--- Log opened Tue Apr 13 00:00:17 2010
--- Day changed Tue Apr 13 2010
20100413 00:00:17< gabba> I have a whole 3 weeks for bugfixing
20100413 00:00:27< gabba> ah before midterm ok
20100413 00:00:40< Crab_> gabba: yes, but they're after the midterm. I suggest moving one of them before.
20100413 00:01:50< gabba> ok. and does the priority system make sense after my explanation?
20100413 00:02:36< Crab_> gabba: it made sense before, too. 
20100413 00:03:16< gabba> Crab_: does the timeline still seem strange then ? :P
20100413 00:04:14< Crab_> gabba: yes :-P. you see, I suggest pushing low-priority and optional things towards the end. e.g., you can go 'on time' and then devote time to optional things thinking that all is ok, then you'll get hit by a underestimation and will have a hard time completing some of the more important things.
20100413 00:05:38< Crab_> gabba: that's why I think that 'going for the hardest and most important things first' will be a good idea
20100413 00:05:48-!- Darkas [~quassel@ppp-88-217-107-192.dynamic.mnet-online.de] has quit [Remote host closed the connection]
20100413 00:06:22< Crab_> gabba: since there'll be more time to react if things go not the way they should.
20100413 00:07:11< gabba> Crab_: well, you got a point. Even though it did make more sense to group "Implement the merge algorithm for planned_action vectors from other clients [low-priority]" with the backend work, IMO.
20100413 00:07:49< gabba> Crab_: so, do you suggest I move optional stuff toward the very end of the project, or at the end of the respective milestones?
20100413 00:07:59< Crab_> gabba: well you'll be familiar with that code inside&out, there'll be no problems switching back and forth :)
20100413 00:08:30< gabba> Crab_: and is the intermediate milestone after 4 weeks useful in your opinion?
20100413 00:08:50< Crab_> gabba: very useful
20100413 00:09:29< Crab_> gabba: another thing that I suggest is to mark gui and engine parts cleanly
20100413 00:10:48< Crab_> there'll be little engine things there , but there'll be some things which are like 'engine part for the gui', I think
20100413 00:11:15< Crab_> e.g. you'll most likely need to code a certain 'framework' for your arrows/markers stuff, extending existing gui capabilities
20100413 00:12:13< gabba> Crab_: I looked at that a bit, and it's more a matter of adding images to the right hex layer at the right time
20100413 00:13:44< loonycyborg> Espreon: Updated.
20100413 00:13:50< Espreon> Thank you.
20100413 00:13:52< gabba> So, for my other question above, "do you suggest I move optional stuff toward the very end of the project, or at the end of the respective milestones?"
20100413 00:13:58< loonycyborg> np
20100413 00:14:51< Crab_> gabba: yes, but it's always better to zoom things out a little, then zoom back in. e.g., if you add code which will allow you to speak in the terms of 'drawing planned moves', it would be less prone to bugs than the code that speaks in terms of 'adding images to the right hexes'.
20100413 00:15:34< Crab_> gabba: move the optional stuff to the point where the 'definition of done' for the milestone is reached.
20100413 00:15:50< gabba> i.e. wrapping instead of directly attaching myself to that code... I agree
20100413 00:16:04< Crab_> gabba: yes, setting up a system for yourself to use
20100413 00:16:39< Crab_> gabba: then, you'll have less dependencies, and in case of bugs, it will be possible to see if it's 'above' or 'below' your code layer, thus making it easier to fix it.
20100413 00:16:59< gabba> Crab_: ok, back to the drawing board I go then (gah!)
20100413 00:17:10< Crab_> gabba: e.g., 'hack,hack,hack, look at what is done, cleanup and fix optional cute things, pass milestone' process.
20100413 00:20:08-!- _jbx_ [~jbailey@12.190.80.225] has quit [Quit: It was a new day yesterday, but it's an old day now.]
20100413 00:22:09< AI0867> boucman, upthorn: an MP campaign is started like an MP scenario, and any side that doesn't specify its recruit list gets one from the era
20100413 00:22:52< AI0867> additionally, the era may contain (or consist entirely) of events, in order to modify gameplay
20100413 00:22:53< Gambit> Crab_: Was passive_leader=bool removed in 1.8's AI?
20100413 00:23:05< Gambit> The wiki says no, but it isn't working as of 1.8.
20100413 00:23:13< Gambit> *doesn't appear to be working
20100413 00:23:28-!- mjs-de [~mjs-de@wh.uni-dortmund.de] has quit [Remote host closed the connection]
20100413 00:26:01< Crab_> Gambit: hi. I guess 'no, it was meant to stay'
20100413 00:26:52< Crab_> Gambit: how-to-reproduce ?
20100413 00:26:57< Gambit> Crab_: http://wesnoth.pastebin.com/Y6zgT2qC
20100413 00:27:05< Gambit> It moves straight to the nearest open keep.
20100413 00:27:27-!- YogiHH [YogiHH@wesnoth/developer/yogihh] has left #wesnoth-dev []
20100413 00:27:49< Crab_> will take a look atm
20100413 00:29:20< Gambit> It was working on 1.7.11 that's the last version I tested on. I had the release client, but I didn't play that particular scenario on it.
20100413 00:31:59< AI0867> I think passive_leader might be defined as "keep leader on keep"
20100413 00:32:11< Gambit> It wasn't before.
20100413 00:32:14< AI0867> so, if it's not on one, for some reason, move it there
20100413 00:32:18< Gambit> It was "stay seated plox"
20100413 00:32:30< Crab_> AI0867: no, that case can be handled by RCA with just 'move to keep' CA.
20100413 00:32:58< AI0867> well, not before RCA existed ;)
20100413 00:34:14< Gambit> AI0867: no. Before it definitly stayed put no matter where it was.
20100413 00:34:58< Crab_> AI0867: and there was passive_leader_shares_keep for allied leaders which are to be passive but give way to player to recruit and return back.
20100413 00:35:04-!- meric [~Eric@124-168-142-121.dyn.iinet.net.au] has joined #wesnoth-dev
20100413 00:35:12< Crab_> e.g. in HttT 1
20100413 00:38:00< Crab_> Gambit: yes, RCA ai doesn't really honor passive_leader in a few cases where it should do it. can you submit it as a bug, please ?
20100413 00:38:28< Crab_> I'll fix it
20100413 00:39:06< Crab_> (it should still be passive for attacking and strategic movement)
20100413 00:39:14< Gambit> Just "RCA ignores passive_leader"?
20100413 00:40:40< Crab_> Gambit: no, it's better to describe the behavior you're seeing. e.g. 'ai leader moves to keep, ignoring passive_leader=yes'
20100413 00:41:15< Gambit> Just meant the title. I wasn't sure what to call it.
20100413 00:41:19< Crab_> ok
20100413 00:41:30< Crab_> yes, I've about the title
20100413 00:41:53< Crab_> the 'main' movement map of the ai correctly removes the leader's moves, but some candidate actions use their own mini-maps
20100413 00:43:55< Gambit> Crab_: Thanks. https://gna.org/bugs/index.php?15861
20100413 00:44:01< Crab_> ok!
20100413 00:44:23< Gambit> Bye.
20100413 00:44:27< Crab_> bye
20100413 01:00:32-!- happygrue_ [~George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20100413 01:00:47-!- happygrue [~George@wesnoth/developer/wintermute] has quit [Read error: Connection reset by peer]
20100413 01:07:59-!- ancestral [~ancestral@c-24-118-198-88.hsd1.mn.comcast.net] has quit [Quit: ancestral]
20100413 01:19:05-!- Sargeant [~Major@121-73-94-210.cable.telstraclear.net] has joined #wesnoth-dev
20100413 01:21:24-!- Sargeant [~Major@121-73-94-210.cable.telstraclear.net] has quit [Quit: Leaving]
20100413 01:23:03-!- meric [~Eric@124-168-142-121.dyn.iinet.net.au] has quit [Quit: meric]
20100413 01:34:26-!- qemqemqem [~quassel@urwireless-dhcp-128-151-138-208.wireless.rochester.edu] has quit [Ping timeout: 260 seconds]
20100413 01:37:14-!- Sapient [~patrickp@wesnoth/developer/sapient] has joined #wesnoth-dev
20100413 01:40:52-!- Blarumyrran [~Blarumyrr@81-20-159-197.levira.ee] has quit [Quit: Lahkun]
20100413 01:48:17-!- ancestral [~ancestral@97-116-115-135.mpls.qwest.net] has joined #wesnoth-dev
20100413 01:50:07-!- Crab_ [~Crab_@wesnoth/developer/crab] has quit [Quit: Leaving.]
20100413 01:50:59-!- wesbot changed the topic of #wesnoth-dev to: GSoC: Thanks to all GSoC candidates for submitting proposals to Google. Accepted student applications announced on April 26. | 88 bugs, 266 feature requests, 18 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100413 01:55:53-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz]
20100413 01:56:34-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20100413 01:59:47-!- Sapient [~patrickp@wesnoth/developer/sapient] has left #wesnoth-dev []
20100413 02:11:08-!- qemqemqem [~quassel@cpe-74-74-155-246.rochester.res.rr.com] has joined #wesnoth-dev
20100413 02:19:10-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has quit [Quit: Leaving]
20100413 02:31:35< Upthorn> at some point while I was talking in here about eras, I set my laptop safely away and fell asleep
20100413 02:31:55< Upthorn> and kept dreaming that I was in here talking about eras while trying to stay awake.
20100413 02:32:48< Upthorn> AI0867: thank you for the succinct clarification, that is somewhat helpful.
20100413 02:37:07-!- Blueblaze [~nick@adsl-99-188-132-247.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100413 02:49:22-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Quit: This computer has gone to sleep]
20100413 02:49:49-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100413 02:50:41-!- deekay [~dk@wesnoth/developer/dragonking] has quit [Quit: deekay]
20100413 03:02:57-!- gabba [~gabba@212.75-ppp.3menatwork.com] has left #wesnoth-dev []
20100413 03:03:08-!- gabba [~gabba@212.75-ppp.3menatwork.com] has joined #wesnoth-dev
20100413 03:04:57< gabba> mordante, Ivanovic, Crab_: I've updated my calendar again. I think I have to call it final now ;) (at least until april 26)
20100413 03:05:01-!- gabba [~gabba@212.75-ppp.3menatwork.com] has left #wesnoth-dev []
20100413 03:09:35< Upthorn> hrm.
20100413 03:19:26< Upthorn> that calendar is nifty, I should steal it when I have the time.
20100413 03:30:43-!- Appleman1234 [~Appleman1@131.181.47.1] has quit [Ping timeout: 260 seconds]
20100413 03:41:57-!- Zarel [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100413 04:06:22-!- Appleman1234 [~Appleman1@131.181.47.4] has joined #wesnoth-dev
20100413 04:48:23-!- Gambit [~Gambit@pa-67-234-73-7.dhcp.embarqhsd.net] has quit [Quit: Leaving.]
20100413 04:50:52-!- Ivanovic_ [~ivanovic@dtmd-4db2b8bb.pool.mediaWays.net] has joined #wesnoth-dev
20100413 04:52:21-!- aranair_work [~chatzilla@nusnet-195-156.dynip.nus.edu.sg] has joined #wesnoth-dev
20100413 04:54:20-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 258 seconds]
20100413 04:54:50-!- Ivanovic_ is now known as Ivanovic
20100413 05:07:13-!- endercoaster [~endercoas@132.161.243.60] has joined #wesnoth-dev
20100413 05:14:06-!- Espreon [~espreon@wesnoth/developer/espreon] has quit [Quit: WRYYYYYYYYYYYYYYYYYYYY!]
20100413 05:49:00-!- aranair_work [~chatzilla@nusnet-195-156.dynip.nus.edu.sg] has quit [Ping timeout: 265 seconds]
20100413 06:09:59-!- billynux [~billy@239-226-126-200.fibertel.com.ar] has joined #wesnoth-dev
20100413 06:10:37-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Quit: Leaving]
20100413 06:11:39-!- pokhbocee [~pokhbocee@ws40.cs.drexel.edu] has joined #wesnoth-dev
20100413 06:13:50-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Quit: This computer has gone to sleep]
20100413 06:51:19-!- billynux [~billy@239-226-126-200.fibertel.com.ar] has quit [Quit: Leaving]
20100413 07:02:17-!- endercoaster [~endercoas@132.161.243.60] has left #wesnoth-dev []
20100413 07:09:21-!- iwaim [~iwaim@2001:2c0:40e::1] has quit [Ping timeout: 268 seconds]
20100413 07:10:49-!- Appleman1234 [~Appleman1@131.181.47.4] has quit [Ping timeout: 264 seconds]
20100413 07:13:37-!- Elsewise [~Dmitry@80.233.231.12] has joined #wesnoth-dev
20100413 07:16:48-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Quit: crimson_penguin]
20100413 07:29:40-!- Appleman1234 [~Appleman1@131.181.47.4] has joined #wesnoth-dev
20100413 07:31:28-!- Elsewise [~Dmitry@80.233.231.12] has quit [Ping timeout: 265 seconds]
20100413 07:33:34-!- Netsplit *.net <-> *.split quits: wesbot, ettin, mordocai, lukjad86, Ivanovic, Vetinari
20100413 07:33:43-!- Netsplit *.net <-> *.split quits: aranair, apoi, freim, teaser, Appleman1234, ctrlfreak, Polarina, Bocom, Deiz
20100413 07:35:32-!- dtiger [~dtiger@dynamic-vpdn-93-125-13-17.telecom.by] has joined #wesnoth-dev
20100413 07:35:38-!- Netsplit over, joins: Ivanovic, teaser, Appleman1234, mordocai, ettin, Polarina, lukjad86, Vetinari, apoi, wesbot (+5 more)
20100413 07:36:35-!- Netsplit *.net <-> *.split quits: esr, shikadibot, AnMaster, Smar, noy, isaac, Soliton, AI0867
20100413 07:37:11-!- Netsplit over, joins: noy, AI0867, Soliton, AnMaster, esr, shikadibot, isaac, Smar
20100413 07:37:45-!- mjs-de [~mjs-de@vpw.wh.Uni-Dortmund.DE] has joined #wesnoth-dev
20100413 07:38:30-!- pokhbocee [~pokhbocee@ws40.cs.drexel.edu] has quit [Quit: Ex-Chat]
20100413 07:45:51-!- mjs-de [~mjs-de@vpw.wh.Uni-Dortmund.DE] has quit [Remote host closed the connection]
20100413 07:54:54-!- timotei [timotei@2001:b30:5000:11:f5cd:f746:93fb:b60c] has joined #wesnoth-dev
20100413 08:12:11-!- billynux [~billy@239-226-126-200.fibertel.com.ar] has joined #wesnoth-dev
20100413 08:27:38-!- StandYourGround [~Adium@230.121.55.224.guam.net] has joined #wesnoth-dev
20100413 08:31:05< billynux> hi Ivanovic, you there?
20100413 08:32:01< noy> he's asleep for another few hours
20100413 08:32:11-!- timotei [timotei@2001:b30:5000:11:f5cd:f746:93fb:b60c] has quit [Ping timeout: 248 seconds]
20100413 08:33:44-!- StandYourGround [~Adium@230.121.55.224.guam.net] has quit [Quit: Leaving.]
20100413 08:35:44-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100413 08:36:28-!- zookeeper [~l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20100413 08:38:16< zookeeper> urgh. no forum admins around?
20100413 08:39:31< billynux> I guess its a good time for sleeping wherever they are :)
20100413 08:39:52-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100413 08:40:11-!- noy [~Noy@wesnoth/developer/noy] has quit [Client Quit]
20100413 08:40:19< zookeeper> yeah, well, i'd settle for anyone who could make the spam stop
20100413 08:40:55-!- StandYourGround [~Adium@230.121.55.224.guam.net] has joined #wesnoth-dev
20100413 08:47:34-!- StandYourGround [~Adium@230.121.55.224.guam.net] has left #wesnoth-dev []
20100413 08:48:38-!- fakedrake [~ece7361@YK1-PC44.ece.upatras.gr] has joined #wesnoth-dev
20100413 08:56:16-!- Crab_ [~Crab@wesnoth/developer/crab] has joined #wesnoth-dev
20100413 08:59:51< billynux> hi Crab_ 
20100413 08:59:57< Crab_> hi, billynux
20100413 09:00:10< billynux> I updated the wiki with some answers for Ivanovic 
20100413 09:00:17< billynux> (mainly documentation stuff)
20100413 09:00:25< Crab_> ok
20100413 09:00:33< billynux> I'm now extending with some more technical information
20100413 09:00:40< Crab_> I will take a look at it, too
20100413 09:00:54< billynux> when I'm done I'll let you guys know through a GSoC comment
20100413 09:01:20< billynux> Crab_, no need to look at it now, I'll let you know via that comment
20100413 09:01:50< Crab_> ok. but I've looked at it already :))
20100413 09:01:59< Crab_> will wait till you ping us on google's site, then.
20100413 09:05:48-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100413 09:13:58< billynux> ok Crab_ , standing by...
20100413 09:16:53< Upthorn> Hi, Crab_
20100413 09:17:11< Upthorn> I've made a bunch of changes to my wiki proposal (and submitted a patch) since I've last heard comment from you. Have you had time to look at any of these yet? (answer after you're done with billynux)
20100413 09:18:35< billynux> Oh Upthorn, no need to take turns, I'm sure that Crab_ has multithreading capacities :) , I don't have any pending questions ATM
20100413 09:18:47-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20100413 09:19:55< Upthorn> well I'm not sure my questions are thread-safe and I don't want to cause heap corruption
20100413 09:20:07< billynux> :))
20100413 09:21:21< billynux> These GSoC proposals must be causing a lot of mentors' CPUs to overheat --> (recommends beer)
20100413 09:23:01< Upthorn> Yeah. Crab_ particularly seems to be kept quite busy at them
20100413 09:24:08< Crab_> hi, Upthorn
20100413 09:24:42-!- ancestral [~ancestral@97-116-115-135.mpls.qwest.net] has quit [Quit: And that’s the end of THAT chapter.]
20100413 09:24:49< Upthorn> Though I guess Ivanovic and boucman are putting in a lot of time as well
20100413 09:25:06< Crab_> well, stayed awake for 34+ hours in sunday&yesterday, fo various reasons. still, feels quite good.
20100413 09:26:07< Upthorn> That would explain quite a bit.
20100413 09:28:47< Crab_> Upthorn: I've looked through your wiki page and comments at google site, will take a look at the new patch atm.
20100413 09:29:16< Upthorn> thanks
20100413 09:30:46< Crab_> comment about the patch: I suggest rewriting is_valid_namespace to use a whitelist, not a blacklist. this is safer.
20100413 09:31:29< Upthorn> yeah, i was wondering what a safer way to do that would be
20100413 09:31:55< Upthorn> because I only know the invalid characters for NTFS and Fat32.
20100413 09:32:45< Crab_> Upthorn: we already depend on boost::regex, you can use it, or (simpler) use http://www.cplusplus.com/reference/string/string/find_first_not_of/
20100413 09:33:31< Upthorn> I know string::find_first_not_of. What I meant was that whitelisting hadn't occured to me.
20100413 09:33:56< Crab_> Upthorn: and note that '!str.empty()' is, imo, a more readable way to check if string is empty, compared to  (str != "")
20100413 09:34:04< Crab_> Upthorn: ok
20100413 09:34:14< Upthorn> Oh, yeah, that's a good point.
20100413 09:34:29-!- fabi [~fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20100413 09:34:39-!- fendrin [~fabi@wesnoth/developer/fendrin] has quit [Ping timeout: 260 seconds]
20100413 09:39:36-!- Blueblaze [~nick@adsl-99-188-132-247.dsl.hstntx.sbcglobal.net] has quit [Remote host closed the connection]
20100413 09:46:31-!- euschn [~chatzilla@wesnoth/developer/euschn] has joined #wesnoth-dev
20100413 09:46:47< Crab_> Upthorn:  you don't check if global or local variable names are empty/nonexistent, I think that it's better to check them and log a error if they are. for example, to catch misspelled attributes earlier and output saner error messages
20100413 09:49:01< Crab_> Upthorn: and, do you want the cases for 'global variable is empty' and 'there is no global variable' to be handled in the same way or in different way ?
20100413 09:49:25< Upthorn> the same way.
20100413 09:50:03< Upthorn> I think.
20100413 09:50:23< Upthorn> I want global variables to work approximately the same way as local ones.
20100413 09:50:52< Crab_> Upthorn: then, the call to set_global_variable to the empty value should result in a call to 'clear_global_variable' ?
20100413 09:50:56< Crab_> or not ?
20100413 09:51:50< Upthorn> Crab_: I was thinking that set_global_variable with no value should be a synonym for [clear_global_variable], but I didn't implement it yet.
20100413 09:52:29< Crab_> Upthorn: ok. also you have an error in clear_global_variable - in std::string global = pcfg["to_global"];    it should be std::string global = pcfg["global"];
20100413 09:52:57< Crab_> Upthorn: (probably a copypaste thing)
20100413 09:52:59< Upthorn> err, we changed the attribute name?
20100413 09:53:11< Crab_> http://wiki.wesnoth.org/SummerOfCodeProposal_JodyNorthup#WML_Syntax
20100413 09:53:16< Crab_> no, we havent
20100413 09:53:24< Crab_> [clear_global_variable] has    global=my_variable_name
20100413 09:53:25< Upthorn> oh. I misread it from the beginning.
20100413 09:53:48< Crab_> so, either the documentation or the code is wrong :)
20100413 09:54:01< Upthorn> I copy-pasted that from http://wiki.wesnoth.org/SoC_Ideas_Persistent_Gameworld#WML_Syntax without reading it
20100413 09:54:52< Crab_> Upthorn: also, in clear_global_variable, why do config var = cfg.child_or_empty(global); ?
20100413 09:54:54< Upthorn> honestly, I have been considering proposing that "from_" and "to_" in the attribute names be dropped entirely.
20100413 09:55:56< Crab_> Upthorn: also good. but it will mean more bug where we use get_ instead of set_
20100413 09:56:18< Crab_> using different from and to would mean that wmllint and engine would be able to spot such things immediately
20100413 09:56:27< Upthorn> Ah.
20100413 09:56:42< Crab_> if we use the same syntax, then we're generally unable to spot this by syntax checking
20100413 09:58:45< Crab_> note that 'config var = cfg.child_or_empty(global);' will do a copy of the config child, which is not generally wanted.
20100413 09:58:48-!- fakedrake [~ece7361@YK1-PC44.ece.upatras.gr] has left #wesnoth-dev []
20100413 09:59:11< Upthorn> Crab_: the reason I use child_or_empty there is because at that point, I don't know if the specified variable is a structure, is empty, or does not exist.
20100413 09:59:39< Crab_> e.g., you've loaded the entire namespace.cfg into a config cfg, you shouldn't do any more copies as you can change that cfg.
20100413 10:00:01< Crab_> you can get a child and test it for validity
20100413 10:00:15< Crab_> config (in boolean context), can be either valid or invalid
20100413 10:00:18< Crab_> (true/false)
20100413 10:00:40< Upthorn> when I used find_child(global,"","") it threw an error when the child didn't exist.
20100413 10:01:15< Upthorn> what should I use to get a child and test it for validity?
20100413 10:02:33< Upthorn> ... oh. it looks like cfg.child() will do that.
20100413 10:03:44< Crab_> things like   " if (config &resistance = cfg_.child("resistance")) "
20100413 10:04:12< Crab_> note that the if would be false if the child is absent
20100413 10:04:26< Crab_> (note that the 'if' would be false if the child is absent)
20100413 10:05:40< Upthorn> I see.
20100413 10:06:25< Crab_> also, if we cleared a persistent variable that hasn't existed, then you'd call save_persist_data without reason to do so.
20100413 10:08:16-!- DDR_ [~chatzilla@66.183.125.196] has joined #wesnoth-dev
20100413 10:09:11< Upthorn> I see that.
20100413 10:10:31< Upthorn> I meant to add {}s to the if(exists), but I got distracted with testing and forgot.
20100413 10:14:58-!- DDR_ [~chatzilla@66.183.125.196] has quit [Ping timeout: 268 seconds]
20100413 10:17:25< Crab_> Upthorn: I think that it would be a good idea to make that information 'managed', e.g., make code to track if the 'information' is dirty for you.
20100413 10:17:56< Upthorn> Which information?
20100413 10:18:01< Crab_> Upthorn: or, alternatively, double-check that the code works ok in all situations
20100413 10:18:48< Crab_> the 'config for a particular namespace'
20100413 10:19:22< Crab_> note that there most probably will be need for 'transaction blocks', to set/clear/get multiple things during a single all-or-nothing run.
20100413 10:21:20< Crab_> e.g., it might be a good idea to make the code layered - first layer will take the information from game_events.cpp, create a persistence context, and ask the second layer to do the actual dirty work in that context.
20100413 10:21:44< Crab_> that way, your actual get/set implementations will not have to deal with files/dirty checks/etc
20100413 10:22:25-!- Ivanovic [~ivanovic@dtmd-4db2b8bb.pool.mediaWays.net] has quit [Changing host]
20100413 10:22:25-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20100413 10:23:09< Crab_> upthorn: also, we should try to abstract the 'get storage location for a particular variable' part, since eventually we might want to store it elsewhere, not inside a 'config' taken from file, but, say, as a DB row.
20100413 10:24:29< Crab_> Upthorn: I've seen that you've said that the patch is simple. yes, it's indeed simple to code. but we need to make it bulletproof because it'll be a foundation of our entire persistence system.
20100413 10:25:59< Upthorn> You're absolutely right. But do you want all of this done in the patches.wesnoth.org submission?
20100413 10:26:49< Crab_> Upthorn: we can create a svn branch for that, as well
20100413 10:27:21< Crab_> Upthorn: since the changes are well-localized (as you've done it the right way, in a separate file), merging should be easy.
20100413 10:27:45< Crab_> Upthorn: you're up to it and have time this week to work on it?
20100413 10:28:18< Ivanovic> moin
20100413 10:29:43< Crab_> hi, Ivanovic:
20100413 10:31:30< zookeeper> Ivanovic, the forum needs you
20100413 10:32:03< zookeeper> take a look at "view new posts" ...
20100413 10:32:44< Ivanovic> you mean 'anthonylau' ?
20100413 10:32:46< zookeeper> yeah
20100413 10:35:22< Ivanovic> anotherone?
20100413 10:37:02< Upthorn> Crab_: I am not sure how much time I have this week, but I should be able to put a few hours here and there into it, yes.
20100413 10:38:56< zookeeper> Ivanovic, seems to have been the only account used
20100413 10:39:15 * Ivanovic loves the 'delete all posts from user' feature that admins got...
20100413 10:39:27< Crab_> Upthorn: then let's do it this way: fix those issues and make the code as good as possible without introducing any new features, submit a new patch with it, and, if it's ok, we'll start a branch for it, and we'll be able to work on it directly.
20100413 10:40:44-!- deekay [~dk@wesnoth/developer/dragonking] has joined #wesnoth-dev
20100413 10:41:34-!- euschn [~chatzilla@wesnoth/developer/euschn] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.0.19/2010040121]]
20100413 10:41:50-!- euschn1 [~euschn@tmp6.logic.tuwien.ac.at] has joined #wesnoth-dev
20100413 10:42:59< Crab_> Upthorn: will it be ok ?
20100413 10:43:50-!- euschn1 [~euschn@tmp6.logic.tuwien.ac.at] has quit [Client Quit]
20100413 10:44:53-!- euschn [~euschn@wesnoth/developer/euschn] has joined #wesnoth-dev
20100413 10:47:00< fabi> Ivanovic: I wish one could sue spammers more easily. Or send them letterbombs.
20100413 10:47:01< Upthorn> that sounds ok, but I have a couple of questions.
20100413 10:47:43< Upthorn> I'm still not entirely sure what you mean about making information "managed"
20100413 10:48:57< Upthorn> (I guess that's not really a question)
20100413 10:50:10< Upthorn> And do you want abstraction and management in the patch, or can those wait for SVN?
20100413 10:51:00< Upthorn> Crab_ also, I apologize for the delayed reply. I was looking through the logs to compile a list of tasks.
20100413 10:51:22< Crab_> Upthorn: about the reply - all is ok
20100413 10:51:57< Crab_> Upthorn: 'managed' I mean that the code that modifies/clears the variable should not care about the storage location, it should only do the changes and raise a 'dirty' flag, if necessary.
20100413 10:53:05< Upthorn> oaky, so management and abstraction are the same thing
20100413 10:53:38< Upthorn> *okay
20100413 10:54:40< Upthorn> Ivanovic: Boucman suggested that I should ask you if my timeline estimates have enough detail, etc
20100413 10:54:49< Upthorn> http://wiki.wesnoth.org/SummerOfCodeProposal_JodyNorthup#Include_an_estimated_timeline_for_your_work_on_the_project._Don.27t_forget_to_mention_special_things_like_.22I_booked_holidays_between_A_and_B.22_and_.22I_got_an_exam_at_ABC_and_won.27t_be_doing_much_then.22.
20100413 10:55:24< Crab_> Upthorn: e.g., I was thinking of providing a c++ 'persistence_context &'  inout parameter to the get/set/clear functions, so they'll be able to call context.set_dirty() if they change something.
20100413 10:55:39< Upthorn> that link is longer than I realized.
20100413 10:56:03< Rhonda> Wow, that's a link. :)
20100413 10:56:23< Upthorn> Crab_: I see, so that the persistence data can be loaded once per session, instead of once per WML event?
20100413 10:57:37< Crab_> Upthorn: then we can change the 'the persistence data can be loaded once per session, instead of once per WML event' without changing the code of the get/set functions.
20100413 10:57:59< Ivanovic> Upthorn: to be honest i prefer having a table marking some thing as optional instead of having two timelines which you will most likely not follow
20100413 10:58:00< Crab_> Upthorn: so, we'll be able to try various approaches, e.g., transaction blocks or once-per-turn or other saves
20100413 10:58:22< Upthorn> I see. Okay, yeah that does seem reasonable.
20100413 10:58:26< billynux> Ok Crab_ , I'm done for the time being: * Updated wiki and * Sent a comment through GSoC
20100413 10:58:51< fabi> Upthorn: Is Jody a male or female name?
20100413 10:58:54< billynux> Gotta sleep now! See you later
20100413 10:59:09< billynux> ... Jody Foster? -> Female... I think
20100413 10:59:11< Ivanovic> billynux: my comment regarding "hard to comment" was more like "okay, sounds rather good so far, but somthing seems to be missing..."
20100413 10:59:33< billynux> ok Ivanovic , it would be great if you can pinpoint whats missing :)
20100413 10:59:40< Ivanovic> billynux: without me being able to pinpoint it *at all*
20100413 10:59:42< fabi> billynux: I thought female as well, but Upthorn sounds more male.
20100413 10:59:44< billynux> :D
20100413 11:00:03< Ivanovic> billynux: so i trust in the other devs to pinpoint those details
20100413 11:00:04< Ivanovic> ;)
20100413 11:00:10< billynux> great, thanks!
20100413 11:00:44< Crab_> billynux: ok. thanks!
20100413 11:00:51< billynux> boucman was also vague in some points, and it is tough to stop being "light on details" if you don't know where to add them
20100413 11:01:25< billynux> so... off now, thanks again, good luck Upthorn 
20100413 11:01:33-!- billynux [~billy@239-226-126-200.fibertel.com.ar] has quit [Quit: Leaving]
20100413 11:02:53< Upthorn> fabi: the short version is that there are about as many males with the name Jody as there are females
20100413 11:03:31< fabi> Upthorn: I see. So I guess you are male.
20100413 11:03:33< Upthorn> And I, specifically, am male.
20100413 11:04:05< Upthorn> However, there are also two other spellings of the name "Jodi" and "Jodie"
20100413 11:04:12< Upthorn> which are exclusively female.
20100413 11:04:37< Upthorn> And there is no spelling that is exclusively male.
20100413 11:05:17< fabi> I see.
20100413 11:05:42< Crab_> Upthorn: note http://wesnoth.pastebin.com/5vKYwZcA
20100413 11:05:57 * Ivanovic thinks that names really are something strange...
20100413 11:05:58< Ivanovic> ;)
20100413 11:06:07< Crab_> one of the ways to do that 'abstraction', but ther are other, as well.
20100413 11:06:07-!- Appleman1234 [~Appleman1@131.181.47.4] has quit [Ping timeout: 246 seconds]
20100413 11:07:08< Upthorn> Why not have cfg be a member of persistence_context?
20100413 11:07:27< Upthorn> oh right, database.
20100413 11:07:49< Crab_> because cfg is what we got from user, a bunch of commands. pcfg in your patch
20100413 11:07:59< Upthorn> oh.
20100413 11:08:18< Crab_> and persistence_context is all the c++ info we 'associate' with this request or a group of requests.
20100413 11:08:41-!- loonybot [~loonybot@ppp79-139-136-187.pppoe.spdop.ru] has joined #wesnoth-dev
20100413 11:08:41-!- loonybot [~loonybot@ppp79-139-136-187.pppoe.spdop.ru] has quit [Changing host]
20100413 11:08:41-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20100413 11:09:31< Crab_> in that pseudocode, that context dealt with the namespace itself, leaving get and set to deal with packing/unpacking of the value of the variable
20100413 11:09:35-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100413 11:09:51< Crab_> Upthorn hat is to cleanly separate the responsibilites of various pieces of code.
20100413 11:09:55< Crab_> s/hat/that
20100413 11:10:31< Upthorn> But also, if I had a hat, the reason for the hat would be to cleanly separate the responsibilities of various pieces of code.
20100413 11:10:41< Upthorn> :)
20100413 11:10:44< Crab_> :))
20100413 11:11:02< Upthorn> since you said about making it layered, it seemed like it might be good to have the outer function build a structure which gets passed to the worker functions.
20100413 11:11:08< Crab_> yes
20100413 11:11:23< Crab_> so, I think that it is a good idea to deal with all that 'wml arrays' stuff, and such. in one place, and in 'find a good place to save/load the resulting config' in another place
20100413 11:11:50< Upthorn> so instead of "const config &cfg" we could have "const persist_request &req"
20100413 11:12:14< Crab_> well, config is good, too. because it's easy to store anything in it
20100413 11:12:35< Crab_> so, that's why I was supplying both a 'original config' and a 'persistence context'
20100413 11:12:56< Crab_> there's no reason to introduce a new c++ class if you simply want to use it as a DTO
20100413 11:12:59< Upthorn> Good point. If we add something to the event syntax, we would have to modify the persist_request structure every time
20100413 11:14:06< Crab_> as config is very good for that. but we want the c++ context to hide that 'find a current serialized config of the variable' and 'store this serialized representation in the right place' methods, with dirty tracking, transaction blocks, and other such fancies.
20100413 11:14:06< Upthorn> and there's no need if we just make sure always to sanitize inputs to the work functions beforehand.
20100413 11:14:47< Crab_> yes. actually, the point of 'sanitizing input' is to spawn the best error message possible
20100413 11:15:33< Upthorn> and if an error is encountered, there is no reason even to attempt to perform the work
20100413 11:18:16< Crab_> yes
20100413 11:58:51< Crab_> Upthorn: : about your timeline...
20100413 12:01:05< Crab_> your timeline is fine, but I want additional section to be added to the list. A section which will represent your commitment to specific deliverables, like http://wesnoth.pastebin.com/8v9Hb5Es
20100413 12:02:27-!- Chusslove [~Chusslove@unaffiliated/chusslove] has quit [Remote host closed the connection]
20100413 12:02:53< Crab_> (in case you have worries that things-to-do will run out, just add more items to optional list. but, there are big things on it already :) )
20100413 12:03:03-!- Appleman1234 [~Appleman1@CPE-124-191-176-143.oxqn1.cha.bigpond.net.au] has joined #wesnoth-dev
20100413 12:03:42< Upthorn> Crab_: thanks. I will modify the timeline tomorrow, if I have time.
20100413 12:04:04< Crab_> Upthorn: no reason to 'modify' your timeline, your timeline is fine for me.
20100413 12:04:48< Crab_> Upthorn: just add a new short-list-of-things-that-are-to-be-done (required things), and short-list-of-things-that-will-be-cool-to-do (optional things you want to try)
20100413 12:11:53-!- Appleman1234 [~Appleman1@CPE-124-191-176-143.oxqn1.cha.bigpond.net.au] has quit [Ping timeout: 246 seconds]
20100413 12:14:53-!- iwaim [~iwaim@2001:2c0:40e::1] has joined #wesnoth-dev
20100413 12:33:49< Crab_> icelus: btw, a funny thing to check: what will happen if "player A has got an 'select advance' screen, other players wait for his input, and then, player's A wesnoth segfaults or he drops connection" - will the game continue ? will the unit advance ? if so, how it will advance ?
20100413 12:53:59-!- ancestral [~ancestral@97-116-115-135.mpls.qwest.net] has joined #wesnoth-dev
20100413 12:55:53-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has joined #wesnoth-dev
20100413 13:03:28< ancestral> Do I need to be in a certain directory when I run wmlparser?
20100413 13:05:19< Ivanovic> ancestral: most likely
20100413 13:05:33< Ivanovic> ancestral: that is: there is probably a --help entry stating where you should start
20100413 13:05:43< ancestral> Yes, I have done that
20100413 13:06:26< ancestral> Okay I might be missing an argument
20100413 13:09:13< ancestral> :x Doesn't say which ones are mandatory!
20100413 13:09:13-!- Appleman1234 [~Appleman1@CPE-124-191-176-143.oxqn1.cha.bigpond.net.au] has joined #wesnoth-dev
20100413 13:09:29< ancestral> Should be obvious, I guess
20100413 13:11:28< ancestral> Okay so I'm using -u which I assume must be mandatory
20100413 13:11:46< ancestral> oh -p my mistake
20100413 13:20:55-!- qemqemqem [~quassel@cpe-74-74-155-246.rochester.res.rr.com] has quit [Ping timeout: 265 seconds]
20100413 13:21:16< ancestral> Alright, it's working now :) AI0867 thank you for adding in JSON output
20100413 13:36:33-!- Lastmerlin [~Lastmerli@kalypso.csn.tu-chemnitz.de] has joined #wesnoth-dev
20100413 13:50:43-!- ancestral [~ancestral@97-116-115-135.mpls.qwest.net] has quit [Quit: ancestral]
20100413 13:50:58-!- wesbot changed the topic of #wesnoth-dev to: GSoC: Thanks to all GSoC candidates for submitting proposals to Google. Accepted student applications announced on April 26. | 89 bugs, 266 feature requests, 18 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100413 13:56:22-!- Ken_Oh [~briang@static-71-178-174-220.washdc.fios.verizon.net] has joined #wesnoth-dev
20100413 14:31:51-!- Johannes13 [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20100413 14:36:41< euschn> can anyone tell me what the purpose of unit_map::num_iters_ is? its a private field that never seems to be modified
20100413 14:37:29< euschn> Im looking at bug #15791 atm and sometimes I get an assertion failure containing num_iters_ when testing
20100413 14:40:50< Crab_> hi, euschn
20100413 14:41:01< euschn> hey Crab_
20100413 14:42:40< Crab_> well, the map has to keep track of the number of iterators that are still pointed to it
20100413 14:43:02< Crab_> since when the unit_map is destroyed, the number of iterators should be 0
20100413 14:43:13< Crab_> that's a 'safety net'
20100413 14:43:20-!- shadowm_laptop [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100413 14:44:42-!- shadowm_laptop [~ignacio@wesnoth/developer/shadowmaster] has quit [Client Quit]
20100413 14:47:08< Crab_> euschn: so, it seems, that, that bug is about some iterators not being killed on reset to beginning
20100413 14:47:27< Crab_> s/bug/assertion failure
20100413 14:50:46< euschn> Crab_: I see - what confuses me however is, that I dont see num_iters_ modified at any point - add_iter() or remove_iter() are never called
20100413 14:50:59< euschn> at least thats what my ide's search function says
20100413 14:52:28-!- Gambit [~Gambit@pa-67-234-73-7.dhcp.embarqhsd.net] has joined #wesnoth-dev
20100413 14:56:38< Crab_> euschn: unit_map.hpp:75 ?
20100413 14:56:58-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100413 14:57:06< Crab_> euschn: it's done by iterators
20100413 14:58:11< euschn> ah thanks, I overlooked that
20100413 14:59:18-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100413 15:00:40-!- Zarel [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100413 15:14:39-!- happygrue_ is now known as happygrue
20100413 15:15:32< Ivanovic> Crab_: have you experienced this "wml child missing" bug already?
20100413 15:15:56< Ivanovic> Crab_: that is: silene had a look at it and was not able to far to find the likely cause in 1.8
20100413 15:16:06< Ivanovic> (fixed on trunk specific case though)
20100413 15:16:21< Ivanovic> IIRC he made it crash now so that we can get an assert
20100413 15:16:35< Ivanovic> Johannes13: were you not trying to get us some debug output?
20100413 15:16:40< Ivanovic> Johannes13: any news on this?
20100413 15:17:51-!- Appleman1234 [~Appleman1@CPE-124-191-176-143.oxqn1.cha.bigpond.net.au] has quit [Quit: Leaving]
20100413 15:18:15< Johannes13> yes...
20100413 15:18:25< Johannes13> but the error doed not occour...
20100413 15:18:34< Johannes13> I don't know why
20100413 15:18:34-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Quit: This computer has gone to sleep]
20100413 15:18:44< Ivanovic> lovely...
20100413 15:18:51< Johannes13> But I also don't know why the error occours...
20100413 15:19:34< Johannes13> ? svn
20100413 15:19:40< Johannes13> wesbot: wvn
20100413 15:19:43< Johannes13> wesbot: svn
20100413 15:19:46< Johannes13> wesbot: svn?
20100413 15:19:53< Ivanovic> Johannes13: what do you want to know?
20100413 15:20:03< Johannes13> the link for the svn
20100413 15:20:10< Ivanovic> *which* link for svn?
20100413 15:20:11< shadowmaster> shikadibot: repo
20100413 15:20:12< shikadibot> Repository URI: http://svn.gna.org/svn/wesnoth
20100413 15:20:18< Ivanovic> howto compile it? where the repo is?
20100413 15:20:19< Johannes13> yeah
20100413 15:20:29< Ivanovic> in general: try svn.wesnoth.org to find the info
20100413 15:20:30< shadowmaster> Johannes13: see above, svn co <that URI>/trunk wesnoth
20100413 15:20:31< Johannes13> where the repo is
20100413 15:20:37< Johannes13> yes
20100413 15:20:49< Ivanovic> shadowmaster: in fact testing branches/1.8 would be better
20100413 15:21:07< shadowmaster> in such case replace /trunk with /branches/1.8
20100413 15:21:14< Ivanovic> shadowmaster: since trunk already has many changes which might (by "accident" or sheer luck) fix the main cause for the missing wml child stuff
20100413 15:21:34< Johannes13> should I checkout 1.8 or trunk?
20100413 15:21:51< Ivanovic> in general it depends on what you want to do
20100413 15:22:03< Ivanovic> if you mainly want to play with other users: get branches/1.8
20100413 15:22:03< Johannes13> look at the code
20100413 15:22:15< Ivanovic> that is the latest version and will eventually be a new 1.8.x release
20100413 15:22:34< Ivanovic> where trunk is the development version with myriads of new features and often as many bugs
20100413 15:22:40< Johannes13> I know
20100413 15:22:51-!- icelus [~ed@cpc9-sgyl11-0-0-cust29.sgyl.cable.virginmedia.com] has joined #wesnoth-dev
20100413 15:22:59< Ivanovic> branches/1.8 on the other hand is bugfixing only
20100413 15:23:06< Ivanovic> (at least meant to be)
20100413 15:23:40< Johannes13> changes in 1.8 should not break compat for other 1.8 versions
20100413 15:24:03< Ivanovic> correct
20100413 15:24:34< Johannes13> :) Usualy I prefer java or tcl, but I also know a little bit c/c++
20100413 15:29:28-!- FAAB [~huajie@219.142.240.43] has joined #wesnoth-dev
20100413 15:30:57< FAAB> Crab_: hi, I have file a feature request, could you check it when you have time [https://gna.org/bugs/?15850] ?
20100413 15:31:10< Crab_> Ivanovic: no, I've not experienced it.
20100413 15:31:32< Crab_> Ivanovic: that's sad, because I'm usually interested in funny bugs.
20100413 15:31:34< Ivanovic> okay, now the interesting question is: how comes that this many users seem to get it?
20100413 15:32:04< Ivanovic> considering the amount of posts in the forums that are about the missing child it seems to be a rather common problem
20100413 15:32:07< FAAB> deekay: BTW I tried both 'end' and 'continue' and it didn't work
20100413 15:32:33< Crab_> FAAB: Yes, I've read that FR.
20100413 15:32:52< FAAB> what do you think ?
20100413 15:33:46< Crab_> FAAB: you're doing things slightly wrong.
20100413 15:34:28< FAAB> ok, how should I do it ?
20100413 15:34:38< Crab_> FAAB: you see, if a certain candidate action scores highest and gets selected for execution, BUT then does nothing (e.g., moves on spot), then nothing will change, and, in the next evaluation, the SAME move will be selected for execution, leading to the not-no-nice infinite loop
20100413 15:35:03< Crab_> that's why partial move from me.loc to me.loc is an error.
20100413 15:35:17-!- Elsewise [~Dmitry@80.233.231.12] has joined #wesnoth-dev
20100413 15:35:52< Crab_> FAAB: as of now, the engine will prevent that infinite loop, by blacklisting your candidate action till end of turn. but that's, too, not the result which is wanted.
20100413 15:36:42< Crab_> the usual cause of this is 'your evaluation function is not strict enough to check if there's a 'real' move available
20100413 15:36:57< Crab_> doing it the 'right way' is easier with lua ai and c++ ai, but is possible with formula ai, too
20100413 15:37:10< FAAB> well the things on the map haven't changed but the stack of candidate actions has changed as the last one has been executed
20100413 15:37:33< Crab_> after each move, all candidate actions in a current stage are considered for execution
20100413 15:37:34< FAAB> so how could it be executed again ?
20100413 15:37:41< Crab_> so, the stack of candidate actions remains the same
20100413 15:37:49< FAAB> I see
20100413 15:38:25< Crab_> you should calculate the move in evaluate part, and return >0 only if you have a 'good, valid' move on hand
20100413 15:38:54< Crab_> (lua and c++ can more easily pass state from evaluate to execute, but fai can use set_var or just evaluate the same thing 2 times)
20100413 15:39:08< FAAB> ok, I think I got it
20100413 15:39:51< FAAB> what am I writing when I write a formula AI ? which language ?
20100413 15:39:52< Crab_> so, your evaluate function needs to check 'there is a village in my moves'
20100413 15:39:58< Ivanovic> Crab_: in general this error does seem to only occur in multiplayer games when units do (try to) attack
20100413 15:40:09< Ivanovic> and it seems like mainly the host if affected
20100413 15:40:32< Ivanovic> so far there were many reporters with windows, some with linux and mac though, too
20100413 15:40:39< Crab_> Ivanovic: is there a 'dual replay' (replay from both sides) available for at least one of the cases?
20100413 15:40:43< Ivanovic> so probably not platform specifc
20100413 15:40:48< Ivanovic> Crab_: i don't think so
20100413 15:42:21< Crab_> ok, will try some running-around-in-a-debugger things, maybe I'll get lucky.
20100413 15:42:46< Crab_> FAAB: 'formula ai' (fai) is a domain-specific functional language.
20100413 15:44:00< FAAB> which one from fai or lua will be supported on the long term ?
20100413 15:44:14< Crab_> lua
20100413 15:44:24< Crab_> unless someone appears who will care more about formula ai.
20100413 15:44:48< Crab_> formula ai will not get dropped, but it's unlikely that it would be improved apart from bugfixes.
20100413 15:46:12< Crab_> FAAB: I expect that, especially if 'lua ai' project gets selected for GSoC, lua ai will be ready in 2-3 months (there's already prototype code in 1.8, which works and can use the full power of LuaWML, and do stages and candidate actions)
20100413 15:46:49< Crab_> FAAB: so, to fix your particular problem, you should move your 'If currently considered unit has a move to village'  part to evaluate
20100413 15:47:40< FAAB> Crab_: understood, I will look how to do that
20100413 15:48:13< Crab_> ok. ask questions, if you'll need to know more about anything ai-related.
20100413 15:48:31< FAAB> I will, thanks !
20100413 15:49:45< Ivanovic> okay, opened a bug report for the missing wml child problem
20100413 15:49:55< Ivanovic> just that we can keep track of things more easily
20100413 15:50:01< Ivanovic> https://gna.org/bugs/index.php?15865
20100413 15:51:44< Crab_> ok
20100413 15:59:00< FAAB> Crab_: it work much better like this, thanks a bunch !
20100413 16:00:51< FAAB> feel free to reject the feature request ;)
20100413 16:02:47< Crab_> FAAB: thanks :))
20100413 16:10:34-!- FAAB [~huajie@219.142.240.43] has left #wesnoth-dev []
20100413 16:28:46-!- fabi [~fabi@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20100413 16:30:10-!- euschn [~euschn@wesnoth/developer/euschn] has quit [Quit: Leaving.]
20100413 16:34:14-!- iwaim [~iwaim@2001:2c0:40e::1] has quit [Ping timeout: 268 seconds]
20100413 16:34:38-!- timotei [~timotei@193.34.191.4] has joined #wesnoth-dev
20100413 16:37:22< Crab_> hi, timotei
20100413 16:37:26< timotei> hello everyone!
20100413 16:37:28< timotei> hello Crab_ 
20100413 16:37:35< timotei> at last caught you online:D
20100413 16:37:39< Crab_> how it's going ? :)
20100413 16:37:57< timotei> well, wanted yesterday to talk with you about the wesnoth way of start
20100413 16:38:05< timotei> but in the meantime, looked over the codebase
20100413 16:38:50< timotei> I will continue working on starting a specified campaign from the command line 
20100413 16:39:05< timotei> btw, talked with mordante, and he told me the thing you wanted me to do is "impossible"
20100413 16:39:08< timotei> if you read the logs
20100413 16:40:51< Crab_> well, I see " I still think it would be a good thing to do, but not any time soon" in the logs :)
20100413 16:41:02-!- Elsewise [~Dmitry@80.233.231.12] has quit [Ping timeout: 246 seconds]
20100413 16:41:40< timotei> well, that too
20100413 16:41:54< Crab_> so ok, if you cant quickly hack it it the way I've said (I suspect that it's easier than mordante suspects, but ok), do it by extending the current system :)
20100413 16:42:08< timotei> I think I misunderstood it a little
20100413 16:42:25< timotei> quote: but at the it's not possible
20100413 16:43:04< timotei> well, the start campaign is already there, I need just to make it for a scenario, right?
20100413 16:43:13< timotei> that's what I saw already
20100413 16:43:48< Crab_> timotei: yes
20100413 16:44:33< timotei> right now, I'm doing the thing of starting a campaign, by name from the command line.
20100413 16:44:40< timotei> I'll do that scenario thingy after it, ok?
20100413 16:45:05-!- Blarumyrran [~Blarumyrr@81-20-159-197.levira.ee] has joined #wesnoth-dev
20100413 16:47:56< Crab_> ok
20100413 16:48:37< timotei> god, can't belive it :))
20100413 16:48:42< Crab_> it worked ;) ?
20100413 16:48:48< Crab_> or it crashed :) ?
20100413 16:48:50< timotei> I succesfully opened a spam email :((
20100413 16:48:53< Crab_> ah :)
20100413 16:48:58< timotei> I thought it's from gna.org :((((((
20100413 16:49:06< timotei> my entire life didn't have any spam, until now
20100413 16:49:34-!- meric [~Eric@124-168-169-14.dyn.iinet.net.au] has joined #wesnoth-dev
20100413 16:50:32< timotei> damn, I forgot to uncheck: show my email on gna.org
20100413 16:58:00< Crab_> timotei: going away, will be online in +2 h
20100413 16:58:07< timotei> ok
20100413 16:58:11-!- Crab_ [~Crab@wesnoth/developer/crab] has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
20100413 17:15:23-!- _jbx_ [~jbailey@12.190.80.225] has joined #wesnoth-dev
20100413 17:15:57-!- fendrin [~fabi@88-134-186-30-dynip.superkabel.de] has joined #wesnoth-dev
20100413 17:15:57-!- fendrin [~fabi@88-134-186-30-dynip.superkabel.de] has quit [Changing host]
20100413 17:15:57-!- fendrin [~fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20100413 17:23:59-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has quit [Ping timeout: 276 seconds]
20100413 17:24:02-!- iwaim_ [~iwaim@2001:2c0:40e::1] has joined #wesnoth-dev
20100413 17:40:14-!- meric [~Eric@124-168-169-14.dyn.iinet.net.au] has quit [Quit: meric]
20100413 17:49:55-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20100413 17:53:23-!- Ishan [~Ishan@123.201.150.71] has joined #wesnoth-dev
20100413 17:53:28-!- Ishan [~Ishan@123.201.150.71] has left #wesnoth-dev []
20100413 17:54:01-!- Lastmerlin [~Lastmerli@kalypso.csn.tu-chemnitz.de] has left #wesnoth-dev ["Kopete 0.12.7 : http://kopete.kde.org"]
20100413 17:54:59-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has joined #wesnoth-dev
20100413 18:12:13-!- Crab_ [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20100413 18:12:15-!- Sapient [~patrickp@wesnoth/developer/sapient] has joined #wesnoth-dev
20100413 18:16:35-!- norbert_ [~norbert@82-171-70-54.ip.telfort.nl] has joined #wesnoth-dev
20100413 18:51:10-!- shadowm_laptop [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100413 18:51:43< timotei> Crab_, LOG_GENERAL uses stdout.txt ?
20100413 18:51:49< timotei> the macro in game.cpp
20100413 18:52:09< shadowmaster> timotei: no, stderr.txt
20100413 18:52:31< timotei> so it's just like std::cerr ?
20100413 18:52:49< shadowmaster> all log messages using the lg::* log facility get sent through std::cerr since it doesn't have buffered output by default
20100413 18:53:03< timotei> okey
20100413 18:53:05< timotei> thanks
20100413 18:53:15< shadowmaster> which means that an unexpected signal will not mean that some errors don't appear in the console/stderr.xt
20100413 18:56:06-!- Ken_Oh [~briang@static-71-178-174-220.washdc.fios.verizon.net] has quit [Read error: Connection reset by peer]
20100413 19:06:37-!- Blueblaze [~nick@adsl-99-188-132-247.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100413 19:06:57-!- shadowm [~quassel@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100413 19:10:06< norbert_> website is live at wsrsw.org
20100413 19:10:22< shadowmaster> first complaint:
20100413 19:10:27< norbert_> but for some reason some pages are painfully slow to access
20100413 19:10:29< shadowmaster> the hostname doesn't have any vowels :P
20100413 19:10:58< norbert_> on my pc it's all super fast
20100413 19:11:03< norbert_> so something is clearly wrong right now
20100413 19:12:10< norbert_> may have something to do with sessions, but I have no clue
20100413 19:13:55< timotei> lets say I have a "config" that contains a campaign
20100413 19:14:01< timotei> how can I get the campaign's ID
20100413 19:14:08< timotei> cause I can't access config.values
20100413 19:14:50< timotei> should I use: attribute_range?
20100413 19:15:47< shadowmaster> um config["key"] isn't supported anymore?
20100413 19:16:00< shadowmaster> (where "key" would be "id" in this case assuming config represents a [campaign] node)
20100413 19:16:06< timotei> ok, let's see
20100413 19:16:31< shadowmaster> norbert_: will you announce this at the forums?
20100413 19:16:35< norbert_> no
20100413 19:16:40< norbert_> it works like crap
20100413 19:16:54< norbert_> if it works, I will :)
20100413 19:17:01< norbert_> but right now it only pisses me off
20100413 19:17:01< shadowmaster> ...what?
20100413 19:17:17< shadowmaster> could you be a bit more specific? If you have problems with the forums, I'm supposed to help you.
20100413 19:17:30< norbert_> no, not the forum, my website
20100413 19:17:43< shadowmaster> ahhh. well.
20100413 19:17:50< norbert_> works fine on my pc, but not on-line
20100413 19:17:56< norbert_> if I press the Login button it takes centuries
20100413 19:18:05< norbert_> something is wrong and I don't know what
20100413 19:19:07< timotei> the connection to the db?
20100413 19:19:08< norbert_> if I go from search.php to login.php and back and forth, and so on, it gets slower and slower
20100413 19:19:18< norbert_> yes, it has to do with the db
20100413 19:19:29< norbert_> maybe connections aren't closed properly or something
20100413 19:20:49-!- shadowm [~quassel@wesnoth/developer/shadowmaster] has quit [Quit: do not like]
20100413 19:26:11-!- orfest [~chatzilla@16.ngu.ac-tel.ru] has joined #wesnoth-dev
20100413 19:26:31< orfest> Hi all!
20100413 19:26:42< orfest> Crab_: I'm not ready with a design proposal today
20100413 19:26:46< timotei> hello orfest 
20100413 19:27:17< orfest> Crab_: I can't figure out why it's not possible to implement "Game master" client using current network layer implementation. Could you give a hint about it?
20100413 19:28:31< orfest> Crab_: I see there are several static variables like static vector<> pending_connections, but it doesn't look like a big problem
20100413 19:30:15< stikonas> Crab_, timotei: Hi, I was browsing and accidentaly stumbled upon http://www.fsf.org/blogs/licensing/using-the-gpl-for-eclipse-plug-ins Have you seen it?
20100413 19:30:54< timotei> stikonas, didn't see that. I will read it
20100413 19:31:26< stikonas> I've just noticed it myseld
20100413 19:31:51< Crab_> hi, orfest. yes, it's possible to do so 
20100413 19:32:07< Crab_> but, there's the timeout problem, as well
20100413 19:32:10-!- EdB [~EdB@tss37-1-89-84-18-220.dsl.club-internet.fr] has joined #wesnoth-dev
20100413 19:32:55< timotei> Crab_, well, it works starting a campaign from command line :D
20100413 19:33:00< timotei> I'm so happy. My first modification
20100413 19:33:18< timotei> should I make a case for selecting the difficulty also?
20100413 19:33:35< timotei> from command line?
20100413 19:33:39< Crab_> timotei: cool :) yes, it would be nice. it all boils down to setting a 'define', yes ?
20100413 19:33:48< Crab_> timotei: (wrt difficulty)
20100413 19:33:49< orfest> Crab_: Also I'd like to clarify, is it true that the network layer itself is incapsulated in just 2 files: network.cpp and network_worker.cpp. The rest of code doesn't deal with send/recv but are just actively using all kind of network API (including send_buffer, etc.)?
20100413 19:34:00< timotei> wrt?
20100413 19:34:09< Crab_> shikadibot: wrt ?
20100413 19:34:09< shikadibot> WRT: with respect to
20100413 19:34:28< timotei> oh
20100413 19:34:36< timotei> well, I need to think, what's the best way:D
20100413 19:34:42-!- Chusslove [~Chusslove@unaffiliated/chusslove] has joined #wesnoth-dev
20100413 19:34:59< Crab_> orfest: afair, yes. you can ask Soliton to be sure.
20100413 19:35:19< Crab_> orfest: the design proposal will be before friday ?
20100413 19:36:09< orfest> Crab_: could you tell more about timeouts? Who is going to define timeouts? Are not timeouts configured once for the whole game? Which level of timeouts flexibility is desired?
20100413 19:36:25< AI0867> norbert_: $ret = "SELECT s.scenario_id, s.op_name, m.name, if(s.op_name = '', m.name, s.op_name) as orderby FROM `scenarios` s INNER JOIN `mapp` m ON s.mapp_id = m.mapp_id ORDER BY orderby;"; <-- seriously?
20100413 19:37:12< AI0867> ah, wait, that's not javascript, it's php
20100413 19:37:35< orfest> Crab_: I'll definitely submit it not later than on Thursday
20100413 19:38:54< orfest> Crab_: could you please tell more about the desired format of the proposal? Should it be in the form of UML, text description? Should it include implementation details from very high point of view?
20100413 19:40:01< Crab_> orfest: whatever is easiest for you to write
20100413 19:40:38< norbert_> problem is the mysql server of my provider
20100413 19:40:55< norbert_> 100% memory usage and I just called them, no one can look at it before tomorrow
20100413 19:40:56< norbert_> bleh
20100413 19:41:16-!- shadowm_laptop is now known as ShikadiLord
20100413 19:41:25< Crab_> orfest: and it should include a 'timeline which will allow to track progress', 'list of pieces of code that you will need to touch', 'extra optional items that you might do if there's time', 'high-level design overview and estimation of cpu and memory usage compared to current'
20100413 19:41:29-!- Gambit is now known as Grickit
20100413 19:42:04< Crab_> orfest: basically, we want to see all those things that you studied while preparing to implement and design
20100413 19:45:49< orfest> Crab_: thanks. could you tell more about timeouts? :)
20100413 19:46:37< Crab_> basically, the same issue you've seen with network uploader - not all SDL.net methods allow to set timeouts
20100413 19:46:48< Crab_> s/network/replay
20100413 19:48:25< orfest> it's not clear for me. does it cause application to hang? Where timeouts are defined? Can they be modified while the application is running? Can timeouts differ inside the application?
20100413 19:49:17-!- ShikadiLord [~ignacio@wesnoth/developer/shadowmaster] has quit [Read error: error:1408F10B:SSL routines:SSL3_GET_RECORD:wrong version number]
20100413 19:50:06-!- ShikadiLord [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100413 19:50:58-!- wesbot changed the topic of #wesnoth-dev to: GSoC: Thanks to all GSoC candidates for submitting proposals to Google. Accepted student applications announced on April 26. | 90 bugs, 265 feature requests, 18 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100413 19:52:57-!- norbert_ [~norbert@82-171-70-54.ip.telfort.nl] has quit [Quit: Leaving]
20100413 19:59:37< Crab_> orfest: (sorry, not really near the keyboard atm)
20100413 19:59:50< Crab_> will be in +1 to 2 h
20100413 19:59:59-!- Blueblaze [~nick@adsl-99-188-132-247.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 276 seconds]
20100413 20:00:26-!- _jbx_ [~jbailey@12.190.80.225] has quit [Quit: _jbx_]
20100413 20:00:27< Crab_> afair, it can cause hang if a client gets 'unreachable' during the game, when we want to send something to him
20100413 20:01:53-!- Elsewise [~Dmitry@80.233.231.12] has joined #wesnoth-dev
20100413 20:05:05-!- YogiHH [YogiHH@d098013.adsl.hansenet.de] has joined #wesnoth-dev
20100413 20:05:21-!- YogiHH [YogiHH@d098013.adsl.hansenet.de] has quit [Changing host]
20100413 20:05:21-!- YogiHH [YogiHH@wesnoth/developer/yogihh] has joined #wesnoth-dev
20100413 20:05:22< YogiHH> hello
20100413 20:05:33< fendrin> Hi YogiHH
20100413 20:06:29< fendrin> YogiHH: Good start in the forum. Nice posts. I am just watching the non violent conversation video. I guess I have a great minus in that area.
20100413 20:07:01< YogiHH> fendrin: :)
20100413 20:07:41-!- Ken_Oh [~briang@static-71-178-174-220.washdc.fios.verizon.net] has joined #wesnoth-dev
20100413 20:11:22-!- _jbx_ [~jbailey@12.190.80.225] has joined #wesnoth-dev
20100413 20:12:23-!- Sapient1 [~patrickp@97.67.113.210] has joined #wesnoth-dev
20100413 20:13:37< shadowmaster> ShikadiLord: log HEAD
20100413 20:13:45< shadowmaster> shikadibot: log HEAD
20100413 20:13:48< shikadibot> Revision 42122 (mordante, 2010-04-12 19:26:09 +0000 (Mon, 12 Apr 2010)):
20100413 20:13:51< shikadibot> Update design documentation.
20100413 20:13:54< shikadibot> Web interface URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=42122
20100413 20:16:05-!- Sapient [~patrickp@wesnoth/developer/sapient] has quit [Ping timeout: 265 seconds]
20100413 20:18:14< timotei> hmm, I'm having an issue
20100413 20:18:26< timotei> trying to convert from char to int, and thought of using lexical_cast<>
20100413 20:18:43< timotei> is that way ok? or I should use "ascii" converting?
20100413 20:18:48< timotei> jump_to_campaign_diff = lexical_cast<int,char>(val[3]);
20100413 20:18:52< timotei> val[3] is a char
20100413 20:18:55< shadowmaster> mordante: btw, no, there's nothing wrong with HEAD, I am just too lazy to make shikadibot resolve it for me without the full log output
20100413 20:18:56< timotei> val is string
20100413 20:19:10-!- EdB [~EdB@tss37-1-89-84-18-220.dsl.club-internet.fr] has quit [Remote host closed the connection]
20100413 20:19:11< shadowmaster> an I'm an idiot who doesn't know about svnlook
20100413 20:20:14-!- Sapient [~patrickp@agentek.com] has joined #wesnoth-dev
20100413 20:21:11-!- Sapient1 [~patrickp@97.67.113.210] has quit [Ping timeout: 268 seconds]
20100413 20:23:58-!- Sapient [~patrickp@agentek.com] has quit [Read error: Connection reset by peer]
20100413 20:28:45< AI0867> timotei: you don't have to specify char, the compiler can infer that from the argument
20100413 20:29:06< timotei> so... just: jump_to_campaign_diff = val[3]?
20100413 20:29:23< timotei> but this way, will get the ascii code
20100413 20:29:32< timotei> currently I use: val[3]-'0'
20100413 20:29:48< timotei> and it works, but maybe *is required* to use lexical_cast / static_cast
20100413 20:30:34< AI0867> no
20100413 20:30:42< timotei> ok
20100413 20:30:47< AI0867> jump_to_campaign_diff = lexical_cast<int>(val[3]);
20100413 20:31:28< AI0867> or, you could use the c stdlib function itoa()
20100413 20:31:40< AI0867> er atoi() in this case
20100413 20:32:02< AI0867> wait, that operators on a string
20100413 20:32:16< AI0867> go with the lexical cast
20100413 20:34:03< timotei> well, why is better lexical cast than plain: val[3]-'0';
20100413 20:34:28< fendrin> timotei: Lexical cast is easier to read.
20100413 20:34:31< timotei> ok
20100413 20:34:31< timotei> ;)
20100413 20:35:18< shadowmaster> not only easier, but also more standard within wesnoth's code
20100413 20:35:33< timotei> understood
20100413 20:35:39< shadowmaster> if one can stick to conventions without hurting functionality, well.
20100413 20:40:01< Crab_> re
20100413 20:42:28-!- elias [~elias@allegro/developer/allefant] has joined #wesnoth-dev
20100413 20:42:49-!- pokhbocee [~pokhbocee@ws34.cs.drexel.edu] has joined #wesnoth-dev
20100413 20:43:05< Ivanovic> re, too
20100413 20:45:30< timotei> I have to admit, that, apart from new data structures (stl,boost), wesnoth's code base is really cool:D
20100413 20:45:55< timotei> well, something is done stupid here, lexical cast doesn't work as wanted:|
20100413 20:47:25< shadowmaster> timotei: well, if you feed it the char '2' you'll get 50, according to the character's actual value
20100413 20:47:30-!- ancestral [~ancestral@12.145.225.25] has joined #wesnoth-dev
20100413 20:47:31< shadowmaster> er, wait.
20100413 20:47:48< shadowmaster> no, you'll get 2. If you want 50 you'll need to convert it directly
20100413 20:47:59< timotei> yes
20100413 20:48:22< timotei> Crab_, it works jumping to the campaign+ optional difficulty:D
20100413 20:48:57-!- ShikadiLord [~ignacio@wesnoth/developer/shadowmaster] has quit [Quit: Lost terminal]
20100413 20:49:11< Crab_> timotei: good :)
20100413 20:49:40< timotei> testing more the command line,
20100413 20:50:04< Crab_> how it behaves if you specify nonexisting difficulty or scenario ?
20100413 20:50:09< Crab_> (or campaign)
20100413 20:50:38< timotei> well, for now, it starts only the campaign's first scenario
20100413 20:50:50< timotei> nonexisting difficutly is checked (if between 1-3)
20100413 20:51:01< timotei> as I saw there is easy medium hard, right?
20100413 20:52:01< Ivanovic> timotei: some campaigns have 4 levels
20100413 20:52:06< Ivanovic> and others only 2
20100413 20:52:08< Crab_> no, there can be more :)
20100413 20:52:13< Crab_> or less, just as Ivanovic says
20100413 20:52:13< Ivanovic> example for more: northern rebirth
20100413 20:52:15< timotei> oh, didn't know that
20100413 20:52:20< Ivanovic> example for less: two brothers
20100413 20:52:27< timotei> well, ok, I'll check that 
20100413 20:52:33< timotei> thanks for telling me:D
20100413 20:52:57< Ivanovic> and the names can differ from those "common" EASY, NORMAL and HARD
20100413 20:53:29< timotei> well, for now I require only a number from 1 to maximum campaign's difficulty
20100413 20:53:32< timotei> isn't ok like that?
20100413 20:53:42< timotei> it's faster than typing the whole words
20100413 20:53:58< Ivanovic> no idea how this selection stuff does work internally
20100413 20:54:08< Ivanovic> though numbers *should* be a nice representation
20100413 20:54:17< timotei> no, I'm telling you: number vs word representation
20100413 20:54:18< timotei> yes
20100413 20:55:37-!- ilor [~ilor@wesnoth/developer/ilor] has joined #wesnoth-dev
20100413 20:56:52< AI0867> timotei: they're defines
20100413 20:57:00< AI0867> so you can do #ifdef HARD... in WML
20100413 20:57:09< timotei> yes
20100413 20:57:15< AI0867> #ifdef 1 would be rather silly
20100413 20:57:33< timotei> I'm not "reworking" the current difficulty system :P
20100413 20:57:38< timotei> just the command line arguments:)
20100413 20:57:50< AI0867> ah
20100413 20:57:55< AI0867> so, indexing the list?
20100413 20:58:05< timotei> Crab_, Ivanovic , if a user tries to open an invalid campaign, from cmd-line, should I write an error to log, or present the campaign list?
20100413 20:58:10< AI0867> still, it's probably more clear just to use the define
20100413 20:58:28< Ivanovic> show an errormessage in stdout
20100413 20:58:40< Crab_> timotei: show a easy to parse/understand error message
20100413 20:58:53< timotei> well, I've already got a "difficulty" list. I'm just getting the respective difficulty from that list
20100413 20:59:07< Ivanovic> (as in "clear abort with exist state 1" (or whatever we use for "error occured")
20100413 20:59:41-!- Ken_Oh [~briang@static-71-178-174-220.washdc.fios.verizon.net] has quit [Read error: Connection reset by peer]
20100413 20:59:48< Ivanovic> i think for the mp server or something like this we had some specific error states defined, don't think we have this for the game itself
20100413 21:03:14-!- Darkas [~quassel@ppp-93-104-141-195.dynamic.mnet-online.de] has joined #wesnoth-dev
20100413 21:17:39-!- Grickit [~Gambit@pa-67-234-73-7.dhcp.embarqhsd.net] has quit [Quit: Leaving.]
20100413 21:25:41-!- ancestral [~ancestral@12.145.225.25] has quit [Quit: ancestral]
20100413 21:28:23< timotei> Crab_, well, ok, fixed the nonexisting campaign/difficulty
20100413 21:28:28< Crab_> good
20100413 21:28:29< timotei> now for the scenario part.
20100413 21:28:37< timotei> or should I send you the diff?
20100413 21:28:42< timotei> to look over it? (pastebin)
20100413 21:29:10< timotei> I don't know if I will finish the scenario start tonight, have some projects due to tomorrow.
20100413 21:29:12< Crab_> patches.wesnoth.org
20100413 21:29:36< timotei> well, I'll prefer to submit the patch with the scenario too :P
20100413 21:29:39< timotei> isn't it better?
20100413 21:30:41< timotei> starting the scenario... should we specifiy a scenario from current campaign only?
20100413 21:30:50< timotei> or a separat "standalone" scenario - if that's possible
20100413 21:30:50< timotei> ?
20100413 21:31:01< Crab_> no, scenario from within campaign
20100413 21:31:17< timotei> oky
20100413 21:31:22< Crab_> the former is possible but more work (e.g., test scenarios are handled somewhat differently)
20100413 21:31:29< timotei> yes
20100413 21:31:48-!- pokhbocee [~pokhbocee@ws34.cs.drexel.edu] has quit [Remote host closed the connection]
20100413 21:32:04< timotei> in the case of test scenarios... we should use the [test] instead [scenario] ?
20100413 21:32:18< timotei> the sample test scenario uses that
20100413 21:32:29-!- gabba [~gabba@212.75-ppp.3menatwork.com] has joined #wesnoth-dev
20100413 21:32:41< gabba> hi there
20100413 21:32:48< timotei> but Ok, i'll do it from current campaign for now
20100413 21:32:52-!- schumi [~a11849c4@gateway/web/freenode/x-ofqvrrraeqaaduhz] has joined #wesnoth-dev
20100413 21:32:57< timotei> hi gabba :D
20100413 21:33:02< timotei> whassup?
20100413 21:33:42< gabba> hello timotei
20100413 21:33:56< gabba> final exams is what's up mainly lol
20100413 21:34:12< timotei> oh, but you'll soon finish them :P
20100413 21:34:25< gabba> sure
20100413 21:34:42< timotei> my exams are right in the *middle* of gsoc :))
20100413 21:35:17< gabba> exams in mid-summer are strange to me: european tradition I guess?
20100413 21:35:38< timotei> well, it's not so "mid-summer"
20100413 21:35:46< timotei> it's just the start of summer's holiday:P
20100413 21:36:01< gabba> but the timing is better for your GSoC preparation
20100413 21:36:01< timotei> I mean, it depends how you take it.
20100413 21:36:05< gabba> ah ok
20100413 21:36:11< timotei> yes, could be :P
20100413 21:36:39< timotei> btw, congrats for your proposals ;)
20100413 21:36:55< timotei> I mean, I'm a bit jealous cause you're so good :D
20100413 21:37:14< gabba> well thanks
20100413 21:37:31< Soliton> orfest: yes, direct network code should only be in src/network*.
20100413 21:37:50< gabba> last year I couldn't get in because I competed against a brilliant PhD student
20100413 21:37:53< timotei> well, Crab_ I will go now
20100413 21:37:59< timotei> I'll catch you tomorrow
20100413 21:38:02< gabba> so this time I sweared I'd get it right ;)
20100413 21:38:11< timotei> gabba, well, good luck
20100413 21:38:16< gabba> you too
20100413 21:38:21< timotei> thnx
20100413 21:38:28< timotei> good night everyone
20100413 21:38:32< gabba> bye
20100413 21:38:36-!- timotei [~timotei@193.34.191.4] has quit [Quit: Leaving]
20100413 21:38:36< Crab_> bye
20100413 21:42:03< Soliton> zookeeper: http://gna.org/bugs/?15863  something for you?
20100413 21:54:52-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100413 21:55:37-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100413 21:56:11-!- Gambit [~Gambit@pa-67-234-73-7.dhcp.embarqhsd.net] has joined #wesnoth-dev
20100413 22:00:52-!- dtiger [~dtiger@dynamic-vpdn-93-125-13-17.telecom.by] has quit [Remote host closed the connection]
20100413 22:02:40< Soliton> fendrin: http://forum.wesnoth.org/viewtopic.php?p=421291#p421291 can you tell me what you're talking about there? i see no experimental compile flag.
20100413 22:03:06< fendrin> Soliton: Yes, it was removed again.
20100413 22:03:41< Soliton> so can you answer the question then?
20100413 22:04:03-!- icelus [~ed@cpc9-sgyl11-0-0-cust29.sgyl.cable.virginmedia.com] has quit [Quit: Leaving]
20100413 22:04:38< fendrin> Soliton: I branched Wesnoth before the experimental flag was removed. So I have all my features now in a single branch. That is the problem.
20100413 22:05:24< Soliton> you can't make a patch for that feature?
20100413 22:06:13< fendrin> Soliton: I can. And I will do so. If the feature is wanted. 
20100413 22:07:24< fendrin> And I need a developer that takes time and reviews it. My c++ is still not perfect.
20100413 22:09:41< Soliton> the first step to achieve that is posting a patch.
20100413 22:10:42< Soliton> why do you think the feature might not be wanted? doesn't it fix any of the open FRs regarding keyboard use?
20100413 22:11:49< fendrin> Soliton: It's the fact that I don't have much luck with getting my c++ code into mainline. 
20100413 22:12:27< fendrin> It had already been in for a short amount of time, but together with my other features. 
20100413 22:12:49< Soliton> sounds like seperating would be the way to go then.
20100413 22:13:35< fendrin> Soliton: Right, I will take the time and submit it at patches.wesnoth.org during the next week. 
20100413 22:14:32< Soliton> you said you have all your features in a single branch but i see 3 fendrin_* branches. can 2 of those be removed?
20100413 22:15:12< fendrin> Soliton: Yes, but please let me make sure that I have not forgot something in there.
20100413 22:16:00< Soliton> sure.
20100413 22:16:09< fendrin> :-)
20100413 22:16:18< zookeeper> Soliton, it's not really a bug, i'll mark as invalid but assign to jetrel anyway in case he wants to change it...
20100413 22:16:30< Soliton> ok.
20100413 22:16:49-!- Elsewise [~Dmitry@80.233.231.12] has quit [Ping timeout: 264 seconds]
20100413 22:21:18-!- schumi [~a11849c4@gateway/web/freenode/x-ofqvrrraeqaaduhz] has quit [Quit: Page closed]
20100413 22:24:29-!- Darkas [~quassel@ppp-93-104-141-195.dynamic.mnet-online.de] has quit [Ping timeout: 276 seconds]
20100413 22:35:20-!- Gambit [~Gambit@pa-67-234-73-7.dhcp.embarqhsd.net] has quit [Ping timeout: 276 seconds]
20100413 22:40:41-!- Gambit [~Gambit@pa-67-234-73-7.dhcp.embarqhsd.net] has joined #wesnoth-dev
20100413 22:53:09-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20100413 22:53:17-!- stikonas [~and@2002:4f84:b38b:b:226:9eff:fe0a:3e55] has joined #wesnoth-dev
20100413 22:53:17-!- stikonas [~and@2002:4f84:b38b:b:226:9eff:fe0a:3e55] has quit [Changing host]
20100413 22:53:18-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100413 22:53:19-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Client Quit]
20100413 22:53:28-!- stikonas [~and@2002:4f84:b38b:b:226:9eff:fe0a:3e55] has joined #wesnoth-dev
20100413 22:53:28-!- stikonas [~and@2002:4f84:b38b:b:226:9eff:fe0a:3e55] has quit [Changing host]
20100413 22:53:28-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100413 22:59:01< Crab_> yes! I've got a backtrace for the 'missing wml child in 1.8'
20100413 22:59:22< Crab_> seems related to the host sending/receiving a chat message during the attack
20100413 22:59:37< Ivanovic> interesting
20100413 22:59:52< Crab_> http://wesnoth.pastebin.com/eRTpdkRT
20100413 22:59:56< Ivanovic> so intermediate solution: tell all players not to chat!
20100413 23:00:07< Crab_> (ai code because it's from ai-vs-ai matchup on wesnoth mp server)
20100413 23:09:32< Crab_> yes, I see, it looks like there's a race condition in the code ilor used to add SRNG to the ai's attacks
20100413 23:10:01< Crab_> ilor: around ?
20100413 23:14:39-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit []
20100413 23:17:22< ilor> Crab_: ?
20100413 23:17:27< Crab_> ilor: there's a race condition in there - http://wesnoth.pastebin.com/PmTAdF1h
20100413 23:18:01< Crab_> ilor: the code in recorder.add_seed incorrectly assumes that the current_ replay element is related to the attack
20100413 23:18:39< Crab_> ilor: but, it the host send a chat message during that 'wait for srng', then this won't be the case. so, it'll get a 'missing wml  child' screen.
20100413 23:19:35< ilor> ahh, the random seed will get added to the chat message
20100413 23:19:44< Crab_> ilor: it will try and fail.
20100413 23:20:10< ilor> even if it didn't things would break down sooner or later
20100413 23:20:21< Crab_> no, since it's at the host.
20100413 23:20:47< ilor> well, the host would end up with a corrupt replay
20100413 23:20:59< Crab_> yes
20100413 23:21:19< Crab_> so, the question is, is the 'ui part' affected, too ? you use a different solution there, with a callback from rng
20100413 23:21:47< ilor> hmm not sure
20100413 23:21:52< Crab_> e.g. 'ai part' has a larger window of opportunity due to  SDL_Delay(10);, but maybe the 'user attacks' part are affected as well ?
20100413 23:22:07< Crab_> from the reports we see that only host is affected, so I think there are no problems there
20100413 23:22:26< Crab_> but, can you check if there's a possibility of a similar f**up in there?
20100413 23:22:49< ilor> looking
20100413 23:22:53< Crab_> thanks
20100413 23:23:31< ilor> generally I guess yes
20100413 23:23:54< Crab_> ok, thanks
20100413 23:24:07< ilor> the code goes back from the attack_enemy_ to the main loop where a message might come in
20100413 23:24:22< ilor> and then the callback is called and sets the seed on the wrong item
20100413 23:24:37-!- Johannes13 [~Johannes@unaffiliated/johannes13] has quit [Ping timeout: 264 seconds]
20100413 23:24:49< ilor> a soulution could be to remember where the seed should go instead of using the last item
20100413 23:25:19< Crab_> yes, I was thinking about that. or passing that 'where' to the replay code when adding seed.
20100413 23:25:45< ilor> at any rate, good catch
20100413 23:26:21< ilor> now if you could find the root cause of the lobby bugs that would be cool ;p
20100413 23:26:24-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Read error: Operation timed out]
20100413 23:27:44< Crab_> well, I think/hope that you and mordante have fixed most of them already
20100413 23:31:01-!- gabba [~gabba@212.75-ppp.3menatwork.com] has left #wesnoth-dev []
20100413 23:41:01< ilor> wesbot: seen mordante
20100413 23:41:02< wesbot> ilor: The person with the nick mordante last spoke 1d 1h ago.  1d 1h ago was here and on the channel #wesnoth-de with the message: Quit: Leaving
20100413 23:42:05-!- allefant [~elias@allegro/developer/allefant] has joined #wesnoth-dev
20100413 23:43:05< ilor> mordante: I could work around one major lobby issue -- the gamelist jumping around -- if:
20100413 23:43:15-!- Espreon [~espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20100413 23:43:28-!- _jbx_ [~jbailey@12.190.80.225] has quit [Quit: It was a new day yesterday, but it's an old day now.]
20100413 23:43:33< ilor> 1.  there were a way for my code to get notified on a mouse down / up on the scrollbar so I don't update the gamelist while the user is drag-scrolling
20100413 23:43:54< ilor> 2. there were a way to access the top/bottom visible element in the gamelist]
20100413 23:44:47-!- elias [~elias@allegro/developer/allefant] has quit [Ping timeout: 258 seconds]
20100413 23:53:52< YogiHH> night everyone
20100413 23:53:58< Crab_> night, YogiHH
20100413 23:54:28-!- YogiHH [YogiHH@wesnoth/developer/yogihh] has left #wesnoth-dev []
--- Log closed Wed Apr 14 00:00:05 2010