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

[PATCH v3 14/17] midx: use chunk-format read API

From
Derrick Stolee via GitGitGadget <gitgitgadget@gmail.com>
Date
Feb 5, 2021, 14:30 UTC
Message-ID
<6bddd9e63b9b41d196db1bb74c60ad66566f66c7.1612535453.git.gitgitgadget@gmail.com>
In-Reply-To
<pull.848.v3.git.1612535452.gitgitgadget@gmail.com>
From: Derrick Stolee <dstolee@microsoft.com>

Instead of parsing the table of contents directly, use the chunk-format API methods read_table_of_contents() and pair_chunk(). In particular, we can use the return value of pair_chunk() to generate an error when a required chunk is missing.

Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
---
 midx.c                      | 72 ++++++++++++++-----------------------
 t/t5319-multi-pack-index.sh |  6 ++--
 2 files changed, 29 insertions(+), 49 deletions(-)
diff --git a/midx.c b/midx.c
index d9c7411b083b..aee9ed832d52 100644
--- a/midx.c
+++ b/midx.c
@@ -54,6 +54,19 @@ static char *get_midx_filename(const char *object_dir)
 	return xstrfmt("%s/pack/multi-pack-index", object_dir);
 }
 
+static int midx_read_oid_fanout(const unsigned char *chunk_start,
+				size_t chunk_size, void *data)
+{
+	struct multi_pack_index *m = data;
+	m->chunk_oid_fanout = (uint32_t *)chunk_start;
+
+	if (chunk_size != 4 * 256) {
+		error(_("multi-pack-index OID fanout is of the wrong size"));
+		return 1;
+	}
+	return 0;
+}
+
 struct multi_pack_index *load_multi_pack_index(const char *object_dir, int local)
 {
 	struct multi_pack_index *m = NULL;
@@ -65,6 +78,7 @@ struct multi_pack_index *load_multi_pack_index(const char *object_dir, int local
 	char *midx_name = get_midx_filename(object_dir);
 	uint32_t i;
 	const char *cur_pack_name;
+	struct chunkfile *cf = NULL;
 
 	fd = git_open(midx_name);
 
@@ -114,58 +128,23 @@ struct multi_pack_index *load_multi_pack_index(const char *object_dir, int local
 
 	m->num_packs = get_be32(m->data + MIDX_BYTE_NUM_PACKS);
 
-	for (i = 0; i < m->num_chunks; i++) {
-		uint32_t chunk_id = get_be32(m->data + MIDX_HEADER_SIZE +
-					     MIDX_CHUNKLOOKUP_WIDTH * i);
-		uint64_t chunk_offset = get_be64(m->data + MIDX_HEADER_SIZE + 4 +
-						 MIDX_CHUNKLOOKUP_WIDTH * i);
-
-		if (chunk_offset >= m->data_len)
-			die(_("invalid chunk offset (too large)"));
-
-		switch (chunk_id) {
-			case MIDX_CHUNKID_PACKNAMES:
-				m->chunk_pack_names = m->data + chunk_offset;
-				break;
-
-			case MIDX_CHUNKID_OIDFANOUT:
-				m->chunk_oid_fanout = (uint32_t *)(m->data + chunk_offset);
-				break;
-
-			case MIDX_CHUNKID_OIDLOOKUP:
-				m->chunk_oid_lookup = m->data + chunk_offset;
-				break;
-
-			case MIDX_CHUNKID_OBJECTOFFSETS:
-				m->chunk_object_offsets = m->data + chunk_offset;
-				break;
-
-			case MIDX_CHUNKID_LARGEOFFSETS:
-				m->chunk_large_offsets = m->data + chunk_offset;
-				break;
-
-			case 0:
-				die(_("terminating multi-pack-index chunk id appears earlier than expected"));
-				break;
-
-			default:
-				/*
-				 * Do nothing on unrecognized chunks, allowing future
-				 * extensions to add optional chunks.
-				 */
-				break;
-		}
-	}
+	cf = init_chunkfile(NULL);
 
-	if (!m->chunk_pack_names)
+	if (read_table_of_contents(cf, m->data, midx_size,
+				   MIDX_HEADER_SIZE, m->num_chunks))
+		goto cleanup_fail;
+
+	if (pair_chunk(cf, MIDX_CHUNKID_PACKNAMES, &m->chunk_pack_names) == CHUNK_NOT_FOUND)
 		die(_("multi-pack-index missing required pack-name chunk"));
-	if (!m->chunk_oid_fanout)
+	if (read_chunk(cf, MIDX_CHUNKID_OIDFANOUT, midx_read_oid_fanout, m) == CHUNK_NOT_FOUND)
 		die(_("multi-pack-index missing required OID fanout chunk"));
-	if (!m->chunk_oid_lookup)
+	if (pair_chunk(cf, MIDX_CHUNKID_OIDLOOKUP, &m->chunk_oid_lookup) == CHUNK_NOT_FOUND)
 		die(_("multi-pack-index missing required OID lookup chunk"));
-	if (!m->chunk_object_offsets)
+	if (pair_chunk(cf, MIDX_CHUNKID_OBJECTOFFSETS, &m->chunk_object_offsets) == CHUNK_NOT_FOUND)
 		die(_("multi-pack-index missing required object offsets chunk"));
 
+	pair_chunk(cf, MIDX_CHUNKID_LARGEOFFSETS, &m->chunk_large_offsets);
+
 	m->num_objects = ntohl(m->chunk_oid_fanout[255]);
 
 	m->pack_names = xcalloc(m->num_packs, sizeof(*m->pack_names));
@@ -191,6 +170,7 @@ struct multi_pack_index *load_multi_pack_index(const char *object_dir, int local
 cleanup_fail:
 	free(m);
 	free(midx_name);
+	free(cf);
 	if (midx_map)
 		munmap(midx_map, midx_size);
 	if (0 <= fd)
diff --git a/t/t5319-multi-pack-index.sh b/t/t5319-multi-pack-index.sh
index 297de502a94f..ad4e878b65b8 100755
--- a/t/t5319-multi-pack-index.sh
+++ b/t/t5319-multi-pack-index.sh
@@ -314,12 +314,12 @@ test_expect_success 'verify bad OID version' '
 
 test_expect_success 'verify truncated chunk count' '
 	corrupt_midx_and_verify $MIDX_BYTE_CHUNK_COUNT "\01" $objdir \
-		"missing required"
+		"final chunk has non-zero id"
 '
 
 test_expect_success 'verify extended chunk count' '
 	corrupt_midx_and_verify $MIDX_BYTE_CHUNK_COUNT "\07" $objdir \
-		"terminating multi-pack-index chunk id appears earlier than expected"
+		"terminating chunk id appears earlier than expected"
 '
 
 test_expect_success 'verify missing required chunk' '
@@ -329,7 +329,7 @@ test_expect_success 'verify missing required chunk' '
 
 test_expect_success 'verify invalid chunk offset' '
 	corrupt_midx_and_verify $MIDX_BYTE_CHUNK_OFFSET "\01" $objdir \
-		"invalid chunk offset (too large)"
+		"improper chunk offset(s)"
 '
 
 test_expect_success 'verify packnames out of order' '
-- 
gitgitgadget
Previous: Derrick Stolee via GitGitGadgetNext: Derrick Stolee via GitGitGadget
Message 85 of 120 in “Refactor chunk-format into an API”
  1. 00/17 Refactor chunk-format into an APIDerrick Stolee via GitGitGadget, Jan 26, 2021
  2. 02/17 chunk-format: create chunk format write APIDerrick Stolee via GitGitGadget, Jan 26, 2021
  3. Taylor BlauJan 27, 2021
  4. Derrick StoleeJan 27, 2021
  5. 01/17 commit-graph: anonymize data in chunk_write_fnDerrick Stolee via GitGitGadget, Jan 26, 2021
  6. Chris TorekJan 27, 2021
  7. Taylor BlauJan 27, 2021
  8. 05/17 midx: use context in write_midx_pack_names()Derrick Stolee via GitGitGadget, Jan 26, 2021
  9. 06/17 midx: add entries to write_midx_contextDerrick Stolee via GitGitGadget, Jan 26, 2021
  10. 04/17 midx: rename pack_info to write_midx_contextDerrick Stolee via GitGitGadget, Jan 26, 2021
  11. Taylor BlauJan 27, 2021
  12. 03/17 commit-graph: use chunk-format write APIDerrick Stolee via GitGitGadget, Jan 26, 2021
  13. Taylor BlauJan 27, 2021
  14. 14/17 midx: use chunk-format read APIDerrick Stolee via GitGitGadget, Jan 26, 2021
  15. Taylor BlauJan 27, 2021
  16. Derrick StoleeJan 27, 2021
  17. 17/17 chunk-format: add technical docsDerrick Stolee via GitGitGadget, Jan 26, 2021
  18. 11/17 midx: use chunk-format API in write_midx_internal()Derrick Stolee via GitGitGadget, Jan 26, 2021
  19. 16/17 chunk-format: restore duplicate chunk checksDerrick Stolee via GitGitGadget, Jan 26, 2021
  20. 15/17 midx: use 64-bit multiplication for chunk sizesDerrick Stolee via GitGitGadget, Jan 26, 2021
  21. 07/17 midx: add pack_perm to write_midx_contextDerrick Stolee via GitGitGadget, Jan 26, 2021
  22. 10/17 midx: drop chunk progress during writeDerrick Stolee via GitGitGadget, Jan 26, 2021
  23. 09/17 midx: return success/failure in chunk write methodsDerrick Stolee via GitGitGadget, Jan 26, 2021
  24. 08/17 midx: add num_large_offsets to write_midx_contextDerrick Stolee via GitGitGadget, Jan 26, 2021
  25. 13/17 commit-graph: use chunk-format read APIDerrick Stolee via GitGitGadget, Jan 26, 2021
  26. 12/17 chunk-format: create read chunk APIDerrick Stolee via GitGitGadget, Jan 26, 2021
  27. Taylor BlauJan 27, 2021
  28. Taylor BlauJan 27, 2021
  29. Junio C HamanoJan 26, 2021
  30. 00/17 Refactor chunk-format into an APIDerrick Stolee via GitGitGadget, Jan 27, 2021
  31. 05/17 midx: use context in write_midx_pack_names()Derrick Stolee via GitGitGadget, Jan 27, 2021
  32. 03/17 commit-graph: use chunk-format write APIDerrick Stolee via GitGitGadget, Jan 27, 2021
  33. 07/17 midx: add pack_perm to write_midx_contextDerrick Stolee via GitGitGadget, Jan 27, 2021
  34. 02/17 chunk-format: create chunk format write APIDerrick Stolee via GitGitGadget, Jan 27, 2021
  35. Junio C HamanoFeb 4, 2021
  36. Junio C HamanoFeb 4, 2021
  37. Derrick StoleeFeb 5, 2021
  38. Junio C HamanoFeb 5, 2021
  39. 15/17 midx: use 64-bit multiplication for chunk sizesDerrick Stolee via GitGitGadget, Jan 27, 2021
  40. Junio C HamanoFeb 5, 2021
  41. Chris TorekFeb 5, 2021
  42. Junio C HamanoFeb 5, 2021
  43. Chris TorekFeb 6, 2021
  44. Derrick StoleeFeb 5, 2021
  45. Junio C HamanoFeb 5, 2021
  46. SZEDER GáborFeb 7, 2021
  47. Junio C HamanoFeb 8, 2021
  48. 17/17 chunk-format: add technical docsDerrick Stolee via GitGitGadget, Jan 27, 2021
  49. Junio C HamanoFeb 5, 2021
  50. 14/17 midx: use chunk-format read APIDerrick Stolee via GitGitGadget, Jan 27, 2021
  51. 13/17 commit-graph: use chunk-format read APIDerrick Stolee via GitGitGadget, Jan 27, 2021
  52. 16/17 chunk-format: restore duplicate chunk checksDerrick Stolee via GitGitGadget, Jan 27, 2021
  53. Junio C HamanoFeb 5, 2021
  54. Derrick StoleeFeb 5, 2021
  55. 11/17 midx: use chunk-format API in write_midx_internal()Derrick Stolee via GitGitGadget, Jan 27, 2021
  56. 12/17 chunk-format: create read chunk APIDerrick Stolee via GitGitGadget, Jan 27, 2021
  57. Junio C HamanoFeb 4, 2021
  58. Derrick StoleeFeb 5, 2021
  59. Junio C HamanoFeb 5, 2021
  60. Junio C HamanoFeb 8, 2021
  61. Derrick StoleeFeb 9, 2021
  62. Junio C HamanoFeb 9, 2021
  63. 09/17 midx: return success/failure in chunk write methodsDerrick Stolee via GitGitGadget, Jan 27, 2021
  64. Junio C HamanoFeb 4, 2021
  65. Derrick StoleeFeb 5, 2021
  66. 10/17 midx: drop chunk progress during writeDerrick Stolee via GitGitGadget, Jan 27, 2021
  67. 06/17 midx: add entries to write_midx_contextDerrick Stolee via GitGitGadget, Jan 27, 2021
  68. 08/17 midx: add num_large_offsets to write_midx_contextDerrick Stolee via GitGitGadget, Jan 27, 2021
  69. 04/17 midx: rename pack_info to write_midx_contextDerrick Stolee via GitGitGadget, Jan 27, 2021
  70. 01/17 commit-graph: anonymize data in chunk_write_fnDerrick Stolee via GitGitGadget, Jan 27, 2021
  71. Taylor BlauJan 27, 2021
  72. Junio C HamanoFeb 5, 2021
  73. Derrick StoleeFeb 5, 2021
  74. 00/17 Refactor chunk-format into an APIDerrick Stolee via GitGitGadget, Feb 5, 2021
  75. 10/17 midx: drop chunk progress during writeDerrick Stolee via GitGitGadget, Feb 5, 2021
  76. 17/17 chunk-format: add technical docsDerrick Stolee via GitGitGadget, Feb 5, 2021
  77. 09/17 midx: return success/failure in chunk write methodsDerrick Stolee via GitGitGadget, Feb 5, 2021
  78. 15/17 midx: use 64-bit multiplication for chunk sizesDerrick Stolee via GitGitGadget, Feb 5, 2021
  79. 02/17 chunk-format: create chunk format write APIDerrick Stolee via GitGitGadget, Feb 5, 2021
  80. SZEDER GáborFeb 7, 2021
  81. Derrick StoleeFeb 8, 2021
  82. SZEDER GáborFeb 11, 2021
  83. 08/17 midx: add num_large_offsets to write_midx_contextDerrick Stolee via GitGitGadget, Feb 5, 2021
  84. 05/17 midx: use context in write_midx_pack_names()Derrick Stolee via GitGitGadget, Feb 5, 2021
  85. 14/17 midx: use chunk-format read APIDerrick Stolee via GitGitGadget, Feb 5, 2021
  86. 07/17 midx: add pack_perm to write_midx_contextDerrick Stolee via GitGitGadget, Feb 5, 2021
  87. 06/17 midx: add entries to write_midx_contextDerrick Stolee via GitGitGadget, Feb 5, 2021
  88. 16/17 chunk-format: restore duplicate chunk checksDerrick Stolee via GitGitGadget, Feb 5, 2021
  89. 13/17 commit-graph: use chunk-format read APIDerrick Stolee via GitGitGadget, Feb 5, 2021
  90. 01/17 commit-graph: anonymize data in chunk_write_fnDerrick Stolee via GitGitGadget, Feb 5, 2021
  91. 03/17 commit-graph: use chunk-format write APIDerrick Stolee via GitGitGadget, Feb 5, 2021
  92. 04/17 midx: rename pack_info to write_midx_contextDerrick Stolee via GitGitGadget, Feb 5, 2021
  93. 11/17 midx: use chunk-format API in write_midx_internal()Derrick Stolee via GitGitGadget, Feb 5, 2021
  94. 12/17 chunk-format: create read chunk APIDerrick Stolee via GitGitGadget, Feb 5, 2021
  95. SZEDER GáborFeb 7, 2021
  96. Derrick StoleeFeb 8, 2021
  97. 00/17 Refactor chunk-format into an APIDerrick Stolee via GitGitGadget, Feb 18, 2021
  98. 07/17 midx: add pack_perm to write_midx_contextDerrick Stolee via GitGitGadget, Feb 18, 2021
  99. 05/17 midx: use context in write_midx_pack_names()Derrick Stolee via GitGitGadget, Feb 18, 2021
  100. 03/17 commit-graph: use chunk-format write APIDerrick Stolee via GitGitGadget, Feb 18, 2021
  101. SZEDER GáborFeb 24, 2021
  102. Taylor BlauFeb 24, 2021
  103. Derrick StoleeFeb 24, 2021
  104. Junio C HamanoFeb 24, 2021
  105. 04/17 midx: rename pack_info to write_midx_contextDerrick Stolee via GitGitGadget, Feb 18, 2021
  106. 06/17 midx: add entries to write_midx_contextDerrick Stolee via GitGitGadget, Feb 18, 2021
  107. 02/17 chunk-format: create chunk format write APIDerrick Stolee via GitGitGadget, Feb 18, 2021
  108. 01/17 commit-graph: anonymize data in chunk_write_fnDerrick Stolee via GitGitGadget, Feb 18, 2021
  109. 08/17 midx: add num_large_offsets to write_midx_contextDerrick Stolee via GitGitGadget, Feb 18, 2021
  110. 09/17 midx: return success/failure in chunk write methodsDerrick Stolee via GitGitGadget, Feb 18, 2021
  111. 10/17 midx: drop chunk progress during writeDerrick Stolee via GitGitGadget, Feb 18, 2021
  112. 11/17 midx: use chunk-format API in write_midx_internal()Derrick Stolee via GitGitGadget, Feb 18, 2021
  113. 12/17 chunk-format: create read chunk APIDerrick Stolee via GitGitGadget, Feb 18, 2021
  114. 17/17 chunk-format: add technical docsDerrick Stolee via GitGitGadget, Feb 18, 2021
  115. Junio C HamanoFeb 18, 2021
  116. Derrick StoleeFeb 19, 2021
  117. 15/17 midx: use 64-bit multiplication for chunk sizesDerrick Stolee via GitGitGadget, Feb 18, 2021
  118. 14/17 midx: use chunk-format read APIDerrick Stolee via GitGitGadget, Feb 18, 2021
  119. 13/17 commit-graph: use chunk-format read APIDerrick Stolee via GitGitGadget, Feb 18, 2021
  120. 16/17 chunk-format: restore duplicate chunk checksDerrick Stolee via GitGitGadget, Feb 18, 2021

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.