[revised PATCH stable] document 'default' and 'default-push' in paths docstring

[Faheem Mitha]

Hi Martin,

On Fri, 16 Apr 2010, Martin Geisler wrote:

> On Fri, Apr 16, 2010 at 23:12, Faheem Mitha <faheem at email.unc.edu> wrote:

> Oh, sorry about that, I was just happy to finally push a patch again :-)

You mean, because you've been out of the loop with moving to Switzerland? 
I hope you are having a nice time there.

> I sounds like the tests cover some cases and the rest (pull with
> default set) is covered by our daily usage. So I don't see a big need
> for such tests. But if you write them, then I'll of course be happy to
> push them.

Ok, well, lets not worry about it for now then.

>> -    The names 'default' and 'default-push' have a special meaning.
>> -    They are the locations used when pulling and pushing respectively
>> -    unless a location is specified. When cloning a repository, the
>> -    clone source is written as 'default' in .hg/hgrc.
>> +    For operations requiring a location, if one is not specified, the
>> +    path names 'default' and 'default-push', if defined, are used as
>> +    fallbacks for incoming and outgoing operations, respectively.
>
> I don't like the flow in that sentence, I think it is chopped up into
> too many pieces. What about:
>
>  The names 'default' and 'default-push' have a special meaning. Doing
> 'hg pull' or 'hg incoming' will use 'default' when no location is
> given whereas 'hg push' and 'hg outgoing' will use 'default-push'. If
> 'default-push' is not defined, then 'default' is used as a fallback
> for both incoming and outgoing operations.

Does that mean you think the rest is Ok? :-)

The second sentence doesn't refer to the other commands like 'bundle', 
'email' etc. (news to me, btw) and doesn't say anything about 'default' 
and 'default-push' needing to be defined. Ok, perhaps that's 
redundant/understood. The third sentence uses 'incoming' and 'outgoing' 
which are also names of hg subcommands. Perhaps it is Ok because they are 
not surrounded by quotes.

I agree your version has better flow. If people think this is an 
improvement (possibly I'm being petty and nitpicking) we can go with that 
instead.

                                                           Regards, Faheem.