--- Log opened Fri Mar 18 00:00:53 2016
20160318 00:04:07< SigurdFD> Should I have marked this bug as important? https://gna.org/bugs/index.php?24383
20160318 00:05:09< celticminstrel> Doesn't sound super-important to me, but definitely something to be looked at...
20160318 00:05:25< celticminstrel> I dunno how much people care about the stats.
20160318 00:06:59< Ravana_> just enough to feel justified to be complaining about luck
20160318 00:07:53< SigurdFD> :D Forgot about the luck crowd.
20160318 00:08:39< SigurdFD> I've been using them to compare recruit/recall/advancements/losses between & during scenarios.
20160318 00:09:47< SigurdFD> The bug probably extends to mp campaigns, I've only tested it in a single scenario vs computer so far.
20160318 00:13:24< fabi> celticminstrel: About FormulaAI: Isn't it mostly replaced by lua for AI purposes but only used for inline WML arithmetic nowadays?
20160318 00:13:47< fabi> celticminstrel: Where can I see the changes to FormulaAI?
20160318 00:19:09-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20160318 00:19:23-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160318 00:19:38< Ravana_> yes, thats why it has been renamed recently
20160318 00:28:11-!- Greg-Bog_ [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160318 00:31:02-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 250 seconds]
20160318 00:32:24-!- Greg-Bog_ [~greg_bogg@173.240.241.83] has quit [Ping timeout: 246 seconds]
20160318 00:42:54-!- prkc [~prkc@108.61.123.89] has quit [Ping timeout: 246 seconds]
20160318 00:56:34-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 268 seconds]
20160318 00:57:34-!- prkc [~prkc@192.40.89.11] has joined #wesnoth-dev
20160318 01:03:49-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160318 01:09:16< celticminstrel> fabi: It can't really be said that it was replaced with Lua in general, but for AI, yeah, it's been largely supplanted. In mainline, only LoW uses AI formulas, and I think only in one scenario.
20160318 01:09:32< celticminstrel> fabi: The PR is here, though I'm still working on it a bit so some new stuff will probably be added soon.
20160318 01:09:33< celticminstrel> variant get_value(const std::string& key) const
20160318 01:09:35< celticminstrel> ...
20160318 01:09:42< celticminstrel> My copy didn't register.
20160318 01:09:45< celticminstrel> https://github.com/wesnoth/wesnoth/pull/624
20160318 01:10:00< fabi> thank you
20160318 01:10:58< celticminstrel> If you have any input in particular on commits 08734e6 and ca2da8b in particular, that'd be great; I'm starting to think the latter is not really necessary and the former would be better eliminated in favour of an actual boolean type.
20160318 01:11:16< celticminstrel> But any input is appreciated, of course.
20160318 01:12:52< wedge009> vultraz, Aginor, and anyone else working on graphics: Something has broken recently, but I presume there is work ongoing for this - https://gna.org/bugs/?24528
20160318 01:13:52< wedge009> I'm not sure what's happening, but I know the story screens were working previously... (:
20160318 01:14:20 * celticminstrel also draws their attention to PR562 for you.
20160318 01:22:28-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 268 seconds]
20160318 01:24:01-!- louis94 [~~louis94@91.178.242.45] has quit [Ping timeout: 252 seconds]
20160318 01:26:50-!- iceiceice [~chris@ext-74.ias.edu] has joined #wesnoth-dev
20160318 01:26:50-!- iceiceice [~chris@ext-74.ias.edu] has quit [Changing host]
20160318 01:26:50-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20160318 02:06:57< SigurdFD> Can someone set an item group for https://gna.org/bugs/?24529 I forgot to set it?
20160318 02:07:20< celticminstrel> I have no idea what that means.
20160318 02:07:47< SigurdFD> Also, I'm pretty sure that https://gna.org/bugs/?24527 should at least rate as 'Important' 
20160318 02:08:35-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has joined #wesnoth-dev
20160318 02:08:40< SigurdFD> item group is stuff such as 'WML' 'Graphics', 'Campaign' It's a way of labeling what part of wesnoth the bug is in.
20160318 02:10:36< SigurdFD> I guess it doesn't really matter.
20160318 02:10:56< celticminstrel> Ah.
20160318 02:11:03-!- irker054 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160318 02:13:49-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 248 seconds]
20160318 02:15:02-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20160318 02:17:34-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160318 02:21:55-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 248 seconds]
20160318 02:26:44< SigurdFD> shadowm, vultrav or anyone else: is there anything anywhere about forum titles and the guidelines for qualifying?
20160318 02:26:52< SigurdFD> vultraz ^
20160318 02:27:42-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160318 02:29:28< celticminstrel> wedge009: Have you tried SigurdFD's saved game yet?
20160318 02:29:54< wedge009> Just about to - still recovering from a BSOD.
20160318 02:31:05< celticminstrel> wml-tags.lua:657 is in [store_unit].
20160318 02:31:53< celticminstrel> Hmm, it's possible it might've been caused by my location-related stuff.
20160318 02:32:13< SigurdFD> I retested, I'm still getting the error, and I removed all add-on manually before starting (instead of --noaddons, which I did before)
20160318 02:32:27< SigurdFD> ... and created a new save, and still same thing.
20160318 02:32:35< celticminstrel> Yup, I think this is my fault.
20160318 02:34:42< wedge009> At least you know where to look. Cool.
20160318 02:35:27< wedge009> I'll skip trying to replay from S1 then.
20160318 02:35:43< celticminstrel> wedge009: Did you try the saved game though?
20160318 02:35:57< wedge009> Yes I did, I said as much in the report.
20160318 02:37:01< SigurdFD> wedge009: I just recruited a few units, create some with debug to kill the orc to move to s2
20160318 02:37:19< wedge009> Ah.
20160318 02:37:41< wedge009> Well, I think there may still be not much difference between that and playing normally.
20160318 02:39:07< SigurdFD> ugh... If I just start s1, and straight up kill the orc with debug kill, s2 works normally
20160318 02:39:56-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160318 02:40:21< SigurdFD> hmmm... creating units with debug seems to be involved
20160318 02:41:13-!- irker163 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160318 02:41:13< irker163> wesnoth: Celtic Minstrel wesnoth:master f02a038879ec / src/scripting/game_lua_kernel.cpp: Fix wesnoth.erase_unit for recall units https://github.com/wesnoth/wesnoth/commit/f02a038879ec366fac643ee3b5a6756b78b7e0e1
20160318 02:41:15< celticminstrel> See if that fixes it ^
20160318 02:44:19-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 252 seconds]
20160318 02:49:49< SigurdFD> That seems to have fixed it.
20160318 02:50:07< celticminstrel> Yay.
20160318 02:50:13< celticminstrel> I guess that means I should close it?
20160318 02:51:28< SigurdFD> thanks :) . I think so. That's what it says at the bottom of: https://wiki.wesnoth.org/ReportingBugs
20160318 02:52:54 * celticminstrel should probably do a quick review of the location stuff to make sure nothing else similar slipped through.
20160318 03:01:38< iceiceice> i mean ideally there could be a test for stuff like that
20160318 03:02:01< iceiceice> if its an issue iwth the lua api and not an issue with the dialog
20160318 03:02:12< celticminstrel> ?
20160318 03:02:43< iceiceice> maybe i didnt understand the issue
20160318 03:02:52< iceiceice> it looked like a minor logic issue in impl of erase_unit?
20160318 03:02:57< celticminstrel> Yes.
20160318 03:03:23< iceiceice> usually what i did was, when i find a regression like that, fix it and also make a regression test
20160318 03:03:47< celticminstrel> So you were talking about a unit test.
20160318 03:03:51< iceiceice> yeah
20160318 03:04:07< iceiceice> otherwise its easy like to have, one step forward, one step back fixes / refactors / etc.
20160318 03:04:13< celticminstrel> A test for erase_unit specifically wouldn't be too hard, I think. I might do that later.
20160318 03:05:26-!- Necrosporus_ [~Necrospor@unaffiliated/necrosporus] has joined #wesnoth-dev
20160318 03:08:15-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has quit [Disconnected by services]
20160318 03:08:17-!- Necrosporus_ is now known as Necrosporus
20160318 03:09:44 * celticminstrel poke fabi
20160318 03:15:03< shadowm> SigurdFD: Group membership or special titles? The latter are given at the sole discretion of the forum administration.
20160318 03:17:15< SigurdFD> I guess group membership then. specificially 'Code & WML Contributor' & 'UMC Pioneer', also 'Music & Art' & Translators
20160318 03:17:30< shadowm> gfgtdf, celticminstrel: I use inline formulas. If the formula language is removed I don't think I'll have any motivation left to maintain my add-ons anymore.
20160318 03:18:04< shadowm> SigurdFD: UMC Pioneer and Forum Regulars are currently in limbo after the last person in charge of them left.
20160318 03:18:55< celticminstrel> I'm against removing it.
20160318 03:19:04< shadowm> Code & WML Contributor is given to people who've had pull requests accepted on GitHub. However, because the people actually in charge of accepting pull requests have never bothered to report to me and I don't have the time to track them down, it's also unmaintained.
20160318 03:19:22< celticminstrel> shadowm: By inline do you mean $(...)? Stuff more complicated than just simple arithmetic?
20160318 03:19:24< shadowm> You can still ask me to add you if you can provide me with links.
20160318 03:19:49< celticminstrel> Sigurd has had PRs accepted.
20160318 03:19:54< SigurdFD> ok. So you're one one I'd send a pm to about Code & WML specifically?
20160318 03:20:05< shadowm> Music & Art Contributors have the same issue more or less. Membership is decided by our Art Director, however.
20160318 03:20:20< celticminstrel> That's Jetrel, right?
20160318 03:20:27< shadowm> Translators have the same issue because Ivanovic is lazy and almost never tells me about new translators with forum accounts.
20160318 03:20:32< shadowm> celticminstrel: Yes.
20160318 03:20:37< celticminstrel> Heh
20160318 03:20:44< shadowm> SigurdFD: Yes.
20160318 03:21:09< celticminstrel> So shadowm, do you use $(...) for things more complicated than just arithmetic?
20160318 03:21:29< shadowm> celticminstrel: In one or two places. My plan going forward was to do it more often.
20160318 03:21:38< celticminstrel> I see.
20160318 03:22:07 * celticminstrel has no problem with this, in fact it's sort of nice to hear that it's not a completely unused feature.
20160318 03:22:33 * celticminstrel wonders if anyone has ever run into the fact that exponentiation is left-associative in the formula engine.
20160318 03:23:24< SigurdFD> thanks, shadowm.
20160318 03:27:21< SigurdFD> shadowm: so I guess it's unwritten convention governing all those?
20160318 03:33:45-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20160318 03:37:45< shadowm> SigurdFD: Yes.
20160318 03:39:36< shadowm> SigurdFD: You are now member of the C&W Contributors group.
20160318 03:44:37< iceiceice> celticminstrel, i think shadowm is probably one of the only people that does this
20160318 03:44:49< iceiceice> given that the documentation for the more advanced stuff barely exists
20160318 03:45:06< iceiceice> most umc htat i've encountered uses only lua
20160318 03:45:07< iceiceice> *sorry
20160318 03:45:09< iceiceice> only wml
20160318 03:45:12< iceiceice> and no lua or formula
20160318 03:45:26< iceiceice> which i guess is understandable
20160318 03:45:45< celticminstrel> iceiceice: The documentation exists now.
20160318 03:45:48< iceiceice> really?
20160318 03:45:51< celticminstrel> Yes.
20160318 03:46:07< iceiceice> did you like reverse engineer it?
20160318 03:46:15 * iceiceice is interested in this
20160318 03:46:23< celticminstrel> Most of it was already around, just packed away in the FormulaAI pages.
20160318 03:46:39< celticminstrel> But yes, there was some "reverse engineering" involved too.
20160318 03:46:50< iceiceice> how do i find it?
20160318 03:46:54< celticminstrel> https://wiki.wesnoth.org/Wesnoth_Formula_Language
20160318 03:47:04< SigurdFD> shadowm: thanks :)
20160318 03:49:56< iceiceice> that is some nice work, kudos
20160318 03:50:21-!- Samual [~Samual@xonotic/core-team/Samual] has quit [Ping timeout: 248 seconds]
20160318 03:56:23-!- Samual [~Samual@xonotic/core-team/Samual] has joined #wesnoth-dev
20160318 04:19:15< shadowm> "I think 1.12.5 is honestly going to be the last 1.12 release." I assume that you reached this conclusion based on 1.12's bug report rate and not its bug fixing rate.
20160318 04:20:28< shadowm> Or maybe I did just too damn good a job with those 1.12.x releases.
20160318 04:21:06-!- Jetrel_bot is now known as scarlet_pinguin
20160318 04:21:35-!- scarlet_pinguin is now known as ddddRRR
20160318 04:21:46-!- ddddRRR is now known as Jetrel_bot
20160318 04:24:19< celticminstrel> shadowm: Do you think it would be good if $(...) formulas could directly access the event context as formula variables rather than through WML variable substitution? eg, you might write $(unit.max_hitpoints - unit.hitpoints) instead of $($unit.max_hitpoints - $unit.hitpoints) -- I guess that example doesn't really show much of an advantage, though...
20160318 04:25:09< shadowm> Yeah, I don't see much of a point given the effort involved in implementing that.
20160318 04:25:37< shadowm> After all, more of then than not you'll need to access custom variables that aren't part of the event context.
20160318 04:25:54< shadowm> Er.
20160318 04:25:58< shadowm> *more often than not
20160318 04:26:04< celticminstrel> I was only thinking of the event context, not WML variables in general.
20160318 04:26:22< celticminstrel> I haven't yet investigated the difficulty of it.
20160318 04:28:58< celticminstrel> I guess what you're saying is that it's not so useful if it's just the event contex.
20160318 04:31:09< shadowm> And it seems like extra effort for little gain.
20160318 04:31:34< shadowm> By the way, it's shadowm, not shadowmaster.
20160318 04:31:58< celticminstrel> ...oh, are you referencing that forum post from awhile ago?
20160318 04:32:38< shadowm> If less than 24 hours ago count as a 'while ago', yes.
20160318 04:32:47< celticminstrel> It feels like awhile. >_>
20160318 04:32:55< celticminstrel> I guess it was this morning then.
20160318 04:33:06< shadowm> Incidentally, I checked and there isn't a phpBB 3.1 syntax highlighting extension in the official database yet.
20160318 04:33:49< shadowm> It seems like it was considered as a feature for 3.2 but it has already missed the freeze I believe.
20160318 04:34:14< celticminstrel> Does "as a feature" mean built-in to the software?
20160318 04:34:25< shadowm> Yes.
20160318 04:34:52 * celticminstrel wonders if that would use Pygments though.
20160318 04:35:10< shadowm> That's what was proposed.
20160318 04:35:18< celticminstrel> Okay.
20160318 04:35:28< shadowm> The older syntax highlighting mod for 3.0 used GeShi, which apparently is unmaintained or something.
20160318 04:35:57< celticminstrel> It's surprisingly convenient that MediaWiki, phpBB, and apparently even GitHub use the same syntax highlighting framework... though I suspect it wouldn't be installed on github since it literally only applies to one project.
20160318 04:35:58< shadowm> I reckon that'st motivated the authors of the MW extension to switchi to Pygments too.
20160318 04:36:02< shadowm> *that's what
20160318 04:36:52< shadowm> I don't know what GitHub's syntax highlighting backend uses, but if it's pygmnts then all the more reason for you to try getting your lexer accepted upstream!
20160318 04:37:00< shadowm> ;)
20160318 04:38:31< shadowm> (Although if you do that, you might want to keep in mind we are not the only claimants to the "WML" name: https://en.wikipedia.org/wiki/Wireless_Markup_Language )
20160318 04:39:09< celticminstrel> Hmm, actually, if they do use Pygments, then I wonder if a Linguist override file would be able to select it even if Linguist itself has no entry for it...
20160318 04:53:44-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has quit [Read error: Connection reset by peer]
20160318 04:54:06-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has joined #wesnoth-dev
20160318 04:57:39-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Ping timeout: 276 seconds]
20160318 04:59:53-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20160318 05:00:52< celticminstrel> Oh huh, apparently someone's working on the map diff idea.
20160318 05:05:38-!- SigurdFD [~SigurdFD@dynamic-acs-72-23-176-151.zoominternet.net] has quit [Read error: No route to host]
20160318 05:40:10-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160318 05:42:48-!- irker163 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160318 05:44:21-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160318 05:47:00-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160318 06:04:59-!- Kwandulin [~Miranda@p200300760F7D73F6216D4A46081159BD.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160318 07:06:17< vultraz> I'm back
20160318 07:07:06< celticminstrel> Why am I not in bed yet...
20160318 07:07:28< vultraz> sorry about storming off earlier
20160318 07:07:36< celticminstrel> Ugh, my Travis build errored.
20160318 07:09:23< vultraz> btw, that storyscreen issue probably happened when I refactored out the display class from the storyscreen
20160318 07:11:39< vultraz> shadowm: former
20160318 07:18:41< celticminstrel> Though also see my response on the PR thread.
20160318 07:18:55-!- celticminstrel is now known as celmin|sleep
20160318 07:25:09< vultraz> might need Aginor for this one
20160318 07:49:57-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20160318 08:01:14< vultraz> wait, I got it
20160318 08:05:54-!- EdB [~edb@89.193.129.77.rev.sfr.net] has joined #wesnoth-dev
20160318 08:06:52-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160318 08:06:52-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Read error: Connection reset by peer]
20160318 08:07:14-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160318 08:07:36-!- irker116 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160318 08:07:36< irker116> wesnoth: Charles Dang wesnoth:master d47336e41d74 / src/storyscreen/render.cpp: Fix storyscreen buttons disappearing when drawing title (bug #24528) https://github.com/wesnoth/wesnoth/commit/d47336e41d745268782fb1541e02d290ff33da93
20160318 08:12:04-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 260 seconds]
20160318 08:14:48-!- boucman_work [~boucman@193.84.89.18] has joined #wesnoth-dev
20160318 08:35:43-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has joined #wesnoth-dev
20160318 08:39:16-!- travis-ci [~travis-ci@ec2-54-167-158-117.compute-1.amazonaws.com] has joined #wesnoth-dev
20160318 08:39:17< travis-ci> wesnoth/wesnoth#8925 (master - d47336e : Charles Dang): The build has errored.
20160318 08:39:17< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/116839635
20160318 08:39:17-!- travis-ci [~travis-ci@ec2-54-167-158-117.compute-1.amazonaws.com] has left #wesnoth-dev []
20160318 08:41:14-!- Kwandulin [~Miranda@p200300760F7D73F6216D4A46081159BD.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160318 08:53:35-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160318 09:55:33-!- mjs-de [~mjs-de@x4db62f58.dyn.telefonica.de] has joined #wesnoth-dev
20160318 10:06:16-!- louis94 [~~louis94@91.178.242.151] has joined #wesnoth-dev
20160318 10:10:08-!- louis94 [~~louis94@91.178.242.151] has quit [Client Quit]
20160318 10:19:33-!- mjs-de [~mjs-de@x4db62f58.dyn.telefonica.de] has quit [Ping timeout: 240 seconds]
20160318 10:22:08-!- mjs-de [~mjs-de@x4db62f58.dyn.telefonica.de] has joined #wesnoth-dev
20160318 10:25:45-!- mjs-de [~mjs-de@x4db62f58.dyn.telefonica.de] has quit [Remote host closed the connection]
20160318 10:26:57-!- mjs-de [~mjs-de@x4db62f58.dyn.telefonica.de] has joined #wesnoth-dev
20160318 10:30:45-!- Appleman1234 [~Appleman1@KD106161155152.au-net.ne.jp] has left #wesnoth-dev ["Leaving"]
20160318 10:32:39-!- mjs-de [~mjs-de@x4db62f58.dyn.telefonica.de] has quit [Ping timeout: 244 seconds]
20160318 10:34:19-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160318 10:38:48-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160318 10:39:16-!- mjs-de [~mjs-de@x4db62f58.dyn.telefonica.de] has joined #wesnoth-dev
20160318 10:39:58-!- mjs-de [~mjs-de@x4db62f58.dyn.telefonica.de] has quit [Remote host closed the connection]
20160318 11:12:37-!- irker116 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160318 11:34:54-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160318 11:44:15-!- EdB [~edb@89.193.129.77.rev.sfr.net] has quit [Quit: Konversation terminated!]
20160318 11:58:18< vultraz> Aginor: I've been gradually phasing out the use of set_volatile() in GUI1. Basically, in cases where refactoring has changed a call from display::flip() to CVideo::flip(), you lose the call to raise_volatile_draw_event, meaning that set_volatile() actually makes the widget not show up.
20160318 11:58:54< vultraz> but I'd like your approval before removing it entirely
20160318 12:06:06-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160318 12:09:02< vultraz> I've tested without the final usecases (in display::layout_buttons() and floating_textbox::update_location()) and it seems to work fine without it
20160318 12:38:51< vultraz> ah wait
20160318 12:38:59< vultraz> no, the second case breaks :|
20160318 12:39:22< vultraz> Aginor: ok, so I think it can be removed from layout_buttons() but not the second
20160318 12:59:45-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has quit [Read error: Connection reset by peer]
20160318 13:30:56< celmin|sleep> I honestly can't tell the difference with that contrast bump (without looking really closely).
20160318 13:41:20-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has joined #wesnoth-dev
20160318 13:43:12-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20160318 13:43:30-!- Nobun [~nobun@host62-49-dynamic.12-79-r.retail.telecomitalia.it] has joined #wesnoth-dev
20160318 13:44:04< mattsc> celmin|sleep: I am rushing through, I just have a few quick comments, questions on your latest commits to the AI branch for now.
20160318 13:44:07< celmin|sleep> Hi.
20160318 13:44:19< mattsc> hi
20160318 13:44:27< celmin|sleep> I still need to fix FormulaAI recruitment.
20160318 13:44:51< mattsc> 1. I wouldn’t have made this part of the branch, since it’s really not a refactor, but it’ll still show up as a separate commit, so, okay.
20160318 13:45:08< celmin|sleep> ?
20160318 13:45:35< mattsc> 2. I don’t like the commit title, as it’s really not a fix to the MAIs, but a specific MAI.  Also, there’s another MAI that does the same.
20160318 13:45:57< celmin|sleep> I missed something?`Did I miss something?
20160318 13:46:05< mattsc> 3. Have you actually tested this?  I am asking because what you have now is what I originally had and then I changed it.
20160318 13:46:17< celmin|sleep> Ack escape didn't register.
20160318 13:46:24< mattsc> ?
20160318 13:46:35< mattsc> And I clearly remember that I changed it because it did not work at first.
20160318 13:46:43< celmin|sleep> Is this all about the MAI commit?
20160318 13:46:49< mattsc> yes
20160318 13:47:37< mattsc> And it might not have worked because the rest of the code was different and that might not be the case any more, but I just don’t remember.  So I think it needs to be tested.  Do I still need to do that or did you?
20160318 13:48:24< celmin|sleep> I did not.
20160318 13:49:17< mattsc> ok
20160318 13:49:27-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20160318 13:50:10< celmin|sleep> BTW I will be changing [params] to [args] like gfgtdf recommended.
20160318 13:50:24< mattsc> ok
20160318 13:51:04< celmin|sleep> What's the other MAI that does this?
20160318 13:51:46< mattsc> Hang Out  (you can just use grep for that)
20160318 13:52:06< mattsc> And I just checked the code, that one is the reason why it had to be done that way.
20160318 13:52:38< mattsc> And I guess then I must have just copied the method over to the Fast MAI, where it is not needed.
20160318 13:53:06< celmin|sleep> Doing it that way is still not good though.
20160318 13:53:13-!- celmin|sleep is now known as celticminstrel
20160318 13:53:35< mattsc> Yes; but you cannot just use get_avoid()
20160318 13:54:56< celticminstrel> I don't understand why this case is any different... it looks exactly the same to me...
20160318 13:55:10< celticminstrel> Oh wait. "if neither is given"?
20160318 13:55:18< celticminstrel> Is that what the difference is?
20160318 13:55:41< mattsc> yes
20160318 13:55:56< mattsc> It has a default value that’s not [not][/not]
20160318 13:56:10< celticminstrel> Okay, I think what could be done is, use that method to check if the aspect is set, but then use get_avoid() to actually fetch the value.
20160318 13:56:20< mattsc> agreed
20160318 13:56:59< mattsc> I still think that the commit subject should be change as that (to me) implies that now all MAIs respect the avoid aspect.
20160318 13:57:16< celticminstrel> I'll change the commit message too.
20160318 13:57:22< mattsc> Which they don’t (and I think they shouldn’t).
20160318 13:57:25< mattsc> Thanks.
20160318 13:57:37< mattsc> And I’ll test the changes afterwards, just to be sure.
20160318 13:57:43< celticminstrel> Well, there are others that possibly could make sense to respect it, but probably not all, yeah.
20160318 13:57:44-!- EdB [~edb@89.193.129.77.rev.sfr.net] has joined #wesnoth-dev
20160318 13:57:59< mattsc> yes, agreed on that too
20160318 14:01:25< vultraz> celticminstrel: do a slide comparison on github
20160318 14:01:29< celticminstrel> I keep getting tempted to end Lua statements with a semicolon.
20160318 14:01:50< celticminstrel> vultraz: I did. Admittedly only with one image though.
20160318 14:02:43< vultraz> well the effect is only supposed to be subtle enough to improve the color but not change the look significantly
20160318 14:06:08-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20160318 14:06:14-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has joined #wesnoth-dev
20160318 14:08:34< celticminstrel> I almost deleted the default without realizing it.
20160318 14:10:37-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has quit [Ping timeout: 248 seconds]
20160318 14:13:20< celticminstrel> Pushed it. Haven't fixed the unit test failure though.
20160318 14:13:46< celticminstrel> Actually, this situation might be an argument in favour of reverting the mutability of [default].
20160318 14:16:24< celticminstrel> I also haven't yet implemented the [params] -> [args] change.
20160318 14:17:45< celticminstrel> Oh, vultraz, I think the new tooltips might be worthy of mention in not just changelog but also players_changelog.
20160318 14:17:56< vultraz> alright
20160318 14:20:50-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160318 14:21:54-!- irker791 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160318 14:21:55< irker791> wesnoth: Charles Dang wesnoth:master 5dcb578c3d0f / data/gui/default/macros/_initial.cfg: Brightened the standard border color a little https://github.com/wesnoth/wesnoth/commit/5dcb578c3d0f775a2fbe9f8167b2c8b858732ee1
20160318 14:22:21< vultraz> celticminstrel: next time you pull, tell me if you think that makes borders a little more clearer
20160318 14:23:30< celticminstrel> I have a feeling I won't notice the difference... >_>
20160318 14:23:39< celticminstrel> But if I do, I'll let you know.
20160318 14:24:03-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 240 seconds]
20160318 14:24:04-!- wedge010 is now known as wedge009
20160318 14:32:01-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has quit [Ping timeout: 240 seconds]
20160318 14:36:43< irker791> wesnoth: Charles Dang wesnoth:master d299ab129246 / data/gui/default/macros/_initial.cfg: Reduce brightness of default font color slightly https://github.com/wesnoth/wesnoth/commit/d299ab129246a27ea1765c6deeb124fd8af5d979
20160318 14:39:29-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has joined #wesnoth-dev
20160318 14:48:38< irker791> wesnoth: Charles Dang wesnoth:master d764bba7e46d / data/gui/default/macros/_initial.cfg: Procedurally generate selected listbox cell borders https://github.com/wesnoth/wesnoth/commit/d764bba7e46dc4465506ee7c02be5401b93b4f87
20160318 14:59:12-!- EdB [~edb@89.193.129.77.rev.sfr.net] has quit [Quit: Konversation terminated!]
20160318 15:19:36< celticminstrel> mattsc: Should I remove commit b476490 ("Change syntax of fallback stage") from the PR? It's pointless if we plan to remove ai::interface and the fallback stage by 1.13.5. I don't think removing ai::interface would even break anything, though it's a big enough change that I can't be sure (which is why I didn't end up including it in the PR - I started, then gave up).
20160318 15:22:05-!- boucman_work [~boucman@193.84.89.18] has quit [Ping timeout: 248 seconds]
20160318 15:29:39< mattsc> celticminstrel: doesn’t really matter IMO
20160318 15:29:44< mattsc> whatever you prefer is fine by me
20160318 15:30:11< celticminstrel> Well, part of that question is whether we should plan to remove them by 1.13.5.
20160318 15:33:47< mattsc> I’d do whatever cleanup we (you) want to do by 1.13.5.
20160318 15:34:08< mattsc> And in my opinion, the more unnecessary and/or duplicate stuff we get rid of the better.
20160318 15:34:37< mattsc> So that’s the long-winded way to say: I would, but you’re the one who’ll have to do it, so it’s really up to you.
20160318 15:38:09-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160318 15:40:55-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: mattsc]
20160318 15:43:41-!- timotei_ [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 244 seconds]
20160318 15:49:15-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160318 15:56:36-!- travis-ci [~travis-ci@ec2-54-167-158-117.compute-1.amazonaws.com] has joined #wesnoth-dev
20160318 15:56:37< travis-ci> wesnoth/wesnoth#8928 (master - 5dcb578 : Charles Dang): The build passed.
20160318 15:56:37< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/116912867
20160318 15:56:37-!- travis-ci [~travis-ci@ec2-54-167-158-117.compute-1.amazonaws.com] has left #wesnoth-dev []
20160318 16:03:52-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20160318 16:05:46-!- gfgtdf [~chatzilla@f054175005.adsl.alicedsl.de] has joined #wesnoth-dev
20160318 16:07:14-!- Kwandulin [~Miranda@p200300760F7D7365959E83CD65929EDC.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160318 16:08:20-!- timotei_ [~timotei@188.24.18.173] has joined #wesnoth-dev
20160318 16:08:21-!- timotei_ [~timotei@188.24.18.173] has quit [Changing host]
20160318 16:08:21-!- timotei_ [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20160318 16:08:56-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has joined #wesnoth-dev
20160318 16:09:50-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160318 16:10:05< mattsc> ancestral: hi
20160318 16:13:33-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has quit [Ping timeout: 268 seconds]
20160318 16:14:50-!- travis-ci [~travis-ci@ec2-54-197-104-64.compute-1.amazonaws.com] has joined #wesnoth-dev
20160318 16:14:51< travis-ci> wesnoth/wesnoth#8929 (master - d299ab1 : Charles Dang): The build passed.
20160318 16:14:51< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/116916847
20160318 16:14:51-!- travis-ci [~travis-ci@ec2-54-197-104-64.compute-1.amazonaws.com] has left #wesnoth-dev []
20160318 16:19:46-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160318 16:21:22< celticminstrel> mattsc: I have no particular problem with removing anything that's unused before 1.13.5. Of course that does not include FormulaAI though, as that's used byt LoW and several UMC.
20160318 16:21:24< celticminstrel> ^by
20160318 16:21:50< celticminstrel> So I guess I'll go ahead and remove ai::interface, but I'll wait until after this PR is merged.
20160318 16:21:58< shadowm> Anyone here familiarized with GCC's sanitizer options?
20160318 16:22:06< celticminstrel> Sanitizer?
20160318 16:22:13< shadowm> That means you aren't.
20160318 16:22:19< celticminstrel> Probably, yeah.
20160318 16:22:58< shadowm> For anyone who actually is: why is my program dying before even getting to main()? https://gist.githubusercontent.com/shikadilord/25ad9c594053d520df63/raw
20160318 16:24:15< shadowm> I don't think it's even gotten to run static constructors yet.
20160318 16:25:30< mattsc> celticminstrel: and sorry for being so slow, but I don’t have much time and the MAIs are giving more problems than I expected.
20160318 16:26:02< mattsc> *okay; and sorry ...
20160318 16:26:12< mattsc> (not sure who ate that okay)
20160318 16:26:20< shadowm> Okay, I think the documentation is trying to say in the most oblique fashion possible that I can't combine asan and lsan.
20160318 16:26:41< celticminstrel> That's way too close to asin.
20160318 16:28:05< shadowm> " This option only matters for linking of executables and if neither -fsanitize=address nor -fsanitize=thread is used. In that case the executable is linked against a library that overrides malloc and other allocator functions."
20160318 16:28:24< shadowm> Oh, I think I get it, the issue stems from the fact that I'm forcing the linker's hand since otherwise it can't link crap.
20160318 16:28:52< celticminstrel> mattsc: I actually had the weird idea of reimplementing the FormulaAI engine in Lua. (I won't do it because there's no real point, but I think it would be possible with minor changes to the general composite AI framework.)
20160318 16:29:21< celticminstrel> (If it's not clear, I mean the logic of the FormulaAI stages and how CAs are called - the Lua would call the formula code.)
20160318 16:30:54< mattsc> celticminstrel: sounds like an interesting exercise, but I agree that there’s no point in doing so
20160318 16:31:34< celticminstrel> For the same reason there's no real point in porting all WML tags to Lua.
20160318 16:31:56< vultraz> well, if we do, we can remove wml :P
20160318 16:32:01< mattsc> indeed
20160318 16:32:12< celticminstrel> Uh, no, vultraz, I'm pretty sure that's not the case.
20160318 16:32:21< mattsc> the ‘indeed’ was on celticminstrel’s comment, not vultraz’s  ;)
20160318 16:32:32< celticminstrel> Note, I really meant all ActionWML tags specifically. (And I guess ConditionalWML tags.)
20160318 16:32:42< celticminstrel> Not all the WML data tags.
20160318 16:33:09< celticminstrel> Speaking of ConditionalWML, I wonder if you can override "true"... or "variables"...
20160318 16:34:20< celticminstrel> The former has no use, but the latter might.
20160318 16:38:18-!- travis-ci [~travis-ci@ec2-54-204-134-202.compute-1.amazonaws.com] has joined #wesnoth-dev
20160318 16:38:19< travis-ci> wesnoth/wesnoth#8930 (master - d764bba : Charles Dang): The build passed.
20160318 16:38:19< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/116920016
20160318 16:38:19-!- travis-ci [~travis-ci@ec2-54-204-134-202.compute-1.amazonaws.com] has left #wesnoth-dev []
20160318 16:38:26< celticminstrel> Yay.
20160318 16:41:35< celticminstrel> vultraz, gfgtdf: Any objections to merging the formula PR now?
20160318 16:41:45< irker791> wesnoth: ln-zookeeper wesnoth:master e5b3d8139b11 / data/core/terrain-graphics/new-macros.cfg: Slightly improved the beach waves rules https://github.com/wesnoth/wesnoth/commit/e5b3d8139b1148c03b7c8c702513fd8a5377c795
20160318 16:42:16< gfgtdf> celticminstrel: hmm no, but i still think we should rmeove formula, or least switch $(...) to use lua.
20160318 16:44:49< shadowm> I think we should remove all mainline campaigns and MP maps.
20160318 16:48:06< zookeeper> i can't see how removing formulas could even be considered seriously unless it actually decayed into complete brokenness.
20160318 16:48:40< zookeeper> assuming that removing would mean breaking existing formulas, of course.
20160318 16:51:12< celticminstrel> Merging PR now, prepare for botspam (it's 44 commits).
20160318 16:52:12< irker791> wesnoth: Celtic Minstrel wesnoth:master 26a2638f2cce / / (4 files in 4 dirs): Make the formula and Lua views of a unit coincide more closely https://github.com/wesnoth/wesnoth/commit/26a2638f2cce0fd66dddc5b26eb9707c06c99c7d
20160318 16:52:14< irker791> wesnoth: Celtic Minstrel wesnoth:master 240b5f61a7c6 / / (3 files in 3 dirs): Output formula errors in unit filters to WML error stream https://github.com/wesnoth/wesnoth/commit/240b5f61a7c6be0f886fefb5e7b6af0348fa8fb2
20160318 16:52:16< irker791> wesnoth: Celtic Minstrel wesnoth:master 03bbc793678f / data/gui/default/widget/toggle_button_orb.cfg src/formula.cpp: Formula engine: change undocumented string substitution syntax https://github.com/wesnoth/wesnoth/commit/03bbc793678f38fbd1cd4974da6375eadf92e7a7
20160318 16:52:18< irker791> wesnoth: Celtic Minstrel wesnoth:master 1ae426c51c60 / src/ (formula.cpp formula_tokenizer.cpp variant.cpp variant.hpp): Formula engine: new concatenate and range operators https://github.com/wesnoth/wesnoth/commit/1ae426c51c6038013b8d9ba491337816d5541fd0
20160318 16:52:20< irker791> wesnoth: Celtic Minstrel wesnoth:master 0674f2a8443a / src/ (formula.cpp formula_tokenizer.cpp): Formula engine: support wfl/wflend instead of fai/faiend https://github.com/wesnoth/wesnoth/commit/0674f2a8443a32aa4344ad4f42a380216f4f1b0f
20160318 16:52:22< irker791> wesnoth: Celtic Minstrel wesnoth:master 279ced9b5545 / src/ (variant.cpp variant.hpp): Formula engine: allow lists to be indexed by a list https://github.com/wesnoth/wesnoth/commit/279ced9b5545bdeb9e3bbf509c5252baeb0c3ce8
20160318 16:52:24< irker791> wesnoth: Celtic Minstrel wesnoth:master e4b944d83be7 / src/ (variant.cpp variant.hpp): Make variant iterator bidirectional and stdlib compatible https://github.com/wesnoth/wesnoth/commit/e4b944d83be7fe133270ec4bd247477cc0336512
20160318 16:52:26< irker791> wesnoth: Celtic Minstrel wesnoth:master 1c9eb0cdbb7e / src/formula_function.cpp: Formula engine: add tan and tail functions, extra argument to head function https://github.com/wesnoth/wesnoth/commit/1c9eb0cdbb7e158cc2ad9bb76c0c6daadaf17bce
20160318 16:52:28< irker791> wesnoth: Celtic Minstrel wesnoth:master 6490b535a3e2 / src/ (ai/formula/ai.cpp menu_events.cpp variant.cpp): Tweak output of formula evaluation command prompt https://github.com/wesnoth/wesnoth/commit/6490b535a3e2c3be10e1815a24152a4d10e98bed
20160318 16:52:30< irker791> wesnoth: Celtic Minstrel wesnoth:master ad6528d13cad / src/ (formula.cpp formula.hpp): Formula engine: Support function definitions everywhere https://github.com/wesnoth/wesnoth/commit/ad6528d13cad19bcaf67d3dbba234e9203f2e770
20160318 16:52:32< irker791> wesnoth: Celtic Minstrel wesnoth:master 4aa725ac29e2 / src/formula.cpp: Formula engine: When possible, use synced RNG for the dice roll operator https://github.com/wesnoth/wesnoth/commit/4aa725ac29e2e22af652c8ffb9d022022dadfa13
20160318 16:52:34< irker791> wesnoth: Celtic Minstrel wesnoth:master fe05f51c9e3a / src/ (variant.cpp variant.hpp): Formula engine: Fix decimal values not being convertible to integers https://github.com/wesnoth/wesnoth/commit/fe05f51c9e3a3346c59186991e9247242ddad589
20160318 16:52:36< irker791> wesnoth: Celtic Minstrel wesnoth:master c1ca72f1dc2e / src/ (formula_function.cpp variant.cpp): Formula engine: fix abs() and modulus not working on decimals https://github.com/wesnoth/wesnoth/commit/c1ca72f1dc2e1255877f89100783c198866281d9
20160318 16:52:38< irker791> wesnoth: Celtic Minstrel wesnoth:master 59dd9cc6af93 / src/formula_function.cpp: Formula engine: Add optional base argument to reduce() https://github.com/wesnoth/wesnoth/commit/59dd9cc6af93431dd8e395a35eceecdaea0a3919
20160318 16:52:40< irker791> wesnoth: Celtic Minstrel wesnoth:master 2637f23afd82 / src/ (formula_function.cpp variant.cpp variant.hpp): Formula engine: New function to reverse a string https://github.com/wesnoth/wesnoth/commit/2637f23afd821034b23c63498ee9789a16c2e5b7
20160318 16:52:42< irker791> wesnoth: Celtic Minstrel wesnoth:master 390baf7d11d0 / src/formula_function.cpp: Formula engine: New function to zip a list of lists https://github.com/wesnoth/wesnoth/commit/390baf7d11d0c04af64acc6ae4c449c5bbf676e8
20160318 16:52:44< irker791> wesnoth: Celtic Minstrel wesnoth:master 159dbcf28e1e / src/formula_function.cpp: Formula engine: New take_while function https://github.com/wesnoth/wesnoth/commit/159dbcf28e1e38aed555eadc9877bc8d055c7ad3
20160318 16:52:46< irker791> wesnoth: Celtic Minstrel wesnoth:master 1cf9f616ecf1 / src/ (formula_function.cpp tests/test_formula_function.cpp): Formula engine: Substring function now accepts negative size https://github.com/wesnoth/wesnoth/commit/1cf9f616ecf1f4791e6f16f7bc68a83ea6aa9106
20160318 16:52:48< irker791> wesnoth: Celtic Minstrel wesnoth:master c589298e0f90 / src/formula_function.cpp: Formula engine: Add inverse trig functions https://github.com/wesnoth/wesnoth/commit/c589298e0f90efaae23ca72dede907a8d6ca0e48
20160318 16:52:50< irker791> wesnoth: Celtic Minstrel wesnoth:master f32b51b02219 / src/formula_function.cpp: Formula engine: Add root functions https://github.com/wesnoth/wesnoth/commit/f32b51b022191922085ea44a4cc6d1977c404d81
20160318 16:52:52< irker791> wesnoth: Celtic Minstrel wesnoth:master 523154f8071c / src/formula_function.cpp: Formula engine: Add several more math functions https://github.com/wesnoth/wesnoth/commit/523154f8071c1913049268777d54a49c1f79989c
20160318 16:52:54< irker791> wesnoth: Celtic Minstrel wesnoth:master a40ffc0351b3 / src/formula_function.cpp: Formula engine: Catch cases of NaN resulting from math functions https://github.com/wesnoth/wesnoth/commit/a40ffc0351b3a480ce46e84a78ad56f53298cd1b
20160318 16:52:56< irker791> wesnoth: Celtic Minstrel wesnoth:master 6aed97911398 / src/formula_function.cpp: Reformat the formula function classes https://github.com/wesnoth/wesnoth/commit/6aed9791139802c6e799bdf32417056aac709ddd
20160318 16:52:58< irker791> wesnoth: Celtic Minstrel wesnoth:master bb0d00f34864 / src/ (formula_function.cpp tests/test_formula_function.cpp variant.cpp variant.hpp): Formula engine: Fix rounding issues with math functions https://github.com/wesnoth/wesnoth/commit/bb0d00f34864cf224fde4b29cfc214a56ce7bc86
20160318 16:53:00< irker791> wesnoth: Celtic Minstrel wesnoth:master b239dd5fd109 / src/formula_function.cpp: Formula engine: New string functions find_string and replace https://github.com/wesnoth/wesnoth/commit/b239dd5fd109a03a8b2feb4263801df3040fe7da
20160318 16:53:02< irker791> wesnoth: Celtic Minstrel wesnoth:master 38f6e8c4f127 / src/ (formula.cpp formula_tokenizer.cpp variant.cpp): Formula engine: Add string escapes mechanism https://github.com/wesnoth/wesnoth/commit/38f6e8c4f127ad8b1c3ea2bbed2f05ea22578eae
20160318 16:53:04< irker791> wesnoth: Celtic Minstrel wesnoth:master 0b84677ee91a / src/variant.cpp: Formula engine: Lists can be indexed with negative numbers https://github.com/wesnoth/wesnoth/commit/0b84677ee91af528cffc781e58e24b32412998a1
20160318 16:53:06< irker791> wesnoth: Celtic Minstrel wesnoth:master a4469210d425 / src/formula.cpp: Formula engine: fix exponentiation being left-associative https://github.com/wesnoth/wesnoth/commit/a4469210d425b5e6e649d662d4272c8a6fb85604
20160318 16:53:08< irker791> wesnoth: Celtic Minstrel wesnoth:master 4a3fce393a8b / src/formula.cpp: Formula engine: fix index operator having lowest precedence https://github.com/wesnoth/wesnoth/commit/4a3fce393a8b0eebdb426c2db12347a6884400e8
20160318 16:53:10< irker791> wesnoth: Celtic Minstrel wesnoth:master de4186c47b17 / src/ (formula_function.cpp variant.cpp variant.hpp): Formula engine: add type() function https://github.com/wesnoth/wesnoth/commit/de4186c47b17ee601944c37373d03e6912aeb418
20160318 16:53:12< irker791> wesnoth: Celtic Minstrel wesnoth:master 6da744a37a54 / src/ (formula.cpp variant.cpp): Formula engine: add empty map literal https://github.com/wesnoth/wesnoth/commit/6da744a37a54e5043d175329aedc577bbf6f476c
20160318 16:53:14< irker791> wesnoth: Celtic Minstrel wesnoth:master b53ee994c98c / src/variant.cpp: Formula engine: now exponentiation returns null on NaN https://github.com/wesnoth/wesnoth/commit/b53ee994c98cfabd8ff6ff67bffb5cc79cd79d7a
20160318 16:53:16< irker791> wesnoth: Celtic Minstrel wesnoth:master e78b5d7bafc5 / src/variant.cpp: Formula engine: fix comparison trying to cast everything to decimal https://github.com/wesnoth/wesnoth/commit/e78b5d7bafc5d795431256377fbc8f6a06603599
20160318 16:53:18< irker791> wesnoth: Celtic Minstrel wesnoth:master f6be40bb71af / / (6 files in 3 dirs): Transfer original formula unit tests to Boost framework https://github.com/wesnoth/wesnoth/commit/f6be40bb71aff2a42acfea4950fadf97666c2384
20160318 16:53:20< irker791> wesnoth: Celtic Minstrel wesnoth:master a9a29a11ccf8 / src/tests/test_formula_function.cpp: Add a few more formula math tests https://github.com/wesnoth/wesnoth/commit/a9a29a11ccf813e210df008fcfbc7890e9da6b6b
20160318 16:53:22< irker791> wesnoth: Celtic Minstrel wesnoth:master 928ccbfc06da / src/ (formula_function.cpp tests/test_formula_function.cpp): Formula engine: fix max() and min() not working on decimals https://github.com/wesnoth/wesnoth/commit/928ccbfc06da2795a3a57eede424c0d78fead4ce
20160318 16:53:24< irker791> wesnoth: Celtic Minstrel wesnoth:master f0abbf111149 / src/tests/test_formula_core.cpp: Add a few more formula string tests https://github.com/wesnoth/wesnoth/commit/f0abbf1111495ae2a3f6d93150bd1dd61c9ab7ff
20160318 16:53:26< irker791> wesnoth: Celtic Minstrel wesnoth:master e814bb48be96 / src/formula.cpp: Formula engine: Allow dot notation on strings and maps https://github.com/wesnoth/wesnoth/commit/e814bb48be967c9789900ab1522adae01589691d
20160318 16:53:28< irker791> wesnoth: Celtic Minstrel wesnoth:master 50c42caf1129 / src/ (formula.cpp formula_tokenizer.cpp variant.cpp variant.hpp): Formula engine: Add "in" operator for map/list containment https://github.com/wesnoth/wesnoth/commit/50c42caf11290f4c95909439eb3d48e2d6aa95e5
20160318 16:53:30< irker791> wesnoth: Celtic Minstrel wesnoth:master 2e412df9b948 / src/ (ai/formula/function_table.cpp formula_function.cpp): Formula_engine: make distance_between function a core function https://github.com/wesnoth/wesnoth/commit/2e412df9b94883997edf00ef7944f6351d76045c
20160318 16:53:32< irker791> wesnoth: Celtic Minstrel wesnoth:master 69024df4f53f / src/formula_function.cpp: Formula engine: Allow two-argument if() - returns null if false https://github.com/wesnoth/wesnoth/commit/69024df4f53f804839db36128249fd4dd3b20075
20160318 16:53:34< irker791> wesnoth: Celtic Minstrel wesnoth:master a02b1c13a1c2 / src/ (6 files in 2 dirs): Formula engine: Improve debugger https://github.com/wesnoth/wesnoth/commit/a02b1c13a1c2da2076c937516c09c9f58722b806
20160318 16:53:36< irker791> wesnoth: Celtic Minstrel wesnoth:master f8476368d323 / src/formula_function.cpp: Formula engine: debug_print() now shows in chat if debug mode on https://github.com/wesnoth/wesnoth/commit/f8476368d323b3c71a75dc4b3c55a2ac08f524c0
20160318 16:53:38< irker791> wesnoth: Celtic Minstrel wesnoth:master 229e8881ee17 / changelog: Update changelog https://github.com/wesnoth/wesnoth/commit/229e8881ee1759cf8d50480c56a5697f521090f5
20160318 16:53:40< irker791> wesnoth: Celtic Minstrel wesnoth:master bb510a58af3e / / (26 files in 9 dirs): Merge pull request #624 from CelticMinstrel/lua_formula_bridge https://github.com/wesnoth/wesnoth/commit/bb510a58af3e1b01d96c0c72b87ff1d4277f59b3
20160318 16:53:46< celticminstrel> And it's over.
20160318 16:53:50< celticminstrel> Sorry about that.
20160318 16:55:28< zookeeper> i've been musing on some kind of third changelog dedicated solely for UMC compatibility affecting changes. core unit stat changes, WML API changes (not additions), removed images, etc. i'm just not sure what format it should take. i guess even just an informal bullet point list on the forums would be quite sufficient.
20160318 16:55:32< mattsc> celticminstrel: how much are you itching to get the AI refactor merged?
20160318 16:55:46< mattsc> Are you okay to wait a few more days?
20160318 16:56:21< celticminstrel> Yeah, that should be fine. I still have the [params] -> [args] change to make, a unit test to fix, and probably several more unit tests to make.
20160318 16:56:25< mattsc> zookeeper: there are topics like that already, aren’t there?
20160318 16:56:53< zookeeper> mattsc, well, yeah that's true, but i think only one for WML API changes
20160318 16:57:18< shadowm> zookeeper: Good luck getting people to maintain even a single changelog properly.
20160318 16:57:19< mattsc> zookeeper: sounds like a good idea to me
20160318 16:57:29< mattsc> in theory
20160318 16:58:05< mattsc> celticminstrel: okay; I’m just completely swamped by RL at the moment.  And I think I should go back to master and fix all the problems with the MAIs that have nothing to do with your PR first.
20160318 16:58:09< celticminstrel> Hey, I think I've been pretty good at maintaining the changelog for my own stuff.
20160318 16:58:12< zookeeper> shadowm, i know. but the number of people that need squeaky-malleting due to forgetting to log their changes is small
20160318 16:58:23< celticminstrel> mattsc: Like the Lua API changes? That might help a bit, yeah.
20160318 16:58:38< shadowm> zookeeper: It's not sufficiently smaller than the number of active devs.
20160318 16:58:49< celticminstrel> mattsc: Then I could rebase onto master and you (or I) could fix the remaining issues.
20160318 16:59:10< mattsc> celticminstrel: That; amnd things like MP map dimensions changing which are used for test scenarios, and probably some stuff I have not found yet.
20160318 16:59:30< shadowm> I facetiously pointed out some time ago that the 1.12 branch had no bug fixes since 1.12.5. Guess what the commit log actually says.
20160318 17:00:10< shadowm> I presume the culprits are operating on the assumption that they won't forget or be hit by a bus tomorrow, but it's still bad form.
20160318 17:00:15< mattsc> celticminstrel: yes; I also thinnk it would make it easier to decide whether any potential problem from the refactoring might only be an issue for the MAIs, or a more general problem.
20160318 17:00:30< mattsc> Not impossible to do all that together, of course, but I think this would be cleaner.
20160318 17:00:50< celticminstrel> I do see a couple of fixes after 1.12.5, yeah.
20160318 17:01:03< celticminstrel> mattsc: Makes sense.
20160318 17:01:20< zookeeper> anyway, even properly updated changelogs aren't really a substitute for a single stable-to-next-stable list of compatibility changes. and of course that kind of list can be updated anytime, whereas the regular changelogs need to be updated before release.
20160318 17:01:37< mattsc> celticminstrel: I’ll do that then.  I think I’ll have some time this weekend, but probably not as much as I would like.
20160318 17:01:49< zookeeper> i should lead by example and start a 1.12->1.13/1.14 thread myself
20160318 17:01:51< gfgtdf> zookeeper: well, in most cases we know, people use only very simple formulas like '$($a/$b + 6)' whihc would still work the same way if it was switched to use lua instead. In all oother cases i've seen in umc porting the code to lua would be quiet easy. The advantage of using lua instead is that 1) more people know lua so more people would be able to make use of $(...) and that 2) lua can do...
20160318 17:01:52< gfgtdf> ...more things, specailly using the whole lua api which allows. things like reading files, calculating shortest paths, accessing the gamestate and calling user defined lua functions.
20160318 17:02:05< shadowm> zookeeper: I proposed such a list around the time we were working on the 1.12 announcement, actually.
20160318 17:02:09< celticminstrel> zookeeper: I was planning on putting announcements about any compatibility breakages (eg in the AI branch) into RELEASE_NOTES.
20160318 17:02:32< shadowm> You were there, I don't remember what your opinion was.
20160318 17:03:22< celticminstrel> gfgtdf: Formula has a shortest_path function already.
20160318 17:03:33< celticminstrel> gfgtdf: It's currently restricted to FormulaAI, but could easily be made global.
20160318 17:03:42< celticminstrel> gfgtdf: I already did that with the distance_between function.
20160318 17:03:47< zookeeper> shadowm, something other than what ended up becoming https://forums.wesnoth.org/viewtopic.php?f=21&t=36057 ?
20160318 17:04:40< shadowm> zookeeper: I guess you updated it then.
20160318 17:04:42< celticminstrel> gfgtdf: I'm also working on something that could allow formulas to call functions implemented in Lua.
20160318 17:05:26< shadowm> zookeeper: So what are you suggesting again and how is it different to the solution that we are already purportedly using?
20160318 17:06:28< celticminstrel> Come to think of it, a lot of the FormulaAI functions could be useful outside the AI ... almost all the ones that return anything other than AI data or actions.
20160318 17:06:53< zookeeper> shadowm, i'm not really suggesting anything fundamentally different from that thread. mainly i'd just add some more sections (such as image removals) and try to keep an eye on changes that should be added to the list and then squeaky-mallet people when they forget.
20160318 17:08:09< zookeeper> shadowm, so, yeah, no help or input needed from anyone really, i can start it myself whenever i please
20160318 17:08:31< shadowm> I hope you don't mind having grey hair.
20160318 17:09:23-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160318 17:09:23-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Read error: Connection reset by peer]
20160318 17:09:55-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has joined #wesnoth-dev
20160318 17:10:28< zookeeper> yeah, i expect that when i try to mallet some people who don't think their change needs to be added, they'll just tell me to do it myself and then suddenly all of it becomes my job and either i have to do it myself or let the thing rot
20160318 17:10:49< shadowm> Even things intended to make people's lives easier tend to get ignored a lot around here unless you are very in-your-face about them.
20160318 17:10:59< vultraz> please not another changelog
20160318 17:11:14< shadowm> "Hey all! GitHub has labels and assignments and stuff for PRs, let's use them!" "<crickets>"
20160318 17:11:37< vultraz> Yeah, that backfires
20160318 17:11:50< vultraz> People don't work on what you assign them if they don't want it anyway
20160318 17:11:52< zookeeper> could we add metal spikes to the squeaky mallet and coat it with poison?
20160318 17:12:21< shadowm> vultraz: Assigning people was intended to prevent devs from stepping on each others' toes.
20160318 17:12:45< shadowm> _And_ if someone no longer wants a PR/bug they should obviously not do the mordante thing and hoard it for several years.
20160318 17:13:14< shadowm> People on the receiving end are supposed to visibly complain about these cases too.
20160318 17:13:34< zookeeper> ...now i want to see a secret magical mallet that a unit can get and which would make a funny squeaky sound :/
20160318 17:14:06-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has quit [Ping timeout: 246 seconds]
20160318 17:14:14 * celticminstrel applies the squeaky mallet to uh... *rolls die* vultraz.
20160318 17:14:36< shadowm> A lot of the time it's just people believing that they'll remember to update the changelog 1 hour before the release is tagged.
20160318 17:15:07< shadowm> I don't mean to be fatalistic, but I've always used the "hit by a bus" rule of thumb for a reason.
20160318 17:15:38< celticminstrel> I think I've seen people to that, but yeah, updating it right after the change is much better.
20160318 17:15:43< shadowm> It's supposed to illustrate that no matter how much in control you think you are, there is always a possibility that something or someone will prevent you from doing tomorrow what you could've done today.
20160318 17:15:53< celticminstrel> ^do
20160318 17:17:23< shadowm> It could be anything from a traffic jam to a schedule conflict to someone outright running you over with a motorized vehicle.
20160318 17:17:43< vultraz> 45 issues still assigned to mordante
20160318 17:18:08< shadowm> Or just you deciding "nope I'm done with this PoC, I'm out".
20160318 17:19:05< celticminstrel> That thread zookeeper listed is actually quite useful.
20160318 17:19:09< celticminstrel> ^linked
20160318 17:19:34< celticminstrel> I agree there should definitely be one for 1.12->1.14. Dunno if it matters for 1.12->1.13 though.
20160318 17:19:54< shadowm> Yes, yes it does.
20160318 17:20:02< celticminstrel> 'kay
20160318 17:20:10< shadowm> The last known state of 1.13.x becomes 1.14.0.
20160318 17:20:32< shadowm> You probably don't want to spend the last 30 days leading up to 1.14.0 figuring out what needs to go into such a changelog.
20160318 17:20:46< celticminstrel> So zookeeper is taking responsibility for creating it at some point?
20160318 17:22:48< zookeeper> let's say that i was acknowledging that i won't demand anyone else to do it or complain that it hasn't been done as long as i can do it myself
20160318 17:23:07< zookeeper> which hopefully leads me to doing it sometime soon
20160318 17:25:57< vultraz> ancient fr from zookeeper https://gna.org/bugs/index.php?5821 still valid?
20160318 17:27:06-!- louis94 [~~louis94@91.178.242.151] has joined #wesnoth-dev
20160318 17:28:16< vultraz> note to self: implement smooth scrolling in GUI2
20160318 17:28:57< shadowm> Please no. :\
20160318 17:29:22< vultraz> please yes
20160318 17:29:24< shadowm> Or at least make it an option. Smooth scrolling is actually painful for me.
20160318 17:29:30< vultraz> how so?
20160318 17:29:39< shadowm> It makes my head hurt.
20160318 17:29:50< vultraz> huh
20160318 17:30:36< shadowm> That's why the first thing I do after installing Firefox in a VM is to disable smooth scrolling.
20160318 17:31:43< celticminstrel> Smooth scrolling means...?
20160318 17:32:13< celticminstrel> The scrollpane contents can move in pixel increments?
20160318 17:32:34< vultraz> yeah
20160318 17:32:38 * celticminstrel would consider something like that to be very low priority.
20160318 17:32:39< vultraz> as opposed to jumps
20160318 17:32:52< vultraz> jump scrolling is outdated
20160318 17:32:58< celticminstrel> There is absolutely nothing wrong with jumps. (Unless they're ridiculously large.)
20160318 17:32:59< vultraz> low prio, yes 
20160318 17:33:04< shadowm> No, it isn't oudated.
20160318 17:33:05< vultraz> well, no
20160318 17:33:09< vultraz> but it's not the New Thing 
20160318 17:33:40< shadowm> You know what else is the New Thing? Freemium.
20160318 17:35:39< shadowm> Oh, and smartwatches.
20160318 17:36:09< celticminstrel> Watch Quest almost (but not quite) makes me want to get one of those. (An Apple Watch, that is.)
20160318 17:36:14< shadowm> But I think I could suggest making a smartwatch port of Wesnoth and you'd take the suggestion totally seriously.
20160318 17:36:25< celticminstrel> Haha.
20160318 17:36:55< vultraz> I would
20160318 17:37:02< vultraz> Smartwatches are awesome
20160318 17:37:04< vultraz> I need to get one
20160318 17:37:06< shadowm> You have no shame.
20160318 17:37:08< celticminstrel> That's impossible.
20160318 17:37:12< shadowm> (That's not a good thing.)
20160318 17:37:14< celticminstrel> Wesnoth on a smartwatch, I mean.
20160318 17:38:28< vultraz> that's true
20160318 17:38:38< shadowm> People said humans couldn't fly either.
20160318 17:38:49< zookeeper> vultraz, hrhrhhm. i guess i should have the knowhow to either implement that or find out whether it's a fundamentally too complicated way to go about it.
20160318 17:38:54< shadowm> Or get to the moon.
20160318 17:39:27< shadowm> Give a delusional person a sufficiently large budget and time and you'd be surprised what they can do with it.
20160318 17:39:32< vultraz> there's just not enough space on a watch for games
20160318 17:39:34< vultraz> however
20160318 17:39:47< vultraz> if we combined smartwatches with VR tech...
20160318 17:41:00< shadowm> But more seriously, I don't have a problem with smooth scrolling in e.g. the game map because I'm not _reading_ it.
20160318 17:41:03-!- midzer [~quassel@p5B312EF2.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds]
20160318 17:41:12< zookeeper> vultraz, well, i assigned it to myself until i decide.
20160318 17:41:22< shadowm> It's when I'm reading substantial amounts of text that it actually hurts me.
20160318 17:43:16< shadowm> So it probably wouldn't affect me in the Recruit list, but it wouldn't be great to see in a control that displays large chunks of text like the Help browser's right area or an unusually large message box.
20160318 17:43:59< shadowm> Or the Gamestate Inspector or a log report element of some sort.
20160318 17:44:05< gfgtdf> does anyoe know how i prevent c++ form doing macro expansions in macro parmeters?
20160318 17:44:16< gfgtdf> i want to make strign weitch macro
20160318 17:44:22< gfgtdf> what ui currently have looks liek this: http://pastebin.com/ETqnCsGx
20160318 17:44:46< gfgtdf> te problme is that i cannot use comma expression inside case paremter for the macro
20160318 17:45:30< celticminstrel> gfgtdf: If you want a comma inside a macro parameter, surround the parameter in parentheses.
20160318 17:45:41< zookeeper> so smooth scrolling is when a scrolling happens not in one instantaneous jump, but by scrolling that same distance pixel by pixel in a short moment?
20160318 17:45:44< gfgtdf> hmmsry this is the corret paste: http://pastebin.com/zBJmdqrH
20160318 17:45:48< gfgtdf> celticminstrel: i i tried that
20160318 17:46:42< zookeeper> if so, then i think i mildly dislike smooth scrolling it because i feel like i'm not in control during the scroll. it's like if i pressed a button and am stuck watching a pointless animation of the computer completing the task i gave it.
20160318 17:46:44< gfgtdf> celticminstrel: if i simply try addn parensis i get cmpile erorrs: http://pastebin.com/mkcgKFKT
20160318 17:46:50< celticminstrel> Hmm, it's possible it only works if parentheses would actually be valid in that situation.
20160318 17:47:15< celticminstrel> Ah, I see your problem.
20160318 17:47:28< celticminstrel> I assume it doesn't work with braces either.
20160318 17:47:45< celticminstrel> I think I would suggest a different approach to this.
20160318 17:48:22< celticminstrel> An approach that allows the code to go after the macro call instead of inside it.
20160318 17:49:13< gfgtdf> celticminstrel: when i try to remove the brackets inside the STRING_SWITCH mactos  liek this i again have teh problem with the commas http://pastebin.com/KDN3LAYy
20160318 17:49:28< gfgtdf> celticminstrel: hmm the problem is taht i need to have code for each case
20160318 17:49:52< gfgtdf> celticminstrel: so i cannot put the code after teh macro liek boost_foerach or similar thing do it
20160318 17:49:54 * celticminstrel has to go suddenly, back in ~10-15 minutes.
20160318 18:03:18< celticminstrel> Back.
20160318 18:03:59< celticminstrel> Okay, so you are, in fact, switching on the hash.
20160318 18:04:26-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160318 18:04:35< celticminstrel> gfgtdf: Two things. 1) I would suggest making the macro end with the "switch(itor->second)".
20160318 18:05:08-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160318 18:05:53< celticminstrel> 2) Have a second macro that expands to the appropriate case label.
20160318 18:06:38< gfgtdf> celticminstrel: the problem is that the second macro wont know then about which sstrings are asucoiates to which hashes/indexes
20160318 18:06:39< celticminstrel> Ah, I see, you're not actually switching on the hash...?
20160318 18:06:59< celticminstrel> gfgtdf: How does it currently know about that?
20160318 18:07:32< gfgtdf> celticminstrel: well i afaik cannot even calculate the hash of a strign at compiletime to use in for a case.
20160318 18:07:49< celticminstrel> You probably can if your compiler supports constexpr.
20160318 18:08:20< gfgtdf> celticminstrel: well it has a hashmap tha map the strign value of the case number i to i and than switches on i
20160318 18:09:26< celticminstrel> gfgtdf: I wonder if case switch_string_static_map["aaa"]: can work.
20160318 18:09:38< celticminstrel> Possibly with added const or constexpr.
20160318 18:09:46< celticminstrel> (So I guess use .at() instead of [])
20160318 18:09:59< celticminstrel> Oh wait, there's still a problem though.
20160318 18:10:00< celticminstrel> Hmm.
20160318 18:10:22< gfgtdf> celticminstrel: hmm in currently traing to macro varargs to fix teh comma issue in theucrrent implementation
20160318 18:10:25< gfgtdf> celticminstrel: seems to work
20160318 18:10:41< celticminstrel> Yeah, okay, varargs macros should indeed do the trick.
20160318 18:11:00< celticminstrel> There's two ways to do that - C99 style and GNU style.
20160318 18:11:18< celticminstrel> I don't actually remember what the GNU style is anymore.
20160318 18:12:17-!- Kwandulin [~Miranda@p200300760F7D7365959E83CD65929EDC.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160318 18:18:20-!- louis94 [~~louis94@91.178.242.151] has quit [Ping timeout: 250 seconds]
20160318 18:25:53< gfgtdf> celticminstrel: i'll try whether this spped up impl_unit_get
20160318 18:26:18< celticminstrel> Is impl_unit_get currently noticably slow?
20160318 18:26:35< celticminstrel> ^noticeably
20160318 18:26:55< celticminstrel> There are also zillions of other similar cases littered through the code, both in the Lua API and in the formula API.
20160318 18:28:26< celticminstrel> I guess impl_unit_get is kind of inefficient right now since it might have to compare against quite a few strings before finding a match.
20160318 18:28:29< gfgtdf> celticminstrel: hmm didnt text but it has ~40 possible atributes and do if you want to quers the last attribute it'd need to do 40 string comparisions before returning valuie
20160318 18:28:37< gfgtdf> test*
20160318 18:28:50< celticminstrel> gfgtdf: Worst case is when you try to query an attribute that doesn't exist. :)
20160318 18:29:04< gfgtdf> celticminstrel: actually > 50 if you coint the function like u:to_map
20160318 18:29:07< gfgtdf> count*
20160318 18:29:09< celticminstrel> gfgtdf: Also, don't forget that there are function attributes, which are dealt with in a separate function.
20160318 18:29:15< celticminstrel> And just as I say that, you mention it too.
20160318 18:29:43< celticminstrel> There's no particular reason that I know of to prevent those from being moved into the same function.
20160318 18:33:50< celticminstrel> BTW gfgtdf, should I take this to mean that you now have C++11 support?
20160318 18:34:51< gfgtdf> celticminstrel: hmm no still using msvc 2010 which has very limitedc++11 support
20160318 18:35:04< celticminstrel> But you're using unordered_map there?
20160318 18:35:12< gfgtdf> celticminstrel: boost::unordered_map
20160318 18:35:15< celticminstrel> Ah, right.
20160318 18:35:24 * celticminstrel clearly wasn't paying enough attention. >_>
20160318 18:40:23< zookeeper> surprising: even with SSF's and all, it seems impossible to write a SUF which checks whether there's enemy units within radius of >1.
20160318 18:40:42< zookeeper> units that are enemies of that unit, that is.
20160318 18:41:18< celticminstrel> That is kind of surprising.
20160318 18:41:44< zookeeper> of course you can probably do it with lua, but still
20160318 18:42:04< celticminstrel> Or [store_unit] and find_in= or something.
20160318 18:42:30< zookeeper> "a SUF" was the key part here :p
20160318 18:42:55< celticminstrel> Well, that does involve a SUF (or two of them, I guess).
20160318 18:43:11< zookeeper> ...that's not even remotely a reasonable interpretation
20160318 18:44:01< celticminstrel> I was proposing storing all enemy units in an array and then using the find_in= key in the SUF to filter to units in that array... but that's probably not very efficient.
20160318 18:44:18-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20160318 18:44:27< zookeeper> and i said "a SUF" was the key part of what i was saying.
20160318 18:44:43< celticminstrel> I guess you're trying to say that singularity is part of the key part.
20160318 18:46:17< zookeeper> that's exactly how it was worded
20160318 18:47:37< zookeeper> my point is that "to write a SUF which checks whether there's enemy units within radius of >1" is very unambiguous
20160318 18:47:50< zookeeper> there is no way it can be interpreted as something which allows for [store_unit]
20160318 18:48:02< zookeeper> well, no reasonable way
20160318 19:02:07-!- louis94 [~~louis94@91.178.242.151] has joined #wesnoth-dev
20160318 19:05:02-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160318 19:28:41-!- mjs-de [~mjs-de@x4db62f58.dyn.telefonica.de] has joined #wesnoth-dev
20160318 19:31:05< zookeeper> i take it that it's not possible to use lua_function= (in a SUF) with all the code inlined right there and then? the function must have been defined elsewhere beforehand?
20160318 19:32:42< celticminstrel> Yup.
20160318 19:32:49< celticminstrel> That's why formula= is still cleaner.
20160318 19:33:18< celticminstrel> Also, the function has to be defined as a global function.
20160318 19:34:52-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160318 19:35:14< zookeeper> yeah...
20160318 19:37:18< gfgtdf> celticminstrel: hmm i tested and the string_wtch implementation seems to be slower than the current implementation
20160318 19:37:31< gfgtdf> but i don't understand why
20160318 19:39:31-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 268 seconds]
20160318 19:54:00-!- irker791 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160318 19:54:36< zookeeper> i can only presume that this breaks awfully in scenarios with multiple sides, but in a 2-side scenario it seems to work? :x http://pastebin.com/TyK76CGx
20160318 19:54:55< zookeeper> i can't really replicate the thought process that went into that, and it's probably hideously overcomplicated for the simple 2-side usecase
20160318 19:56:59< zookeeper> oh, yeah, it checks for units of all other sides, not just enemies.
20160318 19:57:38< mattsc> celticminstrel: was it you who changed something about messages?
20160318 20:02:06< celticminstrel> I would definitely say that I changed something about messages, but I don't think I was the only one.
20160318 20:03:42< mattsc> celticminstrel: The messages in the MAI test scenarios don’t have portraits and I am trying to figure out why.
20160318 20:04:14< mattsc> On a different topic, when you start the “normal” test scenario from the CL, there’s a put_unit() error message.
20160318 20:06:11< celticminstrel> An error, or a deprecation message?
20160318 20:06:23< celticminstrel> How are the portraits specified?
20160318 20:06:32< celticminstrel> (Or are they just default ones?)
20160318 20:06:36-!- louis94 [~~louis94@91.178.242.151] has quit [Quit: Konversation terminated!]
20160318 20:06:42< mattsc> Default units.
20160318 20:06:56< celticminstrel> So no image= in the [message]?
20160318 20:07:04< mattsc> No error message at all, and the units are highlighted correctly.  Just no portrait.
20160318 20:07:21< celticminstrel> That comment was referring to the put_unit thing.
20160318 20:08:01< mattsc> Oh.  Error.
20160318 20:08:40< mattsc> As for the messages, that’s when using the MESSAGE macro with “” for the image.
20160318 20:09:19< celticminstrel> Ah. That could be it then.
20160318 20:09:51< celticminstrel> You might even be able to fix that yourself.
20160318 20:09:54< mattsc> For the messages for which an image is specified, it shows up correctly.
20160318 20:10:06< celticminstrel> It's implemented at something like data/lua/wml/wml-tags/message.lua
20160318 20:10:19< celticminstrel> The get_image function is the relevant place.
20160318 20:10:33-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160318 20:11:18< mattsc> Ugh.  Ok.  Maybe I’ll look into that later.  Right now my lunch break is over.
20160318 20:12:16< mattsc> Btw, the same happens with () in place of “”.
20160318 20:13:01< celticminstrel> Yeah, I'm guessing the problem is that it's set to an empty string rather than not set at all.
20160318 20:14:49< mattsc> Yeah, that’s certainly the behavior, since everything else is working.
20160318 20:15:24-!- irker727 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160318 20:15:24< irker727> wesnoth: mattsc wesnoth:master b347d2b2d6ab / data/ai/micro_ais/scenarios/ (scenario_micro_ai.cfg simple_attack.cfg): Micro AI scenarios: adapt coordinates to changed MP map https://github.com/wesnoth/wesnoth/commit/b347d2b2d6ab7b98e385d68b085af13baea0bb33
20160318 20:17:14< mattsc> celticminstrel: can you reproduce the error in the test scenario?
20160318 20:17:29< celticminstrel> I'm not sure. I can try.
20160318 20:17:40< celticminstrel> Last time I tried the test scenario, Wesnoth crashed. >_>
20160318 20:18:13< mattsc> It starts for me, but there are put_unit error messages right at start-up.
20160318 20:19:32< mattsc> This is on master, I’m not in the ai branch any more.
20160318 20:20:17< celticminstrel> I gathered, based on the fact that you just pushed a commit. :P
20160318 20:20:38< mattsc> :)
20160318 20:21:38< fabi> mattsc: What are this scenarios used for? Are they tests?
20160318 20:21:51< mattsc> Hi fabi
20160318 20:21:57< fabi> hi mattsc
20160318 20:22:06< mattsc> Yeah, they are just test scenarios for the Micro AIs.
20160318 20:22:17< fabi> are they non interactive?
20160318 20:22:52< mattsc> Some are, some are not.  But they are not suitable for unit tests, if that’s what you’re aming at.  :)
20160318 20:23:07< fabi> yes
20160318 20:23:12< mattsc> fabi: this is how you get to them:  https://wiki.wesnoth.org/Micro_AIs#Micro_AI_Test_and_Demo_Scenarios
20160318 20:23:14< fabi> you read my mind
20160318 20:24:04< mattsc> :D  I just saw you (plural you, but I think you were involved also) talk about unit tests very recently. 
20160318 20:24:08< fabi> mattsc: Your work is mostly lua not c++ , right?
20160318 20:24:20< mattsc> Yeah, pretty much entirely.
20160318 20:24:36< mattsc> I’ve done the occasional change to C++ code, but always by copying it from somewhere else.
20160318 20:24:56< mattsc> I really don’t know C++ very much at all.
20160318 20:25:01< fabi> Well, I am searching for unit tests to confirm that my engine is indeed a working Wesnoth clone.
20160318 20:25:23< fabi> But AI is somehow beyond my scope.
20160318 20:25:36< fabi> I am coding the server, AI is a client.
20160318 20:25:53< mattsc> The problem with the MAI test scenarios (for that purpose) is that they don’t really have a determinstic outcome.
20160318 20:26:02< fabi> Still a client that plays on my server without the need to have a full graphical client would be cool.
20160318 20:26:49< mattsc> Yeah; I don’t think I can help you with that.
20160318 20:40:38< mattsc> celticminstrel: the code that produces the put_unit error looks good to me and I confirmed that the two variables containing the location corredinates are defined, but I get an “invalid location” error anyway.
20160318 20:40:53< mattsc> Now I’m really done for now though.
20160318 20:41:30< mattsc> corredinates?  That’s a good one.
20160318 20:44:19-!- aidanhs [~aidanhs@81.4.110.234] has quit [Ping timeout: 248 seconds]
20160318 20:44:43-!- aidanhs [~aidanhs@81.4.110.234] has joined #wesnoth-dev
20160318 20:50:08< celticminstrel> Okay, that looks like the same issue I had with erase_unit().
20160318 20:50:57< celticminstrel> Actually, it might be slightly different, but caused by the same commit(s) at least.
20160318 20:51:39< celticminstrel> It's basically a typo. :/
20160318 20:52:05< celticminstrel> I put 1 where I should have put 2.
20160318 20:58:12< celticminstrel> If anyone wants to fix it immediately, it's game_lua_kernel.cpp, lines 2313 and 2315, but I'll probably commit a fix later today.
20160318 21:04:53< zookeeper> heh. when i really arrange surrounding (core) terrains unoptimally, i can get one hex to have at most 33 terrain images drawn on it.
20160318 21:05:11< celticminstrel> Wow.
20160318 21:05:27< zookeeper> 34!
20160318 21:05:37< zookeeper> 36 O.o
20160318 21:06:18< zookeeper> 38! i spoke way too soon
20160318 21:09:14-!- louis94 [~~louis94@91.178.242.151] has joined #wesnoth-dev
20160318 21:10:20< zookeeper> https://dl.dropboxusercontent.com/u/63964618/wesnoth/38_terrain_images.png
20160318 21:10:32< zookeeper> i guess that's a pretty rare case
20160318 21:11:57< celticminstrel> Wow.
20160318 21:12:23< celticminstrel> Yeah, trees don't normally go in those places. Nor does light or windmills.
20160318 21:13:19< zookeeper> well, to be fair you could do it with a much more monotonous terrains
20160318 21:13:34< zookeeper> no need to use specifically light, windmills or forests, any single kind of overlay would do
20160318 21:14:17< celticminstrel> Well, that doesn't really change it though - most of those base terrains normally do not have overlays.
20160318 21:14:30< celticminstrel> The windmill is on mountains though, right? So that one's an exception.
20160318 21:14:35< zookeeper> wow, i got one further, to 39
20160318 21:16:08< zookeeper> 40 \o/ now i'm satisfied
20160318 21:18:14< celticminstrel> XD
20160318 21:21:03< zookeeper> not a completely unreasonable arrangement: https://dl.dropboxusercontent.com/u/63964618/wesnoth/40_terrain_images.png (lightbeams on all surrounding hexes)
20160318 21:25:24< celticminstrel> Yeah, okay, that one is something that might actually occur in a scenario, though lightbeams probably aren't used much on non-cave terrains...
20160318 21:25:24< celticminstrel> Is that a volcano in the middle?
20160318 21:25:24< zookeeper> no, tropical water with a campfire
20160318 21:25:24< celticminstrel>  Okay, that part is weird then. :P
20160318 21:25:24-!- mattsc_ [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20160318 21:25:24< zookeeper> well it can just as well be gray deep if you'd prefer :p
20160318 21:25:24< celticminstrel> The weird part is that a campfire is on water.
20160318 21:25:24< zookeeper> well, any overlay will work
20160318 21:25:24< zookeeper> rubble, lilies, lightbeam, anything
20160318 21:25:25-!- heirecka [~heirecka@exherbo/developer/heirecka] has quit [Excess Flood]
20160318 21:25:39-!- frainz [~Frainz@mmisc.de] has quit [Ping timeout: 276 seconds]
20160318 21:25:56-!- frainz [~Frainz@mmisc.de] has joined #wesnoth-dev
20160318 21:26:15< zookeeper> ahhh. if i put a lit mushroom grove on top of gray/tropical water, i get 41
20160318 21:26:18-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Ping timeout: 276 seconds]
20160318 21:26:18-!- {V} [~V@105-70-ftth.on.nl] has quit [Ping timeout: 276 seconds]
20160318 21:26:18-!- Jetrel_bot [~Jetrel@ec2.happyspork.com] has quit [Ping timeout: 276 seconds]
20160318 21:26:18-!- mattsc_ is now known as mattsc
20160318 21:26:35-!- {V} [~V@105-70-ftth.on.nl] has joined #wesnoth-dev
20160318 21:26:55-!- Jetrel_bot [~Jetrel@ec2.happyspork.com] has joined #wesnoth-dev
20160318 21:27:06-!- heirecka [~heirecka@exherbo/developer/heirecka] has joined #wesnoth-dev
20160318 21:30:41< zookeeper> 42. this is a great waste of time.
20160318 21:31:47< celticminstrel> Heh.
20160318 21:32:14< celticminstrel> Okay, if you replaced the campfire with lilies, your 40-image one seems like a plausible thing that might occur in a scenario.
20160318 21:34:42 * zookeeper stops
20160318 21:57:39< mattsc> celticminstrel: so the offending line is this one: https://github.com/wesnoth/wesnoth/blob/master/data/lua/wml/message.lua#L17
20160318 21:58:53< mattsc> The fix is, in principle, easy.  But did you have in mind that image=“” should _not_ mean “no image”?
20160318 21:59:36< celticminstrel> Given line 21, I think the intent is that you should use image=none to mean "no image".
20160318 21:59:38< mattsc> Looking a few lines later, that does not seem to be the case, but rather that image=none should be used in that case.
20160318 21:59:43 * celticminstrel nods.
20160318 21:59:46< mattsc> :D
20160318 21:59:58< mattsc> Let’s see what the wiki says ...
20160318 22:01:56< mattsc> So, just change to (image == nil or image == “”) then?
20160318 22:02:33 * celticminstrel nods.
20160318 22:02:43< mattsc> Oh, just saw the time.  I need to go.  Please go ahead make that change, unless you specifically want me to do it …
20160318 22:02:59< celticminstrel> I can, but I probably won't get to it for awhile.
20160318 22:03:00< mattsc> I’l be back online … probably tomorrow.  I don’t think I’ll have time today any more.
20160318 22:03:12< mattsc> Bye for now.
20160318 22:03:21< celticminstrel> Bye.
20160318 22:03:24-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20160318 22:22:45< gfgtdf> hmm i still don't understand why my string_switch is so slow.. trying with a std::map now
20160318 22:31:23< celticminstrel> Formula functions support a "precondition", but I don't understand the point of it.
20160318 22:31:50< celticminstrel> It's evaluated just before the function is called, and if it's false, an error message is printed; however, the function is still called despite that.
20160318 22:32:31< gfgtdf> celticminstrel: this looks lieka debuggn feature
20160318 22:32:40< gfgtdf> sounds*
20160318 22:33:14< celticminstrel> Yeah, that does seem to be its only use.
20160318 22:37:50< gfgtdf> celticminstrel: with std::map its even slower than with  boost::unordered map
20160318 22:38:02< celticminstrel> Makes sense.
20160318 22:38:24< gfgtdf> celticminstrel: and the boost::unordered_map is even slowe than the if(strcmp()) else if chain
20160318 22:38:29< gfgtdf> celticminstrel: i mean how can this be
20160318 22:38:31< celticminstrel> map is O(log n) on average, unordered_map is O(1) on average.
20160318 22:38:40< celticminstrel> I think that's right.
20160318 22:38:48< gfgtdf> celticminstrel: becasue std::map sill most likeley use also strcmp internally
20160318 22:38:59< gfgtdf> celticminstrel: it calls it less often
20160318 22:39:03< celticminstrel> However, unordered_map is O(n) in worst case.
20160318 22:39:15< gfgtdf> celticminstrel: so how can th std::map baed implementation be wlower 
20160318 22:39:19< celticminstrel> (Worst case is if call the strings have the same hash.)
20160318 22:39:23< gfgtdf> than the if else chain
20160318 22:39:36< gfgtdf> based implementation*
20160318 22:39:43< celticminstrel> I think the if-else chain would be O(n) too.
20160318 22:39:54< celticminstrel> So there's clearly something weird going on here...
20160318 22:41:18< Nobun> an if-else should check case-by-case
20160318 22:41:26< Nobun> while std::map not
20160318 22:41:40< Nobun> it seems strange to me that std::map is slower
20160318 22:41:50< celticminstrel> Yes, it's very strange that std::map is slower.
20160318 22:42:00< celticminstrel> The if-else case is equivalent to if you used std::vector.
20160318 22:42:20< celticminstrel> Like the way I handle the functions, it's essentially "loop through until you find one that matches".
20160318 22:42:30< celticminstrel> std::map on the other hand is a binary tree.
20160318 22:42:33< Nobun> yep... it is more or less the same if you iterate a std::vector element by element, indeed
20160318 22:42:39< celticminstrel> So it should be faster than linear search.
20160318 22:43:15< Nobun> however... the unordered map could be unefficient if the hash table is bad written, probably
20160318 22:43:27< celticminstrel> Yes, but this is Boost, so that's kinda unlikely.
20160318 22:43:28< Nobun> I would suggest to try to use std::map instead of unordered
20160318 22:44:16< Nobun> what is the element type you are trying to iterate?
20160318 22:44:29< Nobun> (or, better, to map?)
20160318 22:44:31< celticminstrel> Since std::map is essentially guaranteed to be faster (assuming you don't do something weird, which it seems you have), I'd suggest starting with that. Then once it works, switch out to unordered_map, which on averager should be even faster.
20160318 22:49:44-!- ancestral [~ancestral@209.181.254.220] has joined #wesnoth-dev
20160318 22:51:08< celticminstrel> ^on average
20160318 22:51:56-!- Nobun [~nobun@host62-49-dynamic.12-79-r.retail.telecomitalia.it] has quit [Quit: Salve a tutti]
20160318 22:52:43< celticminstrel> gfgtdf: I assume you've ignored the first access in determining which is faster.
20160318 22:53:02< celticminstrel> Since the first access will definitely be slower for either case.
20160318 22:53:57< gfgtdf> celticminstrel: hmm  i tested ina  loop and teh difference was ~8 seconds so i doint the first access will hvae  anoticable difference
20160318 22:54:38< celticminstrel> It wouldn't hurt to have a dummy access outside the loop though, just in case.
20160318 22:55:10-!- Nobun [~nobun@host62-49-dynamic.12-79-r.retail.telecomitalia.it] has joined #wesnoth-dev
20160318 22:55:27< celticminstrel> The first access will always be O(n) regardless of your choice of container.
20160318 22:55:58< gfgtdf> celticminstrel: i just noticed that the compiler inlines the strcmp() calls. But why wouldnt he do the same for the strcmp calls in the map baes implementatuion.
20160318 22:57:09< celticminstrel> I dunno.
20160318 23:01:14< celticminstrel> I would think that if it's inlining them in the one case it's likely doing it in the other case too...
20160318 23:07:16-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160318 23:09:21-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160318 23:13:15-!- ancestral [~ancestral@209.181.254.220] has quit [Quit: i go nstuf kthxbai]
20160318 23:13:56-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160318 23:15:44-!- irker727 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160318 23:16:56< Nobun> celticminstrel: I had the ok from ancestral to try to study how to support static build for wesnoth
20160318 23:17:11< Nobun> but I need to know something
20160318 23:17:24< celticminstrel> How to support what now?
20160318 23:17:35< celticminstrel> You mean building with all the libs statically linked, or something else?
20160318 23:17:56< Nobun> I mean a complete static build, with all dependencies solved
20160318 23:18:25< celticminstrel> Okay...
20160318 23:18:30< celticminstrel> Why from ancestral?
20160318 23:18:35< celticminstrel> And what do you need to know?
20160318 23:18:42< Nobun> the basic idea is that the static build could help you to solve all issues you are ecountering on MAC packages
20160318 23:18:59< Nobun> I contacted him since he is the mac packager
20160318 23:19:23< celticminstrel> Ah, yes, it could solve those issues.
20160318 23:19:42< Nobun> I need only to know some basic things on your CMakeLIsts.txt (that it is quite complex to read)
20160318 23:19:45< celticminstrel> At least, the issues of the distributed package not working on other computers.
20160318 23:19:54< Nobun> yep... this is my idea
20160318 23:20:01< celticminstrel> I don't know if I can help with CMake stuff.
20160318 23:20:08< Nobun> hmm
20160318 23:20:19< Nobun> ok I will do a weird workaround then
20160318 23:20:21< celticminstrel> Basically I know nothing about CMake.
20160318 23:20:23< celticminstrel> Sorry.
20160318 23:20:28< Nobun> don't worry
20160318 23:20:46< Nobun> I know CMake a bit, so I will try to figure a solution
20160318 23:21:00< celticminstrel> I could set up a static build in XCode without much trouble though. Probably. I haven't actually tried it.
20160318 23:21:16< Nobun> I have a starting idea (an uncomplete CMakeLists.txt for static linking starting from shared build)
20160318 23:21:46< Nobun> celticminstrel: I have a bit of experience on static build... it is not an easy task and you need a lot of workarounds
20160318 23:22:09< Nobun> this why I will try to support it in my os before going further
20160318 23:22:33< celticminstrel> "it is not an easy task"?
20160318 23:22:55< celticminstrel> I got a new project working with statically-linked Boost in a couple of minutes awhile back.
20160318 23:23:14< Nobun> boost is a library that it is thinked to be staticly linked
20160318 23:23:23< Nobun> even if it is shared linked in linux
20160318 23:23:30< celticminstrel> ?
20160318 23:23:54< Nobun> I will you pastebin a thing to explain where the things starts to be complex
20160318 23:24:47< vultraz> shadowm: out of curiosity, do you know what the Matrix widget was supposed to do?
20160318 23:25:51 * celticminstrel would assume it's supposed to be a matrix/table.
20160318 23:26:12< vultraz> it looks like a xox, ooo, xox 3x3 layout of viewport widgets, where o is a viewport and x is a spacer
20160318 23:26:21< vultraz> again, no idea what the viewport widget does either...
20160318 23:26:43< gfgtdf> hmm using a boost flat_map instead of std::map makes it perform worse
20160318 23:27:10< shadowm> vultraz: No.
20160318 23:27:23< gfgtdf> but if a use cusom comperator to boost::flat_set that used a cusom strcmp it performs better
20160318 23:27:33< Nobun> http://pastebin.com/p4gYCxci
20160318 23:27:51< Nobun> this is a part of a component of a Cmakelist of my own projects
20160318 23:27:54< vultraz> "A viewport is an special widget used to view only a part of the widget it `holds'." well this tells me nothing
20160318 23:28:02< Nobun> it is not important you understand or not Cmake
20160318 23:28:18< shadowm> vultraz: Sounds like what you'd use to implement a scrollable pane/listbox.
20160318 23:28:40< Nobun> look at the line 1-26--- those are the linking flags you require for two shared libraries
20160318 23:29:09< Nobun> when you static link (the last case - all following lines) you need to explicitly tells all single dependencies
20160318 23:29:49< Nobun> so a complete static link may be a very frustrating thing to obtain
20160318 23:30:19< vultraz> shadowm: potentially... it's used in tlistbox
20160318 23:30:24< vultraz> (viewport, that is)
20160318 23:30:36< loonycyborg> Nobun: I know a couple things about static linking, but I don't do cmake :P
20160318 23:30:43< Nobun> xD
20160318 23:30:45< shadowm> vultraz: Then you already solved it! Good job.
20160318 23:30:58< vultraz> viewpoer
20160318 23:31:00< vultraz> port*
20160318 23:31:03< shadowm> You were looking at the code while all I did was use some basic logic.
20160318 23:31:08< shadowm> *to use
20160318 23:31:12< vultraz> still don't know what matrix does.
20160318 23:31:25< shadowm> 20:26:22 <vultraz> again, no idea what the viewport widget does either...
20160318 23:31:29< Nobun> however I will try to understand your CMakeLists.txt, or I will rewrite another one from scratch
20160318 23:31:45< Nobun> is there a very complete list of dependencies used by every module?
20160318 23:31:47-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Ping timeout: 244 seconds]
20160318 23:31:49< Nobun> I mean
20160318 23:32:15< vultraz> tgenerator_
20160318 23:32:25< loonycyborg> anyway, I'm not sure it's even worth bothering with
20160318 23:32:30< vultraz> #nope #nope #nope 
20160318 23:33:05< Nobun> however in the next days I will ask something more detailed, perhaps
20160318 23:33:13< Nobun> now it is the time to go bed
20160318 23:33:32< Nobun> loonycyborg: let me know if you find other bugs on my wmlxgettext :)
20160318 23:33:35< loonycyborg> like even if you need to do something like distro-neutral release you can just package all .so files in a dir
20160318 23:34:38< Nobun> loonycyborg: good point
20160318 23:35:30< Nobun> I will think about it too (even if I never used the flag options like rpath or similar linking options)
20160318 23:35:46< Nobun> (so I have less experience in this)
20160318 23:36:09< Nobun> time to go bed now... good night all :)
20160318 23:36:28< loonycyborg> nn
20160318 23:36:58-!- Nobun [~nobun@host62-49-dynamic.12-79-r.retail.telecomitalia.it] has quit [Quit: Salve a tutti]
20160318 23:39:58< celticminstrel> Nobun: If I understand correctly, you're saying that for static linking you need to know the dependencies of your dependencies, right? However, that's still the case for dynamic linking - you don't need that information for building, but you still need it for distribution.
20160318 23:40:00< celticminstrel> vultraz: That description of a viewport sounds pretty effective to me.
20160318 23:40:08 * celticminstrel hopes Nobun sees that in the logs.
20160318 23:40:09< vultraz> not to me
20160318 23:40:31-!- mjs-de [~mjs-de@x4db62f58.dyn.telefonica.de] has quit [Ping timeout: 252 seconds]
20160318 23:40:43< celticminstrel> It contains a widget and displays a subset of that widget. How is that not clear?
20160318 23:40:47< celticminstrel> ^subrect
20160318 23:40:55 * shadowm seconds celticminstrel.
20160318 23:41:10< celticminstrel> Heck, that's clear from the name alone.
20160318 23:45:54< loonycyborg> celticminstrel: the issue here is that dynamic linking automatically handles some of indirect dependencies without requiring appropriate -l flags
20160318 23:46:11< loonycyborg> so static build might break unless you explictly maintain it
20160318 23:51:10< celticminstrel> The registration system for formula functions is confusing and keeps making me go in circles.
--- Log closed Sat Mar 19 00:00:54 2016