@matheusbsbs / matheus-bagliano / commits / 7be182045a

backfill: work with prefix pathspecs

The previous change allowed specifying revision arguments over the 'git backfill' command-line. This created the opportunity for restricting the initial commit set by filtering the revision walk through a pathspec. Other than filtering the commit set (and thereby the root trees), this did not restrict the path-walk implementation of 'git backfill' and did not restrict the blobs that were downloaded to only those matching the pathspec. Update the path-walk API to accept certain kinds of pathspecs and to silently ignore anything too complex, for now. We will update this in the next change to properly restrict to even complex pathspecs. The current behavior focuses on pathspecs that match paths exactly. This includes exact filenames, including directory names as prefixes. Pathspecs containing wildcards or magic are cleared so the path walk downloads all blobs, as before. The reason for this restriction is to allow for a faster execution by pruning the path walk to only trees that could contribute towards one of those paths as a parent directory. The test directory 'd/f/' (next to 'd/file*.txt') was prepared in a previous commit to exercise the subtlety in prefix matching. Signed-off-by: Derrick Stolee <stolee@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>

Derrick Stolee committed Mar 26, 2026 at 15:14 UTC 7be182045a6a113b118982fc81296d5b9746779e

4 files changed +52 -11

path-walk.c

+39

@@ -11,6 +11,7 @@

11 #include "list-objects.h"

12 #include "object.h"

13 #include "oid-array.h"

14 +#include "path.h"

15 #include "prio-queue.h"

16 #include "repository.h"

17 #include "revision.h"

@@ -206,6 +207,33 @@ static int add_tree_entries(struct path_walk_context *ctx,

207 match != MATCHED)

208 continue;

209 }

210 + if (ctx->revs->prune_data.nr) {

211 + struct pathspec *pd = &ctx->revs->prune_data;

212 + bool found = false;

213 + int did_strip_suffix = strbuf_strip_suffix(&path, "/");

214 +

215 +

216 + for (int i = 0; i < pd->nr; i++) {

217 + struct pathspec_item *item = &pd->items[i];

218 +

219 + /*

220 + * Continue if either is a directory prefix

221 + * of the other.

222 + */

223 + if (dir_prefix(path.buf, item->match) ||

224 + dir_prefix(item->match, path.buf)) {

225 + found = true;

226 + break;

227 + }

228 + }

229 +

230 + if (did_strip_suffix)

231 + strbuf_addch(&path, '/');

232 +

233 + /* Skip paths that do not match the prefix. */

234 + if (!found)

235 + continue;

236 + }

237

238 add_path_to_list(ctx, path.buf, type, &entry.oid,

239 !(o->flags & UNINTERESTING));

@@ -481,6 +509,17 @@ int walk_objects_by_path(struct path_walk_info *info)

509 if (info->tags)

510 info->revs->tag_objects = 1;

511

512 + if (ctx.revs->prune_data.nr) {

513 + /*

514 + * Only exact prefix pathspecs are currently supported.

515 + * Clear any wildcard or magic pathspecs to avoid

516 + * incorrect prefix matching.

517 + */

518 + if (ctx.revs->prune_data.has_wildcard ||

519 + ctx.revs->prune_data.magic)

520 + clear_pathspec(&ctx.revs->prune_data);

521 + }

522 +

523 /* Insert a single list for the root tree into the paths. */

524 CALLOC_ARRAY(root_tree_list, 1);

525 root_tree_list->type = OBJ_TREE;

path.c

+1 -1

@@ -56,7 +56,7 @@ static void strbuf_cleanup_path(struct strbuf *sb)

56 strbuf_remove(sb, 0, path - sb->buf);

57 }

59 -static int dir_prefix(const char *buf, const char *dir)

59 +int dir_prefix(const char *buf, const char *dir)

60 {

61 size_t len = strlen(dir);

62 return !strncmp(buf, dir, len) &&

path.h

@@ -112,6 +112,12 @@ const char *repo_submodule_path_replace(struct repository *repo,

112 const char *fmt, ...)

113 __attribute__((format (printf, 4, 5)));

114

115 +/*

116 + * Given a directory name 'dir' (not ending with a trailing '/'),

117 + * determine if 'buf' is equal to 'dir' or has prefix 'dir'+'/'.

118 + */

119 +int dir_prefix(const char *buf, const char *dir);

120 +

121 void report_linked_checkout_garbage(struct repository *r);

122

123 /*

t/t5620-backfill.sh

+6 -10

@@ -273,13 +273,11 @@ test_expect_success 'backfill with prefix pathspec' '

273 git -C backfill-path rev-list --quiet --objects --missing=print HEAD >missing &&

274 test_line_count = 48 missing &&

275

276 - # TODO: The pathspec should limit the downloaded blobs to

277 - # only those matching the prefix "d/f", but currently all

278 - # blobs are downloaded.

279 - git -C backfill-path backfill HEAD -- d/f &&

276 + git -C backfill-path backfill HEAD -- d/f 2>err &&

277 + test_must_be_empty err &&

278

279 git -C backfill-path rev-list --quiet --objects --missing=print HEAD >missing &&

282 - test_line_count = 0 missing

280 + test_line_count = 40 missing

281 '

282

283 test_expect_success 'backfill with multiple pathspecs' '

@@ -292,13 +290,11 @@ test_expect_success 'backfill with multiple pathspecs' '

290 git -C backfill-path rev-list --quiet --objects --missing=print HEAD >missing &&

291 test_line_count = 48 missing &&

292

295 - # TODO: The pathspecs should limit the downloaded blobs to

296 - # only those matching "d/f" or "a", but currently all blobs

297 - # are downloaded.

298 - git -C backfill-path backfill HEAD -- d/f a &&

293 + git -C backfill-path backfill HEAD -- d/f a 2>err &&

294 + test_must_be_empty err &&

295

296 git -C backfill-path rev-list --quiet --objects --missing=print HEAD >missing &&

301 - test_line_count = 0 missing

297 + test_line_count = 16 missing

298 '

299

300 test_expect_success 'backfill with wildcard pathspec' '