[PATCH 1 of 2] notify: various fixes to docstring

[David Champion]

* On 12 Jun 2012, Matt Mackall wrote: 
> On Tue, 2012-05-29 at 07:58 -0500, David Champion wrote:
> > # HG changeset patch
> > # User David Champion <dgc at uchicago.edu>
> > # Date 1338293230 18000
> > # Node ID 222934f53810943a6e7eb74ebd756ea8cafa810e
> > # Parent  2ac08d8b21aa7b6e0a062afed5a3f357ccef67f9
> > notify: various fixes to docstring
> > 
> > Typo corrections, grammar adjustments, clarity improvements.
> 
> This broke two tests, so I'm dropping it for now.
> 
> Most importantly, there's an RST issue buried in here somewhere that's
> breaking gendoc.

I've fixed the tests, but that causes a new problem.

AFAICT the correct way to do an indented dictionary list is:

   Space-separated list of change sources. Notifications are activated only
   when a changeset's source is in this list. Sources may be:

   :``serve``: changesets received via http or ssh
   :``pull``: changesets received via ``hg pull``
   :``unbundle``: changesets received via ``hg unbundle``
   :``push``: changesets sent or received via ``hg push``
   :``bundle``: changesets sent via ``hg unbundle``

   ...

That is, the list should be indented equally with the surrounding
indent.  When I do that the test passes.  The reason I initially did not
do that is that when hg processes the docstring to produce 'hg help'
output, it does *not* parse the RST.  This block comes out as literal
RST.  When I unindent it 'hg help' DWIM.

So... Martin?  Someone who knows mercurial's rst?  What's the right
solution, other than just not using a dictionary list?

-- 
David Champion • dgc at uchicago.edu • IT Services • University of Chicago