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 /* See Documentation/technical/pack-format.txt in Git. */
18 
19 struct got_pack_obj_id {
20 	u_int8_t sha1[SHA1_DIGEST_LENGTH];
21 } __attribute__((__packed__));
22 
23 struct got_packidx_trailer {
24 	u_int8_t	pack_file_sha1[SHA1_DIGEST_LENGTH];
25 	u_int8_t	pack_idx_sha1[SHA1_DIGEST_LENGTH];
26 } __attribute__((__packed__));
27 
28 /* Ignore pack index version 1 which is no longer written by Git. */
29 #define GOT_PACKIDX_VERSION 2
30 
31 struct got_packidx_v2_hdr {
32 	uint32_t	magic;		/* big endian */
33 #define GOT_PACKIDX_V2_MAGIC 0xff744f63	/* "\377t0c" */
34 	uint32_t	version;
35 
36 	/* 
37 	 * Each entry N in the fanout table contains the number of objects in
38 	 * the packfile whose SHA1 begins with a byte less than or equal to N.
39 	 * The last entry (index 255) contains the number of objects in the
40 	 * pack file whose first SHA1 byte is <= 0xff, and thus records the
41 	 * total number of objects in the pack file. All pointer variables
42 	 * below point to tables with a corresponding number of entries.
43 	 */
44 	uint32_t	fanout_table[0xff + 1];	/* values are big endian */
45 
46 	/* Sorted SHA1 checksums for each object in the pack file. */
47 	struct got_pack_obj_id *sorted_ids;
48 
49 	/* Offset into the pack file for each object. */
50 	uint32_t	*offsets;		/* values are big endian */
51 #define GOT_PACKIDX_OFFSET_VAL_MASK		0x7fffffff
52 #define GOT_PACKIDX_OFFSET_VAL_IS_LARGE_IDX	0x80000000
53 
54 	/* CRC32 of the packed representation of each object. */
55 	uint32_t	*crc32;
56 
57 	/* Large offsets table is empty for pack files < 2 GB. */
58 	uint64_t	*large_offsets;		/* values are big endian */
59 
60 	struct got_packidx_trailer trailer;
61 };
62 
63 struct got_packfile_hdr {
64 	uint32_t	signature;
65 #define GOT_PACKFILE_SIGNATURE	0x5041434b	/* 'P' 'A' 'C' 'K' */
66 	uint32_t	version;	/* big endian */
67 #define GOT_PACKFILE_VERSION 2
68 	uint32_t	nobjects;	/* big endian */
69 };
70 
71 struct got_packfile_obj_hdr {
72 	/* 
73 	 * The object size field uses a variable length encoding:
74 	 * size0...sizeN form a 4+7+7+...+7 bit integer, where size0 is the
75 	 * least significant part and sizeN is the most significant part.
76 	 * If the MSB of a size byte is set, an additional size byte follows.
77 	 * Of the 7 remaining bits of size0, the first 3 bits indicate the
78 	 * object's type, and the remaining 4 bits contribute to the size.
79 	 */
80 	uint8_t *size;		/* variable length */
81 #define GOT_PACK_OBJ_SIZE_MORE		0x80
82 #define GOT_PACK_OBJ_SIZE0_TYPE_MASK	0x70 /* See struct got_object->type */
83 #define GOT_PACK_OBJ_SIZE0_VAL_MASK	0x0f
84 #define GOT_PACK_OBJ_SIZEN_VAL_MASK	0x7f
85 };
86 
87 /* If object is not a DELTA type. */
88 struct got_packfile_object_data {
89 	uint8_t *data;	/* compressed */
90 };
91 
92 /* If object is of type	GOT_OBJ_TYPE_REF_DELTA. */
93 struct got_packfile_object_data_ref_delta {
94 	struct got_pack_obj_id id;
95 	uint8_t *delta_data;		/* compressed */
96 };
97 
98 /* If object is of type GOT_OBJ_TYPE_OFFSET_DELTA. */
99 struct got_packfile_object_data_offset_delta {
100 	/* 
101 	 * This offset is interpreted as a negative offset from
102 	 * the got_packfile_obj_hdr corresponding to this object.
103 	 * The size provided in the header specifies the amount
104 	 * of compressed delta data that follows.
105 	 *
106 	 * This field uses a variable length encoding of N bytes,
107 	 * where the MSB is always set except for the last byte.
108 	 * The value is encoded as a series of N 7 bit integers,
109 	 * which are concatenated, and if N > 1 the value 2^7 +
110 	 * 2^14 + ... + 2^(7 * (n-1)) is added to the result.
111 	 */
112 	uint8_t *offset;	/* variable length */
113 };
114 
115 struct got_packfile_obj_data {
116 	union {
117 		struct got_packfile_object_data data;
118 		struct got_packfile_object_data_ref_delta ref_delta;
119 		struct got_packfile_object_data_offset_delta offset_delta;
120 	} __attribute__((__packed__));
121 } __attribute__((__packed__));
122 
123 const struct got_error *got_packidx_open(struct got_packidx_v2_hdr **,
124     const char *);
125 void got_packidx_close(struct got_packidx_v2_hdr *);