1 
2 /*
3  * Copyright (c) 2017 Stefan Sperling <stsp@openbsd.org>
4  *
5  * Permission to use, copy, modify, and distribute this software for any
6  * purpose with or without fee is hereby granted, provided that the above
7  * copyright notice and this permission notice appear in all copies.
8  *
9  * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
10  * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
11  * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
12  * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
13  * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
14  * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
15  * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
16  */
17 
18 /* See Documentation/technical/pack-format.txt in Git. */
19 
20 struct got_pack_obj_id {
21 	u_int8_t sha1[SHA1_DIGEST_LENGTH];
22 } __attribute__((__packed__));
23 
24 struct got_pack_idx_trailer {
25 	u_int8_t	pack_file_sha1[SHA1_DIGEST_LENGTH];
26 	u_int8_t	pack_idx_sha1[SHA1_DIGEST_LENGTH];
27 } __attribute__((__packed__));
28 
29 /* Ignore pack index version 1 which is no longer written by Git. */
30 #define GOT_PACKIDX_VERSION 2
31 
32 struct got_packidx_v2_hdr {
33 	uint32_t	magic;		/* big endian */
34 #define GOT_PACKIDX_V2_MAGIC 0xff744f63	/* "\377t0c" */
35 	uint32_t	version;
36 
37 	/* 
38 	 * Each entry N in the fanout table contains the number of objects in
39 	 * the packfile whose SHA1 begins with a byte less than or equal to N.
40 	 * The last entry (index 255) contains the number of objects in the
41 	 * pack file whose first SHA1 byte is <= 0xff, and thus records the
42 	 * total number of objects in the pack file. All pointer variables
43 	 * below point to tables with a corresponding number of entries.
44 	 */
45 	uint32_t	fanout_table[0xff];	/* values are big endian */
46 
47 	/* Sorted SHA1 checksums for each object in the pack file. */
48 	struct got_pack_obj_id *sorted_ids;
49 
50 	/* Offset into the pack file for each object. */
51 	uint32_t	*offsets;		/* values are big endian */
52 #define GOT_PACKIDX_OFFSET_VAL_MASK		0x7fffffff
53 #define GOT_PACKIDX_OFFSET_VAL_IS_LARGE_IDX	0x80000000
54 
55 	/* CRC32 of the packed representation of each object. */
56 	uint32_t	*crc32;
57 
58 	/* Large offsets table is empty for pack files < 2 GB. */
59 	uint64_t	*large_offsets;		/* values are big endian */
60 
61 	struct got_pack_idx_trailer trailer;
62 };
63 
64 struct got_packfile_hdr {
65 	uint32_t	signature;
66 #define GOT_PACKFILE_SIGNATURE	0x4041434b	/* 'P' 'A' 'C' 'K' */
67 	uint32_t	version;	/* big endian */
68 #define GOT_PACKFILE_VERSION 2
69 	uint32_t	nobjects;	/* big endian */
70 };
71 
72 struct got_packfile_obj_hdr {
73 	/* 
74 	 * The object size field uses a variable length encoding:
75 	 * size0...sizeN form a 4+7+7+...+7 bit integer, where size0 is the
76 	 * least significant part and sizeN is the most significant part.
77 	 * If the MSB of a size byte is set, an additional size byte follows.
78 	 * Of the 7 remaining bits of size0, the first 3 bits indicate the
79 	 * object's type, and the remaining 4 bits contribute to the size.
80 	 */
81 	uint8_t *size;		/* variable length */
82 #define GOT_PACK_OBJ_SIZE_MORE		0x80
83 #define GOT_PACK_OBJ_SIZE0_TYPE_MASK	0x70 /* See struct got_object->type */
84 #define GOT_PACK_OBJ_SIZE0_VAL_MASK	0x0f
85 #define GOT_PACK_OBJ_SIZEN_VAL_MASK	0x7f
86 };
87 
88 /* If object is not a DELTA type. */
89 struct got_packfile_object_data {
90 	uint8_t *data;	/* compressed */
91 };
92 
93 /* If object is of type	GOT_OBJ_TYPE_REF_DELTA. */
94 struct got_packfile_object_data_ref_delta {
95 	struct got_pack_obj_id id;
96 	uint8_t *delta_data;		/* compressed */
97 };
98 
99 /* If object is of type GOT_OBJ_TYPE_OFFSET_DELTA. */
100 struct got_packfile_object_data_offset_delta {
101 	/* 
102 	 * This offset is interpreted as a negative offset from
103 	 * the got_packfile_obj_hdr corresponding to this object.
104 	 * The size provided in the header specifies the amount
105 	 * of compressed delta data that follows.
106 	 *
107 	 * This field uses a variable length encoding of N bytes,
108 	 * where the MSB is always set except for the last byte.
109 	 * The value is encoded as a series of N 7 bit integers,
110 	 * which are concatenated, and if N > 1 the value 2^7 +
111 	 * 2^14 + ... + 2^(7 * (n-1)) is added to the result.
112 	 */
113 	uint8_t *offset;	/* variable length */
114 };
115 
116 struct got_packfile_obj_data {
117 	union {
118 		struct got_packfile_object_data;
119 		struct got_packfile_object_data_ref_delta;
120 		struct got_packfile_object_data_offset_delta;
121 	} __attribute__((__packed__));
122 } __attribute__((__packed__));