On Thu, 2008-04-03 at 16:16 +0200, Jelmer Vernooij wrote:
> On Do, 2008-04-03 at 16:10 +0200, Daniel Schierbeck wrote:
> > A regression was introduced which would raise an exception when
> > VerifyText returned an empty key string. This fixes that issue.
> bb:resubmit
> 
> We shoudln't be calling discover() at all imho. There's a separate
> setting for finding keys automatically in seahorse.

I talked with Adam Schreiber from the Seahorse project, and he told me
that I needed to call DiscoverKeys(), or else GetKeyField() would fail
on unknown keys. This is separate from VerifyText() returning an empty
key string, which was apparently caused by the key being marked as
missing in Seahorse. Furthermore, the discover will not fail if there's
no network connection, and doesn't introduce any significant overhead.
I asked if there was another way to solve this issue, but none has been
found.

The API will change in the future, though, based partially on my
conversation with Adam, so we can use a saner solution then.


Cheers,
Daniel

Attachment: signature.asc
Description: This is a digitally signed message part

-- 
bzr-gtk mailing list
[email protected]
Modify settings or unsubscribe at: 
https://lists.canonical.com/mailman/listinfo/bzr-gtk

Reply via email to