[PATCH 3 of 3] cat: use the topmost repo when expanding node based format strings for output

Pierre-Yves David pierre-yves.david at ens-lyon.org
Mon Apr 14 00:16:36 CDT 2014



On 03/23/2014 11:04 PM, Matt Harbison wrote:
> Siddharth Agarwal wrote:
>> On 03/22/2014 10:08 AM, Matt Harbison wrote:
>>> # HG changeset patch
>>> # User Matt Harbison <matt_harbison at yahoo.com>
>>> # Date 1395456880 14400
>>> # Node ID f4b974de9c952b027d8f137eb4d302d72eb7ea60
>>> # Parent 35686f3835c908e3e681a0abd0c805476ccdfddc
>>> cat: use the topmost repo when expanding node based format strings for
>>> output
>>
>> Your first two patches look fine, but I'm not sure going to such
>> contortions to fix undocumented features is worth it -- do we want to
>> document them or not?
>>
>
> I'm not sure- I wasn't aware these worked either, and only stumbled
> across them accidentally while looking at something else.  I was really
> surprised how long they have been available and not documented, but
> maybe cat'ing to stdout and using shell redirection is how most people
> use it and nobody else thought to try these either?
>
> I don't feel strongly about this one- I just submitted it because if
> somebody does document these someday, I wasn't sure if we would be able
> to make this (or equivalent) change in the future due to backward
> compatibility rules.  It's not likely anybody would think to test
> subrepos when simply documenting what already works.
>
> It also looks easy enough to disable these- just don't pass the node to
> makefileobj().  But the hash and rev expansions at least seem marginally
> useful.

Gentle bump about this patches. Do we want this fixed or not?


-- 
Pierre-Yves


More information about the Mercurial-devel mailing list