| 1 | #ifndef ODB_SOURCE_H |
| 2 | #define ODB_SOURCE_H |
| 3 | |
| 4 | #include "object.h" |
| 5 | #include "odb.h" |
| 6 | #include "odb/transaction.h" |
| 7 | |
| 8 | enum odb_source_type { |
| 9 | /* |
| 10 | * The "unknown" type, which should never be in use. This type mostly |
| 11 | * exists to catch cases where the type field remains zeroed out. |
| 12 | */ |
| 13 | ODB_SOURCE_UNKNOWN, |
| 14 | |
| 15 | /* The "files" backend that uses loose objects and packfiles. */ |
| 16 | ODB_SOURCE_FILES, |
| 17 | |
| 18 | /* The "loose" backend that uses loose objects, only. */ |
| 19 | ODB_SOURCE_LOOSE, |
| 20 | |
| 21 | /* The "packed" backend that uses packfiles. */ |
| 22 | ODB_SOURCE_PACKED, |
| 23 | |
| 24 | /* The "in-memory" backend that stores objects in memory. */ |
| 25 | ODB_SOURCE_INMEMORY, |
| 26 | }; |
| 27 | |
| 28 | /* |
| 29 | * Convert between the enum and its name. Returns the equivalent of "unknown" |
| 30 | * for unknown types. |
| 31 | */ |
| 32 | const char *odb_source_type_to_name(enum odb_source_type type); |
| 33 | |
| 34 | struct object_id; |
| 35 | struct odb_read_stream; |
| 36 | struct strvec; |
| 37 | |
| 38 | /* |
| 39 | * The source is the part of the object database that stores the actual |
| 40 | * objects. It thus encapsulates the logic to read and write the specific |
| 41 | * on-disk format. An object database can have multiple sources: |
| 42 | * |
| 43 | * - The primary source, which is typically located in "$GIT_DIR/objects". |
| 44 | * This is where new objects are usually written to. |
| 45 | * |
| 46 | * - Alternate sources, which are configured via "objects/info/alternates" or |
| 47 | * via the GIT_ALTERNATE_OBJECT_DIRECTORIES environment variable. These |
| 48 | * alternate sources are only used to read objects. |
| 49 | */ |
| 50 | struct odb_source { |
| 51 | struct odb_source *next; |
| 52 | |
| 53 | /* Object database that owns this object source. */ |
| 54 | struct object_database *odb; |
| 55 | |
| 56 | /* The type used by this source. */ |
| 57 | enum odb_source_type type; |
| 58 | |
| 59 | /* |
| 60 | * Figure out whether this is the local source of the owning |
| 61 | * repository, which would typically be its ".git/objects" directory. |
| 62 | * This local object directory is usually where objects would be |
| 63 | * written to. |
| 64 | */ |
| 65 | bool local; |
| 66 | |
| 67 | /* |
| 68 | * This object store is ephemeral, so there is no need to fsync. |
| 69 | */ |
| 70 | int will_destroy; |
| 71 | |
| 72 | /* |
| 73 | * Path to the source. If this is a relative path, it is relative to |
| 74 | * the current working directory. |
| 75 | */ |
| 76 | char *path; |
| 77 | |
| 78 | /* |
| 79 | * This callback is expected to free the underlying object database source and |
| 80 | * all associated resources. The function will never be called with a NULL pointer. |
| 81 | */ |
| 82 | void (*free)(struct odb_source *source); |
| 83 | |
| 84 | /* |
| 85 | * This callback is expected to close any open resources, like for |
| 86 | * example file descriptors or connections. The source is expected to |
| 87 | * still be usable after it has been closed. Closed resources may need |
| 88 | * to be reopened in that case. |
| 89 | */ |
| 90 | void (*close)(struct odb_source *source); |
| 91 | |
| 92 | /* |
| 93 | * This callback is expected to create on-disk data structures that are |
| 94 | * required for this source to operate. |
| 95 | * |
| 96 | * The callback is expected to return 0 on success, a negative error |
| 97 | * code otherwise. |
| 98 | * |
| 99 | * This callback may be NULL in case the source does not need any |
| 100 | * on-disk setup. |
| 101 | */ |
| 102 | int (*create_on_disk)(struct odb_source *source); |
| 103 | |
| 104 | /* |
| 105 | * This callback is expected to prepare the source so that it becomes |
| 106 | * ready for use. It optionally clears underlying caches of the object |
| 107 | * database source. |
| 108 | */ |
| 109 | void (*prepare)(struct odb_source *source, |
| 110 | enum odb_prepare_flags flags); |
| 111 | |
| 112 | /* |
| 113 | * This callback is expected to read object information from the object |
| 114 | * database source. The object info will be partially populated with |
| 115 | * pointers for each bit of information that was requested by the |
| 116 | * caller. |
| 117 | * |
| 118 | * The flags field is a combination of `OBJECT_INFO` flags. Only the |
| 119 | * following fields need to be handled by the backend: |
| 120 | * |
| 121 | * - `OBJECT_INFO_QUICK` indicates it is fine to use caches without |
| 122 | * re-verifying the data. |
| 123 | * |
| 124 | * - `OBJECT_INFO_SECOND_READ` indicates that the initial object |
| 125 | * lookup has failed and that the object sources should check |
| 126 | * whether any of its on-disk state has changed that may have |
| 127 | * caused the object to appear. Sources are free to ignore the |
| 128 | * second read in case they know that the first read would have |
| 129 | * already surfaced the object without reloading any on-disk state. |
| 130 | * |
| 131 | * The callback is expected to return a negative error code in case |
| 132 | * reading the object has failed, 0 otherwise. |
| 133 | */ |
| 134 | int (*read_object_info)(struct odb_source *source, |
| 135 | const struct object_id *oid, |
| 136 | struct object_info *oi, |
| 137 | enum object_info_flags flags); |
| 138 | |
| 139 | /* |
| 140 | * This callback is expected to create a new read stream that can be |
| 141 | * used to stream the object identified by the given ID. |
| 142 | * |
| 143 | * The callback is expected to return a negative error code in case |
| 144 | * creating the object stream has failed, 0 otherwise. |
| 145 | */ |
| 146 | int (*read_object_stream)(struct odb_read_stream **out, |
| 147 | struct odb_source *source, |
| 148 | const struct object_id *oid); |
| 149 | |
| 150 | /* |
| 151 | * This callback is expected to iterate over all objects stored in this |
| 152 | * source and invoke the callback function for each of them. It is |
| 153 | * valid to yield the same object multiple time. A non-zero exit code |
| 154 | * from the object callback shall abort iteration. |
| 155 | * |
| 156 | * The optional `request` structure should serve as a template for |
| 157 | * looking up object info for every individual iterated object. It |
| 158 | * should not be modified directly and should instead be copied into a |
| 159 | * separate `struct object_info` that gets passed to the callback. If |
| 160 | * the caller passes a `NULL` pointer then the object itself shall not |
| 161 | * be read. |
| 162 | * |
| 163 | * The callback is expected to return a negative error code in case the |
| 164 | * iteration has failed to read all objects, 0 otherwise. When the |
| 165 | * callback function returns a non-zero error code then that error code |
| 166 | * should be returned. |
| 167 | */ |
| 168 | int (*for_each_object)(struct odb_source *source, |
| 169 | const struct object_info *request, |
| 170 | odb_for_each_object_cb cb, |
| 171 | void *cb_data, |
| 172 | const struct odb_for_each_object_options *opts); |
| 173 | |
| 174 | /* |
| 175 | * This callback is expected to count objects in the given object |
| 176 | * database source. The callback function does not have to guarantee |
| 177 | * that only unique objects are counted. The result shall be assigned |
| 178 | * to the `out` pointer. |
| 179 | * |
| 180 | * Accepts `enum odb_count_objects_flag` flags to alter the behaviour. |
| 181 | * |
| 182 | * The callback is expected to return 0 on success, or a negative error |
| 183 | * code otherwise. |
| 184 | */ |
| 185 | int (*count_objects)(struct odb_source *source, |
| 186 | enum odb_count_objects_flags flags, |
| 187 | unsigned long *out); |
| 188 | |
| 189 | /* |
| 190 | * This callback is expected to find the minimum required length to |
| 191 | * make the given object ID unique. |
| 192 | * |
| 193 | * The callback is expected to return a negative error code in case it |
| 194 | * failed, 0 otherwise. |
| 195 | */ |
| 196 | int (*find_abbrev_len)(struct odb_source *source, |
| 197 | const struct object_id *oid, |
| 198 | unsigned min_length, |
| 199 | unsigned *out); |
| 200 | |
| 201 | /* |
| 202 | * This callback is expected to freshen the given object so that its |
| 203 | * last access time is set to the current time. This is used to ensure |
| 204 | * that objects that are recent will not get garbage collected even if |
| 205 | * they were unreachable. |
| 206 | * |
| 207 | * Returns 0 in case the object does not exist, 1 in case the object |
| 208 | * has been freshened. |
| 209 | */ |
| 210 | int (*freshen_object)(struct odb_source *source, |
| 211 | const struct object_id *oid, |
| 212 | const time_t *mtime); |
| 213 | |
| 214 | /* |
| 215 | * This callback is expected to persist the given object into the |
| 216 | * object source. In case the object already exists it shall be |
| 217 | * freshened. |
| 218 | * |
| 219 | * The flags field is a combination of `WRITE_OBJECT` flags. |
| 220 | * |
| 221 | * The resulting object ID (and optionally the compatibility object ID) |
| 222 | * shall be written into the out pointers. The callback is expected to |
| 223 | * return 0 on success, a negative error code otherwise. |
| 224 | */ |
| 225 | int (*write_object)(struct odb_source *source, |
| 226 | const void *buf, size_t len, |
| 227 | enum object_type type, |
| 228 | const struct object_id *oid, |
| 229 | const struct object_id *compat_oid, |
| 230 | const time_t *mtime, |
| 231 | enum odb_write_object_flags flags); |
| 232 | |
| 233 | /* |
| 234 | * This callback is expected to persist the given object stream into |
| 235 | * the object source. |
| 236 | * |
| 237 | * The resulting object ID shall be written into the out pointer. The |
| 238 | * callback is expected to return 0 on success, a negative error code |
| 239 | * otherwise. |
| 240 | */ |
| 241 | int (*write_object_stream)(struct odb_source *source, |
| 242 | struct odb_write_stream *stream, size_t len, |
| 243 | struct object_id *oid); |
| 244 | |
| 245 | /* |
| 246 | * This callback is expected to create a new transaction that can be |
| 247 | * used to write objects to. The objects shall only be persisted into |
| 248 | * the object database when the transcation's commit function is |
| 249 | * called. Otherwise, the objects shall be discarded. |
| 250 | * |
| 251 | * Returns 0 on success, in which case the `*out` pointer will have |
| 252 | * been populated with the object database transaction. Returns a |
| 253 | * negative error code otherwise. |
| 254 | */ |
| 255 | int (*begin_transaction)(struct odb_source *source, |
| 256 | struct odb_transaction **out, |
| 257 | enum odb_transaction_flags flags); |
| 258 | |
| 259 | /* |
| 260 | * This callback is expected to read the list of alternate object |
| 261 | * database sources connected to it and write them into the `strvec`. |
| 262 | * |
| 263 | * The result is expected to be paths to the alternates. All paths must |
| 264 | * be resolved to absolute paths. |
| 265 | * |
| 266 | * The callback is expected to return 0 on success, a negative error |
| 267 | * code otherwise. |
| 268 | */ |
| 269 | int (*read_alternates)(struct odb_source *source, |
| 270 | struct strvec *out); |
| 271 | |
| 272 | /* |
| 273 | * This callback is expected to persist the singular alternate passed |
| 274 | * to it into its list of alternates. Any pre-existing alternates are |
| 275 | * expected to remain active. Subsequent calls to `read_alternates` are |
| 276 | * thus expected to yield the pre-existing list of alternates plus the |
| 277 | * newly added alternate appended to its end. |
| 278 | * |
| 279 | * The callback is expected to return 0 on success, a negative error |
| 280 | * code otherwise. |
| 281 | */ |
| 282 | int (*write_alternate)(struct odb_source *source, |
| 283 | const char *alternate); |
| 284 | |
| 285 | /* |
| 286 | * This callback is expected to optimize the object database source. |
| 287 | * Returns 0 on success, a negative error code otherwise. |
| 288 | */ |
| 289 | int (*optimize)(struct odb_source *source, |
| 290 | const struct odb_optimize_options *opts); |
| 291 | |
| 292 | /* |
| 293 | * This callback is expected to check whether optimization of the |
| 294 | * object database source is required given the provided options. |
| 295 | * Returns true if optimization should be performed, false otherwise. |
| 296 | */ |
| 297 | bool (*optimize_required)(struct odb_source *source, |
| 298 | const struct odb_optimize_options *opts); |
| 299 | }; |
| 300 | |
| 301 | /* |
| 302 | * Allocate and initialize a new source for the given object database located |
| 303 | * at `path`. `local` indicates whether or not the source is the local and thus |
| 304 | * primary object source of the object database. |
| 305 | */ |
| 306 | struct odb_source *odb_source_new(struct object_database *odb, |
| 307 | const char *path, |
| 308 | bool local); |
| 309 | |
| 310 | /* |
| 311 | * Initialize the source for the given object database located at `path`. |
| 312 | * `local` indicates whether or not the source is the local and thus primary |
| 313 | * object source of the object database. |
| 314 | * |
| 315 | * This function is only supposed to be called by specific object source |
| 316 | * implementations. |
| 317 | */ |
| 318 | void odb_source_init(struct odb_source *source, |
| 319 | struct object_database *odb, |
| 320 | enum odb_source_type type, |
| 321 | const char *path, |
| 322 | bool local); |
| 323 | |
| 324 | /* |
| 325 | * Free the object database source, releasing all associated resources and |
| 326 | * freeing the structure itself. |
| 327 | */ |
| 328 | void odb_source_free(struct odb_source *source); |
| 329 | |
| 330 | /* |
| 331 | * Release the object database source, releasing all associated resources. |
| 332 | * |
| 333 | * This function is only supposed to be called by specific object source |
| 334 | * implementations. |
| 335 | */ |
| 336 | void odb_source_release(struct odb_source *source); |
| 337 | |
| 338 | /* |
| 339 | * Close the object database source without releasing he underlying data. The |
| 340 | * source can still be used going forward, but it first needs to be reopened. |
| 341 | * This can be useful to reduce resource usage. |
| 342 | */ |
| 343 | static inline void odb_source_close(struct odb_source *source) |
| 344 | { |
| 345 | source->close(source); |
| 346 | } |
| 347 | |
| 348 | /* |
| 349 | * Create on-disk data structures that are required for this source to operate |
| 350 | * correctly. Returns 0 on success, a negative error code otherwise. |
| 351 | */ |
| 352 | static inline int odb_source_create_on_disk(struct odb_source *source) |
| 353 | { |
| 354 | if (!source->create_on_disk) |
| 355 | return 0; |
| 356 | return source->create_on_disk(source); |
| 357 | } |
| 358 | |
| 359 | /* |
| 360 | * Prepare the object database source and clear any caches. Depending on the |
| 361 | * backend used this may have the effect that concurrently-written objects |
| 362 | * become visible. |
| 363 | */ |
| 364 | static inline void odb_source_prepare(struct odb_source *source, |
| 365 | enum odb_prepare_flags flags) |
| 366 | { |
| 367 | source->prepare(source, flags); |
| 368 | } |
| 369 | |
| 370 | /* |
| 371 | * Read an object from the object database source identified by its object ID. |
| 372 | * Returns 0 on success, a negative error code otherwise. |
| 373 | */ |
| 374 | static inline int odb_source_read_object_info(struct odb_source *source, |
| 375 | const struct object_id *oid, |
| 376 | struct object_info *oi, |
| 377 | enum object_info_flags flags) |
| 378 | { |
| 379 | return source->read_object_info(source, oid, oi, flags); |
| 380 | } |
| 381 | |
| 382 | /* |
| 383 | * Create a new read stream for the given object ID. Returns 0 on success, a |
| 384 | * negative error code otherwise. |
| 385 | */ |
| 386 | static inline int odb_source_read_object_stream(struct odb_read_stream **out, |
| 387 | struct odb_source *source, |
| 388 | const struct object_id *oid) |
| 389 | { |
| 390 | return source->read_object_stream(out, source, oid); |
| 391 | } |
| 392 | |
| 393 | /* |
| 394 | * Iterate through all objects contained in the given source and invoke the |
| 395 | * callback function for each of them. Returning a non-zero code from the |
| 396 | * callback function aborts iteration. There is no guarantee that objects |
| 397 | * are only iterated over once. |
| 398 | * |
| 399 | * The optional `request` structure serves as a template for retrieving the |
| 400 | * object info for each individual iterated object and will be populated as if |
| 401 | * `odb_source_read_object_info()` was called on the object. It will not be |
| 402 | * modified, the callback will instead be invoked with a separate `struct |
| 403 | * object_info` for every object. Object info will not be read when passing a |
| 404 | * `NULL` pointer. |
| 405 | * |
| 406 | * The flags is a bitfield of `ODB_FOR_EACH_OBJECT_*` flags. Not all flags may |
| 407 | * apply to a specific backend, so whether or not they are honored is defined |
| 408 | * by the implementation. |
| 409 | * |
| 410 | * Returns 0 when all objects have been iterated over, a negative error code in |
| 411 | * case iteration has failed, or a non-zero value returned from the callback. |
| 412 | */ |
| 413 | static inline int odb_source_for_each_object(struct odb_source *source, |
| 414 | const struct object_info *request, |
| 415 | odb_for_each_object_cb cb, |
| 416 | void *cb_data, |
| 417 | const struct odb_for_each_object_options *opts) |
| 418 | { |
| 419 | return source->for_each_object(source, request, cb, cb_data, opts); |
| 420 | } |
| 421 | |
| 422 | /* |
| 423 | * Count the number of objects in the given object database source. |
| 424 | * |
| 425 | * Returns 0 on success, a negative error code otherwise. |
| 426 | */ |
| 427 | static inline int odb_source_count_objects(struct odb_source *source, |
| 428 | enum odb_count_objects_flags flags, |
| 429 | unsigned long *out) |
| 430 | { |
| 431 | return source->count_objects(source, flags, out); |
| 432 | } |
| 433 | |
| 434 | /* |
| 435 | * Determine the minimum required length to make the given object ID unique in |
| 436 | * the given source. Returns 0 on success, a negative error code otherwise. |
| 437 | */ |
| 438 | static inline int odb_source_find_abbrev_len(struct odb_source *source, |
| 439 | const struct object_id *oid, |
| 440 | unsigned min_len, |
| 441 | unsigned *out) |
| 442 | { |
| 443 | return source->find_abbrev_len(source, oid, min_len, out); |
| 444 | } |
| 445 | |
| 446 | /* |
| 447 | * Freshen an object in the object database by updating its timestamp. |
| 448 | * Returns 1 in case the object has been freshened, 0 in case the object does |
| 449 | * not exist. |
| 450 | */ |
| 451 | static inline int odb_source_freshen_object(struct odb_source *source, |
| 452 | const struct object_id *oid, |
| 453 | const time_t *mtime) |
| 454 | { |
| 455 | return source->freshen_object(source, oid, mtime); |
| 456 | } |
| 457 | |
| 458 | /* |
| 459 | * Write an object into the object database source. Returns 0 on success, a |
| 460 | * negative error code otherwise. Populates the given out pointers for the |
| 461 | * object ID and the compatibility object ID, if non-NULL. |
| 462 | */ |
| 463 | static inline int odb_source_write_object(struct odb_source *source, |
| 464 | const void *buf, unsigned long len, |
| 465 | enum object_type type, |
| 466 | const struct object_id *oid, |
| 467 | const struct object_id *compat_oid, |
| 468 | const time_t *mtime, |
| 469 | enum odb_write_object_flags flags) |
| 470 | { |
| 471 | return source->write_object(source, buf, len, type, oid, |
| 472 | compat_oid, mtime, flags); |
| 473 | } |
| 474 | |
| 475 | /* |
| 476 | * Write an object into the object database source via a stream. The overall |
| 477 | * length of the object must be known in advance. |
| 478 | * |
| 479 | * Return 0 on success, a negative error code otherwise. Populates the given |
| 480 | * out pointer for the object ID. |
| 481 | */ |
| 482 | static inline int odb_source_write_object_stream(struct odb_source *source, |
| 483 | struct odb_write_stream *stream, |
| 484 | size_t len, |
| 485 | struct object_id *oid) |
| 486 | { |
| 487 | return source->write_object_stream(source, stream, len, oid); |
| 488 | } |
| 489 | |
| 490 | /* |
| 491 | * Read the list of alternative object database sources from the given backend |
| 492 | * and populate the `strvec` with them. The listing is not recursive -- that |
| 493 | * is, if any of the yielded alternate sources has alternates itself, those |
| 494 | * will not be yielded as part of this function call. |
| 495 | * |
| 496 | * Return 0 on success, a negative error code otherwise. |
| 497 | */ |
| 498 | static inline int odb_source_read_alternates(struct odb_source *source, |
| 499 | struct strvec *out) |
| 500 | { |
| 501 | return source->read_alternates(source, out); |
| 502 | } |
| 503 | |
| 504 | /* |
| 505 | * Write and persist a new alternate object database source for the given |
| 506 | * source. Any preexisting alternates are expected to stay valid, and the new |
| 507 | * alternate shall be appended to the end of the list. |
| 508 | * |
| 509 | * Returns 0 on success, a negative error code otherwise. |
| 510 | */ |
| 511 | static inline int odb_source_write_alternate(struct odb_source *source, |
| 512 | const char *alternate) |
| 513 | { |
| 514 | return source->write_alternate(source, alternate); |
| 515 | } |
| 516 | |
| 517 | /* |
| 518 | * Create a new transaction that can be used to write objects into a temporary |
| 519 | * staging area. The objects will only be persisted when the transaction is |
| 520 | * committed. |
| 521 | * |
| 522 | * Returns 0 on success, a negative error code otherwise. |
| 523 | */ |
| 524 | static inline int odb_source_begin_transaction(struct odb_source *source, |
| 525 | struct odb_transaction **out, |
| 526 | enum odb_transaction_flags flags) |
| 527 | { |
| 528 | return source->begin_transaction(source, out, flags); |
| 529 | } |
| 530 | |
| 531 | /* |
| 532 | * Optimize the object database source. Returns 0 on success, a negative error |
| 533 | * code otherwise. |
| 534 | */ |
| 535 | static inline int odb_source_optimize(struct odb_source *source, |
| 536 | const struct odb_optimize_options *opts) |
| 537 | { |
| 538 | return source->optimize(source, opts); |
| 539 | } |
| 540 | |
| 541 | /* |
| 542 | * Check whether optimization of the object database source is required given |
| 543 | * the provided options. Returns true if optimization should be performed, |
| 544 | * false otherwise. |
| 545 | */ |
| 546 | static inline bool odb_source_optimize_required(struct odb_source *source, |
| 547 | const struct odb_optimize_options *opts) |
| 548 | { |
| 549 | return source->optimize_required(source, opts); |
| 550 | } |
| 551 | |
| 552 | #endif |