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

[PATCH v2 3/5] match-trees: use hashcpy to splice trees

From
brian m. carlson <sandals@crustytoothpaste.net>
Date
Jan 15, 2019, 00:39 UTC
Message-ID
<20190115003946.932078-4-sandals@crustytoothpaste.net>
In-Reply-To
<20190115003946.932078-1-sandals@crustytoothpaste.net>

When we splice trees together, we operate in place on the tree buffer. If we're using SHA-1 for the hash algorithm, we may not have a full GIT_MAX_RAWSZ (32) bytes to copy. Consequently, it doesn't logically make sense for us to use a struct object_id to represent this type, since it isn't a complete object.

Represent this value as a unsigned char pointer instead and copy it when necessary.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
---
 match-trees.c | 24 ++++++++++++++++++------
 1 file changed, 18 insertions(+), 6 deletions(-)
diff --git a/match-trees.c b/match-trees.c
index feca48a5fd..c2b7329e09 100644
--- a/match-trees.c
+++ b/match-trees.c
@@ -179,7 +179,7 @@ static int splice_tree(const struct object_id *oid1, const char *prefix,
 	char *buf;
 	unsigned long sz;
 	struct tree_desc desc;
-	struct object_id *rewrite_here;
+	unsigned char *rewrite_here;
 	const struct object_id *rewrite_with;
 	struct object_id subtree;
 	enum object_type type;
@@ -206,9 +206,19 @@ static int splice_tree(const struct object_id *oid1, const char *prefix,
 			if (!S_ISDIR(mode))
 				die("entry %s in tree %s is not a tree", name,
 				    oid_to_hex(oid1));
-			rewrite_here = (struct object_id *)(desc.entry.path +
-							    strlen(desc.entry.path) +
-							    1);
+
+			/*
+			 * We cast here for two reasons:
+			 *
+			 *   - to flip the "char *" (for the path) to "unsigned
+			 *     char *" (for the hash stored after it)
+			 *
+			 *   - to discard the "const"; this is OK because we
+			 *     know it points into our non-const "buf"
+			 */
+			rewrite_here = (unsigned char *)(desc.entry.path +
+							 strlen(desc.entry.path) +
+							 1);
 			break;
 		}
 		update_tree_entry(&desc);
@@ -217,14 +227,16 @@ static int splice_tree(const struct object_id *oid1, const char *prefix,
 		die("entry %.*s not found in tree %s", toplen, prefix,
 		    oid_to_hex(oid1));
 	if (*subpath) {
-		status = splice_tree(rewrite_here, subpath, oid2, &subtree);
+		struct object_id tree_oid;
+		hashcpy(tree_oid.hash, rewrite_here);
+		status = splice_tree(&tree_oid, subpath, oid2, &subtree);
 		if (status)
 			return status;
 		rewrite_with = &subtree;
 	} else {
 		rewrite_with = oid2;
 	}
-	oidcpy(rewrite_here, rewrite_with);
+	hashcpy(rewrite_here, rewrite_with->hash);
 	status = write_object_file(buf, sz, tree_type, result);
 	free(buf);
 	return status;
Previous: brian m. carlsonNext: Junio C Hamano
Message 36 of 41 in “What's cooking in git.git (Jan 2019, #01; Mon, 7)”
  1. Junio C HamanoJan 7, 2019
  2. tg/checkout-no-overlay, was Re: What's cooking in git.git (Jan 2019, #01; Mon, 7)Thomas Gummerer, Jan 8, 2019
  3. Junio C HamanoJan 8, 2019
  4. ag/sequencer-reduce-rewriting-todo Re: What's cooking in git.git (Jan 2019, #01; Mon, 7)Alban Gruin, Jan 8, 2019
  5. sb/more-repo-in-api, was Re: What's cooking in git.git (Jan 2019, #01; Mon, 7)Jonathan Tan, Jan 8, 2019
  6. Junio C HamanoJan 8, 2019
  7. Stefan BellerJan 9, 2019
  8. Martin ÅgrenJan 9, 2019
  9. Martin ÅgrenJan 9, 2019
  10. brian m. carlsonJan 10, 2019
  11. Junio C HamanoJan 10, 2019
  12. Martin ÅgrenJan 10, 2019
  13. 0/5 tree-walk object_id refactorbrian m. carlson, Jan 10, 2019
  14. 1/5 tree-walk: copy object ID before usebrian m. carlson, Jan 10, 2019
  15. 2/5 match-trees: compute buffer offset correctly when splicingbrian m. carlson, Jan 10, 2019
  16. 3/5 match-trees: use hashcpy to splice treesbrian m. carlson, Jan 10, 2019
  17. Jeff KingJan 10, 2019
  18. brian m. carlsonJan 10, 2019
  19. Jeff KingJan 11, 2019
  20. Jeff KingJan 11, 2019
  21. brian m. carlsonJan 14, 2019
  22. Jeff KingJan 14, 2019
  23. 4/5 tree-walk: store object_id in a separate memberbrian m. carlson, Jan 10, 2019
  24. Jeff KingJan 10, 2019
  25. brian m. carlsonJan 10, 2019
  26. 5/5 cache: make oidcpy always copy GIT_MAX_RAWSZ bytesbrian m. carlson, Jan 10, 2019
  27. Jeff KingJan 10, 2019
  28. Jeff KingJan 10, 2019
  29. brian m. carlsonJan 11, 2019
  30. Jeff KingJan 11, 2019
  31. 0/5 tree-walk object_id refactorbrian m. carlson, Jan 15, 2019
  32. 1/5 tree-walk: copy object ID before usebrian m. carlson, Jan 15, 2019
  33. 2/5 match-trees: compute buffer offset correctly when splicingbrian m. carlson, Jan 15, 2019
  34. 4/5 tree-walk: store object_id in a separate memberbrian m. carlson, Jan 15, 2019
  35. 5/5 cache: make oidcpy always copy GIT_MAX_RAWSZ bytesbrian m. carlson, Jan 15, 2019
  36. 3/5 match-trees: use hashcpy to splice treesbrian m. carlson, Jan 15, 2019
  37. Junio C HamanoJan 15, 2019
  38. Jeff KingJan 9, 2019
  39. Junio C HamanoJan 10, 2019
  40. Junio C HamanoJan 10, 2019
  41. Stefan BellerJan 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.