[PATCH] identify: add template support

[Matt Harbison]

On Sun, 25 Jun 2017 20:41:21 -0400, Yuya Nishihara <yuya at tcha.org> wrote:

> On Sun, 25 Jun 2017 16:20:46 -0400, Matt Harbison wrote:
>> On Sun, 25 Jun 2017 08:19:29 -0400, Yuya Nishihara <yuya at tcha.org>  
>> wrote:
>>
>> > On Sun, 25 Jun 2017 01:23:41 -0400, Matt Harbison wrote:
>> >> # HG changeset patch
>> >> # User Matt Harbison <matt_harbison at yahoo.com>
>> >> # Date 1498360161 14400
>> >> #      Sat Jun 24 23:09:21 2017 -0400
>> >> # Node ID 1e917448c8d5c36f19c41bc83793df3e92477ede
>> >> # Parent  8299eb9b08c79699f496717d626842b72fa5ca4f
>> >> identify: add template support
>> >
>> ...
>>
>> >
>> >> Yuya mentioned using fm.nested() in that thread, so I did for the
>> >> parent nodes.
>> >> I'm not sure if the plan is to fill in all of the context attributes  
>> in
>> >> these
>> >> items, or if these nested items should simply be made {p1node} and
>> >> {p1rev}.
>> >
>> > [...]
>> >
>> >> +            for i, p in enumerate(parents):
>> >> +                fn = fm.nested('p%d' % (i + 1))
>> >> +                fn.startitem()
>> >> +                fn.data(rev=p.rev())
>> >> +                fn.data(node=p.hex())
>> >> +                fn.end()
>> >
>> > I meant this could be a {parents} list, which is
>> >
>> >   fn = fm.nested('parents')
>> >   for ...
>> >       fn.startitem()
>> >       ...
>> >
>> > Each item could have fn.context(ctx=p) so that all template keywrods  
>> are
>> > available.
>>
>> I saw that, and tried it, but I didn't think it worked.  I was expecting
>> lines in the json output I guess.  But after digging in a bit more,  
>> maybe
>> it's just a bug in nested formatters?
>
> Yeah, it's the bug of formatter. I'll send a fix shortly.

Actually, it seems to work if fm.context(ctx=ctx) is set too.  But now  
there's an off by one issue:

$ ../hg log -r . -T "{latesttag % '{changes}'}
913
$ ../hg id -T "{p1 % '{latesttag() % \'{changes}\'}'}"
914

(Unless the bug is fn is only looking at fm's ctx?)