From 518120e3487a00148f8001454f1e76e484442a22 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Santi=20B=C3=A9jar?= <sbejar@gmail.com>
Date: Mon, 25 Feb 2008 10:43:33 +0100
Subject: [PATCH 01/53] git-describe: --long shows the object name even for a
 tagged commit
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This is useful when you want to see parts of the commit object name
in "describe" output, even when the commit in question happens to be
a tagged version.  Instead of just emitting the tag name, it will
describe such a commit as v1.2-0-deadbeef (0th commit since tag v1.2
that points at object deadbeef....).

Signed-off-by: Santi Béjar <sbejar@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 Documentation/git-describe.txt |  9 +++++++++
 builtin-describe.c             | 11 ++++++++++-
 t/t6120-describe.sh            |  2 ++
 3 files changed, 21 insertions(+), 1 deletion(-)

diff --git a/Documentation/git-describe.txt b/Documentation/git-describe.txt
index 1c3dfb40c6..270b808755 100644
--- a/Documentation/git-describe.txt
+++ b/Documentation/git-describe.txt
@@ -51,6 +51,15 @@ OPTIONS
 	being employed to standard error.  The tag name will still
 	be printed to standard out.
 
+--long::
+	Always output the long format (the tag, the number of commits
+	and the abbreviated commit name) even when it matches a tag.
+	This is useful when you want to see parts of the commit object name
+	in "describe" output, even when the commit in question happens to be
+	a tagged version.  Instead of just emitting the tag name, it will
+	describe such a commit as v1.2-0-deadbeef (0th commit since tag v1.2
+	that points at object deadbeef....).
+
 --match <pattern>::
 	Only consider tags matching the given pattern (can be used to avoid
 	leaking private tags made from the repository).
diff --git a/builtin-describe.c b/builtin-describe.c
index 3428483134..3fd2e7371f 100644
--- a/builtin-describe.c
+++ b/builtin-describe.c
@@ -17,6 +17,7 @@ static const char * const describe_usage[] = {
 static int debug;	/* Display lots of verbose info */
 static int all;	/* Default to annotated tags only */
 static int tags;	/* But allow any tags if --tags is specified */
+static int longformat;
 static int abbrev = DEFAULT_ABBREV;
 static int max_candidates = 10;
 const char *pattern = NULL;
@@ -155,7 +156,11 @@ static void describe(const char *arg, int last_one)
 
 	n = cmit->util;
 	if (n) {
-		printf("%s\n", n->path);
+		if (!longformat)
+			printf("%s\n", n->path);
+		else
+			printf("%s-0-g%s\n", n->path,
+				find_unique_abbrev(cmit->object.sha1, abbrev));
 		return;
 	}
 
@@ -254,6 +259,7 @@ int cmd_describe(int argc, const char **argv, const char *prefix)
 		OPT_BOOLEAN(0, "debug",      &debug, "debug search strategy on stderr"),
 		OPT_BOOLEAN(0, "all",        &all, "use any ref in .git/refs"),
 		OPT_BOOLEAN(0, "tags",       &tags, "use any tag in .git/refs/tags"),
+		OPT_BOOLEAN(0, "long",       &longformat, "always use long format"),
 		OPT__ABBREV(&abbrev),
 		OPT_INTEGER(0, "candidates", &max_candidates,
 			    "consider <n> most recent tags (default: 10)"),
@@ -270,6 +276,9 @@ int cmd_describe(int argc, const char **argv, const char *prefix)
 
 	save_commit_buffer = 0;
 
+	if (longformat && abbrev == 0)
+		die("--long is incompatible with --abbrev=0");
+
 	if (contains) {
 		const char **args = xmalloc((6 + argc) * sizeof(char*));
 		int i = 0;
diff --git a/t/t6120-describe.sh b/t/t6120-describe.sh
index ae8ee11183..a7557bdc79 100755
--- a/t/t6120-describe.sh
+++ b/t/t6120-describe.sh
@@ -94,4 +94,6 @@ check_describe D-* --tags HEAD^^
 check_describe A-* --tags HEAD^^2
 check_describe B --tags HEAD^^2^
 
+check_describe B-0-* --long HEAD^^2^
+
 test_done

From 355885d531568e144ce5c9a2f65792a3aa8b5528 Mon Sep 17 00:00:00 2001
From: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Date: Mon, 25 Feb 2008 22:46:04 +0100
Subject: [PATCH 02/53] add generic, type aware object chain walker

The requirements are:
* it may not crash on NULL pointers
* a callback function is needed, as index-pack/unpack-objects
  need to do different things
* the type information is needed to check the expected <-> real type
  and print better error messages

Signed-off-by: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 Makefile |  4 +--
 cache.h  |  1 +
 fsck.c   | 91 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 fsck.h   | 23 ++++++++++++++
 4 files changed, 117 insertions(+), 2 deletions(-)
 create mode 100644 fsck.c
 create mode 100644 fsck.h

diff --git a/Makefile b/Makefile
index 92341c4bbe..33c367dfb1 100644
--- a/Makefile
+++ b/Makefile
@@ -297,7 +297,7 @@ LIB_H = \
 	run-command.h strbuf.h tag.h tree.h git-compat-util.h revision.h \
 	tree-walk.h log-tree.h dir.h path-list.h unpack-trees.h builtin.h \
 	utf8.h reflog-walk.h patch-ids.h attr.h decorate.h progress.h \
-	mailmap.h remote.h parse-options.h transport.h diffcore.h hash.h
+	mailmap.h remote.h parse-options.h transport.h diffcore.h hash.h fsck.h
 
 DIFF_OBJS = \
 	diff.o diff-lib.o diffcore-break.o diffcore-order.o \
@@ -320,7 +320,7 @@ LIB_OBJS = \
 	alloc.o merge-file.o path-list.o help.o unpack-trees.o $(DIFF_OBJS) \
 	color.o wt-status.o archive-zip.o archive-tar.o shallow.o utf8.o \
 	convert.o attr.o decorate.o progress.o mailmap.o symlinks.o remote.o \
-	transport.o bundle.o walker.o parse-options.o ws.o archive.o
+	transport.o bundle.o walker.o parse-options.o ws.o archive.o fsck.o
 
 BUILTIN_OBJS = \
 	builtin-add.o \
diff --git a/cache.h b/cache.h
index 43ba6a3ba5..d3b898708b 100644
--- a/cache.h
+++ b/cache.h
@@ -189,6 +189,7 @@ enum object_type {
 	/* 5 for future expansion */
 	OBJ_OFS_DELTA = 6,
 	OBJ_REF_DELTA = 7,
+	OBJ_ANY,
 	OBJ_MAX,
 };
 
diff --git a/fsck.c b/fsck.c
new file mode 100644
index 0000000000..98fb41af68
--- /dev/null
+++ b/fsck.c
@@ -0,0 +1,91 @@
+#include "cache.h"
+#include "object.h"
+#include "blob.h"
+#include "tree.h"
+#include "tree-walk.h"
+#include "commit.h"
+#include "tag.h"
+#include "fsck.h"
+
+static int fsck_walk_tree(struct tree *tree, fsck_walk_func walk, void *data)
+{
+	struct tree_desc desc;
+	struct name_entry entry;
+	int res = 0;
+
+	if (parse_tree(tree))
+		return -1;
+
+	init_tree_desc(&desc, tree->buffer, tree->size);
+	while (tree_entry(&desc, &entry)) {
+		int result;
+
+		if (S_ISGITLINK(entry.mode))
+			continue;
+		if (S_ISDIR(entry.mode))
+			result = walk(&lookup_tree(entry.sha1)->object, OBJ_TREE, data);
+		else if (S_ISREG(entry.mode) || S_ISLNK(entry.mode))
+			result = walk(&lookup_blob(entry.sha1)->object, OBJ_BLOB, data);
+		else {
+			result = error("in tree %s: entry %s has bad mode %.6o\n",
+					sha1_to_hex(tree->object.sha1), entry.path, entry.mode);
+		}
+		if (result < 0)
+			return result;
+		if (!res)
+			res = result;
+	}
+	return res;
+}
+
+static int fsck_walk_commit(struct commit *commit, fsck_walk_func walk, void *data)
+{
+	struct commit_list *parents;
+	int res;
+	int result;
+
+	if (parse_commit(commit))
+		return -1;
+
+	result = walk((struct object *)commit->tree, OBJ_TREE, data);
+	if (result < 0)
+		return result;
+	res = result;
+
+	parents = commit->parents;
+	while (parents) {
+		result = walk((struct object *)parents->item, OBJ_COMMIT, data);
+		if (result < 0)
+			return result;
+		if (!res)
+			res = result;
+		parents = parents->next;
+	}
+	return res;
+}
+
+static int fsck_walk_tag(struct tag *tag, fsck_walk_func walk, void *data)
+{
+	if (parse_tag(tag))
+		return -1;
+	return walk(tag->tagged, OBJ_ANY, data);
+}
+
+int fsck_walk(struct object *obj, fsck_walk_func walk, void *data)
+{
+	if (!obj)
+		return -1;
+	switch (obj->type) {
+	case OBJ_BLOB:
+		return 0;
+	case OBJ_TREE:
+		return fsck_walk_tree((struct tree *)obj, walk, data);
+	case OBJ_COMMIT:
+		return fsck_walk_commit((struct commit *)obj, walk, data);
+	case OBJ_TAG:
+		return fsck_walk_tag((struct tag *)obj, walk, data);
+	default:
+		error("Unknown object type for %s", sha1_to_hex(obj->sha1));
+		return -1;
+	}
+}
diff --git a/fsck.h b/fsck.h
new file mode 100644
index 0000000000..ba5514a818
--- /dev/null
+++ b/fsck.h
@@ -0,0 +1,23 @@
+#ifndef GIT_FSCK_H
+#define GIT_FSCK_H
+
+/*
+ * callback function for fsck_walk
+ * type is the expected type of the object or OBJ_ANY
+ * the return value is:
+ *     0	everything OK
+ *     <0	error signaled and abort
+ *     >0	error signaled and do not abort
+ */
+typedef int (*fsck_walk_func)(struct object *obj, int type, void *data);
+
+/* descend in all linked child objects
+ * the return value is:
+ *    -1	error in processing the object
+ *    <0	return value of the callback, which lead to an abort
+ *    >0	return value of the first sigaled error >0 (in the case of no other errors)
+ *    0		everything OK
+ */
+int fsck_walk(struct object *obj, fsck_walk_func walk, void *data);
+
+#endif

From 271b8d25b25e49b367087440e093e755e5f35aa9 Mon Sep 17 00:00:00 2001
From: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Date: Mon, 25 Feb 2008 22:46:05 +0100
Subject: [PATCH 03/53] builtin-fsck: move away from object-refs to fsck_walk

Signed-off-by: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 builtin-fsck.c | 99 ++++++++++++++++++++++++++++++++++++--------------
 1 file changed, 71 insertions(+), 28 deletions(-)

diff --git a/builtin-fsck.c b/builtin-fsck.c
index 2a6e94deaf..7321ab236b 100644
--- a/builtin-fsck.c
+++ b/builtin-fsck.c
@@ -8,6 +8,7 @@
 #include "pack.h"
 #include "cache-tree.h"
 #include "tree-walk.h"
+#include "fsck.h"
 #include "parse-options.h"
 
 #define REACHABLE 0x0001
@@ -63,13 +64,73 @@ static int objwarning(struct object *obj, const char *err, ...)
 	return -1;
 }
 
+static int mark_object(struct object *obj, int type, void *data)
+{
+	struct tree *tree = NULL;
+	struct object *parent = data;
+	int result;
+
+	if (!obj) {
+		printf("broken link from %7s %s\n",
+			   typename(parent->type), sha1_to_hex(parent->sha1));
+		printf("broken link from %7s %s\n",
+			   (type == OBJ_ANY ? "unknown" : typename(type)), "unknown");
+		errors_found |= ERROR_REACHABLE;
+		return 1;
+	}
+
+	if (type != OBJ_ANY && obj->type != type)
+		objerror(parent, "wrong object type in link");
+
+	if (obj->flags & REACHABLE)
+		return 0;
+	obj->flags |= REACHABLE;
+	if (!obj->parsed) {
+		if (parent && !has_sha1_file(obj->sha1)) {
+			printf("broken link from %7s %s\n",
+				 typename(parent->type), sha1_to_hex(parent->sha1));
+			printf("              to %7s %s\n",
+				 typename(obj->type), sha1_to_hex(obj->sha1));
+			errors_found |= ERROR_REACHABLE;
+		}
+		return 1;
+	}
+
+	if (obj->type == OBJ_TREE) {
+		obj->parsed = 0;
+		tree = (struct tree *)obj;
+		if (parse_tree(tree) < 0)
+			return 1; /* error already displayed */
+	}
+	result = fsck_walk(obj, mark_object, obj);
+	if (tree) {
+		free(tree->buffer);
+		tree->buffer = NULL;
+	}
+	if (result < 0)
+		result = 1;
+
+	return result;
+}
+
+static void mark_object_reachable(struct object *obj)
+{
+	mark_object(obj, OBJ_ANY, 0);
+}
+
+static int mark_used(struct object *obj, int type, void *data)
+{
+	if (!obj)
+		return 1;
+	obj->used = 1;
+	return 0;
+}
+
 /*
  * Check a single reachable object
  */
 static void check_reachable_object(struct object *obj)
 {
-	const struct object_refs *refs;
-
 	/*
 	 * We obviously want the object to be parsed,
 	 * except if it was in a pack-file and we didn't
@@ -82,25 +143,6 @@ static void check_reachable_object(struct object *obj)
 		errors_found |= ERROR_REACHABLE;
 		return;
 	}
-
-	/*
-	 * Check that everything that we try to reference is also good.
-	 */
-	refs = lookup_object_refs(obj);
-	if (refs) {
-		unsigned j;
-		for (j = 0; j < refs->count; j++) {
-			struct object *ref = refs->ref[j];
-			if (ref->parsed ||
-			    (has_sha1_file(ref->sha1)))
-				continue;
-			printf("broken link from %7s %s\n",
-			       typename(obj->type), sha1_to_hex(obj->sha1));
-			printf("              to %7s %s\n",
-			       typename(ref->type), sha1_to_hex(ref->sha1));
-			errors_found |= ERROR_REACHABLE;
-		}
-	}
 }
 
 /*
@@ -414,6 +456,8 @@ static int fsck_sha1(const unsigned char *sha1)
 	if (obj->flags & SEEN)
 		return 0;
 	obj->flags |= SEEN;
+	if (fsck_walk(obj, mark_used, 0))
+		objerror(obj, "broken links");
 	if (obj->type == OBJ_BLOB)
 		return 0;
 	if (obj->type == OBJ_TREE)
@@ -538,13 +582,13 @@ static int fsck_handle_reflog_ent(unsigned char *osha1, unsigned char *nsha1,
 		obj = lookup_object(osha1);
 		if (obj) {
 			obj->used = 1;
-			mark_reachable(obj, REACHABLE);
+			mark_object_reachable(obj);
 		}
 	}
 	obj = lookup_object(nsha1);
 	if (obj) {
 		obj->used = 1;
-		mark_reachable(obj, REACHABLE);
+		mark_object_reachable(obj);
 	}
 	return 0;
 }
@@ -574,7 +618,7 @@ static int fsck_handle_ref(const char *refname, const unsigned char *sha1, int f
 		error("%s: not a commit", refname);
 	default_refs++;
 	obj->used = 1;
-	mark_reachable(obj, REACHABLE);
+	mark_object_reachable(obj);
 
 	return 0;
 }
@@ -660,7 +704,7 @@ static int fsck_cache_tree(struct cache_tree *it)
 			      sha1_to_hex(it->sha1));
 			return 1;
 		}
-		mark_reachable(obj, REACHABLE);
+		mark_object_reachable(obj);
 		obj->used = 1;
 		if (obj->type != OBJ_TREE)
 			err |= objerror(obj, "non-tree in cache-tree");
@@ -693,7 +737,6 @@ int cmd_fsck(int argc, const char **argv, const char *prefix)
 {
 	int i, heads;
 
-	track_object_refs = 1;
 	errors_found = 0;
 
 	argc = parse_options(argc, argv, fsck_opts, fsck_usage, 0);
@@ -741,7 +784,7 @@ int cmd_fsck(int argc, const char **argv, const char *prefix)
 				continue;
 
 			obj->used = 1;
-			mark_reachable(obj, REACHABLE);
+			mark_object_reachable(obj);
 			heads++;
 			continue;
 		}
@@ -773,7 +816,7 @@ int cmd_fsck(int argc, const char **argv, const char *prefix)
 				continue;
 			obj = &blob->object;
 			obj->used = 1;
-			mark_reachable(obj, REACHABLE);
+			mark_object_reachable(obj);
 		}
 		if (active_cache_tree)
 			fsck_cache_tree(active_cache_tree);

From 7914053ba9901be1f1530f46e8e2e6ee6f4ae5b1 Mon Sep 17 00:00:00 2001
From: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Date: Mon, 25 Feb 2008 22:46:06 +0100
Subject: [PATCH 04/53] Remove unused object-ref code

Signed-off-by: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 Makefile               |  2 +-
 builtin-fetch-pack.c   |  1 -
 builtin-pack-objects.c |  1 -
 builtin-rev-list.c     |  1 -
 commit.c               | 11 ------
 object-refs.c          | 87 ------------------------------------------
 object.h               |  8 ----
 tag.c                  |  6 ---
 tree.c                 | 48 -----------------------
 upload-pack.c          |  1 -
 walker.c               |  1 -
 11 files changed, 1 insertion(+), 166 deletions(-)
 delete mode 100644 object-refs.c

diff --git a/Makefile b/Makefile
index 33c367dfb1..8ff0c772f1 100644
--- a/Makefile
+++ b/Makefile
@@ -312,7 +312,7 @@ LIB_OBJS = \
 	patch-ids.o \
 	object.o pack-check.o pack-write.o patch-delta.o path.o pkt-line.o \
 	sideband.o reachable.o reflog-walk.o \
-	quote.o read-cache.o refs.o run-command.o dir.o object-refs.o \
+	quote.o read-cache.o refs.o run-command.o dir.o \
 	server-info.o setup.o sha1_file.o sha1_name.o strbuf.o \
 	tag.o tree.o usage.o config.o environment.o ctype.o copy.o \
 	revision.o pager.o tree-walk.o xdiff-interface.o \
diff --git a/builtin-fetch-pack.c b/builtin-fetch-pack.c
index e68e01592d..410414d91f 100644
--- a/builtin-fetch-pack.c
+++ b/builtin-fetch-pack.c
@@ -385,7 +385,6 @@ static int everything_local(struct ref **refs, int nr_match, char **match)
 	int retval;
 	unsigned long cutoff = 0;
 
-	track_object_refs = 0;
 	save_commit_buffer = 0;
 
 	for (ref = *refs; ref; ref = ref->next) {
diff --git a/builtin-pack-objects.c b/builtin-pack-objects.c
index d3efeff03f..6f8f388bdf 100644
--- a/builtin-pack-objects.c
+++ b/builtin-pack-objects.c
@@ -2009,7 +2009,6 @@ static void get_object_list(int ac, const char **av)
 
 	init_revisions(&revs, NULL);
 	save_commit_buffer = 0;
-	track_object_refs = 0;
 	setup_revisions(ac, av, &revs, NULL);
 
 	while (fgets(line, sizeof(line), stdin) != NULL) {
diff --git a/builtin-rev-list.c b/builtin-rev-list.c
index de80158fd4..14e86ceabc 100644
--- a/builtin-rev-list.c
+++ b/builtin-rev-list.c
@@ -605,7 +605,6 @@ int cmd_rev_list(int argc, const char **argv, const char *prefix)
 		usage(rev_list_usage);
 
 	save_commit_buffer = revs.verbose_header || revs.grep_filter;
-	track_object_refs = 0;
 	if (bisect_list)
 		revs.limited = 1;
 
diff --git a/commit.c b/commit.c
index 22ce776863..6684c4e731 100644
--- a/commit.c
+++ b/commit.c
@@ -290,17 +290,6 @@ int parse_commit_buffer(struct commit *item, void *buffer, unsigned long size)
 	}
 	item->date = parse_commit_date(bufptr, tail);
 
-	if (track_object_refs) {
-		unsigned i = 0;
-		struct commit_list *p;
-		struct object_refs *refs = alloc_object_refs(n_refs);
-		if (item->tree)
-			refs->ref[i++] = &item->tree->object;
-		for (p = item->parents; p; p = p->next)
-			refs->ref[i++] = &p->item->object;
-		set_object_refs(&item->object, refs);
-	}
-
 	return 0;
 }
 
diff --git a/object-refs.c b/object-refs.c
deleted file mode 100644
index 5345671569..0000000000
--- a/object-refs.c
+++ /dev/null
@@ -1,87 +0,0 @@
-#include "cache.h"
-#include "object.h"
-#include "decorate.h"
-
-int track_object_refs = 0;
-
-static struct decoration ref_decorate;
-
-struct object_refs *lookup_object_refs(struct object *base)
-{
-	return lookup_decoration(&ref_decorate, base);
-}
-
-static void add_object_refs(struct object *obj, struct object_refs *refs)
-{
-	if (add_decoration(&ref_decorate, obj, refs))
-		die("object %s tried to add refs twice!", sha1_to_hex(obj->sha1));
-}
-
-struct object_refs *alloc_object_refs(unsigned count)
-{
-	struct object_refs *refs;
-	size_t size = sizeof(*refs) + count*sizeof(struct object *);
-
-	refs = xcalloc(1, size);
-	refs->count = count;
-	return refs;
-}
-
-static int compare_object_pointers(const void *a, const void *b)
-{
-	const struct object * const *pa = a;
-	const struct object * const *pb = b;
-	if (*pa == *pb)
-		return 0;
-	else if (*pa < *pb)
-		return -1;
-	else
-		return 1;
-}
-
-void set_object_refs(struct object *obj, struct object_refs *refs)
-{
-	unsigned int i, j;
-
-	/* Do not install empty list of references */
-	if (refs->count < 1) {
-		free(refs);
-		return;
-	}
-
-	/* Sort the list and filter out duplicates */
-	qsort(refs->ref, refs->count, sizeof(refs->ref[0]),
-	      compare_object_pointers);
-	for (i = j = 1; i < refs->count; i++) {
-		if (refs->ref[i] != refs->ref[i - 1])
-			refs->ref[j++] = refs->ref[i];
-	}
-	if (j < refs->count) {
-		/* Duplicates were found - reallocate list */
-		size_t size = sizeof(*refs) + j*sizeof(struct object *);
-		refs->count = j;
-		refs = xrealloc(refs, size);
-	}
-
-	for (i = 0; i < refs->count; i++)
-		refs->ref[i]->used = 1;
-	add_object_refs(obj, refs);
-}
-
-void mark_reachable(struct object *obj, unsigned int mask)
-{
-	const struct object_refs *refs;
-
-	if (!track_object_refs)
-		die("cannot do reachability with object refs turned off");
-	/* If we've been here already, don't bother */
-	if (obj->flags & mask)
-		return;
-	obj->flags |= mask;
-	refs = lookup_object_refs(obj);
-	if (refs) {
-		unsigned i;
-		for (i = 0; i < refs->count; i++)
-			mark_reachable(refs->ref[i], mask);
-	}
-}
diff --git a/object.h b/object.h
index 397bbfa090..036bd66fe9 100644
--- a/object.h
+++ b/object.h
@@ -35,14 +35,11 @@ struct object {
 	unsigned char sha1[20];
 };
 
-extern int track_object_refs;
-
 extern const char *typename(unsigned int type);
 extern int type_from_string(const char *str);
 
 extern unsigned int get_max_object_index(void);
 extern struct object *get_indexed_object(unsigned int);
-extern struct object_refs *lookup_object_refs(struct object *);
 
 /** Internal only **/
 struct object *lookup_object(const unsigned char *sha1);
@@ -61,11 +58,6 @@ struct object *parse_object_buffer(const unsigned char *sha1, enum object_type t
 /** Returns the object, with potentially excess memory allocated. **/
 struct object *lookup_unknown_object(const unsigned  char *sha1);
 
-struct object_refs *alloc_object_refs(unsigned count);
-void set_object_refs(struct object *obj, struct object_refs *refs);
-
-void mark_reachable(struct object *obj, unsigned int mask);
-
 struct object_list *object_list_insert(struct object *item,
 				       struct object_list **list_p);
 
diff --git a/tag.c b/tag.c
index 38bf9134f9..d19f56d60c 100644
--- a/tag.c
+++ b/tag.c
@@ -84,12 +84,6 @@ int parse_tag_buffer(struct tag *item, void *data, unsigned long size)
 		item->tagged = NULL;
 	}
 
-	if (item->tagged && track_object_refs) {
-		struct object_refs *refs = alloc_object_refs(1);
-		refs->ref[0] = item->tagged;
-		set_object_refs(&item->object, refs);
-	}
-
 	return 0;
 }
 
diff --git a/tree.c b/tree.c
index 8c0819fa72..113e1bb0a4 100644
--- a/tree.c
+++ b/tree.c
@@ -202,52 +202,6 @@ struct tree *lookup_tree(const unsigned char *sha1)
 	return (struct tree *) obj;
 }
 
-/*
- * NOTE! Tree refs to external git repositories
- * (ie gitlinks) do not count as real references.
- *
- * You don't have to have those repositories
- * available at all, much less have the objects
- * accessible from the current repository.
- */
-static void track_tree_refs(struct tree *item)
-{
-	int n_refs = 0, i;
-	struct object_refs *refs;
-	struct tree_desc desc;
-	struct name_entry entry;
-
-	/* Count how many entries there are.. */
-	init_tree_desc(&desc, item->buffer, item->size);
-	while (tree_entry(&desc, &entry)) {
-		if (S_ISGITLINK(entry.mode))
-			continue;
-		n_refs++;
-	}
-
-	/* Allocate object refs and walk it again.. */
-	i = 0;
-	refs = alloc_object_refs(n_refs);
-	init_tree_desc(&desc, item->buffer, item->size);
-	while (tree_entry(&desc, &entry)) {
-		struct object *obj;
-
-		if (S_ISGITLINK(entry.mode))
-			continue;
-		if (S_ISDIR(entry.mode))
-			obj = &lookup_tree(entry.sha1)->object;
-		else if (S_ISREG(entry.mode) || S_ISLNK(entry.mode))
-			obj = &lookup_blob(entry.sha1)->object;
-		else {
-			warning("in tree %s: entry %s has bad mode %.6o\n",
-			     sha1_to_hex(item->object.sha1), entry.path, entry.mode);
-			obj = lookup_unknown_object(entry.sha1);
-		}
-		refs->ref[i++] = obj;
-	}
-	set_object_refs(&item->object, refs);
-}
-
 int parse_tree_buffer(struct tree *item, void *buffer, unsigned long size)
 {
 	if (item->object.parsed)
@@ -256,8 +210,6 @@ int parse_tree_buffer(struct tree *item, void *buffer, unsigned long size)
 	item->buffer = buffer;
 	item->size = size;
 
-	if (track_object_refs)
-		track_tree_refs(item);
 	return 0;
 }
 
diff --git a/upload-pack.c b/upload-pack.c
index d1d2c2a1f7..ba9ba59976 100644
--- a/upload-pack.c
+++ b/upload-pack.c
@@ -392,7 +392,6 @@ static int get_common_commits(void)
 	char hex[41], last_hex[41];
 	int len;
 
-	track_object_refs = 0;
 	save_commit_buffer = 0;
 
 	for(;;) {
diff --git a/walker.c b/walker.c
index adc3e80ce1..c10eca8826 100644
--- a/walker.c
+++ b/walker.c
@@ -256,7 +256,6 @@ int walker_fetch(struct walker *walker, int targets, char **target,
 	int i;
 
 	save_commit_buffer = 0;
-	track_object_refs = 0;
 
 	for (i = 0; i < targets; i++) {
 		if (!write_ref || !write_ref[i])

From 45163382437c3862d3beb88134b7a975a3a26443 Mon Sep 17 00:00:00 2001
From: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Date: Mon, 25 Feb 2008 22:46:07 +0100
Subject: [PATCH 05/53] builtin-fsck: reports missing parent commits

parse_commit ignores parent commits with certain errors
(eg. a non commit object is already loaded under the sha1 of
the parent). To make fsck reports such errors, it has to compare
the nummer of parent commits returned by parse commit with the
number of parent commits in the object or in the graft/shallow file.

Signed-off-by: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 builtin-fsck.c | 24 ++++++++++++++++++++++++
 commit.c       |  2 +-
 commit.h       |  1 +
 3 files changed, 26 insertions(+), 1 deletion(-)

diff --git a/builtin-fsck.c b/builtin-fsck.c
index 7321ab236b..727310afc2 100644
--- a/builtin-fsck.c
+++ b/builtin-fsck.c
@@ -397,6 +397,8 @@ static int fsck_commit(struct commit *commit)
 {
 	char *buffer = commit->buffer;
 	unsigned char tree_sha1[20], sha1[20];
+	struct commit_graft *graft;
+	int parents = 0;
 
 	if (verbose)
 		fprintf(stderr, "Checking commit %s\n",
@@ -411,6 +413,28 @@ static int fsck_commit(struct commit *commit)
 		if (get_sha1_hex(buffer+7, sha1) || buffer[47] != '\n')
 			return objerror(&commit->object, "invalid 'parent' line format - bad sha1");
 		buffer += 48;
+		parents++;
+	}
+	graft = lookup_commit_graft(commit->object.sha1);
+	if (graft) {
+		struct commit_list *p = commit->parents;
+		parents = 0;
+		while (p) {
+			p = p->next;
+			parents++;
+		}
+		if (graft->nr_parent == -1 && !parents)
+			; /* shallow commit */
+		else if (graft->nr_parent != parents)
+			return objerror(&commit->object, "graft objects missing");
+	} else {
+		struct commit_list *p = commit->parents;
+		while (p && parents) {
+			p = p->next;
+			parents--;
+		}
+		if (p || parents)
+			return objerror(&commit->object, "parent objects missing");
 	}
 	if (memcmp(buffer, "author ", 7))
 		return objerror(&commit->object, "invalid format - expected 'author' line");
diff --git a/commit.c b/commit.c
index 6684c4e731..94d5b3d261 100644
--- a/commit.c
+++ b/commit.c
@@ -193,7 +193,7 @@ static void prepare_commit_graft(void)
 	commit_graft_prepared = 1;
 }
 
-static struct commit_graft *lookup_commit_graft(const unsigned char *sha1)
+struct commit_graft *lookup_commit_graft(const unsigned char *sha1)
 {
 	int pos;
 	prepare_commit_graft();
diff --git a/commit.h b/commit.h
index 10e2b5d4cf..3ad3dd9af1 100644
--- a/commit.h
+++ b/commit.h
@@ -101,6 +101,7 @@ struct commit_graft {
 struct commit_graft *read_graft_line(char *buf, int len);
 int register_commit_graft(struct commit_graft *, int);
 int read_graft_file(const char *graft_file);
+struct commit_graft *lookup_commit_graft(const unsigned char *sha1);
 
 extern struct commit_list *get_merge_bases(struct commit *rev1, struct commit *rev2, int cleanup);
 

From ba002f3b28ab9febea432d4c415dbe581836d9a0 Mon Sep 17 00:00:00 2001
From: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Date: Mon, 25 Feb 2008 22:46:08 +0100
Subject: [PATCH 06/53] builtin-fsck: move common object checking code to
 fsck.c

Signed-off-by: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 builtin-fsck.c | 274 +++++++------------------------------------------
 fsck.c         | 215 ++++++++++++++++++++++++++++++++++++++
 fsck.h         |   7 ++
 3 files changed, 259 insertions(+), 237 deletions(-)

diff --git a/builtin-fsck.c b/builtin-fsck.c
index 727310afc2..d545650c8c 100644
--- a/builtin-fsck.c
+++ b/builtin-fsck.c
@@ -55,13 +55,13 @@ static int objerror(struct object *obj, const char *err, ...)
 	return -1;
 }
 
-static int objwarning(struct object *obj, const char *err, ...)
+static int fsck_error_func(struct object *obj, int type, const char *err, ...)
 {
 	va_list params;
 	va_start(params, err);
-	objreport(obj, "warning", err, params);
+	objreport(obj, (type == FSCK_WARN) ? "warning" : "error", err, params);
 	va_end(params);
-	return -1;
+	return (type == FSCK_WARN) ? 0 : 1;
 }
 
 static int mark_object(struct object *obj, int type, void *data)
@@ -246,229 +246,6 @@ static void check_connectivity(void)
 	}
 }
 
-/*
- * The entries in a tree are ordered in the _path_ order,
- * which means that a directory entry is ordered by adding
- * a slash to the end of it.
- *
- * So a directory called "a" is ordered _after_ a file
- * called "a.c", because "a/" sorts after "a.c".
- */
-#define TREE_UNORDERED (-1)
-#define TREE_HAS_DUPS  (-2)
-
-static int verify_ordered(unsigned mode1, const char *name1, unsigned mode2, const char *name2)
-{
-	int len1 = strlen(name1);
-	int len2 = strlen(name2);
-	int len = len1 < len2 ? len1 : len2;
-	unsigned char c1, c2;
-	int cmp;
-
-	cmp = memcmp(name1, name2, len);
-	if (cmp < 0)
-		return 0;
-	if (cmp > 0)
-		return TREE_UNORDERED;
-
-	/*
-	 * Ok, the first <len> characters are the same.
-	 * Now we need to order the next one, but turn
-	 * a '\0' into a '/' for a directory entry.
-	 */
-	c1 = name1[len];
-	c2 = name2[len];
-	if (!c1 && !c2)
-		/*
-		 * git-write-tree used to write out a nonsense tree that has
-		 * entries with the same name, one blob and one tree.  Make
-		 * sure we do not have duplicate entries.
-		 */
-		return TREE_HAS_DUPS;
-	if (!c1 && S_ISDIR(mode1))
-		c1 = '/';
-	if (!c2 && S_ISDIR(mode2))
-		c2 = '/';
-	return c1 < c2 ? 0 : TREE_UNORDERED;
-}
-
-static int fsck_tree(struct tree *item)
-{
-	int retval;
-	int has_full_path = 0;
-	int has_empty_name = 0;
-	int has_zero_pad = 0;
-	int has_bad_modes = 0;
-	int has_dup_entries = 0;
-	int not_properly_sorted = 0;
-	struct tree_desc desc;
-	unsigned o_mode;
-	const char *o_name;
-	const unsigned char *o_sha1;
-
-	if (verbose)
-		fprintf(stderr, "Checking tree %s\n",
-				sha1_to_hex(item->object.sha1));
-
-	init_tree_desc(&desc, item->buffer, item->size);
-
-	o_mode = 0;
-	o_name = NULL;
-	o_sha1 = NULL;
-	while (desc.size) {
-		unsigned mode;
-		const char *name;
-		const unsigned char *sha1;
-
-		sha1 = tree_entry_extract(&desc, &name, &mode);
-
-		if (strchr(name, '/'))
-			has_full_path = 1;
-		if (!*name)
-			has_empty_name = 1;
-		has_zero_pad |= *(char *)desc.buffer == '0';
-		update_tree_entry(&desc);
-
-		switch (mode) {
-		/*
-		 * Standard modes..
-		 */
-		case S_IFREG | 0755:
-		case S_IFREG | 0644:
-		case S_IFLNK:
-		case S_IFDIR:
-		case S_IFGITLINK:
-			break;
-		/*
-		 * This is nonstandard, but we had a few of these
-		 * early on when we honored the full set of mode
-		 * bits..
-		 */
-		case S_IFREG | 0664:
-			if (!check_strict)
-				break;
-		default:
-			has_bad_modes = 1;
-		}
-
-		if (o_name) {
-			switch (verify_ordered(o_mode, o_name, mode, name)) {
-			case TREE_UNORDERED:
-				not_properly_sorted = 1;
-				break;
-			case TREE_HAS_DUPS:
-				has_dup_entries = 1;
-				break;
-			default:
-				break;
-			}
-		}
-
-		o_mode = mode;
-		o_name = name;
-		o_sha1 = sha1;
-	}
-	free(item->buffer);
-	item->buffer = NULL;
-
-	retval = 0;
-	if (has_full_path) {
-		objwarning(&item->object, "contains full pathnames");
-	}
-	if (has_empty_name) {
-		objwarning(&item->object, "contains empty pathname");
-	}
-	if (has_zero_pad) {
-		objwarning(&item->object, "contains zero-padded file modes");
-	}
-	if (has_bad_modes) {
-		objwarning(&item->object, "contains bad file modes");
-	}
-	if (has_dup_entries) {
-		retval = objerror(&item->object, "contains duplicate file entries");
-	}
-	if (not_properly_sorted) {
-		retval = objerror(&item->object, "not properly sorted");
-	}
-	return retval;
-}
-
-static int fsck_commit(struct commit *commit)
-{
-	char *buffer = commit->buffer;
-	unsigned char tree_sha1[20], sha1[20];
-	struct commit_graft *graft;
-	int parents = 0;
-
-	if (verbose)
-		fprintf(stderr, "Checking commit %s\n",
-			sha1_to_hex(commit->object.sha1));
-
-	if (memcmp(buffer, "tree ", 5))
-		return objerror(&commit->object, "invalid format - expected 'tree' line");
-	if (get_sha1_hex(buffer+5, tree_sha1) || buffer[45] != '\n')
-		return objerror(&commit->object, "invalid 'tree' line format - bad sha1");
-	buffer += 46;
-	while (!memcmp(buffer, "parent ", 7)) {
-		if (get_sha1_hex(buffer+7, sha1) || buffer[47] != '\n')
-			return objerror(&commit->object, "invalid 'parent' line format - bad sha1");
-		buffer += 48;
-		parents++;
-	}
-	graft = lookup_commit_graft(commit->object.sha1);
-	if (graft) {
-		struct commit_list *p = commit->parents;
-		parents = 0;
-		while (p) {
-			p = p->next;
-			parents++;
-		}
-		if (graft->nr_parent == -1 && !parents)
-			; /* shallow commit */
-		else if (graft->nr_parent != parents)
-			return objerror(&commit->object, "graft objects missing");
-	} else {
-		struct commit_list *p = commit->parents;
-		while (p && parents) {
-			p = p->next;
-			parents--;
-		}
-		if (p || parents)
-			return objerror(&commit->object, "parent objects missing");
-	}
-	if (memcmp(buffer, "author ", 7))
-		return objerror(&commit->object, "invalid format - expected 'author' line");
-	free(commit->buffer);
-	commit->buffer = NULL;
-	if (!commit->tree)
-		return objerror(&commit->object, "could not load commit's tree %s", tree_sha1);
-	if (!commit->parents && show_root)
-		printf("root %s\n", sha1_to_hex(commit->object.sha1));
-	if (!commit->date)
-		printf("bad commit date in %s\n",
-		       sha1_to_hex(commit->object.sha1));
-	return 0;
-}
-
-static int fsck_tag(struct tag *tag)
-{
-	struct object *tagged = tag->tagged;
-
-	if (verbose)
-		fprintf(stderr, "Checking tag %s\n",
-			sha1_to_hex(tag->object.sha1));
-
-	if (!tagged) {
-		return objerror(&tag->object, "could not load tagged object");
-	}
-	if (!show_tags)
-		return 0;
-
-	printf("tagged %s %s", typename(tagged->type), sha1_to_hex(tagged->sha1));
-	printf(" (%s) in %s\n", tag->tag, sha1_to_hex(tag->object.sha1));
-	return 0;
-}
-
 static int fsck_sha1(const unsigned char *sha1)
 {
 	struct object *obj = parse_object(sha1);
@@ -480,20 +257,43 @@ static int fsck_sha1(const unsigned char *sha1)
 	if (obj->flags & SEEN)
 		return 0;
 	obj->flags |= SEEN;
+
+	if (verbose)
+		fprintf(stderr, "Checking %s %s\n",
+			typename(obj->type), sha1_to_hex(obj->sha1));
+
 	if (fsck_walk(obj, mark_used, 0))
 		objerror(obj, "broken links");
-	if (obj->type == OBJ_BLOB)
-		return 0;
-	if (obj->type == OBJ_TREE)
-		return fsck_tree((struct tree *) obj);
-	if (obj->type == OBJ_COMMIT)
-		return fsck_commit((struct commit *) obj);
-	if (obj->type == OBJ_TAG)
-		return fsck_tag((struct tag *) obj);
+	if (fsck_object(obj, check_strict, fsck_error_func))
+		return -1;
 
-	/* By now, parse_object() would've returned NULL instead. */
-	return objerror(obj, "unknown type '%d' (internal fsck error)",
-			obj->type);
+	if (obj->type == OBJ_TREE) {
+		struct tree *item = (struct tree *) obj;
+
+		free(item->buffer);
+		item->buffer = NULL;
+	}
+
+	if (obj->type == OBJ_COMMIT) {
+		struct commit *commit = (struct commit *) obj;
+
+		free(commit->buffer);
+		commit->buffer = NULL;
+
+		if (!commit->parents && show_root)
+			printf("root %s\n", sha1_to_hex(commit->object.sha1));
+	}
+
+	if (obj->type == OBJ_TAG) {
+		struct tag *tag = (struct tag *) obj;
+
+		if (show_tags && tag->tagged) {
+			printf("tagged %s %s", typename(tag->tagged->type), sha1_to_hex(tag->tagged->sha1));
+			printf(" (%s) in %s\n", tag->tag, sha1_to_hex(tag->object.sha1));
+		}
+	}
+
+	return 0;
 }
 
 /*
diff --git a/fsck.c b/fsck.c
index 98fb41af68..0680862151 100644
--- a/fsck.c
+++ b/fsck.c
@@ -89,3 +89,218 @@ int fsck_walk(struct object *obj, fsck_walk_func walk, void *data)
 		return -1;
 	}
 }
+
+/*
+ * The entries in a tree are ordered in the _path_ order,
+ * which means that a directory entry is ordered by adding
+ * a slash to the end of it.
+ *
+ * So a directory called "a" is ordered _after_ a file
+ * called "a.c", because "a/" sorts after "a.c".
+ */
+#define TREE_UNORDERED (-1)
+#define TREE_HAS_DUPS  (-2)
+
+static int verify_ordered(unsigned mode1, const char *name1, unsigned mode2, const char *name2)
+{
+	int len1 = strlen(name1);
+	int len2 = strlen(name2);
+	int len = len1 < len2 ? len1 : len2;
+	unsigned char c1, c2;
+	int cmp;
+
+	cmp = memcmp(name1, name2, len);
+	if (cmp < 0)
+		return 0;
+	if (cmp > 0)
+		return TREE_UNORDERED;
+
+	/*
+	 * Ok, the first <len> characters are the same.
+	 * Now we need to order the next one, but turn
+	 * a '\0' into a '/' for a directory entry.
+	 */
+	c1 = name1[len];
+	c2 = name2[len];
+	if (!c1 && !c2)
+		/*
+		 * git-write-tree used to write out a nonsense tree that has
+		 * entries with the same name, one blob and one tree.  Make
+		 * sure we do not have duplicate entries.
+		 */
+		return TREE_HAS_DUPS;
+	if (!c1 && S_ISDIR(mode1))
+		c1 = '/';
+	if (!c2 && S_ISDIR(mode2))
+		c2 = '/';
+	return c1 < c2 ? 0 : TREE_UNORDERED;
+}
+
+static int fsck_tree(struct tree *item, int strict, fsck_error error_func)
+{
+	int retval;
+	int has_full_path = 0;
+	int has_empty_name = 0;
+	int has_zero_pad = 0;
+	int has_bad_modes = 0;
+	int has_dup_entries = 0;
+	int not_properly_sorted = 0;
+	struct tree_desc desc;
+	unsigned o_mode;
+	const char *o_name;
+	const unsigned char *o_sha1;
+
+	init_tree_desc(&desc, item->buffer, item->size);
+
+	o_mode = 0;
+	o_name = NULL;
+	o_sha1 = NULL;
+	if (!desc.size)
+		return error_func(&item->object, FSCK_ERROR, "empty tree");
+
+	while (desc.size) {
+		unsigned mode;
+		const char *name;
+		const unsigned char *sha1;
+
+		sha1 = tree_entry_extract(&desc, &name, &mode);
+
+		if (strchr(name, '/'))
+			has_full_path = 1;
+		if (!*name)
+			has_empty_name = 1;
+		has_zero_pad |= *(char *)desc.buffer == '0';
+		update_tree_entry(&desc);
+
+		switch (mode) {
+		/*
+		 * Standard modes..
+		 */
+		case S_IFREG | 0755:
+		case S_IFREG | 0644:
+		case S_IFLNK:
+		case S_IFDIR:
+		case S_IFGITLINK:
+			break;
+		/*
+		 * This is nonstandard, but we had a few of these
+		 * early on when we honored the full set of mode
+		 * bits..
+		 */
+		case S_IFREG | 0664:
+			if (!strict)
+				break;
+		default:
+			has_bad_modes = 1;
+		}
+
+		if (o_name) {
+			switch (verify_ordered(o_mode, o_name, mode, name)) {
+			case TREE_UNORDERED:
+				not_properly_sorted = 1;
+				break;
+			case TREE_HAS_DUPS:
+				has_dup_entries = 1;
+				break;
+			default:
+				break;
+			}
+		}
+
+		o_mode = mode;
+		o_name = name;
+		o_sha1 = sha1;
+	}
+
+	retval = 0;
+	if (has_full_path)
+		retval += error_func(&item->object, FSCK_WARN, "contains full pathnames");
+	if (has_empty_name)
+		retval += error_func(&item->object, FSCK_WARN, "contains empty pathname");
+	if (has_zero_pad)
+		retval += error_func(&item->object, FSCK_WARN, "contains zero-padded file modes");
+	if (has_bad_modes)
+		retval += error_func(&item->object, FSCK_WARN, "contains bad file modes");
+	if (has_dup_entries)
+		retval += error_func(&item->object, FSCK_ERROR, "contains duplicate file entries");
+	if (not_properly_sorted)
+		retval += error_func(&item->object, FSCK_ERROR, "not properly sorted");
+	return retval;
+}
+
+static int fsck_commit(struct commit *commit, fsck_error error_func)
+{
+	char *buffer = commit->buffer;
+	unsigned char tree_sha1[20], sha1[20];
+	struct commit_graft *graft;
+	int parents = 0;
+
+	if (!commit->date)
+		return error_func(&commit->object, FSCK_ERROR, "invalid author/committer line");
+
+	if (memcmp(buffer, "tree ", 5))
+		return error_func(&commit->object, FSCK_ERROR, "invalid format - expected 'tree' line");
+	if (get_sha1_hex(buffer+5, tree_sha1) || buffer[45] != '\n')
+		return error_func(&commit->object, FSCK_ERROR, "invalid 'tree' line format - bad sha1");
+	buffer += 46;
+	while (!memcmp(buffer, "parent ", 7)) {
+		if (get_sha1_hex(buffer+7, sha1) || buffer[47] != '\n')
+			return error_func(&commit->object, FSCK_ERROR, "invalid 'parent' line format - bad sha1");
+		buffer += 48;
+		parents++;
+	}
+	graft = lookup_commit_graft(commit->object.sha1);
+	if (graft) {
+		struct commit_list *p = commit->parents;
+		parents = 0;
+		while (p) {
+			p = p->next;
+			parents++;
+		}
+		if (graft->nr_parent == -1 && !parents)
+			; /* shallow commit */
+		else if (graft->nr_parent != parents)
+			return error_func(&commit->object, FSCK_ERROR, "graft objects missing");
+	} else {
+		struct commit_list *p = commit->parents;
+		while (p && parents) {
+			p = p->next;
+			parents--;
+		}
+		if (p || parents)
+			return error_func(&commit->object, FSCK_ERROR, "parent objects missing");
+	}
+	if (memcmp(buffer, "author ", 7))
+		return error_func(&commit->object, FSCK_ERROR, "invalid format - expected 'author' line");
+	if (!commit->tree)
+		return error_func(&commit->object, FSCK_ERROR, "could not load commit's tree %s", sha1_to_hex(tree_sha1));
+
+	return 0;
+}
+
+static int fsck_tag(struct tag *tag, fsck_error error_func)
+{
+	struct object *tagged = tag->tagged;
+
+	if (!tagged)
+		return error_func(&tag->object, FSCK_ERROR, "could not load tagged object");
+	return 0;
+}
+
+int fsck_object(struct object *obj, int strict, fsck_error error_func)
+{
+	if (!obj)
+		return error_func(obj, FSCK_ERROR, "no valid object to fsck");
+
+	if (obj->type == OBJ_BLOB)
+		return 0;
+	if (obj->type == OBJ_TREE)
+		return fsck_tree((struct tree *) obj, strict, error_func);
+	if (obj->type == OBJ_COMMIT)
+		return fsck_commit((struct commit *) obj, error_func);
+	if (obj->type == OBJ_TAG)
+		return fsck_tag((struct tag *) obj, error_func);
+
+	return error_func(obj, FSCK_ERROR, "unknown type '%d' (internal fsck error)",
+			  obj->type);
+}
diff --git a/fsck.h b/fsck.h
index ba5514a818..9bd37c9f1c 100644
--- a/fsck.h
+++ b/fsck.h
@@ -1,6 +1,9 @@
 #ifndef GIT_FSCK_H
 #define GIT_FSCK_H
 
+#define FSCK_ERROR 1
+#define FSCK_WARN 2
+
 /*
  * callback function for fsck_walk
  * type is the expected type of the object or OBJ_ANY
@@ -11,6 +14,9 @@
  */
 typedef int (*fsck_walk_func)(struct object *obj, int type, void *data);
 
+/* callback for fsck_object, type is FSCK_ERROR or FSCK_WARN */
+typedef int (*fsck_error)(struct object *obj, int type, const char *err, ...);
+
 /* descend in all linked child objects
  * the return value is:
  *    -1	error in processing the object
@@ -19,5 +25,6 @@ typedef int (*fsck_walk_func)(struct object *obj, int type, void *data);
  *    0		everything OK
  */
 int fsck_walk(struct object *obj, fsck_walk_func walk, void *data);
+int fsck_object(struct object *obj, int strict, fsck_error error_func);
 
 #endif

From d6ffc8d784d41977397a49ce5333cfe7db1e9c2c Mon Sep 17 00:00:00 2001
From: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Date: Mon, 25 Feb 2008 22:46:09 +0100
Subject: [PATCH 07/53] add common fsck error printing function

Signed-off-by: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 fsck.c | 29 +++++++++++++++++++++++++++++
 fsck.h |  2 ++
 2 files changed, 31 insertions(+)

diff --git a/fsck.c b/fsck.c
index 0680862151..6883d1bd68 100644
--- a/fsck.c
+++ b/fsck.c
@@ -304,3 +304,32 @@ int fsck_object(struct object *obj, int strict, fsck_error error_func)
 	return error_func(obj, FSCK_ERROR, "unknown type '%d' (internal fsck error)",
 			  obj->type);
 }
+
+int fsck_error_function(struct object *obj, int type, const char *fmt, ...)
+{
+	va_list ap;
+	int len;
+	struct strbuf sb;
+
+	strbuf_init(&sb, 0);
+	strbuf_addf(&sb, "object %s:", obj->sha1?sha1_to_hex(obj->sha1):"(null)");
+
+	va_start(ap, fmt);
+	len = vsnprintf(sb.buf + sb.len, strbuf_avail(&sb), fmt, ap);
+	va_end(ap);
+
+	if (len < 0)
+		len = 0;
+	if (len >= strbuf_avail(&sb)) {
+		strbuf_grow(&sb, len + 2);
+		va_start(ap, fmt);
+		len = vsnprintf(sb.buf + sb.len, strbuf_avail(&sb), fmt, ap);
+		va_end(ap);
+		if (len >= strbuf_avail(&sb))
+			die("this should not happen, your snprintf is broken");
+	}
+
+	error(sb.buf);
+	strbuf_release(&sb);
+	return 1;
+}
diff --git a/fsck.h b/fsck.h
index 9bd37c9f1c..990ee02335 100644
--- a/fsck.h
+++ b/fsck.h
@@ -17,6 +17,8 @@ typedef int (*fsck_walk_func)(struct object *obj, int type, void *data);
 /* callback for fsck_object, type is FSCK_ERROR or FSCK_WARN */
 typedef int (*fsck_error)(struct object *obj, int type, const char *err, ...);
 
+int fsck_error_function(struct object *obj, int type, const char *fmt, ...);
+
 /* descend in all linked child objects
  * the return value is:
  *    -1	error in processing the object

From 47e83e3cc0b5b5e2402afb34b7617515f905b7a7 Mon Sep 17 00:00:00 2001
From: Sebastian Noack <sebastian.noack@gmail.com>
Date: Mon, 25 Feb 2008 15:56:28 +0100
Subject: [PATCH 08/53] git-svn: Don't prompt for client cert password
 everytime.

Acked-by: Eric Wong <normalperson@yhbt.net>

Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 git-svn.perl | 1 +
 1 file changed, 1 insertion(+)

diff --git a/git-svn.perl b/git-svn.perl
index 75e97cc72f..a4ab05155d 100755
--- a/git-svn.perl
+++ b/git-svn.perl
@@ -3632,6 +3632,7 @@ sub _auth_providers () {
 	  SVN::Client::get_ssl_client_cert_file_provider(),
 	  SVN::Client::get_ssl_client_cert_prompt_provider(
 	    \&Git::SVN::Prompt::ssl_client_cert, 2),
+	  SVN::Client::get_ssl_client_cert_pw_file_provider(),
 	  SVN::Client::get_ssl_client_cert_pw_prompt_provider(
 	    \&Git::SVN::Prompt::ssl_client_cert_pw, 2),
 	  SVN::Client::get_username_provider(),

From 311e552e7674be620bd930ddada307c15140c6ac Mon Sep 17 00:00:00 2001
From: Jakub Narebski <jnareb@gmail.com>
Date: Tue, 26 Feb 2008 13:22:06 +0100
Subject: [PATCH 09/53] gitweb: Change parse_commits signature to allow for
 multiple options

Change order of parameters in parse_commits() to have $filename
before @args (extra options), to allow for multiple extra options,
for example both '--grep=<pattern>' and '--fixed-strings'.

Change all callers to follow new calling convention.

Originally by Petr Baudis, in http://repo.or.cz/git/gitweb.git:

    b98f0a7c gitweb: Clearly distinguish regexp / exact match searches

Signed-off-by: Petr Baudis <pasky@suse.cz>
Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 gitweb/gitweb.perl | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/gitweb/gitweb.perl b/gitweb/gitweb.perl
index fc95e2ca85..3b4b15a5f0 100755
--- a/gitweb/gitweb.perl
+++ b/gitweb/gitweb.perl
@@ -2079,7 +2079,7 @@ sub parse_commit {
 }
 
 sub parse_commits {
-	my ($commit_id, $maxcount, $skip, $arg, $filename) = @_;
+	my ($commit_id, $maxcount, $skip, $filename, @args) = @_;
 	my @cos;
 
 	$maxcount ||= 1;
@@ -2089,7 +2089,7 @@ sub parse_commits {
 
 	open my $fd, "-|", git_cmd(), "rev-list",
 		"--header",
-		($arg ? ($arg) : ()),
+		@args,
 		("--max-count=" . $maxcount),
 		("--skip=" . $skip),
 		@extra_options,
@@ -5172,7 +5172,7 @@ sub git_history {
 		$ftype = git_get_type($hash);
 	}
 
-	my @commitlist = parse_commits($hash_base, 101, (100 * $page), "--full-history", $file_name);
+	my @commitlist = parse_commits($hash_base, 101, (100 * $page), $file_name, "--full-history");
 
 	my $paging_nav = '';
 	if ($page > 0) {
@@ -5255,7 +5255,7 @@ sub git_search {
 			$greptype = "--committer=";
 		}
 		$greptype .= $search_regexp;
-		my @commitlist = parse_commits($hash, 101, (100 * $page), $greptype);
+		my @commitlist = parse_commits($hash, 101, (100 * $page), undef, $greptype);
 
 		my $paging_nav = '';
 		if ($page > 0) {
@@ -5507,7 +5507,7 @@ sub git_feed {
 
 	# log/feed of current (HEAD) branch, log of given branch, history of file/directory
 	my $head = $hash || 'HEAD';
-	my @commitlist = parse_commits($head, 150, 0, undef, $file_name);
+	my @commitlist = parse_commits($head, 150, 0, $file_name);
 
 	my %latest_commit;
 	my %latest_date;

From 0270cd0eeaff06f6bee1b52d5cdffb1a6e69c15a Mon Sep 17 00:00:00 2001
From: Jakub Narebski <jnareb@gmail.com>
Date: Tue, 26 Feb 2008 13:22:07 +0100
Subject: [PATCH 10/53] gitweb: Simplify fixed string search

Use '--fixed-strings' option to git-rev-list to simplify and improve
searching commit messages (commit search).  It allows to search for
example for "don't" successfully from gitweb.

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 gitweb/gitweb.perl | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/gitweb/gitweb.perl b/gitweb/gitweb.perl
index 3b4b15a5f0..90cf78e436 100755
--- a/gitweb/gitweb.perl
+++ b/gitweb/gitweb.perl
@@ -5254,14 +5254,16 @@ sub git_search {
 		} elsif ($searchtype eq 'committer') {
 			$greptype = "--committer=";
 		}
-		$greptype .= $search_regexp;
-		my @commitlist = parse_commits($hash, 101, (100 * $page), undef, $greptype);
+		$greptype .= $searchtext;
+		my @commitlist = parse_commits($hash, 101, (100 * $page), undef,
+		                               $greptype, '--fixed-strings');
 
 		my $paging_nav = '';
 		if ($page > 0) {
 			$paging_nav .=
 				$cgi->a({-href => href(action=>"search", hash=>$hash,
-				                       searchtext=>$searchtext, searchtype=>$searchtype)},
+				                       searchtext=>$searchtext,
+				                       searchtype=>$searchtype)},
 				        "first");
 			$paging_nav .= " &sdot; " .
 				$cgi->a({-href => href(-replay=>1, page=>$page-1),

From 0e55991987e490a69bc549f6f789b34489c953a7 Mon Sep 17 00:00:00 2001
From: Petr Baudis <pasky@ucw.cz>
Date: Tue, 26 Feb 2008 13:22:08 +0100
Subject: [PATCH 11/53] gitweb: Clearly distinguish regexp / exact match
 searches

This patch does a couple of things:

* Makes commit/author/committer search case insensitive

  To be consistent with the grep search; I see no convincing
  reason for the search to be case sensitive, and you might
  get in trouble especially with contributors e.g. from Japan
  or France where they sometimes like to uppercase their last
  name.

* Makes grep search by default search for fixed strings.

* Introduces 're' checkbox that enables POSIX extended regexp searches

  This works for all the search types. The idea comes from Jakub.

It does not make much sense (and is not easy at all) to untangle most
of these changes from each other, thus they all go in a single patch.

[jn: Cherry-picked from Pasky's http://repo.or.cz/git/gitweb.git]

Signed-off-by: Petr Baudis <pasky@suse.cz>
Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 gitweb/gitweb.perl | 44 ++++++++++++++++++++++++++++++--------------
 1 file changed, 30 insertions(+), 14 deletions(-)

diff --git a/gitweb/gitweb.perl b/gitweb/gitweb.perl
index 90cf78e436..20dc5d59c2 100755
--- a/gitweb/gitweb.perl
+++ b/gitweb/gitweb.perl
@@ -472,13 +472,15 @@ if (defined $searchtype) {
 	}
 }
 
+our $search_use_regexp = $cgi->param('sr');
+
 our $searchtext = $cgi->param('s');
 our $search_regexp;
 if (defined $searchtext) {
 	if (length($searchtext) < 2) {
 		die_error(undef, "At least two characters are required for search parameter");
 	}
-	$search_regexp = quotemeta $searchtext;
+	$search_regexp = $search_use_regexp ? $searchtext : quotemeta $searchtext;
 }
 
 # now read PATH_INFO and use it as alternative to parameters
@@ -608,6 +610,7 @@ sub href(%) {
 		searchtype => "st",
 		snapshot_format => "sf",
 		extra_options => "opt",
+		search_use_regexp => "sr",
 	);
 	my %mapping = @mapping;
 
@@ -2584,6 +2587,10 @@ EOF
 		      $cgi->sup($cgi->a({-href => href(action=>"search_help")}, "?")) .
 		      " search:\n",
 		      $cgi->textfield(-name => "s", -value => $searchtext) . "\n" .
+		      "<span title=\"Extended regular expression\">" .
+		      $cgi->checkbox(-name => 'sr', -value => 1, -label => 're',
+		                     -checked => $search_use_regexp) .
+		      "</span>" .
 		      "</div>" .
 		      $cgi->end_form() . "\n";
 	}
@@ -5256,7 +5263,8 @@ sub git_search {
 		}
 		$greptype .= $searchtext;
 		my @commitlist = parse_commits($hash, 101, (100 * $page), undef,
-		                               $greptype, '--fixed-strings');
+		                               $greptype, '--regexp-ignore-case',
+		                               $search_use_regexp ? '--extended-regexp' : '--fixed-strings');
 
 		my $paging_nav = '';
 		if ($page > 0) {
@@ -5300,8 +5308,9 @@ sub git_search {
 		my $git_command = git_cmd_str();
 		my $searchqtext = $searchtext;
 		$searchqtext =~ s/'/'\\''/;
+		my $pickaxe_flags = $search_use_regexp ? '--pickaxe-regex' : '';
 		open my $fd, "-|", "$git_command rev-list $hash | " .
-			"$git_command diff-tree -r --stdin -S\'$searchqtext\'";
+			"$git_command diff-tree -r --stdin -S\'$searchqtext\' $pickaxe_flags";
 		undef %co;
 		my @files;
 		while (my $line = <$fd>) {
@@ -5365,7 +5374,9 @@ sub git_search {
 		my $alternate = 1;
 		my $matches = 0;
 		$/ = "\n";
-		open my $fd, "-|", git_cmd(), 'grep', '-n', '-i', '-E', $searchtext, $co{'tree'};
+		open my $fd, "-|", git_cmd(), 'grep', '-n',
+			$search_use_regexp ? ('-E', '-i') : '-F',
+			$searchtext, $co{'tree'};
 		my $lastfile = '';
 		while (my $line = <$fd>) {
 			chomp $line;
@@ -5395,7 +5406,7 @@ sub git_search {
 				print "<div class=\"binary\">Binary file</div>\n";
 			} else {
 				$ltext = untabify($ltext);
-				if ($ltext =~ m/^(.*)($searchtext)(.*)$/i) {
+				if ($ltext =~ m/^(.*)($search_regexp)(.*)$/i) {
 					$ltext = esc_html($1, -nbsp=>1);
 					$ltext .= '<span class="match">';
 					$ltext .= esc_html($2, -nbsp=>1);
@@ -5430,27 +5441,31 @@ sub git_search_help {
 	git_header_html();
 	git_print_page_nav('','', $hash,$hash,$hash);
 	print <<EOT;
+<p><strong>Pattern</strong> is by default a normal string that is matched precisely (but without
+regard to case, except in the case of pickaxe). However, when you check the <em>re</em> checkbox,
+the pattern entered is recognized as the POSIX extended
+<a href="http://en.wikipedia.org/wiki/Regular_expression">regular expression</a> (also case
+insensitive).</p>
 <dl>
 <dt><b>commit</b></dt>
-<dd>The commit messages and authorship information will be scanned for the given string.</dd>
+<dd>The commit messages and authorship information will be scanned for the given pattern.</dd>
 EOT
 	my ($have_grep) = gitweb_check_feature('grep');
 	if ($have_grep) {
 		print <<EOT;
 <dt><b>grep</b></dt>
 <dd>All files in the currently selected tree (HEAD unless you are explicitly browsing
-    a different one) are searched for the given
-<a href="http://en.wikipedia.org/wiki/Regular_expression">regular expression</a>
-(POSIX extended) and the matches are listed. On large
-trees, this search can take a while and put some strain on the server, so please use it with
-some consideration.</dd>
+    a different one) are searched for the given pattern. On large trees, this search can take
+a while and put some strain on the server, so please use it with some consideration. Note that
+due to git-grep peculiarity, currently if regexp mode is turned off, the matches are
+case-sensitive.</dd>
 EOT
 	}
 	print <<EOT;
 <dt><b>author</b></dt>
-<dd>Name and e-mail of the change author and date of birth of the patch will be scanned for the given string.</dd>
+<dd>Name and e-mail of the change author and date of birth of the patch will be scanned for the given pattern.</dd>
 <dt><b>committer</b></dt>
-<dd>Name and e-mail of the committer and date of commit will be scanned for the given string.</dd>
+<dd>Name and e-mail of the committer and date of commit will be scanned for the given pattern.</dd>
 EOT
 	my ($have_pickaxe) = gitweb_check_feature('pickaxe');
 	if ($have_pickaxe) {
@@ -5458,7 +5473,8 @@ EOT
 <dt><b>pickaxe</b></dt>
 <dd>All commits that caused the string to appear or disappear from any file (changes that
 added, removed or "modified" the string) will be listed. This search can take a while and
-takes a lot of strain on the server, so please use it wisely.</dd>
+takes a lot of strain on the server, so please use it wisely. Note that since you may be
+interested even in changes just changing the case as well, this search is case sensitive.</dd>
 EOT
 	}
 	print "</dl>\n";

From 95775e5377960409a1b3ac4082dded7d17e14a46 Mon Sep 17 00:00:00 2001
From: Junio C Hamano <gitster@pobox.com>
Date: Wed, 27 Feb 2008 13:50:44 -0800
Subject: [PATCH 12/53] git-remote: do not complain on multiple URLs for a
 remote

Having more than one URL for a remote is perfectly normal when
the remote is defined to push to multiple places.  Get rid of
the annoying "Warning" message.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 git-remote.perl | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/git-remote.perl b/git-remote.perl
index 5cd69513cf..b30ed734e7 100755
--- a/git-remote.perl
+++ b/git-remote.perl
@@ -7,10 +7,10 @@ my $git = Git->repository();
 sub add_remote_config {
 	my ($hash, $name, $what, $value) = @_;
 	if ($what eq 'url') {
-		if (exists $hash->{$name}{'URL'}) {
-			print STDERR "Warning: more than one remote.$name.url\n";
+		# Having more than one is Ok -- it is used for push.
+		if (! exists $hash->{'URL'}) {
+			$hash->{$name}{'URL'} = $value;
 		}
-		$hash->{$name}{'URL'} = $value;
 	}
 	elsif ($what eq 'fetch') {
 		$hash->{$name}{'FETCH'} ||= [];

From 22665bbaab799b1f20a23039a5c759cd35d36939 Mon Sep 17 00:00:00 2001
From: Johannes Sixt <j.sixt@viscovery.net>
Date: Mon, 25 Feb 2008 14:25:20 +0100
Subject: [PATCH 13/53] daemon: send more error messages to the syslog

There were a number of die() calls before the syslog was opened; hence,
these error messages would have been sent to /dev/null in detached mode.
Now we install the daemon-specific die routine before any error message is
generated so that these messages go to the syslog.

Signed-off-by: Johannes Sixt <johannes.sixt@telecom.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 daemon.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/daemon.c b/daemon.c
index 41a60af624..dd0177f48d 100644
--- a/daemon.c
+++ b/daemon.c
@@ -1149,6 +1149,11 @@ int main(int argc, char **argv)
 		usage(daemon_usage);
 	}
 
+	if (log_syslog) {
+		openlog("git-daemon", 0, LOG_DAEMON);
+		set_die_routine(daemon_die);
+	}
+
 	if (inetd_mode && (group_name || user_name))
 		die("--user and --group are incompatible with --inetd");
 
@@ -1176,11 +1181,6 @@ int main(int argc, char **argv)
 		}
 	}
 
-	if (log_syslog) {
-		openlog("git-daemon", 0, LOG_DAEMON);
-		set_die_routine(daemon_die);
-	}
-
 	if (strict_paths && (!ok_paths || !*ok_paths))
 		die("option --strict-paths requires a whitelist");
 

From 20632071560ad4915f4e620d3c053e5ee3af80f3 Mon Sep 17 00:00:00 2001
From: Johannes Sixt <j.sixt@viscovery.net>
Date: Tue, 26 Feb 2008 13:00:55 +0100
Subject: [PATCH 14/53] daemon: ensure that base-path is an existing directory

Any request to the daemon would fail if base-path (if specified) is not
a directory. We now check for this condition early.

Signed-off-by: Johannes Sixt <johannes.sixt@telecom.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 daemon.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/daemon.c b/daemon.c
index dd0177f48d..2b4a6f145c 100644
--- a/daemon.c
+++ b/daemon.c
@@ -1184,6 +1184,14 @@ int main(int argc, char **argv)
 	if (strict_paths && (!ok_paths || !*ok_paths))
 		die("option --strict-paths requires a whitelist");
 
+	if (base_path) {
+		struct stat st;
+
+		if (stat(base_path, &st) || !S_ISDIR(st.st_mode))
+			die("base-path '%s' does not exist or "
+			    "is not a directory", base_path);
+	}
+
 	if (inetd_mode) {
 		struct sockaddr_storage ss;
 		struct sockaddr *peer = (struct sockaddr *)&ss;

From 9fc6440d783ca773353880aa97c23ed9c640d3c0 Mon Sep 17 00:00:00 2001
From: Mike Hommey <mh@glandium.org>
Date: Wed, 27 Feb 2008 21:35:50 +0100
Subject: [PATCH 15/53] Set proxy override with http_init()

In transport.c, proxy setting (the one from the remote conf) was set through
curl_easy_setopt() call, while http.c already does the same with the
http.proxy setting. We now just use this infrastructure instead, and make
http_init() now take the struct remote as argument so that it can take the
http_proxy setting from there, and any other property that would be added
later.

At the same time, we make get_http_walker() take a struct remote argument
too, and pass it to http_init(), which makes remote defined proxy be used
for more than get_refs_via_curl().

We leave out http-fetch and http-push, which don't use remotes for the
moment, purposefully.

Signed-off-by: Mike Hommey <mh@glandium.org>
Acked-by: Daniel Barkalow <barkalow@iabervon.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 builtin-http-fetch.c |  2 +-
 http-push.c          |  2 +-
 http-walker.c        |  4 ++--
 http.c               | 10 +++++++++-
 http.h               |  3 ++-
 transport.c          |  9 ++++-----
 walker.h             |  4 +++-
 7 files changed, 22 insertions(+), 12 deletions(-)

diff --git a/builtin-http-fetch.c b/builtin-http-fetch.c
index 7f450c61d9..48128c610e 100644
--- a/builtin-http-fetch.c
+++ b/builtin-http-fetch.c
@@ -59,7 +59,7 @@ int cmd_http_fetch(int argc, const char **argv, const char *prefix)
 		url = rewritten_url;
 	}
 
-	walker = get_http_walker(url);
+	walker = get_http_walker(url, NULL);
 	walker->get_tree = get_tree;
 	walker->get_history = get_history;
 	walker->get_all = get_all;
diff --git a/http-push.c b/http-push.c
index b2b410df90..869c01c75b 100644
--- a/http-push.c
+++ b/http-push.c
@@ -2233,7 +2233,7 @@ int main(int argc, char **argv)
 
 	memset(remote_dir_exists, -1, 256);
 
-	http_init();
+	http_init(NULL);
 
 	no_pragma_header = curl_slist_append(no_pragma_header, "Pragma:");
 
diff --git a/http-walker.c b/http-walker.c
index 2c3786870e..7bda34d914 100644
--- a/http-walker.c
+++ b/http-walker.c
@@ -902,13 +902,13 @@ static void cleanup(struct walker *walker)
 	curl_slist_free_all(data->no_pragma_header);
 }
 
-struct walker *get_http_walker(const char *url)
+struct walker *get_http_walker(const char *url, struct remote *remote)
 {
 	char *s;
 	struct walker_data *data = xmalloc(sizeof(struct walker_data));
 	struct walker *walker = xmalloc(sizeof(struct walker));
 
-	http_init();
+	http_init(remote);
 
 	data->no_pragma_header = curl_slist_append(NULL, "Pragma:");
 
diff --git a/http.c b/http.c
index 5925d07478..8e554c0969 100644
--- a/http.c
+++ b/http.c
@@ -218,13 +218,16 @@ static CURL* get_curl_handle(void)
 	return result;
 }
 
-void http_init(void)
+void http_init(struct remote *remote)
 {
 	char *low_speed_limit;
 	char *low_speed_time;
 
 	curl_global_init(CURL_GLOBAL_ALL);
 
+	if (remote && remote->http_proxy)
+		curl_http_proxy = xstrdup(remote->http_proxy);
+
 	pragma_header = curl_slist_append(pragma_header, "Pragma: no-cache");
 
 #ifdef USE_CURL_MULTI
@@ -314,6 +317,11 @@ void http_cleanup(void)
 
 	curl_slist_free_all(pragma_header);
 	pragma_header = NULL;
+
+	if (curl_http_proxy) {
+		free(curl_http_proxy);
+		curl_http_proxy = NULL;
+	}
 }
 
 struct active_request_slot *get_active_slot(void)
diff --git a/http.h b/http.h
index 9bab2c8821..04169d5f9c 100644
--- a/http.h
+++ b/http.h
@@ -7,6 +7,7 @@
 #include <curl/easy.h>
 
 #include "strbuf.h"
+#include "remote.h"
 
 /*
  * We detect based on the cURL version if multi-transfer is
@@ -83,7 +84,7 @@ extern void add_fill_function(void *data, int (*fill)(void *));
 extern void step_active_slots(void);
 #endif
 
-extern void http_init(void);
+extern void http_init(struct remote *remote);
 extern void http_cleanup(void);
 
 extern int data_received;
diff --git a/transport.c b/transport.c
index 497f853721..97c59dce60 100644
--- a/transport.c
+++ b/transport.c
@@ -442,7 +442,8 @@ static struct ref *get_refs_via_curl(struct transport *transport)
 	struct ref *last_ref = NULL;
 
 	if (!transport->data)
-		transport->data = get_http_walker(transport->url);
+		transport->data = get_http_walker(transport->url,
+						transport->remote);
 
 	refs_url = xmalloc(strlen(transport->url) + 11);
 	sprintf(refs_url, "%s/info/refs", transport->url);
@@ -453,9 +454,6 @@ static struct ref *get_refs_via_curl(struct transport *transport)
 	curl_easy_setopt(slot->curl, CURLOPT_WRITEFUNCTION, fwrite_buffer);
 	curl_easy_setopt(slot->curl, CURLOPT_URL, refs_url);
 	curl_easy_setopt(slot->curl, CURLOPT_HTTPHEADER, NULL);
-	if (transport->remote->http_proxy)
-		curl_easy_setopt(slot->curl, CURLOPT_PROXY,
-				 transport->remote->http_proxy);
 
 	if (start_active_slot(slot)) {
 		run_active_slot(slot);
@@ -509,7 +507,8 @@ static int fetch_objs_via_curl(struct transport *transport,
 				 int nr_objs, struct ref **to_fetch)
 {
 	if (!transport->data)
-		transport->data = get_http_walker(transport->url);
+		transport->data = get_http_walker(transport->url,
+						transport->remote);
 	return fetch_objs_via_walker(transport, nr_objs, to_fetch);
 }
 
diff --git a/walker.h b/walker.h
index ea2c363f4e..e1d40deaff 100644
--- a/walker.h
+++ b/walker.h
@@ -1,6 +1,8 @@
 #ifndef WALKER_H
 #define WALKER_H
 
+#include "remote.h"
+
 struct walker {
 	void *data;
 	int (*fetch_ref)(struct walker *, char *ref, unsigned char *sha1);
@@ -32,6 +34,6 @@ int walker_fetch(struct walker *impl, int targets, char **target,
 
 void walker_free(struct walker *walker);
 
-struct walker *get_http_walker(const char *url);
+struct walker *get_http_walker(const char *url, struct remote *remote);
 
 #endif /* WALKER_H */

From 6eaf40608dfe0c23aa990a8c7d4a135df1af3cab Mon Sep 17 00:00:00 2001
From: Clemens Buchacher <drizzd@aon.at>
Date: Wed, 27 Feb 2008 20:27:53 +0100
Subject: [PATCH 16/53] http-push: push <remote> :<branch> deletes remote
 branch

This mirrors current ssh/git push syntax.

Signed-off-by: Clemens Buchacher <drizzd@aon.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 http-push.c | 17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)

diff --git a/http-push.c b/http-push.c
index 0beb7406c3..4b31070eb9 100644
--- a/http-push.c
+++ b/http-push.c
@@ -2138,6 +2138,8 @@ static int delete_remote_branch(char *pattern, int force)
 
 	/* Send delete request */
 	fprintf(stderr, "Removing remote branch '%s'\n", remote_ref->name);
+	if (dry_run)
+		return 0;
 	url = xmalloc(strlen(remote->url) + strlen(remote_ref->name) + 1);
 	sprintf(url, "%s%s", remote->url, remote_ref->name);
 	slot = get_active_slot();
@@ -2311,6 +2313,16 @@ int main(int argc, char **argv)
 
 		if (!ref->peer_ref)
 			continue;
+
+		if (is_zero_sha1(ref->peer_ref->new_sha1)) {
+			if (delete_remote_branch(ref->name, 1) == -1) {
+				error("Could not remove %s", ref->name);
+				rc = -4;
+			}
+			new_refs++;
+			continue;
+		}
+
 		if (!hashcmp(ref->old_sha1, ref->peer_ref->new_sha1)) {
 			if (push_verbosely || 1)
 				fprintf(stderr, "'%s': up-to-date\n", ref->name);
@@ -2342,11 +2354,6 @@ int main(int argc, char **argv)
 			}
 		}
 		hashcpy(ref->new_sha1, ref->peer_ref->new_sha1);
-		if (is_zero_sha1(ref->new_sha1)) {
-			error("cannot happen anymore");
-			rc = -3;
-			continue;
-		}
 		new_refs++;
 		strcpy(old_hex, sha1_to_hex(ref->old_sha1));
 		new_hex = sha1_to_hex(ref->new_sha1);

From faa4bc35a05ddb1822f3770cd8c51859e3b929ee Mon Sep 17 00:00:00 2001
From: Clemens Buchacher <drizzd@aon.at>
Date: Wed, 27 Feb 2008 20:28:45 +0100
Subject: [PATCH 17/53] http-push: add regression tests

http-push tests require a web server with WebDAV support.

This commit introduces a HTTPD test library, which can be configured using
the following environment variables.

GIT_TEST_HTTPD		enable HTTPD tests
LIB_HTTPD_PATH		web server path
LIB_HTTPD_MODULE_PATH	web server modules path
LIB_HTTPD_PORT		listening port
LIB_HTTPD_DAV		enable DAV
LIB_HTTPD_SVN		enable SVN
LIB_HTTPD_SSL		enable SSL

Signed-off-by: Clemens Buchacher <drizzd@aon.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 t/lib-httpd.sh          | 96 +++++++++++++++++++++++++++++++++++++++++
 t/lib-httpd/apache.conf | 34 +++++++++++++++
 t/lib-httpd/ssl.cnf     |  8 ++++
 t/t5540-http-push.sh    | 73 +++++++++++++++++++++++++++++++
 t/test-lib.sh           |  9 +++-
 5 files changed, 218 insertions(+), 2 deletions(-)
 create mode 100644 t/lib-httpd.sh
 create mode 100644 t/lib-httpd/apache.conf
 create mode 100644 t/lib-httpd/ssl.cnf
 create mode 100755 t/t5540-http-push.sh

diff --git a/t/lib-httpd.sh b/t/lib-httpd.sh
new file mode 100644
index 0000000000..7f206c56cf
--- /dev/null
+++ b/t/lib-httpd.sh
@@ -0,0 +1,96 @@
+#!/bin/sh
+#
+# Copyright (c) 2008 Clemens Buchacher <drizzd@aon.at>
+#
+
+if test -z "$GIT_TEST_HTTPD"
+then
+	say "skipping test, network testing disabled by default"
+	say "(define GIT_TEST_HTTPD to enable)"
+	test_done
+	exit
+fi
+
+LIB_HTTPD_PATH=${LIB_HTTPD_PATH-'/usr/sbin/apache2'}
+LIB_HTTPD_PORT=${LIB_HTTPD_PORT-'8111'}
+
+TEST_PATH="$PWD"/../lib-httpd
+HTTPD_ROOT_PATH="$PWD"/httpd
+HTTPD_DOCUMENT_ROOT_PATH=$HTTPD_ROOT_PATH/www
+
+if ! test -x "$LIB_HTTPD_PATH"
+then
+        say "skipping test, no web server found at '$LIB_HTTPD_PATH'"
+        test_done
+        exit
+fi
+
+HTTPD_VERSION=`$LIB_HTTPD_PATH -v | \
+	sed -n 's/^Server version: Apache\/\([0-9]*\)\..*$/\1/p; q'`
+
+if test -n "$HTTPD_VERSION"
+then
+	if test -z "$LIB_HTTPD_MODULE_PATH"
+	then
+		if ! test $HTTPD_VERSION -ge 2
+		then
+			say "skipping test, at least Apache version 2 is required"
+			test_done
+			exit
+		fi
+
+		LIB_HTTPD_MODULE_PATH='/usr/lib/apache2/modules'
+	fi
+else
+	error "Could not identify web server at '$LIB_HTTPD_PATH'"
+fi
+
+HTTPD_PARA="-d $HTTPD_ROOT_PATH -f $TEST_PATH/apache.conf"
+
+prepare_httpd() {
+	mkdir -p $HTTPD_DOCUMENT_ROOT_PATH
+
+	ln -s $LIB_HTTPD_MODULE_PATH $HTTPD_ROOT_PATH/modules
+
+	if test -n "$LIB_HTTPD_SSL"
+	then
+		HTTPD_URL=https://127.0.0.1:$LIB_HTTPD_PORT
+
+		RANDFILE_PATH="$HTTPD_ROOT_PATH"/.rnd openssl req \
+			-config $TEST_PATH/ssl.cnf \
+			-new -x509 -nodes \
+			-out $HTTPD_ROOT_PATH/httpd.pem \
+			-keyout $HTTPD_ROOT_PATH/httpd.pem
+		export GIT_SSL_NO_VERIFY=t
+		HTTPD_PARA="$HTTPD_PARA -DSSL"
+	else
+		HTTPD_URL=http://127.0.0.1:$LIB_HTTPD_PORT
+	fi
+
+	if test -n "$LIB_HTTPD_DAV" -o -n "$LIB_HTTPD_SVN"
+	then
+		HTTPD_PARA="$HTTPD_PARA -DDAV"
+
+		if test -n "$LIB_HTTPD_SVN"
+		then
+			HTTPD_PARA="$HTTPD_PARA -DSVN"
+			rawsvnrepo="$HTTPD_ROOT_PATH/svnrepo"
+			svnrepo="http://127.0.0.1:$LIB_HTTPD_PORT/svn"
+		fi
+	fi
+}
+
+start_httpd() {
+	prepare_httpd
+
+	trap 'stop_httpd; die' exit
+
+	"$LIB_HTTPD_PATH" $HTTPD_PARA \
+		-c "Listen 127.0.0.1:$LIB_HTTPD_PORT" -k start
+}
+
+stop_httpd() {
+	trap 'die' exit
+
+	"$LIB_HTTPD_PATH" $HTTPD_PARA -k stop
+}
diff --git a/t/lib-httpd/apache.conf b/t/lib-httpd/apache.conf
new file mode 100644
index 0000000000..a4473462d1
--- /dev/null
+++ b/t/lib-httpd/apache.conf
@@ -0,0 +1,34 @@
+PidFile httpd.pid
+DocumentRoot www
+ErrorLog error.log
+
+<IfDefine SSL>
+LoadModule ssl_module modules/mod_ssl.so
+
+SSLCertificateFile httpd.pem
+SSLCertificateKeyFile httpd.pem
+SSLRandomSeed startup file:/dev/urandom 512
+SSLRandomSeed connect file:/dev/urandom 512
+SSLSessionCache none
+SSLMutex file:ssl_mutex
+SSLEngine On
+</IfDefine>
+
+<IfDefine DAV>
+	LoadModule dav_module modules/mod_dav.so
+	LoadModule dav_fs_module modules/mod_dav_fs.so
+
+	DAVLockDB DAVLock
+	<Location />
+		Dav on
+	</Location>
+</IfDefine>
+
+<IfDefine SVN>
+	LoadModule dav_svn_module modules/mod_dav_svn.so
+
+	<Location /svn>
+		DAV svn
+		SVNPath svnrepo
+	</Location>
+</IfDefine>
diff --git a/t/lib-httpd/ssl.cnf b/t/lib-httpd/ssl.cnf
new file mode 100644
index 0000000000..6dab2579cb
--- /dev/null
+++ b/t/lib-httpd/ssl.cnf
@@ -0,0 +1,8 @@
+RANDFILE                = $ENV::RANDFILE_PATH
+
+[ req ]
+default_bits            = 1024
+distinguished_name      = req_distinguished_name
+prompt                  = no
+[ req_distinguished_name ]
+commonName              = 127.0.0.1
diff --git a/t/t5540-http-push.sh b/t/t5540-http-push.sh
new file mode 100755
index 0000000000..7372439164
--- /dev/null
+++ b/t/t5540-http-push.sh
@@ -0,0 +1,73 @@
+#!/bin/sh
+#
+# Copyright (c) 2008 Clemens Buchacher <drizzd@aon.at>
+#
+
+test_description='test http-push
+
+This test runs various sanity checks on http-push.'
+
+. ./test-lib.sh
+
+ROOT_PATH="$PWD"
+LIB_HTTPD_DAV=t
+
+. ../lib-httpd.sh
+
+if ! start_httpd >&3 2>&4
+then
+	say "skipping test, web server setup failed"
+	test_done
+	exit
+fi
+
+test_expect_success 'setup remote repository' '
+	cd "$ROOT_PATH" &&
+	mkdir test_repo &&
+	cd test_repo &&
+	git init &&
+	: >path1 &&
+	git add path1 &&
+	test_tick &&
+	git commit -m initial &&
+	cd - &&
+	git clone --bare test_repo test_repo.git &&
+	cd test_repo.git &&
+	git --bare update-server-info &&
+	chmod +x hooks/post-update &&
+	cd - &&
+	mv test_repo.git $HTTPD_DOCUMENT_ROOT_PATH
+'
+	
+test_expect_success 'clone remote repository' '
+	cd "$ROOT_PATH" &&
+	git clone $HTTPD_URL/test_repo.git test_repo_clone
+'
+
+test_expect_success 'push to remote repository' '
+	cd "$ROOT_PATH"/test_repo_clone &&
+	: >path2 &&
+	git add path2 &&
+	test_tick &&
+	git commit -m path2 &&
+	git push
+'
+
+test_expect_success 'create and delete remote branch' '
+	cd "$ROOT_PATH"/test_repo_clone &&
+	git checkout -b dev &&
+	: >path3 &&
+	git add path3 &&
+	test_tick &&
+	git commit -m dev &&
+	git push origin dev &&
+	git fetch &&
+	git push origin :dev &&
+	git branch -d -r origin/dev &&
+	git fetch &&
+	! git show-ref --verify refs/remotes/origin/dev
+'
+
+stop_httpd
+
+test_done
diff --git a/t/test-lib.sh b/t/test-lib.sh
index 83889c4f46..9d9cb8d5a1 100644
--- a/t/test-lib.sh
+++ b/t/test-lib.sh
@@ -80,7 +80,7 @@ do
 	-q|--q|--qu|--qui|--quie|--quiet)
 		quiet=t; shift ;;
 	--no-color)
-	    color=; shift ;;
+		color=; shift ;;
 	--no-python)
 		# noop now...
 		shift ;;
@@ -142,7 +142,12 @@ test_count=0
 test_fixed=0
 test_broken=0
 
-trap 'echo >&5 "FATAL: Unexpected exit with code $?"; exit 1' exit
+die () {
+	echo >&5 "FATAL: Unexpected exit with code $?"
+	exit 1
+}
+
+trap 'die' exit
 
 test_tick () {
 	if test -z "${test_tick+set}"

From e82447b1dfbda6ecfc101381c1295c444c73c903 Mon Sep 17 00:00:00 2001
From: Junio C Hamano <gitster@pobox.com>
Date: Tue, 26 Feb 2008 23:18:38 -0800
Subject: [PATCH 18/53] Fix "git log --merge --left-right"

The command did not reject the combination of these options, but
did not show left/right markers.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 revision.c | 12 ++++--------
 1 file changed, 4 insertions(+), 8 deletions(-)

diff --git a/revision.c b/revision.c
index 6e85aaa3fb..a399f27144 100644
--- a/revision.c
+++ b/revision.c
@@ -749,14 +749,9 @@ static void prepare_show_merge(struct rev_info *revs)
 	add_pending_object(revs, &head->object, "HEAD");
 	add_pending_object(revs, &other->object, "MERGE_HEAD");
 	bases = get_merge_bases(head, other, 1);
-	while (bases) {
-		struct commit *it = bases->item;
-		struct commit_list *n = bases->next;
-		free(bases);
-		bases = n;
-		it->object.flags |= UNINTERESTING;
-		add_pending_object(revs, &it->object, "(merge-base)");
-	}
+	add_pending_commit_list(revs, bases, UNINTERESTING);
+	free_commit_list(bases);
+	head->object.flags |= SYMMETRIC_LEFT;
 
 	if (!active_nr)
 		read_cache();
@@ -775,6 +770,7 @@ static void prepare_show_merge(struct rev_info *revs)
 			i++;
 	}
 	revs->prune_data = prune;
+	revs->limited = 1;
 }
 
 int handle_revision_arg(const char *arg, struct rev_info *revs,

From 212945d4a85dfa172ea55ec73b1d830ef2d8582f Mon Sep 17 00:00:00 2001
From: "Shawn O. Pearce" <spearce@spearce.org>
Date: Thu, 28 Feb 2008 01:22:36 -0500
Subject: [PATCH 19/53] Teach git-describe to verify annotated tag names before
 output

If an annotated tag describes a commit we want to favor the name
listed in the body of the tag, rather than whatever name it has
been stored under locally.  By doing so it is easier to converse
about tags with others, even if the tags happen to be fetched to
a different name than it was given by its creator.

To avoid confusion when a tag is stored under a different name
(and thus is not readable via git-rev-parse --verify, etc.) we show
a warning message if the name of the tag does not match the ref
we found it under and if that tag was also selected for output.
For example:

  $ git tag -a -m "i am a test" testtag
  $ mv .git/refs/tags/testtag .git/refs/tags/bobbytag

  $ ./git-describe HEAD
  warning: tag 'testtag' is really 'bobbytag' here
  testtag

  $ git tag -d testtag
  error: tag 'testtag' not found.
  $ git tag -d bobbytag
  Deleted tag 'bobbytag'

Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 builtin-describe.c | 38 ++++++++++++++++++++++++++++++--------
 1 file changed, 30 insertions(+), 8 deletions(-)

diff --git a/builtin-describe.c b/builtin-describe.c
index 05e309f5ad..08d18507ac 100644
--- a/builtin-describe.c
+++ b/builtin-describe.c
@@ -22,7 +22,9 @@ static int max_candidates = 10;
 const char *pattern = NULL;
 
 struct commit_name {
+	struct tag *tag;
 	int prio; /* annotated tag = 2, tag = 1, head = 0 */
+	unsigned char sha1[20];
 	char path[FLEX_ARRAY]; /* more */
 };
 static const char *prio_names[] = {
@@ -31,14 +33,17 @@ static const char *prio_names[] = {
 
 static void add_to_known_names(const char *path,
 			       struct commit *commit,
-			       int prio)
+			       int prio,
+			       const unsigned char *sha1)
 {
 	struct commit_name *e = commit->util;
 	if (!e || e->prio < prio) {
 		size_t len = strlen(path)+1;
 		free(e);
 		e = xmalloc(sizeof(struct commit_name) + len);
+		e->tag = NULL;
 		e->prio = prio;
+		hashcpy(e->sha1, sha1);
 		memcpy(e->path, path, len);
 		commit->util = e;
 	}
@@ -89,7 +94,7 @@ static int get_name(const char *path, const unsigned char *sha1, int flag, void
 		if (!tags && prio < 2)
 			return 0;
 	}
-	add_to_known_names(all ? path + 5 : path + 10, commit, prio);
+	add_to_known_names(all ? path + 5 : path + 10, commit, prio, sha1);
 	return 0;
 }
 
@@ -146,6 +151,22 @@ static unsigned long finish_depth_computation(
 	return seen_commits;
 }
 
+static void display_name(struct commit_name *n)
+{
+	if (n->prio == 2 && !n->tag) {
+		n->tag = lookup_tag(n->sha1);
+		if (!n->tag || !n->tag->tag)
+			die("annotated tag %s not available", n->path);
+		if (strcmp(n->tag->tag, n->path))
+			warning("tag '%s' is really '%s' here", n->tag->tag, n->path);
+	}
+
+	if (n->tag)
+		printf("%s", n->tag->tag);
+	else
+		printf("%s", n->path);
+}
+
 static void describe(const char *arg, int last_one)
 {
 	unsigned char sha1[20];
@@ -170,7 +191,8 @@ static void describe(const char *arg, int last_one)
 
 	n = cmit->util;
 	if (n) {
-		printf("%s\n", n->path);
+		display_name(n);
+		printf("\n");
 		return;
 	}
 
@@ -252,12 +274,12 @@ static void describe(const char *arg, int last_one)
 				sha1_to_hex(gave_up_on->object.sha1));
 		}
 	}
-	if (abbrev == 0)
-		printf("%s\n", all_matches[0].name->path );
-	else
-		printf("%s-%d-g%s\n", all_matches[0].name->path,
-		       all_matches[0].depth,
+
+	display_name(all_matches[0].name);
+	if (abbrev)
+		printf("-%d-g%s", all_matches[0].depth,
 		       find_unique_abbrev(cmit->object.sha1, abbrev));
+	printf("\n");
 
 	if (!last_one)
 		clear_commit_marks(cmit, -1);

From 2add1e6db44c81e19cd4263317b53f3c1339e61b Mon Sep 17 00:00:00 2001
From: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Date: Mon, 25 Feb 2008 22:46:10 +0100
Subject: [PATCH 20/53] unpack-object: cache for non written objects

Preventing objects with broken links entering the repository
means, that write of some objects must be delayed.

This patch adds a cache to keep the object data in memory. The delta
resolving code must also search in the cache.

Signed-off-by: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 builtin-unpack-objects.c | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/builtin-unpack-objects.c b/builtin-unpack-objects.c
index 1e51865c52..50e07faa12 100644
--- a/builtin-unpack-objects.c
+++ b/builtin-unpack-objects.c
@@ -8,6 +8,7 @@
 #include "tag.h"
 #include "tree.h"
 #include "progress.h"
+#include "decorate.h"
 
 static int dry_run, quiet, recover, has_errors;
 static const char unpack_usage[] = "git-unpack-objects [-n] [-q] [-r] < pack-file";
@@ -18,6 +19,18 @@ static unsigned int offset, len;
 static off_t consumed_bytes;
 static SHA_CTX ctx;
 
+struct obj_buffer {
+	char *buffer;
+	unsigned long size;
+};
+
+static struct decoration obj_decorate;
+
+static struct obj_buffer *lookup_object_buffer(struct object *base)
+{
+	return lookup_decoration(&obj_decorate, base);
+}
+
 /*
  * Make sure at least "min" bytes are available in the buffer, and
  * return the pointer to the buffer.
@@ -189,6 +202,7 @@ static void unpack_delta_entry(enum object_type type, unsigned long delta_size,
 	void *delta_data, *base;
 	unsigned long base_size;
 	unsigned char base_sha1[20];
+	struct object *obj;
 
 	if (type == OBJ_REF_DELTA) {
 		hashcpy(base_sha1, fill(20));
@@ -252,6 +266,15 @@ static void unpack_delta_entry(enum object_type type, unsigned long delta_size,
 		}
 	}
 
+	obj = lookup_object(base_sha1);
+	if (obj) {
+		struct obj_buffer *obj_buf = lookup_object_buffer(obj);
+		if (obj_buf) {
+			resolve_delta(nr, obj->type, obj_buf->buffer, obj_buf->size, delta_data, delta_size);
+			return;
+		}
+	}
+
 	base = read_sha1_file(base_sha1, &type, &base_size);
 	if (!base) {
 		error("failed to read delta-pack base object %s",

From d5ef408b9afb5b4417f4e7e1593a96302d666650 Mon Sep 17 00:00:00 2001
From: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Date: Mon, 25 Feb 2008 22:46:11 +0100
Subject: [PATCH 21/53] unpack-objects: prevent writing of inconsistent objects

This patch introduces a strict mode, which ensures that:
- no malformed object will be written
- no object with broken links will be written

The patch ensures this by delaying the write of all non blob object.
These object are written, after all objects they link to are written.

An error can only result in unreferenced objects.

Signed-off-by: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 Documentation/git-unpack-objects.txt |   3 +
 builtin-unpack-objects.c             | 110 +++++++++++++++++++++++++--
 2 files changed, 106 insertions(+), 7 deletions(-)

diff --git a/Documentation/git-unpack-objects.txt b/Documentation/git-unpack-objects.txt
index b79be3fd4c..3697896a06 100644
--- a/Documentation/git-unpack-objects.txt
+++ b/Documentation/git-unpack-objects.txt
@@ -40,6 +40,9 @@ OPTIONS
 	and make the best effort to recover as many objects as
 	possible.
 
+--strict::
+	Don't write objects with broken content or links.
+
 
 Author
 ------
diff --git a/builtin-unpack-objects.c b/builtin-unpack-objects.c
index 50e07faa12..9d2a854950 100644
--- a/builtin-unpack-objects.c
+++ b/builtin-unpack-objects.c
@@ -7,11 +7,13 @@
 #include "commit.h"
 #include "tag.h"
 #include "tree.h"
+#include "tree-walk.h"
 #include "progress.h"
 #include "decorate.h"
+#include "fsck.h"
 
-static int dry_run, quiet, recover, has_errors;
-static const char unpack_usage[] = "git-unpack-objects [-n] [-q] [-r] < pack-file";
+static int dry_run, quiet, recover, has_errors, strict;
+static const char unpack_usage[] = "git-unpack-objects [-n] [-q] [-r] [--strict] < pack-file";
 
 /* We always read in 4kB chunks. */
 static unsigned char buffer[4096];
@@ -31,6 +33,16 @@ static struct obj_buffer *lookup_object_buffer(struct object *base)
 	return lookup_decoration(&obj_decorate, base);
 }
 
+static void add_object_buffer(struct object *object, char *buffer, unsigned long size)
+{
+	struct obj_buffer *obj;
+	obj = xcalloc(1, sizeof(struct obj_buffer));
+	obj->buffer = buffer;
+	obj->size = size;
+	if (add_decoration(&obj_decorate, object, obj))
+		die("object %s tried to add buffer twice!", sha1_to_hex(object->sha1));
+}
+
 /*
  * Make sure at least "min" bytes are available in the buffer, and
  * return the pointer to the buffer.
@@ -134,9 +146,58 @@ static void add_delta_to_list(unsigned nr, unsigned const char *base_sha1,
 struct obj_info {
 	off_t offset;
 	unsigned char sha1[20];
+	struct object *obj;
 };
 
+#define FLAG_OPEN (1u<<20)
+#define FLAG_WRITTEN (1u<<21)
+
 static struct obj_info *obj_list;
+unsigned nr_objects;
+
+static void write_cached_object(struct object *obj)
+{
+	unsigned char sha1[20];
+	struct obj_buffer *obj_buf = lookup_object_buffer(obj);
+	if (write_sha1_file(obj_buf->buffer, obj_buf->size, typename(obj->type), sha1) < 0)
+		die("failed to write object %s", sha1_to_hex(obj->sha1));
+	obj->flags |= FLAG_WRITTEN;
+}
+
+static int check_object(struct object *obj, int type, void *data)
+{
+	if (!obj)
+		return 0;
+
+	if (obj->flags & FLAG_WRITTEN)
+		return 1;
+
+	if (type != OBJ_ANY && obj->type != type)
+		die("object type mismatch");
+
+	if (!(obj->flags & FLAG_OPEN)) {
+		unsigned long size;
+		int type = sha1_object_info(obj->sha1, &size);
+		if (type != obj->type || type <= 0)
+			die("object of unexpected type");
+		obj->flags |= FLAG_WRITTEN;
+		return 1;
+	}
+
+	if (fsck_object(obj, 1, fsck_error_function))
+		die("Error in object");
+	if (!fsck_walk(obj, check_object, 0))
+		die("Error on reachable objects of %s", sha1_to_hex(obj->sha1));
+	write_cached_object(obj);
+	return 1;
+}
+
+static void write_rest(void)
+{
+	unsigned i;
+	for (i = 0; i < nr_objects; i++)
+		check_object(obj_list[i].obj, OBJ_ANY, 0);
+}
 
 static void added_object(unsigned nr, enum object_type type,
 			 void *data, unsigned long size);
@@ -144,9 +205,36 @@ static void added_object(unsigned nr, enum object_type type,
 static void write_object(unsigned nr, enum object_type type,
 			 void *buf, unsigned long size)
 {
-	if (write_sha1_file(buf, size, typename(type), obj_list[nr].sha1) < 0)
-		die("failed to write object");
 	added_object(nr, type, buf, size);
+	if (!strict) {
+		if (write_sha1_file(buf, size, typename(type), obj_list[nr].sha1) < 0)
+			die("failed to write object");
+		free(buf);
+		obj_list[nr].obj = 0;
+	} else if (type == OBJ_BLOB) {
+		struct blob *blob;
+		if (write_sha1_file(buf, size, typename(type), obj_list[nr].sha1) < 0)
+			die("failed to write object");
+		free(buf);
+
+		blob = lookup_blob(obj_list[nr].sha1);
+		if (blob)
+			blob->object.flags |= FLAG_WRITTEN;
+		else
+			die("invalid blob object");
+		obj_list[nr].obj = 0;
+	} else {
+		struct object *obj;
+		int eaten;
+		hash_sha1_file(buf, size, typename(type), obj_list[nr].sha1);
+		obj = parse_object_buffer(obj_list[nr].sha1, type, size, buf, &eaten);
+		if (!obj)
+			die("invalid %s", typename(type));
+		/* buf is stored via add_object_buffer and in obj, if its a tree or commit */
+		add_object_buffer(obj, buf, size);
+		obj->flags |= FLAG_OPEN;
+		obj_list[nr].obj = obj;
+	}
 }
 
 static void resolve_delta(unsigned nr, enum object_type type,
@@ -163,7 +251,6 @@ static void resolve_delta(unsigned nr, enum object_type type,
 		die("failed to apply delta");
 	free(delta);
 	write_object(nr, type, result, result_size);
-	free(result);
 }
 
 static void added_object(unsigned nr, enum object_type type,
@@ -193,7 +280,8 @@ static void unpack_non_delta_entry(enum object_type type, unsigned long size,
 
 	if (!dry_run && buf)
 		write_object(nr, type, buf, size);
-	free(buf);
+	else
+		free(buf);
 }
 
 static void unpack_delta_entry(enum object_type type, unsigned long delta_size,
@@ -336,7 +424,8 @@ static void unpack_all(void)
 	int i;
 	struct progress *progress = NULL;
 	struct pack_header *hdr = fill(sizeof(struct pack_header));
-	unsigned nr_objects = ntohl(hdr->hdr_entries);
+
+	nr_objects = ntohl(hdr->hdr_entries);
 
 	if (ntohl(hdr->hdr_signature) != PACK_SIGNATURE)
 		die("bad pack file");
@@ -347,6 +436,7 @@ static void unpack_all(void)
 	if (!quiet)
 		progress = start_progress("Unpacking objects", nr_objects);
 	obj_list = xmalloc(nr_objects * sizeof(*obj_list));
+	memset(obj_list, 0, nr_objects * sizeof(*obj_list));
 	for (i = 0; i < nr_objects; i++) {
 		unpack_one(i);
 		display_progress(progress, i + 1);
@@ -382,6 +472,10 @@ int cmd_unpack_objects(int argc, const char **argv, const char *prefix)
 				recover = 1;
 				continue;
 			}
+			if (!strcmp(arg, "--strict")) {
+				strict = 1;
+				continue;
+			}
 			if (!prefixcmp(arg, "--pack_header=")) {
 				struct pack_header *hdr;
 				char *c;
@@ -407,6 +501,8 @@ int cmd_unpack_objects(int argc, const char **argv, const char *prefix)
 	unpack_all();
 	SHA1_Update(&ctx, buffer, offset);
 	SHA1_Final(sha1, &ctx);
+	if (strict)
+		write_rest();
 	if (hashcmp(fill(20), sha1))
 		die("final sha1 did not match");
 	use(20);

From 0153be05ae332b8df9bb21d8d249881323e30725 Mon Sep 17 00:00:00 2001
From: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Date: Mon, 25 Feb 2008 22:46:12 +0100
Subject: [PATCH 22/53] index-pack: introduce checking mode

Adds strict option, which bails out if the pack would
introduces broken object or links in the repository.

Signed-off-by: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 Documentation/git-index-pack.txt |  3 ++
 index-pack.c                     | 88 +++++++++++++++++++++++++++++++-
 2 files changed, 90 insertions(+), 1 deletion(-)

diff --git a/Documentation/git-index-pack.txt b/Documentation/git-index-pack.txt
index 72b5d00116..a7825b6144 100644
--- a/Documentation/git-index-pack.txt
+++ b/Documentation/git-index-pack.txt
@@ -75,6 +75,9 @@ OPTIONS
 	to force the version for the generated pack index, and to force
 	64-bit index entries on objects located above the given offset.
 
+--strict::
+	Die, if the pack contains broken objects or links.
+
 
 Note
 ----
diff --git a/index-pack.c b/index-pack.c
index 9fd6982a97..9c0c27813f 100644
--- a/index-pack.c
+++ b/index-pack.c
@@ -7,9 +7,10 @@
 #include "tag.h"
 #include "tree.h"
 #include "progress.h"
+#include "fsck.h"
 
 static const char index_pack_usage[] =
-"git-index-pack [-v] [-o <index-file>] [{ ---keep | --keep=<msg> }] { <pack-file> | --stdin [--fix-thin] [<pack-file>] }";
+"git-index-pack [-v] [-o <index-file>] [{ ---keep | --keep=<msg> }] [--strict] { <pack-file> | --stdin [--fix-thin] [<pack-file>] }";
 
 struct object_entry
 {
@@ -31,6 +32,9 @@ union delta_base {
  */
 #define UNION_BASE_SZ	20
 
+#define FLAG_LINK (1u<<20)
+#define FLAG_CHECKED (1u<<21)
+
 struct delta_entry
 {
 	union delta_base base;
@@ -44,6 +48,7 @@ static int nr_deltas;
 static int nr_resolved_deltas;
 
 static int from_stdin;
+static int strict;
 static int verbose;
 
 static struct progress *progress;
@@ -56,6 +61,48 @@ static SHA_CTX input_ctx;
 static uint32_t input_crc32;
 static int input_fd, output_fd, pack_fd;
 
+static int mark_link(struct object *obj, int type, void *data)
+{
+	if (!obj)
+		return -1;
+
+	if (type != OBJ_ANY && obj->type != type)
+		die("object type mismatch at %s", sha1_to_hex(obj->sha1));
+
+	obj->flags |= FLAG_LINK;
+	return 0;
+}
+
+/* The content of each linked object must have been checked
+   or it must be already present in the object database */
+static void check_object(struct object *obj)
+{
+	if (!obj)
+		return;
+
+	if (!(obj->flags & FLAG_LINK))
+		return;
+
+	if (!(obj->flags & FLAG_CHECKED)) {
+		unsigned long size;
+		int type = sha1_object_info(obj->sha1, &size);
+		if (type != obj->type || type <= 0)
+			die("object of unexpected type");
+		obj->flags |= FLAG_CHECKED;
+		return;
+	}
+}
+
+static void check_objects(void)
+{
+	unsigned i, max;
+
+	max = get_max_object_index();
+	for (i = 0; i < max; i++)
+		check_object(get_indexed_object(i));
+}
+
+
 /* Discard current buffer used content. */
 static void flush(void)
 {
@@ -341,6 +388,41 @@ static void sha1_object(const void *data, unsigned long size,
 			die("SHA1 COLLISION FOUND WITH %s !", sha1_to_hex(sha1));
 		free(has_data);
 	}
+	if (strict) {
+		if (type == OBJ_BLOB) {
+			struct blob *blob = lookup_blob(sha1);
+			if (blob)
+				blob->object.flags |= FLAG_CHECKED;
+			else
+				die("invalid blob object %s", sha1_to_hex(sha1));
+		} else {
+			struct object *obj;
+			int eaten;
+			void *buf = (void *) data;
+
+			/*
+			 * we do not need to free the memory here, as the
+			 * buf is deleted by the caller.
+			 */
+			obj = parse_object_buffer(sha1, type, size, buf, &eaten);
+			if (!obj)
+				die("invalid %s", typename(type));
+			if (fsck_object(obj, 1, fsck_error_function))
+				die("Error in object");
+			if (fsck_walk(obj, mark_link, 0))
+				die("Not all child objects of %s are reachable", sha1_to_hex(obj->sha1));
+
+			if (obj->type == OBJ_TREE) {
+				struct tree *item = (struct tree *) obj;
+				item->buffer = NULL;
+			}
+			if (obj->type == OBJ_COMMIT) {
+				struct commit *commit = (struct commit *) obj;
+				commit->buffer = NULL;
+			}
+			obj->flags |= FLAG_CHECKED;
+		}
+	}
 }
 
 static void resolve_delta(struct object_entry *delta_obj, void *base_data,
@@ -714,6 +796,8 @@ int main(int argc, char **argv)
 				from_stdin = 1;
 			} else if (!strcmp(arg, "--fix-thin")) {
 				fix_thin_pack = 1;
+			} else if (!strcmp(arg, "--strict")) {
+				strict = 1;
 			} else if (!strcmp(arg, "--keep")) {
 				keep_msg = "";
 			} else if (!prefixcmp(arg, "--keep=")) {
@@ -812,6 +896,8 @@ int main(int argc, char **argv)
 			    nr_deltas - nr_resolved_deltas);
 	}
 	free(deltas);
+	if (strict)
+		check_objects();
 
 	idx_objects = xmalloc((nr_objects) * sizeof(struct pack_idx_entry *));
 	for (i = 0; i < nr_objects; i++)

From 28f72a0f232dfc71b3be726e7e71d0a6d5f9ebba Mon Sep 17 00:00:00 2001
From: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Date: Mon, 25 Feb 2008 22:46:13 +0100
Subject: [PATCH 23/53] receive-pack: use strict mode for unpacking objects

Signed-off-by: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 Documentation/config.txt |  6 ++++++
 receive-pack.c           | 36 +++++++++++++++++++++++-------------
 2 files changed, 29 insertions(+), 13 deletions(-)

diff --git a/Documentation/config.txt b/Documentation/config.txt
index 6d8cca46ab..72d3a345ec 100644
--- a/Documentation/config.txt
+++ b/Documentation/config.txt
@@ -860,6 +860,12 @@ imap::
 	The configuration variables in the 'imap' section are described
 	in linkgit:git-imap-send[1].
 
+receive.fsckObjects::
+	If it is set to true, git-receive-pack will check all received
+	objects. It will abort in the case of a malformed object or a
+	broken link. The result of an abort are only dangling objects.
+	The default value is true.
+
 receive.unpackLimit::
 	If the number of objects received in a push is below this
 	limit then the objects will be unpacked into loose object
diff --git a/receive-pack.c b/receive-pack.c
index 3267495832..f5440ff4d4 100644
--- a/receive-pack.c
+++ b/receive-pack.c
@@ -10,6 +10,7 @@
 static const char receive_pack_usage[] = "git-receive-pack <git-dir>";
 
 static int deny_non_fast_forwards = 0;
+static int receive_fsck_objects = 1;
 static int receive_unpack_limit = -1;
 static int transfer_unpack_limit = -1;
 static int unpack_limit = 100;
@@ -35,6 +36,11 @@ static int receive_pack_config(const char *var, const char *value)
 		return 0;
 	}
 
+	if (strcmp(var, "receive.fsckobjects") == 0) {
+		receive_fsck_objects = git_config_bool(var, value);
+		return 0;
+	}
+
 	return git_default_config(var, value);
 }
 
@@ -367,11 +373,13 @@ static const char *unpack(void)
 			ntohl(hdr.hdr_version), ntohl(hdr.hdr_entries));
 
 	if (ntohl(hdr.hdr_entries) < unpack_limit) {
-		int code;
-		const char *unpacker[3];
-		unpacker[0] = "unpack-objects";
-		unpacker[1] = hdr_arg;
-		unpacker[2] = NULL;
+		int code, i = 0;
+		const char *unpacker[4];
+		unpacker[i++] = "unpack-objects";
+		if (receive_fsck_objects)
+			unpacker[i++] = "--strict";
+		unpacker[i++] = hdr_arg;
+		unpacker[i++] = NULL;
 		code = run_command_v_opt(unpacker, RUN_GIT_CMD);
 		switch (code) {
 		case 0:
@@ -392,8 +400,8 @@ static const char *unpack(void)
 			return "unpacker exited with error code";
 		}
 	} else {
-		const char *keeper[6];
-		int s, status;
+		const char *keeper[7];
+		int s, status, i = 0;
 		char keep_arg[256];
 		struct child_process ip;
 
@@ -401,12 +409,14 @@ static const char *unpack(void)
 		if (gethostname(keep_arg + s, sizeof(keep_arg) - s))
 			strcpy(keep_arg + s, "localhost");
 
-		keeper[0] = "index-pack";
-		keeper[1] = "--stdin";
-		keeper[2] = "--fix-thin";
-		keeper[3] = hdr_arg;
-		keeper[4] = keep_arg;
-		keeper[5] = NULL;
+		keeper[i++] = "index-pack";
+		keeper[i++] = "--stdin";
+		if (receive_fsck_objects)
+			keeper[i++] = "--strict";
+		keeper[i++] = "--fix-thin";
+		keeper[i++] = hdr_arg;
+		keeper[i++] = keep_arg;
+		keeper[i++] = NULL;
 		memset(&ip, 0, sizeof(ip));
 		ip.argv = keeper;
 		ip.out = -1;

From fbbbc362ab9d3a8d76eb4273e65fb1fb26849d75 Mon Sep 17 00:00:00 2001
From: "Philippe Bruhat (BooK" <book@cpan.org>
Date: Thu, 28 Feb 2008 11:18:21 +0100
Subject: [PATCH 24/53] cvsimport: have default merge regex allow for dashes in
 the branch name

The default value of @mergerx uses \w, which matches word
character; a branch name like policy-20050608-br will not be
matched.

Signed-off-by: Philippe Bruhat (BooK) <book@cpan.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 git-cvsimport.perl | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/git-cvsimport.perl b/git-cvsimport.perl
index 9516242338..3d013a7d62 100755
--- a/git-cvsimport.perl
+++ b/git-cvsimport.perl
@@ -164,7 +164,7 @@ if ($#ARGV == 0) {
 
 our @mergerx = ();
 if ($opt_m) {
-	@mergerx = ( qr/\b(?:from|of|merge|merging|merged) (\w+)/i );
+	@mergerx = ( qr/\b(?:from|of|merge|merging|merged) ([-\w]+)/i );
 }
 if ($opt_M) {
 	push (@mergerx, qr/$opt_M/);

From bc434e829c6f1757d2761c50e06a28278dc2f2c7 Mon Sep 17 00:00:00 2001
From: "Philippe Bruhat (BooK" <book@cpan.org>
Date: Thu, 28 Feb 2008 11:18:22 +0100
Subject: [PATCH 25/53] cvsimport: allow for multiple -M options

Use Getopt::Long instead of Getopt::Std to handle multiple -M options,
for all the cases when having a single custom regex is not enough.

Signed-off-by: Philippe Bruhat (BooK) <book@cpan.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 git-cvsimport.perl | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)

diff --git a/git-cvsimport.perl b/git-cvsimport.perl
index 3d013a7d62..47f116f37e 100755
--- a/git-cvsimport.perl
+++ b/git-cvsimport.perl
@@ -15,7 +15,7 @@
 
 use strict;
 use warnings;
-use Getopt::Std;
+use Getopt::Long;
 use File::Spec;
 use File::Temp qw(tempfile tmpnam);
 use File::Path qw(mkpath);
@@ -29,7 +29,7 @@ use IPC::Open2;
 $SIG{'PIPE'}="IGNORE";
 $ENV{'TZ'}="UTC";
 
-our ($opt_h,$opt_o,$opt_v,$opt_k,$opt_u,$opt_d,$opt_p,$opt_C,$opt_z,$opt_i,$opt_P, $opt_s,$opt_m,$opt_M,$opt_A,$opt_S,$opt_L, $opt_a, $opt_r);
+our ($opt_h,$opt_o,$opt_v,$opt_k,$opt_u,$opt_d,$opt_p,$opt_C,$opt_z,$opt_i,$opt_P, $opt_s,$opt_m,@opt_M,$opt_A,$opt_S,$opt_L, $opt_a, $opt_r);
 my (%conv_author_name, %conv_author_email);
 
 sub usage(;$) {
@@ -112,7 +112,12 @@ sub read_repo_config {
 
 my $opts = "haivmkuo:d:p:r:C:z:s:M:P:A:S:L:";
 read_repo_config($opts);
-getopts($opts) or usage();
+Getopt::Long::Configure( 'no_ignore_case', 'bundling' );
+
+# turn the Getopt::Std specification in a Getopt::Long one,
+# with support for multiple -M options
+GetOptions( map { s/:/=s/; /M/ ? "$_\@" : $_ } split( /(?!:)/, $opts ) )
+    or usage();
 usage if $opt_h;
 
 if (@ARGV == 0) {
@@ -166,8 +171,8 @@ our @mergerx = ();
 if ($opt_m) {
 	@mergerx = ( qr/\b(?:from|of|merge|merging|merged) ([-\w]+)/i );
 }
-if ($opt_M) {
-	push (@mergerx, qr/$opt_M/);
+if (@opt_M) {
+	push (@mergerx, map { qr/$_/ } @opt_M);
 }
 
 # Remember UTC of our starting time

From 3c832a78b15f8ffcb4e6f1f6f6d8b7a607ba5d06 Mon Sep 17 00:00:00 2001
From: "Philippe Bruhat (BooK" <book@cpan.org>
Date: Thu, 28 Feb 2008 11:18:23 +0100
Subject: [PATCH 26/53] cvsimport: document that -M can be used multiple times

Also document the capture behaviour (source branch name in $1)

Signed-off-by: Philippe Bruhat (BooK) <book@cpan.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 Documentation/git-cvsimport.txt | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/Documentation/git-cvsimport.txt b/Documentation/git-cvsimport.txt
index 6f91b9ea2a..58eefd42e5 100644
--- a/Documentation/git-cvsimport.txt
+++ b/Documentation/git-cvsimport.txt
@@ -102,13 +102,17 @@ If you need to pass multiple options, separate them with a comma.
 
 -m::
 	Attempt to detect merges based on the commit message. This option
-	will enable default regexes that try to capture the name source
+	will enable default regexes that try to capture the source
 	branch name from the commit message.
 
 -M <regex>::
 	Attempt to detect merges based on the commit message with a custom
 	regex. It can be used with '-m' to enable the default regexes
 	as well. You must escape forward slashes.
++
+The regex must capture the source branch name in $1.
++
+This option can be used several times to provide several detection regexes.
 
 -S <regex>::
 	Skip paths matching the regex.

From 3449f8c4cb93a0ec445db22ee7549d0b051446d6 Mon Sep 17 00:00:00 2001
From: Nicolas Pitre <nico@cam.org>
Date: Thu, 28 Feb 2008 00:25:17 -0500
Subject: [PATCH 27/53] factorize revindex code out of builtin-pack-objects.c

No functional change. This is needed to fix verify-pack in a later patch.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 Makefile               |   5 +-
 builtin-pack-objects.c | 160 +++--------------------------------------
 pack-revindex.c        | 142 ++++++++++++++++++++++++++++++++++++
 pack-revindex.h        |  12 ++++
 4 files changed, 167 insertions(+), 152 deletions(-)
 create mode 100644 pack-revindex.c
 create mode 100644 pack-revindex.h

diff --git a/Makefile b/Makefile
index a5b6eebf1a..1acac02766 100644
--- a/Makefile
+++ b/Makefile
@@ -304,7 +304,8 @@ LIB_H = \
 	run-command.h strbuf.h tag.h tree.h git-compat-util.h revision.h \
 	tree-walk.h log-tree.h dir.h path-list.h unpack-trees.h builtin.h \
 	utf8.h reflog-walk.h patch-ids.h attr.h decorate.h progress.h \
-	mailmap.h remote.h parse-options.h transport.h diffcore.h hash.h
+	mailmap.h remote.h parse-options.h transport.h diffcore.h hash.h \
+	pack-revindex.h
 
 DIFF_OBJS = \
 	diff.o diff-lib.o diffcore-break.o diffcore-order.o \
@@ -328,7 +329,7 @@ LIB_OBJS = \
 	color.o wt-status.o archive-zip.o archive-tar.o shallow.o utf8.o \
 	convert.o attr.o decorate.o progress.o mailmap.o symlinks.o remote.o \
 	transport.o bundle.o walker.o parse-options.o ws.o archive.o branch.o \
-	alias.o
+	alias.o pack-revindex.o
 
 BUILTIN_OBJS = \
 	builtin-add.o \
diff --git a/builtin-pack-objects.c b/builtin-pack-objects.c
index 1bba6e6a64..c7834bb368 100644
--- a/builtin-pack-objects.c
+++ b/builtin-pack-objects.c
@@ -8,6 +8,7 @@
 #include "tree.h"
 #include "delta.h"
 #include "pack.h"
+#include "pack-revindex.h"
 #include "csum-file.h"
 #include "tree-walk.h"
 #include "diff.h"
@@ -92,158 +93,12 @@ static unsigned long window_memory_limit = 0;
 static int *object_ix;
 static int object_ix_hashsz;
 
-/*
- * Pack index for existing packs give us easy access to the offsets into
- * corresponding pack file where each object's data starts, but the entries
- * do not store the size of the compressed representation (uncompressed
- * size is easily available by examining the pack entry header).  It is
- * also rather expensive to find the sha1 for an object given its offset.
- *
- * We build a hashtable of existing packs (pack_revindex), and keep reverse
- * index here -- pack index file is sorted by object name mapping to offset;
- * this pack_revindex[].revindex array is a list of offset/index_nr pairs
- * ordered by offset, so if you know the offset of an object, next offset
- * is where its packed representation ends and the index_nr can be used to
- * get the object sha1 from the main index.
- */
-struct revindex_entry {
-	off_t offset;
-	unsigned int nr;
-};
-struct pack_revindex {
-	struct packed_git *p;
-	struct revindex_entry *revindex;
-};
-static struct  pack_revindex *pack_revindex;
-static int pack_revindex_hashsz;
-
 /*
  * stats
  */
 static uint32_t written, written_delta;
 static uint32_t reused, reused_delta;
 
-static int pack_revindex_ix(struct packed_git *p)
-{
-	unsigned long ui = (unsigned long)p;
-	int i;
-
-	ui = ui ^ (ui >> 16); /* defeat structure alignment */
-	i = (int)(ui % pack_revindex_hashsz);
-	while (pack_revindex[i].p) {
-		if (pack_revindex[i].p == p)
-			return i;
-		if (++i == pack_revindex_hashsz)
-			i = 0;
-	}
-	return -1 - i;
-}
-
-static void prepare_pack_ix(void)
-{
-	int num;
-	struct packed_git *p;
-	for (num = 0, p = packed_git; p; p = p->next)
-		num++;
-	if (!num)
-		return;
-	pack_revindex_hashsz = num * 11;
-	pack_revindex = xcalloc(sizeof(*pack_revindex), pack_revindex_hashsz);
-	for (p = packed_git; p; p = p->next) {
-		num = pack_revindex_ix(p);
-		num = - 1 - num;
-		pack_revindex[num].p = p;
-	}
-	/* revindex elements are lazily initialized */
-}
-
-static int cmp_offset(const void *a_, const void *b_)
-{
-	const struct revindex_entry *a = a_;
-	const struct revindex_entry *b = b_;
-	return (a->offset < b->offset) ? -1 : (a->offset > b->offset) ? 1 : 0;
-}
-
-/*
- * Ordered list of offsets of objects in the pack.
- */
-static void prepare_pack_revindex(struct pack_revindex *rix)
-{
-	struct packed_git *p = rix->p;
-	int num_ent = p->num_objects;
-	int i;
-	const char *index = p->index_data;
-
-	rix->revindex = xmalloc(sizeof(*rix->revindex) * (num_ent + 1));
-	index += 4 * 256;
-
-	if (p->index_version > 1) {
-		const uint32_t *off_32 =
-			(uint32_t *)(index + 8 + p->num_objects * (20 + 4));
-		const uint32_t *off_64 = off_32 + p->num_objects;
-		for (i = 0; i < num_ent; i++) {
-			uint32_t off = ntohl(*off_32++);
-			if (!(off & 0x80000000)) {
-				rix->revindex[i].offset = off;
-			} else {
-				rix->revindex[i].offset =
-					((uint64_t)ntohl(*off_64++)) << 32;
-				rix->revindex[i].offset |=
-					ntohl(*off_64++);
-			}
-			rix->revindex[i].nr = i;
-		}
-	} else {
-		for (i = 0; i < num_ent; i++) {
-			uint32_t hl = *((uint32_t *)(index + 24 * i));
-			rix->revindex[i].offset = ntohl(hl);
-			rix->revindex[i].nr = i;
-		}
-	}
-
-	/* This knows the pack format -- the 20-byte trailer
-	 * follows immediately after the last object data.
-	 */
-	rix->revindex[num_ent].offset = p->pack_size - 20;
-	rix->revindex[num_ent].nr = -1;
-	qsort(rix->revindex, num_ent, sizeof(*rix->revindex), cmp_offset);
-}
-
-static struct revindex_entry * find_packed_object(struct packed_git *p,
-						  off_t ofs)
-{
-	int num;
-	int lo, hi;
-	struct pack_revindex *rix;
-	struct revindex_entry *revindex;
-	num = pack_revindex_ix(p);
-	if (num < 0)
-		die("internal error: pack revindex uninitialized");
-	rix = &pack_revindex[num];
-	if (!rix->revindex)
-		prepare_pack_revindex(rix);
-	revindex = rix->revindex;
-	lo = 0;
-	hi = p->num_objects + 1;
-	do {
-		int mi = (lo + hi) / 2;
-		if (revindex[mi].offset == ofs) {
-			return revindex + mi;
-		}
-		else if (ofs < revindex[mi].offset)
-			hi = mi;
-		else
-			lo = mi + 1;
-	} while (lo < hi);
-	die("internal error: pack revindex corrupt");
-}
-
-static const unsigned char *find_packed_object_name(struct packed_git *p,
-						    off_t ofs)
-{
-	struct revindex_entry *entry = find_packed_object(p, ofs);
-	return nth_packed_object_sha1(p, entry->nr);
-}
 
 static void *delta_against(void *buf, unsigned long size, struct object_entry *entry)
 {
@@ -510,7 +365,7 @@ static unsigned long write_object(struct sha1file *f,
 		}
 		hdrlen = encode_header(obj_type, entry->size, header);
 		offset = entry->in_pack_offset;
-		revidx = find_packed_object(p, offset);
+		revidx = find_pack_revindex(p, offset);
 		datalen = revidx[1].offset - offset;
 		if (!pack_to_stdout && p->index_version > 1 &&
 		    check_pack_crc(p, &w_curs, offset, datalen, revidx->nr))
@@ -1162,8 +1017,11 @@ static void check_object(struct object_entry *entry)
 				die("delta base offset out of bound for %s",
 				    sha1_to_hex(entry->idx.sha1));
 			ofs = entry->in_pack_offset - ofs;
-			if (!no_reuse_delta && !entry->preferred_base)
-				base_ref = find_packed_object_name(p, ofs);
+			if (!no_reuse_delta && !entry->preferred_base) {
+				struct revindex_entry *revidx;
+				revidx = find_pack_revindex(p, ofs);
+				base_ref = nth_packed_object_sha1(p, revidx->nr);
+			}
 			entry->in_pack_header_size = used + used_0;
 			break;
 		}
@@ -1240,9 +1098,11 @@ static void get_object_details(void)
 		sorted_by_offset[i] = objects + i;
 	qsort(sorted_by_offset, nr_objects, sizeof(*sorted_by_offset), pack_offset_sort);
 
-	prepare_pack_ix();
+	init_pack_revindex();
+
 	for (i = 0; i < nr_objects; i++)
 		check_object(sorted_by_offset[i]);
+
 	free(sorted_by_offset);
 }
 
diff --git a/pack-revindex.c b/pack-revindex.c
new file mode 100644
index 0000000000..a8aa2cd6ca
--- /dev/null
+++ b/pack-revindex.c
@@ -0,0 +1,142 @@
+#include "cache.h"
+#include "pack-revindex.h"
+
+/*
+ * Pack index for existing packs give us easy access to the offsets into
+ * corresponding pack file where each object's data starts, but the entries
+ * do not store the size of the compressed representation (uncompressed
+ * size is easily available by examining the pack entry header).  It is
+ * also rather expensive to find the sha1 for an object given its offset.
+ *
+ * We build a hashtable of existing packs (pack_revindex), and keep reverse
+ * index here -- pack index file is sorted by object name mapping to offset;
+ * this pack_revindex[].revindex array is a list of offset/index_nr pairs
+ * ordered by offset, so if you know the offset of an object, next offset
+ * is where its packed representation ends and the index_nr can be used to
+ * get the object sha1 from the main index.
+ */
+
+struct pack_revindex {
+	struct packed_git *p;
+	struct revindex_entry *revindex;
+};
+
+static struct pack_revindex *pack_revindex;
+static int pack_revindex_hashsz;
+
+static int pack_revindex_ix(struct packed_git *p)
+{
+	unsigned long ui = (unsigned long)p;
+	int i;
+
+	ui = ui ^ (ui >> 16); /* defeat structure alignment */
+	i = (int)(ui % pack_revindex_hashsz);
+	while (pack_revindex[i].p) {
+		if (pack_revindex[i].p == p)
+			return i;
+		if (++i == pack_revindex_hashsz)
+			i = 0;
+	}
+	return -1 - i;
+}
+
+void init_pack_revindex(void)
+{
+	int num;
+	struct packed_git *p;
+
+	for (num = 0, p = packed_git; p; p = p->next)
+		num++;
+	if (!num)
+		return;
+	pack_revindex_hashsz = num * 11;
+	pack_revindex = xcalloc(sizeof(*pack_revindex), pack_revindex_hashsz);
+	for (p = packed_git; p; p = p->next) {
+		num = pack_revindex_ix(p);
+		num = - 1 - num;
+		pack_revindex[num].p = p;
+	}
+	/* revindex elements are lazily initialized */
+}
+
+static int cmp_offset(const void *a_, const void *b_)
+{
+	const struct revindex_entry *a = a_;
+	const struct revindex_entry *b = b_;
+	return (a->offset < b->offset) ? -1 : (a->offset > b->offset) ? 1 : 0;
+}
+
+/*
+ * Ordered list of offsets of objects in the pack.
+ */
+static void create_pack_revindex(struct pack_revindex *rix)
+{
+	struct packed_git *p = rix->p;
+	int num_ent = p->num_objects;
+	int i;
+	const char *index = p->index_data;
+
+	rix->revindex = xmalloc(sizeof(*rix->revindex) * (num_ent + 1));
+	index += 4 * 256;
+
+	if (p->index_version > 1) {
+		const uint32_t *off_32 =
+			(uint32_t *)(index + 8 + p->num_objects * (20 + 4));
+		const uint32_t *off_64 = off_32 + p->num_objects;
+		for (i = 0; i < num_ent; i++) {
+			uint32_t off = ntohl(*off_32++);
+			if (!(off & 0x80000000)) {
+				rix->revindex[i].offset = off;
+			} else {
+				rix->revindex[i].offset =
+					((uint64_t)ntohl(*off_64++)) << 32;
+				rix->revindex[i].offset |=
+					ntohl(*off_64++);
+			}
+			rix->revindex[i].nr = i;
+		}
+	} else {
+		for (i = 0; i < num_ent; i++) {
+			uint32_t hl = *((uint32_t *)(index + 24 * i));
+			rix->revindex[i].offset = ntohl(hl);
+			rix->revindex[i].nr = i;
+		}
+	}
+
+	/* This knows the pack format -- the 20-byte trailer
+	 * follows immediately after the last object data.
+	 */
+	rix->revindex[num_ent].offset = p->pack_size - 20;
+	rix->revindex[num_ent].nr = -1;
+	qsort(rix->revindex, num_ent, sizeof(*rix->revindex), cmp_offset);
+}
+
+struct revindex_entry *find_pack_revindex(struct packed_git *p, off_t ofs)
+{
+	int num;
+	int lo, hi;
+	struct pack_revindex *rix;
+	struct revindex_entry *revindex;
+
+	num = pack_revindex_ix(p);
+	if (num < 0)
+		die("internal error: pack revindex uninitialized");
+
+	rix = &pack_revindex[num];
+	if (!rix->revindex)
+		create_pack_revindex(rix);
+	revindex = rix->revindex;
+
+	lo = 0;
+	hi = p->num_objects + 1;
+	do {
+		int mi = (lo + hi) / 2;
+		if (revindex[mi].offset == ofs) {
+			return revindex + mi;
+		} else if (ofs < revindex[mi].offset)
+			hi = mi;
+		else
+			lo = mi + 1;
+	} while (lo < hi);
+	die("internal error: pack revindex corrupt");
+}
diff --git a/pack-revindex.h b/pack-revindex.h
new file mode 100644
index 0000000000..c3527a7565
--- /dev/null
+++ b/pack-revindex.h
@@ -0,0 +1,12 @@
+#ifndef PACK_REVINDEX_H
+#define PACK_REVINDEX_H
+
+struct revindex_entry {
+	off_t offset;
+	unsigned int nr;
+};
+
+void init_pack_revindex(void);
+struct revindex_entry *find_pack_revindex(struct packed_git *p, off_t ofs);
+
+#endif

From 340814636dde3cbd2e461b12f9ae832d2100766a Mon Sep 17 00:00:00 2001
From: Nicolas Pitre <nico@cam.org>
Date: Thu, 28 Feb 2008 00:25:18 -0500
Subject: [PATCH 28/53] make verify_one_pack() a bit less wrong wrt packed_git
 structure

Simply freeing it is wrong.  There are many things attached to this
structure that are not cleaned up.  In practice this doesn't matter much
since this happens just before the program exits, but it is still
a bit more "correct" to leak it implicitly rather than explicitly.

And therefore it is also a good idea to register it with
install_packed_git().  Not only might it have better chance of being
properly cleaned up if such functionality is implemented for the general
case, but some functions like init_revindex() expect all packed_git
instances to be globally accessible.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 builtin-verify-pack.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/builtin-verify-pack.c b/builtin-verify-pack.c
index 4e31c273f4..4958bbbf11 100644
--- a/builtin-verify-pack.c
+++ b/builtin-verify-pack.c
@@ -40,8 +40,8 @@ static int verify_one_pack(const char *path, int verbose)
 	if (!pack)
 		return error("packfile %s not found.", arg);
 
+	install_packed_git(pack);
 	err = verify_pack(pack, verbose);
-	free(pack);
 
 	return err;
 }

From 70f5d5d31cbf0ba273083805124ad67135142527 Mon Sep 17 00:00:00 2001
From: Nicolas Pitre <nico@cam.org>
Date: Thu, 28 Feb 2008 00:25:19 -0500
Subject: [PATCH 29/53] fix unimplemented packed_object_info_detail() features

Since commit eb32d236df0c16b936b04f0c5402addb61cdb311, there was a TODO
comment in packed_object_info_detail() about the SHA1 of base object to
OBJ_OFS_DELTA objects.  So here it is at last.

While at it, providing the actual storage size information as well is now
trivial.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 pack-check.c |  3 +++
 sha1_file.c  | 10 +++++++---
 2 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/pack-check.c b/pack-check.c
index d7dd62bb83..ae25685036 100644
--- a/pack-check.c
+++ b/pack-check.c
@@ -1,5 +1,6 @@
 #include "cache.h"
 #include "pack.h"
+#include "pack-revindex.h"
 
 struct idx_entry
 {
@@ -101,8 +102,10 @@ static int verify_packfile(struct packed_git *p,
 static void show_pack_info(struct packed_git *p)
 {
 	uint32_t nr_objects, i, chain_histogram[MAX_CHAIN+1];
+
 	nr_objects = p->num_objects;
 	memset(chain_histogram, 0, sizeof(chain_histogram));
+	init_pack_revindex();
 
 	for (i = 0; i < nr_objects; i++) {
 		const unsigned char *sha1;
diff --git a/sha1_file.c b/sha1_file.c
index 1ddb96bb82..445a871db3 100644
--- a/sha1_file.c
+++ b/sha1_file.c
@@ -14,6 +14,7 @@
 #include "tag.h"
 #include "tree.h"
 #include "refs.h"
+#include "pack-revindex.h"
 
 #ifndef O_NOATIME
 #if defined(__linux__) && (defined(__i386__) || defined(__PPC__))
@@ -1367,11 +1368,15 @@ const char *packed_object_info_detail(struct packed_git *p,
 	unsigned long dummy;
 	unsigned char *next_sha1;
 	enum object_type type;
+	struct revindex_entry *revidx;
 
 	*delta_chain_length = 0;
 	curpos = obj_offset;
 	type = unpack_object_header(p, &w_curs, &curpos, size);
 
+	revidx = find_pack_revindex(p, obj_offset);
+	*store_size = revidx[1].offset - obj_offset;
+
 	for (;;) {
 		switch (type) {
 		default:
@@ -1381,14 +1386,13 @@ const char *packed_object_info_detail(struct packed_git *p,
 		case OBJ_TREE:
 		case OBJ_BLOB:
 		case OBJ_TAG:
-			*store_size = 0; /* notyet */
 			unuse_pack(&w_curs);
 			return typename(type);
 		case OBJ_OFS_DELTA:
 			obj_offset = get_delta_base(p, &w_curs, &curpos, type, obj_offset);
 			if (*delta_chain_length == 0) {
-				/* TODO: find base_sha1 as pointed by curpos */
-				hashclr(base_sha1);
+				revidx = find_pack_revindex(p, obj_offset);
+				hashcpy(base_sha1, nth_packed_object_sha1(p, revidx->nr));
 			}
 			break;
 		case OBJ_REF_DELTA:

From 5f4347bba39ddb147b06913ac263fc46954d2d0b Mon Sep 17 00:00:00 2001
From: Nicolas Pitre <nico@cam.org>
Date: Thu, 28 Feb 2008 00:25:20 -0500
Subject: [PATCH 30/53] add storage size output to 'git verify-pack -v'

This can possibly break external scripts that depend on the previous
output, but those script can't possibly be critical to Git usage, and
fixing them should be trivial.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 Documentation/git-verify-pack.txt | 4 ++--
 contrib/stats/packinfo.pl         | 2 +-
 pack-check.c                      | 8 ++++----
 3 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/Documentation/git-verify-pack.txt b/Documentation/git-verify-pack.txt
index db019a2b8d..ba2a157299 100644
--- a/Documentation/git-verify-pack.txt
+++ b/Documentation/git-verify-pack.txt
@@ -32,11 +32,11 @@ OUTPUT FORMAT
 -------------
 When specifying the -v option the format used is:
 
-	SHA1 type size offset-in-packfile
+	SHA1 type size size-in-pack-file offset-in-packfile
 
 for objects that are not deltified in the pack, and
 
-	SHA1 type size offset-in-packfile depth base-SHA1
+	SHA1 type size size-in-packfile offset-in-packfile depth base-SHA1
 
 for objects that are deltified.
 
diff --git a/contrib/stats/packinfo.pl b/contrib/stats/packinfo.pl
index aab501ea08..f4a7b62cd9 100755
--- a/contrib/stats/packinfo.pl
+++ b/contrib/stats/packinfo.pl
@@ -93,7 +93,7 @@ my %depths;
 my @depths;
 
 while (<STDIN>) {
-    my ($sha1, $type, $size, $offset, $depth, $parent) = split(/\s+/, $_);
+    my ($sha1, $type, $size, $space, $offset, $depth, $parent) = split(/\s+/, $_);
     next unless ($sha1 =~ /^[0-9a-f]{40}$/);
     $depths{$sha1} = $depth || 0;
     push(@depths, $depth || 0);
diff --git a/pack-check.c b/pack-check.c
index ae25685036..0f8ad2c00f 100644
--- a/pack-check.c
+++ b/pack-check.c
@@ -128,11 +128,11 @@ static void show_pack_info(struct packed_git *p)
 						 base_sha1);
 		printf("%s ", sha1_to_hex(sha1));
 		if (!delta_chain_length)
-			printf("%-6s %lu %"PRIuMAX"\n",
-			       type, size, (uintmax_t)offset);
+			printf("%-6s %lu %lu %"PRIuMAX"\n",
+			       type, size, store_size, (uintmax_t)offset);
 		else {
-			printf("%-6s %lu %"PRIuMAX" %u %s\n",
-			       type, size, (uintmax_t)offset,
+			printf("%-6s %lu %lu %"PRIuMAX" %u %s\n",
+			       type, size, store_size, (uintmax_t)offset,
 			       delta_chain_length, sha1_to_hex(base_sha1));
 			if (delta_chain_length <= MAX_CHAIN)
 				chain_histogram[delta_chain_length]++;

From 009c98ee170b36598548cc67c54826544cced108 Mon Sep 17 00:00:00 2001
From: Junio C Hamano <gitster@pobox.com>
Date: Sat, 1 Mar 2008 18:18:16 -0800
Subject: [PATCH 31/53] CodingGuidelines: spell out how we use grep in our
 scripts

Our scripts try to stick to fairly limited subset of POSIX BRE for
portability.  It is unclear from manual page from GNU grep which is GNU
extension and which is portable, so let's spell it out to help new people
to keep their contributions from hurting porters.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 Documentation/CodingGuidelines | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/Documentation/CodingGuidelines b/Documentation/CodingGuidelines
index 3b042db624..994eb9159a 100644
--- a/Documentation/CodingGuidelines
+++ b/Documentation/CodingGuidelines
@@ -53,6 +53,18 @@ For shell scripts specifically (not exhaustive):
  - We do not write the noiseword "function" in front of shell
    functions.
 
+ - As to use of grep, stick to a subset of BRE (namely, no \{m,n\},
+   [::], [==], nor [..]) for portability.
+
+   - We do not use \{m,n\};
+
+   - We do not use -E;
+
+   - We do not use ? nor + (which are \{0,1\} and \{1,\}
+     respectively in BRE) but that goes without saying as these
+     are ERE elements not BRE (note that \? and \+ are not even part
+     of BRE -- making them accessible from BRE is a GNU extension).
+
 For C programs:
 
  - We use tabs to indent, and interpret tabs as taking up to

From f32086becc2b46bda0680ef178935546a64ad693 Mon Sep 17 00:00:00 2001
From: Mike Hommey <mh@glandium.org>
Date: Sat, 1 Mar 2008 12:39:52 +0100
Subject: [PATCH 32/53] Documentation/git-rebase.txt: Add --strategy to
 synopsys

Signed-off-by: Mike Hommey <mh@glandium.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 Documentation/git-rebase.txt | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/git-rebase.txt b/Documentation/git-rebase.txt
index c11c6453ea..4b10304740 100644
--- a/Documentation/git-rebase.txt
+++ b/Documentation/git-rebase.txt
@@ -9,6 +9,7 @@ SYNOPSIS
 --------
 [verse]
 'git-rebase' [-i | --interactive] [-v | --verbose] [-m | --merge]
+	[-s <strategy> | --strategy=<strategy>]
 	[-C<n>] [ --whitespace=<option>] [-p | --preserve-merges]
 	[--onto <newbase>] <upstream> [<branch>]
 'git-rebase' --continue | --skip | --abort

From 48411d2233aec65b612ac2b1f50540ec60f0f568 Mon Sep 17 00:00:00 2001
From: Mike Hommey <mh@glandium.org>
Date: Sat, 1 Mar 2008 11:32:14 +0100
Subject: [PATCH 33/53] git rebase --abort: always restore the right commit

Previously, --abort would end by git resetting to ORIG_HEAD, but some
commands, such as git reset --hard (which happened in git rebase --skip,
but could just as well be typed by the user), would have already modified
ORIG_HEAD.

Just use the orig-head we store in $dotest instead.

Signed-off-by: Mike Hommey <mh@glandium.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 git-rebase.sh           | 5 ++---
 t/t3407-rebase-abort.sh | 2 +-
 2 files changed, 3 insertions(+), 4 deletions(-)

diff --git a/git-rebase.sh b/git-rebase.sh
index bdcea0ed70..6b9af962a9 100755
--- a/git-rebase.sh
+++ b/git-rebase.sh
@@ -208,16 +208,15 @@ do
 		if test -d "$dotest"
 		then
 			move_to_original_branch
-			rm -r "$dotest"
 		elif test -d .dotest
 		then
 			dotest=.dotest
 			move_to_original_branch
-			rm -r .dotest
 		else
 			die "No rebase in progress?"
 		fi
-		git reset --hard ORIG_HEAD
+		git reset --hard $(cat $dotest/orig-head)
+		rm -r "$dotest"
 		exit
 		;;
 	--onto)
diff --git a/t/t3407-rebase-abort.sh b/t/t3407-rebase-abort.sh
index 94bdd72330..3417138a80 100755
--- a/t/t3407-rebase-abort.sh
+++ b/t/t3407-rebase-abort.sh
@@ -29,7 +29,7 @@ test_expect_success 'rebase --abort' '
 	test $(git rev-parse to-rebase) = $(git rev-parse pre-rebase)
 '
 
-test_expect_failure 'rebase --abort after --skip' '
+test_expect_success 'rebase --abort after --skip' '
 	# Clean up the state from the previous one
 	git reset --hard pre-rebase
 	rm -rf .dotest

From b66fde9a287f64ed81d4d4a1639997ad5eedb432 Mon Sep 17 00:00:00 2001
From: Junio C Hamano <gitster@pobox.com>
Date: Sat, 1 Mar 2008 23:35:32 -0800
Subject: [PATCH 34/53] find_unique_abbrev(): redefine semantics

The function returned NULL when no object that matches the name
was found, but that made the callers more complicated, as nobody
used that NULL return as an indication that no object with such
a name exists.  They (at least the careful ones) instead took
the full 40-hexdigit and used in such a case, and the careless
ones segfaulted.

With this "git rev-parse --short 5555555555555555555555555555555555555555"
would stop segfaulting.

This is based on Jeff King's rewrite to my RFC patch, but "missing"
logic swapped to "exists".  The final logic reads:

    For existing objects, make sure the abbreviated string uniquely
    identifies it.  Otherwise, make sure the abbreviated string is
    long enough so that it would not name any existing object.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 sha1_name.c | 13 ++++++-------
 1 file changed, 6 insertions(+), 7 deletions(-)

diff --git a/sha1_name.c b/sha1_name.c
index 9d088cc2ca..8358ba2069 100644
--- a/sha1_name.c
+++ b/sha1_name.c
@@ -192,26 +192,25 @@ static int get_short_sha1(const char *name, int len, unsigned char *sha1,
 
 const char *find_unique_abbrev(const unsigned char *sha1, int len)
 {
-	int status, is_null;
+	int status, exists;
 	static char hex[41];
 
-	is_null = is_null_sha1(sha1);
+	exists = has_sha1_file(sha1);
 	memcpy(hex, sha1_to_hex(sha1), 40);
 	if (len == 40 || !len)
 		return hex;
 	while (len < 40) {
 		unsigned char sha1_ret[20];
 		status = get_short_sha1(hex, len, sha1_ret, 1);
-		if (!status ||
-		    (is_null && status != SHORT_NAME_AMBIGUOUS)) {
+		if (exists
+		    ? !status
+		    : status == SHORT_NAME_NOT_FOUND) {
 			hex[len] = 0;
 			return hex;
 		}
-		if (status != SHORT_NAME_AMBIGUOUS)
-			return NULL;
 		len++;
 	}
-	return NULL;
+	return hex;
 }
 
 static int ambiguous_path(const char *path, int len)

From 2efb3b061709bd204f11d54c177dae89e4af0f67 Mon Sep 17 00:00:00 2001
From: Junio C Hamano <gitster@pobox.com>
Date: Sat, 1 Mar 2008 23:43:32 -0800
Subject: [PATCH 35/53] Clean up find_unique_abbrev() callers

Now find_unique_abbrev() never returns NULL, there is no need for callers
to prepare for seeing NULL and fall back to giving the full 40-hexdigits.

While we are at it, drop "..." in the "git reset" output that reports the
location of the new HEAD, between the abbreviated commit object name and
the one line commit summary.  Because we are always showing the HEAD
(which cannot be missing!), we never had a case where we show the full 40
hexdigits that is not followed by three dots, and these three dots were
stealing 3 columns from the precious horizontal screen real estate out of
80 that can better be used for the one line commit summary.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 builtin-reset.c     | 8 ++------
 builtin-send-pack.c | 4 +---
 diff.c              | 2 --
 3 files changed, 3 insertions(+), 11 deletions(-)

diff --git a/builtin-reset.c b/builtin-reset.c
index af0037ec6e..bb3e19240a 100644
--- a/builtin-reset.c
+++ b/builtin-reset.c
@@ -63,14 +63,10 @@ static int reset_index_file(const unsigned char *sha1, int is_hard_reset)
 
 static void print_new_head_line(struct commit *commit)
 {
-	const char *hex, *dots = "...", *body;
+	const char *hex, *body;
 
 	hex = find_unique_abbrev(commit->object.sha1, DEFAULT_ABBREV);
-	if (!hex) {
-		hex = sha1_to_hex(commit->object.sha1);
-		dots = "";
-	}
-	printf("HEAD is now at %s%s", hex, dots);
+	printf("HEAD is now at %s", hex);
 	body = strstr(commit->buffer, "\n\n");
 	if (body) {
 		const char *eol;
diff --git a/builtin-send-pack.c b/builtin-send-pack.c
index b0cfae83fc..930e0fb3fd 100644
--- a/builtin-send-pack.c
+++ b/builtin-send-pack.c
@@ -264,9 +264,7 @@ static void print_ref_status(char flag, const char *summary, struct ref *to, str
 
 static const char *status_abbrev(unsigned char sha1[20])
 {
-	const char *abbrev;
-	abbrev = find_unique_abbrev(sha1, DEFAULT_ABBREV);
-	return abbrev ? abbrev : sha1_to_hex(sha1);
+	return find_unique_abbrev(sha1, DEFAULT_ABBREV);
 }
 
 static void print_ok_ref_status(struct ref *ref)
diff --git a/diff.c b/diff.c
index ad16164232..b80f656fb9 100644
--- a/diff.c
+++ b/diff.c
@@ -2581,8 +2581,6 @@ const char *diff_unique_abbrev(const unsigned char *sha1, int len)
 		return sha1_to_hex(sha1);
 
 	abbrev = find_unique_abbrev(sha1, len);
-	if (!abbrev)
-		return sha1_to_hex(sha1);
 	abblen = strlen(abbrev);
 	if (abblen < 37) {
 		static char hex[41];

From c8c16f2865f7c9c0d59b31ce66d50a4ecae72fd0 Mon Sep 17 00:00:00 2001
From: Junio C Hamano <gitster@pobox.com>
Date: Sun, 2 Mar 2008 00:57:26 -0800
Subject: [PATCH 36/53] diff-lib.c: constness strengthening

The internal implementation of diff-index codepath used to use non const
pointer to pass sha1 around, but it did not have to.  With this, we can
also lose the private no_sha1[] array, as we can use the public null_sha1[]
array that exists exactly for the same purpose.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 diff-lib.c | 13 ++++++-------
 1 file changed, 6 insertions(+), 7 deletions(-)

diff --git a/diff-lib.c b/diff-lib.c
index 94b150e830..4581b594d0 100644
--- a/diff-lib.c
+++ b/diff-lib.c
@@ -472,22 +472,21 @@ int run_diff_files(struct rev_info *revs, unsigned int option)
 static void diff_index_show_file(struct rev_info *revs,
 				 const char *prefix,
 				 struct cache_entry *ce,
-				 unsigned char *sha1, unsigned int mode)
+				 const unsigned char *sha1, unsigned int mode)
 {
 	diff_addremove(&revs->diffopt, prefix[0], mode,
 		       sha1, ce->name, NULL);
 }
 
 static int get_stat_data(struct cache_entry *ce,
-			 unsigned char **sha1p,
+			 const unsigned char **sha1p,
 			 unsigned int *modep,
 			 int cached, int match_missing)
 {
-	unsigned char *sha1 = ce->sha1;
+	const unsigned char *sha1 = ce->sha1;
 	unsigned int mode = ce->ce_mode;
 
 	if (!cached) {
-		static unsigned char no_sha1[20];
 		int changed;
 		struct stat st;
 		if (lstat(ce->name, &st) < 0) {
@@ -501,7 +500,7 @@ static int get_stat_data(struct cache_entry *ce,
 		changed = ce_match_stat(ce, &st, 0);
 		if (changed) {
 			mode = ce_mode_from_stat(ce, st.st_mode);
-			sha1 = no_sha1;
+			sha1 = null_sha1;
 		}
 	}
 
@@ -514,7 +513,7 @@ static void show_new_file(struct rev_info *revs,
 			  struct cache_entry *new,
 			  int cached, int match_missing)
 {
-	unsigned char *sha1;
+	const unsigned char *sha1;
 	unsigned int mode;
 
 	/* New file in the index: it might actually be different in
@@ -533,7 +532,7 @@ static int show_modified(struct rev_info *revs,
 			 int cached, int match_missing)
 {
 	unsigned int mode, oldmode;
-	unsigned char *sha1;
+	const unsigned char *sha1;
 
 	if (get_stat_data(new, &sha1, &mode, cached, match_missing) < 0) {
 		if (report_missing)

From 2b459b483cb264d3e9a39cd7857da2ccab08179f Mon Sep 17 00:00:00 2001
From: Junio C Hamano <gitster@pobox.com>
Date: Sun, 2 Mar 2008 00:07:59 -0800
Subject: [PATCH 37/53] diff: make sure work tree side is shown as 0{40} when
 different

Ping Yin noticed that "git diff-index --raw" shows 0{40} when work tree
has submodule difference, but "git diff --raw" didn't correctly do so.

There was a mistake in the diffcore_skip_stat_unmatch() that was meant to
clean up the stat-only difference for running diff between the index and
work tree and diff between the tree and the work tree, to cause it re-read
from the submodule repository HEAD.  When ce_stat_match() says work tree
is different, we should always say 0{40} on the work tree side.

This patch fixes the issue, and adds tests.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 diff.c                    |  7 ++----
 t/t4027-diff-submodule.sh | 53 +++++++++++++++++++++++++++++++++++++++
 2 files changed, 55 insertions(+), 5 deletions(-)
 create mode 100755 t/t4027-diff-submodule.sh

diff --git a/diff.c b/diff.c
index b80f656fb9..00e1590c6e 100644
--- a/diff.c
+++ b/diff.c
@@ -3187,11 +3187,8 @@ static void diffcore_apply_filter(const char *filter)
 static int diff_filespec_is_identical(struct diff_filespec *one,
 				      struct diff_filespec *two)
 {
-	if (S_ISGITLINK(one->mode)) {
-		diff_fill_sha1_info(one);
-		diff_fill_sha1_info(two);
-		return !hashcmp(one->sha1, two->sha1);
-	}
+	if (S_ISGITLINK(one->mode))
+		return 0;
 	if (diff_populate_filespec(one, 0))
 		return 0;
 	if (diff_populate_filespec(two, 0))
diff --git a/t/t4027-diff-submodule.sh b/t/t4027-diff-submodule.sh
new file mode 100755
index 0000000000..3d2d0816a3
--- /dev/null
+++ b/t/t4027-diff-submodule.sh
@@ -0,0 +1,53 @@
+#!/bin/sh
+
+test_description='difference in submodules'
+
+. ./test-lib.sh
+. ../diff-lib.sh
+
+_z40=0000000000000000000000000000000000000000
+test_expect_success setup '
+	test_tick &&
+	test_create_repo sub &&
+	(
+		cd sub &&
+		echo hello >world &&
+		git add world &&
+		git commit -m submodule
+	) &&
+
+	test_tick &&
+	echo frotz >nitfol &&
+	git add nitfol sub &&
+	git commit -m superproject &&
+
+	(
+		cd sub &&
+		echo goodbye >world &&
+		git add world &&
+		git commit -m "submodule #2"
+	) &&
+
+	set x $(
+		cd sub &&
+		git rev-list HEAD
+	) &&
+	echo ":160000 160000 $3 $_z40 M	sub" >expect
+'
+
+test_expect_success 'git diff --raw HEAD' '
+	git diff --raw --abbrev=40 HEAD >actual &&
+	diff -u expect actual
+'
+
+test_expect_success 'git diff-index --raw HEAD' '
+	git diff-index --raw HEAD >actual.index &&
+	diff -u expect actual.index
+'
+
+test_expect_success 'git diff-files --raw' '
+	git diff-files --raw >actual.files &&
+	diff -u expect actual.files
+'
+
+test_done

From e371a4c648e72673fcfdf0b45192857dd149e2f5 Mon Sep 17 00:00:00 2001
From: Johannes Schindelin <Johannes.Schindelin@gmx.de>
Date: Sun, 2 Mar 2008 07:40:33 +0000
Subject: [PATCH 38/53] Fix make_absolute_path() for parameters without a slash

When passing "xyz" to make_absolute_path(), make_absolute_path()
erroneously tried to chdir("xyz"), and then append "/xyz".  Instead,
skip the chdir() completely when no slash was found.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 path.c           | 4 +++-
 t/t0000-basic.sh | 2 ++
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/path.c b/path.c
index 42609524a5..af2716199b 100644
--- a/path.c
+++ b/path.c
@@ -311,8 +311,10 @@ const char *make_absolute_path(const char *path)
 			if (last_slash) {
 				*last_slash = '\0';
 				last_elem = xstrdup(last_slash + 1);
-			} else
+			} else {
 				last_elem = xstrdup(buf);
+				*buf = '\0';
+			}
 		}
 
 		if (*buf) {
diff --git a/t/t0000-basic.sh b/t/t0000-basic.sh
index 92de088227..27b54cbb12 100755
--- a/t/t0000-basic.sh
+++ b/t/t0000-basic.sh
@@ -304,6 +304,8 @@ test_expect_success 'absolute path works as expected' '
 	test "$dir" = "$(test-absolute-path $dir2)" &&
 	file="$dir"/index &&
 	test "$file" = "$(test-absolute-path $dir2/index)" &&
+	basename=blub &&
+	test "$dir/$basename" = $(cd .git && test-absolute-path $basename) &&
 	ln -s ../first/file .git/syml &&
 	sym="$(cd first; pwd -P)"/file &&
 	test "$sym" = "$(test-absolute-path $dir2/syml)"

From a3647bee1a91d930c4c3e17e36232b5c4a51533d Mon Sep 17 00:00:00 2001
From: Denis Cheng <crquan@gmail.com>
Date: Sun, 2 Mar 2008 18:05:05 +0800
Subject: [PATCH 39/53] cleanup: remove unused git_checkout_config

Directly call git_default_config instead.

Signed-off-by: Denis Cheng <crquan@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 builtin-checkout.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

diff --git a/builtin-checkout.c b/builtin-checkout.c
index b0cd788564..6b08016228 100644
--- a/builtin-checkout.c
+++ b/builtin-checkout.c
@@ -484,11 +484,6 @@ static int switch_branches(struct checkout_opts *opts, struct branch_info *new)
 	return post_checkout_hook(old.commit, new->commit, 1);
 }
 
-static int git_checkout_config(const char *var, const char *value)
-{
-	return git_default_config(var, value);
-}
-
 int cmd_checkout(int argc, const char **argv, const char *prefix)
 {
 	struct checkout_opts opts;
@@ -510,7 +505,7 @@ int cmd_checkout(int argc, const char **argv, const char *prefix)
 	memset(&opts, 0, sizeof(opts));
 	memset(&new, 0, sizeof(new));
 
-	git_config(git_checkout_config);
+	git_config(git_default_config);
 
 	opts.track = git_branch_track;
 

From 56b6d01d8497436e6b2545482ce2c86e614fc73c Mon Sep 17 00:00:00 2001
From: Jakub Narebski <jnareb@gmail.com>
Date: Sun, 2 Mar 2008 15:11:35 +0100
Subject: [PATCH 40/53] Documentation: Remove --{min,max}-age option from
 git-log(1)

The --max-age=<timestamp> and --min-age=<timestamp> are now shown only
in the git-rev-list manpage (plumbing).

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 Documentation/rev-list-options.txt | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/rev-list-options.txt b/Documentation/rev-list-options.txt
index 259072c078..2648a55085 100644
--- a/Documentation/rev-list-options.txt
+++ b/Documentation/rev-list-options.txt
@@ -130,9 +130,11 @@ limiting may be applied.
 
 	Show commits older than a specific date.
 
+ifdef::git-rev-list[]
 --max-age='timestamp', --min-age='timestamp'::
 
 	Limit the commits output to specified time range.
+endif::git-rev-list[]
 
 --author='pattern', --committer='pattern'::
 

From 6dfbb304be5a5e260321c5f47626582dbe2c62f0 Mon Sep 17 00:00:00 2001
From: Jakub Narebski <jnareb@gmail.com>
Date: Sun, 2 Mar 2008 16:57:14 +0100
Subject: [PATCH 41/53] gitweb: Mark first match when searching commit messages

Due to greediness of a pattern, gitweb used to mark (show) last match
in line, if there are more than one match in line. Now it shows first.
Showing all matches in a line would require further work.

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 gitweb/gitweb.perl | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/gitweb/gitweb.perl b/gitweb/gitweb.perl
index 20dc5d59c2..922dee98b9 100755
--- a/gitweb/gitweb.perl
+++ b/gitweb/gitweb.perl
@@ -3837,7 +3837,7 @@ sub git_search_grep_body {
 		              chop_and_escape_str($co{'title'}, 50) . "<br/>");
 		my $comment = $co{'comment'};
 		foreach my $line (@$comment) {
-			if ($line =~ m/^(.*)($search_regexp)(.*)$/i) {
+			if ($line =~ m/^(.*?)($search_regexp)(.*)$/i) {
 				my ($lead, $match, $trail) = ($1, $2, $3);
 				$match = chop_str($match, 70, 5, 'center');
 				my $contextlen = int((80 - length($match))/2);

From 5d02294c776c46f0d454470c66c16fe9f08fad3d Mon Sep 17 00:00:00 2001
From: Johannes Schindelin <Johannes.Schindelin@gmx.de>
Date: Sun, 2 Mar 2008 15:53:04 +0000
Subject: [PATCH 42/53] format-patch: use the diff options for the cover
 letter, too

Earlier, when you called "git format-patch --cover-letter -M", the
diffstat in the cover letter would not inherit the "-M".  Now it does.

While at it, add a few "|| break" statements in the test's loops;
otherwise, breakages inside the loops would not be caught.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 builtin-log.c           |  4 ++--
 t/t4014-format-patch.sh | 18 ++++++++++++++----
 2 files changed, 16 insertions(+), 6 deletions(-)

diff --git a/builtin-log.c b/builtin-log.c
index bbadbc0de2..fd9b3ae95e 100644
--- a/builtin-log.c
+++ b/builtin-log.c
@@ -694,8 +694,8 @@ static void make_cover_letter(struct rev_info *rev, int use_stdout,
 	if (!origin)
 		return;
 
-	diff_setup(&opts);
-	opts.output_format |= DIFF_FORMAT_SUMMARY | DIFF_FORMAT_DIFFSTAT;
+	memcpy(&opts, &rev->diffopt, sizeof(opts));
+	opts.output_format = DIFF_FORMAT_SUMMARY | DIFF_FORMAT_DIFFSTAT;
 
 	diff_setup_done(&opts);
 
diff --git a/t/t4014-format-patch.sh b/t/t4014-format-patch.sh
index 16aa99dc0d..6d86b7dd72 100755
--- a/t/t4014-format-patch.sh
+++ b/t/t4014-format-patch.sh
@@ -147,7 +147,7 @@ test_expect_success 'thread' '
 	for i in patches/0002-* patches/0003-*
 	do
 	  grep "References: $FIRST_MID" $i &&
-	  grep "In-Reply-To: $FIRST_MID" $i
+	  grep "In-Reply-To: $FIRST_MID" $i || break
 	done
 '
 
@@ -160,7 +160,7 @@ test_expect_success 'thread in-reply-to' '
 	for i in patches/*
 	do
 	  grep "References: $FIRST_MID" $i &&
-	  grep "In-Reply-To: $FIRST_MID" $i
+	  grep "In-Reply-To: $FIRST_MID" $i || break
 	done
 '
 
@@ -173,7 +173,7 @@ test_expect_success 'thread cover-letter' '
 	for i in patches/0001-* patches/0002-* patches/0003-* 
 	do
 	  grep "References: $FIRST_MID" $i &&
-	  grep "In-Reply-To: $FIRST_MID" $i
+	  grep "In-Reply-To: $FIRST_MID" $i || break
 	done
 '
 
@@ -186,7 +186,7 @@ test_expect_success 'thread cover-letter in-reply-to' '
 	for i in patches/*
 	do
 	  grep "References: $FIRST_MID" $i &&
-	  grep "In-Reply-To: $FIRST_MID" $i
+	  grep "In-Reply-To: $FIRST_MID" $i || break
 	done
 '
 
@@ -201,4 +201,14 @@ test_expect_success 'excessive subject' '
 	ls patches/0004-This-is-an-excessively-long-subject-line-for-a-messa.patch
 '
 
+test_expect_success 'cover-letter inherits diff options' '
+
+	git mv file foo &&
+	git commit -m foo &&
+	git format-patch --cover-letter -1 &&
+	! grep "file => foo .* 0 *$" 0000-cover-letter.patch &&
+	git format-patch --cover-letter -1 -M &&
+	grep "file => foo .* 0 *$" 0000-cover-letter.patch
+
+'
 test_done

From 859c4fbef589841585f8f21cc567b7016681df7a Mon Sep 17 00:00:00 2001
From: Johannes Schindelin <Johannes.Schindelin@gmx.de>
Date: Sun, 2 Mar 2008 15:53:39 +0000
Subject: [PATCH 43/53] format-patch: wrap cover-letter's shortlog sensibly

Earlier, overly-long onelines would not be wrapped at all, and indented
with 6 spaces.

Instead, we now wrap around at 72 characters, with a first-line indent
of 2 spaces, and the rest with 4 spaces.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 builtin-log.c                                 |  4 ++++
 ...-stdout_--cover-letter_-n_initial..master^ |  4 ++--
 t/t4014-format-patch.sh                       | 19 +++++++++++++++++++
 3 files changed, 25 insertions(+), 2 deletions(-)

diff --git a/builtin-log.c b/builtin-log.c
index fd9b3ae95e..fe8fc6f22a 100644
--- a/builtin-log.c
+++ b/builtin-log.c
@@ -683,6 +683,10 @@ static void make_cover_letter(struct rev_info *rev, int use_stdout,
 	strbuf_release(&sb);
 
 	shortlog_init(&log);
+	log.wrap_lines = 1;
+	log.wrap = 72;
+	log.in1 = 2;
+	log.in2 = 4;
 	for (i = 0; i < nr; i++)
 		shortlog_add_commit(&log, list[i]);
 
diff --git a/t/t4013/diff.format-patch_--stdout_--cover-letter_-n_initial..master^ b/t/t4013/diff.format-patch_--stdout_--cover-letter_-n_initial..master^
index 0151453b73..8dab4bf93e 100644
--- a/t/t4013/diff.format-patch_--stdout_--cover-letter_-n_initial..master^
+++ b/t/t4013/diff.format-patch_--stdout_--cover-letter_-n_initial..master^
@@ -7,8 +7,8 @@ Subject: [DIFFERENT_PREFIX 0/2] *** SUBJECT HERE ***
 *** BLURB HERE ***
 
 A U Thor (2):
-      Second
-      Third
+  Second
+  Third
 
  dir/sub |    4 ++++
  file0   |    3 +++
diff --git a/t/t4014-format-patch.sh b/t/t4014-format-patch.sh
index 6d86b7dd72..b2b7a8db85 100755
--- a/t/t4014-format-patch.sh
+++ b/t/t4014-format-patch.sh
@@ -211,4 +211,23 @@ test_expect_success 'cover-letter inherits diff options' '
 	grep "file => foo .* 0 *$" 0000-cover-letter.patch
 
 '
+
+cat > expect << EOF
+  This is an excessively long subject line for a message due to the
+    habit some projects have of not having a short, one-line subject at
+    the start of the commit message, but rather sticking a whole
+    paragraph right at the start as the only thing in the commit
+    message. It had better not become the filename for the patch.
+  foo
+
+EOF
+
+test_expect_success 'shortlog of cover-letter wraps overly-long onelines' '
+
+	git format-patch --cover-letter -2 &&
+	sed -e "1,/A U Thor/d" -e "/^$/q" < 0000-cover-letter.patch > output &&
+	git diff expect output
+
+'
+
 test_done

From 733ee2b7a95f6f4ac03f6f8db4899bcd54d0017e Mon Sep 17 00:00:00 2001
From: Jean-Luc Herren <jlh@gmx.ch>
Date: Thu, 28 Feb 2008 23:29:54 +0100
Subject: [PATCH 44/53] fast-import: exit with proper message if not a git dir

git fast-import expects to be run from an existing (possibly
empty) repository.  It was dying with a suboptimal message if that
wasn't the case.

Signed-off-by: Jean-Luc Herren <jlh@gmx.ch>
Acked-by: Shawn O. Pearce <spearce@spearce.org>
---
 fast-import.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/fast-import.c b/fast-import.c
index 0d3449f2ce..7f197d5e36 100644
--- a/fast-import.c
+++ b/fast-import.c
@@ -2377,6 +2377,7 @@ int main(int argc, const char **argv)
 {
 	unsigned int i, show_stats = 1;
 
+	setup_git_directory();
 	git_config(git_pack_config);
 	if (!pack_compression_seen && core_compression_seen)
 		pack_compression_level = core_compression_level;

From 34cd62eb91600109378c8121c1fecd924a9af177 Mon Sep 17 00:00:00 2001
From: Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
Date: Mon, 3 Mar 2008 00:07:47 +0100
Subject: [PATCH 45/53] Fix doc typos.

Signed-off-by: Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 Documentation/git-grep.txt                  | 2 +-
 Documentation/technical/api-run-command.txt | 6 +++---
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/Documentation/git-grep.txt b/Documentation/git-grep.txt
index 71a73354f8..a97f0557f4 100644
--- a/Documentation/git-grep.txt
+++ b/Documentation/git-grep.txt
@@ -78,7 +78,7 @@ OPTIONS
 -l | --files-with-matches | --name-only | -L | --files-without-match::
 	Instead of showing every matched line, show only the
 	names of files that contain (or do not contain) matches.
-	For better compatability with git-diff, --name-only is a
+	For better compatibility with git-diff, --name-only is a
 	synonym for --files-with-matches.
 
 -c | --count::
diff --git a/Documentation/technical/api-run-command.txt b/Documentation/technical/api-run-command.txt
index dfbf9ac5d0..fde3b45321 100644
--- a/Documentation/technical/api-run-command.txt
+++ b/Documentation/technical/api-run-command.txt
@@ -49,7 +49,7 @@ Functions
 
 `finish_async`::
 
-	Wait for the completeion of an asynchronous function that was
+	Wait for the completion of an asynchronous function that was
 	started with start_async().
 
 
@@ -121,8 +121,8 @@ string pointers (NULL terminated) in .env:
 . If the string is of the form "VAR=value", i.e. it contains '='
   the variable is added to the child process's environment.
 
-. If the string does not contain '=', it names an environement
-  variable that will be removed from the child process's envionment.
+. If the string does not contain '=', it names an environment
+  variable that will be removed from the child process's environment.
 
 To specify a new initial working directory for the sub-process,
 specify it in the .dir member.

From 5c09f321729c2f1ce6718a0cfefa5e647fa808fc Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Bj=C3=B6rn=20Steinbrink?= <B.Steinbrink@gmx.de>
Date: Mon, 3 Mar 2008 05:08:43 +0100
Subject: [PATCH 46/53] receive-pack: Initialize PATH to include exec-dir.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

511707d (use only the $PATH for exec'ing git commands) made it a
requirement to call setup_path() to include the git exec-dir in PATH
before spawning any other git commands. git-receive-pack was not yet
adapted to do this and therefore fails to spawn git-unpack-objects if that
is not in the standard PATH.

Signed-off-by: Björn Steinbrink <B.Steinbrink@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 receive-pack.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/receive-pack.c b/receive-pack.c
index 3267495832..c90ec7dde2 100644
--- a/receive-pack.c
+++ b/receive-pack.c
@@ -469,6 +469,8 @@ int main(int argc, char **argv)
 	if (!dir)
 		usage(receive_pack_usage);
 
+	setup_path(NULL);
+
 	if (!enter_repo(dir, 0))
 		die("'%s': unable to chdir or not a git archive", dir);
 

From fcbcfe707ae23b37d72025a229f31c450fb4d3b3 Mon Sep 17 00:00:00 2001
From: Ping Yin <pkufranky@gmail.com>
Date: Mon, 3 Mar 2008 10:03:18 +0800
Subject: [PATCH 47/53] git-submodule: Fix typo 'url' which should be '$url'

Fix typo in 'test -z "url"' when checking whether a submodule url is
empty. "url" should be "$url".

Signed-off-by: Ping Yin <pkufranky@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 git-submodule.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/git-submodule.sh b/git-submodule.sh
index a6aaf40b0a..67d3224c8c 100755
--- a/git-submodule.sh
+++ b/git-submodule.sh
@@ -362,7 +362,7 @@ cmd_status()
 	do
 		name=$(module_name "$path") || exit
 		url=$(git config submodule."$name".url)
-		if test -z "url" || ! test -d "$path"/.git
+		if test -z "$url" || ! test -d "$path"/.git
 		then
 			say "-$sha1 $path"
 			continue;

From 90d0ed96b76ee51f8ae6f32923b92e7b20ba73c0 Mon Sep 17 00:00:00 2001
From: Junio C Hamano <gitster@pobox.com>
Date: Thu, 28 Feb 2008 13:09:30 -0800
Subject: [PATCH 48/53] tests: introduce test_must_fail

When we expect a git command to notice and signal errors, we
carelessly wrote in our tests:

    test_expect_success 'reject bogus request' '
        do something &&
        do something else &&
        ! git command
    '

but a non-zero exit could come from the "git command" segfaulting.

A new helper function "tset_must_fail" is introduced and it is
meant to be used to make sure the command gracefully fails (iow,
dying and exiting with non zero status is counted as a failure
to "gracefully fail").  The above example should be written as:

    test_expect_success 'reject bogus request' '
        do something &&
        do something else &&
        test_must_fail git command
    '

Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 t/t2008-checkout-subdir.sh | 16 ++++++++--------
 t/test-lib.sh              | 17 +++++++++++++++++
 2 files changed, 25 insertions(+), 8 deletions(-)

diff --git a/t/t2008-checkout-subdir.sh b/t/t2008-checkout-subdir.sh
index f78945ed8e..3e098ab31e 100755
--- a/t/t2008-checkout-subdir.sh
+++ b/t/t2008-checkout-subdir.sh
@@ -67,16 +67,16 @@ test_expect_success 'checkout with simple prefix' '
 
 '
 
-test_expect_failure 'relative path outside tree should fail' \
-	'git checkout HEAD -- ../../Makefile'
+test_expect_success 'relative path outside tree should fail' \
+	'test_must_fail git checkout HEAD -- ../../Makefile'
 
-test_expect_failure 'incorrect relative path to file should fail (1)' \
-	'git checkout HEAD -- ../file0'
+test_expect_success 'incorrect relative path to file should fail (1)' \
+	'test_must_fail git checkout HEAD -- ../file0'
 
-test_expect_failure 'incorrect relative path should fail (2)' \
-	'( cd dir1 && git checkout HEAD -- ./file0 )'
+test_expect_success 'incorrect relative path should fail (2)' \
+	'( cd dir1 && test_must_fail git checkout HEAD -- ./file0 )'
 
-test_expect_failure 'incorrect relative path should fail (3)' \
-	'( cd dir1 && git checkout HEAD -- ../../file0 )'
+test_expect_success 'incorrect relative path should fail (3)' \
+	'( cd dir1 && test_must_fail git checkout HEAD -- ../../file0 )'
 
 test_done
diff --git a/t/test-lib.sh b/t/test-lib.sh
index 142540e1b1..c0c5e21adf 100644
--- a/t/test-lib.sh
+++ b/t/test-lib.sh
@@ -257,6 +257,23 @@ test_expect_code () {
 	echo >&3 ""
 }
 
+# This is not among top-level (test_expect_success | test_expect_failure)
+# but is a prefix that can be used in the test script, like:
+#
+#	test_expect_success 'complain and die' '
+#           do something &&
+#           do something else &&
+#	    test_must_fail git checkout ../outerspace
+#	'
+#
+# Writing this as "! git checkout ../outerspace" is wrong, because
+# the failure could be due to a segv.  We want a controlled failure.
+
+test_must_fail () {
+	"$@"
+	test $? -gt 0 -a $? -le 128
+}
+
 # Most tests can use the created repository, but some may need to create more.
 # Usage: test_create_repo <directory>
 test_create_repo () {

From 0f2d4476c12c15106be20d53c507035e157b66f1 Mon Sep 17 00:00:00 2001
From: Jeff King <peff@peff.net>
Date: Mon, 3 Mar 2008 01:30:56 -0500
Subject: [PATCH 49/53] revert: actually check for a dirty index

The previous code mistakenly used wt_status_prepare to check whether the
index had anything commitable in it; however, that function is just an
init function, and will never report a dirty index.

The correct way with wt_status_* would be to call wt_status_print with the
output pointing to /dev/null or similar. However, that does extra work by
both examining the working tree and spewing status information to nowhere.

Instead, let's just implement the useful subset of wt_status_print as an
"is_index_dirty" function.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 builtin-revert.c              | 20 ++++++++++++++++----
 t/t3501-revert-cherry-pick.sh |  9 +++++++++
 2 files changed, 25 insertions(+), 4 deletions(-)

diff --git a/builtin-revert.c b/builtin-revert.c
index 358af53747..64f0d0ee02 100644
--- a/builtin-revert.c
+++ b/builtin-revert.c
@@ -8,6 +8,8 @@
 #include "exec_cmd.h"
 #include "utf8.h"
 #include "parse-options.h"
+#include "diff.h"
+#include "revision.h"
 
 /*
  * This implements the builtins revert and cherry-pick.
@@ -245,6 +247,17 @@ static char *help_msg(const unsigned char *sha1)
 	return helpbuf;
 }
 
+static int index_is_dirty(void)
+{
+	struct rev_info rev;
+	init_revisions(&rev, NULL);
+	setup_revisions(0, NULL, &rev, "HEAD");
+	DIFF_OPT_SET(&rev.diffopt, QUIET);
+	DIFF_OPT_SET(&rev.diffopt, EXIT_WITH_STATUS);
+	run_diff_index(&rev, 1);
+	return !!DIFF_OPT_TST(&rev.diffopt, HAS_CHANGES);
+}
+
 static int revert_or_cherry_pick(int argc, const char **argv)
 {
 	unsigned char head[20];
@@ -273,12 +286,11 @@ static int revert_or_cherry_pick(int argc, const char **argv)
 		if (write_tree(head, 0, NULL))
 			die ("Your index file is unmerged.");
 	} else {
-		struct wt_status s;
-
 		if (get_sha1("HEAD", head))
 			die ("You do not have a valid HEAD");
-		wt_status_prepare(&s);
-		if (s.commitable)
+		if (read_cache() < 0)
+			die("could not read the index");
+		if (index_is_dirty())
 			die ("Dirty index: cannot %s", me);
 		discard_cache();
 	}
diff --git a/t/t3501-revert-cherry-pick.sh b/t/t3501-revert-cherry-pick.sh
index 2dbe04fb20..6da212825a 100755
--- a/t/t3501-revert-cherry-pick.sh
+++ b/t/t3501-revert-cherry-pick.sh
@@ -59,4 +59,13 @@ test_expect_success 'revert after renaming branch' '
 
 '
 
+test_expect_success 'revert forbidden on dirty working tree' '
+
+	echo content >extra_file &&
+	git add extra_file &&
+	test_must_fail git revert HEAD 2>errors &&
+	grep "Dirty index" errors
+
+'
+
 test_done

From d3df4271b900c9bf529cf495afeb77fbbf621221 Mon Sep 17 00:00:00 2001
From: Junio C Hamano <gitster@pobox.com>
Date: Sun, 2 Mar 2008 21:49:28 -0800
Subject: [PATCH 50/53] Update draft release notes for 1.5.4.4

Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 Documentation/RelNotes-1.5.4.4.txt | 22 +++++++++++++++++++++-
 1 file changed, 21 insertions(+), 1 deletion(-)

diff --git a/Documentation/RelNotes-1.5.4.4.txt b/Documentation/RelNotes-1.5.4.4.txt
index 5bfdb35376..5635977c93 100644
--- a/Documentation/RelNotes-1.5.4.4.txt
+++ b/Documentation/RelNotes-1.5.4.4.txt
@@ -4,6 +4,10 @@ GIT v1.5.4.4 Release Notes
 Fixes since v1.5.4.3
 --------------------
 
+ * Building and installing with an overtight umask such as 077 made
+   installed templates unreadable by others, while the rest of the install
+   are done in a way that is friendly to umask 022.
+
  * "git cvsexportcommit -w $cvsdir" misbehaved when GIT_DIR is set to a
    relative directory.
 
@@ -17,10 +21,26 @@ Fixes since v1.5.4.3
 
  * "git send-email" in 1.5.4.3 issued a bogus empty In-Reply-To: header.
 
+ * "git bisect" showed mysterious "won't bisect on seeked tree" error message.
+   This was leftover from Cogito days to prevent "bisect" starting from a
+   cg-seeked state.  We still keep the Cogito safety, but running "git bisect
+   start" when another bisect was in effect will clean up and start over.
+
+ * "git push" with an explicit PATH to receive-pack did not quite work if
+   receive-pack was not on usual PATH.  We earlier fixed the same issue
+   with "git fetch" and upload-pack, but somehow forgot to do so in the
+   other direction.
+
+ * git-gui's info dialog was not displayed correctly when the user tries
+   to commit nothing (i.e. without staging anything).
+
+ * "git revert" did not properly fail when attempting to run with a
+   dirty index.
+
 Also included are a handful documentation updates.
 
 ---
 exec >/var/tmp/1
 echo O=$(git describe maint)
-O=v1.5.4.3
+O=v1.5.4.3-32-g0f2d447
 git shortlog --no-merges $O..maint

From 2d3539e87a75b5f93f9e53bf13a60ef67a28ea62 Mon Sep 17 00:00:00 2001
From: Junio C Hamano <gitster@pobox.com>
Date: Sun, 2 Mar 2008 22:04:59 -0800
Subject: [PATCH 51/53] Update draft release notes for 1.5.5

Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 Documentation/RelNotes-1.5.5.txt | 28 +++++++++++++++++++++++++++-
 1 file changed, 27 insertions(+), 1 deletion(-)

diff --git a/Documentation/RelNotes-1.5.5.txt b/Documentation/RelNotes-1.5.5.txt
index 849b6b9604..b57fa1eb1a 100644
--- a/Documentation/RelNotes-1.5.5.txt
+++ b/Documentation/RelNotes-1.5.5.txt
@@ -4,6 +4,10 @@ GIT v1.5.5 Release Notes
 Updates since v1.5.4
 --------------------
 
+(subsystems)
+
+ * Comes with git-gui 0.9.3
+
 (performance)
 
  * On platforms with suboptimal qsort(3) implementation, there
@@ -57,6 +61,8 @@ Updates since v1.5.4
 
  * "git add -i" behaves better even before you make an initial commit.
 
+ * "git am" refused to run from a subdirectory without a good reason.
+
  * After "git apply --whitespace=fix" fixes whitespace errors in a patch,
    a line before the fix can appear as a context or preimage line in a
    later patch, causing the patch not to apply.  The command now knows to
@@ -81,6 +87,8 @@ Updates since v1.5.4
    inspect what is going to be committed and prepare the commit
    log message template to be edited.
 
+ * "git cvsimport" can now take more than one -M options.
+
  * "git describe" learned to limit the tags to be used for
    naming with --match option.
 
@@ -89,6 +97,12 @@ Updates since v1.5.4
 
  * "git describe --exact-match" describes only commits that are tagged.
 
+ * "git describe --long" describes a tagged commit as $tag-0-$sha1,
+   instead of just showing the exact tagname.
+
+ * "git describe" warns when using a tag whose name and path contradict
+   with each other.
+
  * "git diff" learned "--relative" option to limit and output paths
    relative to the current directory when working in a subdirectory.
 
@@ -98,6 +112,8 @@ Updates since v1.5.4
  * "git format-patch" learned --cover-letter option to generate a cover
    letter template.
 
+ * "git gc" learned --quiet option.
+
  * "git grep" now knows "--name-only" is a synonym for the "-l" option.
 
  * "git help <alias>" now reports "'git <alias>' is alias to <what>",
@@ -144,8 +160,18 @@ Fixes since v1.5.4
 All of the fixes in v1.5.4 maintenance series are included in
 this release, unless otherwise noted.
 
+ * "git-daemon" did not send early errors to syslog.
+
+ * "git-http-push" did not allow deletion of remote ref with the usual
+   "push <remote> :<branch>" syntax.
+
+ * "git-log --merge" did not well work with --left-right option.
+
+ * "git-rebase --abort" did not go back to the right location if
+   "git-reset" was run during the "git-rebase" session.
+
 ---
 exec >/var/tmp/1
-O=v1.5.4.3-339-g7cf7f54
+O=v1.5.4.3-428-g6b48990
 echo O=`git describe refs/heads/master`
 git shortlog --no-merges $O..refs/heads/master ^refs/heads/maint

From f3ec549481827b10609a43bf504517a0e8063a12 Mon Sep 17 00:00:00 2001
From: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Date: Mon, 3 Mar 2008 07:31:23 +0100
Subject: [PATCH 52/53] fetch-pack: check parse_commit/object results

Signed-off-by: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 builtin-fetch-pack.c | 17 ++++++++++-------
 1 file changed, 10 insertions(+), 7 deletions(-)

diff --git a/builtin-fetch-pack.c b/builtin-fetch-pack.c
index b23e886d75..29b38e4650 100644
--- a/builtin-fetch-pack.c
+++ b/builtin-fetch-pack.c
@@ -41,7 +41,8 @@ static void rev_list_push(struct commit *commit, int mark)
 		commit->object.flags |= mark;
 
 		if (!(commit->object.parsed))
-			parse_commit(commit);
+			if (parse_commit(commit))
+				return;
 
 		insert_by_date(commit, &rev_list);
 
@@ -83,7 +84,8 @@ static void mark_common(struct commit *commit,
 			if (!ancestors_only && !(o->flags & POPPED))
 				non_common_revs--;
 			if (!o->parsed && !dont_parse)
-				parse_commit(commit);
+				if (parse_commit(commit))
+					return;
 
 			for (parents = commit->parents;
 					parents;
@@ -103,20 +105,20 @@ static const unsigned char* get_rev(void)
 
 	while (commit == NULL) {
 		unsigned int mark;
-		struct commit_list* parents;
+		struct commit_list *parents = NULL;
 
 		if (rev_list == NULL || non_common_revs == 0)
 			return NULL;
 
 		commit = rev_list->item;
 		if (!(commit->object.parsed))
-			parse_commit(commit);
+			if (!parse_commit(commit))
+				parents = commit->parents;
+
 		commit->object.flags |= POPPED;
 		if (!(commit->object.flags & COMMON))
 			non_common_revs--;
 
-		parents = commit->parents;
-
 		if (commit->object.flags & COMMON) {
 			/* do not send "have", and ignore ancestors */
 			commit = NULL;
@@ -212,7 +214,8 @@ static int find_common(int fd[2], unsigned char *result_sha1,
 				if (!lookup_object(sha1))
 					die("object not found: %s", line);
 				/* make sure that it is parsed as shallow */
-				parse_object(sha1);
+				if (!parse_object(sha1))
+					die("error in object: %s", line);
 				if (unregister_shallow(sha1))
 					die("no shallow found: %s", line);
 				continue;

From c374b91cf295f437d438a103bfd2cf3fffcce580 Mon Sep 17 00:00:00 2001
From: Junio C Hamano <gitster@pobox.com>
Date: Mon, 3 Mar 2008 09:24:17 -0800
Subject: [PATCH 53/53] git-describe: use tags found in packed-refs correctly

When your refs are packed, "git-describe" can find the tag that is the
best match without ever parsing the tag itself.  But lookup_tag() in
display_name() says "I've never seen it", creates an empty shell, and
returns it.  We need to make sure that we actually have parsed the tag
data into it.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 builtin-describe.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/builtin-describe.c b/builtin-describe.c
index 2f1e7ba150..c549d362b2 100644
--- a/builtin-describe.c
+++ b/builtin-describe.c
@@ -156,7 +156,7 @@ static void display_name(struct commit_name *n)
 {
 	if (n->prio == 2 && !n->tag) {
 		n->tag = lookup_tag(n->sha1);
-		if (!n->tag || !n->tag->tag)
+		if (parse_tag(n->tag) || !n->tag || !n->tag->tag)
 			die("annotated tag %s not available", n->path);
 		if (strcmp(n->tag->tag, n->path))
 			warning("tag '%s' is really '%s' here", n->tag->tag, n->path);