summaryrefslogtreecommitdiffstats
path: root/qemu/roms/ipxe/src/core/uri.c
blob: 3b5f270fe61396a41044c67f57e388c66482a681 (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
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
/*
 * Copyright (C) 2007 Michael Brown <mbrown@fensystems.co.uk>.
 *
 * This program 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; either version 2 of the
 * License, or any later version.
 *
 * This program 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, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
 * 02110-1301, USA.
 *
 * You can also choose to distribute this program under the terms of
 * the Unmodified Binary Distribution Licence (as given in the file
 * COPYING.UBDL), provided that you have satisfied its requirements.
 */

FILE_LICENCE ( GPL2_OR_LATER_OR_UBDL );

/** @file
 *
 * Uniform Resource Identifiers
 *
 */

#include <stdint.h>
#include <stdlib.h>
#include <string.h>
#include <libgen.h>
#include <ctype.h>
#include <ipxe/vsprintf.h>
#include <ipxe/params.h>
#include <ipxe/uri.h>

/**
 * Decode URI field (in place)
 *
 * @v string		String
 *
 * URI decoding can never increase the length of a string; we can
 * therefore safely decode in place.
 */
static void uri_decode ( char *string ) {
	char *dest = string;
	char hexbuf[3];
	char *hexbuf_end;
	char c;
	char decoded;
	unsigned int skip;

	/* Copy string, decoding escaped characters as necessary */
	do {
		c = *(string++);
		if ( c == '%' ) {
			snprintf ( hexbuf, sizeof ( hexbuf ), "%s", string );
			decoded = strtoul ( hexbuf, &hexbuf_end, 16 );
			skip = ( hexbuf_end - hexbuf );
			string += skip;
			if ( skip )
				c = decoded;
		}
		*(dest++) = c;
	} while ( c );
}

/**
 * Check if character should be escaped within a URI field
 *
 * @v c			Character
 * @v field		URI field index
 * @ret escaped		Character should be escaped
 */
static int uri_character_escaped ( char c, unsigned int field ) {

	/* Non-printing characters and whitespace should always be
	 * escaped, since they cannot sensibly be displayed as part of
	 * a coherent URL string.  (This test also catches control
	 * characters such as CR and LF, which could affect the
	 * operation of line-based protocols such as HTTP.)
	 *
	 * We should also escape characters which would alter the
	 * interpretation of the URL if not escaped, i.e. characters
	 * which have significance to the URL parser.  We should not
	 * blindly escape all such characters, because this would lead
	 * to some very strange-looking URLs (e.g. if we were to
	 * always escape '/' as "%2F" even within the URI path).
	 *
	 * We do not need to be perfect.  Our primary role is as a
	 * consumer of URIs rather than a producer; the main situation
	 * in which we produce a URI string is for display to a human
	 * user, who can probably tolerate some variance from the
	 * formal specification.  The only situation in which we
	 * currently produce a URI string to be consumed by a computer
	 * is when constructing an HTTP request URI, which contains
	 * only the path and query fields.
	 *
	 * We can therefore sacrifice some correctness for the sake of
	 * code size.  For example, colons within the URI host should
	 * be escaped unless they form part of an IPv6 literal
	 * address; doing this correctly would require the URI
	 * formatter to be aware of whether or not the URI host
	 * contained an IPv4 address, an IPv6 address, or a host name.
	 * We choose to simplify and never escape colons within the
	 * URI host field: in the event of a pathological hostname
	 * containing colons, this could potentially produce a URI
	 * string which could not be reparsed.
	 *
	 * After excluding non-printing characters, whitespace, and
	 * '%', the full set of characters with significance to the
	 * URL parser is "/#:@?".  We choose for each URI field which
	 * of these require escaping in our use cases.
	 */
	static const char *escaped[URI_FIELDS] = {
		/* Scheme: escape everything */
		[URI_SCHEME]	= "/#:@?",
		/* Opaque part: escape characters which would affect
		 * the reparsing of the URI, allowing everything else
		 * (e.g. ':', which will appear in iSCSI URIs).
		 */
		[URI_OPAQUE]	= "/#",
		/* User name: escape everything */
		[URI_USER]	= "/#:@?",
		/* Password: escape everything */
		[URI_PASSWORD]	= "/#:@?",
		/* Host name: escape everything except ':', which may
		 * appear as part of an IPv6 literal address.
		 */
		[URI_HOST]	= "/#@?",
		/* Port number: escape everything */
		[URI_PORT]	= "/#:@?",
		/* Path: escape everything except '/', which usually
		 * appears within paths.
		 */
		[URI_PATH]	= "#:@?",
		/* Query: escape everything except '/', which
		 * sometimes appears within queries.
		 */
		[URI_QUERY]	= "#:@?",
		/* Fragment: escape everything */
		[URI_FRAGMENT]	= "/#:@?",
	};

	return ( /* Always escape non-printing characters and whitespace */
		 ( ! isprint ( c ) ) || ( c == ' ' ) ||
		 /* Always escape '%' */
		 ( c == '%' ) ||
		 /* Escape field-specific characters */
		 strchr ( escaped[field], c ) );
}

/**
 * Encode URI field
 *
 * @v uri		URI
 * @v field		URI field index
 * @v buf		Buffer to contain encoded string
 * @v len		Length of buffer
 * @ret len		Length of encoded string (excluding NUL)
 */
size_t uri_encode ( const char *string, unsigned int field,
		    char *buf, ssize_t len ) {
	ssize_t remaining = len;
	size_t used;
	char c;

	/* Ensure encoded string is NUL-terminated even if empty */
	if ( len > 0 )
		buf[0] = '\0';

	/* Copy string, escaping as necessary */
	while ( ( c = *(string++) ) ) {
		if ( uri_character_escaped ( c, field ) ) {
			used = ssnprintf ( buf, remaining, "%%%02X", c );
		} else {
			used = ssnprintf ( buf, remaining, "%c", c );
		}
		buf += used;
		remaining -= used;
	}

	return ( len - remaining );
}

/**
 * Dump URI for debugging
 *
 * @v uri		URI
 */
static void uri_dump ( const struct uri *uri ) {

	if ( ! uri )
		return;
	if ( uri->scheme )
		DBGC ( uri, " scheme \"%s\"", uri->scheme );
	if ( uri->opaque )
		DBGC ( uri, " opaque \"%s\"", uri->opaque );
	if ( uri->user )
		DBGC ( uri, " user \"%s\"", uri->user );
	if ( uri->password )
		DBGC ( uri, " password \"%s\"", uri->password );
	if ( uri->host )
		DBGC ( uri, " host \"%s\"", uri->host );
	if ( uri->port )
		DBGC ( uri, " port \"%s\"", uri->port );
	if ( uri->path )
		DBGC ( uri, " path \"%s\"", uri->path );
	if ( uri->query )
		DBGC ( uri, " query \"%s\"", uri->query );
	if ( uri->fragment )
		DBGC ( uri, " fragment \"%s\"", uri->fragment );
	if ( uri->params )
		DBGC ( uri, " params \"%s\"", uri->params->name );
}

/**
 * Free URI
 *
 * @v refcnt		Reference count
 */
static void uri_free ( struct refcnt *refcnt ) {
	struct uri *uri = container_of ( refcnt, struct uri, refcnt );

	params_put ( uri->params );
	free ( uri );
}

/**
 * Parse URI
 *
 * @v uri_string	URI as a string
 * @ret uri		URI
 *
 * Splits a URI into its component parts.  The return URI structure is
 * dynamically allocated and must eventually be freed by calling
 * uri_put().
 */
struct uri * parse_uri ( const char *uri_string ) {
	struct uri *uri;
	struct parameters *params;
	char *raw;
	char *tmp;
	char *path;
	char *authority;
	size_t raw_len;
	unsigned int field;

	/* Allocate space for URI struct and a copy of the string */
	raw_len = ( strlen ( uri_string ) + 1 /* NUL */ );
	uri = zalloc ( sizeof ( *uri ) + raw_len );
	if ( ! uri )
		return NULL;
	ref_init ( &uri->refcnt, uri_free );
	raw = ( ( ( void * ) uri ) + sizeof ( *uri ) );

	/* Copy in the raw string */
	memcpy ( raw, uri_string, raw_len );

	/* Identify the parameter list, if present */
	if ( ( tmp = strstr ( raw, "##params" ) ) ) {
		*tmp = '\0';
		tmp += 8 /* "##params" */;
		params = find_parameters ( *tmp ? ( tmp + 1 ) : NULL );
		if ( params ) {
			uri->params = claim_parameters ( params );
		} else {
			/* Ignore non-existent submission blocks */
		}
	}

	/* Chop off the fragment, if it exists */
	if ( ( tmp = strchr ( raw, '#' ) ) ) {
		*(tmp++) = '\0';
		uri->fragment = tmp;
	}

	/* Identify absolute/relative URI */
	if ( ( tmp = strchr ( raw, ':' ) ) ) {
		/* Absolute URI: identify hierarchical/opaque */
		uri->scheme = raw;
		*(tmp++) = '\0';
		if ( *tmp == '/' ) {
			/* Absolute URI with hierarchical part */
			path = tmp;
		} else {
			/* Absolute URI with opaque part */
			uri->opaque = tmp;
			path = NULL;
		}
	} else {
		/* Relative URI */
		path = raw;
	}

	/* If we don't have a path (i.e. we have an absolute URI with
	 * an opaque portion, we're already finished processing
	 */
	if ( ! path )
		goto done;

	/* Chop off the query, if it exists */
	if ( ( tmp = strchr ( path, '?' ) ) ) {
		*(tmp++) = '\0';
		uri->query = tmp;
	}

	/* If we have no path remaining, then we're already finished
	 * processing.
	 */
	if ( ! path[0] )
		goto done;

	/* Identify net/absolute/relative path */
	if ( strncmp ( path, "//", 2 ) == 0 ) {
		/* Net path.  If this is terminated by the first '/'
		 * of an absolute path, then we have no space for a
		 * terminator after the authority field, so shuffle
		 * the authority down by one byte, overwriting one of
		 * the two slashes.
		 */
		authority = ( path + 2 );
		if ( ( tmp = strchr ( authority, '/' ) ) ) {
			/* Shuffle down */
			uri->path = tmp;
			memmove ( ( authority - 1 ), authority,
				  ( tmp - authority ) );
			authority--;
			*(--tmp) = '\0';
		}
	} else {
		/* Absolute/relative path */
		uri->path = path;
		authority = NULL;
	}

	/* If we don't have an authority (i.e. we have a non-net
	 * path), we're already finished processing
	 */
	if ( ! authority )
		goto done;

	/* Split authority into user[:password] and host[:port] portions */
	if ( ( tmp = strchr ( authority, '@' ) ) ) {
		/* Has user[:password] */
		*(tmp++) = '\0';
		uri->host = tmp;
		uri->user = authority;
		if ( ( tmp = strchr ( authority, ':' ) ) ) {
			/* Has password */
			*(tmp++) = '\0';
			uri->password = tmp;
		}
	} else {
		/* No user:password */
		uri->host = authority;
	}

	/* Split host into host[:port] */
	if ( ( uri->host[ strlen ( uri->host ) - 1 ] != ']' ) &&
	     ( tmp = strrchr ( uri->host, ':' ) ) ) {
		*(tmp++) = '\0';
		uri->port = tmp;
	}

	/* Decode fields in-place */
	for ( field = 0 ; field < URI_FIELDS ; field++ ) {
		if ( uri_field ( uri, field ) )
			uri_decode ( ( char * ) uri_field ( uri, field ) );
	}

 done:
	DBGC ( uri, "URI parsed \"%s\" to", uri_string );
	uri_dump ( uri );
	DBGC ( uri, "\n" );

	return uri;
}

/**
 * Get port from URI
 *
 * @v uri		URI, or NULL
 * @v default_port	Default port to use if none specified in URI
 * @ret port		Port
 */
unsigned int uri_port ( const struct uri *uri, unsigned int default_port ) {

	if ( ( ! uri ) || ( ! uri->port ) )
		return default_port;

	return ( strtoul ( uri->port, NULL, 0 ) );
}

/**
 * Format URI
 *
 * @v uri		URI
 * @v buf		Buffer to fill with URI string
 * @v size		Size of buffer
 * @ret len		Length of URI string
 */
size_t format_uri ( const struct uri *uri, char *buf, size_t len ) {
	static const char prefixes[URI_FIELDS] = {
		[URI_OPAQUE] = ':',
		[URI_PASSWORD] = ':',
		[URI_PORT] = ':',
		[URI_PATH] = '/',
		[URI_QUERY] = '?',
		[URI_FRAGMENT] = '#',
	};
	char prefix;
	size_t used = 0;
	unsigned int field;

	/* Ensure buffer is NUL-terminated */
	if ( len )
		buf[0] = '\0';

	/* Special-case NULL URI */
	if ( ! uri )
		return 0;

	/* Generate fields */
	for ( field = 0 ; field < URI_FIELDS ; field++ ) {

		/* Skip non-existent fields */
		if ( ! uri_field ( uri, field ) )
			continue;

		/* Prefix this field, if applicable */
		prefix = prefixes[field];
		if ( ( field == URI_HOST ) && ( uri->user != NULL ) )
			prefix = '@';
		if ( ( field == URI_PATH ) && ( uri->path[0] == '/' ) )
			prefix = '\0';
		if ( prefix ) {
			used += ssnprintf ( ( buf + used ), ( len - used ),
					    "%c", prefix );
		}

		/* Encode this field */
		used += uri_encode ( uri_field ( uri, field ), field,
				     ( buf + used ), ( len - used ) );

		/* Suffix this field, if applicable */
		if ( ( field == URI_SCHEME ) && ( ! uri->opaque ) ) {
			used += ssnprintf ( ( buf + used ), ( len - used ),
					    "://" );
		}
	}

	if ( len ) {
		DBGC ( uri, "URI formatted" );
		uri_dump ( uri );
		DBGC ( uri, " to \"%s%s\"\n", buf,
		       ( ( used > len ) ? "<TRUNCATED>" : "" ) );
	}

	return used;
}

/**
 * Format URI
 *
 * @v uri		URI
 * @ret string		URI string, or NULL on failure
 *
 * The caller is responsible for eventually freeing the allocated
 * memory.
 */
char * format_uri_alloc ( const struct uri *uri ) {
	size_t len;
	char *string;

	len = ( format_uri ( uri, NULL, 0 ) + 1 /* NUL */ );
	string = malloc ( len );
	if ( string )
		format_uri ( uri, string, len );
	return string;
}

/**
 * Copy URI fields
 *
 * @v src		Source URI
 * @v dest		Destination URI, or NULL to calculate length
 * @ret len		Length of raw URI
 */
static size_t uri_copy_fields ( const struct uri *src, struct uri *dest ) {
	size_t len = sizeof ( *dest );
	char *out = ( ( void * ) dest + len );
	unsigned int field;
	size_t field_len;

	/* Copy existent fields */
	for ( field = 0 ; field < URI_FIELDS ; field++ ) {

		/* Skip non-existent fields */
		if ( ! uri_field ( src, field ) )
			continue;

		/* Calculate field length */
		field_len = ( strlen ( uri_field ( src, field ) )
			      + 1 /* NUL */ );
		len += field_len;

		/* Copy field, if applicable */
		if ( dest ) {
			memcpy ( out, uri_field ( src, field ), field_len );
			uri_field ( dest, field ) = out;
			out += field_len;
		}
	}
	return len;
}

/**
 * Duplicate URI
 *
 * @v uri		URI
 * @ret uri		Duplicate URI
 *
 * Creates a modifiable copy of a URI.
 */
struct uri * uri_dup ( const struct uri *uri ) {
	struct uri *dup;
	size_t len;

	/* Allocate new URI */
	len = uri_copy_fields ( uri, NULL );
	dup = zalloc ( len );
	if ( ! dup )
		return NULL;
	ref_init ( &dup->refcnt, uri_free );

	/* Copy fields */
	uri_copy_fields ( uri, dup );

	/* Copy parameters */
	dup->params = params_get ( uri->params );

	DBGC ( uri, "URI duplicated" );
	uri_dump ( uri );
	DBGC ( uri, "\n" );

	return dup;
}

/**
 * Resolve base+relative path
 *
 * @v base_uri		Base path
 * @v relative_uri	Relative path
 * @ret resolved_uri	Resolved path
 *
 * Takes a base path (e.g. "/var/lib/tftpboot/vmlinuz" and a relative
 * path (e.g. "initrd.gz") and produces a new path
 * (e.g. "/var/lib/tftpboot/initrd.gz").  Note that any non-directory
 * portion of the base path will automatically be stripped; this
 * matches the semantics used when resolving the path component of
 * URIs.
 */
char * resolve_path ( const char *base_path,
		      const char *relative_path ) {
	size_t base_len = ( strlen ( base_path ) + 1 );
	char base_path_copy[base_len];
	char *base_tmp = base_path_copy;
	char *resolved;

	/* If relative path is absolute, just re-use it */
	if ( relative_path[0] == '/' )
		return strdup ( relative_path );

	/* Create modifiable copy of path for dirname() */
	memcpy ( base_tmp, base_path, base_len );
	base_tmp = dirname ( base_tmp );

	/* Process "./" and "../" elements */
	while ( *relative_path == '.' ) {
		relative_path++;
		if ( *relative_path == 0 ) {
			/* Do nothing */
		} else if ( *relative_path == '/' ) {
			relative_path++;
		} else if ( *relative_path == '.' ) {
			relative_path++;
			if ( *relative_path == 0 ) {
				base_tmp = dirname ( base_tmp );
			} else if ( *relative_path == '/' ) {
				base_tmp = dirname ( base_tmp );
				relative_path++;
			} else {
				relative_path -= 2;
				break;
			}
		} else {
			relative_path--;
			break;
		}
	}

	/* Create and return new path */
	if ( asprintf ( &resolved, "%s%s%s", base_tmp,
			( ( base_tmp[ strlen ( base_tmp ) - 1 ] == '/' ) ?
			  "" : "/" ), relative_path ) < 0 )
		return NULL;

	return resolved;
}

/**
 * Resolve base+relative URI
 *
 * @v base_uri		Base URI, or NULL
 * @v relative_uri	Relative URI
 * @ret resolved_uri	Resolved URI
 *
 * Takes a base URI (e.g. "http://ipxe.org/kernels/vmlinuz" and a
 * relative URI (e.g. "../initrds/initrd.gz") and produces a new URI
 * (e.g. "http://ipxe.org/initrds/initrd.gz").
 */
struct uri * resolve_uri ( const struct uri *base_uri,
			   struct uri *relative_uri ) {
	struct uri tmp_uri;
	char *tmp_path = NULL;
	struct uri *new_uri;

	/* If relative URI is absolute, just re-use it */
	if ( uri_is_absolute ( relative_uri ) || ( ! base_uri ) )
		return uri_get ( relative_uri );

	/* Mangle URI */
	memcpy ( &tmp_uri, base_uri, sizeof ( tmp_uri ) );
	if ( relative_uri->path ) {
		tmp_path = resolve_path ( ( base_uri->path ?
					    base_uri->path : "/" ),
					  relative_uri->path );
		tmp_uri.path = tmp_path;
		tmp_uri.query = relative_uri->query;
		tmp_uri.fragment = relative_uri->fragment;
		tmp_uri.params = relative_uri->params;
	} else if ( relative_uri->query ) {
		tmp_uri.query = relative_uri->query;
		tmp_uri.fragment = relative_uri->fragment;
		tmp_uri.params = relative_uri->params;
	} else if ( relative_uri->fragment ) {
		tmp_uri.fragment = relative_uri->fragment;
		tmp_uri.params = relative_uri->params;
	} else if ( relative_uri->params ) {
		tmp_uri.params = relative_uri->params;
	}

	/* Create demangled URI */
	new_uri = uri_dup ( &tmp_uri );
	free ( tmp_path );
	return new_uri;
}

/**
 * Construct TFTP URI from next-server and filename
 *
 * @v next_server	Next-server address
 * @v port		Port number, or zero to use the default port
 * @v filename		Filename
 * @ret uri		URI, or NULL on failure
 *
 * TFTP filenames specified via the DHCP next-server field often
 * contain characters such as ':' or '#' which would confuse the
 * generic URI parser.  We provide a mechanism for directly
 * constructing a TFTP URI from the next-server and filename.
 */
struct uri * tftp_uri ( struct in_addr next_server, unsigned int port,
			const char *filename ) {
	char buf[ 6 /* "65535" + NUL */ ];
	struct uri uri;

	memset ( &uri, 0, sizeof ( uri ) );
	uri.scheme = "tftp";
	uri.host = inet_ntoa ( next_server );
	if ( port ) {
		snprintf ( buf, sizeof ( buf ), "%d", port );
		uri.port = buf;
	}
	uri.path = filename;
	return uri_dup ( &uri );
}