Re: [PATCH] git-send-email: add ~/.authinfo parsing

On Wed, 06 Feb 2013 17:41:01 +0100 Matthieu Moy [off-list ref] wrote: 

MM> Ted Zlatanov [off-list ref] writes:

quoted

- sort the output tokens (after 'url' is extracted) so the output is consistent and testable

MM> Why not, if you want to use the output of credential_write in tests. But
MM> credential_write is essentially used to talk to "git credential", so the
MM> important information is the content of the hash before credential_write
MM> and after credential_read. They are unordered, but consistent and
MM> testable.

I like testing output (especially when it's part of an API), so we
should make the externally observable output consistent and testable.

The change is tiny, just sort the keys instead of calling each(), so I
hope it makes it in the final version.

quoted

Yup.  But what you call "read" and "write" are, to the credential
helper, "write" and "read" but it's the same protocol :)  So maybe the
names should be changed to reflect that, e.g. "query" and "response."

MM> I don't think that would be a better naming. Maybe "serialize" and
MM> "parse" would be better, but "query" would sound like it establishes the
MM> connection and possibly reads the response to me.

I'm OK with anything unambiguous.

Thanks!
Ted