summaryrefslogtreecommitdiff
path: root/content/fetchers/data.c
blob: 0d92ad2822a2b7238dc8bdbde999706809bc09be (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
/*
 * Copyright 2008 Rob Kendrick <rjek@netsurf-browser.org>
 *
 * This file is part of NetSurf.
 *
 * NetSurf is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; version 2 of the License.
 *
 * NetSurf is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program.  If not, see <http://www.gnu.org/licenses/>.
 */

/**
 * \file
 * data scheme handling.  See http://tools.ietf.org/html/rfc2397
 */

#include <stdbool.h>
#include <string.h>
#include <stdlib.h>
#include <libwapcaplet/libwapcaplet.h>
#include <nsutils/base64.h>

#include "utils/url.h"
#include "utils/nsurl.h"
#include "utils/corestrings.h"
#include "utils/log.h"
#include "utils/utils.h"
#include "utils/ring.h"

#include "content/fetch.h"
#include "content/fetchers.h"
#include "content/fetchers/data.h"

struct fetch_data_context {
	struct fetch *parent_fetch;
	char *url;
	char *mimetype;
	char *data;
	size_t datalen;
	bool base64;

	bool aborted;
	bool locked;
	
	struct fetch_data_context *r_next, *r_prev;
};

static struct fetch_data_context *ring = NULL;

static bool fetch_data_initialise(lwc_string *scheme)
{
	LOG("fetch_data_initialise called for %s", lwc_string_data(scheme));

	return true;
}

static void fetch_data_finalise(lwc_string *scheme)
{
	LOG("fetch_data_finalise called for %s", lwc_string_data(scheme));
}

static bool fetch_data_can_fetch(const nsurl *url)
{
	return true;
}

static void *fetch_data_setup(struct fetch *parent_fetch, nsurl *url,
		 bool only_2xx, bool downgrade_tls, const char *post_urlenc,
		 const struct fetch_multipart_data *post_multipart,
		 const char **headers)
{
	struct fetch_data_context *ctx = calloc(1, sizeof(*ctx));
	
	if (ctx == NULL)
		return NULL;
		
	ctx->parent_fetch = parent_fetch;

	/* TODO: keep as nsurl to avoid copy */
	ctx->url = malloc(nsurl_length(url) + 1);
	if (ctx->url == NULL) {
		free(ctx);
		return NULL;
	}
	memcpy(ctx->url, nsurl_access(url), nsurl_length(url) + 1);

	RING_INSERT(ring, ctx);
	
	return ctx;
}

static bool fetch_data_start(void *ctx)
{
	return true;
}

static void fetch_data_free(void *ctx)
{
	struct fetch_data_context *c = ctx;

	free(c->url);
	free(c->data);
	free(c->mimetype);
	RING_REMOVE(ring, c);
	free(ctx);
}

static void fetch_data_abort(void *ctx)
{
	struct fetch_data_context *c = ctx;

	/* To avoid the poll loop having to deal with the fetch context
	 * disappearing from under it, we simply flag the abort here. 
	 * The poll loop itself will perform the appropriate cleanup.
	 */
	c->aborted = true;
}

static void fetch_data_send_callback(const fetch_msg *msg, 
		struct fetch_data_context *c) 
{
	c->locked = true;
	fetch_send_callback(msg, c->parent_fetch);
	c->locked = false;
}

static bool fetch_data_process(struct fetch_data_context *c)
{
	nserror res;
	fetch_msg msg;
	char *params;
	char *comma;
	char *unescaped;
	size_t unescaped_len;
	
	/* format of a data: URL is:
	 *   data:[<mimetype>][;base64],<data>
	 * The mimetype is optional.  If it is missing, the , before the
	 * data must still be there.
	 */
	
	LOG("url: %.140s", c->url);
	
	if (strlen(c->url) < 6) {
		/* 6 is the minimum possible length (data:,) */
		msg.type = FETCH_ERROR;
		msg.data.error = "Malformed data: URL";
		fetch_data_send_callback(&msg, c);
		return false;
	}
	
	/* skip the data: part */
	params = c->url + SLEN("data:");
	
	/* find the comma */
	if ( (comma = strchr(params, ',')) == NULL) {
		msg.type = FETCH_ERROR;
		msg.data.error = "Malformed data: URL";
		fetch_data_send_callback(&msg, c);
		return false;
	}
	
	if (params[0] == ',') {
		/* there is no mimetype here, assume text/plain */
		c->mimetype = strdup("text/plain;charset=US-ASCII");
	} else {	
		/* make a copy of everything between data: and the comma */
		c->mimetype = strndup(params, comma - params);
	}
	
	if (c->mimetype == NULL) {
		msg.type = FETCH_ERROR;
		msg.data.error = 
			"Unable to allocate memory for mimetype in data: URL";
		fetch_data_send_callback(&msg, c);
		return false;
	}
	
	if (strcmp(c->mimetype + strlen(c->mimetype) - 7, ";base64") == 0) {
		c->base64 = true;
		c->mimetype[strlen(c->mimetype) - 7] = '\0';
	} else {
		c->base64 = false;
	}
	
	/* URL unescape the data first, just incase some insane page
	 * decides to nest URL and base64 encoding.  Like, say, Acid2.
	 */
	res = url_unescape(comma + 1, 0, &unescaped_len, &unescaped);
	if (res != NSERROR_OK) {
		msg.type = FETCH_ERROR;
		msg.data.error = "Unable to URL decode data: URL";
		fetch_data_send_callback(&msg, c);
		return false;
	}
	
	if (c->base64) {
		if ((nsu_base64_decode_alloc((uint8_t *)unescaped,
					     unescaped_len,
					     (uint8_t **)&c->data,
					     &c->datalen) != NSUERROR_OK) ||
		    (c->data == NULL)) {
			msg.type = FETCH_ERROR;
			msg.data.error = "Unable to Base64 decode data: URL";
			fetch_data_send_callback(&msg, c);
			free(unescaped);
			return false;
		}
	} else {
		c->data = malloc(unescaped_len);
		if (c->data == NULL) {
			msg.type = FETCH_ERROR;
			msg.data.error =
				"Unable to allocate memory for data: URL";
			fetch_data_send_callback(&msg, c);
			free(unescaped);
			return false;
		}
		c->datalen = unescaped_len;
		memcpy(c->data, unescaped, unescaped_len);
	}
	
	free(unescaped);
	
	return true;
}

static void fetch_data_poll(lwc_string *scheme)
{
	fetch_msg msg;
	struct fetch_data_context *c, *next;
	
	if (ring == NULL) return;
	
	/* Iterate over ring, processing each pending fetch */
	c = ring;
	do {
		/* Ignore fetches that have been flagged as locked.
		 * This allows safe re-entrant calls to this function.
		 * Re-entrancy can occur if, as a result of a callback,
		 * the interested party causes fetch_poll() to be called 
		 * again.
		 */
		if (c->locked == true) {
			next = c->r_next;
			continue;
		}

		/* Only process non-aborted fetches */
		if (c->aborted == false && fetch_data_process(c) == true) {
			char header[64];

			fetch_set_http_code(c->parent_fetch, 200);
			LOG("setting data: MIME type to %s, length to %" PRIsizet,
			    c->mimetype, c->datalen);
			/* Any callback can result in the fetch being aborted.
			 * Therefore, we _must_ check for this after _every_
			 * call to fetch_data_send_callback().
			 */
			snprintf(header, sizeof header, "Content-Type: %s",
					c->mimetype);
			msg.type = FETCH_HEADER;
			msg.data.header_or_data.buf = (const uint8_t *) header;
			msg.data.header_or_data.len = strlen(header);
			fetch_data_send_callback(&msg, c); 

			if (c->aborted == false) {
				snprintf(header, sizeof header, 
					 "Content-Length: %" PRIsizet,
					 c->datalen);
				msg.type = FETCH_HEADER;
				msg.data.header_or_data.buf = 
						(const uint8_t *) header;
				msg.data.header_or_data.len = strlen(header);
				fetch_data_send_callback(&msg, c);
			}

			if (c->aborted == false) {
				msg.type = FETCH_DATA;
				msg.data.header_or_data.buf = 
						(const uint8_t *) c->data;
				msg.data.header_or_data.len = c->datalen;
				fetch_data_send_callback(&msg, c);
			}

			if (c->aborted == false) {
				msg.type = FETCH_FINISHED;
				fetch_data_send_callback(&msg, c);
			}
		} else {
			LOG("Processing of %s failed!", c->url);

			/* Ensure that we're unlocked here. If we aren't, 
			 * then fetch_data_process() is broken.
			 */
			assert(c->locked == false);
		}

		/* Compute next fetch item at the last possible moment as
		 * processing this item may have added to the ring.
		 */
		next = c->r_next;

		fetch_remove_from_queues(c->parent_fetch);
		fetch_free(c->parent_fetch);

		/* Advance to next ring entry, exiting if we've reached
		 * the start of the ring or the ring has become empty
		 */
	} while ( (c = next) != ring && ring != NULL);
}

nserror fetch_data_register(void)
{
	lwc_string *scheme = lwc_string_ref(corestring_lwc_data);
	const struct fetcher_operation_table fetcher_ops = {
		.initialise = fetch_data_initialise,
		.acceptable = fetch_data_can_fetch,
		.setup = fetch_data_setup,
		.start = fetch_data_start,
		.abort = fetch_data_abort,
		.ffree = fetch_data_free,
		.poll = fetch_data_poll,
		.finalise = fetch_data_finalise
	};

	return fetcher_add(scheme, &fetcher_ops);
}