Re: [PATCH 3/3] remote: introduce and fill branch->pushremote

From: Ramkumar Ramachandra <artagnon@gmail.com>
Date: Jan 13, 2014, 11:22 UTC
Message-ID: <CALkWK0ncSLza3Q0PSZ0oTZqB2YxjgGSqA7QYxk2+rN_77BKZMA@mail.gmail.com>
In-Reply-To: <20140113083421.GA18531@sigill.intra.peff.net>

Jeff King wrote:

Show 13 quoted lines

>   2. If the current branch has a branch.*.pushremote set, but we want to
>      know where a _different_ branch would be pushed, we have no way to
>      access remote.pushdefault (it gets overwritten in the hunk above).
>
>      @{upstream} does not have this problem, because it is _only_
>      defined if branch.*.remote is set. There is no such thing as
>      defaulting to a "remote.default" (or "origin") there, and we never
>      need to look at default_remote_name.
>
>      For @{publish}, though, I think we will want that default. The
>      common config will be to simply set "remote.pushdefault", rather
>      than setting up "branch.*.pushremote" for each branch, and we would
>      want @{publish} to handle that properly.

Not sure I understand what the problem is. Let's say we have two branches: "master", and "side" with remote.pushdefault = ram, branch.*.remote = origin, and branch.side.pushremote = peff. Now, when I query master's pushremote, I get "ram" and when I query side's pushremote, I get "peff"; all the logic for falling-back from branch.*.pushremote to remote.pushdefault to branch.*.remote is in branch_get(), so I need to do nothing extra on the caller-side. From the caller's perspective, why does it matter if the pushremote of a particular branch is due to branch.*.pushremote or remote.pushdefault?