summaryrefslogtreecommitdiff
path: root/content/fetchcache.c
blob: 4714d8002459d88d1347f54917c49344d78a766a (plain)
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
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
/*
 * This file is part of NetSurf, http://netsurf.sourceforge.net/
 * Licensed under the GNU General Public License,
 *                http://www.opensource.org/licenses/gpl-license
 * Copyright 2004 James Bursa <bursa@users.sourceforge.net>
 */

/** \file
 * High-level fetching, caching and conversion (implementation).
 *
 * The implementation checks the cache for the requested URL. If it is not
 * present, a content is created and a fetch is initiated. As the status of the
 * fetch changes and data is received, the content is updated appropriately.
 */

#define _GNU_SOURCE  /* for strndup */
#include <assert.h>
#include <string.h>
#include <sys/types.h>
#include <regex.h>
#include "netsurf/utils/config.h"
#include "netsurf/content/content.h"
#include "netsurf/content/fetchcache.h"
#include "netsurf/content/fetch.h"
#include "netsurf/content/url_store.h"
#include "netsurf/utils/log.h"
#include "netsurf/utils/messages.h"
#include "netsurf/utils/url.h"
#include "netsurf/utils/utils.h"


static char error_page[1000];
static regex_t re_content_type;
static void fetchcache_callback(fetch_msg msg, void *p, const char *data,
		unsigned long size);
static char *fetchcache_parse_type(const char *s, char **params[]);
static void fetchcache_error_page(struct content *c, const char *error);


/**
 * Retrieve a URL or prepare to fetch, convert, and cache it.
 *
 * The caller must supply a callback function which is called when anything
 * interesting happens to the content which is returned. See content.h.
 *
 * \param  url       address to fetch
 * \param  callback  function to call when anything interesting happens to
 *                   the new content
 * \param  p1 user   parameter for callback
 * \param  p2 user   parameter for callback
 * \param  width     available space
 * \param  height    available space
 * \param  no_error_pages if an error occurs, send CONTENT_MSG_ERROR instead
 *                   of generating an error page
 * \param  post_urlenc     url encoded post data, or 0 if none
 * \param  post_multipart  multipart post data, or 0 if none
 * \param  cookies   send and accept cookies
 * \param  download  download, rather than render content
 * \return  a new content, or 0 on memory exhaustion
 *
 * On success, call fetchcache_go() to start work on the new content.
 */

struct content * fetchcache(const char *url,
		void (*callback)(content_msg msg, struct content *c, void *p1,
			void *p2, union content_msg_data data),
		void *p1, void *p2,
		int width, int height,
		bool no_error_pages,
		char *post_urlenc,
		struct form_successful_control *post_multipart,
		bool cookies,
		bool download)
{
	struct content *c;
	char *url1;
	char *hash;
	
	if ((url1 = strdup(url)) == NULL)
		return NULL;

	/* strip fragment identifier */
	if ((hash = strchr(url1, '#')) != NULL)
		*hash = 0;

	LOG(("url %s", url1));

	if (!post_urlenc && !post_multipart && !download) {
		if ((c = content_get(url1)) != NULL) {
			free(url1);
			if (!content_add_user(c, callback, p1, p2))
				return NULL;
			else
				return c;
		}
	}

	c = content_create(url1);
	free(url1);
	if (!c)
		return NULL;
	if (!content_add_user(c, callback, p1, p2)) {
		return NULL;
	}

	if (!post_urlenc && !post_multipart && !download)
		c->fresh = true;

	c->width = width;
	c->height = height;
	c->no_error_pages = no_error_pages;
	c->download = download;

	return c;
}


/**
 * Start fetching and converting a content.
 *
 * \param  content   content to fetch, as returned by fetchcache()
 * \param  referer   referring URL, or 0
 * \param  callback  function to call when anything interesting happens to
 *                   the new content
 * \param  p1 user   parameter for callback
 * \param  p2 user   parameter for callback
 * \param  width     available space
 * \param  height    available space
 * \param  post_urlenc     url encoded post data, or 0 if none
 * \param  post_multipart  multipart post data, or 0 if none
 * \param  cookies   send and accept cookies
 *
 * Errors will be sent back through the callback.
 */

void fetchcache_go(struct content *content, char *referer,
		void (*callback)(content_msg msg, struct content *c, void *p1,
			void *p2, union content_msg_data data),
		void *p1, void *p2,
		int width, int height,
		char *post_urlenc,
		struct form_successful_control *post_multipart,
		bool cookies)
{
	char error_message[500];
	union content_msg_data msg_data;

	LOG(("url %s, status %s", content->url,
			content_status_name[content->status]));

	if (content->status == CONTENT_STATUS_TYPE_UNKNOWN && content->fetch) {
		/* fetching, but not yet received any response:
		 * no action required */

        } else if (content->status == CONTENT_STATUS_TYPE_UNKNOWN) {
        	/* brand new content: start fetch */
		content->fetch = fetch_start(content->url, referer,
				fetchcache_callback, content,
				content->no_error_pages,
				post_urlenc, post_multipart, cookies);
		if (!content->fetch) {
			LOG(("warning: fetch_start failed"));
			snprintf(error_message, sizeof error_message,
					messages_get("InvalidURL"),
					content->url);
			if (content->no_error_pages) {
				content->status = CONTENT_STATUS_ERROR;
				msg_data.error = error_message;
				content_broadcast(content, CONTENT_MSG_ERROR,
						msg_data);
			} else {
				fetchcache_error_page(content, error_message);
			}
		}

	/* in these remaining cases, we have to 'catch up' with the content's
	 * status, ie. send the same messages as if the content was
	 * gradually getting to the current status from TYPE_UNKNOWN */
	} else if (content->status == CONTENT_STATUS_LOADING) {
		callback(CONTENT_MSG_LOADING, content, p1, p2, msg_data);

	} else if (content->status == CONTENT_STATUS_READY) {
		callback(CONTENT_MSG_LOADING, content, p1, p2, msg_data);
		if (content_find_user(content, callback, p1, p2))
			callback(CONTENT_MSG_READY, content, p1, p2, msg_data);

	} else if (content->status == CONTENT_STATUS_DONE) {
		callback(CONTENT_MSG_LOADING, content, p1, p2, msg_data);
		if (content->available_width != width)
			content_reformat(content, width, height);
		if (content_find_user(content, callback, p1, p2))
			callback(CONTENT_MSG_READY, content, p1, p2, msg_data);
		if (content_find_user(content, callback, p1, p2))
			callback(CONTENT_MSG_DONE, content, p1, p2, msg_data);

	} else if (content->status == CONTENT_STATUS_ERROR) {
		/* shouldn't usually occur */
		msg_data.error = messages_get("MiscError");
		callback(CONTENT_MSG_ERROR, content, p1, p2, msg_data);

	}
}


/**
 * Callback function for fetch.
 *
 * This is called when the status of a fetch changes.
 */

void fetchcache_callback(fetch_msg msg, void *p, const char *data,
		unsigned long size)
{
	struct url_content *url_content;
	bool res;
	struct content *c = p;
	content_type type;
	char *mime_type, *url;
	char **params;
	unsigned int i;
	union content_msg_data msg_data;
	url_func_result result;

	switch (msg) {
		case FETCH_TYPE:
			c->total_size = size;
			mime_type = fetchcache_parse_type(data, &params);
			if (!mime_type) {
				msg_data.error = messages_get("NoMemory");
				content_broadcast(c, CONTENT_MSG_ERROR,
						msg_data);
				fetch_abort(c->fetch);
				c->fetch = 0;
				return;
			}
			type = content_lookup(mime_type);
			LOG(("FETCH_TYPE, type %u", c->download ?
						CONTENT_OTHER : type));
			res = content_set_type(c,
					c->download ? CONTENT_OTHER : type,
					mime_type, params);
			free(mime_type);
			for (i = 0; params[i]; i++)
				free(params[i]);
			free(params);
			if (!res) {
				fetch_abort(c->fetch);
				c->fetch = 0;
			}
			break;

		case FETCH_PROGRESS:
			if (size)
				content_set_status(c,
						messages_get("RecPercent"),
						data, (unsigned int)size);
			else
				content_set_status(c,
						messages_get("Received"),
						data);
			content_broadcast(c, CONTENT_MSG_STATUS, msg_data);
			break;

		case FETCH_DATA:
			LOG(("FETCH_DATA"));
/*			if (c->total_size)
				content_set_status(c,
						messages_get("RecPercent"),
						human_friendly_bytesize(c->source_size + size),
						human_friendly_bytesize(c->total_size),
						(unsigned int) ((c->source_size + size) * 100 / c->total_size));
			else
				content_set_status(c,
						messages_get("Received"),
						human_friendly_bytesize(c->source_size + size));
			content_broadcast(c, CONTENT_MSG_STATUS, msg_data);
*/			if (!content_process_data(c, data, size)) {
				fetch_abort(c->fetch);
				c->fetch = 0;
			}
			break;

		case FETCH_FINISHED:
			url_content = url_store_find(c->url);
			if (url_content)
				url_content->requests++;

			LOG(("FETCH_FINISHED"));
			c->fetch = 0;
			content_set_status(c, messages_get("Converting"),
					c->source_size);
			content_broadcast(c, CONTENT_MSG_STATUS, msg_data);
			content_convert(c, c->width, c->height);
			break;

		case FETCH_ERROR:
			LOG(("FETCH_ERROR, '%s'", data));
			c->fetch = 0;
			if (c->no_error_pages) {
				c->status = CONTENT_STATUS_ERROR;
				msg_data.error = data;
				content_broadcast(c, CONTENT_MSG_ERROR,
						msg_data);
			} else {
				content_reset(c);
				fetchcache_error_page(c, data);
			}
			break;

		case FETCH_REDIRECT:
			LOG(("FETCH_REDIRECT, '%s'", data));
			c->fetch = 0;
			/* redirect URLs must be absolute by HTTP/1.1, but many sites send
			 * relative ones: treat them as relative to requested URL */
			result = url_join(data, c->url, &url);
			if (result == URL_FUNC_OK) {
				msg_data.redirect = url;
				content_broadcast(c, CONTENT_MSG_REDIRECT, msg_data);
				free(url);
			} else {
				msg_data.error = messages_get("BadRedirect");
				content_broadcast(c, CONTENT_MSG_ERROR, msg_data);
			}
			/* set the status to ERROR so that the content is
			 * destroyed in content_clean() */
			c->status = CONTENT_STATUS_ERROR;
			break;
#ifdef WITH_AUTH
		case FETCH_AUTH:
			/* data -> string containing the Realm */
			LOG(("FETCH_AUTH, '%s'", data));
			c->fetch = 0;
			msg_data.auth_realm = data;
			content_broadcast(c, CONTENT_MSG_AUTH, msg_data);
			/* set the status to ERROR so that the content is
			 * destroyed in content_clean() */
			c->status = CONTENT_STATUS_ERROR;
			break;
#endif
		default:
			assert(0);
	}
}


/**
 * Initialise the fetchcache module.
 */

void fetchcache_init(void)
{
	regcomp_wrapper(&re_content_type,
			"^([-0-9a-zA-Z_.]+/[-0-9a-zA-Z_.]+)[ \t]*"
			"(;[ \t]*([-0-9a-zA-Z_.]+)="
			"([-0-9a-zA-Z_.]+|\"([^\"]|[\\].)*\")[ \t]*)*$",
			REG_EXTENDED);
}


/**
 * Parse a Content-Type header.
 *
 * \param  s       a Content-Type header
 * \param  params  updated to point to an array of strings, ordered attribute,
 *                 value, attribute, ..., 0
 * \return  a new string containing the MIME-type, or 0 on memory exhaustion
 */

#define MAX_ATTRS 10

char *fetchcache_parse_type(const char *s, char **params[])
{
	char *type = 0;
	unsigned int i;
	int r;
	regmatch_t pmatch[2 + MAX_ATTRS * 3];

	*params = malloc((MAX_ATTRS * 2 + 2) * sizeof (*params)[0]);
	if (!*params)
		goto no_memory;
	for (i = 0; i != MAX_ATTRS * 2 + 2; i++)
		(*params)[i] = 0;

	r = regexec(&re_content_type, s, 2 + MAX_ATTRS * 3, pmatch, 0);
	if (r) {
		LOG(("failed to parse content-type '%s'", s));
		type = strdup(s);
		if (!type)
			goto no_memory;
		return type;
	}

	type = strndup(s + pmatch[1].rm_so, pmatch[1].rm_eo - pmatch[1].rm_so);
	if (!type) {
		free(*params);
		return 0;
	}

	/* parameters */
	for (i = 0; i != MAX_ATTRS && pmatch[2 + 3 * i].rm_so != -1; i++) {
		(*params)[2 * i] = strndup(s + pmatch[2 + 3 * i + 1].rm_so,
				pmatch[2 + 3 * i + 1].rm_eo -
				pmatch[2 + 3 * i + 1].rm_so);
		(*params)[2 * i + 1] = strndup(s + pmatch[2 + 3 * i + 2].rm_so,
				pmatch[2 + 3 * i + 2].rm_eo -
				pmatch[2 + 3 * i + 2].rm_so);
		if (!(*params)[2 * i] || !(*params)[2 * i + 1])
			goto no_memory;
	}
	(*params)[2 * i] = 0;

	return type;

no_memory:
	for (i = 0; i != MAX_ATTRS * 2 + 2; i++)
		free((*params)[i]);
	free(*params);
	free(type);

	return 0;
}


/**
 * Generate an error page.
 *
 * \param c empty content to generate the page in
 * \param error message to display
 */

void fetchcache_error_page(struct content *c, const char *error)
{
	const char *params[] = { 0 };
	int length;

	if ((length = snprintf(error_page, sizeof(error_page),
			messages_get("ErrorPage"), error)) < 0)
		length = 0;
	if (!content_set_type(c, CONTENT_HTML, "text/html", params))
		return;
	if (!content_process_data(c, error_page, length))
		return;
	content_convert(c, c->width, c->height);
}


#ifdef TEST

#include <unistd.h>

void callback(fetchcache_msg msg, struct content *c, void *p, char *error)
{
	switch (msg) {
		case FETCHCACHE_OK:
			LOG(("FETCHCACHE_OK, url '%s'", p));
			break;
		case FETCHCACHE_BADTYPE:
			LOG(("FETCHCACHE_BADTYPE, url '%s'", p));
			break;
		case FETCHCACHE_ERROR:
			LOG(("FETCHCACHE_ERROR, url '%s', error '%s'", p, error));
			break;
		default:
			assert(0);
	}
}

char *test[] = {"http://www.google.co.uk/", "http://www.ox.ac.uk/", "blah://blah/"};

int main(void)
{
	int i;

	cache_init();
	fetch_init();

	for (i = 0; i != sizeof(test) / sizeof(test[0]); i++)
		fetchcache(test[i], 0, callback, test[i], 800, 0);
	for (i = 0; i != 5; i++) {
		fetch_poll();
		sleep(1);
	}
	for (i = 0; i != sizeof(test) / sizeof(test[0]); i++)
		fetchcache(test[i], 0, callback, test[i], 800, 0);
	for (i = 0; i != 20; i++) {
		fetch_poll();
		sleep(1);
	}
	return 0;
}

#endif