--- Log opened Sat Apr 09 00:00:48 2016
20160409 00:05:11-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160409 00:06:54-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160409 00:07:24-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20160409 00:17:01-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20160409 00:19:21-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160409 00:31:04-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20160409 00:31:10-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160409 00:46:27-!- Ravana_ [~Ravana@unaffiliated/ravana/x-2327071] has quit [Disconnected by services]
20160409 00:46:51-!- Ravana__ [~Ravana@1-169-50-84.dyn.estpak.ee] has joined #wesnoth-dev
20160409 00:47:20-!- Ravana__ is now known as Ravana_
20160409 00:47:22-!- Ravana_ [~Ravana@1-169-50-84.dyn.estpak.ee] has quit [Changing host]
20160409 00:47:22-!- Ravana_ [~Ravana@unaffiliated/ravana/x-2327071] has joined #wesnoth-dev
20160409 00:58:35-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160409 01:03:49-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160409 01:07:34-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 250 seconds]
20160409 01:08:47 * celticminstrel pokes vultraz about changelog things.
20160409 01:09:48< celticminstrel> I see you added to lines to the changelog, but do you think maybe some of that should go in the players_changelog?
20160409 01:09:50< celticminstrel> ^two
20160409 01:11:55< celticminstrel> Hmm... I think we did put the new IPFs in the changelog already...
20160409 01:13:09< celticminstrel> Oh, the [message] thing should probably be in the changelog...
20160409 01:13:14< celticminstrel> https://github.com/wesnoth/wesnoth/commit/b512ad2143af67c02d1dc2951de68543848c1423
20160409 01:13:19< celticminstrel> Maybe the small_profile thing too?
20160409 01:13:33-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 240 seconds]
20160409 01:13:36< celticminstrel> The [message] thing is essentially a bugfix.
20160409 01:13:41< celticminstrel> Aww, he's gone?
20160409 01:15:05-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 268 seconds]
20160409 01:21:12-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160409 01:32:58-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20160409 01:39:58-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160409 01:41:28-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160409 01:43:06< vultraz_iOS> celticminstrel: my laptop is going crazy
20160409 01:43:12< vultraz_iOS> I cannot commit anything right now
20160409 01:43:17< vultraz_iOS> Use your discretion
20160409 01:43:34< celticminstrel> vultraz_iOS: What's your opinion on mentioning the GUI changes in players_changelog, though?
20160409 01:44:14< vultraz_iOS> Not really necessary since they're obvious but it might be worth it
20160409 02:01:26-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The computer falls into a deep sleep, to awake again some other day!]
20160409 02:06:56-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160409 02:17:04-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160409 02:21:57-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 260 seconds]
20160409 02:53:53-!- Waste [~Cracker@blk-138-68-122.eastlink.ca] has joined #wesnoth-dev
20160409 03:24:38-!- Waste [~Cracker@blk-138-68-122.eastlink.ca] has quit [Quit: Leaving]
20160409 04:01:03-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 240 seconds]
20160409 04:04:27-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20160409 04:06:25-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has joined #wesnoth-dev
20160409 04:49:13-!- vultraz [~chatzilla@124.109.10.167] has joined #wesnoth-dev
20160409 04:49:13-!- vultraz [~chatzilla@124.109.10.167] has quit [Changing host]
20160409 04:49:13-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160409 05:17:28-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160409 05:21:12-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Quit: ChatZilla 0.9.92-rdmsoft [XULRunner 35.0.1/20150122214805]]
20160409 05:22:42-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 276 seconds]
20160409 05:23:34-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160409 05:32:21-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160409 05:54:42-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has joined #wesnoth-dev
20160409 06:39:51-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20160409 06:40:07-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20160409 06:41:23-!- Kwandulin [~Miranda@p200300760F0D013CDC9155CD72A0B0A1.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160409 06:48:46-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160409 06:55:12< Aginor> thanks loonycyborg 
20160409 07:34:46-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160409 07:44:42-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160409 07:47:32-!- Kwandulin [~Miranda@p200300760F0D013CDC9155CD72A0B0A1.dip0.t-ipconnect.de] has quit [Ping timeout: 250 seconds]
20160409 07:48:48-!- Kwandulin [~Miranda@p200300760F0D013CDC9155CD72A0B0A1.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160409 07:49:45-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 268 seconds]
20160409 07:59:47-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160409 08:10:38< loonycyborg> vultraz: you were working on new lobby guy right?
20160409 08:12:45< loonycyborg> *new lobby gui
20160409 08:13:09-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20160409 08:49:14-!- Appleman1234 [~Appleman1@KD059138162076.au-net.ne.jp] has quit [Ping timeout: 244 seconds]
20160409 08:59:18-!- Kwandulin [~Miranda@p200300760F0D013CDC9155CD72A0B0A1.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160409 09:12:00-!- mjs-de [~mjs-de@x4db60fb0.dyn.telefonica.de] has joined #wesnoth-dev
20160409 09:34:39-!- gfgtdf [~chatzilla@x4e3698e2.dyn.telefonica.de] has joined #wesnoth-dev
20160409 09:35:03< gfgtdf> is wesnoth.org down ?
20160409 09:37:16< vultraz_iOS> loonycyborg: I was improving it, yes
20160409 09:37:17< vultraz_iOS> Why?
20160409 09:37:53< loonycyborg> vultraz_iOS: did you plan to use rooms too?
20160409 09:38:37< loonycyborg> gfgtdf: wesnoth.org seems up for me
20160409 09:39:28< gfgtdf> loonycyborg: hmm now it woroksfor me too
20160409 09:39:30< loonycyborg> vultraz_iOS: I mean those that are implemented with /room_join /room_part
20160409 09:39:58< vultraz_iOS> loonycyborg: I only know of rooms joined with /join room
20160409 09:40:07< gfgtdf> loonycyborg: it didnt work  aminute ago (all other pages worked)
20160409 09:40:44< loonycyborg> it is a feature specific to new lobby that we didn't end up fully adopting
20160409 09:41:01< vultraz_iOS> Elaborate
20160409 09:41:03-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 240 seconds]
20160409 09:42:27-!- gfgtdf [~chatzilla@x4e3698e2.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 45.0.1/20160315153207]]
20160409 09:43:00< loonycyborg> vultraz_iOS: lobby can be subdivided into rooms that contain a subset of players
20160409 09:43:10< vultraz_iOS> Right
20160409 09:43:27< loonycyborg> but this functionality isn't invoked by client gui currently
20160409 09:43:42< loonycyborg> only available via /room_join and /room_part commands
20160409 09:43:53< vultraz_iOS> So is this something different from /join room
20160409 09:44:16< vultraz_iOS> I've never heard of /room_join
20160409 09:44:40-!- Appleman1234 [~Appleman1@KD111237163141.au-net.ne.jp] has joined #wesnoth-dev
20160409 09:45:34< loonycyborg> so if you're not emitting <room_join> and <room_part> wml tags you're not using it
20160409 09:46:03< vultraz_iOS> I've only done interface-level work on the new lobby
20160409 09:46:17< vultraz_iOS> So anything that wasn't there before internally likely isn't there now
20160409 09:47:11-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160409 09:47:14< loonycyborg> Just to be sure: it's based on current lobby gui or mordante's new lobby that didn't get adopted?
20160409 09:47:30< vultraz_iOS> What?
20160409 09:47:51< vultraz_iOS> I've been working on the gui2 "experimental" lobby
20160409 09:48:09< vultraz_iOS> Which hopefully will become default this dev cycle
20160409 09:49:54< loonycyborg> mordante tried to implement own gui2 new lobby too long time ago
20160409 09:50:03< loonycyborg> but it ended up being too slow
20160409 09:50:17< vultraz_iOS> Is that the experimental lobby we have now?
20160409 09:50:37-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 268 seconds]
20160409 09:50:37-!- wedge010 is now known as wedge009
20160409 09:50:47< loonycyborg> I'm not sure, maybe it's still here, maybe removed
20160409 09:51:13< loonycyborg> But it seems only that project of his used <room_join>
20160409 09:52:01< vultraz_iOS> It's still there
20160409 09:56:59< loonycyborg> So did you use that project of his or made your own?
20160409 10:11:49-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160409 10:11:50-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Read error: Connection reset by peer]
20160409 10:12:18-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160409 10:16:32-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160409 10:28:07-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160409 10:29:04-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Client Quit]
20160409 10:29:41-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160409 10:42:30-!- Kwandulin [~Miranda@p200300760F0D013C801FCF00591C1B4C.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160409 10:47:29< vultraz_iOS> loonycyborg: I was working on the experimental lobby code that's already in the repo
20160409 10:47:46< vultraz_iOS> Gui/dialogs/lobby/
20160409 11:03:15-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160409 11:33:27-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160409 11:55:00-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20160409 12:27:07< loonycyborg> vultraz_iOS: seems that gui/dialogs/lobby/lobby.cpp is using those <room*> tags
20160409 12:29:03< loonycyborg> also be aware that mordante's last effort to revamp lobby was rejected by community, people even asked me to make special old lobby installers until new lobby got disabled.
20160409 12:29:19< loonycyborg> so I hope you'll fix whatever issues those people were having
20160409 12:36:30-!- Kwandulin_2 [~Miranda@p200300760F0D01519C009E9C23CD66BD.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160409 12:38:14-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160409 12:38:21-!- Kwandulin [~Miranda@p200300760F0D013C801FCF00591C1B4C.dip0.t-ipconnect.de] has quit [Ping timeout: 268 seconds]
20160409 12:44:25-!- Kwandulin_2 [~Miranda@p200300760F0D01519C009E9C23CD66BD.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160409 12:55:40-!- AI0867 [~ai@wesnoth/developer/ai0867] has quit [Ping timeout: 244 seconds]
20160409 12:55:47-!- AI0867 [~ai@wesnoth/developer/ai0867] has joined #wesnoth-dev
20160409 12:57:14-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has quit [Quit: Connection closed for inactivity]
20160409 12:59:00-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160409 13:09:30-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 250 seconds]
20160409 13:09:34-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160409 13:19:09-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 260 seconds]
20160409 13:23:50-!- Kwandulin [~Miranda@p200300760F0D015135969ED0D2C9FEC4.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160409 13:58:04-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160409 14:11:22-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20160409 14:11:29-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20160409 14:51:50-!- gfgtdf [~chatzilla@x4e3698e2.dyn.telefonica.de] has joined #wesnoth-dev
20160409 15:00:23-!- Dugi [b28f8f9e@gateway/web/freenode/ip.178.143.143.158] has joined #wesnoth-dev
20160409 15:01:50< Dugi> Hello.
20160409 15:04:44< Dugi> Looks like this isn't a time of activity.
20160409 15:17:36< pydsigner> It's still morning on a weekend
20160409 15:18:16-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20160409 15:19:27< pydsigner> Although there was a lot more activity 5 hours ago so maybe it's something else
20160409 15:19:39< pydsigner> Like people just coding on stuff
20160409 15:19:54< pydsigner> Instead of active-lurking on IRC
20160409 15:24:26-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has joined #wesnoth-dev
20160409 15:33:49< Dugi> Aha, thanks.
20160409 15:34:56< Dugi> I just had made corrections to my pull request according to comments, I was wondering when will someone look at it if there's something else to correct or it's okay and can be accepted.
20160409 15:35:33< celticminstrel> I didn't see your PR in my notifications, so I didn't look at it just now. But I don't think it notifies when new commits are pushed...
20160409 15:36:02< celticminstrel> Build passed, yay.
20160409 15:36:31< celticminstrel> The only new commit I see is fixed indentation...
20160409 15:36:53< celticminstrel> ...oh.
20160409 15:37:04< celticminstrel> You mixed the fixed indentation in with the new stuff. That's not good.
20160409 15:38:15< celticminstrel> I think you misspelled "vowel" in the human name generator.
20160409 15:38:47< Dugi> Hm, I thought it's spelled wowel, dammit.
20160409 15:38:53< celticminstrel> Also, generated_prefix is defined twice for female, is that intentional? (Given how CFG is supposed to work I suppose that could be the same as using |)
20160409 15:39:26< celticminstrel> So this parser is completely whitespace-sensitive after the =?
20160409 15:40:27< celticminstrel> Okay, I guess mixing the indentation fix with the new stuff didn't really hurt in this case, but generally, it's better to make a separate commit for anything like indentation fixes.
20160409 15:40:29< Dugi> The generated_prefix was a copying error. I was testing it, then copied the corrected part back... and did it badly.
20160409 15:40:43< Dugi> Okay, I will split it next time.
20160409 15:41:22< Dugi> It is space sensitive, because you spaces are inseparable parts of names.
20160409 15:41:44< celticminstrel> I wonder whether dwarven names are the same for male and female.
20160409 15:41:54< celticminstrel> Ditto for saurians...
20160409 15:42:04< Dugi> They were all male only.
20160409 15:42:05< celticminstrel> Khalifate though probably wouldn't be...
20160409 15:42:11< celticminstrel> Yeah, I know, that's why I'm wondering.
20160409 15:42:29< Dugi> There are neither female dwarves nor female saurians nor female khalifates.
20160409 15:42:42< celticminstrel> That's obviously false, lore-wise. :P
20160409 15:43:04< celticminstrel> Do female drakes actually occur? The unit type definitions don't reference gender at all.
20160409 15:43:15< celticminstrel> Ogres too are missing female names...
20160409 15:43:17< Dugi> Gliders' line can be female.
20160409 15:43:19< celticminstrel> And orcs.
20160409 15:43:28< celticminstrel> Really? 
20160409 15:43:43-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has quit [Quit: End Transmission.]
20160409 15:44:05-!- mjs-de [~mjs-de@x4db60fb0.dyn.telefonica.de] has quit [Remote host closed the connection]
20160409 15:44:07< celticminstrel> (My complaining right now is obviously nothing to do with your PR.)
20160409 15:44:08< Dugi> Yep, the names were never defined for female units of races that are always male.
20160409 15:44:20-!- fendrin [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20160409 15:44:25< celticminstrel> I want to add at least female Khalifate names.
20160409 15:44:51< celticminstrel> For dwarves, orcs, saurians, and ogres, I could believe that females don't have different naming conventions.
20160409 15:45:25< celticminstrel> Hmm, I don't see any reference to gender in the sky drake...
20160409 15:46:28< Dugi> Well, in real life, there are nations where male names don't differ from female ones. Like Israel. You can find there women named Adrien or men named Ariel. They see no reason not to name a boy after an important woman or vice versa.
20160409 15:47:11< celticminstrel> I don't believe this is the case in Arab nations, though, and since Khalifate names look Arabic...
20160409 15:47:31< Dugi> I have no idea what are name conventions for Arab women. Any memorable Arab is a man.
20160409 15:47:41-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Ping timeout: 248 seconds]
20160409 15:47:56< celticminstrel> Really now.
20160409 15:48:52< celticminstrel> So, regarding the drakes, one of two things is true. a) The default gender value is "male or female", and thus all the drake units can be either gender. b)  The default gender is "male", and thus all the drake units are male.
20160409 15:49:43< Dugi> I am sure I have seen female drakes somewhere. Maybe they were removed when their spritework was renamed around 1.9.
20160409 15:49:49< celticminstrel> BTW, does the output of your generator produce all the names that are input for the Markov chains?
20160409 15:50:38< celticminstrel> Ah, you did add a generator for the village names. Okay then. And I guess that actually works, since the map generator uses the race class as a generator.
20160409 15:51:08< Dugi> No, the grammar is inspired by the old names so the output is similar, but the set of names it can produce is different. And smaller except for gryphon females.
20160409 15:51:42< celticminstrel> I think you misunderstood a bit. I mean, is every name in the input list for the Markov chains a possible output from the new generator?
20160409 15:52:09< celticminstrel> I feel like villages should maybe have more suffixes like "shire" or "combe" or "ton" or whatever... but that's unrelated to what you've done.
20160409 15:52:11< Dugi> However, the probabilistic distribution of the old names wasn't very good, I have generated 100 000 names for each race and found that despite the number of possible names, you need only around 50 units to find a namesake.
20160409 15:53:01< Dugi> The village names were short because it adds some suffix. Try a generated map and look what it does with these names.
20160409 15:53:10< celticminstrel> Ah. Okay then.
20160409 15:53:26< celticminstrel> Maybe that suffix could be rolled into the generator later.
20160409 15:53:42< celticminstrel> Although... I guess the suffix might depend on external parameters...
20160409 15:53:54< celticminstrel> So maybe there's a good reason to leave it how it is.
20160409 15:53:54< Dugi> Not every name is a possible output, but most of them are. I may have forgotten to include some and some did not fit.
20160409 15:54:25< Dugi> I don't know if it depends, but yeah, it could be added into it later.
20160409 15:54:38< celticminstrel> I'm curious about some that did not fit.
20160409 15:55:29< celticminstrel> Oh huh, bumbadadabum is a wose.
20160409 15:55:38< celticminstrel> I assumed he just made up that name.
20160409 15:57:01< celticminstrel> Looking at woses, it seems "Carnimirië" is one that can't be an output of the new grammar.
20160409 15:57:03< Dugi> Sometimes, there were some letter groups that would sound really bad with some other letters around and making that combination possible would require changing the grammar rules a lot. Also, some names were too short, so they were made possible only as a part of a longer name.
20160409 15:57:31< celticminstrel> Though similar names are possible.
20160409 15:57:33-!- fendrin [~quassel@wesnoth/developer/fendrin] has quit [Ping timeout: 240 seconds]
20160409 15:57:44< celticminstrel> Like "Carnimidrum".
20160409 15:57:48< Dugi> Well, Carnimirië does not sound very wosish.
20160409 15:57:52< zookeeper> it doesn't really matter whether the new generators can crank out the same names, as long as the general aesthetic of most names remains more or less similar.
20160409 15:57:55< celticminstrel> Maybe.
20160409 15:57:59-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20160409 15:58:03< celticminstrel> zookeeper: Yeah, I was just curious.
20160409 15:58:21< celticminstrel> The maybe was to not sounding very wosish, not to zookeeper 
20160409 15:59:10< celticminstrel> BTW Dugi, what happens if you do define a nonterminal twice?
20160409 15:59:38< celticminstrel> Or indeed, main twice.
20160409 15:59:58< Dugi> Also, I have no idea who wrote these lists, but if it was a human, he had to be really mischievous, I have found names like 'Analus' there.
20160409 16:00:22< celticminstrel> That honestly doesn't sound super-terrible to me, though I can see why you'd think it does.
20160409 16:00:32< Dugi> If a nonterminal is defined twice, either the latter one works or they are merged, I am not sure, lemme check the code.
20160409 16:01:01< celticminstrel> I think merging with | makes the most sense.
20160409 16:01:27< pydsigner> Sky Drakes are described as male
20160409 16:01:33< pydsigner> In the unit desc
20160409 16:01:54< Dugi> Double definition of a nonterminal leads to appending the new list of possibilities to the old one.
20160409 16:02:03< celticminstrel> Excellent.
20160409 16:02:40< celticminstrel> I think it could be merged, unless you want to fix the spelling of vowel in various places, or do something about that double generated_prefix that I mentioned. zookeeper, do you agree?
20160409 16:03:10< Dugi> I will fix these mistakes.
20160409 16:03:15< zookeeper> i didn't yet look at the name output examples that were posted
20160409 16:03:27< celticminstrel> So you want a chance to look at that first?
20160409 16:04:53< zookeeper> yeah
20160409 16:05:17< celticminstrel> 'kay
20160409 16:05:30< Dugi> I have committed these corrections.
20160409 16:06:17< celticminstrel> I suppose it's bad for a nonterminal to reference itself...
20160409 16:06:30< Dugi> Where does it happen?
20160409 16:07:04< celticminstrel> The one you just fixed in female human names, before the fix.
20160409 16:08:06< celticminstrel> So, nowhere anymore.
20160409 16:08:24< celticminstrel> Would that lead to infinite loops or just potentially very long names?
20160409 16:08:35< Dugi> It was a copying error. It can be used to generate texts of random lengths, with exponentially decreasing probability with length.
20160409 16:08:45< celticminstrel> So it's not bad?
20160409 16:08:51< Dugi> But if you write it badly, it can keep expanding.
20160409 16:08:56< celticminstrel> 'kay
20160409 16:09:06< Dugi> Depends how it's used. But I didn't want to use it.
20160409 16:09:18< celticminstrel> Yeah, better to avoid it, certainly.
20160409 16:11:50< Dugi> gfgtdf replied to the last commit that I should merge the last two, it's done with git rebase something, right? And when doing it, I could just merge all three commits, no?
20160409 16:12:44< celticminstrel> You could, yes.
20160409 16:12:55-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20160409 16:12:58< celticminstrel> It's git rebase -i 
20160409 16:13:03< celticminstrel> Or --interactive
20160409 16:13:12< celticminstrel> And I think you need to specify a base too.
20160409 16:13:23< celticminstrel> In this case something like HEAD~3 should work.
20160409 16:13:42< celticminstrel> Lately I've been using main/master (which for you would be origin/master if I recall correctly).
20160409 16:13:54< celticminstrel> Or just master, if I'm on a branch.
20160409 16:14:04< gfgtdf> i actually meant the first 2 commit, (the previous = the one before the other one = the first one) but you coudl also merge all three commits
20160409 16:14:32< Dugi> Is this what I want? http://pastebin.com/7YydDvws
20160409 16:14:45< celticminstrel> You could also split out the indentation portion of the second commit and merge that with the previous commit, while keeping the other portion of it separate (and change the commit message).
20160409 16:15:15< gfgtdf> Dugi: yes, but it also possible tuse fixup instead of squash here
20160409 16:15:19< celticminstrel> That'll merge the three commits and let you make a new message.
20160409 16:16:21< Dugi> That's what I want. Make one commit of the three, because the latter two are just corrections of the first one, there's no need for them to stand in the history.
20160409 16:17:01< celticminstrel> You can use fixup instead of squash if you want the first commit message to stand; if you use squash you get to edit the commit messages and combine them however you want.
20160409 16:17:01< Dugi> Maybe fixup instead of squash could be better.
20160409 16:17:20< Dugi> Okay, so fixup it is.
20160409 16:19:11< Dugi> Okay, looks correct now.
20160409 16:23:36< Dugi> Anything else? I will be going to have dinner soon.
20160409 16:25:58-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160409 16:32:16< celticminstrel> Dugi: What about using the custom random generator instead of rand()?
20160409 16:32:41< celticminstrel> Same as in race.cpp
20160409 16:33:34-!- Dugi_ [b28f8f9e@gateway/web/freenode/ip.178.143.143.158] has joined #wesnoth-dev
20160409 16:34:03< celticminstrel> Though apparently that also causes problems... hmm...
20160409 16:34:25< Dugi_> I have crashed. This unstable browser is... irritating.
20160409 16:34:34< Dugi_> I have read the log of what I missed.
20160409 16:35:22< Dugi_> The custom rand function was removed at some time from the name generation before.
20160409 16:35:37< Dugi_> I am off to dinner.
20160409 16:35:50 * celticminstrel is looking near the top of markov_generate_name where it generates a large numner of random numbers and stores them in a lookup table.
20160409 16:36:05< celticminstrel> ^number
20160409 16:36:52-!- Dugi [b28f8f9e@gateway/web/freenode/ip.178.143.143.158] has quit [Ping timeout: 250 seconds]
20160409 16:46:48< celticminstrel> I wonder why the Markov generator is using UCS-4 strings...
20160409 16:47:06< celticminstrel> I guess maybe it simplifies it.
20160409 16:49:52-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has joined #wesnoth-dev
20160409 16:54:49-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160409 17:02:10-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160409 17:08:31-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160409 17:08:51-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160409 17:13:16-!- gfgtdf [~chatzilla@x4e3698e2.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 45.0.1/20160315153207]]
20160409 17:13:31-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has joined #wesnoth-dev
20160409 17:23:22-!- ancestral [~ancestral@75-168-27-21.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160409 17:26:07-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Remote host closed the connection]
20160409 17:28:20< Dugi_> Imagine you have them stored in UTF-8 and you stumble upon a letter made of two bits, like š or ž. That would mess it up if it used UTF-8. I am not back yet.
20160409 17:41:55-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20160409 17:42:01-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160409 17:42:05-!- stikonas_ is now known as stikonas
20160409 17:50:56< Dugi_> That random_new is some sort of singleton or what?
20160409 17:51:07< celticminstrel> Think so...
20160409 17:53:29< Dugi_> So I can call it just like I would call rand(). Since unit naming is only a part of unit creation that is synced, it won't cause OoS, right? I am still wondering why is it needed when unit names are not synced.
20160409 17:54:14< celticminstrel> It sounds like the problem wasn't in synching of the names but rather in synching the state of the random generator itself.
20160409 17:57:11< Dugi_> Well, but someone has a different language, the algorithm uses a different number of random calls and it goes OoS.
20160409 17:57:14-!- lipkab [~the_new_l@host-91-147-210-58.biatv.hu] has joined #wesnoth-dev
20160409 17:57:26< celticminstrel> Something like that.
20160409 17:58:42< Dugi_> That's why the markov generator prepares a fixed number of them to handle this. But I think it would be safer just to avoid using that thing completely in things that need not to be synched.
20160409 17:59:11< Dugi_> On the other hand, players might chat about a unit with different names on each client.
20160409 18:03:30< Dugi_> Though the names will be different if they have a different language set, I have no idea if unit names are mentioned in players' chat because of this.
20160409 18:04:00< celticminstrel> I think it's good for unit names to be synched. The actual name might depend on each user's locale, but all players should see the same names.
20160409 18:05:54< Dugi_> I think it would be far safer to create local copies of the rng class for operations like this. It starts synched, if it goes OoS, it will cause the non-synched thing to evolve differently, but keep the game intact.
20160409 18:06:56< celticminstrel> This sounds like a possible thing... I think it's also possible to save its state and restore it later...
20160409 18:09:31< Dugi_> I haven't found such a method.
20160409 18:14:48< Dugi_> Well, I can change my code to generate 20 numbers using that random generator and circle through them instead of rand() calls if you want. My complains are mostly aimed at the OoS vulnerability of the rng.
20160409 18:15:16< Dugi_> Or I can generate one number and use srand on it.
20160409 18:16:09< Dugi_> Syncing rand() before any action that need not to be synced could be an easy way to avoid such a vulnerability.
20160409 18:17:16-!- mjs-de [~mjs-de@x4db60fb0.dyn.telefonica.de] has joined #wesnoth-dev
20160409 18:17:31-!- gfgtdf [~chatzilla@x4e3698e2.dyn.telefonica.de] has joined #wesnoth-dev
20160409 18:18:30< celticminstrel> I think rand() is potentially a worse random generator than the random_new thing (which I think is a mersenne twister), but I suppose that's not super-important since this isn't a cryptographical application. It bugs me a little to be seeding one random generator from another, but pre-generating a bunch of random numbers bugs me more.
20160409 18:21:04< Dugi_> So I will go for the srand() possibility?
20160409 18:21:43< gfgtdf> Dugi_: what do you ewant to do with srand ?
20160409 18:24:53< Dugi_> gfgtdf: If you use the default rng for generating the names, different language can cause a different number of calls and the game will go OoS. rand() alone would generate names without OoS, but the names would be different even if the language is the same. There are two solutions to it, either generating a group of random numbers to use before generating the name or using srand() to sync the calls to rand().
20160409 18:25:30< gfgtdf> Dugi_: so form where do you want to take te value to pass to srand ?
20160409 18:26:20< Dugi_> From a single call to the rng at the start. So that there would be one call per generation, regardless of the grammar used.
20160409 18:27:37< gfgtdf> Dugi_: hmm what i don't liek abut this is that the rand() rng might also be usee by other codes
20160409 18:27:37-!- prkc [~prkc@46.166.190.185] has quit [Remote host closed the connection]
20160409 18:28:25< celticminstrel> You could check if that's the case.
20160409 18:28:40< celticminstrel> Unless you mean things less in Wesnoth's control.
20160409 18:30:10< Dugi_> That shouldn't break anything. They are not executed concurrently.
20160409 18:32:09< Dugi_> rand() is used in the code a couple of times. srand() too. I don't think that adding additional calls to that would change anything.
20160409 18:34:52< Dugi_> However, I am not sure if the implementation of rand() does not depend on the compiler. That would cause the unit names to be different anyway.
20160409 18:35:51< celticminstrel> The implementation of rand() is unspecified, I think.
20160409 18:37:04< Dugi_> So maybe generating a fixed quantity of names is unavoidable?
20160409 18:38:17< celticminstrel> ^numbers
20160409 18:38:52< Dugi_> Yeah, numbers.
20160409 18:39:17< celticminstrel> I've forgotten what problem we're trying to solve now...
20160409 18:40:24< Dugi_> How to generate the same names if possible (if the grammar is similar enough) without causing the rng to go OoS if the grammars are different and lead to a different number of calls.
20160409 18:41:47< Dugi_> Generating a fixed quantity of numbers (high enough) and circling between them instead of rand() calls seems to be the only way at the moment.
20160409 18:49:58< celticminstrel> Ehh...
20160409 18:51:31< Dugi_> Okay, I will do it. Let's not let this end in a stalemate.
20160409 19:15:52-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160409 19:17:27< Dugi_> I have edited the last commit with this.
20160409 19:21:35< celticminstrel> Waiting on zookeeper then, I think.
20160409 19:29:39< Dugi_> Btw, what's the difference between 'waiting for xx' and 'waiting on xx'? I have discussed this with an Ukrainian girl and we never came to a satisfying explanation.
20160409 19:30:22< vultraz> Grammar-wise?
20160409 19:31:01< celticminstrel> I don't think there's any difference...
20160409 19:31:16< celticminstrel> At least in this context. "waiting on" has another meaning.
20160409 19:31:17< Dugi_> vultraz: Yeah, grammar-wise.
20160409 19:32:34< vultraz> Not really any difference.
20160409 19:33:03< vultraz> "waiting on" can have another meaning, though.
20160409 19:33:13< Dugi_> And that is what?
20160409 19:34:53< vultraz> Kinda like "to serve" someone.
20160409 19:35:37< celticminstrel> Like a waiter waiting on a table in a restaurant.
20160409 19:35:41< vultraz> Ever heard of the expression "wait on them hand and foot"?
20160409 19:35:55< Dugi_> Never heard that expression.
20160409 19:36:33< vultraz> https://english.stackexchange.com/questions/36384/origin-of-waited-on-hand-and-foot
20160409 19:36:43< Dugi_> So 'wait on xx' can mean 'serve xx' without having to mean that the person is actually waiting for something?
20160409 19:37:02< vultraz> Yes.
20160409 19:37:06< celticminstrel> Yeah.
20160409 19:37:55< Dugi_> Thanks. That link explained it quite well.
20160409 19:42:38< Dugi_> That clang check in travis seems to succeed and fail quite randomly. Any idea what can be happening:
20160409 19:43:49< vultraz> Nope
20160409 19:43:54< vultraz> Don't worry about it
20160409 19:45:27-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160409 19:45:40-!- boucman [~rosen@wesnoth/developer/boucman] has left #wesnoth-dev []
20160409 19:49:28-!- irker026 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160409 19:49:28< irker026> wesnoth: Charles Dang wesnoth:master 1702f10360d2 / data/gui/widget/button_default.cfg: tbutton: fix borders of disabled state buttons not being gray https://github.com/wesnoth/wesnoth/commit/1702f10360d245d21443ab02c6c3a1efab9235b5
20160409 19:49:29< irker026> wesnoth: Charles Dang wesnoth:master 47bb358003ef / data/gui/macros/_initial.cfg: GUI2: removed extra + 2 offset from GUI__TEXT_VERTICALLY_CENTRED https://github.com/wesnoth/wesnoth/commit/47bb358003efa3d19571ecd0970c1adfa3866d95
20160409 19:51:16< irker026> wesnoth: Charles Dang wesnoth:master 25e8d4f07c0c / data/gui/window/lobby_main.cfg: tlobby_main: use large buttons for main controls https://github.com/wesnoth/wesnoth/commit/25e8d4f07c0cb3e15e60d5570f01e6eddc62cd25
20160409 19:57:39< irker026> wesnoth: Charles Dang wesnoth:master 344955e4c11a / data/gui/ (macros/_initial.cfg window/unit_attack.cfg): GUI2: added macro for vertical spacer line https://github.com/wesnoth/wesnoth/commit/344955e4c11a17dcb7237b0ecc4490cf97c2c3e4
20160409 20:07:07< irker026> wesnoth: Charles Dang wesnoth:master 3bfc17a9e9bd / data/gui/window/lobby_main.cfg: tlobby_main: fix misalignment of game listbox borders https://github.com/wesnoth/wesnoth/commit/3bfc17a9e9bdb8f42f4b96ac948e191bf2589afc
20160409 20:07:28 * vultraz checks a few things of TODO list
20160409 20:07:33< vultraz> off*
20160409 20:10:02< vultraz> still don't know why the chatbox insists on growing now
20160409 20:13:16< vultraz> still getting occasional random crashes on the loadscreen
20160409 20:15:02< irker026> wesnoth: Charles Dang wesnoth:master f77b3d7a6fde / data/gui/window/lobby_main.cfg: tlobby_main: improved formula for game listbox min height https://github.com/wesnoth/wesnoth/commit/f77b3d7a6fde05cceb3ac123d2eef38edb364b05
20160409 20:15:19< vultraz> we really need a 'force widget to always be this size' setting
20160409 20:18:28-!- horrowind [~Icedove@2a02:810a:83c0:1c18:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20160409 20:20:26< Dugi_> It makes me wonder why CEGUI or some other common GUI library is not used. The current GUI implementation behaves unpredictably and getting some things to work as indented includes tons of voodoo.
20160409 20:20:34-!- fendrin [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20160409 20:22:54< celticminstrel> If someone can write a wrapper to CEGUI that reads WML and has the same basic functionality as GUI2, I would not be opposed to accepting it.
20160409 20:23:35< celticminstrel> The key point about reading WML would be that a valid GUI2 WML should still be valid, even if it doesn't quite produce the same result.
20160409 20:24:06< celticminstrel> That doesn't mean there couldn't also be simpler ways of doing things though.
20160409 20:24:15-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Ping timeout: 276 seconds]
20160409 20:24:33< celticminstrel> vultraz: So what about changelogs?
20160409 20:25:07< celticminstrel> Also, that +2 was there for a reason, so I might revert that one.
20160409 20:26:30< Dugi_> The only GUI system using the weird logic of the WML used to bind GUI2 I have ever seen is Wt (and that one would not be very applicable).
20160409 20:27:32< celticminstrel> Not sure what you mean by "weird logic".
20160409 20:28:59< Dugi_> celticminstrel: Declaring the grids and growths instead of coordinates and their scaling. I played around CEGUI a bit and it seems much better to me.
20160409 20:29:18< celticminstrel> Declaring coordinats is far less flexible.
20160409 20:29:25< celticminstrel> ^coordinates
20160409 20:29:31< celticminstrel> Have you ever worked with Swing?
20160409 20:29:39< Dugi_> Nope.
20160409 20:29:41< celticminstrel> It does things a little similarly, I think.
20160409 20:29:53< celticminstrel> I've worked with it a little.
20160409 20:30:36< Dugi_> Well, Wt does it too, I have created the website I use besides other things to show the name generator using that one.
20160409 20:30:37< celticminstrel> Mind you, I never used any kind of declarative language (XML, whatever) to build an interface.
20160409 20:30:51< celticminstrel> I just constructed the whole thing in the Java code.
20160409 20:30:59< celticminstrel> It was a relatively simple interface.
20160409 20:34:54< Dugi_> I am not using any XML, in both cases I construct it right from the C++ code that uses it, I find it more flexible, I can use lambdas and macros to make later changes easier.
20160409 20:35:56-!- mjs-de [~mjs-de@x4db60fb0.dyn.telefonica.de] has quit [Remote host closed the connection]
20160409 20:44:36< celticminstrel> Well, the key point about not changing the data format is that addons can also use GUI2, and we'd like to not break them. The other reason for having the dialog structure stored as WML is to make it themable.
20160409 20:45:04-!- travis-ci [~travis-ci@ec2-107-22-52-114.compute-1.amazonaws.com] has joined #wesnoth-dev
20160409 20:45:05< travis-ci> wesnoth/wesnoth#9244 (master - 344955e : Charles Dang): The build was broken.
20160409 20:45:05< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/121955298
20160409 20:45:05-!- travis-ci [~travis-ci@ec2-107-22-52-114.compute-1.amazonaws.com] has left #wesnoth-dev []
20160409 20:45:15< celticminstrel> I think it would actually be fairly simple to add the necessary things to make GUI2 themable, though I haven't investigated in detail.
20160409 20:45:30< celticminstrel> I suspect most of the framework for it is already in place, though.
20160409 20:46:11< celticminstrel> Mind you, with how it currently is, defining a new theme would mean completely redefining every dialog, which I guess is less than ideal...
20160409 20:47:20< Dugi_> Backwards compatibility is a huge problem, right. I guess there weren't many libraries to choose from back in the day when GUI2 was introduced.
20160409 20:47:33< celticminstrel> I dunno.
20160409 20:47:40< celticminstrel> Maybe.
20160409 20:53:21< fendrin> Dugi_: There have been enough. GUI2 was introduced in 2008. You should consider GUI2 to be the experiment of a developer. He abused Wesnoth for his dream of writting an own GUI system. It is the shame of the development team that this was not stopped until the current day.
20160409 20:53:40-!- horrowind [~Icedove@2a02:810a:83c0:1c18:21b:fcff:fee3:c3ff] has quit [Quit: horrowind]
20160409 20:54:49< vultraz> celticminstrel: why was the +2 there?
20160409 20:54:56< vultraz> celticminstrel: it makes the button text off-center
20160409 20:55:29< Dugi_> fendrin: That explains a lot. Thanks.
20160409 20:55:39< celticminstrel> I think originally it made it on-centre.
20160409 20:55:50< celticminstrel> I'll check again, maybe you're right.
20160409 20:55:52< vultraz> Probably with DVS
20160409 20:55:58< vultraz> Not Lato
20160409 20:57:16< vultraz> As for GUI compatibility, it depends on what we do for Wesnoth 2
20160409 20:57:33< celticminstrel> Let's not talk about an imaginary Wesnoth 2.
20160409 20:57:53< vultraz> Might be a bit less imaginary than you think.
20160409 20:59:07< fendrin> vultraz: Do you know something we don't?
20160409 21:00:17< vultraz> Once we release 1.14, we will need to decide to either proceed on our current path and start on 1.15, or adopt this prototype and develop it further: https://github.com/anura-engine/wesnoth2 
20160409 21:00:50< celticminstrel> I don't see any reason for that to be a good move.
20160409 21:00:56< fendrin> Neither me.
20160409 21:01:00< vultraz> And that's not something we can just declare outright 
20160409 21:01:02< celticminstrel> That breaks all backwards compatibility, after all.
20160409 21:01:09< vultraz> That's kinda the point.
20160409 21:01:22< vultraz> But again, I don't have time to deal with this right now.
20160409 21:01:24< celticminstrel> Yeah, breaking all backwards compatibility is bad.
20160409 21:01:33< vultraz> I'll need to come up with a list of pros and cons in a few months
20160409 21:02:07< vultraz> I've spoken to Aginor about it, and he brought up some good points against. I've also heard some good points in favor.
20160409 21:02:32< celticminstrel> I think the lack of backwards compatibility is a major point against it.
20160409 21:03:16< vultraz> Right now, the focus should be on completing 1.14 and getting it on Steam.
20160409 21:03:53< fendrin> I think you should backport what is worth and release more fixes for the 1.12 version.
20160409 21:04:04< vultraz> Too difficult
20160409 21:04:06< fendrin> And never release 1.14 based on the old engine.
20160409 21:04:25< fendrin> Every work on the old codebase is wasted time.
20160409 21:04:29< celticminstrel> I think it could very well be worthwhile to release a 1.12.6, but I think fendrin also just said something totally crazy.
20160409 21:04:44< vultraz> ^
20160409 21:05:01< celticminstrel> So I'll just pretend he didn't suggest dumping everything and starting over.
20160409 21:05:40< vultraz> fendrin: we all want wesnoth 2 in some form or another, but what you suggested is more certainly not the way to get there.
20160409 21:06:16< vultraz> most*
20160409 21:07:14-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has quit [Quit: Connection closed for inactivity]
20160409 21:07:31< fendrin> vultraz: Do you have arguments for that claim?
20160409 21:07:39< vultraz> ...
20160409 21:07:48< celticminstrel> It's pretty self-evident that it's a terrible idea.
20160409 21:07:56< vultraz> ^
20160409 21:08:00< celticminstrel> Unless you intend to destroy Wesnoth altogether.
20160409 21:08:16< fendrin> I intend to.
20160409 21:08:35< celticminstrel> Dropping the entire codebase and starting over would alienate all the devs, all the players, everyone.
20160409 21:08:46< celticminstrel> So it's not happening.
20160409 21:08:49< fendrin> I guess the players don't care.
20160409 21:08:54< vultraz> Perhaps we will have to do that eventually.
20160409 21:09:02< vultraz> But certainly not without careful thought
20160409 21:09:05< vultraz> And planning
20160409 21:09:18< vultraz> And most certainly not right now
20160409 21:09:25< fendrin> The earlier the better.
20160409 21:09:56< vultraz> I said
20160409 21:10:05< vultraz> 1.14, then we'll see
20160409 21:10:09< Dugi_> I would like lua replaced with Haskell.
20160409 21:10:23< vultraz> And I would like a Small Loan Of A Million Dollars
20160409 21:10:23< fendrin> Dugi_: Yeah, haskell is nice, i love it.
20160409 21:10:39< celticminstrel> Meh.
20160409 21:11:20< fendrin> Wesnoth is a bunch of GUI systems. Drop them all and take a library that is ready to use.
20160409 21:11:39< fendrin> Then there is support for lua, wml, formula ai, and what not
20160409 21:11:50< fendrin> drop it all together with the prprocessor crap
20160409 21:11:56< fendrin> and replace it with lua only
20160409 21:12:23< fendrin> That brings down the codebase to ~80000 lines.
20160409 21:12:25< fendrin> maybe less
20160409 21:12:39< fendrin> A medium sized project.
20160409 21:12:47< fendrin> Can be done by 5 persons in a year.
20160409 21:12:54< vultraz> THAT MAY BE
20160409 21:13:04< vultraz> And we may drop everything for wesnoth 2
20160409 21:13:13< celticminstrel> Tearing out all the GUI code in favour of something new is a far cry from dropping the entire codebase and starting over.
20160409 21:13:29< vultraz> Or parts of it
20160409 21:13:31< celticminstrel> But replacing the preprocessor and WML with Lua is not a good idea. That breaks all backwards compatibility.
20160409 21:13:31< vultraz> Or none of it
20160409 21:13:56< vultraz> How many times do I have to reiterate that any such decisions will require careful consideration
20160409 21:13:58< celticminstrel> Slowly phasing out the preprocessor is not a terrible idea, though I'm not sure it's worthwhile.
20160409 21:14:22< fendrin> the preprocessor is the reason that there is not real linter 
20160409 21:14:24< celticminstrel> There are a few things that I wish were done differently so as to avoid macros, such as unit abilities and weapon specials.
20160409 21:14:32< vultraz> Say we drop everything and switch to Anura's prototype
20160409 21:14:34< celticminstrel> A real what now?
20160409 21:14:37< vultraz> We'd lose compatibility 
20160409 21:14:40< vultraz> What would we gain?
20160409 21:14:44< fendrin> nothing
20160409 21:14:49< vultraz> A lot, actually
20160409 21:14:51< fendrin> anura is just not suited for wesnoth
20160409 21:15:00< vultraz> What would we gain by using Haskell in place of Lua?
20160409 21:15:01< fendrin> and a huge and clumsy engine itself
20160409 21:15:09< fendrin> I do not want that
20160409 21:15:16< fendrin> It is Dugi_'s idea.
20160409 21:15:39< fendrin> I just want to express love towards Heskell, I don't think it is suited for Wesnoth.
20160409 21:15:47< vultraz> Doesn't matter whose idea
20160409 21:15:57< vultraz> It's a radical idea.
20160409 21:16:03< fendrin> haskel?
20160409 21:16:07< vultraz> Wesnoth 2 will be radically different
20160409 21:16:38< celticminstrel> If that's all internal differences, fine, but if it's truly radically different, I think calling it Wesnoth 2 might be a bit of a stretch.
20160409 21:16:58< Dugi_> I suggested Haskell because you can write an entire algorithm in one line that isn't very long. It's quicker to write and more readable once you have the experience.
20160409 21:17:29< vultraz> And that might be something we want.
20160409 21:17:31< vultraz> *I don't know*
20160409 21:17:48< vultraz> I have not looked into this fully because it would require time *I do not have right now*
20160409 21:17:56< vultraz> And I don't mean today
20160409 21:17:57< vultraz> or this week
20160409 21:18:23< celticminstrel> Dugi_: We have a functional language in Wesnoth already, though it can't change the gamestate (except for AI moves).
20160409 21:18:51< celticminstrel> Also, despite being basically a functional language, functions aren't currently first-class values.
20160409 21:19:13< vultraz> So the plan is to work on 1.14
20160409 21:19:16< vultraz> Release that on Steam
20160409 21:19:30< lipkab> No Haskell please. Nobody's going create content if they need to learn Haskell for that.
20160409 21:19:32< vultraz> and once that's done, we'll sit down and decide what to do.
20160409 21:19:37< celticminstrel> Heh.
20160409 21:20:14< celticminstrel> I think any radical change (with the possible exception of eliminating GUI2) is most likely a bad one.
20160409 21:20:39< fendrin> I think my replacement for WML is a sane one.
20160409 21:20:45< Dugi_> celticminstrel: What functional language? Lua? That one is as much a functional language as C++.
20160409 21:21:04< celticminstrel> fendrin: But it's not backward compatible, so sane or not, it's not a good choice.
20160409 21:21:11< celticminstrel> Dugi_: I mean the formula language.
20160409 21:21:17< celticminstrel> Lua is procedural, obviously.
20160409 21:21:26< celticminstrel> Though does support a functional paradigm, kind of.
20160409 21:21:26< fendrin> celticminstrel: It is so similar that I would call it backwards compatible.
20160409 21:21:46< celticminstrel> fendrin: I seem to recall it being totally different.
20160409 21:21:55< zookeeper> vultraz, "we" don't need to decide anything. whichever devs want to work on wesnoth 2 will work on wesnoth 2. there is no collective decision to make.
20160409 21:22:04< celticminstrel> So either we're talking about different things, or our concept of "similar" is different.
20160409 21:22:26< Dugi_> celticminstrel: What is that formula language? Something related to formulaWML?
20160409 21:22:49< celticminstrel> Dugi_: FormulaAI, [filter]formula="...", $(...)
20160409 21:22:56< vultraz> zookeeper: It'd be better if wesnoth 2 were not a spinoff
20160409 21:23:09< celticminstrel> http://wiki.wesnoth.org/Wesnoth_Formula_Language
20160409 21:23:34< zookeeper> vultraz, yeah and i'd like that million dollars too. of course it's a spinoff.
20160409 21:23:39< vultraz> zookeeper: the collective decision would be about the future of the project
20160409 21:23:56< fendrin> celticminstrel: No, the syntax is very similar. I expect every current WML developers to easily adopt it.
20160409 21:24:03< vultraz> zookeeper: because we cannot keep rehashing these same damn points every time it comes up in chat
20160409 21:24:07< celticminstrel> "the syntax is very similar"
20160409 21:24:20< vultraz> zookeeper: some kind of general goal has to be set
20160409 21:24:21< celticminstrel> Show me a one-line example of your new syntax, then.
20160409 21:24:38< fendrin> celticminstrel: What do you want the line to do?
20160409 21:25:02< celticminstrel> I dunno, maybe something like a simple unit filter with id or x.y?
20160409 21:25:09< vultraz> zookeeper: IMO, it would be most unfortunate if we ended up with two different wesnoths being developed by two different teams
20160409 21:25:17< celticminstrel> ^x,y
20160409 21:25:30< Dugi_> celticminstrel: Oh, I have never seen how deep this can go. It's pretty cool. Far better than the one I have made myself.
20160409 21:25:31< celticminstrel> vultraz: Honestly, that currently feels inevitable to me.
20160409 21:26:16< celticminstrel> Dugi_: Until recently it was poorly documented and known only as FormulaAI, which gives the illusion that it's only usable in AI code. Admittedly, there aren't currently many hooks into it, but one (the $(...) syntax) is pretty widely used for very simple things.
20160409 21:26:24< vultraz> celticminstrel: that's exactly why I want 1.14 finished before anything is decided. So *if* that happens, there's a stable "old wesnoth".
20160409 21:26:54< celticminstrel> vultraz: Honestly, I think that anyone trying to make a spinoff of Wesnoth should have the decency to at least change the name, but I suppose there's no way to force them to.
20160409 21:26:58< Dugi_> I also had an idea that WML could be replaced by a proper programming language, like Java. And it could be allowed to do much more stuff to allow add-ons doing much cooler tricks without having to resort to dirty hacks.
20160409 21:27:01< fendrin> https://forums.wesnoth.org/viewtopic.php?p=595829#p595829
20160409 21:27:02< vultraz> celticminstrel: and if that doesn't happen, there STILL is a stable "old wesnoth"
20160409 21:27:03< zookeeper> vultraz, whether there's a wesnoth 2 or not, and regardless of how many devs move to working on it, the old wesnoth project will still remain. there's nothing to decide about its future except who stays.
20160409 21:27:34< vultraz> zookeeper: yes, exactly, see point above
20160409 21:28:08< fendrin> zookeeper: Or it just shares the fate of xfree86 vs xorg
20160409 21:28:09< celticminstrel> Dugi_: I sort of agree; ActionWML could be definitely be replaced with a proper programming language, though given how Wesnoth currently is, I think that'd most likely be Lua or WFL. WML in general though, as a declarative data language, I don't see any need to replace it with anything.
20160409 21:29:11-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20160409 21:29:39< vultraz> zookeeper: you're exactly right that the decision has to be about who stays.
20160409 21:29:43< fendrin> celticminstrel: I have converted the complete AoI campaign into my new WSL. So you can see a lot of example code.
20160409 21:29:45< celticminstrel> Terrain graphics and unit animations could be replaced with some new syntax, I suppose. Both are pretty complicated in WML form. That said, I'm not sure if there's a good syntax that would be simpler.
20160409 21:29:58< celticminstrel> fendrin: The fact that you needed to convert it is already a major alarm bell.
20160409 21:30:13< celticminstrel> And proof that it is not backwards compatible.
20160409 21:30:35< celticminstrel> If it were backwards compatible, you wouldn't need to convert anything. Maybe tweak a few minor things, but not a wholesale conversion.
20160409 21:30:46< fendrin> I think it is compatible enough.
20160409 21:30:50< zookeeper> vultraz, indeed, but if you frame it as "we need to decide the future of the project" then that sounds like something completely different.
20160409 21:31:11< celticminstrel> Yeah, seeing that thread, there's absolutely no way you can call that backwards compatible.
20160409 21:31:21< celticminstrel> There's no compatibility whatsoever.
20160409 21:31:56< fendrin> no?
20160409 21:32:28< celticminstrel> "Backwards compatible" means "it reads WML".
20160409 21:32:46< celticminstrel> What else would it mean, anyway?
20160409 21:32:53< fendrin> I think it is good enough to have a conversion tool.
20160409 21:33:02< celticminstrel> A conversion tool does help, admittedly.
20160409 21:33:17< fendrin> And to have a syntax and api that every wml developer can use in a short time.
20160409 21:33:24< celticminstrel> It's not as good as true backwards compatibility, but it's better than no compatibility at all.
20160409 21:33:35< vultraz> zookeeper: basically, do the majority of us want to work on wesnoth 2 or not? I would think if so, we'd quietly cease any new dev cycles on wesnoth 1 and only maintain it with critical bugfixes while we work on 2
20160409 21:33:53< celticminstrel> I think your syntax is about equal to WML in terms of basic usability, though, so there's not much point in switching to it.
20160409 21:34:04< celticminstrel> (Ignoring ActionWML.)
20160409 21:34:25< fendrin> no
20160409 21:34:32< celticminstrel> (But switching ActionWML to use Lua/WFL would be easier and more backwards compatible.)
20160409 21:34:36< fendrin> my syntax is a full featured turing complete language
20160409 21:34:58< fendrin> you can use lua and the whole power of it in every config file
20160409 21:35:05< celticminstrel> So it's like having a Turing complete preprocessor.
20160409 21:35:23< zookeeper> vultraz, sure, that would most likely happen if most devs decided to switch.
20160409 21:35:56< fendrin> celticminstrel: You don't need a preprocessor, you can simple use lua.
20160409 21:36:09< celticminstrel> But it's conceptually the same as a preprocessor.
20160409 21:36:26< fendrin> no
20160409 21:36:32< fendrin> not a little bit
20160409 21:36:38< celticminstrel> I don't see any difference.
20160409 21:37:39< lipkab> Oh, what's this Wesnoth 2 thing? Can I get a link?
20160409 21:37:58< vultraz> li
20160409 21:38:01< vultraz> lipkab: https://github.com/anura-engine/wesnoth2
20160409 21:39:14< fendrin> celticminstrel: A preprocessor is a text concatenator. It can't do anything else. Lua is a programming language.
20160409 21:39:22< fendrin> celticminstrel: The difference is huge.
20160409 21:39:42< celticminstrel> That's a difference in implementation, not in concept.
20160409 21:41:16< vultraz> celticminstrel, lipkab: right now it just looks like this https://drive.google.com/file/d/0B-mR9s8FduLLZUt4N3o2Q0NreUk/view?usp=sharing
20160409 21:41:42< Dugi_> When speaking about Wesnoth 2, everybody is speaking about better API and implementation, but what 99% of users is interested in are features. And I see almost no features added into wesnoth. Better add-on API, incremental art improvements and regressions are relatively good, but most of the new features are niche. It would need something major, like skeletal animations for units, tessellation shaders, particle effects, post-pro
20160409 21:42:09< celticminstrel> Skeletal animations? That sounds either weird or impossible.
20160409 21:42:09< vultraz> Dugi_: particle effects will definitely be a part of wesnoth 2
20160409 21:42:47< celticminstrel> BTW, your message was cut off at "post-pro".
20160409 21:42:59< fendrin> celticminstrel: Then please explain what concept the Wesnoth preprocessor is following and why any concept in my approach is similar to that.
20160409 21:43:44< celticminstrel> Defining portions of the config based on conditions, for example.
20160409 21:43:59< celticminstrel> Or defining portions of the config to be the return value of a function.
20160409 21:44:37< fendrin> okay, I guess I know how you think now
20160409 21:44:58< fendrin> yes, then the lua/moonscript is used that way.
20160409 21:45:13< celticminstrel> And the current preprocessor does that as well.
20160409 21:45:27< fendrin> But it can't offer functions.
20160409 21:45:32< celticminstrel> The Lua does admittedly add more opportunities - loops, for example. Though I'm not sure you could actually do that.
20160409 21:45:41< fendrin> And it can't be linted properly because of the ugly macros.
20160409 21:45:50< celticminstrel> The preprocessor does sort of offer functions - macros are function-like.
20160409 21:45:52< Dugi_> celticminstrel: I have not seen any cutoff. It is a problem of your client. I thought the skeletal animations like an alternative way of animating units, you bind pixels to bones you define and move the bones. It will have enough frames, so the movement will look smooth and hide the oddness of pixel interpolation. It might be also used as blueprint for actual animation frames.
20160409 21:45:57< vultraz> honestly, the preprocessor should go
20160409 21:46:10< fendrin> ^
20160409 21:46:28< celticminstrel> Dugi_: It's not a problem in the client. It's a problem on the IRC server.
20160409 21:46:40< zookeeper> ^
20160409 21:46:47< celticminstrel> I do agree that the preprocessor isn't really great.
20160409 21:47:04< vultraz> or a problem on YOUR client. Some split extra-long messages in two
20160409 21:47:08< celticminstrel> Dugi_: The reason I said skeletal movement sounds impossible is because that's for 3D imaging, not sprites.
20160409 21:47:09-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160409 21:47:31< celticminstrel> But maybe it can be used for sprites too... I dunno...
20160409 21:47:40< celticminstrel> Seems dubious though.
20160409 21:48:00< celticminstrel> Anyway, I do agree that the preprocessor isn't really great, and it would not be bad to phase it out.
20160409 21:48:17< celticminstrel> But I'm not sure that eliminating it altogether is the right thing to do, either.
20160409 21:48:33< vultraz> I wouldn't mind keeping certain parts of  WML if we didn't have the preprocessor
20160409 21:48:54< lipkab> Er, how many active devs there are these days? I mean, you can dream about rewriting the whole game and everything, but Wesnoth has 10 years of development to catch up with, and last I checked the project was pretty low on manpower...
20160409 21:49:00< fendrin> A clever conversion tool can translate 98% of all macro usage into lua functions or lua variables.
20160409 21:49:20< Dugi_> WML preprocessor is the #define BLAHBLAHBLAH thing? I don't like it and it's abused a lot, but it's useful in some cases.
20160409 21:49:32< celticminstrel> ^
20160409 21:49:38< vultraz> lipkab: as far as I know, both dave and jet want to work on wesnoth 2 personally
20160409 21:49:46< celticminstrel> Dugi_: Also #ifdef and such, and {}
20160409 21:50:04< celticminstrel> lipkab: Maybe, uhh... four or five? I'm not too sure.
20160409 21:50:30< fendrin> Dugi_: But real functions are a good replacement for it.
20160409 21:50:30< fendrin> real functions that can go into recursion
20160409 21:50:44< fendrin> you should not need the macro system anymore
20160409 21:51:02< vultraz> Dugi_: the most annoying part of the preprocessor is having to quit to the main menu to refresh the config and then reload a start-of-scenario save whenever you change one line of WML in your addon
20160409 21:51:08< fendrin> And the power of lua everywhere
20160409 21:51:14< fendrin> sharing the same syntax everywhere
20160409 21:51:14< vultraz> it's such a huge waste of time for umc authors
20160409 21:51:19-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 260 seconds]
20160409 21:51:19-!- wedge010 is now known as wedge009
20160409 21:51:34< Dugi_> fendrin: Not if it's used when declaring data. Stuff like {WEAPON_SPECIAL_MARKSMAN} can hardly be replaced by any functions. The implementation would have to be much more complex without it.
20160409 21:52:23< vultraz> not necessarily 
20160409 21:52:30< fendrin> I have already converted WEAPON_SPECIAL_MARKSMAN 
20160409 21:52:31< vultraz> not if we had a way to define data set objects
20160409 21:52:33< fendrin> in my codebase
20160409 21:53:01< fendrin> It is not a function but a variable that holds a table iirc
20160409 21:53:04< Dugi_> I agree that it sucks and I am trying to put something together without it and it works, but the number of changes to the logic of WML it would bring...
20160409 21:53:35< fendrin> can be kept to a minimum
20160409 21:54:09< vultraz> I very much prefer coding in lua for my wesnoth umc, since I can just reload a turn save in 2 seconds and all the changes are there.
20160409 21:54:25< fendrin> I am working on this since roughly a year.
20160409 21:54:54< vultraz> fendrin: the problem is, last I checked you also wanted to make wesnoth playable like Dwarf Fortress :\
20160409 21:55:26< Dugi_> vultraz: You mean like in Terminal?
20160409 21:55:38< vultraz> yes
20160409 21:55:51< vultraz> that would be horrible
20160409 21:56:01< Dugi_> That would make it super nerdy. Good for times when you break your X server.
20160409 21:56:21< fendrin> I have written an ascii client for my engine.
20160409 21:56:21< fendrin> To have some way to see the map.
20160409 21:56:21< fendrin> hihi
20160409 21:56:21< fendrin> But it is more a tool for me to develop
20160409 21:56:31< fendrin> than a serious attempt at writting a display client for my game server.
20160409 21:58:29< fendrin> vultraz: My proposal focuses on a game server. I like to get one thing done right. If I would start at every end of the problem the project would not reach an end.
20160409 22:01:05-!- Kwandulin [~Miranda@p200300760F0D015135969ED0D2C9FEC4.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160409 22:03:08< Dugi_> It's too late for me. Good night.
20160409 22:03:11-!- Dugi_ [b28f8f9e@gateway/web/freenode/ip.178.143.143.158] has quit []
20160409 22:06:05< lipkab> Gah, some people can afford to get to bed at midnight. What a lucky man.
20160409 22:19:13-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160409 22:30:03-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 240 seconds]
20160409 22:59:53-!- lipkab [~the_new_l@host-91-147-210-58.biatv.hu] has quit [Quit: Leaving]
20160409 23:15:32-!- irker026 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160409 23:34:42< Aginor> good morning
20160409 23:35:06< celticminstrel> Hello.
20160409 23:35:39< celticminstrel> What about PR 631?
20160409 23:35:51< celticminstrel> Or others that you have opinions on.
20160409 23:39:56 * Aginor has a lot of opinions
20160409 23:40:04< Aginor> I wsa just forming a few new actually :D
20160409 23:40:39< Aginor> I've just started a new job, again, so I've been pretty mentally exhausted in the evenings
20160409 23:42:11 * Aginor reads 631 again
20160409 23:43:23< Aginor> celticminstrel: I think your comment in play_controller makes a great deal of sense
20160409 23:43:28< Aginor> has this been tested in the editor?
20160409 23:49:58< Aginor> I've got to be somewhere in 40 minutes, I'll check out and test later
20160409 23:50:49< Aginor> celticminstrel: I do not think #641 should be merged
--- Log closed Sun Apr 10 00:00:16 2016