--- Log opened Sun Dec 07 00:00:05 2008 --- Day changed Sun Dec 07 2008 20081207 00:00:05 yeah, much better than subversion's silly interface 20081207 00:00:42 svn revert is a joke compared to git revert :) 20081207 00:01:19 although I haven't learned yet how to reset files in git (individual files) 20081207 00:01:20 Not to mention that those are totally different things :) 20081207 00:03:57 git checkout -- file 20081207 00:03:58 for instance, to revert a configure.ac change made by kdevelop 20081207 00:04:27 Soliton: oh, many thanks :) 20081207 00:04:36 had been trying to figure that out for months 20081207 00:08:46 wesbot: topic 20081207 00:08:48 -!- wesbot changed the topic of #wesnoth-dev to: 1.4.7 planned for 12.12. | soft stringfreeze for most campaigns: http://www.wesnoth.org/wiki/TextdomainStatus | 1.6 release plan: http://www.wesnoth.org/wiki/Wesnoth1.6ReleasePlan | 56 bugs, 209 feature requests, 7 patches | logs: http://wesnoth.org/irclogs | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com, or for images: http://imagebin.org/index.php?page=add 20081207 00:09:20 -!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev 20081207 00:20:03 baufo * r31324 /trunk/data/multiplayer/ (8 files in 2 dirs): adding 4 mp maps 20081207 00:23:49 what does "adding 4 mp maps" mean?= 20081207 00:24:25 -!- Shadow_Master is now known as Obvious_Master 20081207 00:24:31 that he added 4 multiplayer maps 20081207 00:24:37 -!- Obvious_Master is now known as Shadow_Master 20081207 00:24:48 -!- grzywacz [n=grzywacz@moinmoin/developer/karol] has quit [Remote closed the connection] 20081207 00:25:22 -!- Elvish_Pillage2 [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)] 20081207 00:25:26 -!- Baufo_ [n=thomas@62-47-139-78.adsl.highway.telekom.at] has quit ["Leaving"] 20081207 00:26:42 oh. 20081207 00:27:22 i guess i didn't even consider that someone would add 4 new maps in the same commit 20081207 00:28:06 -!- fnaek [n=fnaek@athedsl-4507772.home.otenet.gr] has joined #wesnoth-dev 20081207 00:28:18 hi santi/kalenz/me/fnaek 20081207 00:28:28 hi Shadow Master 20081207 00:29:09 how's life going? :) 20081207 00:30:34 ok, busy with work. How about you 20081207 00:31:02 buried at home, bored like a rock 20081207 00:31:14 the fnaek should have been implemented everywhere, did I miss something? 20081207 00:31:19 and trying to salvage my addon before xmas 20081207 00:31:34 what's an fnaek? 20081207 00:32:04 an old account I had in an IBM 3270(shows how old I am!) 20081207 00:32:16 IftU will be a nice xmas gift to unsuspecting wesnoth players :) 20081207 00:32:23 I don't know how to implement an "fnaek" :/ 20081207 00:33:08 maybe that's the point. Are you really gettuing IftU mainlined by Xmas? 20081207 00:33:23 no, I'm getting it released in the 1.5.x add-on server for the first time 20081207 00:33:32 Maybe "fnaek" is an opcode in some long forgotten assembly language? :) 20081207 00:33:39 ok, way to go! 20081207 00:33:52 fnaek: were you asking about your nickname change? 20081207 00:34:13 Yes, it should have been implemented everywhere, no? 20081207 00:34:26 no. nobody did it. 20081207 00:34:54 btw aek is the name of a football team, 'fn' was a standard prefix for all accounts in the place I was working at the time 20081207 00:35:11 but I put that in DM as well as in Low trunk 20081207 00:35:11 well, I guess I can do it for you 20081207 00:36:04 thanks 20081207 00:36:56 shadowmaster * r31325 /trunk/data/ (campaigns/Legend_of_Wesmere/_main.cfg core/about.cfg): Added Santi's IRC nick, "fnaek" to credits. 20081207 00:38:25 I tried an svn update from DM but got a message certificate not issued by a trusted authority. Any problem there? 20081207 00:39:04 I think you should be able to tell SSVN "yes, I still want to continue with this" in such situations 20081207 00:39:44 hi fnaek 20081207 00:39:46 otherwise, I haven't seen anything like that ever 20081207 00:40:03 fnaek: commit your nickname to DM too 20081207 00:40:07 *commited 20081207 00:40:35 you should join #wesnoth-umc-dev if you want to dicuss something in detail to avoid bothering the idling people here :) 20081207 00:41:16 yes, but I'll have to split. thanks for the commits 20081207 00:41:24 np 20081207 00:42:54 -!- grrrX [n=_@p57BBE44B.dip.t-dialin.net] has quit ["Ex-Chat"] 20081207 00:45:18 "split" has a very precise meaning in irc. it doesn't mean leaving. 20081207 00:45:55 Yes. It does. split makes lots of people leave :) 20081207 00:46:31 err, well, yeah ;) 20081207 00:46:39 i stand corrected 20081207 00:47:12 you mean netsplit 20081207 00:47:39 yep 20081207 00:48:18 http://www.thefreedictionary.com/split 20081207 00:48:27 5. Slang to leave a place 20081207 00:49:07 oh well 20081207 00:53:34 accouterments <-- from the dwarvish description. My dictionary speaks only of 'accoutrements' 20081207 00:53:53 -!- zookeeper [n=l@dsl-tkubrasgw1-fe9ff800-215.dhcp.inet.fi] has quit [] 20081207 00:53:58 possible en_US/en_UK difference, like scepter vs. sceptre? 20081207 00:54:40 I believe dutch dictionaries try to adhere to UK spelling, so yes 20081207 00:54:49 but which one should we standardise on? 20081207 00:55:15 I think US, since we already have a en_UK/en_GB locale translation 20081207 00:55:22 right 20081207 00:56:11 http://en.wiktionary.org/wiki/accouterment : accoutrement (Common Wealth) 20081207 01:04:44 -!- CIA-8 [n=CIA@208.69.182.149] has quit [] 20081207 01:05:41 -!- fnaek [n=fnaek@athedsl-4507772.home.otenet.gr] has quit [Read error: 110 (Connection timed out)] 20081207 01:08:02 those are a fair amount of maps, I'll have to check it out... 20081207 01:08:05 *Them 20081207 01:08:29 .... thanks for forcing me to recompile everything loonycyborg 20081207 01:10:01 -!- CIA-8 [n=CIA@208.69.182.149] has joined #wesnoth-dev 20081207 01:18:19 -!- Netsplit wolfe.freenode.net <-> irc.freenode.net quits: isaac 20081207 01:19:14 -!- Netsplit over, joins: isaac 20081207 01:20:00 Shadow_Master: I hope that it'll also fix those slowdowns :) 20081207 01:20:26 your full recompile is triggered by the standard CXXFLAGS (CCFLAGS?) change 20081207 01:20:45 And image.hpp change 20081207 01:20:54 why C++98? 20081207 01:21:19 For clarity :) 20081207 01:21:55 ic 20081207 01:23:12 * loonycyborg specially waited with committing flags change to commit it along with a header change 20081207 01:23:46 to avoid forcing people to do unnecessary recompiles :) 20081207 01:38:45 is it just me or does the new naga fighter image have a case of creeping biggerism ? 20081207 01:39:30 depends on what you call "creeping". there was a time on which all units looked like dwarves 20081207 01:40:11 wesnoth’s units tend to grow over time... 20081207 01:40:28 besides, those are... nagas. 20081207 01:40:33 Soon they'll stop fitting in hexes :) 20081207 01:40:54 the former ones were a gift from the 0.x times, so no surprise if they were too small for other factions 20081207 01:40:55 4-hex units... 20081207 01:43:11 it just looks like the lvl 2 is smaller than the lvl 1 20081207 01:43:16 and that looks wrong 20081207 01:43:24 the old nagas had size consistency 20081207 01:43:31 the level 2 will be updated as well. 20081207 01:43:37 as cartoony as they were 20081207 01:43:43 Soliton, thanks :) 20081207 01:45:36 it's called Work-IN-Progress 20081207 01:47:13 Shadow_Master, I am just clarifying ...... 20081207 01:47:48 and I would have been satisified with GSAP or IIRWIIR 20081207 01:49:42 -!- loonycyborg [n=sergey@79.139.246.148] has quit ["Zzzzzzzzzzzzzzzzzzzz"] 20081207 01:50:22 shikadibot: GSAP 20081207 01:50:23 * shikadibot is offended 20081207 01:50:24 shikadibot: GSAP? 20081207 01:50:27 Sorry, I don't know what 'GSAP' means... 20081207 01:51:15 looks like one can crash wesnoth when right clicking exactly when the turn switches. presumably some problem with player_number_ being out of range. 20081207 01:51:41 AI0867, GSAP = go submit a patch 20081207 01:52:11 AI0867, the more aggressive version of PW or patches welcome 20081207 01:55:05 Appleman1234: GSAP would be great, please do so 20081207 01:55:14 but you'll first need some graphics :P 20081207 01:55:32 Shadow_Master, exactly 20081207 01:58:11 <[Relic]> :) 20081207 02:01:29 -!- Sapient [n=sapien-x@c-71-59-39-246.hsd1.ga.comcast.net] has joined #wesnoth-dev 20081207 02:11:00 -!- thespaceinvader [n=chatzill@91.108.177.229] has quit ["ChatZilla 0.9.84 [Firefox 3.0.4/2008102920]"] 20081207 02:13:08 -!- noy [n=Noy@70.70.128.133] has joined #wesnoth-dev 20081207 02:15:54 -!- boucman [n=rosen@159.83.102-84.rev.gaoland.net] has quit [Remote closed the connection] 20081207 02:16:21 -!- BenUrban [n=benurban@unaffiliated/benurban] has quit ["Power failu"] 20081207 02:24:55 -!- BenUrban [n=benurban@c-68-49-10-243.hsd1.md.comcast.net] has joined #wesnoth-dev 20081207 02:25:49 -!- Shadow_Master [n=shadowm@unaffiliated/shadowmaster/x-275924] has quit ["Battle for Wesnoth [www.wesnoth.org]"] 20081207 02:26:09 -!- Shadow_Master [n=shadowm@unaffiliated/shadowmaster/x-275924] has joined #wesnoth-dev 20081207 02:28:01 -!- Shadow_Master [n=shadowm@unaffiliated/shadowmaster/x-275924] has quit [Client Quit] 20081207 02:28:23 -!- Shadow_Master [n=shadowm@unaffiliated/shadowmaster/x-275924] has joined #wesnoth-dev 20081207 02:28:33 -!- Shadow_Master [n=shadowm@unaffiliated/shadowmaster/x-275924] has quit [Client Quit] 20081207 02:29:29 -!- Shadow_Master [n=shadowm@unaffiliated/shadowmaster/x-275924] has joined #wesnoth-dev 20081207 02:29:54 jetryl * r31326 /trunk/src/asserts.hpp: Added a set of assert macros. 20081207 02:30:17 ? 20081207 02:30:19 shikadibot: log 31326 20081207 02:30:40 Revision 31326 (jetryl 2008-12-07 01:29:28 +0000 (Sun, 07 Dec 2008)): 20081207 02:30:40 Added a set of assert macros. 20081207 02:30:40 Web interface URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=31326 20081207 02:30:51 assert(jetryl != crazy); 20081207 02:31:07 Assertion failed: jetryl != crazy 20081207 02:31:12 Aborted (core dumped) 20081207 02:31:28 * Appleman1234 pulls out gdb 20081207 02:31:44 * Shadow_Master suspects he brought those from Frogatto 20081207 02:33:49 he forgot to define __FILE__ for all platforms 20081207 02:34:25 ai0867 * r31327 /trunk/data/campaigns/Under_the_Burning_Suns/units/Desert_Hunter.cfg: Fix botched search/replace by espreon. 20081207 02:34:32 __FILE__ is defined by some compilers, including GCC and MSVC++ AFAIK 20081207 02:34:33 hmm... no I think __FUNCTION__ was the unportable one 20081207 02:35:00 right, there's and A src/asserts.hpp in Frogatto's source code too 20081207 02:35:05 nevermind 20081207 02:35:33 I deserve a prize for my awesome prediction 20081207 02:36:25 -!- Jetrel [n=Jetrel@mn-10k-dhcp1-151.dsl.hickorytech.net] has joined #wesnoth-dev 20081207 02:36:32 so how is a line in stderr.txt better than having the failed assertion appear onscreen? 20081207 02:36:49 er... huh? 20081207 02:37:05 If you want to improve it to do that, or if you have something to do that, please do. 20081207 02:37:12 Dave just told me to commit those. 20081207 02:37:19 abort() does not show the failed assertion onscreen 20081207 02:37:27 :( I'm not trying to step on anyone's toes. 20081207 02:37:56 Sapient: feel free to do whatever you see fit, as far as I'm concerned. 20081207 02:38:02 assert((a)==(b)); //better than abort(); 20081207 02:38:32 I, personally, don't care - I made those for frogatto, and dave+soliton said they'd be useful here. 20081207 02:38:41 Jetrel: I need to talk to youu 20081207 02:39:01 is there any particular reason for the staff placement in units that have staffs? 20081207 02:39:16 staff placement? 20081207 02:39:31 or you could throw an exception with the error description and display it using a handler in game.cpp 20081207 02:39:40 Jetrel: i.e. the Elvish Sorceress has it on her right hand, but most of her body is facing in the player's opposite direction 20081207 02:40:03 Jetryl: but all other units with staff face in the player's same direction and have their staffs in the left hand 20081207 02:40:07 *on 20081207 02:40:10 Shadow-master, that's to make it more dynamic. 20081207 02:40:28 I'm confused about this. 20081207 02:40:32 -!- noy [n=Noy@70.70.128.133] has quit [] 20081207 02:41:15 well, it's just a pattern I noticed with the mages, Elvish magic units and saurian units. Just curious if there was an artistic decission behind that pattern 20081207 02:41:42 Usually, I make the choice on a per-unit basis. 20081207 02:42:01 But with the mages, I was cloning large parts of the mages to each other, so they kinda all inherited it from the L1. 20081207 02:42:45 aha, but in general, I can put a staff on any hand and the sprite should not look inherently awkward? 20081207 02:44:19 Nope. 20081207 02:44:32 er, yes, it won't look inherently awkward. 20081207 02:44:56 Basically, it's _all_ a matter of composition, and of the individual pose of the individual unit. 20081207 02:45:09 2 things to consider. 20081207 02:45:10 iokay 20081207 02:46:19 1] The pose should look anatomically possible, and not have the unit in an uncomfortable position. 20081207 02:46:51 That said, if you can put it in a more "heroic"/"dynamic" pose, it helps. 20081207 02:47:27 Like the difference between having a baseball batter standing there with the bat hanging from one hand down to his ankle (blah), to being "at bat" (dynamic). 20081207 02:47:45 2] Composition is all about balance across lines of symmetry. 20081207 02:48:30 It's balance of "the area of solid masses" versus "empty space" 20081207 02:49:13 it works a lot like a teeter totter, really. 20081207 02:49:45 A small mass further away from a line of symmetry, can balance a large mass close to the line of symmetry (but obviously on the other side of it). 20081207 02:52:03 Jetrel: instead of abort(), please throw an exception from game_errors.hpp 20081207 02:52:34 This will allow it to be displayed by the GUI as well 20081207 02:53:07 Sapient: ... okay. I'll want you to spot me on this, because I don't quite know what I'm doing, having not used them before. 20081207 02:53:57 ok, instead of streaming to std::cerr, you make a std::stringstream and stream to that 20081207 02:54:26 then, std::cerr << mystream.str(); 20081207 02:54:33 But, aren't these supposed to make the game immediately exit? 20081207 02:55:25 then, throw game::game_error(mystream.str()); 20081207 02:56:11 throwing an exception is preferrable because the scope can decide if it wants to handle it, display it to the user, or just continue aborting 20081207 02:56:31 Jetryl: roger 20081207 02:56:43 (oh sorry, I so much prefer Jetryl to Jetrel) :P 20081207 02:57:12 Sapient: actually, because I'm totally clueless about -how- to do these, partly because I'm bad at handling strings, would you care to do the modifications? 20081207 02:57:53 I'd like to look at that afterwards and learn from it. 20081207 02:58:01 * Shadow_Master retreats back into his lair of "artwork" crafting 20081207 02:58:02 I could do that; but I can't right now. Got some company over 20081207 02:58:13 Okay 20081207 02:58:14 I am about to show them Wesnoth, though ;) 20081207 02:58:39 Sapient: one big thing is - I don't have a compile setup for wesnoth right now, so I'm kinda scared of modifying code that I haven't tried somewhere else. 20081207 02:58:42 Jetrel: also, please add the /* $Id$ */ line at the top, and assign the svn:keywords property as "Author Id Date Revision" 20081207 02:59:10 if that complicates you, I or someone else can do that later 20081207 02:59:33 Shadow_Master: uh... please do educate me on what that does. :) 20081207 03:00:29 it adds a nifty line with something like /* $Id: game.cpp 31247 2008-12-03 13:56:40Z loonycyborg $ */ that gets updated every time the file is modified 20081207 03:02:55 every source code file in Wesnoth has that (or at least, that should be the rule) 20081207 03:03:20 -!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev 20081207 03:03:40 What's the second part, the svn:keywords thing? 20081207 03:03:48 a svn property 20081207 03:03:53 That tells svn to actually do that 20081207 03:04:03 How do I set that? 20081207 03:04:14 it's set per-file, btw? 20081207 03:04:20 svn propset svn:keywords "Author Id Date Revision" filename 20081207 03:04:28 yeah, per-file 20081207 03:04:56 if you were using the vanilla CLI subversion client you could teach it to automatically set the correct properties on added files 20081207 03:05:21 Actually, I am using the plain, official CLI client. 20081207 03:05:25 :] 20081207 03:05:29 ah good :) 20081207 03:05:49 okay I'm off to get on an airplane to Australia......won't be around much for three weeks.... 20081207 03:05:52 but I don't expect to have more than an extremely minimal contribution to wesnoth's -code- 20081207 03:06:01 Sirp: nooo 20081207 03:06:06 Sirp: aaargh. 20081207 03:06:09 * Jetrel sobs 20081207 03:06:14 good luck ! 20081207 03:06:22 Sirp: watch out for the drop bears 20081207 03:07:06 Jetrel but , in order to it to work properly, you must as a $Id$ somewhere in the file; we prefer to add it at the top in a C-style comment (/* $Id$ */) 20081207 03:07:17 ahh no the drop bears are something that attractive female tourists are meant to be afraid of...so then Australian men can 'protect' them. ;) 20081207 03:07:32 hehe 20081207 03:10:16 Shadow_Master: thanks 20081207 03:10:32 Now you get to play sensei. :) 20081207 03:11:18 -!- DDR_ [n=chatzill@66.183.125.196] has quit [Read error: 104 (Connection reset by peer)] 20081207 03:11:46 -!- Sirp [n=user@c-76-102-104-121.hsd1.ca.comcast.net] has quit ["leaving"] 20081207 03:11:48 hehe, drop bears. I didn't know about those before reading the wikipedia article now. 20081207 03:12:09 * Shadow_Master hopes nobody touches the AI code during Sirp's absence... 20081207 03:13:31 -!- DDR_ [n=chatzill@66.183.125.196] has joined #wesnoth-dev 20081207 03:14:40 * Shadow_Master off 20081207 03:15:05 for real D:< 20081207 03:16:30 Shadow_Master: just did the commit 20081207 03:20:59 * AI0867 pokes CIA-8 20081207 03:23:19 Rise robot rise? 20081207 03:23:34 -!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has quit ["Hi! I'm a quit message virus vaccine. If you see a quit message virus, don't replace your quit message with it!"] 20081207 03:30:29 -!- CIA-8 [n=CIA@208.69.182.149] has quit [] 20081207 03:32:14 killer poke! 20081207 03:35:57 -!- CIA-8 [n=CIA@208.69.182.149] has joined #wesnoth-dev 20081207 03:37:19 me pokes CIA-8 again 20081207 03:37:22 * AI0867 pokes CIA-8 again 20081207 03:52:25 AI0867: Wesnoth:irc is not a porn movie. 20081207 04:11:10 -!- Psyche^ [n=Psyche@e177229220.adsl.alicedsl.de] has joined #wesnoth-dev 20081207 04:27:41 -!- Patterner [n=Psyche@e177230229.adsl.alicedsl.de] has quit [Read error: 110 (Connection timed out)] 20081207 04:27:41 -!- Psyche^ is now known as Patterner 20081207 04:38:57 http://www.wesnoth.org/forum/viewtopic.php?p=324922#p324922 what's he talking about forum passwords? 20081207 04:47:58 -!- DDR_ [n=chatzill@66.183.125.196] has quit [Read error: 104 (Connection reset by peer)] 20081207 04:58:03 Soliton: maybe he's saying that some people may use the same forum pw and pbl passphrase 20081207 04:58:29 though that's not clear at all 20081207 04:59:02 or he may just be confused about the mp registration 20081207 04:59:07 It isn't? oh... 20081207 05:07:09 -!- CIA-8 [n=CIA@208.69.182.149] has quit [Remote closed the connection] 20081207 05:24:51 -!- DDR [n=chatzill@66.183.125.196] has joined #wesnoth-dev 20081207 05:36:31 -!- CIA-8 [n=CIA@208.69.182.149] has joined #wesnoth-dev 20081207 06:04:03 cya later, have fun. 20081207 06:04:03 -!- Sapient [n=sapien-x@c-71-59-39-246.hsd1.ga.comcast.net] has left #wesnoth-dev [] 20081207 06:27:38 -!- [Relic] [n=[Relic]@adsl-76-229-179-215.dsl.milwwi.sbcglobal.net] has quit ["Leaving"] 20081207 07:40:31 -!- Jetrel [n=Jetrel@mn-10k-dhcp1-151.dsl.hickorytech.net] has quit [Read error: 145 (Connection timed out)] 20081207 07:43:49 'night 20081207 07:44:17 -!- DDR [n=chatzill@66.183.125.196] has quit [Remote closed the connection] 20081207 07:55:49 -!- Jetrel [n=Jetrel@mn-10k-dhcp1-151.dsl.hickorytech.net] has joined #wesnoth-dev 20081207 08:21:32 -!- Appleman1234 [n=Appleman@123.211.191.172] has quit [Read error: 110 (Connection timed out)] 20081207 09:19:52 -!- mordante [n=chatzill@roadie.xs4all.nl] has joined #wesnoth-dev 20081207 09:20:29 morning 20081207 09:20:33 ping Jetrel 20081207 09:21:18 mordante: pong? 20081207 09:21:48 Jetrel I just see your asserts.hpp and I like the idea 20081207 09:22:02 Uh, that's ... 20081207 09:22:02 only I rather see the macros in uppercase 20081207 09:22:20 ;_: seriously folks, just do whatever you want with it. 20081207 09:22:25 I don't care. 20081207 09:22:38 I made it for frogatto, like dave told me to. 20081207 09:23:23 mordante: sorry, sapient was giving me a bunch of instructions on how he wanted it modified, earlier, and I'm ... I don't really have an opinion on it. :) 20081207 09:24:05 I haven't read the log yet ;) 20081207 09:24:08 You know how I'm usually very particular about art? This is the opposite. Go ahead and do anything you want to it. 20081207 09:24:26 I just saw the commit in the log and I have a good usage case for it 20081207 09:24:41 (and I was too lazy to write the macros myself ;) ) 20081207 09:25:24 I'm an opinionated asshole when it comes to art, but in programming, I try to spare myself from getting involved in religious wars. 20081207 09:25:24 -!- mordante changed the topic of #wesnoth-dev to: 1.4.7 planned for 12.12. | soft stringfreeze for most campaigns: http://www.wesnoth.org/wiki/TextdomainStatus | 1.6 release plan: http://www.wesnoth.org/wiki/Wesnoth1.6ReleasePlan | 56 bugs, 209 feature requests, 7 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com, or for images: http://imagebin.org/index.php?page=add 20081207 09:25:36 Same for music. 20081207 09:26:48 -!- yann [n=dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has joined #wesnoth-dev 20081207 09:27:08 mordante: so don't worry - I don't think of that code as being "mine" and I don't desire to have any authorial control over it. 20081207 09:27:37 If you see an improvement, or some change for consistency with our coding styles, by all means do it. 20081207 09:27:44 I still think it's a good idea to ask the author, since s/he may have a different opinion as I have 20081207 09:28:01 mordante: very true. ;) 20081207 09:28:03 -!- Baufo [n=thomas@62-47-140-175.adsl.highway.telekom.at] has joined #wesnoth-dev 20081207 09:28:25 Also, I'm afraid to screw with it in any significant way, -myself-, because I don't actually have a working wesnoth compile-setup. 20081207 09:30:11 that is a big handicap 20081207 09:30:37 how is your programming for frogatto going? 20081207 09:39:15 Actually pretty good. 20081207 09:39:28 I've been doing lots of little oddjobs. 20081207 09:39:45 Although I recently did a big one in adding 2xSaI interpolation. 20081207 09:40:49 I've been doing lots of "petty programming tasks" that dave hasn't gotten around to yet, and have been learning from doing them. 20081207 09:41:17 One item I did was adding the ability to shake the screen, callable from WML. 20081207 09:41:35 what's 2xSaI interpolation? 20081207 09:42:10 So you can have something happen in the game, like a big explosion, and it makes the screen shake - thanks to dave's wizardry, this can happen in tandem with other things in the game - it doesn't "lock" execution. 20081207 09:42:27 Uh, 2xSaI is a special "pixel art scaling algorithm". 20081207 09:43:04 It detects certain patterns in -very- low color count pixel art, which are usually intended to represent lines or edges, and intelligently treats them as such. 20081207 09:43:30 It preserves the "crispness" of nearest-neighbor interpolation. 20081207 09:43:48 sounds like an interesting algorithm 20081207 09:44:11 It's a much better fit for -pure- pixel art games, than "billinear" or "bicubic" interpolation, because it doesn't blur the art. 20081207 09:44:26 Unfortunately, it would not be useful for wesnoth because: 20081207 09:44:41 1] over half of our game - the terrain, is not pixel art. 20081207 09:45:22 2] it's a fixed size increase - it gives you a version 2x the size. If you need something different than that, you need to take that, and then interpolate it again to resize it to something else. 20081207 09:45:43 (Many emulators actually do this, to allow freeform resizing of windows.) 20081207 09:46:24 When I say "interpolate it again", the second pass would be done with some normal scaling algorithm, like bilinear 20081207 09:46:39 3] Wesnoth doesn't resize it's art. 20081207 09:47:22 Frogatto is different, because frogatto does all its art at a native resolution of 400x300. 20081207 09:47:40 The actual art is small, and we upsize it to twice the size. 20081207 09:48:25 This actually makes art resources vastly easier to create, because you're only working with 1/4 of the pixels, and as a bonus, it helps to make it look like the nintendo games of yore. 20081207 09:48:56 (In reality, it's actually _still_, even at 400x300, nearly twice the resolution of a super nintendo. Or maybe 1.5 times.) 20081207 09:49:16 Wait, acutally, I'm looking at a nintendo screenshot right now. 20081207 09:49:29 SNES = 256x223 pixels. 20081207 09:49:43 Frogatto = 400x300 20081207 09:50:38 can't remember it had such a low resolution :/ 20081207 09:50:48 mordante: please forgive my soliloquy. 20081207 09:51:22 an example of an SNES screenshot: http://mmxz.zophar.net/rpg/mana2/images/dragon_battle.gif 20081207 09:51:29 no problem, you answer most of my questions before I finish typing ;) 20081207 09:53:39 frogatto seems to have some really nice features, maybe I should look at it again 20081207 09:53:59 The WML engine is apparently a major improvement over dave's earlier efforts. 20081207 09:54:23 As, in fact, is the graphics engine ... dave has basically done everything right, that I accused him of having done wrong with wesnoth. 20081207 09:55:04 now we only need to get the better engine back in Wesnoth ;) 20081207 09:55:05 But of course, so much of it is incrementally built off of wesnoth, and/or silvertree. 20081207 09:55:55 Well, the big thing he did right in frogatto was basing it on opengl, and having the whole thing be hardware accelerated. 20081207 09:56:16 yes of course, it's hard to get something 100% good the first time 20081207 09:56:23 It's liquid-smooth, and the big thing that's necessary, is that the camera is very fluid as a result. 20081207 09:56:41 yes the OGL makes it not run on my machine 20081207 09:56:54 Wesnoth, by necessity, tries to minimize camera movement. 20081207 09:57:36 Because it only draws what's new. 20081207 09:58:17 indeed, but it also limits the effects we have at the moment ;) 20081207 09:58:45 But frogatto redraws everything in every frame, so not only are there no restrictions on moving the camera around, but it also means that there are no visual redraw artifacts like we occasionally get in wesnoth. 20081207 09:59:04 That said ... it has one ugly restriction.. 20081207 09:59:52 No direct bit-level access to the frame buffer (unless the programmer is intent on castrating performance). Meaning it's hard to do a lot of the tricks they did on SNES games. 20081207 10:00:27 One such trick being, for example, making the screen "wave/ripple" like you're underwater. 20081207 10:00:39 mordante * r31329 /trunk/src/asserts.hpp: Capitalize the macros and add the eol property. 20081207 10:00:49 By sliding the rows/columns back and forth. 20081207 10:01:01 rows/columns of pixels. 20081207 10:01:49 yes that's the disadvantage of OGL, but doesn't OGL have ways to emulate that? 20081207 10:01:53 Shadow_Master next time also explain the svn:eol-style :P 20081207 10:02:27 mordante: OGL has ways to achieve specific effects. Many of these are in the newer, and of course, less widely-supported versions of OGL. 20081207 10:02:51 We're shooting for a pretty basic usage of OGL - we don't do any shaders or anything. 20081207 10:03:00 Not yet, at least, though we have no plans to. 20081207 10:03:07 -!- ettin_ [n=jorda@139.206.219.87.dynamic.jazztel.es] has joined #wesnoth-dev 20081207 10:03:10 Jetrel the svn:eol-style native makes sure if you download a file with svn (regardless of the platform) that the proper end of line character is used 20081207 10:03:39 -!- zookeeper [n=l@dsl-tkubrasgw1-fe9ff800-215.dhcp.inet.fi] has joined #wesnoth-dev 20081207 10:03:50 ah ok 20081207 10:03:56 Can't we all just be stand-over men, and go to redmond and chop off toes until MS changes their EoL to the unix standard? 20081207 10:04:36 you know how many toes they have? ;) 20081207 10:05:11 and mac has been converted without any toe being hurt :) 20081207 10:05:12 mordante: enough to make a lot of McDonald's Burger patties. 20081207 10:05:18 Mmmm... burgers. 20081207 10:06:02 mordante: actually, one nice "platform inconsistency" like that was recently resolved, was that macs have now dropped their own style of Screen Gamma setting, and now use the PC standard in the unreleased OS 10.6. 20081207 10:06:20 Which means WOOHOOO - the whole bloody world will be using the same standard! 20081207 10:06:27 Ding dong the witch is dead! 20081207 10:06:40 not sure whether I want those McMicrosoft burgers 20081207 10:06:59 they might have bugs in them 20081207 10:07:08 :D 20081207 10:07:41 well I think it takes a while before MS will be converted to the same standard 20081207 10:08:15 mordante: give me a pair of scissors, 10 hours, and full diplomatic immunity, and I can solve that problem. 20081207 10:08:36 but I like it that rather mainstream news channels notice that apple gets bigger and MS smaller 20081207 10:09:23 so more people will get aware there are alternatives 20081207 10:09:50 * Jetrel nods. 20081207 10:10:15 * mordante loves this picture http://www.nu.nl/foto_popup.jsp?p=1872775&t=Marktaandeel_Windows_bereikt_historisch_dieptepunt 20081207 10:10:31 -!- boucman [n=rosen@159.83.102-84.rev.gaoland.net] has joined #wesnoth-dev 20081207 10:10:48 it's about the news MS got below the 90% marketshare 20081207 10:10:49 yo 20081207 10:10:54 hi zookeeper 20081207 10:10:59 mordante: aha 20081207 10:11:30 Actually, MS just dipped below 70% browser market share too, recently. 20081207 10:12:03 yes that was the same report OS < 90% browser < 70% 20081207 10:12:57 All we need is a version of linux that can do two things, and it'll spread like wildfire amongst "normal people" - 20081207 10:13:19 one that can 1] Run any normal windows program, especially games 20081207 10:13:28 moin 20081207 10:13:37 uh... I forget what 2] was. 20081207 10:13:40 hi Ivanovic 20081207 10:13:51 Jetrel: 1) is already not possible 20081207 10:14:04 -!- ettin [n=jorda@234.207.219.87.dynamic.jazztel.es] has quit [Read error: 110 (Connection timed out)] 20081207 10:14:14 well Ubuntu already seems to get much acceptance 20081207 10:14:16 that is: you would have to reprogram windows to have this any many problems directly build onto the (not "specified") bugs 20081207 10:14:35 s/problems/programs 20081207 10:14:41 Ivanovic: if someone makes a FLOSS virtualization program that works as well as parallels, it's possible. 20081207 10:14:57 (okay, those programs tend to be a real problem for end users...) 20081207 10:15:05 And all people need to do then is install their "legal" copy of windows into it. 20081207 10:15:14 Jetrel: uhm, you would still be running *windows* inside your linux to run those programs 20081207 10:15:56 Ivanovic: yes, but it solves the problem, because then there would actually be a wide group of people using linux 20081207 10:16:07 hence a market for programs written for the API. 20081207 10:16:36 Jetrel: but the same group of people would also have no problem if the prog was windows only, since they would run it "their usual way" 20081207 10:16:49 so this does not really help with getting progs natively to linux either 20081207 10:17:20 Ivanovic: Really, the problem is that there are several market segments which probably are never going to be OSS, as much as I might like them to be. 20081207 10:17:25 New games are a big one. 20081207 10:17:56 We need some way to attract them to write their software for the linux api, instead of the win32 api. 20081207 10:18:15 Or we have to make the win32 api run. 20081207 10:18:25 and we're back at open standards and using them ;) 20081207 10:18:34 Jetrel, has anyone asked you whether you think these are just fine for mainline use (UtBS, as desert prowler and sentinel), or whether they should be tweaked somehow (and whether you could do it)? 20081207 10:18:35 http://wesnoth-umc-dev.svn.sourceforge.net/viewvc/wesnoth-umc-dev/trunk/Invasion_from_the_Unknown/images/units/elves-wood/trapper.png?revision=2705 and http://wesnoth-umc-dev.svn.sourceforge.net/viewvc/wesnoth-umc-dev/trunk/Invasion_from_the_Unknown/images/units/elves-wood/prowler.png?revision=2704 20081207 10:18:50 I've had about 20-30 people who I could have converted to linux, but -games- were the clincher. 20081207 10:18:55 POSIX and OGL are both standards and if used would make porting much easier 20081207 10:19:52 mordante: jupp 20081207 10:19:58 * mordante wonders how long it will be before the gaming industry kills games on the PCs 20081207 10:19:58 -!- Dragonking [n=dk@dedikerad/dragonking] has joined #wesnoth-dev 20081207 10:19:59 zookeeper: those are just fine for mainline UtBS - they're not perfect, but they don't need to be. 20081207 10:20:07 add openal into this mix, too 20081207 10:20:39 mordante: basically never. Until, at least, normal people stop owning PCs. 20081207 10:20:59 mordante: i think what will happen first is that frustrated users bring their "protected" games back to the stores 20081207 10:21:03 Jetrel well they try really hard with their DRM games 20081207 10:21:08 since they got too many problems making this bananaware run 20081207 10:21:12 Jetrel, ok, great 20081207 10:21:19 mordante: yes, but you can't make mods for console games. 20081207 10:21:20 a great current example is gta4 20081207 10:21:40 Jetrel: publishers might not *want* there to be mods 20081207 10:21:41 yes and spore also took a lot of flaming 20081207 10:22:04 since mods are for free and you got "extra content" this way without paying the publisher extra money 20081207 10:22:07 Ivanovic: well, then they're not going to be very popular, comparatively. 20081207 10:22:08 come on, how many people actually are bothered by DRM? 5%? 20081207 10:22:11 so you are unlikely to buy a new game 20081207 10:22:27 i bet the masses still just don't care about it 20081207 10:22:31 zookeeper: to be honest, i would really get fed up if i had bought gta4 20081207 10:22:36 Jetrel true but some games cripple your CD burning software and uninstalling doesn't seem to help 20081207 10:22:38 first you start the normal installer 20081207 10:22:47 Ivanovic, who cares? you're in that 5%. 20081207 10:22:53 okay, this is normal and while take ~25mins since you install 2DVDs 20081207 10:23:02 then you got to install "rockstar social club" 20081207 10:23:18 then you got to install "games for windows live" 20081207 10:23:25 "click ok to everything" 20081207 10:23:26 for this you need a recent .net 20081207 10:23:38 this you got to find yourself at microsoft 20081207 10:23:41 of course most games of those have little replay value so people don't want to play it again in 5 years ;) 20081207 10:24:00 All I'm saying is that people don't buy into DRM. Even my sister knows what DRM is. 20081207 10:24:08 oh? well, yeah, that's annoying for the average joe too if he needs to go find stuff himself 20081207 10:24:12 okay, if you are through those installed progs, you got to create an account at rockstar social club and games for windows live 20081207 10:24:14 and discover the publisher turned of its DRM servers 20081207 10:24:16 You know how many songs on the average iPod are DRMed? 20081207 10:24:21 Anyone want to guess? 20081207 10:24:26 3% 20081207 10:24:30 and at least games for windows live has to be running while playing *just to be able to save the game!* 20081207 10:25:06 combine this with extreme slowdowns in the game due to the copy protection malfunctioning 20081207 10:25:20 option 2 is go to TPB download the DRM free version and play 20081207 10:25:35 jupp 20081207 10:25:39 mordante: which is actually what most people do. 20081207 10:25:54 and as soon as joe sixpack discovers those games work much better, they no longer buy them 20081207 10:25:54 especially because this one works without the "copy protection related slowdowns" 20081207 10:25:59 Like really - who the hell actually BUYS windows these days? 20081207 10:26:20 everybody how buys a new computer 20081207 10:26:33 Jetrel: companies (who are really checked) and "average joe" as oem version with a new computer 20081207 10:26:34 who* 20081207 10:26:50 But that's just the thing. 20081207 10:27:01 Besides those who are forced - no one else does. 20081207 10:27:16 no but everybody is forced ;) 20081207 10:27:22 I'm not forced. 20081207 10:27:32 mordante: uhm, with some work, it is possible to get a refund for windows 20081207 10:27:36 Basically, the emperor has no clothes. 20081207 10:27:39 well it seems to get easier to buy a PC without OS 20081207 10:27:46 that is i was able to get the money for windows with my table back from lenovo 20081207 10:27:56 Ivanovic yes but still you bought Windows ;) 20081207 10:28:01 ayes, it requires some efford and most people are too lazy for this 20081207 10:28:11 It's the same reason apple tries to make money off the hardware, rather than off their software. Because they _know_ everyone's gonna steal their software. 20081207 10:28:46 yup 20081207 10:28:56 It's funny too, I kind of like it. 20081207 10:29:03 and the funny thing is people do buy OSX updates 20081207 10:29:08 Back to making money by selling the razors, rather than the blades. 20081207 10:30:13 mordante: honestly, I actually don't buy any apple software. The only OS update I paid for was 10.5, and that was pooling with some friends. 20081207 10:30:43 I do buy software from small independent publishers, though. 20081207 10:31:04 Like, I bought this: http://www.panic.com/transmit/ 20081207 10:32:13 Best FTP client I've ever used. And more to the point, I knew the money was going to the guys who actually programmed it, rather than some executives. 20081207 10:32:47 I still think more people buy a new OSX version as a new Windows version 20081207 10:32:54 mordante: probably 20081207 10:33:01 of course for a new Windows version you need to wait ages 20081207 10:33:20 mordante: it helps that they come out regularly, and are priced in more bite-sized pieces. 20081207 10:33:46 For windows they expect you to swallow an entire steak. 20081207 10:34:04 hey that's doing proper marketing ;) 20081207 10:34:12 Vista Supreme - $350. @_@ 20081207 10:34:24 Jetrel: noone buys those versions 20081207 10:34:40 the oem versions are a lot cheaper, more like 80 to 100 bucks 20081207 10:34:42 jup and in the same time apple sold 4 $100 updates 20081207 10:35:04 that is: if a windows user bought a license, it would probably be such a "system builder" edition 20081207 10:35:28 Vista Los Brazos Gigantes - $40000 20081207 10:36:37 vista home premium costs about € 80 20081207 10:36:39 okay, vista ultimate system builder is at 165€ at alternate where osx 10.5.4 is at 110€ 20081207 10:36:42 OEM 20081207 10:36:53 Well, dispensing with some other things, it is nice to see some significant victories happening. 20081207 10:37:06 the "normal" vista ultimate package would cost lovely 290€ 20081207 10:37:21 Modern "non-IE" browsers really are 100% compatible with IE, these days. 20081207 10:37:39 sorry, not 100% 20081207 10:37:55 that is: once MS starts with its activex pages, things will blow up 20081207 10:38:09 Ivanovic: you know what. I've never run into one. In my life. 20081207 10:38:14 We've won. 20081207 10:38:16 (though everyone who loves security deactivates activex in internet explorer) 20081207 10:38:32 Jetrel: the windows update page that you start right out of the start menu is built on it 20081207 10:38:47 Ivanovic: I don't use windows. 20081207 10:38:55 and some pages really do display strangely in eg konqueror or firefox 20081207 10:39:06 I have not seen any in years. 20081207 10:39:13 because the companies coding them really screwed up and used some strange special function 20081207 10:39:27 Jetrel: most bigger companies can not afford this anymore 20081207 10:39:55 Ugh, I don't have time for this. :P 20081207 10:40:02 I'm supposed to be drawing. 20081207 10:40:32 or, uh ... sleeping. 20081207 10:40:32 especially since a lot of companies disable ActiveX due to security holes and rogue applications 20081207 10:40:37 yes, sleeping. 20081207 10:40:40 * mordante is supposed to code 20081207 10:54:47 mordante * r31330 /trunk/src/ (asserts.hpp gui/widgets/settings.cpp): Use the new assert macros and fixed a bug in them. 20081207 10:56:54 * Ivanovic is supposed to wake up and get some tea... 20081207 10:57:12 baufo * r31331 /trunk/data/multiplayer/ (8 files in 2 dirs): setting correct svn properties 20081207 11:10:00 * zookeeper suggest renaming the editor button in the title screen button to "Map Editor" 20081207 11:10:04 -!- Baufo [n=thomas@62-47-140-175.adsl.highway.telekom.at] has quit [Remote closed the connection] 20081207 11:10:23 otherwise every newbie thinks there exists an "editor" instead of just a map editor. 20081207 11:10:33 zookeeper: yes, please do so 20081207 11:11:20 i'm not touching C++ 20081207 11:19:40 zookeeper * r31332 /trunk/data/core/terrain-graphics.cfg: Fixed river ford not transitioning to void. 20081207 11:21:57 -!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev 20081207 11:32:40 -!- loonybot [n=loonybot@79.139.246.148] has joined #wesnoth-dev 20081207 11:32:55 -!- loonycyborg [n=sergey@79.139.246.148] has joined #wesnoth-dev 20081207 11:56:05 -!- grzywacz [n=grzywacz@moinmoin/developer/karol] has joined #wesnoth-dev 20081207 12:01:46 -!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has quit [] 20081207 12:13:00 boucman: what do forum passwords have to do with the issue here? http://www.wesnoth.org/forum/viewtopic.php?p=324922#p324922 20081207 12:13:26 lemme check what I wrote 20081207 12:14:27 ok, I messed up between pbl passwords and game passwords in 1.5 20081207 12:14:31 everyone interested in coming to fosdem 2009: http://www.wesnoth.org/wiki/Fosdem2009 20081207 12:15:05 boucman: ah, ok. 20081207 12:15:10 sry 20081207 12:17:16 * zookeeper is adding forested hills to HttT maps 20081207 12:18:01 elf lover :P 20081207 12:19:04 i didn't say i was only turning existing hills to them :o 20081207 12:23:19 zookeeper * r31333 /trunk/data/campaigns/Heir_To_The_Throne/maps/ (14 files): Made HttT maps use the forested hills terrain. 20081207 12:26:10 soliton * r31334 /trunk/src/titlescreen.cpp: renamed Editor to Map Editor in the titlescreen 20081207 12:27:37 thanks 20081207 12:35:38 zookeeper * r31335 /trunk/data/campaigns/Son_Of_The_Black_Eye/maps/ (11 files): Made SotBE maps use the forested hills terrain. 20081207 12:47:43 ivanovic * r31336 /trunk/data/multiplayer/scenarios/ (2p_Dark_Forecast.cfg 4p_A_New_Land.cfg 8p_Mokena_Prairie.cfg): make sure that teamnames are all set (and rather uniform regarding capitalization) 20081207 12:49:13 hi 20081207 12:50:34 -!- Ivanovic changed the topic of #wesnoth-dev to: 1.4.7 planned for 12.12. | soft stringfreeze for most campaigns: http://www.wesnoth.org/wiki/TextdomainStatus | wesconf@FOSDEM2009 http://www.wesnoth.org/wiki/Fosdem2009 | 1.6 release plan: http://www.wesnoth.org/wiki/Wesnoth1.6ReleasePlan | 56 bugs, 209 feature requests, 7 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com, or for images: http://imagebin.org/index.php?page 20081207 12:52:31 zookeeper * r31337 /trunk/data/campaigns/Sceptre_of_Fire/maps/ (6 files): Made SoF maps use the forested hills terrain. 20081207 12:53:31 -!- Baufo [n=thomas@62-47-140-175.adsl.highway.telekom.at] has joined #wesnoth-dev 20081207 12:54:18 -!- EdB [n=EdB@52.101.196-77.rev.gaoland.net] has joined #wesnoth-dev 20081207 12:54:29 ivanovic * r31338 /trunk/data/multiplayer/scenarios/ (2p_Dark_Forecast.cfg 4p_A_New_Land.cfg): tabs -> spaces 20081207 12:54:50 boucman: coming to fosdem? then please add yourself in the wiki, so that the list grows a little 20081207 12:54:51 http://www.wesnoth.org/wiki/Fosdem2009 20081207 12:55:14 mordante: do not forget to kidnap neorice 20081207 12:55:16 ;) 20081207 12:56:26 done 20081207 12:58:03 I first need to convince him to give his address ;) 20081207 12:58:08 :) 20081207 12:58:26 mordante: have you already asked him if he would like to come over? 20081207 12:58:49 no not yet, first want to know when I got 20081207 12:58:54 go* 20081207 12:59:16 when we go to bruegel yh, at least one of us has to be there before 4pm 20081207 12:59:35 that is: for me this will be damn difficult when picking DK up at dortmund airport at 1pm 20081207 13:01:56 several people seem to think that the mouse scrolling area is too small now that scrolling stops when the pointer exits the window, so i think it needs to be enlarged. maybe double it, seems to be something like 5px now. 20081207 13:05:21 yes it's 5 pixels 20081207 13:05:30 what do you think is a good threshold? 20081207 13:05:54 test it with 10 for the moment 20081207 13:05:56 or should we make it a hidden preference? 20081207 13:06:01 this should be possible to hit 20081207 13:06:40 i'd go with 10 too, but of course it'd be nice if it wasn't actually hardcoded.. 20081207 13:07:00 okay, i am afk for an hour (or something like this), cu 20081207 13:07:08 bye Ivanovic 20081207 13:16:59 might also be nice if mouse scrolling accelerated and decelerated a bit just like how scrolling presumably works elsewhere (when scrolling to unit, speaker, etc) 20081207 13:19:24 -!- Baufo [n=thomas@62-47-140-175.adsl.highway.telekom.at] has quit [Remote closed the connection] 20081207 13:20:07 for now I only add the preference and let it default to 10 20081207 13:21:02 -!- EdB [n=EdB@52.101.196-77.rev.gaoland.net] has quit ["Konversation terminated!"] 20081207 13:22:25 cool 20081207 13:35:33 hmm not sure whether 10 is enough, still feels rather small 20081207 13:43:43 -!- loonycyborg [n=sergey@79.139.246.148] has left #wesnoth-dev ["In Soviet Russia, channel /parts YOU!!"] 20081207 13:44:14 -!- loonybot [n=loonybot@79.139.246.148] has quit [Remote closed the connection] 20081207 13:46:29 zookeeper ^ 20081207 13:46:37 mordante * r31339 /trunk/ (5 files in 2 dirs): 20081207 13:46:37 Improved the map scrolling feature. 20081207 13:46:37 After r30918 fixed the scrolling outside the window, several people noticed the 20081207 13:46:37 threshold of 5 pixels is rather small. Added a hidden preference named 20081207 13:46:37 'scroll_threshold' which is used to determine the number of pixels. The 20081207 13:46:38 threshold also defaults to 10 instead of 5, more testing is needed to get a 20081207 13:46:41 zookeeper ^ 20081207 13:46:42 good default value. 20081207 13:58:26 afk now 20081207 14:03:51 re 20081207 14:26:03 -!- EdB [n=EdB@52.101.196-77.rev.gaoland.net] has joined #wesnoth-dev 20081207 14:33:32 -!- miq_ [n=miq@krlh-5d856e4c.pool.einsundeins.de] has joined #wesnoth-dev 20081207 14:41:11 -!- Baufo [n=thomas@62-47-140-175.adsl.highway.telekom.at] has joined #wesnoth-dev 20081207 15:02:56 -!- EdB [n=EdB@52.101.196-77.rev.gaoland.net] has quit ["Konversation terminated!"] 20081207 16:09:33 mordante * r31340 /trunk/src/gui/widgets/grid.hpp: 20081207 16:09:33 Import find_widget from twidget. 20081207 16:09:33 This loads the templated version, which can be handy at times. 20081207 16:09:37 mordante * r31341 /trunk/src/gui/widgets/ (container.cpp container.hpp grid.cpp grid.hpp): Enable the horizontal scrollbar. 20081207 16:12:10 -!- loonycyborg [n=sergey@79.139.246.148] has joined #wesnoth-dev 20081207 16:12:11 -!- loonybot [n=loonybot@79.139.246.148] has joined #wesnoth-dev 20081207 16:20:18 So, does someone still get animation slowdowns after I reverted Sirp's image caching changes? 20081207 16:20:53 I.e. did I fix same slowdowns as everyone else have? :) 20081207 16:21:02 no opinion :) 20081207 16:21:38 loonycyborg: I'm no longer seeing slowdown. 20081207 16:21:51 esr: Great. 20081207 16:22:53 The annoying thing here that Sirp not only bundled those image caching changes with allocator changes 20081207 16:22:56 -!- BenUrban_ [n=benurban@unaffiliated/benurban] has joined #wesnoth-dev 20081207 16:23:09 he even didn't mention them in commit message.. 20081207 16:23:11 -!- zookeeper [n=l@dsl-tkubrasgw1-fe9ff800-215.dhcp.inet.fi] has quit [] 20081207 16:35:20 -!- Elvish_Pillage2 [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev 20081207 16:36:51 -!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)] 20081207 16:38:56 -!- BenUrban [n=benurban@unaffiliated/benurban] has quit [Read error: 110 (Connection timed out)] 20081207 16:39:30 I note that I am using the new poolalloc code and it sseems to work. 20081207 16:40:09 -!- Baufo [n=thomas@62-47-140-175.adsl.highway.telekom.at] has quit [Read error: 110 (Connection timed out)] 20081207 16:40:45 -!- Baufo [n=thomas@62-47-135-69.adsl.highway.telekom.at] has joined #wesnoth-dev 20081207 16:42:18 I am, however, getting occasional audop dropouts. 20081207 16:42:35 s/audop/audio/ 20081207 16:46:11 * loonycyborg always has his speakers turned off 20081207 16:47:15 sweet, wesnoth mentioned on http://www.opensourcewindows.org/page2.php 20081207 16:47:52 -!- ettin_ [n=jorda@139.206.219.87.dynamic.jazztel.es] has quit [Read error: 60 (Operation timed out)] 20081207 16:49:46 Rhonda: Is image next to wesnoth incorrectly resized for you too? 20081207 16:50:29 I have finished some macros and hacks to recall an ai side. Shall I commit them to core/macros/side-utils.cfg? 20081207 16:54:07 -!- zookeeper [n=l@dsl-tkubrasgw1-fe9ff800-215.dhcp.inet.fi] has joined #wesnoth-dev 20081207 16:58:27 loonycyborg: Yes 20081207 17:01:14 zookeeper: may I commit new macros to side-utils.cfg? 20081207 17:02:41 loonycyborg yes that's why I also wonder whether or not it was committed by accident 20081207 17:05:59 -!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev 20081207 17:06:29 fendrin, if they're good macros, sure :p 20081207 17:07:10 zookeeper: would you review them for me, please? 20081207 17:07:54 sure 20081207 17:19:50 looks like there are some OOS bugs in 1.5.6. 20081207 17:21:17 might have something to do with units appearing to have full MPs when they used them all up and thus getting rest healing when they shouldn't. 20081207 17:25:21 oh, if that means that bug 12467 affects mainline, i'm happy :) 20081207 17:40:23 -!- wesbot changed the topic of #wesnoth-dev to: 1.4.7 planned for 12.12. | soft stringfreeze for most campaigns: http://www.wesnoth.org/wiki/TextdomainStatus | wesconf@FOSDEM2009 http://www.wesnoth.org/wiki/Fosdem2009 | 1.6 release plan: http://www.wesnoth.org/wiki/Wesnoth1.6ReleasePlan | 57 bugs, 210 feature requests, 7 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com, or for images: http://imagebin.org/index.php?page 20081207 17:42:16 soliton * r31342 /trunk/src/asserts.hpp: fixed several inconsistencies 20081207 17:42:56 Jetrel: you probably want to backport this version back to frogatto. 20081207 17:46:14 -!- BenUrban_ is now known as BenUrban 20081207 17:47:05 probably one of the last backports :) 20081207 17:49:02 ? 20081207 17:50:43 btw Jetrel I also did a minor fix to asserts.hpp 20081207 17:52:04 esr * r31343 /trunk/data/multiplayer/scenarios/ANL_utils/ANL_general_macros.cfg: Remove an ANL macro that simply duplicates a mainline definition. 20081207 17:59:32 fendrin * r31344 /trunk/data/core/macros/side-utils.cfg: Added macros for handling persistent ai sides. 20081207 18:00:05 zookeeper: please go after r31344 20081207 18:01:44 esr * r31345 /trunk/data/multiplayer/scenarios/ (4 files): canrecruit=yes is better style than canrecruuit=1 20081207 18:08:21 baufo * r31346 /trunk/data/multiplayer/maps/2p_Thalins_Stronghold.map: removing off-map villages to avoid potential confusion 20081207 18:08:50 -!- Espreon [n=Espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev 20081207 18:10:04 Hello people. 20081207 18:10:53 hi Espreon 20081207 18:11:08 Did I miss anything? 20081207 18:11:18 Did they kill the slowdown yet? 20081207 18:11:30 Yes. 20081207 18:12:05 Jetrel: ping? 20081207 18:12:19 They were caused by possibly unintended changes to image caching by Sirp. 20081207 18:12:38 Yay. 20081207 18:12:55 Now we can all rejoice. 20081207 18:28:12 Baufo: Are you the Mines of Mallorod maintainer? 20081207 18:32:12 fendrin: I've had an idea for DM. I think Delf's acquisition of the Staff of Power should have more drama to it. While I was showering I came up with a little vignette in which the chamber where the Staff lies is "the tomb of An-Khurusar, greatest of all mages, who bestrode Irdya in the morning of the World, and sleeps now in a death beyond death..." Delf then has to answer questions from... 20081207 18:32:14 ...a tomb guardian before being allowed to take up the Staff. 20081207 18:34:47 esr, kind of, yes 20081207 18:35:45 wmllint is issuing a warning: "../../data/multiplayer/maps/4p_Mines_of_Mallorod.map", line 5: warning, fog in map file 20081207 18:36:35 hmm, I'll look into this 20081207 18:36:37 UI don't know what that means, but the fact that sommeone asked me to test for it in wmllint means someone tyhought it was a problem. 20081207 18:40:15 esr: What if the player gives a wrong answer? 20081207 18:41:47 Defeat. The purpose isn't to be a real puzzle, just to attach some interesting story to simething that is otherwise "ho-hum, Delf just granbbed the most powerful magical artifact in existence like a sandwich off a lunch ciunter". 20081207 18:42:03 Defeat and delete saves. 20081207 18:42:15 Of course if you can come up with something more interesting... 20081207 18:42:41 s/lunch ciunter/lunch counter 20081207 18:43:43 We could have it spawn Yetis with Fire Dragons surrounding the Yetis... 20081207 18:43:52 ... around the player. 20081207 18:46:21 I am going to make Olurf's side persistent. But I don't know how to handle the player. Should he be told that Olurf's side is persistent? 20081207 18:47:20 At the moment the recall macro trys to look like normal recruiting. 20081207 18:47:23 You could put "Until we meet again" in the dialog. 20081207 18:47:41 baufo * r31347 /trunk/data/multiplayer/maps/4p_Mines_of_Mallorod.map: using correct terrain string for void 20081207 18:48:34 esr: I can't dialogue it because of the string freeze. And I don't want that. I will have to talk about game rules which is a bad idea between a dense story. 20081207 18:49:31 Guess there's not much we can do, then, other than leave FIXME comment there for 1.7. 20081207 18:49:37 The question is: Just recall the dwarf before the game starts? Like the loyal units that are recalled for Kalenz. The player would notice that easier. 20081207 18:50:13 Could work. I don't have a strong opinion about the matter. 20081207 18:50:36 I am happy with that new feature. I also want to make the side of Galtrid persistent. 20081207 18:50:50 This gives the player a reason for taking care of the ally. 20081207 18:52:18 Seems reasonable. 20081207 18:52:58 Shall I write that vignette around the Staff? Do you have any other suggestions about it? 20081207 18:53:27 Yes, write it. 20081207 18:56:46 esr: Give a hint in the speech of the oracles to the new content. 20081207 19:01:30 * loonycyborg wonders how did that staff end up there and why didn't anyone take it before 20081207 19:01:34 fendrin: OK, I have a few things to do first, though. wmllint is throwing up a mess of errors. 20081207 19:02:40 fendrin * r31348 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/03_Kalian.cfg: LoW 03: Adjusted the ai parameters and redid village ownership assignments. 20081207 19:04:19 fendrin * r31349 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/04_Elvish_Treasury.cfg: LoW 4: redid starting villages. 20081207 19:06:25 -!- fnaek [n=fnaek@athedsl-4496785.home.otenet.gr] has joined #wesnoth-dev 20081207 19:06:38 hi fnaek 20081207 19:06:43 hi fabi 20081207 19:07:04 fnaek: I am working on making Olurf's party persistent. 20081207 19:07:11 do we have a problem with Low scenario 2? 20081207 19:07:37 fnaek: I believe not. I did playtest it on Hard yesterday, It worked good for me. 20081207 19:08:14 I also posted the replay, there was no problem; if anything the trolls did not target the elves enough 20081207 19:09:46 fnaek: Maybe we avoid that by making the dwarves allies of the trolls. 20081207 19:10:15 Until a troll steps on dwarven land of course. 20081207 19:10:58 and the avoid would disallow them going after the elves unless they cross, right? 20081207 19:11:07 yes 20081207 19:11:16 -!- thespaceinvader [n=chatzill@91.108.177.229] has joined #wesnoth-dev 20081207 19:11:19 Makes sense. 20081207 19:12:13 Hello thespaceinvader. 20081207 19:12:26 hey espreon 20081207 19:12:41 I love your Saurian Augur portrait. 20081207 19:13:19 thanks =) 20081207 19:13:46 fnaek: That would also make the scenario more a challenge. Currently you only have to be far away enough to cause the troll's ai to go after the dwarves. What triggers them to attack. 20081207 19:14:34 fnaek: after the change this is more difficult to archieve. 20081207 19:15:32 If trolls will be allied with dwarves, they're highly unlikely to cross the river. 20081207 19:15:48 Unless player crosses it too.. 20081207 19:15:54 loonycyborg: yes, that's intended. 20081207 19:16:09 loonycyborg: the player has to lure them there. 20081207 19:16:15 It's intended to make player cross the river? 20081207 19:16:32 loonycyborg: no it is intended to lure the trolls to go there. 20081207 19:16:38 without crossing it. 20081207 19:16:47 how does the player do that? 20081207 19:16:52 How could player lure trolls there without crossing? 20081207 19:17:28 place a unit in the river, the trolls chooses the land to gain better terrain bonus. bingo 20081207 19:18:03 That's unlikely to work since trolls are slow in water.. 20081207 19:19:01 That is how it was before I rewrote that scenario. It happenend often enough by accident. 20081207 19:36:00 esr * r31350 /trunk/data/campaigns/The_Rise_Of_Wesnoth/scenarios/09_Fallen_Lich_Point.cfg: Replace an obsolete image with the correct macro. 20081207 19:36:51 fendrin, ok, looking at your macros now 20081207 19:36:57 We could also employ a [target] ai thing. 20081207 19:37:26 fnaek: we already have one. 20081207 19:37:56 but it's adjusted quite low. I gave elves a target value of 1.5 20081207 19:38:07 fendrin, instead of the ai turn event you now use, i'd suggest making a prerecruit event instead, which randomly replaces the recruited unit with a recalled one (and of course refunds the recruit and checks if there's enough gold to recall) 20081207 19:38:59 zookeeper: good plan. I will work on that. 20081207 19:41:18 it should be pretty simple to do 20081207 19:41:40 -!- fendrin [n=fabi@e179206158.adsl.alicedsl.de] has quit [Remote closed the connection] 20081207 19:42:53 esr, r31350 ain't good 20081207 19:43:54 he gets the leader crown overlay because he doesn't get canrecruit=yes initially, because it's annoying when the view scrolls to him at the start of his turn when he's stoned 20081207 19:44:01 he's not supposed to get the hero icon 20081207 19:44:18 zookeeper: Is there a better fix? 20081207 19:44:28 -!- fendrin [n=fabi@e179206158.adsl.alicedsl.de] has joined #wesnoth-dev 20081207 19:45:31 fendrin r31344 looks odd [+side] around line 255 20081207 19:45:59 esr, to prevent scrolling? well, yeah, an event hack, but that'd be even uglier :p 20081207 19:46:22 hmh, actually i have a core macro for that. should work fine in a campaign 20081207 19:46:42 I'd settle fthe right image name at that spot, but I don't know what it is. 20081207 19:46:52 mordante: what is wrong with it? 20081207 19:47:07 is the + sign really wanted? 20081207 19:47:28 esr, you're saying the image path was wrong? 20081207 19:47:35 mordante: I believe it is necessary. 20081207 19:48:13 [+tag] inserts the contents to the previous [tag] 20081207 19:48:21 ok wasn't sure, I'm not too familiar with the + syntax 20081207 19:48:34 zookeeper: Jetryl removed a "obsolete" image without running wmlscope to check for breakage, a mistake about which I shall have words with him. 20081207 19:49:30 i know 20081207 19:49:58 Is there another image we should plug in there? 20081207 19:50:46 -!- Espreon [n=Espreon@c-76-117-239-33.hsd1.pa.comcast.net] has quit ["ChatZilla 0.9.84 [Firefox 3.0.3/2008092417]"] 20081207 19:50:56 (He also broke two campaigns by removing "obsolete" lightning-bolt images. I'm trying to figure out how to get those back now.) 20081207 19:51:10 no, everything was presumably fine there until your commit 20081207 19:51:33 eh, i'll replace the mainline references to the bolts when i get around to it, that's no problem 20081207 19:51:55 i've been thinking i should just do a unique effect for the storm tridents 20081207 19:52:10 zookeeper: No, it *wasn't* fine. The image reference was broken! 20081207 19:52:25 how? 20081207 19:52:40 overlays=misc/leader-crown.png AFAIK 20081207 19:53:26 Jetryl removed that image, apparently. And the reference broke. I didn't change that line because I felt like it. 20081207 19:53:42 uh, i have that image here just fine.. 20081207 19:54:04 Probably from before he did the celettion. 20081207 19:54:21 deletion, I mean. 20081207 19:54:45 mordante * r31351 /trunk/src/gui/widgets/ (23 files): 20081207 19:54:45 First steps for the new drawing engine. 20081207 19:54:45 The new engine draws in layers like the units and terrains. The engine 20081207 19:54:45 is not ready for genaral usage and is enabled by adding -DNEW_DRAW to 20081207 19:54:45 the compiler. Most normal widgets (except the minimap) already work. 20081207 19:54:46 Scrollbar containers don't work yet, since they need a redesign. 20081207 19:55:29 May I commit wml code that causes a engine bug for debug reasons? 20081207 19:55:45 that image hasn't been touched by anyone, check the svn logs 20081207 19:56:57 s/touched/removed 20081207 19:58:49 -!- fnaek [n=fnaek@athedsl-4496785.home.otenet.gr] has quit ["using sirc version 2.211+KSIRC/1.3.12"] 20081207 19:59:11 zookeeper: If that's so, I can't imagine why wmlscope would have started compaining about it. 20081207 19:59:59 fendrin * r31352 /trunk/data/campaigns/Legend_of_Wesmere/ (4 files in 2 dirs): 20081207 19:59:59 LoW: Made the side of Olurf persistent. 20081207 19:59:59 This commit causes the engine to drop out with an assertion error when LoW reaches Scenario 7. 20081207 20:00:14 esr, bug :)? 20081207 20:04:08 -!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev 20081207 20:05:47 I'm off bye 20081207 20:06:08 -!- mordante [n=chatzill@roadie.xs4all.nl] has quit ["ChatZilla 0.9.84 [Firefox 3.0.4/2008102920]"] 20081207 20:07:19 I haven't actually changed wmlscope in three weeks, so this seems unlikely. 20081207 20:08:29 zookeeper * r31353 /trunk/data/core/macros/utils.cfg: Fixed the NO_SCROLL_TO_AI_LEADER macro. 20081207 20:37:27 zookeeper: leader-crown is not present in my checkout. 20081207 20:38:51 esr, it's in the main images dir, not in core 20081207 20:39:30 Aw, fer cripes sake...WML isn't supposed to refer to stuff in there! 20081207 20:39:48 err, wmlscope doesn't try to match paths to those images? well, that explains it 20081207 20:39:57 Of course that's going to break WMLscope. It's a bug. 20081207 20:40:53 (i've always said it's silly to try to make a hard distinction between images used from C++ and images used from WML) 20081207 20:41:12 zookeeper: You don't remember, do you? The whole point of data/core/images was so wmlscope wouldn't constantly be throwing up false negatives for stuff that only the C++ code references. 20081207 20:41:56 If we *don't* make that distinction, my auditing tools break. 20081207 20:42:12 esr: Make them parse C++ to :P 20081207 20:43:14 esr: wmllint isn't happy about my +side tag. 20081207 20:43:59 fendrin: What's the error message? 20081207 20:44:14 "../../data/core/macros/side-utils.cfg", line 258: [side] without type attribute 20081207 20:44:26 esr: http://www.gccxml.org/HTML/Index.html <- wonder if there are python bindings for that :) 20081207 20:45:08 -!- isaac [n=isaac@debian/developer/isaac] has quit [Read error: 60 (Operation timed out)] 20081207 20:45:08 -!- Espreon [n=Espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev 20081207 20:45:30 Wait.. It doesn't need python bindings. Python already has XML parser.. 20081207 20:46:03 shadowmaster * r31354 /trunk/ (changelog src/dialogs.cpp src/dialogs.hpp): 20081207 20:46:03 All dialogs using the units_list_preview_pane and 20081207 20:46:03 unit_types_preview_pane facilities (create-unit, recall-unit, 20081207 20:46:03 recruit-unit and unit-list) display the selection's race. 20081207 20:46:44 fendrin: Wrap it in no-validate comments. I'll put better handling of this on my to-do list. 20081207 20:46:55 -!- isaac [n=isaac@debian/developer/isaac] has joined #wesnoth-dev 20081207 20:47:10 shadowmaster * r31355 /trunk/ (changelog src/widgets/progressbar.cpp): 20081207 20:47:10 Port alink's loadscreen progress-bar shading to the old GUI progressbar 20081207 20:47:10 widget code. 20081207 20:48:21 esr: okay 20081207 20:48:34 * loonycyborg wonders why he wasn't slapped with a fish for his silly ideas.. 20081207 20:50:35 shadowmaster: I am annoyed with you. Whernever you make an image-reference-change to mainline WML, you sough run "make" in data/tools to make sure you haven't broken anything. Otherwise I end up having to clean up a mess I didn't make. You and Jetrel have made such messes recently. 20081207 20:51:05 * esr was saving a fish for ShadowMaster, and a bgger one for Jetrel. 20081207 20:51:10 because the tools must addapt to the man, not the man to the tools 20081207 20:52:04 esr, yes, i remember. and exactly because of cases like this i was saying back then that the checking should be smarter so we wouldn't need to make that split :) 20081207 20:52:33 otherwise it sounds like a MS typical UI design in which the user must adapt to alien habits from the applications 20081207 20:52:35 No, you must adapt a little to the tools, otherwise the mainline maintainence overhead goes through the roof. I am pretty rapid in responding to wmltools change requests, but *they need to be made*. 20081207 20:52:40 but anyway, this particular case can be solved using my event hack i guess 20081207 20:53:21 okay , then you can forget about me touching the mainline WML ....ç 20081207 20:53:49 zookeeper: If there were a way to make the checking "smarter" I would have done it. The only way out was an explicit ignore list, which would have been an even bigger pain in the ass. 20081207 20:54:32 I can also resign from my position if you want , but I'm certainly annoyed that I must fufill a tool's and its creator's whims 20081207 20:54:49 that's all for now, now off 20081207 20:54:53 With thirteen campaigns in mainline these tools *have to work*. Otherwise we'll be smamped. 20081207 20:55:07 * loonycyborg does some real research and decides that GCC-XML is totally inadequate for this particular purpose 20081207 20:56:37 esr: the wmllint tool is a workaround not having an adequate interpreter for wml. I don't like it either for that reason. 20081207 20:57:33 * Espreon is scared. 20081207 20:58:26 fendrin: Yeah, but we ghave to have automated checking and it has to work. The manual overhead was tolerable when there were only a handful of campaigns. It isn't now. 20081207 20:59:07 the auutomated checking is important for you and= 20081207 20:59:29 esr: I do agree. But can't python be fast enough to replace the ingame wml evaluation? I would like to have a tool that is in sync with the game. 20081207 21:01:05 fendrin: The problem there would be (a) specifying the actual semantics of the ingame eveluation, and (b) keeping them in sync as the C++ core changes. Hideous problem. Probably intractable. 20081207 21:01:39 (Trust me, I've thought about this a lot. The design of wmllint and wmlscope is not random or whimsical.) 20081207 21:01:44 esr: I want to replace the c++ core wml part with an external python tool that is a wml interpreter. 20081207 21:02:00 if the hack didn't work in-game, be annoyed with me. if it does work, then ... meh 20081207 21:03:01 fendrin: I heartily agree with that design direction. But until you can actually do it, wmllint and wmldcope are still needed. 20081207 21:03:04 it would be possibly better to concentrate efforts in libaryzing the C++ WML engine and then reusing it for a new checking tool 20081207 21:03:39 Shadow_Master: aggreed. That is also a good solution. 20081207 21:03:42 * loonycyborg agrees with Shadow_Master 20081207 21:03:53 you'd kill most of the tools maintenance overhead 20081207 21:04:09 and make WML reusable to other games 20081207 21:04:32 Until one of those things are done, we need wmllint and wmlscope to work -- otherwise we're going to ship buggy WML. 20081207 21:04:51 esr: agreed 20081207 21:04:54 esr: sure. 20081207 21:05:13 still it could/should be a goal to make the wml parser of the game the checking and conversion tool 20081207 21:05:43 Ivanovic: I would like to volunteer to such a project. 20081207 21:05:57 That means accepting a few small constraints that are artificial and due to my tools. I try to make those is minimal as I can, there is no "whim" involved here. 20081207 21:06:04 fendrin: just work on it, noone will hold you back 20081207 21:06:21 esr: your tools for itself really are great 20081207 21:06:24 Ivanovic: I believe I can't even handle the planing phase alone. 20081207 21:06:42 but having something closer to the games engine might reduce the workload with the tools themselves 20081207 21:06:51 fendrin: maybe esr can give you a helping hand 20081207 21:06:58 esr: I don't see that someone asserts that those constraints are whimsical. 20081207 21:07:06 since he has *lots* of experience with tools 20081207 21:07:17 or which part of the planning phase do you mean? 20081207 21:07:39 loonycyborg: Shadow_Master apparently thought "whims" were involved. 20081207 21:07:40 Ivanovic: first, we would need a language specification of wml. 20081207 21:07:51 -!- Espreon [n=Espreon@c-76-117-239-33.hsd1.pa.comcast.net] has quit ["ChatZilla 0.9.84 [Firefox 3.0.3/2008092417]"] 20081207 21:07:54 hehe, that is the most fun there is, isn't it? 20081207 21:08:23 Is it? 20081207 21:08:23 hmm, at least partly we got some kind of specification with the WML reference in the wiki 20081207 21:09:04 not sure how accurate this is regarding the lovely corner cases, but at least the basics are well covered 20081207 21:09:22 beside this: the specification is the existing sourcecode 20081207 21:09:24 ;) 20081207 21:09:41 A specification like in a functional language. 20081207 21:09:59 Look at this crap: http://rafb.net/p/wxA4j067.html 20081207 21:10:36 a formal specification of WML is probably entirely doable, except for terrain graphics WML :] 20081207 21:11:25 This is what I am presently dealing with because WML maintainers aren't checking with these ttools before we commit. With all those bugs and/or false positives, there's no way I can certify that our mainline WML is in shape to ship! 20081207 21:11:48 esr: I am going to fix the errors right now. 20081207 21:12:07 esr: the ones in LoW and UtbS 20081207 21:12:40 fendrin: Thank you. 20081207 21:13:29 zookeeper: We could start a wiki page about a formal wml specification. 20081207 21:13:59 fendrin: In the future, check your work before committing, please? Constantly sifting through this sort of muck gets wearying. 20081207 21:15:12 I spend a significant portion of my Wesnoth time chasing stuff like this. 20081207 21:17:09 esr: I am sorry. But I don't have the same output from wmllint. Maybe I should do an upgrade. 20081207 21:17:37 Or svn up at least. 20081207 21:20:42 we could setup a webpage with wmllint output that gets updated automatically so those that can't check for themselves can do that at least after commit. 20081207 21:21:49 Soliton: not needed in my opinion. It's just a new rule commit only code that is wmllint silenced. If wmllint has a bug silence it. 20081207 21:22:11 That will cause esr to have less feedback. 20081207 21:22:56 if by silence you mean turn it off for part of the WML that seems counter productive. 20081207 21:23:52 Soliton: but that is mostly the fix for my code style. wmllint isn't very fendrin compatible. 20081207 21:24:01 :-) 20081207 21:24:42 fendrin: I'd like to improve that. 20081207 21:25:17 such a webpage would actually be useful. i might check what wmllint says of my commits then ;) 20081207 21:25:45 It could probably be written as part of the commit hook. 20081207 21:26:07 i don't think we have that kind of access. 20081207 21:26:19 I couldn't do it myself, though, I don't have the right kind of server access. 20081207 21:26:41 You'll need access to gna to hack commit hooks.. 20081207 21:26:54 Right. 20081207 21:27:39 esr: "../../data/campaigns/Legend_of_Wesmere/scenarios/07_Elves_last_stand.cfg", line 455: unbalanced [dwarf] closed with [/unstore_unit]. 20081207 21:27:44 well, i'd imagine it'd work so that when the page is accessed, wmllint is run and the output saved somewhere. then if it's accessed again but no new commits have been made since the last time, it just displays the previous output. 20081207 21:27:54 (or if the output is identical to the previous one) 20081207 21:28:15 Running wmllint is *expensive* 20081207 21:28:28 yeah, that would be a bit much, i think. 20081207 21:28:37 yeah, so running it after every commit would be pretty insane 20081207 21:28:50 (in case anyone actually suggested that) 20081207 21:29:02 After every commit no, on every page access yes. 20081207 21:29:14 Hm, evey commit on the data tree. That's the right granularity. 20081207 21:29:18 i didn't say on every page access 20081207 21:29:45 esr: Wmllint believes that [dwarf] is a wml tag. But it is a variable without the $. Maybe you can improve the output. 20081207 21:29:53 Most likely wesnoth.org will have enough horsepower to run it on every commit.. 20081207 21:30:06 fendrin: I'l llook. 20081207 21:30:21 If not then cron job could run it at regular intervals. 20081207 21:30:38 fendrin, err 20081207 21:30:42 add the dollar sign? 20081207 21:30:50 also, wth is that [message] :P 20081207 21:31:37 (you're leaving out dollar signs when indexing the array there) 20081207 21:32:13 ...which is probably why it's being mistaken for a tag. 20081207 21:32:37 This is the sort of thing wmllint is *supposed* to catch :-) 20081207 21:33:11 * Soliton watches wmllint catch the tumbling [dwarf]. 20081207 21:33:29 -!- EdB [n=EdB@52.101.196-77.rev.gaoland.net] has joined #wesnoth-dev 20081207 21:34:09 I'm not surprised that Shadow_Master doesn't run wmllint to check his commits. It would surely turn his laptop into molten slag. 20081207 21:35:00 zookeeper: this code is broken by a reason. It's a wml thing that causes an assert error in c++. I filled a bug report about it. 20081207 21:35:41 a wrong array index causing an assert doesn't sound surprising. 20081207 21:36:32 Soliton: so the assert is meant to be an assert? I thought that should be a error message that is more human readable. 20081207 21:37:23 i'm not saying it can't be improved. i haven't looked at the WML or the assert. 20081207 21:37:58 It's just if I commit working code I destroy the testcase for the bug to be fixed. 20081207 21:38:40 well, if the bug is only exposed by invalid WML i'm afraid we'll have to live with that. 20081207 21:41:24 esr * r31356 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/06_Acquaintance_in_Need.cfg: Fix a wmllint false positive related to making Olurf persistatent. 20081207 21:42:20 esr: please don't commit when I promised to take care about the bugs. 20081207 21:44:37 esr: we don't need to work on the same thing at the same time and produce merge conflicts. 20081207 21:45:36 OK, sorry. 20081207 21:47:18 Is running "make" in data/tools supposed to take almost as long as full recompile or I'm doing soething wrong? :) 20081207 21:48:27 s/soething/something 20081207 21:48:48 fendrin * r31357 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/ (06_Acquaintance_in_Need.cfg 07_Elves_last_stand.cfg): Fixed some bugs reported by wmllint. 20081207 21:49:19 esr: make lint-low gives me 2 errors I am not sure what to do with. 20081207 21:56:01 esr: And I have an macro call to a undefined symbol. I guess wmllint should catch that. 20081207 21:57:12 When I'm finished replying to an email I'll svn up and look. 20081207 22:18:30 -!- ettin [n=jorda@156.206.219.87.dynamic.jazztel.es] has joined #wesnoth-dev 20081207 22:29:19 esr: the problem I'm running into is that I don't have the time to set up and compile wesnoth. :( Is there any way for me to check for broken image references without compiling and running the game? 20081207 22:29:32 namely - is data/tools/make precisely that? 20081207 22:29:55 Exactly that. All you need is Python installed. 20081207 22:30:05 Okay. 20081207 22:30:11 maybe i'm ignorant, but i wonder what the heck is it supposed to "make" 20081207 22:30:15 I had no idea this existed. 20081207 22:31:15 -!- fendrin_ [n=fabi@g227022136.adsl.alicedsl.de] has joined #wesnoth-dev 20081207 22:31:23 zookeeper: In Unix-land "make" is where you commonly drop all kinds of maintainence recipes, even if they don't have any actual output other than a bunch of diagnostic messages. 20081207 22:31:23 esr: how exactly does it get the info back to me. It's a CLI program, but ... does it just output lines to the terminal it's in to denote missing images? 20081207 22:31:38 Jetrel: Yes, exactly. 20081207 22:32:04 Okay, so this is probably going to overlap with a bunch of other potentially-broken stuff WRT building a copy of the game? 20081207 22:32:27 esr, i've figured as much, i just don't see why it's called "make" if it doesn't make anything :| 20081207 22:33:04 It makes abstract concepts :) 20081207 22:33:19 Also known as phony targets. 20081207 22:33:52 zookeeper: "make" is the name of the program tha, given a target, tries to build all its dependencies. Without a tatrget argument it just tries to build the first target in the file -- in this case, the full sanity check. 20081207 22:34:15 esr: ah, thank you, that answered a -lot- of questions. 20081207 22:34:55 Jetrel: Good, because I didn't understand your previous question :-) 20081207 22:34:59 esr, uh oh. 20081207 22:35:05 ? 20081207 22:35:15 nevermind 20081207 22:35:24 esr: the biggest problem I've had these past 12-?? months, is that I've been doing all my work on wesnoth mostly blind. 20081207 22:35:43 gotta be hard for someone who just makes art :P 20081207 22:35:52 wmlsco[e and wmllint will, at leasst, be the equivalent of a white cane. 20081207 22:36:21 Besides testing animations is a distributed "official release" of 1.5 or something, I don't have the ability to do other testing. 20081207 22:36:46 Jetrel: Sirp almost managed to compile trunk on Mac :) 20081207 22:36:54 So when I do a commit, I don't have the ability to build beforehand and make sure there are no warnings. 20081207 22:37:19 Yeah, and loonycyborg's statement right there is exactly why I don't touch building wesnoth with a ten-foot pole. 20081207 22:37:34 Jetrel: That make production is the reason we haven't been getting broken-image reports from users for most of two years. With this, I catch 'em before we ship. 20081207 22:37:51 Because if I did, I wouldn't get any art done. 20081207 22:38:07 But yes, now that I'm aware of this new tool, this should be excellent. 20081207 22:38:23 Perhas you setup multiboot configuration with ubuntu? 20081207 22:38:59 There are large classes of other WML errors also susceoptble to automated detection. Whenever I figure out how to test for one, I add logic to wmllint. It's pretty formidable at this point. 20081207 22:39:32 Jetrel, btw, i figured we should make a custom animation for storm tridents 20081207 22:39:40 i'm thinking something like a trident-shaped lightning 20081207 22:40:19 zookeeper: uh, actually I had a different, more difficult but perhaps more interesting project I'd like to look at. 20081207 22:40:24 yes? 20081207 22:40:59 Modular lightning, composed of multiple parts that get composited together to make a final lightning bolt. 20081207 22:42:07 yeah you mentioned that earlier 20081207 22:42:21 i just didn't quite get what and where did you mean that to be used 20081207 22:42:34 So that every lightning strike is different (not really, but it'll increase the probability space for them by enough orders of magnitude that it effectively would be). 20081207 22:42:39 as an attack anim, where you get a random combo of parts every time? 20081207 22:42:41 right 20081207 22:42:50 can't be done without ridiculously verbose WML 20081207 22:42:55 zookeeper: well, yeah. It would replace our current lightning. 20081207 22:42:59 Damn. 20081207 22:43:43 I'll take your word for that. 20081207 22:43:56 i don't really see the need either, if we have 5-6 variations it'd be almost as good 20081207 22:44:31 the only way to do any animation variations is to specify multiple versions of the whole animation 20081207 22:44:57 What I'd like to do then, is - instead of making custom versions of them (such as for the mermen, which in this rare case, I think is not necessary, or could be done with an additional overlay). 20081207 22:45:09 it's not possible to tell the game to "as the first frame, pick one randomly from these three and then as the second frame, pick one randomly from these two" 20081207 22:46:11 I'd like to both improve the current lightning, slightly, and make 5-6 variations. 20081207 22:46:51 The improvement to the current lightning would be to make it "twitch" a bit between frames. Like electricity arcing. 20081207 22:46:54 -!- fendrin [n=fabi@e179206158.adsl.alicedsl.de] has quit [Read error: 110 (Connection timed out)] 20081207 22:46:57 zookeeper: r31300 causes wmllint to be angry. I don't understand what you have done there. Can you have a look, please? 20081207 22:47:15 Right now, the whole thing is a still image that just has different opacities swept over it. 20081207 22:47:40 The other improvement would be make an "impact" at the bottom when it hits. 20081207 22:48:37 hmh, frankly i don't think a twitching lightning would look so good :/ i mean, the current one has a very nice...well, what do you call it? a snes-like look to it due to it being static like that, and a twitching lightning would need quite a few frames to make it look good. just my guess. 20081207 22:49:09 zookeeper: maybe. I could easily prototype it with variant, and see how it looks. 20081207 22:49:16 sure 20081207 22:49:17 You may well be right. 20081207 22:49:53 on an unrelated note, i thought that maybe healing anims should include a classic little-sparklings-raining-down-and-healing-you effect on the target unit 20081207 22:50:06 zookeeper: that would rock 20081207 22:50:40 fendrin_, the macro only takes the contents of a single [effect], and i needed the advancement to give him two [effect]s 20081207 22:50:40 zookeeper: in fact, since I'm currently screwing around with little to do, I'd like to make some sparkles for that in the next hour. 20081207 22:50:45 cool 20081207 22:51:42 fendrin_, i guess i should have realized it'd result in wmllint complaints 20081207 22:51:54 esr: Also, just for clarification; I know wmllint, and wmlscope are tools designed to automate the maintenance of WML, but I'm fuzzy on the specifics. Just to get things straight so I can actually do this right, which ones should I be running on a new commit I'm prepping, and what do they do? 20081207 22:52:12 please someone add whatever wmllint directives it might need, i don't remember them really (nor do i have time right now) 20081207 22:54:43 Jetrel: Explanation follows... 20081207 22:57:41 -!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has quit [] 20081207 22:57:51 wmlscioe iis a cross-referencer. Basicaklly, what is does is spot references to macros and images or souind files (um, also map files) that aren't satisfied. It can be called in some non-defayltt modes that check for duplicate or unused resources. Thus, for example, I can flip a switch and tell you exactly what cliques of identical images exist at different locations in mainline. 20081207 22:58:12 Sorry, that's "wmlscope". 20081207 22:59:05 wmllint is more like a grammer checker. What it looks for is ill-formed WML. It also spellchecks storyline text. 20081207 22:59:26 -!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev 20081207 23:00:28 wmllint messages are not likely to concern you very much. Usually it's wmlscope that gripes after art changes, and usually because not all references to an old image have been spotted and fixed. 20081207 23:01:16 -!- Baufo [n=thomas@62-47-135-69.adsl.highway.telekom.at] has quit [Remote closed the connection] 20081207 23:01:31 Both are Python scripts provided you have Python installed, they should just work. 20081207 23:01:34 -!- [Relic] [n=[Relic]@adsl-76-236-130-144.dsl.milwwi.sbcglobal.net] has joined #wesnoth-dev 20081207 23:01:52 esr: how can I silence wmllint about unclosed tags? 20081207 23:02:10 Checking... 20081207 23:02:34 <[Relic]> Hello :) 20081207 23:02:53 esr: that unclosed wml tags are in utbs. units/Kaleh.cfg 20081207 23:03:44 # wmllint: validate-off will do it. Try to # wmllint: validate-on as soon as you can afterwards. 20081207 23:04:05 esr: line 387 both /effect and effect are disliked. i tried validate-off. Maybe a caching problem. 20081207 23:04:19 Checking... 20081207 23:05:58 Jetrel: You don't actually have to run either tool by hand to check mainline. Look at the targets in data/tools/Makefile; they're reasonably well commented, I think. 20081207 23:07:02 Maybe Jetrel even be able to run make. 20081207 23:08:37 * loonycyborg wonders whether make is installed on Mac. Provided Jetrel is still using a Mac.. 20081207 23:13:04 loonycyborg: I'm gonna be using a mac till they pry it from my cold, dead fingers. 20081207 23:13:22 fendrin_, maybe use # wmllint: unbalanced-on ? 20081207 23:13:57 Jetrel: But does it have make installed by default? 20081207 23:13:58 feeding seems to use that since it also has unbalanced tags 20081207 23:14:17 loonycyborg: and no, "make" is not installed in distributed mac (or windows, AFAIK) copies of wesnoth 20081207 23:14:46 loonycyborg: however, make is always distributed in svn, and svn is the same for all platforms. 20081207 23:15:18 -!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev 20081207 23:15:18 Yeah, the "unbanced" pragmas should work. 20081207 23:15:19 Jetrel: make is distributed in svn? Why should it? 20081207 23:15:26 loonycyborg: and I couldn't commit without having a downloaded svn repo 20081207 23:15:29 I meant whether Unix "make" utility is distributed along with Mac OS X 20081207 23:15:46 loonycyborg: oh 20081207 23:15:53 yeah, of course it is. 20081207 23:15:57 Dude, it's Unix. 20081207 23:16:18 fendrin_: I meant the script to run against it 20081207 23:16:26 ah 20081207 23:16:31 Apple guys could think that it's too technical for most Mac users :) 20081207 23:17:00 loonycyborg: no, fool. that whole policy disappears after you leave the gui layer 20081207 23:17:34 (fool meant in the camaraderie, Mr. T sense) 20081207 23:17:43 Not trying to be insulting. 20081207 23:18:07 You have to realize the mac only insulates people in the gui layer of the system. 20081207 23:18:17 * loonycyborg is sure that Mr. T is currently not envies him :) 20081207 23:18:54 When people are screwing with the terminal, or anything in the Unix layer, the kid gloves are off. It's totally normal unix. 20081207 23:19:03 Jetrel: That feeling is wholly unfamiliar to windows ;) 20081207 23:19:13 ...and that is *GOOD*! :-) 20081207 23:19:27 And the reason for that, is that they assume that the computer-illiterate won't be using anything at that layer. 20081207 23:19:44 I meamn the fact that's a totally normal Unix under the hood is good. 20081207 23:19:56 yeah, i could even pay money for an OS which was normal unix under the hood but with a good GUI and capable of running every windows app. 20081207 23:20:48 zookeeper: That won't come cheap :) 20081207 23:20:56 In fact, as of 10.4, or maybe 10.5, they actually updated enough certain things that they finally got official unix certification 20081207 23:22:10 So there are curious things you can do because of this. For example, you can actually compile traditional linux window managers, and run them on top of OS X. 20081207 23:22:40 The only thing that prevented Sirp from compiling wesnoth on his Mac was his inability to get pango with cairo backend from fink. 20081207 23:22:53 -!- zookeeper [n=l@dsl-tkubrasgw1-fe9ff800-215.dhcp.inet.fi] has quit [] 20081207 23:23:13 loonycyborg: I wonder if he's seen MacPorts.... and I wonder if that would have a copy 20081207 23:23:42 MacPorts is a port (no pun intended) of the BSD (?) ports system 20081207 23:24:33 It's another contender in the same space that fink serves. 20081207 23:24:37 I think. 20081207 23:25:17 http://trac.macports.org/browser/trunk/dports/x11/pango/Portfile 20081207 23:25:45 It depends on cairo, therefore most likely it has cairo backend. 20081207 23:25:56 Jetrel: If you use Emacs, the report formats of both wmlscope and wmllint are compatible with Emacs compilation mode. So you can use it to step through the error locations. 20081207 23:26:29 I'm afraid I don't. :/ 20081207 23:26:30 But Sirp already got a laptop with ubuntu, so he no longer needs it :) 20081207 23:27:26 esr: how is that be used? 20081207 23:29:39 Anyways, though, probably the primary reason I like using a mac is that it's kinda like paying someone else to do the icky job of managing my distro. No recompiling the kernel, no scary dependency issues when installing software (I may dislike the way mac software distribution wastes space, but there truly are no dependency problems, ever). 20081207 23:30:16 No driver issues, etc. And the ability to run certain commercial titles I unfortunately need to (or in the case of games, want to). 20081207 23:31:23 fendrin_: Run 'make' as your compilation command from within Emacs. Use C-x ` to step forward through the error locations. Also, just pressing Enter on a message line will pop up a buffer visiting the location. 20081207 23:31:26 -!- Elvish_Pillage2 [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)] 20081207 23:31:33 * loonycyborg prefers to manage his distro himself. Recompiling the kernel and dependency issues are fun :) 20081207 23:32:14 -!- noy [n=Noy@70.70.128.133] has joined #wesnoth-dev 20081207 23:32:30 Jetrel: It sounds like Ubuntu might not suck for you. They handle the repo-dependency thing really well. 20081207 23:32:42 esr: The run make part is my problem. 20081207 23:33:09 fendrin_: Ho;d on while I query my Emacs... 20081207 23:33:34 esr: it's definitely something I recommend to non-computer literate people who've been sucked into using windows for what's basically just a web/email terminal for them. 20081207 23:33:47 esr: I love what that distro has been doing. 20081207 23:33:48 fendrin_: Try M-x compile 20081207 23:34:23 Jetrel: Yeah, me too. 20081207 23:35:04 esr: seems to work, thank you 20081207 23:36:31 Jetrel: My 74-tear-old mother is now using Ubuntu in the style you describe. 20081207 23:37:13 :) 20081207 23:37:14 i am off to bed now, n8 20081207 23:37:19 Good night. 20081207 23:38:03 But, surprisingly, getting all wesnoth's dependencies is much easier on ubuntu than on mac :P 20081207 23:38:20 Jetrel: I think I will work on changing Familiar's portrait to match his new form. 20081207 23:38:43 Espreon: are you talking about a color-change from red -> black? 20081207 23:38:50 Jetrel: ...which is actually kind of cool since she's actually acquired some clue about what I've beenb doing with my life. "Hey, mom, my code renders images onto the screen for that browser you're using." 20081207 23:39:03 Jetrel: Yes. 20081207 23:39:53 Espreon: cool. This is just a "dialogue" unit; not something that actually fights in the game, right? So there'd be no reason to preserve the current portrait for a lower-level of it, because it never levels up? 20081207 23:40:22 I think he does fight in the scenario where you fight Jevyan. 20081207 23:40:23 -!- wesbot changed the topic of #wesnoth-dev to: 1.4.7 planned for 12.12. | soft stringfreeze for most campaigns: http://www.wesnoth.org/wiki/TextdomainStatus | wesconf@FOSDEM2009 http://www.wesnoth.org/wiki/Fosdem2009 | 1.6 release plan: http://www.wesnoth.org/wiki/Wesnoth1.6ReleasePlan | 58 bugs, 210 feature requests, 7 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com, or for images: http://imagebin.org/index.php?page 20081207 23:40:52 Espreon: but it's not something that you, the player, work to advance from one level to the next, though, is what I'm saying? 20081207 23:41:29 Yes. 20081207 23:42:29 esr * r31358 /trunk/data/tools/wmllint: Shut wmllint up about a place qhere it shouldn't be complaining. 20081207 23:42:59 fendrin_: I just quieted wmllint about [+side] without type; it will check that only inside [scenario] now. 20081207 23:43:51 zookeeper: so anyways, looking at the output of make, it seems we've got a number of mainline breakages WRT to the ugly lightning imgs I removed. Uhm... is the delfador-mage's lightning attack in such a state that it's a macro we can just drop into any units who use it, or is that not the case right now? 20081207 23:45:14 esr: #wmllint: unbalanced-on didn't solve the problem of Kaleh.cfg 20081207 23:45:19 zookeeper: I don't know if we ever got around to updating that. 20081207 23:46:12 fendrin_: I'll look at it in a moment. 20081207 23:46:25 Jetrel: I looked at this earlier. 20081207 23:47:11 No, the lightning attack is not a re-usable macro. I tried top repackage it that way erarlier today when I noticed this, but... 20081207 23:48:04 ...the Elder Mage attack frames you adviced replacing it with have a problem in that they assume the casting unit is, in fact, an Elder Mage. 20081207 23:48:12 Yeah. 20081207 23:48:58 It's an old, old macro from before we had modern, proper support for placing these over the enemy via "offset". Basically, it's a hack to make it appear to hit the enemy directly. 20081207 23:49:11 Made before we had support in the engine to do that. 20081207 23:49:28 It might be possible to macro-package the missile-frames you used for that, but at that point I came to my senses and asked "Why I am *I* doing this?" Me making a big stink on IRC followed within minutes. 20081207 23:49:35 In -fact-, the images are probably individually and awkwardly made off-center to facilitate exactly that. 20081207 23:50:10 -!- boucman [n=rosen@159.83.102-84.rev.gaoland.net] has quit [Remote closed the connection] 20081207 23:50:53 Yeah, I should be able to take care of this this afternoon. The WML should end up nearly identical to the mage-of-light's attack, with a different frame count of course. 20081207 23:52:34 There's a related question, outside the art domain, about whether rthe Elder Mage unit should have an intrinic lightning attack or whrther that's a property of Delf's staff. I favor the latter, as we also want to re-use Elder Mage for Crelanu in LoW. 20081207 23:53:33 esr: I have no clue. 20081207 23:53:33 * loonycyborg hopes that Elder Mage will get attacks appropriate to level5 unit 20081207 23:53:34 If it were up to me, I'd actuually associate the zap graphics with Delf's staff as an artifact. Then he just starts out with one in HttT. 20081207 23:54:27 loonycyborg: actually, one thing I'd like to see done to delfador, would be to give him *other*, very unusual abilities, to make him feel more the part of an "elder mage". 20081207 23:54:45 Jetrel: Absolutely agree. 20081207 23:54:57 It would be nice if the solution were re-usable for the Storm Trident and Rod of Justice as well as Delf's staff in DM. 20081207 23:55:04 -!- noy [n=Noy@70.70.128.133] has quit [Success] 20081207 23:55:09 esr: that is precisely the intent. 20081207 23:55:22 Jetrel: Good. 20081207 23:55:26 The question is what abilities that should be.. 20081207 23:55:47 esr: wmllint abourts because of the .#filename temp files of emacs. Can you ignore them? 20081207 23:55:48 esr: every unit with a lightning attack not only should be able to, but should, use the revised macro. 20081207 23:56:03 s/abourts/aborts 20081207 23:56:04 As to Delf special abilities...one obvious one might be village teleport, like a Silver Mage. 20081207 23:56:15 Possibly. 20081207 23:56:17 fendrin_: I'll look into that now. 20081207 23:56:37 Here are a bunch of interesting abilities I've had bouncing around my head: 20081207 23:57:20 1] curse v1 - you hit an opponent, and it 'sticks' to them like slow for a turn. It reduces their chance-to-dodge by some amount. 20081207 23:57:37 2] curse v2 - same thing, but reduces their resistances in some way 20081207 23:58:32 3] empower v1 - an aura effect like leadership. Causes nearby units to have higher resistances 20081207 23:58:51 4] empower v2 - causes nearby units to have higher defense 20081207 23:58:59 That's already present in IftU 20081207 23:59:19 Wait.. It increases resistances.. 20081207 23:59:39 Jetrel: I am done. 20081207 23:59:45 I am gonna slap it on imagebin. 20081207 23:59:52 esr * r31359 /trunk/data/tools/wesnoth/wmltools.py: Make wmllint and wmlscope ignore Emacs tempfiles. --- Log closed Mon Dec 08 00:00:02 2008