--- Log opened Sat Feb 21 00:00:52 2009
20090221 00:02:20< esr> I've fixed or confirmed fixed on eight bugs in the last two days.  Still trying to get the bug count down to the previous low again...
20090221 00:04:25< zookeeper> is the gamma slider supposed to get either restored or entirely removed for 1.6?
20090221 00:05:30< Shadow_Master> see the commit message for revision 32813
20090221 00:06:13< zookeeper> wesbot, log 32813
20090221 00:06:15< wesbot> mordante * r32813 : Removed the gamma correction option.
20090221 00:06:21< wesbot> URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=32813
20090221 00:06:37< zookeeper> well it's not properly removed then, it still takes up space in the UI.
20090221 00:07:27< Shadow_Master> yes, Mordante mentioned that needed fixing.
20090221 00:07:38< Shadow_Master> zookeeper: read the entire message. Wesbot throws only the first line
20090221 00:07:41< zookeeper> ok, no problem then
20090221 00:10:39< esr> zookeeper: Thanks for educating me on the music macros, BTW.  I got that bug fixed.
20090221 00:13:44-!- stikonas [n=and@ctv-213-164-126-243.vinita.lt] has quit [Read error: 110 (Connection timed out)]
20090221 00:14:19-!- Noyga [n=noyga@AVelizy-151-1-27-8.w82-124.abo.wanadoo.fr] has left #wesnoth-dev ["Quitte"]
20090221 00:25:40-!- cib0 [n=cib@p4FD0DADE.dip.t-dialin.net] has quit [Remote closed the connection]
20090221 00:28:39< thespaceinvader> esr: re. the sylph portrait, check the big TC thread where Jet provided the sprite - he specifically asked for the mask.
20090221 00:29:34< esr> Oh.  Well, I still don't like it.  Aesthetically speaking.
20090221 00:30:22< thespaceinvader> mm, fair enough
20090221 00:30:39< thespaceinvader> but it might be better to send to comment to jet who designed the unit, rather than kitty =)
20090221 00:34:20-!- cjhopman_ [n=chris@wireless33.cs.wisc.edu] has joined #wesnoth-dev
20090221 00:34:21-!- cjhopman [n=chris@wireless33.cs.wisc.edu] has quit [Read error: 54 (Connection reset by peer)]
20090221 00:34:27< esr> thespaceinvader: I've update the post to so note.
20090221 00:34:30-!- ilor [n=user@213-238-122-211.adsl.inetia.pl] has quit []
20090221 00:35:00< thespaceinvader> cool, thanks =)
20090221 00:35:24< thespaceinvader> i'm just aware that kitty's kinda burned out on the elves, and just wants to see the back of them as quickly as possible...
20090221 00:36:37-!- grzywacz [n=grzywacz@moinmoin/developer/karol] has quit [Read error: 110 (Connection timed out)]
20090221 00:41:55-!- Sapient [n=sapien-x@c-24-126-133-155.hsd1.ga.comcast.net] has joined #wesnoth-dev
20090221 00:41:58-!- BenUrban [n=benurban@pool-71-178-139-190.washdc.east.verizon.net] has joined #wesnoth-dev
20090221 00:42:14< Sapient> Shadow_Master: problem with wmliterator?
20090221 00:42:20< Shadow_Master> esr: ^
20090221 00:42:34-!- cjhopman_ [n=chris@wireless33.cs.wisc.edu] has quit [Remote closed the connection]
20090221 00:42:36< Shadow_Master> Sapient: yes, it is ignoring the "wmllint: unbalanced-on" pragma and its implications
20090221 00:42:51-!- cjhopman_ [n=chris@wireless33.cs.wisc.edu] has joined #wesnoth-dev
20090221 00:42:59< Sapient> that's because wmliterator doesn't check for unbalanced tags
20090221 00:43:02< esr> We tripped over it while examining a problem with Kaleh's AMLAs in UtBS.  
20090221 00:43:12< Shadow_Master> so, I haven't marked data/campaigns/<UtBS>/units/Kaleh.cfg as I should because it'd be kind of useless at,
20090221 00:43:15< Shadow_Master> *atm
20090221 00:43:34< Sapient> so... since wmliterator doesn't check for unbalanced tags, there may be something seriously wrong with your WML
20090221 00:43:55< Shadow_Master> Sapient: the only thing wrong with my *not mine* WML is that it uses macros.
20090221 00:44:12< Shadow_Master> and that esr ruined Kaleh's AMLA trying to fix that.
20090221 00:44:13< Sapient> simple macros obviously aren't the issue
20090221 00:44:21< esr> There is.  Ugly horrible macro abuse.  I told Shadow_Master it really needs to be cleaned up or removed.
20090221 00:44:25< Sapient> but, can I reproduce it by running wmliterator on trunk?
20090221 00:44:30< esr> Yes.
20090221 00:44:35< Shadow_Master> or wmllint on the single file.
20090221 00:44:40< Sapient> which?
20090221 00:44:53< Shadow_Master> data/campaigns/Under.../units/Kaleh.cfg
20090221 00:45:10< Sapient> ok, I'll have a look
20090221 00:45:10-!- cjhopman_ is now known as cjhopman
20090221 00:46:20 * Sapient runs svn update
20090221 00:46:29< esr> cjhopman: Are you a long-tme dev recently reactivating or a new guy?  I don't recall having seen your name on things before recently, but you've been pretty active.  Just curious.
20090221 00:48:11-!- zookeeper [n=l@dsl-tkubrasgw1-fe9ff800-215.dhcp.inet.fi] has quit []
20090221 00:48:46< cjhopman> i applied for gsoc last year
20090221 00:48:51< Sapient> he's a fine coder and kindly helped us out with some unit_map improvements
20090221 00:49:01< esr> Welcome (back).
20090221 00:49:21< cjhopman> and have been ridiculously busy since
20090221 00:49:27< cjhopman> til this semester
20090221 00:49:56< esr> Makes me happy whenevber we fet someone who's willing to dive into the C++ guts, because I sure don't want to.
20090221 00:50:25< cjhopman> some of us like pain, i guess
20090221 00:50:29< Sapient> lol
20090221 00:50:37-!- mjs-de [n=mjs-de@vpw.wh.Uni-Dortmund.DE] has quit ["On the road again"]
20090221 00:50:53< Shadow_Master> I love it!
20090221 00:51:04< Sapient> yes, the best computer science departments are designed to turn you into a "glutton for pain" ;)
20090221 00:51:06< esr> I love this project, but the one major C++ refactoring run I did on rthis code soured me on the language forever.
20090221 00:51:14< Ivanovic> time to head off to bed, n8
20090221 00:52:25< esr> Anybody know when :cl was added to debug mode?  I noticed it's not documented on the wiki.
20090221 00:52:39< Shadow_Master> git log
20090221 00:52:41< Shadow_Master> er
20090221 00:52:46< Sapient> esr: hows that C++ hate-book coming along?
20090221 00:53:14< Shadow_Master> wesbot: log 27965
20090221 00:53:15< wesbot> alink * r27965 : Add a new debug command ":choose_level" or ":cn" to open a simple menu with all
20090221 00:53:19< Shadow_Master> esr:  
20090221 00:53:19< esr> Sapient: My collaborator flew in gwo days ago.  We're working on it now.
20090221 00:53:19< wesbot> URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=27965
20090221 00:54:02< alink> wesbot: log 27986
20090221 00:54:03< wesbot> alink * r27986 : Use ":cl" as predefined alias of ":choose_level" instead of ":cn"(?)
20090221 00:54:06< wesbot> URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=27986
20090221 00:55:04< Sapient> ah, well I think the worst thing about C++ is it has so many "gotchas;" I wish I had written them all down as I was learning the language (without any book)
20090221 00:55:16< esr> OK, so that's definitely dev branch only.  That's what I needed.
20090221 00:55:28< Shadow_Master> yes.
20090221 00:55:29< esr> Sapient: Yes, that's a running theme.
20090221 00:55:55< Shadow_Master> indeed.
20090221 00:56:11< Sapient> in fact, in #c++ they'll tell you you're insane if you try to pick up the language without a *good* book
20090221 00:56:23< esr> In fact, he just walked in, ttold me he'd shipped me a draft, and said "Your turn." Gonna be unavailable for a while...
20090221 00:56:25< Shadow_Master> ##c++ actually
20090221 00:56:32 * Sapient same
20090221 00:57:03 * esr joins #c++
20090221 00:57:18< Shadow_Master> #c++ is just an ugly redirect
20090221 00:58:08< Sirp> ##C++ and #C++ are both ugly. ;)
20090221 00:58:21< Sirp> in my experience, at least, they severely abuse people in there.
20090221 00:58:38< Shadow_Master> that makes such a bad image of C++ users <sighs>
20090221 00:58:52< Sirp> though they are right that you are insane if you try to learn C++ without a good book.
20090221 00:58:55< esr> Sirp: You know you're on the short-list to getan early draft of this thing.
20090221 00:59:01< Shadow_Master> I read that the freebsd people also flame people tho
20090221 00:59:11< Shadow_Master> s/flame people/abuse newcomers/
20090221 00:59:15< Sirp> esr: uh-huh. :)
20090221 01:00:03< esr> No no, that's OpenBSD you're thinking of. Theo deRaadt.  The man's insane.
20090221 01:01:18< Espreon> How so?
20090221 01:01:58-!- Turuk_ [n=Turuk@FUSE-WS1-208-102-202-188.fuse.net] has joined #wesnoth-dev
20090221 01:02:45< Shadow_Master> "vampire hunting pro..." lol
20090221 01:04:56-!- noy_ [n=Noy@70.70.128.133] has quit ["GO, GET TO THE CHOPPAH!!!"]
20090221 01:05:24-!- Shadow_Master [n=imorelle@unaffiliated/shadowmaster/x-275924] has quit ["brb"]
20090221 01:06:08-!- Shadow_Master [n=imorelle@unaffiliated/shadowmaster/x-275924] has joined #wesnoth-dev
20090221 01:08:10-!- Shadow_Master [n=imorelle@unaffiliated/shadowmaster/x-275924] has quit [Client Quit]
20090221 01:09:20-!- Shadow_Master [n=imorelle@unaffiliated/shadowmaster/x-275924] has joined #wesnoth-dev
20090221 01:12:23< Sapient> yes, obvious reference to the anime of same name ;)
20090221 01:12:28 * loonycyborg is disappointed that esr didn't start a flamewar on ##c++ :P
20090221 01:12:36< Sapient> which was pretty good IMO
20090221 01:13:15-!- cjhopman [n=chris@wireless33.cs.wisc.edu] has quit ["Konversation terminated!"]
20090221 01:21:03-!- Turuk [n=Turuk@FUSE-WS1-208-102-202-188.fuse.net] has quit [Connection timed out]
20090221 01:21:05 * Shadow_Master is disappointed at himself.
20090221 01:21:41< Shadow_Master> I should no longer spend time on mainline coding/bug fixing. I should be concentrating on my UMC work...
20090221 01:23:33-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has quit [Remote closed the connection]
20090221 01:25:49< Shadow_Master> wesbot: topic
20090221 01:25:51< wesbot> Shadow_Master: Don't be impatient, wait until something changes...  *tsk*
20090221 01:25:58< Shadow_Master> k
20090221 01:26:21< Shadow_Master> so, (In Progress == wesbot doesn't care) too?
20090221 01:27:12< Shadow_Master> wow at ##c++ ;-)
20090221 01:28:47 * Sirp has to go in to see what's going on there.... :)
20090221 01:29:36< Sapient> just having some fun wiith nolyc, the resident bot :p
20090221 01:30:01-!- happygrue [n=Wintermu@c-67-176-145-41.hsd1.in.comcast.net] has quit ["grok 'noth!"]
20090221 01:30:59-!- Shadow_Master [n=imorelle@unaffiliated/shadowmaster/x-275924] has quit ["Reconnecting"]
20090221 01:31:02-!- Shadow_Master [n=imorelle@unaffiliated/shadowmaster/x-275924] has joined #wesnoth-dev
20090221 01:31:47-!- cjhopman [n=chris@wireless33.cs.wisc.edu] has joined #wesnoth-dev
20090221 01:33:27 * Shadow_Master wonders why Mordante takes time to do C++ formatting fixes, and not use proper capitalization in the GUI2 source code files' headings
20090221 01:37:44-!- happygrue [n=Wintermu@c-67-176-145-41.hsd1.in.comcast.net] has joined #wesnoth-dev
20090221 01:39:34< Sirp> okay, "evilgeek" reminded me very quickly why I do not visit ##C++
20090221 01:40:16< Sirp> ...with the smug "everyone is beneath me and everything is obvious; if you don't get it straight away you are a moron" attitude.
20090221 01:41:02-!- Turuk_ [n=Turuk@FUSE-WS1-208-102-202-188.fuse.net] has quit [Read error: 60 (Operation timed out)]
20090221 01:41:36< Shadow_Master> I don't know why it sounds familiar to me though.
20090221 01:42:48-!- loonybot [n=loonybot@79.139.247.150] has quit [Remote closed the connection]
20090221 01:43:10-!- loonybot [n=loonybot@79.139.247.150] has joined #wesnoth-dev
20090221 01:44:36-!- ikarius [n=ross@smtp.gridironsystems.com] has quit []
20090221 01:47:35-!- loonybot [n=loonybot@79.139.247.150] has quit [Remote closed the connection]
20090221 01:47:45-!- Turuk [n=Turuk@FUSE-WS1-208-102-202-188.fuse.net] has joined #wesnoth-dev
20090221 01:47:57-!- loonybot [n=loonybot@79.139.247.150] has joined #wesnoth-dev
20090221 01:51:39-!- loonycyborg [n=sergey@79.139.247.150] has quit ["Zzzzzzzzzzzzzzzzzzzzzzzzzz"]
20090221 01:52:17-!- loonybot [n=loonybot@79.139.247.150] has quit [Remote closed the connection]
20090221 01:54:19-!- Shadow_Master [n=imorelle@unaffiliated/shadowmaster/x-275924] has quit ["brb"]
20090221 01:54:39-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Read error: 104 (Connection reset by peer)]
20090221 01:55:12-!- ABCD [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090221 01:56:06-!- Shadow_Master [n=imorelle@unaffiliated/shadowmaster/x-275924] has joined #wesnoth-dev
20090221 02:05:47-!- cjhopman [n=chris@wireless33.cs.wisc.edu] has quit [Read error: 60 (Operation timed out)]
20090221 02:10:31-!- Elvish_Pillager [n=eli@66-189-14-42.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090221 02:10:52-!- Elvish_Pillage2 [n=eli@66-189-14-42.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090221 02:35:39-!- cjhopman [n=chris@68-118-159-7.dhcp.mdsn.wi.charter.com] has joined #wesnoth-dev
20090221 02:40:36-!- Dragonking [n=dk@dedikerad/dragonking] has quit []
20090221 02:44:31-!- crimson_penguin [n=ben@CPE0013f7fa2460-CM0013f7fa245c.cpe.net.cable.rogers.com] has joined #wesnoth-dev
20090221 02:48:14-!- AndreLuiz [n=AndreLui@189.71.43.112] has quit ["Leaving"]
20090221 02:48:21< Sapient> Shadow_Master: ok I see what the problem with Kaleh is
20090221 02:48:35-!- AndreLuiz [n=AndreLui@189.71.43.112] has joined #wesnoth-dev
20090221 02:48:45< Shadow_Master> I'm playing a game. No answers.
20090221 02:48:50< Sapient> wmliterator allows you to have unbalanced tags in #define blocks
20090221 02:49:13< Sapient> but in this case, it's an unbalanced tag in a macro (parameter)
20090221 02:49:48< Sapient> which is, frankly, so bizarre I'm not sure it should even be allowed
20090221 02:50:39< Sapient> but since it apparently works, I guess I could teach wmliterator to ignore it...
20090221 02:51:36-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev
20090221 03:10:43-!- Elvish_Pillager [n=eli@66-189-14-42.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090221 03:11:19-!- Elvish_Pillager [n=eli@66-189-14-42.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090221 03:16:37< Shadow_Master> <+CIA-26> shikadilord * r3917 [...] Overriding the melee attack strengths for Cockatrice and Chimera on offense, to make Glaze more likely to be used.
20090221 03:16:52< Shadow_Master> Invasion from the Unknown, the add-on where Cookie-Eater monsters rejoice.
20090221 03:17:02-!- esr [n=chatzill@static-71-162-243-5.phlapa.fios.verizon.net] has quit [Remote closed the connection]
20090221 03:17:48-!- Sirp_ [n=me@c-71-198-3-104.hsd1.ca.comcast.net] has joined #wesnoth-dev
20090221 03:18:11< Shadow_Master> the conversation in #wesnoth-umc-dev is tasty.
20090221 03:18:51< Sirp_> ?
20090221 03:19:19< Shadow_Master> I like to mispell "gaze" as "glaze", that's al.
20090221 03:19:31-!- thespaceinvader [n=chatzill@91.108.177.73] has quit ["ChatZilla 0.9.84 [Firefox 3.0.6/2009011913]"]
20090221 03:19:42-!- cookiebot [n=sh314001@190.22.106.231] has joined #wesnoth-dev
20090221 03:21:39-!- cookiebot [n=sh314001@190.22.106.231] has quit [Client Quit]
20090221 03:22:03-!- Shadow_Master is now known as Sugar_Master
20090221 03:22:14 * Sapient throws sugar packet at Sugar ------------>#
20090221 03:22:32 * Sugar_Master assimilates it to increase in size.
20090221 03:23:02< Sapient> do most Chileans have an odd sense of humor?
20090221 03:23:13-!- Espreon is now known as SkittleLord
20090221 03:23:21< Sugar_Master> no
20090221 03:27:00< Sapient> what about Argentinians?
20090221 03:27:15-!- Sugar_Master is now known as ShikadiLord
20090221 03:27:27< ShikadiLord> they have plagued the chilean TV market.
20090221 03:27:36< ShikadiLord> they aren't any good for our TV programs.
20090221 03:27:44-!- SkittleLord is now known as Espreon
20090221 03:28:42 * ShikadiLord throws a smoke bomb to the floor
20090221 03:28:49-!- ShikadiLord [n=imorelle@unaffiliated/shadowmaster/x-275924] has quit ["bye, n00bs!"]
20090221 03:30:45-!- Genjuro [n=elliott@189.71.66.49] has joined #wesnoth-dev
20090221 03:33:33-!- [Relic] [n=[Relic]@adsl-76-229-202-137.dsl.milwwi.sbcglobal.net] has joined #wesnoth-dev
20090221 03:34:50< [Relic]> Hello :)
20090221 03:38:20-!- alink [n=alink@78.129.18.215] has quit [Remote closed the connection]
20090221 03:40:33-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has quit [Remote closed the connection]
20090221 03:49:24-!- Elvish_Pillager [n=eli@66-189-14-42.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!"]
20090221 03:53:29-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev
20090221 03:57:31-!- Genjuro [n=elliott@189.71.66.49] has quit ["Leaving"]
20090221 03:59:43-!- Corvvs [n=michaelm@pool-71-163-211-230.washdc.east.verizon.net] has joined #wesnoth-dev
20090221 04:04:15-!- ikarius [n=ross@216.27.182.3] has joined #wesnoth-dev
20090221 04:12:29-!- BenUrban [n=benurban@unaffiliated/benurban] has quit [Remote closed the connection]
20090221 04:20:02-!- ikarius [n=ross@216.27.182.3] has quit []
20090221 04:21:20-!- JW1 [n=X@c-71-57-85-168.hsd1.il.comcast.net] has joined #wesnoth-dev
20090221 04:30:48< Sirp_> Turuk: I wonder, do you think there's some way we can easily let people know that you're the default "go to guy" for help with the forums?
20090221 04:30:53< Sirp_> I take it you are happy with that role? :)
20090221 04:31:12< Turuk> Haha sure, I don't mind it at all, Deciero found me in game and I fixed his issue
20090221 04:31:40< Sapient> Turuk, the forum guy :)
20090221 04:31:48< Turuk> I'm not sure how to easily convey that to individuals though
20090221 04:31:58< Turuk> Hmm.... I could change my signature
20090221 04:32:23< Espreon> Custom title perhaps?
20090221 04:33:13< Turuk> What if I kept the green so people knew I was a Sith but just changed my title to Administrator?
20090221 04:33:53-!- BenUrban [n=benurban@c-68-49-10-243.hsd1.md.comcast.net] has joined #wesnoth-dev
20090221 04:34:19< Espreon> Turuk: I think that they would fear you more instead of wanting to seek help from you...
20090221 04:35:10< Turuk> I'm open to ideas :)
20090221 04:35:24< Corvvs> he could be a jedi instead... :)
20090221 04:35:51< Sapient> no, the jedi are relentless... there could be civil war without end
20090221 04:35:54< Espreon> "Totally Awesome Guy to Whom You Can Easily Talk" -- just an idea...
20090221 04:36:05< Sirp_> Turuk: you could put "Forum Administrator. Please send any administrative problems with the forum to me." in your signature
20090221 04:36:30< Sirp_> Espreon: then he'll get messages like, "So my girlfriend dumped me, and I don't know what to do, and....."
20090221 04:36:35< Turuk> Ah, that is perfect actually, I was thinking along those lines.  Thanks Sirp.
20090221 04:36:44< Sirp_> :)
20090221 04:37:27< Espreon> Sirp_: Hmm.... I did not consider that...
20090221 04:37:47< Sapient> also, "if you think all the developers are corrupting Dave's pure vision of Wesnoth with lunacy, please contact me"
20090221 04:38:51< Espreon> LOL.
20090221 04:39:08< Turuk> Haha
20090221 04:39:23< Turuk> My inbox would be overflowing with random emails
20090221 04:40:40< Turuk> Adjusted my signature, with a nod back to the old forum days ;)
20090221 04:41:44< Corvvs> good color choice for that red. Very eye-catching. Actually made me sit back in my seat.
20090221 04:42:11< Turuk> Yeah, this way people will take notice.  The Admins used to have orange titles back in the day
20090221 04:47:15-!- Ivanovic_ [n=ivanovic@dtmd-4db26e4e.pool.einsundeins.de] has joined #wesnoth-dev
20090221 04:51:02-!- JW1 [n=X@c-71-57-85-168.hsd1.il.comcast.net] has quit []
20090221 05:03:45-!- Ivanovic [n=ivanovic@dtmd-4db2a4c3.pool.einsundeins.de] has quit [Read error: 110 (Connection timed out)]
20090221 05:05:13-!- Ivanovic_ is now known as Ivanovic
20090221 05:13:37-!- esr [n=chatzill@static-71-162-243-5.phlapa.fios.verizon.net] has joined #wesnoth-dev
20090221 05:29:16< Espreon> ilor: It seems that r32885 did fix bug #12711...
20090221 05:29:26 * Espreon hopes that ilors reads the logs.
20090221 05:30:49-!- Corvvs [n=michaelm@pool-71-163-211-230.washdc.east.verizon.net] has quit ["Cheers."]
20090221 05:48:11< Espreon> ilor: Wait, I meant it did not...
20090221 05:49:40< Espreon> esr: ^
20090221 05:49:46 * Espreon is exhausted...
20090221 05:51:28< Soliton> you could add a comment to the bug report.
20090221 05:51:28< esr> Fuck.
20090221 05:52:10< Espreon> Soliton: I did.
20090221 05:52:27 * Espreon is ashamed due to the fact that he spread incorrect info...
20090221 05:53:10< esr> I've changed the status  back to "None".
20090221 05:56:22-!- Turuk [n=Turuk@FUSE-WS1-208-102-202-188.fuse.net] has left #wesnoth-dev []
20090221 05:58:15< CIA-70> soliton * r32950 /trunk/ (changelog src/game_display.cpp): * Fixed bug #13034: "/me" messages cut of first character in multiplayer game chat.
20090221 06:04:04-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev
20090221 06:09:52-!- Lord_Aether [n=castle@207.212.136.10] has joined #wesnoth-dev
20090221 06:12:04< Sapient> esr: the good news is I fixed wmliterator's handling of Kaleh.cfg
20090221 06:12:13< Sapient> the bad news is, wmllint still chokes on it
20090221 06:13:49< CIA-70> sapient * r32951 /trunk/data/tools/wesnoth/wmliterator.py: wmliterator now allows unbalanced tags inside macro-call (parameters).
20090221 06:16:03< Sapient> "campaigns\Under_the_Burning_Suns\units\Kaleh.cfg", line 621: tag stack nonempty
20090221 06:19:06< Sapient> so you can't blame wmliterator for your woes anymore :p
20090221 06:19:14< Sapient> cya later, have fun.
20090221 06:19:14-!- Sapient [n=sapien-x@c-24-126-133-155.hsd1.ga.comcast.net] has left #wesnoth-dev []
20090221 06:21:12-!- Corvvs [n=michaelm@pool-71-163-211-230.washdc.east.verizon.net] has joined #wesnoth-dev
20090221 06:23:41-!- ikarius [n=ross@c-98-207-225-178.hsd1.ca.comcast.net] has joined #wesnoth-dev
20090221 06:30:05< esr> wesbot: topic
20090221 06:50:13-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has quit [Remote closed the connection]
20090221 06:53:29-!- boucman [n=rosen@140.117.71-86.rev.gaoland.net] has quit [Remote closed the connection]
20090221 06:57:44-!- [Relic] [n=[Relic]@adsl-76-229-202-137.dsl.milwwi.sbcglobal.net] has quit ["Leaving"]
20090221 07:04:15-!- Mordante [n=chatzill@roadie.xs4all.nl] has joined #wesnoth-dev
20090221 07:04:31< Mordante> hi
20090221 07:07:03-!- happygrue [n=Wintermu@c-67-176-145-41.hsd1.in.comcast.net] has quit ["grok 'noth!"]
20090221 07:07:57< esr> Mordante: Hi.  I'm in full bug-hunt mode.  About ten disposed of today and yesterday.
20090221 07:08:21< Mordante> nice :-) I also dismissed quite some bugs last week
20090221 07:08:22< esr> You own, like, 19 of the remaining 57.  Poke, poke...
20090221 07:08:38< Mordante> how many of them are postponed?
20090221 07:08:51< esr> Oh, about half.
20090221 07:09:25< esr> I postponed the Python 2.6 compilation problem, we don't need that before 1.6.
20090221 07:09:26< Mordante> postponed won't be fixed before 1.6
20090221 07:10:15< dfranke> esr: I noticed you haven't assigned my security bug.  Does someone have it under control?  If not I can probably squash it myself.
20090221 07:11:08< Mordante> esr ready for test means we think the bug is fixed so we don't count it anymore, thus ready for test -> fixed doesn't change the bug count
20090221 07:13:11< esr> dfranke: I haven't assigned it because no=body particularly ownbs that code, though Mordante was the last personm to work on it seriously.  If you can fix it, by all means do.
20090221 07:13:59 * esr Tests's Mordante's claim...
20090221 07:14:12< esr> Ghod I'm illiterate tonight...
20090221 07:14:19< esr> wesbot: topic
20090221 07:14:21< wesbot> esr: Don't be impatient, wait until something changes...  *tsk*
20090221 07:15:10< esr> Hm.  One mve from Ready To Test to Confirm didn't change the stats.  Maybe somebody else nipped in and fixed one during the window>
20090221 07:15:11< Mordante> dfranke I think that problem can be triggered in various ways due to the gzip compression
20090221 07:15:13< esr> ?
20090221 07:15:57< Mordante> wesbot: topic
20090221 07:15:59< wesbot> Mordante: Don't be impatient, wait until something changes...  *tsk*
20090221 07:16:43< Mordante> no new items... odd
20090221 07:17:02< dfranke> Mordante: fixing the particular problem I reported is just a matter of adding a bounds check to the parsing code.  But yeah, there may be other related attacks.
20090221 07:18:06< Mordante> bounds checks are an ugly bandaid and doesn't help for lowmem machines
20090221 07:18:30< esr> Mordante: Agreed, but it's unclear what else might serve here.
20090221 07:21:29< dfranke> is this problem minor enough that I can go ahead and speak freely about it on this channel?  I'd rather not have it leak out to the script kiddies before there's a patch.
20090221 07:21:37< Mordante> I agree it's a problem only fixing it is hard unless you add hard-coded limits
20090221 07:21:41< esr> I think so.
20090221 07:21:59< Mordante> uhm yes afaik everybody logged in to gna can see the bug
20090221 07:22:32< dfranke> really?  I assumed it was just me and project members.
20090221 07:23:10< esr> No.  Anonymous access to the bug tracker is allowed.  You can, however, declare a bug "private".
20090221 07:23:21< dfranke> and I did.
20090221 07:23:37< esr> That might restrict access, I dunno.
20090221 07:23:39< dfranke> the above is my interpretation of private.
20090221 07:24:22< Mordante> not 100% sure
20090221 07:24:27< esr> dfranke: It would be mine, too, but I don't know if the tracker actually behaves that way.
20090221 07:26:38< dfranke> anyway, the map editor UI currently allows maps up to 200x200.  I take that as implying a policy decision that that's the maximum reasonable size for a map.  So given that policy, it's not a bug if users don't have enough memory to render that.
20090221 07:27:20< Mordante> the problem is we have creative people who sometimes go beyond the limits
20090221 07:27:43< Mordante> so it wouldn't be the first time we think something is not used and breaking user stuff
20090221 07:28:13< Mordante> and yes I wouldn't be surprised if somebody has a script to combine four maps to one 
20090221 07:29:37< Mordante> shadow_master for some things I have a script for others not :-P
20090221 07:29:38< dfranke> how about if I have the bounds check be a function of available memory?
20090221 07:33:06< Mordante> how do you mean that?
20090221 07:35:30< dfranke> Nevermind, screw it.  I really think we should just make a policy decision on this.
20090221 07:36:06< dfranke> and 200x200 seems like a reasonable policy.
20090221 07:36:31< dfranke> AFAIK no existing UMC violates that.
20090221 07:38:31< dfranke> Arbitrary limits are theoretically ugly, but the bottom line is that real computers are finite state machines, and if you try too hard to pretend they're Turing machines then reality is going to smack you for it.
20090221 07:47:13< Mordante> true, let's say I have bad experience with assuming nobody violated some rules
20090221 07:47:24< Mordante> and opened a can of worms with it
20090221 07:47:55< Soliton> just issue a warning rather than disallowing certain map sizes.
20090221 07:49:23< dfranke> that would be bad UI, because you can encounter it just by entering the game lobby and having it try to render a preview.
20090221 07:49:58< esr> dfranke, Mordante: Looks like making the bug private has the effect you expcted.  I logged out and looked at the tracker anonymously to check.
20090221 07:50:14< Mordante> ?
20090221 07:50:30< dfranke> if anything, it needs to be an option you set ahead of time in preferences.
20090221 07:50:41< esr> The  bug dfranke declared 'private' isn't visible.
20090221 07:50:46< Soliton> you mean the mini map would cause trouble?
20090221 07:50:53< dfranke> yeah
20090221 07:51:02< cjhopman> is it visible to logged in people who are not project members?
20090221 07:51:14< cjhopman> i doubt it
20090221 07:52:07< dfranke> cjhopman: was 'i doubt it' in reference to the mini-map or in reference to the visibility of the bug?
20090221 07:52:12< Soliton> well, not showing the mini map at a certain size is no problem of course.
20090221 07:52:28< esr> cjhopman: Can'tvtell.
20090221 07:52:48< cjhopman> dfranke: to bug visibility
20090221 07:53:50< Mordante> that was my question about the bug as well
20090221 07:54:14-!- Corvvs [n=michaelm@pool-71-163-211-230.washdc.east.verizon.net] has quit []
20090221 07:56:43< Mordante> afk now, please do not assign that bug to me my plate is already quite full
20090221 07:56:51< dfranke> I think Mordante is correct in pointing out that there's a larger problem with compressed data in general.
20090221 07:57:09< dfranke> I've seen bz2 files a few bytes long that uncompress to terabytes.
20090221 07:57:37< Mordante> which is not hard to do with compression
20090221 08:00:28-!- Lord_Aether [n=castle@207.212.136.10] has quit []
20090221 08:12:47-!- ABCD_ [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090221 08:13:13-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Read error: 113 (No route to host)]
20090221 08:15:09< CIA-70> cjhopman * r32952 /trunk/src/ (actions.cpp actions.hpp): 
20090221 08:15:09< CIA-70> - begin refactoring of attack class
20090221 08:15:09< CIA-70> - add a small structure for unit-specific data and move data into it
20090221 08:15:09< CIA-70> - also use unit_map::iterator::valid() rather than the roll-your-own validity checks that used to be required (e.g. at beginning of refresh_bc() line 826)
20090221 08:37:21< esr> cjhopman: I don't mean to be annoying, because I'm generally in favor of refactoring cleanups, but...is this one likely to fix a pre-1.6 bug?
20090221 08:40:21< Mordante> Soliton you fixed this recently, didn't you? http://www.wesnoth.org/forum/viewtopic.php?f=4&t=24193&
20090221 08:40:43< Mordante> the second post
20090221 08:41:34-!- Noyga [n=noyga@AVelizy-151-1-27-8.w82-124.abo.wanadoo.fr] has joined #wesnoth-dev
20090221 08:45:34< Soliton> Mordante: yes.
20090221 08:46:14< Mordante> thanks
20090221 08:46:34< cjhopman> esr:  well, the goal is to fix 12428 ( and there are probably similar unreported bugs )... but it would be easy to convince me that its a lot of superfluous code to change pre 1.6 for that one bug
20090221 08:47:05< esr> Looking....
20090221 08:47:53< esr> Right.  Given there's a solid test case this does seem kuike a reasonable thing to try.
20090221 08:48:22< esr> s/kuike/like/
20090221 08:49:52< cjhopman> i don't know... i could still fix it without changing so much.
20090221 08:51:02< esr> I couldn't substitute my judgment for yours even if I were inclined to try -- don't know the relevant code well enough.  Just test the hell out of it, OK?
20090221 08:58:09-!- BenUrban [n=benurban@unaffiliated/benurban] has quit [Read error: 104 (Connection reset by peer)]
20090221 08:58:10-!- BenUrban_ [n=benurban@unaffiliated/benurban] has joined #wesnoth-dev
20090221 08:59:58-!- BenUrban_ is now known as BenUrban
20090221 09:06:21< Mordante> alink when running the test suite the output is flooded with "Could not backup background for report" 
20090221 09:06:33< Mordante> guess we need to test for a dummy video somewhere
20090221 09:18:11-!- ABCD_ is now known as ABCD
20090221 09:34:42-!- zookeeper [n=l@dsl-tkubrasgw1-fe9ff800-215.dhcp.inet.fi] has joined #wesnoth-dev
20090221 09:35:42-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has quit []
20090221 10:05:05-!- DDR [n=chatzill@66.183.125.196] has quit [Read error: 110 (Connection timed out)]
20090221 10:12:55-!- boucman [n=rosen@140.117.71-86.rev.gaoland.net] has joined #wesnoth-dev
20090221 10:18:26< dfranke> Mordante: well, you're right about the compression issues.  I just wrote a working exploit that crashes the MP server.
20090221 10:19:13< Mordante> dfranke I feared you would manage to do that ;-)
20090221 10:20:03< Mordante> it's always a problem with untrusted input and especially with compression
20090221 10:20:43< dfranke> (After first crashing my client umpteen times because there's a bug in network::send_raw_data())
20090221 10:21:06< boucman> dfranke: did you manage to crash a remote client too ?
20090221 10:21:29< Mordante> why crash a client when you can crash the server ;-) :-P
20090221 10:21:56< boucman> Mordante: to get yourself a place in a particular MP game :P
20090221 10:22:26< Mordante> but I don't want to play MP :-P
20090221 10:22:40< dfranke> boucman: I didn't actually develop a working exploit for that, but yes.  See #13031
20090221 10:23:06< boucman> wesbot: but 13031
20090221 10:23:13< boucman> wesbot: bug 13031
20090221 10:23:14< wesbot> boucman: Are you sure with that number?  Please check again...
20090221 10:23:17-!- silene [n=plouf@ASte-Genev-Bois-152-1-37-233.w82-121.abo.wanadoo.fr] has joined #wesnoth-dev
20090221 10:23:20< dfranke> it's private.
20090221 10:23:28< silene> hi
20090221 10:23:32< boucman> dfranke: you've got yourself a palindromic bug :P
20090221 10:23:37< boucman> hey silene
20090221 10:23:43< boucman> how's lua going ?
20090221 10:24:59< silene> boucman: i have been quite busy with work lately, so it didn't progress much: only wml events are fully supported, wml filters need a bit more work
20090221 10:25:18< boucman> k
20090221 10:25:30-!- ABCD_ [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090221 10:26:45-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Read error: 104 (Connection reset by peer)]
20090221 10:27:33< dfranke> Mordante: anyway, I read the RFC for gzip and it looks like it's possible to determine the uncompressed size of a compressed stream using O(1) space.
20090221 10:29:02< Mordante> dfranke also when using streaming compression?
20090221 10:29:37< dfranke> Mordante: well, obviously if you want to determine the uncompressed size and then do the uncompression you need to do two passes.
20090221 10:30:04< dfranke> Mordante: but you can always just monitor the size of the uncompressed stream as it grows.
20090221 10:30:14< Mordante> which means changing the server code quite a bit
20090221 10:31:34< dfranke> yeah, this is starting to seem fairly serious and non-trivial to fix.  It might have been a mistake to discuss it in public.
20090221 10:32:22< Mordante> yup the can of worms I mentioned earlier today
20090221 10:33:06< silene> what's the issue?
20090221 10:33:08< boucman> dfranke, Mordante: you might want to discuss technical details on #wesnoth-mentor
20090221 10:33:49< dfranke> I'm not saying any more about it on any channel with public logs.
20090221 10:34:05< boucman> -mentor is not logged iirc
20090221 10:34:26< dfranke> k
20090221 10:35:41-!- Sirp_ [n=me@c-71-198-3-104.hsd1.ca.comcast.net] has quit ["leaving"]
20090221 10:37:08< Mordante> silene can you also join #wesnoth-mentor
20090221 10:38:28-!- ettin [n=jorda@199.207.219.87.dynamic.jazztel.es] has joined #wesnoth-dev
20090221 10:50:08-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090221 10:50:30-!- jorda [n=jorda@132.207.219.87.dynamic.jazztel.es] has quit [Read error: 110 (Connection timed out)]
20090221 11:02:06-!- loonybot [n=loonybot@79.139.247.150] has joined #wesnoth-dev
20090221 11:02:50-!- loonycyborg [n=sergey@79.139.247.150] has joined #wesnoth-dev
20090221 11:09:35< Ivanovic> moin
20090221 11:10:07< boucman> moinmoin
20090221 11:11:21< Mordante> hi Ivanovic 
20090221 11:12:54< Ivanovic> dfranke: uhm, somehow i do not see any messages you write
20090221 11:13:01< Ivanovic> that is: they don't appear in my logs either
20090221 11:13:33< dfranke> Ivanovic: are you seeing this?
20090221 11:13:50< Ivanovic> yes
20090221 11:14:11< dfranke> odd...
20090221 11:15:03< Ivanovic> interesting, my bouncer got those messages in its log
20090221 11:15:22-!- ilor [n=user@213-238-122-211.adsl.inetia.pl] has joined #wesnoth-dev
20090221 11:15:23< Ivanovic> just konversation has not gotten your messages when replaying the incoming stuff from the bouncer
20090221 11:16:25< dfranke> that's... pretty whacked.
20090221 11:16:56< dfranke> I'm not setting the evil bit, I swear :-)
20090221 11:19:53< Ivanovic> and yes, bugs reported as private are only shown to the reporter (when he was logged in, not 100% sure about that part) and to members of the project)
20090221 11:20:54< dfranke> ok, I'll file a bug for the new exploit then
20090221 11:21:14< Mordante> thanks
20090221 11:21:22< boucman> I think we got ouselves a couple release blockers...
20090221 11:21:23< Ivanovic> jupp, when not logged in at gna, i don't see the private report
20090221 11:22:00< Ivanovic> so, uhm, fix the blockers and i can get 1.5.11 out on sunday
20090221 11:22:02< Ivanovic> ;)
20090221 11:22:27< dfranke> why are there any blackhats out there, when it's so much more fun just to drive the devs crazy? :-)
20090221 11:23:03< Mordante> it proves again security is hard
20090221 11:23:21< Mordante> you want to release tomorrow?
20090221 11:24:12< dfranke> but I wouldn't call these blockers since they aren't regressions.  They're probably as old as multiplayer.
20090221 11:24:32< Ivanovic> Mordante: if it makes sense because "lots of stuff" was changed/fixed: yeah
20090221 11:25:28< Ivanovic> and looking at the tracker it seems like *many* bugs were fixed in the last week
20090221 11:25:41< Mordante> dfranke they can get blocker once discovered, we had more security issues which took years to surface
20090221 11:26:03< Mordante> yes a lot was fixed, made a lot of progress on the dialogs :-)
20090221 11:26:23< Mordante> still have a big fix pending, but need to think about how to fix it
20090221 11:26:30< Mordante> (won't manage this weekend)
20090221 11:29:21< boucman> and our network layer has been heavily reworked in 1.5 too
20090221 11:30:53< Mordante> yep
20090221 11:38:34-!- yann [n=dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has joined #wesnoth-dev
20090221 11:39:10< boucman> zookeeper: btw you didn't open the bug for [item] i'd be very interested in a description of what you want it to actually look like
20090221 11:40:05< dfranke> alright, posted.  #13037
20090221 11:40:52< Mordante> what to do with http://gna.org/bugs/?12638 Andrea has a point, do we have any dev using python 2.6?
20090221 11:41:15< Mordante> I expect 1.6 to be used while using Python 2.6
20090221 11:41:28< loonycyborg> Mordante: I'm not going to use it until it's unmasked.
20090221 11:41:38< ilor> I have python 2.6 on windows
20090221 11:41:50< Mordante> loonycyborg any eta for that?
20090221 11:42:34< loonycyborg> Mordante: No idea. I don't know what Gentoo people doo good enough.
20090221 11:43:04< Mordante> ilor can you reproduce the compilation problem on Windows?
20090221 11:43:09< loonycyborg> I only know that it's masked because it breaks other packages, and will remain so until they're fixed.
20090221 11:43:44< boucman> i think we should fix this pre 1.6
20090221 11:43:50-!- Dragonking [n=dk@dedikerad/dragonking] has joined #wesnoth-dev
20090221 11:44:03< Mordante> ok, that means nobody knows when 2.6 will be in Gentoo
20090221 11:44:08< boucman> I expect python 2.6 should be standard in the not so distant future, and we need to be as ready as we can
20090221 11:44:25< Dragonking> Hi everyone
20090221 11:44:29< Mordante> hi Dragonking 
20090221 11:45:01< Mordante> boucman agreeed
20090221 11:45:06< Mordante> agreed*
20090221 11:46:28< Ivanovic> i think it might take a while till this python version hits gentoo
20090221 11:46:51< Ivanovic> since IIRC portage is built upon python and they will probably have to adjust "some things" to make it work
20090221 11:47:16< Ivanovic> http://gentoo-portage.com/dev-lang/python
20090221 11:47:47< Ivanovic> it is in the tree, but hardmasked at the moment (that is you can only get it when manually activating it, hardmasking tends to have a "good reason")
20090221 11:47:59< zookeeper> boucman, i'll try to do that today. not that i see what could possibly be unclear about it :x
20090221 11:48:39< Ivanovic> regarding the python 2.6 status in gentoo: http://bugs.gentoo.org/show_bug.cgi?id=230205
20090221 11:48:44< zookeeper> (move a merman to the trident in BoP and don't pick it up for an easy test case)
20090221 11:52:21< Mordante> ok really seems not to get into gentoo soon
20090221 11:58:51< ilor> Mordante: no warnings on VC9 or gcc 4.2.1/mingw
20090221 12:06:44-!- Elvish_Pillager [n=eli@66-189-14-42.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090221 12:07:39< Mordante> ok
20090221 12:15:01-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090221 12:16:53< ilor> Mordante: most likely a gcc4.3/python2.6 thing, I don't have a 4.3 around in mingw but I'll try to test later today in linux
20090221 12:20:15< esr> Ivanovic: We have 30 fixed bugs since 1.5.10 (I'm in full bug-hunt mode now and accounted for 10 of them myself).  I think that is ample justfication for shipping another point release.
20090221 12:21:23< Mordante> ilor ok you also run python 2.6 there?
20090221 12:21:32< ilor> I think so
20090221 12:21:40< ilor> if not, I'll upgrade
20090221 12:21:53< Mordante> can I assign the bug to you?
20090221 12:22:22< ilor> go ahead, but I can't promise to be able to fix a weird gcc behavior, only look into it :)
20090221 12:23:04< boucman> esr: if you have some time, could you test and commit the patch in bug 13036
20090221 12:23:28< boucman> I assigned it to me, but i might be a little bit short on time today
20090221 12:24:51< esr> Furthermore, I've been looking through the remaining ones re-prioritizing them by how they affect user experience . (In one case this involved downgrading one I reported; I upgraded one or two others.)   I find that we now have just 6 bugs rated Blocker or Important; if wee could fix those, and clean up another couple minor issues like getting the whole tree wmllint and wmlscope-clean, I'd...
20090221 12:24:53< esr> ...sthen say 1.6 is rready.
20090221 12:25:41< boucman> :)
20090221 12:25:48< boucman> 6 is a good number
20090221 12:25:56< boucman> did you count 13031 ?
20090221 12:27:02< esr> That's not on my hot-list. Let me look at it.  And yes, I'll merge dfranke's access.  
20090221 12:27:45< dfranke> Like I said before, I don't think either of my security bugs should block release.
20090221 12:27:54< dfranke> because they're already present in past stable versions.
20090221 12:27:57< boucman> it's new, and a security vulnerability, so it should be in
20090221 12:28:52< esr> Counting security bugs, that's 8.  Still not a bad place to be.
20090221 12:29:16< esr> I meant "I'll merge dfranke's patch".
20090221 12:29:53< esr> Freudian slip, I was thinking that if he keeps generating fixes like this we're going to have to give him dev access :-)
20090221 12:30:45< nital> is anything being done about that glitch? http://www.wesnoth.org/forum/viewtopic.php?f=4&t=24082
20090221 12:31:19< boucman> dfranke: what/how many patches did you alredy submit and get commited ?
20090221 12:31:37-!- thespaceinvader [n=chatzill@91.108.177.73] has joined #wesnoth-dev
20090221 12:31:44< boucman> nital: first time I hear of it, will look
20090221 12:31:45< nital> I'm asking because I can't see it on the bugtracker and zookeeper says in that topic it should be done before 1.6
20090221 12:31:52< dfranke> boucman: I think this is only the second patch I've submitted, but I've generated a slew of bug reports.
20090221 12:32:15< dfranke> the other was a fix for UtBS a while back.
20090221 12:32:46< esr> boucman: They're high-quality bug reports - dfranke has a keen eye for trouble.
20090221 12:32:48< dfranke> maybe I contributed some others during 1.3 development.  I forget.
20090221 12:32:56< zookeeper> uh. just got an astrarsearch.cpp:148 assert with r32928
20090221 12:33:06< zookeeper> on enemy turn
20090221 12:33:38< zookeeper> nital, boucman, that's the gryphon glitch.
20090221 12:33:47< esr> zookeeper: Yeah, we've still got pathfinding trouble. That's why I upgraded the known patfinding problem to "bBocker".
20090221 12:33:59< zookeeper> if boucman has forgotten that then he's got a severe case of dementia ;)
20090221 12:34:11< Mordante> yeah I pinged Sirp about it last weekend
20090221 12:34:50< esr> zookeeper: I tested the gryphon glitch (with an acttual gyphon) in THoT a few days ago. I couldn't reproduce it.
20090221 12:35:13< Ivanovic> esr: can you fix+merge http://www.wesnoth.org/wiki/SpellingMistakes#The_Legend_of_Wesmere ?
20090221 12:35:20< Ivanovic> that is: pofix stuff
20090221 12:35:31< Ivanovic> afterwards i will run a pot-update
20090221 12:35:45< esr> Ivanovic: Can do.  
20090221 12:36:21< esr> Give me a few minutes, though, I want to merge dfranke's patch and kill that one.
20090221 12:36:22< Ivanovic> and what about this patch assigned to you, esr? https://gna.org/patch/?1119
20090221 12:36:23< boucman> nital: if it's the gryphon but zookeeper mentionned (I can't see the video) then it's supposed to be fixed
20090221 12:36:25< Mordante> boucman shadow_master also left a glitch in the log yesterday night 22:36 <Shadow_Master> boucman, Mordante: http://shadowm.nfshost.com/foggy.png
20090221 12:36:33< esr> Ivanovic: I'll get to it.
20090221 12:36:52< boucman> yeah, I saw that one
20090221 12:37:00< boucman> not sure what's going on here...
20090221 12:37:10< boucman> what layer is fog supposed to be drawn on ?
20090221 12:37:20< boucman> much higher iirc
20090221 12:38:14< Mordante> over the unit itself
20090221 12:38:34< ilor> boucman: it's a foggy mountrain, not actial fog over a terrain I think
20090221 12:39:24< boucman> ilor: no, the fog of impassable mountain is drawn in the terrain
20090221 12:41:44< CIA-70> esr * r32953 /trunk/src/network.cpp: Fix bug #13036, patch by dfranke.
20090221 12:41:51< Mordante> I can reproduce it here and it's the normal fog
20090221 12:42:20< boucman> thx esr
20090221 12:42:41< boucman> btw, dfranke I didn't ask, but you are in the about.cfg, arn't you ?
20090221 12:42:59< dfranke> not AFAIK.
20090221 12:43:04< boucman> hmm
20090221 12:43:09< boucman> you should be :P
20090221 12:44:17< Mordante> if you supply a patch you certainly have two patches in :-P
20090221 12:45:47< dfranke> I could go in "Miscellaneous contributors", I guess.  I don't think my volume justifies being listed as a programmer yet.
20090221 12:48:22< Mordante> boucman it seems sm's problem is fixed by adding a new layer group :-)
20090221 12:57:06< dfranke> here you go, apply at your leisure I guess: http://wesnoth.pastebin.com/m101426fc
20090221 12:57:19< CIA-70> esr * r32954 /trunk/ (51 files in 4 dirs): Apply LOW spelling fix via pofix.
20090221 12:57:39< esr> Ivanovic: You can run the pot-update now, I guess.
20090221 12:59:04-!- cib0 [n=cib@p4FD0EEC2.dip.t-dialin.net] has joined #wesnoth-dev
20090221 13:01:29< esr> So: Further good news - looking at the 8 serious bugs we have left, our prospects of fixing 7 of them in a fairly short time look good to me (there's one that we'll nail just as soon as I can corner fendrin and beat him into un-breaking a map).  The only one that really worries me is the pathfinding lossage..
20090221 13:01:54< silene> pathfinding lossage?
20090221 13:02:39< esr> Note, this is a dramatic change from just a week and a half ago, at which time I tiold Ivanovic the buglist was a mess and we were nowhere near a releasable state. 
20090221 13:03:03< esr> silene:  #12632
20090221 13:03:17< esr> There may be other bugs near that as well.
20090221 13:03:23< cib0> ah, hey silene 
20090221 13:03:42< dfranke> oh gawds, I just realized it's 4AM
20090221 13:03:51 * dfranke mv /proc/self ~/bed && sleep
20090221 13:04:11< Mordante> night dfranke 
20090221 13:04:36< esr> Night, dfranke.
20090221 13:04:46< silene> cib0: i see you have tried lua; how did it go?
20090221 13:06:07< esr> loonycyborg: Do you have any insight into  bug #12888?  Any prospect of a near-term fix?
20090221 13:06:43< cib0> silene, the preload event didn't seem to work
20090221 13:06:59< cib0> it would work on start, but not when i loaded the game
20090221 13:07:50< CIA-70> boucman * r32955 /trunk/data/core/about.cfg: add dfranke in misc contributors
20090221 13:08:05< cib0> other than that i didn't find any bugs with the lua-tutorial
20090221 13:08:38< silene> cib0: that's strange, i'm sure i tested it, i will test again
20090221 13:09:05< cib0> also, is it possible for a lua tag to contain another lua tag?
20090221 13:09:06< Mordante> zookeeper did you find time for the mountains?
20090221 13:09:09< esr> boucman: I know dfranke from elsewhere.  If he sticks around I think he's going to make a steady, competent dev.
20090221 13:09:28< cib0> there are examples in the tutorial where an event is created in another event, is it possible to do that in lua fully?
20090221 13:09:29< loonycyborg> esr: Have you noticed my comment in that bug?
20090221 13:09:51< esr> loonycyborg: Looking...
20090221 13:10:35< boucman> esr: I think so too, but as a principle, I want a "real code" patch in, not just one liners, but I'm not too worried, he'll get one whenever he finds something that hits his fancy
20090221 13:10:50< esr> boucman: Agreed on all counts.
20090221 13:10:53< loonycyborg> Perhaps some later windows update broken it..
20090221 13:11:08 * loonycyborg is too lazy to install windows updates.
20090221 13:11:15< silene> cib0: yes, it is possible; but it's a bit unfortunate that the inner lua code is just a string containing lua code and not an actual lua code
20090221 13:11:34< esr> loonycyborg: In that case I'll mark it "Works For Me" on your behalf.  
20090221 13:12:18< silene> cib0: i tried with real lua code, but it has an unfortunate effect: you couldn't save a game and then load it on a computer with a different endianness, hence why it has to be a string for now
20090221 13:14:18< zookeeper> Mordante, no, but probably will today. although i wouldn't mind if you refreshed my memory about what was it that's wrong with them, since your previous imagebins have expired :p
20090221 13:14:18< cib0> so that inner string containing lua code will be triggered once the sub-event is triggered? and what is the disadvantage of using a string rather than special syntax?
20090221 13:15:20< ilor> esr: about https://gna.org/bugs/?12711 , I think it's fixed and Espreon confused it with a separate issue
20090221 13:16:04< esr> ilor: OK, mark it Fixed then.  That's good, it's one of the big 8 release blockers.
20090221 13:16:55< CIA-70> mordante * r32956 /trunk/src/gui/widgets/scrollbar_container.cpp: 
20090221 13:16:55< CIA-70> Fix show_content_rect for top and left side.
20090221 13:16:55< CIA-70> The function didn't work as advertised and moved to the bottom right
20090221 13:16:55< CIA-70> instead of top left. This has been fixed.
20090221 13:17:08< CIA-70> mordante * r32957 /trunk/ (changelog src/display.cpp): 
20090221 13:17:08< CIA-70> Fix a drawing glitch with the fog and reachmap.
20090221 13:17:08< CIA-70> The out of hex part of the Gryphon was drawn over the fog and reachmap
20090221 13:17:08< CIA-70> instead of underneath it. Spotted by Shadow_Master.
20090221 13:18:01< silene> cib0: the disadvantage is that you have no upvalues in the inner code; i have alleviated this issue by adding an "args" tag to the lua tag, so it is as powerful but it looks a bit ugly
20090221 13:18:05< boucman> Mordante: quick question, did you fix the menu bug that prevents wesband from wroking in latest dev release ?
20090221 13:18:36< Mordante> zookeeper the gryphon is drawn wrongly over mountains
20090221 13:18:50< Mordante> part of the terrain should be background but is foreground
20090221 13:18:50< cib0> silene, i seem to have missed something, what are up-values? =)
20090221 13:19:02< Mordante> boucman which bug? 
20090221 13:19:22< Mordante> I fixed quite a few and only the ugly looking listboxes remain
20090221 13:19:38< zookeeper> Mordante, right, i just tried it and it looked fine to me...i'll take a closer look a bit later then since i'm afk now
20090221 13:20:44< boucman> Mordante: http://www.wesnoth.org/forum/viewtopic.php?p=340068#p340068 but apparently you fixed it
20090221 13:21:31< silene> cib0: it is the context of a function closure; imagine you have this code: "local b=3; return function() return b end", so this is a function that returns a function that returns b; but b no longer exists once the function has been returned, so it has to be stored in the context of the created function; so the b in the inner function becomes an upvalue
20090221 13:21:56< boucman> zookeeper: i'll assign bug 13007 and patch 1125 to you (the patch fix the bug)
20090221 13:22:04< boucman> could you quick check and commit ?
20090221 13:22:57< CIA-70> esr * r32958 /trunk/data/campaigns/Eastern_Invasion/scenarios/03.An_Unexpected_Appearance.cfg: Apply patch #1121 by turuk, fixing cosmetic recruitment error in EI.
20090221 13:23:01< boucman> who's responsible for EI (assigning patch 1121)
20090221 13:23:21< cib0> silene: so how do you get around that with an args tag?
20090221 13:23:31< esr> boucman: I'm a campaign maintainer.  I just took care of it.
20090221 13:23:40< Mordante> boucman yes I fixed it this week
20090221 13:23:55< boucman> ok, could you assign/mark as fied the patch ? thx
20090221 13:24:12< Mordante> I'm off now, will be back this evening
20090221 13:24:35< esr> boucman: I've marked #1121 done, was there something else you wanted?
20090221 13:24:42-!- Mordante [n=chatzill@roadie.xs4all.nl] has quit ["ChatZilla 0.9.84 [Iceweasel 3.0.6/2009020409]"]
20090221 13:25:13< boucman> nope, that's it
20090221 13:27:31< silene> cib0: something like that: local b = 3; do_something_with { "lua", { code="local arg = ...; return arg.toto", [1] = { "args", { toto = b }}}}
20090221 13:28:55< silene> in other words, it's as if you had typed [lua]code=... [args]toto=3[/args][/lua]
20090221 13:29:18< esr> boucman, zookeeperL I can address bug #13007 with a wmllint change that will automatucally fix future WML.  Shall I do so?
20090221 13:29:59< boucman> no opinion, that's for zookeeper to discuss
20090221 13:31:56< cib0> silene: i think i understand what that code does, but i have no idea what that [1] is doing there
20090221 13:33:32< cib0> also, upvalues in event code or something not even used in WML yet
20090221 13:33:40< CIA-70> esr * r32959 /trunk/data/campaigns/The_Hammer_of_Thursagan/maps/mages_and_drakes.map: 
20090221 13:33:40< CIA-70> Apply patch #1119, THoT map fix by turuk.
20090221 13:33:40< CIA-70> --This line,and those below, will be ignored--
20090221 13:33:40< CIA-70> M trunk/data/campaigns/The_Hammer_of_Thursagan/maps/mages_and_drakes.map
20090221 13:34:05-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090221 13:34:13< CIA-70> mordante * r32960 /trunk/data/core/about.cfg: Fix an invalid about entry.
20090221 13:34:18< cib0> that is, a mix of dynamically inserted variables and variables declared in the current scope
20090221 13:34:55< silene> cib0: it's due to the way wml objects are encoded in lua (note that i will provide some helper functions so that it looks better in most cases); since wml allows you to have several children with the same tag, you can't just type args={toto=3}, so there is an indirection first through a number
20090221 13:35:31< cib0> i see
20090221 13:37:05< cib0> i would have expected something like args = [{toto=3}], but i guess your syntax gets closer to WML
20090221 13:37:31< boucman> cib0: actually, it's very close to perl...
20090221 13:38:16< silene> cib0: in some cases, your syntax wouldn't work, since you would not be able to differentiate [a][/a][b][/b][a][/a] from [a][/a][a][/a][b][/b]
20090221 13:39:09< cib0> right, i even didnt remember you would
20090221 13:39:40< cib0> i think when i first read config.cpp i must have misinterpreted something
20090221 13:40:51< cib0> i had the impression that all tags with the same name would be dumped into one vector and there indeed would be no way to differentiate something like that.. on the other hand, it is supposed to be possible in eventWML
20090221 13:41:50< silene> cib0: there are indeed dumped in a vector, but there is an auxiliary structure that keeps track of the overall order of the tags
20090221 13:42:21< cib0> silene, i see
20090221 13:44:00< cib0> anyway, having lua available in WML seems to be really nice, but it's really important to keep it simple
20090221 13:44:20< cib0> well, WML itself already isnt very simple in some areas
20090221 13:44:31< silene> note that this order is actually used a lot in scenarios: during an event, you never have all the messages first and then all the units moving
20090221 13:44:34< cib0> e.g. scalar and container variables
20090221 13:45:14< silene> yes, it's a bit of a pity
20090221 13:46:58< cib0> i think there are many great WML authors who still don't know how it works
20090221 13:47:14< cib0> simply because theyre no coders
20090221 13:47:20< CIA-70> esr * r32961 /trunk/data/ (3 files in 3 dirs): 
20090221 13:47:20< CIA-70> Address bug #13007. Ordinarily I'd be worried about the risk of UMC breakage
20090221 13:47:20< CIA-70> from renaming macros, but one of these was broken for a long time until recently
20090221 13:47:20< CIA-70> so we're probably not increasing any lossage here.
20090221 13:49:58< CIA-70> silene * r32962 /trunk/src/about.cpp: Removed useless translation markers.
20090221 13:50:34< esr> wesbot: topic
20090221 13:50:36< cib0> but i think lua would be great to make development of advanced stuff easier
20090221 13:50:49< esr> Yay!
20090221 13:51:04< silene> cib0: i hope so :-)
20090221 13:51:09< cib0> e.g. shop utilities, advanced stuff like magic, etc.
20090221 13:53:18< silene> even without going that far; i have read some forum topics where campaign designers had a hard time having dynamic campaign objectives; lua removes this difficulty without having to add a new lua tag (as was suggested in the topics)
20090221 13:53:32< silene> s/lua tag/wml tag/
20090221 13:53:43< cib0> true
20090221 13:54:10< cib0> though thats a problem that could be solved by better WML utilities, too
20090221 13:54:45 * esr likes the idea of lua estensions too.
20090221 13:55:02< esr> s/estensions/extensions/
20090221 13:55:34< boucman> i think a lua type language was long overdue in wesnoth
20090221 13:55:55< boucman> wml is great for structure description, but we need a real programming language here...
20090221 13:57:15< esr> boucman: Heck, I'd like to see lua infiltrate the codebase and replace large chubnks of C++.  Almost as good as Python from my POV.
20090221 13:59:27< silene> esr: for the pathfinding bug, how do you reproduce it?
20090221 13:59:43< cib0> from what ive seen, i >would< prefer lua
20090221 13:59:50< cib0> IF it had OOP :p
20090221 14:01:01< esr> cib0: : The lia book complains you canbn build OOP in Lua pretty trivially by using its table feature the right way.  I believe that -- it could be done in Tcl, and in a lot of ways lua is a better Tcl.
20090221 14:01:12< silene> object-oriented programing? lua even has multiple inheritance, if needed
20090221 14:01:51< esr> silene: There's a savefile and a recipe attached to the bug.
20090221 14:02:57< silene> esr: there are two savegames, i hadn't tried the good one; i can reproduce it now
20090221 14:03:01< cib0> hm.. interesting
20090221 14:03:19< cib0> i never tried myself, but someone i know said OOP in lua would be very messy
20090221 14:04:46< silene> it's messy in the sense that you use a metatable that directly encodes your inheritance rules; but once your object is created, there is no differences with other oo languages
20090221 14:05:27< cib0> well, in that case..
20090221 14:06:01< cib0> yay, another reproducible MP campaign bug found
20090221 14:06:12< cib0> well its not even with MP campaigns because it happens on the first scenario
20090221 14:24:19< Ivanovic> silene: the pathfinding bugs are currently probably those assigned to Sirp
20090221 14:24:33< Ivanovic> just look in the bugtracker for reports assigned to "dave" and you should find them
20090221 14:27:20< Ivanovic> silene: i think esr means this report: https://gna.org/bugs/?12632
20090221 14:27:50< Ivanovic> boucman: you were talking to Sirp about this one, any progress on it? https://gna.org/bugs/?12817
20090221 14:28:09< boucman> not that I know of
20090221 14:28:41< silene> Ivanovic: yes, i know, i'm already on it :-) (my issue was with the savegame, not the report)
20090221 14:28:47< Ivanovic> ah, good
20090221 14:29:25< esr> Ivanovic: Yes that is good, it's the only remaining bug that really worries me.
20090221 14:29:56< Ivanovic> esr: the two newly reported security issues dfranke found worry me a lot more
20090221 14:30:13< Ivanovic> and i would love to see some fixes in before getting out the next beta release
20090221 14:30:26< esr> I think zookeeper tripped over the pathfinding in a different context  a few hours ago.
20090221 14:30:28< Ivanovic> that is: i would like to get it out tomorrow and head for the first rc the week after
20090221 14:30:56< Ivanovic> then get out another rc a week later and try to get 1.6 out the week after
20090221 14:30:59< esr> Ivanovic: I think dfranke might fix those if you giver him a few days.  Just a hunch.
20090221 14:31:25< Ivanovic> so: from now on one beta, two rc and then final 1.6 in about 3 weeks if everything goes as planned and the bugs are wiped out
20090221 14:32:12< esr> Ivanovic: That seems reasonable.
20090221 14:32:40< cib0> 1.6 in 3 weeks?
20090221 14:32:52< esr> If all goes well.
20090221 14:33:02< Ivanovic> cib0: if the open bugs get tackled and everything goes well: yes
20090221 14:33:05< cib0> id give it another 6 months for bug fixing :p
20090221 14:33:35< cib0> https://gna.org/bugs/index.php?13038 is this one duplicate?
20090221 14:33:45< esr> Ten days ahgo I might have said likewise, but we've cleared a *lot* of bugs since.
20090221 14:34:00-!- malumalu [n=malu@hnvr-4dbbffeb.pool.einsundeins.de] has joined #wesnoth-dev
20090221 14:34:09< cib0> esr, whenever you clear a bug, another one slips out of the darkness =)
20090221 14:34:11< esr> Duplicate of what?
20090221 14:34:19< cib0> another bug
20090221 14:34:31< esr> Hmmm....
20090221 14:34:35< cib0> because im thinking someone must have stumbled across it
20090221 14:35:22< cib0> it seems to trigger whenever someone leaves, is droided, and rejoins
20090221 14:35:27< esr> Hmmm...possibly  #11114?
20090221 14:36:14< cib0> esr, nope, that is something different which has been fixed
20090221 14:36:23< cib0> so i guess it's not duplicate
20090221 14:37:00< Ivanovic> running a pot-update now
20090221 14:37:19< cib0> when is the next release planned?
20090221 14:37:27< esr> Bug #11114 has been marked Ready For Test.  If you *know* it's fixed, mark it so.
20090221 14:37:33< Ivanovic> *if* the two reported security bugs are fixed then: tomorrow
20090221 14:37:50< cib0> Ivanovic, ok
20090221 14:41:34-!- malumalu [n=malu@hnvr-4dbbffeb.pool.einsundeins.de] has quit [Remote closed the connection]
20090221 14:44:03-!- Elvish_Pillage2 [n=eli@66-189-14-42.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090221 14:44:35 * Ivanovic is currently busy updating the "release schedule" with those plans
20090221 14:45:34< cib0> Soliton, there?
20090221 14:46:41< Ivanovic> everyone: please check if you are fine with the *new* release schedule / plans for 1.6: http://www.wesnoth.org/wiki/Wesnoth1.6ReleasePlan
20090221 14:48:14-!- Elvish_Pillager [n=eli@66-189-14-42.dhcp.oxfr.ma.charter.com] has quit [Read error: 60 (Operation timed out)]
20090221 14:48:17-!- malumalu [n=malu@hnvr-4dbbffeb.pool.einsundeins.de] has joined #wesnoth-dev
20090221 14:49:09< esr> Ivanovic: Looks OK to me.
20090221 14:49:47< esr> I guess my only to-do left other than bug hunting is the release announcement.  
20090221 14:50:08< Ivanovic> jupp
20090221 14:50:20< Ivanovic> and i would like to have it ready next week
20090221 14:50:34< esr> I think that is doable.
20090221 14:50:34< Ivanovic> then translators at least got two weeks to work on it, which is probably fine
20090221 14:50:58< esr> I've about run out if things I can fix.
20090221 14:51:12< Ivanovic> :)
20090221 14:51:35< esr> (Unless a new bug that's in one of my areas pops up.)
20090221 14:52:32< cib0> esr, make MP campaigns one of your areas and your thirst for bugs shall be satisfied =)
20090221 14:52:34< Ivanovic> esr: one request
20090221 14:52:40< esr> ?
20090221 14:52:44< Ivanovic> with pofix do *not* touch commented out strings
20090221 14:53:20< Ivanovic> that is omit all lines starting with this: #~ 
20090221 14:53:35< esr> I'll have to rewrite it to accomplish that.
20090221 14:53:45-!- Mythological [i=Mytholog@77.28.68.46] has joined #wesnoth-dev
20090221 14:53:50< silene> esr: i found where the bug is, it's in one of the things you committed on behalf of suollo
20090221 14:54:09< esr> silene: I was half expecting that.
20090221 14:54:57< Ivanovic> esr: no way to have some "wrapper" before the lines and not cinsidering them at all when starting with #~ ?
20090221 14:55:06< Ivanovic> should in theory be a rather simple regexp
20090221 14:55:08< Ivanovic> ;)
20090221 14:55:20< silene> (more precisely, there is a division by zero at ai_move:154)
20090221 14:55:56< esr> What I'll have to do is have a little state machine that does subsitutions line by line rather than on the whole file content as now.  This can be done.
20090221 14:56:10< CIA-70> ivanovic * r32963 /trunk/ (530 files in 13 dirs): 
20090221 14:56:10< CIA-70> pot-update (string addition in the editor!!!)
20090221 14:56:10< CIA-70> regenerated doc files
20090221 14:58:17< ilor> Ivanovic: it was a string modification iirc, necessary to fix a documentation bug
20090221 14:59:01< Ivanovic> ilor: it was enough of a change to not appear as "fuzzy" in the po file
20090221 14:59:27< Ivanovic> that is why i explicitly named it as "new string", just to be sure that translators reading the commit mails (not many...) see it
20090221 15:01:43-!- Baufo_laptop [n=thomas@62-47-145-124.adsl.highway.telekom.at] has joined #wesnoth-dev
20090221 15:02:23< ilor> anyone know of a tool that could scan c++ files and see which includes are completely unnecesary?
20090221 15:02:51< zookeeper> esr, r32958 fixes nothing.
20090221 15:03:19< esr> wesbot: log 32958
20090221 15:03:20< wesbot> esr * r32958 : Apply patch #1121 by turuk, fixing cosmetic recruitment error in EI.
20090221 15:03:23< wesbot> URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=32958
20090221 15:03:37< zookeeper> and i already talked about it with someone a while ago, i thought the patch was already marked as invalid
20090221 15:03:49< esr> Wasn't when I saw it.
20090221 15:04:29< esr> I have no objection to it being reverted, I was just trying to clear pending stuff.
20090221 15:04:39< zookeeper> (esr, and if you want to be sure that the persistant misspelling fixes don't break anything, just define both, with one calling the other)
20090221 15:05:09< esr> zookeeper: I'd have done that...but them wmllint would clobber the old version! :-)
20090221 15:05:51< CIA-70> ilor * r32964 /trunk/src/ (dialogs.hpp game.cpp menu_events.cpp): reduce header dependencies (filechooser.hpp)
20090221 15:07:04< ilor> 3 changed lines and suddenly editing filechooser's headers doesn't result in recompiling game events or the ai, or half the codebase
20090221 15:07:45< esr> ilor: I have built a Pythobn script that does that, and I ran it on Wesbnoth.  Want a copy?
20090221 15:07:46< Ivanovic> :)
20090221 15:07:55< ilor> esr: of course!
20090221 15:08:11< loonycyborg> ilor: esr wrote such tool, and it caused a lot of trouble on various platforms :)
20090221 15:09:02< esr> ilor: http://rafb.net/p/17C85e45.html
20090221 15:10:35< ilor> esr: when did you run this on Wesnoth?
20090221 15:10:56< esr> Couple months back.
20090221 15:11:30< ilor> seems worth to try and run it again, I'll have a go 
20090221 15:12:01< CIA-70> zookeeper * r32965 /trunk/data/campaigns/Eastern_Invasion/scenarios/03.An_Unexpected_Appearance.cfg: Revert the non-functional fix of r32958.
20090221 15:13:13< ilor> anyway, what would really be cool would be something that'd give you used symbols count for each include, so it would be easy to spot situations where a forward declaration would be enough, or an extracted smaller header 
20090221 15:13:26-!- alink [n=alink@78.129.18.215] has joined #wesnoth-dev
20090221 15:13:35< alink> hi
20090221 15:13:39< Ivanovic> hi alink
20090221 15:13:55< alink> 'lo Ivanovic
20090221 15:20:09 * ilor kicks CIA-70
20090221 15:20:09< CIA-70> ow
20090221 15:20:23< alink> Mordante: thanks for the warning about the dummy video case
20090221 15:21:02< alink> I think I will move my optimization into to the old get_portion function and remove the new one
20090221 15:21:56< CIA-70> ilor * r32966 /trunk/src/ (4 files in 2 dirs): proper fix for bug #12711 (disable auto-complete in the "save" dialog)
20090221 15:22:15< alink> the sad thing is that i added a protection against missing video surface, thinking "yeah i must take care of fake screen" and wondering why it worked before
20090221 15:23:04< alink> thus completly forgetting that fake screen is not equal to "no-screen" but instead use a 16x16 dummy mini screen
20090221 15:27:18< cib0> ahah.. so the bug doesn't seem to be a bug
20090221 15:28:17< cib0> do you guys think, when a player takes over a side, it should be made human by default or remain AI if it has been droided?
20090221 15:28:40< cib0> personally i think that it should be made human when you use :control explicitly
20090221 15:28:46< cib0> anything else doesn't make sense
20090221 15:59:31< loonycyborg> Soliton: Do you mind if I convert /usr/src/svn-checkouts/wesnoth-mingw32 into an svn checkout?
20090221 15:59:45 * loonycyborg got a permission error again.
20090221 16:02:27< zookeeper> mordante, i don't see anything wrong with gryphs on mountains, unless you count what happens to the ellipses.
20090221 16:14:15< Ivanovic> zookeeper: duplicate, right? https://gna.org/bugs/?13039
20090221 16:20:25-!- nital [n=nital@public963.xdsl.centertel.pl] has quit [Remote closed the connection]
20090221 16:23:39-!- Sapient [n=sapien-x@c-24-126-133-155.hsd1.ga.comcast.net] has joined #wesnoth-dev
20090221 16:23:55< cib0> Ivanovic, https://gna.org/bugs/?13038
20090221 16:24:59-!- malumalu_ [n=malu@hnvr-4dbbffeb.pool.einsundeins.de] has joined #wesnoth-dev
20090221 16:24:59< cib0> Ivanovic, can my fix go into 1.6? it could be considered a bug with the UI, as i dont think it's intended to work that way
20090221 16:25:00< Ivanovic> cib0: you mean if this breaks the feature freeze?
20090221 16:25:06< cib0> yeah
20090221 16:25:10< Ivanovic> yes, i consider this a bug, too
20090221 16:25:27< Ivanovic> when handing over control, i assume that the one getting control will handle stuff "himself"
20090221 16:25:43< Ivanovic> if the receiver wants the ai to play, he should enter the command himself
20090221 16:30:12-!- malumalu [n=malu@hnvr-4dbbffeb.pool.einsundeins.de] has quit [Read error: 110 (Connection timed out)]
20090221 16:30:35< cib0> hm, now that im thinking about it..
20090221 16:31:15< cib0> i made some extra code to have it set to human for :control exclusively
20090221 16:31:25< cib0> but when would you EVER want to have it that way?
20090221 16:31:47< cib0> i mean, when someone leaves, the host gets asked whom to give control
20090221 16:32:05< cib0> and when he choses another player, i dont think he wants it to remain droided, either
20090221 16:33:04< cib0> i might just remove the check and have it set to human player in general, but im not sure whether that would break anything
20090221 16:36:16< Soliton> loonycyborg: could you just move stuff over to wesnoth-git and get rid of wesnoth-mingw32?
20090221 16:36:31< Soliton> cib0: there now.
20090221 16:36:42< cib0> hey Soliton 
20090221 16:36:47< cib0> i could use some help with this
20090221 16:37:05< cib0> im not too familiar with server stuff
20090221 16:37:48< loonycyborg> Soliton: That can be done if noone wants to run native compiles in wesnoth-git.
20090221 16:38:27< Soliton> loonycyborg: can't you setup different targets for that?
20090221 16:38:40< Soliton> or build modes or whatever.
20090221 16:38:52< cib0> Soliton: server/game.cpp:464
20090221 16:39:34< cib0> do you think that line is responsible for sides remaining droided even when control is changed?
20090221 16:40:22< loonycyborg> Soliton: I can, but I'd rather not to
20090221 16:40:27 * loonycyborg is lazy :P
20090221 16:42:19< loonycyborg> But this won't fix the permission problem anyway..
20090221 16:42:55< Soliton> cib0: yes.
20090221 16:43:22< cib0> Soliton, what is it needed for? it doesn't make any sense for the cases i can think of
20090221 16:43:25 * loonycyborg is getting permission errors in wesnoth-git too
20090221 16:44:21< Soliton> cib0: to leave a side droided when you give it to someone else naturally.
20090221 16:45:03< cib0> Soliton, why would you ever want that?
20090221 16:45:54< Soliton> why wouldn't you?
20090221 16:46:21< cib0> because i give sides to be played by other players
20090221 16:46:42< Soliton> then undroid them.
20090221 16:46:44< cib0> if i want the side to be controlled by AI, there is no need to give it to someone else
20090221 16:48:30< cib0> i don't really consider this solution practical and i think most players will agree with me on that
20090221 16:49:43< zookeeper> Ivanovic, maybe it is. i don't know of which report, though.
20090221 16:50:18-!- mjs-de [n=mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20090221 16:51:04< Soliton> i don't think replacing an AI with a player is a common occurence.
20090221 16:51:11< Ivanovic> ah, okay, it was "only" in the forums
20090221 16:51:14< Ivanovic> http://www.wesnoth.org/forum/viewtopic.php?f=4&t=24180&p=340784&hilit=kamikaze.py#p340784
20090221 16:51:26< Ivanovic> i think Dragonking wanted to look at it, right?
20090221 16:51:41< cib0> when a player uses :control, he wants to give his side to another player, not another client, he might not even be aware of the difference between between local and networked AIs
20090221 16:51:43< Soliton> also if you control AI sides of a different team you will get their team message.
20090221 16:51:50< Ivanovic> Dragonking: so this one is for you (as in "iinstead of python ai, use formula stuff)? https://gna.org/bugs/index.php?13039
20090221 16:52:55< Dragonking> Ivanovic: Seems so
20090221 16:53:13< cib0> Soliton, that is a different issue though
20090221 16:53:13< Dragonking> I looked at it, I'll need to rewrite move formula function
20090221 16:53:43< Dragonking> Problem is: my PC is dieing :/  Either mobo or CPU, I'm afraid it's CPU...
20090221 16:53:54< Dragonking> So I'm working on my notebook now and it's so sloooooooow
20090221 16:54:18< Ivanovic> Dragonking: okay, assigned it to you
20090221 16:54:41< Dragonking> ok
20090221 16:54:50< Ivanovic> Dragonking: which mainboard/cpu do you have in your normal box?
20090221 16:55:14< Dragonking> Gigabyte mobo with AM2+ slot, and CPU is Athlon 3800X2
20090221 16:55:36< Dragonking> And I think it's cpu fault, when I stress CPU PC just dies
20090221 16:55:51< cib0> Soliton, i think the player should be able to ignore on what client the AI runs
20090221 16:55:55< Ivanovic> have you tried raising the voltage for the ram?
20090221 16:56:06< Ivanovic> this is not sooo uncommon a problem
20090221 16:56:24< Dragonking> Ivanovic: No, should I?  But I have been using this PC for a while and it is fresh problem
20090221 16:56:45< Ivanovic> Dragonking: in general ram voltage tends to be lower than the modules really need
20090221 16:56:58< Ivanovic> have a look at the sticker on you ram and how much voltage it needs
20090221 16:57:05< Ivanovic> set this amount +0.1V
20090221 16:57:13< Dragonking> ok, will try
20090221 16:57:32< Ivanovic> (i had to do so to be able to play gothic 2 ages ago, it was stressing ram too much)
20090221 16:57:47< Dragonking> But memory tests are ok
20090221 16:57:53< cib0> it happens quite often that a player side is droided because someone left and then the side is given back to a player later
20090221 16:57:53< Dragonking> All memtest stuff is passed
20090221 16:58:04< Ivanovic> Dragonking: all tests there passed for me, too
20090221 16:58:11< cib0> it happens less often that you want someone else to control your AI side
20090221 16:58:13< Ivanovic> though increasing ram voltage slightly fixed the problems
20090221 16:58:18< cib0> in fact, it never happened to me at all
20090221 16:58:19< Dragonking> ok
20090221 16:58:53-!- TusharGhosh [i=d2d40558@gateway/web/ajax/mibbit.com/x-1f82287037c48c90] has joined #wesnoth-dev
20090221 16:59:56< Soliton> cib0: i wouldn't want my side back if the host droided while i was gone.
20090221 17:01:11< cib0> Soliton, do you play stuff like survivals, RPGs or MP campaigns?
20090221 17:01:30< Soliton> yes.
20090221 17:01:51< cib0> im just saying it happens, otherwise it would never have been a problem in the first place
20090221 17:01:52-!- Corvvs [n=michaelm@pool-71-163-211-230.washdc.east.verizon.net] has joined #wesnoth-dev
20090221 17:02:15-!- TusharGhosh [i=d2d40558@gateway/web/ajax/mibbit.com/x-1f82287037c48c90] has left #wesnoth-dev []
20090221 17:02:17-!- TusharGhosh [i=d2d40558@gateway/web/ajax/mibbit.com/x-1f82287037c48c90] has joined #wesnoth-dev
20090221 17:02:23< loonycyborg> Soliton: Would you at least run chmod 775 -R /usr/src/svn-checkouts/wesnoth-git/.git/objects/* ? :P
20090221 17:02:30-!- TusharGhosh is now known as tushar_hailstone
20090221 17:02:35< cib0> whereas you wanting to give your AI to someone else does almost never happen
20090221 17:02:54< cib0> and if it happens, the player who takes over just droids on his turn, what's the big deal?
20090221 17:03:26< cib0> like this, though, if you forget to droid, the player will lose another turn to the AI
20090221 17:05:05< cib0> by the way, a very simple situation where it can happen is.. a player's connection breaks, the host is confused and just picks the first item on the list(AI) and then the player reconnects and gets back his side
20090221 17:05:51< cib0> which will then be played by AI if the host forgot to droid it beforehand
20090221 17:06:41< Soliton> well, then make the client undroid his side when he gets a new one.
20090221 17:07:18< cib0> the client as in the program or as in the player?
20090221 17:07:47< tushar_hailstone> Ivanovic, i am interested in becoming a Summer of Code student for GSOC 2009 and get involoved in Battle of Wesnoth for the same
20090221 17:08:13< Soliton> cib0: the program.
20090221 17:08:22< Ivanovic> tushar_hailstone: that is a good thing, for sure
20090221 17:08:37< Ivanovic> tushar_hailstone: have you played the game already a little?
20090221 17:08:50< Ivanovic> which areas are you in general interested in working at?
20090221 17:09:07< tushar_hailstone> Yes, although not the multiplayer one!!
20090221 17:09:46< Ivanovic> there are different player types, no matter if multiplayer only or single player only or both
20090221 17:10:02< tushar_hailstone> My general interests lie in innovating new techniques ivanovic
20090221 17:10:29< tushar_hailstone> I although dont have much experience in involvement in big projects as this
20090221 17:11:03< Ivanovic> hehe, innovating new techniques is, uhm, rather broad, this can be anything
20090221 17:11:14< Baufo_laptop> btw, do we have an ideas page for soc already?
20090221 17:11:18< Ivanovic> but in terms of wesnoth, what area are you interested in
20090221 17:11:24< Ivanovic> Baufo_laptop: we have started working on it
20090221 17:11:40< tushar_hailstone> Wesnoth has a potential for beating most closed source games
20090221 17:12:04< tushar_hailstone> I infact have started to implement it in our LAN ivanovic
20090221 17:13:08< tushar_hailstone> I frankly state that I am here to gain experience
20090221 17:13:16< Ivanovic> that is a good approach
20090221 17:13:22< tushar_hailstone> I have gone through and joined your Forums
20090221 17:13:59< Ivanovic> you might want to have a look at this wiki page as of "possible big ideas" for summer of code projects: http://www.wesnoth.org/wiki/SummerOfCodeIdeas
20090221 17:14:22< tushar_hailstone> I also read about the latest developments that you are planning at present
20090221 17:14:31< Ivanovic> that is: especially have a look at the easyCoding page since some "easier" tasks are listed there as a good start to get into the code
20090221 17:14:59< Baufo_laptop> tushar_hailstone, what do you mean by "implement it in our LAN"?
20090221 17:16:00< tushar_hailstone> I went through the your Summer of Code page where there which tells that I gethave my forum nick marked as a Summer of code student 
20090221 17:16:20< Ivanovic> tushar_hailstone: i won't do so before we are accepted in this years summer of code
20090221 17:16:35< Ivanovic> that is: you are *really* welcome to have a look at possible things to do
20090221 17:16:52< Ivanovic> and a good start is for example having a look at a feature that you think wesnoth does miss
20090221 17:17:03< tushar_hailstone> Actually we group of friends have started a hub and we provide opensource games and softwares in our college 
20090221 17:17:14< Ivanovic> or help fixing a bug that annoys you, stuff like this
20090221 17:17:50< tushar_hailstone> Yes and what about the wiki page listed there
20090221 17:18:23< Ivanovic> the stuff on the wiki page with "summerofcodeideas" are "big, long term projects"
20090221 17:18:43< Ivanovic> they do all take a lot of time and thus are bad as "initial contact" with the wesnoth code
20090221 17:19:02< tushar_hailstone> No about the idea of starting your own wikim page
20090221 17:19:29< Ivanovic> first you should know what you want to work on before really starting a wiki page
20090221 17:19:30< Ivanovic> ;)
20090221 17:20:15< Ivanovic> and it might well be that wesnoth will not get into SoC, so far the project that will participate are not announced (signing up has not even started so far!) and it is only known that SoC this year will be smaller than last year
20090221 17:20:37< Soliton> cib0: well, i guess we can do it implicitely as well but maybe add a note that the side is not droided anymore. i dislike having the server care about AIs in the first place though. it's not like there is any guarantee the client isn't lying...
20090221 17:21:02< tushar_hailstone> Wesnoth does deserve to be a part of Soc though
20090221 17:21:30< Ivanovic> if you have me decide on the projects getting in: sure
20090221 17:21:35< Ivanovic> but i am not the one deciding
20090221 17:21:54< tushar_hailstone> What are the areas of development ivanovic, I mean how can students like me help out in a concrete way?
20090221 17:21:55< Ivanovic> and so far nothing is decided
20090221 17:22:28< Ivanovic> tushar_hailstone: like i said: have a look at the easy coding tasks, have a look at a bug that annoys you, have a look at a feature you feel is missing
20090221 17:22:40< Ivanovic> in general you have to know what area you want to work on
20090221 17:22:45< Ivanovic> this can basically be anything
20090221 17:23:29< Ivanovic> it can be something with WML or it could be helping with the lua implementation silene is working on or it could be working on a scenario editor or improving the map editor or makting the AI better/easier scriptable, ...
20090221 17:23:57< tushar_hailstone> Yes those are areas of my liking
20090221 17:23:58< cib0> Soliton, i did something like that at first, but it didn't work and it was sort of out of place, so im doing it at the client now
20090221 17:24:01< Ivanovic> you could help with improving the addon server or the mp server or the statistic system shown on the website
20090221 17:24:16< Ivanovic> as you see, there is *lots* of stuff possible
20090221 17:24:33< cib0> i think the AI logic has to be improved upon in the long run
20090221 17:24:40< Ivanovic> it just depends on what you like as well as what you know already (programming and other skills) and want to learn/improve
20090221 17:25:01< tushar_hailstone> Yes, but that requires high proficiency cib0
20090221 17:25:13< Soliton> cib0: what AI logic?
20090221 17:25:35< tushar_hailstone> What are the duties of a mentor, ivanovic?
20090221 17:25:40< boucman> cib0: the AI as a whole is too much work for a GSOC, we have to cut it in manageable pieces
20090221 17:26:28< cib0> er, oops, i meant the logic how AI controllers are handled =)
20090221 17:26:47< Ivanovic> tushar_hailstone: what do you mean with this?
20090221 17:26:56< Ivanovic> in general the mentors will provide a helping hand
20090221 17:27:02< Soliton> what's to improve besides what you're doing nowß
20090221 17:27:14< Soliton> s/ß/?/
20090221 17:27:26< Ivanovic> they will help you when you got an obvious thinking mistake and will also help you find your way in the wesnoth code
20090221 17:27:51< tushar_hailstone> all this right here in the irc chat?
20090221 17:27:56< Ivanovic> jupp
20090221 17:28:18< Ivanovic> since this is a worldwide thingie visting is difficult
20090221 17:28:28< tushar_hailstone> thats true
20090221 17:28:36< Ivanovic> and doing stuff via phone can be a lot more problematic due to language problems/dialects
20090221 17:28:54< Ivanovic> our main channel of communications for wesnoth is irc
20090221 17:29:05< Ivanovic> beside this of course the forums and mails can be used
20090221 17:29:16< Ivanovic> but for "coding" most stuff is in general discussed in irc
20090221 17:29:21< tushar_hailstone> Yes, I have joined your forum
20090221 17:29:39< tushar_hailstone> How to join your Mail list?
20090221 17:29:57< Ivanovic> the forums are in general mainly a working ground for artists as well as "content creators"
20090221 17:29:59< alink> also don't forget that irc char are logged
20090221 17:30:06< alink> s/char/chat
20090221 17:30:08< cib0> Soliton, i don't know, but the fact that what im doing now doesn't fit within the system at all, even though its just changing what one command does, makes me think that theres something wrong
20090221 17:31:17< Soliton> how does it not fit?
20090221 17:31:51< Ivanovic> tushar_hailstone: the mailling lists are listed here: https://gna.org/mail/?group=wesnoth
20090221 17:32:12< Ivanovic> that is: wesnoth-releases is not relevant at all (no idea why it is there at all, probably some gna thingie)
20090221 17:32:39< Ivanovic> the one where "some" dev stuff is posted on (like release plans and "broader" discussions) is wesnoth-dev
20090221 17:34:04< Dragonking> Ivanovic: ok, set to 1,9V  ( 1.8 + 0.1 )
20090221 17:34:10< Dragonking> Will try to do some testing now
20090221 17:34:32< Ivanovic> Dragonking: good luck, i hope it helps
20090221 17:34:45< Ivanovic> (never underestimate the impact of ram problems, they are nasty...)
20090221 17:34:52< Dragonking> Yup
20090221 17:35:10< Ivanovic> tushar_hailstone: i hope i was able to help you a little
20090221 17:35:24< Ivanovic> tushar_hailstone: in general just look at something you want to work on and ask about it in here
20090221 17:35:26< zookeeper> tushar_hailstone, so how long have you been playing wesnoth?
20090221 17:36:13< tushar_hailstone> Actually, it has been the last 3 months
20090221 17:36:27< tushar_hailstone> I liked Age of Empires quite a lot
20090221 17:37:01< tushar_hailstone> But since I switched to Open Source completely, I queried for an alternate strategy game
20090221 17:37:25< zookeeper> ok, i was just curious, since it's good to be familiar with what you're intending on working with ;)
20090221 17:37:35< boucman> tushar_hailstone: did you already manage to compile trunk ?
20090221 17:38:05< tushar_hailstone> Well not yet, I need to do a little bit of research on it
20090221 17:38:33< tushar_hailstone> In fact I always have been wishing to be a contributor to such games ivanovic
20090221 17:39:12< CIA-70> silene * r32968 /trunk/src/ai_move.cpp: Reorganized code and removed a division by zero which induced an assertion failure later. Fix for bug #12632.
20090221 17:39:44< Sapient> thanks, silene
20090221 17:40:39< tushar_hailstone> well ivanovic are you available throughout the day 
20090221 17:41:49< CIA-70> torangan * r32969 /trunk/ (changelog po/wesnoth-units/he.po): hebrew translation update
20090221 17:42:06< tushar_hailstone> can anybody elaborate on complie trunk?
20090221 17:42:20< Soliton> what OS?
20090221 17:42:32< tushar_hailstone> Fedora 10
20090221 17:42:40< Ivanovic> *easy*
20090221 17:42:45< boucman> trivial
20090221 17:43:01< Ivanovic> http://www.wesnoth.org/wiki/CompilingWesnoth#Prerequisites
20090221 17:43:07< CIA-70> alink * r32970 /trunk/src/ (6 files in 2 dirs): Mainly revert r32931 (problem with fake screen) but keep the small optimization
20090221 17:43:07< CIA-70> ilor * r32971 /trunk/src/widgets/file_menu.cpp: revert an improper fix from r32885 and reorder an operation to avoid triggering a vector iterator assert in msvc debug mode
20090221 17:43:07< CIA-70> ilor * r32972 /trunk/src/filechooser.cpp: default to the ".." item in the filebrowser when entering a new directory, fixing bug #13033
20090221 17:43:09< boucman> (we all fear the "windows" answer when asking that question)
20090221 17:44:20< cib0> =)
20090221 17:46:28< tushar_hailstone> Ivanovic if even though wesnoth doesn't get through GSOC, cant I be a part of its development?
20090221 17:46:48< Ivanovic> tushar_hailstone: of course you can
20090221 17:46:56< Ivanovic> tushar_hailstone: we *always* welcome new people
20090221 17:47:10< Ivanovic> summer of code is for us basically a way to get some extra attention
20090221 17:47:19< cib0> right, you can actually be open source and not welcome people..
20090221 17:47:32< tushar_hailstone> I bet you got the attention
20090221 17:48:24< tushar_hailstone> Sirp how did the idea actually come to your mind?
20090221 17:48:50< cib0> i think there is an article about it, but i forgot where
20090221 17:50:24< Soliton> loonycyborg: "<_igel_> I am using gentoo, do you know if there is an up-to-date portage overlay for getting the newest version of BfW?" (#wesnoth)
20090221 17:54:56< cib0> i give up
20090221 17:55:11< cib0> the client solution will lead to the host not being able to choose AI
20090221 17:55:53< Soliton> huh?
20090221 17:55:56-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev
20090221 18:01:49< Ivanovic> okay, updated the compilingwesnoth instructions since it is well possible to use cmake as well as scons...
20090221 18:02:06-!- Mordante [n=chatzill@roadie.xs4all.nl] has joined #wesnoth-dev
20090221 18:02:19< Mordante> hi
20090221 18:02:31< Ivanovic> hi Mordante
20090221 18:02:41< Mordante> hi Ivanovic 
20090221 18:03:03< tushar_hailstone> Ivanovic are you online throughout the day
20090221 18:03:37< boucman> tushar_hailstone: when he's not, there are always people that can answer questions
20090221 18:03:49< boucman> there is a list of people to bug about GSOC in the wiki
20090221 18:04:48< Ivanovic> tushar_hailstone: in general you should ask questions in here
20090221 18:05:08< Ivanovic> that is, have a look at the "people to bug" page and also ping them when asking your questions, just to be sure that they read them
20090221 18:05:22< tushar_hailstone> Got it sir
20090221 18:05:46< Ivanovic> and yeah, it might take some time until you get a reply (since eg i am *often* away for some hours and every night i tend to even sleep for severl hours...)
20090221 18:05:59< tushar_hailstone> Fine sir
20090221 18:06:21< Ivanovic> that is: some of us got their irc clients running all day and will just read and answer things every now and then (like me), others only join when they really are around (like Mordante)
20090221 18:07:04< tushar_hailstone> Sir how would it be if i plan to host a wesnoth game challenge in my college among us students
20090221 18:07:07< Mordante> ha you think I'm around when I'm joined ;-)
20090221 18:07:33< Ivanovic> no need to call me sir at all (me being still a student myself)
20090221 18:07:53< Ivanovic> and of course you can host a wesnoth game challenge at your college
20090221 18:07:55< Mordante> tushar_hailstone and we have a log which most developers read
20090221 18:07:57< Ivanovic> there is no problem with this at all
20090221 18:08:04< tushar_hailstone> which college are you from ivanovic
20090221 18:09:03< Ivanovic> i am at the university in dortmund
20090221 18:09:15< Mordante> Ivanovic I've two real blockers left: fixing the tooltip and fixing the alignment of multiple columns in a listbox
20090221 18:09:29< tushar_hailstone> Germany, right
20090221 18:09:35< Mordante> and one minor issue but I commit a fix after dinner
20090221 18:09:44< Ivanovic> Mordante: okay
20090221 18:10:01< Mordante> the listbox bug is a bit nasty :-(
20090221 18:10:19< Ivanovic> Mordante: if you really want to have them fixed and if they are even listed in the tracker, please mark them as "important" with the comment "blocker for 1.6 for me!"
20090221 18:10:23< Ivanovic> just that i don't miss it
20090221 18:10:45< Mordante> it's on the 1.6 page which I will polish later today again
20090221 18:11:01< Mordante> both are not in the tracker
20090221 18:11:16< Mordante> seems nobody misses the tooltips :-/
20090221 18:12:18< ilor> Mordante: which tooltips?
20090221 18:12:36< Mordante> the big ones in the MP menu
20090221 18:12:47< Mordante> s/menu/dialog/
20090221 18:13:03-!- Baufo [n=thomas@62-47-145-124.adsl.highway.telekom.at] has joined #wesnoth-dev
20090221 18:13:41-!- Baufo_laptop [n=thomas@62-47-145-124.adsl.highway.telekom.at] has quit [Remote closed the connection]
20090221 18:14:09< Sapient> You mean the menu help strings which show at bottom of screen?
20090221 18:14:17< Mordante> Ivanovic we also need to put a verdict on the gamma stuff
20090221 18:14:19< Mordante> Sapient yes
20090221 18:15:08< Sapient> cya later, have fun.
20090221 18:15:08-!- Sapient [n=sapien-x@c-24-126-133-155.hsd1.ga.comcast.net] has left #wesnoth-dev []
20090221 18:15:52< alink> Mordante: I fixed the flood that you reported, tx
20090221 18:16:15< Mordante> thx for fixing it
20090221 18:16:44< Ivanovic> Mordante: so far i have seen no real complaints about the missing gamma slider
20090221 18:16:50< Ivanovic> so i'd say: gone it is!
20090221 18:17:12< Mordante> I haven't heard anything from bruno yet, like to wait for that
20090221 18:17:29< alink> about the gamma stuff, if it's only for dark map at night, maybe an option for using a less dark ToD ? (like boosting it, or using minimum values etc..)
20090221 18:17:47< Mordante> I'll add it to my todo list on the 1.6
20090221 18:18:13< Mordante> I can postpone it a bit, since the revert or fix is unlikely to break anything
20090221 18:18:46-!- tushar_hailstone [i=d2d40558@gateway/web/ajax/mibbit.com/x-1f82287037c48c90] has left #wesnoth-dev []
20090221 18:18:55< Mordante> then I can do it after fixing my two blockers
20090221 18:19:46< Ivanovic> Mordante: i hope you can cope with the release plan
20090221 18:20:07< Ivanovic> since we *really* should be out of feature freeze mid march, so that possible SoC students can get in patches which add features
20090221 18:20:58< Mordante> I hope so too
20090221 18:21:17< Mordante> of course we could branch as well (I know not the best option)
20090221 18:23:28< boucman> at some point, it's better than freezing everything
20090221 18:24:45 * Ivanovic thinks torturing Mordante to finish his stuff is the fastest way
20090221 18:24:47< Ivanovic> ;)
20090221 18:26:16< alink> or make them polish their patches to the point of being a super clean big rewrite of half the engine :p
20090221 18:26:16-!- alink [n=alink@78.129.18.215] has quit [Remote closed the connection]
20090221 18:26:38-!- alink [n=alink@78.129.18.215] has joined #wesnoth-dev
20090221 18:27:47< Ivanovic> that is: those changes by Mordante are the main blocker for 1.6
20090221 18:28:50< Ivanovic> silene: so https://gna.org/bugs/?12632 can be marked as "fixed"?
20090221 18:29:37< silene> once the next version is released, yes
20090221 18:29:49< alink> Mordante: btw, later, when you have a little time, i will have small questions about windows.cpp:218 " SDL_AddTimer(draw_interval, draw_timer, NULL);"
20090221 18:30:00< silene> or do you first "fixed" and lated "closed"?
20090221 18:30:08< Ivanovic> silene: marking should happen directly when the issue is fixed
20090221 18:30:12< Ivanovic> closing after a release
20090221 18:30:21< Ivanovic> ilor: what is the status of this one? https://gna.org/bugs/?12625
20090221 18:30:39< ilor> Ivanovic: in progress
20090221 18:30:46< Ivanovic> ilor: okay
20090221 18:31:01< Mordante> alink what's the question
20090221 18:31:22< Mordante> Ivanovic I always warned about the release schedule :-P
20090221 18:32:32< alink> Mordante: mainly why using a timer. Is it wanted that it often triggers 2 draw events and why wait before sending the first draw event.
20090221 18:33:16< Mordante> I want to make the entire drawing timer based, doesn't really work yet since the upper level items aren't handled yet
20090221 18:33:46< alink> ah ok, so it's for future plans.
20090221 18:33:52-!- cjhopman [n=chris@68-118-159-7.dhcp.mdsn.wi.charter.com] has quit [Connection timed out]
20090221 18:34:03< Mordante> yes but the entire thing is still a WIP
20090221 18:34:29< Mordante> but at least some parts are getting done and polished now :-)
20090221 18:34:55< Mordante> but this project will by my main project for 1.7 and maybe even 1.9...
20090221 18:35:24< Ivanovic> okay, off to hunt for some food as lunch
20090221 18:35:34< Ivanovic> cu
20090221 18:36:08 * Mordante thinks of Ivanovic's torture option...
20090221 18:36:35< Mordante> I think it will also be my main project for 1.11, 1.13, 1.15 and maybe a bit longer ;-)
20090221 18:36:50< Ivanovic> Mordante: WHIMP!!!
20090221 18:36:50< Mordante> cu Ivanovic 
20090221 18:36:52< Ivanovic> ;)
20090221 18:37:38< Mordante> hey I imagine you forcing me to drive to Austria today, that would be horror :-P
20090221 18:37:51< alink> Mordante: lol, in the meantime, what do you think of my 2 suggestions (for now or later): directly push a draw event before launching the first timer, and using SDL_RemoveTimer  when setting draw_interval to 0
20090221 18:38:02< Mordante> 80 km of traffic jams...
20090221 18:38:39< Baufo> Mordante, really, where?
20090221 18:38:49< Mordante> alink let me think of it, dinners ready well be back in about 10 minutes
20090221 18:39:14< alink> Mordante: ok, no hurry, that's silly tiny things ;)
20090221 18:39:50< Mordante> Baufo from Bavaria to Austria
20090221 18:41:21 * alink still trying to optimize wml dialogs (that timer thing often delay it of  ~30 ms, which is huuuuge <:o)  )
20090221 18:42:06< Dragonking> Ivanovic: Hm, so far every test is passed... tho I'm still far from announcing success cause it was really a random error
20090221 18:42:14< Dragonking> More tests are needed :)
20090221 18:44:00-!- happygrue [n=Wintermu@c-67-176-145-41.hsd1.in.comcast.net] has joined #wesnoth-dev
20090221 18:48:58< alink> mmh, i still have my speed move problem with trunk
20090221 18:49:28< boucman> alink: ??
20090221 18:49:29< alink> does someone can use slow speed ?
20090221 18:49:53< boucman> yeh, I have no problem
20090221 18:50:07< alink> in preference > general > accelerate speed ON   + set speed on minimum
20090221 18:50:31< alink> more are then super speedy instead of slow  here
20090221 18:50:42< alink> s/more/moves
20090221 18:50:51< boucman> hmm
20090221 18:50:58< boucman> I see it now, that is very weird
20090221 18:51:12< Mordante> esr the header cleanup script does it already work on header files as well?
20090221 18:52:23< boucman> alink: I have an idea, let me check
20090221 18:53:03< alink> boucman: ok thqnks, note that undo/redo moves seems slowed correctly
20090221 18:53:19< isaac> p
20090221 18:53:48< boucman> hey isaac
20090221 18:54:51< Mordante> zookeeper regarding the mountains it was a skeletal dragon instead of a gryphon, sorry. Just move one in the test scenario to 3,2
20090221 18:56:58-!- busfahrer` [n=busfahre@dsl72bA207.sdtnet.de] has joined #wesnoth-dev
20090221 19:01:03-!- tushar_hailstone [i=d2d40558@gateway/web/ajax/mibbit.com/x-e43e934d7d277c06] has joined #wesnoth-dev
20090221 19:03:14< Mordante> alink I wanted to keep the timer code simple since it's supposed to run constantly in the future
20090221 19:05:20< Mordante> but we could do this trick http://wesnoth.pastebin.com/m71ad820f to make the first drawing faster
20090221 19:05:30< alink> mmmk ok, i think i am talking about one more line before starting it and another for stopping it, but i didn't tested it, so maybe i forget something
20090221 19:05:59-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit [Read error: 145 (Connection timed out)]
20090221 19:06:13< alink> yes should probably work
20090221 19:06:35< alink> but may push several draw events
20090221 19:06:36< Mordante> alink what do you mean then?
20090221 19:07:03< boucman> Mordante: what's that timer thing ?
20090221 19:07:06< Mordante> 1 draw event should be enough
20090221 19:07:14< alink> something like adding draw_timer(0,0) but not sure about the parameters
20090221 19:07:24< Mordante> timer to trigger draw events in the new gui code
20090221 19:07:41-!- Shadow_Master [n=imorelle@190.22.106.231] has joined #wesnoth-dev
20090221 19:07:54-!- tushar_hailstone [i=d2d40558@gateway/web/ajax/mibbit.com/x-e43e934d7d277c06] has left #wesnoth-dev []
20090221 19:08:03< boucman> Mordante: would that replace the 10ms SDL_Delay ?
20090221 19:08:46-!- grzywacz [n=grzywacz@moinmoin/developer/karol] has joined #wesnoth-dev
20090221 19:08:55< Mordante> no that should be replaced with a wait for event call (can't remember the exact SDL name)
20090221 19:09:02< boucman> ok
20090221 19:09:17< Shadow_Master> boucman: re your suggestion for kitty. Note that the new sprite's wings are not translucent at all
20090221 19:09:30< Mordante> but I ran in some problems while testing and seems I forgot about it until now ;-)
20090221 19:09:37< boucman> oops
20090221 19:10:03< Mordante> so thanks for the reminder 
20090221 19:10:13< alink> Mordante: i am not sure about using 1ms timer, it will cycle at high rate and thus may pushing several events
20090221 19:10:57< Mordante> well several events don't really hurt and the race window is rather small
20090221 19:11:06< Mordante> so we can try with it
20090221 19:12:00< alink> ok indeed, and i see that it has a granularity of 10ms
20090221 19:12:13-!- Sirp_ [n=me@c-71-198-3-104.hsd1.ca.comcast.net] has joined #wesnoth-dev
20090221 19:12:28< Mordante> yup
20090221 19:12:39< Sirp_> hi
20090221 19:12:43< Mordante> hi Sirp_ 
20090221 19:16:36-!- cjhopman [n=chris@wireless33.cs.wisc.edu] has joined #wesnoth-dev
20090221 19:19:21< Shadow_Master> CONFLICT (content): Merge conflict in changelog
20090221 19:19:22< Shadow_Master> CONFLICT (content): Merge conflict in data/tools/wmllint
20090221 19:19:48-!- alink_ [n=alink@78.129.18.215] has joined #wesnoth-dev
20090221 19:19:51 * Shadow_Master . O o (why people always edit the most unexpected files in the most unexpected manners at the most unexpected times?!)
20090221 19:20:01< zookeeper> Mordante, roger
20090221 19:20:10< Shadow_Master> esr: what is your latest change in wmllint about?
20090221 19:20:36< alink_> mmh my connection is flacky today
20090221 19:22:00< Shadow_Master> # Changed just before 1.5.11 ("LABEL_PERSISTANT", "LABEL_PERSISTENT"),
20090221 19:22:14< Shadow_Master> ......
20090221 19:23:00< zookeeper> Mordante, oh, that one...
20090221 19:23:10< Mordante> yes that one 
20090221 19:23:35 * zookeeper scratches head
20090221 19:24:22< alink_> Mordante: still not important, just remark, but about sending several draw events, i notice that twindow::draw() has no framerate control, and thus may update+flip at high rate
20090221 19:26:18< Mordante> only with coloured cursor every frame is a flip
20090221 19:26:27< zookeeper> Mordante, AFAICT they're given a terrain graphics layer of 0, but i don't know how that interacts with the whole drawing system
20090221 19:26:28< Mordante> otherwise something needs to be dirty
20090221 19:27:08< zookeeper> Mordante, what layer should they be in (ignore that comment about 0)?
20090221 19:27:29< Mordante> the background terrain so the unit is drawn in front
20090221 19:27:45< zookeeper> number?
20090221 19:28:02< alink_> aah ok sry, i assumed that get_rect() was always something.
20090221 19:28:10 * alink_ will stop to annoy Mordante and will study more the code before speaking ;)
20090221 19:28:21-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev
20090221 19:28:34< Mordante> alink_ well I like feedback :-)
20090221 19:28:40< Mordante> zookeeper 0
20090221 19:29:07< zookeeper> Mordante, ok, i'll fiddle with them a bit
20090221 19:29:14< Mordante> thanks
20090221 19:33:08< Sirp_> does anyone know, do ink cartridges on a printer really 'expire' or is that just a money making scheme by manufacturers to get people to buy more?
20090221 19:33:56< alink_> Sirp_ well at least i hope that it doesn't expire after being printed <:o)
20090221 19:34:20< Sirp_> alink_: well the claim is that using it can damage the printer
20090221 19:34:46< loonycyborg> Sirp_: I remember reading something about that on slashdot..
20090221 19:34:59< Sirp_> loonycyborg: and what was the conclusion? :)
20090221 19:35:34< Soliton> maybe the ink can dry out and clog stuff?
20090221 19:35:41< ilor> Sirp: money making scheme
20090221 19:35:44< Ivanovic> re
20090221 19:35:49< alink_> loonycyborg: i remember something similar but about the fill value of ink cartridges
20090221 19:36:08-!- stikonas [n=and@ctv-213-164-101-51.vinita.lt] has joined #wesnoth-dev
20090221 19:36:15< Ivanovic> hi Sirp_
20090221 19:36:17-!- alink [n=alink@78.129.18.215] has quit [Read error: 110 (Connection timed out)]
20090221 19:36:35< alink_> like the machine declare it's empty before that it's really empty
20090221 19:36:41< ilor> Sirp: worst thing that can happen is that the uh, bit where the ink comes out, gets dry and doesn't work well which will make the prints look pretty bad
20090221 19:36:56< Ivanovic> Sirp_: it would be *really* great if you could have a look at fixing this blocker over the weekend: https://gna.org/bugs/index.php?12817
20090221 19:37:25< Ivanovic> i think boucman already gave you an idea where to look, right?
20090221 19:37:28-!- JW1 [n=X@c-71-57-85-168.hsd1.il.comcast.net] has joined #wesnoth-dev
20090221 19:37:45< Ivanovic> that is: is it fixable *at all* on short terms or is some larger scale rewrite needed?
20090221 19:38:05< zookeeper> Mordante, while i'm working on that, there's another not-so-glaring but serious terrain layering issue:
20090221 19:39:14< zookeeper> eh, actually nevermind
20090221 19:39:29< zookeeper> it's a WML-only issue and i don't see how a code workaround could possibly be made anyway
20090221 19:40:40< alink_> was this story ^  http://hardware.slashdot.org/article.pl?sid=07/06/21/128228
20090221 19:40:45< zookeeper> Mordante, ok, i got the mountains to stay behind the units...but now they don't overlap the other mountains north of them properly...urgh
20090221 19:40:50 * zookeeper works on in a bit more
20090221 19:41:28< Sirp_> Ivanovic: I will see what I can do. Though I am rather busy in r/l right now, so we will have to see.
20090221 19:41:39< Ivanovic> okay
20090221 19:41:57< Ivanovic> Sirp_: at least it is one of the blockers for 1.6
20090221 19:42:23< alink_> Sirp _ lol seems that some HP ink cartridges are secretly programmed to expire on a certain date  http://news.bbc.co.uk/1/hi/technology/4293427.stm
20090221 19:42:52< CIA-70> mordante * r32973 /trunk/ (3 files in 2 dirs): Keep the horizontal position when a listbox scrolls.
20090221 19:42:54-!- crimson_penguin [n=ben@CPE0013f7fa2460-CM0013f7fa245c.cpe.net.cable.rogers.com] has quit ["Leaving"]
20090221 19:46:22< zookeeper> Mordante, does the mountain problem occur with any other terrains?
20090221 19:46:42< zookeeper> i think i have a fix, but how i'm going to fix it depends..
20090221 19:47:08< Mordante> zookeeper no idea haven't played wesnoth for a while
20090221 19:47:21< Mordante> but can't remember anybody mentioning it on the forum
20090221 19:48:32< Mordante> Ivanovic updated items are in order of priority http://www.wesnoth.org/wiki/Wesnoth1.6Features#Will_complete
20090221 19:48:54< Mordante> the last two of will do are not blockers
20090221 19:50:36-!- stikonas_ [n=and@ctv-213-164-101-51.vinita.lt] has joined #wesnoth-dev
20090221 19:51:58< Ivanovic> Mordante: okay, looks reasonable
20090221 19:53:02-!- stikonas [n=and@ctv-213-164-101-51.vinita.lt] has quit [Read error: 60 (Operation timed out)]
20090221 19:53:04-!- cookiebot [n=sh314001@190.22.106.231] has joined #wesnoth-dev
20090221 19:53:08-!- cookiebot is now known as shikadibot
20090221 19:53:17< Shadow_Master> shikadibot: file /trunk/data/tools/wmllint
20090221 19:53:23< shikadibot> Web interface URL to file /trunk/data/tools/wmllint: http://svn.gna.org/viewcvs/wesnoth/trunk/data/tools/wmllint?view=log
20090221 19:53:44< Shadow_Master> shikadibot: file /trunk/changelog
20090221 19:53:47< shikadibot> Web interface URL to file /trunk/changelog: http://svn.gna.org/viewcvs/wesnoth/trunk/changelog?view=log
20090221 19:54:16-!- busfahrer` is now known as busfahrer
20090221 19:54:24< Shadow_Master> esr: thanks for your answer! ;-)
20090221 19:56:45< zookeeper> hmm, am i right in assuming that terrain graphics WML isn't actually retained in memory in a/the grand big WML document, but instead after parsing the terrain rules at startup the actual WML is just discarded (and the rules kept in some other, more efficient format for the engine to use)?
20090221 19:57:15< zookeeper> or does the question even make much sense at all?
20090221 19:57:50< boucman> zookeeper: I would assume you're right
20090221 19:58:04< boucman> parsing wml is costly, we probably don't do it at every redraw
20090221 19:58:22< Mordante> yes it gets moved in another structure
20090221 19:58:42 * shikadibot wonders if Mordante wants to edit the changelog again or not.
20090221 19:58:49< Mordante> of course the WML could be freed from the memory
20090221 19:59:03< Mordante> shikadibot of course I want that, but not today
20090221 19:59:10< Shadow_Master> thanks!
20090221 19:59:30< zookeeper> Mordante, you mean that it might be freed already, or just that it could be made to be freed?
20090221 20:00:06< CIA-70> shadowmaster * r32974 /trunk/data/core/about.cfg: Update entries for Shadow_Master/ShikadiLord in credits.
20090221 20:00:08< Mordante> could be made
20090221 20:00:14-!- shikadibot [n=sh314001@190.22.106.231] has quit ["Lost terminal"]
20090221 20:00:38< zookeeper> ok...i wonder if anyone has checked how much memory it takes?
20090221 20:00:52< Mordante> at the moment almost no wml or even none at all is freed 
20090221 20:01:31< Mordante> guess not, but we want to optimize wml loading and unloading for quite a while but nobody really worked on it
20090221 20:01:50< Mordante> we had some talks at the fosdem 2008 about it
20090221 20:02:23< boucman> and sirp want to mentor a gsoc about it
20090221 20:02:55-!- nital [n=nital@public758.xdsl.centertel.pl] has joined #wesnoth-dev
20090221 20:03:06< boucman> hey nital
20090221 20:03:11< nital> hello
20090221 20:03:13< zookeeper> might it be a useful debugging tool if there was a command-line switch to make wesnoth output the WML document into a simple .cfg file on startup? so that you could check the final preprocessed and expanded WML
20090221 20:03:16< boucman> nital: did you see my replies wrt your patch ?
20090221 20:03:24< Mordante> hi nital 
20090221 20:03:26< CIA-70> shadowmaster * r32975 /trunk/data/core/about.cfg: 
20090221 20:03:26< CIA-70> Since santi is a collective of the three Alexiou people, compressed them
20090221 20:03:26< CIA-70> under their single nickname for core about.cfg. This also solves the
20090221 20:03:26< CIA-70> alphabetical ordering issue.
20090221 20:04:32< CIA-70> shadowmaster * r32976 /trunk/ (12 files in 6 dirs): 
20090221 20:04:32< CIA-70> Per Kitty's and Dave's request, the Great Continent (central) map is now
20090221 20:04:32< CIA-70> the only titlescreen background again.
20090221 20:04:32< CIA-70> The old titlescreen files were moved into story/ and renamed as well;
20090221 20:04:32< CIA-70> wmllint has been updated to update existing content due to these
20090221 20:04:33< CIA-70> changes. Fixed mainline WML referencing those files, and added a release
20090221 20:04:37< CIA-70> note along with a quiet changelog entry.
20090221 20:05:00< boucman> ow :(
20090221 20:05:01< nital> boucman: ok, read it, working on it
20090221 20:05:14< boucman> Shadow_Master: why did they want them removed ?
20090221 20:05:30-!- Corvvs [n=michaelm@pool-71-163-211-230.washdc.east.verizon.net] has quit []
20090221 20:05:38< alink_> zookeeper: I just did a little test, i think i have 2010 [terrain_graphics] parsed by the terrain buillder code
20090221 20:05:42< Shadow_Master> boucman: Kitty strongly insisted that she never intended those "crappy pictures" to be used for the titlescreen
20090221 20:05:52< Shadow_Master> besides the JPEG artifacts.
20090221 20:06:06< boucman> ok
20090221 20:06:37< zookeeper> alink, sounds realistic
20090221 20:07:00< alink_> mmh and internaly we create rotations of these rules
20090221 20:07:26< alink_> but that's not WML anymore
20090221 20:09:12< zookeeper> Mordante, layer=0 didn't work, so i'm using layer=-10
20090221 20:09:44< CIA-70> boucman * r32977 /trunk/src/unit_animation.cpp: fix animation jumpiness reported by alink. The bug was much more serious than that, but it's fixed now
20090221 20:09:48< Mordante> zookeeper the layers in my code have nothing to do with the layers in the terrain code
20090221 20:10:14< zookeeper> right
20090221 20:10:29< Ivanovic> Shadow_Master: in the changelog you forgot to mention titlescreen/landscapebattlefield.jpg
20090221 20:10:30< Mordante> all terrain layers end up at two layers in my code, before or behind a unit
20090221 20:10:33< Shadow_Master> boucman: jumpiness fixed, whooo!
20090221 20:10:40< Ivanovic> only the two other ones are mentioned
20090221 20:10:52< Shadow_Master> Ivanovic: oh, I nuked it accidentally while resolving the merge conflict
20090221 20:11:19< Ivanovic> that is: this file change is missing in the changelog, the release notes and wmllint
20090221 20:11:25< alink_> zookeeper: outputing the whole wml document to a text file gave me a 2.6 Mo file, but in memory there is no tab charcter or [,],= etc..
20090221 20:12:04< Shadow_Master> Ivanovic: ohhh. Then I nuked it accidentally only fro mthe changelog :-)
20090221 20:12:13< alink_> well there is also info about the data structure
20090221 20:12:34< zookeeper> alink_, cool. so that means it doesn't really take up a noticeable amount of memory in-game?
20090221 20:12:35< Mordante> Shadow_Master in that case I should have modified the changelog :-P
20090221 20:12:48< Ivanovic> Shadow_Master: uhm, i don't understand what you mean
20090221 20:12:49< Shadow_Master> ?
20090221 20:12:52< Ivanovic> what i said is:
20090221 20:13:01< Shadow_Master> Ivanovic: i know i know
20090221 20:13:13< Ivanovic> you forgot to mention that you also moved titlescreen/landscapebattlefield.jpg to story/landscape-battlefield.jpg
20090221 20:13:13< Shadow_Master> I'm already working on fixing it.
20090221 20:13:20< Shadow_Master> I know!
20090221 20:13:24< Ivanovic> okay, good
20090221 20:13:24< alink_> zookeeper: not sure how this data is fragmented/stored by c++ , probably must ask Sirp about that 
20090221 20:13:36< Mordante> Shadow_Master there's a reason why I rebase before changing the changelog and commit directly afterwards ;-)
20090221 20:13:51< Shadow_Master> uh... uh.
20090221 20:17:22< Sirp_> alink_ / zookeeper: what is it you want to know about what data (I'm finding reading up hard...
20090221 20:17:27< CIA-70> shadowmaster * r32978 /trunk/ (RELEASE_NOTES changelog data/tools/wmllint): 
20090221 20:17:27< CIA-70> Forgot to mention the titlescreen/landscapebattlefield.jpg ->
20090221 20:17:27< CIA-70> story/landscape-battlefield.jpg step in the changelog and Rel. Notes
20090221 20:17:27< CIA-70> entries, and in wmllint's code.
20090221 20:17:27< CIA-70> Thanks Ivanovic for spotting this.
20090221 20:17:57< alink_> pfff, almost crashed my system by scrolling this big file in my text editor
20090221 20:17:57 * Shadow_Master hands Ivanovic a glazed cookie
20090221 20:18:05-!- ABCD_ is now known as ABCD
20090221 20:18:34< Shadow_Master> alink_: yeah, I once tried to open a 2.2 GB unused sector dump in vim
20090221 20:18:42 * Ivanovic eats the cookie
20090221 20:18:43< Shadow_Master> the results were awesome.
20090221 20:18:50< alink_> Sirp_ size in memory of the main wml document and the proportion used by terrain rule
20090221 20:18:55< zookeeper> yeah
20090221 20:19:24< Ivanovic> afk, watching some tv, cu
20090221 20:19:34< alink_> zookeeper: from what i saw, seems that half of (in the text output) is unit animation stuff
20090221 20:20:30< alink_> but more i think about it , more i think it doesn't reflect well the real size in memory 
20090221 20:21:11< silene> don't worry too much about the size of wml, it's negligible wrt to graphics; a bigger issue would probably be the cost of handling it
20090221 20:22:43< zookeeper> Mordante, i think i got my fix finished. i'll commit it, but i wouldn't mind if other folks tested it in various situations, like magic and other halo effects and so on
20090221 20:23:39< Mordante> thanks zookeeper can you also add that request to the RELEASE_NOTES?
20090221 20:24:41< ilor> zookeeper: I think it would be nice, especially if it behaved like gcc -E (so it'd preserve the original filenames that the separate bits came from as comments)
20090221 20:25:00< ilor> argh. scroll fail, I replied to an old message -_-
20090221 20:25:06< zookeeper> oh, wait, it causes glitches with forests.
20090221 20:28:39< CIA-70> ilor * r32979 /trunk/src/filechooser.cpp: work around for bug #12625
20090221 20:28:50< thespaceinvader> http://www.wesnoth.org/forum/viewtopic.php?p=341203#p341203 who would be appropriate to approach in this instance? Dave?
20090221 20:29:58< Mordante> the project is GPL so as long as he honours the GPL he's free to make money of Wesnoth
20090221 20:30:10< Shadow_Master> where's Turuk and Ivanovic when you need to ban a spambot?
20090221 20:30:36< Sirp_> well he could do it but he'd have to release it under GPL
20090221 20:30:39< Mordante> if he doesn't want to honour he'll need to have an agreement of all contributors (good luck with locating them all)
20090221 20:30:47< Shadow_Master> the bot is making spam in the range of 2 minutes
20090221 20:30:59< Sirp_> meaning someone else could then take the code and release it for free.
20090221 20:31:12< Shadow_Master> Sirp_: do you know how to handle spambot banning?
20090221 20:31:55< Sirp_> Shadow_Master: where? On forums? IRC? Wesnoth server? ...
20090221 20:32:01< Shadow_Master> Sirp_: forums
20090221 20:32:01-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090221 20:32:16< Shadow_Master> if you know how to do it, please do before it kills my last bits of sanity.
20090221 20:32:29< Shadow_Master> I already moved two posts to the hidden place
20090221 20:33:34< Sirp_> Shadow_Master: banned.
20090221 20:33:44< Shadow_Master> thanks
20090221 20:34:05< ilor> wesbot: topic
20090221 20:34:14< ilor> yay, below 50 ;)
20090221 20:34:24< Mordante> :-)
20090221 20:34:25< Espreon> Awesome!
20090221 20:34:26-!- Baufo [n=thomas@62-47-145-124.adsl.highway.telekom.at] has quit ["Leaving"]
20090221 20:34:29< Mordante> that has been a while ago
20090221 20:34:34< Shadow_Master> that's some nice work! are we going to have an RC now?
20090221 20:35:24< ilor> Shadow_Master: probably not until the "security" bugs are fixed
20090221 20:35:24< Mordante> nope but Ivanovic probably tags beta 3 tomorrow
20090221 20:36:14< Shadow_Master> is anyone handling AI0867's patch in the hidden forum btw?
20090221 20:36:39< silene> Sirp_: note that there is a particularity with the iphone; while he would have to release the source to people buying the application and asking for it, the source itself would be quite useless to other people unless they have unlocked their iphone
20090221 20:36:44-!- Turuk [n=Turuk@FUSE-WS1-208-102-202-188.fuse.net] has joined #wesnoth-dev
20090221 20:36:59< stikonas_> what should be the expected number of bugs in 1.6?
20090221 20:37:05-!- stikonas_ is now known as stikonas
20090221 20:37:36< Mordante> silene the Tivo backdoor
20090221 20:37:57< Mordante> stikonas no fixed number we just have to feel well about it
20090221 20:38:32< Mordante> the last weeks quite some bugs got filed which have been in the engine much longer so the bug count only tells the number of known bugs
20090221 20:39:15< silene> Mordante: is this backdoor used by the majority of the iphone users? (i have no clue how widely it is available)
20090221 20:39:17< Mordante> if somebody files 100 valid bugs today, Wesnoth didn't get worse only we know better what's wrong
20090221 20:39:31< stikonas> btw is there a developer who knows cache code well?
20090221 20:39:59< Mordante> silene no the backdoor that you can release the source but the user can't do much with it, which caused RMS to look at the GPL 3
20090221 20:40:54< silene> Mordante: oh i see, i had misunderstood
20090221 20:41:56< silene> does it apply there? it's not like an iphone developer can do anything about apple locking the thing
20090221 20:42:08-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has quit []
20090221 20:42:14< Mordante> no and it's allowed with the GPL2
20090221 20:42:58< Mordante> so he could develop and release Wesnoth and asking money for it and people can't really use the source
20090221 20:43:17< Mordante> unless they buy and apple sdk or unlock their iPhone
20090221 20:43:26< silene> agreed
20090221 20:43:41< Mordante> of course somebody could buy the SDK and sell it cheaper 
20090221 20:43:58< Mordante> (no idea whether apple will step in at that point)
20090221 20:44:58< Sirp_> silene: but then you could take the source and release the application yourself on the app store for free, no?
20090221 20:45:16< stikonas> ant he will have to remove bruteforce.py from the game
20090221 20:45:26< stikonas> because it is not available ander GPL 2
20090221 20:45:51< voris> I'm pretty sure free software is pornographic under the Apple TOS. ;)
20090221 20:46:12< Mordante> Sirp afaik you need to buy a SDK to be able to upload things to the store
20090221 20:46:18< Mordante> stikonas why?
20090221 20:46:24< ilor> is anyone working on https://gna.org/bugs/?13031 ?
20090221 20:46:33< stikonas> ##    Copyright 2006 by Michael Schmahl
20090221 20:46:34< stikonas> ##    This code is available under the latest version of the GNU Public License.
20090221 20:46:41-!- Corvvs [n=michaelm@pool-71-163-211-230.washdc.east.verizon.net] has joined #wesnoth-dev
20090221 20:47:03< ilor> stikonas: that's a nonstandard clause that I can't say I like
20090221 20:47:10< Mordante> uhm that's weird since all contributions to wesnoth should be GPL 2
20090221 20:47:24< Mordante> 2+
20090221 20:47:34< Mordante> 2 or 2+
20090221 20:47:39< ilor> 2+
20090221 20:47:39< stikonas> ilor: yes
20090221 20:47:49< Mordante> since in the old days we never really defined 2 or 2+
20090221 20:47:54< zookeeper> Mordante, i might try to simplify the mountain WML a bit while i'm at this.
20090221 20:48:01< stikonas> because if FSF is hijacked and GPLv 26 is non free, then bruteforce.py becomes non-free
20090221 20:48:02< Mordante> great :-)
20090221 20:48:02< silene> Sirp_: apple controls what's on the store, and they are not intested in a free program, since that means less money for them
20090221 20:48:30< ilor> stikonas: I don't thin you can release something and say that it's under the latest license *only*
20090221 20:48:32< Mordante> ilor I'm not working on that bug
20090221 20:48:55< stikonas> ilor: maybe, I am not a lawyer
20090221 20:49:42< ilor> Mordante: I'm adding a hard limit in terrain_translation, I'll later make it configurable somehow so it doesn't take a rebuild to change it
20090221 20:49:51< Mordante> also the copyright is stated as 2006 at which time GPL 2 was the most current one
20090221 20:50:24< Mordante> ilor like I said before I'm not really fond of hardcoded limits
20090221 20:51:00< ilor> Mordante: in that case we should clarify the statement, change it to GPL2+ witha  commit comment that this was the current version then 
20090221 20:51:13< ilor> Mordante: me neither, but what would be a "better" way?
20090221 20:51:42< ilor> since there's a lot code further down that grabs more memory when the map is bigger
20090221 20:51:48< Mordante> not really sure, but there's also no guarantee a 200x200 map isn't too big for certain devices
20090221 20:52:21< Mordante> what's the real problem slow or running out of memory?
20090221 20:52:43< ilor> I think it runs out of memory
20090221 20:53:04< Mordante> can't we catch std::bad_alloc in a clean way?
20090221 20:53:26< Sirp_> silene: my understanding was that there are plenty of free programs for the iPhone. It is A Good Thing for Apple since it makes the iPhone itself more attractive
20090221 20:54:15< Shadow_Master> I deleted a stupid response by a certain user btw
20090221 20:54:36< Sirp_> Mordante: it's not guaranteed that memory exhaustion will result in std::bad_alloc on most modern platforms
20090221 20:54:50< silene> Sirp_: my mistake then
20090221 20:54:58< Shadow_Master> "
20090221 20:55:01< Shadow_Master> "Most likely badly. Most of Wesnoth's people are volunteers."
20090221 20:56:10-!- Elvish_Pillage2 [n=eli@66-189-14-42.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090221 20:56:26< Sirp_> there is still the concern that Apple could start getting freaked out about there being two essentially identical apps, one for money the other free, wondering "what's going on here?"
20090221 20:56:27< Mordante> Sirp_ :-(
20090221 20:56:41-!- Elvish_Pillage2 [n=eli@66-189-14-42.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090221 20:56:47< Mordante> that was my worry about apple as well
20090221 20:57:02< Mordante> especially if the first author starts to complain
20090221 20:58:01< ilor> Mordante: I'm testing a huge map in the editor, and first of all it's taking up the cpu and allocating memory, taking a lot of time
20090221 20:59:25< ilor> it's somewhat painful to watch a million of 3-element vectors get allocated
20090221 21:00:40-!- Mythological [i=Mytholog@77.28.68.46] has quit []
20090221 21:02:24< ilor> one million wasn't enough though ;)
20090221 21:03:10< Mordante> it managed?
20090221 21:03:25< Shadow_Master> how big is the map in hexes? :-)
20090221 21:04:20< ilor> it managed a three by one million
20090221 21:04:29< ilor> taking around a gigabyte of ram total
20090221 21:04:54< Mordante> which I see more as a problem since it's slow but succeeds
20090221 21:05:33-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has quit [Remote closed the connection]
20090221 21:10:42-!- Lord_Aether [n=castle@207.212.136.10] has joined #wesnoth-dev
20090221 21:10:58< zookeeper> Mordante, i presume we'll still want the mountain peaks to go over units north of them.
20090221 21:11:16< Mordante> ilor in that case I can't think of a better solution, still doesn't mean I like the hardcoded limit ;-)
20090221 21:11:31< Mordante> zookeeper yes
20090221 21:13:52-!- cjhopman [n=chris@wireless33.cs.wisc.edu] has quit [Remote closed the connection]
20090221 21:16:29< Mordante> I'm off night
20090221 21:16:39-!- Mordante [n=chatzill@roadie.xs4all.nl] has quit ["ChatZilla 0.9.84 [Iceweasel 3.0.6/2009020409]"]
20090221 21:16:41-!- PK [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has joined #wesnoth-dev
20090221 21:17:08< ilor> I tried a 3-million one now, and it died with a bad_alloc
20090221 21:17:49< loonycyborg> Did it run out of address space? :)
20090221 21:18:17-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090221 21:24:53< ilor> most likely
20090221 21:43:39< Sirp_> alink_, zookeeper: sorry got distracted before. :(
20090221 21:44:48< Sirp_> alink_, zookeeper: read this for some details on how much memory WML documents take: http://dave.wesnoth.org/?p=7
20090221 21:45:37< Sirp_> alink_, zookeeper: basically I think each WML attribute has around 80 bytes of overhead in addition to the actual data
20090221 21:45:49< Sirp_> and an element has more
20090221 21:46:14< Shadow_Master> 80 bytes? that's a lot for my campaign's WML.
20090221 21:46:56< Shadow_Master> my WML is 1.2 MB aprox., before doing macro expansion
20090221 21:47:12< Ivanovic> after the expension it gets interesting...
20090221 21:47:19< Shadow_Master> yeah
20090221 21:47:58< Ivanovic> silene: regarding memory usage: https://mail.gna.org/public/wesnoth-dev/2008-02/msg00079.html
20090221 21:48:26< Ivanovic> okay, this is data from "shortly before 1.4", but you can easily calculate how much memory usage there is without the images
20090221 21:48:39< CIA-70> ilor * r32980 /trunk/src/ (map.cpp terrain_translation.hpp): use const char* instead of a std::string in t_translation::error
20090221 21:48:59< Ivanovic> that is: in the tiny builds only a quarter of the amount of the "normal" build is for images (since scaled down versions are used)
20090221 21:50:09< Ivanovic> so wesnoth without image would use almost 40MB for a 9 turn replay without sound, music or translations
20090221 21:50:30< Ivanovic> short guess is that of those 40MB at least 30, more likely more, is for WML
20090221 21:51:29< Ivanovic> and this was from a normal multiplayer skirmish game replay, nothing fancy
20090221 21:52:13< Ivanovic> with other words: for campaigns with *lots* of WML there would probably lots of space be used for the content
20090221 21:52:38< ilor> I found an interesting memory issue in the editor
20090221 21:53:10< ilor> when I load the huge-but-not-too-much map (3*500K), memory usage goes from 100 to 500M
20090221 21:53:23< ilor> then after I close just the map, it drops to 250 (!)
20090221 21:53:45< ilor> but when I close the editor, go back to the main menu and load it again, it's at the 100M mark again
20090221 21:54:48< silene> Ivanovic: i agree with the 40MB value for the game without graphics, but i think you are underestimating the size of the binary itself ;-)
20090221 21:54:49< Ivanovic> ilor: yeah, we got some places where not all memory is freeed when it could be
20090221 21:55:33< ilor> Ivanovic: what's puzzling is that the momery is freed eventually so it's not an outright leak
20090221 21:55:37< Sirp_> I am hoping that Dragonking will work on something to reduce memory usage for SoC
20090221 21:55:54 * Ivanovic first hopes that we get into SoC
20090221 21:56:02< Ivanovic> then i hope that we get several good proposals
20090221 21:56:08 * Dragonking would like to and hopes we get into SoC too
20090221 21:56:22< Ivanovic> s/good proposals/good proposals with promising students to work on them
20090221 21:58:06-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev
20090221 21:59:34-!- happygrue [n=Wintermu@c-67-176-145-41.hsd1.in.comcast.net] has quit ["grok 'noth!"]
20090221 22:04:46-!- Sapient [n=sapien-x@c-24-126-133-155.hsd1.ga.comcast.net] has joined #wesnoth-dev
20090221 22:07:36< Sapient> hey there PK
20090221 22:08:00< Ivanovic> wesbot: seen fabi
20090221 22:08:01< wesbot> Ivanovic: The person with the nick fabi last spoke 2d 22h ago.  2d 20h ago was here and on the channels #wesnoth and #wesnoth-umc-dev with the message: Remote closed the connection
20090221 22:08:02< Ivanovic> wesbot: seen fabi_
20090221 22:08:05< wesbot> Ivanovic: The person with the nick fabi_ last spoke 5d 1h ago.  4d 23h ago was here and on the channels #wesnoth and #wesnoth-umc-dev with the message: Remote closed the connection
20090221 22:08:16< PK> I was wondering, although its probably possible, if it was possible to make complex traits for the purpose of making it simpler to know what group of units do what kind of stuff?
20090221 22:13:50-!- Shadow_Master [n=imorelle@unaffiliated/shadowmaster/x-275924] has quit ["BRB"]
20090221 22:14:40< CIA-70> ivanovic * r32981 /trunk/ (14 files in 10 dirs): updated German, Polish and Russian translation (sorry that it took a little longer...)
20090221 22:16:03-!- Shadow_Master [n=imorelle@unaffiliated/shadowmaster/x-275924] has joined #wesnoth-dev
20090221 22:21:46-!- PK [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has quit ["Java user signed off"]
20090221 22:24:26-!- ^Noyga^ [n=noyga@AVelizy-151-1-27-8.w82-124.abo.wanadoo.fr] has joined #wesnoth-dev
20090221 22:24:57-!- ^Noyga^ [n=noyga@AVelizy-151-1-27-8.w82-124.abo.wanadoo.fr] has left #wesnoth-dev ["Quitte"]
20090221 22:28:13-!- Lord_Aether [n=castle@207.212.136.10] has quit []
20090221 22:28:56-!- crimson_penguin [n=ben@32.60.108.73] has joined #wesnoth-dev
20090221 22:33:38< Ivanovic> Sirp_: got such stats available for public somewhere? http://www.wesnoth.org/forum/viewtopic.php?f=11&t=24204&start=0&st=0&sk=t&sd=a
20090221 22:34:09< Soliton> awstats.wesnoth.org
20090221 22:34:39< Turuk> I knew that was somewhere
20090221 22:35:14< Soliton> well, i've actually just set it up last week. ;-)
20090221 22:35:58< crimson_penguin> Ivanovic: This font thing on Mac should really be fixed at least partially before 1.6 :-/ I mean, right now it's likely that people who don't have MS Office installed can't see dialogs
20090221 22:35:59< Turuk> Ah, well then. :)  Was there an older one though, or previous statistics floating around somewhere?
20090221 22:36:39< Ivanovic> crimson_penguin: so for the moment at least a different font that is shipped by default and useable by "western languages" has to be found
20090221 22:36:41< Soliton> Sirp posted some occasionally from some google thingy which is not quite public though afair.
20090221 22:36:52< Ivanovic> crimson_penguin: that is i won't be of any help there since i don't have osx
20090221 22:37:08< Turuk> Ah ok, I think that's what I was thinking of
20090221 22:37:10< Ivanovic> crimson_penguin: you could also talk to the pangocairo dudes asking if they know a way around this stuff
20090221 22:37:26< crimson_penguin> Ivanovic: yeah
20090221 22:37:49< crimson_penguin> Ivanovic: I just checked this computer next to me (I'm at the Apple Store), and it has it... but it also has Office, so I can't tell if that's what installed it
20090221 22:38:00< crimson_penguin> I wonder if there's any computers here without Office
20090221 22:38:09< Ivanovic> ask
20090221 22:38:44< crimson_penguin> I'll probably look around on my way out
20090221 22:38:58< voris> Last time I was at the Apple store I could find one without Office on it.
20090221 22:38:59< crimson_penguin> There's a lot of computers ;)
20090221 22:39:06< crimson_penguin> Cool
20090221 22:43:18< Sapient> cya later, have fun.
20090221 22:43:18-!- Sapient [n=sapien-x@c-24-126-133-155.hsd1.ga.comcast.net] has left #wesnoth-dev []
20090221 22:51:06< Shadow_Master> wesbot: seen esr 
20090221 22:51:07< wesbot> Shadow_Master: Queried user last spoke 3h 33m ago.  esr is currently here and on the channel #wesnoth-umc-dev.
20090221 22:51:19< esr> I'm here, what's up?
20090221 22:52:36< Shadow_Master> esr: remember I asked you to review a patch at #wesnoth-umc-dev?
20090221 22:52:47< Shadow_Master> (since you are the O Master of English etc.)
20090221 22:53:18< esr> 1121, I thought it was.  I applied it and zookeeper later reverted it, clainming it didn't fix anything.
20090221 22:53:54< Shadow_Master> no?!
20090221 22:54:07< Shadow_Master> I asked you to review a patch for *wesnoth-umc-dev* , not for mainline.
20090221 22:54:16< Shadow_Master> and review, not apply. It's my campaign still.
20090221 22:54:23< esr> OK, where is it?
20090221 22:54:41< Shadow_Master> http://www.wesnoth.org/forum/download/file.php?id=28490
20090221 22:54:59< Shadow_Master> it touches your text, therefore I want you to check it carefully before applying.
20090221 22:55:07< Shadow_Master> any further comments should go to #wesnoth-umc-dev
20090221 22:55:34< esr> POh, right.  I'll look at iyt after I finish composing a bug summary for the dev list.
20090221 22:56:02< Shadow_Master> esr: I have no problems on you postponing the review as long as you tell me how much time it will take
20090221 22:56:12< Shadow_Master> days, weeks, etc. No hurries.
20090221 22:56:38< esr> Dunno.  I'm really focused on bug-hunting right now.
20090221 22:57:09< esr> I don't *think* it will be weeks.
20090221 22:57:11< ilor> sigh, I just remembered the (not reported) editor hotkey list bug that I have to fix before the next release
20090221 22:57:43< Shadow_Master> esr: okay, just don't forget to pling me when you have time ;-)
20090221 22:58:57< ilor> at least I think I found the reason for the pseudo-leak
20090221 23:00:21< ilor> there's a std::vector deep inside the display class that was being resized() on a map change, and that won't shrink it
20090221 23:01:52< Shadow_Master> by the way, is there anyway to optimize the chat log dialog in MP replays?
20090221 23:02:04< ilor> optimize?
20090221 23:02:07-!- DDR [n=chatzill@66.183.125.196] has joined #wesnoth-dev
20090221 23:02:11< Shadow_Master> yes, optimize
20090221 23:02:24< ilor> in which direction? ;)
20090221 23:02:28< Shadow_Master> as it is now, it may take a while to pop up when you have average amounts of chatter in the replay
20090221 23:02:38< ilor> ah, speed then
20090221 23:02:43< Shadow_Master> yes
20090221 23:02:59< ilor> most likely it is possible
20090221 23:03:25< ilor> do you know where is the relevant code?
20090221 23:03:38< Shadow_Master> checking
20090221 23:04:46< Shadow_Master> game_display.?pp
20090221 23:05:01< Shadow_Master> it's a vector of chat_message's
20090221 23:05:37< Shadow_Master> uh, but apparently the real messages are stored elsewhere
20090221 23:06:42< Soliton> when clearing and resizing a vector of ints all elements should be 0, yes?
20090221 23:07:08-!- crimson_penguin [n=ben@32.60.108.73] has quit ["Leaving"]
20090221 23:07:17< CIA-70> ilor * r32982 /trunk/src/builder.cpp: use vector.swap(vector()) instead of vector.resize() in terrain_builder cache when the map size changes to avoid a leak-like behavior when a very large map is replaced by a small one
20090221 23:07:19< Shadow_Master> I think so, since int() == 0
20090221 23:07:46< Shadow_Master> loonybot: { std::cerr << (int() == 0 ? "good" : "not good) << '\n'; }
20090221 23:07:47< loonybot> error: Unexpected end of request.
20090221 23:08:02< Shadow_Master> loonybot: { std::cerr << (int() == 0 ? "good" : "not good") << '\n'; }
20090221 23:08:08< loonybot> good
20090221 23:08:46< Sirp_> Soliton: yes, in a vector of ints all elements will be 0
20090221 23:09:22< Sirp_> loonybot: { int x; std::cerr << (x == 0 ? "good" : "not good") << '\n'; }
20090221 23:09:23< loonybot> warning: 'x' is used uninitialized in this function
20090221 23:09:34< Sirp_> well gee. :p
20090221 23:09:46< Shadow_Master> int() != <uninitialized int> !
20090221 23:09:56< Sirp_> Shadow_Master: I know.
20090221 23:09:58< Shadow_Master> the STL containers default-initialize their members AFAIK.
20090221 23:10:04< Sirp_> Shadow_Master: they do.
20090221 23:10:16< Sirp_> Shadow_Master: however, I think the most *likely* thing is for a stack variable to be initialized to 0
20090221 23:10:36< Shadow_Master> but then you fall in the "what will happen in platform XYZ?" problem
20090221 23:10:36< Sirp_> Shadow_Master: so I was just going to demonstrate how though your point is correct, your test doesn't really show anything. :)
20090221 23:10:43< Sirp_> Shadow_Master: absolutely.
20090221 23:10:55< ilor> loonybot: -w { int x; std::cerr << (x == 0 ? "good" : "not good") << '\n'; }
20090221 23:10:56< loonybot> good
20090221 23:11:07< Sirp_> Shadow_Master: I was just pointing out that this kind of thing has to be decided by reading the documentation/standard rather than by running code.
20090221 23:11:13< Sirp_> ahhh thanks ilor
20090221 23:11:30< ilor> sitting in ##c++ pays off finally ;)
20090221 23:11:32< Sirp_> your understanding of what actually happens is correct on all counts though.
20090221 23:11:40< Shadow_Master> I was just double-checking that a default-initializing integer was zero :-)
20090221 23:12:20< ilor> msvc's debug mode has a nice feature of setting uninitialized stack space to a particular bit pattern, uninitialized new()ed memory to another etc
20090221 23:12:41< Shadow_Master> I think valgrind and gdb can do the same
20090221 23:13:10< Sirp_> ilor: indeed.
20090221 23:13:11< Shadow_Master> at least for malloc()'d blocks
20090221 23:13:14-!- stikonas [n=and@ctv-213-164-101-51.vinita.lt] has quit [Read error: 110 (Connection timed out)]
20090221 23:13:38< Sirp_> the malloc()'d blocks thing is a feature of the malloc implementation I believe
20090221 23:13:40< zookeeper> Sapient, i happened to try attacking the orcish warlord with the dark adept in the test scenario, caused the scenario to crash
20090221 23:13:52< zookeeper> (with a WML error message of some kind)
20090221 23:15:42 * Shadow_Master is waiting for the message
20090221 23:18:02< Shadow_Master> "private" bug reports are not available for view unless you have logged in to Gna!, or is it unless you are a tracker technician or the original reporter?
20090221 23:18:31< alink_> Shadow_Master, Sirp_: at least an uninitilializsed bool has a random value here (linux, gcc 4.3)
20090221 23:19:47< alink_> and not fully random, same code gives the same value, change something somewhere and the value changes
20090221 23:21:22< Shadow_Master> https://gna.org/bugs/?12997 uhhh
20090221 23:21:39< alink_> correction: can't make the value change now
20090221 23:21:43< Shadow_Master> I think this used to be possible in 1.0, so that people like me could save-reload when a fight was going very well
20090221 23:21:51< Shadow_Master> was not*
20090221 23:22:51< alink_> ah just execute some STL code, change the value
20090221 23:23:07< Shadow_Master> boucman, Sirp_: how's https://gna.org/bugs/?12817 going by the way?
20090221 23:23:11-!- mjs-de [n=mjs-de@vpw.wh.uni-dortmund.de] has quit ["On the road again"]
20090221 23:23:26< Soliton> Shadow_Master: which is why it was disabled.
20090221 23:24:06< Shadow_Master> Soliton: mind to post a comment there?
20090221 23:24:19< Shadow_Master> and mark it as 'Won't Fix' or something..
20090221 23:24:57< Soliton> yes, too lazy. and the proper fix is to just disable saving.
20090221 23:25:08< Ivanovic> okay, time for me to head off to bed, n8
20090221 23:25:09< CIA-70> ilor * r32983 /trunk/src/editor2/editor_controller.cpp: style fix
20090221 23:25:50< Shadow_Master> loading?
20090221 23:26:15< Ivanovic> Shadow_Master: if there is no way to save a game, there is no way to load it either
20090221 23:26:20< Ivanovic> ;)
20090221 23:26:27< Shadow_Master> since we are nitpicking, maybe I should post a bug report on "Wesnoth should react to SIGTERM during loading screens".
20090221 23:26:42< Ivanovic> but there were some really ugly problems with saving while in a fight and other stuff like this
20090221 23:26:42< alink_> Sirp: quick technical question, does creating a lot of uninitialized data allow to read memory used by another program and thus spying something ?
20090221 23:26:53< Shadow_Master> Ivanovic: I remember.
20090221 23:26:56-!- yann [n=dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has quit [Remote closed the connection]
20090221 23:27:12< Shadow_Master> HPs changed, but not the amount of attacks_left
20090221 23:27:47< Sirp_> alink_: depends on the program.
20090221 23:27:50< Sirp_> sorry
20090221 23:27:54< Sirp_> on the platform
20090221 23:27:58< zookeeper> meh, gna isn't working for me atm.
20090221 23:28:07< zookeeper> i finally got the mountain layering sorted out.
20090221 23:28:24< Soliton> would the OS assign the same memory to different programs?
20090221 23:28:28< Sirp_> alink_: on Linux, all memory is zeroed
20090221 23:28:33< Sirp_> when it is allocated to a program
20090221 23:28:38< Sirp_> see the man page for mmap
20090221 23:28:48< Sirp_> Soliton: yes. But it zeroes it first
20090221 23:28:53< Soliton> or do you mean spying on programs shortly after they terminated?
20090221 23:29:00< Shadow_Master> I think he meant that
20090221 23:29:18< Sirp_> Soliton: if program A maps memory and then unmaps it, and then program B maps memory, it may get some of the memory program A just unmapped.
20090221 23:29:28< Soliton> right, of course.
20090221 23:29:29< Sirp_> however the OS will zero the memory before letting program B access it
20090221 23:29:50< Sirp_> however, *most* of the time when you call malloc() in a program you don't get memory from the OS
20090221 23:29:53< alink_> tx sirp was typing that (but i was much less clear :) )
20090221 23:29:59< Shadow_Master> man malloc
20090221 23:30:23< Sirp_> you get memory that was previously malloc()ed by your program and then free()ed
20090221 23:30:24< Shadow_Master> "optimistic"... "available"... "overcomitting". Right.
20090221 23:30:37< Sirp_> so 'uninitialized' means that it may contain data that your program previously wrote to it
20090221 23:31:18< Shadow_Master> using gparted to resize a 30 GB partition after disabling the "optimistic" allocator was a nice experience here
20090221 23:31:34< Shadow_Master> (specially since there was no swap space to use)
20090221 23:32:05< Sirp_> ...in which case not initializing is an optimization, but not a security problem.
20090221 23:33:11< CIA-70> zookeeper * r32984 /trunk/data/core/ (terrain-graphics/mountains.cfg terrain-graphics.cfg): Made the mountains not layer on top of units except where it makes sense. Also tried to simplify some of the mountains WML a little bit.
20090221 23:33:57< Soliton> what about malicious plugins? :-)
20090221 23:34:25< Shadow_Master> or Python!
20090221 23:34:46< Sirp_> I don't think Python can read uninitialized memory
20090221 23:35:29< Sirp_> but if you have code already inside the process, it doesn't even need to use malloc. It can just read all the memory mapped into the process and start reading it directly if it wants
20090221 23:35:41< loonycyborg> Under DOS you can freely read other program's data :)
20090221 23:36:16< silene> under linux too
20090221 23:36:21< Sirp_> sure, it also depends on the platform
20090221 23:36:36< Sirp_> just read /proc/self/maps to find all the memory your program has mapped
20090221 23:36:40< Sirp_> silene: how under Linux?
20090221 23:37:04< silene> Sirp_: the same way on-the-fly gdb does it
20090221 23:37:07< loonycyborg> Linux uses processor's protected mode at least..
20090221 23:37:35< Sirp_> silene: only if the process has the correct permissions though, no?
20090221 23:37:53-!- malumalu_ [n=malu@hnvr-4dbbffeb.pool.einsundeins.de] has quit ["Verlassend"]
20090221 23:38:45< silene> Sirp_: sure, you won't be able to read processes from other users; but a malicious plugin could well read your own unencrypted ssh keys from ssh-agent for instance
20090221 23:39:06 * Sirp_ nods
20090221 23:39:19< CIA-70> ilor * r32985 /trunk/src/builder.cpp: fix a compile error on gcc caused by a recent commit
20090221 23:39:23< Shadow_Master> Sirp_: is Python able to read memory directly?
20090221 23:39:44< ilor> Shadow_Master: python can call C code, therefore it is able to ;)
20090221 23:39:53< Shadow_Master> ilor: arbitrary C code?
20090221 23:40:14< loonycyborg> ilor: So can any other scripting language :)
20090221 23:41:29-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090221 23:44:09< Sirp_> Shadow_Master: well BBBIAAAAAA d
20090221 23:44:13< Sirp_> ugh
20090221 23:44:31< Sirp_> Shadow_Master: well I don't know if it has any standard library feature that allows it to; one could certainly make it possible.
20090221 23:44:45< ilor> nice typo, Sirp ;)
20090221 23:44:48< Sirp_> e.g. you could easily make some function read_bytes_from_address
20090221 23:44:53< alink_> the Sirp_ bot had some uninitialized string <:o)
20090221 23:44:54< Sirp_> ilor: kids. :)
20090221 23:45:35< loonycyborg> Sirp_: Can python be secured by overriding __import__?
20090221 23:46:42< loonycyborg> Say, if we allow only import known safe functions?
20090221 23:47:30< loonycyborg> Of course, malicious code could override __import__ too, but where will it get another __import__ function? :)
20090221 23:52:08< zookeeper> uh, i wonder why the basic forests have now started to spill over on top of mountains...they didn't do that before (unrelated to what i just committed, though).
20090221 23:52:31< zookeeper> oh, wait, actually it is related. duh.
20090221 23:54:43< zookeeper> i'm not entirely sure if it's a problem...anyone fancy trying it out in the editor? just draw forests nw/ne of some mountains and you'll see some trees on the mountain sides
20090221 23:57:30< CIA-70> zookeeper * r32986 /trunk/data/multiplayer/maps/ (18 files): Updates to forest variation usage by Doc Paterson.
--- Log closed Sun Feb 22 00:00:01 2009