All times shown according to UTC.

Time Nick Message
00:37 paulproteus deardiary: spec'ing out tiers end-to-end on my user page on the Develop wiki.
00:59 zanoi deardiary: slow progress on per display shuffle..
02:38 odenthaa joined #miro-hackers
02:46 paulproteus deardiary: I think I have convinced the PayPal API to do what I want. Unless I'm mistaken, that is.
03:04 willkg paulproteus: w00t!  bash it into submission!
03:04 deardiary: i closed a couple of bugs that have been fixed in the last couple of weeks.
03:08 deardiary: i think i'll work on reworking fastresume tomorrow.  plus i want to look into libtorrent build flags on osx.
03:08 night all!
03:08 willkg left #miro-hackers
04:11 odenthaa left #miro-hackers
04:39 odenthaa joined #miro-hackers
04:45 DGMurdockIII joined #miro-hackers
06:27 DGMurdockIII left #miro-hackers
09:01 arpu joined #miro-hackers
10:01 epitron_ is now known as _epitron_
10:03 janetPCF joined #miro-hackers
10:33 arpu left #miro-hackers
10:47 arpu joined #miro-hackers
12:01 Jarrhead joined #miro-hackers
12:01 janetPCF1 joined #miro-hackers
12:04 janetPCF left #miro-hackers
13:30 z3p morning
14:02 willkg joined #miro-hackers
14:28 z3p deardiary: working on OS X TableView issues
14:28 willkg z3p: what osx tableview issues?
14:28 z3p willkg: #16157
14:29 willkg ahhh...  right.
14:31 z3p: i'm looking into the xgettext weirdness now.
14:32 z3p willkg: I don't think it's that big a deal; amazon.py just decrypts stuff
14:35 willkg interesting.  i can fix it by specifying the encoding of the file.
14:35 z3p willkg: isn't the encoding for that one 'ascii'?
14:38 willkg um...  yes.  the file is ascii.  however, since these are bytes that xgettext is having problems with, marking the file as ascii doesn't help since those values are outside the ascii range.
14:39 er, let me rephrase that because it's not clear that there's "file encoding" and "string encoding" here.
14:40 zanoi morning
14:40 willkg if i make the file encoding iso-8859-1 (which is a superset of ascii), then the xgettext error goes away.
14:40 which doesn't make a whole lot of sense.  so i'm still poking around.
14:52 z3p: no dice.  i don't really understand why xgettext is doing what it's doing.  seems like a bug, but it's hard to tell.
14:54 Jarrhead left #miro-hackers
15:07 Jarrhead joined #miro-hackers
15:07 willkg afk for pants, breakfast and coffee.
15:29 zanoi is the procedure and the policies of adding new stuff to the database documented anywhere?
15:42 Jarrhead left #miro-hackers
15:52 willkg zanoi: i don't know.
15:53 ajonas joined #miro-hackers
15:53 arpu left #miro-hackers
15:53 zanoi willkg: are there any policies or should i just try?
15:54 willkg i don't know if there are any policies.  i haven't done database changes in a while.  best to ping someone who has like ben, kaz, or z3p.
15:55 z3p zanoi: I'm not sure what you mean; have you looked at databaseupgrade.py?
15:56 zanoi z3p: with policies i mean if i can just "randomly" change databaseupgrade and other database files and push it
15:57 z3p zanoi: as long as your numbering doesn't conflict with other changesets, you should be fine
15:58 zanoi z3p: ok
15:58 z3p: so i guess there is no documented way of how to add stuff to the db and expose it in the ORM? or is that done automagically?
15:59 z3p zanoi: schema.py has all of the ORM stuff
16:00 willkg schema.py has the models and databaseupgrade.py allows you to do migrations from one database version to the next.
16:00 zanoi: what're you trying to do specifically?
16:01 zanoi i'm trying to get widgetstatestore to save the shuffle/repeat state per display
16:01 though I am very confused about which messages classes all have to be altered to get that to happen
16:02 willkg what do you mean by "which message classes"?
16:02 zanoi my latest conclusion is that i need to add it to the db before it can be tracked by widgetstatestore
16:02 sorry i meant just classes
16:03 willkg ok, let me walk you through it, then.
16:04 looks like there are two things for widgetstatestore: ViewState and DisplayState.  both of these are in schema.py.
16:05 they have corresponding *Info classes which get passed from the backend to the frontend.
16:05 the *Info classes are defined in messages.py.
16:06 zanoi so would i create a databaseupgrade expose it in the DisplayState class and DisplayInfo message class and access that from WidgetStateStore
16:06 willkg: ok
16:06 willkg so i think you need to add your propert(y|ies) to either ViewState or DisplayState, add a databaseupgrade migration, tweak the *Info classes and then at that point you have the data you want in the frontend.
16:07 then you need to create a message class (messages.py) and handler (messagehandler.py) to handle changing the state when the user presses the button.
16:07 zanoi wouldn't i use the DisplayInfo class for that?
16:08 willkg no.  if you look at widgetstatestore.py, you'll see that there are SaveDisplayState and SaveViewState classes.  you might be able to use one of those.
16:08 the *Info classes don't _do_ anything--they're data classes.
16:09 when you asked earlier about procedure and policies of adding new stuff to the database, i didn't realize you were talking about this.
16:09 zanoi I meant using the DisplayInfo class to send shuffle/repeat status to the backend
16:09 willkg i thought you were asking about workflow and sign-off kinds of things.
16:10 to the backend to do what?  to save it to the database?
16:10 zanoi willkg: yes
16:10 willkg take a look at the widgetstatestore code and see how it works.
16:10 because you're probably going to want to piggy-back off of that.
16:10 zanoi yup, that's what i'm currently doing
16:11 willkg ok.
16:11 because you asked about using DisplayInfo and that's a data class--not an action class.  so i thought you sounded confused.
16:11 if that's what you're doing, then, it sounds like you're doing fine.
16:11 zanoi well I am confused :)
16:12 willkg if you have questions about whether your implementation is right, make sure to get someone to peer review it before pushing it to the database.
16:12 paulproteus Morning, all.
16:12 zanoi i'll definitely do that
16:12 willkg paulproteus: hi!
16:12 zanoi paulproteus: morning
16:12 willkg zanoi: what are you still confused about?
16:14 kcw joined #miro-hackers
16:14 zanoi willkg:  i'll have a look at the instructions you just gave me and then i'll get back to you :)
16:14 willkg zanoi: as an aside, there is a widgetstate design document: http://develop.participatorycu[…]x.php/WidgetState
16:14 zanoi willkg: yes, I'm having a hard time understanding it though
16:14 willkg zanoi: talk to kcw who wrote it.
16:15 zanoi willkg: will do
16:16 willkg and feel free to keep asking questions.  if the questions are in question form, they're easier to answer.
16:16 zanoi willkg: ok, i'll try to ask questions as questions :)
16:19 z3p deardiary: figured out that NSOutlineView is lying to us; time to figure out what to do about it
16:19 willkg deardiary: working on reworking fastresume.  getting into code i've never fiddled with before.  it's like walking through a haunted mansion.
16:43 deardiary: sent out a "only 3 weeks till code deadline" email.
16:46 z3p deardiary: back to UI tix (sync screen, then connect pane)
17:14 zanoi afk for dinner
17:17 willkg afk -- baby.
17:44 mattjohnson joined #miro-hackers
18:01 zanoi back from dinner
18:14 willkg back.
18:15 baby is strapped to my front.  now he is forced to watch me fix bugs for a while.  muah ha ha ha ha ha!
18:15 as a side note, i think baby likes org-mode.  but, then again, who doesn't?
18:17 it'd be really funny if his first words were "C-x occurs 'talk'; Shift - rightarrow"
18:17 oops--that should be "M-x occurs RET talk ; Shift - rightarrow"
18:46 janetPCF1: i'm seriously thinking about adding a "see crash report" button to the crash reporter dialog.
19:01 janetPCF1: wrote up a bug for it.  16169.
19:04 glee: you on?
19:04 glee: seems like miro on osx doesn't catch exceptions in the front end right.
19:08 paulproteus ajonas: Lunch, then I will work on the pcf7 issues and finishing the paypal chain.
19:55 deardiary: pcf7 is now a little happier.
19:57 deardiary: Time to polish up the PayPal integration work from the weekend.
20:05 janetPCF1 willkg:  for the crash dialog - one really helpful thing would maybe to be nudge to people to give us their email contact.
20:05 Sometimes we get those weird crash reports, with no way to contact the person and ask what happened.
20:06 willkg janetPCF1: if by "sometimes" you mean "regularly", then i totally agree.
20:06 janetPCF1 I usually drop my initial in a crash report, to make it easier for me to find,
20:06 yes sometimes does really me everytime or similar
20:06 * willkg nods.
20:07 willkg janetPCF1: can you write that enhancement up in a new bug report?  i think it's an interesting idea and probably worth having its own field in the crash report.
20:07 janetPCF1 sure -
20:08 brb - have to chase Katerina from bath to bed and switch machines
20:08 janetPCF1 left #miro-hackers
20:08 * willkg nods.
20:14 janet_ joined #miro-hackers
20:29 zanoi so per display shuffle works nows, but it's not saving to the database
20:29 I have added it to database (schema.py & databaseupgrade)
20:30 I have added it to the DisplayInfo message class
20:30 I have added it to the DisplayState class
20:30 but it doesn't seem to get saved with the other display state stuff
20:30 anyone have an idea what I have missed?
20:32 willkg zanoi: pastebin the patch you have so far.  it's hard to know what you missed without seeing what you did.
20:33 zanoi willkg: ok, i thought maybe there was some obvious class I had missed
20:37 willkg: http://kde.pastey.net/145818
20:41 willkg zanoi: http://kde.pastey.net/145818/127  that line is puzzling.  isn't shuffle an integer?  also, you're not setting repeat here.
20:41 zanoi: seems like both self.shuffle and self.repeat should be None or some default value.
20:43 zanoi: also, you're missing repeat here: http://kde.pastey.net/145818/89:93
20:43 zanoi willkg: oops, thx that might be the mistake, it's an integer in the db
20:43 it's supposed to be a boolean, and repeat i haven't added yet except for in the db
20:45 kcw zanoi: did you update handle_save_display_state in messagehandler?
20:46 willkg zanoi: i'm not seeing anything that updates the DisplayInfo when the user presses the shuffle button.  what code does that?
20:46 zanoi kcw: ahhh, that's the one I had missed. Thx!
20:48 willkg: the shuffle button is connected to PlaybackManager.toggle_shuffle
20:49 and whenenver shuffle is updated in PlaybackManager the signal "update-shuffle" is emited
20:49 kcw zanoi: also, really minor tweak: I've been defaulting values to None in messages.DisplayInfo and then applying the default value in the WSS getter if if the value in the message is None - I think it might make it easier to tell where the value is coming from when it uses the default
20:49 zanoi and ItemlistController._handle_update_shuffle is connected to that signal
20:50 willkg ok.
20:50 zanoi kcw: ok, sure
20:52 yay, it works now
20:52 despite the db value still being int not bool.
20:52 thx guys
21:02 do i understand it correctly that what it added as SchemaBool() in schema.py is an integer in databaseupgrade.py?
21:04 kcw zanoi: yes - sqlitedb doesn't do bools, I think
21:06 zanoi kcw: ok, thanks
21:21 maggie_s joined #miro-hackers
22:08 odenthaa left #miro-hackers
22:36 willkg gah...  it's 5:30pm already?
22:36 ajonas willkg: didn't today go by crazy fast?
22:36 willkg deardiary: wrote up some bugs, added a basic "see crash report" button thingy, and worked out all the changes i think i need to make for moving fastresume storage to the downloader.
22:37 ajonas: yes.  way too fast.
22:37 maybe someone put me in some kind of montage or something without telling me.  except this would have been one of those montages where time goes by, but not a whole lot of stuff is accomplished.
22:38 this should definitely be a scene in Miro, The Movie.
22:38 ajonas willkg: there was an episode of buffy the vampire slayer where that happened
22:38 maggie_s left #miro-hackers
22:39 paulproteus deardiary: After printing out a copy of the PayPal documentation and using a super simple API wrapper, I got the "hard part" working in about twenty minutes.
22:39 deardiary: In conclusion, drink more strawberry milkshakes.
22:40 willkg paulproteus: did you make your shake?  or did you go out and buy it somewhere?
22:40 paulproteus Made it here.
22:40 Bought the strawberries at haymarket on Saturday.
22:40 It was kind of astonishing.
22:40 willkg can you send that recipe to the develop list?  i think that's a key component to the "what can we do to be faster/better developing so we hit the Miro 4.0 code freeze deadline" mission.
22:40 paulproteus I was feeling super lame and frustrated, and had a sip of the milkshake hoping it would make me feel better.
22:40 No luck. Then I had three more sips and suddenly I felt *great*.
22:42 ajonas paulproteus: that's awesome!
22:43 the milkshake, but mainly the paypal thing
22:43 paulproteus heh. (-:
22:43 ajonas night all
22:43 paulproteus Night, ajonas!
22:43 kcw code freeze deadline... or brain freeze deadline?
22:43 ajonas left #miro-hackers
22:44 willkg kcw: oo!  i wonder if brain freeze can fight the code freeze and win!
22:45 paulproteus No fighting, please! This is the War Room.
22:45 willkg i bet that's the silver bullet that Fred Brooks was searching for, but never found....
22:45 paulproteus: i _love_ that line.
22:45 * paulproteus chortles.
22:56 willkg afk a bit.

← Previous day | Index | Server Index | Channel Index | Today | Next day → | Atom Feed | Search | Google Search | Plain-Text | plain, newest first