1 /*
2  * Copyright (c) 2018 Stefan Sperling <stsp@openbsd.org>
3  *
4  * Permission to use, copy, modify, and distribute this software for any
5  * purpose with or without fee is hereby granted, provided that the above
6  * copyright notice and this permission notice appear in all copies.
7  *
8  * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
9  * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
10  * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
11  * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
12  * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
13  * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
14  * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
15  */
16 
17 struct got_delta {
18 	SIMPLEQ_ENTRY(got_delta) entry;
19 	char *path_packfile;
20 	off_t offset;
21 	size_t tslen;
22 	int type;
23 	size_t size;
24 	off_t data_offset;
25 };
26 
27 struct got_delta_chain {
28 	int nentries;
29 	SIMPLEQ_HEAD(, got_delta) entries;
30 };
31 
32 struct got_delta *got_delta_open(const char *, off_t, size_t, int, size_t,
33     off_t);
34 void got_delta_close(struct got_delta *);
35 const struct got_error *got_delta_chain_get_base_type(int *,
36     struct got_delta_chain *);
37 const struct got_error *got_delta_apply(FILE *, const uint8_t *, size_t,
38     FILE *);
39 
40 /*
41  * The amount of result data we may keep in RAM while applying deltas.
42  * Data larger than this is written to disk during delta application (slow).
43  */
44 #define GOT_DELTA_RESULT_SIZE_CACHED_MAX	(32 * 1024 * 1024) /* bytes */
45 
46 /*
47  * Definitions for delta data streams.
48  */
49 
50 #define GOT_DELTA_STREAM_LENGTH_MIN	4	/* bytes */
51 
52 /*
53  * A delta stream begins with two size fields. The first specifies the
54  * size of the delta base, and the second describes the expected size of
55  * the data which results from applying the delta to the delta base.
56  *
57  * Each size field uses a variable length encoding:
58  * size0...sizeN form a 7+7+7+...+7 bit integer, where size0 is the
59  * least significant part and sizeN is the most significant part.
60  * If the MSB of a size byte is set, an additional size byte follows.
61  */
62 #define GOT_DELTA_SIZE_VAL_MASK	0x7f
63 #define GOT_DELTA_SIZE_SHIFT	7
64 #define GOT_DELTA_SIZE_MORE	0x80
65 
66 /*
67  * The rest of the delta stream contains copy instructions.
68  *
69  * A base copy instruction copies N bytes starting at offset X from the delta
70  * base to the output. Base copy instructions begin with a byte which has its
71  * MSB set. The remaining bits of this byte describe how many offset and
72  * length value bytes follow.
73  * The offset X is encoded in 1 to 4 bytes, and the length N is encoded in
74  * 1 to 3 bytes. For both values, the first byte contributes the least
75  * significant part and the last byte which is present contributes the
76  * most significant part.
77  * If the offset value is omitted, an offset of zero is implied.
78  * If the length value is omitted, a default length of 65536 bytes is implied.
79  *
80  * An inline copy instruction copies data from the delta stream to the output.
81  * Such instructions begin with one byte which does not have the MSB set
82  * and which specifies the length of the inline data which follows (i.e.
83  * at most 127 bytes). A length value of zero is invalid.
84  */
85 
86 #define GOT_DELTA_BASE_COPY	0x80
87 
88 #define GOT_DELTA_COPY_OFF1	0x01	/* byte 1 of offset is present */
89 #define GOT_DELTA_COPY_OFF2	0x02	/* byte 2 of offset is present */
90 #define GOT_DELTA_COPY_OFF3	0x04	/* byte 3 of offset is present */
91 #define GOT_DELTA_COPY_OFF4	0x08	/* byte 4 of offset is present */
92 
93 #define GOT_DELTA_COPY_LEN1	0x10	/* byte 1 of length is present */
94 #define GOT_DELTA_COPY_LEN2	0x20	/* byte 2 of length is present */
95 #define GOT_DELTA_COPY_LEN3	0x40	/* byte 3 of length is present */
96 
97 #define GOT_DELTA_COPY_DEFAULT_OFF	0x0	/* default offset if omitted */
98 #define GOT_DELTA_COPY_DEFAULT_LEN	0x10000 /* default length if omitted */