From: Taylor Blau <me@ttaylorr•com>
To: Jeff King <peff@peff•net>
Cc: Junio C Hamano <gitster@pobox•com>,
Vasil Dimov via GitGitGadget <gitgitgadget@gmail•com>,
git@vger•kernel.org, Vasil Dimov <vd@freebsd•org>
Subject: Re: [PATCH 1/2] range-diff: fix a crash in parsing git-log output
Date: Wed, 15 Apr 2020 16:02:42 -0600 [thread overview]
Message-ID: <20200415220242.GA45241@syl.local> (raw)
In-Reply-To: <20200415162326.GG2464307@coredump.intra.peff.net>
Hi Peff,
On Wed, Apr 15, 2020 at 12:23:26PM -0400, Jeff King wrote:
> On Wed, Apr 15, 2020 at 08:31:39AM -0700, Junio C Hamano wrote:
>
> > "Vasil Dimov via GitGitGadget" <gitgitgadget@gmail•com> writes:
> >
> > > From: Vasil Dimov <vd@FreeBSD•org>
> > >
> > > `git range-diff` calls `git log` internally and tries to parse its
> > > output. But `git log` output can be customized by the user in their
> > > git config and for certain configurations either an error will be
> > > returned by `git range-diff` or it will crash.
> > >
> > > To fix this explicitly set the output format of the internally
> > > executed `git log` with `--pretty=medium`. Because that cancels
> > > `--notes`, add explicitly `--notes` at the end.
> >
> > Good finding.
> >
> > Shouldn't we also disable customizations that come from the
> > configuration variables like diff.external, diff.<driver>.command?
>
> If range-diff were a script, I would say it should be using the
> "rev-list | diff-tree --stdin" plumbing under the hood, rather than
> "log".
>
> The read_patches() function does let callers pass options to git-log,
> but I don't _think_ this is exposed to the user. We only allow a few
> --notes options to be passed, and we should be able to apply those to
> diff-tree. So converting it to use plumbing might be an option.
>
> Though I think there is another bug:
>
> $ git rev-list HEAD | git diff-tree --stdin --pretty=medium --notes
> commit 8f3d9f354286745c751374f5f1fcafee6b3f3136
> git: notes.c:1308: format_display_notes: Assertion `display_notes_trees' failed.
> Aborted
Nice find. I think that I have a patch addressing this below. Please let
me know what you think:
--- >8 ---
Subject: [PATCH] diff-tree.c: load notes machinery with '--notes'
Since its introduction in 7249e91 (revision.c: support --notes
command-line option, 2011-03-29), combining '--notes' with '--pretty'
causes 'git diff-tree' to fail a runtime assertion:
$ git rev-list HEAD | git diff-tree --stdin --pretty=medium --notes
commit 8f3d9f354286745c751374f5f1fcafee6b3f3136
git: notes.c:1308: format_display_notes: Assertion `display_notes_trees' failed.
Aborted
This failure is due to diff-tree not calling 'load_display_notes' to
initialize the notes machinery.
Ordinarily, this failure isn't triggered, because it requires passing
both '--notes' and '--pretty'. Specifically, passing '--pretty' sets
'opt->verbose_header', causing 'show_log()' to eventually call
'format_display_notes()', which expects a non-NULL 'display_note_trees'.
Without initializing the notes machinery, 'display_note_trees' remains
NULL, and thus triggers an assertion failure. This doesn't occur without
'--pretty' since we never call 'format_display_notes()' without it.
Fix this by initializing the notes machinery after parsing our options,
and harden this behavior against regression with a test in t4013.
Reported-by: Jeff King <peff@peff•net>
Signed-off-by: Taylor Blau <me@ttaylorr•com>
---
builtin/diff-tree.c | 2 ++
t/t4013-diff-various.sh | 1 +
2 files changed, 3 insertions(+)
diff --git a/builtin/diff-tree.c b/builtin/diff-tree.c
index cb9ea79367..17c1cc8c3c 100644
--- a/builtin/diff-tree.c
+++ b/builtin/diff-tree.c
@@ -126,6 +126,8 @@ int cmd_diff_tree(int argc, const char **argv, const char *prefix)
precompose_argv(argc, argv);
argc = setup_revisions(argc, argv, opt, &s_r_opt);
+ if (opt->show_notes)
+ load_display_notes(&opt->notes_opt);
while (--argc > 0) {
const char *arg = *++argv;
diff --git a/t/t4013-diff-various.sh b/t/t4013-diff-various.sh
index dde3f11fec..6ae8cfb271 100755
--- a/t/t4013-diff-various.sh
+++ b/t/t4013-diff-various.sh
@@ -398,6 +398,7 @@ diff --no-index --raw --no-abbrev dir2 dir
diff-tree --pretty --root --stat --compact-summary initial
diff-tree --pretty -R --root --stat --compact-summary initial
+diff-tree --pretty --notes initial
diff-tree --stat --compact-summary initial mode
diff-tree -R --stat --compact-summary initial mode
EOF
--
2.26.0.106.g9fadedd637
next prev parent reply other threads:[~2020-04-15 22:02 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-04-15 14:28 [PATCH 0/2] range-diff: fix a crash in parsing git-log output Vasil Dimov via GitGitGadget
2020-04-15 14:28 ` [PATCH 1/2] " Vasil Dimov via GitGitGadget
2020-04-15 15:31 ` Junio C Hamano
2020-04-15 16:16 ` Vasil Dimov
2020-04-15 16:23 ` Jeff King
2020-04-15 22:02 ` Taylor Blau [this message]
2020-04-15 22:29 ` Jeff King
2020-04-15 16:13 ` Taylor Blau
2020-04-15 14:28 ` [PATCH 2/2] range-diff: avoid negative string precision Vasil Dimov via GitGitGadget
2020-04-15 16:20 ` Taylor Blau
2020-04-15 20:19 ` Vasil Dimov
2020-04-15 20:32 ` [PATCH v2 0/2] range-diff: fix a crash in parsing git-log output Vasil Dimov via GitGitGadget
2020-04-15 20:32 ` [PATCH v2 1/2] " Vasil Dimov via GitGitGadget
2020-04-15 20:32 ` [PATCH v2 2/2] range-diff: avoid negative string precision Vasil Dimov via GitGitGadget
2020-04-16 1:07 ` [PATCH v2 0/2] range-diff: fix a crash in parsing git-log output Taylor Blau
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=20200415220242.GA45241@syl.local \
--to=me@ttaylorr$(echo .)com \
--cc=git@vger$(echo .)kernel.org \
--cc=gitgitgadget@gmail$(echo .)com \
--cc=gitster@pobox$(echo .)com \
--cc=peff@peff$(echo .)net \
--cc=vd@freebsd$(echo .)org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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