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 /* An open pack file. */
18 struct got_pack {
19 	char *path_packfile;
20 	int fd;
21 	size_t filesize;
22 };
23 
24 void got_pack_close(struct got_pack *);
25 
26 /* See Documentation/technical/pack-format.txt in Git. */
27 
28 struct got_packidx_trailer {
29 	u_int8_t	packfile_sha1[SHA1_DIGEST_LENGTH];
30 	u_int8_t	packidx_sha1[SHA1_DIGEST_LENGTH];
31 } __attribute__((__packed__));
32 
33 /* Ignore pack index version 1 which is no longer written by Git. */
34 #define GOT_PACKIDX_VERSION 2
35 
36 struct got_packidx_v2_hdr {
37 	uint32_t	magic;		/* big endian */
38 #define GOT_PACKIDX_V2_MAGIC 0xff744f63	/* "\377t0c" */
39 	uint32_t	version;
40 
41 	/* 
42 	 * Each entry N in the fanout table contains the number of objects in
43 	 * the packfile whose SHA1 begins with a byte less than or equal to N.
44 	 * The last entry (index 255) contains the number of objects in the
45 	 * pack file whose first SHA1 byte is <= 0xff, and thus records the
46 	 * total number of objects in the pack file. All pointer variables
47 	 * below point to tables with a corresponding number of entries.
48 	 */
49 	uint32_t	fanout_table[0xff + 1];	/* values are big endian */
50 
51 	/* Sorted SHA1 checksums for each object in the pack file. */
52 	struct got_object_id *sorted_ids;
53 
54 	/* CRC32 of the packed representation of each object. */
55 	uint32_t	*crc32;
56 
57 	/* Offset into the pack file for each object. */
58 	uint32_t	*offsets;		/* values are big endian */
59 #define GOT_PACKIDX_OFFSET_VAL_MASK		0x7fffffff
60 #define GOT_PACKIDX_OFFSET_VAL_IS_LARGE_IDX	0x80000000
61 
62 	/* Large offsets table is empty for pack files < 2 GB. */
63 	uint64_t	*large_offsets;		/* values are big endian */
64 
65 	struct got_packidx_trailer trailer;
66 };
67 
68 /* An open pack index file. */
69 struct got_packidx {
70 	char *path_packidx; /* actual on-disk path */
71 	struct got_packidx_v2_hdr hdr;
72 };
73 
74 struct got_packfile_hdr {
75 	uint32_t	signature;
76 #define GOT_PACKFILE_SIGNATURE	0x5041434b	/* 'P' 'A' 'C' 'K' */
77 	uint32_t	version;	/* big endian */
78 #define GOT_PACKFILE_VERSION 2
79 	uint32_t	nobjects;	/* big endian */
80 };
81 
82 struct got_packfile_obj_hdr {
83 	/* 
84 	 * The object size field uses a variable length encoding:
85 	 * size0...sizeN form a 4+7+7+...+7 bit integer, where size0 is the
86 	 * least significant part and sizeN is the most significant part.
87 	 * If the MSB of a size byte is set, an additional size byte follows.
88 	 * Of the 7 remaining bits of size0, the first 3 bits indicate the
89 	 * object's type, and the remaining 4 bits contribute to the size.
90 	 */
91 	uint8_t *size;		/* variable length */
92 #define GOT_PACK_OBJ_SIZE_MORE		0x80
93 #define GOT_PACK_OBJ_SIZE0_TYPE_MASK	0x70 /* See struct got_object->type */
94 #define GOT_PACK_OBJ_SIZE0_TYPE_MASK_SHIFT	4
95 #define GOT_PACK_OBJ_SIZE0_VAL_MASK	0x0f
96 #define GOT_PACK_OBJ_SIZE_VAL_MASK	0x7f
97 };
98 
99 /* If object is not a DELTA type. */
100 struct got_packfile_object_data {
101 	uint8_t *data;	/* compressed */
102 };
103 
104 /* If object is of type	GOT_OBJ_TYPE_REF_DELTA. */
105 struct got_packfile_object_data_ref_delta {
106 	uint8_t sha1[SHA1_DIGEST_LENGTH];
107 	uint8_t *delta_data;		/* compressed */
108 };
109 
110 /* If object is of type GOT_OBJ_TYPE_OFFSET_DELTA. */
111 struct got_packfile_object_data_offset_delta {
112 	/* 
113 	 * This offset is interpreted as a negative offset from
114 	 * the got_packfile_obj_hdr corresponding to this object.
115 	 * The size provided in the header specifies the amount
116 	 * of compressed delta data that follows.
117 	 *
118 	 * This field uses a variable length encoding of N bytes,
119 	 * where the MSB is always set except for the last byte.
120 	 * The value is encoded as a series of N 7 bit integers,
121 	 * which are concatenated, and if N > 1 the value 2^7 +
122 	 * 2^14 + ... + 2^(7 * (n-1)) is added to the result.
123 	 */
124 	uint8_t *offset;	/* variable length */
125 #define GOT_PACK_OBJ_DELTA_OFF_MORE		0x80
126 #define GOT_PACK_OBJ_DELTA_OFF_VAL_MASK		0x7f
127 	uint8_t *delta_data;		/* compressed */
128 };
129 
130 struct got_packfile_obj_data {
131 	union {
132 		struct got_packfile_object_data data;
133 		struct got_packfile_object_data_ref_delta ref_delta;
134 		struct got_packfile_object_data_offset_delta offset_delta;
135 	} __attribute__((__packed__));
136 } __attribute__((__packed__));
137 
138 const struct got_error *got_packidx_open(struct got_packidx **,
139     const char *);
140 void got_packidx_close(struct got_packidx *);
141 
142 const struct got_error *got_packfile_open_object(struct got_object **,
143     struct got_object_id *, struct got_repository *);
144 const struct got_error *got_packfile_extract_object(FILE **,
145     struct got_object *, struct got_repository *);
146 const struct got_error *got_packfile_extract_object_to_mem(uint8_t **, size_t *,
147     struct got_object *, struct got_repository *);