xref: /PHP-5.3/ext/standard/iptc.c (revision d4b67896)
1 /*
2    +----------------------------------------------------------------------+
3    | PHP Version 5                                                        |
4    +----------------------------------------------------------------------+
5    | Copyright (c) 1997-2013 The PHP Group                                |
6    +----------------------------------------------------------------------+
7    | This source file is subject to version 3.01 of the PHP license,      |
8    | that is bundled with this package in the file LICENSE, and is        |
9    | available through the world-wide-web at the following url:           |
10    | http://www.php.net/license/3_01.txt                                  |
11    | If you did not receive a copy of the PHP license and are unable to   |
12    | obtain it through the world-wide-web, please send a note to          |
13    | license@php.net so we can mail you a copy immediately.               |
14    +----------------------------------------------------------------------+
15    | Author: Thies C. Arntzen <thies@thieso.net>                          |
16    +----------------------------------------------------------------------+
17  */
18 
19 /* $Id$ */
20 
21 /*
22  * Functions to parse & compse IPTC data.
23  * PhotoShop >= 3.0 can read and write textual data to JPEG files.
24  * ... more to come .....
25  *
26  * i know, parts of this is now duplicated in image.c
27  * but in this case i think it's okay!
28  */
29 
30 /*
31  * TODO:
32  *  - add IPTC translation table
33  */
34 
35 #include "php.h"
36 #include "php_iptc.h"
37 #include "ext/standard/head.h"
38 
39 #include <sys/stat.h>
40 
41 
42 /* some defines for the different JPEG block types */
43 #define M_SOF0  0xC0            /* Start Of Frame N */
44 #define M_SOF1  0xC1            /* N indicates which compression process */
45 #define M_SOF2  0xC2            /* Only SOF0-SOF2 are now in common use */
46 #define M_SOF3  0xC3
47 #define M_SOF5  0xC5            /* NB: codes C4 and CC are NOT SOF markers */
48 #define M_SOF6  0xC6
49 #define M_SOF7  0xC7
50 #define M_SOF9  0xC9
51 #define M_SOF10 0xCA
52 #define M_SOF11 0xCB
53 #define M_SOF13 0xCD
54 #define M_SOF14 0xCE
55 #define M_SOF15 0xCF
56 #define M_SOI   0xD8
57 #define M_EOI   0xD9            /* End Of Image (end of datastream) */
58 #define M_SOS   0xDA            /* Start Of Scan (begins compressed data) */
59 #define M_APP0  0xe0
60 #define M_APP1  0xe1
61 #define M_APP2  0xe2
62 #define M_APP3  0xe3
63 #define M_APP4  0xe4
64 #define M_APP5  0xe5
65 #define M_APP6  0xe6
66 #define M_APP7  0xe7
67 #define M_APP8  0xe8
68 #define M_APP9  0xe9
69 #define M_APP10 0xea
70 #define M_APP11 0xeb
71 #define M_APP12 0xec
72 #define M_APP13 0xed
73 #define M_APP14 0xee
74 #define M_APP15 0xef
75 
76 /* {{{ php_iptc_put1
77  */
php_iptc_put1(FILE * fp,int spool,unsigned char c,unsigned char ** spoolbuf TSRMLS_DC)78 static int php_iptc_put1(FILE *fp, int spool, unsigned char c, unsigned char **spoolbuf TSRMLS_DC)
79 {
80 	if (spool > 0)
81 		PUTC(c);
82 
83 	if (spoolbuf) *(*spoolbuf)++ = c;
84 
85   	return c;
86 }
87 /* }}} */
88 
89 /* {{{ php_iptc_get1
90  */
php_iptc_get1(FILE * fp,int spool,unsigned char ** spoolbuf TSRMLS_DC)91 static int php_iptc_get1(FILE *fp, int spool, unsigned char **spoolbuf TSRMLS_DC)
92 {
93 	int c;
94 	char cc;
95 
96 	c = getc(fp);
97 
98 	if (c == EOF) return EOF;
99 
100 	if (spool > 0) {
101 		cc = c;
102 		PUTC(cc);
103 	}
104 
105 	if (spoolbuf) *(*spoolbuf)++ = c;
106 
107 	return c;
108 }
109 /* }}} */
110 
111 /* {{{ php_iptc_read_remaining
112  */
php_iptc_read_remaining(FILE * fp,int spool,unsigned char ** spoolbuf TSRMLS_DC)113 static int php_iptc_read_remaining(FILE *fp, int spool, unsigned char **spoolbuf TSRMLS_DC)
114 {
115   	while (php_iptc_get1(fp, spool, spoolbuf TSRMLS_CC) != EOF) continue;
116 
117 	return M_EOI;
118 }
119 /* }}} */
120 
121 /* {{{ php_iptc_skip_variable
122  */
php_iptc_skip_variable(FILE * fp,int spool,unsigned char ** spoolbuf TSRMLS_DC)123 static int php_iptc_skip_variable(FILE *fp, int spool, unsigned char **spoolbuf TSRMLS_DC)
124 {
125 	unsigned int  length;
126 	int c1, c2;
127 
128     if ((c1 = php_iptc_get1(fp, spool, spoolbuf TSRMLS_CC)) == EOF) return M_EOI;
129 
130     if ((c2 = php_iptc_get1(fp, spool, spoolbuf TSRMLS_CC)) == EOF) return M_EOI;
131 
132 	length = (((unsigned char) c1) << 8) + ((unsigned char) c2);
133 
134 	length -= 2;
135 
136 	while (length--)
137 		if (php_iptc_get1(fp, spool, spoolbuf TSRMLS_CC) == EOF) return M_EOI;
138 
139 	return 0;
140 }
141 /* }}} */
142 
143 /* {{{ php_iptc_next_marker
144  */
php_iptc_next_marker(FILE * fp,int spool,unsigned char ** spoolbuf TSRMLS_DC)145 static int php_iptc_next_marker(FILE *fp, int spool, unsigned char **spoolbuf TSRMLS_DC)
146 {
147     int c;
148 
149     /* skip unimportant stuff */
150 
151     c = php_iptc_get1(fp, spool, spoolbuf TSRMLS_CC);
152 
153 	if (c == EOF) return M_EOI;
154 
155     while (c != 0xff) {
156         if ((c = php_iptc_get1(fp, spool, spoolbuf TSRMLS_CC)) == EOF)
157             return M_EOI; /* we hit EOF */
158     }
159 
160     /* get marker byte, swallowing possible padding */
161     do {
162         c = php_iptc_get1(fp, 0, 0 TSRMLS_CC);
163 		if (c == EOF)
164             return M_EOI;       /* we hit EOF */
165 		else
166 		if (c == 0xff)
167 			php_iptc_put1(fp, spool, (unsigned char)c, spoolbuf TSRMLS_CC);
168     } while (c == 0xff);
169 
170     return (unsigned int) c;
171 }
172 /* }}} */
173 
174 static char psheader[] = "\xFF\xED\0\0Photoshop 3.0\08BIM\x04\x04\0\0\0\0";
175 
176 /* {{{ proto array iptcembed(string iptcdata, string jpeg_file_name [, int spool])
177    Embed binary IPTC data into a JPEG image. */
PHP_FUNCTION(iptcembed)178 PHP_FUNCTION(iptcembed)
179 {
180 	char *iptcdata, *jpeg_file;
181 	int iptcdata_len, jpeg_file_len;
182 	long spool = 0;
183 	FILE *fp;
184 	unsigned int marker, done = 0, inx;
185 	unsigned char *spoolbuf = NULL, *poi = NULL;
186 	struct stat sb;
187 	zend_bool written = 0;
188 
189 	if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "ss|l", &iptcdata, &iptcdata_len, &jpeg_file, &jpeg_file_len, &spool) != SUCCESS) {
190 		return;
191 	}
192 
193 	if (strlen(jpeg_file) != jpeg_file_len) {
194 		RETURN_FALSE;
195 	}
196 
197 	if (PG(safe_mode) && (!php_checkuid(jpeg_file, NULL, CHECKUID_CHECK_FILE_AND_DIR))) {
198 		RETURN_FALSE;
199 	}
200 
201 	if (php_check_open_basedir(jpeg_file TSRMLS_CC)) {
202 		RETURN_FALSE;
203 	}
204 
205 	if ((fp = VCWD_FOPEN(jpeg_file, "rb")) == 0) {
206 		php_error_docref(NULL TSRMLS_CC, E_WARNING, "Unable to open %s", jpeg_file);
207 		RETURN_FALSE;
208 	}
209 
210 	if (spool < 2) {
211 		fstat(fileno(fp), &sb);
212 
213 		poi = spoolbuf = safe_emalloc(1, iptcdata_len + sizeof(psheader) + sb.st_size + 1024, 1);
214 		memset(poi, 0, iptcdata_len + sizeof(psheader) + sb.st_size + 1024 + 1);
215 	}
216 
217 	if (php_iptc_get1(fp, spool, poi?&poi:0 TSRMLS_CC) != 0xFF) {
218 		fclose(fp);
219 		if (spoolbuf) {
220 			efree(spoolbuf);
221 		}
222 		RETURN_FALSE;
223 	}
224 
225 	if (php_iptc_get1(fp, spool, poi?&poi:0 TSRMLS_CC) != 0xD8) {
226 		fclose(fp);
227 		if (spoolbuf) {
228 			efree(spoolbuf);
229 		}
230 		RETURN_FALSE;
231 	}
232 
233 	while (!done) {
234 		marker = php_iptc_next_marker(fp, spool, poi?&poi:0 TSRMLS_CC);
235 
236 		if (marker == M_EOI) { /* EOF */
237 			break;
238 		} else if (marker != M_APP13) {
239 			php_iptc_put1(fp, spool, (unsigned char)marker, poi?&poi:0 TSRMLS_CC);
240 		}
241 
242 		switch (marker) {
243 			case M_APP13:
244 				/* we are going to write a new APP13 marker, so don't output the old one */
245 				php_iptc_skip_variable(fp, 0, 0 TSRMLS_CC);
246 				php_iptc_read_remaining(fp, spool, poi?&poi:0 TSRMLS_CC);
247 				done = 1;
248 				break;
249 
250 			case M_APP0:
251 				/* APP0 is in each and every JPEG, so when we hit APP0 we insert our new APP13! */
252 			case M_APP1:
253 				if (written) {
254 					/* don't try to write the data twice */
255 					break;
256 				}
257 				written = 1;
258 
259 				php_iptc_skip_variable(fp, spool, poi?&poi:0 TSRMLS_CC);
260 
261 				if (iptcdata_len & 1) {
262 					iptcdata_len++; /* make the length even */
263 				}
264 
265 				psheader[ 2 ] = (iptcdata_len+28)>>8;
266 				psheader[ 3 ] = (iptcdata_len+28)&0xff;
267 
268 				for (inx = 0; inx < 28; inx++) {
269 					php_iptc_put1(fp, spool, psheader[inx], poi?&poi:0 TSRMLS_CC);
270 				}
271 
272 				php_iptc_put1(fp, spool, (unsigned char)(iptcdata_len>>8), poi?&poi:0 TSRMLS_CC);
273 				php_iptc_put1(fp, spool, (unsigned char)(iptcdata_len&0xff), poi?&poi:0 TSRMLS_CC);
274 
275 				for (inx = 0; inx < iptcdata_len; inx++) {
276 					php_iptc_put1(fp, spool, iptcdata[inx], poi?&poi:0 TSRMLS_CC);
277 				}
278 				break;
279 
280 			case M_SOS:
281 				/* we hit data, no more marker-inserting can be done! */
282 				php_iptc_read_remaining(fp, spool, poi?&poi:0 TSRMLS_CC);
283 				done = 1;
284 				break;
285 
286 			default:
287 				php_iptc_skip_variable(fp, spool, poi?&poi:0 TSRMLS_CC);
288 				break;
289 		}
290 	}
291 
292 	fclose(fp);
293 
294 	if (spool < 2) {
295 		RETVAL_STRINGL(spoolbuf, poi - spoolbuf, 0);
296 	} else {
297 		RETURN_TRUE;
298 	}
299 }
300 /* }}} */
301 
302 /* {{{ proto array iptcparse(string iptcdata)
303    Parse binary IPTC-data into associative array */
PHP_FUNCTION(iptcparse)304 PHP_FUNCTION(iptcparse)
305 {
306 	unsigned int inx = 0, len, tagsfound = 0;
307 	unsigned char *buffer, recnum, dataset, key[ 16 ];
308 	char *str;
309 	int str_len;
310 	zval *values, **element;
311 
312 	if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "s", &str, &str_len) != SUCCESS) {
313 		return;
314 	}
315 
316 	buffer = (unsigned char *)str;
317 
318 	while (inx < str_len) { /* find 1st tag */
319 		if ((buffer[inx] == 0x1c) && ((buffer[inx+1] == 0x01) || (buffer[inx+1] == 0x02))){
320 			break;
321 		} else {
322 			inx++;
323 		}
324 	}
325 
326 	while (inx < str_len) {
327 		if (buffer[ inx++ ] != 0x1c) {
328 			break;   /* we ran against some data which does not conform to IPTC - stop parsing! */
329 		}
330 
331 		if ((inx + 4) >= str_len)
332 			break;
333 
334 		dataset = buffer[ inx++ ];
335 		recnum = buffer[ inx++ ];
336 
337 		if (buffer[ inx ] & (unsigned char) 0x80) { /* long tag */
338             if((inx+6) >= str_len) {
339                 break;
340             }
341 			len = (((long) buffer[ inx + 2 ]) << 24) + (((long) buffer[ inx + 3 ]) << 16) +
342 				  (((long) buffer[ inx + 4 ]) <<  8) + (((long) buffer[ inx + 5 ]));
343 			inx += 6;
344 		} else { /* short tag */
345 			len = (((unsigned short) buffer[ inx ])<<8) | (unsigned short)buffer[ inx+1 ];
346 			inx += 2;
347 		}
348 
349 		if ((len < 0) || (len > str_len) || (inx + len) > str_len) {
350 			break;
351 		}
352 
353 		snprintf(key, sizeof(key), "%d#%03d", (unsigned int) dataset, (unsigned int) recnum);
354 
355 		if (tagsfound == 0) { /* found the 1st tag - initialize the return array */
356 			array_init(return_value);
357 		}
358 
359 		if (zend_hash_find(Z_ARRVAL_P(return_value), key, strlen(key) + 1, (void **) &element) == FAILURE) {
360 			MAKE_STD_ZVAL(values);
361 			array_init(values);
362 
363 			zend_hash_update(Z_ARRVAL_P(return_value), key, strlen(key) + 1, (void *) &values, sizeof(zval*), (void **) &element);
364 		}
365 
366 		add_next_index_stringl(*element, buffer+inx, len, 1);
367 		inx += len;
368 		tagsfound++;
369 	}
370 
371 	if (! tagsfound) {
372 		RETURN_FALSE;
373 	}
374 }
375 /* }}} */
376 
377 /*
378  * Local variables:
379  * tab-width: 4
380  * c-basic-offset: 4
381  * End:
382  * vim600: sw=4 ts=4 fdm=marker
383  * vim<600: sw=4 ts=4
384  */
385