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 /* A pack file segment mapped with mmap(2). */
18 struct got_pack_mapping {
19 	TAILQ_ENTRY(got_pack_mapping) entry;
20 	int fd;
21 	uint8_t *addr;
22 	off_t offset;
23 	size_t len;
24 };
25 
26 #define GOT_PACK_MAX_OPEN_MAPPINGS	512
27 #define GOT_PACK_MAPPING_MIN_SIZE	8192
28 #define GOT_PACK_MAPPING_MAX_SIZE	(2048 * GOT_PACK_MAPPING_MIN_SIZE)
29 
30 /* An open pack file. */
31 struct got_pack {
32 	char *path_packfile;
33 	FILE *packfile;
34 	size_t filesize;
35 	int nmappings;
36 
37 	TAILQ_HEAD(, got_pack_mapping) mappings;
38 };
39 
40 const struct got_error *got_pack_close(struct got_pack *);
41 
42 /* See Documentation/technical/pack-format.txt in Git. */
43 
44 struct got_packidx_trailer {
45 	u_int8_t	packfile_sha1[SHA1_DIGEST_LENGTH];
46 	u_int8_t	packidx_sha1[SHA1_DIGEST_LENGTH];
47 } __attribute__((__packed__));
48 
49 /* Ignore pack index version 1 which is no longer written by Git. */
50 #define GOT_PACKIDX_VERSION 2
51 
52 struct got_packidx_v2_hdr {
53 	uint32_t	magic;		/* big endian */
54 #define GOT_PACKIDX_V2_MAGIC 0xff744f63	/* "\377t0c" */
55 	uint32_t	version;
56 
57 	/* 
58 	 * Each entry N in the fanout table contains the number of objects in
59 	 * the packfile whose SHA1 begins with a byte less than or equal to N.
60 	 * The last entry (index 255) contains the number of objects in the
61 	 * pack file whose first SHA1 byte is <= 0xff, and thus records the
62 	 * total number of objects in the pack file. All pointer variables
63 	 * below point to tables with a corresponding number of entries.
64 	 */
65 	uint32_t	fanout_table[0xff + 1];	/* values are big endian */
66 
67 	/* Sorted SHA1 checksums for each object in the pack file. */
68 	struct got_object_id *sorted_ids;
69 
70 	/* CRC32 of the packed representation of each object. */
71 	uint32_t	*crc32;
72 
73 	/* Offset into the pack file for each object. */
74 	uint32_t	*offsets;		/* values are big endian */
75 #define GOT_PACKIDX_OFFSET_VAL_MASK		0x7fffffff
76 #define GOT_PACKIDX_OFFSET_VAL_IS_LARGE_IDX	0x80000000
77 
78 	/* Large offsets table is empty for pack files < 2 GB. */
79 	uint64_t	*large_offsets;		/* values are big endian */
80 
81 	struct got_packidx_trailer trailer;
82 };
83 
84 struct got_packfile_hdr {
85 	uint32_t	signature;
86 #define GOT_PACKFILE_SIGNATURE	0x5041434b	/* 'P' 'A' 'C' 'K' */
87 	uint32_t	version;	/* big endian */
88 #define GOT_PACKFILE_VERSION 2
89 	uint32_t	nobjects;	/* big endian */
90 };
91 
92 struct got_packfile_obj_hdr {
93 	/* 
94 	 * The object size field uses a variable length encoding:
95 	 * size0...sizeN form a 4+7+7+...+7 bit integer, where size0 is the
96 	 * least significant part and sizeN is the most significant part.
97 	 * If the MSB of a size byte is set, an additional size byte follows.
98 	 * Of the 7 remaining bits of size0, the first 3 bits indicate the
99 	 * object's type, and the remaining 4 bits contribute to the size.
100 	 */
101 	uint8_t *size;		/* variable length */
102 #define GOT_PACK_OBJ_SIZE_MORE		0x80
103 #define GOT_PACK_OBJ_SIZE0_TYPE_MASK	0x70 /* See struct got_object->type */
104 #define GOT_PACK_OBJ_SIZE0_TYPE_MASK_SHIFT	4
105 #define GOT_PACK_OBJ_SIZE0_VAL_MASK	0x0f
106 #define GOT_PACK_OBJ_SIZE_VAL_MASK	0x7f
107 };
108 
109 /* If object is not a DELTA type. */
110 struct got_packfile_object_data {
111 	uint8_t *data;	/* compressed */
112 };
113 
114 /* If object is of type	GOT_OBJ_TYPE_REF_DELTA. */
115 struct got_packfile_object_data_ref_delta {
116 	uint8_t sha1[SHA1_DIGEST_LENGTH];
117 	uint8_t *delta_data;		/* compressed */
118 };
119 
120 /* If object is of type GOT_OBJ_TYPE_OFFSET_DELTA. */
121 struct got_packfile_object_data_offset_delta {
122 	/* 
123 	 * This offset is interpreted as a negative offset from
124 	 * the got_packfile_obj_hdr corresponding to this object.
125 	 * The size provided in the header specifies the amount
126 	 * of compressed delta data that follows.
127 	 *
128 	 * This field uses a variable length encoding of N bytes,
129 	 * where the MSB is always set except for the last byte.
130 	 * The value is encoded as a series of N 7 bit integers,
131 	 * which are concatenated, and if N > 1 the value 2^7 +
132 	 * 2^14 + ... + 2^(7 * (n-1)) is added to the result.
133 	 */
134 	uint8_t *offset;	/* variable length */
135 #define GOT_PACK_OBJ_DELTA_OFF_MORE		0x80
136 #define GOT_PACK_OBJ_DELTA_OFF_VAL_MASK		0x7f
137 	uint8_t *delta_data;		/* compressed */
138 };
139 
140 struct got_packfile_obj_data {
141 	union {
142 		struct got_packfile_object_data data;
143 		struct got_packfile_object_data_ref_delta ref_delta;
144 		struct got_packfile_object_data_offset_delta offset_delta;
145 	} __attribute__((__packed__));
146 } __attribute__((__packed__));
147 
148 const struct got_error *got_packidx_open(struct got_packidx_v2_hdr **,
149     const char *);
150 void got_packidx_close(struct got_packidx_v2_hdr *);
151 
152 const struct got_error *got_packfile_open_object(struct got_object **,
153     struct got_object_id *, struct got_repository *);
154 const struct got_error *got_packfile_extract_object(FILE **,
155     struct got_object *, struct got_repository *);
156 const struct got_error *got_packfile_extract_object_to_mem(uint8_t **, size_t *,
157     struct got_object *, struct got_repository *);