--- Log opened Sat Mar 01 00:00:59 2014
20140301 00:02:39-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140301 00:07:12-!- kex [~kex@89.205.75.19] has quit [Ping timeout: 240 seconds]
20140301 00:08:44-!- mattsc [~mattsc@154.20.32.245] has joined #wesnoth-dev
20140301 00:11:58< Aishiko_laptop> ahhh I don't do anysort of forum hosting so that is all good to know stuff for me as I try to figure out if the idea is feasible
20140301 00:15:51-!- Coffee_irc [~david@ppp118-210-31-67.lns20.adl2.internode.on.net] has joined #wesnoth-dev
20140301 00:29:19-!- Velensk [~Velensk@cpe-75-187-86-211.woh.res.rr.com] has joined #wesnoth-dev
20140301 00:29:36< Velensk> I've made an announcement in the Multiplayer Development server, could someone sticky it for me?
20140301 00:32:27< fabi> AI0867: ping
20140301 00:35:00< shadowm> Velensk: Done.
20140301 00:36:30< Velensk> thanks
20140301 00:40:14-!- sachith500|2 [~kvirc@112.135.13.143] has joined #wesnoth-dev
20140301 00:41:21-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20140301 00:45:45< shadowm> iceiceice: Oh, any news about iceiceice2?
20140301 00:46:03< iceiceice> actually no
20140301 00:46:04< shadowm> Or iceiceiceN in general.
20140301 00:46:15< iceiceice> i'm really not sure what happened there
20140301 00:46:20< iceiceice> i guess i should pm the account and see what happens
20140301 00:46:41< iceiceice> is there some way i could have accidentally registered it from command line in the wesnoth client or something?
20140301 00:46:48< iceiceice> oh also is there a way to check if it is activated?
20140301 00:47:05< shadowm> That's physically impossible since the registration functionality in the FUH is a stub.
20140301 00:47:13< iceiceice> ok good to know
20140301 00:47:15< shadowm> Also, that email address doesn't look lik you.
20140301 00:47:18< happygrue> JunJM: I don't personally have any impementation in mind.  I would imagine that the details would depend a bit the strengths each student would bring.  I think at this stage there are really two things that are important: 1) A proposal that clearly grasps the intent of the project and addresses a realistic looking way to tackle them
20140301 00:47:31< iceiceice> alright i'm going to pm the account and see what happens
20140301 00:47:46< iceiceice> if i dont get a response i think we should deactivate it
20140301 00:48:04< happygrue> and 2) some demonstration of ability, in the form of using Git, easy coding tasks or bugfixes - showing that the student can do something, in other words
20140301 00:48:56< happygrue> re 1) by tackle them I meant to have "indicates the challenges and propses a way to tackle them"
20140301 00:49:15< happygrue> I but have like 5 hours sleep in the past two days and my brain...
20140301 00:49:19< happygrue> it is not so hot right now :D
20140301 00:49:51 * happygrue mumbles about kids
20140301 00:50:05< Aishiko_laptop> shadowm, can I bounce an idea I have on the encrypting of the passwords on the client side?
20140301 00:50:36< shadowm> Aishiko_laptop: I don't understand encryption in general, so it's unlikely I can be of much help there. :(
20140301 00:51:27-!- spoffy [~spoffy@152.78.175.8] has quit [Quit: Leaving]
20140301 00:51:40< happygrue> fabi: I was wondering if the choice to have a faded font for the alignment stuff was aesthetic?  Personally I have some trouble reading it, but since I know what words to look for it doesn't really slow me down
20140301 00:52:02< happygrue> but it seems to be about the only faded font that I know of in the game, so it does stand out a bit I guess?
20140301 00:52:49< happygrue> I think the placement is fine, though I think the +/-% would still be best as a tooltip, I can see the advantage of having it displayed too
20140301 00:53:24< happygrue> but since it's already highligted in the attack data it also seems redundent to me
20140301 00:53:26< Aishiko_laptop> well, I'll put it out there and you tell me if you think it'll work, as a starting point at any rate.  There is a C++ library to do 2 way salted encryption, I was thinking use that to take the password salt it, and then encrypt it.  Then when its needed to be sent, decrypt it and send it through the current login protocols that expect plaintext
20140301 00:54:10< fabi> happygrue: The story is a little longer. The color of the font was hardcoded in c++. Then I discovered that undocumented font_rgb attribute. I do not know if it is broken or if I just don't get how the correct syntax and format for it is. 
20140301 00:54:12< happygrue> I also know nothing useful about encrypting stuff, but it seems like a great thing to be working on.
20140301 00:54:21< happygrue> ah, I see
20140301 00:54:46< fabi> 255,255,255,255 gives me a non transparent blue color.
20140301 00:55:25< shadowm> Aishiko_laptop: We generally try to keep external dependencies to a minimum (new dependencies ought to be discussed in the ML first), and try to stick to software and versions from Debian stable, if that helps. (We run Debian stable on wesnoth.org.)
20140301 00:55:34< Aishiko_laptop> with overwriting of the memory location that held the plaintext so that it can't be grabbed out of memory (if its left there after the function has done its job), not a perfect solution but, its better then plaintext storage I think, the biggest downside is the requirement of yet another lib
20140301 00:57:03< shadowm> As for your idea, IIUC what you are suggesting is to encrypt the password on the client side using a parameter given by the MP server, and then let the MP server decrypt it?
20140301 00:57:04< Aishiko_laptop> shadowm, okay I was thinking the added dependency could be a problem, and I have no problem checking to see if it is in debian stable before I do anything else
20140301 00:58:30< Aishiko_laptop> more like, the password is stored encrypted, and only decrypted for the login functions to do what they need to do with the plaintext
20140301 00:59:15< shadowm> I gather that while that will certainly eliminate the problem of having a password in clear text on the client-side, it'd still be possible for someone to break that encryption easily.
20140301 00:59:16< Aishiko_laptop> it was just an idea
20140301 01:01:06< Aishiko_laptop> but I see what your saying, there isn't much that can be done if the key to the lock is right there sitting next to the lock
20140301 01:02:44< Coffee_irc> Aishiko_laptop: just to chime in, encryption that needs to be secure across a channel need be 2 factor
20140301 01:03:01< happygrue> perhaps even if it's not great it would still be worth doing something so that passwords aren't just stored as they are now though...
20140301 01:03:19< Coffee_irc> you encrypt on the end that knows the password, and then again with a generated random seed on the server side
20140301 01:03:28< Aishiko_laptop> yeah, that was all I was really looking at doing at first, remove the storage from plaintext to encrypted
20140301 01:03:34< happygrue> even a chain link fence is better than no fence!
20140301 01:03:35< Coffee_irc> this is how rsa works over the net
20140301 01:04:52< Coffee_irc> the server sends the seed number encrypted with a known encryptino to the client nd should be generated uniquely for each new transaction
20140301 01:05:04< Coffee_irc> for banking, etc.
20140301 01:05:58-!- Gallaecio [~quassel@84.120.219.139.dyn.user.ono.com] has joined #wesnoth-dev
20140301 01:06:33< Coffee_irc> you could always use rot13 to encrypt if you just want to encrypt stored passwords ;)
20140301 01:06:40-!- Gallaecio_ [~quassel@84.120.219.139.dyn.user.ono.com] has quit [Ping timeout: 244 seconds]
20140301 01:06:59< iceiceice> y, just to reiterate the thing i was mentioning earlier,
20140301 01:07:16< iceiceice> you can use the same key exchange procedures, like diffie hellman etc., or just a hash function
20140301 01:07:24< iceiceice> to exchange the seed of the rng for an mp game
20140301 01:07:42< sachith500|2> oh the rng :P
20140301 01:08:01< Coffee_irc> DON'T MENTION THE RNG :P
20140301 01:08:05< sachith500|2> hahaha
20140301 01:08:14< iceiceice> i dont think we do that currently, and if in fact the server is generating them then it doesn't matter :)
20140301 01:08:26< iceiceice> but if it does then it fixes a possible cheat
20140301 01:08:48< Aishiko_laptop> or we could do something like chap does and have the server send a password request with a salt and the client has to take the password and hash it and send back the hash,  but that's not the best way on a non ppp connection anyway
20140301 01:08:57< iceiceice> hehe the rng code is like ancient though.... who knows what is lurking there :p
20140301 01:09:08< sachith500|2> yeah, I remember you saying that in game
20140301 01:09:28< sachith500|2> Thanks for that, btw, it made me realise why I was losing so much ;)
20140301 01:09:31< Coffee_irc> I suspect it is slightly weighted between rounds, although the average over many rounds is relatively uniformly distributed
20140301 01:11:38< iceiceice> y i'm not saying i think theres anything wrong with the rng, i assume whoever made it did a good job...
20140301 01:11:50< iceiceice> i'm just saying about host recompile cheat
20140301 01:11:53< Coffee_irc> :)
20140301 01:12:08< iceiceice> if its even actually possible
20140301 01:12:35< iceiceice> idk, i like that sort of crypto thing, "how to flip a coin over the telephone"
20140301 01:12:36< Coffee_irc> iceiceice: you could avoid problems of cheating with 2 factor encryption of seeds
20140301 01:12:42< shadowm> "The RNG sucks, it's not really random. I just missed three times in a row with my mages!"
20140301 01:12:48< iceiceice> yeah but i guess thats overkill
20140301 01:12:50< Coffee_irc> because you know that the server side cannot be tampered with
20140301 01:13:07< Coffee_irc> but really, who would cheat at Wesnoth to win?
20140301 01:13:12< iceiceice> idk we have tournaments
20140301 01:13:29< iceiceice> there are definitely people who have admitted to cheating and been banned from ladder
20140301 01:13:40< sachith500|2> One thing I've learnt with online games is there will always be people who try to cheat
20140301 01:13:52< happygrue> it's part of the reason there is no formal and official ranking system, as surely there would be some motivated folks who would try.
20140301 01:14:04< iceiceice> i would honestly be surprised if no one had *ever* tried to cheat during a TGT match for example
20140301 01:14:14< iceiceice> even though i think the wesnoth community is better than other games
20140301 01:14:37< happygrue> there are very easy ways to cheat in multiplayer, and you have already fixed one of them iceiceice ;)
20140301 01:14:48< happygrue> "cheat" meaning gain some small advantage
20140301 01:14:50< iceiceice> i dont care so much about that, i just like the academic coinflipping solution that doesnt use a trusted third party :)
20140301 01:15:05< iceiceice> *don't care so much about whether people actually cheat
20140301 01:15:38< Coffee_irc> a 2 factor authentication type scheme, with unique seed generation for each game on the server, would eliminate the possiblity of cheating
20140301 01:15:59 * happygrue eyes GSoC 2015
20140301 01:17:08< iceiceice> yeah thats true
20140301 01:17:12< Aishiko_laptop> wow amazing where the topic goes sometimes
20140301 01:17:28< iceiceice> happygrue: its true that there are many ways to cheat, but manipulating the rng is a pretty insidious one
20140301 01:17:42< happygrue> yes, I agree, I was off on a tangent
20140301 01:17:45< iceiceice> if u just have a magic luck button that you turn on to get favorable roles when they are crucial, then turn it off
20140301 01:18:41< Coffee_irc> iceiceice: the game quickly becomes dull if you save and reload as well
20140301 01:18:53< iceiceice> y but you cant do that in an mp game
20140301 01:18:55< shadowm> zookeeper: Yeah... I see now why it sucks.
20140301 01:19:16< shadowm> It feels like I could just stay at home and let the elves fight for me.
20140301 01:22:08< irker854> wesnoth: Chris Beck wesnoth:master c9a4f17a4389 / src/minimap.cpp: implement minimap blindfold http://git.io/9X7C9A
20140301 01:22:10< irker854> wesnoth: Chris Beck wesnoth:master b0023b7e9c16 / src/ (playmp_controller.cpp playmp_controller.hpp): apply a blindfold during skip replays http://git.io/Sl7xWw
20140301 01:22:12< irker854> wesnoth: Chris Beck wesnoth:master 41ee06930a6b / src/ (minimap.cpp playmp_controller.cpp playmp_controller.hpp): Merge pull request #113 from cbeck88/blindfold_minimap http://git.io/Fs0UAA
20140301 01:22:28< happygrue> shadowm: can we give Velensk the developer title in the forums?  I'm not sure when the call gets made for commit access, but it would be nice to have a title to launch such a thing, is that possible?
20140301 01:22:40< happygrue> or the title, not developer, what was it you wanted Velensk?
20140301 01:23:01< happygrue> I forget now, but I recall you had something in mind.
20140301 01:23:13< shadowm> happygrue: I haven't seen any commits from him since the map update.
20140301 01:23:30< happygrue> and 3 is the magic number needed?
20140301 01:23:59< shadowm> It really depends on how trivial the patches are.
20140301 01:24:15< happygrue> I see.  And the forum title can't go on without the other?
20140301 01:24:27< shadowm> Not without a very good justification.
20140301 01:24:30< happygrue> okay
20140301 01:26:59< Aishiko_laptop> That makes sense, you don't want to give the title (and the supposed knowledge and skills set that usually goes with it) to just anyone.
20140301 01:27:11-!- wesbot changed the topic of #wesnoth-dev to: string+feature freeze active on master | 214 bugs, 350 feature requests, 28 patches | Logs: http://irclogs.wesnoth.org | Alternate logs: http://wesnoth.debian.net | Don't paste on IRC! Use a pastebin: http://pastebin.com | http://imagebin.org
20140301 01:27:39< shadowm> Developers in the forums also have moderator powers.
20140301 01:29:27< happygrue> Yep.  I was wondering if just changing title to something a bit more offical were possible without rushing the trial process or even granting any real power, but I understand the rational for the system.
20140301 01:31:00< shadowm> I don't think I've been informed of the extent of his contributions outside the pull requests system yet.
20140301 01:31:47< shadowm> So, one map and several updates to it. Any other things I may be unaware of since I barely pay attention to the MP stuff?
20140301 01:31:47< Aishiko_laptop> personally I'd be hesident to accept that, worried I might abuse the power, plus  I really stink at the email/forum thing.
20140301 01:32:52< shadowm> (If I _have_ been informed, I probably forgot since my memory is pretty much useless these days.)
20140301 01:33:00< happygrue> He's going to maintain mp maps.  Eventually as changes get made map pull requests will add up to something and we can revisit it then I guess
20140301 01:34:02 * shadowm loves how his email client notifies him twice in a row every time iceiceice does stuff. :p
20140301 01:34:27< shadowm> I probably should create a separate folder and rules for github notifications...
20140301 01:34:54< iceiceice> hehe
20140301 01:35:20< happygrue> or you could get iceiceice to stop doing stuff...
20140301 01:35:58< iceiceice> i asked before if people preferred i dont use PRs, if you are chaning your mind i'll just push to master,
20140301 01:36:04< iceiceice> but i think i make less mistakes this way
20140301 01:36:34< shadowm> zookeeper: "Undead are coming! Let's abandon our duty and let Wesnoth figure out what to do while we pretend we are fleeing to figure out a solution from the outside!"
20140301 01:37:16< iceiceice> or i can just halt all development work, that is also an option :p
20140301 01:37:20< happygrue> ;)
20140301 01:37:38< sachith500|2> :P
20140301 01:37:59< shadowm> Nah, bad idea.
20140301 01:38:35< AI0867> fabi: pong
20140301 01:39:05< fabi> AI0867: Do you remember, we talked about the font_rgb attribute.
20140301 01:39:07< fabi> ?
20140301 01:39:10< AI0867> right
20140301 01:39:25< AI0867> you said something about grey looking blue
20140301 01:39:29< AI0867> and I couldn't reproduce that
20140301 01:40:09< irker854> wesnoth: fendrin wesnoth:master 56001398e741 / src/reports.cpp: Disable hardcoded trait colors, disabled "Attack(s)" label. http://git.io/WWor9g
20140301 01:40:11< irker854> wesnoth: fendrin wesnoth:master 257908cd0116 / data/themes/default.cfg: Add a font color for the traits. Some cleanups. http://git.io/zlk_lQ
20140301 01:40:46< fabi> AI0867: It is now in the default theme. 255,255,255,255 results in some blue as well.
20140301 01:41:22< shadowm> Hm?
20140301 01:42:42< AI0867> what's the fourth 255?
20140301 01:42:47< shadowm> This doesn't have anything to do with the nigh-unreadable alignment label, does it?
20140301 01:43:57< fabi> AI0867: Transparent thing, I think. The problem is that the attribute is undocumented, I have no idea which format it expects.
20140301 01:44:30< fabi> shadowm: if the unreadability is color caused, then yes.
20140301 01:45:47< shadowm> The color is incredibly dark and green.
20140301 01:45:59< shadowm> You were around when I posted the screenshots.
20140301 01:49:14< Coffee_irc> shadowm: web accessibility checkers can check contrast for readability http://juicystudio.com/services/luminositycontrastratio.php
20140301 01:49:39< Coffee_irc> I know because I recently sat through a 3 hour lecture on this stuff
20140301 01:50:11< shadowm> It's notable that nothing of this is mentioned in the commit messages.
20140301 01:50:21< shadowm> fabi: What code needs to be examined?
20140301 01:50:49-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140301 01:53:45< happygrue> shadowm: tweet worthy?  http://forums.wesnoth.org/viewtopic.php?f=6&t=39947
20140301 01:55:25-!- kex [~kex@89.205.75.19] has quit [Ping timeout: 264 seconds]
20140301 01:56:57< happygrue> maybe even could go on the front page this month?
20140301 01:57:16< happygrue> Or maybe we don't care?  I think it's fun anyway. :D
20140301 01:58:13< happygrue> obviously sample size is low... but it's always fun to win!
20140301 01:59:15< irker854> wesnoth: Chris Beck wesnoth:master 47a3f9d9b237 / src/playturn.cpp: save and abort generates an autosave http://git.io/qqThHQ
20140301 01:59:17< irker854> wesnoth: Chris Beck wesnoth:master a7d7f7c8fa4b / src/playturn.cpp: Merge branch 'save_and_abort_autosave' http://git.io/gx_jcw
20140301 02:01:40< iceiceice> happygrue: its definitely cool :) also sample size isn't that low i guess compared to like electoral politics polls? 
20140301 02:01:46< iceiceice> but i guess theres not much comparison anyways
20140301 02:02:34< shadowm> Hm.
20140301 02:11:47< iceiceice> oh btw: 
20140301 02:12:05< iceiceice> when i started working on the blindfold thing, jamit commented that i should make sure i dont reduce performance for low end machines
20140301 02:12:16< iceiceice> since i'm basically adding a bunch of extra checks to all this vision code
20140301 02:12:30< iceiceice> imo we probably should refactor that stuff anyways
20140301 02:12:43< iceiceice> but, i wanted to ask, is there a preferred profiling tool?
20140301 02:12:56< iceiceice> have we ever profiled the c++?
20140301 02:13:03< iceiceice> i didnt find any tools in the repo anywhere
20140301 02:13:21< iceiceice> or mention of this
20140301 02:13:22< shadowm> Some people have done that at various points, but I don't know anything about the specifics. We certainly don't have a formal profiling procedure.
20140301 02:13:29< iceiceice> ok just checking
20140301 02:13:59< iceiceice> i will keep an eye on it for sure and if people report issues in the beta i will revert all the blindfold stuff until i figure out a better implementation
20140301 02:16:25-!- iceiceice [~chris@207-237-132-90.ny.subnet.cable.rcn.com] has quit [Quit: Leaving]
20140301 02:20:41< fabi> shadowm: theme.cpp and display.cpp
20140301 02:21:12< shadowm> fabi: But where exactly?
20140301 02:22:59< Aishiko_laptop> shadowm, since there isn't an encryption in any of the libs being used (that will do what I was thinking of doing), I'm putting that on the back burner, but that doesn't mean I've given up
20140301 02:23:43< fabi> shadowm: theme.cpp line 459
20140301 02:24:17< shadowm> Aishiko_laptop: Sure.
20140301 02:24:53< irker854> wesnoth: JaMiT wesnoth:master 406afeab79fb / src/game_events/handlers.cpp: Indentation fix. http://git.io/m4FOQw
20140301 02:25:39 * shadowm pauses his game to take a look at theme.cpp:459.
20140301 02:26:10< Aishiko_laptop> I might try my hand at porting something over, perhaps as a python script, eliminating the need for a new dependency.
20140301 02:26:11< shadowm> Oh my god it's a bitwise operation.
20140301 02:26:51< fabi> Yes, it is from hell.
20140301 02:35:16< Aishiko_laptop> could be worse.  it could be a modifier of all the images used doing a bitwise operation the pixels of each (excepting the background) so everything looks funny when you play the game
20140301 02:36:21< Aishiko_laptop> yes I did that once for a class
20140301 02:36:24< shadowm> Is & commutative?
20140301 02:36:44< shadowm> Okay, it seems it is.
20140301 02:40:45< shadowm> Waaait, this code isn't even reached.
20140301 02:42:27< Aishiko_laptop> ohhh isn't that always it?  You spend hours on a function trying to debug it only to in the end find it wasn't being called at all, nope I've never done it either
20140301 02:42:35-!- sachith500 [~kvirc@112.135.10.210] has joined #wesnoth-dev
20140301 02:42:40< shadowm> fabi: Are you sure you gave me the correct location to check?
20140301 02:44:07< fabi> shadowm:  display.cpp 1439 may also be a point of failure. 
20140301 02:45:12-!- sachith500|2 [~kvirc@112.135.13.143] has quit [Ping timeout: 240 seconds]
20140301 02:45:50< shadowm> I *think* I was supposed to look at theme.cpp:500 instead.
20140301 02:46:23-!- happygrue [~happygrue@wesnoth/developer/wintermute] has quit [Ping timeout: 244 seconds]
20140301 02:46:36< shadowm> That said, I inserted a couple of  { std::cerr << "! " << std::hex << font_rgb_ << std::endl; } probes, and the numbers look right, so yeah, probably display.
20140301 02:47:35< shadowm> What.
20140301 02:48:00< shadowm> So the theme code goes through the hassle of packing the color values only for this function to unpack them again?
20140301 02:48:12 * shadowm is not impressed.
20140301 02:49:27< fabi> Well, undocumented feature from 2005.
20140301 02:50:24< shadowm> Once again I get the correct numbers, sort of. RGB 200,200,200.
20140301 02:50:36-!- Gambit [~derek@wesnoth/developer/grickit] has quit [Remote host closed the connection]
20140301 02:52:29< shadowm> Yep, it's a red herring.
20140301 02:52:48< shadowm> At no point text is "neutral" or "cures, heals +8" or anything I'm seeing onscreen.
20140301 02:53:05< shadowm> Only the top bar labels.
20140301 02:53:29-!- cib0 [~cib@p20030067CE0E5A01267703FFFEE75B84.dip0.t-ipconnect.de] has quit [Ping timeout: 245 seconds]
20140301 02:54:10< irker854> wesnoth: Chris Beck wesnoth:master eb9274bd4c78 / src/game_display.cpp: linux dbus notifications now remember only last 5 messages http://git.io/LYr0fA
20140301 02:54:12< irker854> wesnoth: Chris Beck wesnoth:master c7a239416882 / changelog: update changelog http://git.io/tanLoQ
20140301 02:54:24< shadowm> Now, if I were the sidebar, where would I hide my rendering code?
20140301 02:54:48< shadowm> virtual void draw_sidebar() {}
20140301 02:54:52< shadowm> Pfft.
20140301 02:55:22< shadowm> Implemented in game_display, uses refresh_report().
20140301 02:55:55< shadowm> ... which is in display.
20140301 02:56:45< shadowm> Relevant: display.cpp:2828.
20140301 02:57:56< shadowm> I'm getting 16777215 for everything here, which is 0x00FFFFFF.
20140301 02:58:47-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20140301 02:59:09< shadowm> I think I know what's going on here.
20140301 02:59:56< shadowm> ttext uses a different color format. Instead of 0xUURRGGBB (where UU = unused), it expects 0xRRGGBBAA.
20140301 03:00:01< shadowm> Where AA = alpha.
20140301 03:00:33-!- iceiceice [~chris@pool-108-21-12-72.nycmny.east.verizon.net] has joined #wesnoth-dev
20140301 03:01:16< shadowm> Yep, I had figured that out back when I rewrote the storyscreen code.
20140301 03:02:14< iceiceice> anyone who runs wesnoth under linux: i pushed a fix to the linux dbus notifications, if you have them off, please turn them on and let me know at some point if you think it is still broken
20140301 03:02:43< shadowm> fabi: text.set_foreground_color(0x000000FF + (item->font_rgb() << 8)); ought to do the trick.
20140301 03:02:51< shadowm> No idea about big endian systems, though. :(
20140301 03:03:24< shadowm> This is for display.cpp:2829, by the way, where it currently reads text.set_foreground_color(item->font_rgb());
20140301 03:03:54< shadowm> The addition of 0x000000FF is supposed to set the alpha channel to 0xFF (ttext is alpha-aware) on LE.
20140301 03:04:33< shadowm> The value retrieved with item->font_rgb() has no alpha channel, hence this is necessary.
20140301 03:04:59< shadowm> And yes, your macros FONT_RGB_LEVEL1 and FONT_RGB_LEVEL2 have an extra unused value that the engine disregards.
20140301 03:05:41< shadowm> Let me see if I can find how GUI2 deals with converting RR,GG,BB,AA strings to the format ttext needs.
20140301 03:06:16< shadowm> But first I'll go to the kitchen to make some tea.
20140301 03:07:28< iceiceice> while you're at it, you might consider buying yourself a whole new kitchen, right here on the wesnoth forums!
20140301 03:09:24< fabi> shadowm: tea is a good idea. Some Earl Gray here.
20140301 03:09:41< vultraz> iceiceice: xDDD
20140301 03:17:47< Aishiko_laptop> fabi+earl gray=...... Picard?
20140301 03:18:17-!- Ivanovic_ [~ivanovic@x2f51b61.dyn.telefonica.de] has joined #wesnoth-dev
20140301 03:18:17-!- Ivanovic_ [~ivanovic@x2f51b61.dyn.telefonica.de] has quit [Changing host]
20140301 03:18:17-!- Ivanovic_ [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20140301 03:18:21< fabi> Aishiko_laptop: Yes, I think he prefers this sort of black tea.
20140301 03:19:20< Aishiko_laptop> fabi, I only know that because my brother was a trekie I must have seen every episode of every series of every season.  Some things you pick up by osmosis.... IPv6 is NOT one of those things
20140301 03:20:21< vultraz> Who's Picard? I only saw Star Trek: Into Darkness
20140301 03:21:01< Aishiko_laptop> Captain in the next generation, aka the bald guy
20140301 03:21:30< Aishiko_laptop> vultraz, ^^
20140301 03:21:41< vultraz> Oh that guy
20140301 03:21:49< vultraz> I've seen the meme
20140301 03:22:23-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 252 seconds]
20140301 03:24:17-!- Ivanovic_ is now known as Ivanovic
20140301 03:24:38< shadowm> As I suspected, GUI2 has its own decode_color() function... which takes a string.
20140301 03:26:12< shadowm> But.
20140301 03:27:01< shadowm> Does it really matter whether the code is running on BE or LE? I'm not looking at the word arrangement directly, so...
20140301 03:27:52< shadowm> And neither does int_to_color() in sdl_utils.cpp, for that matter.
20140301 03:28:13< shadowm> mattsc: Is the Mac bundle still built for PPC too?
20140301 03:28:25-!- Aishiko_laptop [~unknown@cpe-065-191-176-226.nc.res.rr.com] has quit [Ping timeout: 264 seconds]
20140301 03:28:58< vultraz> PPC?
20140301 03:29:47< shadowm> PowerPC, a BE processor that used to be the cornerstone of the Mac platform before Apple switched to x86.
20140301 03:30:00-!- iceiceice [~chris@pool-108-21-12-72.nycmny.east.verizon.net] has quit [Ping timeout: 240 seconds]
20140301 03:30:10< vultraz> IIRC no
20140301 03:30:13< shadowm> I think it was also a RISC platform.
20140301 03:35:01< JunJM> Soliton: I have a question about the implementation to decide winner of a MP game
20140301 03:37:28-!- Velensk [~Velensk@cpe-75-187-86-211.woh.res.rr.com] has quit [Quit: Leaving]
20140301 03:37:35-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has joined #wesnoth-dev
20140301 03:38:43< iceiceice> shadowm: I guess that, whethe you are big endian or little endian, you can always assert e.g. x == ((x << 8) >> 8) ?
20140301 03:39:01-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140301 03:39:13< shadowm> I don't know what that means!!!
20140301 03:39:27< shadowm> I only know it's a bitwise operation and it scares me.
20140301 03:39:28< iceiceice> i think it means rotate 8 bits to left, rotate 8 bits to right?
20140301 03:39:49< iceiceice> and you should get the same thing back at the end?
20140301 03:39:54< _8680_> Shift, not rotate.
20140301 03:39:58< iceiceice> actually i think theres two version of rtotat
20140301 03:40:03< iceiceice> yeah thats what i meant
20140301 03:40:22< shadowm> I'm committing the fix.
20140301 03:41:23< iceiceice> hmm yeah maybe shift is different on big endian vs little endian, its hard for me to say...
20140301 03:41:44-!- Aishiko_laptop [~unknown@cpe-065-191-176-226.nc.res.rr.com] has joined #wesnoth-dev
20140301 03:42:40< _8680_> ((x << 8) >> 8) = (x, with the leftmost 8 bits zeroed)
20140301 03:43:31-!- goblinThing [44bd8c2c@gateway/web/freenode/ip.68.189.140.44] has joined #wesnoth-dev
20140301 03:43:50-!- kex [~kex@89.205.75.19] has quit [Ping timeout: 252 seconds]
20140301 03:43:52< shadowm> _8680_: Okay, since you know what you are talking about, would this operation yield different results on BE and LE: text.set_foreground_color((item->font_rgb() << 8) + 0xFF);
20140301 03:44:16< shadowm> Let's assume that item->font_rgb() yields the same value on both first.
20140301 03:44:55< vultraz> _8680_, iceiceice: you know that function looks really funny, since chatzilla inserts smiley faces with sunglasses in the middle of it
20140301 03:45:08< iceiceice> lol
20140301 03:45:16< shadowm> Yeah, you should disable that.
20140301 03:45:20< Aishiko_laptop> lol vultraz, lucky you!
20140301 03:45:32< _8680_> vultraz: Ew, like phpBB? Turn it off.
20140301 03:48:07< shadowm> _8680_: So?
20140301 03:50:10< iceiceice> http://stackoverflow.com/questions/7184789/does-bit-shift-depends-on-endianness
20140301 03:50:11< _8680_> shadowm: I don't know how endianness would affect that, I'm afraid.
20140301 03:50:32< vultraz> _8680_: https://dl.dropboxusercontent.com/u/95558676/smileyfunction.PNG
20140301 03:51:39< shadowm> Well, let's go with my fix then°!
20140301 03:51:58< iceiceice> 8) 8) 8)
20140301 03:52:32< mattsc> shadowm: I don't think so.  I seem to remember that Alarantalara told me that the libraries for PPC are still in the build package (or something along those lines), but we're not building for it any more.
20140301 03:52:51< irker854> wesnoth: Ignacio R. Morelle wesnoth:master 4523fb087540 / changelog src/display.cpp: Fix for unusual [theme] font_rgb attribute behavior with unit status labels http://git.io/edmyHg
20140301 03:53:05< shadowm> If the sidebar looks funny on one of those rare platforms, you know who to blame.
20140301 03:53:17< shadowm> fabi: ^^
20140301 03:53:56< mattsc> shadowm: "Not intentional at all (though it looks like the change was related to building for PowerPC, which you likely can't build for, though I don't have any problems dropping support for it—I should use lipo to get rid of the PowerPC half of the libraries to save space too)." -- Alarantalara
20140301 03:54:11< fabi> shadowm: Thanks. Would you also document  font_rgb in the wml reference wiki?
20140301 03:54:34< shadowm> fabi: No, not really, I don't know for sure where it's accepted inside [theme] and where it's not.
20140301 03:54:47-!- goblinThing [44bd8c2c@gateway/web/freenode/ip.68.189.140.44] has quit [Quit: Page closed]
20140301 03:54:47< shadowm> I do know that it only takes three values, not four like you are using.
20140301 03:54:51< JunJM> I was wondering if someone could explain what scons/build environments are
20140301 03:56:09< mattsc> Rats, you folks have been busy...  I can't catch up on all that right now.
20140301 03:57:00< shadowm> JunJM: Do you need to know something in particular about scons et al, or was that a more general question?
20140301 03:57:40< JunJM> general
20140301 03:57:59< fabi> mattsc: You mean the irc log is huge? 
20140301 03:58:08< JunJM> like a quick run down. I am new to the open source projects like git etc..
20140301 03:58:08< shadowm> JunJM: Hm, what platform are you on? (Windows, OS X, ...)
20140301 03:58:12< JunJM> windows
20140301 03:58:29< JunJM> also, i cant seem to download the game for windows as the files may be damaged?
20140301 03:58:48< mattsc> fabi: yeah, I was out for a jog and dinner and a drink and I come back to gigabytes of irc backlog  :P
20140301 03:59:11< shadowm> JunJM: Are you trying to build from source for coding?
20140301 03:59:23< JunJM> shadowm: yes
20140301 03:59:25< fabi> mattsc: Is this your first gsoc as a developer?
20140301 03:59:39< mattsc> fabi: no, I know the drill ...
20140301 03:59:44< JunJM> fabi: yes. excited but nervous. never done anything like this
20140301 04:00:01-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has quit [Quit: tomreyn]
20140301 04:00:04< mattsc> It might only be the second time though
20140301 04:00:04< shadowm> JunJM: In that case you should make sure you download the source code tarball (.tar.bz2), not the Windows installer (.exe).
20140301 04:00:26< JunJM> shadowm: well i have already downloaded the tar so we got that going xD
20140301 04:00:36< shadowm> But why do you say it's corrupt?
20140301 04:00:58< JunJM> well i didnt know which to download so i did both. the .exe wouldnt go through
20140301 04:01:11< JunJM> some error msg. idk if it was cause i downloaded the tar also but yea
20140301 04:02:09< shadowm> No, that wouldn't cause that. If the file seems to be corrupt, it's corrupt and you'll need to download it again. But if the tarball is not corrupt , then you won't need the Windows installer, anyway.
20140301 04:02:26< JunJM> ok. thanks
20140301 04:02:35< shadowm> _But_ if you are going to be contributing code, you really should clone the Git repository instead, since that's where development takes place.
20140301 04:03:12< JunJM> lol i was trying to find a way aroudn that for now. I am new to git and dont understand majority of the lingo
20140301 04:03:15< shadowm> But if you have the tarball there already, you might as well try building Wesnoth from source with your IDE of choice. We supply project files and dependencies for MSVC++ and Code::Blocks.
20140301 04:03:24< JunJM> shadowm: in the process of reading up now
20140301 04:03:38-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Ping timeout: 252 seconds]
20140301 04:04:02< shadowm> THe instructions for Code::Blocks are in projectfiles/CodeBlocks/README.txt in the tarball. For MSVC++, see: http://wiki.wesnoth.org/CompilingWesnothOnWindows
20140301 04:06:23< shadowm> fabi: Can I remove the superfluous fourth value from the WML, though?
20140301 04:06:41< shadowm> Or do you have a patch in your queue there that could conflict with such a change?
20140301 04:07:07< fabi> shadowm: Yes, I am busy with it.
20140301 04:07:48< fabi> shadowm: I am inserting the correct gray color.
20140301 04:08:11< shadowm> Ah, okay.
20140301 04:10:28< fabi> shadowm: And it works :-)
20140301 04:11:54< fabi> shadowm: I saw your screenshot with the side number being too near to the flag icon.
20140301 04:12:16< fabi> You have a pretty large flag there.
20140301 04:12:33< shadowm> HttT uses the same flag for side 1.
20140301 04:14:04< shadowm> iceiceice: A different buffer is used for server notifications, right?
20140301 04:14:23< iceiceice> actually every speaker gets their own buffer it seems
20140301 04:14:30< shadowm> Huh.
20140301 04:14:34< iceiceice> and the server is just one of the speakers
20140301 04:14:49< iceiceice> additionally, before i made changes, the buffers were never cleared
20140301 04:14:53< iceiceice> even when the game ended and another began
20140301 04:15:08< iceiceice> and the messages were displayed *oldest* first and then the most recent were truncated...
20140301 04:15:20< iceiceice> i remember there were some times i would play a game and then later spec a game, 
20140301 04:15:28< iceiceice> and then i would get notifications "iceiceice has taken control"
20140301 04:15:28< shadowm> Anyway, it looks fine to me: https://dl.dropboxusercontent.com/u/21371130/screenshots/wesnoth-1.11.11-notification-test.png
20140301 04:15:37< iceiceice> but it was a lie.
20140301 04:15:50< shadowm> Maybe just a bit too long.
20140301 04:15:59< iceiceice> could shorten to 3 or 4 messages i guess
20140301 04:16:12< shadowm> I don't know for sure. It may be KDE being KDE.
20140301 04:16:15< iceiceice> its just changin a number in video.cpp
20140301 04:16:39< iceiceice> err sry game_display.cpp
20140301 04:17:26< shadowm> I wonder, does the protocol used support more application information like an icon or something?
20140301 04:18:26< iceiceice> actually, yes:
20140301 04:19:27< iceiceice> https://github.com/wesnoth/wesnoth/blob/master/src/game_display.cpp#L833
20140301 04:19:39< iceiceice> it seems we are already passing it a string for "app_icon" but i have no idea why we dont get icon
20140301 04:20:20< JunJM> shadowm: im guessing msvc 2012 is fine 
20140301 04:20:22< shadowm> Because that image doesn't exist.
20140301 04:21:14< shadowm> JunJM: I don't know for sure. Those instructions appear to have been written for MSVC++ 2008 and 2010 only, and I'm on Linux so I don't get to play around with that kind of thing. ;)
20140301 04:21:58< JunJM> shadowm: haha. fair enough. 
20140301 04:22:08< iceiceice> hmm i see we have wesnoth-icon.png in data/core/images/
20140301 04:22:14< iceiceice> maybe should change to point there?
20140301 04:22:20< iceiceice> or is that icon too large? idk
20140301 04:22:44< shadowm> This is dbus, right? Does it recommend any dimensions in particular?
20140301 04:22:48< JunJM> shadowm: out of curiosity how competitive is the gsoc for wesnoth?
20140301 04:23:18< sachith500> :D
20140301 04:23:32< sachith500> I'd like to hear the answer to that too :p
20140301 04:23:39 * Aishiko_laptop likely doesn't want to know
20140301 04:23:47< sachith500> haha
20140301 04:23:57< shadowm> Hm, it depends on how many students apply for the same task I guess. I haven't really paid much attention to the student selection process in the past, I'm afraid.
20140301 04:24:07< sachith500> so far JunJM is the only other guy who I saw asking about the project I'm interested in
20140301 04:24:25< sachith500> shadowm, are multiple applications allowed?
20140301 04:24:28< Aishiko_laptop> last year I missed out because my proposal writing skills were, and likely still are not that great.
20140301 04:24:42< Aishiko_laptop> sachith500, according to Google they are
20140301 04:24:53< sachith500> yeah but it differs from org to org
20140301 04:24:53< Aishiko_laptop> at least it was that way last year
20140301 04:25:10< shadowm> sachith500: Do you mean within Wesnoth, or in GSoC in general? I don't know about the first, but for GSoC students are only allowed to participate on one project at a time I believe.
20140301 04:25:15< Aishiko_laptop> ohh if the org will allow more then 1 proposal?
20140301 04:25:16< sachith500> within wesnoth
20140301 04:25:25< sachith500> yeah
20140301 04:25:38< iceiceice> https://wiki.ubuntu.com/NotificationDevelopmentGuidelines#If_a_notification.2BIBk-s_icon_looks_blurry_or_jagged
20140301 04:25:42< iceiceice> shadowm: ^
20140301 04:25:49< Aishiko_laptop> well you can apply to many but you only get 1 acceptance is how I was told it worked
20140301 04:26:03< iceiceice> oh wait thats not relevant, nm
20140301 04:26:21< iceiceice> err at least i'm not sure yet
20140301 04:26:22< shadowm> http://www.google-melange.com/gsoc/document/show/gsoc_program/google/gsoc2014/help_page#6._Can_a_student_submit_more_than_one
20140301 04:26:41< shadowm> That applies to proposals in a single project.
20140301 04:27:03< sachith500> oh
20140301 04:27:14< sachith500> well I was a little interested in the AI project as well
20140301 04:27:33< sachith500> I've already got the code built and did a few PRs, so I thought I'd look around a bit
20140301 04:28:05< shadowm> iceiceice: Boo, we don't have a SVG icon yet because nobody has been able to contact Sgt Groovy with the right words, it seems.
20140301 04:28:30< JunJM> man i feel like i aint gonna get any proposals accepted -_-
20140301 04:28:44< iceiceice> urgh thats too bad
20140301 04:28:57< iceiceice> (@shadowm)
20140301 04:29:01< shadowm> iceiceice: However, icons/*-Mac.png are 128x128 bitmap icons. You can't rely on them existing, though, so you'd need to copy them over to images/.
20140301 04:29:21< Aishiko_laptop> sachith500, I'm applying for the spritesheets because I've worked with them before, and have an idea of what it wants, I even have an idea for a utility to convert all the current sprite images into spritesheets and hope to implement it in addition converting the code for the sprites
20140301 04:29:28< iceiceice> ok i guess we can try copying them to images/wesnoth-icon-small.png and seeing how they look
20140301 04:29:35< shadowm> At least I don't think icons/ is included in the Windows distribution.
20140301 04:29:46< iceiceice> sorry, images/
20140301 04:29:54< iceiceice> wait that is what i wrote
20140301 04:30:01< Aishiko_laptop> if they aren
20140301 04:30:23< sachith500> ahh
20140301 04:30:24< Aishiko_laptop> t I can zip them and email them to whomever needs them until that changes
20140301 04:30:35< sachith500> cool
20140301 04:30:56< sachith500> I wanted to work on the multiplayer data cause I think it'll be valuable to the game
20140301 04:31:04< shadowm> iceiceice: Oh. Yeah, I can try that.
20140301 04:31:14< sachith500> My initial impression was that it was more valuable than the AI project
20140301 04:31:18< sachith500> hence my reason for picking it
20140301 04:32:06< shadowm> iceiceice: Nothing.
20140301 04:32:12< Aishiko_laptop> well I've got limited time until May (I'm taking 18 credits) so I have to stick with something that I already have an understanding of the basics of, otherwise I'll not be able to do a good job on anything
20140301 04:32:31< iceiceice> shadowm: yeah i'm not getting anything either
20140301 04:32:58< sachith500> ah :\
20140301 04:33:28< Aishiko_laptop> sachith500, plus I like the idea of doing something that will benefit all players of the game, which in this case is speeding up loading and reducing overhead, so instead of 20 files getting loaded for each unit only one is
20140301 04:33:47< shadowm> It purportedly would also benefit artists greatly.
20140301 04:34:08< JunJM> sachith500: did you have your heart set on the multiplayer data portion
20140301 04:34:13< iceiceice> my 2 cents: i dont think any development work is "more important" than other work, most of its incomparable, as it will benefit different groups of users / developers
20140301 04:34:18< Aishiko_laptop> shadowm, my brother is an artist but he doesn't do sprites (at least not that I know of!)
20140301 04:34:38< sachith500> yeah iceiceice I know
20140301 04:34:43< shadowm> Aishiko_laptop: Yeah, unit artists, of course.
20140301 04:34:50< sachith500> JunJM yeah pretty much
20140301 04:35:05< shadowm> I have worked a lot on unit art myself, but not with animations because I am afraid of animations.
20140301 04:35:40< sachith500> I would love to work with evaluation functions and the like
20140301 04:35:49< sachith500> it's just I figure someone better will come along
20140301 04:35:49< shadowm> So my opinion in that regard leans more toward the storage logistics.
20140301 04:35:52< sachith500> for the AI project
20140301 04:36:07< Aishiko_laptop> iceiceice, I aggree, if the network devs don't do their thing then the Multiplayer dev's work goes unused, if the AI guys/gals stop, then things don't get better, its a matter of picking what you know or want to learn or what your good at.  And then doing it!
20140301 04:38:07< shadowm> All that said, I've encountered plenty of performance er... 'situations' here that would really benefit from spritesheets.
20140301 04:39:27< shadowm> It's just too evident at times that the game is loading lots of small separate files on the spot without a disk access time guarantee.
20140301 04:39:38< AI0867> alignment and traits look bright cyan here
20140301 04:39:43< Aishiko_laptop> shadowm, I'm sure of that!  I saw all those images and was thinking of the all the things that it could improve from less disk space to faster loading, etc
20140301 04:39:55< shadowm> AI0867: You're too late and several commits behind, I already fixed it.
20140301 04:41:17< iceiceice> shadowm: https://developer.gnome.org/notification-spec/#icons-and-images-formats
20140301 04:41:30< iceiceice> i guess at least there it looks like theres some part of the icon spec we aren't doing?
20140301 04:43:00< Coffee_irc> shadowm: spritesheets might make it harder to make art?
20140301 04:43:17< shadowm> Coffee_irc: Is that a question?
20140301 04:43:23< Coffee_irc> and cause issues with large images that stretch out of the hexes
20140301 04:43:29< iceiceice> hmm no maybe i am misreading it
20140301 04:43:42< Coffee_irc> well, it is something I remember Jetrel mentioning in a forum thread once
20140301 04:43:49< shadowm> Coffee_irc: Just have people draw on 72x72 boxes.
20140301 04:43:52< Coffee_irc> although he seemed to be for spritesheets in general as well
20140301 04:43:59< iceiceice> oh hmm
20140301 04:44:08< iceiceice> maybe we are suppsoed to have file:// in front of our path?
20140301 04:44:10< shadowm> I think Frogatto's approach involves literal colored bounding boxes.
20140301 04:44:14< Aishiko_laptop> Coffee_irc, Thank you for mentioning that, I'll have to include tests to make sure that doesn't happen!
20140301 04:45:36< iceiceice> shadowm: do you know what 'game_config::path' looks like at runtime?
20140301 04:46:07< shadowm> No, but I'm pretty sure it's not an URL.
20140301 04:46:59< shadowm> /home/shadowm/src/wesnoth
20140301 04:47:39< iceiceice> hmm according to grep it look slike "/usr/local/games/wesnoth"
20140301 04:48:07< shadowm> I mean, of course it differs from invocation to invocation.
20140301 04:48:15< iceiceice> ok maybe i will try putting file://in front of this or something
20140301 04:48:30< shadowm> It should be whatever "Data directory:" says on startup, which means it may also not be an absolute path.
20140301 04:48:41< iceiceice> or i guess jsut "file:"
20140301 04:49:32< shadowm> A proper URL should look like file:///home/shadowm, I believe.
20140301 04:49:41< Aishiko_laptop> shadowm, mordate (I hope I spelled that right) suggested looking at how Frogatto does it and I'm reviewing those before I talk to him so I can ask intelliegent questions (which I sometimes don't ask)
20140301 04:49:51< shadowm> No, it's mordante with an n.
20140301 04:50:23< iceiceice> i see, my browser agrees with you shadowm
20140301 04:50:31< shadowm> Aishiko_laptop: Yeah, I think I saw that.
20140301 04:51:01< shadowm> iceiceice: Aded file:///, now I see the "unknown icon" placeholder from KDE.
20140301 04:51:19< shadowm> Oh wait.
20140301 04:51:32< shadowm> I should add file:// instead because the path already begins with a slash.
20140301 04:51:45 * Aishiko_laptop is really really bad with names, and is usually to *something* to ask someone to remind her what their name is after seeing them in class for 2 months
20140301 04:51:59< iceiceice> yeah "file://" + ... is what i am compiling now
20140301 04:52:03< shadowm> Nope, still the unknown icon placeholder.
20140301 04:52:47< iceiceice> hmm
20140301 04:52:57< iceiceice> does it need to be like "data/core/images/..." ?
20140301 04:53:14< shadowm> No, because that wouldn't be a path that exists.
20140301 04:53:52< shadowm> As far as everyone outside Wesnoth cares, data/core/images/ and images/ are different and not special in the least, just two different paths.
20140301 04:54:15< iceiceice> i guess i'm not sure exactly what path we are giving to dbus riht now
20140301 04:54:27< iceiceice> i think i might insert some debugging output for that 
20140301 04:54:27< shadowm> I dropped wesnoth-icon-small.png in the latter and Wesnoth is passing "file:///home/shadowm/src/wesnoth/images/wesnoth-icon-small.png
20140301 04:54:41< iceiceice> i see
20140301 04:54:52< shadowm> -rw-r--r-- 1 shadowm shadowm 28355 Mar 28  2013 /home/shadowm/src/wesnoth/images/wesnoth-icon-small.png
20140301 04:55:13< iceiceice> so shouldnt it be bassing wesnoth/data/core/images/wesnoth-icon-small.png ?
20140301 04:55:17< iceiceice> *be passing?
20140301 04:55:29< shadowm> The receiver has no way to know what that means.
20140301 04:56:13< iceiceice> i see i think im confusing the repo with the final result
20140301 04:57:47< shadowm> const char***     !
20140301 04:57:50< shadowm> Gah!!!
20140301 04:58:22< shadowm> This is the first time I've seen triple-indirection in use in a legitimate... thing!
20140301 04:58:43< iceiceice> release the hounds
20140301 04:59:00< Aishiko_laptop> of the baskervilles!
20140301 05:00:10< shadowm> Okay, I give up, I don't intend to learn how to dbus right now.
20140301 05:00:48< Aishiko_laptop> sort of like me and AI, I'm interested but don't have the time to devote to it right now
20140301 05:00:54< shadowm> It's worth noting that, since game_config::path is not necessarily absolute, you may need to use normalize_path() (from filesystem.hpp) on it to get something that can be safely used to build an URL.
20140301 05:01:26< shadowm> But here it's absolute, and it still doesn't work like that, so bah.
20140301 05:02:54< shadowm> OTOH it looks like normalize_path() is dumb, not sure if it'd be able to deal with a path like "../wesnoth"  correctly.
20140301 05:03:31< shadowm> Eh, not sure. I really should stop looking at code now.
20140301 05:05:42< iceiceice> yeah i have no idea why its not working
20140301 05:08:12< iceiceice> it seems like this will be a pain to test since it will be different in the compile from source vs. install from debian package scenarios
20140301 05:09:36< sachith500> did someone say triple indirection?
20140301 05:09:38< sachith500> where?
20140301 05:09:49< irker854> wesnoth: Ignacio R. Morelle wesnoth:master 99b1a1f6bdad / src/ (filesystem.cpp filesystem.hpp): Make is_path_sep() a public function and document it http://git.io/nXiDiQ
20140301 05:09:52< irker854> wesnoth: Ignacio R. Morelle wesnoth:master 5bc77e325fd6 / src/gui/dialogs/wml_error.cpp: gui2/twml_error: Use filesystem.cpp's is_path_sep() instead of our own thing http://git.io/-MA4oQ
20140301 05:18:27-!- JunJM [60fc984b@gateway/web/freenode/ip.96.252.152.75] has quit [Quit: Page closed]
20140301 05:27:16-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140301 05:31:37-!- kex [~kex@89.205.75.19] has quit [Ping timeout: 240 seconds]
20140301 06:10:27-!- exciton_ [chuck-the-@95.73.189.186] has joined #wesnoth-dev
20140301 06:14:38-!- exciton [chuck-the-@89.208.170.132] has quit [Ping timeout: 240 seconds]
20140301 07:07:29< shadowm> fabi: Hm, starting location labels in the editor disappear whenever I switch open maps or load a map, until I change a side's starting location.
20140301 07:08:05< fabi> shadowm: Noted, thank you.
20140301 07:08:33< shadowm> And this problem was introduced between 1.11.10 and HEAD... wait.
20140301 07:08:49< shadowm> Hm, didn't someone else change something about labels since 1.11.10, actually?
20140301 07:11:44< vultraz> fabi: could you please work on making the custom tod dialog actually save
20140301 07:12:08< irker854> wesnoth: fendrin wesnoth:master 91b16cfefd2e / src/reports.cpp: Add a white space between the side flag and number in the sidebar. http://git.io/Psbqlw
20140301 07:12:10< irker854> wesnoth: fendrin wesnoth:master 9008b167553a / data/scenario-test.cfg: Add a testcase for units with level >9. http://git.io/ZiEKBQ
20140301 07:12:12< irker854> wesnoth: fendrin wesnoth:master 4e03757bbe16 / data/themes/default.cfg: Adjust space for the unit side flag and colors for attributes. http://git.io/ZSU1Mw
20140301 07:12:33< shadowm> Okay, let me take a look at these commits first...
20140301 07:14:07< shadowm> !log bd09e01992e333a10a87d5bb5d520efca03f607b
20140301 07:14:08< shikadibot> shadowm: Revision bd09e01992e333a10a87d5bb5d520efca03f607b (<Anonymous>) on Thu Jan  1 00:00:00 1970:
20140301 07:14:11< shikadibot> shadowm: <Invalid commit>
20140301 07:14:14< shikadibot> shadowm: Web interface URL: https://github.com/wesnoth/wesnoth-old/commit/bd09e01992e333a10a87d5bb5d520efca03f607b
20140301 07:14:17< shadowm> Pfft.
20140301 07:15:18-!- exciton_ [chuck-the-@95.73.189.186] has quit [Read error: Connection reset by peer]
20140301 07:15:27-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140301 07:15:32-!- exciton [chuck-the-@89.208.170.132] has joined #wesnoth-dev
20140301 07:15:37< shadowm> !log bd09e01992e333a10a87d5bb5d520efca03f607b
20140301 07:15:38< shikadibot> shadowm: Revision bd09e01992e333a10a87d5bb5d520efca03f607b (<Anonymous>) on Thu Jan  1 00:00:00 1970:
20140301 07:15:41< shikadibot> shadowm: <Invalid commit>
20140301 07:15:44< shikadibot> shadowm: Web interface URL: https://github.com/wesnoth/wesnoth/commit/bd09e01992e333a10a87d5bb5d520efca03f607b
20140301 07:15:52< shadowm> Well, either GitHub changed something, or I don't know.
20140301 07:16:07< shadowm> jamit, fabi: This commit caused it: https://github.com/wesnoth/wesnoth/commit/bd09e01992e333a10a87d5bb5d520efca03f607b
20140301 07:16:38< fabi> vultraz: It is not urgent. There is the need to add some translatable strings which means it can make it in 1.12.1 but not in 1.12.0 anymore.
20140301 07:17:33< shadowm> I... I probably should stay away from that commit. :|
20140301 07:17:42< shadowm> *sigh*
20140301 07:17:59< fabi> shadowm: What do you think?
20140301 07:18:25< shadowm> fabi: About the commit? I don't really understand the code it changes.
20140301 07:18:46< vultraz> fabi: then it should be disabled for 1.12.0
20140301 07:19:15< shadowm> My first guess is that the editor does something it doesn't take into account.
20140301 07:19:42< shadowm> (Does label visibility even make sense while in the editor? I'd say no, it doesn't.)
20140301 07:20:01-!- kex [~kex@89.205.75.19] has quit [Ping timeout: 240 seconds]
20140301 07:20:49< fabi> vultraz: Yes.
20140301 07:20:54< shadowm> Would you prefer that I file a bug for it? But who should I assign it to?
20140301 07:21:28< fabi> shadowm: I would love to have a bug report for the issue. It is fine to assign it to me.
20140301 07:27:39< shadowm> fabi: Done. https://gna.org/bugs/index.php?21732
20140301 07:28:56< fabi> shadowm: Thank you
20140301 07:32:42< shadowm> It appears the spammers targeting the #Wesnoth forums want me to turn the blacklist into a list of cities in the UK, or at least England.
20140301 07:34:06< shadowm> I can't help but wonder when selling kitchens became so profitable.
20140301 07:34:54< fabi> yeah, the kitchen spam is strange.
20140301 07:47:43 * fabi wonders if there is a way to determine if the game or the editor is running.
20140301 07:48:01< fabi> From out of the help browser.
20140301 07:52:54< fabi> hmmm, I can abuse the hotkey scope system.
20140301 07:59:43< irker854> wesnoth: fendrin wesnoth:master 1814f21321c5 / src/help.cpp: Abuse the hotkey scope system to show undiscovered units in the editor. http://git.io/jmO6Dg
20140301 08:02:56< shadowm> Or you could have added a flag somewhere visible that's set when the editor is running.
20140301 08:03:57< shadowm> Or dynamic_casted the closest display object pointer you could find to editor_controller and checked against a null pointer? N
20140301 08:04:59< shadowm> I once considered publishing an editor_controller pointer in resources.hpp, but meh.
20140301 08:05:55< shadowm> Oh wait, scrap the dynamic_cast idea, editor_controller is a controller_base, not a display.
20140301 08:07:29< shadowm> In any case, that kind of clever abuse is likely to accidentally break (or at least give someone looking at the undocumented code a migraine) at some point in the future if the underlying semantics change.
20140301 08:09:12< shadowm> game_config::in_editor would be far more self explanatory in a conditional expression than hotkey::is_scope_active(hotkey::SCOPE_EDITOR) to the outsider ("what are hotkey scopes?", "how do they behave normally?", etc.).
20140301 08:09:59< shadowm> Incidentally, there is a  game_config::editor. No idea what it's actually used for, though.
20140301 08:12:08-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20140301 08:13:35< shadowm> It's read by [cursor.cpp] use_color_cursors()... and set by no-one.
20140301 08:14:03< shadowm> But it might be worth considering making use of it again!
20140301 08:20:11< irker854> wesnoth: fendrin wesnoth:master d4494d7e79b6 / data/themes/default.cfg: Fix bug #21542: Observers icon on the top bar overlaps the Menu button http://git.io/x7KA-A
20140301 08:20:49< zookeeper> shadowm, you never played EI before? :p
20140301 08:21:26< zookeeper> also, i'm tempted to make a forum account named ReferenceWML so i could use it to smugly post reference links to threads like this: http://forums.wesnoth.org/viewtopic.php?f=21&t=40099
20140301 08:32:11-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20140301 08:44:22-!- mordante [~mordante@roadie.xs4all.nl] has joined #wesnoth-dev
20140301 08:44:22-!- mordante [~mordante@roadie.xs4all.nl] has quit [Changing host]
20140301 08:44:22-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20140301 08:44:39< mordante> servus
20140301 08:45:22< irker854> wesnoth: fendrin wesnoth:master 14da08a39117 / src/editor/map/ (context_manager.cpp map_context.cpp): fix bug #21485: Time Schedule Menu button crashes game http://git.io/YAEJ7w
20140301 08:49:35< fabi> hi mordante
20140301 08:50:26< mordante> servus fabi 
20140301 09:00:00< mordante> fabi, how long have you been a developer? Finally adding yourself to the credits ;-)
20140301 09:01:41< shadowm> zookeeper: I have, three times or so.
20140301 09:03:38-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140301 09:08:20-!- kex [~kex@89.205.75.19] has quit [Ping timeout: 252 seconds]
20140301 09:12:17< zookeeper> shadowm, but still the quality of the plot and dialogue still manages to surprise? yeah that's my experience too, and i've played it more than thrice... :>
20140301 09:12:36< zookeeper> although luckily not _all_ the dialogue is horrible anymore
20140301 09:13:54-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Ping timeout: 245 seconds]
20140301 09:14:13-!- sachith500 [~kvirc@112.135.10.210] has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/]
20140301 09:14:59-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20140301 09:15:03< shadowm> I'm not all that surprised, but every time I play it it feels worse, plot-wise.
20140301 09:18:32< zookeeper> well, it can't have actually gotten worse, of that i'm pretty sure :p
20140301 09:18:58< zookeeper> but the plot can't really be made to make any sense without a heavy rewrite
20140301 09:23:08< fabi> mordante: ~5 Years.
20140301 09:27:37-!- Boldi [~lipkab@apn-185-10-127-25.vodafone.hu] has joined #wesnoth-dev
20140301 09:28:04-!- Boldi is now known as lipkab
20140301 09:28:31< mordante> nice timing to add yourself to the credits ;-P
20140301 09:28:42< mordante> fabi, ^
20140301 09:29:17-!- lipkab [~lipkab@apn-185-10-127-25.vodafone.hu] has quit [Read error: Connection reset by peer]
20140301 09:29:50< fabi> mordante: I have been in the credits of LoW and DM before. I just missed to add me as a coder when my first c++ stuff went mainline.
20140301 09:30:09-!- Guest65903 is now known as root_
20140301 09:30:12-!- root_ is now known as SoulOfTheInterne
20140301 09:30:18-!- SoulOfTheInterne is now known as ToBeFree
20140301 09:30:59-!- lipkab [~the_new_l@apn-185-10-127-25.vodafone.hu] has joined #wesnoth-dev
20140301 09:31:24< lipkab> mordante: Hello.
20140301 09:34:02< lipkab> mordante: I wrote a summary of my researches on SDL2: http://pastebin.com/4qHTuLNz
20140301 09:36:29-!- mjs-de [~mjs-de@g228200163.adsl.alicedsl.de] has joined #wesnoth-dev
20140301 09:40:33-!- trademark_ [~trademark@nsg93-8-88-175-59-164.fbx.proxad.net] has joined #wesnoth-dev
20140301 09:41:39-!- thunderstruck [~zaibotren@cpc13-sgyl31-2-0-cust696.18-2.cable.virginm.net] has joined #wesnoth-dev
20140301 09:47:24< shadowm> Ugh, I hate it that the order in which gettext.hpp and <boost/bind.hpp> may be included isn't obvious.
20140301 09:47:38-!- vorobeez [~quassel@85.142.148.12] has joined #wesnoth-dev
20140301 09:48:52< shadowm> Hm, and now I'm getting similarly oblique error messages for a different reason.
20140301 09:50:30< shadowm> Oh, it believes my class is a... function.
20140301 09:51:19< shadowm> ... Somehow.
20140301 09:51:26< mordante> shadowm, the most vexing parse?
20140301 09:51:43< irker854> wesnoth: fendrin wesnoth:master 1871e9fca41c / src/editor/editor_controller.cpp: Don't show music tracks which can't be found and spit out a warning. http://git.io/fhAWjA
20140301 09:53:10< mordante> lipkab, thanks, if mostly fits with my observations, only hadn't looked at the audio yet
20140301 09:53:33-!- timotei_ is now known as timotei
20140301 09:53:53< shadowm> I kind of doubt it's the case in context: http://pastebin.com/rbYYcDwv    http://pastebin.com/0Hek52wL
20140301 09:54:16< mordante> lipkab, I first want to get Wesnoth compiling with 2.0 (no matter how bad the compiled version looks) and then start to work on the getting it to work properly
20140301 09:54:50< shadowm> Oh wait, is it because I'm binding an argument that no overload of copy_to_clipboard_callback() expects.
20140301 09:55:17< shadowm> Yep, that's it.
20140301 09:55:34< lipkab> mordante: Sure. So, what should I focus on if I wanted to submit a proposal?
20140301 09:57:03-!- Crab_ [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20140301 09:58:37< vultraz> fabi: BTW, did you fix the thing were selecting a music track in-editor causes it to scroll back to the top of the list?
20140301 09:58:51< fabi> vultraz: I can't fix that.
20140301 09:59:05< vultraz> Internal bug?
20140301 10:01:22< fabi> vultraz: The code just calls the menu to open again. But I can't tell the menu to scroll to the previous position. It is not a bug, just a missing feature. Or maybe you can call it a flow because I abuse the menu system in a way it is not meant to be used.
20140301 10:01:29< mordante> lipkab, what area are you interested in?
20140301 10:01:57< fabi> s/flow/flaw
20140301 10:02:10< vultraz> fabi: oh, I get it. Yeah, I ran into that when creating my custom GUI2 inventory. Had to save position on exit.
20140301 10:02:13< lipkab> mordante: Mostly the rendering/texture stuff.
20140301 10:02:44< fabi> vultraz: Did you manage to scroll down?
20140301 10:04:00< vultraz> well it was a listbox, so I just told it to select the row selected before it closed
20140301 10:05:22< fabi> mordante: Is there a reason why our menu and context menus are still gui1 ones?
20140301 10:06:11< fabi> vultraz: Okay, you used some gui2 widget.
20140301 10:06:21< lipkab> fabi: There's no such thing as a GUI2 menu.
20140301 10:06:24< vultraz> Yes
20140301 10:06:43< vultraz> A listbox
20140301 10:07:06< mordante> lipkab, I expected that, but that area is rather large, anything specifict
20140301 10:07:09< fabi> I guess a window with just a listbox is pretty much a GUI2 menu.
20140301 10:07:10< mordante> specific*
20140301 10:08:00< mordante> fabi, yes I did some work with them, but there were several unrender issues
20140301 10:08:32< mordante> however if we move to SDL2, that problem should fix itself
20140301 10:10:02< fabi> vultraz: It seems the editor's music playlist menu must wait to be fixed.
20140301 10:10:23< vultraz> Alright. Just that is is a bit annoying the way it is now
20140301 10:12:07< fabi> Indeed
20140301 10:13:07< shadowm> Oooookay... [filesystem.cpp] file_size() grabs the stat struct's st_size member, which is an off_t, and throws it back as an int.
20140301 10:13:18< shadowm> Here, sizeof(off_t) > sizeof(int).
20140301 10:14:43< shadowm> Although I guess for what I want to do file_size() is insufficient anyway since there's not much point in getting the size of a directory.
20140301 10:18:30< lipkab> mordante: Well, I'd definitely like to work on refitting sdl_utils for SDL2. Since that's almost the only place where actual SDL calls are used, I thought that if I get it done right, it won't be that a tremendous amount of work to port the other unit/terrain/etc rendering code to use it.
20140301 10:19:15< lipkab> Granted, I'm not really familiar with any of those subsystems.
20140301 10:20:15< lipkab> I looked a bit on the unit rendering code however and it looked heavily layered with SDL-specific stuff restricted to the bottom layers.
20140301 10:21:32-!- stevedes [0e61f21d@gateway/web/freenode/ip.14.97.242.29] has joined #wesnoth-dev
20140301 10:24:54-!- stevedes [0e61f21d@gateway/web/freenode/ip.14.97.242.29] has left #wesnoth-dev []
20140301 10:27:00< shadowm> fabi: Shouldn't that [binary_path] be in #ifdef EDITOR? (Re <http://forums.wesnoth.org/viewtopic.php?f=14&t=37141>).
20140301 10:27:44< fabi> shadowm: I thought it would make sense to keep it outside. 
20140301 10:27:55< fabi> But I am not sure.
20140301 10:28:50< shadowm> It's bad for add-ons to be pushy like that with their stuff, it may result in name conflicts.
20140301 10:29:54< fabi> Yes I know. I thought in case of the umc music addons this is okay.
20140301 10:29:57< shadowm> Hm, it's hard to believe nowadays that MP add-ons needed an unconditional global binary_path until 1.5.x.
20140301 10:30:17< shadowm> 1.5.1.
20140301 10:31:14< shadowm> Okay, what with the Divine Incarnation unit in UtBS? It looks like a broken derivative unit type.
20140301 10:32:00-!- Aishiko_laptop [~unknown@cpe-065-191-176-226.nc.res.rr.com] has quit [Ping timeout: 240 seconds]
20140301 10:32:08< vorobeez> Crab_: Hello. i want to work on AI at GSoC. What task would you advise me to begin understanding the code structure and logic of AI?
20140301 10:32:24< Crab_> vorobeez: hello!
20140301 10:32:47< shadowm> Okay, next time I merge someone's WML pull request I'll make sure to run wmllint before committing. *facepalm*
20140301 10:33:09< shadowm> And from now on all WML pull requests are required to be wmllint-clean.
20140301 10:33:11< lipkab> shadowm: What's with the forums again?
20140301 10:33:14< Crab_> vorobeez: the first thing is, probably, to read http://wiki.wesnoth.org/Customizing_AI_in_Wesnoth_1.8#how_AI_works.2C_a_short_developer-oriented_overview
20140301 10:33:21< shadowm> lipkab: I don't know, they look fine to me?
20140301 10:33:58< vorobeez> Crab_: thank you
20140301 10:34:05< lipkab> shadowm: I can't reach it and status.w.o says it's offline.
20140301 10:34:06< Crab_> vorobeez: and then, read http://wiki.wesnoth.org/Customizing_AI_in_Wesnoth_1.8#Customizing_AI_turn_sequence
20140301 10:34:41< Crab_> vorobeez: and then take a look at some of the 'candidate actions', for example here - https://github.com/wesnoth/wesnoth/blob/master/src/ai/testing/ca.cpp
20140301 10:34:48< shadowm> lipkab: Probably just a transient timeout because of the backup job.
20140301 10:35:03< lipkab> Ah. Thanks.
20140301 10:35:23< Crab_> vorobeez: the goal is to understand the AI turn structure, without going far into details of how each individual action works.
20140301 10:35:38-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20140301 10:36:15< Crab_> vorobeez: and ask questions,  if you'll have any. it's better to ask here since it'll benefit other students asking the same
20140301 10:36:59< shadowm> UtBS spoilers: https://dl.dropboxusercontent.com/u/21371130/screenshots/broken-eloh.png     (already fixed locally)
20140301 10:37:44< vorobeez> Crab_; ok, thanks. I will read all of it and try to understand. 
20140301 10:37:49< Crab_> vorobeez: during the course of the project, you'll most likely have to add new candidate actions ("IF this DO that WITH PRIORITY such") and modify existing ones.
20140301 10:42:12< shadowm> fabi: wmllint on LoW throws a load of errors I need to figure out at a later point before 1.12.0, but these three might be relevant to you: http://pastebin.com/HykmhCrQ
20140301 10:42:57< irker854> wesnoth: Ignacio R. Morelle wesnoth:master c55d5344eb34 / data/campaigns/Delfadors_Memoirs/scenarios/12_Terror_at_the_Ford_of_Parthyn.cfg: DM S12: Fix typo 'enemey' -> 'enemy', spotted by wmllint http://git.io/4GhwdA
20140301 10:43:00< irker854> wesnoth: Ignacio R. Morelle wesnoth:master cb934cf3e9ce / / (3 files in 2 dirs): UtBS S12: Fixed broken Divine Incarnation unit type due to missing [base_unit] http://git.io/CcxiDg
20140301 10:43:03< irker854> wesnoth: Ignacio R. Morelle wesnoth:master 9dca522a7803 / data/campaigns/Eastern_Invasion/scenarios/11_Captured.cfg: EI S11: Exempt a bat's line from wmllint spellchecking http://git.io/weNnRA
20140301 10:43:27-!- stevedes [0e61f21d@gateway/web/freenode/ip.14.97.242.29] has joined #wesnoth-dev
20140301 10:46:37< vorobeez> Crab_: priority will depend of global factors, right?
20140301 10:47:12< Crab_> in current candidate actions, priorities are usually "constant if action is good to do now or -1 if it's bad to do now"
20140301 10:48:21< fabi> shadowm: okay
20140301 10:48:21< Crab_> vorobeez: see default AI configuration - https://github.com/wesnoth/wesnoth/blob/master/data/ai/ais/ai_default_rca.cfg - actions are listed according to their priority (it's not the order here this matters, but the score in the macro, but they're just listed in descending priority order for convenience)
20140301 10:50:39< vultraz> !log 769320e2085d4d5412ac8522f5e78bead7f60a89
20140301 10:50:39< shikadibot> vultraz: Revision 769320e2085d4d5412ac8522f5e78bead7f60a89 (<Anonymous>) on Thu Jan  1 00:00:00 1970:
20140301 10:50:44< shikadibot> vultraz: <Invalid commit>
20140301 10:50:45< shikadibot> vultraz: Web interface URL: https://github.com/wesnoth/wesnoth/commit/769320e2085d4d5412ac8522f5e78bead7f60a89
20140301 10:50:49< vorobeez> Crab_: ok, and sorry for my bad english. I'm Russian student.
20140301 10:50:57< vultraz> shadowm: what's wrong here :/
20140301 10:51:09< shadowm> vultraz: Follow the link.
20140301 10:51:14< Crab_> vorobeez: don't worry, it's ok. Russian is one of my native languages, btw.
20140301 10:51:59-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140301 10:53:09< shadowm> Re shikadibot: my $content = eval{$curl->get("https://api.github.com/repos/wesnoth/wesnoth-old/commits/$commit")};
20140301 10:53:45-!- lipkab [~the_new_l@apn-185-10-127-25.vodafone.hu] has quit [Quit: Vannak idők, mikor menni kell]
20140301 10:54:18< vorobeez> Crab_: this is cool 
20140301 10:55:52< shadowm> Okay, fixed the API URL.
20140301 10:56:24-!- kex [~kex@89.205.75.19] has quit [Ping timeout: 240 seconds]
20140301 10:56:43< shadowm> Lesson secured -- GitHub repository rename redirects don't cover API calls.
20140301 10:58:39-!- mjs-de [~mjs-de@g228200163.adsl.alicedsl.de] has quit [Remote host closed the connection]
20140301 11:00:43< vultraz> !log 769320e2085d4d5412ac8522f5e78bead7f60a89
20140301 11:00:43< shikadibot> vultraz: Revision 769320e2085d (Charles Dang) on Sat Feb 22 10:35:43 2014:
20140301 11:00:47< shikadibot> vultraz: Defeat all enemies -> Defeat all enemy leaders
20140301 11:00:49< shikadibot> vultraz: Web interface URL: https://github.com/wesnoth/wesnoth/commit/769320e2085d
20140301 11:00:56-!- Gambit [~derek@wesnoth/developer/grickit] has joined #wesnoth-dev
20140301 11:00:58< shadowm> You still could have followed the previous link.
20140301 11:01:06< vultraz> I did
20140301 11:01:13< vultraz> This was to check it was fixed
20140301 11:01:27< shadowm> I did so first in /query.
20140301 11:01:48< mordante> lipkab it's not just the SDL calls, but IIRC accelerated rendering uses a OGL based drawing method, so redraw everything every frame
20140301 11:02:18< mordante> lipkab, so the code no longer needs to unrender things in the GUI(2) and halo code
20140301 11:13:04-!- spoffy [~spoffy@152.78.175.8] has joined #wesnoth-dev
20140301 11:28:02-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 252 seconds]
20140301 11:29:06< Soliton> JunJM: ask it then.
20140301 11:37:25-!- spoffy [~spoffy@152.78.175.8] has quit [Ping timeout: 264 seconds]
20140301 11:37:42-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has joined #wesnoth-dev
20140301 11:46:40-!- aquileia [52d4193f@gateway/web/freenode/ip.82.212.25.63] has joined #wesnoth-dev
20140301 11:48:35-!- sachith500 [~kvirc@112.135.10.210] has joined #wesnoth-dev
20140301 11:57:55-!- ALourenco [0252303e@gateway/web/freenode/ip.2.82.48.62] has joined #wesnoth-dev
20140301 12:06:13-!- shadowm [~ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 264 seconds]
20140301 12:06:35-!- shadowm [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20140301 12:25:25-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20140301 12:27:29-!- spoffy [~spoffy@152.78.175.8] has joined #wesnoth-dev
20140301 12:36:17-!- Aishiko_laptop [~unknown@cpe-065-191-176-226.nc.res.rr.com] has joined #wesnoth-dev
20140301 12:40:03-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140301 12:42:57-!- sachith500|2 [~kvirc@112.135.37.90] has joined #wesnoth-dev
20140301 12:44:24-!- kex [~kex@89.205.75.19] has quit [Ping timeout: 240 seconds]
20140301 12:45:36-!- sachith500 [~kvirc@112.135.10.210] has quit [Ping timeout: 240 seconds]
20140301 12:52:49-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Ping timeout: 240 seconds]
20140301 12:59:12-!- lipkab [~lipkab@apn-94-44-255-246.vodafone.hu] has joined #wesnoth-dev
20140301 13:03:20-!- lipkab [~lipkab@apn-94-44-255-246.vodafone.hu] has quit [Read error: Connection reset by peer]
20140301 13:21:38-!- serados [Junwei@nusnet-196-209.dynip.nus.edu.sg] has joined #wesnoth-dev
20140301 13:21:41-!- seralol [Junwei@nusnet-196-209.dynip.nus.edu.sg] has joined #wesnoth-dev
20140301 13:21:45-!- seralol [Junwei@nusnet-196-209.dynip.nus.edu.sg] has quit [Client Quit]
20140301 13:27:10-!- bwrsandman [~bwrsandma@13-73-252-216.dsl.colba.net] has joined #wesnoth-dev
20140301 13:27:11-!- wesbot changed the topic of #wesnoth-dev to: string+feature freeze active on master | 213 bugs, 350 feature requests, 28 patches | Logs: http://irclogs.wesnoth.org | Alternate logs: http://wesnoth.debian.net | Don't paste on IRC! Use a pastebin: http://pastebin.com | http://imagebin.org
20140301 13:30:16< bwrsandman> I have a question regarding sdl -> sdl2, mainly what is the most sensible approach to switching libraries. Do you try to get the two to coexist while phasing one out? Do you refactor all the old api calls at once? How do you debug when it seems like it would be broken until you're done. What's the strategy?
20140301 13:39:53< mordante> bwrsandman, for now I want to get Wesnoth to compile with both then test how well this code works before looking at the best way to proceed
20140301 13:39:56-!- Matei [~razvan@p16.eregie.pub.ro] has joined #wesnoth-dev
20140301 13:40:21< mordante> at the moment compiling with SDL2 causes a lot of compiler errors
20140301 13:41:29< stevedes> by that, you mean you just want a runnable build, which isn't necessarilt optimised for SDL2?
20140301 13:41:52< mordante> more like compilable first ;-)
20140301 13:47:41< bwrsandman> so the general strategy is to completely replace the api calls of the old library and replace with the new library calls until it compiles?
20140301 13:49:08< bwrsandman> How do you deal with completely broken environments that still build? Do you disable features to test just the basics and then open up the features when the basics work?
20140301 13:50:38< mordante> well it's more than just API calls, we also want to switch from surfaces to textures
20140301 13:51:13< mordante> what do you mean with environments? Operating systems?
20140301 13:52:35< bwrsandman> mordante: I mean just say you build it with the new API calls but you get a blank screen or textures are messy
20140301 13:52:44< bwrsandman> -> broken
20140301 13:53:49< bwrsandman> I am not suggesting blindly replacing api calls one to one, but that sort of break is still possible
20140301 13:54:26-!- irker854 [~irker@ai0867.net] has quit [Quit: transmission timeout]
20140301 13:56:00< mordante> well I first want to get Wesnoth compiling, then I've to see how bad things are
20140301 13:56:15< mordante> I expect bad, but I might be mistaken...
20140301 13:56:27< bwrsandman> haha, ok that's the approach I would take too. 
20140301 13:56:41< bwrsandman> Seems very similar to porting games to other OSs
20140301 13:56:52< mordante> then I'm going to look about how to proceed
20140301 13:56:56< bwrsandman> just seems very prone to regressions
20140301 13:57:14< bwrsandman> Maybe pre-emptive unittests
20140301 13:57:24< bwrsandman> could help
20140301 13:59:10< mordante> any idea how to write a good unit test for graphics?
20140301 14:01:15< bwrsandman> Never done it myself, the wine devs do, but they're pretty low level
20140301 14:02:19< bwrsandman> I guess another way would be to abstract graphic calls to a class within wesnoth
20140301 14:02:39< bwrsandman> Idk if that's already the case
20140301 14:04:21< mordante> not really, also you need to test when a unit moves there are no glitches
20140301 14:04:44< mordante> which sounds to me you almost need to capture the screen and compare that with a reference
20140301 14:06:13< bwrsandman> so: unittest are doable, but probably not worth it
20140301 14:06:36< bwrsandman> mordante: thank you so much for answering all my questions
20140301 14:07:07< mordante> I think they're worth it, but I don't see how they can be made properly
20140301 14:08:04< mordante> especially when we more certain operations to the GPU there might be minor differences between GPUs
20140301 14:08:28< aquileia> shadowm: 11:31 <shadowm> Okay, what with the Divine Incarnation unit in UtBS? It looks like a broken derivative unit type.
20140301 14:08:39< aquileia> Well, I ran wmllint myself...
20140301 14:08:50< mordante> bwrsandman, you're welcome
20140301 14:10:10< aquileia> shadowm: I somehow missed to add those 3 lines, sorry
20140301 14:15:31< mattsc> Crab_: hi - did you get my PM about the CA deletion problem?
20140301 14:16:36-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20140301 14:23:48< aquileia> shadowm: I ran it again and now it complains over the missing race, but back then I only had that bunch of errors you had already corrected somewhere else...
20140301 14:24:56-!- bwrsandman [~bwrsandma@13-73-252-216.dsl.colba.net] has left #wesnoth-dev ["Leaving"]
20140301 14:25:23< mordante> aquileia, what did you want to know about the child map stuff, do you want to change something in the config code or just use it?
20140301 14:26:51< aquileia> Well, I'd like to understand it because YAMG has to pass the result as a config
20140301 14:28:15-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140301 14:32:48-!- kex [~kex@89.205.75.19] has quit [Ping timeout: 240 seconds]
20140301 14:45:00< Aishiko_laptop> good morning mordante,
20140301 14:45:27< mordante> good afternoon Aishiko_laptop 
20140301 14:48:59< Aishiko_laptop> I put out a draft, very rough draft of my application for the spritesheets, and if you have time, would like to brainstorm with you on, how it could/should be implemented.
20140301 14:49:56-!- Yasin [~Yasin@41.129.76.234] has quit [Read error: Connection reset by peer]
20140301 14:52:15< Aishiko_laptop> I think I learn better that way, with the other person(s) bringing up things I either didn't think or or realize could be affected by the changes, or any number of other things
20140301 15:05:04-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140301 15:06:17< mordante> Aishiko_laptop, do you have a link to your draft?
20140301 15:09:13-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 264 seconds]
20140301 15:14:02< aquileia> mordante: I guess it's http://wiki.wesnoth.org/SpriteSheetApplicationSAB
20140301 15:14:29< mordante> thanks aquileia 
20140301 15:15:49-!- Matei [~razvan@p16.eregie.pub.ro] has quit [Read error: Operation timed out]
20140301 15:24:36< Aishiko_laptop> yes I do ohh and aquileia did while I was dealing with the dog (she started barking for no reason)
20140301 15:24:48-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20140301 15:25:07< mordante> Aishiko_laptop, which project did you apply for last year?
20140301 15:27:11< Aishiko_laptop> Syster's Mailman and LibreOffice, though on the later I didn't do so well, and screwed up my approach (in fact it was a pretty wretched approach and ill suited) and with Gnome's OPW I applied to the Kernel Newbies.
20140301 15:27:37< Aishiko_laptop> I guess I should start adding mordante before each line from now on
20140301 15:27:38< mordante> Aishiko_laptop, regarding the timeline we expect it to be fleshed out based on discussing the idea in IRC
20140301 15:27:56< mordante> not per se, I just do it sometimes
20140301 15:29:03< Aishiko_laptop> mordante, I totally understand, but I'm eager to figure out that timeline, but more then that the technical needs and steps needed to implement, test, and error handle it.
20140301 15:31:39-!- mjs-de [~mjs-de@f048145047.adsl.alicedsl.de] has joined #wesnoth-dev
20140301 15:33:18< Aishiko_laptop> mordante, as you can see some of the libraries a bit new to me, however, I'm confident that I'll be able to get it to work in a clean efficient manner that maintaining it will be easy.
20140301 15:39:53< mordante> I'm still reading
20140301 15:40:16< mordante> normally the timeline will form itself when the technical details are fleshed out more
20140301 15:40:37< Aishiko_laptop> of course, sorry I'll be quiet until your finished reading
20140301 15:41:19< mordante> no problem, you can't know when I'm finished (especially not since I'm talking to several people at the same time ;-) )
20140301 15:42:34< Aishiko_laptop> I figure you'll make some sort of comment or ask me a question(s) when your done, or straight up say something like "geez are you sure your a college student?"
20140301 15:42:56< mordante> I think for you SDL and WML are the most important things to learn
20140301 15:43:45< Aishiko_laptop> I agree, I've downloaded and installed SDL and have been reading up on how it does things.
20140301 15:43:58< mordante> we expect students and not fulltime developers, don't worry about that too much
20140301 15:44:08< sachith500|2> hello mordante :)
20140301 15:44:15< sachith500|2> I'm interested in a project as well
20140301 15:44:30< mordante> and in general I'm impressed how much students already know
20140301 15:44:32< Aishiko_laptop> What I was thinking of doing is using the SDL to get the frames, and then pass them to WML
20140301 15:44:41< mordante> hi sachith500|2 
20140301 15:44:45< sachith500|2> I can't make up my mind between the AI project and the multiplayer data collection project :)
20140301 15:45:00< sachith500|2> I got the initial impression
20140301 15:45:03< sachith500|2> from the mailing list
20140301 15:45:15< sachith500|2> that the AI project wasn't as high a priority as the data collection one
20140301 15:45:18< sachith500|2> was I wrong?
20140301 15:45:19 * Aishiko_laptop is less impressed with her lack of knowledge in certain areas... like error control, and input validation
20140301 15:45:37< mordante> sachith500|2, for the AI project you can best ask Crab_ details and for the MP Soliton 
20140301 15:45:46< sachith500|2> yeah, I know
20140301 15:45:49-!- kex [~kex@89.205.75.19] has quit [Remote host closed the connection]
20140301 15:45:56< sachith500|2> I already talked to Soliton and happygrue about that
20140301 15:46:04< mordante> sachith500|2, good
20140301 15:46:09< sachith500|2> I'm just curious about the AI project as well
20140301 15:46:14< sachith500|2> in terms of priority
20140301 15:46:24-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140301 15:46:44< sachith500|2> I would be more interested in that project, however I could easily get into the data collection project as well :)
20140301 15:47:01< sachith500|2> when I initially checked ideas out (it was a while back) I didnt see the AI project
20140301 15:47:07< mordante> sachith500|2, priority depends on who you ask. I prefer single player so like a better AI. However I think the MP project is also rather wanted
20140301 15:47:25< sachith500|2> hmm all right :D
20140301 15:47:32< sachith500|2> I meant in the project as a whole
20140301 15:47:41< sachith500|2> regarding slot allocations and such :)
20140301 15:47:53< Aishiko_laptop> mordante, I would think the average player does more MP then SP, however I'm like you I don't do the MP unless its against the computer
20140301 15:48:11< aquileia> Aishiko_laptop: lack of knowledge in error control? Welcome in the team... I have no idea of the details
20140301 15:48:12< mordante> Aishiko_laptop, WML is a text based language (a bit like XML) so it describes how things look SDL is about handling the imates
20140301 15:49:25< mordante> Aishiko_laptop, well it's always hard to judge I've played SP a long time before joining the forum etc. so there might be a lot of 'hidden' players
20140301 15:49:43< Aishiko_laptop> sachith500|2, I have a friend that  has worked the last I think 7 years as admin or mentor and he said they usually take the top proposals, or the students that they feel are most likely to stay on and complete their project though I would love to hear how BFW does it
20140301 15:49:58< aquileia> Aishiko_laptop: so we are two at least
20140301 15:50:13< sachith500|2> Well I guess I'll stick with the multiplayer data collection project, since I don't want all my previous research to go to waste :D
20140301 15:50:14< mordante> sachith500|2, we generally don't discuss that, nor the number of slots we have
20140301 15:50:25< sachith500|2> oh all right :D
20140301 15:50:30< sachith500|2> I kinda thought as much
20140301 15:50:36< sachith500|2> was expecting a reply along those lines
20140301 15:50:48< sachith500|2> I'll stick to the data collection one
20140301 15:50:51< sachith500|2> thanks for your help :D
20140301 15:51:21< Aishiko_laptop> mordante, lol me too, but mostly because I'm not good at getting the point across on a forum or email, usually someone takes it as an attack and I didn't feel it was worth the effort to correct impressions or even start a lot of the time
20140301 15:51:33< sachith500|2> aishiko_laptop, yeah that's true :)
20140301 15:52:01-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 240 seconds]
20140301 15:52:49< sachith500|2> mordante, would it be all right to submit 2 proposals?
20140301 15:52:53< sachith500|2> would that count against me?
20140301 15:53:35< mordante> sachith500|2, no you're allowed to and people have done it in the past
20140301 15:53:51< sachith500|2> all right thanks :D
20140301 15:54:18< sachith500|2> I'll focus on the data collection project, and if I have any brainwaves about the AI one, I'll send one for that as well :)
20140301 15:54:28< sachith500|2> you've been very helpful thank you
20140301 15:54:37-!- ALourenco [0252303e@gateway/web/freenode/ip.2.82.48.62] has quit [Quit: Page closed]
20140301 15:57:02-!- lipkab [~the_new_l@apn-185-10-124-224.vodafone.hu] has joined #wesnoth-dev
20140301 15:57:03< mordante> you're welcome
20140301 15:58:29< mordante> Aishiko_laptop, finised, did you have a chance to look at Frogatto?
20140301 15:59:15-!- sachith500|2 [~kvirc@112.135.37.90] has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/]
20140301 15:59:23-!- sachith500 [~kvirc@112.135.37.90] has joined #wesnoth-dev
20140301 15:59:44< Aishiko_laptop> I did, I have downloaded it as well and am going through it, the way it handles everything is how I was conceptualizing the back end of things to work, with regards to getting the frames
20140301 16:00:07< lipkab> mordante: Erm, I'm not sure what "unrender" means. If you rendercopy something onto the screen, it's going to stay there until you overwrite it, just like when blitting.
20140301 16:00:35< lipkab> SDL2 can use DirectX as a backend on Windows, by the way.
20140301 16:01:47< mordante> lipkab, when I draw a dialogue over an area and want to remove it I have to restore what was at that area before the dialog was placed over it
20140301 16:01:51< mordante> that's unrendering
20140301 16:02:25< mordante> Aishiko_laptop, good, it's no problem to borrow ideas ;-)
20140301 16:03:17< Aishiko_laptop> or to have similar ideas, history is rife with 2 people having the same idea at about the same time, and having the race to the patent office being the determiner of who "Owns" the idea
20140301 16:03:22< Aishiko_laptop> great minds and all that
20140301 16:03:56< lipkab> mordante: Ah-ha. So we'll instead just render both layers for each frame.
20140301 16:04:17< Aishiko_laptop> lipkab, I was looking at the addon manager and saw some strange fixmes that I couldn't figure out like /** @todo FIXME: get rid of this legacy "campaign"/"campaigns" silliness */ which is line 77 in the client.cpp
20140301 16:04:33< mordante> that's the idea of OGL and the z-buffer of the image determines what is drawn
20140301 16:04:38-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20140301 16:04:56< mordante> so when the top layer disappears the layer below is automatically visible
20140301 16:05:08< lipkab> Aishiko_laptop: And what do I have to fo with that? :S
20140301 16:05:11< lipkab> *do
20140301 16:05:54< mordante> of course borrowing can be a problem at times, but since the Frogatto engine is also open source it's not a problem
20140301 16:06:10< mordante> not sure whether they use WML
20140301 16:06:22< Aishiko_laptop> lipkab, sorry! I thought you might know since you work with the dialogs
20140301 16:06:22< lipkab> They don't.
20140301 16:06:33< sachith500> hello mordante, regarding the AI, does the current AI possess an evaluation function that maps the state of the map to a "score" for it?
20140301 16:06:52< sachith500> something to use with minimax perhaps :)
20140301 16:06:56< lipkab> Aishiko_laptop: No, that dialog is shadowm's stuff.
20140301 16:07:18< Aishiko_laptop> lipkab, ahhh thank you!  I'll ask him then the next time I see him!
20140301 16:07:45< mordante> sachith500, I'm not an AI expert, but I doubt there is such a function. Also note that our maps can have a lot of states so a value tree can be expensive to create
20140301 16:08:19< sachith500> yeah, I agree
20140301 16:08:20< mordante> I think some people did work on some minimax ideas, but on a small number of states
20140301 16:08:20< sachith500> but perhaps a simple one that looks only a few steps ahead?
20140301 16:08:21< sachith500> exactly
20140301 16:08:44< sachith500> hmm I will try to contact crab regarding it, but he seems rather busy at the moment :)
20140301 16:09:03< mordante> depends on how much variables you want to count
20140301 16:09:20< mordante> you can also try to ask mattsc
20140301 16:09:23< Aishiko_laptop> mordante, about my process I semi-outlined is it a good direction to be going or is there another direction I should be looking at?
20140301 16:09:24< sachith500> oh right
20140301 16:09:35< mordante> mattsc, did several AI things in Lua
20140301 16:09:39< sachith500> I read that he'd been playing about
20140301 16:09:54< sachith500> right I will ask him if I see him as well :)
20140301 16:10:26< mordante> Aishiko_laptop, well I'd like to discuss the project in more detail so you can add more technical details
20140301 16:10:55< Aishiko_laptop> alright, I'd be happy to
20140301 16:11:00< mordante> the technical details are the most important part of the application (and we are aware it takes a few weeks to get them to a good state)
20140301 16:11:34< mordante> that's why we like to discuss with students on IRC, to help them writing the proposal and get to know them
20140301 16:11:44< sachith500> :D
20140301 16:12:24 * Aishiko_laptop is what you see is what you get =P
20140301 16:12:24< mordante> Aishiko_laptop, do you have any specific questions? Need pointers where to look in the source code?
20140301 16:13:36< Aishiko_laptop> well yes I've not found it in the source code but that's more from lack of time to look and not sure where to start, though I think I'd find all or parts in the src and data directory trees
20140301 16:14:29-!- happygrue [~happygrue@c-66-30-155-184.hsd1.ma.comcast.net] has joined #wesnoth-dev
20140301 16:14:29-!- happygrue [~happygrue@c-66-30-155-184.hsd1.ma.comcast.net] has quit [Changing host]
20140301 16:14:29-!- happygrue [~happygrue@wesnoth/developer/wintermute] has joined #wesnoth-dev
20140301 16:14:48-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 240 seconds]
20140301 16:16:03< Aishiko_laptop> right now, from what I understand is that the individual image files are passed one at a time to the WML to be rendered on the screen, unless the units are in motion they remain in a static state (IE there is no X/Y movement to where it is drawn)
20140301 16:16:33-!- vorobeez [~quassel@85.142.148.12] has quit [Remote host closed the connection]
20140301 16:18:30< mordante> not exactly
20140301 16:18:47< mordante> WML defines where to find an image
20140301 16:19:04< mordante> that image is loaded from 1 file
20140301 16:19:13< mordante> optionally there are some modifications
20140301 16:19:26< mordante> then that image is rendered on the screen
20140301 16:19:53< mordante> the goal of the sprite sheets it to have 1 file with multiple images
20140301 16:20:10< mordante> so the WML needs to be modified to point to a file and an area
20140301 16:20:11< Aishiko_laptop> ohh, and right now the WML is not capable of taking a part of an image and only displaying that part
20140301 16:20:24< mordante> actually it is
20140301 16:20:34< sachith500> hello happygrue :D
20140301 16:20:52< sachith500> I've been reading up on the data analysis project :)
20140301 16:20:57< Aishiko_laptop> ohh this will be easier then I thought! at least in that respect
20140301 16:20:58< mordante> that happens at the modification level
20140301 16:21:00< happygrue> hi sachith500
20140301 16:21:02< happygrue> ah good
20140301 16:21:06< aquileia> shadowm's crop function could alreay do spritesheets
20140301 16:21:06< sachith500> I have an idea that might be useful
20140301 16:21:09< sachith500> what do you think
20140301 16:21:23< sachith500> oh wait
20140301 16:21:27< sachith500> it might already be there
20140301 16:21:30< sachith500> well my idea was
20140301 16:21:35< sachith500> along with the winning status
20140301 16:21:47< sachith500> we could also provide information about recruited units
20140301 16:21:49< sachith500> dead units etc
20140301 16:21:58< aquileia> But that would not bring loading speed advantages
20140301 16:22:04< sachith500> I know those stats are independently maintained already
20140301 16:22:07< sachith500> for each player
20140301 16:22:18-!- stevedes [0e61f21d@gateway/web/freenode/ip.14.97.242.29] has quit [Ping timeout: 245 seconds]
20140301 16:22:19< mordante> aquileia, exactly
20140301 16:22:42< happygrue> that would also be great information to have
20140301 16:22:44< sachith500> what do you think about providing that as well?
20140301 16:22:45< sachith500> yeah
20140301 16:22:47< sachith500> my idea was
20140301 16:22:53< Aishiko_laptop> mordante, so using SDL to define the area and passing it to the WML would?
20140301 16:22:55< sachith500> it could be great for data mining and machine learning
20140301 16:23:03< sachith500> and also for general knowledge
20140301 16:23:58< sachith500> for example for improving the ai recruitment, it might provide insight into which units on which map are more effective
20140301 16:24:07< sachith500> what do you think? :D
20140301 16:24:40< aquileia> mordante, Aishiko_laptop: Perhaps the syntax change could be minimal invasive? IIRC there are no cases where different heights of the frames of one animation occur, and all are square. Perhaps read the height and crop in increments of it?
20140301 16:24:56< happygrue> it would be most useful it it could also be paired on a faction by faction basis, so that we could see the unit spread of northerners vs. drakes, and so on
20140301 16:25:27< happygrue> but even if that part comes down the road, including the collection of unit data is great IMO
20140301 16:25:33< sachith500> yeah :D
20140301 16:25:38< happygrue> several other ways it could be uses also I guess
20140301 16:25:43< happygrue> *used
20140301 16:25:45< sachith500> yeah :D
20140301 16:25:45< Aishiko_laptop> aquileia, I agree that the less changes made to stable code the better in getting it working
20140301 16:25:55< mattsc> sachith500: Hi.  I'm not really around for the next few hours, just staying online and stopping by from time to time.
20140301 16:26:06< sachith500> oh hey mattsc :)
20140301 16:26:13< mordante> Aishiko_laptop, somehow you get it backwards, WML should define the area and SDL should only render that part of the image
20140301 16:26:18< sachith500> I was just curious about the ai project as well
20140301 16:26:20< mattsc> As for a map state parameter, no that doesn't really exist.
20140301 16:26:28< sachith500> hmm all right
20140301 16:26:39< sachith500> has anyone worked on that before?
20140301 16:26:48< Aishiko_laptop> ohhh, mordante that sometimes happens.
20140301 16:27:07< mattsc> Also, I'm not a CS person at all (important disclaimer!), but doesn't min-max imply playing through (all, ideally) the possibilities and looking several turns ahead?
20140301 16:27:16< mordante> aquileia, there are some snaggs, we have an image cache which also keeps team coloured images
20140301 16:27:21< sachith500> yeah
20140301 16:27:28< sachith500> but we could potentially prune the tree
20140301 16:27:30< sachith500> as we go along
20140301 16:27:39< mattsc> I doubt that that is possible in Wesnoth, there are just too many different possible states.
20140301 16:27:49< sachith500> yeah
20140301 16:27:59< sachith500> in that case we could limit it to a really low depth
20140301 16:28:02< mordante> aquileia, I'm quite sure this cache will not work nicely with the current crop code
20140301 16:28:09< sachith500> even 2 or 3 would be plenty i feel
20140301 16:28:10< mattsc> like half a move?  ;)
20140301 16:28:14< sachith500> hahaa
20140301 16:28:16< mattsc> s/move/turn
20140301 16:28:24< sachith500> even 1 move might work
20140301 16:28:28< sachith500> in wesnoth 1 move is a long time
20140301 16:28:29< mattsc> Seriously, that's the most I have managed to do so far (in Lua, at least)
20140301 16:28:29< sachith500> ;)
20140301 16:28:35< sachith500> oh?
20140301 16:28:38< sachith500> half a move?
20140301 16:28:42< aquileia> mordante: I'm mot saying the current crop should be used - I just said it's there
20140301 16:28:46< sachith500> is it efficiency issues?
20140301 16:28:49< mattsc> yep, what might the enemy do on its move.
20140301 16:28:52< mattsc> turn
20140301 16:29:02< mattsc> just too many options
20140301 16:29:12< Aishiko_laptop> so WML should send the starting pixel to the SDL along with the size of the area to take (in the case of a unit sprite 72x72) and then display that image, until it is time to display the next image in the same spot creating the animation effect.
20140301 16:29:21< sachith500> doesnt having the ai in the scripting code cause performance issues?
20140301 16:29:26< mordante> aquileia, I know ;-)
20140301 16:29:43< mattsc> but of course, as you say, you don't have to try out _all_ the options, so figuring out the "serious" ones is an important task
20140301 16:29:58< sachith500> yeah
20140301 16:30:10< mattsc> sachith500: sure, but putting it into C++ does not gain you that much.
20140301 16:30:19< sachith500> really?
20140301 16:30:22< mordante> Aishiko_laptop, yes, but the sprite-sheet project is mainly about the loading part
20140301 16:30:30< sachith500> assuming linear behaviour can you gimme a factor perhaps?
20140301 16:30:36< mattsc> 2
20140301 16:30:40< sachith500> really
20140301 16:30:41< sachith500> wow
20140301 16:30:48< sachith500> well that's not much for consideration then
20140301 16:30:53< sachith500> hmm
20140301 16:31:00< mattsc> Just a really rough guess.  And it really depends on what you are doing.
20140301 16:31:11< sachith500> yeah :D
20140301 16:31:18< mattsc> I actually have Lua code that is faster than it's C++ equivalent because it's "smarter".  :)
20140301 16:31:24< mordante> Aishiko_laptop, but the part which changes from one image to the next should not know about the sprite-sheets
20140301 16:31:30< aquileia> There are animations with 90x90 or 100x100, so putting them side on side to have the height for that would be the least syntax work, I guess
20140301 16:31:37< mattsc> So of course, putting that into C++ would make it faster yet.
20140301 16:31:42< sachith500> exactly
20140301 16:32:02< mattsc> But to be able to play through all possibilities, you'd need a couple orders of magnitude, and that's just not there, I think.
20140301 16:32:11< mattsc> So ... pruning the tree it is.  :)
20140301 16:32:12< sachith500> yup
20140301 16:32:32< mattsc> Anyways, I need to get going again.  Just wanted to say hi, for the most part.
20140301 16:32:33< sachith500> plus a bit of Dynamic programming might give a bit of efficiency
20140301 16:32:44< sachith500> yeah :D good talk 
20140301 16:33:18< mattsc> The efficient coding part is something I have to leave to others, I just don't know anything about that.  I know more about the AI behavior side of things ...
20140301 16:33:21< Aishiko_laptop> mordante, ohh, I've been focusing on the wrong aspect then.  But I'm not sure how to go about it then.
20140301 16:33:47< sachith500> oh
20140301 16:33:57< sachith500> algorithmic efficiency is kind of my strong point
20140301 16:34:06< mattsc> cool
20140301 16:34:09< sachith500> I have about 4-5 years of contest coding experience
20140301 16:34:18< mattsc> Anyways, off for now.  Talk to you later.  I'll stay online though and even if I'm not, I always check the logs later and (at least) search for my nick.
20140301 16:34:19< sachith500> codeforces, topcoder :D
20140301 16:34:28< sachith500> oh ok :D
20140301 16:34:29< mattsc> :)
20140301 16:34:32< sachith500> I intend to camp here anyways
20140301 16:34:34< sachith500> ;)
20140301 16:36:05< Aishiko_laptop> mordante, so basically the idea is to use the sprite sheets without telling SDL that its using a sprite sheet?
20140301 16:36:36< aquileia> Aishiko_laptop: I think he speaks about WML not knowing...
20140301 16:38:07< Aishiko_laptop> aquileia, if that is the case, then my understanding is that the WML tells SDL to display say the "attack" sequence and then SDL just displays the attack sequence with the proper timing and order of images
20140301 16:38:36< aquileia> Do you know the WML side code?
20140301 16:39:08< mordante> Aishiko_laptop, we tell SDL, but the part of Wesnoth that is aware of the is very close to SDL
20140301 16:39:22< aquileia> You specify like     image=path/unit-animation-[1~12].png
20140301 16:39:31< mordante> higher levels are not aware, they just ask to draw image foo somewhare
20140301 16:39:37< mordante> somewhere*
20140301 16:39:37< Aishiko_laptop> Somewhat it is still new to me, though I do know it passes, the type of image (or specific) and the timing
20140301 16:40:01-!- DCW [~Thunderbi@cpc66863-finc15-2-0-cust393.4-2.cable.virginmedia.com] has joined #wesnoth-dev
20140301 16:40:24< Coffee_irc> Aishiko_laptop: you've probably seen the reference wiki, but http://wiki.wesnoth.org/AnimationWML
20140301 16:40:38< mordante> Aishiko_laptop, for unit animations it does, but for that function calls a load function for some image
20140301 16:40:39< Aishiko_laptop> Coffee_irc, yes that is exactly where I started
20140301 16:40:48< aquileia> So currently this expands to a call of 12 images. If it instead would be 12 offsets for your spritesheet, no real syntax change is needed
20140301 16:41:21< mordante> Aishiko_laptop, how long will you be around?
20140301 16:41:23-!- sachith500 [~kvirc@112.135.37.90] has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/]
20140301 16:41:32-!- sachith500 [~kvirc@112.135.37.90] has joined #wesnoth-dev
20140301 16:41:49< Aishiko_laptop> I can be around for as long as needed (if you mean for today/this weekend) 
20140301 16:42:24-!- Matei [~razvan@p16.eregie.pub.ro] has joined #wesnoth-dev
20140301 16:43:06< mordante> ok, well I'm going to hunt for some food, will be back afterwards, then maybe we should discuss some more details
20140301 16:43:28-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20140301 16:43:33< Aishiko_laptop> thank you, I'll take care of some business too and see you when we're both back
20140301 16:44:43< happygrue> sachith500: just FYI improving AI performance when lots of units are involved (say 150-200 or more) is something I'd thrown out as a GSoC idea but we settled on others - partly because I don't think anyone was interested in mentoring it
20140301 16:44:59< happygrue> but I think if gains could be made there someday it would be good.  :D
20140301 16:45:19< sachith500> oh :D
20140301 16:45:44< sachith500> I think I saw that on the mailing list
20140301 16:45:56< sachith500> problem now is I cant decide on which project to pick
20140301 16:45:59< sachith500> >.<
20140301 16:46:29< sachith500> I'm pretty bad at picking from among interesting things :S
20140301 16:46:46< Aishiko_laptop> sachith500, welcome to the Club!
20140301 16:46:59< sachith500> aishiko_laptop your handle is so long xD
20140301 16:47:15< sachith500> but you're lucky
20140301 16:47:23< Coffee_irc> sachith500: you can use tab completion on nicknames
20140301 16:47:24< sachith500> at least you made up your mind :)
20140301 16:47:30< Aishiko_laptop> sachith500, don't you have tab-complete in your client
20140301 16:47:31< sachith500> DOH
20140301 16:47:44 * sachith500 feels like a complete idiot
20140301 16:47:48< sachith500> haha
20140301 16:47:55< Aishiko_laptop> start typing and then hit tab and then again and again if its cycling through more then one
20140301 16:48:01< sachith500> yeah i just did it :P
20140301 16:48:02< sachith500> works
20140301 16:48:05-!- prkc [~negusnyul@catv-188-142-184-108.catv.broadband.hu] has joined #wesnoth-dev
20140301 16:48:13< sachith500> does it work through the web client too
20140301 16:48:18-!- prkc [~negusnyul@catv-188-142-184-108.catv.broadband.hu] has left #wesnoth-dev []
20140301 16:48:25< sachith500> I'm usually on that because our university has a proxy
20140301 16:48:59< Aishiko_laptop> sachith500, most likely not, because then tab usually jumps you to the next field/button/etc
20140301 16:49:35-!- sachith500dummy [7087255a@gateway/web/freenode/ip.112.135.37.90] has joined #wesnoth-dev
20140301 16:49:38< sachith500dummy> easy way to confirm
20140301 16:49:50< sachith500dummy> no it works
20140301 16:50:01 * sachith500dummy pokes sachith500
20140301 16:50:06-!- sachith500dummy [7087255a@gateway/web/freenode/ip.112.135.37.90] has quit [Client Quit]
20140301 16:50:34< sachith500> Aishiko_laptop: what's your favorite faction (default) ?
20140301 16:50:35< sachith500> :D
20140301 16:51:54< Aishiko_laptop> faction?  I've done the campaigns so I have no clue what the factions are
20140301 16:52:00< sachith500> oh
20140301 16:52:14< sachith500> elves, undead, loyalist etc
20140301 16:52:15< Aishiko_laptop> sorry!
20140301 16:52:20< sachith500> :D
20140301 16:52:23< sachith500> i love undead
20140301 16:52:25< Aishiko_laptop> I'd have to say the elves
20140301 16:52:35< sachith500> i like to go all zombie apocalypse on my enemies :D
20140301 16:52:49< Aishiko_laptop> but I tend to pick the units based on what the enemy is when recruiting/recalling
20140301 16:52:58< sachith500> yeah me too :D
20140301 16:53:16< sachith500> we should play a game sometime :D
20140301 16:54:05< Aishiko_laptop> you'd trounce me for sure
20140301 16:54:12< vultraz> Undead coolest units
20140301 16:55:09< vultraz> are the*
20140301 16:55:26< sachith500> yeah
20140301 16:55:28< sachith500> dat adept
20140301 16:55:50< vultraz> But shadowm's custom units are even better
20140301 16:56:05< sachith500> oh? ageless units?
20140301 16:56:23< vultraz> no, no, After The Storm units
20140301 16:56:25< vultraz> UMC campaign
20140301 16:56:28< sachith500> oh
20140301 16:56:30< lipkab> There's no such thing as ageless units :P
20140301 16:56:32< sachith500> :D
20140301 16:56:39< sachith500> right i call them ageless
20140301 16:56:47< lipkab> Ageless is just a compilation.
20140301 16:56:50< sachith500> because they arent default
20140301 16:56:50< sachith500> :P
20140301 16:56:58< lipkab> Heh.
20140301 16:57:02< sachith500> I have trouble keeping track
20140301 16:57:04< sachith500> of all the names
20140301 16:57:08< sachith500> so many names >.>
20140301 17:00:19< aquileia> zookeeper, vultraz, Soliton: mordante suggested using one of the exosting RNG's for YAMG. The question is - should it be system independant (so I'd need the Wenoth RNG) or is it acceptable that seeds would no longer map to maps 1 to 1?
20140301 17:00:43< zookeeper> former's of course better
20140301 17:00:43< aquileia> I think it's an issue, but I'd like to verify
20140301 17:03:02< aquileia> So I have to read about seeding Wesnoth RNG?
20140301 17:03:29< aquileia> Oh and one more thing - this could then be used to cheat...
20140301 17:03:48< aquileia> Using the editor to manipulate the RNG
20140301 17:03:50< mattsc> sachith500: btw, a very simple map state could simply be something like summed value of each sides units.  Something like: sum( gold * HP/max_HP * (1 + XP/max_XP) )
20140301 17:04:00< aquileia> except if there was a second instance
20140301 17:04:12< sachith500> yeah
20140301 17:04:13< mattsc> It has its limitations of course, but it works amazingly well for some things
20140301 17:04:20< sachith500> yeah it should
20140301 17:04:29< sachith500> power of simplicity
20140301 17:04:33< mattsc> yep.
20140301 17:04:42< mattsc> And now I actually have to be off for real.  TTYL.
20140301 17:04:44< sachith500> simple evaluation too
20140301 17:04:47< sachith500> all right :D
20140301 17:05:48< lipkab> aquileia: Not really. The RNG is regularly reseeded (IIRC after each attack sequence).
20140301 17:06:14< aquileia> Three options: 1. leave pyrophorous's custom RNG  2. take std::rand and loose system independance  3. use Wesnoth RNG
20140301 17:06:18-!- mattsc [~mattsc@154.20.32.245] has quit [Quit: Ciao]
20140301 17:07:00< Aishiko_laptop> how does pyrophorous's custom work?
20140301 17:07:25< aquileia> lipkab: But if you can set a specific seed, that's very powerful on cheating - even if just for a single attack
20140301 17:07:49< lipkab> aquileia: Then we, uh, reseed on game start? :P
20140301 17:07:50< aquileia> Aishiko_laptop: Mersenne-Twister
20140301 17:08:14< Aishiko_laptop> I mean if there was a way to generically ask the system for say the time, could that be sufficiently random for seeding and still keep system independence??
20140301 17:08:38< aquileia> Setting the seed is a key feature
20140301 17:08:53< aquileia> It's like the ID of your map
20140301 17:09:26< Aishiko_laptop> something better left to those that are not knowledgeable about the ins and outs gotcha =)
20140301 17:09:30< aquileia> But introducing a third RNG is a bad idea, so mardante is right it should be swapped
20140301 17:09:38< sachith500> mersenne?
20140301 17:09:41< sachith500> like the primes?
20140301 17:10:28< aquileia> I don' understand the implementation, so... no idea
20140301 17:11:26< aquileia> Aishiko_laptop: No, it's just in this very specific case. Any input is welcome (you surely have more experience than me)
20140301 17:12:10< aquileia> That's the first time I use C++ instead of C, I'm struggling with lots of things
20140301 17:14:15< Aishiko_laptop> aquileia, I usually seed the random number generator with a whatever the system time (usually the milliseconds  or seconds) and then use that
20140301 17:14:46< sachith500> if the code gets called regularly
20140301 17:14:53< sachith500> that can be problematic right?
20140301 17:15:05< sachith500> if it gets called every 25 miliseconds ;)
20140301 17:15:24< sachith500> usually that's not the case though
20140301 17:16:19< aquileia> lipkab: Is it already reseeded?
20140301 17:16:34< lipkab> I don't know. Perhaps.
20140301 17:18:52< aquileia> I don't want to mess with that part of the code an explain in the commit message - I'm too lazy to call a second instance for the map generator. shadowm probaybly already hates me for my last commits...
20140301 17:18:53-!- lipkab [~the_new_l@apn-185-10-124-224.vodafone.hu] has quit [Quit: Vannak idők, mikor menni kell]
20140301 17:20:16< Aishiko_laptop> aquileia, he'll likely not be too enthusiastic about my ideas either
20140301 17:21:11< aquileia> Well, you didn't delay the last release by some minutes because your editor messed up the UTF8 encoding of a file...
20140301 17:24:44< aquileia> A single additional caracter did wreak havok (ok, now I'm exaggerating, it was just a warning message)
20140301 17:25:23< aquileia> s/caracter/character
20140301 17:25:27< Aishiko_laptop> aquileia, I'm usually the only coder on a project, so I don't let anyone down when I'm delayed at all
20140301 17:26:22< Aishiko_laptop> even the group projects usually their like "Let Aishiko, handle it, lets do this part instead." and run off with no plan to play with the physical part of the project...
20140301 17:26:49< aquileia> nice...
20140301 17:27:36-!- Matei [~razvan@p16.eregie.pub.ro] has quit [Ping timeout: 240 seconds]
20140301 17:27:36< aquileia> Wesnoth is definitely different from that :p
20140301 17:27:43< Aishiko_laptop> though this week I have done more of the guide someone new on how to code.  The biggest issue I see in them is the lack of attention to detail, and failure to understand what a function call does
20140301 17:28:13< Aishiko_laptop> aquileia, ohhh I'm well aware of that, and I'm eager to not have the whole coding thing rest on my shoulders.
20140301 17:30:31< mordante> aquileia, I prefer to simply use rand(), all other random map generators do the same
20140301 17:30:50-!- ancestral [~ancestral@74.174.236.84] has joined #wesnoth-dev
20140301 17:31:04< mordante> aquileia, how do you want to cheat if the map generator uses rand() ?
20140301 17:31:15< sachith500> mordante, is the gsoc questionnaire essential or can we just structure the same information how we like?
20140301 17:31:19< aquileia> But YAMG is different - the RNG should shape a map reproduceably
20140301 17:31:46< aquileia> with rand() there's no cheating issue
20140301 17:32:13< happygrue> sachith500: re keeping unit names and such straight, this is useful:  http://units.wesnoth.org/
20140301 17:33:15< sachith500> happygrue ok thanks :D
20140301 17:33:42< sachith500> btw can I structure the information in the questionnaire, how I like?
20140301 17:34:16< mordante> sachith500, depends on how you do it, IIRC some people did it in the past, but feel free to do it, if we don't like we'll let you know :-P
20140301 17:34:18< happygrue> I don't know actually, probably waiting for mordante or someone with more GSoC experience to answer is best.
20140301 17:34:33< sachith500> OK :D
20140301 17:34:41< sachith500> thanks
20140301 17:34:49< sachith500> happygrue it says animation stats?
20140301 17:35:48< mordante> aquileia, yeah but IMO only locally and not forever
20140301 17:35:50< sachith500> I'm trying to make sense of the numbers
20140301 17:36:13< happygrue> sachith500: click the first links, those are only useful to people looking at animations
20140301 17:36:15< mordante> we're not going to guarantee the generator will never be changed
20140301 17:36:22< happygrue> eg, here: http://units.wesnoth.org/trunk/mainline/en_US/mainline.html
20140301 17:36:26< mordante> Aishiko_laptop, around?
20140301 17:36:32< Aishiko_laptop> mordante, yes I am
20140301 17:36:41< mordante> happygrue, which question?
20140301 17:36:55< mordante> Aishiko_laptop, you have time to look at the code?
20140301 17:36:57< happygrue> you answered it already ;)  re: form of info on GSoC sheet
20140301 17:37:23< mordante> ah ok
20140301 17:37:44< aquileia> mordante: But then it would break with a commit message stating a reason, not because you have a new laptop
20140301 17:37:49< Aishiko_laptop> mordante, I have all the time you need I cloned the repo  recently so it should be the same as yours unless changes have been made in the last 24-48 hours
20140301 17:41:28-!- TCD [~TheCommie@152.78.235.20] has joined #wesnoth-dev
20140301 17:41:42< TCD> Hey.
20140301 17:42:06< mordante> aquileia, I expect it to break by accident, because somebody calls m_rand more or less times
20140301 17:42:10< mordante> hi TCD 
20140301 17:42:38-!- DCW [~Thunderbi@cpc66863-finc15-2-0-cust393.4-2.cable.virginmedia.com] has quit [Quit: DCW]
20140301 17:43:05< mordante> Aishiko_laptop, you assume I update my copy regularly :-P but the code is rather old and hasn't been modified for a long time
20140301 17:43:19< happygrue> sachith500: did you see what I mean?  It grabs the era data from the add-on server so you can see what eras are available and their unit tree in a nice format.  Click the unit for more info.  You can also do the same for campaigns
20140301 17:43:42< sachith500> yeah
20140301 17:43:54< sachith500> it has all the stuff :)
20140301 17:44:09< sachith500> wait where are the campaigns
20140301 17:44:23< sachith500> oh right
20140301 17:44:24< sachith500> found it
20140301 17:44:34< mordante> Aishiko_laptop, have a look at image.cpp
20140301 17:45:46< mordante> this code is the heart of the image loading code
20140301 17:46:30< sachith500> happygrue, are you thinking about adding the stats to the unit info page itself?
20140301 17:46:41< Aishiko_laptop> alright
20140301 17:47:14< happygrue> you mean from the data analysis project?  No, I hadn't really considered that.
20140301 17:47:27< sachith500> yeah
20140301 17:47:42< mordante> it also caches the images loaded, best first to read the comment in the header file
20140301 17:47:44< sachith500> just crossed my mind that it's possible :D
20140301 17:47:54< happygrue> sure, it could be at some point.
20140301 17:48:03< sachith500> although we should probably provide it in a more parsable format
20140301 17:48:25< happygrue> It's more important to me to get some way for UMC authors and MP folks to look at how factions are doing on a map by map basis or against other factions
20140301 17:48:30< happygrue> everything after that is bonus :D
20140301 17:48:48< sachith500> yeah :D
20140301 17:49:00< Aishiko_laptop> alright opening the header files as well
20140301 17:49:20< sachith500> also when the project says database
20140301 17:49:28< sachith500> were you thinking something like mysql?
20140301 17:49:38< happygrue> and yes sachith500, I think ideally a good proposal will show forward thinking regarding how this data collection and presentation could be expanded in the future.
20140301 17:50:02< sachith500> :D
20140301 17:50:06< ancestral> sachith500: This is a little outdated, but If it helps you in any way, feel free to loot/pillage/pilfer: http://mproud.com/wesnoth/bestiary/#spearman (links don't work, and currently, the only way to load a page is to put #unitname in the URL)
20140301 17:50:29< sachith500> ancestral, thanks I'll take a look :D
20140301 17:51:27< happygrue> I honestly have no opinion on that... since they are archived now it could be just parsing the replays somehow, or it would be better to transfer things to a more easily accessable database but what can be done is also somewhat based on if the student can make it work
20140301 17:52:02< sachith500> hmm well i suppose that can be towards the end
20140301 17:52:03< happygrue> I have no database experience, perhaps Soliton has a better idea of what would be ideal since he has a lot of experience with our servers
20140301 17:52:05< sachith500> optional maybe
20140301 17:52:14< sachith500> I know mysql pretty well
20140301 17:52:23< sachith500> I will ask him as well
20140301 17:52:24< sachith500> :)
20140301 17:53:19-!- aquileia [52d4193f@gateway/web/freenode/ip.82.212.25.63] has quit [Quit: Page closed]
20140301 17:53:42< happygrue> I do almost entirely just unit tweaking stuff, so I obviously have a Phase 1: start project, Phase 2: ??, Phase 3: Profit!  mentality torward this.  ;)
20140301 17:56:29< Aishiko_laptop> mordante, I've taken a brief over view, ( read the comments and got an idea of what I think its doing) but I'm not sure what the commit messages are, unless they are the /// prefixed comments
20140301 17:56:58< sachith500> hahaha ok :D
20140301 17:57:30< Aishiko_laptop> happygrue, would that be ??=find sucker to do the heavy lifting? =P
20140301 17:57:35< mordante> commit messages? they are used when committing code, reading the comments should be enough
20140301 17:57:55< mordante> are the comments clear?
20140301 17:58:10< sachith500> hahaha
20140301 17:58:15< happygrue> Aishiko_laptop: shhhh, you will give away all our secrets...
20140301 17:58:29< sachith500> too late! I already saw :P
20140301 17:58:51< Aishiko_laptop> mordante, yes, most of them are clear.
20140301 17:59:07-!- ancestral_ [~ancestral@74.174.236.84] has joined #wesnoth-dev
20140301 17:59:36-!- ancestral [~ancestral@74.174.236.84] has quit [Ping timeout: 240 seconds]
20140301 17:59:37-!- ancestral_ is now known as ancestral
20140301 18:01:53< mordante> the important class is the image locator, this loads a file and its modifications
20140301 18:03:12< Aishiko_laptop> mordante, I'll have to walk through each part one at a time to truely understand how its doing what its doing but the purpose and with that I can follow along like lines 111-121 in the images.hpp where it reads the image file then checks that it is in cache, followed by what I think is an access check (IE the program can access the cached image),  then applies a const on the image, followed finally by giving it to the cache
20140301 18:03:13< Aishiko_laptop>  the game uses for play.
20140301 18:03:23< mordante> the modifications are listed here http://wiki.wesnoth.org/ImagePathFunctionWML
20140301 18:04:51< mordante> the code can always access the cache, but it expires items based on their last access time
20140301 18:07:40< mordante> when the code wants an image it can get it in different flavours (enum TYPE)
20140301 18:09:51-!- ancestral_ [~ancestral@74.174.236.84] has joined #wesnoth-dev
20140301 18:10:23-!- sk94 [~sk@77-254-169-239.adsl.inetia.pl] has joined #wesnoth-dev
20140301 18:10:50< Aishiko_laptop> Okay, 
20140301 18:11:41< mordante> if you look at display.cpp it creates the image using different types
20140301 18:13:10-!- ancestral [~ancestral@74.174.236.84] has quit [Ping timeout: 252 seconds]
20140301 18:13:30-!- ancestral_ is now known as ancestral
20140301 18:15:22< Aishiko_laptop> and different overlays as well
20140301 18:15:38< mordante> Aishiko_laptop, do you understand why display.cpp:2660 »image::get_image("terrain/foreground.png", image_type));« works
20140301 18:16:02< mordante> just looking at the code and the wanted parameters for image::get_image
20140301 18:16:22< sachith500> well I'm off for now guys
20140301 18:16:28< mordante> bye
20140301 18:16:28 * sachith500 waves.
20140301 18:16:40-!- sachith500 [~kvirc@112.135.37.90] has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/]
20140301 18:17:33< Aishiko_laptop> mordante, I think so let me try to explain it, and you tell me if I got it right.
20140301 18:17:53-!- TCD [~TheCommie@152.78.235.20] has left #wesnoth-dev []
20140301 18:19:05< Aishiko_laptop> odd my copy of display.cpp has a different line of code at 2660
20140301 18:19:33< mordante> maybe my copy is outdated ;-)
20140301 18:21:04< Aishiko_laptop> my copy has that line at 2685 =) so not too far off but, it looks like its getting a particular file in this case forground.png and applying a type to it in the case of an error
20140301 18:21:31< Aishiko_laptop> no wait not an error 
20140301 18:21:43< mordante> what's the function signature of image::get_image?
20140301 18:24:35< Aishiko_laptop> surface get_image(const image::locator& i_locator, TYPE type) in image.cpp
20140301 18:25:09< mordante> what kind of parameters are we using in display.cpp?
20140301 18:25:26< Aishiko_laptop> it takes the ref of the locator and then the type value
20140301 18:26:22< Aishiko_laptop> one is an constant image (never changes) and the other is the a type datatype?  that seems a bit odd
20140301 18:28:43< Aishiko_laptop> it appears to return the cached location of the rendered image
20140301 18:29:18< Aishiko_laptop> am I at least close?
20140301 18:30:43< mordante> well it gets a const reference to an image locator, not a copy
20140301 18:31:28< mordante> but what kind of parameters do we use in display.cpp?
20140301 18:32:21< Aishiko_laptop> right, thats what I said but in two statements what I meant was the first var was a ref the second a value and the first var was const image.  I've got I'm just not being a clear as I should be
20140301 18:32:47< Aishiko_laptop> at what point?
20140301 18:32:58-!- stevedes [0e61725b@gateway/web/freenode/ip.14.97.114.91] has joined #wesnoth-dev
20140301 18:33:34< mordante> image::get_image("terrain/foreground.png", image_type));
20140301 18:33:56-!- stevedes [0e61725b@gateway/web/freenode/ip.14.97.114.91] has left #wesnoth-dev []
20140301 18:35:19< Aishiko_laptop> we sent an image locator (it looks like a path) and then a value to tell get image what type it is or what type type to apply to the image when it returns the value
20140301 18:36:26< mordante> well I see a const char* type being used, how is it turned into an image locator?
20140301 18:38:22< Aishiko_laptop> when get_image has taken the const image sent, copies into a new variable (res) and then applies the type to it, it places that in the cache and sends back the image locator
20140301 18:38:43< Aishiko_laptop> which points to the location in cache that has that image
20140301 18:39:21< Aishiko_laptop> no wait
20140301 18:42:39< Aishiko_laptop> it checks to see if the image_locator is void if its not then it calls  simply_type(image locator, type) and passes the simplified type to type (ges that could get confusing fast)
20140301 18:43:20-!- goblinThing [44bd8c2c@gateway/web/freenode/ip.68.189.140.44] has joined #wesnoth-dev
20140301 18:44:16< goblinThing> I'm writing my SoC proposal for the wesnoth website. Is it better to go for a proffessional or personal tone?
20140301 18:45:15< mordante> well I wonder more how can it be I call with a »const char*« and the called function wants a »const locator&« and still the compiler seems to be happy
20140301 18:45:28< Aishiko_laptop> it then goes through a switch statement to assign imap to a ref to some other  ohhh
20140301 18:46:24< AI0867> const& can bind a temporary, and image::locator has a const char* constructor?
20140301 18:46:34< AI0867> or at least an std::string one?
20140301 18:46:48< mordante> AI0867, please don't explain it
20140301 18:46:57< Aishiko_laptop> I was going to say I thought the string being given was a locator
20140301 18:47:02< AI0867> okay
20140301 18:47:14< mordante> goblinThing, I prefer personal
20140301 18:47:20 * AI0867 reads back a bit
20140301 18:47:29< AI0867> oops, sorry
20140301 18:47:38-!- sk94 [~sk@77-254-169-239.adsl.inetia.pl] has quit [Quit: Wychodzi]
20140301 18:47:43< Aishiko_laptop> AI0867, mordante is working with me to get the details of displaying the sprites
20140301 18:48:16< goblinThing> @mordante: Thanks. Good to know... Professional feels just a little dishonest, so I prefer personal, too.
20140301 18:48:53< Aishiko_laptop> I've honestly not had a similar situation I think, that during compile phase the image path given is converted to a locator value, maybe
20140301 18:49:21< Aishiko_laptop> in that way its more user friendly to us humans
20140301 18:49:53< goblinThing> Sorry if I interrupted anything. I waited a few seconds to see if people were doing dev stuff... I obviously should have checked the logs instead. I'll get out of your guys's hair now.
20140301 18:50:05< Aishiko_laptop> no no goblinThing 
20140301 18:50:09< mordante> goblinThing, not a problem
20140301 18:50:33< mordante> goblinThing, at times there can be three of four discussions at the same time ;-)
20140301 18:50:41< Aishiko_laptop> we can have several converstations going at once, its just that mordante and I were having one and everyone else was sort of not talking about anything
20140301 18:51:06< mordante> Aishiko_laptop, whether it's friendly to humans remains to be seen, it confuses several of them :-P
20140301 18:51:12< goblinThing> 0_0     ... how do you focus on any one?
20140301 18:51:25< mordante> task-switching ;-)
20140301 18:51:41< Aishiko_laptop> goblinThing, we might put the other users name at the front of the line so its highlighted in someway on their screen
20140301 18:52:58< Aishiko_laptop> mordante, I don't have to know exactly how it works if I'm just using it, just like the end user of the game that doesn't work on it doesn't really care how it displays a sprite or connects to the server as long as it does and they know how to make it do that
20140301 18:54:13< Aishiko_laptop> all I need to know is that if I pass a file path in and it converts it to a locator, I'm happy with that, now if I need to modify how it does that, then I start to care how it goes about doing that, and I think I'm going to have to start caring =)
20140301 18:54:34< mordante> Aishiko_laptop, well it's part of the C++ language so I wondered whether you know why it works
20140301 18:55:30< Aishiko_laptop> mordante, no sorry I don't I know it does work but why, no.
20140301 18:56:26< mordante> well AI0867 already asked the right question
20140301 18:56:56< mordante> the constructor of locator can convert a const char* to a locator
20140301 18:57:31< mordante> since the constructor is not marked as explicit the conversion can be done implicitly
20140301 18:57:55< Aishiko_laptop> ohh that makes sense
20140301 18:58:03< mordante> whether this is a good thing or not is another question
20140301 18:58:18< mordante> implicit conversions can be nasty and buggy
20140301 18:58:26 * mordante waves at t_string
20140301 18:58:38< Aishiko_laptop> I would think both ways (explicit and implicit) have their own up and downsides
20140301 18:58:56 * AI0867 marks t_string::t_string(std::string) as explicit and breaks compilation for every other source file
20140301 18:59:08< mordante> :-)
20140301 18:59:17< iceiceice> quick question: is there a 1.11.11 release happening today or am i confused about that?
20140301 18:59:26< AI0867> somewhere today was the plan
20140301 18:59:27< mordante> iceiceice, ask Ivanovic 
20140301 18:59:33< AI0867> it is ready when ivanovic is ready
20140301 18:59:52< iceiceice> ok thanks
20140301 19:00:12-!- anonymissimus [~chatzilla@HSI-KBW-149-172-228-192.hsi13.kabel-badenwuerttemberg.de] has joined #wesnoth-dev
20140301 19:00:16< Aishiko_laptop> AI0867, most I think because in the past it was implicit and so no one expects to have to be explicitly done
20140301 19:00:20< mordante> AI0867, it's more that the explicit std::string operator is nasty
20140301 19:00:46< Aishiko_laptop> most=mostly
20140301 19:02:14< AI0867> you mean implicit?
20140301 19:02:29< mordante> yes
20140301 19:02:47< anonymissimus> mordante: does smart_list.hpp perhaps not compile with clang for you ? (MSVC does not)
20140301 19:03:03< mordante> both explicit and implicit have their advantages, but I just wondered whether you knew about it
20140301 19:03:23< anonymissimus> (so I could lift some work off of me)
20140301 19:03:24< mordante> anonymissimus, no idea, I assume it does since travis isn't unhappy
20140301 19:03:34< mordante> anonymissimus, can you pastebin your errors?
20140301 19:04:31< Aishiko_laptop> no, I'm learning a lot though and I think that is a good thing
20140301 19:05:06-!- ancestral_ [~ancestral@74.174.236.84] has joined #wesnoth-dev
20140301 19:05:20< AI0867> anonymissimus: I've looked into the errors you pastebinned
20140301 19:05:38< mordante> yes it is, GSoC is about learning to work at a real-world project and working with experienced developers
20140301 19:06:12< AI0867> #3 and #4 are just results of MSVC not parsing the function declaration properly
20140301 19:06:26< Aishiko_laptop> and then being the experienced developer helping developing developers
20140301 19:06:29< AI0867> I can't figure out what's going wrong with #2
20140301 19:06:33< Aishiko_laptop> well hopefully that is
20140301 19:06:39-!- Crab_ [~Crab_@wesnoth/developer/crab] has left #wesnoth-dev []
20140301 19:06:48-!- ancestral [~ancestral@74.174.236.84] has quit [Ping timeout: 240 seconds]
20140301 19:06:49-!- ancestral_ is now known as ancestral
20140301 19:07:20< mordante> Aishiko_laptop, well we have several mentors who started as student, so yes it's possible
20140301 19:07:21< anonymissimus> AI0867: i suspect id jokes somehow about the double templates
20140301 19:07:42< AI0867> and #1 means it doesn't recognise smart_list<Data>::iterator_base<Value, Reversed> as a valid templated name
20140301 19:08:02< AI0867> yeah, I'd suggest shotgun debugging stuff like that
20140301 19:08:13< AI0867> can't really help there as I don't have the compiler myself
20140301 19:08:15< mordante> anonymissimus, where is the pastebin?
20140301 19:08:20< AI0867> http://pastebin.com/zyzYTXwE
20140301 19:08:24< mordante> thanks
20140301 19:08:52< mordante> Aishiko_laptop, do you have more questions regarding the code?
20140301 19:11:39< Aishiko_laptop> I think I have it, in the image.cpp will have to be changed to accept the sprite sheets and then pass off locators to the individual sprite images instead of loading all the images and passing locators for each.  is that about right?
20140301 19:12:22< Aishiko_laptop> I'm going to spend some time reading the image and display files to get a better handle on how they are doing everything
20140301 19:15:10 * Aishiko_laptop is trying to relate whats been learned to what the task is
20140301 19:15:35< mordante> Aishiko_laptop, well I think it needs something like »image::get_image("terrain/foreground.png[72,72,4]", image_type));« to get the 4th image of the sprite sheet with 72x72 images
20140301 19:16:12< mordante> this syntax is just what I think up now, not per se what we need to use
20140301 19:16:53< Aishiko_laptop> ahh then get_image gets the image out of the sheet and then goes through the steps to prepare it and return it for display
20140301 19:17:59-!- ancestral [~ancestral@74.174.236.84] has quit [Quit: i go nstuf kthxbai]
20140301 19:19:00< mordante> yes and the code using the image doesn't know whether it came from a single image or a sprite-sheet
20140301 19:19:29< mordante> that was the part earlier today, where I said higher levels of the code do not know or care about sprite-sheets
20140301 19:19:48< mordante> they just send a string and get an image
20140301 19:20:09< mordante> just like the modifications like crop
20140301 19:20:43< mordante> but the lowlevel code needs to know about it since once the sheet is loaded it doesn't need to reload that sheet
20140301 19:21:07< Aishiko_laptop> now i'm getting it, though if the higher levels keep the same names like attack_1 then there has to be some way to relate each name to a frame on the sheet when each sprite has different numbers and some have different names like elves don't have fly_1
20140301 19:22:08< mordante> well the names of the files are normally in WML, let me get an example
20140301 19:23:01< Ivanovic> iceiceice: not today
20140301 19:23:01< mordante> for example data/core/units/bats/Bat_Blood.cfg
20140301 19:23:04< Aishiko_laptop> mordante, right, thats what I mean a way to relate the file names to frames
20140301 19:23:14< Aishiko_laptop> ohhh there!
20140301 19:23:16-!- Ivanovic changed the topic of #wesnoth-dev to: release of 1.11.11 maybe on Sunday | string+feature freeze active on master | 213 bugs, 350 feature requests, 28 patches | Logs: http://irclogs.wesnoth.org | Alternate logs: http://wesnoth.debian.net | Don't paste on IRC! Use a pastebin: http://pastebin.com | http://imagebin.org
20140301 19:23:30< iceiceice> Ivanovic: thanks for letting me know
20140301 19:23:45< mordante> Aishiko_laptop, line 49 »image="units/undead/bloodbat-se-[3,4].png:[1,100]"«
20140301 19:24:00-!- serados [Junwei@nusnet-196-209.dynip.nus.edu.sg] has quit [Read error: Connection reset by peer]
20140301 19:24:22< Aishiko_laptop> right
20140301 19:24:27< mordante> if the image was in a sprite-sheet it would become  »image="units/undead/bloodbat-se-[3,4].png[72,72,4]:[1,100]"«
20140301 19:25:25< Ivanovic> shadowm: did you add the typofix to pofix.py and apply that one to all po files?
20140301 19:26:30< mordante> (obviously the syntax needs to change a bit)
20140301 19:27:31< Aishiko_laptop> I think it would be more like   »image="units/undead/bloodbat[start,end].png[72,72,end]:[1,100]"«  where start is the current frame and end is the next frame
20140301 19:28:09< mordante> well that's something to work on in your proposal ;-)
20140301 19:28:37< mordante> do you have a better picture of what the job entails now?
20140301 19:28:39< Aishiko_laptop> because all the bloodbat images will be in one file likely called bloodbat.png (or something like that) and it would have to reference frames and not the old files
20140301 19:28:54< mordante> yup
20140301 19:29:26< mordante> but obviously we need to be able to both use single images and sprite-sheets it will take a while before all images are converted
20140301 19:29:33< Aishiko_laptop> I think I do, and I have a much better point to revise my proposal and I hope to do so today, before I forget everything l learned.
20140301 19:29:39-!- justinzane [~justinzan@tiny.justinzane.com] has quit []
20140301 19:29:54< mordante> (it would be great if there is enough time to create a tool for it, or maybe Frogatto has it)
20140301 19:29:57< anonymissimus> mordante: http://pastebin.com/J8ekAxJL
20140301 19:30:07< anonymissimus> I'm almost not here so to say
20140301 19:30:35< mordante> anonymissimus, can you give this patch a spin http://paste.debian.net/84807/
20140301 19:31:23< mordante> Aishiko_laptop, good, if you have more questions just ask
20140301 19:31:37< Aishiko_laptop> mordante, I was actually thinking that converting all the images would be part of my project.  I know how to do it, I could easily automate all the currently created images into spritesheets in a few days (I want to say an afternoon but want to give myself some breathing room)  infact I already planed on having a conditional to check to see if the old images are needed or the new spritesheets
20140301 19:31:53< Aishiko_laptop> thank you mordante I will of course let you know
20140301 19:33:03< mordante> I see converting existing images as a secondary task, first to get the main task done
20140301 19:33:35< mordante> (my experience is that a lot of projects turn out to be larger than expected)
20140301 19:33:53< Aishiko_laptop> I'll have to convert at least one to unit for testing purposes
20140301 19:33:57< mordante> you're welcome
20140301 19:34:18< mordante> you can make a new dummy unit with only a few sprites to test
20140301 19:34:39< mordante> we have a test-scenario, start wesnoth -t to have a look at it
20140301 19:35:08< Aishiko_laptop> at the command prompt type "wesnoth -t" right?
20140301 19:35:10< mordante> that unit could also be an item, which would probably be even easier
20140301 19:35:12< mordante> yes
20140301 19:35:38< Aishiko_laptop> thank you I'm writing that down now
20140301 19:37:22< mordante> have a look at data/scenario-test.cfg starting at line 2372, there you can see items
20140301 19:37:53< mordante> probably the easiest place to test with sprite-sheets you just add new items, or replace existing items
20140301 19:38:02-!- justinzane [~justinzan@tiny.justinzane.com] has joined #wesnoth-dev
20140301 19:38:10< Aishiko_laptop> I'll do that.
20140301 19:38:34-!- Matei [~razvan@188.26.63.174] has joined #wesnoth-dev
20140301 19:41:46-!- jamit [~jamit@wesnoth/developer/jamit] has joined #wesnoth-dev
20140301 19:42:09< Aishiko_laptop> I'll do that, maybe even do a custom map that makes each character go through their sequences
20140301 19:43:20< mordante> there are also units in that scenario and you're free to modify it
20140301 19:43:50< mordante> and once the code is tested with some units I've no doubt it will work with all of them
20140301 19:44:15< Aishiko_laptop> ohh I'm sure of that too!
20140301 19:44:35< mordante> the easiest to test would be with units with constant animations, flying units that hover for example
20140301 19:44:37< Aishiko_laptop> and just slightly more images then I thought
20140301 19:45:07< Aishiko_laptop> or that scepter from heir to the throne that is always sparkling
20140301 19:47:44< jamit> anonymissimus: Sorry. I'll take a look at your MSVC errors later today. I can get a quick fix in for a different issue now, but then I have to be off.
20140301 19:48:33-!- irker924 [~irker@ai0867.net] has joined #wesnoth-dev
20140301 19:48:33< irker924> wesnoth: JaMiT wesnoth:master 2fb13f89daa2 / src/map_label.cpp: Make all labels visible in the editor. http://git.io/tBS_4Q
20140301 19:52:19< Aishiko_laptop> OHHHH I nearly forgot a very important question!!!  is there a file that gives the coding conventions that wesnoth devs need to use?
20140301 19:52:26< iceiceice> yes
20140301 19:52:41< iceiceice> well there may be several but you can look at "developers guide" i think on the wiki
20140301 19:52:51< iceiceice> or perhaps developers resources
20140301 19:53:41< iceiceice> anyways poke around on the wiki ;)
20140301 19:53:43< Aishiko_laptop> iceiceice, I have a feeling I that odds are I saw it when "Ohh I need to read that!"  got distracted and forgot I saw it
20140301 19:54:04< mordante> wiki is slow, will post the link in a sec
20140301 19:54:33< Aishiko_laptop> I saw it and got it http://wiki.wesnoth.org/CodingStandards
20140301 19:54:50< mordante> yes that's the one
20140301 19:56:22< iceiceice> question: after poking around on wiki myself ^^ i found this under "Future:" 
20140301 19:56:23< iceiceice> http://wiki.wesnoth.org/Barbarians
20140301 19:56:40< iceiceice> is this still planned for the future?
20140301 19:57:20< iceiceice> quoting: "What I propose is three unit lines, a sort of barbarian light infantry light cavalry, and light archers that would be to the Loyalists as Huns or Vikings were to the high-medieval plate-armored fighters the Wesnothians are modeled on. "
20140301 19:57:35< zookeeper> at best it could be someone's pipedream
20140301 19:57:35< iceiceice> tbh it sounds like perhaps some of the khalifate units have filled this niche?
20140301 19:58:54< zookeeper> but no, no one's voiced any plans to actually work on that.
20140301 20:00:20-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20140301 20:00:40< happygrue> such things making their way into a multiplayer era has no plan, and given how hard it can be to mess with default I would put it in the pipe dream status as well.  But they could always end up in a campaign if the art and stats turn up. 
20140301 20:00:56< mordante> anonymissimus, any luck wit the patch?
20140301 20:01:28< Aishiko_laptop> I volunteer my brother...  he needs to do something besides sleep at his desk and function as a "Life-support system for a computer" all day
20140301 20:01:28< happygrue> Given the struggle with the khalifate and the length of time it's been in development, I would not expect to ever consider another MP faction until such time as it is ALREADY developed in a campaign
20140301 20:03:21< iceiceice> :)
20140301 20:06:32< shadowm> mordante: IIRC the image cache keeps more than just team colored images, it also keeps cropped images (which is why CROP is only good as a proof of concept or for non-spritesheet tasks), blurred images, and everything else that deals with image path functions.
20140301 20:07:10-!- Matei [~razvan@188.26.63.174] has quit [Read error: Operation timed out]
20140301 20:07:17< irker924> wesnoth: Chris Beck wesnoth:master ba6d58c2f1ca / src/ (playmp_controller.cpp playmp_controller.hpp): Revert "apply a blindfold during skip replays" http://git.io/BFHv7Q
20140301 20:07:36< shadowm> Aishiko_laptop: ^ too I guess.
20140301 20:08:13< Aishiko_laptop> shadowm, thank you thats good to know
20140301 20:08:25< mordante> shadowm, yes it also keeps all modified images
20140301 20:08:38< shadowm> aquileia: I wouldn't hate you for making a few mistakes. o O
20140301 20:08:57< anonymissimus> mordante: I'll check it out but probably not while you're still here, read logs
20140301 20:09:07< shadowm> My first patch to mainline had considerably worse mistakes and had way too many iterations.
20140301 20:10:35< mordante> anonymissimus, ok, and since I'm about to leave you're right about me not being here ;-)
20140301 20:10:36< Aishiko_laptop> shadowm, I wanted to ask you, if I wanted to add an "update" button to the addons list dialog where you press the button and it updates all your currently installed addons where an update is available, would it be hard to have the upgradeable status message add those addons to a list that is used to upgrade them?
20140301 20:10:45< Espreon> (Mine was probably far worse)
20140301 20:11:03< shadowm_desktop> http://wiki.wesnoth.org/ImagePathFunctionWML  <- for those who haven't heard of the term "image path functions" before, this is the reference page.
20140301 20:13:21< mordante> I'm off bye
20140301 20:13:39< Aishiko_laptop> bye and thank you again mordante 
20140301 20:13:58< shadowm> Ivanovic: It's not needed, the fixed string was already in the textdomain.
20140301 20:13:58< mordante> Aishiko_laptop, no problem and you're welcome
20140301 20:14:06< Ivanovic> shadowm: okay
20140301 20:14:21-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20140301 20:14:40 * shadowm is all caught up with the logs!
20140301 20:15:21< shadowm> Aishiko_laptop: Hm, I'm not sure I understand the second part of the question.
20140301 20:16:36-!- trademark_ [~trademark@nsg93-8-88-175-59-164.fbx.proxad.net] has quit [Quit: Leaving]
20140301 20:17:20< shadowm> Right now most of the grunt work in the add-ons manager is done by a big ugly monolithic function I haven't figured out how to refactor nicely yet.
20140301 20:18:19< shadowm> <unnamed>::show_addons_manager_dialog() in src/addon/manager_ui.cpp.
20140301 20:18:57< shadowm> This ugly design allows it to have unimpeded access to all the things it needs at a given time, which is both an advantage and a disadvantage (readability).
20140301 20:19:36< happygrue> also: there already is an update button on the connect to add-on server dialog, but I don't understand the second part of the question either.
20140301 20:19:41< shadowm> If you choose Options in the dialog and select to view add-ons with the Upgradable installation status only, an Update All button appears.
20140301 20:19:46-!- trademark_ [~trademark@nsg93-8-88-175-59-164.fbx.proxad.net] has joined #wesnoth-dev
20140301 20:19:47< Aishiko_laptop> oh, I saw that its supposed to colour code the addons based in if its installed, newer version on server, etc and I figured that if it could detect that stuff already, making a list (array) of the ones that can be upgraded would be possible with a little extra work leveraging that function that is already being implemented
20140301 20:19:52-!- trademark_ [~trademark@nsg93-8-88-175-59-164.fbx.proxad.net] has quit [Client Quit]
20140301 20:20:06-!- trademark_ [~trademark@nsg93-8-88-175-59-164.fbx.proxad.net] has joined #wesnoth-dev
20140301 20:20:13< Aishiko_laptop> ohhh
20140301 20:20:40< Aishiko_laptop> I missed that
20140301 20:21:16< happygrue> One thing that I've wished for a while is that the buttons said "download" and "quit" or something rather than "OK" and "cancel", but I guess it's fine.
20140301 20:21:18< iceiceice> i'm also out, bb
20140301 20:21:20< shadowm> Arguably, that button should be there for all views sans Not Installed. I think I restricted it to Upgradable only because before converting the dialog to use this filtering approach, there were two separate dialogs for displaying all add-ons, and only add-ons that can be upgraded.
20140301 20:21:21< happygrue> o/
20140301 20:21:40-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has quit [Quit: Leaving]
20140301 20:21:53< shadowm> happygrue: Yeah, someone made it Download and Close once, I reverted that during the 1.11.x refactor.
20140301 20:22:01< happygrue> ah
20140301 20:22:13< happygrue> I have memory of filing a feature request and then thinking it was fixed
20140301 20:22:19< shadowm> Why? Well, "Downloading" a Publish entry doesn't really make sense.
20140301 20:22:32< shadowm> Neither does "Downloading" a Delete (from server) entry.
20140301 20:22:48< happygrue> maybe another word would be better, but I can't imagine "OK" and "cancel" are ideal words either?
20140301 20:23:03< shadowm> I had vague plans to move Publish/Delete options to a separate dialog, but that never really happened in time for 1.12.
20140301 20:23:07< happygrue> install then?
20140301 20:23:40< happygrue> but I see, it's because you have to have the same button right now for both uploading and downloading
20140301 20:24:05< happygrue> well, that makes some case for "OK" I guess.
20140301 20:24:37< happygrue> I think exit or quit would be better than cancel still, but maybe that's just me
20140301 20:24:53< shadowm> It may be possible to detect the type of the selected entry and change the OK label accordingly each time, but it'd require a rather intrusive patch since gui::dialog doesn't expose its default buttons.
20140301 20:25:33< shadowm> "Quit" sounds too global in my opinion.
20140301 20:26:51< Aishiko_laptop> I have to aggre Quit does sound too global too
20140301 20:27:48< happygrue> "Back" or "Return" would also work, but since we use cancel so many places there is some reason to keep it.  I just don't like it in that context.
20140301 20:28:10< shadowm> Aishiko_laptop: Anyway, as I was saying, Update All is restricted to a specific view mostly for legacy reasons/laziness. If you wanted to make it appear in all views where at least one upgradable add-on exists, that would not be too hard, but it would require some reengineering such as a separate vector of upgradable add-ons ids.
20140301 20:28:32< shadowm> Right now it looks like:    if(update_everything) { ids_to_install = option_ids; }
20140301 20:28:57-!- cib0 [~cib@p20030067CE0E5A01267703FFFEE75B84.dip0.t-ipconnect.de] has joined #wesnoth-dev
20140301 20:29:01< shadowm> Which for the regular view would result in downloading all add-ons at once regardless of upgradability (definitely not a good idea).
20140301 20:29:37< shadowm> But it definitely wouldn't require changing anything beyond the scope of the aforementioned function, no.
20140301 20:30:05-!- matrixMinion [67157f4c@gateway/web/freenode/ip.103.21.127.76] has joined #wesnoth-dev
20140301 20:30:09-!- kex [~kex@89.205.75.19] has quit [Remote host closed the connection]
20140301 20:30:46< shadowm> The fact that the conditional looks like that attests to my laziness.
20140301 20:30:47< Aishiko_laptop> I see
20140301 20:31:54< Aishiko_laptop> in theory I'd have a for_each loop go through an array of option_ids that are eligible and ignore the rest.
20140301 20:33:31< shadowm> That is an option if you are not too crazy about optimization, otherwise the previous loop at line 605 is where add-on states are determined (e.g. ADDON_INSTALLED_UPGRADABLE) first for attaching those fancy indicators to them.
20140301 20:34:43< shadowm> Of course, having two loops instead of one would only be a concern with a large numbers of add-ons, I guess -- somewhere around 9000, maybe.
20140301 20:45:17< shadowm> jamit: Save a map as a "scenario" (see the options in the editor menu), then you will get all those tools enabled.
20140301 20:45:40< Aishiko_laptop> perhaps create the list when its assigning those fancy ids and then iterate through that list if you press "upgrade all"
20140301 20:46:13< Aishiko_laptop> ids/indicators
20140301 20:47:17< Aishiko_laptop> but there really isn't a need for it
20140301 20:47:53< Aishiko_laptop> though perhaps you can tell me what that "campaign/campaigns sillness" is
20140301 20:48:37< shadowm> The (legacy, production) add-ons server is called campaignd, the WML protocol it uses involves a list of [campaign] children in a [campaigns] node, you send requests about campaigns to it, and so on.
20140301 20:48:38-!- ALourenco [0252303e@gateway/web/freenode/ip.2.82.48.62] has joined #wesnoth-dev
20140301 20:49:20< ALourenco> Hello. I have already been here because I'm very interested in the spritesheet project sugested for GSoC 2014
20140301 20:49:22< shadowm> I'm not sure if campaignd was written before Wesnoth had more add-on types than just campaigns (e.g. before multiplayer support was first implemented), or whoever did it wasn't good at names.
20140301 20:49:44< Aishiko_laptop> sounds like it would be a pain to change it at this point but might be necessary at some point for readability
20140301 20:50:23< ALourenco> I already know how it can be done. 
20140301 20:50:46< shadowm> Aishiko_laptop: I refactored all client-side code that has to deal with that kind of thing into the addons_client class precisely so hide this obtuse implementation detail from view.
20140301 20:50:51< ALourenco> I also played the game and I would like to give some sugestions
20140301 20:51:40< shadowm> I considered changing the protocol at the same time, but that would break backwards compatibility with previous add-ons server instances (I needed to connect to 1.10 at the time to test my code) or clients.
20140301 20:52:15< shadowm> ALourenco: What kind of suggestions?
20140301 20:53:38< ALourenco> I believe they are not Great features, just somethings I noticed as a gamer and I think it would be good changes
20140301 20:53:58< ALourenco> first of all, It would be good to skip or speed up the AI's moves
20140301 20:54:20< ALourenco> it gets kind of boring for waiting for them to move and attack
20140301 20:54:46< shadowm> There is an option to do so, Skip AI Moves in Preferences -> General, as well as options to change the animation speed for everyone.
20140301 20:54:56< ALourenco> ok thats good
20140301 20:55:12< ALourenco> also, some sound from the environment in the background
20140301 20:55:15< ALourenco> like birds winds
20140301 20:55:20< ALourenco> owls at night 
20140301 20:55:29< ALourenco> would be good
20140301 20:56:20< shadowm> Hm, that's currently an opt-in thing that scenario coders must do themselves. Some scenarios like the test scenario or one of the Legend of Wesmere scenarios use it.
20140301 20:57:18< ALourenco> ok ;) 
20140301 20:58:14< ALourenco> I'll see if I haver some ideas and get some people playing the game to know what they think. 
20140301 20:58:30< shadowm> Having done it in my campaign as well, the hardest part is probably finding suitable sounds (mainline Wesnoth only ships with a few looping bird sounds) and figuring out the "sound source" parameters (radius, visibility, etc.).
20140301 21:01:26< ALourenco> I still haven't seen the code, but maybe an improved sound system would help. I never did it before, but if you guys are interested in I can search a bit and propose something
20140301 21:03:32< ALourenco> Altough, I'm very interested in the SpriteSheet. I have somthing done, not with SDL but with SFML/OpenGL. It's not that different
20140301 21:08:36-!- jamit [~jamit@wesnoth/developer/jamit] has quit [Read error: Connection timed out]
20140301 21:08:54< ALourenco> To submit an application in the wiki do I have to wait until 10th March?
20140301 21:13:59< shadowm> I am not sure, but it seems people have started preparing their proposals already.
20140301 21:15:38< happygrue> Since there are a number of options for projects, having a strong proposal already attached to a project might help to push others into applying for other ideas...  but yes you don't have to have it completed just yet.
20140301 21:22:43< Aishiko_laptop> happygrue, I'd not thought of that, that having an application out there might be discouraging to others.  But having said that I'm hoping I'll make a strong proposal and be able to be selected.
20140301 21:25:37< happygrue> it's just one factor, obviously, but if you see a great proposal for one idea but not another and you are trying to choose which basket to put your eggs in... ;)
20140301 21:26:45< ALourenco> I already have registered in the forums, is there any admin here that could give me GSoC membership?
20140301 21:27:46< Aishiko_laptop> ohh I need to do that too, perhaps that is why my proposal is not being detected and added to the proposal page
20140301 21:28:12-!- jamit [~jamit@wesnoth/developer/jamit] has joined #wesnoth-dev
20140301 21:28:55< shadowm> I'm a forum admin, I need your account names.
20140301 21:29:19< Aishiko_laptop> Aishiko
20140301 21:29:41 * Aishiko_laptop knows she's quite imaginative using the same name here as on the forums
20140301 21:30:23< shadowm> ALourenco: Account name?
20140301 21:30:31< ALourenco> ALourenco
20140301 21:30:37< Aishiko_laptop> thank you shadowm 
20140301 21:30:38< shadowm> Aishiko_laptop: Yeah, sure, it's mostly to make sure I don't get the wrong person.
20140301 21:31:00< shadowm> ALourenco, Aishiko_laptop : Done.
20140301 21:31:08< ALourenco> thank you :D
20140301 21:31:21< shadowm> Well, not done for Aishiko_laptop, hm.
20140301 21:31:24< Aishiko_laptop> lol I figured =)  don't want so and so getting that done to them when you meant someone else's account
20140301 21:31:40< shadowm> Why am I getting a "user does not exist" error?
20140301 21:31:41< Aishiko_laptop> ohhh I broke something=P
20140301 21:32:09< shadowm> Ah, it's Ashiko, not Aishiko?
20140301 21:32:51< Aishiko_laptop> ohh drat I mispelled it grr
20140301 21:33:00< Aishiko_laptop> I need to fix that
20140301 21:33:13< shadowm> Do you want me to rename your account as well?
20140301 21:33:22< Aishiko_laptop> if you can and it'd not be too much trouble
20140301 21:33:44< shadowm> Aishiko_laptop: Done.
20140301 21:34:27< Aishiko_laptop> thank you.  Now I just need to sit down and try my hand at fixing something =)  I need the practice
20140301 21:35:09< shadowm> zookeeper: Why do we need two Great River ford scenarios in mainline? :x
20140301 21:35:34< shadowm> It's so infuriating! Owaec is even slower than Gweddry on water.
20140301 21:36:57< zookeeper> shadowm, i'm sure there's a very good reason for it that we just don't see.
20140301 21:37:06< shadowm> Oh well, I got a loyal Ogre.
20140301 21:37:17< zookeeper> ohh. then you missed the ogre scenario.
20140301 21:37:27< shadowm> Uh... maybe?
20140301 21:37:43-!- ALourenco [0252303e@gateway/web/freenode/ip.2.82.48.62] has quit [Ping timeout: 245 seconds]
20140301 21:37:46< shadowm> I killed the orcs first in Two Paths.
20140301 21:38:01< shadowm> Then crossed the Great River because why the hell not.
20140301 21:38:23< shadowm> And now I'm in Xenophobia.
20140301 21:38:25< zookeeper> nah, the only way to skip the ogre scenario is to get the loyal one
20140301 21:38:41< shadowm> The farmland bits here are ... on grass?
20140301 21:38:48< shadowm> How does that make sense?
20140301 21:39:09-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140301 21:39:16< Aishiko_laptop> the feilds are laying fallow?
20140301 21:39:30< shadowm> zookeeper: Should I go back and debug-kill the Ogre, or I'm not missing out on much?
20140301 21:40:40-!- matrixMinion is now known as recycler
20140301 21:40:54-!- trewe [~trewe@124.164.54.77.rev.vodafone.pt] has joined #wesnoth-dev
20140301 21:41:13< zookeeper> shadowm, well i wouldn't mind more playtesting on it, but i'm 99% certain it can't be worse than the old version so up to you.
20140301 21:42:01-!- kex [~kex@89.205.75.19] has quit [Read error: Operation timed out]
20140301 21:43:11< Aishiko_laptop> I'd actually like it if each time one of the healing units cast a heal on a unit it got an experience point.
20140301 21:45:11-!- mattsc [~mattsc@154.20.32.245] has joined #wesnoth-dev
20140301 21:47:58< zookeeper> Aishiko_laptop, then that'd very much incentivize allowing as many of your units to get damaged so healers could harm XP off of them
20140301 21:49:34< zookeeper> eh
20140301 21:49:37< zookeeper> s/harm/farm
20140301 21:49:50< Aishiko_laptop> I'd not thought of that, but then I don't think that way.  Perhaps allowing it but limiting it to 1or 2 points per turn
20140301 21:50:34< Aishiko_laptop> the units usually get a point for each successful attack right?
20140301 21:51:20< zookeeper> units get 1xp per level of the enemy unit from combat, yes
20140301 21:52:20-!- eldruz [~eldruz@lav35-1-82-236-137-179.fbx.proxad.net] has joined #wesnoth-dev
20140301 21:52:25-!- eldruz [~eldruz@lav35-1-82-236-137-179.fbx.proxad.net] has quit [Client Quit]
20140301 21:53:14< Aishiko_laptop> perhaps the healers could use the same method only they get 1xp per level but only from the highest level unit they heal that round
20140301 21:53:47< goblinThing> Posted my SoC page. I hope I did it right.
20140301 21:53:59< Aishiko_laptop> so they heal 5 lvl 1 characters and a lvl 2 character they only get 2 points even though they cast 6 heals.
20140301 21:54:37< Aishiko_laptop> its merely an idea
20140301 21:54:54-!- ancestral [~ancestral@74.174.236.84] has joined #wesnoth-dev
20140301 21:56:31< mattsc> anonymissimus: I'm trying to work my way through the logs so please excuse if I missed something: Xcode uses clang (well, Apple llvm) and I have had no problems compiling lately.
20140301 21:59:17-!- ancestral [~ancestral@74.174.236.84] has quit [Ping timeout: 252 seconds]
20140301 22:00:32< Aishiko_laptop> goblinThing, which project are you applying for?
20140301 22:02:56< goblinThing> Eh... I don't know C++ yet, so I created my own idea. I might submit several other ideas as they come to me. I REALLY want this, so if anyone ends up proposing stuff I can do, I'll probably jump on it. (Anyone: Anyone but other SoC applicants. It would be stupid to steal other people's projects.)
20140301 22:04:52-!- ancestral [~ancestral@74.174.236.84] has joined #wesnoth-dev
20140301 22:05:44< Aishiko_laptop> or take someone's research and bundle it into your own application.
20140301 22:05:54< goblinThing> ?
20140301 22:08:37< Aishiko_laptop> well lets say we have a student, not a great coder, yet, they go through and come up with a detailed todo list, goals and puts all the steps in order, and then a great coder comes and goes, "ohh all the research is done", lifts it, changes it around and then submits it.  I mean who do you with the great coder with the great proposal or the not great coder with the great proposal?
20140301 22:08:45< shadowm> zookeeper: The map is certainly far better than I remember it.
20140301 22:08:53< Aishiko_laptop> I've actually seen stuff like that happen in the real world
20140301 22:09:45< Aishiko_laptop> though thankfully here they have a way of tracking who's done the research and is actually involved
20140301 22:10:00< goblinThing> Yeah, I would not do that.
20140301 22:10:33< goblinThing> I have to leave. Catch you guys later.
20140301 22:10:39-!- goblinThing [44bd8c2c@gateway/web/freenode/ip.68.189.140.44] has quit [Quit: Page closed]
20140301 22:11:37-!- thunderstruck [~zaibotren@cpc13-sgyl31-2-0-cust696.18-2.cable.virginm.net] has quit [Ping timeout: 264 seconds]
20140301 22:12:58-!- ancestral_ [~ancestral@74.174.236.84] has joined #wesnoth-dev
20140301 22:14:01-!- ancestral [~ancestral@74.174.236.84] has quit [Ping timeout: 240 seconds]
20140301 22:14:02-!- ancestral_ is now known as ancestral
20140301 22:17:34< shadowm> Well, three non-loyal ogres vs. 1 loyal ogre.
20140301 22:18:05< Aishiko_laptop> ohh sweet loyal ogre, we hardly knew ye!
20140301 22:18:13< shadowm> This is a campaign where every piece of gold counts, so...
20140301 22:18:25-!- thunderstruck [~zaibotren@cpc13-sgyl31-2-0-cust696.18-2.cable.virginm.net] has joined #wesnoth-dev
20140301 22:18:39< Aishiko_laptop> protect the ogre?
20140301 22:19:21< shadowm> Nah, it's two different paths. In one I get a variable amount of non-loyal ogres if I manage to "capture" them using ZoC-pinning. In the other path I only get one loyal ogre.
20140301 22:19:33< shadowm> But guess what, I actually care about my income, so I'll go back to the path with the loyal ogre.
20140301 22:20:02< shadowm> zookeeper: It was fun to some extent, I guess. It doesn't seem doable without spamming cavalrymen on the first turn.
20140301 22:20:22< mattsc> shadowm: life just isn't fair...  On a related note, any comments on the AI in Capturing the Ogres?
20140301 22:20:50< mattsc> As in, suggestions for improvements?
20140301 22:20:51< shadowm> It'd have been nice if I was given the choice to end the scenario at any time I wanted, since after capturing the first three ogres I had already calculated (MP-wise) that capturing any more would not be feasible with my current units.
20140301 22:21:18< shadowm> mattsc: I don't know, they seem to be doing exactly what they are supposed to, so... mission accomplished? :p
20140301 22:21:22-!- ancestral_ [~ancestral@74.174.236.84] has joined #wesnoth-dev
20140301 22:21:26< zookeeper> shadowm, fair enough about the income, i'd probably rather take the loyal one too
20140301 22:21:32< mattsc> shadowm: cool.  :)
20140301 22:21:44< zookeeper> yay
20140301 22:22:14< shadowm> I will probably steal the AI later.
20140301 22:23:01< mattsc> Great.  And as always, let me know if you'd like modifications/improvements or whatever for it.
20140301 22:23:07< shadowm> Waaaait.
20140301 22:23:13< shadowm> I had this loyal Ogre, all right.
20140301 22:23:18< shadowm> And two non-loyal Ogres in my recall list.
20140301 22:23:32< shadowm> So the loyal Ogre actually _is_ the best deal?
20140301 22:24:27< shadowm> Suddenly the choice seems very obvious.
20140301 22:25:08-!- ancestral [~ancestral@74.174.236.84] has quit [Ping timeout: 252 seconds]
20140301 22:25:09-!- ancestral_ is now known as ancestral
20140301 22:29:37-!- jamit [~jamit@wesnoth/developer/jamit] has quit [Ping timeout: 264 seconds]
20140301 22:32:19< shadowm> Xenophobia feels like such a filler scenario.
20140301 22:34:44 * mattsc goes to check out which one's Xenophobia
20140301 22:35:16< shadowm> The cheap one with the 1v1v1v1 situation.
20140301 22:35:26< shadowm> Humans vs elves vs orcs vs dwarves.
20140301 22:35:37< shadowm> Come to think of it, it sounds even more ridiculous when put into words.
20140301 22:37:01< mattsc> Oh, right, that one ...
20140301 22:37:18< mattsc> My memory's so bad when it comes to those kinds of things.  :P
20140301 22:37:53< Aishiko_laptop> sounds like your waiting for someone to come out into the middle of the map and declare "LETS GET READY TO RUMMMMBBBBLLLLLEEEEE!!!!!!!!!!!!!!!!!"    before you start
20140301 22:40:41-!- ancestral [~ancestral@74.174.236.84] has quit [Remote host closed the connection]
20140301 22:40:57< mattsc> Well, they _are_ ready to rumble.  You can either let them fight, wait it out and kill them off easily; or go after them actively, possibly lose some unit and go for the bonus gold
20140301 22:41:00-!- ancestral [~ancestral@74.174.236.84] has joined #wesnoth-dev
20140301 22:41:12< mattsc> s/unit/units
20140301 22:41:52-!- jamit [~jamit@68-115-35-102.dhcp.mdsn.wi.charter.com] has joined #wesnoth-dev
20140301 22:41:59-!- jamit [~jamit@68-115-35-102.dhcp.mdsn.wi.charter.com] has quit [Changing host]
20140301 22:41:59-!- jamit [~jamit@wesnoth/developer/jamit] has joined #wesnoth-dev
20140301 22:49:41< shadowm> Oh, there's an abandoned library in Lake Vrug now? I think that's new.
20140301 22:50:44< shadowm> "The island ahead is mountainous; it would be wise to select troops that do well on such terrain."  Yeah, easier said than done!
20140301 22:52:42< shadowm> Drakes!
20140301 22:52:46< shadowm> I thought you said gryphons?!
20140301 22:53:33-!- recycler [67157f4c@gateway/web/freenode/ip.103.21.127.76] has quit [Ping timeout: 245 seconds]
20140301 22:53:48< shadowm> Drakes *and* gryphons. Great.
20140301 22:54:46< Aishiko_laptop> where's draco when you need him? ohh right dead.
20140301 22:55:32< mattsc> I liked that scenario.
20140301 23:05:01-!- trademark_ [~trademark@nsg93-8-88-175-59-164.fbx.proxad.net] has quit [Ping timeout: 264 seconds]
20140301 23:06:40< zookeeper> shadowm, yeah, loci's rewrites added some plot stuff like that
20140301 23:08:31< zookeeper> so there's at least some farfetched point in them going all the way there... as opposed to before, when the master plan was apparently to run away to the most inhospitable location they could find and sit there waiting for a wonder
20140301 23:11:17< Aishiko_laptop> what no we must go here and get the rare flower before the X get it?
20140301 23:17:30-!- ancestral_ [~ancestral@74.174.236.84] has joined #wesnoth-dev
20140301 23:19:51< shadowm> Wheee!
20140301 23:19:56< shadowm> 20140301 20:19:46 error display: could not open image 'misc/ellipse-hero-leader-top.png'
20140301 23:19:59< shadowm> 20140301 20:19:46 error display: could not open image 'misc/ellipse-hero-leader-bottom.png'
20140301 23:20:29< Aishiko_laptop> shadowm, ohhh joy!
20140301 23:20:37< shadowm> zookeeper: Because you did promoted Dacyn in Captured without removing his ellipse.
20140301 23:20:43-!- ancestral [~ancestral@74.174.236.84] has quit [Ping timeout: 252 seconds]
20140301 23:20:43-!- ancestral_ is now known as ancestral
20140301 23:21:10< shadowm> Why is Dacyn saying "I'm sure the commander will come and save us?"
20140301 23:21:15< zookeeper> uh oh
20140301 23:21:47< shadowm> Actually, am I even supposed to be able to move Dacyn?
20140301 23:22:03< zookeeper> wanna hand me the beginning-of-scenario save?
20140301 23:22:20< shadowm> Sure, hang on a minute...
20140301 23:22:21< zookeeper> no, you're not. you shouldn't even be able to see him.
20140301 23:22:40< zookeeper> something's going wrong with the escapee selection there, although i thought i made it pretty foolproof... i guess not
20140301 23:23:03< shadowm> zookeeper: https://dl.dropboxusercontent.com/u/21371130/junk/EI-Captured.gz
20140301 23:23:22< zookeeper> oh right, i see the problem
20140301 23:24:22< zookeeper> i can commit a fix in a few minutes, hang on...
20140301 23:27:17-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140301 23:27:39< irker924> wesnoth: ln-zookeeper wesnoth:master 455bb9b55ab5 / data/campaigns/Eastern_Invasion/scenarios/11_Captured.cfg: Fixed possibility of getting Dacyn as escapee sidekick 2. http://git.io/EOOBsw
20140301 23:27:43< zookeeper> ought to be fixed now
20140301 23:29:02< zookeeper> ...and yeah, i know the first half of the scenario is very boring, but hopefully the latter half isn't
20140301 23:29:21< shadowm> Is it changelog-worthy?
20140301 23:30:03< shadowm> Okay, fix confirmed, let's see how the rest goes.
20140301 23:32:16-!- kex [~kex@89.205.75.19] has quit [Ping timeout: 252 seconds]
20140301 23:36:05< zookeeper> dunno, i guess it's unlikely to trigger that bug. i think you'd need to have none of Spearman,Mage,Bowman,Mage,Red Mage,Longbowman,Pikeman,Swordsman,Javelineer,Cavalryman,Horseman on your recall list after the other two escapees have been chosen
20140301 23:36:26< shadowm> zookeeper: Hm, how do I release the prisoners?
20140301 23:36:38< zookeeper> move next to the door(s)
20140301 23:36:50< shadowm> I moved next to the cell gates and they changed to side 1 and everything, but the gates are still impassable.
20140301 23:37:55< zookeeper> huh? doesn't happen for me
20140301 23:38:17< shadowm> It works for 17,10 but not for the rest.
20140301 23:38:47< shadowm>     terrain=Re^Xo
20140301 23:38:55< shadowm> Because the rest has a different base terrain.
20140301 23:39:04< zookeeper> wtf's happened to the map
20140301 23:40:03< zookeeper> buuuuuuuuumbadadabum
20140301 23:40:27< zookeeper> 0efe12da4fe596324c1be9008d39f66b96cc9aa1 broke stuff
20140301 23:40:27 * shadowm lua wesnoth.wml_actions.terrain { x = "21,29", y = "8,9", terrain = "Re^Xo" } wesnoth.wml_actions.redraw {}
20140301 23:42:27< zookeeper> so, you have to revert that commit to fix the scenario
20140301 23:42:45< shadowm> bumbadadabum: ^
20140301 23:43:16< shadowm> zookeeper: Why not just change the filter and terrain replacement?
20140301 23:43:35< bumbadadabum> oh
20140301 23:43:45< bumbadadabum> I was unaware there was filtering for a specific terrain
20140301 23:43:57< bumbadadabum> I'm sorry
20140301 23:44:26< zookeeper> shadowm, well i'm not sure i like the mix of regular and earthy caves either, and the mixing of road and dirt
20140301 23:45:07< bumbadadabum> revert it if you really don't like it
20140301 23:45:12< zookeeper> bumbadadabum, well at least it was after the release, so no harm done except taking up shadowm's time :p
20140301 23:45:28< bumbadadabum> I should have looked it through better
20140301 23:45:30< bumbadadabum> oh well
20140301 23:45:37< bumbadadabum> zookeeper: If you don't like it, revert
20140301 23:46:07< bumbadadabum> or revert that map alone, if it's possible
20140301 23:46:12< bumbadadabum> I think there were more maps in the commit
20140301 23:46:21< zookeeper> the ogre scenario's map too, yeah
20140301 23:46:38< bumbadadabum> yeah
20140301 23:46:39-!- ancestral [~ancestral@74.174.236.84] has quit [Quit: i go nstuf kthxbai]
20140301 23:46:46< bumbadadabum> I think that one turned out a bit better anyway
20140301 23:48:30< zookeeper> i could just copypaste the old map over it and commit that and call it a revert, but i'm sure that'd be considered an unholy practise and that it should really be done with some arcane git lore
20140301 23:49:22< bumbadadabum> I think that's not THAT taboo
20140301 23:49:26< bumbadadabum> shadowm: is it?
20140301 23:50:29< zookeeper> the ogre map looks ok, although some of the added hill padding does affect how fast cavalry can get to NE/NW
20140301 23:50:58< bumbadadabum> cavalry can't move over mountains
20140301 23:51:05< bumbadadabum> so it shouldn't be too much of a problem
20140301 23:51:41< shadowm> git revert is the equivalent of applying a patch in reverse.
20140301 23:52:11< zookeeper> bumbadadabum, that doesn't make any sense
20140301 23:52:22< zookeeper> added hills slow down cavalry. mountains didn't change.
20140301 23:52:35< bumbadadabum> well, but I mean
20140301 23:52:43< bumbadadabum> a lot of the movement is restricted by cavalry
20140301 23:52:46< shadowm> Oh, so that means I played a version of the ogres scenario that's harder than it was intended to. :x
20140301 23:52:49< shadowm> *to be
20140301 23:53:01< bumbadadabum> I don't think I blocked anything completely with hills
20140301 23:53:02-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20140301 23:53:26< zookeeper> shadowm, marginally, i believe. unless you recruited cavalrymen/etc immediately and they didn't make it to NW/NE in time to catch some ogres
20140301 23:54:17< bumbadadabum> I tried to make it look better without changing gameplay too much
20140301 23:54:39< shadowm> zookeeper: I did recruit cavalrymen immediately on the first turn, 5 of them.
20140301 23:55:05< shadowm> No, 4.
20140301 23:55:22< shadowm> I figured it was the only way to catch up with the ogres since they are rather fast.
20140301 23:56:01< zookeeper> yeah
--- Log closed Sun Mar 02 00:00:24 2014