--- Log opened Sat Jun 25 00:00:58 2011
20110625 00:03:21-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20110625 00:04:38< boucman> night all
20110625 00:04:49-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Quit: Leaving.]
20110625 00:10:26-!- Grimling [~FKint@91.181.4.187] has joined #wesnoth-dev
20110625 00:26:02-!- Crab_ [~Crab_@wesnoth/developer/crab] has quit [Quit: Leaving.]
20110625 00:30:11< Soliton> shadowmaster: what are ignored system accounts for in phpbb?
20110625 00:34:34< shadowmaster> they are either used for managing non-registered users' permissions and posts (Anonymous) or bots/crawlers' permissions
20110625 00:37:45< Soliton> doesn't sound relevant to mp so just allow them in.
20110625 00:38:56< shadowmaster> that's the new behavior now
20110625 00:42:43-!- Johannes13_ [~Johannes@unaffiliated/johannes13] has quit [Ping timeout: 246 seconds]
20110625 00:43:54< shadowmaster> theoretically all of this can be backported to 1.8 but unless the client changes to handle [warning] responses are backported too, people would not receive the um, warning about not being able to use their login credentials with an inactive account
20110625 00:44:20< shadowmaster> and that would be terrible
20110625 01:15:38-!- Grimling [~FKint@91.181.4.187] has quit [Quit: Leaving]
20110625 01:16:46< shadowmaster> loonycyborg: what kind of input do you need in regards to the GUI2 network dialog?
20110625 01:18:06< loonycyborg> shadowmaster: I'm not really a gui designer, so I'm not sure how labels etc should be placed around.
20110625 01:18:43< loonycyborg> I can't just copy old design since progressbar can't have a percentage written on it :P
20110625 01:20:54< loonycyborg> Unless stacked widget can achieve that..
20110625 01:21:12< shadowmaster> I'd personally love to have a KiB/second :p
20110625 01:21:20< shadowmaster> and yes, I think stacked_widget was created for that kind of situation
20110625 01:23:45< shadowmaster> Note that in the future I'd like to introduce the ability to display a bit more of information about what the add-ons manager is doing when downloading or updating add-ons (i.e. what the currently downloading add-on is, how many remaining)
20110625 01:24:35< shadowmaster> I imagine a separate dialog could be introduced for that as long as there isn't much code duplication in regards to the network layer interaction.
20110625 01:25:34< loonycyborg> But what *really* bothers me is those fixed size labels.
20110625 01:29:08< loonycyborg> For example, the current initial text for label that displays numeric progress is 'connecting...'
20110625 01:29:36< loonycyborg> Say, if that word get translated to something really short in another language,
20110625 01:30:35< loonycyborg> then numeric display may get truncated.
20110625 01:31:11< loonycyborg> And you'll see something like 12KiB/23K... :P
20110625 01:32:04< loonycyborg> I need to interrogate mordante on subject of what should be used instead of size_text.
20110625 01:37:30-!- Gallaecio [~quassel@92.39.broadband12.iol.cz] has quit [Read error: Connection reset by peer]
20110625 01:39:28< loonycyborg> gn
20110625 01:39:57-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20110625 01:47:49-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20110625 02:06:40-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit [Ping timeout: 258 seconds]
20110625 02:09:31-!- tschmitz_ [626cdb95@gateway/web/freenode/ip.98.108.219.149] has joined #wesnoth-dev
20110625 02:23:49-!- ancestral [~ancestral@mobile-198-228-225-146.mycingular.net] has joined #wesnoth-dev
20110625 02:32:27-!- Blueblaze [~Blueblaze@adsl-99-4-144-59.dsl.hstntx.sbcglobal.net] has quit [Quit: Blueblaze]
20110625 02:34:40-!- ancestral [~ancestral@mobile-198-228-225-146.mycingular.net] has quit [Quit: Smell ya later!]
20110625 02:45:28-!- alkenrinnstet [~alkenrinn@bb121-6-150-206.singnet.com.sg] has quit [Read error: Connection reset by peer]
20110625 02:46:45-!- alkenrinnstet [~alkenrinn@bb121-6-150-206.singnet.com.sg] has joined #wesnoth-dev
20110625 02:47:51-!- Upth [~ogmar@c-76-105-37-245.hsd1.ca.comcast.net] has joined #wesnoth-dev
20110625 02:47:51-!- Upth is now known as Upthorn
20110625 02:54:15-!- alkenrinnstet [~alkenrinn@bb121-6-150-206.singnet.com.sg] has quit [Read error: Connection reset by peer]
20110625 02:55:06-!- alkenrinnstet [~alkenrinn@bb116-14-195-134.singnet.com.sg] has joined #wesnoth-dev
20110625 02:57:12-!- alkenrinnstet [~alkenrinn@bb116-14-195-134.singnet.com.sg] has quit [Read error: Connection reset by peer]
20110625 02:57:24-!- melinath [~stephen@li341-41.members.linode.com] has quit [Remote host closed the connection]
20110625 02:57:30-!- alkenrinnstet [~alkenrinn@bb116-14-195-134.singnet.com.sg] has joined #wesnoth-dev
20110625 02:58:19-!- deekay [~dk@wesnoth/developer/dragonking] has quit [Quit: deekay]
20110625 03:01:40-!- tschmitz_ [626cdb95@gateway/web/freenode/ip.98.108.219.149] has quit [Quit: Page closed]
20110625 03:05:18-!- tschmitz_ [626cdb95@gateway/web/freenode/ip.98.108.219.149] has joined #wesnoth-dev
20110625 03:07:40-!- Vorpal [~AnMaster@unaffiliated/anmaster] has quit [Read error: Operation timed out]
20110625 03:31:03-!- tschmitz_ [626cdb95@gateway/web/freenode/ip.98.108.219.149] has quit [Quit: Page closed]
20110625 03:31:17-!- tschmitz_ [626cdb95@gateway/web/freenode/ip.98.108.219.149] has joined #wesnoth-dev
20110625 03:36:30< tschmitz_> Is there any reason why a call to network::send_data() in one client would *not* eventually result in a "corresponding" call to network::receive_data in another client?
20110625 03:36:34< tschmitz_> (connected to the same game on the same server, and assuming good network connectivity)
20110625 03:37:20< tschmitz_> I'm doing some tests and it seems that sometimes the data is received by the other client, and sometimes it is not
20110625 03:44:51-!- anonymissimus [~chatzilla@HSI-KBW-095-208-215-104.hsi5.kabel-badenwuerttemberg.de] has quit [Quit: ChatZilla 0.9.86.1 [Firefox 3.6.2/20100316074819]]
20110625 03:44:55-!- beetlenaut [~dan@174.32.63.204] has quit [Read error: Connection reset by peer]
20110625 04:26:08-!- tschmitz [626cdb95@gateway/web/freenode/ip.98.108.219.149] has quit [Quit: Page closed]
20110625 04:26:23-!- tschmitz [626cdb95@gateway/web/freenode/ip.98.108.219.149] has joined #wesnoth-dev
20110625 04:49:14-!- knotwork_ is now known as knotwork
20110625 05:20:55-!- beetlenaut [~dan@174.32.63.204] has joined #wesnoth-dev
20110625 05:23:30-!- Alarantalara [~alarantal@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has left #wesnoth-dev []
20110625 05:28:00-!- ancestral [~ancestral@17.232.7.174] has joined #wesnoth-dev
20110625 06:13:27-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 258 seconds]
20110625 06:14:07-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20110625 06:28:06-!- Cookiee [~quassel@unaffiliated/cookiee] has joined #wesnoth-dev
20110625 06:57:06< CIA-95> fendrin * r50029 /trunk/src/ (hotkeys.cpp hotkeys.hpp): Added getters for hotkeys that are based on joybutton and joyhat events.
20110625 07:11:48-!- ancestral [~ancestral@17.232.7.174] has quit [Quit: ancestral]
20110625 07:49:42-!- Cookiee [~quassel@unaffiliated/cookiee] has quit [Ping timeout: 255 seconds]
20110625 07:50:41-!- Cookiee [~quassel@unaffiliated/cookiee] has joined #wesnoth-dev
20110625 07:56:18-!- Cookiee [~quassel@unaffiliated/cookiee] has quit [Remote host closed the connection]
20110625 08:26:01< zaroth> good morning
20110625 08:37:14< tschmitz_> good evening
20110625 09:01:56-!- tschmitz_ [626cdb95@gateway/web/freenode/ip.98.108.219.149] has quit [Quit: Page closed]
20110625 09:15:28-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20110625 09:16:19-!- Gallaecio [~quassel@92.39.broadband12.iol.cz] has joined #wesnoth-dev
20110625 09:20:03-!- Blueblaze [~Blueblaze@adsl-99-4-144-59.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20110625 09:33:05-!- Ivanovic [~ivanovic@dtmd-4d0523c4.pool.mediaWays.net] has quit [Changing host]
20110625 09:33:05-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20110625 09:33:42< Ivanovic> moin
20110625 09:50:51-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20110625 09:51:42-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20110625 09:57:27-!- Gallaecio [~quassel@92.39.broadband12.iol.cz] has quit [Ping timeout: 250 seconds]
20110625 09:58:38< CIA-95> ivanovic * r50030 /trunk/ (5 files in 4 dirs): updated Vietnamese translation
20110625 09:59:23-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Quit: crimson_penguin]
20110625 10:00:21< boucman> morning all
20110625 10:00:42-!- Gallaecio [~quassel@92.39.broadband12.iol.cz] has joined #wesnoth-dev
20110625 10:02:01< zaroth> boucman: Crab_: I read up more on Boost.Statechart, will share my findings now
20110625 10:02:10< boucman> ok
20110625 10:02:46< zaroth> so, the first question is our data providing model
20110625 10:03:05< zaroth> as you may know, the whole statechart is event driven
20110625 10:03:18-!- Blueblaze [~Blueblaze@adsl-99-4-144-59.dsl.hstntx.sbcglobal.net] has quit [Quit: Blueblaze]
20110625 10:03:59< zaroth> and we may store the data necessary for transitions either in the event structures (for example struct load_game_event { string game_name, list_of_players }; etc.)
20110625 10:04:50< zaroth> or set the necessary parameters via setters and have the state verify whether it's ready to transit upon receiving an event, which will be an empty (or almost-empty) structure in this case
20110625 10:05:27< zaroth> both approaches have their advantages, I think, but I'd prefer the 2)
20110625 10:05:48< zaroth> (but of course I'm fine either way, it's just something to discuss)
20110625 10:05:56< boucman> hmm, let me think about it for a minute
20110625 10:06:09-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 264 seconds]
20110625 10:06:30< zaroth> second, if we decide to move our entire wesnoth infrastructure to a state machine
20110625 10:06:45< zaroth> I don't know if the game_controller_new is needed
20110625 10:06:53< boucman> my first idea was that 1 was cleaner, but since the info for transition can come from a lot of sources, 2 might be more handy eventually
20110625 10:06:54< zaroth> we need to discuss class structure here
20110625 10:07:21-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has joined #wesnoth-dev
20110625 10:07:21-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has quit [Changing host]
20110625 10:07:21-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20110625 10:07:34< zaroth> (that will be easier once I provide an UML sketch, which should be shortly)
20110625 10:08:11< zaroth> and responsibilities of each object
20110625 10:08:39< zaroth> it may happen that the game_controller_new will be simply the state machine object
20110625 10:08:47-!- MeccaGod [~majs@h146n3fls33o279.telia.com] has joined #wesnoth-dev
20110625 10:08:58< boucman> i'm perfectly fine with that
20110625 10:09:06< zaroth> or it can be composite of the config and mp_state machine objects 
20110625 10:09:21< zaroth> this brings another problem (when two machines are used), but in a second about that
20110625 10:10:03< zaroth> third, exception handling and load_game_exception
20110625 10:10:22< boucman> when is that one raised ?
20110625 10:10:50< zaroth> hmm... I didn't check that, but quite in a few places
20110625 10:11:07< zaroth> from what I learned up to know, it's the current equivalent of a "rich event"
20110625 10:11:08< boucman> ok, what does it mean ? incorrect config ?
20110625 10:11:24< zaroth> that is, it's not really an exception as in "error" meaning
20110625 10:11:38< zaroth> but rather load_game_event, containing lots of information about what game are we about to load
20110625 10:11:45< zaroth> (and it's the current implementation)
20110625 10:12:04< boucman> not sure I understand, that's a weird implementation
20110625 10:12:22< zaroth> let me check in code then, maybe I misunderstood that as well
20110625 10:12:25-!- hopman- [~chris@c-71-202-164-232.hsd1.ca.comcast.net] has joined #wesnoth-dev
20110625 10:12:39-!- cjhopman [~chris@wesnoth/developer/cjhopman] has quit [Ping timeout: 258 seconds]
20110625 10:13:58< zaroth> from doxygen comment:
20110625 10:14:15-!- stikonas_ [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20110625 10:14:20< zaroth> Exception used to signal that the user has decided to abort a game, and to load another game instead.
20110625 10:14:43< zaroth> members: string game, bool show_replay, bool cancel_orders
20110625 10:14:49< boucman> ok, so it's a "back to stage 2" exception for us
20110625 10:15:05< zaroth> stage 2?
20110625 10:15:27< boucman> the yellow stage
20110625 10:15:34< zaroth> ah
20110625 10:15:48< boucman> yeah, i'll try to call them by their colors :P
20110625 10:15:54< zaroth> yes, it's more like an event than like an exception
20110625 10:16:22< boucman> I guess you are right... an exception was used because there was no statechart like automaton
20110625 10:16:24< zaroth> yet it's next to such exceptions as game_error or save_game_failed
20110625 10:16:32-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 244 seconds]
20110625 10:16:39< zaroth> which are legitimate exceptions
20110625 10:17:34-!- tschmitz [626cdb95@gateway/web/freenode/ip.98.108.219.149] has quit [Quit: Page closed]
20110625 10:17:36< zaroth> well, to go further, the exception handling in Boost.Statechart is also quite fascinating
20110625 10:20:08< zaroth> when an exception goes out of local context in a function and it reaches the boost library context (so is uncaught locally), it either destroys the whole machine (if no exception_manager is specified), or generates an exception_thrown event (if exception_manager is specified) for the machine to handle - and if that event is unhandled by any outer or inner state, it only THEN destroys the whole machine
20110625 10:20:34< zaroth> when done correctly, it allows for quite nice and clean handling of errors imho
20110625 10:20:56< zaroth> but no load_game_exceptions spanning through multiple states are allowed then, sadly ;-)
20110625 10:21:08< zaroth> so, the last issue
20110625 10:21:41< zaroth> about having two separate state machines, one for game config, second for the mp
20110625 10:22:15< boucman> actually I like that idea
20110625 10:22:16< zaroth> well, to put it shortly, it's problematic to have two Boost.Statechart state machines running at the same time AND communicating with each other
20110625 10:22:33< boucman> hmm
20110625 10:22:36< zaroth> there exists this asynchronous_state_machine idea in the library
20110625 10:22:50< zaroth> but it's definitely not KISSy
20110625 10:22:54< zaroth> and it gets complicated
20110625 10:23:24< boucman> zaroth: from my memories of state-machine theories, you can do a state machine where each state is a pair<mp machine state,game config state>
20110625 10:23:37< boucman> then draw the complete diagram and then simplify
20110625 10:24:11< zaroth> so I'd say we should try and make our current graph more KISSy (group things differently and/or create more groups), therefore avoiding the need for two SMs
20110625 10:24:14< zaroth> boucman: correct
20110625 10:24:36< zaroth> that's what I mean, putting all the variables in the correct context can do wonders ;-)
20110625 10:24:50-!- MeccaGod [~majs@h146n3fls33o279.telia.com] has quit [Ping timeout: 260 seconds]
20110625 10:25:15< boucman> we are still in doc 2, are we ?
20110625 10:25:41-!- Blueblaze [~Blueblaze@adsl-99-4-144-59.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20110625 10:25:42< zaroth> so, after I reported on that, I start working on the UML diagram of the whole thing, which is gonna be way more specific and technical than what we have on google docs - unless you have questions/inquiries now
20110625 10:25:49< zaroth> boucman: yes, that's doc 2
20110625 10:26:05< boucman> just a remark, what exactly is MP game configuration ? 
20110625 10:26:24< boucman> is it difficulty selection, or is it stuff like fog of war/use map settings ?
20110625 10:26:54< zaroth> but what I'm saying is reorganizing doc 2 in a way that will allow us to group the specific colours more nicely
20110625 10:27:06< zaroth> I think that it's gonna be both
20110625 10:27:37< zaroth> although it's also something to think about, since we need to consider what information we have at which point
20110625 10:27:38< boucman> hmm
20110625 10:27:46< boucman> shouldn't it be blue then ?
20110625 10:28:13< boucman> or do we want to (eventually) expand the campaign header to be able to specify everything in the header ?
20110625 10:28:17< zaroth> maybe it's necessary to split it into two steps - configuration with only headers (difficulty, other #define-specified game parameters) and configuration with full information
20110625 10:29:02< zaroth> that's a valid question, how much info do we want in the headers
20110625 10:29:23< boucman> esp since things like FoW influence what players see in the lobby*
20110625 10:29:44-!- MeccaGod [~majs@h146n3fls33o279.telia.com] has joined #wesnoth-dev
20110625 10:30:17-!- alkenrinnstet [~alkenrinn@bb116-14-195-134.singnet.com.sg] has quit [Read error: Connection reset by peer]
20110625 10:31:11< zaroth> actually I think it would be only fair to allow players from the lobby see a little more, especially after our changes
20110625 10:31:19< zaroth> such as difficulty settings and all these
20110625 10:31:33-!- alkenrinnstet [~alkenrinn@bb116-14-195-134.singnet.com.sg] has joined #wesnoth-dev
20110625 10:31:43< zaroth> to avoid cluttering of view, it can be done in e.g. a dialog hidden behind "Game details.." button
20110625 10:32:06-!- Elvish_Pillager [~eli@71-10-224-192.dhcp.oxfr.ma.charter.com] has quit [Ping timeout: 246 seconds]
20110625 10:33:03-!- Gallaecio [~quassel@92.39.broadband12.iol.cz] has quit [Ping timeout: 255 seconds]
20110625 10:33:13-!- Blueblaze [~Blueblaze@adsl-99-4-144-59.dsl.hstntx.sbcglobal.net] has quit [Quit: Blueblaze]
20110625 10:33:22-!- crend [~crend@77-22-254-233-dynip.superkabel.de] has joined #wesnoth-dev
20110625 10:35:14-!- Blueblaze [~Blueblaze@adsl-99-4-144-59.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20110625 10:35:18-!- alkenrinnstet [~alkenrinn@bb116-14-195-134.singnet.com.sg] has quit [Read error: Connection reset by peer]
20110625 10:36:30-!- alkenrinnstet [~alkenrinn@bb220-255-178-30.singnet.com.sg] has joined #wesnoth-dev
20110625 10:37:39< boucman> and since mp player connection is already blue, MP game configuration should be a little bit of both :P
20110625 10:38:13< boucman> but I wouldn't be shocked if we went blue after map selection, it would make sense, once we have a map to immediately load it
20110625 10:38:29< boucman> the only problem is that difficulty and other defines need to be set before map loading
20110625 10:40:06-!- negusnyul [~negusnyul@145.236.67.3] has joined #wesnoth-dev
20110625 10:41:38-!- EdB [~edb@89.82.194.231] has joined #wesnoth-dev
20110625 10:42:07-!- crend is now known as Crendgrim
20110625 10:42:56-!- stikonas_ [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 258 seconds]
20110625 10:43:35< zaroth> the lobby starts bugging me now
20110625 10:43:39< zaroth> is it really a game state?
20110625 10:43:45< zaroth> hmmm...
20110625 10:45:57< boucman> no, it's an UI state, and maybe a MP connection state
20110625 10:46:12< boucman> but it's just a yellow state appart from that
20110625 10:46:34< boucman> but we started the drawing as documenting the screen states that's why it's here
20110625 10:47:32< zaroth> I started adding <--> arrows to reflect actions that can be cancelled
20110625 10:49:56-!- stikonas [~and@84.32.245.15] has joined #wesnoth-dev
20110625 10:49:56-!- stikonas [~and@84.32.245.15] has quit [Changing host]
20110625 10:49:56-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20110625 10:50:43< boucman> I think dotted arrows are better since there is a main and a secondary direction
20110625 10:51:12< zaroth> well, the main flow is from up->down anyway
20110625 10:51:46< boucman> true, but I still like dotted arrows better :)
20110625 10:52:46-!- Gallaecio [~quassel@92.39.broadband12.iol.cz] has joined #wesnoth-dev
20110625 10:55:21< Soliton> tschmitz: well, what data are you talking about?
20110625 10:56:08< zaroth> it's still not clear to me how to handle sp special case now... whether the screens should send necessary events or what
20110625 10:56:13< zaroth> hmm...
20110625 10:56:29< boucman> not sure I understand your question
20110625 10:56:57-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20110625 10:58:37< zaroth> for example
20110625 10:59:02< zaroth> in our graph we can go to mp player connection
20110625 10:59:15< zaroth> than go back (via cancel) to mp lobby
20110625 10:59:31< zaroth> which makes no sense in single player case, where we should go back to main menu instead
20110625 11:00:01< zaroth> brb
20110625 11:02:07-!- zookeeper [~l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20110625 11:03:32< boucman> well, a couple of remarks
20110625 11:04:10< boucman> in SP, both mp player connection and mp lobby are non-gui steps (i.e more or less skipped) so it makes sense to skip them too on cancel
20110625 11:04:37< boucman> second, in SP, mp plaeyr connection being non-gui you can't cancel there at all :P
20110625 11:05:22< boucman> the question makes more snse for header mp game configuration though
20110625 11:05:38< boucman> brb too
20110625 11:05:51-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 276 seconds]
20110625 11:08:40-!- stikonas [~and@84.32.245.15] has joined #wesnoth-dev
20110625 11:08:40-!- stikonas [~and@84.32.245.15] has quit [Changing host]
20110625 11:08:40-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20110625 11:13:04< boucman> back
20110625 11:16:21-!- Blueblaze [~Blueblaze@adsl-99-4-144-59.dsl.hstntx.sbcglobal.net] has quit [Quit: Blueblaze]
20110625 11:21:03< zaroth> back too
20110625 11:21:51-!- Crendgrim [~crend@77-22-254-233-dynip.superkabel.de] has quit [Remote host closed the connection]
20110625 11:22:11< zaroth> hm, the mp player connection doesn't necessarily have to be skipped in single player
20110625 11:22:22< zaroth> what if you're playing mp campaign and you have an ally
20110625 11:22:39< zaroth> in this screen you are allowed to choose whether your ally will be a local player or an AI
20110625 11:22:53< boucman> yes, indeed
20110625 11:23:23< boucman> but what I said about mp_lobby being gui-less still stands, 
20110625 11:23:25< zaroth> (there are current WML solution for that, i.e. a dialog pops up whether or not you want to control your ally, but imho it's not something that should be solved by WML)
20110625 11:23:28< zaroth> of course
20110625 11:23:40< zaroth> I have no objections to lobby ;-)
20110625 11:23:44< boucman> int's in mp lobby that we should catch the cancelation, detect we are in SP and rethrow the cancel
20110625 11:24:00< zaroth> well, all the issues here
20110625 11:24:22< zaroth> are bringing us to the point - what is a state, what is a screen and what is the relation between them
20110625 11:24:42-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 276 seconds]
20110625 11:24:43< boucman> ok, i'll expose my ideas
20110625 11:24:51< zaroth> yay :D
20110625 11:24:56 * zaroth listens
20110625 11:25:03< boucman> a state is "the color" in our diagram, i.e how much info we have  in the cfg
20110625 11:25:51< boucman> the different screen don't change the state, they simply require info in the global structure and fill in other info
20110625 11:26:15< boucman> so at any moment we have a set of possible screens, and once all info is filled we can move on to the next state
20110625 11:26:56< boucman> cancel is just a screen that doesn't fill the info it's supposed to, thus returning us in the state we were previously
20110625 11:27:11< zaroth> who calls the screens to existence?
20110625 11:28:48< boucman> hmm
20110625 11:29:10< boucman> my first idea is that game_instance calls whetever can be called, but we need to make sure there are no choice
20110625 11:29:48-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has joined #wesnoth-dev
20110625 11:29:48-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has quit [Changing host]
20110625 11:29:48-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20110625 11:30:42< boucman> let me try something
20110625 11:32:54-!- Crendgrim [~crend@77-22-254-233-dynip.superkabel.de] has joined #wesnoth-dev
20110625 11:33:18< zaroth> ok
20110625 11:34:09< boucman> i'm doing a sort of dual graph, where the states are "what do we know of the config" and the arrows are thus (more or less) the screens that can be displayed
20110625 11:35:07< boucman> what do you think ?
20110625 11:35:44< zaroth> Ivanovic: there are no more releases planned for 1.8, right? I'm thinking whether I can safely assing https://gna.org/bugs/index.php?18269 to myself without messing up someone's plans in case he wants to work on it for 1.8
20110625 11:35:55< zaroth> s/assing/assign/
20110625 11:36:13< Ivanovic> zaroth: yeah, there are no new 1.8 releases planned unless there is some serious security issue found
20110625 11:37:19< shadowmaster> But the labels in the editor...
20110625 11:38:03< zaroth> shadowmaster: console yourself by thinking that users working on maps are most likely to use 1.9 anyway ;-)
20110625 11:39:37-!- loonybot [~loonybot@46.138.87.135] has joined #wesnoth-dev
20110625 11:39:37-!- loonybot [~loonybot@46.138.87.135] has quit [Changing host]
20110625 11:39:37-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20110625 11:42:15-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 276 seconds]
20110625 11:43:21-!- deekay [~dk@89-79-35-236.dynamic.chello.pl] has joined #wesnoth-dev
20110625 11:43:21-!- deekay [~dk@89-79-35-236.dynamic.chello.pl] has quit [Changing host]
20110625 11:43:21-!- deekay [~dk@wesnoth/developer/dragonking] has joined #wesnoth-dev
20110625 11:44:08-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has joined #wesnoth-dev
20110625 11:44:08-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has quit [Changing host]
20110625 11:44:08-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20110625 11:49:59-!- Blueblaze [~Blueblaze@adsl-99-4-144-59.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20110625 11:50:32< boucman> zaroth: so, what do you think of the start of my graph ?
20110625 11:50:54< zaroth> boucman: is it on google docs? I haven't looked yet, was playing with boost.statechart examples
20110625 11:51:14< boucman> it's on the same document, I've added it at the botom
20110625 11:51:45-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 264 seconds]
20110625 11:54:19-!- mjs-de [~mjs-de@f053191219.adsl.alicedsl.de] has joined #wesnoth-dev
20110625 11:54:57< zaroth> yes, now when we keep screens out of it, it makes more sense
20110625 11:55:30< zaroth> I'm still a bit confused in matter of where are the screens created
20110625 11:56:03< zaroth> states names are more like states now ;-)
20110625 11:56:45< boucman> yes, let's work a bit more on those states, the windows will come naturally after that
20110625 11:56:46< zaroth> anyway, I'm happy that we don't have to change anything inside the orange rectangle
20110625 11:56:56< boucman> hehe
20110625 11:57:03-!- Blueblaze [~Blueblaze@adsl-99-4-144-59.dsl.hstntx.sbcglobal.net] has quit [Quit: Blueblaze]
20110625 11:57:05< boucman> I havn't really looked into that yet :P
20110625 11:57:11< zaroth> the need for changes has spread across the project enough as it is
20110625 11:57:30< zaroth> no! we're changing what happens before the game starts, not the game itself, remember ? :P
20110625 11:57:38< zaroth> play_game can stay ;-)
20110625 11:57:54< boucman> :P
20110625 11:58:09< zaroth> I think some of the parts of our state machine are already implemented in multiplayer.cpp
20110625 11:58:20< boucman> yes, but we will have to check that
20110625 11:58:25< zaroth> not really explicit, but screens do are separated from logic
20110625 11:58:54< zaroth> hmm... "do are"? any native English speakers to correct me, or is it right?
20110625 11:59:11< boucman> the thing is the screens are sometime in the states, sometime in the transitions
20110625 11:59:40< boucman> zaroth: that's not correct, remove the "do" and you are ok
20110625 12:00:08< zaroth> but if I want to put emphasis? like in "I do want this car!"
20110625 12:00:45< zaroth> "I do be free today"? "I do am free today"? what's the equivalent of the previous sentence with the "to be" verb?
20110625 12:00:58< zaroth> well, nevermind the grammar ;-)
20110625 12:01:00< zaroth> Espreon: ^
20110625 12:01:15< boucman> zaroth: the do as emphasis sometime works, sometime it doesn't, and i'm not sure what the logic is
20110625 12:01:36< boucman> my guess is that it's ok when it's followed by an infinitive, but i'm not sure
20110625 12:02:19< zaroth> boucman: if you have more ideas for the state diagram, do draw them, I will sketch the UML and play with Boost.Statechart for now a bit more
20110625 12:02:38-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has joined #wesnoth-dev
20110625 12:02:38-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has quit [Changing host]
20110625 12:02:38-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20110625 12:03:08< zaroth> I'll try to create a working example without any current wesnoth code inside on Github, just like the previous one I showed you yesterday
20110625 12:03:41< boucman> not really, but it's interesting to note that the two arrows in "allplayers assigned" are the same screen, and that the dotted arrow correctly displays what needs to be done in case of cancel
20110625 12:03:44< zaroth> when it's working the way we want (and the way it's presented on UML/GDocs) I will copy it to wesnoth trunk and start putting wesnoth code inside
20110625 12:03:54< boucman> ok
20110625 12:19:39< zaroth> Crab_: boucman: until the state machine code is ripe enough to get into Wesnoth SVN, I will develop it here: https://github.com/zaroth/wesnoth-statechart
20110625 12:20:23< zaroth> I will try to keep the doc/wesnoth.uml file up to date with current code (that's not hard since I usually first change UML and then change the code) ;-)
20110625 12:21:43< boucman> zaroth: ok, 
20110625 12:22:34< zaroth> brb in ~40 mins - I need some fresh air after code morning :-)
20110625 12:25:15-!- MeccaGod [~majs@h146n3fls33o279.telia.com] has quit []
20110625 12:31:41-!- alkenrinnstet [~alkenrinn@bb220-255-178-30.singnet.com.sg] has quit [Quit: Leaving.]
20110625 12:36:58-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20110625 12:37:18-!- Crendgrim [~crend@77-22-254-233-dynip.superkabel.de] has quit [Remote host closed the connection]
20110625 12:39:38-!- vultraz [~chatzilla@124.109.10.222] has joined #wesnoth-dev
20110625 12:42:16-!- Crendgrim [~crend@77-22-254-233-dynip.superkabel.de] has joined #wesnoth-dev
20110625 12:46:37-!- Gallifax [~IceChat77@p5DC675FD.dip.t-dialin.net] has joined #wesnoth-dev
20110625 12:51:21-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 276 seconds]
20110625 12:53:56-!- Nephro [~Dmitry@80.233.231.12] has joined #wesnoth-dev
20110625 12:56:40-!- Johannes13_ [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20110625 12:58:38-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20110625 13:11:54-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit [Ping timeout: 255 seconds]
20110625 13:28:08-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 252 seconds]
20110625 13:29:23-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has joined #wesnoth-dev
20110625 13:29:24-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has quit [Changing host]
20110625 13:29:24-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20110625 13:39:04-!- Alarantalara [~alarantal@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has joined #wesnoth-dev
20110625 13:42:05-!- Crendgrim [~crend@77-22-254-233-dynip.superkabel.de] has quit [Remote host closed the connection]
20110625 13:52:53-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Read error: Operation timed out]
20110625 13:56:23-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20110625 13:56:48-!- timotei [~timotei@193.34.190.70] has joined #wesnoth-dev
20110625 13:56:48-!- timotei [~timotei@193.34.190.70] has quit [Changing host]
20110625 13:56:48-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20110625 14:00:22-!- Nephro [~Dmitry@80.233.231.12] has quit [Remote host closed the connection]
20110625 14:09:22-!- knotwork [~markm@unaffiliated/knotwork] has quit [Read error: Connection reset by peer]
20110625 14:20:03-!- hhyloc [~hhyloc@113.166.144.238] has joined #wesnoth-dev
20110625 14:21:41-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 240 seconds]
20110625 14:25:19-!- MeccaGod [~majs@h146n3fls33o279.telia.com] has joined #wesnoth-dev
20110625 14:27:36-!- stikonas [~gentoo@ctv-79-132-175-26.vinita.lt] has joined #wesnoth-dev
20110625 14:27:36-!- stikonas [~gentoo@ctv-79-132-175-26.vinita.lt] has quit [Changing host]
20110625 14:27:36-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20110625 14:36:41-!- zookeeper [~l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20110625 14:51:20-!- ancestral [~ancestral@17.232.7.174] has joined #wesnoth-dev
20110625 14:52:43-!- alkenrinnstet [~alkenrinn@bb220-255-178-30.singnet.com.sg] has joined #wesnoth-dev
20110625 14:55:59-!- timotei21 [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20110625 14:59:22-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 240 seconds]
20110625 15:04:35-!- knotwork [~markm@142.177.234.11] has joined #wesnoth-dev
20110625 15:04:40-!- knotwork [~markm@142.177.234.11] has quit [Changing host]
20110625 15:04:40-!- knotwork [~markm@unaffiliated/knotwork] has joined #wesnoth-dev
20110625 15:14:21-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 258 seconds]
20110625 15:18:39-!- Vorpal [~AnMaster@unaffiliated/anmaster] has joined #wesnoth-dev
20110625 15:19:27< boucman> Espreon: around ?
20110625 15:41:30-!- Appleman1234 [~Appleman1@CPE-60-226-179-130.qld.bigpond.net.au] has quit [Remote host closed the connection]
20110625 15:59:18-!- yenloc [~hhyloc@74.115.6.22] has joined #wesnoth-dev
20110625 16:01:35-!- hhyloc [~hhyloc@113.166.144.238] has quit [Ping timeout: 252 seconds]
20110625 16:01:45-!- ancestral [~ancestral@17.232.7.174] has quit [Quit: ancestral]
20110625 16:02:51-!- hyenloc [~hhyloc@113.166.144.238] has joined #wesnoth-dev
20110625 16:06:01-!- yenloc [~hhyloc@74.115.6.22] has quit [Ping timeout: 240 seconds]
20110625 16:10:51-!- timotei21 [~timotei@wesnoth/developer/timotei] has quit [Read error: Connection reset by peer]
20110625 16:11:05-!- MeccaGod [~majs@h146n3fls33o279.telia.com] has quit [Ping timeout: 258 seconds]
20110625 16:11:40-!- timotei [~timotei@193.34.190.70] has joined #wesnoth-dev
20110625 16:11:40-!- timotei [~timotei@193.34.190.70] has quit [Changing host]
20110625 16:11:40-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20110625 16:13:24-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Read error: Connection reset by peer]
20110625 16:13:53-!- timotei [~timotei@193.34.190.70] has joined #wesnoth-dev
20110625 16:13:53-!- timotei [~timotei@193.34.190.70] has quit [Changing host]
20110625 16:13:53-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20110625 16:15:11-!- MeccaGod [~majs@h146n3fls33o279.telia.com] has joined #wesnoth-dev
20110625 16:15:17-!- StandYourGround [~Adium@dialup-4.239.165.87.Dial1.Philadelphia1.Level3.net] has joined #wesnoth-dev
20110625 16:17:48-!- StandYourGround [~Adium@dialup-4.239.165.87.Dial1.Philadelphia1.Level3.net] has quit [Client Quit]
20110625 16:18:29< CIA-95> timotei * r50031 /trunk/utils/ (767 files in 180 dirs): eclipse plugin: rename 'utils/java' to 'utils/umc_dev'
20110625 16:19:35-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Read error: Connection reset by peer]
20110625 16:20:28-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20110625 16:20:51-!- Cookiee [~quassel@unaffiliated/cookiee] has joined #wesnoth-dev
20110625 16:21:51-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Read error: Connection reset by peer]
20110625 16:21:52< boucman> -+
20110625 16:25:27< CIA-95> timotei * r50032 /trunk/utils/java/: - delete empty directory
20110625 16:25:30-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Read error: Operation timed out]
20110625 16:25:39-!- timotei [~timotei@193.34.190.70] has joined #wesnoth-dev
20110625 16:25:39-!- timotei [~timotei@193.34.190.70] has quit [Changing host]
20110625 16:25:39-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20110625 16:27:50< zaroth> that was huge...
20110625 16:28:47-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has joined #wesnoth-dev
20110625 16:28:47-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has quit [Changing host]
20110625 16:28:47-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20110625 16:43:28-!- melinath [~melinath@li341-41.members.linode.com] has joined #wesnoth-dev
20110625 16:50:30-!- melinath [~melinath@li341-41.members.linode.com] has quit [Quit: leaving]
20110625 16:51:52-!- melinath [~melinath@li341-41.members.linode.com] has joined #wesnoth-dev
20110625 17:05:23-!- anonymissimus [~chatzilla@HSI-KBW-095-208-215-104.hsi5.kabel-badenwuerttemberg.de] has joined #wesnoth-dev
20110625 17:05:29< zaroth> what the heck, argouml? why do you need to stretch so much?
20110625 17:15:56-!- Grimling [~FKint@91.181.4.187] has joined #wesnoth-dev
20110625 17:20:31-!- hyenloc [~hhyloc@113.166.144.238] has quit [Quit: Leaving]
20110625 17:24:32< fendrin> wesbot: seen mordante?
20110625 17:24:32< wesbot> fendrin: The person with the nick mordante last spoke 2d 19h ago.  2d 19h ago was here and on the channel #wesnoth-de with the message: Quit: Leaving
20110625 17:27:57-!- zaroth [~zaroth@wesnoth/developer/zaroth] has quit [Ping timeout: 244 seconds]
20110625 17:33:00-!- Crendgrim [~crend@77-22-254-233-dynip.superkabel.de] has joined #wesnoth-dev
20110625 17:44:20-!- tschmitz [626cdb95@gateway/web/freenode/ip.98.108.219.149] has joined #wesnoth-dev
20110625 18:06:42-!- MeccaGod [~majs@h146n3fls33o279.telia.com] has quit [Ping timeout: 240 seconds]
20110625 18:07:03-!- zaroth [~zaroth@wesnoth/developer/zaroth] has joined #wesnoth-dev
20110625 18:10:51-!- MeccaGod [~majs@h146n3fls33o279.telia.com] has joined #wesnoth-dev
20110625 18:17:50< CIA-95> fendrin * r50033 /trunk/src/gui/auxiliary/event/handler.cpp: Added handling of the joystick events to the gui2 event handler.
20110625 18:18:14< fendrin> mordante: ^
20110625 18:18:55< fendrin> shadowmaster: The last commit should silence your unhandled event output.
20110625 18:21:44< CIA-95> anonymissimus * r50034 /trunk/src/unit_types.cpp: 
20110625 18:21:44< CIA-95> based decision whether a unit type has random traits on whether all of them have [trait]availability=musthave
20110625 18:21:44< CIA-95> (fix for a bug report in the forum http://forums.wesnoth.org/viewtopic.php?f=12&t=34232)
20110625 18:25:35< zaroth> anonymissimus: that was quick :-)
20110625 18:27:28< anonymissimus> timotei: when you do renames, pls make sure you don't just change the case of something
20110625 18:27:41< anonymissimus> that is, such as WML to wml
20110625 18:28:06< anonymissimus> instead, make two revisions, one for WML to WML_ and then WML_ to wml
20110625 18:28:48< zaroth> what's the reasoning for this?
20110625 18:28:55< anonymissimus> IIRC there was an issue with this; tortoise svn wanrs for this, windows cant correctly handle upper/lowercase changes since it's case insensitive
20110625 18:29:25< zaroth> ah, windows
20110625 18:30:11< anonymissimus> (actually, thats good to know for all Linux-only devs; just another of the many ways you can break windows :P)
20110625 18:31:57< anonymissimus> zaroth: what was quick ?
20110625 18:32:05< zaroth> fix for the forum topic
20110625 18:32:11< fendrin> anonymissimus: Windows is broken by default ;-P
20110625 18:32:14< zaroth> that is, the ghoul undoing recruit
20110625 18:32:41< anonymissimus> fendrin: yep almost :|
20110625 18:32:53< anonymissimus> although it got getter I thin
20110625 18:33:23< fendrin> Yes, I must admit that windows7 isn't that bad anymore.
20110625 18:34:02< anonymissimus> I'm always wondering what's the reason that the vast majority of devs is Linux-only ? for very first beginner devs it's probably that wesnoth is easier to compile on Linux
20110625 18:34:44< zaroth> anonymissimus: Wesnoth devs? I suppose that's because Wesnoth is one of better games available for linux ;-)
20110625 18:34:51< fendrin> Linux users get more likely being known to the open source idea.
20110625 18:34:51< anonymissimus> there is however nothing on Linux which is so good as the MSVC debugger and even experienced devs at stackoverflow admit that
20110625 18:34:59< timotei> anonymissimus: I don't change case. Why?
20110625 18:35:09< zaroth> also, I assume that linux devs are more likely to program open-source
20110625 18:35:09< CIA-95> timotei * r50035 /trunk/utils/umc_dev/ (14 files in 4 dirs): eclipse plugin: Move readme to a better place
20110625 18:35:19< timotei> anonymissimus: oh. I'm imune to the casing now
20110625 18:35:26< timotei> since Espreon crashed my repo too many times:P
20110625 18:35:31< timotei> and I've found an option to ignore that!
20110625 18:35:36< CIA-95> timotei * r50036 /trunk/utils/umc_dev/ (LICENSE org.wesnoth/LICENSE): eclipse plugin: Move the license aswell
20110625 18:35:59< CIA-95> timotei * r50037 /trunk/utils/umc_dev/Makefile: eclipse plugin: Update the makefile
20110625 18:36:14< CIA-95> timotei * r50038 /trunk/utils/umc_dev/manual/ (dev_manual.tex user_manual.tex): 
20110625 18:36:14< CIA-95> eclipse plugin: Strip content and leave it
20110625 18:36:14< CIA-95> in the correct section
20110625 18:36:31< timotei> anonymissimus: http://stackoverflow.com/questions/52950/how-to-make-git-ignore-changes-in-case
20110625 18:37:02< loonycyborg> anonymissimus: If you say that MSVC debugger is a good reason to work on windows then you only help us to make our point :P
20110625 18:37:19< loonycyborg> Since, you know, it's not cross-platform.
20110625 18:37:28< timotei> fendrin: no it's not broken ;)
20110625 18:37:55< anonymissimus> loonycyborg: well, of course not
20110625 18:38:02< timotei> zaroth: you can program open-source with msvc aswell
20110625 18:38:17< zaroth> timotei: I never said you can't
20110625 18:38:25< anonymissimus> however, I also can't get gdb to work on win so far so it's also nto cross-platform
20110625 18:38:28< zaroth> just "more likely"
20110625 18:38:33< timotei> oh. ok :P
20110625 18:38:35< fendrin> Hmmm, somehow I made the minimap to response less good.
20110625 18:39:19< fendrin> Pumped too much into the play_slice method maybe.
20110625 18:39:43< loonycyborg> anonymissimus: There's really no reason to even try to lock people into a particular os other than idiocy.
20110625 18:40:44< zaroth> loonycyborg: are you doubting MS business model? because you know, it's working pretty well, people being locked onto many windows only apps and such ;-)
20110625 18:41:13< anonymissimus> loonycyborg: well, I assume that there must be certain technical reasons why most wesnoth devs ar working on Linux; and I am just not experienced enough to realize these reasons
20110625 18:41:15< fendrin> :-)
20110625 18:41:37< loonycyborg> I once was windows-only too.
20110625 18:41:47< loonycyborg> But that was long time ago.
20110625 18:42:01< fendrin> I am glad we have the windows devs here around.
20110625 18:43:06< fendrin> So please, stay at your Windows if it tastes you. Supporting mac is hard as well since we don't have much devs there.
20110625 18:43:25< boucman> zaroth: i'm still around, if you need me, btw
20110625 18:43:41< loonycyborg> I support windows through cross-compilation and multi-booting to it occasionally.
20110625 18:44:09< zaroth> boucman: good, I'm getting a bit frustrated with ArgoUML bugs... but apart from these display glitches, it's a good tool
20110625 18:44:13< fendrin> loonycyborg: Can one test wesnoth in wine?
20110625 18:44:23< loonycyborg> Yes.
20110625 18:45:11< fendrin> loonycyborg: That sounds like an option. Does the cross-compiling need much setup?
20110625 18:45:23< zaroth> I once read that firefox-win+wine works faster than firefox linux version, even ;-)
20110625 18:45:29< zaroth> probably not the case with wesnoth, though
20110625 18:45:46< zaroth> since it's not like wesnoth devs are doing much windows-geared optimizations
20110625 18:45:46< loonycyborg> fendrin: Yes. It does.
20110625 18:45:56< timotei> zaroth: wesnoth works very good in my ubuntu Virtual machine
20110625 18:46:55-!- anonymissimus_ [~chatzilla@HSI-KBW-095-208-215-104.hsi5.kabel-badenwuerttemberg.de] has joined #wesnoth-dev
20110625 18:48:48< timotei> Oh joy, this holiday will rock
20110625 18:49:11< timotei> 'twill be the first holiday I study maths by myself, unobligated
20110625 18:49:22-!- anonymissimus [~chatzilla@HSI-KBW-095-208-215-104.hsi5.kabel-badenwuerttemberg.de] has quit [Ping timeout: 240 seconds]
20110625 18:49:31< tschmitz> timotei: What for?
20110625 18:49:37-!- anonymissimus_ is now known as anonymissimus
20110625 18:49:42< timotei> anonymissimus, fendrin: do you have any ideas what you'd want to see in the developer manual of the eclipse plugin?
20110625 18:49:58< timotei> besides: setting up the environment, and building the zips with the standalone version?
20110625 18:50:22< timotei> tschmitz: well, I want to get deeper into algorithmics and I'll need maths for my Thesis
20110625 18:50:33< timotei> going to do a maths/physics intensive one :P
20110625 18:50:54< tschmitz> timotei: Oh goodie
20110625 18:51:01< tschmitz> timotei: which Thesis?
20110625 18:51:26< timotei> tschmitz: well, for finishing the bachelor degree. Don't you have such big project to do for the final of 4th year?
20110625 18:52:14-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20110625 18:52:20< tschmitz> timotei: Yeah, for my math major I do, although for my computer science major I can just test out of that requirement
20110625 18:52:32< fendrin> timotei: Let's see. The developer manual is for a person that wants to implement new features or fix bugs in the plugin?
20110625 18:52:44< timotei> fendrin: yeah. using the plugin's code
20110625 18:53:03< timotei> fendrin: the 'using the plugin' part will be in the user's manual
20110625 18:54:03< fendrin> timotei: Give also some overview of the architecture. Which eclipse features are used and how this is done, name the code that implements it.
20110625 18:54:51< timotei> k
20110625 18:55:00< zaroth> timotei: consider doing a graph showing all the components and how they interact with each other
20110625 18:55:07< fendrin> timotei: Also document the interfaces. How does the plugin communicate with the engine? How does it use wmltools.
20110625 18:55:09< zaroth> it's really handy for orientation :-)
20110625 18:58:06-!- Grimling [~FKint@91.181.4.187] has quit [Quit: Leaving]
20110625 19:02:09-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20110625 19:13:20-!- EdB [~edb@89.82.194.231] has quit [Remote host closed the connection]
20110625 19:25:44-!- alkenrinnstet [~alkenrinn@bb220-255-178-30.singnet.com.sg] has quit [Quit: Leaving.]
20110625 19:29:32-!- sheraff [~Tri@c-98-196-106-93.hsd1.tx.comcast.net] has joined #wesnoth-dev
20110625 19:38:50-!- Exasperation [4a47319b@gateway/web/freenode/ip.74.71.49.155] has joined #wesnoth-dev
20110625 19:41:36-!- Alarantalara [~alarantal@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has quit [Remote host closed the connection]
20110625 19:44:08< Exasperation> anonymissimus: I got things working for sliders/text_boxes
20110625 19:44:49-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20110625 19:45:03< anonymissimus> Exasperation: ok goodl, submit a patch then ? I'll have a look at it, abd mordante should probably too
20110625 19:46:14< anonymissimus> shadowmaster: I think you can close your help parse bug as wont fix or invalid; for me iit was a vertain addon that caused it too
20110625 19:46:45< Exasperation> You think I should I look at any other widgets first?  Only obvious one I see is progress_bar.
20110625 19:46:55< anonymissimus> certain
20110625 19:48:40< shadowmaster> anonymissimus: it is not *mine* :|
20110625 19:51:44< shadowmaster> anyway Invalid should be more appropriate than Won't Fix for a problem that didn't exist in the first place
20110625 20:31:02-!- Cookiee [~quassel@unaffiliated/cookiee] has quit [Remote host closed the connection]
20110625 21:02:38< Exasperation> And there it is
20110625 21:02:44-!- Crab_1 [~Crab_@62.80.190.34] has joined #wesnoth-dev
20110625 21:03:04< anonymissimus> timotei: another question: when you write eclipse plugin: in your commit messages - do you have some template for this, or do you copy-paste it, or do you write it manually ?
20110625 21:03:30< timotei> anonymissimus: template.http://book.git-scm.com/5_customizing_git.html
20110625 21:03:32< timotei> :P
20110625 21:03:35-!- Crab_1 [~Crab_@62.80.190.34] has quit [Changing host]
20110625 21:03:35-!- Crab_1 [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20110625 21:07:11< anonymissimus> timotei: and what is in the /etc/git-commit-template file
20110625 21:07:22< timotei> anonymissimus: a file. I have it in my home dir
20110625 21:07:48< timotei> anonymissimus: I have written in it: "eclipse plugin: " 
20110625 21:07:49< timotei> :P
20110625 21:08:06< anonymissimus> and how do you use it ? tab in git bash ?
20110625 21:08:23< timotei> tab? no
20110625 21:08:27< timotei> you just git commit
20110625 21:08:35< timotei> and that file will be already written in the commit messag
20110625 21:08:42< timotei> s/messag/message
20110625 21:09:04< anonymissimus> and when you make a commit which isnt related to the eclipse plugin ?
20110625 21:09:21< timotei> I just delete the contents "eclipse plugin: "
20110625 21:09:27< timotei> (dd in vim :P)
20110625 21:11:02< anonymissimus> don't you get tired of deleting/reinserting eclipse plugin: then ? :P
20110625 21:11:16< timotei> anonymissimus: well, what for? By default is eclipse plugin
20110625 21:11:24< timotei> and in exceptional cases I just dd it and that's all :)
20110625 21:12:00< timotei> when I said: "I just delete the contents "eclipse plugin: " I meant, that in the commit message window that opens, I delete the "eclipse plugin"
20110625 21:12:04< timotei> I don't touch the template file
20110625 21:12:16< timotei> just delete that commit's message to not contain eclipse plugin
20110625 21:12:20< anonymissimus> aha ok thx
20110625 21:12:35< timotei> bitte
20110625 21:13:05< anonymissimus> hopefully that file can be set project specific, since i need several such prefixes
20110625 21:13:13< anonymissimus> for each
20110625 21:13:17< timotei> anonymissimus: yes.
20110625 21:13:22< timotei> you can git config it locally or globale
20110625 21:15:07< anonymissimus> ah, timotei have you read about the codeblocks lib package ? just follow the guide in projectfiles/CodeBlocks, no need to compile any dependencies
20110625 21:15:22< anonymissimus> or to get them somehow
20110625 21:15:27< timotei> anonymissimus: today I was forced to use CodeBLocks for the ACM national phase
20110625 21:15:31< timotei> and I hate it more than ever
20110625 21:15:32< timotei> :P
20110625 21:16:09< anonymissimus> what you dont like about it, if yuo like MSVC you should liek codeblocks too
20110625 21:16:40< zaroth> Crab_1: hello, could you catch up on morning discussion and post comments when you have time ? :-)
20110625 21:18:43< Crab_1> hi
20110625 21:18:50< Crab_1> i'll take a look
20110625 21:19:55-!- Crab_1 is now known as Crab____
20110625 21:19:56-!- Crab____ is now known as Crab_1
20110625 21:23:47-!- Crab_1 [~Crab_@wesnoth/developer/crab] has quit [Quit: Leaving.]
20110625 21:23:55-!- Crab_1 [~Crab_@62.80.190.34] has joined #wesnoth-dev
20110625 21:24:14-!- Crab_1 [~Crab_@62.80.190.34] has quit [Changing host]
20110625 21:24:14-!- Crab_1 [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20110625 21:24:18-!- Crab_1 is now known as Crab_
20110625 21:24:34< Crab_> zaroth: ok, I've read
20110625 21:25:26< Crab_> zaroth: I was thinking of making some dummy lua-like code to handle the state machine operations, to see how it'll look like
20110625 21:32:12< Crab_> basically, the role of the script is to receive notifications like 'entered state X' and to set, if necessary, the properties of the state X
20110625 21:32:45< Crab_> the script is optional, and should be set if any degree of automation/state transfer is required
20110625 21:35:22-!- mjs-de [~mjs-de@f053191219.adsl.alicedsl.de] has quit [Remote host closed the connection]
20110625 21:36:30< Crab_> for example, if we want to play, we need to '(1) connect to mp server, (2) select map and difficulty, (3) set map settings, (4) assign players, (5) play game'
20110625 21:36:45< Crab_> if we want to play a SP campaign, we want to skip some of this
20110625 21:37:34< zaroth> so the code would pass a script?
20110625 21:37:40< zaroth> or run it ?
20110625 21:38:14< Crab_> that will mean 'add a script which will, on entering 'connect to mp server' state say that we don't need a mp server, on entering 'select map and diffuculty' tell that we need to use the 'select campaign' dialog instead of 'select mp map' dialog, and on entering 'assign players' it would use the predefined for-sp autoassigner.
20110625 21:39:27< Crab_> the script would be function with a state, so it would be able to remember it's settings, if necessary.
20110625 21:41:45< Crab_> of course, we can pass state as a set of variables
20110625 21:41:51-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Quit: SIGKILL]
20110625 21:42:38< Crab_> like, by attaching a config object. but it would not allow stuff like loops or conditionals in the test control flow
20110625 21:43:22< Crab_> the function, is, basically, for answerring the question 'we are now at state X. do we want to do anything special before normal handling of the state starts?'
20110625 21:44:15< Crab_> we only need to solve the question of nesting such scripts properly
20110625 21:44:38< Crab_> i.e., if wesnoth is launched with the script 'enter help mode three times', but we organized the help mode to be a script 'start a dummy help game'
20110625 21:45:30< zaroth> well, there can be switches in the events triggering transitions
20110625 21:45:44< Crab_> yes, the c++ code would have switches to trigger transitions
20110625 21:45:54< zaroth> like ev_start_game { bool interactive; }
20110625 21:45:58< Crab_> but those have to be conditional switches
20110625 21:46:28-!- Elvish_Pillager [~eli@71-10-224-192.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20110625 21:46:43< Crab_> I think that the simplest way to nest is to keep a stack of such scripts, and let the inner one run to completion before restarting the outer one.
20110625 21:47:52< Crab_> i.e, "if wesnoth is launched with the script 'enter help mode three times', but we organized the help mode to be a script 'start a dummy help game'", we'll have a outer script of "for (i = 0; i < 3; ++i ) { in main_menu: launch help ; in help : quit to main menu ; } "
20110625 21:48:12< Crab_> and a inner script to launch help which will move us from 'main menu' to 'help'
20110625 21:48:55-!- MeccaGod [~majs@h146n3fls33o279.telia.com] has quit [Ping timeout: 240 seconds]
20110625 21:50:38< Crab_> zaroth: so, what do you think about the concept of 'we are now at state X. do we want to do anything special before normal handling of the state starts?' script ?
20110625 21:51:13< Crab_> where 'normal handling' would be in c++ and would use a config object for parameters.
20110625 21:51:36< Crab_> and the script would be able to modify this config object, if so desired, or abort itself if something unexpected happens.
20110625 21:51:52< zaroth> I still don't understand fully where such scripts would be defined, kept and called...
20110625 21:52:37 * boucman doesn't even understand the point of the script
20110625 21:52:39< boucman> :P
20110625 21:52:57-!- MeccaGod [~majs@h146n3fls33o279.telia.com] has joined #wesnoth-dev
20110625 21:53:41< zaroth> I believe the scripts are intended to make the "automated testing of campaigns" idea easy to implement
20110625 21:53:44< Crab_> boucman: if we want to get from state A to state E through states A,B,C,D,E, we might need to skip some states or pick some parameters for other states without user interaction. 
20110625 21:54:13< Crab_> boucman: and only the caller (in A) knows fully what we want to do.
20110625 21:54:27< Crab_> boucman: i.e., if we are in main menu, and want to start an SP campaign
20110625 21:54:29< boucman> yes, I guess so, but that would be part of the info we pass to the intermediate state, wouldn't it ?
20110625 21:55:33< Crab_> boucman: there are two ways to do that, (1) pass enough info downstream to B,C,D,E to let them sort things out;  (2) remain in control while we pass through B,C,D,E and tell what to do. since we are ruled by the state machine, the only way to 'remain in control' is to use callback like 'on entering state B', 'on entering state C', ....
20110625 21:55:44< Crab_> boucman:  'that would be part of the info we pass to the intermediate state, wouldn't it ' is way (1)
20110625 21:56:32< Crab_> it's simpler, but it restricts us somewhat; for example if we want to do 'foreach campaign, start it then immediately exit'
20110625 21:57:11< boucman>  ok, I hadn't thought of way 2
20110625 21:57:14< Crab_> start campaign X then immediately exit' is doable by (1), by passing enough parameters to enter play game and tell that state that we want to go to main menu immediately, skipping the storyscreens.
20110625 21:57:39< boucman> i see, and you want to do these callbacks with lua ?
20110625 21:57:42< zaroth> Crab_: I kind of all the time had in mind idea of an external entity pulling the strings
20110625 21:58:06< zaroth> but it probably doesn't make sense
20110625 21:58:27< CIA-95> timotei * r50039 /trunk/utils/umc_dev/org.wesnoth/build.properties: eclipse plugin: Remove an invalid file from build.properties
20110625 21:58:37< Crab_> but once we will want to do funny things like conditionals and foreach loops, or interact with the state in state-specific way (i.e. 'on entering the MP server, check if any of my friends are there trying to make a game, if yes, join it' ), scripting language starts to rule.
20110625 21:58:38< CIA-95> timotei * r50040 /trunk/utils/umc_dev/manual/ (dev_manual.tex launch_plugin.png): 
20110625 21:58:38< CIA-95> eclipse plugin: Update the dev manual to reflect
20110625 21:58:38< CIA-95> current state
20110625 21:58:47< CIA-95> timotei * r50041 /trunk/utils/umc_dev/manual/ (install_new_software.png user_manual.tex): eclipse plugin: Update the user's manual to reflect the current state
20110625 21:59:07< Crab_> zaroth: external utility ?
20110625 21:59:12< zaroth> also, how are the events sent to the state machine? is the reference to the state machine global (e.g. in resources) and relevant screens send the events necessary for transitions 
20110625 21:59:38< zaroth> Crab_: entity, such as game_instance class or something, sending all the necessary events
20110625 21:59:39< Crab_> zaroth: the state machine itself is C++, the state logic is in c++, the transitions are c++
20110625 22:00:07< Crab_> zaroth: yes, your are right, we can use this approach, but, if our 'external manipulator' is in c++, the users won't be able to provide their own test scripts.
20110625 22:00:50< Crab_> zaroth: but, if we forget about that ability for user modification and only restrict us to hardcoded routines, we can code those 'were are now in state X, what we want to do?' as C++ control strategies.
20110625 22:01:10< Crab_> zaroth: in fact, to simplify things, we can start with c++ only, without having to mix lua here
20110625 22:01:26< Crab_> zaroth: since we can always add a lua-controlled strategy later by proxying it with a c++ one
20110625 22:01:47< Crab_> zaroth: but, we should not send events directly from that external class
20110625 22:01:52< zaroth> That probably would be best, my mind is already boggling at numerous possibilities
20110625 22:02:13< Crab_> zaroth: ok, so sticking to c++ 'scripts' for now
20110625 22:02:26< zaroth> (and I never used lua before on top of that)
20110625 22:02:31< Crab_> zaroth: because each state is supposed to have at least some privacy and encapsulation, we should not tell 'if we are in state X, transition to Y' 
20110625 22:03:06< Crab_> zaroth: we should tell it instead 'if we are in state X, set the 'immediate_transition_to=Y' config attribute, and let the state itself honor it (after doing all the necessary stuff).
20110625 22:03:25< Crab_> zaroth: that way, all the state changes would be inside the 'main' state definitions
20110625 22:03:48< Crab_> but there would be an external c++ strategy which would run as a callback on entering each state
20110625 22:04:06< Crab_> and it would get a non-const ref to a config object used by that state
20110625 22:04:24< Crab_> so, it would be able to set parameters needed by the state
20110625 22:04:33< Crab_> but the state itself would act on them and do dialogs/transitions/etc
20110625 22:04:59< Crab_> does this separation of responsibilities between states and controlling entity makes sense ?
20110625 22:05:31< zaroth> I'll tell you after I tried to draw it
20110625 22:05:36< Crab_> ok 
20110625 22:09:32-!- Bob_The_Mighty [~chatzilla@cpc17-brig16-2-0-cust12.3-3.cable.virginmedia.com] has joined #wesnoth-dev
20110625 22:14:50< CIA-95> ivanovic * r50042 /trunk/ (4 files in 3 dirs): updated Irish translation
20110625 22:27:43< zaroth> Crab_: so far, is it coherent with your idea? or do you already see something out of the place?
20110625 22:27:44< zaroth> http://i.imgur.com/blaf9.png
20110625 22:28:14< zaroth> (I didn't start putting the strategies here yet)
20110625 22:30:22< Crab_> looks good. I'd only add a place to specify the user information
20110625 22:30:43< Crab_> in MP, that's the place to login, in SP, it's skipped (but, if we later add profiles, it would be the place to load them)
20110625 22:31:00< zaroth> that's actually also my problem, kind of
20110625 22:31:03< Crab_> it's somewhen in no_specifics_known
20110625 22:31:08< Crab_> why it's a problem ?
20110625 22:31:17< zaroth> what's the relation between screens and states...
20110625 22:31:43< zaroth> because in the google docs document we used the screens as kinda states
20110625 22:32:06< zaroth> but it occured to me today morning that since we are going to skip the states in quite a few cases
20110625 22:32:13< zaroth> maybe tying them to screens isn't that smart
20110625 22:32:15< Crab_> we have the idea that state is 'config cfg; strategy->act(cfg);  if (cfg["foo"] show_screen_1(cfg(...) ...   '
20110625 22:32:26< Crab_> it all depends on the granularity 
20110625 22:33:02< Crab_> we'll have no problems skipping the dialogs in the state, that way
20110625 22:33:23< Crab_> (where strategy->act(cfg) would actually unwrap the stack of strategies)
20110625 22:33:25-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit [Ping timeout: 260 seconds]
20110625 22:33:40-!- zookeeper [~l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20110625 22:33:52< Crab_> so, '1 state - lots of screens' is ok
20110625 22:34:04< zaroth> but, no screens spanning multiple states
20110625 22:34:07< zaroth> right?
20110625 22:35:00< Crab_> zaroth: yes and no. continuations are ok.
20110625 22:35:13< Crab_> zaroth: i.e., mp lobby -> select map to create -> back to lobby.
20110625 22:35:42< Crab_> zaroth: the lobby is shown 2 times here, and it can be the same screen object, it's just it has to return control
20110625 22:36:06< Crab_> but, note - if we only consult the strategy at the beginning of the state, then our automation is limited to stuff we can set at the start of the state.
20110625 22:36:21< Crab_> in this approach, current wesnoth's code can be thought as having 1 big state
20110625 22:36:25< zaroth> well, what I meant is that the second screen
20110625 22:36:36< zaroth> won't have direct access to the previous screen object
20110625 22:36:51< zaroth> e.g. map_selection state won't have access to lobby 
20110625 22:37:05< Crab_> depends on where the lobby lives
20110625 22:37:16< Crab_> if it's bound to a big outer state, it will be available
20110625 22:37:20< zaroth> yes, yes
20110625 22:37:28< zaroth> I meant parallel states, obviously
20110625 22:37:33< Crab_> if it's only bound to an inner state, it'll be out of scope by this time
20110625 22:37:49< zaroth> s/parallel/neighboring/
20110625 22:37:56< zaroth> not concatenated
20110625 22:38:13< zaroth> s/concatenated/nested/
20110625 22:38:55< zaroth> it seems I'm a bit tired ...
20110625 22:40:33-!- Crendgrim [~crend@77-22-254-233-dynip.superkabel.de] has quit [Remote host closed the connection]
20110625 22:52:52< Crab_> (all ok, I understood the 'parallel = not nested' from the start)
20110625 22:52:53< zaroth> Crab_: ok, started rereading your statements (I *am* tired) - as for the proper nesting of scripts, I imagined it as a one big class with functions run() and handle_state_x() handle_state_y() (of course the handle_* functions behaviour can depend on class variables, i.e. state of the script)
20110625 22:53:24< zaroth> (but that was probably meant for lua-like scripts anyway)
20110625 22:53:27< Crab_> and I imagine it as a stack of those big classes
20110625 22:53:54< Crab_> also note that the script does not have to handle all states, it can handle only 1-2+
20110625 22:55:31< zaroth> yes, but there can be default handlers, can't they? the ones that just display stuff normally
20110625 22:55:51< Crab_> no, the handlers do not have to display stuff at all
20110625 22:55:59< Crab_> the scripts should not contain the control flow at all
20110625 22:56:06< Crab_> they should only modify the config
20110625 22:56:32< Crab_> and the 'display stuff normally OR do something else' logic should be in the states themselves
20110625 22:59:39-!- Exasperation [4a47319b@gateway/web/freenode/ip.74.71.49.155] has quit [Quit: Page closed]
20110625 23:01:43-!- MeccaGod [~majs@h146n3fls33o279.telia.com] has quit []
20110625 23:14:19-!- Gallifax [~IceChat77@p5DC675FD.dip.t-dialin.net] has quit [Quit: Copywight 2007 Elmer Fudd. All wights wesewved.]
20110625 23:18:34-!- Bob_The_Mighty [~chatzilla@cpc17-brig16-2-0-cust12.3-3.cable.virginmedia.com] has quit [Quit: ChatZilla 0.9.87 [Firefox 5.0/20110615151330]]
20110625 23:26:28-!- wesbot changed the topic of #wesnoth-dev to: 218 bugs, 320 feature requests, 27 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://pastebin.com | http://imagebin.org
20110625 23:43:42< Crab_> boucman: see the code that I'm thinking about for that stuff - http://pastebin.mozilla.org/1258233
20110625 23:43:55< boucman> looking
20110625 23:44:43< boucman> ok, sounds interesting
20110625 23:45:04< boucman> so by using cfg, we could script the way these screen go one after the other through WML ?
20110625 23:46:00< Crab_> yes, we can, if we name everything differently. but, I think that cfg should used to pass information from state to next state only.
20110625 23:46:46< zaroth> boucman: a bit of discussion happened in private, here it is: http://pastebin.com/LYM0G8qQ
20110625 23:51:47< boucman> ok, thx zaroth
20110625 23:52:25< zaroth> Crab_: one more question, since the #defined contents go on top of headers, how do we get rid of #defined contents to go back to header only state?
20110625 23:52:53< zaroth> do we get rid of config and reload everything without defines?
20110625 23:53:08< Crab_> take from cache, if we have. else, reload.
20110625 23:56:00< zaroth> btw, I assigned this bug to myself in hope it will solve itself once we design the whole state thingy clearly ;-) http://gna.org/bugs/?18269
20110625 23:57:20< zaroth> all right, I'm too tired to do something productive anymore, good night everyone
20110625 23:57:36< Crab_> ok. I'll post a big example in some minutes...
20110625 23:58:09< zaroth> I will leave IRC on, so highlight me all you need :-)
20110625 23:58:15< Crab_> ok
20110625 23:58:40< zaroth> will read in the morning
20110625 23:58:52< zaroth> by the way, the way I understand it now
20110625 23:59:14< zaroth> basically *everything* happens in state entry actions
20110625 23:59:27< zaroth> (and in event handlers, obviously)
20110625 23:59:32< Crab_> see the first part - http://pastebin.mozilla.org/1258244
--- Log closed Sun Jun 26 00:00:58 2011