--- Log opened Wed Sep 09 00:00:51 2009
20090909 00:05:28-!- DDR [n=chatzill@66.183.125.196] has quit ["ChatZilla 0.9.85 [Firefox 3.5.2/20090729225027]"]
20090909 00:06:04< CIA-62> soliton * r38524 /branches/1.6/src/server/server.cpp: consider the case that a game was started with no players
20090909 00:06:31< CIA-62> soliton * r38525 /trunk/src/server/server.cpp: consider the case that a game was started with no players
20090909 00:10:24-!- noy [n=Noy@wesnoth/developer/noy] has quit [Connection timed out]
20090909 00:25:22-!- YogiHH [n=chatzill@d128107.adsl.hansenet.de] has joined #wesnoth-dev
20090909 00:26:03-!- Appleman1234 [n=Appleman@131.181.101.135] has joined #wesnoth-dev
20090909 00:28:24< YogiHH> Soliton: When reloading an mp save on the server, i get "Invalid WML received". Is that because my checkout is like 3-5 days old?
20090909 00:29:00< YogiHH> Soliton: something like "end of attribute wasn't found"
20090909 00:30:34< Soliton> YogiHH: and the game loads fine otherwise?
20090909 00:30:51< YogiHH> Soliton: yes, it's a simple WML error
20090909 00:31:52< Soliton> well, nothing changed there afaik.
20090909 00:34:38-!- alink [n=alink@wesnoth/developer/alink] has joined #wesnoth-dev
20090909 00:35:03< Soliton> looks like an attribute is missing the closing " quotes.
20090909 00:40:11-!- Espreon [n=espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20090909 00:40:13-!- Sirp [n=user@wesnoth/developer/dave] has joined #wesnoth-dev
20090909 00:45:34< corn> Crab_: now I am
20090909 00:45:46< Crab_> hello
20090909 00:45:53< corn> hi
20090909 00:47:41-!- ardesh [n=ardesh@port-92-195-117-249.dynamic.qsc.de] has joined #wesnoth-dev
20090909 00:50:59< Crab_> corn: for campaign balancing, it's important to know the distribution of 'starting gold','number of L3 units in recall list','number of L2 in recall list' on a specific scenario on a specific difficulty. E.g. "80% players playing LoW scenario 3 on normal have 200 to 300 gold and 2 to 3 L3 units in recall list". what should be needed to collect that info ?
20090909 00:51:39< Crab_> corn: and the effect of that parameters on victory rate.
20090909 00:52:09< Crab_> corn: the first step, is, basically, getting the required data into the upload log. (easy enough, I can do that)
20090909 00:52:26< Crab_> corn: but then, it starts to become somewhat tricky
20090909 00:52:48-!- YogiHH [n=chatzill@d128107.adsl.hansenet.de] has left #wesnoth-dev []
20090909 00:52:56< corn> Crab_: steps to add new data: 1. add to upload log 2. add parsing on server, add new columns or even tables, 3. add new way to view data
20090909 00:53:28< corn> step 1 is easy and the old upload logs actually stored this data, but I removed it
20090909 00:53:31< Crab_> corn: note that the data needs statistical postprocessing to be useful
20090909 00:54:08< corn> what kind? removing outliers?
20090909 00:54:24-!- loonycyborg [n=sergey@wesnoth/developer/loonycyborg] has quit ["Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz"]
20090909 00:54:48< Crab_> corn: if we want to do queries like this "80% players playing LoW scenario 3 on normal have X to Y gold and M to N L3 units in recall list, please show X,Y,M,N"
20090909 00:54:59-!- loonybot [n=loonybot@wesnoth/bot/loonybot] has quit [Remote closed the connection]
20090909 00:55:20< corn> ah I see, something like percentiles
20090909 00:55:36< Crab_> corn: or, alternatively, "80% of players having this starting resources win this scenario. show me the starting resources"
20090909 00:56:34< Crab_> corn: note that even batch processing of those quieries is 'good enough'
20090909 00:56:34< Crab_> corn: so, speed of response is not the issue
20090909 00:56:41< corn> Crab_: can we make any assumptions about the distribution of results?
20090909 00:56:54< corn> ex. normal distribution, completely random distribution
20090909 00:57:19< Crab_> corn: I don't know. but, isn't gold amount logged now ?
20090909 00:57:26< corn> yes, starting gold
20090909 00:57:27< Crab_> corn: can you see how it looks like?
20090909 00:57:46< corn> sure
20090909 00:57:54-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090909 00:58:27-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 00:58:45-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090909 00:58:46< Crab_> corn: I expect it to be bimodal distribution (e.g. grouping about 'players who have lots of gold' and 'players who have little gold')
20090909 00:59:07< Crab_> corn: since if you have little gold, you'll finish scenario in more time, and so, get less gold as bonus
20090909 00:59:30< Crab_> corn: and if you have lots of gold, you'll finish quickly, and get even more gold as bonus
20090909 01:01:18-!- stikonas [n=and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20090909 01:02:25< corn> ok, I am figuring out a way to generate a histogram-like display from the current data
20090909 01:02:28< Crab_> corn: having such results will allow easier balancing. e.g. 1) "we want this scenario to be winnable 80% of the time" 2) we see the resources needed by human to win it 80% of the time 3) we see the resources which humans have when they come to this scenario" 4)  we immediately can calculate the adjustment to the starting gold/resources needed to align (3) with (1)
20090909 01:02:42< corn> there is line graph support but the x scaling is not good right now
20090909 01:03:11-!- ardesh__ [n=ardesh@port-92-195-9-137.dynamic.qsc.de] has quit [Read error: 110 (Connection timed out)]
20090909 01:03:34< Crab_> corn: no fancy displays are needed, tables are good enough if the filter allows to parametrize the query properly 
20090909 01:04:07< corn> at this point I am trying to figure out the correct SQL query for it
20090909 01:04:25< corn> what I am trying to do is SELECT gold FROM GAMES_SP and group by intervals of 10
20090909 01:04:28< corn> or intervals of 100
20090909 01:04:51< corn> counting the occurence of individual values is not as useful as counting a range of values
20090909 01:04:59< corn> ex. 111 is not very different from 110
20090909 01:07:25< Crab_> corn: it depends. 111 can be vastly different from 110 :)
20090909 01:08:38< Crab_> corn: note that the number of leveled units can be small.. 1...2...3, etc
20090909 01:08:46< corn> talking about gold right now
20090909 01:08:51< Crab_> ok
20090909 01:08:51< corn> the other data is not being collected now 
20090909 01:09:23< corn> ok, I made a histogram-like table
20090909 01:09:29< corn> here is the query, I'll put the results on pastebin
20090909 01:09:32< corn> SELECT (gold-(gold%10)),COUNT(gold - (gold%10)) FROM GAMES_SP GROUP BY (gold - (gold%10));
20090909 01:09:33< Crab_> corn: the same with gold. 110 gold can be vastly different from 111, if having 111 gold would allow to recruit 1 unit more...
20090909 01:10:04< Crab_> imagine a L2 unit which costs 37 gold. 110 = 2 units, 111 = 3 units.
20090909 01:10:17< corn> http://pastebin.com/m1eae81ac
20090909 01:10:23< Crab_> corn: ok, thanks
20090909 01:10:55< Crab_> corn: is this for 1 scenario ?
20090909 01:11:03< corn> this is for all scenarios
20090909 01:11:07< corn> do you want it for a particular scenario?
20090909 01:11:14< Crab_> corn: yes. pick any good.
20090909 01:11:27< corn> ok, will pick HttT 2 then
20090909 01:11:32< Crab_> corn: different scenarios have different amount of starting gold
20090909 01:11:37< Crab_> corn: so, 1 scenario 1 difficulty
20090909 01:11:53< Crab_> corn: otherwise we'll get a normal distribution
20090909 01:12:41< Crab_> corn: yes, HttT is good enough
20090909 01:13:15< Soliton> so number of leveled units in the recall list is not logged anymore?
20090909 01:13:27< corn> no
20090909 01:13:34< corn> it's a good idea to readd, but the way it was done before is not good
20090909 01:14:04< corn> right now the current scheme is to add or read data from the replay rather than try to collect it during upload log destruction
20090909 01:15:09< corn> Crab_: http://pastebin.com/m4d193e09
20090909 01:15:35< corn> Crab_: you were right about the two groups of no gold and lots of gold
20090909 01:16:50< Crab_> corn: ok, so what will be the best way to do such statistics ?
20090909 01:18:08< corn> Crab_: do query like this: SELECT gold,COUNT(gold) from GAMES_SP GROUP BY gold, then count from lowest gold while summing the second column COUNT(gold) until the sum reaches 80% of the the sum of the second column
20090909 01:18:29< corn> actually, count from highest gold
20090909 01:18:42< Crab_> corn: is it better to do from a standalone script which will have read-only access to s.w.o DB ?
20090909 01:19:19< corn> Crab_: I would prefer if it was completely integrated into s.w.o but it will be faster to make a standalone script
20090909 01:19:59< Crab_> corn: the problem is that we need to do some fairly complex statistical analysis to get really interesting results
20090909 01:20:45< Crab_> corn: i.e. "what adjustments we should make to the starting gold of this scenario, to set victory rate to 80% for players with gold, and 50% for players without gold?' 
20090909 01:21:26< Crab_> corn: ok, what would be the way of getting ro access to s.w.o DB ?
20090909 01:21:39< corn> it depends on whether it is a one-shot thing to only be used with a particular piece of data and maybe a particular campaign/scenario, or if it can be generalized to multiple columns and scenarios
20090909 01:21:43< corn> Crab_: ask rhonda
20090909 01:21:44< Crab_> corn: we mainly want this for mainline campaigns
20090909 01:22:07< Crab_> corn: another option:
20090909 01:22:22< corn> Crab_: ask rhonda or soliton, soliton can give you ssh access so that you can use mysql locally but rhonda administers mysql stuff
20090909 01:23:29< Crab_> corn: if we split the processing into two parts, first part will be given by s.w.o , and second will be given by offsite script which will query s.w.o
20090909 01:24:13< Crab_> corn: i.e., if we want to show basic gold-and-leveled-units distribution via s.w.o (players might want to see this, too, to compare themselves with other players)
20090909 01:24:38< corn> Crab_: that is silly - why bother making an API if you can just make an 'onsite' script by querying through mysql directly
20090909 01:25:31< Crab_> corn: to do further processing outside wesnoth.org server
20090909 01:27:07< corn> Crab_: it should be ok to do this on server as long as it runs in a nightly cronjob and not realtime
20090909 01:27:20< Crab_> corn: ok, nightly cronjob is good enough, too.
20090909 01:27:25< corn> I am just considering this from the perspective of simplifying the programming
20090909 01:27:58< Crab_> corn: what programming languages are available on w.o ? php/perl/python/c/c++ ?
20090909 01:28:12< corn> yes, all of them afaik
20090909 01:28:35< Crab_> ok, understood
20090909 01:29:25< corn> ok, so lets figure out all the new data that we want collected on the client now
20090909 01:29:43< corn> and after, the proposed table structure for the new data
20090909 01:30:40-!- BenUrban [n=benurban@unaffiliated/benurban] has joined #wesnoth-dev
20090909 01:30:55< corn> is the content of the recall list + the special units enough?
20090909 01:31:23< Crab_> corn: yes. plus a few 'int evaluations' of the recall list
20090909 01:32:24< Crab_> corn: or, do you prefer to log the individual units that are on recall list ?
20090909 01:32:54< corn> Crab_: yes, the latter is what I had in mind
20090909 01:33:44-!- PK7 [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has quit ["Java user signed off"]
20090909 01:34:19< Crab_> corn: then, the script will need to make evaluations on recall list... it'll need unit type, unit type cost, unit level, unit xp, unit type max xp
20090909 01:34:20< corn> although, now that you propose it, just sending a summary of the recall list is enough
20090909 01:34:45< corn> do you want the extra data?
20090909 01:35:29< Crab_> corn: well, it'll be useful
20090909 01:35:35< corn> ok
20090909 01:35:47< Crab_> corn: since that way we'll be able to modify the evaluators without rereleasing wesnoth
20090909 01:36:31< corn> hopefully we won't make upload logs horrendously big though :)
20090909 01:36:33< Crab_> corn: i.e. if suddenly someone asks 'hey, what are the effects of the number of white mages on %-chance of beating this ?'
20090909 01:36:58< corn> sending the _entire_ gamestate would future proof us but it isn't feasible
20090909 01:38:16< corn> anyway, I'll recode in the sending of unit lists now
20090909 01:39:12< Crab_> corn: recall lists are not that big. and we don't need all info to be sent. just 'unit type, level, id, xp, xp_needed_to_level' (if we cannot get level, cost and xp_needed_to_level from somewhere else, like units DB)
20090909 01:40:01< Crab_> corn: so, the overall plan is 1) make upload_log log needed info  2) store it in s.w.o DB 3) ask rhonda for account needed to run a script with access to s.w.o DB 4) I'll try to make up a script to analyze the the s.w.o info and extract something useful from it (in a form of, say, static web pages )
20090909 01:40:32< Crab_> (if you want to try (4), that's good, too. depends on what you want)
20090909 01:43:07< corn> Crab_: do you have a tentative programming language choice yet for generating these pages?
20090909 01:43:28< Crab_> corn: no. the main part is math anyway :)
20090909 01:43:53< corn> ok
20090909 01:44:24< corn> I am interested in seeing more complex analysis on s.w.o so I'll work on (4) in conjunction with you
20090909 01:45:44< Crab_> corn: my present goal is to make the script answer questions like 'hey, I want this scenario to be winnable X % of the time for good players, what adjustments wrt starting gold and/or recall list size I must do?'
20090909 01:46:45< Soliton> that's pretty much what all campaign maintainers will want to know.
20090909 01:46:56< Crab_> it would be extremely interesting to ask around for suitable X, and then run the script on all mainline campaigns, and see the results
20090909 01:47:17< Crab_> note that even with 1.7.5 we can try to do that, but by using gold only.
20090909 01:48:11< corn> those are interesting questions and making results for this is not going to be very easy
20090909 01:48:47< corn> we need to construct models, probably using differential equations, around the data
20090909 01:48:51< corn> do you have experience with it?
20090909 01:49:03< Crab_> corn: yes
20090909 01:49:05-!- Chusslove [n=Chusslov@brsg-d9beffee.pool.mediaWays.net] has quit [Read error: 110 (Connection timed out)]
20090909 01:49:11< corn> ok, cool :)
20090909 01:49:46< Crab_> I've recently graduated from 6 years of applied math, after all ) So, I've got to apply it somewhere ))
20090909 01:55:28< corn> where is the recall list stored within the replay?
20090909 01:55:59-!- Chusslove [n=Chusslov@brsg-d9beec11.pool.mediaWays.net] has joined #wesnoth-dev
20090909 01:58:55< Crab_> corn: somewhat tricky. we need to log all player's starting units and all his recall list at game start
20090909 01:59:01< Crab_> corn: there's some autorecall stuff :)
20090909 02:00:13< Soliton> isn't the log generated on scenario end?
20090909 02:00:42< Soliton> all units that are still there then go on the recall list.
20090909 02:00:48< corn> Soliton: not anymore - log data can be added throughout the game via the replay and then read back on scenario end
20090909 02:00:53< Crab_> Soliton: but we can add various stuff to replay
20090909 02:02:01-!- BenUrban [n=benurban@unaffiliated/benurban] has quit [Read error: 104 (Connection reset by peer)]
20090909 02:02:26< Soliton> well, why jump through hoops on scenario start if it's simple at scenario end?
20090909 02:02:45-!- BenUrban [n=benurban@unaffiliated/benurban] has joined #wesnoth-dev
20090909 02:03:00< Crab_> Soliton: because we're interested in the cause-and-effect relationship of 'game started with X resources' and 'victory?'
20090909 02:03:22< Crab_> Soliton: so, we need to log starting resources
20090909 02:03:25< corn> Crab_, Soliton: your suggestions are not mutually exclusive
20090909 02:04:33< corn> Soliton: it's better for there to be a single entry point into the logging (reading of replay data at scenario start OR end OR anywhere inbetween) rather than generating log at end game + replay reading
20090909 02:05:10< Crab_> corn: the time window (for a specific team) is 'after game_state::build_team finishes', but 'before any prestart/start events are called'
20090909 02:06:40< corn> to log what? recall list info?
20090909 02:06:59< Crab_> corn: yes. note that some units are immediately placed from recall list to the battlefield
20090909 02:07:05< Crab_> corn: e.g. leader
20090909 02:07:54< Crab_> corn: we can, of course, argue that they're always present, so irrelevant. but their level can be relevant, so it still has to be logged.
20090909 02:09:32< corn> ok. playsingle_controller.cpp:327 is a good point to do the logging?
20090909 02:10:58< Crab_> corn: yes. there's some log.start's there, already
20090909 02:12:14< Crab_> corn: it's after play_controller::init which calls game_state::build_team, and it's before fire_prestart()
20090909 02:12:18< corn> ok
20090909 02:12:36< Crab_> corn: note the comment about ' Log before prestart events: they do weird things.' there :)
20090909 02:13:26< Crab_> corn: for *all* human teams, we must 'units on map' and 'units in recall list'
20090909 02:14:19< Crab_> corn: note that even in SP, one human can control different teams. ( for example, NR does that already,and LoW will do that in 1.9)
20090909 02:14:30-!- noy [n=Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090909 02:15:46< Crab_> corn: (note that this means that for each unit we must log 'side', as well)
20090909 02:16:09< Crab_> corn: also note that, it's better to log not just human teams, but all teams with non-empty save_id
20090909 02:16:23< Crab_> corn: in LoW, there are allied teams which persist between scenarios
20090909 02:18:08-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 110 (Connection timed out)]
20090909 02:18:39< Crab_> corn: e.g. "if (team.is_human() || !team.save_id().empty() ) {..}
20090909 02:18:47< corn> Crab_: ok, I'll write code to put all this info into a new section of [upload_log] within the replay using the add_log_data() functions. will you be on in ~8 hours? I am going to bed soon so I'll code this tomorrow
20090909 02:19:15< Crab_> corn: no, I'll be in ~16 hours
20090909 02:20:06< Crab_> corn: and, there's no reason to hurry - 1.7.6 is not near (provided that no critical bugs lurk in 1.7.5 :) )
20090909 02:20:35< corn> ok
20090909 02:22:26-!- BenUrban [n=benurban@unaffiliated/benurban] has quit [Read error: 104 (Connection reset by peer)]
20090909 02:26:08-!- happygrue [n=George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20090909 02:29:26< Crab_> corn: the current (basic) idea for calculation of needed gold adjustments, is: given {scenario,difficulty,X}: 
20090909 02:29:26< Crab_> 1) define functions CHANCE_OF_HAVING_THAT_STARTING_GOLD(x) and CHANCE_OF_VICTORY_HAVING_THAT_STARTING_GOLD(x), precalc them using DB data.
20090909 02:29:26< Crab_> 2) find y, such that sum(x,CHANCE_OF_VICTORY_HAVING_THAT_STARTING_GOLD(x+y,CHANCE_OF_HAVING_THAT_STARTING_GOLD(x))is close enough to X
20090909 02:29:26< Crab_> 3) then, y is needed gold adjustment.
20090909 02:31:30< Crab_> corn: of course, this is only a first step (gold in scenario 2 is not independent from gold on scenario 1, etc). so, a more refined result can be obtained by writing a model of gold carryover system (should be nothing fancy there, just a system of linear equations), and solving it for (y1,y2,y...,yN) for all the scenarios of the campaign, given (X1,X2,..,XN) victory chances
20090909 02:32:05< CIA-62> alink * r38526 /trunk/src/ (4 files): Add a helper function to get red-to-green color scale.
20090909 02:32:38-!- BenUrban [n=benurban@unaffiliated/benurban] has joined #wesnoth-dev
20090909 02:34:43< corn> Crab_: can you explain what step (2) is? is 'x' different from 'X'
20090909 02:34:51< Crab_> corn: yes
20090909 02:35:16< Crab_> corn: X is 0.6 if we want this scenario to result in victory 60% of the time
20090909 02:37:31< Crab_> corn: and,  sum(A,B) is sum of B over all possible values of A
20090909 02:38:44< corn> ok, I think I get the idea
20090909 02:39:03< Crab_> corn: with the added heuristic that 'more gold -> bigger chance of  victory', search for y can be made somewhat faster.
20090909 02:40:18< corn> the end goal is to balance the campaign so that on average, a player has a 60% chance of victory?
20090909 02:40:36< Crab_> corn: (60% is the example)
20090909 02:40:43< corn> I think my misunderstanding was that the goal was to balance so that a player has a 60% chance of victory given a specific gold value
20090909 02:40:50< corn> yeah, I know
20090909 02:41:05-!- isaac [n=isaac@debian/developer/isaac] has quit [Read error: 60 (Operation timed out)]
20090909 02:41:13< Crab_> corn: for example, a campaign creator might say "this particular mid-campaign scenario should be winnable 60% of the time"
20090909 02:41:29< Crab_> corn: or, getting fancier: "this particular mid-campaign scenario should be winnable 60% of the time by top 30% players"
20090909 02:41:55< Crab_> corn: with recalls it'll get even more fancier
20090909 02:42:46< Crab_> corn: since, say, having 20 L3 archers on 'vs undead' dungeon scenario can be less effective than having, say, 4 white mages
20090909 02:42:57< corn> this is just shifting the gold distribution left or right
20090909 02:43:14< Crab_> corn: yes, with 'only gold', I just shift  the distribution left or right
20090909 02:43:25< corn> so if a particular player had a 60% chance of winning that scenario before the rebalancing, and the starting gold is increased, that particular player will now have a 90% chance of winning
20090909 02:43:44< Crab_> corn: no.
20090909 02:44:23< corn> is the player unaffected because he was already above minimum starting gold?
20090909 02:44:25< Crab_> corn: there's a (hidden) cause-and-effect between the player's skill and his starting gold
20090909 02:44:56< Crab_> corn: but, merely giving higher starting gold will not grant him higher skill
20090909 02:45:12-!- happygrue_ [n=George@c-67-176-145-41.hsd1.in.comcast.net] has joined #wesnoth-dev
20090909 02:45:19< Crab_> corn: so, 'so if a particular player had a 60% chance of winning that scenario before the rebalancing, and the starting gold is increased, that particular player will now have a 90% chance of winning' is not true, but it is at least partly true.
20090909 02:46:01< corn> ok, I understand
20090909 02:46:34< Crab_> corn: 'so if a particular player had a 60% chance of winning that scenario before the rebalancing, and the starting gold is increased, that particular player will now have a >60% chance of winning '
20090909 02:46:51< Crab_> corn: so, by modifying the starting gold, we'll at least step in the right direction
20090909 02:47:27< corn> yes, making more complex changes in win distribution requires more complex modelling
20090909 02:47:54< fendrin> Crab_: Scenario 7 already has 0 percent gold carryover.
20090909 02:48:26< corn> going to go now, I'll make a patch with the recall list stuff in the morning
20090909 02:48:31< Crab_> fendrin: good. what about the 'make turns unlimited and buff orc income to make the scenario more interesting'  ?
20090909 02:48:35< Crab_> corn: and, since (ignoring recalls and saveload policy for the time being) 'more skill leads more gold', we'll not overstep 90% in the above example.
20090909 02:49:25< Crab_> corn:  'so if a particular player had a 60% chance of winning that scenario before the rebalancing, and the starting gold is increased, that particular player will now have a somewhere ' >60 % but <90 % ' chance of winning '
20090909 02:49:26-!- isaac [n=isaac@debian/developer/isaac] has joined #wesnoth-dev
20090909 02:49:41< Crab_> corn: so, hopefully, after a few releases, it will be 'close enough'.
20090909 02:54:37< CIA-62> alink * r38527 /trunk/src/dialogs.cpp: fix left race indentation in attack dialog
20090909 02:56:27< CIA-62> alink * r38528 /trunk/src/widgets/menu.cpp: 
20090909 02:56:27< CIA-62> Fix weapons info not aligned in attack dialog when only one has a special.
20090909 02:56:27< CIA-62> A space was already added for that, but some old-gui code started to trim trailing spaces.
20090909 02:56:27< CIA-62> (it is possible that previously hidden trailing spaces become visible in a dialog.
20090909 02:56:27< CIA-62> I didn't see one, but feel free to report them to me)
20090909 03:00:24-!- ardesh_ [n=ardesh@port-92-195-18-156.dynamic.qsc.de] has joined #wesnoth-dev
20090909 03:03:14-!- happygrue [n=George@wesnoth/developer/wintermute] has quit [Read error: 110 (Connection timed out)]
20090909 03:04:48-!- Crab_ [n=Crab_@wesnoth/developer/crab] has quit ["Leaving."]
20090909 03:16:33-!- ardesh [n=ardesh@port-92-195-117-249.dynamic.qsc.de] has quit [Read error: 110 (Connection timed out)]
20090909 03:18:40-!- BenUrban [n=benurban@unaffiliated/benurban] has quit [Read error: 104 (Connection reset by peer)]
20090909 03:25:49-!- BenUrban [n=benurban@unaffiliated/benurban] has joined #wesnoth-dev
20090909 03:31:32-!- wajimba2 [n=Andrew_A@c-24-118-61-250.hsd1.mn.comcast.net] has quit [Read error: 60 (Operation timed out)]
20090909 03:34:41-!- happygrue [n=George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20090909 03:36:49-!- Appleman1234 [n=Appleman@131.181.101.135] has quit [Read error: 104 (Connection reset by peer)]
20090909 03:37:59-!- ettin_ [n=jorda@wesnoth/developer/ettin] has joined #wesnoth-dev
20090909 03:39:50-!- ettin [n=jorda@wesnoth/developer/ettin] has quit [Read error: 110 (Connection timed out)]
20090909 03:44:49-!- wajimba [n=Andrew_A@c-24-118-61-250.hsd1.mn.comcast.net] has joined #wesnoth-dev
20090909 03:47:16-!- happygrue__ [n=George@c-67-176-145-41.hsd1.in.comcast.net] has joined #wesnoth-dev
20090909 03:49:49-!- Appleman1234 [n=Appleman@131.181.101.135] has joined #wesnoth-dev
20090909 03:52:16-!- Espreon [n=espreon@wesnoth/developer/espreon] has quit ["WRYYYYYYYYYYYYYYYYYYYY!"]
20090909 03:52:23-!- happygrue_ [n=George@c-67-176-145-41.hsd1.in.comcast.net] has quit [Read error: 110 (Connection timed out)]
20090909 03:52:40-!- happygrue [n=George@wesnoth/developer/wintermute] has quit [Read error: 104 (Connection reset by peer)]
20090909 04:23:50-!- DDR [n=chatzill@66.183.125.196] has joined #wesnoth-dev
20090909 04:24:33< CIA-62> fendrin * r38529 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/ (6 files): LoW: Side persistence changes all over the campaign.
20090909 04:32:37< alink> btw, I realize that the red-yellow-green color font is a bit too bright, I will try tuning down it a little
20090909 04:33:50-!- alink [n=alink@wesnoth/developer/alink] has quit [Remote closed the connection]
20090909 04:47:24-!- Espreon [n=espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20090909 04:53:28-!- Espreon [n=espreon@wesnoth/developer/espreon] has quit ["WRYYYYYYYYYYYYYYYYYYYY!"]
20090909 04:56:24-!- wesbot changed the topic of #wesnoth-dev to: 1.6.5 released, announcing on Tuesday or Wednesday | 91 bugs, 243 feature requests, 12 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090909 04:59:05-!- Blueblaze [n=nick@adsl-76-202-22-1.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20090909 05:06:11< CIA-62> fendrin * r38531 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/ (09_Bounty_hunters.cfg 22_Northern_Battle.cfg): LoW 9,22: Made wmllint quiet.
20090909 05:06:13< CIA-62> alink * r38530 /trunk/src/mouse_events.cpp: 
20090909 05:06:13< CIA-62> Attempt to improve readability of the attack dialog:
20090909 05:06:13< CIA-62> - Color chance-to-hit (to indicate good/bad, since it's the reverse of def%)
20090909 05:06:13< CIA-62> - Move it to its own line (easier to compare + implementation reason)
20090909 05:06:15< CIA-62> - Damage now have their own line, except that possible special goes there too,
20090909 05:06:17< CIA-62> which works nicely with damage/blows-based specials (almost all of them)
20090909 05:06:19< CIA-62> - Use same darker color for range as in sidebar.
20090909 05:09:37-!- Blueblaze [n=nick@adsl-76-202-22-1.dsl.hstntx.sbcglobal.net] has quit [Read error: 54 (Connection reset by peer)]
20090909 05:09:53-!- Blueblaze [n=nick@adsl-76-202-22-1.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20090909 05:20:27-!- ABCD_ [n=ABCD@gentoo/developer/abcd] has joined #wesnoth-dev
20090909 05:21:04-!- ABCD [n=ABCD@gentoo/developer/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090909 05:24:30-!- ABCD_ is now known as ABCD
20090909 05:25:44< CIA-62> fendrin * r38532 /trunk/data/campaigns/Legend_of_Wesmere/ (9 files in 2 dirs): LoW: Removed the hacky debug macros.
20090909 05:55:50-!- Chusslove [n=Chusslov@brsg-d9beec11.pool.mediaWays.net] has quit [Read error: 110 (Connection timed out)]
20090909 06:01:32-!- Chusslove [n=Chusslov@brsg-d9beee2a.pool.mediaWays.net] has joined #wesnoth-dev
20090909 06:23:52-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has quit []
20090909 06:53:06-!- wajimba [n=Andrew_A@c-24-118-61-250.hsd1.mn.comcast.net] has quit [Remote closed the connection]
20090909 07:21:47-!- Sirp [n=user@wesnoth/developer/dave] has quit ["leaving"]
20090909 07:26:45-!- silene [n=plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20090909 07:38:57-!- Appleman1234 [n=Appleman@131.181.101.135] has quit [Read error: 104 (Connection reset by peer)]
20090909 07:45:33-!- silene [n=plouf@wesnoth/developer/silene] has quit [Read error: 104 (Connection reset by peer)]
20090909 07:54:30-!- Blueblaze [n=nick@adsl-76-202-22-1.dsl.hstntx.sbcglobal.net] has quit [Remote closed the connection]
20090909 08:06:17-!- zookeeper [n=l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20090909 08:06:47-!- ardesh__ [n=ardesh@port-92-195-68-3.dynamic.qsc.de] has joined #wesnoth-dev
20090909 08:19:26-!- stikonas [n=and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20090909 08:21:45-!- DDR [n=chatzill@66.183.125.196] has quit [Read error: 110 (Connection timed out)]
20090909 08:22:26-!- ardesh_ [n=ardesh@port-92-195-18-156.dynamic.qsc.de] has quit [Read error: 110 (Connection timed out)]
20090909 08:23:40-!- silene [n=plouf@ASte-Genev-Bois-152-1-45-187.w83-114.abo.wanadoo.fr] has joined #wesnoth-dev
20090909 08:27:05-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090909 08:27:17-!- ardesh_ [n=ardesh@port-92-195-83-54.dynamic.qsc.de] has joined #wesnoth-dev
20090909 08:28:53-!- ardesh__ [n=ardesh@port-92-195-68-3.dynamic.qsc.de] has quit [Read error: 60 (Operation timed out)]
20090909 08:28:53-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 08:29:08-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090909 08:30:53-!- stikonas [n=and@213.164.125.176] has joined #wesnoth-dev
20090909 08:33:52-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090909 08:35:33-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 08:36:04-!- ardesh__ [n=ardesh@port-92-195-52-219.dynamic.qsc.de] has joined #wesnoth-dev
20090909 08:44:48-!- ardesh [n=ardesh@port-92-195-29-167.dynamic.qsc.de] has joined #wesnoth-dev
20090909 08:46:21-!- Crab_ [n=Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20090909 08:47:39< Crab_> fendrin: you talked about LoW 7 having 0 gold carryover... this is *not* the case - i.e., if I finish LoW 7 with ~900 gold, I start next one with ~1080 gold, if I finish with ~10000 gold, I start next one with ~9000 gold.
20090909 08:50:47-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090909 08:51:02-!- ardesh_ [n=ardesh@port-92-195-83-54.dynamic.qsc.de] has quit [Read error: 110 (Connection timed out)]
20090909 09:01:12-!- ardesh__ [n=ardesh@port-92-195-52-219.dynamic.qsc.de] has quit [Read error: 110 (Connection timed out)]
20090909 09:04:28< Crab_> fendrin: this is an engine bug. I'll try to fix.
20090909 09:09:22-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090909 09:11:14-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 09:12:17-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090909 09:14:07-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 09:15:50< silene> Crab_: this is most probably bug 14160
20090909 09:16:44< Crab_> silene: it is caused by delayed checking of end_level_.  end_level_ is populated correctly, but the actual end_level_exception is fired by check_victory (src/actions.cpp ), which ignores that end_level_
20090909 09:17:35< Crab_> silene: yes, this is 14160
20090909 09:17:39< silene> i know
20090909 09:18:44< Crab_> silene: should I try to fix it ? (i'm thinking about calling victory_conditions::set_foo() in playsingle_controller::force_end_level)
20090909 09:18:51-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090909 09:19:48< silene> that's not the fix i had in mind; i would rather have check_victory delegates to play_controller; there is no reason to duplicate all the informations (and then have them outdated) in actions.cpp
20090909 09:20:38-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 09:21:19-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090909 09:21:44< Crab_> silene: well, we only need to set 'victory_conditions::set_victory_when_enemies_defeated(false);' - no reason to duplicate gold/carryover info.
20090909 09:23:05< silene> hmm... good idea, it may well be a minimal fix for the issue (my long term goal is still to remove the actions.cpp crap, but it would become a lot less urgent)
20090909 09:23:36< Crab_> ok, then I'll fix the issue this way, and leave the bug open.
20090909 09:24:47-!- ardesh_ [n=ardesh@port-92-195-117-42.dynamic.qsc.de] has joined #wesnoth-dev
20090909 09:29:46-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090909 09:32:01-!- ardesh [n=ardesh@port-92-195-29-167.dynamic.qsc.de] has quit [Read error: 110 (Connection timed out)]
20090909 09:34:45-!- ardesh__ [n=ardesh@port-92-195-82-156.dynamic.qsc.de] has joined #wesnoth-dev
20090909 09:37:27-!- silene [n=plouf@ASte-Genev-Bois-152-1-45-187.w83-114.abo.wanadoo.fr] has quit ["Leaving."]
20090909 09:42:15< Ivanovic> moin
20090909 09:48:50-!- ardesh_ [n=ardesh@port-92-195-117-42.dynamic.qsc.de] has quit [Read error: 110 (Connection timed out)]
20090909 09:49:29-!- mjs-de [n=mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20090909 10:00:24-!- ardesh_ [n=ardesh@port-92-195-123-104.dynamic.qsc.de] has joined #wesnoth-dev
20090909 10:04:12< CIA-62> crab * r38533 /trunk/ (changelog src/playsingle_controller.cpp): Fix bug #14160 (in not long-term-optimal way). Fixes issues with LoW 7 gold carryover.
20090909 10:04:19< Crab_> fendrin: ^
20090909 10:15:35-!- ardesh__ [n=ardesh@port-92-195-82-156.dynamic.qsc.de] has quit [Read error: 110 (Connection timed out)]
20090909 10:24:24-!- ardesh_ [n=ardesh@port-92-195-123-104.dynamic.qsc.de] has quit [Read error: 110 (Connection timed out)]
20090909 10:37:29< CIA-62> esr * r38534 /trunk/data/campaigns/Delfadors_Memoirs/ (2 files in 2 dirs): Address a story problem pointed out in the forums.
20090909 10:40:58-!- Crab_ [n=Crab_@wesnoth/developer/crab] has quit ["Leaving."]
20090909 10:49:54< CIA-62> esr * r38535 /trunk/data/campaigns/Delfadors_Memoirs/scenarios/03_road_to_weldyn.cfg: Remove an unneeded TODO tag.
20090909 11:01:08< Ivanovic> esr: could you have a go at unifying the spelling of "Sceptre" over all the campaigns?
20090909 11:01:19< Ivanovic> currently in SoF a different spelling is used than in the rest
20090909 11:01:34< Ivanovic> for the Sceptre of Fire that is
20090909 11:01:42< esr> I'll look into it.
20090909 11:01:48< Ivanovic> great, thanks
20090909 11:02:02< esr> Right now I'm trying to confirm fix on a LoW bug.
20090909 11:02:14< Ivanovic> please also add a pofix directive (since it should just be a plain "Scepter of Fire" -> "Sceptre of Fire" change)
20090909 11:02:21< Ivanovic> sure
20090909 11:02:33-!- loonybot [n=loonybot@79.139.139.133] has joined #wesnoth-dev
20090909 11:03:28-!- loonycyborg [n=sergey@79.139.139.133] has joined #wesnoth-dev
20090909 11:06:52< Ivanovic> uhm, why do i have some strange dejavu when seening this username? http://www.phoronix.com/forums/showthread.php?p=91195
20090909 11:07:08< Ivanovic> okay, at least he points out that "you might end up having problems"
20090909 11:07:09< Ivanovic> ^^
20090909 11:12:36< loonycyborg> Because it matches one of our often vilified former developers :P
20090909 11:13:08< loonycyborg> Are you sure whether it's even the same guy?
20090909 11:15:02< Ivanovic> it might be...
20090909 11:15:22< Ivanovic> and in theory he is capable of stuff like this (his technical knowledge is really fine)
20090909 11:15:32< Ivanovic> he just misses to test the stuff as good as he should do
20090909 11:15:51< Ivanovic> (which is IMO a no go for xorg and kernel related stuff!)
20090909 11:20:26-!- YogiHH [i=d4ca9d15@wesnoth/developer/yogihh] has joined #wesnoth-dev
20090909 11:20:30< YogiHH> hello
20090909 11:30:33 * esr mutters curses in suokko's direction
20090909 11:31:12-!- noy [n=Noy@wesnoth/developer/noy] has quit []
20090909 11:50:32< zookeeper> what's that topic waiting approval thing on the forums?
20090909 11:50:48< zookeeper> who gets to tag them as such and why would it have been done for that thread?
20090909 11:51:31< CIA-62> esr * r38536 /trunk/data/campaigns/Sceptre_of_Fire/scenarios/8_The_Dragon.cfg: Clean up "scepter" spelling.
20090909 11:52:21-!- Blueblaze [n=nick@adsl-76-202-22-1.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20090909 12:01:52-!- Blueblaze [n=nick@adsl-76-202-22-1.dsl.hstntx.sbcglobal.net] has quit [Remote closed the connection]
20090909 12:08:23< CIA-62> esr * r38537 /trunk/data/ (4 files in 3 dirs): Change some macro names to use uniform "scepte" spelling fot Sceptre of Fire.
20090909 12:18:54< CIA-62> esr * r38538 /trunk/data/tools/wmllint: Tell the spellchecker about "sceptre".
20090909 12:22:03-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 12:23:00< CIA-62> esr * r38539 /trunk/data/campaigns/Heir_To_The_Throne/utils/httt_utils.cfg: Safe, string-only Scepter -> Scepter changes.
20090909 12:24:30-!- stikonas changed the topic of #wesnoth-dev to: 1.6.5 released, announcing on Wednesday | 91 bugs, 243 feature requests, 12 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090909 12:40:32-!- Netsplit clarke.freenode.net <-> irc.freenode.net quits: Rhonda, shadowmaster, stikonas, AnMaster, yann, ABCD, zookeeper, kimrhh, nonsensical, fendrin,  (+26 more, use /NETSPLIT to show all of them)
20090909 12:40:44-!- lobby [n=wesnoth@wesnoth/bot/lobby] has joined #wesnoth-dev
20090909 12:40:44-!- Topic for #wesnoth-dev: 1.6.5 released, announcing on Wednesday | 91 bugs, 243 feature requests, 12 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090909 12:40:44-!- Topic set by stikonas [] [Wed Sep  9 12:24:30 2009]
20090909 12:40:44[Users #wesnoth-dev]
20090909 12:40:44[ ABCD        ] [ corn   ] [ fendrin    ] [ lobby      ] [ shadowmaster] [ yann     ] 
20090909 12:40:44[ AI0867      ] [ deekay ] [ happygrue__] [ loonybot   ] [ shikadibot  ] [ YogiHH   ] 
20090909 12:40:44[ Alesis-Novik] [ dfranke] [ ilor       ] [ loonycyborg] [ Smar        ] [ zookeeper] 
20090909 12:40:44[ AnMaster    ] [ Doppp  ] [ isaac      ] [ mjs-de     ] [ Soliton     ] 
20090909 12:40:44[ BenUrban    ] [ erl_   ] [ Ivanovic   ] [ nital      ] [ stikonas    ] 
20090909 12:40:44[ Chusslove   ] [ esr    ] [ kimrhh     ] [ nonsensical] [ Tigge       ] 
20090909 12:40:44[ CIA-62      ] [ ettin_ ] [ knotwork_  ] [ Rhonda     ] [ wesbot      ] 
20090909 12:40:44-!- Irssi: #wesnoth-dev: Total of 38 nicks [0 ops, 0 halfops, 0 voices, 38 normal]
20090909 12:40:53-!- Channel #wesnoth-dev created Tue Jan 27 06:28:41 2009
20090909 12:41:41< Ivanovic> esr: ugh, i feat correcting the sceptre spelling stuff will be a mess
20090909 12:41:50-!- Irssi: Join to #wesnoth-dev was synced in 73 secs
20090909 12:41:51< Ivanovic> since capitalization varies as well as spelling
20090909 12:42:56< Ivanovic> (sure, cases of "Scepter of Fire" -> "Sceptre of Fire" are dead easy, it gets interesting with the plain "scepter" and "Septer" cases...)
20090909 12:43:42< Ivanovic> and: what should the weapon name be, should it be "sceptre of fire" of "Sceptre of Fire"? (talking about the httt units that can get it)
20090909 12:46:29< Ivanovic> that is the usage of "scepter" in trow is fine (only exists once there and there is nothing known about *the* Sceptre there
20090909 12:46:46< Ivanovic> though this one is strange (and should probably get some markup stuff added):
20090909 12:46:48< Ivanovic> data/campaigns/Under_the_Burning_Suns/scenarios/02_Across_the_Harsh_Sands.cfg:                    message= _ "Before the Great Fall, there were huge cities, with great schools of magery, libraries of books, vast repositories of knowledge. Most of it was destroyed in the ensuing chaos and years of decay, but I'm searching for the little fragments that are left. For example, have you ever heard of The Scepter of Fire?"
20090909 12:47:13< Ivanovic> (since the capital 'T' in "The Scepter of Fire" is strange otherwise)
20090909 12:47:35< zookeeper> weapon names are always lower case as you know
20090909 12:50:13< Ivanovic> for NR as well as THoT a plain s/Scepter/Septre will be enough, that is only in one case they don't explicitly add an "of Fire" behind, though it is clear that they are talking about exactly this one
20090909 12:50:41-!- Amu_ [n=smar@a88-113-60-192.elisa-laajakaista.fi] has joined #wesnoth-dev
20090909 12:50:44< Ivanovic> so the only possible mess left is httt
20090909 12:54:55< fendrin> esr: In DM scenario 5 is something strange with the story. Garard tells to Delfador that he shall stay with him at the place. Some sentences above he told that he will ride to the ford of Abez. This are the lines 102 and 114. 
20090909 13:07:42-!- fabi [n=fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20090909 13:24:07-!- fendrin [n=fabi@wesnoth/developer/fendrin] has quit [Read error: 110 (Connection timed out)]
20090909 13:26:57-!- ardesh [n=ardesh@port-92-195-112-21.dynamic.qsc.de] has joined #wesnoth-dev
20090909 13:27:23-!- euschn [n=chatzill@wesnoth/developer/euschn] has joined #wesnoth-dev
20090909 13:28:49< CIA-62> soliton * r38541 /branches/1.4/src/ (5 files): cache the ignores and friends list
20090909 13:28:58< CIA-62> soliton * r38542 /branches/1.4/utils/campaigns_client.py: fixed tarball creation
20090909 13:29:25-!- vigg [n=vigg-vig@jabber.hsdnetwork.net] has joined #wesnoth-dev
20090909 13:29:34< Soliton> oops.. r38541 wasn't planned...
20090909 13:30:29< Soliton> not that anyone cares, i guess. :-)
20090909 13:33:11-!- Zeph [n=ars@209-210-207-246.dslnorthwest.net] has joined #wesnoth-dev
20090909 13:33:39-!- Zeph is now known as Guest7621
20090909 13:37:59-!- vigg [n=vigg-vig@jabber.hsdnetwork.net] has left #wesnoth-dev []
20090909 13:38:03-!- nonsensical [n=ars@209-210-207-246.dslnorthwest.net] has quit [Read error: 60 (Operation timed out)]
20090909 13:49:03-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 110 (Connection timed out)]
20090909 13:56:10-!- loonycyborg [n=sergey@wesnoth/developer/loonycyborg] has quit ["KVIrc 3.4.2 Shiny http://www.kvirc.net/"]
20090909 13:56:40-!- loonybot [n=loonybot@wesnoth/bot/loonybot] has quit [Remote closed the connection]
20090909 14:14:19-!- loonybot [n=loonybot@79.139.139.133] has joined #wesnoth-dev
20090909 14:15:31-!- loonycyborg [n=sergey@79.139.139.133] has joined #wesnoth-dev
20090909 14:21:44-!- vigg [i=596fb03c@gateway/web/freenode/x-jfuijulwyahefplx] has joined #wesnoth-dev
20090909 14:50:35-!- Sapient [n=patrickp@wesnoth/developer/sapient] has joined #wesnoth-dev
20090909 15:04:36-!- vigg [i=596fb03c@gateway/web/freenode/x-jfuijulwyahefplx] has left #wesnoth-dev []
20090909 15:06:15-!- BenUrban [n=benurban@unaffiliated/benurban] has quit ["Power failu"]
20090909 15:06:56-!- euschn [n=chatzill@wesnoth/developer/euschn] has quit [Read error: 60 (Operation timed out)]
20090909 15:07:07< YogiHH> Soliton: Can you try and load an mp savegame on the dev server? I would like to know if it has been my changes to the savegame structure that are causing the errors i face or not.
20090909 15:09:02-!- melinath [n=melinath@132.162.78.236] has joined #wesnoth-dev
20090909 15:09:12< Soliton> YogiHH: works fine.
20090909 15:09:34< YogiHH> Soliton: hmpf, i feared you would say so ;-)
20090909 15:11:45< Soliton> YogiHH: well, the error you get is clear enough, no?
20090909 15:12:44< YogiHH> Soliton: yes, it is. Unfortunately i am not sure where exactly it comes from, as the name of the attribute seems to be empty for whatever reason. But i'll find it eventually.
20090909 15:13:55< YogiHH> Soliton: I first suspected the objectives attribute, as it spans several lines and has translations and markup in it. But if your save loads fine, that's probably not the reason.
20090909 15:14:24< Soliton> zgrep '[^"]$' file
20090909 15:14:39< Soliton> should narrow it down a lot.
20090909 15:27:48-!- Chusslove [n=Chusslov@brsg-d9beee2a.pool.mediaWays.net] has quit [Read error: 110 (Connection timed out)]
20090909 15:34:42-!- Chusslove [n=Chusslov@brsg-d9bef920.pool.mediaWays.net] has joined #wesnoth-dev
20090909 15:37:58-!- melinath [n=melinath@132.162.78.236] has quit ["Verlassend"]
20090909 15:46:29-!- crimson_p [n=irchon@64.201.60.211] has joined #wesnoth-dev
20090909 15:48:29-!- crimson_p [n=irchon@64.201.60.211] has quit [Remote closed the connection]
20090909 15:56:46-!- YogiHH [i=d4ca9d15@wesnoth/developer/yogihh] has left #wesnoth-dev []
20090909 16:04:41-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20090909 16:21:32< CIA-62> esr * r38543 /trunk/data/campaigns/Delfadors_Memoirs/scenarios/05_council_in_weldyn.cfg: 
20090909 16:21:32< CIA-62> Remove a fiossil from the version ot the narrative in which Garard I
20090909 16:21:32< CIA-62> was still alive and sent Garard II north.
20090909 16:47:16-!- rosso_ [n=rosso@dslb-088-070-093-041.pools.arcor-ip.net] has joined #wesnoth-dev
20090909 16:47:33< rosso_> hi ppl!
20090909 16:50:03< rosso_> Is there need for bandwidth, serverspace, mirrors or alike in the project? 
20090909 16:51:26< Soliton> i suppose one more set of alternate mp servers wouldn't hurt. not really a need though.
20090909 16:53:34-!- DDR [n=chatzill@66.183.125.196] has joined #wesnoth-dev
20090909 16:54:23< rosso_> Hm, well, i have a server with a quite good bandwidth and lots of space, of which i gladly would give sth to the community.
20090909 16:56:24-!- wesbot changed the topic of #wesnoth-dev to: 1.6.5 released, announcing on Wednesday | 92 bugs, 243 feature requests, 12 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090909 16:56:39< Soliton> if you don't mind helping an unofficial part out you could ask whether there is a need for a better/additional server for the ladder site (ladder.subversiva.org).
20090909 16:59:12< rosso_> Soliton: You know who to get in touch with by any chance?
20090909 16:59:21< Soliton> there is a forum thread about it or you can ask chains if he is around on the mp server or sometimes here.
20090909 16:59:55< Soliton> http://www.wesnoth.org/forum/viewtopic.php?f=15&t=17941
20090909 17:00:24< rosso_> yeah, i remember him from the MP server, cheers Mate
20090909 17:00:51-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 17:01:13< Soliton> you're welcome.
20090909 17:11:02-!- cib0 [n=cib@p5DD34904.dip.t-dialin.net] has joined #wesnoth-dev
20090909 17:19:50-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 104 (Connection reset by peer)]
20090909 17:30:36-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20090909 17:46:05-!- stikonas [n=and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20090909 17:46:17-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090909 17:46:48-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 17:46:54-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 54 (Connection reset by peer)]
20090909 17:47:28-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 17:55:42< Soliton> -                            name=scepter of fire
20090909 17:55:42< Soliton> +                            name=`sceptre of fire
20090909 17:55:51< Soliton> esr: intentional change?
20090909 17:58:59< esr> Soliton: Argh.  No.
20090909 17:59:15< esr> I wonder why wmllint didn't catch that.
20090909 17:59:25< esr> What file?
20090909 18:00:26< Ivanovic> esr: you saw what i left in here about 5 hours ago regarding how to rename stuff?
20090909 18:00:38< esr> Ivanovic: No.
20090909 18:00:39< Soliton> data/campaigns/Sceptre_of_Fire/scenarios/8_The_Dragon.cfg
20090909 18:00:45< Ivanovic> the conventions for when to capitalize Scepter and all the likes
20090909 18:01:14< esr> I thougt I wrote those up in NoSpellingMistakes.
20090909 18:01:20< Ivanovic> http://wesnoth.org/irclogs/2009/09/%23wesnoth-dev.2009-09-09.log
20090909 18:01:34< Ivanovic> starting at 20090909 12:41:41<
20090909 18:02:29< CIA-62> esr * r38544 /trunk/data/campaigns/Sceptre_of_Fire/scenarios/8_The_Dragon.cfg: Typo fix.
20090909 18:08:04< esr> Ivanovic: I have been using a slightly different rule.  I'm using the -re spelling only where the individual artifactt is being referred to.  Thus, still "scepter" in the one place in TROW where Haldric is speculating.
20090909 18:08:33< Ivanovic> so the rest will be a capitalized Sceptre?
20090909 18:08:41< esr> Yes.
20090909 18:09:56< Ivanovic> will you do this change (probably automated) and add it to pofix.py? (in a way that only changes the mentioned textdomains, so that in trow there is no rename of scepter to Sceptre
20090909 18:10:08< esr> Aaaaargh.
20090909 18:10:54< esr> I think I'm going to *delete* pofix.py if I hear a request like that again! 
20090909 18:11:01< Ivanovic> tststs
20090909 18:11:57< esr> I wrote it as a one-shot - if it adsds to my maintainance burdens it it is not achieving its purpose.
20090909 18:12:31< Ivanovic> uhm, is the textdomain mentioning in there not meant to just keep things to a single domain?
20090909 18:14:05< esr> The logic to do that isn't implemented.  Presently the textdomain stuff only functions as documentation.  *That* I would be willing to fix. 
20090909 18:14:23< Ivanovic> this would be a *great* improvement
20090909 18:14:50< Ivanovic> then it would just be a case of "add the transformation for scepter tot he respective textdomains"
20090909 18:18:58< esr> Ivanovic: Anyway, this change cannot be automated.  The rules for when to use which form are tricky.
20090909 18:19:34< esr> It is almost bad enough to make me advocate dropping the "sceptre" spelling altogether.  I might get there yet...
20090909 18:21:03< Ivanovic> esr: just having the automated replacing per textdomain would already help
20090909 18:21:38< Ivanovic> this would help with the sceptre case, too
20090909 18:21:53< Ivanovic> then automated replacing (for capitalization) can be used, too
20090909 18:24:18-!- vigg [i=57fcffb1@gateway/web/freenode/x-lvkjegzzgqgskmjy] has joined #wesnoth-dev
20090909 18:29:42-!- happygrue [n=George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20090909 18:31:20-!- happygrue__ [n=George@c-67-176-145-41.hsd1.in.comcast.net] has quit [Read error: 104 (Connection reset by peer)]
20090909 18:32:39-!- ^y_emkidzu-it [n=KVIrc@host20-68-dynamic.20-79-r.retail.telecomitalia.it] has joined #wesnoth-dev
20090909 18:33:43-!- zookeeper [n=l@wesnoth/developer/zookeeper] has quit [Read error: 145 (Connection timed out)]
20090909 18:33:45-!- happygrue_ [n=George@c-67-176-145-41.hsd1.in.comcast.net] has joined #wesnoth-dev
20090909 18:35:02-!- zookeeper [n=l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20090909 18:35:32-!- happygrue [n=George@wesnoth/developer/wintermute] has quit [Read error: 104 (Connection reset by peer)]
20090909 18:35:41-!- DDR [n=chatzill@66.183.125.196] has quit [Read error: 145 (Connection timed out)]
20090909 18:36:31-!- happygrue_ [n=George@c-67-176-145-41.hsd1.in.comcast.net] has quit [Read error: 104 (Connection reset by peer)]
20090909 18:39:27-!- Noyga [n=lame-z@wesnoth/developer/noyga] has joined #wesnoth-dev
20090909 19:03:06-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090909 19:03:29-!- ^y_emkidzu-it [n=KVIrc@host20-68-dynamic.20-79-r.retail.telecomitalia.it] has quit [Read error: 104 (Connection reset by peer)]
20090909 19:03:38-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 19:05:19-!- Noyga [n=lame-z@wesnoth/developer/noyga] has left #wesnoth-dev ["Quitte"]
20090909 19:12:13-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 104 (Connection reset by peer)]
20090909 19:14:42-!- boucman [n=rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20090909 19:24:19< Soliton> anyone know what the difference between unit::defense_modifier() and unit_movement_type::defense_modifier() is? is it just that one gets the map and the other uses resources::game_map?
20090909 19:30:09-!- noy [n=Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090909 19:31:48< Soliton> same thing with movement_cost().. a lot of duplicated code. :-/
20090909 19:32:11< Soliton> (and of course it's not exactly in sync.)
20090909 19:34:37-!- Crab_ [n=Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20090909 19:34:59-!- silene [n=plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20090909 19:35:13< silene> hi
20090909 19:37:41< Ivanovic> hi silene
20090909 19:37:50< Ivanovic> Soliton: maybe euschn has an idea, not sure
20090909 19:37:56< Ivanovic> or silene can help
20090909 19:39:35< Ivanovic> silene: any "long term solution" about https://gna.org/bugs/index.php?14160 in view?
20090909 19:39:51< Ivanovic> i assume that there is no update regarding https://gna.org/bugs/index.php?13454 , right?
20090909 19:40:24< silene> Ivanovic: the fix from crab_ is a long term solution to the bug; it's just that the code still needs cleaning
20090909 19:41:12< silene> Ivanovic: no, no update
20090909 19:47:55-!- mordante [n=mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20090909 19:48:16< mordante> servus
20090909 19:50:17< Soliton> where would i put funtions that unit and unit_type share?
20090909 19:51:13< Crab_> Soliton: I'd say ' unit_types.?pp '
20090909 19:52:06< Soliton> yeah, it's actually unit_movement_type that shares with unit and that is in unit_types.
20090909 20:02:03-!- Sapient [n=patrickp@wesnoth/developer/sapient] has left #wesnoth-dev []
20090909 20:02:35< mordante> Ivanovic, somebody already mentioned before that suokko is working on radeon drivers... 
20090909 20:08:58-!- vigg [i=57fcffb1@gateway/web/freenode/x-lvkjegzzgqgskmjy] has quit ["Page closed"]
20090909 20:11:33< Ivanovic> mordante: and it still suprises me...
20090909 20:14:06< shadowmaster> he works on the drm project too
20090909 20:14:52< shadowmaster> both libdrm and the radeon-specific parts
20090909 20:15:20 * fabi fixed an old multiplayer campaign up. Someone around to playtest with?
20090909 20:22:46< shadowmaster> wesbot: seen Aethaeryn
20090909 20:22:47< wesbot> shadowmaster: The person with the nick Aethaeryn last spoke 1d ago.  1d ago was here and on the channels #Wesnoth and #Wesnoth-umc-dev with the message: "Bah!"
20090909 20:22:52< mordante> Ivanovic, me too
20090909 20:28:09< cib0> fabi, sure, i love playing MP campaigns =)
20090909 20:30:26< cib0> fabi: stable, dev version or svn?
20090909 20:39:14< fabi> cib0: dev version
20090909 20:39:21< fabi> corn: better svn
20090909 20:40:09< fabi> cib0: ^
20090909 20:40:11< cib0> fabi: i shall download the dev version then
20090909 20:40:34< fabi> cib0: No, I have only trunk here. I answered too fast.
20090909 20:40:47< cib0> fabi: alright, then trunk(hope i got enough space)
20090909 20:40:58< fabi> cib0: :-)
20090909 20:41:05< mordante> ilor your GUI2_EASY_BUTTON_CALLBACK macro is evil, adding new variables in scope and even using them
20090909 20:41:23< cib0> fabi: 7.2 GB, should suffice, if barely :p
20090909 20:42:36< CIA-62> mordante * r38545 /trunk/src/gui/widgets/ (18 files): 
20090909 20:42:36< CIA-62> find_widget renaming part [1/2].
20090909 20:42:36< CIA-62> The name of find_widget in the twidget class is a bit silly due to the
20090909 20:42:36< CIA-62> double widget part. The overload for the coordinate and id version also
20090909 20:42:36< CIA-62> didn't work to well.
20090909 20:42:37< CIA-62> Rename the coordinate version to find_at.
20090909 20:42:42< CIA-62> mordante * r38546 /trunk/src/gui/ (26 files in 2 dirs): 
20090909 20:42:43< CIA-62> find_widget renaming part [2/2].
20090909 20:42:45< CIA-62> The name of find_widget in the twidget class is a bit silly due to the
20090909 20:42:47< CIA-62> double widget part. The overload for the coordinate and id version also
20090909 20:42:49< CIA-62> didn't work to well.
20090909 20:42:51< CIA-62> Rename the id version to find.
20090909 20:42:53< CIA-62> mordante * r38547 /trunk/src/gui/ (23 files in 2 dirs): 
20090909 20:42:55< CIA-62> Rename NEW_find_widget to find_widget.
20090909 20:42:59< CIA-62> The old find widgets are gone so it's now easy to do the renaming.
20090909 20:43:03< CIA-62> mordante * r38548 /trunk/changelog: Changelog update.
20090909 20:43:15-!- dfranke [i=dfranke@wesnoth/developer/dfranke] has quit ["ZNC - http://znc.sourceforge.net"]
20090909 20:50:42< cib0> wish there was a trunk branch without translations
20090909 20:52:48< Soliton> you can maintain partial checkouts with svn but it's a pain, i think.
20090909 20:53:06-!- grzywacz [n=grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20090909 20:53:22-!- lizard_r [n=Miranda@wesnoth/umc-dev/developer/lizard] has joined #wesnoth-dev
20090909 20:54:54< fabi> I still have problem with the story speech. The id of the leader seems not to be the same as specified in my side tags.
20090909 20:56:17< cib0> fabi: are you sure you are using the right attributes? i remember some changes there, then again, that was a long time ago
20090909 20:57:13< fabi> cib0: It's the first time I am messing around with multiplayer wml.
20090909 20:57:54< cib0> fabi: pastebin your side definition and message tags?
20090909 20:58:12-!- ABCD [n=ABCD@gentoo/developer/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090909 20:58:20-!- ABCD [n=ABCD@gentoo/developer/abcd] has joined #wesnoth-dev
20090909 20:59:51< fabi> cib0: http://wesnoth.pastebin.com/d45bccc2e
20090909 21:01:21< cib0> hm, weird
20090909 21:01:37< cib0> *checks out side documentation*
20090909 21:03:57< cib0> fabi: does anything change if you define the leader in a [unit] tag?
20090909 21:04:03< cib0> i don't see why it wouldn't work
20090909 21:05:15< fabi> cib0: The campaign is both, singleplayer and multiplayer in one code base. The singleplayer mode is working well.
20090909 21:06:06< shadowmaster> aren't leaders' id and names replaced with player names by default?
20090909 21:06:15< shadowmaster> (in MP, of course).
20090909 21:06:19< cib0> probably
20090909 21:06:26< cib0> i never ran into that problem though
20090909 21:06:37< shadowmaster> I even recall cib0 complaining about that for a certain attempt at a MP campaign from HttT
20090909 21:06:37< cib0> if i want the leader, i filter for leadership and side ;)
20090909 21:06:44< cib0> really?
20090909 21:06:56< cib0> i think you're talking about name
20090909 21:07:02< cib0> name is definitely replaced
20090909 21:07:06< cib0> id? not sure
20090909 21:07:12< shadowmaster> save a game and check.
20090909 21:08:37< cib0> shadowmaster: the really annoying thing is that even a reload will replace the name
20090909 21:08:51< cib0> but i think this can be solved in WML now with that initialization tag
20090909 21:09:10< cib0> *event
20090909 21:09:43< shadowmaster> preload?
20090909 21:11:08< cib0> i think so
20090909 21:11:27< shadowmaster> I read it in the changelog.
20090909 21:13:15-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 21:14:00< cib0> ah, yes, lovely, the three different tags that all can store the scenario depending on situation =)
20090909 21:14:18< cib0> those gave me awesome headaches back in the day
20090909 21:16:28< fabi> cib0: Are you ready?
20090909 21:16:54< cib0> fabi: oops, nope, i still need to compile =(
20090909 21:17:05< cib0> and i don't have the right lua headers..
20090909 21:17:57< cib0> do i need lua 5.1.4?
20090909 21:18:38< cib0> ah, got all the deps now, whewt
20090909 21:18:52< cib0> fabi: ill be ready in about 5 minutes
20090909 21:19:15< cib0> i thought.. still missing some boost libs -.-
20090909 21:19:26< cib0> but does scons say anything? nah
20090909 21:20:33< cib0> i have libboost-dev and libboost-iostream-dev.. what am i missing?
20090909 21:23:11< shadowmaster> what os?
20090909 21:23:18< cib0> shadowmaster, debian
20090909 21:23:20< Soliton> there are some dependency problems with boost serialization or so on debian.
20090909 21:23:28< cib0> yeah i noticed
20090909 21:23:38< shadowmaster> don't you need boost 1.35 now?
20090909 21:24:00< cib0> i need an older version of libboost? o.0
20090909 21:24:06< shadowmaster> ah no.
20090909 21:24:16< cib0> debian testing has 1.39
20090909 21:24:20< shadowmaster> you didn't specify whether you were using the stable, testing or unstable distribution, however.
20090909 21:24:28< cib0> yeah, sorry, testing :p
20090909 21:24:33< Soliton> what's the error?
20090909 21:25:03< cib0> thought that'd be obvious for someone who likes to compile sources that are younger than 2 years ;)
20090909 21:25:58< cib0> http://pastebin.ca/1560143
20090909 21:26:50< Soliton> so do you have boost serialization installed?
20090909 21:26:58< CIA-62> thespaceinvader * r38549 /trunk/ (6 files in 5 dirs): Add and wire new Merman Netcaster portrait, update changelogs, portrait credits.
20090909 21:27:54< shadowmaster> cib0: no, not really. I'm a stable user.
20090909 21:28:01< shadowmaster> Wesnoth works just fine on it.
20090909 21:28:38< cib0> shadowmaster, wesnoth is an exception :p
20090909 21:29:38< shadowmaster> wesnoth, linux, radeonhd, qemu, dosbox, hyperion-ircd, etc.
20090909 21:30:02< shadowmaster> oh, and irssi.
20090909 21:32:24< cib0> fabi: i think i have to withdraw my offer. i don't feel like compiling lift right now
20090909 21:32:34< cib0> *boost
20090909 21:32:54< fabi> cib0: What operating system do you use?
20090909 21:33:05< cib0> fabi: debian testing
20090909 21:33:20< shadowmaster> you never answered the question of whether or not you have boost serialization installed btw
20090909 21:34:09< cib0> shadowmaster, oh, no, i overread that. i didn't =) that answers it
20090909 21:34:10-!- ABCD [n=ABCD@gentoo/developer/abcd] has left #wesnoth-dev ["http://quassel-irc.org - Chat comfortably. Anywhere."]
20090909 21:35:05< fabi> cib0: tried apt-get builddep wesnoth ?
20090909 21:35:06< cib0> seems to be working, thanks =)
20090909 21:35:18-!- Baufo [n=thomas@wesnoth/developer/baufo] has joined #wesnoth-dev
20090909 21:35:23 * shadowmaster redirects that to Soliton 
20090909 21:35:47< cib0> fabi: wouldn't that only work for stable?
20090909 21:36:08< cib0> i had all the dependencies for stable, compiled it myself, too
20090909 21:36:32< Soliton> you'd have to add -t experimental.
20090909 21:36:39< cib0> fabi: also, neglecting the chance of any more weird build errors, ill be on the server in 5 minutes =)
20090909 21:36:48< Soliton> and have experimental sources added of course.
20090909 21:38:39-!- happygrue [n=George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20090909 21:38:39< cib0> are substantial parts of the default AI already replaced with the new one? (was it formula?)
20090909 21:38:51< grzywacz> it's not meant to replace it
20090909 21:38:56< cib0> ah, too bad
20090909 21:39:15< cib0> i'd love to see an AI that cycles units, retreats, etc.
20090909 21:39:34< Soliton> those things are not mutually exclusive.
20090909 21:39:52< Crab_> cib0: new changes to ai allow to replace/add 'pieces' to those pieces which are in default ai
20090909 21:39:57< Soliton> the framework to replace parts of the ai is in place now afaiui.
20090909 21:40:22< Crab_> cib0: i.e., you don't have to reimplement the entire ai if you want to add new functionality / new ideas to it
20090909 21:40:40< cib0> Crab_: i see
20090909 21:40:50< Crab_> cib0: this way, we can, for example, mix-in formula_ai snippets (such as recruitment or unit formulas or side formulas) into default ai
20090909 21:41:09< cib0> Crab_: are those changes defined in the scenario WML?
20090909 21:41:26< cib0> or rather, can be defined
20090909 21:41:26< shadowmaster> fabi: why does Kalenz start 1 hex north of his keep in LoW:1?
20090909 21:41:43< Crab_> cib0: yes, the actual pieces from which we construct the ai can be specified in [side][ai] tag
20090909 21:42:10< fabi> shadowmaster: Changed the map. But not his starting position :-P
20090909 21:42:18< cib0> Crab_: neat =) bah, wesnoth is developing great features, we need scenarios to use em =)
20090909 21:42:25< Crab_> cib0: two 'frameworks' are available : first is simple "for each stage, play that stage"
20090909 21:43:13< Crab_> cib0: second is 'candidate action evaluation loop' : "for each candidate action, {check them all} pick best, execute, repeat from start"
20090909 21:43:29-!- yann [n=dwitch@81.57.214.146] has quit [Remote closed the connection]
20090909 21:43:54< Crab_> cib0: second framework allows for dynamic ordering of stages, and is free from an entire class of old default_ai bugs.
20090909 21:44:03-!- YogiHH [n=chatzill@d040070.adsl.hansenet.de] has joined #wesnoth-dev
20090909 21:44:23< Crab_> cib0: there's also some coding towards a third framework (which is based on Sirp's ideas )
20090909 21:44:26< YogiHH> Soliton: do we have some kind of length limit for WML processing on the server?
20090909 21:45:06< Crab_> cib0: now, some parts of LoW use some new ai-related tricks (changes to the ai at runtime)
20090909 21:45:31< Soliton> YogiHH: yes, 40MB is the limit for an uncompressed WML document, i think.
20090909 21:45:39< Soliton> YogiHH: the error will tell you.
20090909 21:45:40< fabi> is there the possibility to define difficult level during multiplayer setup?
20090909 21:45:48< Soliton> no.
20090909 21:46:31< cib0> fabi: most campaigns solve that with a pre-scenario to select the difficulty
20090909 21:47:48< fabi> cib0: Right, the campaign does use that. But it's done poorely.
20090909 21:48:04< cib0> fabi: how so?
20090909 21:48:35-!- Blueblaze [n=nick@adsl-76-202-22-1.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20090909 21:51:06< YogiHH> Soliton: I am getting out of ideas with my savegame problem. Can you have a look at this please? It's the relevant savegame WML and a server log (i made so every attribute gets logged).
20090909 21:51:10< YogiHH> Soliton: http://wesnoth.pastebin.com/dc7a88b7
20090909 21:51:16< YogiHH> any ideas?
20090909 21:54:07< cib0> fabi: done, finally
20090909 21:55:24< cib0> hm, nice, a filter dialogue
20090909 21:56:06< Soliton> YogiHH: that is the data how it is in the savegame or how wesnothd gets it?
20090909 21:56:25< Soliton> should be the former.
20090909 21:56:29< YogiHH> Soliton: It's the savegame data
20090909 21:56:45< Soliton> well, the interesting part is what wesnoth sends.
20090909 21:57:03< Crab_> YogiHH: can you try making a copy, removing half of it, and testing if the problem is still there ?
20090909 21:57:09< YogiHH> Soliton: Could it be because i am not using a forum account?
20090909 21:57:18< Crab_> YogiHH: i.e. try to find offending snippet by binary search
20090909 21:57:42< Soliton> Crab_: the offending WMl might well be added by wesnoth later.
20090909 21:57:48< YogiHH> Crab_: It seems from the log that the "name"-attribute is the offending one
20090909 21:58:01< shadowmaster> um... lua tags broken again?
20090909 21:58:05< Crab_> YogiHH: try removing it :)
20090909 21:58:12< Soliton> YogiHH: so there is an empty name attribute.
20090909 21:58:33< shadowmaster> g20090909 15:57:07 error scripting/lua: /home/shadowm/src/wesnoth/data/lua/wml-tags.lua:165: bad argument #1 to 'gmatch' (string expected, got nil)
20090909 21:58:43< YogiHH> Soliton: That would match the no-forum-account theory, wouldn't it?
20090909 21:58:56< Soliton> YogiHH: not really.
20090909 21:59:15< Soliton> i don't see how that can have anything to do with it.
20090909 21:59:53< YogiHH> Soliton: Anyway, i will try and find out what the server is actually processing.
20090909 22:00:02< mordante> I'm off night
20090909 22:00:10< cib0> fabi: did you proceed? i get a End Turn button that does nothing
20090909 22:00:21< shadowmaster> I got OOS as observer
20090909 22:00:26< Soliton> YogiHH: if you start it with -d it tells you the side WML.
20090909 22:00:27< Crab_> YogiHH: also, <span foreground=""#00ff00"">" has to many ""s, imo...
20090909 22:00:40-!- mordante [n=mordante@wesnoth/developer/mordante] has quit ["Leaving"]
20090909 22:00:41< cib0> i got an OOS, too
20090909 22:00:46< Soliton> Crab_: " are encoded as "".
20090909 22:00:53< Soliton> inside strings.
20090909 22:00:58< cib0> fabi: i got a WML deprecated error because continue_no_save is used
20090909 22:02:08< fabi> cib0: Seems it's not fit for multiplayer.
20090909 22:02:11-!- yann [n=dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has joined #wesnoth-dev
20090909 22:02:33< cib0> fabi: looks like bug to me.
20090909 22:02:53< shadowmaster> fabi: why not  try to fix that deprecation warning?
20090909 22:03:10< Crab_> Soliton: ok
20090909 22:03:14-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090909 22:03:40< Soliton> YogiHH: probably simple_wml should just accept empty values.
20090909 22:03:44-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 22:04:10< Soliton> seems odd that it doesn't.
20090909 22:05:07< fabi> shadowmaster: fixed it.
20090909 22:05:17< Soliton> if it's really empty i wonder why it didn't throw "did not find quotes around attribute" though.
20090909 22:06:04< Soliton> YogiHH: actually you should start the client with debug level so you can see what it sends.
20090909 22:06:04< cib0> fabi: seems to be bugged
20090909 22:06:14< cib0> fabi: getting an "End Turn" button instead of "End Scenario"
20090909 22:06:26< Soliton> YogiHH: since you can't display it on the server side if it errors out before. :-P
20090909 22:06:32< fabi> cib0: Maybe the result must be specified for each side now?
20090909 22:06:40< YogiHH> ok
20090909 22:07:03< cib0> fabi: maybe
20090909 22:07:40< cib0> personally, i think whenever someone makes a change to meta-scenario logic, he should test an MP campaign with it :p
20090909 22:08:01< shadowmaster> oh no.
20090909 22:08:06-!- Espreon [n=espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20090909 22:08:19< Soliton> hard to test things when most logic is broken atm. ;-)
20090909 22:08:40< Soliton> or features simply not there.
20090909 22:08:43< cib0> true
20090909 22:09:00< shadowmaster> cib0: remember you have the power to fix stuff! (as long as you know what you are doing)
20090909 22:11:19< cib0> hm, did the 1.6 server crash? i'm getting timeouts.
20090909 22:11:50< Soliton> nope.
20090909 22:12:39< cib0> alright, works now
20090909 22:13:53-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 104 (Connection reset by peer)]
20090909 22:13:58< YogiHH> Soliton: ok, the WML sent to the server explains a lot :-P. There is quite some crap in it, no wonder the server chokes on that. *Sigh*, just this one time i hoped it wasn't my own fault ;-)
20090909 22:14:19< Soliton> :-)
20090909 22:14:36-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 22:15:19< fabi> cib0: No idea how to fix that.
20090909 22:17:50-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090909 22:19:49-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 22:23:19< lizard_r> boucman: about pixelmind's elipse - I mean, couldn't he use a 100x100 png for the eclipse like you can do with sprites? sprites will get centered, won't elipses get centered, too?
20090909 22:23:29-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090909 22:24:02< boucman> lizard_r: yes, that would work, i was more comenting on the code and what it does
20090909 22:24:49< lizard_r> do you think one can see that I meant that by reading my post?
20090909 22:25:03< boucman> lizard_r: yes, i'm dnot too worried
20090909 22:25:46-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 22:26:05-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090909 22:27:16< Soliton> loonycyborg: could the wesnoth_sources list not be generated by globbing and then files defined in the special lists be removed?
20090909 22:29:30-!- ShikadiLord [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090909 22:30:54-!- busfahrer` [n=busfahre@ulmg-5f70e87c.pool.mediaWays.net] has joined #wesnoth-dev
20090909 22:30:59-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 22:31:53-!- ShikadiLord [n=ignacio@wesnoth/developer/shadowmaster] has quit [Client Quit]
20090909 22:36:30-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 131 (Connection reset by peer)]
20090909 22:36:48-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 22:38:09-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090909 22:42:23-!- zookeeper [n=l@wesnoth/developer/zookeeper] has quit []
20090909 22:43:45-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 22:45:56-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 60 (Operation timed out)]
20090909 22:49:28-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 22:50:07-!- busfahrer` [n=busfahre@ulmg-5f70e87c.pool.mediaWays.net] has quit ["leaving"]
20090909 22:54:41-!- happygrue [n=George@wesnoth/developer/wintermute] has quit [Read error: 104 (Connection reset by peer)]
20090909 22:56:41< fabi> cib0: Is there another multiplayer campaign that could be tested?
20090909 22:56:55< cib0> fabi: world conquest
20090909 22:57:07< fabi> cib0: lets test it.
20090909 22:57:14< cib0> fabi: world conquest works well on 1.6
20090909 22:57:22< cib0> alright
20090909 22:57:32< cib0> not sure whether there is a dev version of it
20090909 22:57:40< Soliton> there is.
20090909 22:57:58< Soliton> not really maintained probably though.
20090909 22:59:01-!- stikonas_ [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 22:59:38-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 104 (Connection reset by peer)]
20090909 23:00:03< fabi> Soliton, cib0: It's not working. @ must be replaced with ~
20090909 23:01:03< Soliton> is that really mandatory?
20090909 23:01:32< Soliton> if support for @ really has already been removed that seems like a bad move..
20090909 23:02:45< fabi> Soliton: Looks like. I have replaced them and now it works. Let's test it.
20090909 23:03:20-!- Baufo [n=thomas@wesnoth/developer/baufo] has quit ["Leaving"]
20090909 23:04:33-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has quit ["night aall"]
20090909 23:05:12-!- stikonas_ [n=and@ctv-213-164-125-176.vinita.lt] has quit [Remote closed the connection]
20090909 23:05:47-!- stikonas [n=and@ctv-213-164-125-176.vinita.lt] has joined #wesnoth-dev
20090909 23:06:13< Soliton> indeed, seems like a poor change to me. *shrugs*
20090909 23:07:29-!- lizard_r [n=Miranda@wesnoth/umc-dev/developer/lizard] has quit [Read error: 110 (Connection timed out)]
20090909 23:09:17< shadowmaster> I can assure you I didn't do it. I didn't cheat in those silly Hurt and Heal threads using my sockpuppet either.
20090909 23:09:47< shadowmaster> (Although I may have proposed to deprecation of @ before)
20090909 23:11:32< Soliton> deprecation is fine but completely removing support for something that is used virtually everywhere... i sure hope the wiki contains no references to it anymore.
20090909 23:12:51< fabi> The @ is not the problem. It seems that multiplayer campaigns are broken in general.
20090909 23:13:32< shadowmaster> @ is not referenced anymore in /wiki/PreprocessorRef at least
20090909 23:14:47< Soliton> i'm not worried about the reference but all those Create pages...
20090909 23:17:49< shadowmaster> um, how do I search for "{~" or "{@" in the wiki? I can't believe there are no matches.
20090909 23:19:22-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 131 (Connection reset by peer)]
20090909 23:20:07< fabi> The campaign does work with local players only. But a remote player doesn't get the end scenario button.
20090909 23:20:17-!- alink [n=alink@wesnoth/developer/alink] has joined #wesnoth-dev
20090909 23:20:48< alink> hi
20090909 23:20:54< Crab_> hi alink
20090909 23:21:06< alink> 'lo Crab_
20090909 23:21:17< alink> wesbot log r35921
20090909 23:21:20< wesbot> ai0867 * r35921 : Deprecate '@' style includes, suggest usage of '~'.
20090909 23:21:22< shadowmaster> maybe I should work with the wiki more often so I could add an anti-MediaWiki rant to my blog.
20090909 23:21:23< wesbot> URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=35921
20090909 23:21:38-!- stikonas [n=and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20090909 23:21:52-!- Ivanovic changed the topic of #wesnoth-dev to: 1.6.5 released, announcing on Thursday | 92 bugs, 243 feature requests, 12 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090909 23:25:06< alink> Soliton: you seems to also have noticed the redundant mess of move_type, which is indeed a waste of memory and diskspace. However note that I think a unit may still have a different movetype than its unit_type (example: after picking artifact)
20090909 23:25:26< Soliton> sure.
20090909 23:26:09< Soliton> i've just refactored the movetype and unit specific functions to use the same internal function since the logic is almost the same.
20090909 23:26:30< alink> IIRC move_type WML data are copied 3 times: in unit, unit_type, and move_type
20090909 23:26:42< alink> Soliton: good
20090909 23:26:55< Soliton> unfortunately handing a reference for the movetype cache works for unit but for some reason not movetype...
20090909 23:27:11< Soliton> no idea why that is but i changed it to a pointer now.
20090909 23:27:53< Soliton> alink: well, movetype is the stuff in unit_types.cpp is there more?
20090909 23:28:34< Soliton> unit_movement_type to be precise.
20090909 23:28:51< alink> again IIRC I think there is pure move_type in memory, and unit_type include it in it's cfg, and unit do the same, but i check
20090909 23:29:03< Soliton> i don't mind the redundancy in the definition just the handling used duplicated code which is bad.
20090909 23:30:30< alink> yes i am mainly speaking about the redundancy in cfg (which end in memory and savegame)
20090909 23:30:58< Soliton> btw, i'm planning to remove all UNREACHABLE definitions from [movetype] so it uses the C++ defaults. any idea if that might cause problems somewhere?
20090909 23:31:33< Soliton> alink: for the savegame things should get unified, indeed.
20090909 23:31:35-!- DDR [n=chatzill@66.183.125.196] has joined #wesnoth-dev
20090909 23:31:52< shadowmaster> I was wondering whether the UNREACHABLE macro could be defined by the game engine instead, but if the defaults are equivalent...
20090909 23:31:58< Soliton> and in memory ideally as well, of course.
20090909 23:32:09< Soliton> shadowmaster: they are mostly.
20090909 23:32:27< Soliton> shadowmaster: the macro is still useful for WML that changes movetypes on the fly though.
20090909 23:32:50< Soliton> but it's pointless for all unit definitions.
20090909 23:32:50< alink> also each unit slowly fills its own little cache instead of copy an already filled one
20090909 23:33:10< Soliton> yes, that's stupid.
20090909 23:33:27< alink> Soliton: does WML can read move_type info ?
20090909 23:33:38< Soliton> at least for unit_types.
20090909 23:33:47< Soliton> for single units it has to be that way i guess.
20090909 23:34:17< Soliton> alink: also write for single units anyway.
20090909 23:34:50< alink> mmh about removing UNREACHABLE, you may want to check if info in help still list these terrain types 
20090909 23:35:00< Soliton> there are a couple of scenarios where units get changed to not be able to move on certain terrain they normally can.
20090909 23:35:12< Soliton> alink: yes, that's indeed what i checked.
20090909 23:35:40< Soliton> alink: it worked mostly fine before my refactoring but revealed that the duplicate code was not in sync anymore.
20090909 23:36:41< Soliton> bah, still shows all impassable and 0% defense in the help...
20090909 23:36:48< Soliton> it works correctly though.
20090909 23:37:33< Soliton> hmm...
20090909 23:38:07< alink> btw, I assume you kept the caching system for move_type (needed for pathfinding)
20090909 23:38:17< alink> i mean unit::move_type
20090909 23:38:39-!- YogiHH [n=chatzill@d040070.adsl.hansenet.de] has left #wesnoth-dev []
20090909 23:39:05< Soliton> i just unified the logic in one function and left the existing functions the same and just act as a wrapper.
20090909 23:39:15-!- DDR [n=chatzill@66.183.125.196] has quit [Remote closed the connection]
20090909 23:40:22-!- DDR [n=chatzill@66.183.125.196] has joined #wesnoth-dev
20090909 23:40:23< cib0> hm, lua support didnt make it into 1.6, right?
20090909 23:40:35< Soliton> that's how the movetype version looks now for example: http://nopaste.com/p/aoEt6vrxcb
20090909 23:41:27< Soliton> i don't really get why it doesn't work right.
20090909 23:41:47< Soliton> i suppose either the map or the moveCosts_ cache get lost somewhere...
20090909 23:42:05< CIA-62> silene * r38550 /trunk/src/scripting/lua.cpp: Tightened metatable names so that they can be used for dynamic typing in scripts.
20090909 23:42:08< CIA-62> silene * r38551 /trunk/src/scripting/lua.cpp: Fixed preallocation of metatables.
20090909 23:42:09< CIA-62> silene * r38552 /trunk/src/scripting/lua.cpp: Added Lua interface to vconfig objects.
20090909 23:42:09< alink> ah ok you pass the cache as parameter, good idea
20090909 23:42:11< CIA-62> silene * r38553 /trunk/src/scripting/lua.cpp: Changed parameters of action handlers from plain tables to vconfig objects.
20090909 23:42:15< CIA-62> silene * r38554 /trunk/src/scripting/lua.cpp: Generalized usage of vconfig objects.
20090909 23:42:18< CIA-62> silene * r38555 /trunk/src/scripting/lua.cpp: Allowed vconfig object inside WML tables.
20090909 23:42:21< CIA-62> silene * r38556 /trunk/data/lua/wml-tags.lua: Updated Lua code for [objectives] to vconfig interface.
20090909 23:42:24< CIA-62> silene * r38557 /trunk/data/scenario-test.cfg: Added check for proper delaying of substitutions in action handlers.
20090909 23:43:19< Soliton> yes, since the unified function is not a member of unit_movement_type or unit i had to pass in a bunch of member variables.
20090909 23:43:51< silene> cib0: 1.6 is a stable branch; there was no good reason for adding lua there
20090909 23:44:26< cib0> silene: well, i was thinking about creating a scenario.. and i wanted to use lua.. but i wanted it to be played, too, so... :p
20090909 23:44:51< alink> small perf remark, I believe that cfg_.child("movement_costs") cause an unwanted config search when the result is cached
20090909 23:45:13< Soliton> yes, how would i avoid it?
20090909 23:45:16< alink> if it's always in child "movement_cost" maybe just pass cfg_
20090909 23:45:25< Soliton> right, ok.
20090909 23:45:34< alink> and let the function search the child when needed
20090909 23:45:47< Soliton> yeah, silly me.
20090909 23:45:59< cib0> silene: did you get the message about the lua error we stumbled upon?
20090909 23:46:28< silene> cib0: no, which error?
20090909 23:46:46< Soliton> alink: any idea why the unit_movement_type part doesn't work right and only returns defaults?
20090909 23:46:56< cib0> silene: <shadowmaster> g20090909 15:57:07 error scripting/lua: /home/shadowm/src/wesnoth/data/lua/wml-tags.lua:165: bad argument #1 to 'gmatch' (string expected, got nil)
20090909 23:47:41< alink> Soliton: badly write in (or read from) the cache ?
20090909 23:47:55< shadowmaster> cib0: I suspect it's a quirk of the scenario anyway
20090909 23:47:56< silene> cib0: that means that someone is using the [clear_variable] tag, but without indicating the name of the variable to clear
20090909 23:48:02< shadowmaster> like ... exactly
20090909 23:48:09< alink> Soliton: possible test : deactivate caching 
20090909 23:48:22< Soliton> alink: i think i didn't change anything there but let me double check...
20090909 23:48:40< loonycyborg> Soliton: Yes. That is possible. Though it would be much nicer if the directory structure required only globs :)
20090909 23:48:48< cib0> silene: does that mean lua validates WML now?
20090909 23:48:57< shadowmaster> cib0: [clear_variaible] is implemented in lua
20090909 23:49:23< cib0> shadowmaster, whuh? interesing o.0
20090909 23:49:25< Soliton> loonycyborg: sure, but it's a lot more work...
20090909 23:49:28< cib0> *interesting
20090909 23:49:29 * shadowmaster lets cib0's mind spin
20090909 23:50:01< silene> cib0: a few wml tag are implemented in lua, either for exercising lua (e.g. [clear_variable]) or to provide new functionalities (e.g. [objectives])
20090909 23:50:04< cib0> let's port wesnoth to lua =D
20090909 23:50:31< cib0> silene: can you point me at the files for that?
20090909 23:50:46-!- lizard_r [n=Miranda@wesnoth/umc-dev/developer/lizard] has joined #wesnoth-dev
20090909 23:52:18< silene> cib0: as a matter of fact, it's the one in the error message ;-) data/lua/wml-tags.lua
20090909 23:52:51< fabi> silene: do you think that the lua error caused the oom in the multiplayer campaign?
20090909 23:53:09< silene> fabi: i doubt it; what kind of oom is it?
20090909 23:53:18< Soliton> alink: actually the logic can't be the issue since it's the same for unit and unit_movement_type.
20090909 23:53:42< fabi> silene: Not sure, cib0 got it. I finished the scenario and cib0 got the error and wasn't able to leave the first scenario.
20090909 23:53:56< Soliton> you probably mean OOS?
20090909 23:54:07< shadowmaster> 20090909 15:59:59 error replay: cannot recruit unit: You must have your leader on a keep to recruit or recall units
20090909 23:54:09< fabi> right
20090909 23:54:11< shadowmaster> that was it.
20090909 23:54:55< cib0> it was caused by the other bug which wouldnt allow me to proceed to the next scenario
20090909 23:54:57< Soliton> either that's not the first error or leaders got spawned on different locations.
20090909 23:54:59< cib0> unlikely it has to do with lua
20090909 23:55:51-!- lizard_r [n=Miranda@wesnoth/umc-dev/developer/lizard] has quit ["Saurian Augur - I'll heal you by 4 hp if you leave next to me"]
20090909 23:56:07< alink> Mmmh i see a lot of "movement_costs" WML key/tags and few "movement_cost" in unit.cpp. I wonder if it's normal
20090909 23:56:42< Soliton> movement_cost sounds wrong.
20090909 23:57:06 * fabi wonders what to do with the bug. I don't even saw it so I can't fill a proper bug report.
20090909 23:57:25< alink> it's in filter and cheksum code
20090909 23:57:26< Soliton> in get_checksum? yeah that looks like a typo.
20090909 23:58:20-!- elynia [n=shyde@wesnoth/umc-dev/misc/elynia] has joined #wesnoth-dev
20090909 23:58:52< Soliton> so you can cheat with movement_costs!
20090909 23:59:19< Soliton> though OOS will just happen later. ;-)
20090909 23:59:49< alink> plus i think people will notice these flying trolls :-)
--- Log closed Thu Sep 10 00:00:00 2009