From: Junio C Hamano <gitster@pobox.com>
To: "Shawn O. Pearce" <spearce@spearce.org>
Cc: Johannes Schindelin <Johannes.Schindelin@gmx.de>, git@vger.kernel.org
Subject: [PATCH] git add --intent-to-add: fix removal of cached emptiness
Date: Fri, 28 Nov 2008 16:15:37 -0800 [thread overview]
Message-ID: <7vk5an4cba.fsf_-_@gitster.siamese.dyndns.org> (raw)
In-Reply-To: <7voczz4cfb.fsf@gitster.siamese.dyndns.org> (Junio C. Hamano's message of "Fri, 28 Nov 2008 16:13:12 -0800")
This uses the extended index flag mechanism introduced earlier to mark
the entries added to the index via "git add -N" with CE_INTENT_TO_ADD.
The logic to detect an "intent to add" entry for the purpose of allowing
"git rm --cached $path" is tightened to check not just for a staged empty
blob, but with the CE_INTENT_TO_ADD bit. This protects an empty blob that
was explicitly added and then modified in the work tree from being dropped
with this sequence:
$ >empty
$ git add empty
$ echo "non empty" >empty
$ git rm --cached empty
An index an "intent to add" entry is blocked. This implies that you
cannot "git commit" from such a state; however "git commit -a" still
works.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
* This applies on top of the result of merging commit 06aaaa0 (Extend
index to save more flags, 2008-10-01) from nd/narrow topic into 'master'.
builtin-rm.c | 11 ++++++-----
builtin-write-tree.c | 2 +-
cache-tree.c | 10 +++++++---
cache.h | 3 ++-
read-cache.c | 2 ++
t/t3600-rm.sh | 4 ++--
t/t3701-add-interactive.sh | 18 ++++++++++++++++++
7 files changed, 38 insertions(+), 12 deletions(-)
diff --git a/builtin-rm.c b/builtin-rm.c
index b7126e3..8debcec 100644
--- a/builtin-rm.c
+++ b/builtin-rm.c
@@ -73,14 +73,15 @@ static int check_local_mod(unsigned char *head, int index_only)
}
if (ce_match_stat(ce, &st, 0))
local_changes = 1;
- if (no_head
- || get_tree_entry(head, name, sha1, &mode)
- || ce->ce_mode != create_ce_mode(mode)
- || hashcmp(ce->sha1, sha1))
+ if (no_head ||
+ ((get_tree_entry(head, name, sha1, &mode)
+ || ce->ce_mode != create_ce_mode(mode)
+ || hashcmp(ce->sha1, sha1)) &&
+ !(ce->ce_flags & CE_INTENT_TO_ADD)))
staged_changes = 1;
if (local_changes && staged_changes &&
- !(index_only && is_empty_blob_sha1(ce->sha1)))
+ !(index_only && (ce->ce_flags & CE_INTENT_TO_ADD)))
errs = error("'%s' has staged content different "
"from both the file and the HEAD\n"
"(use -f to force removal)", name);
diff --git a/builtin-write-tree.c b/builtin-write-tree.c
index 52a3c01..9d64050 100644
--- a/builtin-write-tree.c
+++ b/builtin-write-tree.c
@@ -42,7 +42,7 @@ int cmd_write_tree(int argc, const char **argv, const char *unused_prefix)
die("%s: error reading the index", me);
break;
case WRITE_TREE_UNMERGED_INDEX:
- die("%s: error building trees; the index is unmerged?", me);
+ die("%s: error building trees", me);
break;
case WRITE_TREE_PREFIX_ERROR:
die("%s: prefix %s not found", me, prefix);
diff --git a/cache-tree.c b/cache-tree.c
index 5f8ee87..3d8f218 100644
--- a/cache-tree.c
+++ b/cache-tree.c
@@ -155,13 +155,17 @@ static int verify_cache(struct cache_entry **cache,
funny = 0;
for (i = 0; i < entries; i++) {
struct cache_entry *ce = cache[i];
- if (ce_stage(ce)) {
+ if (ce_stage(ce) || (ce->ce_flags & CE_INTENT_TO_ADD)) {
if (10 < ++funny) {
fprintf(stderr, "...\n");
break;
}
- fprintf(stderr, "%s: unmerged (%s)\n",
- ce->name, sha1_to_hex(ce->sha1));
+ if (ce_stage(ce))
+ fprintf(stderr, "%s: unmerged (%s)\n",
+ ce->name, sha1_to_hex(ce->sha1));
+ else
+ fprintf(stderr, "%s: not added yet\n",
+ ce->name);
}
}
if (funny)
diff --git a/cache.h b/cache.h
index ef2e7f9..f15b3fc 100644
--- a/cache.h
+++ b/cache.h
@@ -176,10 +176,11 @@ struct cache_entry {
/*
* Extended on-disk flags
*/
+#define CE_INTENT_TO_ADD 0x20000000
/* CE_EXTENDED2 is for future extension */
#define CE_EXTENDED2 0x80000000
-#define CE_EXTENDED_FLAGS (0)
+#define CE_EXTENDED_FLAGS (CE_INTENT_TO_ADD)
/*
* Safeguard to avoid saving wrong flags:
diff --git a/read-cache.c b/read-cache.c
index abc627b..fa30a0f 100644
--- a/read-cache.c
+++ b/read-cache.c
@@ -546,6 +546,8 @@ int add_to_index(struct index_state *istate, const char *path, struct stat *st,
ce->ce_flags = namelen;
if (!intent_only)
fill_stat_cache_info(ce, st);
+ else
+ ce->ce_flags |= CE_INTENT_TO_ADD;
if (trust_executable_bit && has_symlinks)
ce->ce_mode = create_ce_mode(st_mode);
diff --git a/t/t3600-rm.sh b/t/t3600-rm.sh
index 5b4d6f7..b7d46e5 100755
--- a/t/t3600-rm.sh
+++ b/t/t3600-rm.sh
@@ -187,8 +187,8 @@ test_expect_success 'but with -f it should work.' '
test_must_fail git ls-files --error-unmatch baz
'
-test_expect_failure 'refuse to remove cached empty file with modifications' '
- touch empty &&
+test_expect_success 'refuse to remove cached empty file with modifications' '
+ >empty &&
git add empty &&
echo content >empty &&
test_must_fail git rm --cached empty
diff --git a/t/t3701-add-interactive.sh b/t/t3701-add-interactive.sh
index e95663d..473ef85 100755
--- a/t/t3701-add-interactive.sh
+++ b/t/t3701-add-interactive.sh
@@ -133,6 +133,24 @@ test_expect_success 'real edit works' '
test_cmp expected output
'
+test_expect_success 'cannot commit with i-t-a entry' '
+ git reset --hard &&
+ echo xyzzy >rezrov &&
+ echo frotz >nitfol &&
+ git add rezrov &&
+ git add -N nitfol &&
+ test_must_fail git commit
+'
+
+test_expect_success 'can commit with an unrelated i-t-a entry in index' '
+ git reset --hard &&
+ echo xyzzy >rezrov &&
+ echo frotz >nitfol &&
+ git add rezrov &&
+ git add -N nitfol &&
+ git commit -m partial rezrov
+'
+
if test "$(git config --bool core.filemode)" = false
then
say 'skipping filemode tests (filesystem does not properly support modes)'
--
1.6.0.4.850.g6bd829
next prev parent reply other threads:[~2008-11-29 0:17 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-11-27 0:28 What's cooking in git.git (Nov 2008, #06; Wed, 26) Junio C Hamano
2008-11-27 22:49 ` Johannes Schindelin
2008-11-28 2:06 ` Junio C Hamano
2008-11-28 11:47 ` Johannes Schindelin
2008-11-28 19:20 ` Shawn O. Pearce
2008-11-29 0:13 ` Junio C Hamano
2008-11-29 0:15 ` Junio C Hamano [this message]
2008-11-29 3:51 ` [PATCH 1/3] builtin-rm.c: explain and clarify the "local change" logic Junio C Hamano
2008-11-29 3:55 ` [PATCH 2/3] git add --intent-to-add: fix removal of cached emptiness Junio C Hamano
2008-11-29 15:38 ` Sverre Rabbelier
2008-11-30 19:21 ` Jeff King
2008-11-29 3:56 ` [PATCH 3/3] git add --intent-to-add: do not let an empty blob committed by accident Junio C Hamano
2008-11-30 19:14 ` Jeff King
2008-12-01 9:24 ` Junio C Hamano
2008-11-29 1:25 ` What's cooking in git.git (Nov 2008, #06; Wed, 26) Daniel Barkalow
2008-11-29 13:02 ` Nguyen Thai Ngoc Duy
2008-11-30 10:29 ` Nguyen Thai Ngoc Duy
2008-11-30 21:26 ` Daniel Barkalow
2008-12-06 17:26 ` Nguyen Thai Ngoc Duy
2008-12-06 18:39 ` Daniel Barkalow
2008-12-07 12:27 ` Nguyen Thai Ngoc Duy
2008-12-07 21:26 ` Daniel Barkalow
2008-12-08 12:51 ` Nguyen Thai Ngoc Duy
2008-12-08 19:41 ` Daniel Barkalow
2008-12-11 13:04 ` Nguyen Thai Ngoc Duy
2008-12-11 20:30 ` Daniel Barkalow
2008-12-12 1:41 ` Junio C Hamano
2008-12-12 2:40 ` Daniel Barkalow
2008-12-12 3:12 ` Junio C Hamano
2008-12-12 3:36 ` Jeff King
2008-12-12 16:13 ` Nguyen Thai Ngoc Duy
2008-12-12 16:45 ` Johannes Sixt
2008-12-12 16:54 ` Nguyen Thai Ngoc Duy
2008-12-13 5:51 ` Junio C Hamano
2008-12-13 5:51 ` Junio C Hamano
2008-12-12 16:08 ` Nguyen Thai Ngoc Duy
2008-12-07 3:45 ` Junio C Hamano
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=7vk5an4cba.fsf_-_@gitster.siamese.dyndns.org \
--to=gitster@pobox.com \
--cc=Johannes.Schindelin@gmx.de \
--cc=git@vger.kernel.org \
--cc=spearce@spearce.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).