All of lore.kernel.org
 help / color / mirror / Atom feed
From: Junio C Hamano <gitster@pobox.com>
To: "Björn Steinbrink" <B.Steinbrink@gmx.de>
Cc: Eric Wong <normalperson@yhbt.net>,
	Anton Gyllenberg <anton@iki.fi>,
	git@vger.kernel.org,
	Linus Torvalds <torvalds@linux-foundation.org>
Subject: [PATCH] match_tree_entry(): a pathspec only matches at directory boundaries
Date: Wed, 01 Apr 2009 21:41:31 -0700	[thread overview]
Message-ID: <7vwsa3llac.fsf_-_@gitster.siamese.dyndns.org> (raw)
In-Reply-To: <7vbprfn0ai.fsf@gitster.siamese.dyndns.org> (Junio C. Hamano's message of "Wed, 01 Apr 2009 21:32:05 -0700")

Previously the code did a simple prefix match, which means that a path in
a directory "frotz/" would have matched with pathspec "f".

Signed-off-by: Junio C Hamano <gitster@pobox.com>
---

 * And this is a companion patch to fix ls-tree.  The test case uses a
   tree that has path3/1.txt and path3/2.txt in it.

   The bug Eric diagnosed and worked around in git-svn makes the current
   code show these two paths when pathspec "pa" and "path3/a" are given.
   The presense of "path3/a" makes the tree walker traverse down to path3
   subtree (in case something that matches "a" is in there---this is a
   correct behaviour), but then in that subtree, "pa" incorrectly matches
   "path3/1.txt".

   This logic dates back to 0ca14a5 (Start adding interfaces to read in
   partial trees, 2005-07-14).  I think it is just a simple oversight and
   we should fix it.

 t/t3101-ls-tree-dirname.sh |    6 ++++++
 tree.c                     |    8 ++++++--
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/t/t3101-ls-tree-dirname.sh b/t/t3101-ls-tree-dirname.sh
index 4dd7d12..51cb4a3 100755
--- a/t/t3101-ls-tree-dirname.sh
+++ b/t/t3101-ls-tree-dirname.sh
@@ -135,4 +135,10 @@ test_expect_success \
 EOF
      test_output'
 
+test_expect_success 'ls-tree filter is leading path match' '
+	git ls-tree $tree pa path3/a >current &&
+	>expected &&
+	test_output
+'
+
 test_done
diff --git a/tree.c b/tree.c
index 03e782a..d82a047 100644
--- a/tree.c
+++ b/tree.c
@@ -60,8 +60,12 @@ static int match_tree_entry(const char *base, int baselen, const char *path, uns
 			/* If it doesn't match, move along... */
 			if (strncmp(base, match, matchlen))
 				continue;
-			/* The base is a subdirectory of a path which was specified. */
-			return 1;
+			/* pathspecs match only at the directory boundaries */
+			if (!matchlen ||
+			    base[matchlen] == '/' ||
+			    match[matchlen - 1] == '/')
+				return 1;
+			continue;
 		}
 
 		/* Does the base match? */
-- 
1.6.2.1.483.gcc994

  reply	other threads:[~2009-04-02  4:43 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-03-26 10:31 svn clone Checksum mismatch question Gilbert Liddell
2009-03-26 13:02 ` Björn Steinbrink
2009-03-26 13:28   ` Gilbert Liddell
2009-03-26 13:54     ` Sverre Rabbelier
2009-03-26 14:18       ` Gilbert Liddell
2009-03-26 14:34       ` Johannes Schindelin
2009-03-26 14:35   ` Anton Gyllenberg
2009-03-27 11:18     ` Anton Gyllenberg
2009-03-29  6:08       ` Eric Wong
2009-03-29  6:10         ` [PATCH] git-svn: fix ls-tree usage with dash-prefixed paths Eric Wong
2009-03-29 20:33           ` Junio C Hamano
2009-03-29 21:56             ` Eric Wong
2009-03-30  6:44               ` Junio C Hamano
2009-03-30 17:41                 ` Eric Wong
2009-03-30 18:05                   ` Junio C Hamano
2009-03-30 22:58                     ` Eric Wong
2009-03-31  7:11                       ` Björn Steinbrink
2009-03-31  7:31                         ` Björn Steinbrink
2009-03-31  9:41                           ` Björn Steinbrink
2009-03-31 15:05                             ` [PATCH] tree_entry_interesting: Only recurse when the pathspec is a leading path component Björn Steinbrink
2009-04-02  4:32                               ` Junio C Hamano
2009-04-02  4:41                                 ` Junio C Hamano [this message]
2009-04-02 16:36                                   ` [PATCH] match_tree_entry(): a pathspec only matches at directory boundaries Linus Torvalds
2009-04-02 11:38                                 ` [PATCH] tree_entry_interesting: Only recurse when the pathspec is a leading path component Björn Steinbrink
2009-04-03 16:25                                   ` Junio C Hamano
2009-03-30  5:28             ` [PATCH] git-svn: fix ls-tree usage with dash-prefixed paths Björn Steinbrink
2009-03-30  7:26         ` svn clone Checksum mismatch question Anton Gyllenberg
2009-03-26 14:34 ` Peter Harris

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=7vwsa3llac.fsf_-_@gitster.siamese.dyndns.org \
    --to=gitster@pobox.com \
    --cc=B.Steinbrink@gmx.de \
    --cc=anton@iki.fi \
    --cc=git@vger.kernel.org \
    --cc=normalperson@yhbt.net \
    --cc=torvalds@linux-foundation.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.