packfile.h

   1 #ifndef PACKFILE_H
   2 #define PACKFILE_H
   3
   4 #include "list.h"
   5 #include "object.h"
   6 #include "odb.h"
   7 #include "oidset.h"
   8
   9 /* in odb.h */
  10 struct object_info;
  11
  12 struct packed_git {
  13         struct hashmap_entry packmap_ent;
  14         struct packed_git *next;
  15         struct list_head mru;
  16         struct pack_window *windows;
  17         off_t pack_size;
  18         const void *index_data;
  19         size_t index_size;
  20         uint32_t num_objects;
  21         size_t crc_offset;
  22         struct oidset bad_objects;
  23         int index_version;
  24         time_t mtime;
  25         int pack_fd;
  26         int index;              /* for builtin/pack-objects.c */
  27         unsigned pack_local:1,
  28                  pack_keep:1,
  29                  pack_keep_in_core:1,
  30                  freshened:1,
  31                  do_not_close:1,
  32                  pack_promisor:1,
  33                  multi_pack_index:1,
  34                  is_cruft:1;
  35         unsigned char hash[GIT_MAX_RAWSZ];
  36         struct revindex_entry *revindex;
  37         const uint32_t *revindex_data;
  38         const uint32_t *revindex_map;
  39         size_t revindex_size;
  40         /*
  41          * mtimes_map points at the beginning of the memory mapped region of
  42          * this pack's corresponding .mtimes file, and mtimes_size is the size
  43          * of that .mtimes file
  44          */
  45         const uint32_t *mtimes_map;
  46         size_t mtimes_size;
  47
  48         /* repo denotes the repository this packfile belongs to */
  49         struct repository *repo;
  50
  51         /* something like ".git/objects/pack/xxxxx.pack" */
  52         char pack_name[FLEX_ARRAY]; /* more */
  53 };
  54
  55 static inline int pack_map_entry_cmp(const void *cmp_data UNUSED,
  56                                      const struct hashmap_entry *entry,
  57                                      const struct hashmap_entry *entry2,
  58                                      const void *keydata)
  59 {
  60         const char *key = keydata;
  61         const struct packed_git *pg1, *pg2;
  62
  63         pg1 = container_of(entry, const struct packed_git, packmap_ent);
  64         pg2 = container_of(entry2, const struct packed_git, packmap_ent);
  65
  66         return strcmp(pg1->pack_name, key ? key : pg2->pack_name);
  67 }
  68
  69 struct pack_window {
  70         struct pack_window *next;
  71         unsigned char *base;
  72         off_t offset;
  73         size_t len;
  74         unsigned int last_used;
  75         unsigned int inuse_cnt;
  76 };
  77
  78 struct pack_entry {
  79         off_t offset;
  80         struct packed_git *p;
  81 };
  82
  83 /*
  84  * Generate the filename to be used for a pack file with checksum "sha1" and
  85  * extension "ext". The result is written into the strbuf "buf", overwriting
  86  * any existing contents. A pointer to buf->buf is returned as a convenience.
  87  *
  88  * Example: odb_pack_name(out, sha1, "idx") => ".git/objects/pack/pack-1234..idx"
  89  */
  90 char *odb_pack_name(struct repository *r, struct strbuf *buf,
  91                     const unsigned char *hash, const char *ext);
  92
  93 /*
  94  * Return the basename of the packfile, omitting any containing directory
  95  * (e.g., "pack-1234abcd[...].pack").
  96  */
  97 const char *pack_basename(struct packed_git *p);
  98
  99 /*
 100  * Parse the pack idx file found at idx_path and create a packed_git struct
 101  * which can be used with find_pack_entry_one().
 102  *
 103  * You probably don't want to use this function! It skips most of the normal
 104  * sanity checks (including whether we even have the matching .pack file),
 105  * and does not add the resulting packed_git struct to the internal list of
 106  * packs. You probably want add_packed_git() instead.
 107  */
 108 struct packed_git *parse_pack_index(struct repository *r, unsigned char *sha1,
 109                                     const char *idx_path);
 110
 111 typedef void each_file_in_pack_dir_fn(const char *full_path, size_t full_path_len,
 112                                       const char *file_name, void *data);
 113 void for_each_file_in_pack_subdir(const char *objdir,
 114                                   const char *subdir,
 115                                   each_file_in_pack_dir_fn fn,
 116                                   void *data);
 117 void for_each_file_in_pack_dir(const char *objdir,
 118                                each_file_in_pack_dir_fn fn,
 119                                void *data);
 120
 121 /*
 122  * Iterate over all accessible packed objects without respect to reachability.
 123  * By default, this includes both local and alternate packs.
 124  *
 125  * Note that some objects may appear twice if they are found in multiple packs.
 126  * Each pack is visited in an unspecified order. By default, objects within a
 127  * pack are visited in pack-idx order (i.e., sorted by oid).
 128  */
 129 typedef int each_packed_object_fn(const struct object_id *oid,
 130                                   struct packed_git *pack,
 131                                   uint32_t pos,
 132                                   void *data);
 133 int for_each_object_in_pack(struct packed_git *p,
 134                             each_packed_object_fn, void *data,
 135                             enum for_each_object_flags flags);
 136 int for_each_packed_object(struct repository *repo, each_packed_object_fn cb,
 137                            void *data, enum for_each_object_flags flags);
 138
 139 /* A hook to report invalid files in pack directory */
 140 #define PACKDIR_FILE_PACK 1
 141 #define PACKDIR_FILE_IDX 2
 142 #define PACKDIR_FILE_GARBAGE 4
 143 extern void (*report_garbage)(unsigned seen_bits, const char *path);
 144
 145 void reprepare_packed_git(struct repository *r);
 146 void install_packed_git(struct repository *r, struct packed_git *pack);
 147
 148 struct packed_git *get_packed_git(struct repository *r);
 149 struct list_head *get_packed_git_mru(struct repository *r);
 150 struct multi_pack_index *get_multi_pack_index(struct repository *r);
 151 struct multi_pack_index *get_local_multi_pack_index(struct repository *r);
 152 struct packed_git *get_all_packs(struct repository *r);
 153
 154 /*
 155  * Give a rough count of objects in the repository. This sacrifices accuracy
 156  * for speed.
 157  */
 158 unsigned long repo_approximate_object_count(struct repository *r);
 159
 160 /*
 161  * Find the pack within the "packs" list whose index contains the object "oid".
 162  * For general object lookups, you probably don't want this; use
 163  * find_pack_entry() instead.
 164  */
 165 struct packed_git *find_oid_pack(const struct object_id *oid,
 166                                  struct packed_git *packs);
 167
 168 void pack_report(struct repository *repo);
 169
 170 /*
 171  * mmap the index file for the specified packfile (if it is not
 172  * already mmapped).  Return 0 on success.
 173  */
 174 int open_pack_index(struct packed_git *);
 175
 176 /*
 177  * munmap the index file for the specified packfile (if it is
 178  * currently mmapped).
 179  */
 180 void close_pack_index(struct packed_git *);
 181
 182 int close_pack_fd(struct packed_git *p);
 183
 184 uint32_t get_pack_fanout(struct packed_git *p, uint32_t value);
 185
 186 struct object_database;
 187
 188 unsigned char *use_pack(struct packed_git *, struct pack_window **, off_t, unsigned long *);
 189 void close_pack_windows(struct packed_git *);
 190 void close_pack(struct packed_git *);
 191 void close_object_store(struct object_database *o);
 192 void unuse_pack(struct pack_window **);
 193 void clear_delta_base_cache(void);
 194 struct packed_git *add_packed_git(struct repository *r, const char *path,
 195                                   size_t path_len, int local);
 196
 197 /*
 198  * Unlink the .pack and associated extension files.
 199  * Does not unlink if 'force_delete' is false and the pack-file is
 200  * marked as ".keep".
 201  */
 202 void unlink_pack_path(const char *pack_name, int force_delete);
 203
 204 /*
 205  * Make sure that a pointer access into an mmap'd index file is within bounds,
 206  * and can provide at least 8 bytes of data.
 207  *
 208  * Note that this is only necessary for variable-length segments of the file
 209  * (like the 64-bit extended offset table), as we compare the size to the
 210  * fixed-length parts when we open the file.
 211  */
 212 void check_pack_index_ptr(const struct packed_git *p, const void *ptr);
 213
 214 /*
 215  * Perform binary search on a pack-index for a given oid. Packfile is expected to
 216  * have a valid pack-index.
 217  *
 218  * See 'bsearch_hash' for more information.
 219  */
 220 int bsearch_pack(const struct object_id *oid, const struct packed_git *p, uint32_t *result);
 221
 222 /*
 223  * Write the oid of the nth object within the specified packfile into the first
 224  * parameter. Open the index if it is not already open.  Returns 0 on success,
 225  * negative otherwise.
 226  */
 227 int nth_packed_object_id(struct object_id *, struct packed_git *, uint32_t n);
 228
 229 /*
 230  * Return the offset of the nth object within the specified packfile.
 231  * The index must already be opened.
 232  */
 233 off_t nth_packed_object_offset(const struct packed_git *, uint32_t n);
 234
 235 /*
 236  * If the object named by oid is present in the specified packfile,
 237  * return its offset within the packfile; otherwise, return 0.
 238  */
 239 off_t find_pack_entry_one(const struct object_id *oid, struct packed_git *);
 240
 241 int is_pack_valid(struct packed_git *);
 242 void *unpack_entry(struct repository *r, struct packed_git *, off_t, enum object_type *, unsigned long *);
 243 unsigned long unpack_object_header_buffer(const unsigned char *buf, unsigned long len, enum object_type *type, unsigned long *sizep);
 244 unsigned long get_size_from_delta(struct packed_git *, struct pack_window **, off_t);
 245 int unpack_object_header(struct packed_git *, struct pack_window **, off_t *, unsigned long *);
 246 off_t get_delta_base(struct packed_git *p, struct pack_window **w_curs,
 247                      off_t *curpos, enum object_type type,
 248                      off_t delta_obj_offset);
 249
 250 void release_pack_memory(size_t);
 251
 252 /* global flag to enable extra checks when accessing packed objects */
 253 extern int do_check_packed_object_crc;
 254
 255 int packed_object_info(struct repository *r,
 256                        struct packed_git *pack,
 257                        off_t offset, struct object_info *);
 258
 259 void mark_bad_packed_object(struct packed_git *, const struct object_id *);
 260 const struct packed_git *has_packed_and_bad(struct repository *, const struct object_id *);
 261
 262 #define ON_DISK_KEEP_PACKS 1
 263 #define IN_CORE_KEEP_PACKS 2
 264
 265 /*
 266  * Iff a pack file in the given repository contains the object named by sha1,
 267  * return true and store its location to e.
 268  */
 269 int find_pack_entry(struct repository *r, const struct object_id *oid, struct pack_entry *e);
 270 int find_kept_pack_entry(struct repository *r, const struct object_id *oid, unsigned flags, struct pack_entry *e);
 271
 272 int has_object_pack(struct repository *r, const struct object_id *oid);
 273 int has_object_kept_pack(struct repository *r, const struct object_id *oid,
 274                          unsigned flags);
 275
 276 struct packed_git **kept_pack_cache(struct repository *r, unsigned flags);
 277
 278 /*
 279  * Return 1 if an object in a promisor packfile is or refers to the given
 280  * object, 0 otherwise.
 281  */
 282 int is_promisor_object(struct repository *r, const struct object_id *oid);
 283
 284 /*
 285  * Expose a function for fuzz testing.
 286  *
 287  * load_idx() parses a block of memory as a packfile index and puts the results
 288  * into a struct packed_git.
 289  *
 290  * This function should not be used directly. It is exposed here only so that we
 291  * have a convenient entry-point for fuzz testing. For real uses, you should
 292  * probably use open_pack_index() instead.
 293  */
 294 int load_idx(const char *path, const unsigned int hashsz, void *idx_map,
 295              size_t idx_size, struct packed_git *p);
 296
 297 /*
 298  * Parse a --pack_header option as accepted by index-pack and unpack-objects,
 299  * turning it into the matching bytes we'd find in a pack.
 300  */
 301 int parse_pack_header_option(const char *in, unsigned char *out, unsigned int *len);
 302
 303 #endif