[PATCH 2 of 2] commitextras: make sure keys are non empty ASCII strings

Kevin Bullock kbullock+mercurial at ringworld.org
Fri Jul 21 11:53:52 EDT 2017


> On Jul 20, 2017, at 09:36, Yuya Nishihara <yuya at tcha.org> wrote:
> 
> On Wed, 19 Jul 2017 20:00:36 +0530, Pulkit Goyal wrote:
>> On Tue, Jul 18, 2017 at 7:02 PM, Yuya Nishihara <yuya at tcha.org> wrote:
>>> On Tue, 18 Jul 2017 17:48:11 +0530, Pulkit Goyal wrote:
>>>> # HG changeset patch
>>>> # User Pulkit Goyal <7895pulkit at gmail.com>
>>>> # Date 1500378367 -19800
>>>> #      Tue Jul 18 17:16:07 2017 +0530
>>>> # Node ID 6cd19a6fd50016034d2faf7462bafdba4b90659e
>>>> # Parent  64137cfad7e30a459bc7fa5b6fc0e8757a6add8e
>>>> # EXP-Topic fbext
>>>> commitextras: make sure keys are non empty ASCII strings
>>>> 
>>>> diff --git a/hgext/commitextras.py b/hgext/commitextras.py
>>>> --- a/hgext/commitextras.py
>>>> +++ b/hgext/commitextras.py
>>>> @@ -52,6 +52,11 @@
>>>>                                 "KEY=VALUE format")
>>>>                         raise error.Abort(msg % raw)
>>>>                     k, v = raw.split('=', 1)
>>>> +                    isascii = all(ord(c) < 128 for c in k)
>>> 
>>> Better to reject non-word characters. Perhaps ascii_letters + digits + '_-'
>>> should be enough.
>>> 
>>> changelog uses ':' as a separator.
>>> 
>>> https://www.mercurial-scm.org/repo/hg/file/4.2.2/mercurial/changelog.py#l66
>> 
>> Can I still follow up on this?
> 
> Maybe okay as long as it is considered a simple bug fix.

Yes, I'd be okay with taking a follow-up during the freeze.

pacem in terris / мир / शान्ति / ‎‫سَلاَم‬ / 平和
Kevin R. Bullock



More information about the Mercurial-devel mailing list