[PATCH] pager: honour aliases (issue3532)

[Matt Mackall]

On Sun, 2013-10-20 at 15:30 -0700, David Soria Parra wrote:
> On 10/20/2013 03:27 PM, Matt Mackall wrote:
> > On Sat, 2013-10-12 at 19:01 -0700, David Soria Parra wrote:
> >> # HG changeset patch
> >> # User David Soria Parra <dsp at experimentalworks.net>
> >> # Date 1381629094 25200
> >> #      Sat Oct 12 18:51:34 2013 -0700
> >> # Node ID 39084a836b0525cbcabc60b5200e83a5cd91577d
> >> # Parent  1b2f9d36953e6ed384a044c1e73cb3a1aa072004
> >> pager: honour aliases (issue3532)
> >>
> >> If paging is configured for a command all it's aliases will be paged as
> >> well. This will make attend=log cause 'hg history' to run the pager as
> >> well as custom aliases defined with [alias].
> > 
> > I think I have to say no to this one. People might be using aliases to
> > avoid/get paging today so this is a clear BC breakage... and offers
> > nothing users can't do for themselves in their config file (which
> > they're already editing), possibly by adding --pager. 
> > 
> 
> Fair enough, I still think that attend on log should also page
> "history". So that predefined aliases in commands should be paged.

Agreed. I'd be surprised if anyone actually uses 'hg history' though.
Most of the other variants are prefixes (in vs incoming) or things you
probably don't care about paging (checkout vs update).

-- 
Mathematics is the supreme nostalgia of our time.