This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: 74 columns for ChangeLogs? [was Re: [PATCH 1/6] gdbserver (linux_debug): Remove extraneous \n from output.]
- From: Pedro Alves <palves at redhat dot com>
- To: Doug Evans <dje at google dot com>
- Cc: gdb-patches <gdb-patches at sourceware dot org>
- Date: Wed, 18 Dec 2013 17:43:26 +0000
- Subject: Re: 74 columns for ChangeLogs? [was Re: [PATCH 1/6] gdbserver (linux_debug): Remove extraneous \n from output.]
- Authentication-results: sourceware.org; auth=none
- References: <CADPb22RCz29jtrtnk5sd+OVP7nB34O-b9j+sig-MhcC-PQU7mg at mail dot gmail dot com>
On 12/18/2013 05:28 PM, Doug Evans wrote:
> On Wed, Dec 18, 2013 at 3:13 AM, Pedro Alves <palves@redhat.com> wrote:
>> On 12/17/2013 09:40 PM, Doug Evans wrote:
>>> 2013-12-17 Doug Evans <dje@google.com>
>>>
>>> * nat/linux-waitpid.c (linux_debug): Remove extraneous \n from output.
>>
>> (Line too long. fill-column is 74 for ChangeLogs)
>
> Is that a hard and fast rule now, or just a nice to have.
It's not a new rule. It's been that way since forever.
It's emacs's default in changelog mode, and also what most
ChangeLog files in the tree have explicitly at the bottom:
fill-column: 74
So M-q in ChangeLog should be doing that for you automatically.
And also mentioned in the ContributionCheckist wiki page.
> I always understood it to be a nice-to-have:
> I don't have to look very far into ChangeLog to find a plethora of
> violations from various people, and I know I didn't review all those
> patches.
Sure, bugs and wrong formatting always pass through review all
the time. What's the point? Honestly, I was just expecting
you'd say something like "whoops, forgot to M-q", but it
sounds like you did it on purpose. This was already discussed
not so long ago, and I didn't mean to open yet another discussion,
so I'm already regretting I pointed it out.
--
Pedro Alves