git@vger.kernel.org mailing list mirror (one of many)
 help / color / mirror / code / Atom feed
From: Jonathan Nieder <jrnieder@gmail.com>
To: Eric Wong <normalperson@yhbt.net>
Cc: "Michael G. Schwern" <schwern@pobox.com>,
	git@vger.kernel.org, gitster@pobox.com, robbat2@gentoo.org,
	bwalton@artsci.utoronto.ca
Subject: [PATCH v3] git svn: work around SVN 1.7 mishandling of svn:special changes
Date: Wed, 10 Oct 2012 13:47:33 -0700	[thread overview]
Message-ID: <20121010204733.GA4517@elie.Belkin> (raw)
In-Reply-To: <20121010201125.GA30952@dcvr.yhbt.net>

Subversion represents symlinks as ordinary files with content starting
with "link " and the svn:special property set to "*".  Thus a file can
switch between being a symlink and a non-symlink simply by toggling
its svn:special property, and new checkouts will automatically write a
file of the appropriate type.  Likewise, in subversion 1.6 and older,
running "svn update" would notice changes in filetype and update the
working copy appropriately.

Starting in subversion 1.7 (issue 4091), changes to the svn:special
property trip an assertion instead:

	$ svn up svn-tree
	Updating 'svn-tree':
	svn: E235000: In file 'subversion/libsvn_wc/update_editor.c' \
	line 1583: assertion failed (action == svn_wc_conflict_action_edit \
	|| action == svn_wc_conflict_action_delete || action == \
	svn_wc_conflict_action_replace)

Revisions prepared with ordinary svn commands ("svn add" and not "svn
propset") don't trip this because they represent these filetype
changes using a replace operation, which is approximately equivalent
to removal followed by adding a new file and works fine.  Follow suit.

Noticed using t9100.  After this change, git-svn's file-to-symlink
changes are sent in a format that modern "svn update" can handle and
tests t9100.11-13 pass again.

Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
---
Eric Wong wrote:

> I needed to filter the patch through:
>
>     s,git-svn\.perl,perl/Git/SVN/Editor.pm,g
>
> though...

Yeah, good catch.  Here's a v3 tested against "master".  Unlike in v2,
it remembers to pass the $deletions parameter to D() and A() --- which
shouldn't make a difference because we are not adding a directory, but
it's nice to be consistent to make reading smoother.

 perl/Git/SVN/Editor.pm |   25 ++++++++++++++++++++++++-
 1 file changed, 24 insertions(+), 1 deletion(-)

diff --git a/perl/Git/SVN/Editor.pm b/perl/Git/SVN/Editor.pm
index 755092fd..178920c8 100644
--- a/perl/Git/SVN/Editor.pm
+++ b/perl/Git/SVN/Editor.pm
@@ -345,7 +345,30 @@ sub M {
 	$self->close_file($fbat,undef,$self->{pool});
 }
 
-sub T { shift->M(@_) }
+sub T {
+	my ($self, $m, $deletions) = @_;
+
+	# Work around subversion issue 4091: toggling the "is a
+	# symlink" property requires removing and re-adding a
+	# file or else "svn up" on affected clients trips an
+	# assertion and aborts.
+	if (($m->{mode_b} =~ /^120/ && $m->{mode_a} !~ /^120/) ||
+	    ($m->{mode_b} !~ /^120/ && $m->{mode_a} =~ /^120/)) {
+		$self->D({
+			mode_a => $m->{mode_a}, mode_b => '000000',
+			sha1_a => $m->{sha1_a}, sha1_b => '0' x 40,
+			chg => 'D', file_b => $m->{file_b}
+		}, $deletions);
+		$self->A({
+			mode_a => '000000', mode_b => $m->{mode_b},
+			sha1_a => '0' x 40, sha1_b => $m->{sha1_b},
+			chg => 'A', file_b => $m->{file_b}
+		}, $deletions);
+		return;
+	}
+
+	$self->M($m, $deletions);
+}
 
 sub change_file_prop {
 	my ($self, $fbat, $pname, $pval) = @_;
-- 
1.7.10.4

  reply	other threads:[~2012-10-10 20:47 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-07-28  9:47 Fix git-svn for SVN 1.7 Michael G. Schwern
2012-07-28  9:47 ` [PATCH 1/8] SVN 1.7 will truncate "not-a%40{0}" to just "not-a" Michael G. Schwern
2012-07-28 14:16   ` Jonathan Nieder
2012-07-28 19:32     ` Michael G Schwern
2012-10-09  8:41       ` [PATCH/RFC] svn test: escape peg revision separator using empty peg rev Jonathan Nieder
2012-10-09  9:47         ` Michael J Gruber
2012-10-09 10:19           ` Jonathan Nieder
2012-10-10 20:37             ` Eric Wong
2012-10-10 21:02               ` Jonathan Nieder
2012-10-10 21:31                 ` Eric Wong
2012-10-10 21:42                   ` Jonathan Nieder
2012-10-10 22:16                     ` Eric Wong
2012-10-10 22:33               ` Junio C Hamano
2012-07-28  9:47 ` [PATCH 2/8] Fix typo in test Michael G. Schwern
2012-07-28  9:47 ` [PATCH 3/8] Improve our URL canonicalization to be more like SVN 1.7's Michael G. Schwern
2012-07-28  9:47 ` [PATCH 4/8] Replace hand rolled URL escapes with canonicalization Michael G. Schwern
2012-07-28  9:47 ` [PATCH 5/8] Canonicalize earlier in a couple spots Michael G. Schwern
2012-07-28  9:47 ` [PATCH 6/8] Add function to append a path to a URL Michael G. Schwern
2012-07-28  9:47 ` [PATCH 7/8] Turn on canonicalization on newly minted URLs Michael G. Schwern
2012-10-06 19:24   ` [PATCH/RFC] test: work around SVN 1.7 mishandling of svn:special changes Jonathan Nieder
2012-10-09 10:12     ` [PATCH/RFC v2] git svn: " Jonathan Nieder
2012-10-10 20:11       ` Eric Wong
2012-10-10 20:47         ` Jonathan Nieder [this message]
2012-07-28  9:47 ` [PATCH 8/8] Remove some ad hoc canonicalizations Michael G. Schwern
2012-07-30 20:38 ` Fix git-svn for SVN 1.7 Eric Wong
2012-07-30 21:10   ` Michael G Schwern
2012-07-30 22:15     ` Eric Wong
2012-07-31  1:04       ` Michael G Schwern
2012-07-31  2:18         ` Eric Wong
2012-07-31  4:30           ` Michael G Schwern
2012-07-31  6:53   ` Junio C Hamano
2012-07-31  9:54     ` Michael G Schwern
2012-07-31 20:01       ` Eric Wong
2012-07-31 23:05         ` Junio C Hamano
2012-07-31 23:28           ` Michael G Schwern
2012-07-31 23:24         ` Michael G Schwern
2012-08-01 21:30           ` Eric Wong
2012-08-02 10:31 ` Eric Wong
2012-08-02 16:07   ` Jonathan Nieder
2012-08-02 18:58     ` Junio C Hamano
2012-08-02 19:50       ` Robin H. Johnson
2012-08-02 22:10         ` Eric Wong
2012-08-21  4:04       ` Junio C Hamano
2012-08-21 21:03         ` Eric Wong
2012-08-21 21:34           ` Junio C Hamano
2012-08-02 20:51     ` Eric Wong
2012-08-02 21:22       ` Junio C Hamano
2012-08-02 21:42         ` Eric Wong
2012-08-02 21:55           ` Eric Wong
2012-08-02 22:05             ` 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

  List information: http://vger.kernel.org/majordomo-info.html

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20121010204733.GA4517@elie.Belkin \
    --to=jrnieder@gmail.com \
    --cc=bwalton@artsci.utoronto.ca \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=normalperson@yhbt.net \
    --cc=robbat2@gentoo.org \
    --cc=schwern@pobox.com \
    /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.
Code repositories for project(s) associated with this public inbox

	https://80x24.org/mirrors/git.git

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 read-only IMAP folder(s) and NNTP newsgroup(s).