(pal->memory, gs_permit_file_reading, fname));
if (f == NULL) {
errprintf(errmem, "Unable to open command line file %s\n", *argstr);
return_error(gs_error_Fatal);
}
pas = &pal->sources[++pal->depth];
pas->is_file = true;
pas->u.file = f;
*argstr = NULL; /* Empty the argument string so we don't return it. */
continue; /* Loop back to parse the first arg from the file. */
}
} while (*argstr == NULL || **argstr == 0); /* Until we get a non-empty arg */
return 1;
}
/* Copy an argument string to the heap. */
char *
arg_copy(const char *str, gs_memory_t * mem)
{
char *sstr = (char *)gs_alloc_bytes(mem, strlen(str) + 1, "arg_copy");
if (sstr == 0) {
lprintf("Out of memory!\n");
return NULL;
}
strcpy(sstr, str);
return sstr;
}
/* Free a previously arg_copy'd string */
void
arg_free(char *str, gs_memory_t * mem)
{
gs_free_object(mem, str, "arg_copy");
}
int arg_strcmp(arg_list *pal, const char *arg, const char *match)
{
int rune, c;
if (!arg || !match)
return 1;
do {
rune = pal->get_codepoint(NULL, &arg);
if (rune == -1)
rune = 0;
c = *match++;
if (rune != c)
return rune - c;
} while (rune && c);
return 0;
}
|