--- Log opened Sun Jun 14 00:00:33 2009
--- Day changed Sun Jun 14 2009
20090614 00:00:33< Blueblaze> Soliton: In Wesnoth campaigns, or just in content in general?
20090614 00:00:43< Soliton> in general.
20090614 00:01:24< Blueblaze> Personally, I use it a decent bit, and have seen it here and there in other people's work
20090614 00:01:36< Soliton> i'm wondering how useful it is or if it is usually rather an accident where x,y are variables that are unset.
20090614 00:01:41< Blueblaze> It certainly has a use
20090614 00:03:16< Soliton> the thing is i'm wondering if there is an automatic solution to accidentally killing all units.
20090614 00:03:32< Blueblaze> Since sometimes you want to kill a particular unit without knowing the exact location of it (of course you could store that unit but then it adds an unnecessary step), and then you could also want to kill all of a certain type etc..
20090614 00:03:44< Blueblaze> ah, I see
20090614 00:03:45< Soliton> yeah, makes sense.
20090614 00:06:25-!- Canadork [n=Canageek@d24-141-252-217.home.cgocable.net] has joined #wesnoth-dev
20090614 00:07:32< Blueblaze> Soliton: I don't see a way to prevent you from killing all units just because it uses a necessary standard unit filter to select the target. A notice on the wiki could warn that using this without proper filters could potentially kill all units in a map.
20090614 00:07:37< Soliton> ok, slightly different question: is it ever useful to have [kill] x,y="" match everything?
20090614 00:08:13< Soliton> Blueblaze: what could be used is x,y being empty but there.
20090614 00:08:48< Blueblaze> I believe x,y="" would be the same as not mentioning x,y at all in the filter
20090614 00:08:48< Soliton> if you don't want to match by x,y you wouldn't ever specify them, right?
20090614 00:09:01< Blueblaze> right
20090614 00:09:09< Blueblaze> That's what I'd (and probably most people) would normally do
20090614 00:09:12< Soliton> from the code side it's not.
20090614 00:09:14-!- wesbot changed the topic of #wesnoth-dev to: 64 bugs, 242 feature requests, 13 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090614 00:09:50< Soliton> it is specifically checked whether the values are empty and if they are a default of true (matches the filter) is used.
20090614 00:10:13< Soliton> i'm not sure what the point of that is.
20090614 00:16:04< silene> Soliton: you mean in map_locatoin::matches_range?
20090614 00:16:55< Soliton> hmm, the default needs to be true, i think. but it could still be specifically checked whether both values are empty and not match then.
20090614 00:17:00< Soliton> silene: yep.
20090614 00:17:26-!- BenUrban [n=benurban@68.55.19.224] has joined #wesnoth-dev
20090614 00:19:22< silene> Soliton: note that the default of being true is used by terrain filters
20090614 00:21:14< Soliton> ok, x,y being empty could already get checked in unit::internal_matches_filter().
20090614 00:23:40-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has quit ["night all"]
20090614 00:25:46-!- Canageek [n=Canageek@d24-141-252-217.home.cgocable.net] has quit [Read error: 110 (Connection timed out)]
20090614 00:26:57 * Soliton pings Sapient.
20090614 00:29:14< Soliton> this is what i'm proposing: http://nopaste.com/p/aZligYlGZ
20090614 00:29:24< zookeeper> Soliton, well, x,y is surely the most common case, but using id= for instance isn't uncommon
20090614 00:29:36< Soliton> zookeeper: yeah, silly question.
20090614 00:29:57< Soliton> zookeeper: but specifying x,y as empty is not used on purpose, right?
20090614 00:30:30< zookeeper> right, i don't think i've ever seen that
20090614 00:30:34< Soliton> zookeeper: what i'm wondering about is this bug: http://gna.org/bugs/?13685
20090614 00:30:54< Soliton> i thought i fixed that a while ago but i guess it is still somehow possible for x and y to be empty.
20090614 00:31:09< Soliton> it could be checked in WML of course, but...
20090614 00:31:19< zookeeper> hmm, yeah
20090614 00:33:28< zookeeper> frankly whoever writes WML like that should realize to take such a case into consideration
20090614 00:34:18< zookeeper> basically i wouldn't mind if an empty/invalid filter matched none instead of all
20090614 00:34:23< Soliton> well, as usual everything should be rewritten nicely, but... :-)
20090614 00:35:18< Blueblaze> zookeeper: what would be the method to kill all then? just side=1-9?
20090614 00:35:19< zookeeper> but i think you'd better talk with sapient about that since it sounds like something that could be changed about filters in general
20090614 00:35:36< zookeeper> Blueblaze, yes, or x,y=1-999,1-999 or whatever
20090614 00:36:07< zookeeper> frankly i don't think i've ever seen a need to kill everything indiscriminately
20090614 00:36:25< Soliton> with the change i propose you can still just leave x,y off completely.
20090614 00:36:37< zookeeper> even in the SoF ending you'd rather kill everyone in order, only displaying the animations of some of the units and so on
20090614 00:38:41< Soliton> i think in most cases an empty value will make the filter not match already. it's just a bit different with locations.
20090614 00:39:10< zookeeper> right
20090614 00:40:01-!- mjs-de [n=mjs-de@vpw.wh.uni-dortmund.de] has quit [Read error: 60 (Operation timed out)]
20090614 00:40:48< zookeeper> well, as i said, please confirm with sapient, it sounds ok to me though.
20090614 00:41:24< Soliton> sure.
20090614 00:42:12-!- giusef [n=giusef@unaffiliated/giusef] has quit ["exit (-1);"]
20090614 00:59:07< Soliton> Ivanovic: how are canned responses for bug reposrts done? wasn't there some feature like that?
20090614 00:59:34< Ivanovic> Soliton: ?
20090614 00:59:48< Soliton> https://gna.org/bugs/admin/field_values.php?group_id=1242&create_canned=1
20090614 00:59:58< Soliton> looks like i have no permission.
20090614 01:00:06< Ivanovic> no idea, we never used this stuff
20090614 01:00:17< Ivanovic> that is: at least i never used it
20090614 01:00:18< Soliton> i'm pretty sure mordante did/does.
20090614 01:00:31< Soliton> i just thought https://gna.org/bugs/?13686 could be a canned response.
20090614 01:00:34< Ivanovic> then please ask him about this stuff
20090614 01:00:40< Soliton> yeah, i will.
20090614 01:01:39< Ivanovic> i added a response with your text
20090614 01:01:43< Ivanovic> no idea how to use it though
20090614 01:02:39< Soliton> there is a drop down box under the post comment field.
20090614 01:03:01< Soliton> i guess whatever you choose there is added to the end of the comment or something.
20090614 01:03:05< Ivanovic> ah, nice
20090614 01:04:36< Ivanovic> heading off to bed now, n8
20090614 01:11:43-!- zookeeper [n=l@wesnoth/developer/zookeeper] has quit []
20090614 01:14:11-!- silene [n=plouf@wesnoth/developer/silene] has quit [Read error: 113 (No route to host)]
20090614 01:24:07< CIA-77> ai0867 * r36185 /trunk/data/tools/wesnoth/wmlgrammar.py: Update wmlgrammar a bit to newer WML.
20090614 01:27:41-!- Canadork [n=Canageek@d24-141-252-217.home.cgocable.net] has left #wesnoth-dev ["As I walk through the valley of the shadow of death/I take a look at my life and realize there's nothing left"]
20090614 01:32:35-!- grzywacz [n=grzywacz@wesnoth/developer/grzywacz] has quit [Remote closed the connection]
20090614 01:36:11-!- gil [n=gil@62.128.50.99] has left #wesnoth-dev []
20090614 01:44:55-!- Michael_Puer [n=michael_@188-23-7-193.adsl.highway.telekom.at] has quit [Read error: 60 (Operation timed out)]
20090614 01:47:12-!- Michael_Puer [n=michael_@188-23-0-178.adsl.highway.telekom.at] has joined #wesnoth-dev
20090614 02:14:16-!- loonycyborg [n=sergey@wesnoth/developer/loonycyborg] has quit ["Zzzzzzzzzzzzzzzzzzzzzzzzzzz"]
20090614 02:14:46-!- loonybot [n=loonybot@wesnoth/bot/loonybot] has quit [Remote closed the connection]
20090614 02:18:59-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090614 02:35:28-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090614 02:53:28-!- ilor_ [n=user@wesnoth/developer/ilor] has quit [Read error: 110 (Connection timed out)]
20090614 02:53:37-!- ilor_ [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090614 02:56:58-!- ilor [n=user@wesnoth/developer/ilor] has quit [Read error: 110 (Connection timed out)]
20090614 03:06:39-!- happygrue [n=George@wesnoth/developer/wintermute] has quit [Read error: 113 (No route to host)]
20090614 03:53:16-!- Sapient [n=sapien-x@wesnoth/developer/sapient] has joined #wesnoth-dev
20090614 03:53:21< Sapient> hi Soliton
20090614 03:53:58< Soliton> greetings human one.
20090614 03:54:02< Sapient> x="" should always fail
20090614 03:54:20< Sapient> if you look at the changelog, you'll see I already made that change in a few places
20090614 03:54:39< Sapient>  I didn't fix it everywhere
20090614 03:55:37< Sapient> the only downside is it may make some macros like something={OPTIONAL_VALUE} fail
20090614 03:55:51< Sapient> but I think that is an acceptable loss
20090614 03:59:51< Soliton> so i'll commit that patch then?
20090614 04:00:44< Sapient> Version 1.5.6:
20090614 04:01:00< Sapient>    * SingleUnitFilters can now check for empty keys such as role=$null
20090614 04:01:10< Sapient> Soliton: yes, sure
20090614 04:05:13< Soliton> Sapient: so basically one could look for all matches_range() calls and make sure the arguments are not empty (if appropriate)? easy coding task?
20090614 04:05:38< Sapient> umm...
20090614 04:05:48< Sapient> not sure what you mean by that last question
20090614 04:06:09< Soliton> whether that would be an easy coding task.
20090614 04:06:09< Sapient> I'm really tired so my brain may not be at full capacity ;)
20090614 04:06:52< Sapient> sounds like a job for grep moreso
20090614 04:07:05< Sapient> easycoding tasks don't get done too speedily
20090614 04:07:15< Sapient> but, then again, I'm not in any hurry
20090614 04:07:34< Soliton> that's what i thought.
20090614 04:11:00< Sapient> what needs to happen is more like this
20090614 04:11:36< Sapient> check cfg.has_key("foo") instead of cfg["foo"].empty()
20090614 04:11:51< Sapient> I'm not sure what you mean about the match_ranges
20090614 04:15:08-!- Ivanovic_ [n=ivanovic@dtmd-4db2d00e.pool.einsundeins.de] has joined #wesnoth-dev
20090614 04:15:40< Soliton> well, for [kill] cfg.has_attribute("x") is already checked but then cfg["x"] is not checked for empty() and simply given to matches_range() which will happily match empty strings.
20090614 04:17:34< Sapient> yeah, there's logic failures like that all over the code
20090614 04:17:45< Sapient> not just matches_range
20090614 04:18:04< Sapient> lots of places.... like in unit_abilities.cpp
20090614 04:18:57< Soliton> well, i don't think the failure is in matches_range() or we could just change it there and not for every caller.
20090614 04:19:33< Sapient> right, sounds like the caller is in error... although I'm not looking at it right now
20090614 04:19:45< Sapient> wesbot: rsapient
20090614 04:19:48< Sapient> lol
20090614 04:20:01< Sapient> wesbot: r30859
20090614 04:20:10< Sapient> wesbot:  log 30859
20090614 04:20:13< wesbot> sapient * r30859 : * SingleUnitFilters can now check for empty keys such as role=$null
20090614 04:20:16< wesbot> URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=30859
20090614 04:21:27< Sapient> I think that revision should be a good example for anyone trying to take it as an EasyCoding task
20090614 04:22:00< Sapient> cya later, have fun.
20090614 04:22:00-!- Sapient [n=sapien-x@wesnoth/developer/sapient] has left #wesnoth-dev []
20090614 04:22:04< Soliton> that commit introduced the problem...
20090614 04:24:58-!- Sapient [n=sapien-x@66.56.60.33] has joined #wesnoth-dev
20090614 04:25:04< Sapient> Soliton: hmmm?
20090614 04:25:42< Soliton> we do not care whether x,y is specified or not if bot are empty we don't want to match.
20090614 04:25:46< Soliton> both*
20090614 04:26:00< Sapient> yes, we do care
20090614 04:26:09< Soliton> i'm sure there are cases where that is different though so in general that surely made sense.
20090614 04:26:22< Soliton> well, i just discussed that we don't.
20090614 04:26:31< Sapient> x="$mistyped_varname"
20090614 04:26:38< Sapient> that should fail
20090614 04:26:44< Soliton> right.
20090614 04:27:02< Sapient> before that revision, it would pass
20090614 04:27:11< Sapient> so I 'fixed' it
20090614 04:27:13< Soliton> if(!cfg_x.empty() || !cfg_y.empty())
20090614 04:27:39< Soliton> if both are empty it does not pass.
20090614 04:28:26< Sapient> if both are empty it passes
20090614 04:28:33< Sapient> because it doesn't check for failure
20090614 04:28:53< Sapient> matbe we are using a different meaning of pass/fail here
20090614 04:29:04< Soliton> looks like.
20090614 04:30:19< Sapient> under the old code, there was no way to check for empty such as role="$empty"
20090614 04:30:49< Sapient> also, it led to really strange behavior when variable names were mistyped but the filter passed for all units
20090614 04:31:12< Sapient> for troubleshooting it is far better to fail when there is a typo in the varname
20090614 04:31:26< Soliton> yeah.
20090614 04:32:08-!- Ivanovic [n=ivanovic@wesnoth/developer/ivanovic] has quit [Read error: 110 (Connection timed out)]
20090614 04:32:09< Sapient> so I think we are in agreement now? good night
20090614 04:32:15 * Sapient zzzs
20090614 04:32:20-!- Sapient [n=sapien-x@wesnoth/developer/sapient] has left #wesnoth-dev []
20090614 04:33:06-!- Ivanovic_ is now known as Ivanovic
20090614 04:35:07< Soliton> right, that commit didn't introduce the problem it just didn't fix it for x,y.
20090614 04:38:48-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090614 04:42:06-!- ilor_ [n=user@wesnoth/developer/ilor] has quit [Read error: 60 (Operation timed out)]
20090614 04:59:00< CIA-77> soliton * r36186 /trunk/src/unit.cpp: do not match on empty x,y keys in [kill] (fixes bug #13685: Dwarvish "Sneak Attack" kills all units on map in UtBS - A Subterranean Struggle)
20090614 05:34:40-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has quit []
20090614 06:09:14-!- wesbot changed the topic of #wesnoth-dev to: 62 bugs, 242 feature requests, 13 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090614 07:38:58-!- ilor_ [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090614 07:46:36-!- ilor [n=user@wesnoth/developer/ilor] has quit [Read error: 60 (Operation timed out)]
20090614 07:51:17-!- Blueblaze [n=nick@c-98-199-143-139.hsd1.tx.comcast.net] has quit [Remote closed the connection]
20090614 07:52:48-!- Blueblaze [n=nick@c-98-199-143-139.hsd1.tx.comcast.net] has joined #wesnoth-dev
20090614 08:00:39-!- maxy [n=maxy@84-74-83-103.dclient.hispeed.ch] has joined #wesnoth-dev
20090614 08:28:21-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 110 (Connection timed out)]
20090614 08:38:07-!- Doppp|EeePC [n=jianxioy@c-67-171-96-240.hsd1.pa.comcast.net] has quit [Read error: 110 (Connection timed out)]
20090614 08:45:03-!- Blueblaze [n=nick@c-98-199-143-139.hsd1.tx.comcast.net] has quit [Remote closed the connection]
20090614 08:55:02-!- Sirp [n=me@wesnoth/developer/dave] has quit ["leaving"]
20090614 08:58:20-!- BenUrban [n=benurban@unaffiliated/benurban] has quit []
20090614 08:59:09-!- BenUrban [n=benurban@68.55.19.224] has joined #wesnoth-dev
20090614 09:04:11-!- mordante [n=mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20090614 09:04:28< mordante> servus
20090614 09:09:14-!- ilor_ [n=user@wesnoth/developer/ilor] has quit [Remote closed the connection]
20090614 09:09:27-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090614 09:22:37-!- stikonas [n=and@ctv-213-164-98-244.vinita.lt] has joined #wesnoth-dev
20090614 09:38:07-!- Michael_Puer [n=michael_@188-23-0-178.adsl.highway.telekom.at] has quit [Read error: 104 (Connection reset by peer)]
20090614 09:46:35-!- Michael_Puer [n=michael_@188-23-98-115.adsl.highway.telekom.at] has joined #wesnoth-dev
20090614 09:52:10-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090614 09:58:10< CIA-77> mordante * r36187 /trunk/src/gui/ (4 files in 2 dirs): 
20090614 09:58:10< CIA-77> Allow to define a maximum width for a window.
20090614 09:58:10< CIA-77> This feature is only used with automatic placement and allows a window
20090614 09:58:10< CIA-77> not to grow beyond a wanted maximum regardless of the window resolution.
20090614 09:58:15< CIA-77> mordante * r36188 /trunk/data/gui/default/window/campaign_dialog.cfg: Set a maximum width for the campaign dialog.
20090614 10:18:33-!- zookeeper [n=l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20090614 11:02:25< Ivanovic> moin
20090614 11:04:50< CIA-77> mordante * r36189 /trunk/data/gui/default/ (widget/panel_title_screen.cfg window/title_screen.cfg): Add a panel for the new title screen.
20090614 11:05:01< mordante> hi Ivanovic 
20090614 11:32:17< stikonas> esr: Delfador Memoirs is broken in trunk. It does not start at all. It complains about scenario8:37, but even after fix it does not start and complains about Oulaw_Rogue.cfg
20090614 11:39:33-!- boucman [n=rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20090614 11:40:41-!- Noyga [n=lame-z@wesnoth/developer/noyga] has joined #wesnoth-dev
20090614 11:44:36-!- loonybot [n=loonybot@79.139.138.177] has joined #wesnoth-dev
20090614 11:45:25-!- loonycyborg [n=sergey@79.139.138.177] has joined #wesnoth-dev
20090614 11:46:06< boucman> morning all
20090614 11:46:47< Ivanovic> hi boucman
20090614 11:46:59< boucman> wesbot: seen Crab_
20090614 11:46:59< wesbot> boucman: The person with the nick Crab_ 12h 38m ago person left:
20090614 11:54:33< mordante> hi boucman 
20090614 11:58:09< CIA-77> mordante * r36190 /trunk/ (4 files in 3 dirs): Implement the tips panel in the new titlescreen.
20090614 12:01:20-!- ettin [n=jorda@wesnoth/developer/ettin] has joined #wesnoth-dev
20090614 12:12:27-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20090614 12:13:57-!- silene [n=plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20090614 12:14:04< silene> hi
20090614 12:14:13< mordante> hi silene 
20090614 12:14:29-!- ettin_ [n=jorda@wesnoth/developer/ettin] has quit [Read error: 110 (Connection timed out)]
20090614 12:14:30< mordante> silene, the campaign dialog has a maximum of 800x600
20090614 12:17:19-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090614 12:19:07-!- anomie3 [n=gare@72.55.205.138] has quit [Read error: 110 (Connection timed out)]
20090614 12:21:02-!- anomie3 [n=gare@72.55.205.138] has joined #wesnoth-dev
20090614 12:26:54-!- Crab_ [n=Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20090614 12:27:11< Crab_> hi 
20090614 12:27:43< Ivanovic> hi Crab_
20090614 12:27:49< Ivanovic> Crab_: boucman was asking wesbot for you
20090614 12:28:10< Crab_> Ivanovic: yes, wesbot has told me about it a second ago :)
20090614 12:28:16< mordante> hi Crab_ 
20090614 12:28:23< Crab_> hi mordante
20090614 12:29:49< boucman> Crab_: afk, just a min
20090614 12:29:55< Crab_> ok
20090614 12:31:17< Ivanovic> everybody: anything speaking against 1.6.3 next sunday?
20090614 12:31:19< mordante> Crab_, regarding r36146 I think g++ was wrong here when it allowed the code ;-)
20090614 12:31:50< mordante> Ivanovic, no
20090614 12:31:58< Ivanovic> and 1.7.1 the week after (though *not* on saturday, there i will be busy...)
20090614 12:31:58< Crab_> mordante: yes, double isn't an integral type
20090614 12:32:58-!- silene1 [n=plouf@ASte-Genev-Bois-152-1-55-10.w82-121.abo.wanadoo.fr] has joined #wesnoth-dev
20090614 12:33:48-!- Ivanovic changed the topic of #wesnoth-dev to: 1.6.3 planned for June 21, string freeze on branches/1.6 | 1.7.1 planned for June 28 | 62 bugs, 242 feature requests, 13 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090614 12:33:48< mordante> nothing against 1.7.1 as well
20090614 12:33:51-!- silene [n=plouf@wesnoth/developer/silene] has quit [Nick collision from services.]
20090614 12:34:01-!- silene1 is now known as silene
20090614 12:35:58< silene> mordante: looks better, thanks
20090614 12:36:11< mordante> you're welcome
20090614 12:39:05-!- ilor_ [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090614 12:47:18< stikonas> Ivanovic: Delfador's Memoirs must be fixed before 1.7.1, this campaign is not working at all now.
20090614 12:47:40< Ivanovic> stikonas: please post a bug report
20090614 12:47:45< Ivanovic> this way it will not be forgotten
20090614 12:48:54< stikonas> Ivanovic: what severity I should select?
20090614 12:48:59< stikonas> 4- important?
20090614 12:49:43< Ivanovic> yes
20090614 12:49:44< silene> if it doesn't even start, "blocker"
20090614 12:49:56< mordante> I also tend to agree with blocker
20090614 12:50:01< stikonas> ok
20090614 12:50:07< Ivanovic> right, since the campaign not working will block the release of 1.7.1
20090614 12:53:55< stikonas> please assingn to esr: https://gna.org/bugs/index.php?13690
20090614 12:54:41< stikonas> I've also filled this minor bug: https://gna.org/bugs/index.php?13689
20090614 12:56:34-!- ilor [n=user@wesnoth/developer/ilor] has quit [Read error: 110 (Connection timed out)]
20090614 12:58:58< Ivanovic> stikonas: not sure if it should be esr or fabi/fendrin
20090614 13:01:16-!- ilor_ [n=user@wesnoth/developer/ilor] has quit [Remote closed the connection]
20090614 13:01:29-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090614 13:01:31< stikonas> Ivanovic: git blame show that the line my patch changes was written by esr
20090614 13:02:01< Ivanovic> ehm, sorry to tell you, but DM starts here
20090614 13:02:24< ilor> mordante: as a status update -- working on server room handling atm, should have solid commit-worthy material later today
20090614 13:02:40< stikonas> Ivanovic: strange...
20090614 13:02:47< Ivanovic> that is using r36190 
20090614 13:03:01< stikonas> I can try to recompile and reinstall
20090614 13:03:14< Ivanovic> please do so
20090614 13:03:23< Ivanovic> and before installing, please remove the old files
20090614 13:03:28< stikonas> sure
20090614 13:03:29< ilor> mordante: the idea currently is to avoid overloading the already-large server.?pp files, so I'm delegating the lobby/rooms processing to a general "lobby" object that will own and manage rooms and their logic
20090614 13:08:30< stikonas> Ivanovic: please mark bug 13690 as invalid. Probably cmake has messed up and hasn't updated all files.
20090614 13:11:54< boucman> Crab_: back
20090614 13:12:21< boucman> Crab_: I had nothing special to discuss, it's just that I had missed you the whole week and was wondering if you were away...
20090614 13:14:16< Crab_> no, I was not 'fully away', and I've done a part of 'combat phase' transition to composite ai, but a lot of time was diverted towards my university diploma (which is due for tomorrow). After that, no more university stuff to do - only the graduation ceremony will remain )
20090614 13:14:50< boucman> ok, no problem
20090614 13:14:53< boucman> just wondering
20090614 13:26:32< Crab_> boucman: also, I've met DK here on irc, so I've got him to ack the formula file renames. plus, after his exams is over, we'll discuss our work plans in detail.
20090614 13:26:51< boucman> sounds good
20090614 13:26:59< boucman> whea are his exams ?
20090614 13:28:14< Crab_> some were this week, and one is around next tuesday
20090614 13:28:24< boucman> k
20090614 13:28:48< Crab_> where next tuesday = 16 Jun
20090614 13:38:18< boucman> k
20090614 13:38:20< boucman> even better
20090614 13:47:39-!- mjs-de [n=mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20090614 13:52:00< mordante> good to hear ilor
20090614 13:52:30< mordante> I added a multi_page which can be used for a tab control
20090614 13:52:57< mordante> it only misses the tabs at the moment, but handles multiple pages
20090614 13:53:06< mordante> you can see it used in the campaign dialog
20090614 13:53:40< mordante> any idea around what time you'll be able to commit?
20090614 13:59:22< ilor> mordante: currently doing some cleanups to make the code compile and work in one-room mode so it can be tested, need to go buy some food too so I'd say 1-2hrs
20090614 14:01:38< mordante> good, then I still be around :-)
20090614 14:08:20< CIA-77> silene * r36191 /trunk/src/ai/default/attack.cpp: 
20090614 14:08:20< CIA-77> Reverted revision 30319 to prevent some overestimation of poisoning units, since
20090614 14:08:20< CIA-77> they do not necessarily poison, they do not necessarily have 70% cth, and they
20090614 14:08:20< CIA-77> certainly do not inflict 8 dmg on their next strike. Cleaned the code along the
20090614 14:08:20< CIA-77> way.
20090614 14:11:01-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has quit ["ChatZilla 0.9.84 [Firefox 3.0.11/2009060215]"]
20090614 14:38:24-!- ilor_ [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
--- Log opened Sun Jun 14 15:18:11 2009
20090614 15:18:20-!- lobby [n=wesnoth@wesnoth/bot/lobby] has joined #wesnoth-dev
20090614 15:18:20-!- Topic for #wesnoth-dev: 1.6.3 planned for June 21, string freeze on branches/1.6 | 1.7.1 planned for June 28 | 62 bugs, 242 feature requests, 13 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090614 15:18:20-!- Topic set by Ivanovic [] [Sun Jun 14 12:33:47 2009]
20090614 15:18:20[Users #wesnoth-dev]
20090614 15:18:20[ ABCD        ] [ CIA-77  ] [ ettin      ] [ maxy        ] [ Rhonda      ] [ Turuk    ] 
20090614 15:18:20[ AI0867      ] [ cjhopman] [ grzywacz   ] [ Mellar      ] [ shadowmaster] [ wesbot   ] 
20090614 15:18:20[ Alesis-Novik] [ corn__  ] [ ilor_      ] [ Michael_Puer] [ shikadibot  ] [ yann     ] 
20090614 15:18:20[ AnMaster    ] [ Crab_   ] [ isaac      ] [ mjs-de      ] [ silene      ] [ zookeeper] 
20090614 15:18:20[ anomie3     ] [ dfranke ] [ Ivanovic   ] [ mordante    ] [ Smar        ] 
20090614 15:18:20[ BenUrban    ] [ Doppp   ] [ lobby      ] [ nital       ] [ Soliton     ] 
20090614 15:18:20[ boucman     ] [ erl     ] [ loonybot   ] [ Noyga       ] [ stikonas    ] 
20090614 15:18:20[ busfahrer   ] [ esr     ] [ loonycyborg] [ Polarina    ] [ Tigge       ] 
20090614 15:18:20-!- Irssi: #wesnoth-dev: Total of 44 nicks [0 ops, 0 halfops, 0 voices, 44 normal]
20090614 15:18:30-!- Channel #wesnoth-dev created Tue Jan 27 06:28:41 2009
20090614 15:19:01< boucman> MP server is almost empty... did it reboot ?
20090614 15:19:14-!- Irssi: Join to #wesnoth-dev was synced in 62 secs
20090614 15:19:18< Soliton> the whole server rebooted.
20090614 15:19:52< boucman> k, thx
20090614 15:20:57< Crab_> so, boucman / silene, can you suggest a good alternative word for [ai] tag name ?
20090614 15:21:10< boucman> not really
20090614 15:21:22< boucman> ai_config :)
20090614 15:21:54-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090614 15:23:02< silene> reusing [ai] is fine to me; what kind of syntax changes do you want to do?
20090614 15:26:21< Crab_> silene: I want to store everything inside a single top-level tag. something like http://wesnoth.pastebin.com/m597dca65
20090614 15:27:28< silene> Crab_: i don't think it is a good idea to store everything; you want a syntax that allows to change a few things, so that it can be used inside events for instance; you con't the scenario designer to duplicate everything in each event
20090614 15:28:08< boucman> silene: IIUC this is a savegame only syntax change
20090614 15:28:17< boucman> not for umc
20090614 15:28:27-!- ilor_ [n=user@wesnoth/developer/ilor] has quit [Read error: 110 (Connection timed out)]
20090614 15:28:33< Crab_> boucman: umc/scenarios too. but the old way will be still readable
20090614 15:29:20< silene> boucman: what is the reason for having a different syntax for savegames and scenarios? i think it should be uniform
20090614 15:29:38< boucman> Crab_: ok, I didn't undertand that
20090614 15:29:40< Crab_> silene: compare that with old (current) syntax: http://wesnoth.pastebin.com/m41dc235c
20090614 15:30:21< Crab_> note: both 'single top-level tag' and current syntax go inside [side]
20090614 15:30:39< silene> Crab_: the new syntax is +33% longer, i definitely prefer the old syntax ;-)
20090614 15:31:31< Crab_> silene: np :) here's the new syntax with some syntax sugar removed: http://wesnoth.pastebin.com/mc04cc85
20090614 15:33:08< silene> that's good; but back to my original point, what happens if a designer were to put an [ai] tag in an event now; would it be appended to the ai with the same id? would it override it?
20090614 15:34:44< silene> speaking of which, please just us "id" instead of "ai_identifier", so that is consistent with existing wml usage
20090614 15:37:06< Crab_> silene: there's no such possibility to make something useful from putting ai tag in an event now. (and current ai parameter handling is a mess). The only thing that it is possible now is to use [modify_side] to reload a new ai from specified file.
20090614 15:37:40< silene> Crab_: i know it isn't possible, but i want it to be possible, and your syntax changes are a good opportunity
20090614 15:41:51< Crab_> silene: does WML support templates to allow something like http://wesnoth.pastebin.com/m584e766a ?
20090614 15:42:28< boucman> not as such
20090614 15:42:31< Crab_> if there's way to do this, I can write the ai in a simular form, and document the 'extension points' of the default ai.
20090614 15:42:39< silene> i don't understand, isn't that just a macro?
20090614 15:42:44< boucman> we usually do that using [+xxx] and macros
20090614 15:42:48< Crab_> basically, it's a macro with parameters
20090614 15:42:52< Crab_> and parameters can be long
20090614 15:43:24< boucman> we do have macros with parameters, but I don't know if we had whole chunks of wml in macro parameters
20090614 15:43:45< silene> wml macros don't care about the length of parameters, it works fine
20090614 15:43:58< Crab_> silene: so, ' whole chunks of wml in macro parameters' is possible ?
20090614 15:44:08< silene> yes
20090614 15:45:36-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit []
20090614 15:45:37< Crab_> good. then, we can just say: "hey, you can write an ai from pieces from scratch, but, if you want to, we have several ai macros that you can use, and here's the documentation on the parameter wml chunks you can supply to them"
20090614 15:46:04-!- ^Noyga^ [n=lame-z@AVelizy-151-1-7-169.w82-120.abo.wanadoo.fr] has joined #wesnoth-dev
20090614 15:46:57< Crab_> then the umc authors will not depend on the actual implementation of the said [ai] macros
20090614 15:47:03-!- Noyga [n=lame-z@wesnoth/developer/noyga] has quit [Nick collision from services.]
20090614 15:47:13-!- ^Noyga^ is now known as Noyga
20090614 15:50:22< Crab_> and regarding in-flight modification of ai (in events, for example), that can be made possible with a tag like [modify_ai]. but this is limited approach, since, for example, formula ai likes to init all its stuff in advance, and it can be time-costly to re-init everything on each config change.
20090614 15:50:49-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20090614 15:50:52< Crab_> so, it is needed to limit and document the things that can be changed by events
20090614 15:58:35< ilor> hmm, were there any start-of-game crashes fixed recently?
20090614 16:02:20< silene> Crab_: that's a defect of formulaai itself, it shouldn't impact the decisions concerning the ai in general; if a designer wants the ai units to flee once the bomb is set up, it should be made possible
20090614 16:03:28< Crab_> silene: that's not the defect of fai. it's just the fact that config is a 'slow-to-access' representation compared with C++ data structures. formula ai is just an extreme case (parsed-formula vs formula-as-text).
20090614 16:03:37< boucman> silene: nope it's more of a design choice : parsing is slow, so we do it as early and as little as possible
20090614 16:03:57< Crab_> silene: we can *allow* to change the ai in flight, but we need to mark 'fast' changes vs 'slow' changes
20090614 16:04:29< boucman> we could drop the whole c++ image on ai-modifying event, but I'd rather have Crab_ focus on the fundamentals rather than being sidetracked on the formula implementation
20090614 16:04:58< Crab_> boucman: I'm thinking more about 'documenting' "this is fast to change" vs "this is slow to change as it requires ai full reinit"
20090614 16:07:02< silene> boucman: i never said we should parse the ai whenever it is about to consider a unit; but we should at least parse it when the wml explicitely asks for it to be modified; i would consider a fundamental defect if it wasn't possible
20090614 16:07:33< boucman> silene: no it would be "a nice feature to have" not "a fundamental deffect"
20090614 16:07:39< boucman> but yes, it would be nice to have
20090614 16:08:23< boucman> the thing is Crab_ already have a lot to do, and I don't want to have him sidetracked, so let's try to focus on the fundamental (a testbed for AI) rather than focusing on a specific AI 
20090614 16:08:42< silene> i stand by my point, if it isn't possible to modify the ai on the fly, i consider Crab-'s design fundamentlly broken
20090614 16:08:49< boucman> lot can already be done through WML events, so we need to organize intelligently
20090614 16:09:05< Crab_> silene:  to allow this to happen, we need only to patch formula_ai::on_create() to make it 're-enterable' - basically, make it clear some of the old stuff before initialization.
20090614 16:09:18< boucman> Crab_'s design is not involved here
20090614 16:09:31< boucman> crab is working on the meta-ai so to speak
20090614 16:09:40< boucman> how different ais interract with each other
20090614 16:09:48< boucman> not how configurable a particular AI is
20090614 16:10:14< boucman> and again, silene it's a matter of priority, as long as we can't test AI properly, everything else is moot
20090614 16:10:40< silene> boucman: this whole discussion is about how to specify an ai in wml; this is definitely something Crab is currently designing
20090614 16:10:57< boucman> indeed
20090614 16:11:08< boucman> how to specify it
20090614 16:11:13< boucman> not how to change it mid-game 
20090614 16:11:30< boucman> you've introduced that mid discussion, and we told you  our priorites were arranged differently
20090614 16:13:01< ilor> mordante: I'm getting a segfault in gui2 tcanvas::blit calling blur_surface
20090614 16:13:49< Crab_> silene: 'changing parts of the ai' in mid game will be possible, but, it requires mostly changes in individual ai implementations, not changes in architecture. Note that 'whole ai' can be hot-redeployed even now.
20090614 16:13:55< mordante> ilor, can you paste a backtrace?
20090614 16:14:22< ilor> mordante: http://codepad.org/RjNYGoPh
20090614 16:14:49< ilor> happens after an addon finishes downloading, but I got a very similar crash when starting a mp gam with trunk
20090614 16:15:22< silene> Crab_: my point is not about the implementation, i never spole about implementation; i'm saying that if you modify the wml syntax, you choose a syntax that makes  sense both at start of scenario and in mid-scenario; it doesn't make change to have two different syntaxes to set an ai
20090614 16:15:49< Crab_> silene: yes
20090614 16:16:00< Crab_> silene: that's why I want to have ai in a single wml tag.
20090614 16:16:26< Crab_> to allow 'serialization' of ai to config, and 'deserialization of ai from config'
20090614 16:16:59< Crab_> for example, the ai can have some things that we need to keep persistent. so, to 'add a config snippet to ai in mid-game' may involve (in the hard case) 1) serialize the ai to config 2) add a snippet to that config 3) replace side ai with the new ai deserialized from that config.
20090614 16:17:54< boucman> Crab_: and I guess it can't be done in an AI-neutral way because we don't know anything about stuff the AI may store internally
20090614 16:18:11< boucman> so to do it in an AI neutral way, we would be redeploying the AI, not modifying it
20090614 16:18:17< boucman> even for small changes
20090614 16:18:27< boucman> (except changes of AI context of course)
20090614 16:19:01< Crab_> boucman: yes. some of the small changes can be done the easy way if we know the implementation.
20090614 16:19:27< Crab_> boucman: for example, if we add some ai parameter handling to ai::readonly_context
20090614 16:19:32< boucman> "if we know the implementation" does mean "not AI neutral" in this case ?
20090614 16:20:09< Crab_> boucman: in that case, "if we know the implementation of a readonly_context" 
20090614 16:20:23< Crab_> boucman: and we'll need to declare this parameter as 'hot-redeployable'
20090614 16:20:36< Crab_> boucman: to make sure each ai author *knows* that it can change in-game.
20090614 16:21:03< Crab_> boucman: for example, imagine that fai is parsing a formula which evaluates to ' "agression parameter"+1 '
20090614 16:21:38< boucman> ok
20090614 16:21:55< Crab_> boucman: if current aggression is 4, then, if fai parsed that formula, evaluated it to 5, and stored that 5, then it won't notice the change to aggression parameter without full redeployment
20090614 16:22:10-!- crimson_penguin [n=ben@64.201.60.211] has joined #wesnoth-dev
20090614 16:22:27< boucman> ok, I see
20090614 16:22:41< boucman> I never thought that AI would consider read-only parameters as const...
20090614 16:22:56< Crab_> boucman: on the other hand, if we said 'hey, this class of parameters can change in-game", then fai will have to store a formula 'aggression+1' and evaluate it each time or on 'ai changed' event.
20090614 16:23:08< boucman> maybe it should be the other way round, having a "static ai context" for data that won't change
20090614 16:23:23< Crab_> boucman: yes, we can add such a thing
20090614 16:23:34< Crab_> between side_context and readonly_context
20090614 16:25:47< Crab_> I'm thinking of a different approach
20090614 16:26:07< boucman> Crab_: yes, i'm not sure if it's worth the trouble and/or if it can be added later
20090614 16:26:39< Crab_> I'm thinking about 'adding ai aspects' - small functions which return a volatile value.
20090614 16:27:36< Crab_> (to explicitly mark the data that can change in-game)
20090614 16:28:05< mordante> ilor, can you post the local variables for the function?
20090614 16:28:19< ilor> mordante: just a sec..
20090614 16:28:21< mordante> I guess some stupid off by x error
20090614 16:28:35< boucman> Crab_: just to be clear: we're talking of future plans here, i'd like to have the "having composite AI completely working" as the next step
20090614 16:29:31< Crab_> boucman: current step is 'make composite_ai::candidate_action implementation of default ai'
20090614 16:29:50< boucman> yes, that's what I meant :)
20090614 16:29:58< Crab_> and, it already requires that 'volatile variables'
20090614 16:30:08< boucman> oh, in that case...
20090614 16:30:19< Crab_> i'm talking about move maps
20090614 16:30:21< boucman> why does it needs them ?
20090614 16:30:59-!- Sirp [n=me@pool-173-74-23-130.dllstx.fios.verizon.net] has joined #wesnoth-dev
20090614 16:31:37< ilor> mordante: http://codepad.org/CSGA48lu
20090614 16:31:44< Crab_> boucman: see http://wesnoth.pastebin.com/m1a153a8a
20090614 16:32:07< Crab_> note that "get_srcdst(), get_dstsrc(), get_enemy_srcdst(), get_enemy_dstsrc()" part
20090614 16:32:40< boucman> ok, I understand
20090614 16:33:07< Crab_> imagine an rca loop having both ai and fai candidate actions
20090614 16:33:27< Crab_> it makes little sense to calculate move maps once per candidate action
20090614 16:33:40< Crab_> and, it makes little sense to calculate move maps once per ai engine
20090614 16:34:20< Crab_> so, it can be done either as part of rca_context, either as a thing which is quite similar to an aspect
20090614 16:34:50< Crab_> ai_default recalculated move maps at start of do_moves
20090614 16:35:41< silene> more precisely, it recalculated them after each move
20090614 16:36:40< Crab_> silene: before each move, at start of ai_default::do_move
20090614 16:37:22< Crab_> formula_ai used a better approach - boolean validity marker move_maps_valid_
20090614 16:37:29< silene> yes, what i meant to say is that do_move is called after each move
20090614 16:37:45< Crab_> ok
20090614 16:38:36-!- ilor_ [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090614 16:39:07< Crab_> so, I'm thinking about something like: function get_srcdst() + validity marker srcdst_valid + event observer which invalidates srcdst when needed.
20090614 16:40:13< boucman> srcdst_valid would be hidden to the users of get_srcdst() iiuc
20090614 16:40:20< Crab_> boucman: yes
20090614 16:40:31< boucman> ok, that approch makes sense, then...
20090614 16:41:12< silene> Crab_: it's a bit complicated; why not just store them into variables and recompute the variables when a unit moves? it will avoid all the overhead
20090614 16:41:33< Crab_> silene: what if we don't need to use the value after the unit moved ?
20090614 16:41:59< Crab_> silene: having a validity marker gives us on-demand initialization
20090614 16:42:17< silene> Crab_: when does that happen? only after the very last unit moved? then it is negligible
20090614 16:42:58< silene> (what i mean is that get_srcdst may be called millions of times per ai turn)
20090614 16:43:34-!- ilor [n=user@wesnoth/developer/ilor] has quit [Read error: 60 (Operation timed out)]
20090614 16:43:56< boucman> silene: for complicated data like reach-map, avoiding the last computation might be worth it
20090614 16:44:48< mordante> ilor_, can you set a breakpoint at line 1081 and restart wesnoth?
20090614 16:45:08< mordante> and reproduce the problem, I'd like to test some things at that line
20090614 16:45:11< ilor_> mordante: sure
20090614 16:46:06< ilor_> the breakpoint is hit a couple of times before the addon dialog, titlescreen I guess
20090614 16:46:12-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090614 16:46:20< ilor_> mordante: ok, I got it
20090614 16:46:48< Crab_> silene: yes, with srcdst maps you're right. but imagine a volatile parameter which can be called only once per few turns. recalculating it after each move is not good.
20090614 16:47:21< mordante> ilor_, I like the values of lock.pixels(),  pixel_offset and y * surf->w
20090614 16:47:32< silene> Crab_: i'm only talking about srcdst here
20090614 16:47:38< mordante> then after stepping 1 line the value of p
20090614 16:48:29< Crab_> silene: srcdst is just one of a 'volatile parameters'.   those ai parameters which can change in-game can be implemented in a similar way: getter function( with parameter ) + hidden content-addressable table of validity markers + event observer which sets the invalidation marker when needed.
20090614 16:48:58< silene> do they all need to be implemented the same way?
20090614 16:49:00< Crab_> silene: no
20090614 16:49:48< ilor_> mordante: http://codepad.org/IOW28hh0
20090614 16:50:40< mordante> well the value of surf->w is not important, x times0 is zero
20090614 16:50:42< mordante> ;-)
20090614 16:51:31< ilor_> mordante: last time it crashed at y=41 or so
20090614 16:53:38< mordante> can you now let it crash and give the value of p?
20090614 16:54:27< ilor_> amusingly, didn't crash this time
20090614 16:55:08< mordante> :-(
20090614 16:56:35< Crab_> silene: also note that not all candidate actions need all those volatile parameters. for example, current 'move_leader_to_keep' implementation uses only enemy_dstsrc
20090614 16:57:08< ilor_> it didn't crash, so I downloaded another addon, and it a) got some slight visual corruption it the bottom 1/5 of the window
20090614 16:57:24< ilor_> b) crashed with a bad_alloc sourcing in some std::string 
20090614 16:59:24< silene> Crab_: i'm not sure that's a good example; it would make sense for move_leader_to_keep to choose the keep with respect to the best support too
20090614 16:59:28< mordante> hmm guess some memory gets silently corruptes
20090614 16:59:32< mordante> corrupted*
20090614 16:59:46< Crab_> silene: if we add a certain volatile parameter which is used infrequently, it may happen (depending on the candidate actions present in the ai) that it's recalculation after each move will be useless.
20090614 16:59:52< ilor_> looks like it, results seem to differ depending on which addon I choose
20090614 17:00:47< silene> Crab_: again i'm only talking about the srcdst maps; for those other parameters, it may indeed be interesting to have validity markers
20090614 17:00:48< boucman> silene: i think that your point shows that we (as in the core AI engine) can't know what variable will be used or not
20090614 17:00:58< boucman> the next implementation might do as you said
20090614 17:01:24< Crab_> silene: you think that for parameters like srcdst,  check   'if (!srcdst_valid) { recalculate_srcdst(); },"  will be a performance loss ?
20090614 17:01:25< mordante> ilor_, guess I can try to run Wesnoth under valgrid and see where things to wrong, but that will take some time
20090614 17:01:28-!- Sirp [n=me@pool-173-74-23-130.dllstx.fios.verizon.net] has quit [Read error: 104 (Connection reset by peer)]
20090614 17:01:49< ilor_> mordante: I'm getting the crash under windows atm, haven't tried on *nix
20090614 17:02:23< ilor_> mordante: I take it you werent able to reproduce?
20090614 17:03:00< mordante> what's the exact procedure?
20090614 17:03:01< silene> Crab_: i don't know; but what i know is that the srcdst maps aren't recalculated more times than the number of ai units, but they are accessed a number of times that is just huge
20090614 17:03:17< ilor_> mordante: start trunk, open addon dialog, download an addon
20090614 17:04:03< Crab_> silene: ok. it is np to do srcdst/dstsrc/enemy_srcdst/enemy_dstsrc maps without validity markers.
20090614 17:05:08-!- stikonas [n=and@ctv-213-164-98-244.vinita.lt] has joined #wesnoth-dev
20090614 17:05:58< mordante> works for me
20090614 17:06:12 * mordante notices we have insanely large addons...
20090614 17:06:39< ilor_> mordante: maybe try a bunch of different addons?
20090614 17:06:51< mordante> I already did
20090614 17:07:07< ilor_> something makes me think I won't get the crash in linux either
20090614 17:07:34< ilor_> let me try a mingw build
20090614 17:09:45< mordante> does this bug cause problems for you gsoc work?
20090614 17:10:57< ilor_> mordante: it crashed for me in the same place when I started a MP game
20090614 17:12:33-!- Sirp [n=me@wesnoth/developer/dave] has joined #wesnoth-dev
20090614 17:12:33< ilor_> mordante: easier to reproduce in the addon, but you can try launching two instances of wesnoth, connecting to trunk server and tryning to start a game
20090614 17:13:40< ilor_> mordante: for me it crashes pretty much right after the map etc is shown, then the other player gets the "other-player-dropped" dialog, and choosing, e.g., replace-with-local makes this wesnoth instance crash as well
20090614 17:15:06< mordante> here's an untested work-around
20090614 17:20:44< ilor_> mordante: where?
20090614 17:23:31< mordante> ilor_, sorry here http://paste.debian.net/39070
20090614 17:45:28-!- giusef [n=giusef@unaffiliated/giusef] has joined #wesnoth-dev
20090614 17:49:19< ilor_> mordante: the change seems to prevent the crash in the addon dialog
20090614 17:49:58< ilor_> mordante: sorry for the delay, mingw build took longer than expected (and I've ran out of hdd space and had to do it again)
20090614 17:50:21-!- Michael_Puer2 [n=michael_@188-23-15-8.adsl.highway.telekom.at] has joined #wesnoth-dev
20090614 17:50:40< ilor_> mordante: anyway, mingw build is less twitchy but managed to crash once already, again the change prevents it
20090614 17:56:31< CIA-77> silene * r36194 /trunk/data/gui/default/widget/window_default.cfg: Typo.
20090614 17:56:36-!- silene [n=plouf@wesnoth/developer/silene] has quit [Read error: 60 (Operation timed out)]
20090614 17:57:50-!- Michael_Puer [n=michael_@188-23-98-115.adsl.highway.telekom.at] has quit [Read error: 110 (Connection timed out)]
20090614 18:05:52< mordante> ilor_, yes the work-around causes the function not to be called ;-)
20090614 18:06:07< ilor_> mordante: it makes me happy anyway :P
20090614 18:06:50< mordante> I expected that, but that's why it's a work-around and not a fix :-P
20090614 18:09:02< CIA-77> ilor * r36195 /trunk/src/gui/widgets/scrollbar_panel.hpp: fix a struct/class warning in msvc
20090614 18:09:13-!- wesbot changed the topic of #wesnoth-dev to: 1.6.3 planned for June 21, string freeze on branches/1.6 | 1.7.1 planned for June 28 | 61 bugs, 242 feature requests, 13 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090614 18:09:19< ilor_> mordante: anyway, I'm commiting a WIP version of the server room handling, it doesn't allow anything new yet, but you should be able to see the general idea
20090614 18:09:40< CIA-77> ilor * r36196 /trunk/src/ (7 files in 3 dirs): new lobby gui stub files
20090614 18:10:00< CIA-77> ilor * r36197 /trunk/src/ (11 files in 2 dirs): initial framework for server-side lobby room support
20090614 18:13:43< CIA-77> mordante * r36198 /trunk/src/gui/widgets/window.hpp: Remove an unused exception class.
20090614 18:13:58< CIA-77> mordante * r36199 /trunk/src/gui/auxiliary/window_builder.hpp: Remove an unneeded include.
20090614 18:14:00< mordante> ok will take a look in about 20 minutes, off for diner now
20090614 18:14:03< CIA-77> mordante * r36200 /trunk/src/gui/widgets/generator.hpp: Reduce the header dependencies.
20090614 18:14:07< CIA-77> mordante * r36201 /trunk/src/gui/widgets/ (control.hpp settings.hpp): Reduce the header dependencies.
20090614 18:14:11< CIA-77> mordante * r36202 /trunk/src/gui/widgets/ (event_handler.cpp event_handler.hpp): Reduce the header dependencies.
20090614 18:14:20< CIA-77> mordante * r36203 /trunk/src/ (ai/game_info.hpp tod_manager.cpp unit.hpp unit_animation.cpp): educe the header dependencies.
20090614 18:14:54-!- silene [n=plouf@ASte-Genev-Bois-152-1-103-192.w86-203.abo.wanadoo.fr] has joined #wesnoth-dev
20090614 18:18:46< grzywacz> mordante, have you been working on the campaigns dialog recently?
20090614 18:24:29< Soliton> yes, it's gui2 now.
20090614 18:26:33< grzywacz> Ok, cool.
20090614 18:34:35< mordante> grzywacz, yes
20090614 18:35:05< CIA-77> mordante * r36204 /trunk/src/server/room.cpp: Fix a few compiler warnings.
20090614 18:35:19< grzywacz> mordante, I'd like to add something to it as an experiment... :)
20090614 18:36:37< mordante> grzywacz, what do you want to add?
20090614 18:36:58< grzywacz> mordante, a small indicator for campaigns that have been won 
20090614 18:37:06< grzywacz> s/won/completed/
20090614 18:37:16< mordante> sounds cool :-)
20090614 18:38:04< mordante> I'm ready with that part of the code at the moment so feel free to play with it
20090614 18:38:27< grzywacz> Ok, that's why I was asking ;)
20090614 18:39:02< mordante> I thought so ;-)
20090614 18:39:08-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090614 18:39:36-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090614 18:41:45-!- AI0867 [n=ai@wesnoth/developer/ai0867] has quit [Remote closed the connection]
20090614 18:41:45-!- shadowmaster [n=ignacio@wesnoth/developer/shadowmaster] has quit [Remote closed the connection]
20090614 18:41:45-!- shikadibot [n=robored@wesnoth/umc-dev/bot/shikadibot] has quit [Remote closed the connection]
20090614 18:42:18-!- AI0867 [n=ai@wesnoth/developer/ai0867] has joined #wesnoth-dev
20090614 18:43:36-!- AI0867 [n=ai@wesnoth/developer/ai0867] has quit [Client Quit]
20090614 18:43:38-!- AI0867 [n=ai@wesnoth/developer/ai0867] has joined #wesnoth-dev
20090614 18:44:19-!- shikadibot [n=robored@wesnoth/umc-dev/bot/shikadibot] has joined #wesnoth-dev
20090614 18:44:27< mordante> ilor_, FYI boost::noncopyable has its own header
20090614 18:45:57< CIA-77> soliton * r36205 /branches/1.6/src/server/server.cpp: don't catch bad_alloc and unknown errors so we get a core dump
20090614 18:46:13< CIA-77> soliton * r36206 /trunk/src/server/server.cpp: don't catch bad_alloc and unknown errors so we get a core dump
20090614 18:46:16< mordante> can't bool room_manager::in_lobby(network::connection player) use a const ref?
20090614 18:47:12< ilor> mordante: a const ref to what? network::connection is a int typedef so not muich reason to pass by ref
20090614 18:47:20< Soliton> network::connection is just an int
20090614 18:47:41< mordante> ah ok, that explains the other copies as well
20090614 18:48:46< mordante> (and I just prefer to use const at variables I don't intent to modify, but that's my personal preference)
20090614 18:50:31-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Client Quit]
20090614 18:50:55-!- ABCD [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090614 18:54:11< mordante> ilor, can it be you forgot to add a file, both grzywacz and I can no longer start Wesnoth
20090614 18:55:12< ilor> mordante: uh. yeah.
20090614 18:55:37< CIA-77> ilor * r36207 /trunk/data/gui/default/window/lobby_main.cfg: missing cfg file
20090614 18:55:49< ilor> mordante, grzywacz: there, sorry for that
20090614 18:56:20-!- ilor_ [n=user@wesnoth/developer/ilor] has quit [Read error: 113 (No route to host)]
20090614 18:56:28< grzywacz> Much better now :)
20090614 18:56:50< mordante> for me to, thanks ilor 
20090614 18:56:54< mordante> too*
20090614 19:00:19< mordante> ilor, have to go now, will look a bit more at it tomorrow
20090614 19:01:03< ilor> ok seeya
20090614 19:03:20-!- mordante [n=mordante@wesnoth/developer/mordante] has quit ["Leaving"]
20090614 19:05:44-!- ardesh [n=ardesh@port-92-206-176-159.dynamic.qsc.de] has joined #wesnoth-dev
20090614 19:13:54-!- Sirp [n=me@wesnoth/developer/dave] has quit ["leaving"]
20090614 19:18:42-!- ancestral [n=ancestra@32.134.134.180] has joined #wesnoth-dev
20090614 19:22:12-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090614 19:25:02< grzywacz> Any idea what ttext_history in gui/widgets/text_box.* is for?
20090614 19:27:14-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit []
20090614 19:33:08-!- ancestral [n=ancestra@32.134.134.180] has quit ["Get Colloquy for iPhone! http://mobile.colloquy.info/"]
20090614 19:33:13-!- loonycyborg [n=sergey@wesnoth/developer/loonycyborg] has quit ["KVIrc 3.4.2 Shiny http://www.kvirc.net/"]
20090614 19:33:40-!- loonybot [n=loonybot@wesnoth/bot/loonybot] has quit [Remote closed the connection]
20090614 19:38:54< cjhopman> grzywacz: i believe it is for history of typed lines... that is, so that up and down arrows will cycle through the lines you've typed in the text box
20090614 19:39:14-!- ilor_ [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090614 19:39:45< Soliton> except it's not up and down anymore but some obscure binding no one can remember.
20090614 19:49:40-!- cjhopman [n=chris@wesnoth/developer/cjhopman] has quit [Remote closed the connection]
--- Log closed Sun Jun 14 19:53:45 2009
--- Log opened Sun Jun 14 19:53:45 2009
20090614 19:55:00-!- cristi_ceata [n=morpheus@92.80.129.50] has joined #wesnoth-dev
20090614 19:55:02< CIA-77> ivanovic * r36208 /trunk/po/ (26 files in 13 dirs): updated Serbian translation
20090614 19:55:05< CIA-77> ivanovic * r36209 /branches/1.6/po/ (18 files in 9 dirs): updated Serbian translation
20090614 19:55:14-!- cristi_ceata [n=morpheus@92.80.129.50] has left #wesnoth-dev []
20090614 19:56:54-!- Blueblaze [n=nick@c-98-199-143-139.hsd1.tx.comcast.net] has joined #wesnoth-dev
20090614 19:57:18-!- ilor [n=user@wesnoth/developer/ilor] has quit [Read error: 113 (No route to host)]
--- Log closed Sun Jun 14 20:00:22 2009
--- Log opened Sun Jun 14 20:00:22 2009
--- Log closed Sun Jun 14 20:03:15 2009
--- Log opened Sun Jun 14 20:03:15 2009
--- Log closed Sun Jun 14 20:04:58 2009
--- Log opened Sun Jun 14 20:04:58 2009
20090614 20:05:49-!- loonybot [n=loonybot@79.139.138.177] has joined #wesnoth-dev
20090614 20:06:36-!- loonycyborg [n=sergey@79.139.138.177] has joined #wesnoth-dev
--- Log closed Sun Jun 14 20:06:43 2009
--- Log opened Sun Jun 14 20:06:43 2009
20090614 20:09:02-!- giusef [n=giusef@unaffiliated/giusef] has quit ["exit (-1);"]
20090614 20:12:21-!- noy [n=Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090614 20:14:44-!- Sirp [n=me@wesnoth/developer/dave] has joined #wesnoth-dev
--- Log closed Sun Jun 14 20:21:57 2009
--- Log opened Sun Jun 14 20:21:57 2009
--- Log closed Sun Jun 14 20:25:49 2009
--- Log opened Sun Jun 14 20:25:49 2009
--- Log closed Sun Jun 14 20:26:24 2009
--- Log opened Sun Jun 14 20:26:24 2009
--- Log closed Sun Jun 14 20:31:57 2009
--- Log opened Sun Jun 14 20:31:57 2009
--- Log closed Sun Jun 14 20:33:00 2009
--- Log opened Sun Jun 14 20:33:00 2009
20090614 20:43:50-!- EdB [n=edb@79.95.12.80] has joined #wesnoth-dev
20090614 21:06:43-!- ancestral [n=ancestra@32.134.80.42] has joined #wesnoth-dev
20090614 21:08:09-!- norbert_ [n=norbert_@82-171-70-54.ip.telfort.nl] has joined #wesnoth-dev
20090614 21:08:26< norbert_> hi all, maybe someone here knows this: are the ulfserkers removed from Colosseum 2.0.8?
20090614 21:09:16< noy> norbert_: its not a mainline concern 
20090614 21:09:32< norbert_> yeah I know, but maybe the dev is/was here
20090614 21:10:59-!- happygrue [n=George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20090614 21:12:14-!- Doppp [n=aassdf@c-67-171-96-240.hsd1.pa.comcast.net] has quit [Connection timed out]
20090614 21:13:08-!- ancestral [n=ancestra@32.134.80.42] has quit [Read error: 104 (Connection reset by peer)]
20090614 21:13:29-!- Doppp [n=aassdf@c-67-171-96-240.hsd1.pa.comcast.net] has joined #wesnoth-dev
20090614 21:13:36-!- Canageek [n=Canageek@d24-141-252-217.home.cgocable.net] has joined #wesnoth-dev
20090614 21:16:32< noy> norbert_: umm... who created it?
20090614 21:16:39< noy> gambit?
20090614 21:19:24< noy> norbert_: it was frezycuz who doesn't come on here. If you have questions I suggest you message him on the forum
20090614 21:19:46< norbert_> k, thanks
20090614 21:20:22-!- Canageek [n=Canageek@d24-141-252-217.home.cgocable.net] has left #wesnoth-dev ["As I walk through the valley of the shadow of death/I take a look at my life and realize there's nothing left"]
20090614 21:20:42< Soliton> i'd guess the ulf is overpowered since a guaranteed kill helps alot in that scenario.
20090614 21:23:22< norbert_> I just noticed the ulf is still there though, but I couldn't pick it as the host - for some reason
20090614 21:23:31< norbert_> nvm
20090614 21:23:45< Blueblaze> lol
20090614 21:30:24< grzywacz> http://pastebin.com/m611b5ff6
20090614 21:38:53-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090614 21:42:05-!- Doppp|EeePC [n=aasdasd@c-67-171-96-240.hsd1.pa.comcast.net] has joined #wesnoth-dev
20090614 21:42:54< silene> grzywacz: which scenario is that? it seems like it is using the [gold] tag without a side attribute
20090614 21:44:12< grzywacz> silene, it died after I finished AOI
20090614 21:45:00-!- ettin_ [n=jorda@wesnoth/developer/ettin] has joined #wesnoth-dev
20090614 21:48:49< CIA-77> silene * r36210 /trunk/data/lua/wml-tags.lua: Allowed the [gold] tag without a side attribute.
20090614 21:55:21-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090614 21:56:52-!- Espreon [n=espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20090614 21:57:17-!- ettin [n=jorda@wesnoth/developer/ettin] has quit [Read error: 110 (Connection timed out)]
20090614 22:03:14-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit []
20090614 22:03:49-!- stikonas [n=and@ctv-213-164-98-244.vinita.lt] has joined #wesnoth-dev
20090614 22:16:49-!- cjhopman [n=chris@wesnoth/developer/cjhopman] has joined #wesnoth-dev
20090614 22:17:00 * grzywacz summons mordante :P
20090614 22:19:46-!- norbert_ [n=norbert_@82-171-70-54.ip.telfort.nl] has quit ["Leaving"]
20090614 22:39:12< Ivanovic> cjhopman: so how is gsoc going for you?
20090614 22:39:16< Ivanovic> everything fine?
20090614 22:39:25-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090614 22:39:38< CIA-77> grzywacz * r36211 /trunk/src/ (game.cpp game_preferences.cpp game_preferences.hpp): Record completed campaigns in game preferences and provide this information to the campaign selection dialog.
20090614 22:41:16< grzywacz> No clue how to display it in a reasonable way with gui2...
20090614 22:41:52-!- maxy [n=maxy@84-74-83-103.dclient.hispeed.ch] has quit []
20090614 22:43:10< grzywacz> The most basic approach would be to add a string like (Completed) after the campaign name, but it's ugly and makes the dialog wider.
20090614 22:43:31< grzywacz> Some kind of a small indicator would be the best...
20090614 22:48:46< grzywacz> A glowing campaign icon would rule, but that's hardly possible. Maybe a special per-campaign "Completed" icon to mark the accomplishment?
20090614 22:49:59< silene> it doesn't have to be on the left, it could be on the right pane, there is enough place there to even make the "completed" per campaign difficulty
20090614 22:50:53< Soliton> it'd be nice to have a quick overview though and not have to click each campaign in order to check.
20090614 22:52:05-!- Crab_ [n=Crab_@wesnoth/developer/crab] has quit ["Leaving."]
20090614 22:54:23< Soliton> could add a ✔ or similar after the campaign name.
20090614 22:56:23< Soliton> as for a special per campaign marker.. you could have a different campaign icon for completed campaigns. i'm not sure that's worth it though. i imagine it's difficult to find intuitive icons.
20090614 22:58:46-!- ilor_ [n=user@wesnoth/developer/ilor] has quit [Read error: 113 (No route to host)]
20090614 23:06:35-!- Noyga [n=lame-z@wesnoth/developer/noyga] has quit [Read error: 110 (Connection timed out)]
20090614 23:07:57-!- Noyga [n=lame-z@wesnoth/developer/noyga] has joined #wesnoth-dev
20090614 23:30:31< cjhopman> Ivanovic: it's going well.
20090614 23:36:36-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090614 23:42:24-!- BenUrban_ [n=benurban@68.55.19.224] has joined #wesnoth-dev
20090614 23:44:16-!- BenUrban [n=benurban@unaffiliated/benurban] has quit [Nick collision from services.]
20090614 23:44:20-!- BenUrban_ is now known as BenUrban
20090614 23:45:04-!- shadowmaster [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090614 23:45:45< shadowmaster> Soliton: um, then perhaps it's just that the python wesnoth_addon_manager doesn't support .ign files / didn't support it at the time I uploaded that campaig
20090614 23:49:45< Soliton> i'm not aware that it does.
20090614 23:52:57-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090614 23:53:21< shadowmaster> .ign files contain shell-like glob patterns specifying which filoes or directories should not be added to the archive sent to the server when uploading an add-on . 
20090614 23:53:27< shadowmaster> example: http://wesnoth-umc-dev.svn.sourceforge.net/viewvc/wesnoth-umc-dev/trunk/Invasion_from_the_Unknown/_server.ign?view=markup
20090614 23:55:03< Soliton> i know.
20090614 23:59:13-!- shikadibot9 [n=shikadi@190.22.65.214] has joined #wesnoth-dev
20090614 23:59:57-!- shikadibot [n=robored@wesnoth/umc-dev/bot/shikadibot] has left #wesnoth-dev ["Leaving (requested by elynia via script)."]
--- Log closed Mon Jun 15 00:00:07 2009