--- Log opened Mon May 08 00:00:28 2017
20170508 00:01:25-!- Greg-Boggs [~greg_bogg@2601:1c2:f00:9780:b8f3:10ec:f389:b4a9] has joined #wesnoth-dev
20170508 00:01:44< vultraz_iOS> celticminstrel: under my proposal, any fixed open bugs would be marked w/ 1.13.8 and closed
20170508 00:03:52-!- Appleman1234 [~Appleman1@pl32183.ag1212.nttpc.ne.jp] has joined #wesnoth-dev
20170508 00:06:17< celticminstrel> vultraz_iOS: So can I start uploading now and we close the fixed bugs once it's done, or do you want the fixed bugs omitted?
20170508 00:06:36< vultraz_iOS> celticminstrel: I concede your objection RE point 3
20170508 00:06:51< celticminstrel> The fixed ones are probably recently fixed...
20170508 00:07:21< vultraz_iOS> Upload, including the fixed open ones, and then apply the protocol 
20170508 00:07:24< vultraz_iOS> ie, close and mark 1.13.8
20170508 00:07:38< vultraz_iOS> (I assume you have no objections to the rest of my proposal?)
20170508 00:07:53< celticminstrel> None.
20170508 00:08:00< vultraz_iOS> Sweet.
20170508 00:08:07< vultraz_iOS> Let me update the text, then.
20170508 00:08:13< celticminstrel> ?
20170508 00:08:17< celticminstrel> You mean on the wiki?
20170508 00:08:27< vultraz_iOS> in my post first, then I can copy it somewhere
20170508 00:09:43< celticminstrel> Okay, it's gonna be roughly one bug per minute, so at least twelve hours. Probably more since abuse limits will almost certainly be hit.
20170508 00:10:00< celticminstrel> There's not much point in doing anything with the bugs until the entire transfer is completed IMO.
20170508 00:10:08< celticminstrel> Though if you want to comment on one or something, that's fine.
20170508 00:10:20< vultraz_iOS> alright
20170508 00:11:06< vultraz_iOS> hm.....I wonder if I should add some text saying that if its a bug known to be master-only, no milestone need be set.
20170508 00:13:59< vultraz_iOS> oh jesus this is going to spam people's inboxes isn't it..
20170508 00:14:18< celticminstrel> Yup.
20170508 00:14:28< vultraz_iOS> any way to disable that..
20170508 00:14:48< celticminstrel> You could disable email notifications until it's done.
20170508 00:14:53< vultraz_iOS> where?
20170508 00:15:04< celticminstrel> But I guess that only helps you, not everyone else...
20170508 00:15:15< celticminstrel> https://github.com/settings/notifications
20170508 00:17:12< vultraz_iOS> done so
20170508 00:18:36< celticminstrel> (I paused the script because I realized I'd forgotten I disabled assigness for the trial run.)
20170508 00:22:19-!- Greg-Boggs [~greg_bogg@2601:1c2:f00:9780:b8f3:10ec:f389:b4a9] has quit [Remote host closed the connection]
20170508 00:25:17< celticminstrel> Gonna pause again because I just thought of a thing.
20170508 00:30:15-!- Greg-Boggs [~greg_bogg@2601:1c2:f00:9780:b8f3:10ec:f389:b4a9] has joined #wesnoth-dev
20170508 00:39:27-!- Greg-Boggs [~greg_bogg@2601:1c2:f00:9780:b8f3:10ec:f389:b4a9] has quit [Remote host closed the connection]
20170508 01:01:08-!- Greg-Boggs [~greg_bogg@2601:1c2:f00:9780:b8f3:10ec:f389:b4a9] has joined #wesnoth-dev
20170508 01:09:10-!- gfgtdf [~chatzilla@x4e36880b.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.93 [Firefox 53.0.2/20170504105526]]
20170508 01:31:29-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20170508 01:36:24< celticminstrel> Alright, assignees work, though perhaps I should've included boucman in the list of people to unassign. Oh well.
20170508 01:42:24< celticminstrel> Oh yeah, I should restore music playlist after Lua calls the storyscreen, right?
20170508 01:42:45< celticminstrel> Should be relatively easy using wesnoth.music_list.all.
20170508 01:43:22< celticminstrel> If I only want it to apply to the WML tag, at least.
20170508 01:50:17< pydsigner> Woah bug inbox spam
20170508 01:51:26< celticminstrel> Yeah sorry. >_>
20170508 01:52:03< celticminstrel> I find it annoying even without email notifications off.
20170508 02:14:31-!- Greg-Boggs [~greg_bogg@2601:1c2:f00:9780:b8f3:10ec:f389:b4a9] has quit [Remote host closed the connection]
20170508 02:20:39-!- irker157 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20170508 02:29:07 * vultraz_iOS is idly reading https://www.mail-archive.com/dri-devel@lists.sourceforge.net/msg39091.html
20170508 02:29:19< vultraz_iOS> (recommend from WesnothRepository)
20170508 02:30:24< vultraz_iOS> this guy makes it sound like it's totally unacceptable to rebase someone's commit ever heh
20170508 02:30:30< vultraz_iOS> even locally
20170508 02:49:25-!- Greg-Boggs [~greg_bogg@2601:1c2:f00:9780:b8f3:10ec:f389:b4a9] has joined #wesnoth-dev
20170508 03:04:36-!- Greg-Boggs [~greg_bogg@2601:1c2:f00:9780:b8f3:10ec:f389:b4a9] has quit [Remote host closed the connection]
20170508 03:10:24< celticminstrel> I think you misread. It sounds like he's saying you should never rebase anything unless it's local (as in, never been pushed), but if it's local, anything goes even if it's someone else's commit.
20170508 03:10:44-!- Appleman1234 [~Appleman1@pl32183.ag1212.nttpc.ne.jp] has quit [Ping timeout: 260 seconds]
20170508 03:13:36< celticminstrel> Aww, it crashed.
20170508 03:25:04-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170508 03:25:13-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170508 03:25:36-!- Greg-Boggs [~greg_bogg@c-76-115-139-154.hsd1.or.comcast.net] has joined #wesnoth-dev
20170508 03:46:35< celticminstrel> Ugh, I thought markup was being translated, but it wasn't?
20170508 03:46:44< celticminstrel> Well, I guess people can just fix it as they notice it. :(
20170508 03:48:04< celticminstrel> Ahhh, I translated it in original summary but forgot in comments. Okay then.
20170508 03:51:25-!- Kwandulin [~Kwandulin@p200300760F6D8083E939901127AC211B.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170508 03:57:44< celticminstrel> For some reason mattsc assigned himself to something and then unassigned himself.
20170508 04:03:40-!- Greg-Boggs [~greg_bogg@c-76-115-139-154.hsd1.or.comcast.net] has quit [Remote host closed the connection]
20170508 04:05:03-!- JyrkiVesterinen [~jyrki@87-100-233-202.bb.dnainternet.fi] has joined #wesnoth-dev
20170508 04:12:34-!- irker171 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20170508 04:12:34< irker171> wesnoth: Charles Dang wesnoth:master dd692343501e / src/ (5 files in 3 dirs): Editor: store map generators with smart pointers https://github.com/wesnoth/wesnoth/commit/dd692343501e871359239d8534efe6aa135ee192
20170508 04:14:30< celticminstrel> If you want to work on more editor stuff, maybe you could fix the autosave issue. >_>
20170508 04:15:12< vultraz_iOS> hm?
20170508 04:15:15< irker171> wesnoth: Charles Dang wesnoth:master 89f8b451dcba / src/editor/map/context_manager.cpp: Editor: auto-save maps starting with visible index 1 https://github.com/wesnoth/wesnoth/commit/89f8b451dcba88107d96f87c1d1c1513aefe26bf
20170508 04:15:22< vultraz_iOS> what autosave issue?
20170508 04:15:42< celticminstrel> When reloading with F5.
20170508 04:15:48< vultraz_iOS> what about it?
20170508 04:15:54< celticminstrel> It auto-saves your maps.
20170508 04:15:57< vultraz_iOS> and?
20170508 04:16:14< celticminstrel> Well, that means it overwrites your map file when you didn't ask it to.
20170508 04:16:31< vultraz_iOS> oh come on
20170508 04:16:47< vultraz_iOS> now that is just pedantic 
20170508 04:16:52< celticminstrel> ...what are you talking about?
20170508 04:16:56< celticminstrel> It's not pedantic.
20170508 04:17:02< vultraz_iOS> what would you have it do
20170508 04:17:05< celticminstrel> It's surprising behaviour and could lead to loss of data.
20170508 04:17:46< vultraz_iOS> really, what would you have it do
20170508 04:18:01< celticminstrel> I'd have it not overwrite the original file.
20170508 04:18:16< celticminstrel> Probably save it to one of the window_i files used for never-saved maps.
20170508 04:18:29< celticminstrel> Of course it also has to somehow remember where the original file was.
20170508 04:18:37 * vultraz_iOS curses
20170508 04:18:39< celticminstrel> It probably won't be easy.
20170508 04:18:44< vultraz_iOS> I'm not implementing that
20170508 04:19:07< celticminstrel> Well okay, I'll go ahead and open the issue I was planning to open then... once this transfer ends so it doesn't get lost in the sea of old issues.
20170508 04:19:55< vultraz_iOS> i mean, it already does window_i + 1 for the unsaved maps
20170508 04:22:30< celticminstrel> One thing I considered was moving the context_manager out of the loop, so that the contexts are not destroyed when the data is reloaded; then the auto-save wouldn't be necessary at all.
20170508 04:23:00< celticminstrel> I haven't looked closely enough to discern how possible that is though.
20170508 04:23:23< vultraz_iOS> you do realize f5 *closes the editor*
20170508 04:23:34< celticminstrel> Yes, I saw that.
20170508 04:33:00< vultraz_iOS> why is wesnoth bugs pinging me
20170508 04:33:14< vultraz_iOS> I see
20170508 04:33:18< vultraz_iOS> a bug was assigned to me
20170508 04:33:43-!- Appleman1234_ [~Appleman1@pl12919.ag1212.nttpc.ne.jp] has joined #wesnoth-dev
20170508 04:48:48-!- JyrkiVesterinen [~jyrki@87-100-233-202.bb.dnainternet.fi] has quit [Quit: .]
20170508 04:48:48-!- Kwandulin [~Kwandulin@p200300760F6D8083E939901127AC211B.dip0.t-ipconnect.de] has quit [Quit: [endlevel] result=novictorynodefeatjustquit [/endlevel]]
20170508 04:49:32< vultraz_iOS> in template deduction is const taken into account.
20170508 04:50:49< vultraz_iOS> since i was wondering
20170508 04:52:28< vultraz_iOS> if you have template<typename T> struct foo { T member; } foo<const type>  im pretty sure member would be const type wouldn it
20170508 04:53:08< vultraz_iOS> in which case..
20170508 04:54:05< celticminstrel> Yes.
20170508 04:54:24< celticminstrel> Everything is taken into account in template type deduction.
20170508 04:54:40< celticminstrel> ...well, maybe not bit fields, I guess.
20170508 04:54:56< celticminstrel> But everything that's a real type at least.
20170508 04:56:04-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20170508 05:00:45< vultraz_iOS> hmm
20170508 05:04:27-!- Appleman1234_ [~Appleman1@pl12919.ag1212.nttpc.ne.jp] has quit [Ping timeout: 240 seconds]
20170508 05:30:31-!- Greg-Boggs [~greg_bogg@2601:1c2:f00:9780:b8f3:10ec:f389:b4a9] has joined #wesnoth-dev
20170508 05:45:22-!- Greg-Boggs [~greg_bogg@2601:1c2:f00:9780:b8f3:10ec:f389:b4a9] has quit [Remote host closed the connection]
20170508 05:48:10< celticminstrel> Almost one-third done, huh...
20170508 05:51:55< wedge009> celticminstrel: I don't have it... looks like it wasn't built. Do we have a list of things that are missing so we can update aquileia's instructions?
20170508 05:54:01< celticminstrel> wedge009: With respect to lib/dll files, I'm pretty sure the unit tests are the only one missing. For headers, I think the best way to check is with grep...
20170508 05:54:32< irker171> wesnoth: Charles Dang wesnoth:master 9c85689914c0 / src/sdl/ (surface.cpp surface.hpp): Use a class template for surface_lock and const_surface_lock https://github.com/wesnoth/wesnoth/commit/9c85689914c0c8cad99e3463a8f99ba5f383e3ab
20170508 06:04:43< vultraz_iOS> wedge009: are you talking about missing files when building boost with aquelia's instructions?
20170508 06:10:06< celticminstrel> Seems like one minute between issues is enough to prevent triggering the abuse limiter.
20170508 06:10:20< celticminstrel> It hasn't been hit even once so far.
20170508 06:10:54< wedge009> vultraz_iOS: celticminstrel: Yes, I'm just trying to build unit_test then I'll add it to the instructions.
20170508 06:11:23< celticminstrel> Still Boost 1.63?
20170508 06:11:26< vultraz_iOS> wedge009: honestly, i just skip the last step and use all of boost
20170508 06:11:39< vultraz_iOS> the last step being the "separate out the required sections" 
20170508 06:11:50< celticminstrel> I don't really have a problem with the last step.
20170508 06:11:54< wedge009> I do too, but it's mainly for copying stuff into the external repository.
20170508 06:11:56< vultraz_iOS> all of boost that was built*
20170508 06:12:01< wedge009> It's the lib that I'm missing.
20170508 06:12:11< wedge009> First part lib, second part headers.
20170508 06:12:14< celticminstrel> Right, the lib and dll is missing.
20170508 06:12:32< wedge009> Because I've built the entire boost library before and it does take a while.
20170508 06:12:47< wedge009> Just building the bits used by Wesnoth seems sensible/.
20170508 06:12:48< wedge009> .
20170508 06:13:00< wedge009> But anyway, it's complaining unit_test is not a library name.
20170508 06:14:59< celticminstrel> Hmm...
20170508 06:16:24< celticminstrel> I definitely got linker errors. I'm rebuilding now to remind myself.
20170508 06:19:28-!- Netsplit *.net <-> *.split quits: janebot
20170508 06:21:37< wedge009> Ah, looks like the library is just 'test'.
20170508 06:21:53< wedge009> libboost_unit_test_framework-vc141-mt-gd-1_64.lib
20170508 06:22:09< wedge009> Will push when compilation done.
20170508 06:23:29< celticminstrel> Yeah, "test" is already in the instructions on master, too.
20170508 06:24:01< wedge009> Uh, not the one I'm seeing.
20170508 06:24:30< wedge009> But it should now, since I just updated.
20170508 06:24:31< wedge009> it.
20170508 06:24:43< wedge009> https://github.com/aquileia/external
20170508 06:25:23< wedge009> Oh I see.
20170508 06:25:30< wedge009> test is included in the headers but not the libs.
20170508 06:26:20< celticminstrel> Ah. Yeah, I was referring to the list of headers.
20170508 06:27:41< celticminstrel> Once that's in, I think we should get Appveyor to run the unit tests too.
20170508 06:28:17< celticminstrel> IIRC it currently runs only the WML unit tests.
20170508 06:28:30-!- JyrkiVesterinen [~JyrkiVest@85.76.66.132] has joined #wesnoth-dev
20170508 06:29:41< wedge009> Push to VC15 branch complete.
20170508 06:30:01< celticminstrel> Yay! But what about VC14 and VC12?
20170508 06:33:10< celticminstrel> I guess it needs to be recompiled with each version.
20170508 06:37:00< wedge009> Probably, but I never had VC12 and unfortunately I got rid of VC14 several weeks ago. Was running out of space on the SSD even though I got a new one now.
20170508 06:37:08-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20170508 06:37:17< celticminstrel> Aw...
20170508 06:38:54< wedge009> BTW, did gfgtdf manage to get config_key_type working with VS2017? I saw some progress being made on the config.* files and when I checked last night (~18 hours ago) it was down to only a couple of compilation errors instead of dozens.
20170508 06:39:15< celticminstrel> No idea.
20170508 06:39:45< wedge009> Just checked the logs, doesn't look like it. That's okay.
20170508 06:42:43< celticminstrel> I don't understand why it's giving "unreferenced variable" at src/tests/gui/test_gui2.cpp:202...
20170508 06:43:08< celticminstrel> Well, the error's on 201, the reference of it is 202.
20170508 06:43:58< celticminstrel> It also gives MSB8012 warnings.
20170508 06:44:06< celticminstrel> And a couple of warnings in lexical_cast.
20170508 06:44:22< celticminstrel> Not to mention gazillions of warnings in Boost itself.
20170508 06:47:46< JyrkiVesterinen> wedge009: I got config_key_type working with VS2017. I verified myself that I was able to build the game.
20170508 06:48:28< JyrkiVesterinen> This commit fixed it: https://github.com/wesnoth/wesnoth/commit/1eca3736f65639d8dfc725c9ceee58ee56abdd88
20170508 06:50:14< JyrkiVesterinen> I see that the latest AppVeyor VS2017 build passed as well: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt
20170508 06:51:28< JyrkiVesterinen> Hm. I see that gfgtdf has sent AppVeyor to auto-build with VS2017 too. :)
20170508 06:51:52< JyrkiVesterinen> *set
20170508 06:51:54< celticminstrel> Should we enable the unit tests on AppVeyor?
20170508 06:52:02< JyrkiVesterinen> They are already enabled.
20170508 06:52:15< celticminstrel> I'm pretty sure they aren't.
20170508 06:52:32< JyrkiVesterinen> To be more specific, WML tests are automatically run.
20170508 06:52:38< celticminstrel> Though I'm also not sure they can be with the current way the project is configured...
20170508 06:53:17< JyrkiVesterinen> Hmm, I haven't even tried to run the Boost unit tests on Windows.
20170508 06:53:23< celticminstrel> To build unit tests you use the Test_Debug or Test_Release configuration.
20170508 06:53:48< celticminstrel> IIRC I got them to build a little while ago...
20170508 06:54:36< celticminstrel> This and the preceding two: https://github.com/wesnoth/wesnoth/commit/0828c2bdf881ba61e884ac89e4765985c77b168f
20170508 06:55:26< celticminstrel> ^preceding three
20170508 06:58:33< wedge009> JyrkiVesterinen: Your commit was a couple of days ago, but I still seemed to get issues. Well, I'm recompiling with today's pull, I'll see how it goes. Thanks.
20170508 07:11:01-!- atarocch [~atarocch@93.56.160.28] has joined #wesnoth-dev
20170508 07:11:28< zookeeper> vultraz_iOS, yeah the editor F5 saving issue isn't pedantic at all, it's caused all sorts of unintended changes for me countless times. i'm just lucky that probably the maps have all been under version control.
20170508 07:17:26< zookeeper> celticminstrel, uh oh, there's an url problem. check the forum link in this one: https://github.com/wesnoth/wesnoth/issues/1299
20170508 07:17:40< zookeeper> (original: https://gna.org/bugs/?19596 )
20170508 07:20:30< celticminstrel> Fixed that specific one.
20170508 07:20:39< celticminstrel> But I wonder how many more there are...
20170508 07:20:44< JyrkiVesterinen> Hmm, I can no longer build Wesnoth with VS2017. Errors in Boost headers.
20170508 07:20:45< JyrkiVesterinen> Error	C2041	illegal digit 'U' for base '10' (compiling source file ..\..\src\tests\test_commandline_options.cpp)	wesnoth	f:\battle for wesnoth\external\include\boost\preprocessor\slot\detail\shared.hpp	137
20170508 07:24:36< JyrkiVesterinen> Hmm, it may actually well be specific to trying to build the _unit tests_. I'll try building the actual game instead.
20170508 07:24:41< wedge009> JyrkiVesterinen: Down to one error for me.
20170508 07:24:56< wedge009> brush.obj : error LNK2001: unresolved external symbol "public: class boost::iterator_range<struct config::const_child_iterator> __thiscall config::child_range(class std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> > const &)const " (?child_range@config@@QBE?AV?$iterator_range@Uconst_child_iterator@config@@@boost@@ABV?$basic_string@DU?$char_traits@D@std@@V?$allocator@D@2@@std@@@Z)
20170508 07:25:09< wedge009> Looks like it's still related to the config classes.
20170508 07:32:48< vultraz_iOS> zookeeper: would a confirmation prompt informing the player that reloading WML will save maps suffice as a temporary stopgap measure
20170508 07:33:52< zookeeper> vultraz_iOS, sure, a prompt that pops up if there are unsaved maps currently open, and lets you either proceed or cancel.
20170508 07:35:49< zookeeper> not sure how it should treat new maps
20170508 07:37:06< zookeeper> of course it should ignore the default starting map, but if you create new ones and haven't saved them, should it ignore those too? i guess i'd lean towards yes
20170508 07:44:30< celticminstrel> Ignore those for which name.empty()?
20170508 07:46:32< JyrkiVesterinen> I get the "illegal digit U" error even when trying to build the game itself...
20170508 07:47:13< JyrkiVesterinen> A difference from when I built it earlier is that I'm at work (i.e. different computer) now, but I don't see how it would result in such an error.
20170508 07:49:10< vultraz_iOS> "The specified character is not a valid digit for the base (such as octal or hex)."
20170508 07:49:30< JyrkiVesterinen> Yes, but U isn't a digit. It means that the constant is unsigned.
20170508 07:50:02< JyrkiVesterinen> It's also strange that VS doesn't complain about other unsigned constants in the same file, until line 137.
20170508 08:08:08< JyrkiVesterinen> Okay, I tried a full rebuild, and now the build succeeded. :)
20170508 08:08:21< JyrkiVesterinen> I'm not getting wedge009's linking error, either.
20170508 08:08:26< wedge009> Was wondering if maybe I should do the same.
20170508 08:08:35< wedge009> But it takes sooooooo long. D:
20170508 08:08:40< wedge009> Even with 12 threads.
20170508 08:09:21< JyrkiVesterinen> At times like this I'm happy that I decided to buy a Core i7 gaming laptop for work...
20170508 08:09:31< wedge009> How long did yours take?
20170508 08:09:39< JyrkiVesterinen> (In fact, my work laptop has a *more* powerful processor than my gaming PC.)
20170508 08:09:53< wedge009> They're still thermally-limited, though, I thought.
20170508 08:09:59< JyrkiVesterinen> It took me 20 minutes.
20170508 08:10:15< wedge009> I'm running i7-4930K, was intended to last me several years.
20170508 08:10:28< wedge009> Fine, I'll rebuild all. ~hmph~
20170508 08:10:43< JyrkiVesterinen> My building time is most likely I/O-bound. I have Wesnoth in an external hard drive (which is exactly why I can also build it here without needing to dowload it again).
20170508 08:11:00< wedge009> celticminstrel: Do you run the wesnoth-bugs account? Is there a limit on issues or are you still in progress of migrating?
20170508 08:11:04< JyrkiVesterinen> For me, it's i5-4430 at home and i7-6700HQ at work.
20170508 08:11:07< wedge009> Fair enough, and good idea.
20170508 08:11:18< wedge009> I'm on VM at work. :/
20170508 08:11:39< celticminstrel> wedge009: Still in progress. There are around 640 issues in total to migrate, and it waits a minute after each.
20170508 08:11:46< wedge009> Oh right. Cool, thanks.
20170508 08:11:58< celticminstrel> To avoid triggering rate-limiting.
20170508 08:12:03< loonycyborg> wow suddenly I get 900+ new messages
20170508 08:12:08< celticminstrel> Sorry. <_<
20170508 08:12:17< wedge009> JyrkiVesterinen: I run other things too, otherwise recompiling wouldn't be a big deal.
20170508 08:12:24< wedge009> celticminstrel: Good idea.
20170508 08:15:14< wedge009> Oh, I also noticed the export copy was a bit old because I marked one of mine as Fixed but the GH copy is still Ready for Test.
20170508 08:17:01< loonycyborg> can github tracker notify people who don't have gh accounts by mail?'
20170508 08:17:25< loonycyborg> that is can one be subscribed on an issue by mail only?'
20170508 08:17:52< JyrkiVesterinen> Unit tests still fail to build.
20170508 08:17:53< JyrkiVesterinen> Error (active)	E0135	namespace "boost::unit_test::runtime_config" has no member "LOG_LEVEL"	wesnoth	f:\Battle for Wesnoth\wesnoth\src\tests\main.cpp	89	
20170508 08:18:07< celticminstrel> loonycyborg: Most likely not, unless you can set up some sort of webhook for it.
20170508 08:19:45< loonycyborg> would that anonymous reporting system handle this?
20170508 08:20:07< celticminstrel> Not sure.
20170508 08:35:43< wedge009> JyrkiVesterinen: Just finished building - yep, all okay now.
20170508 08:37:25< vultraz_iOS> wow, I actually have a newer CPU than jyrki :o (an i7-7500U)
20170508 08:37:43< JyrkiVesterinen> ..how is that surprising?
20170508 08:38:06< vultraz_iOS> I'm used to everyone having newer computers than I :P
20170508 08:38:12< vultraz_iOS> esp in wesnoth
20170508 08:38:47< zookeeper> you have an i7 and you're used to everyone else having newer ones? :p surely you are mistaken
20170508 08:39:46< zookeeper> even mine is not the oldest system that i think someone here is using
20170508 08:40:13< vultraz_iOS> before earlier this year i had an i5 3rd gen
20170508 08:40:48< zookeeper> sounds new and flashy
20170508 08:41:18< JyrkiVesterinen> Just looked it up. While i7-7500U is definitely newer than what any of us has, it's also slower. :P
20170508 08:41:35< JyrkiVesterinen> It's a mere dual-core processor, whereas mine are quadcores.
20170508 08:41:54< celticminstrel> I assume my Mac is the oldest system anyone here uses.
20170508 08:43:02< zookeeper> i wouldn't have thought there are dual-core i7's. not that i know much about cpu's, but still.
20170508 08:51:36< wedge009> I'm also surprised there are dual-core i7s, but these are U-series, presumably for Ultra-low power.
20170508 08:52:30< wedge009> Plus Kaby Lake is the most recent generation of CPUs from Intel - why would you be surprised at having a newer CPU?
20170508 08:55:36 * zookeeper is still happily running on a 2008 CPU
20170508 08:55:53< wedge009> zookeeper: celticminstrel: Well, my Linux system - at least the one that I build Wesnoth on - is my oldest system that's currently still in active daily use. Core 2 T9300 dual-core Penryn laptop. Bought August 2008.
20170508 08:56:09< celticminstrel> Seems like the transfer is approximately half done.
20170508 08:56:14< wedge009> Yay.
20170508 08:56:26< celticminstrel> wedge009: I got my Mac in 2006.
20170508 08:56:32< wedge009> Ah, nice.
20170508 08:57:00< wedge009> My stuff older than 2008 are non-functional - I wear them hard.
20170508 08:57:08< wedge009> Generally non-functional at least.
20170508 08:57:18< wedge009> I still have working Athlon systems.
20170508 08:57:36< wedge009> But not in regular use.
20170508 08:58:17< zookeeper> i barely hobbled along with a very crappy old system until about ~5 years ago when i got this one, and it's been smooth sailing ever since
20170508 08:58:27< celticminstrel> It has been in regular use that entire time, too.
20170508 08:58:33< wedge009> Five years is a reasonable time. Although recent years CPUs have been pretty stable.
20170508 08:58:43< wedge009> As in, not much progress.
20170508 08:58:57< wedge009> 11 years is commitment, though.
20170508 08:59:28< celticminstrel> More like I had no money to upgrade. Macs are expensive.
20170508 08:59:45< celticminstrel> That's why I got a new Windows computer instead (that, and I wanted to play games that didn't run on Mac).
20170508 08:59:48< wedge009> Even my Dad only waited around 8 or so years - I built Intel Ivy Bridge to replace his old Pentium 4.
20170508 09:00:01< wedge009> Heh, that's Apple for you. :/
20170508 09:00:37< wedge009> That said, I'd be interested in seeing how a Ryzen laptop will do, should AMD ever get around to releasing them...
20170508 09:01:43-!- atarocch [~atarocch@93.56.160.28] has quit [Remote host closed the connection]
20170508 09:01:44< JyrkiVesterinen> Speculation is that Ryzen should work extremely well in laptops.
20170508 09:02:14< JyrkiVesterinen> The GlobalFoundries 14 nm manufacturing process doesn't scale well to the clock speeds at which desktop Ryzen operates.
20170508 09:02:20< celticminstrel> Ooh, I like #1413 (GNA 22779)
20170508 09:02:45< JyrkiVesterinen> Desktop Ryzen is at the limits of how fast the process can go, and it suffers from quite poor efficiency as a result.
20170508 09:02:52< celticminstrel> The general idea of a piece of text attached to a unit (rather than a hex) is also good.
20170508 09:03:10< wedge009> Still better than the disappointment of Bulldozer, surely, though?
20170508 09:03:11< celticminstrel> I think some scenarios do that with normal map labels already.
20170508 09:03:32< JyrkiVesterinen> In mobile processors which operate at lower clock speeds, it appears that the 14 nm process is extremely efficient and even significantly ahead of Intel.
20170508 09:04:11< JyrkiVesterinen> Yes, desktop Ryzen is definitely better than Bulldozer and its derivatives were.
20170508 09:04:19< wedge009> Sounds good, but I'll wait for actual results. Too much speculation leads to disappointment.
20170508 09:04:43< wedge009> And sorry, celmin, I did see #1413 but don't have anything to comment on it.
20170508 09:04:47< JyrkiVesterinen> Regardless of the process disadvantage, Ryzen is still the better option on desktop due to the advantages in core count and SMT support.
20170508 09:05:01< JyrkiVesterinen> On mobile the difference should be even larger.
20170508 09:05:03< wedge009> I like APUs in mobile factors.
20170508 09:05:14< wedge009> And AMD still has the edge in graphics area, I believe.
20170508 09:05:15< celticminstrel> Uh, why are you apologizing?
20170508 09:05:25< wedge009> Just so you know we're not ignoring you?
20170508 09:05:33< wedge009> Never mind.
20170508 09:05:38< celticminstrel> Well okay.
20170508 09:05:39< wedge009> Trying to be polite.
20170508 09:05:50< celticminstrel> Anyway, time to snooze,
20170508 09:05:57< wedge009> Bonne nuit.
20170508 09:05:58-!- celticminstrel is now known as celmin|Zzzzzzz
20170508 09:06:12< celmin|Zzzzzzz> G'night
20170508 09:08:05-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170508 09:10:22-!- JyrkiVesterinen [~JyrkiVest@85.76.66.132] has quit [Quit: .]
20170508 09:38:01< vultraz_iOS> zookeeper: 5 years is about the average life of a laptop, I think
20170508 09:38:20< zookeeper> ?
20170508 09:38:33< vultraz_iOS> one would be replacing a laptop ~every 5 years
20170508 09:38:40-!- JyrkiVesterinen [~JyrkiVest@85-76-66-132-nat.elisa-mobile.fi] has joined #wesnoth-dev
20170508 09:39:40< zookeeper> yeah, but why are you telling me that?
20170508 09:42:23< vultraz_iOS> since you mentioned 5 years
20170508 09:42:58< vultraz_iOS> so it popped into my head that that's about the life of a laptop
20170508 09:43:26< zookeeper> umm.... sure, i did say the words "5 years" in a different context, so you're right
20170508 09:56:36-!- sevu [~Unknown@141.57.57.139] has joined #wesnoth-dev
20170508 09:58:03-!- sevu [~Unknown@141.57.57.139] has quit [Remote host closed the connection]
20170508 09:59:13-!- sevu [~Unknown@141.57.57.139] has joined #wesnoth-dev
20170508 10:06:23-!- irker171 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20170508 10:06:43-!- atarocch [~atarocch@93.68.209.42] has joined #wesnoth-dev
20170508 10:08:06-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170508 10:08:38-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170508 10:11:21-!- matthiaskrgr_ is now known as matthiaskrgr
20170508 10:31:41-!- atarocch [~atarocch@93.68.209.42] has quit [Remote host closed the connection]
20170508 10:42:11-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20170508 10:42:31-!- sevu [~Unknown@141.57.57.139] has quit [Remote host closed the connection]
20170508 11:18:02-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170508 11:31:29-!- gfgtdf [~chatzilla@x4e3639cc.dyn.telefonica.de] has joined #wesnoth-dev
20170508 11:37:36-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20170508 11:54:13-!- Kwandulin [~Kwandulin@p200300760F6D80832444895970AEFE5B.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170508 12:24:07-!- Kwandulin [~Kwandulin@p200300760F6D80832444895970AEFE5B.dip0.t-ipconnect.de] has quit [Ping timeout: 272 seconds]
20170508 12:26:08-!- Kwandulin [~Kwandulin@p200300760F6D800C2444895970AEFE5B.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170508 12:45:57-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170508 13:01:46< gfgtdf> celmin|Zzzzzzz: the bugtracker cripts seems to convert the 'release' fieldwhich specifies when the bugs was found to the 'milestone' whihc specifies to which relese we aim to have this bug fixed which is something  different and not neccesarily related..
20170508 13:15:42< esr> Can somebodty explain to me why the entire bug-modification history of Wesnoth is being vomited into my inbox? We're taking thousands up on thousands of messages here.
20170508 13:16:09< vultraz_iOS> heh
20170508 13:16:17< vultraz_iOS> esr: we're moving all bugs to github
20170508 13:16:20< vultraz_iOS> GNA Is closing
20170508 13:16:20< JyrkiVesterinen> esr: The bugs are being moved from Gna to GitHub.
20170508 13:16:22< vultraz_iOS> it must be done
20170508 13:16:32< JyrkiVesterinen> Not sure why you're getting notifications, though.
20170508 13:16:55< JyrkiVesterinen> Maybe people have written @esr in bug reports and that triggers GitHub notifications, or something.
20170508 13:17:24< esr> I think I'm getting the *entire* history.
20170508 13:17:29< vultraz_iOS> JyrkiVesterinen: if you're subbed to the repo and haven't disabled the appropriate settings you will be emailed entirely
20170508 13:17:34< vultraz_iOS> for every change
20170508 13:17:41< JyrkiVesterinen> Okay.
20170508 13:17:43< vultraz_iOS> esr: you're only getting the 600-some open bugs :P
20170508 13:17:58< esr> Lovely.
20170508 13:18:06< vultraz_iOS> feel free to delete them all
20170508 13:18:14< JyrkiVesterinen> Myself, I'm not subscribed to the Wesnoth repo. I don't even get GitHub notifications for the bug reports, let alone emails.
20170508 13:19:24< wedge009> Aye, I haven't been getting all the notifications, only the ones which have been assigned to me.
20170508 13:19:35< vultraz_iOS> yeah, same
20170508 13:19:39< vultraz_iOS> since i disabled emails
20170508 13:19:49< vultraz_iOS> else i'd be inundated with 1000+ emails
20170508 13:19:59< vultraz_iOS> since GH sends 1 email per comment :|
20170508 13:20:41< vultraz_iOS> esr: simple solution is to disable email notifs on your GH account.
20170508 13:21:10< JyrkiVesterinen> Or subscription for the repository.
20170508 13:21:28< JyrkiVesterinen> All future bug reports will go to GitHub too and send notifications to subscribers.
20170508 13:21:48< vultraz_iOS> indeed
20170508 13:41:27-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20170508 13:57:02-!- Kwandulin [~Kwandulin@p200300760F6D800C2444895970AEFE5B.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20170508 13:57:15< vultraz_iOS> looks like we're gone
20170508 13:57:17< vultraz_iOS> ...
20170508 13:57:18< vultraz_iOS> done
20170508 14:00:11-!- gfgtdf [~chatzilla@x4e3639cc.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.93 [Firefox 53.0.2/20170504105526]]
20170508 14:20:14-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20170508 14:22:55-!- atarocch [~atarocch@109.112.49.145] has joined #wesnoth-dev
20170508 14:30:46-!- JyrkiVesterinen [~JyrkiVest@85-76-66-132-nat.elisa-mobile.fi] has quit [Quit: .]
20170508 15:15:39-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Remote host closed the connection]
20170508 15:15:53-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20170508 15:26:12-!- atarocch [~atarocch@109.112.49.145] has quit [Remote host closed the connection]
20170508 15:43:11-!- ChipmunkV[m] [chipmunkvm@gateway/shell/matrix.org/x-pieqnhxwktpyiott] has quit [Remote host closed the connection]
20170508 15:43:21-!- Greg_Boggs[m] [gregboggsm@gateway/shell/matrix.org/x-zibhwanupqcasdhx] has quit [Remote host closed the connection]
20170508 15:47:35-!- Greg_Boggs[m] [gregboggsm@gateway/shell/matrix.org/x-szviqukgarnryanv] has joined #wesnoth-dev
20170508 15:56:51-!- ChipmunkV[m] [chipmunkvm@gateway/shell/matrix.org/x-vcgjewayrkirdkni] has joined #wesnoth-dev
20170508 16:00:21-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170508 16:02:04-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170508 16:02:15-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170508 16:02:20-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170508 16:16:20-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170508 16:23:53< celmin|Zzzzzzz> Alright, next step is to close some bugs, right? :P
20170508 16:24:13< celmin|Zzzzzzz> vultraz_iOS: We can close the ones marked fixed for starters?
20170508 16:25:45-!- Kwandulin [~Kwandulin@p200300760F6D800C2444895970AEFE5B.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170508 16:29:49< celmin|Zzzzzzz> Oh yeah, before that, we should unassign the OS labels from issues where they don't really apply.
20170508 16:29:58-!- celmin|Zzzzzzz is now known as celticminstrel
20170508 16:44:46< celticminstrel> ...and GitHub isn't working for some reason, so I can't.
20170508 16:47:52-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has quit [Quit: Connection closed for inactivity]
20170508 16:48:29< celticminstrel> Seems like one of GitHub's scripts isn't loading.
20170508 16:50:12-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170508 16:57:16-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170508 16:57:52-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170508 16:58:21-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170508 17:13:44-!- JyrkiVesterinen [~jyrki@89.166.112.253] has joined #wesnoth-dev
20170508 17:25:42< JyrkiVesterinen> celticminstrel: I noticed that the login_save_crypt branch doesn't compile with OpenSSL 1.1.0e.
20170508 17:25:56< JyrkiVesterinen> /home/jyrki/wesnoth/src/hash.cpp:102:18: error: ‘SHA_Init’ was not declared in this scope
20170508 17:25:58< JyrkiVesterinen>   SHA_Init(&hasher);
20170508 17:26:16< JyrkiVesterinen> The functions have apparently been renamed when SHA2 support was added.
20170508 17:26:24< JyrkiVesterinen> I'll try to find a fix.
20170508 17:26:42< celticminstrel> When was that version released?
20170508 17:26:51-!- atarocch [~atarocch@31.157.191.162] has joined #wesnoth-dev
20170508 17:26:57< celticminstrel> Maybe I should instead update my OpenSSL and switch to the new names.
20170508 17:27:13< JyrkiVesterinen> 1.1.0e was released on February 16.
20170508 17:27:22< celticminstrel> Ah, so quite recently.
20170508 17:27:30< JyrkiVesterinen> Original 1.1.0 was released some months earlier, I think.
20170508 17:27:40< celticminstrel> I see.
20170508 17:27:51< JyrkiVesterinen> I think we can't require OpenSSL 1.1, it's too new.
20170508 17:27:58< celticminstrel> Okay.
20170508 17:29:24< JyrkiVesterinen> Found the news page. 1.1.0 was released on August 25, 2016. Clearly too new for us to require it.
20170508 17:30:05< celticminstrel> Does OpenSSL not have version macros? I can't see any.
20170508 17:35:19< JyrkiVesterinen> I downloaded OpenSSL 1.0.2k source code. I see that it has SHA1_Init() and other functions (with names beginning with SHA1) which remain in 1.1.0.
20170508 17:35:42< JyrkiVesterinen> So, I'll just switch to the SHA1_xxx() functions.
20170508 17:36:28< celticminstrel> Ah, so it does.
20170508 17:42:45-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20170508 17:43:07-!- mjs-de [~mjs-de@wh.Uni-Dortmund.DE] has joined #wesnoth-dev
20170508 17:51:53< celticminstrel> ...
20170508 17:52:49-!- irker254 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20170508 17:52:49< irker254> wesnoth: Celtic Minstrel wesnoth:master e945755793bc / data/ai/formula/engine.lua: Accidentally committed this https://github.com/wesnoth/wesnoth/commit/e945755793bc09efa2a3ddb4ec021acf94688a30
20170508 17:53:13< celticminstrel> Quite awhile ago, too.
20170508 17:53:18< celticminstrel> Wow.
20170508 17:53:47< celticminstrel> GitHub still broken for me.
20170508 17:55:22< JyrkiVesterinen> It has worked fine for me all evening.
20170508 17:55:29< celticminstrel> Working on Windows though. What's wrong...
20170508 17:58:13< celticminstrel> It seems like the page's Content Security Protocol is blocking one of its scripts for some reason.
20170508 18:03:47-!- markus_ [~mjs-de@wh.Uni-Dortmund.DE] has joined #wesnoth-dev
20170508 18:04:13-!- mjs-de [~mjs-de@wh.Uni-Dortmund.DE] has quit [Ping timeout: 272 seconds]
20170508 18:11:05-!- chiec [~chiec@2a01:c50e:3509:1f00:1f05:155b:5053:7345] has joined #wesnoth-dev
20170508 18:34:30-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has joined #wesnoth-dev
20170508 18:37:37< irker254> wesnoth: Jyrki Vesterinen wesnoth:login_save_crypt 55c37c0313af / src/hash.cpp: Use SHA1_Init() and co. for SHA-1 calculations https://github.com/wesnoth/wesnoth/commit/55c37c0313afffd29be36cefaf39fd4fb596e17a
20170508 18:37:39< irker254> wesnoth: Jyrki Vesterinen wesnoth:login_save_crypt d481380a1989 / CMakeLists.txt: Fix misleading indentation https://github.com/wesnoth/wesnoth/commit/d481380a19890e174a1abdb5c2b1d3a8d87365aa
20170508 18:37:55< JyrkiVesterinen> I pushed the improvements I have made until now.
20170508 18:38:15-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170508 18:38:17< JyrkiVesterinen> Fixing the CMake build is harder than I thought.
20170508 18:39:17< JyrkiVesterinen> CMake does run the check for libcrypto. Also, running `pkg-config --libs libcrypto` returns "-lcrypto", so CMake should know that the game needs to be linked with -lcrypto.
20170508 18:39:28< JyrkiVesterinen> Yet, -lcrypto is not passed when linking.
20170508 18:39:41< JyrkiVesterinen> I think I need to grab strace or something...
20170508 18:47:46-!- Appveyor [~Appveyor@74.205.54.20] has joined #wesnoth-dev
20170508 18:47:46< Appveyor> The Battle for Wesnoth (Visual Studio 2015) - Release Charles Dang 9c85689: Use a class template for surface_lock and const_surface_lock Failed
20170508 18:47:46< Appveyor> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-330
20170508 18:47:50-!- Appveyor [~Appveyor@74.205.54.20] has left #wesnoth-dev []
20170508 18:53:31-!- Appveyor [~Appveyor@74.205.54.20] has joined #wesnoth-dev
20170508 18:53:31< Appveyor> The Battle for Wesnoth (Visual Studio 2015) - Debug Charles Dang 9c85689: Use a class template for surface_lock and const_surface_lock Failed
20170508 18:53:31< Appveyor> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-330
20170508 18:53:36-!- Appveyor [~Appveyor@74.205.54.20] has left #wesnoth-dev []
20170508 18:57:27< JyrkiVesterinen> Strace confirms that CMake queries the correct linking flags (with `pkg-config --libs-only-l libcrypto`). And running that command myself returns -lcrypto.
20170508 18:57:39< JyrkiVesterinen> Do I need to debug CMake myself or what...
20170508 19:01:49< loonycyborg> JyrkiVesterinen: if it runs that command it doesn't mean flags are used
20170508 19:02:04< loonycyborg> result of pkg-config is stored in a variable
20170508 19:02:25< loonycyborg> you should check if it actually added to linkflags of the program it's needed in
20170508 19:03:20< JyrkiVesterinen> Does our CMakeLists.txt have any examples about adding the necessary linkflags?
20170508 19:06:19< loonycyborg> it should, but I don't know details
20170508 19:15:52< JyrkiVesterinen> Aargh. We indeed have code that explicitly adds the necessary linkflags... in the *other* CMakeLists.txt file.
20170508 19:15:54< JyrkiVesterinen> https://github.com/wesnoth/wesnoth/blob/e945755793bc09efa2a3ddb4ec021acf94688a30/src/CMakeLists.txt#L59-L64
20170508 19:16:07< JyrkiVesterinen> I'll try adding libcrypto there.
20170508 19:20:16-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170508 19:20:48-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170508 19:29:31< irker254> wesnoth: Jyrki Vesterinen wesnoth:login_save_crypt fee68eca3087 / src/CMakeLists.txt: Fix CMake build https://github.com/wesnoth/wesnoth/commit/fee68eca3087854719095ce7e6613b763b594c33
20170508 19:29:41< JyrkiVesterinen> That did it. :)
20170508 19:38:02< celticminstrel> \o/
20170508 19:38:04< celticminstrel> Thanks.
20170508 19:43:24-!- JyrkiVesterinen [~jyrki@89.166.112.253] has quit [Quit: Going to bed]
20170508 19:43:59-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170508 19:44:11-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170508 19:48:48-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Quit: KABOOM! It seems that I have exploded. Please wait while I reinstall the universe.]
20170508 19:49:08-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20170508 19:50:25< zookeeper> eh, why does dacyn have that fake_attack animation for killing mal-ravanal, instead of his normal attack anim getting used...
20170508 19:52:26< DeFender1031> Is it different in any way?
20170508 19:52:37< zookeeper> the main difference seems to be that it doesn't have the lightbeam, but what's the point of that when the lightbeam is simply included in mal-ravanal's death anim instead
20170508 19:52:39< DeFender1031> More dramatic?
20170508 19:52:46< DeFender1031> Hahahaha
20170508 19:53:25< DeFender1031> Yeah, that's... not functionally different.
20170508 19:53:53< DeFender1031> See, if *I* were doing it, that particular death animation would have a little more flare to it.
20170508 19:54:03< DeFender1031> (And maybe I WILL do it one of these days...)
20170508 19:54:19< zookeeper> well, that's what i'm doing right now :p
20170508 19:54:36< DeFender1031> Adding flare? I had assumed that you'd just remove it in favor of the existing one.
20170508 19:54:43< zookeeper> longer fadeout with some purple explosions and stuff, nothing _too_ fancy
20170508 19:55:28< DeFender1031> TBH, the ending story sequence gives a pretty elaborate description of what ravanal's death looks like.... it shouldn't be that hard to translate that into animation code.
20170508 19:56:27< zookeeper> ...shouldn't be too hard to make a giant cloud of smoke in the shape of a hand that then dissolves in the wind? err... sure, go ahead, good luck
20170508 19:57:17< zookeeper> it also doesn't specify how the column of smoke gets from the duel chamber out in the open :p
20170508 19:57:21< zookeeper> maybe someone left the window open?
20170508 19:58:10< DeFender1031> "As the lich’s death-scream echoed in the air, a vast and louring column of shadow arose from his corpse like smoke from an unclean fire; then it divided fivefold at its height, reaching for the heavens like a clutching hand."
20170508 19:58:19< DeFender1031> yeah, shouldn't be too hard to do that.
20170508 19:59:02< DeFender1031> I'd use a similar technique to the black cloud that appears for about half a second in my opening cutscene before suddenly spreading out and casting darkness over the area.
20170508 19:59:23< DeFender1031> (meaning, multiple of the same smoke haloes spreading out in the shape I want.)
20170508 19:59:27< zookeeper> ...what black cloud?
20170508 19:59:33< zookeeper> oh
20170508 19:59:36< zookeeper> "my", not "any" :P
20170508 19:59:43< DeFender1031> Yeah... I expected that response.
20170508 19:59:57< DeFender1031> ah, there's the confusion.
20170508 20:00:18< DeFender1031> I had just expected that no one would remember the cloud, since, as I said, it's only there for half a second.
20170508 20:00:35< zookeeper> well, i suppose a rising column of smoke would be doable
20170508 20:01:48< DeFender1031> I'd do the rising column which is a composite of 10-20 haloes and then have them reposition themselves into a hand shape on the way up. Probably time-consuming to get right, but there are no significant technical challenges preventing it.
20170508 20:02:05< zookeeper> but then again it might give the impression that it's actually something important and hinting at him not being completely gone, instead of just being for dramatic effect
20170508 20:02:31< DeFender1031> Okay, so there's sequel fodder. THat's a... bad thing?
20170508 20:02:36< zookeeper> i wouldn't try to make it hand-shaped though, just an unusually large column
20170508 20:02:51< zookeeper> well, not bad i guess
20170508 20:02:56< zookeeper> i
20170508 20:02:58< DeFender1031> Also, the text description of the same right afterward would mitigate that.
20170508 20:03:06< zookeeper> i'll see what my attempts end up looking like
20170508 20:03:22< DeFender1031> [5/8/2017 11:09:02 pm] <zookeeper> i wouldn't try to make it hand-shaped though, just an unusually large column
20170508 20:03:34< DeFender1031> do you mean that you personally wouldn't, or that you don't think it's a good idea?
20170508 20:03:45< zookeeper> i'm personally not going to try that
20170508 20:03:48< DeFender1031> ah
20170508 20:03:59< zookeeper> at least at the moment i don't think i will :p
20170508 20:04:22-!- markus_ [~mjs-de@wh.Uni-Dortmund.DE] has quit [Remote host closed the connection]
20170508 20:04:31< DeFender1031> i might give it a shot one of these days if you don't mind me possibly redoing whatever it is you're doing now.
20170508 20:04:36< zookeeper> i feel like if i did something _too_ fancy like that, it'd set a precedent where we have to start filling mainline with ridiculously complicated fanciness :>
20170508 20:05:21< DeFender1031> And I ask again, what's wrong with that?
20170508 20:05:39< DeFender1031> Cinematics are something that's been sorely lacking.
20170508 20:05:42< zookeeper> sure, feel free to. i do have my doubts though WRT whether it can actually look like a sufficiently impressive hand instead of a clumsy attempt at a hand
20170508 20:06:01< DeFender1031> right, there's always that.
20170508 20:07:35< DeFender1031> On the other hand, unlike with most art forms where the images in my head and the images that I end up creating look nothing alike, so far every attempt I've made to do fancy stuff in wesnoth has come out looking MORE impressive than I imagined.
20170508 20:07:56< DeFender1031> (At least, in my hardly humble opinion)
20170508 20:08:00< zookeeper> eh, i just don't feel like the general style of the game meshes very well with cinematic cutscenes. to me, especially cutscenes where units are moving around the map end up looking a bit... artificial. of course in this case it's just a death animation.
20170508 20:08:10< celticminstrel> BTW zookeeper, Vultraz committed a cutscene theme; should we start using it in mainline campaigns for the cutscene scenarios?
20170508 20:08:22< zookeeper> celticminstrel, sure
20170508 20:08:36< zookeeper> i haven't checked what it looks like though, have a screenshot?
20170508 20:08:52< DeFender1031> zookeeper, re "especially when units are moving around the map": I expect that a large part of that is due to the fact that the engine can only move one unit at a time.
20170508 20:08:53< celticminstrel> I haven't checked either, actually.
20170508 20:09:00< DeFender1031> I have
20170508 20:09:25< DeFender1031> It is literally the battle view and no menubar or sidebar
20170508 20:09:34< DeFender1031> literally just the entire screen is battle view.
20170508 20:10:04< celticminstrel> Not having a menubar seems a bit...
20170508 20:10:16< zookeeper> hmh
20170508 20:10:23< DeFender1031> I could get a screenshot if you really want, but such a screenshot would be no different functionally than cropping out the top half inch and the right three inches.
20170508 20:10:29< celticminstrel> Heh.
20170508 20:10:40< celticminstrel> I suppose the most important menu function is accessible with Esc...
20170508 20:10:41< zookeeper> i'm not sure about not having the sidebar. you don't get any info on the units who are doing the talking.
20170508 20:11:12< celticminstrel> Well, you do see their name and portrait.
20170508 20:11:14< DeFender1031> presumably, you already know who they are.
20170508 20:11:17< celticminstrel> Since they're talking.
20170508 20:11:23< zookeeper> i'm not sure if that's important or not, but it's something to consider
20170508 20:11:29< DeFender1031> yeah, during a cutscene, the exact stats are not really so relevant
20170508 20:11:32< zookeeper> i mean their unit types and stats and sprites and all that
20170508 20:11:49< zookeeper> err, well, you see the sprite on the map of course...
20170508 20:12:13< DeFender1031> oh, there's a second cutscene theme in here which i haven't gotten around to checking out, because it uses some images that I haven't properly transferred yet. The one I'm talking about is cutscene-minimal
20170508 20:12:39< DeFender1031> the sprite is on the map, the unit type should be recognizable from the sprite, the name is in the message.
20170508 20:14:05< DeFender1031> What would be really nice would be the ability to fade from one theme to another. It'd be really cool to open with a cutscene, then right before relinquishing control to the player, the interface fades in nicely on top of the rest. I doubt the engine is ever going to be capable of that though.
20170508 20:15:24< zookeeper> yeah. changing maybe, fading less so.
20170508 20:15:32< DeFender1031> right.
20170508 20:15:55< DeFender1031> i can dream though. ;)
20170508 20:17:58-!- travis-ci [~travis-ci@54.224.250.25] has joined #wesnoth-dev
20170508 20:17:59< travis-ci> wesnoth/wesnoth#13801 (login_save_crypt - d481380 : Jyrki Vesterinen): The build failed.
20170508 20:18:00< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/230070519
20170508 20:18:00-!- travis-ci [~travis-ci@54.224.250.25] has left #wesnoth-dev []
20170508 20:23:10< matthiaskrgr> hm
20170508 20:23:16< matthiaskrgr> on the github issue tracker
20170508 20:23:30< matthiaskrgr> there are open issues which have a "fixed" tag?
20170508 20:24:23< celticminstrel> Those are issues that were marked "fixed" on GNA.
20170508 20:24:43< celticminstrel> We're going to be doing something with them, most likely closing them, but not entirely sure yet.
20170508 20:25:12< celticminstrel> See https://github.com/wesnoth/wesnoth/issues/1031
20170508 20:27:32< vultraz_iOS> close them
20170508 20:28:16< zookeeper> DeFender1031, ah, one possible reason for that weird lightbeam thing would be if [animate_unit] didn't at the time support the [animate] subtag... because without it, you can't simultaneously animate dacyn's attack _and_ specify the target/facing for it
20170508 20:28:54< zookeeper> err, wait i'm not making any sense now
20170508 20:29:10< zookeeper> the only reason would be if [animate_unit] didn't support [facing] back then
20170508 20:29:16< DeFender1031> Btw, (and since I apparently won't be able to make this joke much longer), am I the only one who thinks that GNA is appropriately named because it's probably the noise that its developers make any time you raise the issue of how terrible the interface is?
20170508 20:29:27< DeFender1031> zookeeper, yeah, was going to ask about that.
20170508 20:29:38< zookeeper> wait, i'm not making _any_ sense. the very code that i was looking at already uses an [animate] tag -.-
20170508 20:29:51< zookeeper> forget all i said while i go and jump down a well or something
20170508 20:29:53< DeFender1031> zookeeper, but if it doesn't support facing, then the whole thing would be wrong.
20170508 20:30:13< DeFender1031> zookeeper, it's okay, i've done the same thing.
20170508 20:30:41< zookeeper> i should have looked at what the code actually is like before trying to come up with explanations for why it's as it is :>
20170508 20:30:53< DeFender1031> zookeeper, actually, i've done worse. I've rambled on about what half-understood code seems to be doing without making sense WHEN TALKING ABOUT MY OWN CODE THAT NO ONE ASIDE FROM ME HAD TOUCHED.
20170508 20:36:24< zookeeper> ah, too bad one can't (smoothly and without getting excessively verbose) do progressive IPF parameters... it'd be good to be able to scale up the smoke clouds as they rise.
20170508 20:37:11< zookeeper> or at least i think so
20170508 20:37:49< celticminstrel> Why can't you?
20170508 20:38:45< zookeeper> it just doesn't work
20170508 20:38:54< zookeeper> 20170508 23:36:00 error general: Square bracket lengths do not match up: terrain/misc/smoke-A[01~12].png~SCALE([72~300],144):200
20170508 20:39:17< zookeeper> but also i can see how it wouldn't necessarily make sense... although i find it hard to explain why
20170508 20:39:23< celticminstrel> Ah, I see. What it needs is a notation to specify the step size.
20170508 20:39:48< DeFender1031> zookeeper, you can do progressive IPF, I do it all the time.
20170508 20:40:04< DeFender1031> oh, that's what you mean.
20170508 20:40:07< celticminstrel> [72~300] expands to [72,73,74,...298,299,300]
20170508 20:40:20< celticminstrel> So it has many more elements than [01-12].
20170508 20:40:24< DeFender1031> yeah, you might have to split that into multiple lines to do it.
20170508 20:40:37< zookeeper> right
20170508 20:40:44< zookeeper> so if i split it into 12 steps...
20170508 20:40:45< celticminstrel> You could make it work by picking 12 specific sizes.
20170508 20:40:55< DeFender1031> though, I think the nested bracket syntax I was talking about adding might be able to do that
20170508 20:41:08< celticminstrel> I don't know if that would be smooth though.
20170508 20:41:11-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170508 20:41:19< DeFender1031> or at least get you closer.
20170508 20:41:43-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170508 20:41:44< DeFender1031> celticminstrel, yeah, that would not be smooth.
20170508 20:42:00< DeFender1031> anyway, I should really get off my butt and work on that bracket stuff.
20170508 20:44:12< zookeeper> ah, indeed it works. silly me.
20170508 20:44:49< DeFender1031> i wonder if there's a reason it shouldn't just do a "best fit" for cases where different brackets result in different amounts... like, for example, 300-72 is 228, which is 12*19, so if the expansion would do each of the 1-12 19 times, it would work.
20170508 20:45:12< DeFender1031> zookeeper, how'd you make it work?
20170508 20:45:30< zookeeper> with 12 individual steps
20170508 20:45:37< zookeeper> image="terrain/misc/smoke-A[01~12].png~SCALE([72,90,98,122,140,170,190,240,250,260,370,390],144):200"
20170508 20:47:56< celticminstrel> I was thinking that too.
20170508 20:48:26< DeFender1031> Does that not look jerky?
20170508 20:48:32< celticminstrel> Oh, actually I was thinking of the exact opposite, but similar outcome.
20170508 20:49:44< DeFender1031> I was thinking you could do "terrain/misc/smoke-A[01*19,02*19,03*19,04*19,05*19,06*19,07*19,08*19,09*19,10*19,11*19,12*19].png~SCALE([73~300],144):200"
20170508 20:50:30< zookeeper> DeFender1031, of course it looks jerky with those completely arbitrary test values :P
20170508 20:50:50< DeFender1031> or, if I ever get off my butt and do the stuff I'm planning, "terrain/misc/smoke-A[[01~12]*19].png~SCALE([73~300],144):200"
20170508 20:51:33< celticminstrel> DeFender1031: It's not clear to me whether [[1-5]*3] would expand to [1,1,1,2,2,2,3,3,3,4,4,4,5,5,5] or [1,2,3,4,5,1,2,3,4,5,1,2,3,4,5].
20170508 20:52:19< DeFender1031> celticminstrel, the latter would be [1-5*3]
20170508 20:52:35< celticminstrel> You mean [1~5*3]?
20170508 20:52:50< celticminstrel> Oh you used - instead of ~ the first time too.
20170508 20:52:53< DeFender1031> yes, sorry, you used a hyphen.
20170508 20:52:56< celticminstrel> Wait no, I did.
20170508 20:53:01< celticminstrel> Oops.
20170508 20:53:03< DeFender1031> hehehe
20170508 20:53:09< DeFender1031> i just copied you.
20170508 20:53:29< DeFender1031> but yeah, the point is that each bracket sequence would resolve in place, from inner to outer.
20170508 20:54:03< celticminstrel> And [A,B,C]*3 is [A,A,A,B,B,B,C,C,C]?
20170508 20:54:15< DeFender1031> I really should do that one of these days, I've just become very bad at organizing my time lately.
20170508 20:54:57< DeFender1031> that'd not end up parsing the *3.
20170508 20:55:03< DeFender1031> since it's not in a bracket.
20170508 20:56:27-!- Kwandulin [~Kwandulin@p200300760F6D800C2444895970AEFE5B.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20170508 20:57:12< celticminstrel> I meant when that appears in a bracketed expression. >_>
20170508 20:58:15< DeFender1031> then yes.
20170508 20:58:36< DeFender1031> and [A,B,C*3] would be ABCABCABC
20170508 20:59:00< celticminstrel> Isn't [A,B,C*3] currently [A,B,C,C,C] though?
20170508 20:59:01< DeFender1031> (insert annoying-to-type commas as appropriate)
20170508 20:59:22< DeFender1031> oh crap.
20170508 20:59:24< DeFender1031> you're right.
20170508 20:59:31< DeFender1031> I need to rethink this part of it then.
20170508 20:59:48< celticminstrel> Expanding [A~B*3] ti ABCABCABC might work, at least.
20170508 20:59:51< celticminstrel> ^to
20170508 21:00:11< DeFender1031> right, though that doesn't handle all the cases I want to handle
20170508 21:00:37< celticminstrel> Did your ideas include a way of specifying a step size for the range operator?
20170508 21:01:27< DeFender1031> basically, we need a way to expand the existing syntax, in both the "," and "~" forms, to allow for both "repeat each step of this progression a certain number of times before moving on to the next one" and "repeat this entire progression a certain number of times"
20170508 21:01:40< DeFender1031> ah, I think I did plan for that.
20170508 21:01:57< DeFender1031> ":" I think?
20170508 21:02:04< DeFender1031> hmm...
20170508 21:04:12< DeFender1031> I need to give this more thought. Ideally, there's be a way to do ABCABCABC, AAABBBCCC, QLDQLDQLD, QQQLLLDDD, 123123123, and 111222333, all without the need to include each of those characters more than once for any of them.
20170508 21:04:43< celticminstrel> ":" is currently used to indicate the duration of an animation, but I think that might be separate from the bracketed progressive syntax? So there may be no conflict in reusing it.
20170508 21:05:06< DeFender1031> And then, like you said, we also want a way to do 246246246 and 222444666 without needing to use 2,4,6 explicitly either.
20170508 21:05:21< DeFender1031> celticminstrel, yeah, that's outside the brackets. no interference there.
20170508 21:05:42< celticminstrel> Hmm, I suppose if you wanted palindromic repetition you could do something like [[A~G,F~B]*7,A]...
20170508 21:06:06< DeFender1031> and there are some animations that do that.
20170508 21:06:18< DeFender1031> pulsing spheres of light come to mind as one example.
20170508 21:06:28< celticminstrel> TBH, I think [A~G]*7 looks like "repeat the entire progression". Maybe it's because I'm used to how Pythin uses * as a repetition operator.
20170508 21:07:15< celticminstrel> I mentioned palindromic repetition because I was thinking "you could add a syntax for that too" and then realized "wait, it's not really needed".
20170508 21:07:28< celticminstrel> ^Python
20170508 21:08:14< DeFender1031> ah
20170508 21:08:45< DeFender1031> Ah, yes, Pythin, Python's weightloss module.
20170508 21:08:49< celticminstrel> XD
20170508 21:09:35-!- atarocch [~atarocch@31.157.191.162] has quit [Ping timeout: 240 seconds]
20170508 21:09:59< DeFender1031> truth is, there's no reason [A~G]*7 can't mean repeat the entire progression.
20170508 21:10:09< DeFender1031> hmm
20170508 21:10:14< DeFender1031> what about
20170508 21:10:16< vultraz_iOS> are you finally workingon the string code
20170508 21:10:44< DeFender1031> vultraz_iOS, just bouncing ideas about new syntax options, but it might be enough to get me off my butt to do it :P
20170508 21:10:56< DeFender1031> anyway
20170508 21:10:58< DeFender1031> what about
20170508 21:11:46< DeFender1031> [[A~C]*3] ends up being ABCABCABC and [[A~C]*~3] ends up being AAABBBCCC
20170508 21:12:19< DeFender1031> i'm not entirely sure how much that would complexify the parsing though. It should be doable.
20170508 21:12:48< DeFender1031> but basically "*~" meaning "multiply and intersperse" whereas "*" just means "multiply:
20170508 21:12:50< DeFender1031> "*
20170508 21:13:34< DeFender1031> and that would work just as well with [[t,j,5]*3] and [[t,j,5]*~3]
20170508 21:13:50< DeFender1031> (the former, naturally, being tj5tj5tj5 and the latter being tttjjj555)
20170508 21:14:11< celticminstrel> Well, no major objections.
20170508 21:14:58< DeFender1031> [1~10:2] would be 1,3,5,7,9 and nesting that with a * or *~ would work as expected too.
20170508 21:17:54< DeFender1031> and I'm also thinking of doing an implicit "best fit" spread if amounts don't match up.
20170508 21:19:58-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20170508 21:19:59< DeFender1031> so for the example from above, "terrain/misc/smoke-A[01~12].png~SCALE([72~300],144):200" would be equivalent to something like "terrain/misc/smoke-A[01*20,02*19,03*19,04*19,05*19,06*19,07*19,08*19,09*19,10*19,11*19,12*19].png~SCALE([73~300],144):200"
20170508 21:20:05< DeFender1031> rather than just erroring out.
20170508 21:20:09< celticminstrel> Is the square parenthetical split used anywhere outside of animations?
20170508 21:20:18< DeFender1031> good question
20170508 21:20:48< celticminstrel> IIRC the splut functions take option flags, so if it is used elsewhere, it might be better for the best fit part to be optional.
20170508 21:20:54< celticminstrel> ^split
20170508 21:20:57< DeFender1031> I know it's used in [item] haloes, possibly terrains as well, though those also qualify as animations in this context.
20170508 21:21:09< DeFender1031> that's fair.
20170508 21:21:44< DeFender1031> certainly for animations though, when you give nonmatching sets, a best fit would be the sort of thing you'd want.
20170508 21:22:01< DeFender1031> ooh
20170508 21:22:05< DeFender1031> one problem with that
20170508 21:22:06-!- atarocch [~atarocch@2.43.222.171] has joined #wesnoth-dev
20170508 21:22:12< DeFender1031> the animation durations can be bracketed too
20170508 21:22:31< celticminstrel> They can, yes.
20170508 21:22:44< DeFender1031> you wouldn't want to have a mistake in expansion size leading to an animation you thought should last one duration lasting another.
20170508 21:22:59< DeFender1031> on the other hand, do we really care about that case?
20170508 21:23:26< celticminstrel> I'm not sure?
20170508 21:23:42< celticminstrel> It certainly wouldn't break any existing animations.
20170508 21:24:05< celticminstrel> But I think you have a point that you probably don't want the duration to change to do changing something prior to the colon.
20170508 21:25:10< DeFender1031> then again, it ALREADY changes based on what's before the colon.
20170508 21:25:23< celticminstrel> No?
20170508 21:25:28< celticminstrel> Does it?
20170508 21:25:30< DeFender1031> "blah[1~10]:200" will last 2000 ms, not 200.
20170508 21:25:46< celticminstrel> Oh.
20170508 21:26:00< celticminstrel> IOW :200 is equivalent to :[200*3]
20170508 21:26:11< DeFender1031> if you want to assign a specific duration for the entire progression, you use the duration= property instead.
20170508 21:26:20< DeFender1031> huh?
20170508 21:26:33< celticminstrel> It's 200 per frame, not total.
20170508 21:26:37< DeFender1031> right right.
20170508 21:28:04< DeFender1031> Which means that the duration is already implicitly affected by stuff before the colon, at least in a lot of cases.
20170508 21:28:16< DeFender1031> so I think I lean toward not caring that we're adding a few more.
20170508 21:28:30< DeFender1031> vultraz_iOS, zookeeper, want to weigh in?
20170508 21:28:46< vultraz_iOS> im busy watching the Sally Yates testimony sorry
20170508 21:28:48< celticminstrel> Well, but if you wrote :[200*3[ instead of :200, you'd get an error if the other expansions had a different length, right?
20170508 21:29:01< zookeeper> if you have a specific question that i don't need to read backlog to understand, then sure :p
20170508 21:31:03< DeFender1031> celticminstrel, currently, yes.
20170508 21:31:50< DeFender1031> zookeeper, using a "best fit" for strings that contain multiple expansions with different amounts, like the example you had above.
20170508 21:32:17< DeFender1031> zookeeper, where "terrain/misc/smoke-A[01~12].png~SCALE([72~300],144):200" would be equivalent to something like "terrain/misc/smoke-A[01*20,02*19,03*19,04*19,05*19,06*19,07*19,08*19,09*19,10*19,11*19,12*19].png~SCALE([73~300],144):200"
20170508 21:33:09< DeFender1031> the only drawback we seem to have come up with is that the animation durations could be affected, except that's true already if the duration isn't part of an expansion.
20170508 21:33:40< DeFender1031> celticminstrel, in that case though, they'd all end up with 200 anyway. I'm more concerned with cases where there are varying durations in use.
20170508 21:34:36< celticminstrel> I wonder if it would be acceptable to take a duration= key as a hard duration, so that if the progressive expansion results in something longer, it's scaled down to fit...
20170508 21:34:45-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170508 21:34:51< DeFender1031> celticminstrel, that's already there.
20170508 21:34:54< celticminstrel> Oh?
20170508 21:34:58< celticminstrel> Okay then.
20170508 21:35:17-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170508 21:35:22< DeFender1031> celticminstrel, https://wiki.wesnoth.org/AnimationWML#The_content_of_a_frame
20170508 21:35:33< DeFender1031> duration= is a hard duration for the entire sequence.
20170508 21:36:00< DeFender1031> the :### syntax is when you want to give specific durations to each image in the sequence.
20170508 21:36:38< DeFender1031> (the nomenclature here is somewhat difficult, as [frame] really means image sequence, and a single [frame] can contain multiple frames)
20170508 21:36:54< DeFender1031> (i hope that made sense)
20170508 21:37:00< celticminstrel> I think that's not quite the right way of looking at it?
20170508 21:37:17< celticminstrel> A [frame] is a single frame, but the progressive syntax expands it into a sequence of [frame]s.
20170508 21:37:25< celticminstrel> If I understand correctly.
20170508 21:37:39< DeFender1031> the point is, if you want the [frame] to have a hard duration, use duration=, if you want each actual image frame within it to have a specified duration, you use the : syntax
20170508 21:37:55< celticminstrel> And if you use both, it does scale them down?
20170508 21:38:11-!- gfgtdf [~chatzilla@x4e3639cc.dyn.telefonica.de] has joined #wesnoth-dev
20170508 21:38:50< DeFender1031> celticminstrel, almost accurate, except that the duration= frame applies to everything it's expanded to, the halo= attribute can be expanded to frames with different durations than the image= attribute, and it's generally a mishmash of the concept of a frame and the concept of a progression.
20170508 21:39:52< DeFender1031> if you use both, I think it respects the duration= by truncating the : frame s if they're longer, or freezing on the last image if they're shorter, IIRC.
20170508 21:41:24< celticminstrel> Ah...
20170508 21:41:26< DeFender1031> meaning, if I did duration=700 and image=blah[1~5]:200 if I'm not mistaken, it'd show frames 1-3 for 200 ms each, frame 4 for 100 ms, and skip frame 5 entirely.
20170508 21:41:38< DeFender1031> it's been a while since I tested that though.
20170508 21:41:45< DeFender1031> As one tends to not want that behavior.
20170508 21:41:59< DeFender1031> so one tends to use only one or the other but not both.
20170508 21:42:47< DeFender1031> also, if the total halo= duration and the total image= duration don't match (and there's no duration=), IIRC it shows both, freezing the shorter one on the last frame until the longer one completes.
20170508 21:42:58< DeFender1031> but again, I'm not entirely certain of that.
20170508 21:43:13< DeFender1031> oh, and also the x, y, and offset can take durations.
20170508 21:43:43< DeFender1031> all i remember is that it gets VERY weird when durations start not lining up, and it's best to not make those kinds of mistakes.
20170508 21:45:04< DeFender1031> Honestly, I'd like to see the game error out over that instead.
20170508 21:45:56< DeFender1031> come to think of it, that might solve at least some of the issue here. meaning, if you know this frame should be 2000 ms, then you set the duration=2000. If the expansion ends up making it something else, then you know you made a mistake in the expansion
20170508 21:46:30< DeFender1031> unless there are cases where the durations are made to purposely not line up for some godawful reason.
20170508 21:47:39-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20170508 21:47:44< DeFender1031> there's also the implicit durations, where you can give some images in an expansion or movements in a position an explicit duration while leaving others out, and it'll split whatever time is remaining among the ones which aren't explicit
20170508 21:48:07< DeFender1031> (which, come to think of it, is the same "best fit" idea as I'm proposing)
20170508 21:50:11< celticminstrel> Oh right, isn't that why we can't use ":" for step size?
20170508 21:52:56< DeFender1031> no...?
20170508 21:53:09< DeFender1031> oh... maybe.
20170508 21:53:31< DeFender1031> come to think of it, we've freed up * for step size.
20170508 21:54:21< DeFender1031> if [[1~3]*3] is 123123123 and [[1~3]*~3] is 111222333 then [1~3*2] could be 13
20170508 21:54:54< DeFender1031> it's not the most intuitive, but it does have the advantage of not reserving any characters for special meanings that weren't reserved before.
20170508 21:55:32< DeFender1031> anyway, I've gotta crash soon. I'll sleep on this, maybe see if anyone else has any input before I actually think about how to implement.
20170508 22:03:43< celticminstrel> zookeeper: Does that notation sound good to you?
20170508 22:03:57< zookeeper> i haven't been following
20170508 22:04:07< celticminstrel> The example just a couple of lines up.
20170508 22:04:17< celticminstrel> With commas inserted between all the digits.
20170508 22:04:48< celticminstrel> Blah, still 48 issues with attachments...
20170508 22:05:47< zookeeper> uh, well, [[1~3]*~3] is very arcane and "[1~3*2] could be 13" looks nonsensical
20170508 22:06:11< zookeeper> oh, hmh, i get the latter now
20170508 22:06:25< celticminstrel> The rationale for [1~3*2] -> 1,3 is "sequence from 1 to 3 with a step of 2".
20170508 22:06:29< zookeeper> 1~3 turned into two steps, so 1,3
20170508 22:06:34< zookeeper> yeah
20170508 22:06:37< celticminstrel> So for example [1~7*2] would be [1,3,5,7]
20170508 22:06:52< celticminstrel> The existing notation is a little arcane to though.
20170508 22:06:55< zookeeper> ...and all this would be backwards-compatible?
20170508 22:06:58< celticminstrel> Maybe less so, though.
20170508 22:07:01< celticminstrel> ^too
20170508 22:07:20< celticminstrel> zookeeper: Pretty sure he'd change it if he found he couldn't make it backwards-compatible.
20170508 22:07:28< celticminstrel> IIRC he planned to use the unit tests to make sure it was.
20170508 22:08:44< zookeeper> sounds sane
20170508 22:12:54< celticminstrel> The main thing I'd be worried about is whether [[1~3]*3] is already accepted in some capacity.
20170508 22:13:11< celticminstrel> Since IIRC bracketed expansions can be nested.
20170508 22:26:06-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170508 22:26:19-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170508 22:30:13-!- irker254 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20170508 22:31:57-!- Alkenrinnstet [~alkenrinn@42.61.217.253] has quit [Ping timeout: 240 seconds]
20170508 22:38:01-!- Alkenrinnstet [~alkenrinn@42.61.217.253] has joined #wesnoth-dev
20170508 22:58:07-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 258 seconds]
20170508 23:00:37-!- chiec [~chiec@2a01:c50e:3509:1f00:1f05:155b:5053:7345] has quit [Remote host closed the connection]
20170508 23:15:10-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170508 23:15:15-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20170508 23:15:16-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170508 23:19:20-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170508 23:19:52-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170508 23:44:57-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 272 seconds]
20170508 23:53:14-!- Greg-Boggs [~greg_bogg@c-73-37-6-51.hsd1.or.comcast.net] has joined #wesnoth-dev
20170508 23:54:31< celticminstrel> Wait, why does this need a separate enum? https://github.com/wesnoth/wesnoth/blob/1.13.5/src/units/types.cpp#L825
20170508 23:54:46< celticminstrel> It should be able to just add the "female^" prefix if the unit is female, right?
20170508 23:55:00< celticminstrel> Oh, I guess it is needed for gettext to pick them up, though.
20170508 23:55:50< celticminstrel> Maybe the alignment translations could be moved to english.cfg.
--- Log closed Tue May 09 00:00:18 2017