On Feb 21, 2011, at 5:26 PM, Colin Barrett wrote: > No, the patch queue thing isn't particularly useful. 1 star. > > Do get in touch with Frank (Wixardy). He has been hacking on it a bit, but > ran into some problems. I can forward you the email I have with his changes > in it (just got it on Friday).
Forward it to the list, please, unless there's a strong reason to do otherwise. I can't even figure out how to find your AIFacebookXMPP* files within the qclone checkout. It appears to have checked out HEAD of the adium-1.4 branch, without your changes. I've unfortunately used up the couple hours I had... I'm glad to hear Frank is continuing work. Unless there are useful steps we're both missing, a "patch queue" isn't the right way to collaborate on a feature branch. Could we simply have a branch for this? -Evan > > -Colin > > On Mon, Feb 21, 2011 at 3:23 PM, Evan Schoenberg, M.D. <eva...@dreskin.net> > wrote: > > On Jan 6, 2011, at 4:04 PM, Colin Barrett wrote: > >> https://bitbucket.org/cbarrett/adium-facebook-xmpp >> >> Getting OAuth login working for Facebook is one of the outstanding tasks for >> 1.4.2. I have OAuth itself working fine w/ a test service type -- it just >> needs an acceptable UI hooked up and it needs to be integrated into the >> actual FB XMPP service type. I don't have as much free time as I thought and >> would love some help finishing it. >> >> The bitbucket link above btw is an mq patch repository. Sorry it's in a >> weird format but I was trying it out. This blog post may be able to help you >> figure it out: >> http://ches.nausicaamedia.com/articles/technogeekery/using-mercurial-queues-and-bitbucket-org >> If you need more help, ask me. > > I think it's a rule that any time I try to sit down to do some Adium work, > I'll spend more time dealing with non-Adium things than actually coding. > > In the last 2 days I have: > - Dealt with 2 unrelated build issues on my local system > - Dealt with a long merge issue to maintain a patch that we'd applied to > libpurple's IRC some time ago to provide 'away' status updates > - Subsequently ported it upstream so that others can benefit and we > don't have to worry about that merge issue in the future > - Done it wrong and had to disapprove all the revisions, figure out how > to switch branches in monotone (you can't, I think... you just do a new > checkout on the desired branch), and pluck them to > im.pidgin.pidgin.next.minor (since it adds API) > - Hunted down a change in libpurple (in which the typing flag was cleared > after calling the ui_op for displaying the message) which led to a crash in > Adium when our automatic handling of certain IRC messages caused the > conversation to be closed before the run loop canceled > > And now after reading the requisite how-to blog post (for a system which > appears to think it is perfectly reasonable to have steps like, "To push out > your changes for others, `cd .hg/patches` from the root of the repo you > qclone’d and `hg push`." Obviously!) I'm waiting for a fresh checkout of > adium-1.4 with this patch queue. So, did you decide this patch queue thing > was useful? > > -Evan >
