--- Log opened Thu Aug 20 00:00:21 2009
--- Day changed Thu Aug 20 2009
20090820 00:00:21< Crab_> deekay: note that wolf is the most reliable fighter, with 4 attacks
20090820 00:01:12< shadowmaster> eett
20090820 00:01:19< Crab_> deekay: so, the alternate way is just the same ' hit fast & hard', just with more mixing of units ?
20090820 00:02:13< shadowmaster> ugh, my session kept bouncing around
20090820 00:02:15< deekay> Crab_: Price-wise lvl2 grunts could be useful too, dealing 12-3 at night (13 if strong)
20090820 00:02:33< deekay> Crab_: Poisoners could be nasty too
20090820 00:02:55< boucman> try poisoners and new AI poisoning, and see if it helps :)
20090820 00:03:17< deekay> Crab_: Plus.. getting only meele fighters and attaking with the elvish fighters... better attack fighters with archers
20090820 00:03:32< deekay> AI should try to get the least possible retaliation damage
20090820 00:03:40-!- ilor_ [n=user@wesnoth/developer/ilor] has quit [Read error: 113 (No route to host)]
20090820 00:03:40< deekay> Common sense of attacking :)
20090820 00:03:54< Crab_> deekay: ok
20090820 00:04:46< deekay> Crab_: And what is important..
20090820 00:05:01< deekay> If you mix units, AI may be too fast with some, too slow with other
20090820 00:05:01-!- Chusslove [n=Chusslov@212.124.171.92] has quit [Read error: 131 (Connection reset by peer)]
20090820 00:05:25< deekay> So instead of "hit hard" we get few "just hit" with not too high number of units
20090820 00:05:48< Crab_> deekay: well, this can be (partially) fixed 
20090820 00:05:56< deekay> Would be nice
20090820 00:06:11< corn> boucman, Crab_: are there any automated tools for removing redundant includes
20090820 00:06:28< corn> something like that would be very useful for speeding up compile times
20090820 00:07:30-!- kitty_ [n=kitty@wesnoth/artist/kitty] has quit ["good night"]
20090820 00:08:18< Crab_> corn: if (rd.child("foo")) can be used instead of if (rd.child_count("foo"))
20090820 00:08:22< boucman> would recruiting slow units the first turn and faster units on the second help ?
20090820 00:08:39< deekay> With map size? barely
20090820 00:08:46< silene> corn: redundant (but used) includes have a negligible cost
20090820 00:09:03< Crab_> boucman: depends on speed difference
20090820 00:09:10-!- Appleman1234 [n=Appleman@131.181.102.164] has quit [Remote closed the connection]
20090820 00:09:28< deekay> Slowest are grunts and archers
20090820 00:09:30< deekay> Then comes assassins
20090820 00:09:33< deekay> Then wolves
20090820 00:09:35< Crab_> corn: in ai_testing::log_victory, you do it a bit wrongly. you need to output *w, as well
20090820 00:09:38< corn> silene: is there a way to remove unused includes?
20090820 00:09:54< boucman> Crab_: recruit slow units on turn one, and fast on turn 3/4/whaterver timing difference is needed to hae them arrive together
20090820 00:11:06< silene> corn: not that i know of
20090820 00:11:09< boucman> good night all
20090820 00:11:12< corn> ok
20090820 00:11:14< corn> night
20090820 00:11:19-!- boucman [n=rosen@wesnoth/developer/boucman] has quit ["Leaving."]
20090820 00:11:22-!- ShikadiLord [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090820 00:11:55-!- ShikadiLord [n=ignacio@wesnoth/developer/shadowmaster] has quit [Client Quit]
20090820 00:12:31< Crab_> deekay: there are only 2 turns of recruitment. so, what do you recommend ?
20090820 00:12:47< Crab_> deekay: grunts/archers/assasins on turn 1, wolves on turn 2 ?
20090820 00:13:00< corn> Crab_: ok, I made your suggested changes, recompiling and testing
20090820 00:13:03< deekay> I'd try that
20090820 00:13:19< deekay> And make sure they arrive just before the night
20090820 00:13:28-!- ShikadiLord [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090820 00:14:11< corn> if all goes well I will add tables for this data later tonight
20090820 00:14:11< Crab_> deekay: well, we can't wait - each turn is +2 or +3 pesky elves :)
20090820 00:14:29-!- shadowmaster [n=ignacio@wesnoth/developer/shadowmaster] has quit [SendQ exceeded]
20090820 00:14:33< deekay> Crab_: Then go for village advantage first
20090820 00:14:34-!- shadowmaster [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090820 00:14:52-!- ShikadiLord [n=ignacio@wesnoth/developer/shadowmaster] has quit [Nick collision from services.]
20090820 00:15:02-!- ShikadiLord [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090820 00:15:02-!- ShikadiLord [n=ignacio@wesnoth/developer/shadowmaster] has quit [Client Quit]
20090820 00:15:10< deekay> Crab_: It is player's problem if they won't kill orcs leaders before 25th turn
20090820 00:15:14< Crab_> deekay: well, with 960 starting gold, we go to negative gold for first 10 turns or so, after we recruit
20090820 00:15:24-!- shadowmaster [n=ignacio@wesnoth/developer/shadowmaster] has quit [Client Quit]
20090820 00:15:32< Crab_> deekay: so, what would be best 'let's camp until turn 25' strategy ?
20090820 00:15:38< Crab_> deekay: for those orcs
20090820 00:15:40-!- shadowmaster [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090820 00:15:51< Crab_> deekay: retreat the green player to the other side of the river ?
20090820 00:15:53< deekay> That would be best, but does not makes sense if we look at them as "attackers"
20090820 00:16:09< deekay> Hence my question from the beginning :)
20090820 00:16:11< Crab_> deekay: well, they are an advance force which is waiting for reinforsements
20090820 00:16:23< deekay> Ok, then it makes perfect sense
20090820 00:16:33< deekay> Seize some villages and scout the area
20090820 00:16:49-!- Aethaeryn [n=Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #Wesnoth-dev
20090820 00:16:54< Crab_> (10:05:23 PM) fendrin: Crab_: yes, they have reenforcements arriving later. That is the reason given to the player when turns run out. They should know that.
20090820 00:17:21< Crab_> deekay: what should the green player do in that case ? retreat to other side of the river or camp in the mountains ?
20090820 00:18:02< Crab_> deekay: note that those elves have terrain advantage over all the villages
20090820 00:18:18< deekay> Over the map you meant :)
20090820 00:18:25< deekay> Crab_: But look
20090820 00:18:46< deekay> orc in forest vs elf in forest is better than orc in water vs elf in castle
20090820 00:18:47< Crab_> deekay: I meant over 'all the villages' - there's a part of map which is water/mountains/hills - but there's no villages there
20090820 00:18:51< deekay> From AI point of view
20090820 00:18:51-!- silene [n=plouf@wesnoth/developer/silene] has quit ["Leaving."]
20090820 00:19:04< Crab_> deekay: orc in water vs elf in water is even better :))
20090820 00:19:14< shadowmaster> dammit I missed silene. I was right, my day _could_be worse
20090820 00:19:27< deekay> Crab_: Sadly, unexpected with experienced player ;)
20090820 00:20:01< Crab_> deekay: yes
20090820 00:20:23< Crab_> so, 'join forces and camp in the forest, while trying to get all those villages in the south' ?
20090820 00:20:27< deekay> Crab_: Looking from this POV..
20090820 00:20:32< deekay> Crab_: I am orc
20090820 00:20:36 * deekay becomes ugle
20090820 00:20:41< deekay> ugly even
20090820 00:20:46< fendrin> Fast food isn't even fast anymore.
20090820 00:20:47< grzywacz> Even more ugly?
20090820 00:20:54< deekay> grzywacz: Even more
20090820 00:20:57< grzywacz> :X
20090820 00:21:14 * deekay stares at grzywacz with his ugly orcish eyes
20090820 00:21:29< deekay> Now.. I have my ugly friends coming over to me soon
20090820 00:21:43-!- melinath [n=melinath@pool-72-72-200-231.altnpa.east.verizon.net] has quit [Read error: 113 (No route to host)]
20090820 00:21:47< deekay> There is load of funny looking green tree-lovers ahead of me
20090820 00:22:08< Crab_> deekay: not-so-funny once they start to shoot at you :)
20090820 00:22:12< deekay> And these tree lovers seem to be strong
20090820 00:22:23< fendrin> deekay: tweaking the time of day is a good idea.
20090820 00:22:24< deekay> I am orc, but not a troll
20090820 00:22:35< Soliton> hmm, 100% CPU usage in a story screen.
20090820 00:22:37< deekay> So I won't suicide myself attaking from a water
20090820 00:23:11< deekay> I need to face these fragile but flexible bastards in a smart way
20090820 00:23:34< Crab_> deekay: and even worse, they have some unexpected help (kalenz & company)
20090820 00:23:35< deekay> So I just go for nearest villages, and try to gain advantage over some part of a map
20090820 00:23:45< shadowmaster> Soliton: any one in particular?
20090820 00:23:58< shadowmaster> HttT 1 borders 1% after finishing the effects
20090820 00:24:55< deekay> Crab_: Thing is to not make AI suicide
20090820 00:25:22< deekay> Crab_: Human player expects AI to just kill itself by banging hard with a head into wall of his units
20090820 00:25:35< Soliton> shadowmaster: trow:14 it was busy waiting on something.
20090820 00:25:35< deekay> Crab_: Careful AI may be nasty surprise
20090820 00:25:42< Crab_> deekay: well, it shouldn't suicide if it isn't sure that it's worth it :)
20090820 00:25:44< Soliton> strace flooded with: select(4, [3], NULL, NULL, {0, 0})      = 0 (Timeout)
20090820 00:25:44< Soliton> read(3, 0x128b764, 4096)                = -1 EAGAIN (Resource temporarily unavailable)
20090820 00:25:46< shadowmaster> um, I see 100% on one CPU at the final storyscreen of HttT 1
20090820 00:25:55-!- Appleman1234 [n=Appleman@131.181.102.164] has joined #wesnoth-dev
20090820 00:26:18< Crab_> deekay: but, for more concrete questions... what should the green orc do ?
20090820 00:26:30< deekay> Crab_: Sec, let me look at a map
20090820 00:26:31< Crab_> deekay: in case that we want to hold for 25 turns
20090820 00:26:40< Crab_> on hard, that is
20090820 00:26:55< shadowmaster> read on fd 3? um, isn't that one of the standard file descriptors?
20090820 00:27:15< deekay> Crab_: How much gold he has?
20090820 00:27:20< grzywacz> no
20090820 00:27:22< Soliton> no. there is only 0,1,2
20090820 00:27:35< Crab_> deekay: green orc (north): 460
20090820 00:27:45< Crab_> deekay: purple orc (south): 500
20090820 00:27:46< shadowmaster> oh, 0 is a valid fd :)
20090820 00:27:54< grzywacz> Soliton, /proc/PID/fd and ls -l please
20090820 00:28:08< deekay> Crab_: 5 trolls on hills and wait for opportunity to strike forest
20090820 00:28:17< deekay> Crab_: Some mix of units
20090820 00:28:24< Crab_> kalenz: >=180, elf:=220 (+35 income)
20090820 00:28:27< Soliton> i moved on but i can check now.
20090820 00:28:38< deekay> Crab_: And never hold villages with wolves, only 50% def
20090820 00:28:53< deekay> Crab_: Of course lvl2 trolls
20090820 00:29:04< Soliton> 3 -> socket:[7979050]
20090820 00:29:08< shadowmaster> grzywacz: for me it seems to be an unix socket
20090820 00:29:11< shadowmaster> er, right
20090820 00:29:14< grzywacz> X?
20090820 00:29:39 * shadowmaster asks X
20090820 00:29:45< grzywacz> nestat -apn will tell you
20090820 00:29:52< grzywacz> netstat even
20090820 00:30:08< Crab_> fendrin: ^ those orcs in LoW3 have no trolls in recruitment lists. is it a lore reason / balance reason / 'no one thought about it' reason ?
20090820 00:30:16< Soliton> yeah, X.
20090820 00:30:28< grzywacz> So waiting on input, I guess?
20090820 00:32:04< deekay> Crab_: If AI could lure player into hills...
20090820 00:32:21< fendrin> Crab_: It's because the player is to be confronted with different enemy through the campaign. The first battle at the kalian only includes orcs the second one is against trolls and orcs.
20090820 00:32:34< shadowmaster> the floating images rendering is the issue
20090820 00:32:43< Crab_> deekay: no, i think there's absolutely no reason to venture into hills, for the player...
20090820 00:32:47< shadowmaster> I'm still checking what's wrong with it
20090820 00:32:51< deekay> Crab_: Kill enemy leader?
20090820 00:32:58< Crab_> deekay: except to kill enemy leader, of course :)
20090820 00:33:02< deekay> :)
20090820 00:33:32< deekay> Crab_: AI could "sacrifice some units" and one they are deafeted move bac to hills
20090820 00:33:34< Crab_> deekay: note that if AI gives up map control then, those elves will have (on turn 20), ~1000 gold combined
20090820 00:33:43< deekay> It would be visible by a player as "cleaver retreat"
20090820 00:33:51< deekay> s/visible/seen/
20090820 00:33:59< Crab_> deekay: so, it would be seen by a player as a cheesy retreat :)
20090820 00:34:03< Crab_> deekay: ok, will try :)
20090820 00:34:12< deekay> It may be worth it
20090820 00:34:44< deekay> Crab_: No clue it will really work best, just brainstorming :)
20090820 00:34:51< Crab_> deekay: we can switch strategies depending on initial evaluation of kalenz power
20090820 00:35:00< deekay> yea
20090820 00:35:12< shadowmaster> sssssdddddddddscons -j3
20090820 00:35:21-!- Valkier [n=karasawa@c-71-225-240-156.hsd1.pa.comcast.net] has joined #Wesnoth-dev
20090820 00:37:24< shadowmaster> it's not sleeping after redering the image floating r9
20090820 00:37:49< shadowmaster> it's not sleeping after rendering the floating images , there's broken logi in there
20090820 00:38:59< fendrin> Crab_: LoW's story tells us that the Kalian is under siege for some time. That does make sense since Kalenz is arriving after a longer travelling.
20090820 00:39:36< fendrin> But at the time it was coded that wasn't easy todo. But today it could be done.
20090820 00:40:00< shadowmaster> nah, broken logic my foot. It's just missing a line
20090820 00:40:36< Crab_> fendrin: and what do you think about 'hold position in hills and wait for the reinforcements to come' idea ?
20090820 00:41:13< fendrin> Crab_: That is a good idea for scenario 7 at the same map.
20090820 00:41:36< fendrin> Crab_: That one should stay a smaller battle.
20090820 00:42:37< fendrin> More foresthills or pure hill tiles can be spread over the map to support that orc strategy there. (The map is bigger but mostly the same location)
20090820 00:43:25< Crab_> fendrin: well, 'that orc strategy' is doable on this map, as well. just 'do not attack at all, move onto hills, dig in, and let the elves come'
20090820 00:44:10< Crab_> there's enough hills near the green orc to make it work, (in theory)
20090820 00:44:19< shadowmaster> why did a new portrait have to be committed _this_ _afternoon_ ;(
20090820 00:45:43< Crab_> fendrin: ok, with LoW 3, we'll try some more orc units combinations, and we'll try that 'hold out' strategy (just to see how it works)
20090820 00:46:04< shadowmaster> Soliton: fixed now, thanks for the report
20090820 00:46:33< fendrin> Crab_: Yes, all your experiments will fit at some place in nearly every campaign. So no work will be lost.
20090820 00:46:58< Soliton> shadowmaster: thanks for fixing it!
20090820 00:46:59< Crab_> fendrin: so, be prepared to tell 'what's next' area of experimentation :)
20090820 00:48:00-!- Sirp [n=user@wesnoth/developer/dave] has joined #wesnoth-dev
20090820 00:48:04< fendrin> Crab_: Is there a way to simulate an ongoing battle? Letting the cpu do some turns in quiet and then start the scenario with the player.
20090820 00:48:32< Crab_> fendrin: yes, it's possible (with engine changes, that is)
20090820 00:49:40< Crab_> fendrin: should be used with care :) e.g., imagine a "you lost" message because some important unit died before the player has got even a chance to act :)
20090820 00:50:44< fendrin> Crab_: The story would be: The Kalian is under siege by all kind of orcs and goblins. Some mechanismen must prevent one side to kill the other. At the moment Kalenz arrives the orc commander of the orcs orders the wulf rider into the battle.
20090820 00:51:42< Crab_> fendrin: the problem is not only preventing one side from killing another, but preventing an 'too big' advantage for one of the sides
20090820 00:51:58< fendrin> Crab_: Right, the simulation would need to be stoped after a certain amount of units lost by one of the two sides.
20090820 00:52:15< Crab_> fendrin: doable, through
20090820 00:52:53< fendrin> That would also enhance the replay value of that scenario since it should generate different starting positions every time.
20090820 00:53:08< Crab_> yes
20090820 00:54:05< Crab_> fendrin: note that if we add some more hills/mountains near the orc starting positions, we can either switch to a 'full assault' or a 'hold out till reinforcements' depending on combat outcome
20090820 00:54:34< Crab_> fendrin: imagine a player controlling not kalenz, but a subset of the citadel's defenders
20090820 00:54:38< fendrin> Crab_: Yeah, right. That would be awesome.
20090820 00:54:40< Crab_> fendrin: for 3-5 turns
20090820 00:55:01< Crab_> fendrin: and then, kalenz arrives, and the player starts to control kalenz
20090820 00:55:15< fendrin> Crab_: That would be doable without the need to hack the engine.
20090820 00:55:16< Crab_> fendrin: note that there will not be need for engine changes this way
20090820 00:55:16-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090820 00:55:19< Crab_> yeas
20090820 00:55:37-!- ettin [n=jorda@wesnoth/developer/ettin] has quit [Connection timed out]
20090820 00:56:33< deekay> night everyone
20090820 00:56:38< Crab_> deekay: night
20090820 00:57:04< CIA-62> cornmander * r37958 /trunk/src/ (ai/testing.cpp game.cpp replay.cpp replay.hpp upload_log.cpp): 
20090820 00:57:04< CIA-62> Changed upload_log to record data into replays under the [upload_log] tag. Log
20090820 00:57:04< CIA-62> events can be added throughout the code at all times by using the new
20090820 00:57:04< CIA-62> replay::add_log_data() functions. Changed the --label code to use this new
20090820 00:57:04< CIA-62> functionaity. Added this functionality to src/ai/testing.cpp under the [ai_log]
20090820 00:57:08< CIA-62> category.
20090820 00:57:10< CIA-62> cornmander * r37959 /trunk/src/upload_log.cpp: Reenabled upload log deletion - accidentally disabled in last commit.
20090820 00:57:13< CIA-62> shadowmaster * r37957 /trunk/ (changelog src/storyscreen/render.cpp): Fix a 100% CPU usage on storyscreens without text. Reported by Soliton.
20090820 00:59:44< fendrin> Crab_: Okay, I will do that to scenario 3. Another task would be the sea creatures in scenario. The problem is: If the aggression and caution is to low/high they don't attack. If it is to high they come on Land. I could manage that with avoid tags it it can't be adjusted.
20090820 01:00:01< fendrin> ^scenario 9
20090820 01:00:34< Crab_> ok, doing :cl 9
20090820 01:01:40< fendrin> Crab_: They only appear if you come near the water.
20090820 01:01:47< Crab_> fendrin: I know.
20090820 01:02:04< Crab_> fendrin: since [avoid] accepts a SLF now, it seems trivial to use it to limit their appetite
20090820 01:02:20< Crab_> fendrin: I should code it using a new syntax, to show how it works
20090820 01:02:59< Crab_> fendrin: or, if you want them to come on land if the target is, say, exceptionally tasty, this can be done as well
20090820 01:03:15< Crab_> fendrin: we can even make them stalk for Kalenz or Olurf :)
20090820 01:03:33-!- melinath [n=melinath@pool-72-72-200-231.altnpa.east.verizon.net] has joined #wesnoth-dev
20090820 01:03:40< Crab_> fendrin: i.e., hide in the water and see if they can specifically eat the elf.
20090820 01:04:45-!- grzywacz [n=grzywacz@wesnoth/developer/grzywacz] has quit [Remote closed the connection]
20090820 01:05:32< corn> Crab_: how much longer will you be up today? I think I can finish everything to start recording the new [ai_log] data in a mysql table within an hour
20090820 01:05:52< Crab_> corn: i'm going to sleep quite soon
20090820 01:06:02< corn> ok, then we will work on it tomorrow
20090820 01:06:24< fendrin> Crab_: They would come on land to achieve an easy kill and retreat afterwards. But they should go after the leaders. How should they know that these landcreatures are more important than others.
20090820 01:06:39< fendrin> s/should/shouldn't
20090820 01:16:15-!- Puma [n=karasawa@c-71-225-240-156.hsd1.pa.comcast.net] has joined #Wesnoth-dev
20090820 01:16:29-!- Blueblaze [n=Blueblaz@166.189.219.234] has joined #wesnoth-dev
20090820 01:19:45-!- noy [n=Noy@wesnoth/developer/noy] has quit [Connection timed out]
20090820 01:20:39-!- BenUrban [n=benurban@c-68-50-54-86.hsd1.md.comcast.net] has joined #wesnoth-dev
20090820 01:21:26< Crab_> fendrin: ok. can we add 'submerge' ability to that serpents ?
20090820 01:21:30< Crab_> fendrin: i think that those serpents are best controller by some specific candidate action(s)
20090820 01:21:31-!- xchat_sux [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090820 01:21:32< Crab_> fendrin: to achieve that 'hit & run' behavior
20090820 01:21:33< corn> Crab_: here is my current table structure for ai logs: http://cornmander.com/tbl.html
20090820 01:21:52< Crab_> fendrin: so, you should describe how do you want them to behave. 'They would come on land to achieve an easy kill and retreat afterwards.'. Should they only attack weak creatures ?
20090820 01:22:53< Crab_> corn: where is ai label ?
20090820 01:23:01< corn> the winner column will store the faction name, it will either == faction1 || == faction2
20090820 01:23:07< corn> oh, forgot about that :)
20090820 01:23:35< corn> added
20090820 01:23:43< Crab_> what is 'gold' ?
20090820 01:23:56< corn> the amount of gold the winner has in the end
20090820 01:24:02< Crab_> what is end_gold ?
20090820 01:24:06< corn> oh, it is redundant
20090820 01:24:38< corn> ok. I will remove 'gold'
20090820 01:24:46< Crab_> corn: you should also store the number of units *loser* has at the end
20090820 01:25:03< corn> ok
20090820 01:25:09< Crab_> corn: it's important for spotting suicide leader moves 
20090820 01:25:23< corn> ok, I will store both number of units for winner and loser
20090820 01:25:35< Crab_> ok
20090820 01:26:10< Crab_> ai_config1, ai_config2 can be more than 32 chars long
20090820 01:26:22< corn> is 64 enough?
20090820 01:26:27< Crab_> ai_ident1, ai_ident2 - too :)
20090820 01:26:52< corn> varchar stores these as variable length - so if the actual length is 0, it will not use extra space
20090820 01:26:55-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 60 (Operation timed out)]
20090820 01:27:01< Crab_> corn: imagine: ai/dev/special/test_recruitment_formula/new_recruitment_formula_config.cfg
20090820 01:27:07-!- melinath [n=melinath@pool-72-72-200-231.altnpa.east.verizon.net] has quit [Read error: 113 (No route to host)]
20090820 01:27:14< corn> ouch
20090820 01:27:14< corn> ok
20090820 01:27:22< corn> I will make it 256 :)
20090820 01:27:51< Crab_> for example, now we have 'ai/dev/formula_ai_poisoning.cfg'
20090820 01:28:04-!- Valkier [n=karasawa@c-71-225-240-156.hsd1.pa.comcast.net] has quit [Read error: 110 (Connection timed out)]
20090820 01:28:22< Puma> Oh noes!
20090820 01:28:25-!- Puma is now known as Valkier
20090820 01:29:05< Crab_> corn: also what do you want to long in 'end_gold' during a draw ?
20090820 01:29:12< Crab_> s/long/log
20090820 01:29:44< corn> nothing, it will be 0 and I will just have logic to ignore it if there is a draw
20090820 01:29:48< Crab_> also, why do you want to store  the faction name in winner column ?
20090820 01:30:03< Crab_> drakes (default_ai) vs drakes (formula_ai) is a viable matchup
20090820 01:30:05< corn> updated table format, refresh page
20090820 01:30:31< Valkier> So in 1.7.3 the AI was beefed up a bit?
20090820 01:30:33< Crab_> label -> make 'size' bigger
20090820 01:30:45< corn> oh, I will store it as an enum of {0,1} then
20090820 01:31:13< Crab_> Valkier: no, but the tools we have to develop better AIs have improved a lot
20090820 01:31:28< Valkier> ooo. So in the future then?
20090820 01:31:29< Crab_> Valkier: so, expect some improvements for 1.7.4
20090820 01:31:34< Valkier> Rock on
20090820 01:31:43< Valkier> The AI right now kind of makes me sad to play alone.
20090820 01:32:17< corn> Crab_: ok, I changed label size to 256 as well
20090820 01:32:20< Crab_> Valkier: you can watch the orc ai in ftp://ftp.terraninfo.net/wesnoth/LOW_3_1.gz for an example of what we're trying
20090820 01:32:40< corn> any other suggestions for what I should store?
20090820 01:33:25< Crab_> corn: what's int(2) ?
20090820 01:33:28< Crab_> 0..255 ?
20090820 01:33:53< corn> no
20090820 01:33:55< corn> int 0-99
20090820 01:34:07< corn> maybe -99-99
20090820 01:34:10< Crab_> well, 100 units are definitely possible :)
20090820 01:34:12< fendrin> Crab_: They stay in water and don't attack to the land (because the units would have a good defense there). So they only attack units in water. Except a unit that can be killed in one turn.
20090820 01:34:41< Crab_> fendrin: do they fear death ?
20090820 01:34:41< fendrin> High percentage to be killed in one turn.
20090820 01:35:02< Crab_> fendrin: do they retreat if they're eaten enough ? what's their agenda ?
20090820 01:35:03< corn> Crab_: ok, is -999-999 ok?
20090820 01:35:08< Crab_> corn: should be
20090820 01:35:12< corn> ok
20090820 01:35:59< fendrin> Crab_: Good question. They fear the death and retreat when eaten enough. That is a good idea. 
20090820 01:36:22-!- Blarumyrran [n=minaise@81-20-159-197.levira.ee] has quit []
20090820 01:36:40< Crab_> fendrin: so, they want to take advantage of the situation and eat as much as possible but without taking excessive risks ?
20090820 01:37:23< fendrin> Crab_: Yes. 
20090820 01:37:44< Crab_> ok. so, this is best handled by specific candidate action for them
20090820 01:38:34< Crab_> their AI would include a single stage, candidate action evaluation loop, with some candidate actions. Or, alternatively, this can be done as a unit formulas stage (similar to patrolling wolves)
20090820 01:38:40 * Crab_ puts this in @todo list
20090820 01:38:42-!- allefant [n=allefant@allegro/developer/allefant] has quit [Success]
20090820 01:38:53< Crab_> (engine support is ready, only need to write some formulas)
20090820 01:39:14-!- allefant [n=allefant@allegro/developer/allefant] has joined #wesnoth-dev
20090820 01:39:54< fendrin> Crab_: The last scenario is a most unsatisfying one. Half the army is on Landars side so the player is fighting his former units.
20090820 01:40:23-!- Blarumyrran [n=minaise@81-20-159-197.levira.ee] has joined #wesnoth-dev
20090820 01:40:34< fendrin> But tuning the ai to do a good defense on that map isn't easy.
20090820 01:40:48< fendrin> And the attack hasn't work either.
20090820 01:40:58-!- loonycyborg [n=sergey@wesnoth/developer/loonycyborg] has quit ["Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz"]
20090820 01:41:28-!- loonybot [n=loonybot@wesnoth/bot/loonybot] has quit [Remote closed the connection]
20090820 01:41:31< fendrin> It's a little bit hard to test because you need a savegame in which the players army is stored in an array.
20090820 01:41:57< Crab_> fendrin: can you provide such a save ?
20090820 01:42:19< Crab_> fendrin: ( i can hot-redeploy an AI from file or just edit the save to try changes)
20090820 01:42:42< fendrin> hot redeploy from file?
20090820 01:42:47< Crab_> fendrin: yes
20090820 01:45:28< fendrin> Crab_: I have a savegame from scenario 14
20090820 01:45:37< fendrin> That could be used
20090820 01:45:56< fendrin> ?
20090820 01:46:07< Crab_> fendrin: and which scenario is 'The last scenario ' ?
20090820 01:46:14< fendrin> 23
20090820 01:46:36< Crab_> fendrin: on hard ?
20090820 01:47:55< fendrin> only on easy
20090820 01:48:21< Crab_> fendrin: well, I prefer to modify 'hard' difficulty :) so, I guess I'll need to add a little debug command to make things easier
20090820 01:48:30< Crab_> fendrin: e.g., a right-click command to kill a unit
20090820 01:49:06< Crab_> fendrin: this should allow me to advance from scenario 1 to 23 quickly
20090820 01:49:25< Crab_> fendrin: ok, I'll go to sleep now, will return in ~8 h
20090820 01:49:35< fendrin> Crab_: Good night:-)
20090820 01:49:43-!- Crab_ [n=Crab_@wesnoth/developer/crab] has quit ["Leaving."]
20090820 01:51:24-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 110 (Connection timed out)]
20090820 01:52:53-!- Blueblaze [n=Blueblaz@166.189.219.234] has quit []
20090820 02:15:57-!- Blueblaze [n=Blueblaz@166.189.219.234] has joined #wesnoth-dev
20090820 02:46:59-!- noy [n=Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090820 02:49:57-!- Appleman1234 [n=Appleman@131.181.102.164] has quit [Read error: 145 (Connection timed out)]
20090820 02:50:06-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Connection reset by peer]
20090820 02:54:12-!- ABCD [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090820 03:04:47-!- Blarumyrran [n=minaise@81-20-159-197.levira.ee] has quit []
20090820 03:09:21-!- melinath [n=melinath@pool-72-72-200-231.altnpa.east.verizon.net] has joined #wesnoth-dev
20090820 03:13:38-!- shadowmaster is now known as irmorelle
20090820 03:17:50-!- krzysz00` [n=user@pool-71-252-228-249.dllstx.fios.verizon.net] has joined #wesnoth-dev
20090820 03:18:06< krzysz00`> us there a configure flag that disables the generaton of documentation
20090820 03:19:16-!- irmorelle is now known as Shadow_Master
20090820 03:19:23< krzysz00`> s/us/is
20090820 03:19:56< krzysz00`> can i disable documentation in ./configure? just wondering?
20090820 03:20:41-!- krzysz00 [n=user@pool-71-252-228-249.dllstx.fios.verizon.net] has quit [Remote closed the connection]
20090820 03:21:47< Shadow_Master> just wondering, can I get that secret ability that turns people in a 1-hex radius around me more patient?
20090820 03:22:39-!- Appleman1234 [n=Appleman@131.181.103.31] has joined #wesnoth-dev
20090820 03:23:19< krzysz00`> sorry
20090820 03:25:15< Shadow_Master> krzysz00`: apparently not.
20090820 03:25:18< Soliton> anything that is not in ./configure --help is not there.
20090820 03:26:22< Shadow_Master> you can still disalble generation of translations if it helps 
20090820 03:32:33< krzysz00`> thanks trying to scale down a nokia n800 port allready rm music and sounds, i might put the sounds back later, but in the meantime keep your eyes out for nokia n800 wesnoth port 0.0.1
20090820 03:40:30-!- Appleman1234 [n=Appleman@131.181.103.31] has quit [Read error: 145 (Connection timed out)]
20090820 03:42:01< CIA-62> jetryl * r37960 /trunk/data/core/about.cfg: Slight update to credits.
20090820 03:44:19< Shadow_Master> git show -p HEAD
20090820 03:44:32< Shadow_Master> bah.
20090820 03:50:46-!- Appleman1234 [n=Appleman@131.181.103.31] has joined #wesnoth-dev
20090820 04:00:23-!- cib0 [n=cib@p5DC43427.dip.t-dialin.net] has quit [Remote closed the connection]
20090820 04:16:41-!- Ivanovic_ [n=ivanovic@dtmd-4db222e0.pool.einsundeins.de] has joined #wesnoth-dev
20090820 04:29:01-!- maackey [n=chris@pool-72-91-157-206.tampfl.fios.verizon.net] has joined #wesnoth-dev
20090820 04:29:29-!- ShikadiLord [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090820 04:29:51< maackey> i've a question: is it possible to have a unit advance to another *only* if they have some specially defined trait(s)?
20090820 04:30:51-!- ShikadiLord [n=ignacio@wesnoth/developer/shadowmaster] has quit [Nick collision from services.]
20090820 04:32:42-!- Ivanovic [n=ivanovic@wesnoth/developer/ivanovic] has quit [Read error: 110 (Connection timed out)]
20090820 04:33:04< maackey> nvm
20090820 04:33:51-!- Shadow_Master is now known as shadowmaster
20090820 04:34:39-!- Ivanovic_ is now known as Ivanovic
20090820 04:36:39-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090820 04:38:03-!- ABCD [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090820 04:42:20-!- Valkier [n=karasawa@c-71-225-240-156.hsd1.pa.comcast.net] has quit []
20090820 04:43:32< CIA-62> cornmander * r37961 /trunk/data/core/about.cfg: Updated my entry in about.cfg.
20090820 04:51:54< corn> question: what is the different between putting const before and after the type?
20090820 04:53:05< shadowmaster> corn: I suggest you take a look at http://www.parashift.com/c++-faq-lite/const-correctness.html#faq-18.5 and other FAQs in that same page
20090820 04:53:13< corn> ok
20090820 04:53:24< shadowmaster> it will look confusing at the beginning ;)
20090820 04:54:27< shadowmaster> const type_name& foo   and type_name const& foo are both the same.
20090820 04:55:19< corn> ok, it appears both ways in wesnoth code so I wasn't sure
20090820 04:55:38< Sirp> corn: our standard is const T& foo
20090820 04:55:44< Sirp> I don't think we use T const& foo often
20090820 04:55:56< shadowmaster> Sirp: some ancient code has the latter
20090820 04:56:12< corn> sdl_utils.?pp
20090820 04:56:17< Sirp> shadowmaster: really? I'd have thought it'd be the more modern code
20090820 04:57:47< shadowmaster> Sirp: serialization/string_utils.hpp is rt~t
20090820 04:57:53< shadowmaster> er, using it, for instance.
20090820 04:58:27-!- krzysz00` [n=user@pool-71-252-228-249.dllstx.fios.verizon.net] has quit [Read error: 113 (No route to host)]
20090820 04:58:36< Sirp> ahhh yes it does in some places
20090820 04:58:47< corn> sdl_utils.?pp actually mixes the conventions. should I change it to be consistent?
20090820 04:59:30< Sirp> corn: that would be great.
20090820 04:59:42< corn> ex stretch_surface_vertical(const surface& surf ...) and scale_surface(surface const &surf ...)
20090820 04:59:54< corn> ok
20090820 05:01:08< shadowmaster> the typename &identifier  format looks to me rather ugly
20090820 05:01:25< corn> shadowmaster: I personally like it that way
20090820 05:01:50< shadowmaster> I prefer typename& identifier to make it artificailly different to the address-of operator
20090820 05:02:18< shadowmaster> but that's just me :)
20090820 05:03:28< Sirp> I prefer T& id and T* id
20090820 05:03:48< Sirp> since I see the & and * as conceptually part of the type
20090820 05:05:11< shadowmaster> I think the T* id convention doesn't make  it possible to declare multiple pointers in a  one liner as in T *ptr1, *ptr2; ?
20090820 05:06:01< shadowmaster> I mean, T* ptr1, ptr2; will declare a pointer and an object IIRC
20090820 05:21:57-!- melinath [n=melinath@pool-72-72-200-231.altnpa.east.verizon.net] has quit [Read error: 113 (No route to host)]
20090820 05:57:29< CIA-62> cornmander * r37962 /trunk/src/ (sdl_utils.cpp sdl_utils.hpp): Changed all surface const &var to const surface &var. The syntax was mixed before.
20090820 05:59:16< corn> recompilation is extremely painful on my current box... it takes me 15-30 mins to test a change if there is a change in a header file
20090820 05:59:58-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090820 06:00:18-!- ABCD [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090820 06:29:29-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has quit []
20090820 06:38:48-!- ancestral [n=ancestra@97-116-126-165.mpls.qwest.net] has joined #wesnoth-dev
20090820 07:04:08-!- Aethaeryn [n=Michael@wesnoth/umc-dev/developer/aethaeryn] has quit ["meh"]
20090820 07:11:26-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090820 07:11:39-!- ABCD [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090820 07:13:19-!- silene [n=plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20090820 07:14:03-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090820 07:20:58-!- maackey [n=chris@pool-72-91-157-206.tampfl.fios.verizon.net] has left #wesnoth-dev []
20090820 07:25:14-!- [Relic] [n=[Relic]@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit ["Leaving"]
20090820 07:30:25-!- xchat_sux [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090820 07:37:05-!- Sirp [n=user@wesnoth/developer/dave] has quit [Read error: 113 (No route to host)]
20090820 08:09:33-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090820 08:09:51-!- Blueblaze2 [n=Blueblaz@166.189.219.234] has joined #wesnoth-dev
20090820 08:30:32-!- Blueblaze [n=Blueblaz@166.189.219.234] has quit [Read error: 110 (Connection timed out)]
20090820 08:30:53-!- stikonas [n=and@ctv-213-164-123-205.vinita.lt] has joined #wesnoth-dev
20090820 08:34:56-!- zookeeper [n=l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20090820 08:41:41-!- Blueblaze2 [n=Blueblaz@166.189.219.234] has quit [Remote closed the connection]
20090820 08:42:18-!- Blueblaze [n=Blueblaz@166.189.219.234] has joined #wesnoth-dev
20090820 08:46:39-!- Crab_ [n=Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20090820 08:51:53-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090820 08:55:18-!- ancestral [n=ancestra@97-116-126-165.mpls.qwest.net] has quit ["And that’s the end of THAT chapter."]
20090820 08:56:42-!- stikonas [n=and@ctv-213-164-123-205.vinita.lt] has joined #wesnoth-dev
20090820 08:59:42-!- Blueblaze [n=Blueblaz@166.189.219.234] has quit []
20090820 09:18:34-!- allefant [n=allefant@allegro/developer/allefant] has quit [Read error: 60 (Operation timed out)]
20090820 09:20:48-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090820 09:20:56-!- silene [n=plouf@wesnoth/developer/silene] has quit ["Leaving."]
20090820 09:21:34< Ivanovic> moin
20090820 09:21:42< Crab_> good morning, Ivanovic
20090820 09:22:20< Ivanovic> Crab_: for "good" it is still too early
20090820 09:22:22< Ivanovic> ;)
20090820 09:23:19-!- AnMaster [n=AnMaster@unaffiliated/anmaster] has quit [Nick collision from services.]
20090820 09:30:10-!- AnMaster [n=AnMaster@unaffiliated/anmaster] has joined #wesnoth-dev
20090820 09:31:26-!- Netsplit kubrick.freenode.net <-> irc.freenode.net quits: corn, fendrin
20090820 09:31:28-!- fendrin [n=fabi@f050246205.adsl.alicedsl.de] has joined #wesnoth-dev
20090820 09:31:38< zookeeper> wtf..
20090820 09:32:02-!- Netsplit over, joins: corn
20090820 09:32:37< zookeeper> oh, duh.
20090820 09:33:16-!- allefant [n=allefant@allegro/developer/allefant] has joined #wesnoth-dev
20090820 09:33:48< Crab_> fendrin: LoW2, the dwarves attacked me without provocation (and they are not my enemy - I can't counterattack
20090820 09:34:48< CIA-62> zookeeper * r37963 /trunk/data/campaigns/Heir_To_The_Throne/scenarios/12_Northern_Winter.cfg: Fixed a silly bug with [cases] being used instead of [case], preventing the snowfall from working.
20090820 09:35:18< zookeeper> that was because when i was writing that code, someone had messed up the engine so it actually used [cases] and not [case], and i guess i forgot to change it before committing.
20090820 10:02:57-!- isaac_ [n=isaac@14.Red-88-26-177.staticIP.rima-tde.net] has joined #wesnoth-dev
20090820 10:08:53-!- Chusslove [n=Chusslov@adsl-ppp-1092.yubc.net] has joined #wesnoth-dev
20090820 10:10:29-!- zookeeper [n=l@wesnoth/developer/zookeeper] has quit [Read error: 145 (Connection timed out)]
20090820 10:16:59< fendrin> Crab_: How can that be?
20090820 10:17:26< Crab_> fendrin: they're attacked me for 1 turn
20090820 10:19:21 * stikonas just got admited to Cambridge, yeaah... :)
20090820 10:19:34< Crab_> turn 1: intro: dwarves recruit, I recall; turn 2: dwarves move towards me, I move on top side of the rives; turn 3: dwarves attack me, but stay friendly to me :)
20090820 10:19:41< Crab_> s/rives/river
20090820 10:19:42< corn> stikonas: congratulations
20090820 10:19:51< stikonas> thanks
20090820 10:20:12< Crab_> fendrin: then, on next troll turn 'trolls have stepped on our side of the river' triggers, and dwarves forget about me
20090820 10:20:12< Ivanovic> stikonas: that cambridge in england?
20090820 10:20:18< stikonas> yes, maths
20090820 10:20:22< Ivanovic> congrats
20090820 10:20:53< fendrin> Crab_: very strange. You are playing on hard?
20090820 10:21:09-!- Tigge_ [n=tigge@bacchus.olf.sgsnet.se] has quit [Read error: 60 (Operation timed out)]
20090820 10:21:10< Crab_> fendrin: hard+debug mode 
20090820 10:21:35< Crab_> fendrin: I'm skimming through the campaign to reach that last scenario, and looking at maps / scenarios along the way
20090820 10:22:20< fendrin> Crab_: On hard the dwarves are your enemy but can't cross the border because of an avoid at command
20090820 10:22:29< Crab_> fendrin: ftp://ftp.terraninfo.net/wesnoth/LOW-Hostile_mountains_replay.gz
20090820 10:22:50< Crab_> fendrin: note that I was unable to attack the dwarves
20090820 10:23:25< fendrin> Crab_: Was that before or after 'trolls have stepped on our side of the river'?
20090820 10:23:35< Crab_> fendrin: 1 turn before that
20090820 10:23:44< Crab_> fendrin: dwarves seemed friendly afterwards
20090820 10:23:46< corn> Crab_: if you remember off the top of your head, how do you retrieve the number of units that each AI side has?
20090820 10:23:48< fendrin> oh
20090820 10:23:54< corn> (this is to get # of units per side at end of game)
20090820 10:23:58< Crab_> corn: look at src/ai/testing.cpp
20090820 10:24:18< corn> ah, sorry about that :)
20090820 10:24:34< corn> I thought I was not storing it already
20090820 10:25:22< Crab_> corn: note that I'm not sure if it stores at the 'end of last turn', if game ends without end_turn
20090820 10:26:11< corn> ok. I will just add the code to log_game_end()
20090820 10:28:04< Crab_> fendrin: ok, the AI hasn't got your 'avoid' aspect right
20090820 10:37:02< fendrin> Crab_: It is very huge at the moment. Maybe to huge
20090820 10:37:37< Crab_> fendrin: what is huge ?
20090820 10:38:35< fendrin> Crab_: the x and y arguement list of the avoid tag.
20090820 10:44:30-!- Appleman1234 [n=Appleman@131.181.103.31] has quit ["Leaving"]
20090820 10:47:40< corn> Crab_: how do I log ai type (default, fai)
20090820 10:48:12< Crab_> corn: see src/ai/testing.cpp, note: LOG_AI_TESTING << "AI_IDENTIFIER"<<side<<": " << ai::manager::get_active_ai_identifier_for_side(side) <<std::endl;
20090820 10:48:51< corn> ok. and ai config?
20090820 10:50:31< corn> also, get_active_ai_identifier_for_side() sometimes returns ""
20090820 10:51:18< Crab_> corn: in generic case, ai config location is unreachable at this moment. if it's ai-vs-ai nogui test, and the ai is passed via --ai_config, then the [side] tag has  "ai_config" attribute
20090820 10:51:50< corn> ok
20090820 10:51:53< Crab_> corn: get_active_ai_identifier_for_side()  returns "" if no ai identifier is specified (as this is generally the case with campaigns)
20090820 10:53:15< corn> Crab_: if no ai identifier is specified, is the default ai used?
20090820 10:53:41-!- Crab_ [n=Crab_@wesnoth/developer/crab] has quit ["Leaving."]
20090820 10:53:55-!- Crab_ [n=Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20090820 10:54:16< Crab_> corn: well, ai identifier does not specify the ai
20090820 10:54:32< Crab_> ai identifier is a 'label' that *can* be attached to a particular ai
20090820 10:54:47< Crab_> so, if no ai identifier is used, the ai is considered 'anonymous'
20090820 10:54:49< corn> ah, I see
20090820 10:54:57< Crab_> it can have formula ai parts, default ai parts, etc
20090820 10:55:07< corn> ok
20090820 10:55:57-!- mjs-de [n=mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20090820 10:56:22-!- wesbot changed the topic of #wesnoth-dev to: 77 bugs, 241 feature requests, 10 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090820 10:56:33< CIA-62> crab * r37964 /trunk/src/ai/ (5 files in 2 dirs): minor patch to engine to allow parsing of [goal] ai config elements (which will be interpreted as additional targets)
20090820 10:57:21< CIA-62> cornmander * r37965 /trunk/src/ai/testing.cpp: Added more detailed logging info for end-game. Added logging about faction and ai identifier. Still need to add logging about ai_config.
20090820 10:58:41< corn> going to go now, be back in 10 hrs or so
20090820 11:03:04-!- zookeeper [n=l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20090820 11:03:56-!- loonybot [n=loonybot@79.139.138.234] has joined #wesnoth-dev
20090820 11:04:44-!- loonycyborg [n=sergey@79.139.138.234] has joined #wesnoth-dev
20090820 11:10:05< Crab_> fendrin: ok, it looks like I've found a bug with translation of [avoid] aspect from old-style syntax to new-style syntax.  will fix soon
20090820 11:17:15< fendrin> Crab_: :-)
20090820 11:41:09-!- ilor_ [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090820 11:41:17< fendrin> Crab_: I have changed scenario3 like we discussed yesterday. But Galtrid isn't able to recruit anything.
20090820 11:41:33< ilor_> did a quick pass through the final survey
20090820 11:41:33< Crab_> fendrin: post a diff somewhere, I'll take a look
20090820 11:41:37-!- Kenpachi [n=chatzill@CPE-58-169-219-15.sa.bigpond.net.au] has joined #wesnoth-dev
20090820 11:42:12< fendrin> Crab_: The error is " 20090820 11:28:09 error engine: could not find unit 'dummy=Elves' "
20090820 11:45:03< Crab_> fendrin: ( post a diff somewhere )
20090820 11:45:20< fendrin> Crab_: I have commited the changes.
20090820 11:45:22< CIA-62> fendrin * r37966 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/03_Kalian.cfg: 
20090820 11:45:22< CIA-62> LoW 03:
20090820 11:45:22< CIA-62> Switched player's controll over to Galtrid.
20090820 11:45:25< Crab_> ok
20090820 11:45:43-!- shadowmaster [n=ignacio@wesnoth/developer/shadowmaster] has quit [Remote closed the connection]
20090820 11:45:43-!- shikadibot [n=shikadi@wesnoth/umc-dev/bot/shikadibot] has quit [Remote closed the connection]
20090820 11:47:22< CIA-62> crab * r37967 /trunk/src/ai/configuration.cpp: fixed upgrade of old-style [avoid] tag to new syntax
20090820 11:48:05< Crab_> fendrin: this makes the avoid config parse successfully ( --log-debug='ai/config' is the way to test )
20090820 11:48:09< Crab_> ^
20090820 11:50:14< Ivanovic> boucman: regarding your answer to question 13: somehow the answer in the selection and the comment don't fit
20090820 11:50:25< fendrin> Crab_: Thank you :-)
20090820 11:50:27< Ivanovic> boucman: that is: maybe you meant to tick the answer above the one you have ticked?
20090820 11:51:58< fendrin> Crab_: Please stop investigating the recruit bug. I think it's only a misspelled macro.
20090820 11:53:00< Ivanovic> boucman: and if you don't have three answers to tick in the "top three" section, just use less
20090820 11:53:13< Ivanovic> (like basically all students and other mentors have done)
20090820 11:54:07< Crab_> fendrin: ok. also, what do you think about adding fog to Galtrid's side ?
20090820 11:54:27< Crab_> fendrin: and spawning some orcs near the citadel, from the start
20090820 11:54:43-!- ilor [n=user@wesnoth/developer/ilor] has quit [Read error: 110 (Connection timed out)]
20090820 11:54:51< Crab_> fendrin: this way the player will start 'defending the citadel against unknown number of orcs'
20090820 11:56:47< fendrin> Crab_: Maybe a good idea. I will think about it. The empty recruiting list was a misspelled macro.
20090820 11:57:16< Crab_> fendrin: ok
20090820 11:57:18< fendrin> zookeeper: Does the modify_turns not work in a turns over event?
20090820 11:58:34< CIA-62> fendrin * r37968 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/03_Kalian.cfg: 
20090820 11:58:34< CIA-62> LoW 03:
20090820 11:58:34< CIA-62> Fixed wrong spelled macro that caused Galtrid to have an empty recruit list.
20090820 11:58:50< zookeeper> fendrin, i wouldn't know
20090820 12:05:26< Crab_> fendrin: please see ftp://ftp.terraninfo.net/wesnoth/sc02.png for the 'starting position' I'm thinking about
20090820 12:22:03< fendrin> Crab_: It's good. You can commit it if you want. I have to go away to university for some time. I think I will be back in at least 6 hours.
20090820 12:22:11< Crab_> fendrin: btw, when does King Haldric break the treaty ?
20090820 12:22:31< Crab_> fendrin: 'some time ago' or 'just before the orc attacks' ?
20090820 12:25:42< Crab_> fendrin: I'm judging from the fact that Kalenz, apparently, doesn't know about that. so, I was wondering when it happened...
20090820 12:26:52< fendrin> Crab_: He refused to see the elves' messengers that called for help with the current thread. So it can be only some days maybe a few weeks ago.
20090820 12:27:10< fendrin> s/thread/threat
20090820 12:28:08< Crab_> fendrin: what about the following situation: give elves the goal 'survive until the return of human army', then, on turn ~6, spawn a rider saying 'the help won't come'
20090820 12:28:44< Crab_> fendrin: then, change the objective to 'defeat the orcs until *their* reinforcements come', then spawn kalenz somewhen later
20090820 12:34:26-!- melinath [n=melinath@pool-72-72-200-231.altnpa.east.verizon.net] has joined #wesnoth-dev
20090820 12:36:45< fendrin> Crab_: Not a bad idea but this would clash with the story in later scenarios. See scenario 8. I am off now :-)
20090820 12:36:53< Crab_> ok
20090820 12:36:58-!- fendrin [n=fabi@wesnoth/developer/fendrin] has quit [Remote closed the connection]
20090820 12:51:08-!- euschn [n=chatzill@wesnoth/developer/euschn] has joined #wesnoth-dev
20090820 13:00:12-!- Kenpachi [n=chatzill@CPE-58-169-219-15.sa.bigpond.net.au] has quit [Read error: 110 (Connection timed out)]
20090820 13:38:27-!- noy [n=Noy@wesnoth/developer/noy] has quit []
20090820 13:47:32-!- Kenpachi [n=chatzill@CPE-58-169-219-15.sa.bigpond.net.au] has joined #wesnoth-dev
20090820 13:54:39-!- Chusslove [n=Chusslov@adsl-ppp-1092.yubc.net] has quit [Read error: 110 (Connection timed out)]
20090820 14:01:26-!- Chusslove [n=Chusslov@adsl-ppp-1092.yubc.net] has joined #wesnoth-dev
20090820 14:11:25-!- melinath [n=melinath@pool-72-72-200-231.altnpa.east.verizon.net] has quit [Read error: 113 (No route to host)]
20090820 14:24:28-!- melinath [n=melinath@pool-72-72-200-231.altnpa.east.verizon.net] has joined #wesnoth-dev
20090820 14:25:04-!- Netsplit kubrick.freenode.net <-> irc.freenode.net quits: knotwork__, wesbot, kimrhh, allefant, deekay, Amu_
20090820 14:25:09-!- Netsplit kubrick.freenode.net <-> irc.freenode.net quits: Rhonda, stikonas, yann, ABCD, zookeeper, Chusslove, corn, isaac_, CIA-62, euschn,  (+11 more, use /NETSPLIT to show all of them)
20090820 14:26:04-!- Netsplit over, joins: Soliton, deekay, melinath, ilor_, zookeeper, isaac_, allefant, ABCD, Ivanovic, BenUrban (+17 more)
20090820 14:26:40-!- Netsplit kubrick.freenode.net <-> irc.freenode.net quits: kimrhh, allefant
20090820 14:26:58-!- Netsplit over, joins: allefant, kimrhh
20090820 14:27:30-!- Netsplit kubrick.freenode.net <-> irc.freenode.net quits: knotwork__, wesbot, Amu_
20090820 14:28:08-!- Netsplit over, joins: wesbot, knotwork__, Amu_
20090820 14:28:47-!- Netsplit kubrick.freenode.net <-> irc.freenode.net quits: esr, deekay
20090820 14:29:16-!- Netsplit kubrick.freenode.net <-> irc.freenode.net quits: isaac_, ABCD, melinath, BenUrban, dfranke
20090820 14:29:16-!- Netsplit over, joins: deekay, esr
20090820 14:29:43-!- Netsplit over, joins: melinath, isaac_, ABCD, BenUrban, dfranke
20090820 14:34:00-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Connection timed out]
20090820 14:37:30-!- Blarumyrran [n=minaise@81-20-159-197.levira.ee] has joined #wesnoth-dev
20090820 14:53:19< zookeeper> lots and lots of wiki spam again.
20090820 15:07:23< Crab_> zookeeper: I'll try to upgrade mediawiki next week, after GSoC ends, to install some antispam defense.
20090820 15:10:03-!- melinath [n=melinath@pool-72-72-200-231.altnpa.east.verizon.net] has quit [Read error: 113 (No route to host)]
20090820 15:13:45-!- BenUrban [n=benurban@unaffiliated/benurban] has quit ["Power failu"]
20090820 15:25:27-!- krzysz00 [n=user@pool-71-252-228-249.dllstx.fios.verizon.net] has joined #wesnoth-dev
20090820 15:29:23< krzysz00> if cutting out the (data/)sounds and music, enableing tinygui and disableing nls does not take a nokia n800 port down to a managable size, what can be cut out?
20090820 15:30:10-!- Sirp [n=user@pool-71-164-166-178.dllstx.fios.verizon.net] has joined #wesnoth-dev
20090820 15:30:55< zookeeper> krzysz00, you might want to ask again in case Sirp has an idea of that..
20090820 15:31:10< krzysz00> if cutting out the (data/)sounds and music, enableing tinygui and
20090820 15:31:10< krzysz00> 	   disableing nls does not take a nokia n800 port down to a managable
20090820 15:31:10< krzysz00> 	   size, what can be cut out
20090820 15:32:35< krzysz00> does anyone think that gziping maps would be a good idea (for the next wesnoth)
20090820 15:32:38< zookeeper> well, basically you could cut out all unit animations, too. probably some code changes to help with that would be necessary, though.
20090820 15:33:05-!- Sirp [n=user@pool-71-164-166-178.dllstx.fios.verizon.net] has quit [Read error: 54 (Connection reset by peer)]
20090820 15:33:06< zookeeper> not really, maps and scenario files take very little space.
20090820 15:33:17< deekay> Portaits take 50MB
20090820 15:33:25< zookeeper> oh yeah, definitely portraits.
20090820 15:33:35< krzysz00> wait i could just cut out all the animations disables by lowmem, (if i knew which ones those were)
20090820 15:33:36< zookeeper> or you could drop them to half the size or something.
20090820 15:34:38< krzysz00> tinygui already gets the portraits might help
20090820 15:35:58< krzysz00> i might need to reconcile my port with the official debian packaging setup (modularity) when 1.6.5 comes out
20090820 15:37:22< zookeeper> you trying to cut the overall package size, memory usage, or both?
20090820 15:38:25-!- Sirp [n=user@wesnoth/developer/dave] has joined #wesnoth-dev
20090820 15:41:11< krzysz00> package size, if i need to
20090820 15:41:28< krzysz00> which i'll find out when make install finishes
20090820 15:50:43< zookeeper> well, if you take out portraits and unit animations then that's a pretty big drop already
20090820 15:51:25< krzysz00> can the wesnoth fonts be expected on a usual linux distro
20090820 15:52:11< krzysz00> strike that but i'd like to know what "wqy-zenhei.ttc" is
20090820 15:52:25-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20090820 15:52:43< krzysz00> i'm shrinking portraits
20090820 15:57:05< CIA-62> thespaceinvader * r37969 /trunk/ (6 files in 5 dirs): Add and wire new Master-at-arms portrait. Update changelogs, portrait credits.
20090820 16:10:03-!- krzysz00 [n=user@pool-71-252-228-249.dllstx.fios.verizon.net] has quit [Remote closed the connection]
20090820 16:13:08-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090820 16:13:17-!- ABCD [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090820 16:23:02-!- lizard_r [n=Miranda@wesnoth/umc-dev/developer/lizard] has joined #wesnoth-dev
20090820 16:28:15-!- Elvish_Pillager [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090820 16:28:47-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090820 16:31:27-!- stikonas [n=and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20090820 16:32:26-!- fendrin [n=fabi@g228029125.adsl.alicedsl.de] has joined #wesnoth-dev
20090820 16:38:16< fendrin> Crab_: Hi, I am back.
20090820 16:38:24< Crab_> fendrin: hello
20090820 16:38:34< Crab_> fendrin: working on [modify_ai] tag atm
20090820 16:44:57-!- crimson_penguin [n=ben@64.201.60.211] has joined #wesnoth-dev
20090820 16:55:50-!- Chusslove [n=Chusslov@adsl-ppp-1092.yubc.net] has quit [Read error: 113 (No route to host)]
20090820 16:56:22-!- wesbot changed the topic of #wesnoth-dev to: 78 bugs, 241 feature requests, 10 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090820 17:02:45-!- Chusslove [n=Chusslov@adsl-ppp-3192.yubc.net] has joined #wesnoth-dev
20090820 17:12:23-!- elias [n=allefant@allegro/developer/allefant] has joined #wesnoth-dev
20090820 17:29:02< fendrin> Crab_: I am working on the change of protagonists after kalenz arrived. I got everything to work except that the games tells the player that he is defeated and ends. 
20090820 17:29:37-!- allefant [n=allefant@allegro/developer/allefant] has quit [Read error: 110 (Connection timed out)]
20090820 17:30:33-!- Tigge [n=tigge@bacchus.olf.sgsnet.se] has joined #wesnoth-dev
20090820 17:31:22< CIA-62> fendrin * r37970 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/03_Kalian.cfg: 
20090820 17:31:22< CIA-62> LoW 03:
20090820 17:31:22< CIA-62> Fixed the Protagonist switch from Galtrid to Kalenz.
20090820 17:31:22< CIA-62> The bad is that the games ends after that with a defeat message.
20090820 17:34:39-!- euschn [n=chatzill@wesnoth/developer/euschn] has quit [Read error: 145 (Connection timed out)]
20090820 17:37:54-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090820 17:49:18< Crab_> fendrin: ok, I'll try it out
20090820 17:50:45-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090820 17:51:02-!- Chusslove [n=Chusslov@adsl-ppp-3192.yubc.net] has quit [Read error: 60 (Operation timed out)]
20090820 18:04:48< fendrin> Crab_: That maybe the order of the actions taken. There is a short period in which side 1 has no leader. I guess I need to avoid that.
20090820 18:07:26-!- Elvish_Pillager [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090820 18:07:34< Crab_> fendrin: no
20090820 18:07:53< Crab_> fendrin: the problem is that adding turns from the 'turns are over' event doesn't work
20090820 18:08:16< Crab_> fendrin: 'game has ended' is already signalled
20090820 18:08:40< Crab_> fendrin:  this, for example, fixes it:
20090820 18:08:40< Crab_>  [event]
20090820 18:08:40< Crab_> -        name="time over" 
20090820 18:08:40< Crab_> +        name="turn 2" 
20090820 18:11:01-!- Chusslove [n=Chusslov@adsl-ppp-1200.yubc.net] has joined #wesnoth-dev
20090820 18:11:53< fendrin> Crab_: I see. Can't that be changed?
20090820 18:12:23< Crab_> fendrin: yes. but, IMO, it's better to just fire that event not on 'turns over', but on 'turn X'
20090820 18:13:19< Crab_> fendrin: the elves don't know the turn on which Kalenz will arrive, isn't it ?
20090820 18:15:37< Crab_> fendrin: so, it's better to have original turn timer / objective set to something different. E.g. "defeat orcs in 35 turns"  / "survive for N turns until elven vanguard arrives (Galtrid greets Kalenz with 'Are you our army's vanguard?', so he's obviously expecting that they might come " 
20090820 18:16:56< Crab_> fendrin: meanwhile, I'll patch the engine to check for added turns
20090820 18:22:59< fendrin> Crab_: Indeed, the elves don't have an idea when or even if reenforcements arrive. But it's common to communicate to the player how long he must hold. It only leads to frustration if a level is to be repeated because of the leak of proper scenario goals.
20090820 18:23:14-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has quit []
20090820 18:24:08< Crab_> fendrin: IMO, you can offset/randomize by -1 to -3 turns, this won't change much
20090820 18:24:53< Crab_> fendrin:  on the other hand, the level will not end after "turn 9", so saying '9 turns to play' is a bit misleading
20090820 18:25:38< Crab_> fendrin: IMO, time limit of 'defeat the orcs in 35 turns' from the start, with hints that 'our vanguard should be here in two days' is good
20090820 18:26:19< Crab_> fendrin: that way, the player will know the final goal (= defeat the orcs), and that he's supposed to get reinforcements in two days ( = around turn 12)
20090820 18:40:40< fendrin> I am off for two hours.
20090820 18:46:24< Soliton> zookeeper: trow:15 is pretty mean in cutting your gold since you don't do anything there and get no bonus.
20090820 18:46:41< Soliton> ..and then you should defeat a dragon.
20090820 18:47:49-!- zlefin2 [n=Zlefin@c-24-61-86-218.hsd1.ma.comcast.net] has joined #wesnoth-dev
20090820 19:04:28< zookeeper> Soliton, yeah. wanna redesign it? :p
20090820 19:07:40< Soliton> just don't apply the carryover percentage.
20090820 19:08:40< Soliton> units on lit shallow water don't get submerged.
20090820 19:08:49-!- Aethaeryn [n=Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #Wesnoth-dev
20090820 19:09:01-!- zlefin2 [n=Zlefin@c-24-61-86-218.hsd1.ma.comcast.net] has quit [Read error: 110 (Connection timed out)]
20090820 19:10:01< zookeeper> yeah, terrain layers tend to do that
20090820 19:14:40-!- cib0 [n=cib@p5DC43860.dip.t-dialin.net] has joined #wesnoth-dev
20090820 19:14:40-!- Crab_ [n=Crab_@wesnoth/developer/crab] has quit [Read error: 104 (Connection reset by peer)]
20090820 19:14:43< zookeeper> Soliton, but do you mean that you'd carryover 100% of your gold? sounds a bit...much, considering that can't you still make a profit in that scenario?
20090820 19:14:56-!- Crab_ [n=Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20090820 19:15:15< zookeeper> in any case, i'd really like to see that scenario changed completely. the gameplay is currently very dumb.
20090820 19:18:43< Soliton> sure, but currently it just cuts the players gold in half for no reason.
20090820 19:19:10< Soliton> you can make a bit of gold by not recruiting/recalling and trying to scare the AI off as long as possible.
20090820 19:20:53< CIA-62> crab * r37971 /trunk/src/ (playsingle_controller.cpp tod_manager.cpp tod_manager.hpp): allow 'time over' handling to take into account the possibility of changing the number of turns in the 'time over' event handler
20090820 19:21:06< Crab_> fendrin: ^ fixed
20090820 19:25:16< zookeeper> Soliton, well, i don't mind a change like that if you think it's a good idea
20090820 19:33:31-!- thespaceinvader_ [n=chatzill@cpc2-whit1-0-0-cust621.cdif.cable.ntl.com] has joined #wesnoth-dev
20090820 19:34:05-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has quit [Nick collision from services.]
20090820 19:34:07-!- thespaceinvader_ is now known as thespaceinvader
20090820 19:39:42-!- Crab_ [n=Crab_@wesnoth/developer/crab] has quit [Read error: 54 (Connection reset by peer)]
20090820 19:45:56-!- giusef [n=giusef@unaffiliated/giusef] has joined #wesnoth-dev
20090820 19:55:02-!- Crab_ [n=Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20090820 19:56:41< Soliton> zookeeper: well, basically it's a talking scenario IMO and usually you just keep your gold there.
20090820 20:00:08-!- thespaceinvader_ [n=chatzill@cpc2-whit1-0-0-cust621.cdif.cable.ntl.com] has joined #wesnoth-dev
20090820 20:01:53-!- thespaceinvader_ [n=chatzill@cpc2-whit1-0-0-cust621.cdif.cable.ntl.com] has quit [Read error: 104 (Connection reset by peer)]
20090820 20:02:17-!- thespaceinvader_ [n=chatzill@cpc2-whit1-0-0-cust621.cdif.cable.ntl.com] has joined #wesnoth-dev
20090820 20:02:33-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has quit [Nick collision from services.]
20090820 20:02:37-!- thespaceinvader_ is now known as thespaceinvader
20090820 20:04:24-!- Netsplit kubrick.freenode.net <-> irc.freenode.net quits: erl, Sirp, Aethaeryn
20090820 20:05:00-!- Netsplit over, joins: Aethaeryn, erl, Sirp
20090820 20:05:35-!- Netsplit kubrick.freenode.net <-> irc.freenode.net quits: Crab_, Tigge, yann
20090820 20:06:47-!- Netsplit over, joins: Crab_, Tigge, yann
20090820 20:09:05-!- Blueblaze [n=Blueblaz@166.189.157.120] has joined #wesnoth-dev
20090820 20:12:10-!- kitty_ [n=kitty@wesnoth/artist/kitty] has joined #wesnoth-dev
20090820 20:17:14< Elvish_Pillage2> Hmm. There seems to be a limit of five WML-set menu items. (It's not documented on the wiki.) Is it possible to check how many there are, e.g. if you're in an era and want to know how many menu items the map itself sets?
20090820 20:19:19-!- thespaceinvader_ [n=chatzill@cpc2-whit1-0-0-cust703.cdif.cable.ntl.com] has joined #wesnoth-dev
20090820 20:20:06-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has quit [Nick collision from services.]
20090820 20:20:12-!- thespaceinvader_ is now known as thespaceinvader
20090820 20:21:07-!- grzywacz [n=grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20090820 20:22:10< Blueblaze> Elvish_Pillage2, I believe the limit is per hex you right click, not for the entire map. The others will still exist, but just not be shown. Also, the limit should be 6 or 7, not 5
20090820 20:22:23< Soliton> i can't advance to any scenario the second time i'm at the ka'lian in trow.
20090820 20:22:28< Elvish_Pillage2> Blueblaze, yeah that's what I meant
20090820 20:22:33< Elvish_Pillage2> 6 maybe
20090820 20:23:29< Blueblaze> And no, there is no way to check how many are currently active at a certain location
20090820 20:24:06< Elvish_Pillage2> Blueblaze, it would be useful to check how many there were overall too
20090820 20:24:23-!- noy [n=Noy@d75-157-29-32.bchsia.telus.net] has joined #wesnoth-dev
20090820 20:24:26< Elvish_Pillage2> (I had assumed _that_ wasn't possible)
20090820 20:25:06< Blueblaze> True, it would be useful to have an era check the scenario as a scenario could easily break an era dependent on set menu items
20090820 20:25:36< Elvish_Pillage2> I'm currently modifying EoHS so that it's possible to make it use only 1 menu item if necessary
20090820 20:28:07< Soliton> okay looks like this time the dialog with the choices is in a while loop and the endlevel is not processed right away so i don't get out of it.
20090820 20:28:38< Soliton> zookeeper: any idea why a loop is used there?
20090820 20:29:13< Soliton> that WML is quite redundant anyway..
20090820 20:31:09< fendrin> Crab_: Great :-) I am compiling your fix.
20090820 20:31:25< Crab_> ok
20090820 20:32:47< Soliton> zookeeper: i'll rewrite that WML with switch and remove the loops.
20090820 20:33:42< fendrin> thespaceinvader: Hi, the orcish bowman is new but his idle anim is still the old in trunk. Looks very wrong.
20090820 20:34:31-!- thespaceinvader_ [n=chatzill@cpc3-whit1-0-0-cust741.cdif.cable.ntl.com] has joined #wesnoth-dev
20090820 20:35:13-!- thespaceinvader_ [n=chatzill@cpc3-whit1-0-0-cust741.cdif.cable.ntl.com] has quit [Client Quit]
20090820 20:38:19-!- elias [n=allefant@allegro/developer/allefant] has quit [Remote closed the connection]
20090820 20:41:51-!- elias [n=allefant@allegro/developer/allefant] has joined #wesnoth-dev
20090820 20:43:36< zookeeper> Soliton, sounds good to me
20090820 20:46:40-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has quit [Read error: 110 (Connection timed out)]
20090820 20:49:36-!- deekay [n=dk@wesnoth/developer/dragonking] has quit []
20090820 20:52:45-!- shadowmaster [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090820 20:56:04< Soliton> hmm, has switch issues with numerical values?
20090820 20:56:22< Soliton> oh, typo...
20090820 20:57:49< Soliton> ok, now it works.
20090820 21:00:06< CIA-62> soliton * r37972 /trunk/data/campaigns/The_Rise_Of_Wesnoth/scenarios/16_Elf_Lords.cfg: simplified WML by using [switch] and removing useless loops
20090820 21:00:13-!- shadowmaster [n=ignacio@wesnoth/developer/shadowmaster] has quit [SendQ exceeded]
20090820 21:02:31-!- ShikadiLord [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090820 21:02:57< ShikadiLord> my irssi config in the server went to heck :/
20090820 21:06:16-!- shikadibot [n=shikadi@wesnoth/umc-dev/bot/shikadibot] has joined #wesnoth-dev
20090820 21:06:34-!- ShikadiLord is now known as shadowm_laptop
20090820 21:07:28-!- silene [n=plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20090820 21:07:34< silene> hi
20090820 21:08:16< shadowm_laptop> silene: may I reindent game_events.cpp? it currenty looks rather messy after r35005
20090820 21:10:18< silene> i don't own the code; you have commit access, you do whatever you want; just don't make it hard for people that would like to look at the history later
20090820 21:10:27-!- Deusite [n=Deusire@92-238-120-246.cable.ubr26.newt.blueyonder.co.uk] has joined #wesnoth-dev
20090820 21:10:41< Soliton> silene: endlevel stuff breaks in trow:16 where it is inside a while loop. i've cleaned up the WML since the loop was useless but maybe there are other cases where it makes sense.
20090820 21:10:49< shadowm_laptop> yeah, I was just wondering whether your changes there were intended or not
20090820 21:10:56-!- Deusite [n=Deusire@92-238-120-246.cable.ubr26.newt.blueyonder.co.uk] has left #wesnoth-dev []
20090820 21:12:09-!- Deusite [n=Deusire@92-238-120-246.cable.ubr26.newt.blueyonder.co.uk] has joined #wesnoth-dev
20090820 21:12:14-!- Deusite [n=Deusire@92-238-120-246.cable.ubr26.newt.blueyonder.co.uk] has left #wesnoth-dev []
20090820 21:12:43< Soliton> btw, goto moves can not be undone currently.
20090820 21:15:03< silene> Soliton: right, i wouldn't have imagined a endlevel tag in a loop; but i'm not sure it is worth fixing, since the new behavior also has advantages; for instance, it makes the behavior deterministic when there are several events and one of them ends the level
20090820 21:16:01< Aethaeryn> What's that handy link that gives status of translations?
20090820 21:16:17< Soliton> silene: ah right that was the point of the change. nevermind then.
20090820 21:17:07< Soliton> gettext.wesnoth.org
20090820 21:17:33< Soliton> wiki.wesnoth.org/UsefulLinks
20090820 21:17:53< Soliton> all the handy links you'll ever need! ^
20090820 21:18:00-!- ShikadiLord [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090820 21:18:00< fendrin> Aethaeryn: http://www.wesnoth.org/gettext/index.lang.php?lang=de&version=trunk
20090820 21:18:05< silene> Crab_: have you seen http://www.wesnoth.org/forum/viewtopic.php?f=4&t=26751 ?
20090820 21:18:15< Aethaeryn> If the language is greyed out does it mean it's so horribly abandonned?
20090820 21:18:25< Crab_> silene: thanks, I'll take a look
20090820 21:18:47-!- ShikadiLord [n=ignacio@wesnoth/developer/shadowmaster] has quit [Client Quit]
20090820 21:20:17< Crab_> silene: ok, will be fixed
20090820 21:20:21-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090820 21:21:36< silene> Crab_: there is also https://gna.org/bugs/index.php?14161 (not sure if this is a bug in the old ai or in the new code)
20090820 21:23:17-!- shadowm_laptop [n=ignacio@wesnoth/developer/shadowmaster] has quit [Client Quit]
20090820 21:26:22-!- ShikadiLord [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090820 21:26:27-!- ShikadiLord is now known as shadowm_laptop
20090820 21:27:13-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090820 21:28:06-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Nick collision from services.]
20090820 21:28:26-!- ABCD_ is now known as ABCD
20090820 21:28:57-!- ancestral [n=ancestra@209.181.254.222] has joined #wesnoth-dev
20090820 21:29:33< corn> Ivanovic: I completed the final eval
20090820 21:30:03-!- zookeeper [n=l@wesnoth/developer/zookeeper] has quit []
20090820 21:30:24-!- ancestral_ [n=ancestra@166.205.7.159] has joined #wesnoth-dev
20090820 21:31:01-!- kitty_ [n=kitty@wesnoth/artist/kitty] has quit []
20090820 21:35:45< Ivanovic> good
20090820 21:36:46-!- ancestral [n=ancestra@209.181.254.222] has quit [Read error: 60 (Operation timed out)]
20090820 21:36:48-!- ancestral_ is now known as ancestral
20090820 21:39:18-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has quit [Client Quit]
20090820 21:42:37-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090820 21:43:52-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has quit [Client Quit]
20090820 21:44:10-!- yann [n=dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has quit [Read error: 60 (Operation timed out)]
20090820 21:44:48-!- yann [n=dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has joined #wesnoth-dev
20090820 21:45:03-!- giusef [n=giusef@unaffiliated/giusef] has quit ["exit (-1);"]
20090820 21:45:35-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090820 21:47:16< Soliton> at the saurian beach the saurians say "We must hold the beach until we can call our naga friends." then proceed to recruit nagas...
20090820 21:48:37-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has quit [Client Quit]
20090820 21:50:08-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090820 21:57:07-!- ancestral [n=ancestra@166.205.7.159] has quit ["Get Colloquy for iPhone! http://mobile.colloquy.info/"]
20090820 21:58:24-!- Blueblaze [n=Blueblaz@166.189.157.120] has quit []
20090820 22:05:24-!- crimson_penguin [n=ben@64.201.60.211] has joined #wesnoth-dev
20090820 22:31:13-!- Shadow_Master is now known as shadowmaster
20090820 22:34:49-!- shadowm_laptop [n=ignacio@wesnoth/developer/shadowmaster] has quit ["Lost terminal"]
20090820 22:37:42-!- Crab_ [n=Crab_@wesnoth/developer/crab] has quit [Read error: 54 (Connection reset by peer)]
20090820 22:37:43< Soliton> what would be the use of an empty event with only a filter?
20090820 22:38:51< shadowmaster> absolutely nothing unless there's evil magic hidden in it
20090820 22:38:51< fendrin> Soliton: Is it hot were you live?
20090820 22:39:18-!- Crab_ [n=Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20090820 22:39:37< Soliton> fendrin: somewhat, raining atm though.
20090820 22:39:58< fendrin> Soliton: Lucky you, here it's still over 30 degree and it's already dark.
20090820 22:42:15< Blarumyrran> fendrin: Lucky you, here it's still under 17 degrees and it's already dark
20090820 22:42:44< shadowmaster> under 15 degrees (Celsius) and it's not dark yet
20090820 22:44:20 * fendrin feels not lucky. Can't be nacked enough.
20090820 22:44:33 * shadowmaster weeee... winter!
20090820 22:44:49< fendrin> german hotness is mostly not nice because of the dry air.
20090820 22:45:00< shadowmaster> it's dry here when hot too
20090820 22:45:21< fendrin> shadowmaster: you live in south america, don't you? Chile?
20090820 22:48:45-!- Chusslove [n=Chusslov@adsl-ppp-1200.yubc.net] has quit [Read error: 60 (Operation timed out)]
20090820 22:52:17< fendrin> Crab_: Finished with modifiyng scenario 3. How do you want to make the decission for the orcs how the behave (Retreat to the hills or storm the kalian)?
20090820 22:56:23-!- wesbot changed the topic of #wesnoth-dev to: 79 bugs, 241 feature requests, 10 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090820 22:56:43-!- 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"]
20090820 22:56:54-!- boucman [n=rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20090820 22:57:28< Crab_> fendrin: depends on what do you want from the scenario. I was thinking about the following: 1) 'remove some of the orcs starting gold', 2) turn on fog of war to make it more difficult for the elf to protect the villages 3) instruct the orcs to try to capture the villages and test the citadel's defenses. 
20090820 22:57:37< boucman> hey all
20090820 22:58:03< Crab_> 4) once elven reinforcements (e.g. kalenz) have arrived, do a internal strategy evaluation and select one of several strategies a) focus on killing kalenz b) focus on attacking citadel c) defend till the end of turns
20090820 22:58:35< Crab_> fendrin: that 'internal strategy evaluation' can be done via formula ai
20090820 22:59:21< Crab_> fendrin: and, after the elven reinforcement arrive, I'll return that previously removed gold to orcs (to allow them to recruit again)
20090820 22:59:30< fendrin> Crab_: I like your plans. I suggest we implement it as you suggest and test it.
20090820 23:00:51< Crab_> fendrin: ok. I'm working on [modify_ai] tag atm, once that is finished, I'll work on this.
20090820 23:01:51< Crab_> hi boucman
20090820 23:03:22< esr> fendrin: Are you happy with the LoW dialog changes I did?
20090820 23:03:30-!- shikadibot [n=shikadi@wesnoth/umc-dev/bot/shikadibot] has quit ["Lost terminal"]
20090820 23:05:16< grzywacz> boucman, :)
20090820 23:06:08< fendrin> esr: Hi esr. Yes, like them. Good work like always. There was only one little issue. There is no more speech for the case that velon has died before scenario 1 ends. And I will need more changes before the string freeze since crab is recoding LoW ai settings and this will require more string changes before the freeze.
20090820 23:06:40< esr> OK, insert TODOs as needed and I'll fulfil them.
20090820 23:06:42-!- YogiHH [n=chatzill@c186064.adsl.hansenet.de] has joined #wesnoth-dev
20090820 23:09:05-!- xchat_sux [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090820 23:09:15-!- xchat_sux is now known as Elvish_Pillager
20090820 23:09:59< fendrin> esr: Thank you. The TODOs are already there but not final. I will notify you.
20090820 23:12:38< esr> fendrin: Note, I assigned you a couple of bugs recently. If they're not actually within your scope, please let me know so I can try to find someone else appropriate.
20090820 23:14:31< fendrin> esr: Right, I will have a look at them now.
20090820 23:15:55< fendrin> esr: I believe I can handle them all.
20090820 23:16:51< esr> Good.  The bug list has groewn alarmingly while I was on vacation.
20090820 23:17:36-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090820 23:17:50< Crab_> boucman: can you comment on [modify_ai] syntax ? I'm thinking about one of syntaxes described there - http://wesnoth.pastebin.com/m64e686bb
20090820 23:18:10< boucman> checking
20090820 23:18:46< CIA-62> fendrin * r37973 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/03_Kalian.cfg: LoW 03: Finished the protagonist switching code.
20090820 23:18:58< Crab_> boucman: basically, the idea is to represent the AI for a particular side as a tree of components, and then allow the user to specify one of the elements of the tree, and specify 'what to do with it'
20090820 23:19:17< boucman> hmm
20090820 23:19:37-!- Sapient [n=patrickp@wesnoth/developer/sapient] has joined #wesnoth-dev
20090820 23:19:50< boucman> if I were you i'd try to reuse existing syntax parsers as much as possible, it will avoid compatibility problems if/when the syntax evolves
20090820 23:20:23< Crab_> boucman: it will be exposed to formula ai, almost from the start
20090820 23:20:35 * Sapient removes "if/" from boucman's statement
20090820 23:20:53< boucman> Sapient: hehe
20090820 23:21:19< boucman> Crab_: I was thinking of the syntax parsers you already have, for facets/sides etc...
20090820 23:21:39< boucman> having other edition modes than the simpley "replace" force to have more complicated syntaxe
20090820 23:21:42< boucman> s
20090820 23:21:59< Crab_> boucman: well, those are not a problem. since each facet/aspect/stage/engine will handle all those things 'for itself'
20090820 23:22:09< Crab_> boucman: so, there will be no code duplication in those
20090820 23:22:15< boucman> ok, giid
20090820 23:22:18< boucman> good
20090820 23:22:22< Crab_> boucman: the price to pay is 'ai must be initialized to modify it's config'
20090820 23:22:30-!- ABCD [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090820 23:22:50< Crab_> boucman: so, [modify_ai] will have to either wait around or trigger ai initialization
20090820 23:23:09< CIA-62> fendrin * r37974 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/01_The_Uprooting.cfg: 
20090820 23:23:09< CIA-62> LoW 01:
20090820 23:23:09< CIA-62> Code cleanup.
20090820 23:23:09< CIA-62> TODO for story prose.
20090820 23:23:09< CIA-62> Added a filter for ensuring that one of the orc leaders that survived the battle gives the last orders.
20090820 23:23:34< boucman> how could you call modify_ai before ai is initialized ?
20090820 23:23:51< boucman> i thought ais were part of the generic scenario that is parsed before any event
20090820 23:23:54< Crab_> boucman: from the WML, of course :)
20090820 23:24:06< Crab_> boucman: AI is lazily initialized :)
20090820 23:24:09< Crab_> boucman: on 1st access
20090820 23:24:13< boucman> ok
20090820 23:24:20< CIA-62> fendrin * r37975 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/19_Costly_Revenge.cfg: LoW 19: Removed one of the duplicate note= attributes in the objectives.
20090820 23:24:34< boucman> then modify_ai should init AI
20090820 23:24:34< Crab_> boucman: so, if you start a game, the ai will, generally, get initialized on its first turn
20090820 23:25:01< Crab_> boucman: yes, that's how it'll be done (actually, that part is coded in my local tree)
20090820 23:25:03< boucman> since there could be generic AI marcos/specialisations that would be easier to implement via modify_ai in prestart events
20090820 23:25:14-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090820 23:26:11< Crab_> boucman: I'm intent the following way : a scenario will {include} a generic ai from somewhere else (e.g. from data/ai/ais, or from a campaign-specific file), and, a scenario will add a couple [modify_ai] tags to modify that generic ai.
20090820 23:26:35< CIA-62> fendrin * r37976 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/09_Bounty_hunters.cfg: LoW 09: Added starting gold macro for the player since he no longer carries gold from the Ka'lian treasure.
20090820 23:26:47< boucman> Crab_: yes, that's also how I think people will do it
20090820 23:27:35< CIA-62> fendrin * r37977 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/12_Revelations.cfg: LoW 12: The title of the scenario wasn't displayed.
20090820 23:28:54< Crab_> boucman: the question I'm thinking about, is the following one: for example, if I want to add a facet to a certain aspect, I need to specify both the path to that aspect, and the 'place' to put the facet in (order of the facets - matters). so, how is shall be in WML ? a) [modify_ai] component=aspect['aggression']  property=facet[4]  action=add  [cfg]...[/cfg][/modify_ai]
20090820 23:29:06< Crab_> or  b) [modify_ai] component=aspect['aggression'].facet[4]  action=add  [cfg]...[/cfg][/modify_ai]
20090820 23:29:22< Crab_> or c) [modify_ai] add=aspect['aggression'].facet[4]  [cfg]...[/cfg][/modify_ai]
20090820 23:30:33< Crab_> the code is supposed to mean "add the contents of that [cfg] tag as a 5th facet of 'aggression' aspect "
20090820 23:30:37< boucman> Crab_: b or c, depending on which one is sompler for you from an implementation PoV
20090820 23:30:49-!- Chusslove [n=Chusslov@adsl-ppp-1011.yubc.net] has joined #wesnoth-dev
20090820 23:31:13< boucman> Crab_: however we will probably need a special index to indicate "at the end" (the begining being probably index of 0)
20090820 23:31:16< Crab_> boucman: I like c) more, since it's less verbose
20090820 23:31:23< Crab_> boucman: [] or [-1]
20090820 23:31:29< CIA-62> fendrin * r37978 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/07_Elves_last_stand.cfg: 
20090820 23:31:29< CIA-62> LoW 07: Removed one of the duplicate note attributes in the objectives.
20090820 23:31:29< CIA-62> Added a music title to the playlist.
20090820 23:31:29< CIA-62> Removed the choice between strong and weak ai since it was from a time were computers weren't as smart as humans.
20090820 23:31:29< Crab_> or just 
20090820 23:31:46< boucman> yes, both are fine, just don't forget the feature
20090820 23:33:11< YogiHH> Ivanovic: final evaluation filled and submitted
20090820 23:33:13< Crab_> boucman: also, I want to add a new component to the AI which will handle 'global strategy for this team'. this will be an aspect
20090820 23:33:43< boucman> you mean a "mini-stage" aspect ?
20090820 23:34:17< Crab_> boucman: no. this aspect will be responsible with generation of tactical-level goals for current turn
20090820 23:35:05< boucman> ok, makes sense
20090820 23:35:21< fendrin> Crab_: May I know your Realname for the about credits of LoW?
20090820 23:35:29< Crab_> fendrin: Yurii Chernyi
20090820 23:35:53< Crab_> boucman: basically, I think that is not good to put the lists of 'smaller goals' (e.g. 'kill this, kill that, move there') in scenario WML - it's better to specify a larger goal (e.g. "concentrate on killing side 1") and let the AI find out the best way to do that.
20090820 23:36:48< Crab_> boucman: concentration of efforts is very important - the AI, in general, looks nice if it's actively trying to do something, not just roaming around 
20090820 23:36:50< boucman> Crab_: yes, scenario goals vs tactical goals
20090820 23:37:05< boucman> so this aspect would be for tactical goals ?
20090820 23:37:22< Crab_> boucman: this aspect will be for translation of scenario goals into tactical goals
20090820 23:38:20< boucman> ok
20090820 23:39:37< fendrin> esr: doesn't wmllint check for translate "_" in the about tag?
20090820 23:39:47< Crab_> boucman: that is to allow me to write 'here's the strategy aspect of the orc ai, it has 3 facets 'kill kalenz', 'kill Galtrid', 'try to hold until end of turns',  one of them is active, depending on the evaluation of enemy strength and disposition"
20090820 23:40:15< fendrin> esr: At least all comment attributes in LoW don't have them.
20090820 23:40:23< Crab_> (evaluation - using fai)
20090820 23:40:39< boucman> Crab_: the decision of which one is active is on the scenario side, not the AI side I guess
20090820 23:42:19< Crab_> boucman: as we want to. actually, in some situations, it makes sense to allow the AI to decide for itself. e.g., if the ai went to kill kalenz, the human spotted it and put all forces to defend him, and the ai sees that galtrid is virtually unprotected and perfectly killable in 3-4 turns.... - it'll make sence to switch strategy.
20090820 23:42:21-!- Appleman1234 [n=Appleman@131.181.101.215] has joined #wesnoth-dev
20090820 23:42:41< Crab_> boucman: or, this will allow the ai to do a  'global retreat' if things become not good
20090820 23:43:44< boucman> Crab_: the underlying question was
20090820 23:43:56< Crab_> boucman: scenario creators should pick 'what general options they want to make available to the ai', and setup specific or general switching rules for them..
20090820 23:44:10< Sapient> Crab_: how hard would it be to teach the AI to use custom WML menu items ?
20090820 23:44:28< boucman> should scenarios have an and/or logic for goals, or should they use WML to only set one goal after assessing the situation with formulas
20090820 23:44:29-!- silene [n=plouf@wesnoth/developer/silene] has quit ["Leaving."]
20090820 23:44:45< Crab_> Sapient: for what puprose the AI should use them ?
20090820 23:45:05< Sapient> Crab_: it would differ for each add-on
20090820 23:45:10< shadowmaster> fendrin: yes
20090820 23:45:33< Sapient> for example, there's a mech war era where you can fire missiles at a target in a certain range
20090820 23:45:33< Crab_> Sapient: no,it won't work in such a general way. we need to pick some 'specific points' at which we want to hijack the ai turn sequence
20090820 23:46:05< Crab_> Sapient: e.g., 'here's a mech war era where you can fire missiles at a target in a certain range' - so, we need to code a specific attack selection algorithm which includes those non-standard attacks
20090820 23:46:06< CIA-62> fendrin * r37979 /trunk/data/campaigns/Legend_of_Wesmere/_main.cfg: 
20090820 23:46:06< CIA-62> LoW: Added crab and bera to the credits.
20090820 23:46:06< CIA-62> Added a translation "_" to the comments in the credits.
20090820 23:46:32< shadowmaster> fendrin: what is the translatable mark for?
20090820 23:46:35< Sapient> right, I know there wouldn't be any general solution but I'm just wondering how hard it would be for a UMC author to hack in that kind of behavior
20090820 23:46:44< shadowmaster> comments are only parsed by the script that generates the wiki Credits page
20090820 23:46:52< boucman> Crab_: I think what Sapient means is that the AI<=>game interface should provide to the AI a way to do actions from the menus
20090820 23:46:54< Crab_> Sapient: this will need me or DK to code a small formula_ai function to trigger a specific item
20090820 23:47:03< fendrin> shadowmaster: they are not shown to the player?
20090820 23:47:04< boucman> only era designers could teach the ai how to actually use these menus
20090820 23:47:07< shadowmaster> fendrin: nope.
20090820 23:47:36< Crab_> Sapient: then, formula_ai formula to use this function, can be used a) as part of recruitment  b) as a candidate action c) as unit formula d) as side formula
20090820 23:47:53< Crab_> those a/b/c/d do not require further engine changes
20090820 23:48:05< Sapient> I see. menu items can be fired like any event
20090820 23:48:19< shadowmaster> fendrin: so remove the translatability mark before the next pot update contains useless strings :P
20090820 23:48:27< Sapient> event name is "menu item {id}"
20090820 23:48:37< fendrin> shadowmaster: yes
20090820 23:50:16< Sapient> probably though, you wouldn't want the AI to fire the menu item (which requires user interaction usually) but rather an event where it modifies the gamestate without user interaction
20090820 23:50:42< Sapient> this way it could re-use some code from the human menu item event
20090820 23:51:22< CIA-62> fendrin * r37980 /trunk/data/campaigns/Legend_of_Wesmere/_main.cfg: LoW: removed the translation marks from the comment attributes in LoW's credits.
20090820 23:51:42< fendrin> esr: Forget what I said about wmllint and the translated comments.
20090820 23:52:20< Sapient> Crab_: so I guess a better question that I should have asked is whether you can fire a WML event as a candidate action
20090820 23:53:12< Sapient> obviously, the AI doesn't need to browse through menus
20090820 23:53:12< Crab_> Sapient: that should be added as formula_ai function (and corresponding callable). afterwards, the answer for your question will be 'yes'
20090820 23:54:12< Sapient> ok, I think that's definitely needed then
20090820 23:54:16-!- shadowmaster [n=ignacio@wesnoth/developer/shadowmaster] has quit []
20090820 23:54:16< Crab_> Sapient: then, as I've noted, there will be several different ways of using that function -  a) as part of recruitment  b) in a candidate action c) in unit formula d) in side formula
20090820 23:54:21-!- shadowmaster [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090820 23:54:57< Crab_> boucman: returning to "should scenarios have an and/or logic for goals, or should they use WML to only set one goal after assessing the situation with formulas"..
20090820 23:55:20-!- YogiHH [n=chatzill@c186064.adsl.hansenet.de] has left #wesnoth-dev []
20090820 23:55:44< Crab_> boucman: 'and/or' logic can be implemented as c++ aspect types. but, I don't know if it'll be needed.
20090820 23:55:51< Sapient> Crab_: how hard would it be to make the AI "teleporter-aware"... for example, with a map of earth that has east/west wrap around teleportation
20090820 23:56:00< Elvish_Pillager> hmm, what does [set_variable] [join] do if you give it an empty array?
20090820 23:56:21< Sapient> Elvish_Pillager: should give you empty string as result
20090820 23:56:35< Elvish_Pillager> ah; I'll need to work around that.
20090820 23:56:48< Elvish_Pillager> 'cause using an empty string for the x= of a filter doesn't limit it to nothing.
20090820 23:56:49< Crab_> Sapient: AI nowadays reuses standard movement code. make the pathfinding work, and the ai will happily direct it's units via 'teleporters'
20090820 23:57:13< Crab_> boucman: I intend to improve LoW AI, and I'll know 'what is needed' on case-by-case basis.
20090820 23:57:59< Crab_> boucman: right now, for LoW 3 , we are thinking of 'try A, then pick best from B,C,D" strategy
20090820 23:58:05< fendrin> Crab_: Please post a diff at pastebin for your low 3 start setting ( the unit you placed at startup )
20090820 23:58:09< Crab_> boucman: so, no or/ands are so far required
20090820 23:59:00< Crab_> fendrin: that units which were on that screenshot ?
20090820 23:59:33< Elvish_Pillager> Sapient: https://gna.org/bugs/index.php?14163
20090820 23:59:47< Sapient> wesbot: bug 14163
20090820 23:59:48< wesbot> Bug #14163  Assigned to: None  Status: None  Priority: 5 - Normal
20090820 23:59:48< wesbot> Summary: [kill] to allow specifying the killing unit
20090820 23:59:48< wesbot> Original submission: If a unit is killed by a [kill] with fire_event=yes and t
20090820 23:59:51< wesbot> his triggers a die event, it currently sets the x2, y2, and second_unit to refer
20090820 23:59:54< wesbot> URL: https://gna.org/bugs/?14163
--- Log closed Fri Aug 21 00:00:08 2009