--- Log opened Sun Jul 31 00:00:49 2016
20160731 00:01:48< Aginor> I'll see how I get on later today
20160731 00:01:59< Aginor> I have a bunch of consulting to do first
20160731 00:02:45< mattsc> iceiceice, Aginor: https://wiki.wesnoth.org/ReferenceWML#WML_toplevel_tags
20160731 00:04:16< Ravana_> that list is missing modification
20160731 00:04:57< celmin> I'd argue that savefile doesn't really belong there, too.
20160731 00:05:12< celmin> Anf pbl
20160731 00:05:15< celmin> ^And
20160731 00:05:31< iceiceice> i think lua can appear at top level also
20160731 00:05:36< celmin> Yeah.
20160731 00:05:46< celmin> Save files and pbl are different schemas.
20160731 00:05:47< iceiceice> mattsc, good find
20160731 00:06:15< Ravana_> usually I just ctrl f from that page to tag I want
20160731 00:06:24< celmin> Oh, the logo tags should be added to the core page.
20160731 00:06:27< celmin> ^logo keys
20160731 00:07:44< celmin> Preferences are a fourth schema, and GUI2 a fifth.
20160731 00:07:58< celmin> Not sure if there are others.
20160731 00:09:45< celmin> Hmm, [multiplayer_side] can appear at toplevel, but I guess that doesn't need to be documented… those should go in [era]...
20160731 00:10:18< celmin> [ais] as well.
20160731 00:10:27< celmin> That should probably be listed.
20160731 00:11:44< celmin> [textdomain] is missing
20160731 00:11:55< celmin> Not sure what the page is for that.
20160731 00:12:17< celmin> There's also [advanced_preference], but I doubt that's useful to know about if you're not a dev.
20160731 00:13:21< celmin> …hey, is it possible for a macro expansion to control which macro is expanded?
20160731 00:13:43< celmin> ie, {{BLAH}} resulting in expanding a macro based on the value of the BLAH macro.
20160731 00:14:17< Ravana_> I believe it is
20160731 00:14:55< celmin> I notice that {core/} is unconditionally included in the toplevel _main.cfg
20160731 00:15:43< celmin> Hmm, is it that _main.cfg that's considered to be the default core, rather than core/main.cfg?
20160731 00:17:08< celmin> Oh, I found what appears to be another WML schema - lan_server.cfg
20160731 00:19:50< celmin> Well, anyway, I guess [ais] and [textdomain] at least should be added to that list.
20160731 00:22:51< celmin> [about] is toplevel but really only for devs.
20160731 00:23:40< celmin> Maybe [ais] is only for devs too… I think the [ai] tag can go in [era] and [modification].
20160731 00:24:23< celmin> …eh? The only explanation of the [textdomain] tag is on the WesCamp page? o.o
20160731 00:26:39< iceiceice> awesome, i think my variant is the fastest in the world :D
20160731 00:26:48< celmin> Uhuh
20160731 00:26:58< iceiceice> https://github.com/cbeck88/safe-variant
20160731 00:27:06< iceiceice> i benched it against all the variant implementations that i know of
20160731 00:28:01< celmin> Yeah, that code example should definitely not compile in my opinion.
20160731 00:28:17< celmin> (The first one. The second obviously should.)
20160731 00:30:44-!- irker189 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160731 00:32:46< celmin> I wonder whether boost::variant and std::variant actually use the C union construct.
20160731 00:33:09< celmin> Actually, is it possible to expand a template parameter pack into multiple statements…?
20160731 00:33:40< iceiceice> celmin, it kind of is
20160731 00:33:50< iceiceice> it depends what exaclty you mean
20160731 00:33:58< iceiceice> i dont think any of the variants use a union
20160731 00:34:08< iceiceice> because currently you cant parameter-pack expand into a struct or union like that
20160731 00:34:13< celmin> I'm thinking it's probably impossible, yeah.
20160731 00:34:18< iceiceice> the only one i know that actually uses a union is "eggs::variant"
20160731 00:34:31< iceiceice> but not in the way you expect
20160731 00:34:47< celmin> Maybe a recursive union?
20160731 00:34:59< iceiceice> in that one, hes basically using union as a hack so that it can be a constexpr variable in c++14
20160731 00:35:07< celmin> union {T1 first; variant<T…> tail;};
20160731 00:35:38< iceiceice> i guess you could do that, but you would need to have a "tag" at each level maybe?
20160731 00:36:01< celmin> Yeah, the tagging part would be nontrivial like that, I guess.
20160731 00:36:02< iceiceice> i think in eggs variant it is also using inhereitnace
20160731 00:36:13< iceiceice> eggs::variant is very elaborate :p
20160731 00:39:03< celmin> A discriminated union could also be implemented by storing a void* pointer… I seem to recall Boost doing it that way, in fact.
20160731 00:39:41< iceiceice> and storign the object on the heap?
20160731 00:39:47< celmin> Yeah.
20160731 00:39:52< iceiceice> i think boost::any might do that
20160731 00:40:07< iceiceice> usually variants try to store the object in situ
20160731 00:40:08< celmin> That's probably the one I was thinking of, not variant,.
20160731 00:40:10< iceiceice> so you dont get a cache hit
20160731 00:40:13< iceiceice> *cache miss
20160731 00:40:17< iceiceice> when you visit it
20160731 00:40:36< celmin> Yeah, I think I've used any more than variant… so it was probably any.
20160731 00:41:45< iceiceice> i never actually used any, i think i read some docs though
20160731 00:44:05< celmin> I wanted to change the formula variant to not use heap allocation, but that's incompatible with MSVC 2013.
20160731 00:44:25< celmin> Changing to boost::variant is a much more herculean task (I did attempt it though).
20160731 00:45:00< celmin> Particularly since there are two integer elements.
20160731 00:46:54< iceiceice> i'm looking at it now
20160731 00:46:57< iceiceice> i didnt know it didnt have bool
20160731 00:47:21< celmin> I was thinking about adding bool, but if I'm going to refactor it anyway I'd rather to that before adding a new type.
20160731 00:47:34< iceiceice> oh wierd
20160731 00:47:38< celmin> Hmm?
20160731 00:47:39< iceiceice> it represents decimal as an int?
20160731 00:47:43< iceiceice> i never knew this
20160731 00:47:48< celmin> Yeah, decimals are fixed-point in WFL.
20160731 00:48:02< celmin> Three decimal places exactly.
20160731 00:48:13< celmin> (I also wanted to increase that to five, but same argument as above.)
20160731 00:48:57< iceiceice> yeah that implementation is wierd then
20160731 00:48:59< iceiceice> idk
20160731 00:49:03< celmin> Hm?
20160731 00:49:10< iceiceice> i would rather to see at least like "struct decimal_int { int data; }"
20160731 00:49:15< iceiceice> or something like that
20160731 00:49:31< celmin> Replacing that int with a class was another thing I wanted to do.
20160731 00:49:45-!- mjs-de [~mjs-de@x4db69e37.dyn.telefonica.de] has quit [Remote host closed the connection]
20160731 00:52:34-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 258 seconds]
20160731 01:17:31-!- horrowind [~Icedove@2a02:810a:83c0:404:21b:fcff:fee3:c3ff] has quit [Quit: horrowind]
20160731 01:30:09-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: room bursts into flames]
20160731 01:34:31-!- trewe [~trewe@2001:8a0:d131:8801:b3b6:cbaf:33e3:7418] has quit [Quit: quit]
20160731 01:52:49-!- Bonobo [~Bonobo@ppp118-210-92-241.lns20.adl2.internode.on.net] has joined #wesnoth-dev
20160731 02:06:52-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160731 03:17:09-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: mattsc]
20160731 03:18:23-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20160731 03:35:39-!- atarocch [~atarocch@93.56.160.28] has quit [Ping timeout: 244 seconds]
20160731 03:41:17-!- celmin [~celticmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The minstrel departs, to spread the music to the masses!]
20160731 03:47:52-!- atarocch [~atarocch@93.56.160.28] has joined #wesnoth-dev
20160731 03:59:31-!- hk238 [~kvirc@t224.ip7.netikka.fi] has joined #wesnoth-dev
20160731 04:14:23-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160731 04:16:29< celticminstrel> Huh, trolls get fearless instead of intelligent.
20160731 04:16:38< celticminstrel> I didn't know that.
20160731 04:17:51-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 264 seconds]
20160731 04:18:32-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 240 seconds]
20160731 04:35:00-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has joined #wesnoth-dev
20160731 04:58:45< vultraz> celticminstrel: i don't think it's necessary
20160731 04:58:50< celticminstrel> ???
20160731 04:59:13< vultraz> [09:37:47]	celmin	vultraz: Also, should that bug be in the announcement post?
20160731 04:59:27< celticminstrel> Which one was that?
20160731 05:00:16< vultraz> the ellipse one i think
20160731 05:00:24< celticminstrel> Speaking of which, PR 718?
20160731 05:00:46-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160731 05:01:54< bumbadadabum> hi
20160731 05:02:33< vultraz> what aboutit
20160731 05:02:36-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has joined #wesnoth-dev
20160731 05:03:47< bumbadadabum> anything anyone needs help with
20160731 05:04:23-!- TC02 [~quassel@venus.arosser.com] has quit [Ping timeout: 250 seconds]
20160731 05:04:45-!- TC02 [~quassel@venus.arosser.com] has joined #wesnoth-dev
20160731 05:18:22< vultraz> bumbadadabum: you could work on https://gna.org/bugs/index.php?23995
20160731 05:18:37< vultraz> or someone could
20160731 05:42:25-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160731 05:43:00< vultraz> bumbadadabum: this is simpler https://gna.org/bugs/index.php?24388
20160731 05:44:55< bumbadadabum> yes that latter one is
20160731 05:45:01< bumbadadabum> because that's already fixed
20160731 05:45:47< vultraz> I'm not seeing it
20160731 05:46:18< vultraz> none of the monsters are listing traits
20160731 05:47:03-!- iwaim_ [~iwaim@rasteenie.alib.jp] has quit [Ping timeout: 240 seconds]
20160731 05:47:35< bumbadadabum> wait
20160731 05:47:37< bumbadadabum> oh I misread
20160731 05:47:39< bumbadadabum> well that's odd
20160731 05:48:10< bumbadadabum> I think it's because num_traits=0?
20160731 05:48:27< bumbadadabum> I'll look into that
20160731 05:49:11< celticminstrel> Setting num_traits=1 should fix it then.
20160731 05:49:33< celticminstrel> If it's true that it's 0.
20160731 05:49:50< celticminstrel> I wonder what happens if num_traits is nonzero but no traits are defined...
20160731 05:51:39< bumbadadabum> yeah
20160731 05:51:42< bumbadadabum> I need to fix it regardless
20160731 05:51:51< bumbadadabum> setting num_traits to non-zero is not a SOLUTION
20160731 05:52:21< vultraz> celticminstrel: I think we can close this? https://gna.org/bugs/?22855
20160731 05:52:55< vultraz> and this https://gna.org/bugs/?22777
20160731 05:56:00< vultraz> celticminstrel: please confirm if this is still an issue https://gna.org/bugs/?21294
20160731 05:58:35< vultraz> celticminstrel: also might be invalid https://gna.org/bugs/?20052
20160731 06:01:23< vultraz> bumbadadabum: you might also be interested in pondering the dilemma that is https://gna.org/bugs/?22352 (related https://github.com/wesnoth/wesnoth/pull/536 )
20160731 06:02:11< celticminstrel> Invalid indeed, there's no need for an init function.
20160731 06:02:16< bumbadadabum> ty for links
20160731 06:02:18< celticminstrel> Someone lacks an imagination maybe?
20160731 06:03:56< vultraz> celticminstrel: if any of those four are invalid/fixed, just mark as such and close them
20160731 06:04:04< celticminstrel> I already closed all four.
20160731 06:04:31< celticminstrel> Complete with a comment on each.
20160731 06:05:06< vultraz> sweet
20160731 06:07:25-!- iwaim [~iwaim@rasteenie.alib.jp] has joined #wesnoth-dev
20160731 06:10:30-!- TC02 [~quassel@venus.arosser.com] has quit [Ping timeout: 244 seconds]
20160731 06:10:33 * vultraz is going through the entire bug list and closing old fixes/invalid bugs
20160731 06:10:50< celticminstrel> ...what the heck is {SPECIAL_NOTES_DEFENSE_CAP}?
20160731 06:11:04< celticminstrel> It seems to be used only on human horsemen.
20160731 06:11:15-!- TC02 [~quassel@venus.arosser.com] has joined #wesnoth-dev
20160731 06:11:23< celticminstrel> And doesn't have a wmllint entry.
20160731 06:13:04< vultraz> dunno
20160731 06:13:08< vultraz> what's inside it?
20160731 06:13:31< celticminstrel> _" This unit has a defense cap on certain terrain types — it cannot achieve a higher defense rating on mixed terrains with such terrain types."
20160731 06:13:37< vultraz> ah
20160731 06:14:08< vultraz> bumbadadabum: https://gna.org/bugs/?17982 just marked this as fixed, since that was implemented, but I'm just wondering, when is [delay] used in animations?
20160731 06:14:16< vultraz> I cannot recall such a thing
20160731 06:15:42< bumbadadabum> it isn't
20160731 06:15:54-!- TC02 [~quassel@venus.arosser.com] has quit [Ping timeout: 260 seconds]
20160731 06:16:05< celticminstrel> It's an ActionWML, right?
20160731 06:16:09< bumbadadabum> ^
20160731 06:16:21< celticminstrel> So I guess cutscene-type animations.
20160731 06:16:31< celticminstrel> Not unit animations necessarily.
20160731 06:16:35< bumbadadabum> also, would any of you mind if I unify knife.ogg for the standard short sword sound
20160731 06:16:43< celticminstrel> Though it could include such with [animate_unit].
20160731 06:16:43< bumbadadabum> it's currently used by the dwarvish thunderer line
20160731 06:16:46< bumbadadabum> (and made for it)
20160731 06:16:48-!- TC02 [~quassel@venus.arosser.com] has joined #wesnoth-dev
20160731 06:16:52< celticminstrel> Huh?
20160731 06:16:56< bumbadadabum> but I think it'd fit the orcish crossbowmen as well
20160731 06:17:05< celticminstrel> I have no idea.
20160731 06:17:16< celticminstrel> I guess I have little preference on sounds.
20160731 06:21:21-!- TC02 [~quassel@venus.arosser.com] has quit [Ping timeout: 244 seconds]
20160731 06:21:44-!- Kwandulin [~Miranda@p200300760F6062A518E6FF9992889797.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160731 06:21:51-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The computer falls into a deep sleep, to awake again some other day!]
20160731 06:22:56-!- TC02 [~quassel@venus.arosser.com] has joined #wesnoth-dev
20160731 06:28:30-!- JyrkiVesterinen [~JyrkiVest@87-100-221-13.bb.dnainternet.fi] has joined #wesnoth-dev
20160731 06:30:45< bumbadadabum> vultraz: since you're probably a sound expert with your razer headset
20160731 06:30:55< bumbadadabum> would you mind if I change some attack sounds and stuff
20160731 06:31:06< vultraz> no, that's fine
20160731 06:31:09< bumbadadabum> ok
20160731 06:31:10< bumbadadabum> nice
20160731 06:31:22< vultraz> as long as the new ones fit
20160731 06:31:43< bumbadadabum> idk I feel like knife.ogg fits for short swords imo
20160731 06:32:50< vultraz> as opposed to?
20160731 06:33:20< vultraz> sword-1.ogg?
20160731 06:33:43< bumbadadabum> y
20160731 06:34:17< vultraz> yeah, that does fit
20160731 06:50:37-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has joined #wesnoth-dev
20160731 06:58:36-!- TC02 [~quassel@venus.arosser.com] has quit [Ping timeout: 276 seconds]
20160731 06:59:25< ancestral> So do I release as 1.13.5 or 1.13.5a?
20160731 06:59:38< vultraz> ancestral: former
20160731 07:05:20< vultraz> ancestral: can you add the sha hash and download link to the announcement draft once you have the package up?
20160731 07:05:37< ancestral> Sure
20160731 07:11:42< vultraz> loonycyborg: and can you add the sha for the windows package?
20160731 07:12:31-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160731 07:13:01< vultraz> I think we'll be good to go, then
20160731 07:13:04< shadowm> Do not forget the hash files.
20160731 07:13:32< vultraz> Hash files?
20160731 07:13:33-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160731 07:14:07< shadowm> https://files.wesnoth.org/releases/
20160731 07:14:30< shadowm> Ignore loonycyborg's obsession with MD5. We're not using those, so you don't need to generate them
20160731 07:15:31< vultraz> I can't generate the hashes myself since I don't have the packages.
20160731 07:15:42< shadowm> That wasn't meant for you.
20160731 07:15:49< shadowm> That was meant for ancestral and loonycyborg.
20160731 07:16:04< vultraz> Ah, ok.
20160731 07:17:20 * vultraz makes note to update the Downloads page too
20160731 07:17:32-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Client Quit]
20160731 07:17:35< shadowm> You did not need to use the screenshots in the exact order their filenames specify.
20160731 07:17:45 * vultraz wonders if a script could be written to do that
20160731 07:17:48< shadowm> Whatever.
20160731 07:18:33-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160731 07:19:51-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Remote host closed the connection]
20160731 07:21:24-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160731 07:22:29-!- TC02 [~quassel@venus.arosser.com] has joined #wesnoth-dev
20160731 07:30:45-!- TC02 [~quassel@venus.arosser.com] has quit [Ping timeout: 244 seconds]
20160731 07:32:01-!- TC02 [~quassel@venus.arosser.com] has joined #wesnoth-dev
20160731 07:35:26-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160731 07:36:16-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160731 07:36:24-!- TC02 [~quassel@venus.arosser.com] has quit [Ping timeout: 260 seconds]
20160731 07:47:39-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160731 07:48:24-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160731 07:51:16-!- Appleman1234_ [~Appleman1@KD036012036034.au-net.ne.jp] has joined #wesnoth-dev
20160731 07:52:33-!- Appleman1234 [~Appleman1@KD036012037194.au-net.ne.jp] has quit [Ping timeout: 276 seconds]
20160731 07:53:49-!- TC02 [~quassel@venus.arosser.com] has joined #wesnoth-dev
20160731 07:58:08-!- TC02 [~quassel@venus.arosser.com] has quit [Ping timeout: 250 seconds]
20160731 08:04:35-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 258 seconds]
20160731 08:05:53-!- TC02 [~quassel@venus.arosser.com] has joined #wesnoth-dev
20160731 08:07:54-!- iwaim__ [~iwaim@rasteenie.alib.jp] has joined #wesnoth-dev
20160731 08:07:55-!- iwaim [~iwaim@rasteenie.alib.jp] has quit [Read error: Connection reset by peer]
20160731 08:08:03-!- iwaim__ [~iwaim@rasteenie.alib.jp] has quit [Excess Flood]
20160731 08:08:26-!- iwaim__ [~iwaim@2001:2c0:40e:2002:0:4:14:80] has joined #wesnoth-dev
20160731 08:10:45-!- TC02 [~quassel@venus.arosser.com] has quit [Ping timeout: 276 seconds]
20160731 08:20:00-!- TC02 [~quassel@venus.arosser.com] has joined #wesnoth-dev
20160731 08:46:12-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160731 08:48:47-!- lipkab [~the_new_l@host-91-147-210-58.biatv.hu] has joined #wesnoth-dev
20160731 08:56:52< loonycyborg> vultraz: add sha where?
20160731 08:58:25< vultraz> loonycyborg: to the announcement
20160731 08:58:35-!- mjs-de [~mjs-de@x4db69e37.dyn.telefonica.de] has joined #wesnoth-dev
20160731 08:59:41< loonycyborg> https://files.wesnoth.org/releases/ contains sha hashes for windows installer
20160731 08:59:58< vultraz> oh, I see
20160731 09:00:09< vultraz> thanks
20160731 09:05:28< loonycyborg> actually shadowm asked me only to add windows installer hashes to files.wesnoth.org
20160731 09:05:39< loonycyborg> while tarball hashes are uploaded to frs too
20160731 09:11:08< vultraz> now we just need the macos packages 
20160731 09:14:28< vultraz> package 
20160731 09:15:39-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20160731 09:17:28-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20160731 09:22:16< zookeeper> celticminstrel, the defense cap refers to their forest defense.
20160731 09:24:42< zookeeper> bumbadadabum, actually, what should be done is to replace all dagger-swish.wav with knife.ogg.
20160731 09:29:35-!- Kwandulin_2 [~Miranda@p200300760F6062F20819B6EE52497D9D.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160731 09:29:42< zookeeper> i'd rather have sword-1.ogg on everything labeled as a sword, and knife.ogg on everything labeled as a knife/dagger. the latter wasn't intended only for the thunderer line, applying it to all the other knives was just forgotten.
20160731 09:30:07-!- Kwandulin [~Miranda@p200300760F6062A518E6FF9992889797.dip0.t-ipconnect.de] has quit [Ping timeout: 258 seconds]
20160731 10:01:49-!- JyrkiVesterinen [~JyrkiVest@87-100-221-13.bb.dnainternet.fi] has quit [Quit: .]
20160731 10:04:00< Aginor> ok, finally done with all consulting
20160731 10:04:12< Aginor> and invoicing
20160731 10:33:17-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160731 10:34:51< Aginor> hmm
20160731 10:35:01< Aginor> I have no scrollbar in my savegames dialog
20160731 10:38:39-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160731 10:43:05-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Client Quit]
20160731 10:44:02< Aginor> hmm
20160731 10:44:43< Aginor> the monte carlo simulation yields pretty different results from not using it
20160731 10:51:00-!- Duthlet [~Duthlet@dslb-178-005-055-087.178.005.pools.vodafone-ip.de] has joined #wesnoth-dev
20160731 10:51:12-!- lipkab [~the_new_l@host-91-147-210-58.biatv.hu] has quit [Ping timeout: 244 seconds]
20160731 10:55:48-!- horrowind [~Icedove@2a02:810a:83c0:404:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20160731 11:00:40-!- lipkab [~the_new_l@host-91-147-210-58.biatv.hu] has joined #wesnoth-dev
20160731 11:14:42-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160731 11:20:10< Aginor> JyrkiVesterinen: I've given you some feedback on PR702, I think the discrepancy is sufficiently large that it won't make sense to merge/enable it by default yet
20160731 11:21:39< Aginor> assuming he reads the logs ;)
20160731 11:55:55-!- Kwandulin_2 [~Miranda@p200300760F6062F20819B6EE52497D9D.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160731 12:00:38-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160731 12:03:39-!- JyrkiVesterinen [~JyrkiVest@87-100-209-170.bb.dnainternet.fi] has joined #wesnoth-dev
20160731 12:06:25-!- Appleman1234_ is now known as Appleman1234
20160731 12:20:00< vultraz> [21:35:00]	Aginor	I have no scrollbar in my savegames dialog
20160731 12:20:05< vultraz> Aginor: elaborate, please
20160731 12:26:21< JyrkiVesterinen> Aginor: I responded to your feedback. I do read the IRC logs.
20160731 12:39:56-!- Kwandulin [~Miranda@2003:76:f60:62f2:edd2:2553:c194:181] has joined #wesnoth-dev
20160731 12:49:11< wedge009> vultraz: Scroll bar seems okay to me.
20160731 12:49:31< wedge009> vultraz: I have a list of around 52 bugs which I think can/should be closed.
20160731 12:49:45< wedge009> Some by age, but many because they are fixed.
20160731 12:50:22< vultraz> If you think they're good to close, just close them.
20160731 12:51:05< vultraz> I just closed a few earlier today
20160731 12:52:02< wedge009> vultraz: I know that's why I asked you about it.
20160731 12:52:12< wedge009> I don't think I've ever been granted access to the bug tracker.
20160731 12:52:21< wedge009> I can only submit, not modify anything beyond that.
20160731 12:53:51< vultraz> huh
20160731 12:54:12< vultraz> honestly I wouldn't know how to grant you that
20160731 12:54:17< vultraz> I'm not an admin on the tracker
20160731 12:54:29< vultraz> T'is why we need to get us switched to Github
20160731 12:56:09< vultraz> wedge009: blah, just give me a list.
20160731 12:56:29< wedge009> Like I did last time? It's a big list.
20160731 12:56:41< vultraz> sure
20160731 12:56:48< wedge009> https://gna.org/bugs/?16934
20160731 12:56:48< wedge009> https://gna.org/bugs/?17244
20160731 12:56:48< wedge009> https://gna.org/bugs/?17358
20160731 12:56:48< wedge009> https://gna.org/bugs/?17995
20160731 12:56:48< wedge009> https://gna.org/bugs/?18204
20160731 12:56:48-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Excess Flood]
20160731 12:57:03-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160731 12:57:03< wedge009> Wow, I got booted.
20160731 12:57:03-!- Shiki [~Shiki@141.39.226.227] has joined #wesnoth-dev
20160731 12:57:18< vultraz> durr, not all at once :P
20160731 12:57:21< wedge009> Feel free to leave any open.
20160731 12:57:32< vultraz> I only saw the first 5
20160731 12:57:46< wedge009> Okay, trying again.
20160731 12:57:53< wedge009> https://gna.org/bugs/?16934
20160731 12:57:53< wedge009> https://gna.org/bugs/?17244
20160731 12:57:53< wedge009> https://gna.org/bugs/?17358
20160731 12:57:53< wedge009> https://gna.org/bugs/?17995
20160731 12:57:53< wedge009> https://gna.org/bugs/?18204
20160731 12:57:54< wedge009> https://gna.org/bugs/?18269
20160731 12:57:54< wedge009> https://gna.org/bugs/?18868
20160731 12:57:55< wedge009> https://gna.org/bugs/?19237
20160731 12:58:02< wedge009> https://gna.org/bugs/?19414 OR https://gna.org/bugs/?24411
20160731 12:58:02< wedge009> https://gna.org/bugs/?20388
20160731 12:58:02< wedge009> https://gna.org/bugs/?20938
20160731 12:58:02< wedge009> https://gna.org/bugs/?21157
20160731 12:58:02< wedge009> https://gna.org/bugs/?21567
20160731 12:58:03< wedge009> https://gna.org/bugs/?19544 OR https://gna.org/bugs/?21564
20160731 12:58:03< wedge009> https://gna.org/bugs/?21584
20160731 12:58:04< wedge009> https://gna.org/bugs/?21662
20160731 12:58:04< wedge009> https://gna.org/bugs/?21926
20160731 12:58:16< wedge009> https://gna.org/bugs/?21937
20160731 12:58:16< wedge009> https://gna.org/bugs/?21945
20160731 12:58:16< wedge009> https://gna.org/bugs/?21956
20160731 12:58:16< wedge009> https://gna.org/bugs/?22044
20160731 12:58:16< wedge009> https://gna.org/bugs/?22133
20160731 12:58:17< wedge009> https://gna.org/bugs/?22338
20160731 12:58:17< wedge009> https://gna.org/bugs/?22368
20160731 12:58:18< wedge009> https://gna.org/bugs/?22526
20160731 12:58:18< wedge009> https://gna.org/bugs/?22527
20160731 12:58:20< wedge009> https://gna.org/bugs/?22866
20160731 12:58:20< wedge009> https://gna.org/bugs/?22891
20160731 12:58:20< wedge009> https://gna.org/bugs/?22903
20160731 12:58:20< wedge009> https://gna.org/bugs/?23000
20160731 12:58:21< wedge009> https://gna.org/bugs/?23002
20160731 12:58:35< wedge009> https://gna.org/bugs/?23922
20160731 12:58:35< wedge009> https://gna.org/bugs/?23941
20160731 12:58:35< wedge009> https://gna.org/bugs/?23943
20160731 12:58:35< wedge009> https://gna.org/bugs/?24119
20160731 12:58:35< wedge009> https://gna.org/bugs/?24226
20160731 12:58:36< wedge009> https://gna.org/bugs/?24357
20160731 12:58:36< wedge009> https://gna.org/bugs/?24454
20160731 12:58:37< wedge009> https://gna.org/bugs/?24456
20160731 12:58:37< wedge009> https://gna.org/bugs/?24553 OR https://gna.org/bugs/?24702
20160731 12:58:38< wedge009> Done
20160731 12:58:57< wedge009> We can track bugs on GitHub?
20160731 12:59:06< vultraz> I'm working to get us moved there
20160731 12:59:19< vultraz> Just need to figure out how to export stuff from GNA
20160731 12:59:36< wedge009> Well, you can kill off the old ones to make the job easier. (:
20160731 12:59:37< vultraz> Then how to push it to GH through their API
20160731 12:59:45< vultraz> I'll probably need le Elvish_Hunter 
20160731 12:59:49< lipkab> Didn't we plan to host our own redmine tracker?
20160731 13:00:18< vultraz> And where has that gone?
20160731 13:00:21< vultraz> Much easier to use GH
20160731 13:01:12< lipkab> We *did* have a reason not to use GitHub.
20160731 13:01:27< lipkab> Like no attachments in bug reports.
20160731 13:01:36< lipkab> Is that resolved?
20160731 13:01:52< vultraz> Their attachment support is sufficient now, I believe
20160731 13:07:00 * vultraz wonders if &(*attacker_itor_) is considered messy
20160731 13:15:28-!- Shiki [~Shiki@141.39.226.227] has quit [Quit: Verlassend]
20160731 13:17:17< lipkab> vultraz: I've played around with it and there's still something I miss: anonymous submissions.
20160731 13:17:41< vultraz> IMO it's worth it.
20160731 13:17:52< wedge009> Why would you want to submit anonymously?
20160731 13:18:08< wedge009> Oh, you mean random people from the community who don't necessarily have a GitHub account?
20160731 13:18:12< lipkab> Yep.
20160731 13:18:35< wedge009> Do people bring up bugs on the forums? I don't really go there very often.
20160731 13:18:51< lipkab> vultraz: That's cool but there are other people in this project, you know... we should at least talk this over.
20160731 13:19:30< lipkab> wedge009: I don't visit the forum too often these days either.
20160731 13:19:46< vultraz> No offense, but how long ago was the bug tracker debate? We got nowhere. All we do is talk and go in circles and never come to a consensus.
20160731 13:20:15< wedge009> Just wondering if that could be an alternative for those without GitHub accounts, though I know it's not desirable to split things.
20160731 13:21:24< lipkab> vultraz: We did came to a consensus. It was that GH is not a good enough solution, Redmine is.
20160731 13:21:40< vultraz> lipkab: but then we did nothing about it
20160731 13:22:22< lipkab> So? That automatically means that GH is good idea?
20160731 13:22:56< vultraz> No, it means that since GH is now acceptable eniugh it's better than inaction and sticking with GNA for the next however many years
20160731 13:23:06-!- molt [~molt@37.122.170.190] has joined #wesnoth-dev
20160731 13:23:53< vultraz> Plus, using GH means more centralized project management.
20160731 13:24:25< lipkab> "GH is now acceptable enough" <- Because you decided so.
20160731 13:24:36< lipkab> Redmine can integrate with GH IIRC.
20160731 13:24:48< lipkab> See, I'm not against GH.
20160731 13:26:13< lipkab> But you shouldn't just go and change around elements in our core infrastructure because you think it's "good enough".
20160731 13:26:28-!- hk238 [~kvirc@t224.ip7.netikka.fi] has quit [Quit: http://www.kvirc.net/ 4.9.1 Aria]
20160731 13:27:19< lipkab> Put out an email to the list, collect some opinions, and if we agree that GH is better than Gna, then by all means, go ahead.
20160731 13:32:12< vultraz> I feel if we do that we'll go around in circles again and nothing will happen.
20160731 13:32:18< zookeeper> and/or actually list the pros and cons. i know that people seem to hate gna, but i don't really know what sort of technical issues there are with it... it's pretty convenient for me.
20160731 13:34:39< zookeeper> how _do_ you attach files to a github issue? i don't see an option for that
20160731 13:35:14< vultraz> drag it to the box
20160731 13:35:23< lipkab> zookeeper: Look at the bottom of the textfield :P
20160731 13:35:45< zookeeper> ohh, right
20160731 13:36:00< lipkab> vultraz: I'm not sure why do you think that.
20160731 13:36:07< zookeeper> this is some fancy NEW TECHNOLOGY
20160731 13:36:15< lipkab> Let's assume you send that letter the list.
20160731 13:36:41< lipkab> Nobody replies. Nobody cares. You're free to do as you please.
20160731 13:37:08< zookeeper> are the file format limitations global to all github projects or can each project set their own?
20160731 13:37:16< lipkab> People reply and we agree that GH is a good idea. Again you can go ahead.
20160731 13:37:49< lipkab> People reply and the majority agrees that GH is *not* a good idea, even worse than Gna.
20160731 13:38:04< lipkab> Well, yes, then nothing will happen.
20160731 13:38:39< lipkab> ...but I think we can agree that nothing *should* happen in that case anyway.
20160731 13:38:56< vultraz> Alright
20160731 13:39:04< vultraz> I'll write an email.
20160731 13:39:11< lipkab> Thanks.
20160731 13:48:11-!- markus_ [~mjs-de@x4db5d13b.dyn.telefonica.de] has joined #wesnoth-dev
20160731 13:51:53-!- mjs-de [~mjs-de@x4db69e37.dyn.telefonica.de] has quit [Ping timeout: 244 seconds]
20160731 13:56:36< vultraz> lipkab: sent
20160731 13:56:48< lipkab> Cool.
20160731 14:10:19< wedge009> vultraz: I think I found out how to add myself to the gna group. Aginor is already on it. Maybe that's why I was never able to edit stuff in the bug tracker.
20160731 14:10:26< wedge009> Well, request to add myself.
20160731 14:31:56-!- markus_ is now known as mjs-de
20160731 14:36:49-!- hk238 [~kvirc@t224.ip7.netikka.fi] has joined #wesnoth-dev
20160731 14:39:12< vultraz> hmmmm
20160731 14:39:18< vultraz> I need celticminstrel for this
20160731 14:44:08< vultraz> lipkab: how did you test .gz?
20160731 14:44:26< vultraz> I just tested and they upload fine.
20160731 14:44:36< lipkab> Hmm.
20160731 14:44:40< lipkab> Strange.
20160731 14:44:57< lipkab> I just tested as well and it did not upload fine :D
20160731 14:46:22< vultraz> I just dragged them into the box, what did you do.
20160731 14:46:43< lipkab> Clicked the button and selected the file.
20160731 14:46:59< lipkab> Will try dragging.
20160731 14:47:44< lipkab> Nah.
20160731 14:48:17< vultraz> Just tested by selecting them, it also worked.
20160731 14:48:23< vultraz> Must be something on your end.
20160731 14:49:23< vultraz> you sure you're trying gz files?
20160731 14:49:28< lipkab> More like something on GitHub's end.
20160731 14:50:22< lipkab> Yes and I'm also sure I'm trying to upload them to GitHub.
20160731 14:51:20< lipkab> Oh.
20160731 14:51:26< lipkab> It works in Chrome.
20160731 14:52:05< vultraz> I was also using Chrome
20160731 14:58:17-!- Bonobo [~Bonobo@ppp118-210-92-241.lns20.adl2.internode.on.net] has quit [Quit: Leaving]
20160731 15:04:30< lipkab> Ok, I sent a mail to GH support.
20160731 15:04:49-!- travis-ci [~travis-ci@ec2-54-204-106-148.compute-1.amazonaws.com] has joined #wesnoth-dev
20160731 15:04:50< travis-ci> spixi/wesnoth#6 (chart_engine - 0ffd572 : Spixi): The build failed.
20160731 15:04:50< travis-ci> Build details : https://travis-ci.org/spixi/wesnoth/builds/148676649
20160731 15:04:50-!- travis-ci [~travis-ci@ec2-54-204-106-148.compute-1.amazonaws.com] has left #wesnoth-dev []
20160731 15:09:19< vultraz> lipkab: so, giving that, it seems the only point in disfavor is no anon submissions
20160731 15:10:00< lipkab> ...and cfg-s, though's that admittedly less important than gz-s.
20160731 15:10:32< vultraz> Certainly
20160731 15:12:03< lipkab> Afk now, will be back later.
20160731 15:12:04-!- lipkab [~the_new_l@host-91-147-210-58.biatv.hu] has quit [Quit: Leaving]
20160731 15:17:30-!- irker100 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160731 15:17:30< irker100> wesnoth: Charles Dang wesnoth:master 639548a108eb / / (6 files in 5 dirs): Made Attack dialog use a unit_preview_pane widget https://github.com/wesnoth/wesnoth/commit/639548a108eb05c32596be59359b373fb0248670
20160731 15:30:23-!- Kwandulin [~Miranda@2003:76:f60:62f2:edd2:2553:c194:181] has quit [Read error: Connection reset by peer]
20160731 15:36:53-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160731 16:04:56-!- Kwandulin [~Miranda@p200300760F6062F2E48E7F2D0FF9198C.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160731 16:08:21-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20160731 16:09:40< irker100> wesnoth: Charles Dang wesnoth:master b04083ab9021 / src/ (dialogs.cpp dialogs.hpp): Removed some old GUI1 Recruit dialog code https://github.com/wesnoth/wesnoth/commit/b04083ab90217c09dda67d21235ad0b11cb70cc7
20160731 16:12:24-!- iceiceice [~chris@50.245.222.235] has joined #wesnoth-dev
20160731 16:12:24-!- iceiceice [~chris@50.245.222.235] has quit [Changing host]
20160731 16:12:24-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20160731 16:13:45< iceiceice> lipkab: I think "cfg" attachments is not a significant issue compared to the usability problems of gna
20160731 16:13:56< iceiceice> i don't recall ever seeing a "cfg" attachment in a bug report,
20160731 16:14:04< iceiceice> i only ever saw .gz
20160731 16:14:40< iceiceice> do you know a way to estimate how many `.cfg` we currently have?
20160731 16:14:59< iceiceice> anyways it should be trivial to rename the .cfg as .txt or something
20160731 16:15:19< iceiceice> i think that is a much more straightforward issue for the users than figuring out "what VCS is named after a british slang"
20160731 16:15:35< vultraz> Agreed
20160731 16:15:47< celticminstrel> Heh, I said the exact same thing.
20160731 16:16:08< celticminstrel> Come to think of it, isn't the answer to that "git"?
20160731 16:16:23< vultraz> yes
20160731 16:16:46< vultraz> I really don't think a lack of anon submissions is that much of an issue
20160731 16:16:54< celticminstrel> Though I dunno if it's named after it.
20160731 16:18:36< vultraz> lipkab: btw, don't work on porting the recall dialog to gui2. im doing that.
20160731 16:19:03< irker100> wesnoth: Gregory A Lundberg wesnoth:master a75812d4aa1f / src/units/unit.cpp: unit.cpp Fix bug: Ellipse is lost https://github.com/wesnoth/wesnoth/commit/a75812d4aa1fe8e816165e09d7ca2cf5e240d091
20160731 16:19:05< irker100> wesnoth: Celtic Minstrel wesnoth:master cd858295fec9 / src/units/unit.cpp: Merge pull request #718 from GregoryLundberg/lost-ellipse https://github.com/wesnoth/wesnoth/commit/cd858295fec9a3ca374077e4c1ee6cb6de97d39c
20160731 16:19:48< iceiceice> celticminstrel, supposedly it is, but who knows
20160731 16:19:49< iceiceice> https://en.wikipedia.org/wiki/Git_(software)
20160731 16:20:28< celticminstrel> I think lipkab was looking at some bugs, anyway, vultraz.
20160731 16:23:49< vultraz> celticminstrel: so, I had an idea for a feature.. not sure how to implement it. It occurred to me when moving the attack dialog to gui2 that it might be worth having some kind of post-processing on tcontrol label strings, so something like this could be done: "(text) + 'foo'. That'd append the string 'foo' to whatever the label was at any given time. I looked at the implementation of...
20160731 16:23:50< vultraz> ...tcontro::set_label(), and the label just seems to be a tstring whose value is then set to the formula variable (text). Maybe some extra key could be added... or some on_update thing in the formula engine that post-process a variable after it gets updated? 
20160731 16:24:34< celticminstrel> Does "(text + 'foo')" work?
20160731 16:24:54< vultraz> hm. i hadn't tried that particular syntax.
20160731 16:25:03< celticminstrel> Note though that you shouldn't be doing this with strings that convey meaning.
20160731 16:25:11< celticminstrel> Since those need to be translatable.
20160731 16:25:20< celticminstrel> I could add a _() function to WFL if that helps.
20160731 16:26:02< celticminstrel> Though wmlxgettext would need to be adjusted to account for it. Alternatively, some other syntax for translatable strings.
20160731 16:26:29< vultraz> doesn;t work, no
20160731 16:26:43< celticminstrel> Huh...
20160731 16:26:45< vultraz> label = in WML is basically just a text value
20160731 16:27:10< celticminstrel> I thought you said you had label="(text)" meaning it was a formula value.
20160731 16:27:11< vultraz> so if you have label = "(text + 'foo')" in the WML
20160731 16:27:19< celticminstrel> But I guess not.
20160731 16:27:47< vultraz> I think what happens is it tries to evaluate (text + 'foo') for the value of label=
20160731 16:28:02-!- lipkab [~the_new_l@host-91-147-210-58.biatv.hu] has joined #wesnoth-dev
20160731 16:28:17< vultraz> in gui2, the value of label = in WML is the default label for a tcontrol widget if set_label isn't used in pre_show
20160731 16:28:22< celticminstrel> Oh right, it would be .. for string concatenation, not +
20160731 16:28:34< vultraz> if you call set_label again, it resets the label as a tstring
20160731 16:28:42< vultraz> at least I think that's how it works
20160731 16:28:49< vultraz> .., you say?
20160731 16:28:53< celticminstrel> I'm not sure what you're describing.
20160731 16:28:54< celticminstrel> Yeah.
20160731 16:29:29< vultraz> doesn't work
20160731 16:29:47< celticminstrel> What's the normal value of the label key that does work?
20160731 16:30:05< vultraz> What I'm trying to say is the formula variable 'text' is set to the value of the tstring label passed to set_label()
20160731 16:31:43< vultraz> as opposed to the value of the tstring passed to set_label() being evaluated in a formula using the new text as the variable 'text'.
20160731 16:32:21< vultraz> get it?
20160731 16:32:34< celticminstrel> Kinda of? So why is it put in a variable?
20160731 16:32:40< vultraz> I dunno
20160731 16:33:17< vultraz> I was wondering if a new key could be added with a formula to apply to the text every time set_label() is called..
20160731 16:33:50< vultraz> or, you could have a 'new_text' variable, I guess...
20160731 16:33:58< lipkab> iceiceice: "do you know a way to estimate how many `.cfg` we currently have?" <- No. I do remember seeing simple test cases attached to bug reports, sometimes in cfg attachments.
20160731 16:34:04< vultraz> label = "(new_text = 'foo')"
20160731 16:34:18< vultraz> er,
20160731 16:34:20< vultraz> +
20160731 16:34:22< vultraz> er
20160731 16:34:23< vultraz> ..*
20160731 16:34:26< lipkab> I agree it's not much of a concern.
20160731 16:35:09< celticminstrel> Having a new_text variable is kinda ludicrous.
20160731 16:35:16< vultraz> true
20160731 16:35:25< lipkab> "what VCS is named after a british slang" <- Heh, when did this happen? The old captcha was about Orwell and 1984.
20160731 16:35:46< vultraz> I guess maybe a label_formula= key or something would work...
20160731 16:36:17< vultraz> then apply that formula every time you call set_label()...
20160731 16:36:17< iceiceice> yeah i still get the 1984 one also, idk what the system is
20160731 16:36:20< vultraz> so like
20160731 16:36:22< vultraz> if you had
20160731 16:36:38< vultraz> label = "foo" label_formula = "(text .. 'bar')"
20160731 16:36:46< vultraz> you'd end up with the label showing as  "foobar"
20160731 16:36:49< celticminstrel> There was also the one about duplicating a sheep.
20160731 16:37:25< vultraz> celticminstrel: I think that's probably the most viable solution, here.
20160731 16:37:27< vultraz> but I dunno
20160731 16:37:46< celticminstrel> Still seems a bit weird though...
20160731 16:38:01< vultraz> it does
20160731 16:38:45< vultraz> but there's no way in the formula engine or really in anything to define a constant formula
20160731 16:38:49< celticminstrel> Do you have a specific use case in mind?
20160731 16:39:01< celticminstrel> Constant formula? What do you mean?
20160731 16:39:02< vultraz> it's always just a formula to evaluate to one value
20160731 16:39:44< vultraz> celticminstrel: WFL, at least in GUI2, is used to get values for keys, basically
20160731 16:39:57< celticminstrel> I know.
20160731 16:40:08< vultraz> there's no way to say 'always make something run through this formula'
20160731 16:40:21< vultraz> like
20160731 16:41:21< vultraz> func some_formula = (foo .. bar), label = "(some_formula((text)) where text = [Initial text here])"
20160731 16:41:46< celticminstrel> No need for the double parentheses there.
20160731 16:42:01< vultraz> in that example, label would be evaluated as a formula
20160731 16:42:04< vultraz> not a tstring
20160731 16:42:29< celticminstrel> The facility for this already exists... though WFL does not support tstrings...
20160731 16:42:29< vultraz> it's a very Anura-esq example, really
20160731 16:42:34< celticminstrel> But...
20160731 16:43:01< celticminstrel> I guess you're saying that the formulas are evaluated only when the dialog is built.
20160731 16:43:02< celticminstrel> Again, did you have a specific use case in mind?
20160731 16:44:51< vultraz> celticminstrel: in https://github.com/wesnoth/wesnoth/commit/639548a108eb05c32596be59359b373fb0248670 I converted the attack dialog to use unit_preview_pane. As you can see, I introduced an "image_facing=" key for that widget. However, I had wondered if it were possible, inside the "unit_minimal_right" definition of the widget, to, for the image widget type_image, to do this: label = "(text...
20160731 16:44:52< vultraz> ..... '~FL(horiz)')"
20160731 16:45:24< vultraz> label = "(text .. '~FL(horiz)')" if that went on two lines for you
20160731 16:45:30< celticminstrel> Maybe call it image_mod instead of image_facing.
20160731 16:45:37< celticminstrel> Or mods
20160731 16:45:48< vultraz> well I just do "left" or "right"
20160731 16:46:15< vultraz> I suppose I could write the IPFs directly in the WML
20160731 16:46:19< vultraz> you're right, that might be cleaner
20160731 16:46:26< vultraz> I'll change it tomorrow
20160731 16:46:34< celticminstrel> If it's just about image mods, I think an image_mods key might indeed be easier.
20160731 16:46:55< celticminstrel> If it was about adding a colon or something, I probably would've suggested a label_suffix key.
20160731 16:47:31< vultraz> this is good for this situation, where we want to append a string
20160731 16:47:37< vultraz> but what if, say, I wanted to do..
20160731 16:48:50< vultraz> image_mods = "~RC(magenta>color)". No way I could ever set the 'color' to a variable.
20160731 16:48:56< celticminstrel> Though, leaving even colons untranslated is a little dubious.
20160731 16:49:16< celticminstrel> Is there a color variable?
20160731 16:49:23< vultraz> even if I did "where color =", again, the formula would be run once and evaluate 'color' to ""
20160731 16:50:00< celticminstrel> Hmm.3
20160731 16:50:54< vultraz> celticminstrel: so, building upon this idea, I wonder if it's in any way useful to consider treating WML keys as WFL formulas in general, like Anura does, instead of strings.
20160731 16:51:17< celticminstrel> I think that would have a lot of potential to break things.
20160731 16:52:20-!- molt [~molt@37.122.170.190] has quit [Quit: Leaving]
20160731 16:52:23< vultraz> celticminstrel: or perhaps at least expanding on the () (or is it $()?) syntax to allow something like that
20160731 16:52:35< vultraz> (I dunno why GUI2 uses () but in other WML you need $())
20160731 16:53:10< celticminstrel> I think the trouble you're having here might be because of the fundamental design of GUI2.
20160731 16:53:21< celticminstrel> GUI2's () is different from WML's $().
20160731 16:54:01< vultraz> that's bad
20160731 16:54:15< celticminstrel> Both are formulas, yes, but () has access to context-dependent variables (and only works in contexts that use the tformula class) while $() has access to no variablrs (and only works in contexts where variable substitution is available).
20160731 16:54:34< celticminstrel> You'd use the () syntax if allowing formula in more places in existing keys.
20160731 16:54:56< celticminstrel> The () is necessary to notice that it is, in fact, a formula and not just some string that looks like it could be a formula.
20160731 16:55:03< vultraz> Right
20160731 16:55:12< vultraz> So, what about making () global
20160731 16:55:33< celticminstrel> That entails wiring tformula straight into the config class, which I don't think is a great idea.
20160731 16:55:34< vultraz> ie, expand the config class to use tformula.
20160731 16:55:37< vultraz> heh
20160731 16:56:18< celticminstrel> Besides, the variables that it could have access to are context-dependent, so wiring it into the config class would not be all that useful, honestly.
20160731 16:56:19< vultraz> I think it's worth considering, at least
20160731 16:56:38< celticminstrel> I suppose you could add a .get(key, variables) method, but...
20160731 16:56:50< vultraz> I'm gonna get some sleep
20160731 16:56:55< celticminstrel> Well, bye.
20160731 16:57:06< vultraz> Let's consider this idea further later.
20160731 16:57:26< celticminstrel> It also goes counter to my idea for separating WML into a library.
20160731 16:57:38< vultraz> I think allowing global access to the formula engine has a possibility to expand WML's functionality 
20160731 16:57:39< vultraz> for example
20160731 16:57:45< vultraz> inline lua
20160731 16:58:00< celticminstrel> How do you get from WFL to inline Lua?
20160731 16:58:01< vultraz> key = "eval_lua( LUA STUFF HERE )"
20160731 16:58:04< vultraz> er
20160731 16:58:14< vultraz> key = "(eval_lua())"
20160731 16:58:45< celticminstrel> ActionWML tags can easily be expanded to allow formulas, now that Lua can evaluate them.
20160731 16:59:05< celticminstrel> Though that API is still a bit shaky in some ways.
20160731 17:00:13< vultraz> What I just said would be a generalization and improvement on the SUF lua_function= key 
20160731 17:00:33-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160731 17:00:54< celticminstrel> Eh.
20160731 17:01:48< vultraz> so, for example... key = "(eval_lua( return "foo" ))" would be a very roundabout way of saying key = "foo"
20160731 17:01:55< vultraz> but you see my point
20160731 17:02:13< celticminstrel> I see your point but don't entirely agree.
20160731 17:02:46< vultraz> Well again, we can polish the idea
20160731 17:02:55< celticminstrel> BTW, if you think of a good syntax for translatable strings that doesn't conflict with anything else in WFL, do tell.
20160731 17:03:02< vultraz> anyway
20160731 17:03:04< vultraz> > sleep
20160731 17:03:15< celticminstrel> Anura's syntax won't work because I used ~ for something else.
20160731 17:03:37< vultraz> it might be useful to draw up a document if we're serious about expanding the fundamentals of WML
20160731 17:03:52< celticminstrel> What sort of document?
20160731 17:10:16-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Read error: Connection reset by peer]
20160731 17:10:33-!- iceiceice [~chris@50.245.222.235] has joined #wesnoth-dev
20160731 17:10:33-!- iceiceice [~chris@50.245.222.235] has quit [Changing host]
20160731 17:10:33-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20160731 17:11:13< irker100> wesnoth: Bär Halberkamp wesnoth:master fd1a77699717 / / (5 files in 3 dirs): Change attack sounds for short swords https://github.com/wesnoth/wesnoth/commit/fd1a77699717210f4b235e812a99bea0ed590e7d
20160731 17:17:47< vultraz> bumbadadabum: did you see zookeeper's comments?
20160731 17:17:52< vultraz> celticminstrel: design document 
20160731 17:18:05< bumbadadabum> vultraz: no
20160731 17:18:21< zookeeper> umm: https://github.com/wesnoth/wesnoth/commit/fd1a77699717210f4b235e812a99bea0ed590e7d#diff-6b306396aa9fdbc560492611d41dd56fR287
20160731 17:18:22< vultraz> [20:24:42]	zookeeper	bumbadadabum, actually, what should be done is to replace all dagger-swish.wav with knife.ogg.
20160731 17:18:24< vultraz> [20:29:42]	zookeeper	i'd rather have sword-1.ogg on everything labeled as a sword, and knife.ogg on everything labeled as a knife/dagger. the latter wasn't intended only for the thunderer line, applying it to all the other knives was just forgotten.
20160731 17:18:57< bumbadadabum> zookeeper: I personally like dagger-swish
20160731 17:19:00< bumbadadabum> for daggers, at least
20160731 17:19:19< bumbadadabum> oh shit zookeeper I didn't see that
20160731 17:19:41< celticminstrel> bumbadadabum: You also introduced merge conflict markers into the changelog.
20160731 17:20:31< celticminstrel> Please fix.
20160731 17:20:32< zookeeper> bumbadadabum, but dagger-swish isn't really a hit sound; it's not hitting anything, it's just a swish.
20160731 17:20:54< bumbadadabum> celticminstrel: yeah trying
20160731 17:21:07< bumbadadabum> I looked at the diff but missed that I'm sorry
20160731 17:21:36< bumbadadabum> zookeeper: with a get hit sound it works in my opinion
20160731 17:23:12< irker100> wesnoth: Bär Halberkamp wesnoth:master 51ebabae4d4a / changelog: fix last commit https://github.com/wesnoth/wesnoth/commit/51ebabae4d4a316a1521962b46085f742e3371eb
20160731 17:23:44< celticminstrel> Maybe I should've run spellcheck on the changelog as well as release notes.
20160731 17:25:17< bumbadadabum> zookeeper: idk I wanted to have a little more variety
20160731 17:25:51< bumbadadabum> basically add knife.ogg into a hierarchy of dagger-swish -> knife -> sword-1
20160731 17:25:55< bumbadadabum> when it comes to blade size
20160731 17:26:33< zookeeper> mm-hmh
20160731 17:33:46-!- Kwandulin [~Miranda@p200300760F6062F2E48E7F2D0FF9198C.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160731 17:53:07< celticminstrel> Looks like it builds, but I've probably broken everything.
20160731 17:53:44< celticminstrel> Oh, that was just the server.
20160731 17:53:58< celticminstrel> I forgot that copy precedes compile.
20160731 17:54:14< celticminstrel> ...might be better if it didn't.
20160731 18:05:09< irker100> wesnoth: Jyrki Vesterinen wesnoth:fade-out-music d20f62af06a6 / changelog data/core/macros/sound-utils.cfg players_changelog: Fade out storyscreen music on scenario start https://github.com/wesnoth/wesnoth/commit/d20f62af06a6d2032c2c1fbe2b72633d183ba377
20160731 18:09:34< vultraz> *is still awake*
20160731 18:09:45< vultraz> celticminstrel: btw, is this still an issue? https://gna.org/bugs/?18679
20160731 18:10:09< celticminstrel> Can you quote the title for me?
20160731 18:10:33< vultraz> "If there is more than one [ai] block in a side tag, all of them are ignored. (This is true at least with [avoid]. Most other ai keys and tags have less obvious effects, so it's hard to tell.)"
20160731 18:10:50< celticminstrel> Shouldn't be.
20160731 18:11:30< celticminstrel> Pretty sure there's something like for(config c : cfg.child_range("ai"))
20160731 18:11:52< celticminstrel> If you want to verify it should be uh...  
20160731 18:12:02< celticminstrel> src/ai/configuration.cpp
20160731 18:12:37< celticminstrel> (And of course you could try a test scenario with two ai tags each containing a different avoid, if you want to go that far. :P )
20160731 18:13:13< vultraz> for (const config &ai_configuration : input.child_range("ai")) {
20160731 18:13:15< vultraz> this?
20160731 18:13:25< celticminstrel> Actually, any MP game in a scenario that specifies AI parameters will have more than one [ai] tag, because the MP setup injects one.
20160731 18:13:27< celticminstrel> Yeah, that.
20160731 18:14:03< celticminstrel> That looks like the right place.
20160731 18:16:04< vultraz> I'll have to confirm the fix but I'm guessing it can be closed.
20160731 18:16:23< celticminstrel> Yeah.
20160731 18:25:39< celticminstrel> Well, the shaman still has traits, but (unsurprisingly) is missing heal and slows.
20160731 18:26:01< celticminstrel> I guess the trait portion is commit-worthy then.
20160731 18:26:27< celticminstrel> The loyal trait now shows up.
20160731 18:26:39< celticminstrel> And, heh, says "Availability: Unavailable".
20160731 18:27:02< celticminstrel> But other traits also say that now which shouldn't, like elemental.
20160731 18:27:10< celticminstrel> And fearless.
20160731 18:27:19< celticminstrel> Well... this is why I thought it was a bad idea to put that there, though...
20160731 18:27:57< celticminstrel> I think it'd be better to list the traits available to each race (and, potentially, unit type).
20160731 18:30:16< celticminstrel> ...oh right, I should test a bit more, to make sure that race-specific traits and unit-type-specific traits do work correctly.
20160731 18:30:29< celticminstrel> I think she had dextrous, so maybe the latter before the former.
20160731 18:31:27-!- Shiki [~Shiki@141.39.226.227] has joined #wesnoth-dev
20160731 18:38:10-!- hk238 [~kvirc@t224.ip7.netikka.fi] has quit [Quit: http://www.kvirc.net/ 4.9.1 Aria]
20160731 18:39:36< celticminstrel> Why is Dark Adept not allowed to have the strong trait, but Dark Sorcerer is?
20160731 18:40:21-!- travis-ci [~travis-ci@ec2-54-80-189-234.compute-1.amazonaws.com] has joined #wesnoth-dev
20160731 18:40:22< travis-ci> wesnoth/wesnoth#10033 (fade-out-music - d20f62a : Jyrki Vesterinen): The build passed.
20160731 18:40:22< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/148706215
20160731 18:40:22-!- travis-ci [~travis-ci@ec2-54-80-189-234.compute-1.amazonaws.com] has left #wesnoth-dev []
20160731 18:41:22< celticminstrel> Just an oversight due to it being unusual to have Dark Sorcerers available to recruit?
20160731 18:45:16< zookeeper> because adept has no melee attack
20160731 18:45:30< celticminstrel> So it's intentional that dark sorcerers can be strong?
20160731 18:45:33< zookeeper> yes
20160731 18:45:38< zookeeper> because on them the trait is not wasted
20160731 18:45:39< celticminstrel> Kinda weird, but okay.
20160731 18:45:54< celticminstrel> Well, it's not wasted on dark adepts either, if you get them to level up.
20160731 18:47:40< zookeeper> sure. but it's still enough of a bummer to get a strong adept that it was removed... although i don't recall how much of the rationale was that, and how much was the fact that if they're supposed to be so feeble that they don't even get a melee attack, a strong adept would be a bit of an oxymoron
20160731 18:50:09< bumbadadabum> <celticminstrel> And, heh, says "Availability: Unavailable".
20160731 18:50:09< bumbadadabum> <celticminstrel> But other traits also say that now which shouldn't, like elemental.
20160731 18:50:11< bumbadadabum> wait what
20160731 18:50:22< celticminstrel> It's my fault, don't worry about it too much.
20160731 18:50:25< bumbadadabum> ohh
20160731 18:50:26< bumbadadabum> ok
20160731 18:50:36< celticminstrel> Because I put definitions of those traits at toplevel which have availability=none
20160731 18:58:29< bumbadadabum> did you also fix the thing with num_traits
20160731 18:58:46< bumbadadabum> because I was going to look into that
20160731 19:02:12< bumbadadabum> ok seems like the num_traits thing is just a one-line fix
20160731 19:02:23< bumbadadabum> so first of all, it checks if num_traits > 0
20160731 19:02:40< bumbadadabum> then it iterates all the traits, and checks if type_.num_traits() - must_have_traits.size() > 0;
20160731 19:02:58< bumbadadabum> so if that's 0, the musthave traits will still be displayed, but the others will never
20160731 19:03:20< celticminstrel> I didn't look at the num_traits thing.
20160731 19:12:53< celticminstrel> bumbadadabum: Where is that?
20160731 19:13:03< bumbadadabum> help_topic_generators.cpp
20160731 19:13:08< celticminstrel> Ah.
20160731 19:13:11< bumbadadabum> I'm testing to see if removing the first check works
20160731 19:13:17< bumbadadabum> but full recompile so zzz
20160731 19:13:23< celticminstrel> Not sure if I'll be touching that file.
20160731 19:26:49< vultraz> bumbadadabum: seems to be no trait page for feral
20160731 19:27:01< bumbadadabum> that's weird because I can see it
20160731 19:27:35< vultraz> huh
20160731 19:28:29< bumbadadabum> wait what
20160731 19:28:31< bumbadadabum> now I can't
20160731 19:28:32< bumbadadabum> that's weird
20160731 19:28:35< bumbadadabum> I saw it before
20160731 19:42:50< bumbadadabum> type.possible_traits()
20160731 19:42:58< bumbadadabum> this doesn't seem to contain the traits defined in the unit itself
20160731 19:43:03-!- gfgtdf [~chatzilla@x4e363d84.dyn.telefonica.de] has joined #wesnoth-dev
20160731 19:45:53< bumbadadabum> fixing that as well
20160731 19:45:56< gfgtdf> vultraz: at gh issue migration: i also think that 'No anonymous submissions' is serious problem, but afaik gna also has that problem since october 2015.
20160731 19:47:38< irker100> wesnoth: Bär Halberkamp wesnoth:master 7dbf17f3fea5 / src/ (help/help_impl.cpp help/help_topic_generators.cpp units/types.cpp): Fix traits in help https://github.com/wesnoth/wesnoth/commit/7dbf17f3fea5605556b6d8c1162bd6bae9567890
20160731 19:49:11< bumbadadabum> celticminstrel, vultraz: this should fix the issues with bats and mudcrawlers
20160731 19:50:41< vultraz> bumbadadabum: sweet :D
20160731 19:51:31< vultraz> gfgtdf: why do you think it's a problem?
20160731 19:52:18< gfgtdf> vultraz: becasue if we allowed anynymous reports we will get  informed about more problems.
20160731 19:52:38< vultraz> bumbadadabum: confirmed fix, can you close that bug report about the mudcrawlers
20160731 19:52:46< vultraz> bumbadadabum: well, mark as fixed.
20160731 19:52:48< bumbadadabum> done
20160731 19:52:54< bumbadadabum> DID THAT BEFORE YOU ASKED IT EVEN
20160731 19:52:57< vultraz> :D
20160731 19:53:00< vultraz> kudos
20160731 19:53:07< shadowm> gfgtdf: And also get more spam and hit-and-run posters.
20160731 19:53:07< vultraz> gfgtdf: that is true
20160731 19:53:15< vultraz> that is also true
20160731 19:53:28< vultraz> though I don't think we've gotten many of those
20160731 19:53:42< gfgtdf> shadowm: i sctuall never saw a hit and run poster in the old bugtrakcer
20160731 19:53:47< gfgtdf> actually*
20160731 19:53:49< shadowm> I've seen a lot.
20160731 19:54:04< vultraz> But for purposes of organization as well as contacting people should more info be needed, I think having an account attached to a report is good.
20160731 19:54:12< celticminstrel> That conflicts with what I'm doing... oh well, should be easy to fix.
20160731 19:54:12< gfgtdf> shadowm: wait you mean something differnet than spammers by that ?
20160731 19:54:19< shadowm> Yes, obviously.
20160731 19:54:30< vultraz> oh
20160731 19:54:42< vultraz> ah, then yes
20160731 19:54:47< shadowm> You could probably start by looking at every report that's been marked Need Info (and some Invalid and Won't Fix).
20160731 19:55:07< shadowm> As well as those that have spent too long as Ready For Test.
20160731 19:56:00< shadowm> If people can post without doing as much as giving you an email address that the tracker will notify on every reply/action, they're bound to forget about their own reports and leave the devs talking to themselves.
20160731 19:56:10< vultraz> ^
20160731 19:56:43< irker100> wesnoth: Jyrki Vesterinen wesnoth:fade-out-music 044020a9c601 / changelog data/core/macros/sound-utils.cfg players_changelog src/sound.cpp: Fade out music on scenario start and on disabling it https://github.com/wesnoth/wesnoth/commit/044020a9c6017947911eeeb3f91dc3391ffb5e90
20160731 19:57:00< gfgtdf> shadowm: even if, we can easiyl close then if we cannot repreudce and its still worth for those bug reports that are useful
20160731 19:57:10< gfgtdf> shadowm: also forcing account creating doesnt nessacrily make it better (for iirc my gna account still has a 10minutemail.com emain associated to it)
20160731 19:57:17< shadowm> We used to get more spam back in the day. I'm not sure if that stopped because Gna changed something in their configuration, or because Gna is too shitty even for spammers.
20160731 19:58:33< shadowm> I do remember that they instituted some draconian policies on mailing lists as well, including blocking all @hotmail.com addresses.
20160731 19:58:55< shadowm> So that might be related. (In case you're wondering, I believe they lifted that four or three years ago.)
20160731 19:59:23< iceiceice> omg so much template pain
20160731 19:59:27< celticminstrel> Speaking of mailing lists, where are those hosted?
20160731 19:59:33< celticminstrel> Like wesnoth-dev and stuff.
20160731 19:59:40< shadowm> They are obviously Gna.org addresses. :p
20160731 19:59:52< celticminstrel> Yeah, though so...
20160731 20:00:21< shadowm> We have the facilities to host them on our own, though, it's just that the person in charge of them is almost never around. (AI0867) We even have a half-working replacement for the old commits ML.
20160731 20:01:14< shadowm> I did figure some time ago that the simplest way to get him to do anything is to intentionally botch a system upgrade so that cron emails root@ every single day.
20160731 20:03:17< iceiceice> maybe should request github to host email lists for us :)
20160731 20:11:08-!- lipkab [~the_new_l@host-91-147-210-58.biatv.hu] has quit [Quit: Leaving]
20160731 20:18:59-!- JyrkiVesterinen [~JyrkiVest@87-100-209-170.bb.dnainternet.fi] has quit [Quit: Going to bed]
20160731 20:26:02-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160731 20:26:03-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Read error: Connection reset by peer]
20160731 20:32:16< celticminstrel> Does the game support [lua] tags inside [unit_type]?
20160731 20:37:43-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160731 20:42:08-!- ancestral [~ancestral@189.sub-174-219-6.myvzw.com] has joined #wesnoth-dev
20160731 20:42:30< celticminstrel> I guess it doesn't matter after all.
20160731 20:42:56< celticminstrel> Hi ancestral 
20160731 20:43:02< ancestral> Hi
20160731 20:43:08< ancestral> I should have a build tonight
20160731 20:43:14< ancestral> Yay working Sunday
20160731 20:43:19< ancestral> (day job)
20160731 20:43:27< celticminstrel> And then I'll see if it works on 10.7?
20160731 20:43:32< ancestral> Yeah
20160731 20:43:41< ancestral> One possibility is to just use some of your libs
20160731 20:43:49< ancestral> But I can try rebuilding the pango/cairo stuff
20160731 20:44:28< celticminstrel> If you can point it to 10.7 SDK instead of system-global libs, sure.
20160731 20:45:16< celticminstrel> My libs might not be suitable for general use, but maybe that could be fixed with a little install_name_tool.
20160731 20:47:51-!- ancestral [~ancestral@189.sub-174-219-6.myvzw.com] has quit [Quit: i go nstuf kthxbai]
20160731 20:48:44-!- travis-ci [~travis-ci@ec2-54-80-189-234.compute-1.amazonaws.com] has joined #wesnoth-dev
20160731 20:48:45< travis-ci> wesnoth/wesnoth#10035 (fade-out-music - 044020a : Jyrki Vesterinen): The build has errored.
20160731 20:48:45< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/148720694
20160731 20:48:45-!- travis-ci [~travis-ci@ec2-54-80-189-234.compute-1.amazonaws.com] has left #wesnoth-dev []
20160731 20:57:28-!- atarocch [~atarocch@93.56.160.28] has quit [Remote host closed the connection]
20160731 21:14:53< irker100> wesnoth: Charles Dang wesnoth:master 6080cfb969f6 / / (4 files in 3 dirs): Switch unit_preview_pane image_facing key to image_mods https://github.com/wesnoth/wesnoth/commit/6080cfb969f6cfbe06c59af3a04a60b3c3ea8764
20160731 21:15:32< vultraz> muuuch cleaner 
20160731 21:20:40< celticminstrel> I'm getting vaguely annoyed by the fact that I've made a copy of all the traits when the intent was to avoid that...
20160731 21:21:00< celticminstrel> To fix that I need to do something that avoids needing the availability key at trait generation time.
20160731 21:28:02< iceiceice> celticminstrel, regarding "lua" inside of unit_type
20160731 21:28:07< iceiceice> what would that even mean
20160731 21:28:24< celticminstrel> Same as events inside of unit_type.
20160731 21:28:42< iceiceice> the events inside of unit_type should be majorly refactored also though
20160731 21:28:50< iceiceice> that was a serious blunder the way it was done
20160731 21:29:00< iceiceice> and it doesn't really work properly
20160731 21:29:05< celticminstrel> How should they be refactored in your opinion?
20160731 21:29:09< iceiceice> so right now,
20160731 21:29:13< iceiceice> when unit is constructed,
20160731 21:29:23< iceiceice> it accesses via global variables the event queue
20160731 21:29:35< iceiceice> and registers any events associated to the unit i guess
20160731 21:29:48< celticminstrel> Unless events for that unit had already been registered.
20160731 21:29:48< iceiceice> or maybe inside the type, it's getting a little fuzzy
20160731 21:30:02< iceiceice> yeah theres some kind of like, checklist so that the same unit doesn't get done more than once
20160731 21:30:11< celticminstrel> Yeah.
20160731 21:30:12< iceiceice> but its a really bad idea to make the construction of an object have global side effects
20160731 21:30:29< zookeeper> [event] in [unit_type] doesn't work properly? that's news to me... i think
20160731 21:30:44< iceiceice> yes, tekelili reported all kinds of bugs upon the unit advancing
20160731 21:30:56< iceiceice> presumably because advancing doesn't call a new constructor
20160731 21:31:04< celticminstrel> Oh.
20160731 21:31:04< zookeeper> hrhm
20160731 21:31:12< iceiceice> i just told him, look, whoever implemented this didn't really know C++ that well
20160731 21:31:33< vultraz> iceiceice, celticminstrel: this is exactly the kind of thing we need fixed for 1.14.0
20160731 21:31:36< zookeeper> it's a great feature so i hope it can be simply fixed somehow
20160731 21:32:05< iceiceice> idk exactly how it should work like,
20160731 21:32:09< vultraz> remember, it's imperative we make 1.14 as well-functioning as possible.
20160731 21:32:13< iceiceice> if its a side effect of the unit tag being evaluated
20160731 21:32:17< iceiceice> rathre than the unit being constructed,
20160731 21:32:28< iceiceice> or what, but like,
20160731 21:32:46< iceiceice> i tmeans that if you want to consturct an object of type unit and the event queue doesnt exist, i think you just segfault
20160731 21:32:56< iceiceice> for a long time i wanted to try to implement like, a simulation based AI
20160731 21:33:02< celticminstrel> This would be horribly inefficient probably, but you could loop through all units on the map during event pump and see if any of their type's events should fire,
20160731 21:33:34< iceiceice> but i've basically concluded htat simulation based ai can't work in wesnoth without rewriting all the code
20160731 21:33:38< iceiceice> because its all based on these global variables
20160731 21:33:55< iceiceice> celticminstrel, that cant be less efficient than it was when i started working,
20160731 21:34:01< iceiceice> it used to be that every "unit_ptr" was actually a "unit"
20160731 21:34:18< celticminstrel> Wait, so units were being copied left right and centre?
20160731 21:34:22< iceiceice> and so every time you bring up any dialog that requires rendering a unit , it would call the ctors and presumably go to the event queue and register a ton of events
20160731 21:34:23< iceiceice> yes
20160731 21:34:36< iceiceice> there were also strange things in lua, like,
20160731 21:34:48< iceiceice> someone would say, i just want to make a unit of some type thats not on the map to check its hitpoints or something
20160731 21:34:58< iceiceice> but instead of that being a "private" lua operation
20160731 21:35:03< iceiceice> it actually changes the game state and causes OOS
20160731 21:35:17< celticminstrel> Hmm, so...
20160731 21:35:52< celticminstrel> Events are currently accounted for in advance_to.
20160731 21:37:10< celticminstrel> Which includes [transform_unit].
20160731 21:37:18< celticminstrel> Not sure if it covers variation changes.
20160731 21:38:43< vultraz> trying to figure out why the recall code has this
20160731 21:38:44< vultraz> std::shared_ptr<std::vector<unit_const_ptr> >
20160731 21:38:50< celticminstrel> It does cover variation changes.
20160731 21:39:09< celticminstrel> So at the moment it seems like events are checked for whenever a unit's type changes.
20160731 21:39:19< celticminstrel> I believe the constructor also calls advance_to.
20160731 21:39:40< celticminstrel> It does.
20160731 21:46:29< celticminstrel> Hmm, if I use a two-level split, maybe I can make this work...
20160731 21:48:56< vultraz> why the hell is this a shared_ptr
20160731 21:49:13< celticminstrel> No idea, I'd have to examine the code more closely.
20160731 21:51:09< vultraz> https://github.com/wesnoth/wesnoth/blob/master/src/menu_events.cpp#L617
20160731 21:54:11< vultraz> \what exactly does a shared_ptr even..do
20160731 21:54:33< celticminstrel> So, I can't see any reason why it's a shared pointer in that function.
20160731 21:55:00< celticminstrel> There could be a reason in dialogs::recall_dialog, though I'm inclined to doubt it.
20160731 21:56:12< celticminstrel> As for what a shared_ptr does, it just makes sure it's deleted when you've finished with it.
20160731 21:56:29< celticminstrel> (It can actually be used with things like SDL_Surface if you use a custom deleter.)
20160731 21:58:42< vultraz> oh
20160731 21:58:54< Aginor> hey
20160731 21:58:56< vultraz> I thought it had to do with being shared across... somethings
20160731 21:59:11< Aginor> just to be stirring, why github instead of trying to get a jira instance?
20160731 21:59:15< celticminstrel> Well, the "shared" means that other people might hold a copy of the same pointer.
20160731 21:59:44< vultraz> Aginor: because it's better to consolidate our infrastructure!
20160731 22:00:30< iceiceice> vultraz, i probably wrote that
20160731 22:00:33< iceiceice> as a work around for something
20160731 22:00:39< iceiceice> i vaguely remember doing it :)
20160731 22:00:41< vultraz> behold the Ugly C++ Of The Day:
20160731 22:00:44< vultraz> 	find_widget<tunit_preview_pane>(&window, "unit_details", false)
20160731 22:00:45< vultraz> 		.set_displayed_unit(&*((*recall_list_)[selected_row]));
20160731 22:00:55< vultraz> (written by me just now)
20160731 22:01:17< gfgtdf> vultraz: you are currently porting the recall dialog to gui2 ?
20160731 22:01:22< vultraz> yes
20160731 22:01:41< celticminstrel> BY &*?
20160731 22:01:43< celticminstrel> ...
20160731 22:01:51< celticminstrel> How did I get "BY" from "Why"?
20160731 22:03:21< vultraz> needed pointer
20160731 22:03:29< celticminstrel> Instead of iterator, or something?
20160731 22:03:30< iceiceice> vultraz, heres my nominee
20160731 22:03:30< iceiceice> http://pastebin.com/D0iQQFp9
20160731 22:03:32< vultraz> I think I can simplify that, actually
20160731 22:03:46< celticminstrel> If you have a shared_ptr, get the pointer with .get()
20160731 22:03:47< vultraz> iceiceice: P_P
20160731 22:03:49< vultraz> the fuck is that
20160731 22:04:00< celticminstrel> (Same if it's an intrusive_ptr, I think.)
20160731 22:04:03< vultraz> (pardon the expletive)
20160731 22:04:10< vultraz> celticminstrel: oh?
20160731 22:04:20< vultraz> don't you mean std::get?
20160731 22:04:23< celticminstrel> No.
20160731 22:04:25< celticminstrel> ptr.get()
20160731 22:04:38< celticminstrel> Where ptr is a shared_ptr or a unique_ptr or (probably) an intrusive_ptr.
20160731 22:04:39< iceiceice> it is the thing that is equivalent to "boost::apply_visitor" right now
20160731 22:04:41< iceiceice> in my variant
20160731 22:04:56< iceiceice> mine was not properly respecting R-value ness so i had to rewrite it
20160731 22:04:59< iceiceice> which involved a lot of template spew
20160731 22:05:13< vultraz> heh
20160731 22:05:33< vultraz> celticminstrel: funny enough, this is a shared_ptr containing a vector of intrusive_ptrs :P
20160731 22:05:57< iceiceice> at least you are on C++11 now :)
20160731 22:06:04< celticminstrel> Though I think we established that the shared_ptr wrapper could be dropped.
20160731 22:06:05< vultraz> yes
20160731 22:06:10< iceiceice> in the old code, everytime that vector was resized it would like twiddle all the reference counts
20160731 22:06:11< vultraz> celticminstrel: perhaps
20160731 22:06:23< iceiceice> and in the time before we even had unit_ptr it would like copy all the units
20160731 22:06:27< vultraz> celticminstrel: if iceiceice can tell me why it was a shared_ptr
20160731 22:06:32-!- mjs-de [~mjs-de@x4db5d13b.dyn.telefonica.de] has quit [Remote host closed the connection]
20160731 22:06:37< celticminstrel> You think he would know?
20160731 22:06:42< iceiceice> vultraz, you might need to look at commit history
20160731 22:06:51< celticminstrel> Oh, that might be an idea.
20160731 22:06:52< iceiceice> maybe someone fixed whatever caused it to be a shared_ptr
20160731 22:07:10< celticminstrel> git blame filename -L44,55
20160731 22:07:16< celticminstrel> Where 44 and 55 are line numbers
20160731 22:07:57< gfgtdf> i wonder where the units_list_preview_pane ctor that takes a unit unit_const_ptr u is used?
20160731 22:08:01< celticminstrel> Does Boost somewhere have iterator range concatenation?
20160731 22:08:20< celticminstrel> So that I can produce a single iterator range that's the concatenation of two or more others?
20160731 22:08:59< gfgtdf> celticminstrel: there is boost::join
20160731 22:10:50< vultraz> gah
20160731 22:10:52< vultraz> git blame is useless
20160731 22:11:35< iceiceice> yeah git blame is sometimes hard to use
20160731 22:11:42< iceiceice> i like to just click in github web interface
20160731 22:11:49< Aginor> can we have non-public issues in github?
20160731 22:12:06< iceiceice> if theres not too many commits then thats not too bad
20160731 22:23:38-!- Duthlet [~Duthlet@dslb-178-005-055-087.178.005.pools.vodafone-ip.de] has quit [Quit: leaving]
20160731 22:23:44< vultraz> Aginor: don't think so
20160731 22:24:28< celticminstrel> Join?
20160731 22:24:33 * vultraz now has a preliminary implementation of the recall dialog done.
20160731 22:24:35< celticminstrel> Wher's that?
20160731 22:26:01< Aginor> vultraz: I'd consider that a serious deficency in their issue tracker. It takes away our ability to handle security vulnerabilities without having them public the entire team
20160731 22:26:24< Aginor> I'd also like to see proper issue relationships and priorities which it doesn't look like github has either
20160731 22:26:27< celticminstrel> GitHub does have private repos...
20160731 22:26:51< Aginor> celticminstrel: hidden issues in the tracker, not repos
20160731 22:26:59< celticminstrel> I know
20160731 22:27:39< celticminstrel> I was thinking along the lines of "If they have private repos, then maybe they'd have private issues"
20160731 22:30:33< Aginor> yeah, that's fair enough
20160731 22:30:41< Aginor> still a bit of hassle for administration
20160731 22:30:52< Aginor> not to mention that it suddenly costs money
20160731 22:31:14< vultraz> Aginor: issue relationships?
20160731 22:31:19< vultraz> GH has tags
20160731 22:32:16< Aginor> vultraz: that's not quite the same thing
20160731 22:33:08< vultraz> I'll admit we lose a few features
20160731 22:33:14< vultraz> But I think it's worth it
20160731 22:33:18< gfgtdf> tags are quite nice, for example on github i often don't know in which category to put bugs, so like whe i have a bug in a campaign on github tracker i can just put tag them with both campaign and mp.
20160731 22:33:39< vultraz> BTW, high-priority issues shouldn't be discussed in a public repo.
20160731 22:33:41< vultraz> Just saying.
20160731 22:34:08< gfgtdf> bot sortable categories like 'prority' als also a quiet useful feature too.
20160731 22:34:22< gfgtdf> s/on github i often/on gna i often 
20160731 22:34:50< vultraz> it would be nice if GH allowed issues only visible to certain groups, tho
20160731 22:37:59< celticminstrel> Issue relationships are completely different from tags.
20160731 22:38:22< celticminstrel> It's for saying eg "Bug #45 is a duplicate of #42".
20160731 22:38:43< Aginor> or Bug #56 is blocked by #35
20160731 22:38:44< celticminstrel> I guess you could sorta get the effect on GH by referencing one from the other in a comment.
20160731 22:38:51< celticminstrel> Not sure if that's good enough.
20160731 22:38:56< Aginor> Or #64 and #23 are related
20160731 22:39:12< gfgtdf> celticminstrel: hmm but that soudns liek gh is at lest better than gna on this regard
20160731 22:39:24< celticminstrel> I've used JIRA before, so I wouldn't mind switching to that, either.
20160731 22:39:25< gfgtdf> like*, least*
20160731 22:39:26< Aginor> you don't have any forward and backwards references that way
20160731 22:39:27< celticminstrel> gfgtdf: How so?
20160731 22:39:34< Aginor> Jira would be nice :)
20160731 22:39:39< Aginor> I'd get behind jira
20160731 22:39:47< gfgtdf> celticminstrel: on hitgub you get automatically notifaction 'iths was refecned by issue ...'
20160731 22:40:18< Aginor> https://help.github.com/articles/integrating-jira-with-your-projects/
20160731 22:40:31< celticminstrel> Also gfgtdf, where's boost::join?
20160731 22:41:01< gfgtdf> celticminstrel: boost/range/join.hpp
20160731 22:41:20< celticminstrel> Okay, I'll see if it works...
20160731 22:41:30< vultraz> Aginor: I think something that weighs heavily here is that if we go on Steam, we'll likely get more players. More players = potentially more bug reports. And IMO, it'd be better to have something like GH's tracker for them to report bugs. Most *certainly* not GNA.
20160731 22:42:04< vultraz> User interface is key here.
20160731 22:42:21< celticminstrel> vultraz: JIRA is at least as good as GH in that respect.
20160731 22:43:03< vultraz> perhaps, but I'd still prefer to use GH, if for no other reason than it's a unified management experience between devs and the bugs
20160731 22:43:22< vultraz> hell, you can even have 'Fixes #34' or something in a commit message and it closes that bug!
20160731 22:43:23< celticminstrel> I understand your desire for unification, but that should not be the only consideration.
20160731 22:44:15-!- SeamusFD [~SeamusFD@208.163.165.22] has joined #wesnoth-dev
20160731 22:44:33< celticminstrel> Looks like JIRA supports GH integration, so you might be able to get auto-closing there too.
20160731 22:44:37< vultraz> Aginor: and btw, as gfgtdf said, issues and stuff have an annotation when they're referenced by others.
20160731 22:50:12< shadowm> loonycyborg: Is it you who restarted the 1.13 wesnothd twice in a row or did it die on its own without writing an error for some reason?
20160731 22:50:23< Aginor> vultraz: that only addresses my tertiary concerns though
20160731 22:51:07< shadowm> loonycyborg: Also why on earth does it keep logging these meaningless EOF errors?
20160731 22:52:27< shadowm> Just saying that the client closed the connection unexpectedly or whatever would probably be more helpful.
20160731 22:53:40< shadowm> 17:07:56 <lobby> tail: ‘/home/wesnoth/servers/dev/current.log’ has been replaced;  following end of new file
20160731 22:53:43< shadowm> 17:32:30 <lobby> tail: ‘/home/wesnoth/servers/dev/current.log’ has been replaced;  following end of new file
20160731 22:56:53< shadowm> The first mysteriously restarted instance was servers/dev/logs/wesnothd.20160729-223715.1.13.5-3-g95c4556.log . The second is servers/dev/logs/wesnothd.20160731-210743.1.13.5-3-g95c4556.log .
20160731 22:58:20< iceiceice> i think there should be some acknowledgment that, GNA is totally inadequate, and that to some extent the decision should be a do-ocracy
20160731 22:58:28< iceiceice> i watched for 6 years as people talk this bug tracker thing to death
20160731 22:58:38< iceiceice> and the project suffered as a result
20160731 22:59:00< iceiceice> you dont have to agree wtih GH but if you want to block a move to GH you should be offering to port the bugs somewhere else instead
20160731 22:59:04< iceiceice> IMO
20160731 22:59:33< iceiceice> if vultraz is willing to try to port all the bugs to a new tracker thats a huge effort and a service
20160731 22:59:37< Aginor> iceiceice: I have offered to move us to jira in the past but nobody cared at that time
20160731 22:59:51< iceiceice> i see, i guess i need to look at jira
20160731 22:59:54< iceiceice> i never heard of it before
20160731 23:00:10< vultraz> iceiceice: thanks, but I shall need help to do it. It won't be all me. I'm just trying to get something moving here.
20160731 23:00:30< vultraz> (probably just need one other person, though)
20160731 23:01:01< Aginor> my primary concern is around managing security related issues in a non-public way, if we cannot do that on GH I don't think we should move there.
20160731 23:01:16< vultraz> shadowm: your thoughts? ^
20160731 23:01:33< shadowm> I agree with the last thing Aginor said, haven't been keeping track of the rest and honestly I am not interested in doing that atm.
20160731 23:01:37< Aginor> prioritisation and searching are horrible on both GH and GNA so it wont make any difference ;)
20160731 23:07:28< iceiceice> it sounds like they are working on private issues
20160731 23:07:29< iceiceice> https://github.com/isaacs/github/issues/37
20160731 23:08:04< iceiceice> hmm not sure
20160731 23:08:16< iceiceice> i liked their idea of just encrypting the plaintext :)
20160731 23:08:21< shadowm> I'd like to say I've seen people have very strong opinions on stuff and that's nice because it means they care about the project. But it's not good to let emotions take priority over technical arguments when you're building or working with technology. So, if we're better off running a self-hosted proprietary solution coded in Visual Basic hooked to the slowest RDBMS in existence and keeping ...
20160731 23:08:27< shadowm> ... track of credentials in plain text, that's fine by me as long as it's an informed decision and everyone has agreed to it.
20160731 23:09:24< shadowm> What wouldn't be fine would be if someone in charge made a decision knowing that some people were decidedly unhappy about the possibility.
20160731 23:09:59< shadowm> This has happened before and it's had very negative long-lasting consequences for the project, so it'd be best if it didn't happen before.
20160731 23:10:15< shadowm> s/before\./again./
20160731 23:10:40< iceiceice> what, the move to git?
20160731 23:10:43 * vultraz rubs eyes
20160731 23:11:35< shadowm> If I didn't specify what I'm referring to it's probably because it's not important.
20160731 23:12:00< iceiceice> well i think everyone could understand what you were saying more clearly if you explain what you are referring to
20160731 23:12:50< shadowm> Untrue.
20160731 23:13:14< celticminstrel> Okay, boost::join isn't going to work, because I need to return this from a function and its return type is a super-complicated template class.
20160731 23:13:26< shadowm> I explained the general concept in plain words already, I do not need to give specific examples that are bound to make some people uncomfortable.
20160731 23:13:30< vultraz> Aginor: I'm looking at jira. The interface is nice, I'll grant you.
20160731 23:14:10< celticminstrel> So, the alternatives are: writing my own iterator, copying everything, or using a container of pointers to each trait.
20160731 23:14:42< vultraz> Aginor: doesn't this cost $$ tho?
20160731 23:15:08< shadowm> I believe JIRA is available for free for verified OSS organizations.
20160731 23:15:33-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 276 seconds]
20160731 23:15:58< shadowm> Or maybe it used to be.
20160731 23:16:56< Aginor> it still is
20160731 23:17:15< shadowm> Really? I can't find the page anymore.
20160731 23:17:46< vultraz> Aginor: how's the integration with github?
20160731 23:18:04< vultraz> it says it updates issues if you commit stuff to bitbucket
20160731 23:18:41< Aginor> vultraz: I don't know
20160731 23:19:05< Aginor> I'd suggest we do a study, evaluate them side by side (redmine too for completeness) and then choose the suprior alternative
20160731 23:19:13< Aginor> and pick the criteria that we want
20160731 23:19:16 * vultraz groans
20160731 23:19:18< Aginor> beforehand
20160731 23:19:56< vultraz> can you help with this
20160731 23:20:16-!- celmin [~celticmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20160731 23:20:27< Aginor> https://www.atlassian.com/software/views/open-source-license-request
20160731 23:21:34< vultraz> my plate is full
20160731 23:21:53< vultraz> I can't do a comparison of three trackers, two of which I know nothing about
20160731 23:22:41< Aginor> https://es.atlassian.com/licensing/purchase-licensing#licensing-3
20160731 23:25:43-!- SeamusFD is now known as SeamusFD[Off]
20160731 23:26:06-!- SeamusFD[Off] [~SeamusFD@208.163.165.22] has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
20160731 23:30:24< iceiceice> Aginor, one thing I saw other projects do is just have a special "security" mailing list
20160731 23:30:31< iceiceice> and only people who should know about it are on it
20160731 23:30:40< iceiceice> and users are instructed to just write an email
20160731 23:30:56< iceiceice> i guess thats for project where a security issue is more urgent
20160731 23:31:12< Aginor> we should have something like that too
20160731 23:31:20< iceiceice> if a user reports a security bug on the issue tracker you can just send an email and delete it
20160731 23:31:21< Aginor> so that we can support responsible disclosure
20160731 23:31:44< Aginor> iceiceice: we should still track those issues properly
20160731 23:32:21< iceiceice> you could also retain GNA for those issues i guess
20160731 23:32:39< iceiceice> as long as joe user is not supposed to report general bugs to gna
20160731 23:32:43< iceiceice> i mean gna is fine for developer use i guess
20160731 23:32:49< celmin> Can you delete issues on github?
20160731 23:32:59< celmin> I don't believe you can.
20160731 23:33:04< iceiceice> hmm, i thought you could
20160731 23:33:50< Aginor> iceiceice: two issue trackers is incredibly painful
20160731 23:33:59< Aginor> I've been there in the past and it's really not fun
20160731 23:34:12< Aginor> you end up spending a lot of time grooming them both
20160731 23:34:16< iceiceice> yeah but there are not realisitcally that many security bugs
20160731 23:34:49< Aginor> vultraz: I can help but I'm not going to volunteer to do it all
20160731 23:35:16< iceiceice> my experience has been that even when there is a security bug, no one fixes it :p
20160731 23:35:39-!- gfgtdf [~chatzilla@x4e363d84.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 47.0/20160604131506]]
20160731 23:39:12< shadowm> I found the link via Google, still no idea how you're supposed to get here in the first place or if it's still a thing: https://www.atlassian.com/software/views/open-source-license-request
20160731 23:39:49< shadowm> I'm pretty sure it was far more visible a couple of years ago.
20160731 23:40:34< Aginor> shadowm: look at the links I posted earlier, the second one gives you the trail to find it
20160731 23:48:32-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160731 23:51:05-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 258 seconds]
20160731 23:51:05-!- wedge010 is now known as wedge009
20160731 23:58:27-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
--- Log closed Mon Aug 01 00:00:16 2016