xref: /PHP-5.4/ext/standard/iptc.c (revision d4b67896)
1 /*
2    +----------------------------------------------------------------------+
3    | PHP Version 5                                                        |
4    +----------------------------------------------------------------------+
5    | Copyright (c) 1997-2014 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;
185 	int inx;
186 	unsigned char *spoolbuf = NULL, *poi = NULL;
187 	struct stat sb;
188 	zend_bool written = 0;
189 
190 	if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "sp|l", &iptcdata, &iptcdata_len, &jpeg_file, &jpeg_file_len, &spool) != SUCCESS) {
191 		return;
192 	}
193 
194 	if (php_check_open_basedir(jpeg_file TSRMLS_CC)) {
195 		RETURN_FALSE;
196 	}
197 
198 	if ((fp = VCWD_FOPEN(jpeg_file, "rb")) == 0) {
199 		php_error_docref(NULL TSRMLS_CC, E_WARNING, "Unable to open %s", jpeg_file);
200 		RETURN_FALSE;
201 	}
202 
203 	if (spool < 2) {
204 		fstat(fileno(fp), &sb);
205 
206 		poi = spoolbuf = safe_emalloc(1, iptcdata_len + sizeof(psheader) + sb.st_size + 1024, 1);
207 		memset(poi, 0, iptcdata_len + sizeof(psheader) + sb.st_size + 1024 + 1);
208 	}
209 
210 	if (php_iptc_get1(fp, spool, poi?&poi:0 TSRMLS_CC) != 0xFF) {
211 		fclose(fp);
212 		if (spoolbuf) {
213 			efree(spoolbuf);
214 		}
215 		RETURN_FALSE;
216 	}
217 
218 	if (php_iptc_get1(fp, spool, poi?&poi:0 TSRMLS_CC) != 0xD8) {
219 		fclose(fp);
220 		if (spoolbuf) {
221 			efree(spoolbuf);
222 		}
223 		RETURN_FALSE;
224 	}
225 
226 	while (!done) {
227 		marker = php_iptc_next_marker(fp, spool, poi?&poi:0 TSRMLS_CC);
228 
229 		if (marker == M_EOI) { /* EOF */
230 			break;
231 		} else if (marker != M_APP13) {
232 			php_iptc_put1(fp, spool, (unsigned char)marker, poi?&poi:0 TSRMLS_CC);
233 		}
234 
235 		switch (marker) {
236 			case M_APP13:
237 				/* we are going to write a new APP13 marker, so don't output the old one */
238 				php_iptc_skip_variable(fp, 0, 0 TSRMLS_CC);
239 				php_iptc_read_remaining(fp, spool, poi?&poi:0 TSRMLS_CC);
240 				done = 1;
241 				break;
242 
243 			case M_APP0:
244 				/* APP0 is in each and every JPEG, so when we hit APP0 we insert our new APP13! */
245 			case M_APP1:
246 				if (written) {
247 					/* don't try to write the data twice */
248 					break;
249 				}
250 				written = 1;
251 
252 				php_iptc_skip_variable(fp, spool, poi?&poi:0 TSRMLS_CC);
253 
254 				if (iptcdata_len & 1) {
255 					iptcdata_len++; /* make the length even */
256 				}
257 
258 				psheader[ 2 ] = (iptcdata_len+28)>>8;
259 				psheader[ 3 ] = (iptcdata_len+28)&0xff;
260 
261 				for (inx = 0; inx < 28; inx++) {
262 					php_iptc_put1(fp, spool, psheader[inx], poi?&poi:0 TSRMLS_CC);
263 				}
264 
265 				php_iptc_put1(fp, spool, (unsigned char)(iptcdata_len>>8), poi?&poi:0 TSRMLS_CC);
266 				php_iptc_put1(fp, spool, (unsigned char)(iptcdata_len&0xff), poi?&poi:0 TSRMLS_CC);
267 
268 				for (inx = 0; inx < iptcdata_len; inx++) {
269 					php_iptc_put1(fp, spool, iptcdata[inx], poi?&poi:0 TSRMLS_CC);
270 				}
271 				break;
272 
273 			case M_SOS:
274 				/* we hit data, no more marker-inserting can be done! */
275 				php_iptc_read_remaining(fp, spool, poi?&poi:0 TSRMLS_CC);
276 				done = 1;
277 				break;
278 
279 			default:
280 				php_iptc_skip_variable(fp, spool, poi?&poi:0 TSRMLS_CC);
281 				break;
282 		}
283 	}
284 
285 	fclose(fp);
286 
287 	if (spool < 2) {
288 		RETVAL_STRINGL(spoolbuf, poi - spoolbuf, 0);
289 	} else {
290 		RETURN_TRUE;
291 	}
292 }
293 /* }}} */
294 
295 /* {{{ proto array iptcparse(string iptcdata)
296    Parse binary IPTC-data into associative array */
PHP_FUNCTION(iptcparse)297 PHP_FUNCTION(iptcparse)
298 {
299 	int inx = 0, len;
300 	unsigned int tagsfound = 0;
301 	unsigned char *buffer, recnum, dataset, key[ 16 ];
302 	char *str;
303 	int str_len;
304 	zval *values, **element;
305 
306 	if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "s", &str, &str_len) != SUCCESS) {
307 		return;
308 	}
309 
310 	buffer = (unsigned char *)str;
311 
312 	while (inx < str_len) { /* find 1st tag */
313 		if ((buffer[inx] == 0x1c) && ((buffer[inx+1] == 0x01) || (buffer[inx+1] == 0x02))){
314 			break;
315 		} else {
316 			inx++;
317 		}
318 	}
319 
320 	while (inx < str_len) {
321 		if (buffer[ inx++ ] != 0x1c) {
322 			break;   /* we ran against some data which does not conform to IPTC - stop parsing! */
323 		}
324 
325 		if ((inx + 4) >= str_len)
326 			break;
327 
328 		dataset = buffer[ inx++ ];
329 		recnum = buffer[ inx++ ];
330 
331 		if (buffer[ inx ] & (unsigned char) 0x80) { /* long tag */
332 			if((inx+6) >= str_len) {
333 				break;
334 			}
335 			len = (((long) buffer[ inx + 2 ]) << 24) + (((long) buffer[ inx + 3 ]) << 16) +
336 				  (((long) buffer[ inx + 4 ]) <<  8) + (((long) buffer[ inx + 5 ]));
337 			inx += 6;
338 		} else { /* short tag */
339 			len = (((unsigned short) buffer[ inx ])<<8) | (unsigned short)buffer[ inx+1 ];
340 			inx += 2;
341 		}
342 
343 		if ((len < 0) || (len > str_len) || (inx + len) > str_len) {
344 			break;
345 		}
346 
347 		snprintf(key, sizeof(key), "%d#%03d", (unsigned int) dataset, (unsigned int) recnum);
348 
349 		if (tagsfound == 0) { /* found the 1st tag - initialize the return array */
350 			array_init(return_value);
351 		}
352 
353 		if (zend_hash_find(Z_ARRVAL_P(return_value), key, strlen(key) + 1, (void **) &element) == FAILURE) {
354 			MAKE_STD_ZVAL(values);
355 			array_init(values);
356 
357 			zend_hash_update(Z_ARRVAL_P(return_value), key, strlen(key) + 1, (void *) &values, sizeof(zval*), (void **) &element);
358 		}
359 
360 		add_next_index_stringl(*element, buffer+inx, len, 1);
361 		inx += len;
362 		tagsfound++;
363 	}
364 
365 	if (! tagsfound) {
366 		RETURN_FALSE;
367 	}
368 }
369 /* }}} */
370 
371 /*
372  * Local variables:
373  * tab-width: 4
374  * c-basic-offset: 4
375  * End:
376  * vim600: sw=4 ts=4 fdm=marker
377  * vim<600: sw=4 ts=4
378  */
379