git/list[1] front-page[2] threads[3] people[4] search[5] about
 

[PATCH v2 5/7] multi-pack-index: implement 'expire' verb

From
Derrick Stolee via GitGitGadget <gitgitgadget@gmail.com>
Date
Dec 21, 2018, 16:28 UTC
Message-ID
<1c4af93f5e9833d74a7cf7fc6d38df7fd93c1acc.1545409716.git.gitgitgadget@gmail.com>
In-Reply-To
<pull.92.v2.git.gitgitgadget@gmail.com>
From: Derrick Stolee <dstolee@microsoft.com>

The 'git multi-pack-index expire' command looks at the existing mult-pack-index, counts the number of objects referenced in each pack-file, deletes the pack-fils with no referenced objects, and rewrites the multi-pack-index to no longer reference those packs.

Refactor the write_midx_file() method to call write_midx_internal() which now takes an existing 'struct multi_pack_index' and a list of pack-files to drop (as specified by the names of their pack- indexes). As we write the new multi-pack-index, we drop those file names from the list of known pack-files.

The expire_midx_packs() method removes the unreferenced pack-files after carefully closing the packs to avoid open handles.

Test that a new pack-file that covers the contents of two other pack-files leads to those pack-files being deleted during the expire command. Be sure to read the multi-pack-index to ensure it no longer references those packs.

Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
---
 midx.c                      | 82 +++++++++++++++++++++++++++++++++++--
 t/t5319-multi-pack-index.sh | 15 +++++++
 2 files changed, 93 insertions(+), 4 deletions(-)
diff --git a/midx.c b/midx.c
index 3bd7183a53..043cd1fd97 100644
--- a/midx.c
+++ b/midx.c
@@ -751,7 +751,8 @@ static size_t write_midx_large_offsets(struct hashfile *f, uint32_t nr_large_off
 	return written;
 }
 
-int write_midx_file(const char *object_dir)
+static int write_midx_internal(const char *object_dir, struct multi_pack_index *m,
+			       struct string_list *packs_to_drop)
 {
 	unsigned char cur_chunk, num_chunks = 0;
 	char *midx_name;
@@ -765,6 +766,7 @@ int write_midx_file(const char *object_dir)
 	uint32_t nr_entries, num_large_offsets = 0;
 	struct pack_midx_entry *entries = NULL;
 	int large_offsets_needed = 0;
+	int result = 0;
 
 	midx_name = get_midx_filename(object_dir);
 	if (safe_create_leading_directories(midx_name)) {
@@ -773,7 +775,10 @@ int write_midx_file(const char *object_dir)
 			  midx_name);
 	}
 
-	packs.m = load_multi_pack_index(object_dir, 1);
+	if (m)
+		packs.m = m;
+	else
+		packs.m = load_multi_pack_index(object_dir, 1);
 
 	packs.nr = 0;
 	packs.alloc_list = packs.m ? packs.m->num_packs : 16;
@@ -787,7 +792,25 @@ int write_midx_file(const char *object_dir)
 	ALLOC_ARRAY(packs.perm, packs.alloc_perm);
 
 	if (packs.m) {
+		int drop_index = 0, missing_drops = 0;
 		for (i = 0; i < packs.m->num_packs; i++) {
+			if (packs_to_drop && drop_index < packs_to_drop->nr) {
+				int cmp = strcmp(packs.m->pack_names[i],
+						 packs_to_drop->items[drop_index].string);
+
+				if (!cmp) {
+					drop_index++;
+					continue;
+				} else if (cmp > 0) {
+					error(_("did not see pack-file %s to drop"),
+					      packs_to_drop->items[drop_index].string);
+					drop_index++;
+					i--;
+					missing_drops++;
+					continue;
+				}
+			}
+
 			ALLOC_GROW(packs.list, packs.nr + 1, packs.alloc_list);
 			ALLOC_GROW(packs.names, packs.nr + 1, packs.alloc_names);
 			ALLOC_GROW(packs.perm, packs.nr + 1, packs.alloc_perm);
@@ -798,6 +821,12 @@ int write_midx_file(const char *object_dir)
 			packs.pack_name_concat_len += strlen(packs.names[packs.nr]) + 1;
 			packs.nr++;
 		}
+
+		if (packs_to_drop && (drop_index < packs_to_drop->nr || missing_drops)) {
+			error(_("did not see all pack-files to drop"));
+			result = 1;
+			goto cleanup;
+		}
 	}
 
 	for_each_file_in_pack_dir(object_dir, add_pack_to_midx, &packs);
@@ -932,7 +961,12 @@ int write_midx_file(const char *object_dir)
 	free(packs.perm);
 	free(entries);
 	free(midx_name);
-	return 0;
+	return result;
+}
+
+int write_midx_file(const char *object_dir)
+{
+	return write_midx_internal(object_dir, NULL, NULL);
 }
 
 void clear_midx_file(struct repository *r)
@@ -1034,5 +1068,45 @@ int verify_midx_file(const char *object_dir)
 
 int expire_midx_packs(const char *object_dir)
 {
-	return 0;
+	uint32_t i, *count, result = 0;
+	struct string_list packs_to_drop = STRING_LIST_INIT_DUP;
+	struct multi_pack_index *m = load_multi_pack_index(object_dir, 1);
+
+	if (!m)
+		return 0;
+
+	count = xcalloc(m->num_packs, sizeof(uint32_t));
+	for (i = 0; i < m->num_objects; i++) {
+		int pack_int_id = nth_midxed_pack_int_id(m, i);
+		count[pack_int_id]++;
+	}
+
+	for (i = 0; i < m->num_packs; i++) {
+		char *pack_name;
+
+		if (count[i])
+			continue;
+
+		if (prepare_midx_pack(m, i))
+			continue;
+
+		if (m->packs[i]->pack_keep)
+			continue;
+
+		pack_name = xstrdup(m->packs[i]->pack_name);
+		close_pack(m->packs[i]);
+		FREE_AND_NULL(m->packs[i]);
+
+		string_list_insert(&packs_to_drop, m->pack_names[i]);
+		unlink_pack_path(pack_name, 0);
+		free(pack_name);
+	}
+
+	free(count);
+
+	if (packs_to_drop.nr)
+		result = write_midx_internal(object_dir, m, &packs_to_drop);
+
+	string_list_clear(&packs_to_drop, 0);
+	return result;
 }
diff --git a/t/t5319-multi-pack-index.sh b/t/t5319-multi-pack-index.sh
index 948effc1ee..210279a3cf 100755
--- a/t/t5319-multi-pack-index.sh
+++ b/t/t5319-multi-pack-index.sh
@@ -395,4 +395,19 @@ test_expect_success 'expire does not remove any packs' '
 	)
 '
 
+test_expect_success 'expire removes unreferenced packs' '
+	(
+		cd dup &&
+		git pack-objects --revs .git/objects/pack/pack-combined <<-EOF &&
+		refs/heads/A
+		^refs/heads/C
+		EOF
+		git multi-pack-index write &&
+		ls .git/objects/pack | grep -v -e pack-[AB] >expect &&
+		git multi-pack-index expire &&
+		ls .git/objects/pack >actual &&
+		test_cmp expect actual
+	)
+'
+
 test_done
-- 
gitgitgadget
Previous: Derrick Stolee via GitGitGadgetNext: Derrick Stolee via GitGitGadget
Message 21 of 92 in “Create 'expire' and 'repack' verbs for git-multi-pack-index”
  1. 0/5 Create 'expire' and 'repack' verbs for git-multi-pack-indexDerrick Stolee via GitGitGadget, Dec 10, 2018
  2. 1/5 multi-pack-index: prepare for 'expire' verbDerrick Stolee via GitGitGadget, Dec 10, 2018
  3. Stefan BellerDec 11, 2018
  4. SZEDER GáborDec 11, 2018
  5. Derrick StoleeDec 11, 2018
  6. 3/5 multi-pack-index: implement 'expire' verbDerrick Stolee via GitGitGadget, Dec 10, 2018
  7. 2/5 midx: refactor permutation logicDerrick Stolee via GitGitGadget, Dec 10, 2018
  8. 4/5 multi-pack-index: prepare 'repack' verbDerrick Stolee via GitGitGadget, Dec 10, 2018
  9. Stefan BellerDec 11, 2018
  10. Derrick StoleeDec 11, 2018
  11. 5/5 midx: implement midx_repack()Derrick Stolee via GitGitGadget, Dec 10, 2018
  12. Stefan BellerDec 11, 2018
  13. Derrick StoleeDec 11, 2018
  14. Junio C HamanoDec 12, 2018
  15. Junio C HamanoDec 13, 2018
  16. 0/7 Create 'expire' and 'repack' verbs for git-multi-pack-indexDerrick Stolee via GitGitGadget, Dec 21, 2018
  17. 1/7 repack: refactor pack deletion for future useDerrick Stolee via GitGitGadget, Dec 21, 2018
  18. 2/7 Docs: rearrange subcommands for multi-pack-indexDerrick Stolee via GitGitGadget, Dec 21, 2018
  19. 3/7 multi-pack-index: prepare for 'expire' subcommandDerrick Stolee via GitGitGadget, Dec 21, 2018
  20. 6/7 multi-pack-index: prepare 'repack' subcommandDerrick Stolee via GitGitGadget, Dec 21, 2018
  21. 5/7 multi-pack-index: implement 'expire' verbDerrick Stolee via GitGitGadget, Dec 21, 2018
  22. 7/7 midx: implement midx_repack()Derrick Stolee via GitGitGadget, Dec 21, 2018
  23. 4/7 midx: refactor permutation logicDerrick Stolee via GitGitGadget, Dec 21, 2018
  24. 0/9 Create 'expire' and 'repack' verbs for git-multi-pack-indexDerrick Stolee via GitGitGadget, Jan 9, 2019
  25. 1/9 repack: refactor pack deletion for future useDerrick Stolee via GitGitGadget, Jan 9, 2019
  26. 3/9 multi-pack-index: prepare for 'expire' subcommandDerrick Stolee via GitGitGadget, Jan 9, 2019
  27. 4/9 midx: simplify computation of pack name lengthsDerrick Stolee via GitGitGadget, Jan 9, 2019
  28. 6/9 multi-pack-index: implement 'expire' verbDerrick Stolee via GitGitGadget, Jan 9, 2019
  29. SZEDER GáborJan 9, 2019
  30. Junio C HamanoJan 10, 2019
  31. Jonathan TanJan 23, 2019
  32. Derrick StoleeJan 24, 2019
  33. 5/9 midx: refactor permutation logic and pack sortingDerrick Stolee via GitGitGadget, Jan 9, 2019
  34. Jonathan TanJan 23, 2019
  35. Derrick StoleeJan 24, 2019
  36. Derrick StoleeJan 24, 2019
  37. 8/9 midx: implement midx_repack()Derrick Stolee via GitGitGadget, Jan 9, 2019
  38. Jonathan TanJan 23, 2019
  39. 9/9 multi-pack-index: test expire while adding packsDerrick Stolee via GitGitGadget, Jan 9, 2019
  40. 7/9 multi-pack-index: prepare 'repack' subcommandDerrick Stolee via GitGitGadget, Jan 9, 2019
  41. SZEDER GáborJan 9, 2019
  42. Jonathan TanJan 23, 2019
  43. Derrick StoleeJan 24, 2019
  44. Jonathan TanJan 24, 2019
  45. 2/9 Docs: rearrange subcommands for multi-pack-indexDerrick Stolee via GitGitGadget, Jan 9, 2019
  46. Derrick StoleeJan 17, 2019
  47. Jonathan TanJan 23, 2019
  48. 00/10 Create 'expire' and 'repack' verbs for git-multi-pack-indexDerrick Stolee via GitGitGadget, Jan 24, 2019
  49. 01/10 repack: refactor pack deletion for future useDerrick Stolee via GitGitGadget, Jan 24, 2019
  50. 02/10 Docs: rearrange subcommands for multi-pack-indexDerrick Stolee via GitGitGadget, Jan 24, 2019
  51. 04/10 midx: simplify computation of pack name lengthsDerrick Stolee via GitGitGadget, Jan 24, 2019
  52. 05/10 midx: refactor permutation logic and pack sortingDerrick Stolee via GitGitGadget, Jan 24, 2019
  53. 06/10 multi-pack-index: implement 'expire' subcommandDerrick Stolee via GitGitGadget, Jan 24, 2019
  54. 03/10 multi-pack-index: prepare for 'expire' subcommandDerrick Stolee via GitGitGadget, Jan 24, 2019
  55. 08/10 midx: implement midx_repack()Derrick Stolee via GitGitGadget, Jan 24, 2019
  56. Derrick StoleeJan 26, 2019
  57. Junio C HamanoJan 27, 2019
  58. 09/10 multi-pack-index: test expire while adding packsDerrick Stolee via GitGitGadget, Jan 24, 2019
  59. 10/10 midx: add test that 'expire' respects .keep filesDerrick Stolee via GitGitGadget, Jan 24, 2019
  60. 07/10 multi-pack-index: prepare 'repack' subcommandDerrick Stolee via GitGitGadget, Jan 24, 2019
  61. Josh SteadmonJan 25, 2019
  62. Jonathan TanJan 24, 2019
  63. Josh SteadmonJan 25, 2019
  64. 00/11 Create 'expire' and 'repack' verbs for git-multi-pack-indexDerrick Stolee, Apr 24, 2019
  65. 01/11 repack: refactor pack deletion for future useDerrick Stolee, Apr 24, 2019
  66. 03/11 multi-pack-index: prepare for 'expire' subcommandDerrick Stolee, Apr 24, 2019
  67. 02/11 Docs: rearrange subcommands for multi-pack-indexDerrick Stolee, Apr 24, 2019
  68. 04/11 midx: simplify computation of pack name lengthsDerrick Stolee, Apr 24, 2019
  69. 07/11 multi-pack-index: prepare 'repack' subcommandDerrick Stolee, Apr 24, 2019
  70. 08/11 midx: implement midx_repack()Derrick Stolee, Apr 24, 2019
  71. 09/11 multi-pack-index: test expire while adding packsDerrick Stolee, Apr 24, 2019
  72. 10/11 midx: add test that 'expire' respects .keep filesDerrick Stolee, Apr 24, 2019
  73. 11/11 t5319-multi-pack-index.sh: test batch size zeroDerrick Stolee, Apr 24, 2019
  74. 06/11 multi-pack-index: implement 'expire' subcommandDerrick Stolee, Apr 24, 2019
  75. 05/11 midx: refactor permutation logic and pack sortingDerrick Stolee, Apr 24, 2019
  76. Junio C HamanoApr 25, 2019
  77. Derrick StoleeApr 25, 2019
  78. 00/11 Create 'expire' and 'repack' verbs for git-multi-pack-indexDerrick Stolee, May 14, 2019
  79. 01/11 repack: refactor pack deletion for future useDerrick Stolee, May 14, 2019
  80. 02/11 Docs: rearrange subcommands for multi-pack-indexDerrick Stolee, May 14, 2019
  81. 05/11 midx: refactor permutation logic and pack sortingDerrick Stolee, May 14, 2019
  82. 06/11 multi-pack-index: implement 'expire' subcommandDerrick Stolee, May 14, 2019
  83. 03/11 multi-pack-index: prepare for 'expire' subcommandDerrick Stolee, May 14, 2019
  84. 07/11 multi-pack-index: prepare 'repack' subcommandDerrick Stolee, May 14, 2019
  85. 09/11 multi-pack-index: test expire while adding packsDerrick Stolee, May 14, 2019
  86. 10/11 midx: add test that 'expire' respects .keep filesDerrick Stolee, May 14, 2019
  87. 11/11 t5319-multi-pack-index.sh: test batch size zeroDerrick Stolee, May 14, 2019
  88. 08/11 midx: implement midx_repack()Derrick Stolee, May 14, 2019
  89. 04/11 midx: simplify computation of pack name lengthsDerrick Stolee, May 14, 2019
  90. Derrick StoleeJun 10, 2019
  91. Junio C HamanoJun 10, 2019
  92. Derrick StoleeJun 10, 2019

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.