--- Log opened Thu Jan 01 00:00:01 2009
--- Day changed Thu Jan 01 2009
20090101 00:00:01< Ivanovic> happy new year everyone!
20090101 00:00:12< mordante> yes happy new year
20090101 00:00:50< Kest> happy new year
20090101 00:04:05< crimson_penguin> Ivanovic: just added!
20090101 00:04:14 * crimson_penguin hopes it works
20090101 00:04:31< Ivanovic> okay, then i will write the announcement now
20090101 00:04:39< mordante> :-)
20090101 00:04:53< crimson_penguin> gah! I did something wrong :(
20090101 00:05:01< mordante> :-(
20090101 00:05:16< crimson_penguin> Python won't work on 10.4 (which means it won't work on 10.4)
20090101 00:05:53< crimson_penguin> if I can get someone with faster internet than mine though, with a Mac, it's pretty easy to fix
20090101 00:06:33< crimson_penguin> Ivanovic: I really should've checked this before, but, my binary won't work on 10.4 right now
20090101 00:06:40< Ivanovic> :(
20090101 00:07:01< crimson_penguin> Don't suppose you wanna fix it? :P
20090101 00:07:31< crimson_penguin> Just requires an install_name_tool command, and then remaking with dmg with hdiutil
20090101 00:09:50< CIA-52> zookeeper * r31850 /trunk/data/campaigns/Under_the_Burning_Suns/scenarios/06b_In_the_Domain_of_Dwarves.cfg: Fixed silly lack of proper facing.
20090101 00:11:04< Sirp_> Ivanovic: no response.
20090101 00:11:40< Ivanovic> crimson_penguin: uhm, difficult without osx
20090101 00:11:46< crimson_penguin> Ivanovic: oh :-/
20090101 00:11:46< Ivanovic> since i got no more mac at hand
20090101 00:11:56< crimson_penguin> Sirp: are you on OS X?
20090101 00:13:02< crimson_penguin> Oh... it may not be quite as simple as that...
20090101 00:15:45< CIA-52> zookeeper * r31851 /trunk/data/campaigns/Under_the_Burning_Suns/scenarios/05_A_Subterranean_Struggle.cfg: Fixed something about the cutscenes and renamed a macro.
20090101 00:17:04-!- Noyga [n=lame-z@AVelizy-151-1-70-13.w81-249.abo.wanadoo.fr] has joined #wesnoth-dev
20090101 00:22:04 * crimson_penguin can't figure out why it's linking to the system Python instead of the included one :-/
20090101 00:27:29< Noyga> happy new year
20090101 00:27:49< mordante> you too Noyga 
20090101 00:31:07< crimson_penguin> I did have to fix Python for 1.4.7, with install_name_tool, but the weird thing is it's linking to 2.3 this time, which it looks like it didn't before, because the actual version numbers are different
20090101 00:34:51< Ivanovic> everyone please check the announcement for mistakes: http://www.wesnoth.org/forum/viewtopic.php?f=5&t=23484
20090101 00:35:03< Ivanovic> the frontpage is updated, too
20090101 00:35:16< Ivanovic> crimson_penguin: i already linked the OSX binary anyway
20090101 00:35:24< Ivanovic> since at least the 10.5 users can use it, right?
20090101 00:35:33< crimson_penguin> true
20090101 00:37:17-!- stikonas [n=and@ctv-79-132-188-113.vinita.lt] has quit [Read error: 60 (Operation timed out)]
20090101 00:37:18< CIA-52> ivanovic * r31852 /trunk/RELEASE_NOTES: cleanup the file, release announcement (1.5.7) is done
20090101 00:38:27< mordante> Ivanovic you make a good start in the new year, I've no remarks this time :-)
20090101 01:05:30-!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090101 01:22:05< zookeeper> anyone else get an annoying quirk of new-style dialogs (sometimes/often) disappearing immediately?
20090101 01:22:28< zookeeper> as if the dismissing of the previous dialog somehow dismissed the next new-style dialog too
20090101 01:22:43< zookeeper> half of the time i don't even get to see the dialog
20090101 01:22:58< mordante> loonycyborg had it before but I could reproduce later
20090101 01:23:11-!- Blueblaze [n=Blueblaz@c-98-196-208-51.hsd1.tx.comcast.net] has quit [Read error: 110 (Connection timed out)]
20090101 01:23:45< mordante> but I'll look at it after I slept a bit
20090101 01:25:12< zookeeper> ok
20090101 01:33:13-!- zookeeper [n=l@dsl-tkubrasgw1-fe9ff800-215.dhcp.inet.fi] has quit []
20090101 01:34:03< Ivanovic> time for me to head off to bed, n8 everybody
20090101 01:34:10< mordante> night Ivanovic 
20090101 01:36:30< Espreon> Good night.
20090101 01:38:27< crimson_penguin> goodnight Ivanovic - happy new year!
20090101 01:39:02< crimson_penguin> Ivanovic: by the way, I did fix it on my side, but I have to either upload the whole thing again (4 hours or so), or get someone to replace the executable and remake/upload the .dmg
20090101 01:39:04< Noyga> happy new year crimson_penguin 
20090101 01:39:17< crimson_penguin> God nytt år Noyga!
20090101 01:39:33 * Noyga is still sending some "happy new year mails" ;)
20090101 01:39:49 * crimson_penguin wonders if anyone here has a Mac
20090101 01:40:02< Espreon> I do not.
20090101 01:40:12< crimson_penguin> ohh, Jetrel will be back soon-ish, maybe he can do it
20090101 01:40:36< Ivanovic> crimson_penguin: there are not many with a mac around in here
20090101 01:40:50< Ivanovic> but i think Sirp / Sirp_ also got a macbook
20090101 01:41:05< Ivanovic> maybe he can have a look at this, no idea though
20090101 01:41:05< crimson_penguin> yeah, he has one
20090101 01:41:40< loonycyborg> Unless his son destroyed it too :)
20090101 01:42:37-!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has quit ["Hi! I'm a quit message virus vaccine. If you see a quit message virus, don't replace your quit message with it!"]
20090101 01:42:41< crimson_penguin> Well, I'll ask him if he's around
20090101 01:47:12-!- Turuk [n=JDiSab@pool-71-242-162-195.phlapa.east.verizon.net] has quit ["To err is human.. but to really foul things up requires a computer"]
20090101 01:54:23-!- loonycyborg [n=sergey@79.139.246.35] has quit ["Zzzzzzzzzzzzzzzzzzzzzzzzz"]
20090101 02:03:30< CIA-52> mordante * r31853 /trunk/src/ (sdl_utils.cpp sdl_utils.hpp): Add get_rect_union() for future usage.
20090101 02:03:35< CIA-52> mordante * r31854 /trunk/src/gui/widgets/ (13 files): 
20090101 02:03:35< CIA-52> Add better draw support for invisible items.
20090101 02:03:35< CIA-52> Items in a scrollbar container which are not visible are no longer
20090101 02:03:35< CIA-52> redrawn even it they're dirty. (This was observed if a language which
20090101 02:03:36< CIA-52> was selected and scrolled out of sight still showed up when deselected.)
20090101 02:03:38< CIA-52> Items that are partly visible are still fully redrawn, this will be
20090101 02:03:40< CIA-52> addressed later.
20090101 02:03:42< CIA-52> mordante * r31855 /trunk/src/gui/widgets/ (16 files): 
20090101 02:03:44< CIA-52> Refactor the calls to the drawing routines.
20090101 02:03:46< CIA-52> All calls now go to a wrapper, this wrapper will be modified in the next
20090101 02:03:48< CIA-52> commit to allow partial drawing of widgets.
20090101 02:03:52< CIA-52> mordante * r31856 /trunk/ (4 files in 2 dirs): Fixed the listbox out of bounds drawing.
20090101 02:04:29< mordante> night
20090101 02:04:49-!- mordante [n=chatzill@roadie.xs4all.nl] has quit ["ChatZilla 0.9.84 [Firefox 3.0.5/2008120122]"]
20090101 02:54:11< CIA-52> shadowmaster * r31857 /trunk/src/ (6 files): Update copyright claims for addon_*.?pp, version.?pp
20090101 02:56:14-!- Noyga [n=lame-z@AVelizy-151-1-70-13.w81-249.abo.wanadoo.fr] has quit [Read error: 113 (No route to host)]
20090101 02:58:34< Shadow_Master> I LOVE THE ANNOUNCEMENT!
20090101 02:58:45< Shadow_Master> <oops... caps lock.
20090101 03:18:47-!- ABCD_ [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090101 03:27:43-!- Sapient [n=sapien-x@c-24-126-133-155.hsd1.ga.comcast.net] has joined #wesnoth-dev
20090101 03:27:54< Sapient> Kest: try the :mute command?
20090101 03:28:02< Kest> yes
20090101 03:28:06< Kest> only works on observers
20090101 03:28:15< Sapient> oh, lol
20090101 03:28:22< Kest> for whatever reason
20090101 03:28:52< Sapient> I think muting a player could be deemed 'tactical'
20090101 03:28:56< Shadow_Master> hai
20090101 03:29:09< Shadow_Master> can I has sapient¿?? 
20090101 03:29:18< Kest> a host abusing his abilities like that shouldnt be played with anyways
20090101 03:29:33< Shadow_Master> Sapient: happy new year.
20090101 03:29:36< Kest> he could kick guys he was losing against if he wanted
20090101 03:29:42< Sapient> happy new year, everyone
20090101 03:29:47< Kest> happy new year
20090101 03:30:14< Sapient> Kest: I think the mute feature is right as it is
20090101 03:30:37< Kest> Well, there's a discrepancy there, and i'm point it out
20090101 03:30:50< Shadow_Master> Sapient: what are your plans for 1.7 , if any? :)
20090101 03:30:59< Sapient> better to kick players who ignore warnings than to give hosts a subtly tactical abuse
20090101 03:31:47< Kest> host abuse is kind of irrelevant because they have other ways of abusing, but okay
20090101 03:31:50< Shadow_Master> Kest: think of it as if on IRC: on most networks you only have kick and ban.
20090101 03:32:23< Shadow_Master> only here and a few other ones you have quiet (+q).
20090101 03:33:00< Kest> I don't feel like arguing this. I disagree, that's all.
20090101 03:33:01< Sapient> Shadow_Master: there's always wiki/FutureWML
20090101 03:33:29< Sapient> I don't have any major plans at this point
20090101 03:33:37< Shadow_Master> but I am sure you can persuade a MP dev into implementing it :)
20090101 03:33:50-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Read error: 110 (Connection timed out)]
20090101 03:33:54< Sapient> when I get excited about something I tend to implement it right away ;)
20090101 03:34:21< Shadow_Master> Sapient: [event]  name="move" sounds to me like a possible source of performance issues
20090101 03:34:50< Sapient> I think the synchronization issues are far more worrisome
20090101 03:34:53< Shadow_Master> tjhe engine would have to scan for available event handlers whenever a unit moves, whenever it goes over a single hex of its path
20090101 03:35:13< Sirp_> Shadow_Master: that should not be an issue.
20090101 03:35:34< Sirp_> having to scan for e.g. 10 event handlers during a move should be a piece of cake.
20090101 03:35:36< Shadow_Master> and moving animations are a bit wird  ifyou spend uch time on other tasks
20090101 03:35:38< Sapient> matching against event handler names should be fast; I wrote it ;)
20090101 03:35:59< Sirp_> this is C++ we're talking about, not javascript
20090101 03:36:03< Shadow_Master> i.e. while the engine performs other tasks, the moving animation is playred within the bounds of the same hex over and over
20090101 03:36:29< Sirp_> huh?
20090101 03:36:42< Sapient> speaking of movement animations, dang they are glitchy looking in trunk
20090101 03:37:19< Sapient> that and missing portraits are the two big obvious bugs IMO
20090101 03:37:22< Shadow_Master> Sirp_: it's difficult to describe with words. and ifficult to reproduceatm unless you set "accelerated animations" to a very low speed and intrrupt scrolling somehow when a unit is moving out of the view
20090101 03:37:44< Shadow_Master> ah, and the scrolling speed at the minimum speed.
20090101 03:38:09< Shadow_Master> but then you can notice how movement animations work at the moment, and what would happen if the game performed other tasks while playing them.
20090101 03:39:01< Shadow_Master> unless of course, you block the rendering engine while performing the WML events. Which would be pretty awkward in this laptop already... I often have performance issues with very simple WML.
20090101 03:40:11< Shadow_Master> Sapient: auto-sotred game ingo sounds very interesting.
20090101 03:40:33< Shadow_Master> *storedd, info; But first I'd prefer to see  [modify_unit] ;) 
20090101 03:40:46-!- ABCD_ is now known as ABCD
20090101 03:40:55< Sapient> I think I see what you are saying. If the WML author performs expensive operations during every move to every hex then that is just poorly written WML, though
20090101 03:41:35< Shadow_Master> you already did a great endeavor by implementing formulas in vconfig-managed WML. btw, I read that they do not support floating point numbers yet, is that true? 
20090101 03:41:44< Sapient> that is true
20090101 03:41:56< Shadow_Master> :(
20090101 03:42:17< Sapient> if you bug DragonKing he might change that ;)
20090101 03:42:40< Shadow_Master> okay, nonetheless I plan to convert my campaign to use formulas for all the complex WML arithmetics
20090101 03:42:43< Sapient> I documented $( ... ) on wiki/SyntaxWML
20090101 03:42:52< Shadow_Master> Sapient: or Sirp ? :)
20090101 03:43:09< Sapient> Sirp was the one who said to bug DK :P
20090101 03:43:21< Shadow_Master> I experimented a bit with $() before reading the documentation and I found it pretty trivial.
20090101 03:43:30< Shadow_Master> I'll now read the documentation ;) 
20090101 03:45:31< Shadow_Master> OT: I'd have rather seen powers implemented with a ** operator in FAI, but I know that is not possible to change now
20090101 03:46:15< Sapient> I kinda like the hat :)
20090101 03:46:23< Shadow_Master> there is no ! for not either? 
20090101 03:46:26< Sapient> you wanted to reserve hat for bitwise operations?
20090101 03:46:36< Shadow_Master> Sapient: eh :)
20090101 03:46:52< Shadow_Master> I'm more used to it for that, and see power implemented as some kind of function
20090101 03:47:27< Sapient> I think the "Available variables" are not available to $( ... )
20090101 03:47:45< Shadow_Master> um?? 
20090101 03:47:58< Sapient> http://www.wesnoth.org/wiki/FormulaAI#Available_variables
20090101 03:48:16 * Shadow_Master was still reading about the dice operator
20090101 03:48:35< Sapient> I haven't played too much with FAI "available variables"
20090101 03:48:54< Sirp_> right, they're not available to $(...)
20090101 03:49:07< Sapient> Shadow_Master: I highly suspect that the dice operator is not MP safe
20090101 03:49:09< Sirp_> the way the formula system works is when the calling code calls it, it passes in all available variables.
20090101 03:49:17< Shadow_Master> Sap	me too
20090101 03:51:36< Shadow_Master> since the whole forumla system was imported from ST, which has no idea of MP
20090101 03:51:51< Sapient> ok, eventually we should review which variables we want to make available to $( ... )
20090101 03:52:27< Sapient> (if any)
20090101 03:53:05< Sapient> cya later, have fun.
20090101 03:53:05-!- Sapient [n=sapien-x@c-24-126-133-155.hsd1.ga.comcast.net] has left #wesnoth-dev []
20090101 03:53:18< Shadow_Master> I don't see anything in there athat wouldn't be doable with simple WML though (about getting the same info  variables provided)
20090101 03:53:25< Shadow_Master> ** those variables
20090101 03:59:11-!- ABCD_ [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090101 04:00:28-!- [Relic] [n=[Relic]@adsl-76-229-202-137.dsl.milwwi.sbcglobal.net] has joined #wesnoth-dev
20090101 04:01:25< [Relic]> Hello :)
20090101 04:02:21-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Read error: 113 (No route to host)]
20090101 04:02:44-!- ABCD_ is now known as ABCD
20090101 04:03:00< Shadow_Master> Happy new year to the Battle for Wesnoth project from Chile!
20090101 04:09:58< Kest> happy new year, chile
20090101 04:19:25-!- Ivanovic_ [n=ivanovic@dtmd-4db22081.pool.einsundeins.de] has joined #wesnoth-dev
20090101 04:35:25-!- Ivanovic [n=ivanovic@dtmd-4db24b61.pool.einsundeins.de] has quit [Read error: 110 (Connection timed out)]
20090101 04:37:24-!- Ivanovic_ is now known as Ivanovic
20090101 04:38:12-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has quit [Remote closed the connection]
20090101 05:02:20-!- crimson_penguin [n=ben@64.201.60.218] has quit ["Leaving"]
20090101 05:13:41-!- Shadow_Master_ [n=ignacior@unaffiliated/shadowmaster/x-275924] has joined #wesnoth-dev
20090101 05:22:28-!- Shadow_Master_ [n=ignacior@unaffiliated/shadowmaster/x-275924] has quit [Read error: 104 (Connection reset by peer)]
20090101 06:07:41< ABCD> Happy New Year!
20090101 06:12:41< Kest> happy new year
20090101 06:14:01-!- boucman [n=rosen@159.83.102-84.rev.gaoland.net] has quit [Remote closed the connection]
20090101 06:17:11< Smar> happy 07.17 
20090101 06:23:00< Shadow_Master> happy 1.5.7
20090101 06:23:34< Shadow_Master> (note 07.17 does not parse into anything that is a valid object derived from time)
20090101 06:24:03< Shadow_Master> (fatal: parse error at line 1, near "happy")
20090101 06:30:50< Smar> Shadow_Master: it’s time.
20090101 06:33:18-!- [Relic] [n=[Relic]@adsl-76-229-202-137.dsl.milwwi.sbcglobal.net] has quit ["Leaving"]
20090101 06:36:15< Shadow_Master> 07:17 ?
20090101 06:39:11< Smar> that.
20090101 06:39:17< Smar> . is acceptable in finnish
20090101 06:39:36< Smar> (some say prefferred. atleast easier to write)
20090101 06:41:49< Shadow_Master> aah. I  am used only to Spanish, English and French (and German which I don't understand at all but sounds good)
20090101 06:42:09< Shadow_Master> (German sounds like a really neat language for some reason)
20090101 06:42:38< Smar> :>
20090101 06:43:41< Kest> I'd like to learn German and Spanish
20090101 06:44:05< Smar> I’d like to have more time...
20090101 06:44:10< Kest> they seem like the most practical languages other than maybe Japanese
20090101 06:44:20< Kest> (not including English)
20090101 06:44:28< Smar> and maybe to know why I’m talking on two different chan with Shadow_Master...
20090101 06:44:36< Smar> :)
20090101 06:44:40< Kest> nuts
20090101 06:44:41< Smar> I’m learning japanese
20090101 06:45:30< Smar> I actually hate english
20090101 06:45:41< Shadow_Master> you are talking with my alter ego there. The UMC-making ego.
20090101 06:45:59< Smar> oh :(
20090101 06:46:20< Shadow_Master> here I am mainline developer/breaker/exploiter
20090101 06:46:44< Smar> and I’m...freeloader. I think.
20090101 07:59:59-!- Kest [i=ie@adsl-68-249-238-191.dsl.sfldmi.ameritech.net] has quit ["i go"]
20090101 09:50:23-!- mordante [n=chatzill@roadie.xs4all.nl] has joined #wesnoth-dev
20090101 09:50:50< mordante> morning
20090101 09:52:44< Sirp_> hi mordante 
20090101 09:52:52< mordante> hi Sirp_ 
20090101 09:54:18 * Sirp_ just programmed his first AJAX application.
20090101 09:55:35< mordante> and like AJAX programming?
20090101 09:57:10< Sirp_> it's interesting. :)
20090101 09:57:16< Sirp_> Javascript is actually a reasonably nice language.
20090101 09:57:30< Sirp_> though I find how closures are done a little confusing.
20090101 10:00:31-!- noy_ [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev
20090101 10:00:42-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has quit [Read error: 54 (Connection reset by peer)]
20090101 10:03:39< mordante> I did a few small projects in AJAX but still prefer C++
20090101 10:04:04< mordante> and somehow I still have the feeling it's harder to get things nicely structured in AJAX
20090101 10:05:51< Sirp_> mordante: it's certainly harder to program using AJAX.
20090101 10:06:03< Sirp_> mordante: of course I was using C++ -- wrote the server in C++, with Javascript as the client.
20090101 10:11:22< mordante> I used php and a postgresql server
20090101 10:11:56 * Sirp_ nods
20090101 10:15:18< Ivanovic> moin
20090101 10:15:24< mordante> hi Ivanovic 
20090101 10:16:42-!- loonybot [n=loonybot@79.139.246.35] has joined #wesnoth-dev
20090101 10:17:27-!- loonycyborg [n=sergey@79.139.246.35] has joined #wesnoth-dev
20090101 10:18:51< Sirp_> hi Ivanovic 
20090101 10:19:23< Ivanovic> Sirp_: you have seen the question by crimson_penguin?
20090101 10:19:30< Ivanovic> regarding the mac binary
20090101 10:19:49< Ivanovic> [01:39:03] <crimson_penguin> Ivanovic: by the way, I did fix it on my side, but I have to either upload the whole thing again (4 hours or so), or get someone to replace the executable and remake/upload the .dmg
20090101 10:20:50-!- stikonas [n=and@ctv-79-132-188-113.vinita.lt] has joined #wesnoth-dev
20090101 10:21:07< Sirp_> Ivanovic: fix what? this is the upload_log.cpp thing?
20090101 10:21:34< Ivanovic> Sirp_: no, soemthing is broken in the build, so that it currently is only usable for 10.5 users, not for osx 10.4 users
20090101 10:21:41< Sirp_> oh.
20090101 10:21:45< Ivanovic> and CP has just his rather small line
20090101 10:21:53< Sirp_> Ivanovic: okay....
20090101 10:21:55< Ivanovic> don't ask me how to fix things
20090101 10:22:01< Sirp_> Ivanovic: I have no idea how to fix it.
20090101 10:22:13< Ivanovic> CP knows how to, but he ain't around atm...
20090101 10:22:15< Ivanovic> ;)
20090101 10:28:33< Sirp_> Ivanovic: so according to sourceforge, we had 939,385 downloads in 2008. However, if they didn't have an outage on their stats server in Jun-Aug, I think we'd be very close to a million. :)
20090101 10:28:52< Ivanovic> yeah
20090101 10:29:01< Sirp_> on the other hand, this is still only marginal growth from the 890k we got in 2007
20090101 10:29:12< Ivanovic> we would be well over one million
20090101 10:29:28< Ivanovic> since we got at least 60 to 70k per month (whe stats are working)
20090101 10:29:38< Sirp_> hmmm I think it'd be around a million
20090101 10:30:00< Ivanovic> so it would be a growth of >100k extra
20090101 10:30:11< Ivanovic> so more something like 10 to 15%
20090101 10:30:16< Ivanovic> that ain't too bad
20090101 10:30:18< Ivanovic> ;)
20090101 10:31:05< Sirp_> yeah
20090101 10:31:29< Sirp_> we'd have had to average 83345 downloads/month in June-August to hit a million.
20090101 10:31:53< Sirp_> we got 89953 in May and 82333 in September.
20090101 10:32:14< Sirp_> so I'd say it'd have been very close, though we'd have probably gotten a million.
20090101 10:33:22< Sirp_> 10-15% growth rate is good for a 'mature' project.
20090101 10:33:30< Sirp_> if we could grow more than 10% in 2009 that would be nice.
20090101 10:35:24< Sirp_> okay I'm off to bed
20090101 10:35:26< Sirp_> goodnight.
20090101 10:35:36-!- Sirp_ [n=me@c-76-102-104-121.hsd1.ca.comcast.net] has quit ["leaving"]
20090101 11:05:01-!- zookeeper [n=l@dsl-tkubrasgw1-fe9ff800-215.dhcp.inet.fi] has joined #wesnoth-dev
20090101 11:06:03-!- Noyga [n=lame-z@AVelizy-151-1-70-13.w81-249.abo.wanadoo.fr] has joined #wesnoth-dev
20090101 11:09:00-!- noy_ [n=Noy@d75-157-52-251.bchsia.telus.net] has quit ["GO, GET TO THE CHOPPAH!!!"]
20090101 11:10:47< zookeeper> regarding moveover events: i now think it'd be better if a moveover event always stopped the unit. then we'd also have an auto-stored variable containing the path moved whenever a unit moves, so you could also just flag some variables when moving over a location.
20090101 11:11:28< zookeeper> i can only think of two use cases, 1) making traps or whatever which stop the unit and 2) just internally recording the info that the unit passed over that location
20090101 11:12:06< zookeeper> there's no need to try to do both with the same event, so we could just have a moveover event stop the unit and make people use the auto-stored path variable to figure out the latter.
20090101 11:13:32 * mordante hits Shadow_Master with his new year copyright update script, hard
20090101 11:13:49< mordante> Ivanovic are we still in soft string freeze? I have two small changes to the man pages s/2007/2009/ (and that's not typo ;-) )
20090101 11:15:19< Ivanovic> mordante: FIX IT!!!
20090101 11:15:48< mordante> ok
20090101 11:18:55-!- boucman [n=rosen@159.83.102-84.rev.gaoland.net] has joined #wesnoth-dev
20090101 11:28:16< CIA-52> mordante * r31858 /trunk/ (480 files in 19 dirs): New year copyright update.
20090101 11:28:46< CIA-52> mordante * r31859 /trunk/src/ (250 files in 12 dirs): Remove trailing whitespace.
20090101 11:29:05< mordante> Ivanovic commited in r31858
20090101 11:30:07< zookeeper> ok, so could someone tell me why those copyright notices need to have a date slapped to them anyway?
20090101 11:30:08-!- cib0 [n=cib@p4FD0C396.dip.t-dialin.net] has joined #wesnoth-dev
20090101 11:30:17< cib0> hello
20090101 11:30:23< boucman> morning
20090101 11:30:29< mordante> hi cib0 
20090101 11:30:29< boucman> and happy new year to all
20090101 11:30:38< mordante> yeah to you too
20090101 11:30:48< Noyga> happy new year boucman 
20090101 11:31:04< loonycyborg> zookeeper: For histerical reasons :)
20090101 11:31:43 * mordante wonders whether or not that was a slip of the finger
20090101 11:32:48< loonycyborg> Also it's needed so our grandchildren could find out that the copyright has expired and it entered public domain..
20090101 11:33:41< zookeeper> doesn't SVN hold that info ;)?
20090101 11:34:17< loonycyborg> While SVN does, tarballs don't.
20090101 11:34:19< mordante> not in the tarball we release
20090101 12:06:13-!- Niborelf [n=chatzill@gar13-1-82-229-16-70.fbx.proxad.net] has joined #wesnoth-dev
20090101 12:28:07-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090101 12:37:49< mordante> zookeeper any hint where you miss dialogs?
20090101 12:43:47< mordante> nevermind I found a scenario
20090101 12:55:55< CIA-52> zookeeper * r31860 /trunk/data/campaigns/Under_the_Burning_Suns/scenarios/06b_In_the_Domain_of_Dwarves.cfg: Made the bat swarm at the lake a bit more manageable and commented out an undo-invalidating event.
20090101 13:06:26< CIA-52> mordante * r31861 /trunk/ (4 files in 2 dirs): (log message trimmed)
20090101 13:06:26< CIA-52> Fixed a closing bug in the new dialogs.
20090101 13:06:26< CIA-52> The old style dialogs get dismissed on a mouse down, the next mouse up
20090101 13:06:26< CIA-52> event was most of the time handled by the new dialog. The new dialog
20090101 13:06:26< CIA-52> issued an easy_close event which closed the dialog. Moved the handler in
20090101 13:06:29< CIA-52> the mouse_up code (which already ignores mouse up events without the
20090101 13:06:31< CIA-52> button being down). The effect was percieved as the new dialog not
20090101 13:06:38< mordante> zookeeper loonycyborg ^ that should fix the not showing up problem
20090101 13:08:08< zookeeper> cool
20090101 13:09:08< boucman> zookeeper: about bug 11570 (multiple simultaneous animations)
20090101 13:09:21< boucman> you say you want a secondary unit filter to change the direction 
20090101 13:09:43< boucman> I'm not sure that's the way to go, because it means you can only turn the unit in the direction of another unit
20090101 13:09:52< boucman> wouldn't a direction= be enough ?
20090101 13:10:19< boucman> or do we miss an easy way to get the direction of another hex ?
20090101 13:13:27-!- kitty_ [n=kitty@e180192204.adsl.alicedsl.de] has joined #wesnoth-dev
20090101 13:13:40< kitty_> hi!
20090101 13:13:45< kitty_> mordante: i just saw 1.5.7 for the first time myself - it's really cool, thank you! :D
20090101 13:13:45< kitty_> the resizing looks a lot better than i feared, very small sizes look a bit pixel-ish but it's really ok! i don't think i need to provide inbetween sizes.  what bothers me a bit is the width of the text - the lines can get very long and uncomfortable to read, could you restrict the lines to about 100 characters (including spaces) or somewhat like that?
20090101 13:14:19< mordante> hi kitty_ thanks
20090101 13:14:47< mordante> I can restrict the width in pixels not in number of characters
20090101 13:14:47< kitty_> fendrin: what's going on with Low - why does it display only the generic unit portraits in the dialogues, not the custom ones? apart from olurf, who's big transparent portrait ist shown is the old dialogue box.
20090101 13:15:04< mordante> kitty_ that's a bug in the current dialog engine
20090101 13:15:36< loonycyborg> kitty_: That was reported to mordante many, many times already.
20090101 13:15:39< zookeeper> boucman, well there is an easy way, but it's via a rather long macro
20090101 13:15:46< loonycyborg> Twice by me :)
20090101 13:15:53< kitty_> sorry, then
20090101 13:16:05< mordante> no problem
20090101 13:16:57< boucman> mordante: I would advice around 150 chars
20090101 13:17:14< boucman> that's what's most commonly used nowday that we are not limited to 80 chars anymore
20090101 13:17:30< boucman> zookeeper: yes, but limitating to occupied spaces might be restrictive too...
20090101 13:17:37< boucman> what to do, what to do...
20090101 13:17:43 * Rhonda thanks Shadow_Master :)
20090101 13:17:59 * mordante loves 80 chars so I can add two files next to eachother
20090101 13:18:15< mordante> well I prefer a number of pixels ;-)
20090101 13:18:19< kitty_> mordante: what about 900px max width?
20090101 13:18:20< zookeeper> boucman, hmm, allow an SLF instead :)?
20090101 13:18:29-!- Noyga [n=lame-z@AVelizy-151-1-70-13.w81-249.abo.wanadoo.fr] has quit [Read error: 113 (No route to host)]
20090101 13:18:39< zookeeper> then one can provide both a simple direction, or an adjacent unit, or whatever else
20090101 13:18:55< boucman> zookeeper: good idea, I'll do that...
20090101 13:19:44-!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090101 13:19:52< kitty_> and another question the changelog says lordbob's peasant, spearman and swordsman are in but i can't see them? and why didn't spaceinvaders saurians make it?
20090101 13:20:16< mordante> kitty_ I've seen the spearman
20090101 13:20:16< loonycyborg> I've seen spearman..
20090101 13:20:31< kitty_> hm, then it must be me
20090101 13:20:42< mordante> and I didn't get a greenlight for the saurians
20090101 13:21:27< fendrin> hi
20090101 13:22:20< loonycyborg> mordante: lordbob's portraits are used only in dialogue.
20090101 13:22:24< zookeeper> kitty_, the whole point of the release is AFAIK just to get people to see some of the new dialogs. if there's something missing, just ignore it :p
20090101 13:22:38< loonycyborg> Help system uses old portraits for some reason :?
20090101 13:22:42< kitty_> ok, ok, i just was wondering
20090101 13:22:54< mordante> I haven't change the help system
20090101 13:23:14< kitty_> loonycyborg - that's what i meant, i think lordbob didn't provide small black bg versions
20090101 13:23:25< mordante> no he didn't
20090101 13:24:10< mordante> kitty_ do you know how to modify wesnoth config files?
20090101 13:24:22< kitty_> no :O 
20090101 13:24:40-!- JonW [i=chatzill@cpc3-barn6-0-0-cust235.brnt.cable.ntl.com] has joined #wesnoth-dev
20090101 13:25:17< boucman> kitty_: don't exactly ignore it, you should still bother us so we get it in the next release :P
20090101 13:25:41< mordante> if you click on the wesnoth dmg do you have an option to view the contents of the dmg?
20090101 13:25:42< kitty_> and another question, i'd like to see the different transparent portraits displayed by the new dialogue (to check if the size looks right, effects look what they should look like tec.) - how do i do that? (apart from those which are already in campaigns) is there a way apart from writing wml?
20090101 13:25:44-!- kitty_ [n=kitty@e180192204.adsl.alicedsl.de] has left #wesnoth-dev []
20090101 13:25:53-!- kitty_ [n=kitty@e180192204.adsl.alicedsl.de] has joined #wesnoth-dev
20090101 13:27:17< boucman> hmm
20090101 13:27:30< boucman> zookeeper: is there an easy way in debug mode ?
20090101 13:27:43< boucman> using the test scenario and debug mode to create units, maybe
20090101 13:27:46< mordante> in the test scenario
20090101 13:28:39< boucman> ok, kitty_ what os are you using
20090101 13:28:40< mordante> kitty if you click on the wesnoth dmg do you have an option to view the contents of the dmg?
20090101 13:29:15< boucman> ok, reading mordante, sh'es probably using macosx :)
20090101 13:29:26< mordante> she is
20090101 13:31:31< kitty_> yup it's mac osx and no i think ican't the the contents of the dmg
20090101 13:32:13< mordante> maybe Ivanovic knows how to modify files in the Wesnoth package on a mac
20090101 13:33:13< mordante> and how to start wesnoth with the test scenario
20090101 13:34:36< boucman> in the mean time, is there an easy way to make a unit speak (creating a unit is easy with debug mode)
20090101 13:34:57< boucman> maybe by creating a unit in the tutorial and moving it to a specific, event triggering, location
20090101 13:35:19< mordante> I don't think there's an easy way, except in the test scenario
20090101 13:37:29-!- cib0 [n=cib@p4FD0C396.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)]
20090101 13:38:50< boucman> darn
20090101 13:40:36-!- cib0 [n=cib@p4FD0F6F6.dip.t-dialin.net] has joined #wesnoth-dev
20090101 13:41:18< mordante> indeed, modifying the maximum width is trivial (at least for testing)
20090101 13:41:58< zookeeper> ideally the maximum width would be less when there's not enough text to expand the dialog vertically, and more when there is (to minimize vertical expansion), right?
20090101 13:43:01< mordante> I don't really understand what you mean
20090101 13:45:11< zookeeper> nevermind
20090101 13:45:33< zookeeper> i guess a text width similar to the one in the old dialogs would be fine.
20090101 13:45:41< zookeeper> AFAIK it's wider in the new ones.
20090101 13:46:14< kitty_> woho! i found a way- i just mess with the unit descriptions in existing scenarios.
20090101 13:46:16< mordante> the new width depends on your screen width so on a 1680 wide screen it's about 1100
20090101 13:46:38< mordante> kitty_ you can modify scenarios?
20090101 13:46:59< kitty_> i just taught me.
20090101 13:47:46< mordante> can you look for a file named data/gui/default/widget/label_default.cfg 
20090101 13:48:23< mordante> there's a line max_width = 0 there if you change the 0 to 900 the dialog text will be maximal 900 pixels wide
20090101 13:48:55< boucman> kitty_: be carefull, this is a slippery slope :P
20090101 13:49:14< kitty_> argh, too complicated- i can't find anything named label_default.cfg
20090101 13:49:14< mordante> don't listen to boucman :-P
20090101 13:50:34< mordante> :-(
20090101 13:50:47< kitty_> sorry  :( 
20090101 13:51:02< loonycyborg> Do macs have slocate? :)
20090101 14:13:52-!- Rrenys [n=rrenys@81-20-159-197.levira.ee] has joined #wesnoth-dev
20090101 14:22:42< kitty_> mordante, everything looks fine - but i get a strange white "outline" around the MoL... http://www.box.net/shared/h0zkd8p5zy
20090101 14:29:17< mordante> silly site needs cookies to show an image :-/
20090101 14:29:25< mordante> is that image scaled kitty_ ?
20090101 14:30:29< kitty_> ok, i'll choose another image site next time - but the image isn't scaled by meat all, just like it looks like ingame, mordante
20090101 14:30:56< mordante> I meant is it scaled in game or is this at its real size
20090101 14:31:08< kitty_> how can i see that?
20090101 14:32:01< mordante> uhm yes good question, guess you can't
20090101 14:32:09< boucman> :P
20090101 14:32:20< kitty_> :P 
20090101 14:32:29< mordante> unless you put it next to the original in photoshop :-P
20090101 14:32:47< kitty_> i can d that - but why do you want to know?
20090101 14:32:56< kitty_> *do
20090101 14:33:22< mordante> well if it's scaled it might be a scaling artifact if not it might be another artifact
20090101 14:34:14< kitty_> it's a tad smaller than the original
20090101 14:34:47< mordante> can you get wesnoth bigger on your screen?
20090101 14:35:11< boucman> ESR_: around ?
20090101 14:35:16< kitty_> nope, that's my max
20090101 14:35:35< mordante> :-(
20090101 14:35:40< kitty_> don't you get that effect on your computer?
20090101 14:39:18< mordante> yes, but I'm no artist, so never noticed it before
20090101 14:43:09< CIA-52> zookeeper * r31862 /trunk/data/campaigns/Under_the_Burning_Suns/scenarios/08_Out_of_the_Frying_Pan.cfg: Prevent annoying undo invalidations.
20090101 14:43:19< zookeeper> what i really don't get is how anyone of the half dozen maintainers or fixers of the campaign apparently never notice these kind of things.
20090101 14:53:08< boucman> zookeeper: could you write me a quick slf that matches "any neighbouring hex with a unit on it"
20090101 14:54:15< zookeeper> boucman, [filter_location] [filter][/filter] [filter_adjacent_location] x,y=4,1 [/filter_adjacent_location] [/filter_location]
20090101 14:54:45< zookeeper> matches any neighbour of 4,1 with any unit on it
20090101 14:55:00< boucman> k, cool, I can test my stuff now
20090101 14:56:38< mordante> kitty_ I think this is the full size and I also see a slight white border http://imagebin.ca/view/Y3C0WVY.html
20090101 14:58:07< Rrenys> and next to the white border is a black one
20090101 14:58:49< kitty_> :O  then it is my fault - you'll get a cleaner version! sorry!
20090101 14:59:24< mordante> no problem, I'm happy it can be fixed easily
20090101 15:16:05-!- Turuk [n=JDiSab@pool-71-242-102-166.phlapa.east.verizon.net] has joined #wesnoth-dev
20090101 15:21:42-!- JonW [i=chatzill@cpc3-barn6-0-0-cust235.brnt.cable.ntl.com] has quit ["Don't give up the fun has only just begun!"]
20090101 15:27:46< mordante> kitty_ do you also read the dev-ml?
20090101 15:28:04< kitty_> no - what is dev-ml?
20090101 15:28:41< mordante> a mailing list which the developers (sometimes) use to discuss things
20090101 15:29:11< mordante> archives are here https://mail.gna.org/public/wesnoth-dev/
20090101 15:29:11< kitty_> oh, should i? is it important?
20090101 15:29:41< mordante> most of the time not and not too much traffic
20090101 15:30:15< boucman> kitty_: it's very low traffic, and we usually assume all devs read it, so you might want to subscribe
20090101 15:30:16< mordante> I asked because I'm about to send a question to it and I like your opinion as well
20090101 15:30:26< kitty_> ok, good to know.
20090101 15:38:32-!- stikonas [n=and@ctv-79-132-188-113.vinita.lt] has quit [Remote closed the connection]
20090101 15:41:24< CIA-52> elias * r31863 /trunk/po/wesnoth-utbs/de.po: translation update
20090101 15:56:51-!- cib0 [n=cib@p4FD0F6F6.dip.t-dialin.net] has left #wesnoth-dev []
20090101 15:58:43-!- kitty_ [n=kitty@e180192204.adsl.alicedsl.de] has quit []
20090101 16:09:22-!- loonybot [n=loonybot@79.139.246.35] has quit [Read error: 104 (Connection reset by peer)]
20090101 16:16:44< CIA-52> boucman * r31864 /trunk/src/ (5 files): fix bug 11570 and also allow a [facing] slf in the animate_unit action
20090101 16:19:49< boucman> tada
20090101 16:21:11< CIA-52> boucman * r31865 /trunk/changelog: changelog update
20090101 16:25:21-!- crimson_penguin [n=ben@64.201.60.218] has joined #wesnoth-dev
20090101 16:34:52< fendrin> zookeeper: I did notice that utbs things. That is why I suggested to remove it ;-)
20090101 16:41:38< Shadow_Master> zookeeper: have you made more fixes to UtBS dehydration since  31841 ?
20090101 16:42:19< Shadow_Master> I'm merging 31841, 31840 and 31828 on IftU, just in case...
20090101 16:44:29< crimson_penguin> anyone know what I should do to replace the Mac binary on SF?
20090101 16:44:57< crimson_penguin> should I just add the new one as a new file, and get someone to change the links? or delete the old one, then add the new one?
20090101 16:46:28-!- Baufo [n=thomas@62-47-134-45.adsl.highway.telekom.at] has joined #wesnoth-dev
20090101 16:48:53< Ivanovic> make a new one adding an 'a' behind the version string
20090101 16:48:58< Ivanovic> (or a -1)
20090101 16:49:05< Ivanovic> and ping me once the file is added
20090101 16:50:23< CIA-52> mordante * r31866 /trunk/src/gui/widgets/grid.hpp: 
20090101 16:50:23< CIA-52> Remove an obsolete function.
20090101 16:50:23< CIA-52> The function was only declared in the header, but not defined in the
20090101 16:50:23< CIA-52> implementation.
20090101 16:50:29< CIA-52> mordante * r31867 /trunk/src/gui/widgets/ (window.cpp window.hpp): 
20090101 16:50:29< CIA-52> Remove obsolete resized_ member.
20090101 16:50:29< CIA-52> The need_layout_ member can be used instead so no need for two members
20090101 16:50:30< CIA-52> for the same purpose.
20090101 16:50:34< CIA-52> mordante * r31868 /trunk/ (6 files in 2 dirs): Spacebar closes dialogs without a scrollbar again.
20090101 16:50:44< CIA-52> mordante * r31869 /trunk/ (4 files in 2 dirs): 
20090101 16:50:44< CIA-52> Implemented the escape behaviour in the dialogs.
20090101 16:50:44< CIA-52> The new dialogs didn't honour the escape action, which didn't show the
20090101 16:50:44< CIA-52> following messages. This behaviour is now implemented in the new widgets
20090101 16:50:45< CIA-52> as well.
20090101 16:50:49< ESR_> boucman: You were looking for me?
20090101 16:51:11< boucman> yeah, I reread your campaign howto, I had a remark...
20090101 16:51:15< boucman> but I forgot :P
20090101 16:53:56< crimson_penguin> Ivanovic: ...oh - I just uploaded the new one, with the same name, and it seems like it replaced the old one
20090101 16:54:08 * crimson_penguin is impatient sometimes :(
20090101 16:55:18< crimson_penguin> the mirror I use hasn't updated, but, it didn't list a new file to add to the release
20090101 16:55:27< crimson_penguin> so either it just deleted it, or it replaced the old one
20090101 16:55:35< mordante> Shadow_Master, Ivanovic spacebar and escape now work in the new dialogs (spacebar only if no scrollbar used in the dialog)
20090101 16:56:42< crimson_penguin> Ivanovic: if you'd like, I could download the old one from the non-updated server, upload it over the new one, then upload the new one again with a new name?
20090101 17:03:16< Shadow_Master> iimorgood.
20090101 17:03:25< Shadow_Master> mordante:  ^ mangled message above
20090101 17:03:54< zookeeper> Shadow_Master, if 31841 is the village rehydration thing, then no
20090101 17:05:12< Shadow_Master>  /msg wesbot log $revnum  
20090101 17:05:27< Shadow_Master> but yeah
20090101 17:05:57< ESR_> fendrin: ping
20090101 17:06:28< zookeeper> 31841 is the last one i did, yes
20090101 17:07:37< fendrin> ESR_: pong
20090101 17:09:37< ESR_> fendrin: I'm trying to interpret some wmlscope messages about AI controller macros, and have some questions:
20090101 17:10:03< boucman> ESR_: THAT'S IT
20090101 17:10:23< boucman> in your howto, you should mention the AI controller macros in your thing about alliez AIs
20090101 17:10:25< boucman> :)
20090101 17:10:44< ESR_> 1) Is the call at 07_Elves_Last_Stand.cfg:457 for real?
20090101 17:11:04< ESR_> boucman: Maybe when they're finished and documented...
20090101 17:11:20< boucman> good point
20090101 17:12:43< fendrin> ESR_: 1) How can that call be unreal?
20090101 17:13:18< fendrin> That call is tested if that is what you ask for.
20090101 17:13:50< ESR_> It looks like a typo, frankly.  Perhaps you should add a comment documenting what it's doing?
20090101 17:16:03< fendrin> ESR_: Don't you believe that documenting should be in the ai_controller.cfg file? What let's it look like a typo?
20090101 17:17:05< crimson_penguin> Ivanovic: I'm not sure what to do now - I haven't done anything with SF since I replaced the file, but I have the old and the new one my server, and can upload them quickly to SF
20090101 17:18:01< ESR_> The combination of two blank args and a side set (I'm going to teach the macro artgument type checker about side sets as a distinct type, I think).  I'm not asking you to document AI_CONTROLLER there, just say what this xcall is intended to accomplish.
20090101 17:18:02-!- Noyga [n=lame-z@AVelizy-151-1-70-13.w81-249.abo.wanadoo.fr] has joined #wesnoth-dev
20090101 17:19:56< fendrin> ESR_: Okay, I will do that in the near future. Where is that macro arguement type checker located? In wmllint or in the engine?
20090101 17:20:22< ESR_> In wmllint.  Actually, in wesnoth/wmltools.py
20090101 17:21:56< ESR_> It detects mismatches between actual and expected macro argument types, deducing expected types by applying some rules to the parameter names.  For example, SIDE is expected to be an integer literal.
20090101 17:22:54< ESR_> E.g. if it sees a string in that position it knows something is wrong and complains.
20090101 17:24:10< ESR_> Current set of type-mismatch messages is here: http://rafb.net/p/ScJPiq24.html
20090101 17:24:23< fendrin> ESR_: good thing, wrong place.
20090101 17:24:33< ESR_> ?
20090101 17:25:03-!- Soliton [n=Soliton@rzdspc10.informatik.uni-hamburg.de] has joined #wesnoth-dev
20090101 17:26:35< fendrin> The ID in the ai_controller macro is an ID for that macro call. It's used for having different name spaces for each call. If you do have only one or for the first call to AI_CONTROLLER it's okay to leave that empty.
20090101 17:28:39< ESR_> Ah.
20090101 17:29:12< fendrin> So that ID is a different type to what wmllint expects.
20090101 17:30:20< ESR_> The easy way out would be to change that formal to VALUE, which matches anything.
20090101 17:30:42< ESR_> See the table at http://www.wesnoth.org/wiki/Maintenance_tools
20090101 17:31:29< ESR_> Alternatively, I could define a new optional-ID type, but I'm reluctant to do that if it will only be used once.
20090101 17:33:33< fendrin> ESR_: I don't believe that this hack is used only once. It's a work around for not having procedures with namespaces in wml.
20090101 17:34:22< ESR_> Oh.  Can the ID args in DISABLE_AID_CONTROLLER and DISABLE_AI_CONTROLLER_FOR_SIDE be empty as well?
20090101 17:34:45< fendrin> something like VAR would be fit best because that string will become a part of the variable name.
20090101 17:35:00< fendrin> s/fit/fitting
20090101 17:35:41-!- Sapient [n=sapien-x@c-24-126-133-155.hsd1.ga.comcast.net] has joined #wesnoth-dev
20090101 17:36:02< Sapient> has macro argument type checking ever actually proven useful?
20090101 17:36:14< fendrin> ESR_: Yes, I do believe that. You are talking about the ai_controller.cfg in data/core/macros?
20090101 17:36:15< ESR_> Hm, or I could call it AFFIX (linguist term for prefix/suffix/infix).
20090101 17:36:20< Sapient> it seems more of a maintenance burden than a maintenance tool
20090101 17:36:47< Sapient> i.e. creating more work than it saves
20090101 17:37:17< ESR_> Sapient: Yes, I've could a fair number of bugs with it. Usually calls that were garbled because somebody got argument number or type wrong.
20090101 17:37:30< ESR_> s/could/caught/
20090101 17:37:45< Sapient> define "fair number" ?
20090101 17:38:22< Sapient> btw, I am not talking about argument number here, just type
20090101 17:39:01< ESR_> Oh, I'd say I usually get 2 or 3 per minor release.  They'd be difficult to detect otherwise.
20090101 17:40:04< Sapient> I remember about one commit from dark forecast where there was an incorrect type used
20090101 17:40:07< Sapient> that's all
20090101 17:40:39< Sapient> WML variables don't really have "type" so I think the idea is flawed
20090101 17:41:05< Sapient> maybe you are inordinately attached to it as the creator to see its flaws
20090101 17:41:56< ESR_> They often have expectations about the associated semantics.  A macro that's expecting a coordinate range as a given arg, for example, is unlikely to work right if there's a string at that position in the call.
20090101 17:42:40< Sapient> I think such flaws will be easily enough detected by buggy or unexpected behavior
20090101 17:43:05< ESR_> I'm not sure why, but unit-generator macros used to be particularly likely to be screwed up this way.  
20090101 17:43:06< Sapient> presumably, the WML authors are testing their work
20090101 17:43:39< fendrin> testing wml code is a hard task with the game engine. 
20090101 17:43:55< ESR_> Part of the point is to reduce the testing burden so they (and the maintainers) can concentrate on higher-level issues.
20090101 17:44:19< Sapient> realy type checking is simply impossible
20090101 17:45:00< Sapient> "$(if($bool,123,'alpha'))"
20090101 17:45:13< ESR_> Sapient: No, it's quite feasible *in this environment*.  The fact that we don't have variadic macros helps a lot.
20090101 17:45:31< Sapient> what is the type of my statement above?
20090101 17:45:52< ESR_> Oh, I just pubnt on $-expressions.  They're not that common.
20090101 17:46:58< Sapient> I see
20090101 17:47:03< crimson_penguin> Ivanovic: I'm going now for a bit; if you tell me what to do, I'll do it when I'm back
20090101 17:47:12< Sapient> still, I dislike having so many reserved macro parameter names
20090101 17:47:47< Sapient> and it seems to cause more problems than justified
20090101 17:48:12< Sapient> but since you are the maintainer I guess that's your call
20090101 17:48:14< ESR_> When I set up the conventions I paid careful attention to exiting practice.  I didn't have to change very many calls.
20090101 17:49:21< Sapient> ok, consider this an encouragement not to expand the list of reserved names too much then
20090101 17:50:00< ESR_> Believe me, I don't want to.  It makes the tool more difficult to document.
20090101 17:50:10< fendrin> ESR_: what about VAR_AFFIX instead of id. It's an affix to a variable, you can take the type from VAR and read the fact that it can be empty from the AFFIX string.
20090101 17:50:59< fendrin> s/id/ID
20090101 17:51:59-!- ettin_ [n=jorda@55.207.219.87.dynamic.jazztel.es] has quit [Read error: 110 (Connection timed out)]
20090101 17:52:10< ESR_> I'm already testing AFFIX.  Yet's get the logic right before bikeshedding about the name  (another reason I prefer not to expand the reserved list).
20090101 17:52:16< CIA-52> boucman * r31870 /trunk/src/ (dialogs.cpp unit_display.cpp): slight cleanup in how units return to standing after an animation
20090101 17:54:51< Sapient> BTW, West's new music "Dark Passage" seems very promising
20090101 17:54:51< fendrin> ESR_: by the way, the ai_controller.cfg file you are messing with is not mine. It is written by zookeeper.
20090101 17:55:46< ESR_> Sapient: Good.  If I may dediuce from the name, it will be good for undrground, and wre need more variety for music in those settings.
20090101 17:57:16< Sapient> fendrin: I don't think zookeeper was the original author, but his thread seems lost from the forum crash
20090101 17:57:49-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Read error: 110 (Connection timed out)]
20090101 17:57:52< ESR_> Sapient: FYI, when I initially wrote the type-match checker I found I needed 13 (pseudo-)types to capture existiung practice.  I've since added only one; AFFIX will be the second.
20090101 17:58:46< fendrin> Sapient: I am sure that zookeeper is the original author of that code. I am the original author of that first ai_controller that inspired the one zookooper commit to core/macros.
20090101 17:58:59< Sapient> ah, I see
20090101 18:00:31< zookeeper> yup
20090101 18:01:16< Sapient> ESR_: let me just point out the ones I dislike then ... :)
20090101 18:01:20< CIA-52> esr * r31871 /trunk/data/ (core/macros/ai_controller.cfg tools/wesnoth/wmltools.py): Add AFFIX type to the macro type checker (name may change to VAR_AFFIX later).
20090101 18:01:24< ESR_> OK.
20090101 18:01:26< Sapient> *_SPAN - "melee" or "ranged"
20090101 18:01:51< Sapient> RANGE - "melee" or "ranged"
20090101 18:02:27< Sapient> ALIGN - "lawful" or "neutral" or "chaotic"
20090101 18:02:43 * ESR_ sees that Sapient displikes the fiddly little pseudo-types, which is reasonable.
20090101 18:03:11 * ESR_ doesn't particularly like them himself.
20090101 18:03:17< Sapient> now I could think of many harmless uses of the words above which would clash
20090101 18:03:26< Sapient> ALIGN "center"
20090101 18:03:58< Sapient> not sure if the new dialogs are going to support that or not ;)
20090101 18:05:03< Sapient> I could also use RANGE when describing how far a patrol may go, for example
20090101 18:06:37< fendrin> I thought RANGE would be something like x=23-43
20090101 18:06:53< Sapient> I had forgotten that you found a bunch of bad UNIT calls, so bravo for that
20090101 18:07:18 * Sapient tips hat
20090101 18:08:05< ESR_> Yes, I think bads unit-generators were the reason I wrote the facility originally.
20090101 18:08:41< ESR_> Later I attacked that problem from a different angle by standardizing the argument order of unit generators.
20090101 18:09:19< Sapient> if we really need attack range type checking then I'm afraid someting like ATTACK_RANGE would be better than reserving the word RANGE
20090101 18:09:29< ESR_> You'll notice. for example, that the first argument is now always SIDE and the second always TYPE.
20090101 18:10:04< Sapient> ah, standardization for consistency is a good thing yes
20090101 18:10:34< Sapient> I never used those macros much myself
20090101 18:10:48-!- ilor [n=user@unaffiliated/ilor] has joined #wesnoth-dev
20090101 18:11:03< ESR_> Sapient: I'm flexible about what the special formal names are.  If you were to come up with a proposed list of changes and file it as an FR you'd probably get satisfaction pretty quickly.
20090101 18:11:35< Sapient> kk
20090101 18:15:06< ESR_> You've actually zeroed in on the aspects ogf the type-checking that I myself dislike most.
20090101 18:15:46< ESR_> But I didn't want to rewrite half the macro corpus just to make my categories look a little cleaner.
20090101 18:22:48-!- BenUrban_ [n=benurban@c-68-49-10-243.hsd1.md.comcast.net] has joined #wesnoth-dev
20090101 18:24:39-!- BenUrban [n=benurban@unaffiliated/benurban] has quit [Nick collision from services.]
20090101 18:24:43-!- BenUrban_ is now known as BenUrban
20090101 18:25:13-!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090101 18:25:34-!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090101 18:35:12-!- ettin [n=jorda@145.207.219.87.dynamic.jazztel.es] has joined #wesnoth-dev
20090101 18:37:10-!- Baufo [n=thomas@62-47-134-45.adsl.highway.telekom.at] has quit ["Leaving"]
20090101 18:41:42 * zookeeper remembers to file the bug report about the ai controller bug
20090101 18:41:52< Sapient> yay
20090101 18:44:53< Sapient> let me know when it's up
20090101 18:52:53< CIA-52> esr * r31872 /trunk/data/ (2 files in 2 dirs): Clean up macro argument type mismatches.
20090101 18:54:15< zookeeper> Sapient, done
20090101 18:54:41< Sapient> thanks, zookeeper
20090101 18:54:59< zookeeper> Sapient, using [fire_event] with a name=menu item a_menu_item_id is supposed to be supported, right?
20090101 18:55:32< Sapient> ummm :)
20090101 18:55:43< Sapient> it was undocumented but I would like it to work
20090101 18:55:50< Sapient> so, yes
20090101 18:55:58< zookeeper> great
20090101 18:56:32< Sapient> but I didn't document it because I consider it an implementation detail
20090101 18:56:47< Sapient> as such, it could be dropped in a future version, although it's unlikely
20090101 18:57:05< Sapient> since the current implementation has worked well
20090101 19:00:03< zookeeper> added a perhaps crucial detail to the report
20090101 19:02:09< Sapient> so I can add that to the test scenario as well?
20090101 19:02:30< boucman> we should clean up the test scenario a little at some point...
20090101 19:02:42 * Sapient agrees
20090101 19:03:04< boucman> we should also do a test unit with all mainline powers...
20090101 19:03:46< ESR_> Mainline lints clean again.
20090101 19:04:11< zookeeper> yeah the test scenario is a horrible mess if you ask me ;)
20090101 19:07:01< zookeeper> Sapient, sure, please do
20090101 19:08:01< Sapient> hey I wasn't volunteering to clean it, lol
20090101 19:08:11< boucman> darn :(
20090101 19:08:13< boucman> :P
20090101 19:08:50< zookeeper> i meant adding it to the test scenario :|
20090101 19:09:03< Sapient> oh, yes. I am testing that now
20090101 19:11:03< mordante> ESR_ in the mood to fix another occurance of https://gna.org/bugs/?12632 ?
20090101 19:11:52< ESR_> Oh, fuck.
20090101 19:12:16< ESR_> Tell me about it.
20090101 19:12:46< zookeeper> ah, hmm. Sapient, when i comment out the event name thing, it still crashes when i exit the ai controller menu which pops up after i've selected the location.
20090101 19:13:01< zookeeper> (i think it was crashing more like that anyway when i first experienced it)
20090101 19:13:11< Sapient> I am not getting any AI controller menu appearing at all
20090101 19:13:56< mordante> ESR_ http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=510403 I tested it and it crashes 
20090101 19:14:24< mordante> I couldn't find the problem directly
20090101 19:14:46< mordante> and you looked at it before
20090101 19:14:52< Sapient> I added this line into the test scenario: {AI_CONTROLLER () 1 5,6 ()}
20090101 19:15:46< ESR_> mordante: Yes, but I don't think I actually undetrtand the code any better than you do.  Took Sirp to actually fix it.
20090101 19:15:58< ESR_> Hey, Sirp!  You there?
20090101 19:16:19< mordante> ah ok, you committed a fix so I thought you fixed it
20090101 19:16:47-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090101 19:17:08< ESR_> I committed a fix for a related problem, yes.  
20090101 19:17:09< zookeeper> Sapient, #define AI_CONTROLLER ID PLAYER_SIDE ALLY_SIDES VARIABLES
20090101 19:17:26< zookeeper> so put the correct allied side numbers...well, i think you can figure out where ;)
20090101 19:17:40< Sapient> mordante: your new widgets have broken the right click menu item "Perform Calculation" in the test scenario :(
20090101 19:17:51< ESR_> But my working method was rather too much like voodoo and groping in the dark for my own comfort.
20090101 19:18:26< mordante> Sapient how can I test it?
20090101 19:18:39< Sapient> right click on the silver mage leader
20090101 19:19:02< Sapient> you should be prompted to enter a formula
20090101 19:19:20< Sapient> instead it goes directly to the result of blank formula
20090101 19:19:45< Sapient> last minute breakage before the 1.5.7 release I guess
20090101 19:20:06< Sapient> that is not good
20090101 19:20:16< mordante> breakage yes, last minute no
20090101 19:20:37< Sapient> ok, it worked before I left for Xmas
20090101 19:21:07< zookeeper> meh, 1.5.7 is a breakage release anyway, i'd not recommend that for serious (campaign) play for anyone
20090101 19:21:08< mordante> which already a week ago, but I'll look into it
20090101 19:21:14< Sapient> I just hope we get 1.5.8 soon
20090101 19:21:16< zookeeper> i'm hoping we'll get 1.5.8 out in like a week or two
20090101 19:21:19< zookeeper> echo
20090101 19:22:01< Sapient> yes, I was excited about this release but it seems quite broken
20090101 19:22:41-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev
20090101 19:26:26< mordante> I'm not too surpriced some things broke, but I like that we now can get feedback on the new dialogs
20090101 19:26:33-!- Elvish_Pillage2 [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090101 19:27:29< zookeeper> well, as long as people actually get to see some decent amount of them.
20090101 19:27:32 * mordante compiles a fix
20090101 19:27:32-!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090101 19:28:02< zookeeper> i can only think of HttT where there's immediately some of them visible in the beginning
20090101 19:28:13-!- Psyche^ [n=Psyche@e177239016.adsl.alicedsl.de] has joined #wesnoth-dev
20090101 19:28:19< mordante> TB has one
20090101 19:34:31< Sapient> zookeeper: I don't see "defend a location" in the objectives menu
20090101 19:35:29< CIA-52> mordante * r31873 /trunk/ (4 files in 3 dirs): 
20090101 19:35:29< CIA-52> Added the sunset feature to the new dialogs.
20090101 19:35:29< CIA-52> The sunset feature is used to see which part of a dialog is redrawn.
20090101 19:35:38< CIA-52> mordante * r31874 /trunk/ (changelog src/game_events.cpp): 
20090101 19:35:38< CIA-52> Fix the calculate feature in the test scenario.
20090101 19:35:38< CIA-52> The new dialogs can't show an input so when there's a request for input
20090101 19:35:38< CIA-52> the old dialog needs to be shown.
20090101 19:35:41< mordante> Sapient fixed ^
20090101 19:35:58< zookeeper> Sapient, i don't see how that'd be possible.
20090101 19:36:03< zookeeper> maybe you haven't updated.
20090101 19:36:04< Sapient> thanks, mordante
20090101 19:36:08< mordante> no problem
20090101 19:40:33-!- Patterner [n=Psyche@e177115060.adsl.alicedsl.de] has quit [Read error: 110 (Connection timed out)]
20090101 19:40:34-!- Psyche^ is now known as Patterner
20090101 19:42:41< Sapient> ok, I was using an outdated version of ai_controller. re-trying
20090101 19:48:41-!- Sirp_ [n=me@c-76-102-104-121.hsd1.ca.comcast.net] has joined #wesnoth-dev
20090101 19:49:56-!- Sapient [n=sapien-x@c-24-126-133-155.hsd1.ga.comcast.net] has quit [Nick collision from services.]
20090101 19:50:00-!- Sapient [n=sapien-x@c-24-126-133-155.hsd1.ga.comcast.net] has joined #wesnoth-dev
20090101 19:52:30< boucman> zookeeper: around ?
20090101 19:53:40< zookeeper> yes
20090101 19:53:59< boucman> could you try to attack with the silver mage in the test scenario
20090101 19:54:30< boucman> it's missile anim is "interesting" but I'm not 100% sure it's not intended, and even if it isn't, I'm not sure what is supposed to happen instead
20090101 19:54:53< boucman> he fires white balls that go away from the target then suddenly rush forwar to the target
20090101 19:58:41< zookeeper> yes, it's intended.
20090101 19:58:50< boucman> ok, cool
20090101 19:59:13< boucman> could have been a badly sliding bug... :P
20090101 19:59:49< zookeeper> sliding bugs tend to look like bugs, i don't think the missile anim does.
20090101 20:02:03-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has quit []
20090101 20:07:15-!- crimson_penguin [n=ben@64.201.60.218] has quit ["Leaving"]
20090101 20:07:22< Sapient> yeah, I got the crash. looks like it's trying to evaluate the name of an event handler, but the vconfig is holding a pointer to an expired config
20090101 20:07:40< Sapient> this could take a while
20090101 20:08:21< Sapient> I would have preferred a minimalistic test case :)
20090101 20:08:39< zookeeper> i wouldn't know how to make one :p
20090101 20:19:50< Ivanovic> re
20090101 20:20:15< mordante> hi Ivanovic 
20090101 20:25:16< Sapient> zookeeper: were you aware of this part? warning engine: variable_info: retrieving member of non-existant WML container, ai_controller_..mode
20090101 20:25:52< zookeeper> no
20090101 20:26:11< Sapient> this appears to be wrong:
20090101 20:26:12< Sapient> ai_controller_{AFFIX}.$ai_controller.side_$ally_side|_current_settings.current_behaviour|.mode
20090101 20:26:30< Sapient> somehow that evaluates to ai_controller_..mode
20090101 20:26:35< zookeeper> hmm, let me check
20090101 20:27:09-!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090101 20:27:20< Sapient> I highly recommend --log-warn=engine
20090101 20:28:02< CIA-52> ivanovic * r31875 /trunk/ (4 files in 3 dirs): updated Hungarian translation
20090101 20:28:31< Sapient> it issues that warning four times
20090101 20:28:43< zookeeper> yeah
20090101 20:28:49< Sapient> lines 69-89
20090101 20:29:10-!- Elvish_Pillage2 [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090101 20:30:17< Sapient> aha... then the crash comes when it clears the variable with the inserted tags
20090101 20:31:09< Sapient> or does it... hmm
20090101 20:31:32< zookeeper> fixing the warnings..
20090101 20:36:53< mordante> ESR_ Rhonda noticed that wmldata.py is marked as executable, but can't be run by itself. Is it correct that it should only be imported, if so then I remove the executable flag
20090101 20:36:54-!- Niborelf [n=chatzill@gar13-1-82-229-16-70.fbx.proxad.net] has quit [Read error: 104 (Connection reset by peer)]
20090101 20:37:07< CIA-52> zookeeper * r31876 /trunk/data/core/macros/ai_controller.cfg: Fixed some bad ordering of stuff.
20090101 20:37:52< zookeeper> that should do it, i think
20090101 20:38:30< Sapient> tks
20090101 20:38:57< Sapient> still crashes though?
20090101 20:39:03-!- ilor_ [n=user@unaffiliated/ilor] has joined #wesnoth-dev
20090101 20:39:10< Ivanovic> hi ilor
20090101 20:39:46< ilor_> hi Ivanovic
20090101 20:40:07< Ivanovic> ilor_: as you might have read on the wesnoth-dev ml, the beds are booked
20090101 20:40:11< Ivanovic> but in a different hostel
20090101 20:40:33< zookeeper> Sapient, presumably yes, since it crashes even when it never runs that code
20090101 20:40:35< zookeeper> didn't try though.
20090101 20:41:21< Rhonda> ESR_: wmlgrammar.py that is, not wmldata. Sorry for the confusion.
20090101 20:41:32< ilor_> Ivanovic: ok great
20090101 20:41:57< Rhonda> Ivanovic: Uploading, btw.
20090101 20:42:01< Ivanovic> ilor_: that is (just to clearly state it again): we do *not* have the intention of forming a group with boucman and Noyga in the hostel
20090101 20:42:05< Ivanovic> ;)
20090101 20:42:17< ilor_> of course not!
20090101 20:42:40< ilor_> ;)
20090101 20:42:42< boucman> Ivanovic: most of the hotel will be used by geeks with laptops
20090101 20:42:45< Ivanovic> though we might meet them somewhere in the town
20090101 20:42:47< boucman> we'll manage to blend ;)
20090101 20:43:03< Sapient> dang, gdb is tricky
20090101 20:43:04< Sapient> (gdb) print std::string@0x22d094
20090101 20:43:04< Sapient> No symbol "string" in namespace "std".
20090101 20:45:37-!- ilor [n=user@unaffiliated/ilor] has quit [Read error: 60 (Operation timed out)]
20090101 20:45:46< Sapient> any gdb masters here?
20090101 20:46:28-!- ilor [n=user@unaffiliated/ilor] has joined #wesnoth-dev
20090101 20:46:49 * loonycyborg is not gdb master. He got that error about std::string too.
20090101 20:47:35< loonycyborg> Sapient: What exactly you're trying to achieve?
20090101 20:48:13< Sapient> I want to print the value of the string "name" in stack 9
20090101 20:48:21< Sapient> #9  0x00535993 in game_events::event_handler::matches_name (this=0xae638a8,
20090101 20:48:21< Sapient>     name=@0x22d094) at C:\projects\wesnoth\wesnothd\src\game_events.cpp:3279
20090101 20:49:07< loonycyborg> Try "print name" :)
20090101 20:49:09< Ivanovic> Sirp: you should answer in this thread: http://www.wesnoth.org/forum/viewtopic.php?p=329199#p329199
20090101 20:49:29< Sapient> print name shows: $3 = 8
20090101 20:49:41< loonycyborg> at that stack level.
20090101 20:49:42< boucman> he probably activated the "report my usage" option
20090101 20:49:55< Ivanovic> boucman: IIRC it is active by default now
20090101 20:49:58< Sapient> how do you switch stacks?
20090101 20:50:01< boucman> k
20090101 20:50:07< Sapient> I really know nothing about this
20090101 20:50:10< loonycyborg> Sapient: "up" and "down"
20090101 20:50:29< loonycyborg> So you'll need to do "up" 9 times or so..
20090101 20:51:24< Sapient> ok, name is "apply_ai_params"
20090101 20:51:30< mordante> f number goes directly to the wanted frame
20090101 20:51:53< mordante> Rhonda which file is now marked as executable but shouldn't be?
20090101 20:53:30-!- fendrin_ [n=fabi@g226211073.adsl.alicedsl.de] has joined #wesnoth-dev
20090101 20:56:25< zookeeper> Sapient, so are you saying the apply_ai_params event has something to do with the crash?
20090101 20:57:00< Sapient> not exactly
20090101 20:57:10< Sapient> the apply_ai_params event is executing
20090101 20:57:27< zookeeper> oh, ok. as i said, the crash still occurs even if that event is not called.
20090101 20:57:28< Sapient> but then it cycles through the rest of the handlers and crashes on one
20090101 20:57:59< Sapient> so now I just need to figure out which one and why
20090101 21:01:18< Rhonda> mordante: ./usr/share/games/wesnoth/data/tools/wesnoth/wmlgrammar.py (and IMHO all of the files in the wesnoth directory - I wouldn't know why they should be directly executable, those are libraries only?)
20090101 21:02:16< Rhonda> wescamp.py even gives an error message when executed
20090101 21:02:30< mordante> for me in svn wmlgrammer.py is not marked as executable, wmldata.py is marked as executable for me
20090101 21:02:46< Rhonda> wmlparser can read stdin at least
20090101 21:03:03< Rhonda> It's executable in the tarball Ivanovic uploaded
20090101 21:03:25< mordante> then I think the tarball script might do that
20090101 21:03:45< Rhonda> Hmm, no, it isn't
20090101 21:03:49< Rhonda> Now that's strange.
20090101 21:04:12< mordante> wescamp.py is executable and should be executable I run that almost daily to upload addons to wescamp
20090101 21:04:30< Rhonda> Ah, wait.
20090101 21:04:51< Rhonda> I chmod +x it myself  *hides*
20090101 21:05:08< Rhonda> But it seems to had been proper before. Or I'm confused.
20090101 21:05:14< zookeeper> Sapient, i think i'm not getting the crash anymore after i removed the spawned event
20090101 21:05:41-!- ilor_ [n=user@unaffiliated/ilor] has quit [Connection timed out]
20090101 21:06:01< mordante> no idea for me in svn wescamp.py, wmldata.py and wmlparser.py are executable 
20090101 21:06:01< Sapient> whic spawned event?
20090101 21:06:22< zookeeper> line 567
20090101 21:06:58-!- fendrin [n=fabi@g226201170.adsl.alicedsl.de] has quit [Read error: 110 (Connection timed out)]
20090101 21:06:59< mordante> and those three are also the only ones with a shebang line
20090101 21:08:11< zookeeper> a what?
20090101 21:08:34< zookeeper> oh, right.
20090101 21:12:21< Sapient> ok, so it is an event spawned from an array variable which is causing the problem
20090101 21:13:27< Sapient> I did think of that case but apparently it wasn't tested thoroughly enough
20090101 21:14:00< zookeeper> err, how could the spawning from an array variable specifically cause any sort of a problem?
20090101 21:15:18< Sapient> because variables can be changed at any moment but events need to remain somewhat stable
20090101 21:16:26 * zookeeper doesn't get it
20090101 21:16:42< Sapient> don't worry about it; memory management stuff
20090101 21:16:47< zookeeper> right
20090101 21:17:05< zookeeper> i'd imagine all events to be as stable as any other after they've been created though.
20090101 21:17:36< Sapient> it's like you are driving in a car and then your car decides it's a piece of cheese in another country somewhere, then you crash ;)
20090101 21:19:42< Sapient> events which are spawned from volatile structures are supposed to get an automatically managed chunk of memory
20090101 21:19:46< boucman> Sapient: that's actually pretty clever way to explain how far things can get wrong due to a bug...
20090101 21:20:20< Sapient> lol, glad you liked it
20090101 21:20:34< Shadow_Master> re?
20090101 21:21:07< zookeeper> Sapient, but how is that different from spawning normal simple inner events?
20090101 21:21:46< zookeeper> why aren't those managed in the exact same manner?
20090101 21:22:01< Sapient> zookeeper, normal simple inner events are located in the grand "WML document", which is a constant config. thus guaranteed unchanging
20090101 21:22:36< zookeeper> how strange.
20090101 21:23:41-!- [Relic] [n=[Relic]@adsl-76-229-202-137.dsl.milwwi.sbcglobal.net] has joined #wesnoth-dev
20090101 21:24:16< zookeeper> frankly i wouldn't have expected the handling of any event contents to have been implemented in a manner requiring any manual memory management.
20090101 21:24:38< [Relic]> Hello :)
20090101 21:24:45< Sapient> depends on your definition of "manual"
20090101 21:24:48< zookeeper> i mean manual in terms of other than just destroying stuff not anymore used
20090101 21:25:19< zookeeper> well, anyway
20090101 21:26:05< Sapient> the test scenario does contain events spawned from variables (right click on the avenger)
20090101 21:26:38< Sapient> but apparently, your test case is more, eh... thorough
20090101 21:27:04< mordante> hi [Relic] 
20090101 21:27:11< [Relic]> :)
20090101 21:28:03< Shadow_Master> :)
20090101 21:28:18< Shadow_Master> oh noes full recompile time
20090101 21:28:24< ESR_> mordante, Rhonda: That's not actually my module, I think it's elias's.  But I think you're right and that +x can be removed.
20090101 21:28:36< Sapient> the  cache_key_ = 0x0, which means it was not using vconfig memory management (but should have been)
20090101 21:28:46< Rhonda> ESR_: Was my own error anyway. %-)
20090101 21:29:04< Shadow_Master> mordante: eh script?
20090101 21:29:29< mordante> Shadow_Master yes a bit of sed can do wonders
20090101 21:29:44< Shadow_Master> i'm too lazy.
20090101 21:30:15< Sapient> aha, this bug appears to have been introduced with the delayed_variable_substitution feature
20090101 21:30:25< Shadow_Master> but I think you know that.
20090101 21:32:24-!- fendrin_ [n=fabi@g226211073.adsl.alicedsl.de] has quit [Remote closed the connection]
20090101 21:32:29< Shadow_Master> Rhonda: is the wesnoth-debian.net log rotation working alright?
20090101 21:34:26< Rhonda> No, it's disabled because I haven't yet split the logs from the days I had the server issue.
20090101 21:36:02< Shadow_Master> ah, oky
20090101 21:37:43< mordante> Shadow_Master running a script is much faster as hand editing a few hundred files, so who's lazy?
20090101 21:37:48-!- Blueblaze [n=Blueblaz@c-98-196-208-51.hsd1.tx.comcast.net] has joined #wesnoth-dev
20090101 21:41:45 * Shadow_Master scratches head, confused
20090101 21:43:55< Shadow_Master> who did write terrain_label btw?
20090101 21:45:04< mordante> svn blame/git log
20090101 21:45:38 * mordante prefers git log since it shows all commits of a file and doesn't need a network
20090101 21:45:40< loonycyborg> git svn blame :)
20090101 21:46:06< mordante> loonycyborg what I just said about git log ;-)
20090101 21:47:16< loonycyborg> git svn blame doesn't seem to need network too..
20090101 21:49:31 * Rhonda still hates git svn not even trying to use tags directories as tags
20090101 21:50:14< Shadow_Master> mordante: um, git blame and log don't tell me much and I'm rather lazy to check the output in detail or do more complex history searching
20090101 21:50:47< mordante> git log -p shows all changes as a patch
20090101 21:52:02< Shadow_Master> but I curse the author for their  *gratuitous abuse of const_cast<>()* !
20090101 21:57:36-!- ABCD [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090101 22:03:08< Sapient> svn: command not understood 'curse'
20090101 22:05:10-!- Turuk [n=JDiSab@pool-71-242-102-166.phlapa.east.verizon.net] has left #wesnoth-dev []
20090101 22:13:12< CIA-52> ilor * r31877 /trunk/src/upload_log.cpp: fix a compile error (on vc9 at least)
20090101 22:16:38< Sapient> zookeeper: here is a summary of your test case and it is complicated http://rafb.net/p/VgZ0nG60.html
20090101 22:19:43< zookeeper> neat
20090101 22:20:12< Sapient> now you see why I was whining for a simpler test case ;)
20090101 22:20:55-!- voris [n=voris@c-24-16-107-149.hsd1.wa.comcast.net] has quit ["Leaving"]
20090101 22:22:50< zookeeper> well i haven't been whining about why i need it to be that complex to make it work ;)
20090101 22:24:08< Sapient> lol... quiet you :P
20090101 22:24:54< Sapient> my head is spinning just to figure out where to put my debug statements
20090101 22:26:04< zookeeper> well, i still didn't really understand why there'd need to be a problem in the first place, so i can't help with that :p
20090101 22:43:46< Sapient> ok, rather than try to understand exactly why it is crashing, I see two things that might be wrong, so I'm going to try changing those ;)
20090101 23:04:33-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090101 23:07:44< Sapient> ugh. I figured out the problem. WML menu item commands are volatile but they are using config instead of vconfig
20090101 23:08:41-!- Turuk [n=JDiSab@pool-71-242-102-166.phlapa.east.verizon.net] has joined #wesnoth-dev
20090101 23:08:42< Sapient> fixing this the right way could be ugly so I'm tempted to go with the one line hack
20090101 23:11:52< Sapient> all volatile configs which contain executable WML should really be using vconfig though
20090101 23:12:07< boucman> Sapient: you're the juge, but 1.5.7 has just been release, so now would be a relatively good time for intrusive changes
20090101 23:13:30< Sapient> yeah, I guess so
20090101 23:13:35< zookeeper> a one-line hack isn't a hack :|
20090101 23:13:53< boucman> zookeeper: one line hacks usually mean more bugs later
20090101 23:14:37-!- Rrenys [n=rrenys@81-20-159-197.levira.ee] has quit [Read error: 104 (Connection reset by peer)]
20090101 23:15:04< Sapient> yeah, I could commit the one-line version just to get the monkey off my back then revisit it
20090101 23:15:35< Sapient> the simple solution is just let spawned events always assume they are from a volatile source
20090101 23:16:20< Sapient> I'm testing that hack now
20090101 23:18:06< Patterner> how do I find out why SCons cannot find libfribidi?
20090101 23:19:02< Sapient> I didn't think we were still using fribidi
20090101 23:19:23< Sapient> now we have pango+cairo
20090101 23:19:32< Patterner> font.cpp does...
20090101 23:20:48< Sapient> ah, well you can probably safely disable it
20090101 23:21:44< Sapient> mordante might know more
20090101 23:22:54< loonycyborg> Patterner: In build/config.log
20090101 23:24:25< loonycyborg> fribidi might matter if you're using right-to-left language locale.
20090101 23:25:21< loonycyborg> Otherwise the only place where it's used(language dialog) already uses new widgets and consequently pango.
20090101 23:26:04< loonycyborg> s/where it's used/where right-to-left languages appear
20090101 23:29:14< Patterner> “Building "build/sconf_temp/conftest_13.o" failed in a previous run” isn't really helping...
20090101 23:29:22< CIA-52> sapient * r31878 /trunk/src/game_events.cpp: 
20090101 23:29:22< CIA-52> this is a quick hack to prevent bug #12803 ai_controller crash by letting spawned events always assume they are from a volatile source.
20090101 23:29:22< CIA-52> The real problem is WML menu item commands are volatile but they are using config instead of vconfig.
20090101 23:29:22< CIA-52> All volatile configs which contain executable WML should be using vconfig so this needs to be fixed eventually.
20090101 23:29:31< loonycyborg> Patterner: --conifg=force
20090101 23:29:39< loonycyborg> *--config=force
20090101 23:30:54< Patterner> "error: 'NULL' undeclared"...
20090101 23:30:59< Sapient> zookeeper: I still think spawned events are evil :p
20090101 23:31:26< Sapient> but with that version your ai_controller won't crash
20090101 23:31:45< zookeeper> cool, thanks :)
20090101 23:31:50-!- Kest [i=ie@adsl-68-73-204-15.dsl.sfldmi.ameritech.net] has joined #wesnoth-dev
20090101 23:31:54< Kest> hello
20090101 23:31:58< Sapient> yo Kest
20090101 23:32:04< Kest> how can I stop the camera from centering on another player's units in multiplayer when they move?
20090101 23:32:16< zookeeper> my view on the evilness of spawned events was reduced greatly by delayed_variable_substitution=
20090101 23:32:22< loonycyborg> Patterner: Which fribidi version?
20090101 23:32:23< zookeeper> you can't
20090101 23:32:31< Kest> i cant?
20090101 23:32:35< zookeeper> yep
20090101 23:32:39< Kest> damn
20090101 23:32:44< Kest> this is really irritating
20090101 23:33:00< Patterner> replacing NULL with 0 in SConstruct helps. Now it finds it...
20090101 23:33:01< Kest> we're playing a team time match on a large map
20090101 23:33:11< Sapient> yeah, especially when you are catching up on a long replay
20090101 23:33:21< Kest> so instead when you are planning stuff it is going off to the oither side of them map
20090101 23:33:33< Patterner> loonycyborg: 0.19 with utf8->charset change...
20090101 23:33:37< Sapient> you can right click to pop up a menu, Kest
20090101 23:33:42< Kest> yeah?
20090101 23:33:45< Sapient> that will stop it momentarily
20090101 23:34:16< loonycyborg> Patterner: Yeah. That charset change is the reason why fribidi 2 isn't supported.
20090101 23:34:25< Kest> dpesnt that freeze your game?
20090101 23:34:34< Sapient> pretty much
20090101 23:34:45< Kest> doesnt help then
20090101 23:36:01< loonycyborg> Patterner: So in this case the check fails as intended :)
20090101 23:36:37< loonycyborg> Obviously fribidi2 doesn't #include the header with NULL..
20090101 23:37:40< Kest> thanks though
20090101 23:40:07< Patterner> I thought NULL is a C keyword :)
20090101 23:40:55< loonycyborg> Patterner: No. It's a preprocessor macro :)
20090101 23:41:13< loonycyborg> That's why it's entirely capital letters.
20090101 23:41:28< Sapient> DOH
20090101 23:41:33< loonycyborg> All C keywords are small letters.
20090101 23:43:29< CIA-52> zookeeper * r31879 /trunk/data/campaigns/Under_the_Burning_Suns/scenarios/ (3 files): Removed redundant speaker images, fixed a couple of facing issues and a few minor typos.
20090101 23:44:11-!- zookeeper [n=l@dsl-tkubrasgw1-fe9ff800-215.dhcp.inet.fi] has quit []
20090101 23:51:08-!- boucman [n=rosen@159.83.102-84.rev.gaoland.net] has quit [Remote closed the connection]
--- Log closed Fri Jan 02 00:00:39 2009