[18:51] <dhillon-v10> hi all, I am trying to implement this feature that would allow users to have more than one assignee/approver for one spec. and I know it involves some modification to PublicPersonChoice() method and it could be something similar to the Add Dependancy where I can return a link that a person can add to, but is this the right way ?
[18:52] <dhillon-v10> jamalta: hi :)
[19:25] <mwhudson> good morning
[19:26] <dhillon-v10> mwhudson: hi :) just read your email on the api docs. thanks for that, i was looking for something like that for a while now
[19:26] <mwhudson> dhillon-v10: cool
[19:27] <dhillon-v10> mwhudson: i need some help implementing a new feature (a bug) do you have like 5-10 mins.
[19:27] <mwhudson> dhillon-v10: sure
[19:27] <dhillon-v10> mwhudson: I am trying to implement this feature that would allow users to have more than one assignee/approver for one spec. and I know
[19:27] <dhillon-v10>                      it involves some modification to PublicPersonChoice() method and it could be something similar to the Add Dependancy where I can return
[19:27] <dhillon-v10>                      a link that a person can add to, but is this the right way ?
[19:27] <dhillon-v10> mwhudson: sorry about that
[19:28] <dhillon-v10> mwhudson: copy and paste in irssi fails
[19:28] <mwhudson> dhillon-v10: um, it will require a database change, i think?
[19:28] <dhillon-v10> mwhudson: oh i see because previously there was one column to record that info. now there are more because of additional people
[19:29] <mwhudson> dhillon-v10: right, you'll need a linking table
[19:29] <mwhudson> dhillon-v10: i don't mean to be overly picky, but is there consensus that this is something that should be done?
[19:29] <mwhudson> (a link to a bug report, an email thread, or something would be good)
[19:30] <mwhudson> (i don't know much about blueprints)
[19:30] <dhillon-v10> mwhudson: I don't know, Curtis marked this one as a wishlist, and I thought I might get started on this one, let me get you the link in a sec.
[19:31] <dhillon-v10> mwhudson: https://bugs.launchpad.net/blueprint/+bug/2040
[19:31] <mwhudson> 4 figure bug, awesome!
[19:33] <dhillon-v10> mwhudson: so what do you say, is it even worth doing, because if its a big change in the database then we can  probably not do it or find a workaround
[19:33] <mwhudson> dhillon-v10: it's not a very complicated change in the database really, so i don't think that's not a reason to do it
[19:34] <dhillon-v10> mwhudson: alright then, so where do I get started, I have the dev environment setup and have been playing with it for a while now
[19:38] <mwhudson> dhillon-v10: maybe a ui mockup is the place to start
[19:40] <dhillon-v10> mwhudson: so first focus on the INewSpecification part so I can have a baseline ready then go into the database and make necessary changes
[19:41] <mwhudson> dhillon-v10: i think i meant, "focus on what you want the user experience to be"
[19:41] <mwhudson> i admit i don't use blueprints much
[19:43] <dhillon-v10> mwhudson: can you please rephrase that "focus on what you want the user experience to be" is it basically saying make like a little screenshot of what the end result could look like
[19:44] <mwhudson> dhillon-v10: yeah
[19:44] <mwhudson> dhillon-v10: it can just be a pen and paper drawing scanned, or mutilate the page in firebug or something
[19:44] <mwhudson> dhillon-v10: don't put too much effort into it at this stage
[19:45] <mwhudson> dhillon-v10: is the role of the blueprint approver enforced in code?
[19:46] <dhillon-v10> mwhudson: yeah I understand, a mock-up shouldn't take too much effort, and don't know about approver enforced part, I am still reading the code so sorry about that
[19:47] <mwhudson> dhillon-v10: if you can catch curtis (aka sinzui) online, he might be a good person to talk to about this
[19:47] <dhillon-v10> mwhudson: alright and thanks a *lot* for you help, you rock :)
[19:48] <mwhudson> dhillon-v10: no worries, i hope your first change to launchpad goes well!
[19:48] <dhillon-v10> sinzui: ping :)
[19:48] <dhillon-v10> mwhudson: same here :)
[19:48] <mwhudson> hmm, sinzui is on uk time this week
[19:50] <dhillon-v10> mwhudson: I guess I'll wait if he comes back or otherwise I can just email him
[19:50] <mwhudson> yeah
[20:00] <mwhudson> sinzui: are you actually there?
[22:13] <drubin> Is there a link to launchpadapi docs? I have found the raw webservice doc doesn't match up 100% ant the little changes make it kinda hard to work out what the python object relations are
[22:32] <dhillon-v10> drubin: try this out: http://people.canonical.com/~mwh/canonicalapi/index.html it might help you out
[22:33] <wgrant> dhillon-v10: That's the internal API documentation.
[22:33] <dhillon-v10> wgrant: oops sorry you are right :)
[22:33] <drubin> dhillon-v10: ye, I was a little bit overwelmed
[22:35] <drubin> ye the biggest issue is getting the correct attributes from the python objects. Lets say I want a team members email. I have to do lp.people['teamname'].preferred_email_address.email    where as in the raw API it lists preferred_email_address_link there are a few of those. just wondering what else is different
[22:35] <wgrant> drubin: foo_link -> foo, foo_collection_link -> foo
[22:35] <wgrant> I don't think there's much else.
[22:35] <drubin> wgrant: Is that the only difference?
[22:35] <drubin> If that is the case I am pretty sure I can cope :)
[22:36] <drubin> wgrant: Thanks. Pretty much what I was looking for.
[22:36] <drubin> Is gzip compression turned on by default? If not where would I look to turn it on.
[22:37] <wgrant> drubin: They're just automatic niceties provided by launchpadlib. foo_link webservice attributes contain a URL, but launchpadlib knows to instead produce a foo attribute containing the object itself.
[22:38] <drubin> wgrant: Ye I kinda worked that out, after the first 5mins of using the lib. Was more worried about other stuff I might encounter down the line.
[23:10] <mwhudson> woo, finally down to 0 unread
[23:18] <mwhudson> "Diff against target: 248083 lines (+22639/-108557) 345 files modified"
[23:19] <mwhudson> hard to see how that's ever going to be useful