From: Phillip Wood <phillip.wood123@gmail.com>
To: Junio C Hamano <gitster@pobox.com>, Dominik Maier <dmnk@google.com>
Cc: git@vger.kernel.org, Patrick Steinhardt <ps@pks.im>,
Jeff King <peff@peff.net>, Taylor Blau <me@ttaylorr.com>,
Johannes Schindelin <Johannes.Schindelin@gmx.de>,
dmnk+artist@google.com
Subject: Re: [PATCH] config: skip continued lines when matching section
Date: Fri, 20 Nov 2026 14:33:31 +0000 [thread overview]
Message-ID: <67c5d781-bfe4-41b7-ae1d-0b0c473e9979@gmail.com> (raw)
In-Reply-To: <xmqqbj7ktjp7.fsf@gitster.g>
On 20/11/2026 00:44, Junio C Hamano wrote:
> Dominik Maier <dmnk@google.com> writes:
>
>> When a value in the target section spans multiple lines via a trailing
>> backslash and a continuation line begins with '[other]', removing or
>> renaming the section stops early at '[other]' and leaves the remainder
>> of the continued value behind as an unquoted '[other]' section.
>
> If the behaviour was to treat these as continued values from day
> one, I would have agreed that it would be a reasonably design, but
> offhand I do not know if this is a breaking change in behaviour that
> may negatively affect existing configuration files.
As I understand it, this patch changes the way we parse the config when
we're renaming or copying a section to match the way we parse it when we
read config values. It does not change how we read config settings so I
think it is simply a bug fix for an unlikely corner case when copying or
renaming a section. It's a pity we don't share a common code path for
the parsing and end up implementing it twice.
I had a quick look at the code in this patch and I'm not sure it handles
values and subsection names that contain carriage returns correctly. As
far as I can tell
printf '[a "b\rc"]\rkey=\rv\ralue' >/tmp/test.config
creates a valid config file. For example
git config get -f /tmp/test.config "$(printf 'a.b\rc.key')" |
tr '\r' Q
prints "vQal" but this patch treats '\r' like '\n'
Thanks
Phillip
>> @@ -3434,7 +3525,7 @@ static int repo_config_copy_or_rename_section_in_file(
>> const char *old_name,
>> const char *new_name, int copy)
>> {
>> - int ret = 0, remove = 0;
>> + int ret = 0, remove = 0, continued = 0, in_quote = 0;
>> char *filename_buf = NULL;
>> struct lock_file lock = LOCK_INIT;
>> int out_fd;
>> @@ -3496,9 +3587,12 @@ static int repo_config_copy_or_rename_section_in_file(
>> goto out;
>> }
>>
>> + int was_continued = continued;
>> +
>
> This triggers -Wdeclaration-after-statement doesn't it?
>
> (administrivia)
>
> I do not consider myself an area expert for the configuration file
> editing code. Please do not directly Cc: random people unless you
> know they are specifically familiar with the area or they are
> positively interested in the topic and they will regret if they miss
> your message if they fail to notice.
>
> Thanks.
prev parent reply other threads:[~2026-11-20 14:33 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-19 19:23 [PATCH] config: skip continued lines when matching section Dominik Maier
2026-11-20 0:44 ` Junio C Hamano
2026-11-20 14:33 ` Phillip Wood [this message]
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=67c5d781-bfe4-41b7-ae1d-0b0c473e9979@gmail.com \
--to=phillip.wood123@gmail.com \
--cc=Johannes.Schindelin@gmx.de \
--cc=dmnk+artist@google.com \
--cc=dmnk@google.com \
--cc=git@vger.kernel.org \
--cc=gitster@pobox.com \
--cc=me@ttaylorr.com \
--cc=peff@peff.net \
--cc=phillip.wood@dunelm.org.uk \
--cc=ps@pks.im \
/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.