@matheusbsbs / matheus-bagliano / commits / 190a65f9db

Fork 0

sparse-checkout: respect core.ignoreCase in cone mode

When a user uses the sparse-checkout feature in cone mode, they add patterns using "git sparse-checkout set <dir1> <dir2> ..." or by using "--stdin" to provide the directories line-by-line over stdin. This behaviour naturally looks a lot like the way a user would type "git add <dir1> <dir2> ..." If core.ignoreCase is enabled, then "git add" will match the input using a case-insensitive match. Do the same for the sparse-checkout feature. Perform case-insensitive checks while updating the skip-worktree bits during unpack_trees(). This is done by changing the hash algorithm and hashmap comparison methods to optionally use case- insensitive methods. When this is enabled, there is a small performance cost in the hashing algorithm. To tease out the worst possible case, the following was run on a repo with a deep directory structure: git ls-tree -d -r --name-only HEAD | git sparse-checkout set --stdin The 'set' command was timed with core.ignoreCase disabled or enabled. For the repo with a deep history, the numbers were core.ignoreCase=false: 62s core.ignoreCase=true: 74s (+19.3%) For reproducibility, the equivalent test on the Linux kernel repository had these numbers: core.ignoreCase=false: 3.1s core.ignoreCase=true: 3.6s (+16%) Now, this is not an entirely fair comparison, as most users will define their sparse cone using more shallow directories, and the performance improvement from eb42feca97 ("unpack-trees: hash less in cone mode" 2019-11-21) can remove most of the hash cost. For a more realistic test, drop the "-r" from the ls-tree command to store only the first-level directories. In that case, the Linux kernel repository takes 0.2-0.25s in each case, and the deep repository takes one second, plus or minus 0.05s, in each case. Thus, we _can_ demonstrate a cost to this change, but it is unlikely to matter to any reasonable sparse-checkout cone. Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>

Derrick Stolee committed Dec 13, 2019 at 18:09 UTC 190a65f9db8db9d87d54351429f7879fcb4ad608

4 files changed +42 -5

Documentation/git-sparse-checkout.txt

@@ -150,6 +150,11 @@ expecting patterns of these types. Git will warn if the patterns do not match.

150 If the patterns do match the expected format, then Git will use faster hash-

151 based algorithms to compute inclusion in the sparse-checkout.

152

153 +If `core.ignoreCase=true`, then the pattern-matching algorithm will use a

154 +case-insensitive check. This corrects for case mismatched filenames in the

155 +'git sparse-checkout set' command to reflect the expected cone in the working

156 +directory.

157 +

158 SEE ALSO

159 --------

160

builtin/sparse-checkout.c

+8 -2

@@ -313,7 +313,10 @@ static void insert_recursive_pattern(struct pattern_list *pl, struct strbuf *pat

313 struct pattern_entry *e = xmalloc(sizeof(*e));

314 e->patternlen = path->len;

315 e->pattern = strbuf_detach(path, NULL);

316 - hashmap_entry_init(&e->ent, memhash(e->pattern, e->patternlen));

316 + hashmap_entry_init(&e->ent,

317 + ignore_case ?

318 + strihash(e->pattern) :

319 + strhash(e->pattern));

320

321 hashmap_add(&pl->recursive_hashmap, &e->ent);

322

@@ -329,7 +332,10 @@ static void insert_recursive_pattern(struct pattern_list *pl, struct strbuf *pat

332 e = xmalloc(sizeof(struct pattern_entry));

333 e->patternlen = newlen;

334 e->pattern = xstrndup(oldpattern, newlen);

332 - hashmap_entry_init(&e->ent, memhash(e->pattern, e->patternlen));

335 + hashmap_entry_init(&e->ent,

336 + ignore_case ?

337 + strihash(e->pattern) :

338 + strhash(e->pattern));

339

340 if (!hashmap_get_entry(&pl->parent_hashmap, e, ent, NULL))

341 hashmap_add(&pl->parent_hashmap, &e->ent);

dir.c

+12 -3

@@ -625,6 +625,8 @@ int pl_hashmap_cmp(const void *unused_cmp_data,

625 ? ee1->patternlen

626 : ee2->patternlen;

627

628 + if (ignore_case)

629 + return strncasecmp(ee1->pattern, ee2->pattern, min_len);

630 return strncmp(ee1->pattern, ee2->pattern, min_len);

631 }

632

@@ -665,7 +667,9 @@ static void add_pattern_to_hashsets(struct pattern_list *pl, struct path_pattern

667 translated->pattern = truncated;

668 translated->patternlen = given->patternlen - 2;

669 hashmap_entry_init(&translated->ent,

668 - memhash(translated->pattern, translated->patternlen));

670 + ignore_case ?

671 + strihash(translated->pattern) :

672 + strhash(translated->pattern));

673

674 if (!hashmap_get_entry(&pl->recursive_hashmap,

675 translated, ent, NULL)) {

@@ -694,7 +698,9 @@ static void add_pattern_to_hashsets(struct pattern_list *pl, struct path_pattern

698 translated->pattern = xstrdup(given->pattern);

699 translated->patternlen = given->patternlen;

700 hashmap_entry_init(&translated->ent,

697 - memhash(translated->pattern, translated->patternlen));

701 + ignore_case ?

702 + strihash(translated->pattern) :

703 + strhash(translated->pattern));

704

705 hashmap_add(&pl->recursive_hashmap, &translated->ent);

706

@@ -724,7 +730,10 @@ static int hashmap_contains_path(struct hashmap *map,

730 /* Check straight mapping */

731 p.pattern = pattern->buf;

732 p.patternlen = pattern->len;

727 - hashmap_entry_init(&p.ent, memhash(p.pattern, p.patternlen));

733 + hashmap_entry_init(&p.ent,

734 + ignore_case ?

735 + strihash(p.pattern) :

736 + strhash(p.pattern));

737 return !!hashmap_get_entry(map, &p, ent, NULL);

738 }

739

t/t1091-sparse-checkout-builtin.sh

+17

@@ -304,4 +304,21 @@ test_expect_success 'sparse-checkout (init|set|disable) fails with dirty status'

304 git -C dirty sparse-checkout disable

305 '

306

307 +test_expect_success 'cone mode: set with core.ignoreCase=true' '

308 + git -C repo sparse-checkout init --cone &&

309 + git -C repo -c core.ignoreCase=true sparse-checkout set folder1 &&

310 + cat >expect <<-EOF &&

311 + /*

312 + !/*/

313 + /folder1/

314 + EOF

315 + test_cmp expect repo/.git/info/sparse-checkout &&

316 + ls repo >dir &&

317 + cat >expect <<-EOF &&

318 + a

319 + folder1

320 + EOF

321 + test_cmp expect dir

322 +'

323 +

324 test_done