--- Log opened Sat Oct 11 00:00:18 2014
20141011 00:01:07-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20141011 00:05:09-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20141011 00:05:35< iceiceice> shadowm: re #285 (alpha thresholding)
20141011 00:05:43< iceiceice> i tested it at various points over the past few weeks
20141011 00:05:49< iceiceice> i have some screenshots
20141011 00:05:58< iceiceice> not many
20141011 00:06:07< iceiceice> i might make one that tries to illustrate my comment in the commit
20141011 00:07:32< shadowm> Well, I'll be taking a look at the results once I'm done with my current ask anyway.
20141011 00:07:51< iceiceice> ok, thanks
20141011 00:08:15< iceiceice> gfgtdf: i didn't realize that the boost filesystem branch was pushed out to a public repo
20141011 00:08:38< iceiceice> if you want to try to merge it, i would be willing to help
20141011 00:08:56< iceiceice> i think we should make sure that its the whole branch though and not just part of it
20141011 00:09:04< iceiceice> he might have most of it local or something
20141011 00:09:29< shadowm> There are also two branches in the public repo with similar names for some reason.
20141011 00:09:46< shadowm> boost_filesystem and boost_filesystem_win.
20141011 00:12:12< iceiceice> shadowm: question about gui stuff,
20141011 00:12:22< iceiceice> it is my assumption that we cannot make "modeless" windows
20141011 00:12:34< iceiceice> meaning like, a window or pane that does not grab focus
20141011 00:12:47< iceiceice> i only assume because i never saw one
20141011 00:12:49< iceiceice> in wesnoth
20141011 00:12:59< iceiceice> do you know if this is accurate, or if theres some gui2 way to do this?
20141011 00:13:10< shadowm> It's not possible in GUI2 yet, no.
20141011 00:13:50< iceiceice> ok, thanks
20141011 00:14:46< shadowm> As for GUI1, I'm not completely sure. The thing is that GUI1 widgets are "closer to the wire", to put it in some way, allowing things like the : console in-game to happen by (presumably) wiring one or more widgets into the main event loop.
20141011 00:15:32< shadowm> But GUI2 does its own confusing thing with event handling.
20141011 00:20:44< iceiceice> ok, i think i'm going to request this as a feature
20141011 00:21:28< shadowm> Good luck. ;)
20141011 00:22:24< iceiceice> hehe
20141011 00:22:48< iceiceice> who knows if it will happen, but actually i've found that a large number of features i requested magically got filled
20141011 00:23:14< iceiceice> maybe it will motivate someone, anyways its nice to have a concrete request for something when you start working on something i feel
20141011 00:25:33< shadowm> Hm, did the logger API change between 1.12 and master?
20141011 00:26:49< shadowm> Oh, I see, master added a convenience overload for lg::set_log_domain_severity().
20141011 00:27:58< shadowm> Eh.
20141011 00:28:07< shadowm> Also changed the interface to lg::logger, hm.
20141011 00:39:43< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 a5dd652970cc / src/campaign_server/campaign_server.cpp: campaignd: Use utils::lowercase() instead of a direct call to libc tolower() http://git.io/I-4_6g
20141011 00:39:46< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 33808a290280 / src/campaign_server/ (campaign_server.cpp campaign_server.hpp): campaignd: Rename and refactor campaign_server class into campaignd::server http://git.io/e8CDtg
20141011 00:39:49< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 df6aad4502b4 / src/campaign_server/ (campaign_server.cpp campaign_server.hpp): campaignd: Rename a server class field and some scoped_ostream objects http://git.io/F65dyQ
20141011 00:39:52< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 6df2cfd9b335 / src/campaign_server/ (campaign_server.cpp campaign_server.hpp): campaignd: Refactor config commit boilerplate into a separate method http://git.io/o1R-dQ
20141011 00:39:55< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 29c77cbd611b / src/campaign_server/ (campaign_server.cpp campaign_server.hpp): campaignd: Make input_ a boost::scoped_ptr http://git.io/zh8R3g
20141011 00:39:58< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 52337f93513e / src/campaign_server/campaign_server.cpp: campaignd: Replace write_config() scheduling mechanism http://git.io/6vBIZQ
20141011 00:40:01< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 8f6d82bcb7e4 / src/campaign_server/ (campaign_server.cpp campaign_server.hpp): campaignd: Refactor request handling blocks into separate methods http://git.io/KgoncA
20141011 00:40:04< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 04c97adbb23b / src/campaign_server/ (campaign_server.cpp campaign_server.hpp): campaignd: Refactor construct_message/error() functions and boilerplate http://git.io/9H1XZQ
20141011 00:40:07< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 8a65dfc8597d / src/ (5 files in 2 dirs): campaignd: Move several anonymous namespace members to a new file http://git.io/M67oOw
20141011 00:40:10< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 25846d0bbc1d / src/campaign_server/ (addon_utils.cpp addon_utils.hpp campaign_server.cpp): campaignd: Move and document more ancillary functions http://git.io/uWVmyw
20141011 00:40:13< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 b38e4775cbc3 / src/campaign_server/campaign_server.cpp: campaignd: Wrap request handler registration statements in macros http://git.io/cfZd2g
20141011 00:40:16< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 47c9b63f9c8b / src/campaign_server/campaign_server.cpp: campaignd: Drop pointless use of lexical_cast to set WML attributes http://git.io/oaFwOw
20141011 00:40:19< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 7f39a46e90c9 / src/campaign_server/campaign_server.cpp: campaignd: Use empty C strings instead of t_strings to clear out attributes http://git.io/8IB3mg
20141011 00:40:22< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 5b2ae9c9582d / src/campaign_server/campaign_server.cpp: campaignd: Show RO mode warning in load_config() rather than in run() http://git.io/olaPhg
20141011 00:40:25< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 d9b023e01932 / src/campaign_server/campaign_server.cpp: campaignd: Code indentation and spacing style changes for consistency http://git.io/EJymvw
20141011 00:40:28< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 ec943244caed / src/campaign_server/campaign_server.cpp: campaignd: Use const references or values where applicable http://git.io/atTW_A
20141011 00:40:31< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 69520c44367e / src/campaign_server/campaign_server.cpp: campaignd: Write config to disk every ten minutes, not 60 seconds http://git.io/SWZ5Ag
20141011 00:40:34< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 9e1d1be45e33 / src/campaign_server/campaign_server.cpp: campaignd: Don't use GUI1 markup in [upload] response http://git.io/pgpMCg
20141011 00:40:37< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 37b531b839d0 / src/campaign_server/campaign_server.cpp: campaignd: A few style consistency fixes I missed before http://git.io/oh3tCg
20141011 00:40:40< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 0b4e42a312f5 / src/campaign_server/ (campaign_server.cpp campaign_server.hpp): campaignd: Replace vector<pair<k,v>> with map<k,v> for the handlers table http://git.io/pJZwqQ
20141011 00:40:43< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 61617164bc0a / src/campaign_server/campaign_server.cpp: campaignd: Print the destination IP in the send_error() call http://git.io/wVG37A
20141011 00:40:46< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 b11bce16ae0a / src/campaign_server/campaign_server.hpp: campaignd: Add a typedef for the handlers map http://git.io/Eh2vhw
20141011 00:40:49< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 e6ca14f50d03 / src/campaign_server/campaign_server.cpp: campaignd: Use find() to locate a request handler http://git.io/MF7bGQ
20141011 00:40:52< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 f1dc59f653a0 / src/campaign_server/campaign_server.cpp: campaignd: Reject bad requests with an [error] block http://git.io/jISLvA
20141011 00:40:55< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 cf60d5c249bb / src/campaign_server/campaign_server.hpp: campaignd: Document send_message/send_error's WML output and side-effects http://git.io/4gLInA
20141011 00:40:58< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 e34593b09aa5 / src/campaign_server/campaign_server.cpp: campaignd: Log errno to stderr if deleting an add-on archive failed http://git.io/-ti2OQ
20141011 00:41:01< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 5d744f65a669 / src/campaign_server/campaign_server.cpp: campaignd: Catch boost::bad_function_call exception http://git.io/tP0yFQ
20141011 00:41:04< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 01338720e515 / src/campaign_server/campaign_server.cpp: campaignd: Log messages to a new log domain of our own http://git.io/qMzTLQ
20141011 00:41:07< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 09e4374d7364 / src/campaign_server/campaign_server.cpp: campaignd: Move error messages to the error logger, drop inline "ERROR" label http://git.io/3iT6fw
20141011 00:41:10< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 45f15e6b08ab / src/ (6 files in 2 dirs): Merge branch 'backport/1.12/campaignd-refactoring' into 1.12 http://git.io/mFI7rw
20141011 00:42:04< iceiceice> that's a lot of commits XD
20141011 00:42:14< shadowm> (I'm making use of my "I'm my code's single user" card. Don't try this at home.)
20141011 00:51:08< irker456> wesnoth: Ignacio R. Morelle wesnoth:master fd1ce64d8868 / src/campaign_server/ (campaign_server.cpp campaign_server.hpp): campaignd: Add server* this parameter to server::request_handler type http://git.io/TJ-MUA
20141011 00:51:11< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 d7412ca7b96e / src/campaign_server/ (campaign_server.cpp campaign_server.hpp): campaignd: Add server* this parameter to server::request_handler type http://git.io/7FGutQ
20141011 00:54:33-!- Coffee_irc [~david@ppp118-210-39-35.lns20.adl2.internode.on.net] has quit [Quit: Konversation terminated!]
20141011 00:58:13< shadowm> I.
20141011 00:58:30< shadowm> I'm seeing something involving the Spanish translation and the new UI theme I don't like.
20141011 00:59:30< shadowm> On the sidebar, "MP" -> "Multijugador" ("multiplayer"). The string has no disambiguation marker.
20141011 01:00:25< mattsc> shadowm: beetlenaut made a forum post about that
20141011 01:00:25< shadowm> Well, actually, it seems "MP" in that textdomain is only ever used for the sidebar, so it's the translators' fault.
20141011 01:00:28< mattsc> I think
20141011 01:00:40< shadowm> #: data/themes/default.cfg:151 data/themes/unit_box.cfg:394
20141011 01:01:02< shadowm> If it were used elsewhere it'd be the coder's fault.
20141011 01:01:26< shadowm> So, where's the Spanish translation team so I can complain to them?
20141011 01:01:37< shadowm> :|
20141011 01:02:13< shadowm> Yeah, SpanishTranslation says "there isn't a maintainer, contact Espreon for more info".
20141011 01:02:18< mattsc> the only one active appears to be jorge barroso - you could contact him  ;)
20141011 01:02:33< mattsc> maybe we should make him the maintainer
20141011 01:02:51< shadowm> If he were active then he should be listed as the maintainer, so my suspicion is that that list is outdated.
20141011 01:03:14< mattsc> he was active as recent as Aug 11
20141011 01:03:22< shadowm> (I mean, how do you even get an active translator for a maintainerless translation, how does that even happen.)
20141011 01:03:32< mattsc> anyways, I need to go back to repairing mufflers …
20141011 01:04:28< shadowm> mattsc: I'm afraid that if I email any of these people, they'll proceed to coerce me to maintain their stuff.
20141011 01:05:14< mattsc> yeah, I know how that goes
20141011 01:05:18< shadowm> They'll do the puppy eyes thing and then I will be stuck again translating stuff again.
20141011 01:05:31< mattsc> the one asking the question always ends up the volunteer.
20141011 01:05:49< mattsc> I could ask, since my Spanish is very … rudimentary … :P
20141011 01:06:00< mattsc> anyways, bbl
20141011 01:06:37< shadowm> Yeah, I guess I could pretend I don't speak Spanish and that it was my Spanish-speaking friend who complained about the sidebar label.
20141011 01:08:31< shadowm> iceiceice: I'm on the test scenario with 1.12 and master, and at least terrain zooming looks the same to me.
20141011 01:09:30< iceiceice> i think this was a really subtle effect...
20141011 01:09:38< shadowm> Portrait scaling looks the same too. In which situation your PR was meant to make a difference?
20141011 01:09:41< iceiceice> it should only affect the alpha channel
20141011 01:09:45< iceiceice> and at the edges of images
20141011 01:09:48< iceiceice> basically
20141011 01:10:39< iceiceice> the effect is, when you are resizing an image,
20141011 01:10:44< shadowm> Looking at a scaled-down portrait with the GIMP, it looks the same in both versions.
20141011 01:11:01< iceiceice> i think it should mostly only effect zoomed in images
20141011 01:11:04< shadowm> So maybe I could actually find the differences if I rendered it on a solid color background.
20141011 01:11:33< iceiceice> yeah so its an alpha channel effect
20141011 01:11:46< iceiceice> solid background would probably be a good idea
20141011 01:12:24-!- prkc [~prkc@catv-80-99-60-5.catv.broadband.hu] has quit [Remote host closed the connection]
20141011 01:12:50< shadowm> Health bars get shifted around when zooming in on master but not 1.12, hm.
20141011 01:13:22< iceiceice> yeah, lipkab introduced that
20141011 01:13:31< iceiceice> when he made the respositionable health bars
20141011 01:13:35< iceiceice> i assigned it to him a few weeks ago
20141011 01:13:46< iceiceice> bug #22647
20141011 01:15:47-!- gfgtdf [~chatzilla@e177019093.adsl.alicedsl.de] has joined #wesnoth-dev
20141011 01:16:33-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Ping timeout: 272 seconds]
20141011 01:17:02< shadowm> portraits/elves/transparent/druid.png~BG()~SCALE(700,700) looks the same to me in both branches, zoomed in with the GIMP.
20141011 01:17:29< shadowm> (That's solid background color #000, then scale up to 700x700.)
20141011 01:18:11< shadowm> The portrait does have non-solid edges, so... shrug.
20141011 01:22:20< shadowm> mattsc: Oh yeah, hahaha, it's even in the forum index.
20141011 01:22:29< shadowm> http://forums.wesnoth.org/viewtopic.php?f=7&t=41064
20141011 01:32:33< shadowm> I replied anyway.
20141011 01:38:01-!- matthiaskrgr [matthiaskr@gateway/shell/panicbnc/x-kphggisprsmbfvwz] has quit [Quit: PanicBNC - https://PanicBNC.net - currently sucks]
20141011 01:44:24-!- Necrosporus_ [~Necrospor@176.114.191.31] has joined #wesnoth-dev
20141011 01:47:18-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has quit [Ping timeout: 246 seconds]
20141011 01:49:39-!- matthiaskrgr [matthiaskr@I.Eat.Babies.PanicBNC.net] has joined #wesnoth-dev
20141011 01:50:03-!- matthiaskrgr is now known as Guest2509
20141011 01:51:39< mattsc> I’ll have to see if that does it or if I have to get some wire mesh ...
20141011 01:52:30< mattsc> shadowm: well, if you give it away like that …   (my Spanish is sufficient to understand what you wrote, but that’s where it ends)
20141011 01:53:03< mattsc> fortunately, I don’t have that problem, as my native language is so overrepresented in Wesnoth that people even have their own channel
20141011 01:56:23< ancestral> #wesnoth-lolspeak
20141011 01:57:15< iceiceice> shadowm: i played around with the alpha thing again and made some more screenshots
20141011 01:57:29< iceiceice> so the most differences are notice are when you zoom in,
20141011 01:58:07< iceiceice> pre patch, the edges of units often look strange and "stick out" a bit, its a bit hard to describe
20141011 01:58:18< iceiceice> post patch they blend in a bit better at the edges
20141011 01:58:32-!- ancestral [~ancestral@17.114.45.53] has quit [Quit: ancestral]
20141011 01:58:45-!- LovCAPONE [~LovCAPONE@modemcable052.221-21-96.mc.videotron.ca] has joined #wesnoth-dev
20141011 01:58:48< iceiceice> however, i think post patch and in some situations, there is a "hex tearing artifact" that may be a bit aggravated
20141011 01:59:02< iceiceice> its present in pre and post patch and various zoom levels...
20141011 01:59:23< iceiceice> i'm not sure how much worse it is or if it is actually worse now
20141011 01:59:34< iceiceice> i didnt notice before now
20141011 02:00:28-!- ancestral [~ancestral@17.114.45.53] has joined #wesnoth-dev
20141011 02:01:08-!- ancestral [~ancestral@17.114.45.53] has quit [Client Quit]
20141011 02:01:58< iceiceice> actually, its only really noticeable with fog on
20141011 02:04:51< LovCAPONE> Question: I've pushed an issue (to branch 1.12) that affected only branch 1.12 but I've been asked why I didn't foward-port it to master. Isn't "master" branch 1.13?
20141011 02:05:28< iceiceice> yes
20141011 02:05:53< LovCAPONE> iceiceice: ok that's what I thought. Ok thanks
20141011 02:06:18< iceiceice> shadowm: i'm not so sure committing the patch was a good idea now... its at least debatable
20141011 02:07:38< iceiceice> i will think about what to do, reverting it is definitely on the table
20141011 02:08:45-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20141011 02:08:58-!- LovCAPONE [~LovCAPONE@modemcable052.221-21-96.mc.videotron.ca] has left #wesnoth-dev []
20141011 02:11:12< gfgtdf> iceiceice: i tried to cherrypick AIs first commit from the boost filesystem branch https://github.com/gfgtdf/wesnoth-old/commit/c990956031da897de1fa851de37a4628c2c867e8 and since that was quiet some work, i'd like to push it to master (after travis passed ofc) you see reasons against ?
20141011 02:12:45< iceiceice> no i dont
20141011 02:13:08< iceiceice> it sounds like thats a good thing no matter what happens
20141011 02:13:17< iceiceice> (looking at the commit i mean)
20141011 02:17:27< shadowm> mattsc: There is a #wesnoth-es, but it's pretty dead.
20141011 02:19:44-!- Coffee_irc [~david@ppp118-210-39-35.lns20.adl2.internode.on.net] has joined #wesnoth-dev
20141011 02:20:13< gfgtdf> iceiceice: can it be that the checkkypicked version differs to much from teh original version that, if i try to pull the original versions branch git doesn't recognize that i already have the first commit ?
20141011 02:20:17< mattsc> shadowm: there is?  Didn’t know that ...
20141011 02:20:25< mattsc> ancestral: I wish!
20141011 02:20:51< iceiceice> I actually don't know what the normal behavior is with regards to that
20141011 02:21:40< iceiceice> i actually wouldn't be surprised if it doesn't try to do that, because the cherrypick and the original are different commits
20141011 02:22:22< iceiceice> it might be that rebase is a good way to go here?
20141011 02:22:55< shadowm> gfgtdf: Git won't apply a patch that it guesses has already been applied in its entirety, if that's what you are asking.
20141011 02:23:04< iceiceice> what will happen if you try to merge the two branches, will it give you opportunities to fix conflicts for each commit?
20141011 02:23:11< iceiceice> i guess it wont right, it will make you do them all at once
20141011 02:23:13< shadowm> However, say you have patches A, B, and C, and you merge C first.
20141011 02:23:30< iceiceice> if you rebase then i guess it will prompt you to fix conflicts for each commit
20141011 02:23:31< shadowm> You might be unable to merge A and B correctly if they depend upon the pre-C state.
20141011 02:23:45< shadowm> Or you might in fact be able to merge, but the result won't compile.
20141011 02:24:03< shadowm> (Unless you hand-fix any relevant code in A and B before merging.)
20141011 02:24:07< iceiceice> gfgtdf: maybe you should go to the branch from right after the cherrypicked commit
20141011 02:24:09< iceiceice> and try to use rebase --onto
20141011 02:24:21< iceiceice> to rebase the rest of the branch, or some portion of it onto the cherry-picked part
20141011 02:24:34< iceiceice> and i guess we can work through that over some time / on a separate branch
20141011 02:25:13< iceiceice> hmm idk
20141011 02:25:18< iceiceice> maybe this is a bad idea though
20141011 02:25:36< iceiceice> its breaking naively the rule of, dont rebase someone elses commits...
20141011 02:25:37-!- Ivanovic_ [~ivanovic@frnk-4d016be1.pool.mediaWays.net] has joined #wesnoth-dev
20141011 02:25:57< iceiceice> i dont know it might screw up AI's fork history?
20141011 02:26:03< iceiceice> maybe we have to just cherry-pick the entire branch i guess
20141011 02:26:35< shadowm> iceiceice: I saw that at max zoom hexes get bogus edges after mouse over. Is that what you mean?
20141011 02:27:01< iceiceice> they get that in the pre-patch version too though
20141011 02:27:02< shadowm> For example, village hexes in the top rows of the test scenario start bleeding into one another.
20141011 02:27:07< iceiceice> just not as often
20141011 02:27:20< iceiceice> *more qualifiers ...
20141011 02:27:25< shadowm> Yes, so I don't think that's a reason to revert your patch.
20141011 02:28:04-!- ancestral [~ancestral@12.23.74.29] has joined #wesnoth-dev
20141011 02:28:15< shadowm> If I had to guess, there's probably a floating point approximation error somewhere.
20141011 02:28:25-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 260 seconds]
20141011 02:28:28< shadowm> Hence it doesn't happen at 1:1 zoom.
20141011 02:28:53< iceiceice> yeah so,
20141011 02:28:55< shadowm> (Somewhere in the tile rendering code, to be specific, not the surface scaling algo.)
20141011 02:28:58< iceiceice> here was my reasoning for adding the patch,
20141011 02:29:05< iceiceice> if we are using this scaling function everywhere for all images,
20141011 02:29:14< iceiceice> we don't really want there to be squirrelly stuff happening with the alpha channel
20141011 02:29:24< iceiceice> because it could really hinder someone working on something random
20141011 02:29:31-!- Ivanovic_ is now known as Ivanovic
20141011 02:29:33< iceiceice> like idk if bigmap images need to use alpha for some feature
20141011 02:29:44< iceiceice> and someone is working on this and thinking "why does wesnoth butcher the alpha channel??"
20141011 02:29:50< shadowm> Translation overlays, for one.
20141011 02:30:10< iceiceice> so even if theres rarely a noticeable change i think its better not to have it
20141011 02:30:18< gfgtdf> iceiceice: also does it even make sense to push a _optional_ boost_filesystem based implementation to maker, i mean this could mean we have to update 2 different implementations of filesystem now.
20141011 02:30:24< iceiceice> but on the other hand, i guess the only noticeable changes ive seen are with zoomed in game screens
20141011 02:30:29< iceiceice> and its not clear that this makes it better
20141011 02:30:36< iceiceice> i think its makes some things better and maybe somethings worse
20141011 02:30:47< iceiceice> otoh the game just doesn't look so great zoomed in anyways
20141011 02:30:49< shadowm> I'm pretty sure the new bigmap itself uses a non-binary alpha channel for the parchment edge effects (it's overlaid on top of a solid background so it's not obvious unless you check the image files).
20141011 02:31:40< iceiceice> heres some screenshots: http://imgur.com/a/0dfGK#7
20141011 02:32:15< shadowm> Ideally, for zooming we should use different algorithms: one optimized for terrain and portrait art, and another for pixel art (sprites and health bars).
20141011 02:34:20< iceiceice> gfgtdf: maybe we should make it nonoptional on master
20141011 02:34:33< iceiceice> i guess we should post on the list but,
20141011 02:34:44< gfgtdf> iceiceice: but that raises boost requirement to 1.44
20141011 02:34:51< iceiceice> i think thats okay for 1.13
20141011 02:34:58< iceiceice> we should check with ivanovic though
20141011 02:35:10< iceiceice> if i understood correctly he's not going to make a 1.14 pandora port
20141011 02:35:27< iceiceice> and theres no one else who objects to increasing the boost requirements, or adding boost filesystem or boost random
20141011 02:35:40-!- travis-ci [~travis-ci@ec2-54-87-134-110.compute-1.amazonaws.com] has joined #wesnoth-dev
20141011 02:35:40< travis-ci> gfgtdf/wesnoth-old#255 (AT-boost_filesystem - c990956 : Alexander van Gessel (AI0867)): The build failed.
20141011 02:35:40< travis-ci> Build details : http://travis-ci.org/gfgtdf/wesnoth-old/builds/37659192
20141011 02:35:40-!- travis-ci [~travis-ci@ec2-54-87-134-110.compute-1.amazonaws.com] has left #wesnoth-dev []
20141011 02:35:59 * shadowm has no idea what the branch actually accomplishes.
20141011 02:36:20< shadowm> I've never used Boost.filesystem either, which is part of the reason I have no idea.
20141011 02:36:30< iceiceice> shadowm: i looked at the help pages for boost filesystem a little bit...
20141011 02:36:42< iceiceice> i think the point is, you have a generic "filesystem" abstraction
20141011 02:36:50< iceiceice> that represents paths to directories and files,
20141011 02:37:01< iceiceice> and it works on all platforms
20141011 02:37:11< iceiceice> so you dont have to have separate windows / linux / mac code for this
20141011 02:37:26< iceiceice> and it allows you to check if files exist and such
20141011 02:37:39< iceiceice> without crashing / getting wierd crap, it translates errors into reasonable c++ exceptions
20141011 02:38:16< shadowm> I've heard C++ exceptions are notably slow on Win32.
20141011 02:38:34< iceiceice> well we use them like crazy
20141011 02:38:38< iceiceice> so...
20141011 02:38:53< shadowm> Only for exceptional events AFAIK.
20141011 02:38:57< iceiceice> no,
20141011 02:39:01< shadowm> (Ending a scenario is a exceptional event.)
20141011 02:39:03< iceiceice> every time you click the end turn button 
20141011 02:39:13< iceiceice> it generates about 3 exceptions i think
20141011 02:39:34< iceiceice> since typically its going to pass through lua unless i'm wrong about that
20141011 02:39:50< shadowm> Still more exceptional than rendering a single frame or (to name a possible concern) loading an add-on.
20141011 02:40:13< shadowm> Or to be more specific, reading every single file from an add-on's WML tree.
20141011 02:40:30< iceiceice> i think there should only be exceptions if files are not found
20141011 02:41:17< iceiceice> idk if you generated thousands of exceptions this way i think it's really a corner case
20141011 02:41:19< shadowm> I foresee that impacting the terrain builder case, though for all I know we might already throw exceptions like candy in that case.
20141011 02:41:39< iceiceice> yeah you would know better than me
20141011 02:41:44< iceiceice> if i'm not mistaken AI0867 uses windows?
20141011 02:41:50< iceiceice> i'm sure hes tested this to some extent
20141011 02:42:05< shadowm> (Most terrain graphics macros are generic and refer to nonexistent files which Wesnoth simply ignores rather than complaining about them/loading the debug placeholder. Or so it used to be, anyway.)
20141011 02:42:21< iceiceice> tbh our terrain graphics code is slow as hell anyways
20141011 02:42:50< shadowm> I don't know if AI0867 "uses" Windows, because from memory I believe his primary platform is Debian.
20141011 02:42:52< iceiceice> there was a point in time when i was running valgrind regularly to get a sense of this kind of thing
20141011 02:43:13< iceiceice> when you run a wml unit test, generally half the time is spent loading the config cache, and half the time is spent building the terrain graphics
20141011 02:43:15< shadowm> I assume he at least "uses" Windows more often than me. :P
20141011 02:43:36< iceiceice> when you say "notably slow" how slow are we talking?
20141011 02:44:03< shadowm> That's an excellent question since I'm only stating what I know from hearsay.
20141011 02:44:37< shadowm> I don't think I've seen hard figures anywhere, and I'm allergic to those anyway, so unless I got to experience it in practice some day (which is unlikely)...
20141011 02:45:17< iceiceice> ok, so i think it might be premature optimization to worry about this right now
20141011 02:45:25< iceiceice> if we don't do something, windows users can't really play the game at all
20141011 02:45:25< shadowm> Re parser, I haven't used valgrind's profiling tools, but it is indeed where most time is spent when running with memcheck.
20141011 02:46:10< iceiceice> another point,
20141011 02:46:18< shadowm> -O0 builds are particularly impacted by preprocess/parse activity, even without memcheck clogging allocations.
20141011 02:46:31< iceiceice> it seems unlikely that exceptions are slow on "windows", but maybe for certain compilers, right?
20141011 02:47:09< iceiceice> it depends on what the exact issue is i guess... is there some generic windows C++ debugger that's just a bear at runtime or something?
20141011 02:47:13< shadowm> Maybe it depends on the compiler, maybe it has something to do with how execution normally happens on Windows.
20141011 02:47:36< shadowm> (i.e. how function calls and stack unwinding are performed.)
20141011 02:49:40< shadowm> Regardless, I do agree that it's more important to get Wesnoth to *work* for those people rather than save them a likely marginal performance loss.
20141011 02:58:42-!- gfgtdf [~chatzilla@e177019093.adsl.alicedsl.de] has quit [Read error: Connection reset by peer]
20141011 03:04:06< shadowm> lovcapone: Sorry, I just spammed you with several bug tracker emails while assigning you bugs that were already fixed for book-keeping.
20141011 03:04:32< shadowm> Along the way I noticed I forgot to mark one of them as Fixed the other day, bah.
20141011 03:05:07< shadowm> iceiceice: The feature you alluded to before is #22779?
20141011 03:05:46< iceiceice> i think its this: https://gna.org/bugs/?22778
20141011 03:06:02< shadowm> I can't read.
20141011 03:06:28< shadowm> Yeah, would be great if certain info windows were modeless.
20141011 03:07:55< shadowm> iceiceice: Regarding #22779, I thought I'd note I actually do this *a lot* in one of my campaigns with the tools we were given, namely extra [message] attributes and Pango formatting.
20141011 03:08:45< shadowm> I've found that using portrait-less narrator messages with the character name as capation and the body in italics gets the point across.
20141011 03:17:24< shadowm> OK, I don't think I have anything pending for 1.12 RC1.
20141011 03:17:42< shadowm> Unless I promised mattsc to do something and have forgotten smehow.
20141011 03:21:08< mattsc> shadowm: you promised to fix all remaining bugs in all branches
20141011 03:22:31< shadowm> Well, evidently past me was out of his mind when he said that, so it doesn't count.
20141011 03:27:19< shadowm> Oh.
20141011 03:27:33< shadowm> Yes, I do have something pending involving running wmllint on 1.12 and master.
20141011 03:28:34< mattsc> shadowm, fabi_, thunderstruck, Ayne: here is what I have on my list as “must fix” for 1.12.0: http://pastebin.com/j7SqYrnf
20141011 03:28:49< mattsc> none of those need to be done for 1.12 RC1 though
20141011 03:29:50< mattsc> shadowm: I assume that these should go into the 1.11.17/1.12 RC1 release notes?
20141011 03:30:18< shadowm> I don't think users need to know about these unless they can help somehow.
20141011 03:30:44< shadowm> And #20815 in particular is Private and hidden from users on purpose.
20141011 03:31:03< mattsc> good point about 20815 …
20141011 03:31:13< mattsc> for the other ones though:
20141011 03:32:00< mattsc> aren’t we going to announce this somehow along the lines of: “this is it, last chance to report any serious bugs; and by the way, we do know about the following and will still fix them …”
20141011 03:32:37< shadowm> Well, yes, actually.
20141011 03:33:04< mattsc> (that’s meant as an honest question, as this is my first time I’m involved with a new stable release)
20141011 03:33:56< shadowm> I write those shiny RA bbcodes for this kind of thing. :p
20141011 03:34:38< mattsc> :)
20141011 03:34:39< shadowm> The 1.11.15 announcement includes examples of both the "important notice" block and the "known issues" section: http://forums.wesnoth.org/viewtopic.php?f=5&t=40493
20141011 03:35:24< shadowm> We don't need the first one unless there's some compatibility-breaking change in 1.11.17.
20141011 03:35:36< shadowm> Hopefully not. Because it's our last chance to break compatibility, ever.
20141011 03:35:43< mattsc> I remember.  As much as I like subtlety, sometimes in your face is the right way to go
20141011 03:35:49< mattsc> I don’t think there are
20141011 03:36:53< shadowm> So, who will write the announcement?
20141011 03:37:11< mattsc> remember what I said up there ^
20141011 03:37:33< shadowm> Oh, okay, you'll do it then.
20141011 03:37:35< mattsc> 20141011 01:05:31< mattsc> the one asking the question always ends up the volunteer.
20141011 03:37:50< mattsc> Sue, I’ll do it; with your help
20141011 03:37:56< mattsc> *Sure
20141011 03:38:19< mattsc> Just remember that I’m not all that good with words ...
20141011 03:38:22< shadowm> Remember to keep the "is expected to become available [long series of stricken estimates]" part.
20141011 03:38:58< shadowm> I don't know what comes after 'soon', probably 'in a couple of weeks'.
20141011 03:39:12< shadowm> But then, my draft of the press release says December, hm.
20141011 03:39:37< mattsc> Well, new release every 2 weeks until everything’s good.
20141011 03:40:03< shadowm> Can Ivanovic actually commit to such a release schedule?
20141011 03:40:07< mattsc> The two big ones, IMO, are 20815 and the unicode one
20141011 03:40:28< mattsc> We can always commit him to it…
20141011 03:40:51< mattsc> No, seriously, that’s what he suggested to me himself  [Ivanovic, private communication]
20141011 03:41:14< mattsc> but of course, that sometimes might mean a bit more and sometimes a bit less than 2 weeks.
20141011 03:41:36< mattsc> I’m an astronomer after all, everything within a factor of π is perfectly acceptable.
20141011 03:41:39< shadowm> Also, for reference: Debian jessie freezes on the 5th of November, so that's the last opportunity for Wesnoth 1.11.x/1.12.x release to make it to the next Debian stable outside of the optional backports repository.
20141011 03:41:49< shadowm> *for a
20141011 03:42:03< mattsc> oh, wow, that’s soon ...
20141011 03:42:41< shadowm> At the start of the year I thought we'd be at 1.12.1 or 1.12.2 by then.
20141011 03:43:00< mattsc> well, yeah, the (extended) summer was a bit slow.
20141011 03:43:17< mattsc> I’m happy to see that we have picked up so much momentum in the last week or two ...
20141011 03:43:45< mattsc> Maybe we can get to RC3 by Nov 5, which should be pretty close to done, I hope
20141011 03:43:54< mattsc> be back in a few ...
20141011 03:46:09< shadowm> "../../data/campaigns/Under_the_Burning_Suns/scenarios/04_Descending_into_Darkness.cfg", line 346: [remove_shroud] without "side" attribute is now applied to all sides
20141011 03:46:28< shadowm> Only problem found by wmllint on 1.12 that doesn't involve LoW.
20141011 03:48:03< mattsc> nice
20141011 03:48:55< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 f968070eb041 / data/campaigns/Under_the_Burning_Suns/scenarios/04_Descending_into_Darkness.cfg: UtBS S4: Specify side=1 for [remove_shroud] action http://git.io/SDaegA
20141011 03:48:58< irker456> wesnoth: Ignacio R. Morelle wesnoth:master bd7f87305b3a / data/campaigns/Under_the_Burning_Suns/scenarios/04_Descending_into_Darkness.cfg: UtBS S4: Specify side=1 for [remove_shroud] action http://git.io/Z4c0eQ
20141011 03:49:43< shadowm> wmlindent found nothing to be corrected in 1.12.
20141011 04:01:01-!- fabi_ [~fabi@p548634BC.dip0.t-ipconnect.de] has joined #wesnoth-dev
20141011 04:01:03-!- fabi_ [~fabi@p548634BC.dip0.t-ipconnect.de] has quit [Changing host]
20141011 04:01:03-!- fabi_ [~fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20141011 04:01:48< shadowm> rofl: "TODO include brakes can't be escaped holy **** syntax."
20141011 04:02:59< iceiceice> hmm this is annoying
20141011 04:03:11< iceiceice> i think theres something wierd going on with our sound library,
20141011 04:03:18< iceiceice> i'm trying to debug one of these lobby sounds bugs,
20141011 04:03:37< iceiceice> i've now added a blocking message dialog after the "arrive" sound, and it seems that in some cases it doesn't play
20141011 04:03:46< iceiceice> so its not actually a problem in the convoluted logic of the mp lobby
20141011 04:03:54< iceiceice> which is where i would have bet the farm :p
20141011 04:04:37< shadowm> fabi_: Mind if I commit a few spelling/grammar fixes to data/core/editor/help.cfg to shut wmllint up?
20141011 04:05:33-!- oldlaptop [~quassel@50-108-67-218.adr01.mskg.mi.frontiernet.net] has quit [Ping timeout: 258 seconds]
20141011 04:05:56< shadowm> iceiceice: Do any other sounds play before the arrive sound?
20141011 04:06:23< iceiceice> yes
20141011 04:06:38< shadowm> We have a voice limit for each "channel" (UI, turn bell, sound sources, game sounds, music), so the UI "channel" might be hitting its limit.
20141011 04:07:08< iceiceice> yes, i dont think thats the problem though
20141011 04:07:11< shadowm> (In reality, we have like 32 channels allocated to different purposes. IIRC UI gets only 5 channels or so.)
20141011 04:07:14< iceiceice> because, even if i wait for those sounds to end
20141011 04:07:18< iceiceice> hmm well
20141011 04:07:22< iceiceice> how many channels does a sound require?
20141011 04:07:27< shadowm> One.
20141011 04:07:32< iceiceice> this should not be an issue then
20141011 04:07:44< shadowm> Unless it's the same sound repeated multiple times, then it requires as many channels as concurrent plays.
20141011 04:07:48< iceiceice> although i am getting this: 20141011 00:06:42 info audio: All channels dedicated to sound group(3) are busy, skipping.
20141011 04:08:39-!- GallaecioLaptop [~quassel@105.153.165.83.dynamic.mundo-r.com] has joined #wesnoth-dev
20141011 04:08:40< shadowm> Then it's hitting the limit.
20141011 04:09:27< shadowm> Actually, UI gets a single channel.
20141011 04:10:14< iceiceice> ok
20141011 04:10:25< iceiceice> maybe one of the .wav files has a really long silence at the end or something
20141011 04:10:48< shadowm> 1 for turn bell, 1 for timer warning, 8 for sound sources. 1 for UI sounds, remainder for sound effects.
20141011 04:10:51< iceiceice> actually no i get it
20141011 04:10:55< iceiceice> so the problem is
20141011 04:10:58< shadowm> (From a total of 32 channels.)
20141011 04:11:14< iceiceice> when a player joins you might often have an accompanied server message
20141011 04:11:20< iceiceice> and that competes with the sound for the player i guess
20141011 04:12:10< iceiceice> maybe UI should get 2 or 3?
20141011 04:12:16< shadowm> I don't have any problems with allotting additional channels for UI and/or allocating more channels, seeing as how nobody has complained since I doubled our allocation in 1.11.11 or so.
20141011 04:13:08< iceiceice> ok
20141011 04:13:15< iceiceice> how do i increase the limit?
20141011 04:13:43< shadowm> Sound effects are assigned 21 channels according to my commit messages, so adding 1 for the UI would make for a nice round number.
20141011 04:14:15< shadowm> Ivanovic: I can do that. Do you want +1 or +2 channels?
20141011 04:14:19< shadowm> Er, iceiceice, not Ivanovic.
20141011 04:14:45< iceiceice> so i think, 3 would be nice
20141011 04:14:53< iceiceice> because theres three different sounds that can play close to one another
20141011 04:15:02< iceiceice> join / leave, server message, turn bell
20141011 04:15:11< iceiceice> i'm pretty sure the turn bell in this case is playing on the ui channel
20141011 04:15:20< iceiceice> i guess i could also try to change that...
20141011 04:15:26-!- oldlaptop [~quassel@50-108-67-218.adr01.mskg.mi.frontiernet.net] has joined #wesnoth-dev
20141011 04:15:33< shadowm> No, the turn bell plays on its own channel.
20141011 04:15:48< iceiceice> yeah but this "turn bell" is the "ready to start" sound
20141011 04:16:01< iceiceice> i'm pretty sure i'm using the play_ui_sound function for this, just using the same file path for the sound
20141011 04:16:36< shadowm> Probably worth moving it to the turn bell channel then, unless the risk of overlapping with the real turn bell is too great.
20141011 04:16:49< iceiceice> i think that risk is nonexistant
20141011 04:17:01< shadowm> The sound::play_bell() function would play on the bell channel.
20141011 04:17:04-!- ancestral [~ancestral@12.23.74.29] has quit [Quit: i go nstuf kthxbai]
20141011 04:17:42< shadowm> (That also means it'd play with the bell volume chosen in Preferences -> Sound.)
20141011 04:19:17< iceiceice> yeah...
20141011 04:19:41< iceiceice> i think i might use sound::play_bell but also try to check if the ui sounds are enabled
20141011 04:20:44< shadowm> Oka, so +1 or +2 channels to UI?
20141011 04:21:29< shadowm> It's actually a bit more involved than I thought since there's a few bits that assume there's only one UI channel, hm.
20141011 04:22:06< iceiceice> yeah, i saw that
20141011 04:22:09< iceiceice> +1 channel is probably fine
20141011 04:26:18< iceiceice> hmm so another option is,
20141011 04:26:28< iceiceice> i could just not use the play_ui_sound function for these things
20141011 04:26:33< iceiceice> and treat them as sound effects or something
20141011 04:26:49< shadowm> That'd be a form of namespace abuse.
20141011 04:26:51< iceiceice> but i guess having an extra ui sound channel might help in some toher case
20141011 04:27:46< shadowm> (It'd be better if we had the means to say "hey, we are not in a game, let's allot all our channels for UI usage".)
20141011 04:28:10< shadowm> (But the sounds interface isn't designed with such an usage in mind.)
20141011 04:29:26 * shadowm double-checks his maths.
20141011 04:29:47< iceiceice> i mean putting the "Ready to start" sound on the bell channel is also namespace abuse then
20141011 04:29:52-!- ancestral [~ancestral@12.23.74.29] has joined #wesnoth-dev
20141011 04:30:13< shadowm> Well, it's kind of a bell.
20141011 04:30:24< iceiceice> yeah... i'm not really opposed
20141011 04:30:25< shadowm> It's a notification bell, eh, I don't know.
20141011 04:30:47< shadowm> I wonder if *all* lobby sounds currently play on the UI channel?
20141011 04:30:52< iceiceice> i think they do
20141011 04:30:57< iceiceice> but i'm not sure
20141011 04:31:15< iceiceice> its better to have one "UI" volume control for all of these sounds
20141011 04:31:43< iceiceice> if someone doesn't like the turn bell, they probably won't be willing to mute that
20141011 04:31:58< iceiceice> i could hack it i guess and grab the UI volumes to use for the bell channel before playing it...
20141011 04:32:05< iceiceice> but i would feel bad :)
20141011 04:32:43< shadowm> Yeah, okay, nobody in the C++ in master uses non-UI channels for non-game sounds AFAICT.
20141011 04:33:05< shadowm> So, would you mind testing my next commit with 2 UI channels since I'm kind of busy with other things?
20141011 04:33:15< iceiceice> ok
20141011 04:35:44< irker456> wesnoth: Ignacio R. Morelle wesnoth:master 2295b3b6d013 / data/core/ (help.cfg hotkeys.cfg): wmlindent pass http://git.io/PkPKFw
20141011 04:35:47< irker456> wesnoth: Ignacio R. Morelle wesnoth:master 5d510bd6f3a0 / data/campaigns/Northern_Rebirth/scenarios/01_Breaking_the_Chains.cfg: NR S1: Fix double spaces after sentences http://git.io/glPbWA
20141011 04:35:50< irker456> wesnoth: Ignacio R. Morelle wesnoth:master 569a33b1f595 / data/themes/classic.cfg: Add spelling exception for wmllint http://git.io/dScYXg
20141011 04:35:53< irker456> wesnoth: Ignacio R. Morelle wesnoth:master e39797b6ad61 / changelog src/sound.cpp: sound: Allocate 2 channels for UI sounds and possibly more in the future http://git.io/_7TzXg
20141011 04:36:09< shadowm> iceiceice: ^
20141011 04:36:29-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Ping timeout: 244 seconds]
20141011 04:36:54< shadowm> From now on it's a mater of changing UI_sound_channels, but please try to keep it low -- the reason I increased our overall allocation before was that I was running out of voices in a cutscene in my campaign.
20141011 04:42:23< shadowm> Bah, screw this. I'm going to teach wmllint to parse help markup at a later point.
20141011 04:43:51< shadowm> Oh, text= *is* spellchecked. How do you...
20141011 04:45:43< shadowm> Oh, I see, it specifically ignores help.text, hm.
20141011 04:46:06< shadowm> Well, that's no good. We might have a dozen spell mistakes in help text for all I know.
20141011 04:49:01< fabi_> shadowm: No, please do so.
20141011 04:49:48-!- GallaecioLaptop is now known as Gallaecio
20141011 04:50:26< shadowm> Nice, gna.org ate my +verbatim+ block.
20141011 04:53:40< iceiceice> shadowm: thanks for the patch, it seems to fix everything, after i move the ready to start sound to the turn bell channel
20141011 04:56:38< irker456> wesnoth: Chris Beck wesnoth:master 0ef772257c35 / src/multiplayer_connect.cpp: move the "ready to start" sound to the "turn bell" channel http://git.io/MrWukQ
20141011 04:56:40< irker456> wesnoth: Chris Beck wesnoth:master 84dc376a3625 / changelog: update changelog http://git.io/gpYhzA
20141011 04:57:01< iceiceice> i'm going to backport them also
20141011 04:59:27< shadowm> iceiceice: Oh, I can backport it if you want so you don't have to deal with the changelog mess.
20141011 04:59:39< irker456> wesnoth: Ignacio R. Morelle wesnoth:master 0dadeb37abf9 / data/core/units/orcs/Nightblade.cfg: Remove unused Skirmisher special note from the Orcish Nightblade's description http://git.io/9zZIzQ
20141011 04:59:42< irker456> wesnoth: Ignacio R. Morelle wesnoth:master 57ffe3f27572 / data/core/units/humans/ (Loyalist_Cavalier.cfg Loyalist_Cavalryman.cfg Loyalist_Dragoon.cfg): Silence wmllint warnings about an ability-independent special notes macro http://git.io/gTR2Hw
20141011 04:59:45< irker456> wesnoth: Ignacio R. Morelle wesnoth:master 998f999d9e2d / data/campaigns/Northern_Rebirth/scenarios/01_Breaking_the_Chains.cfg: NR S1: Add spelling exception for 'slavemaster' http://git.io/zUzo5Q
20141011 04:59:48< irker456> wesnoth: Ignacio R. Morelle wesnoth:master 62ae2ff5e314 / data/core/editor/help.cfg: Some spelling/grammar fixes for the editor help text http://git.io/KadFPA
20141011 05:00:45< iceiceice> yeah actually if you could backport your commit that would be good
20141011 05:03:54< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 b21f9d894543 / changelog src/sound.cpp: sound: Allocate 2 channels for UI sounds and possibly more in the future http://git.io/1xcyBQ
20141011 05:03:57< shadowm> iceiceice: ^
20141011 05:07:36< shadowm> (Just tested it, found no regressions in my original [sound] border case.)
20141011 05:10:08-!- Sulfur [~Miranda@p5B009391.dip0.t-ipconnect.de] has joined #wesnoth-dev
20141011 05:16:32< irker456> wesnoth: Chris Beck wesnoth:1.12 a4848b6db82f / src/multiplayer_connect.cpp: move the "ready to start" sound to the "turn bell" channel http://git.io/oGdedQ
20141011 05:16:34< irker456> wesnoth: Chris Beck wesnoth:1.12 8071b9486b37 / changelog: update changelog http://git.io/U_phnA
20141011 05:16:35< iceiceice> yeah i tested again, its all working to my satisfaction
20141011 05:16:37< iceiceice> thanks again
20141011 05:34:30< irker456> wesnoth: Charles Dang website:Vultraz_template_revisions 5e14ce142bb2 / start/1.12/ (images/dl-active.png images/dl.png template.html): Switched from a static image dl button to a css one using the same images http://git.io/zrgt4g
20141011 05:35:38< shadowm> vultraz: Incidentally, you know you could've forked the repository instead of pushing an alternate branch to it?
20141011 05:36:19< vultraz> shadowm: I thought about that, but I figured this would be easier.
20141011 05:37:22< shadowm> Nah, the alternative is just as easy, it just doesn't introduce playground noise to a restricted repo.
20141011 05:37:51< shadowm> (Of course it's too late now.)
20141011 05:42:08< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 ddc29bf7b58a / changelog: campaignd: Changelog entry for refactoring merge for completeness http://git.io/RFzP9g
20141011 05:42:11< irker456> wesnoth: Ignacio R. Morelle wesnoth:master a524452e26dc / changelog: campaignd: Changelog entry for refactoring merge for completeness http://git.io/NWKb0Q
20141011 05:47:24< shadowm> !commit b1715a0c867e835811eb18dd6342cb6a138e06dd
20141011 05:47:25< shikadibot> shadowm: Revision b1715a0c867e (Ignacio R. Morelle) on Sat May 15 22:49:56 2010:
20141011 05:47:29< shikadibot> shadowm: Moving addon_*.?pp files to their own dir, step 1 (move)
20141011 05:47:31< shikadibot> shadowm: Web interface URL: https://github.com/wesnoth/wesnoth/commit/b1715a0c867e
20141011 05:48:17< shadowm> !commit 7abdbceb184463e185115afbc0e67f6dbb66b6f1
20141011 05:48:18< shikadibot> shadowm: Revision 7abdbceb1844 (Daniel Franke) on Sat Feb 28 02:24:31 2009:
20141011 05:48:21< shikadibot> shadowm: Disallow tildes in add-on names, for bug #13080
20141011 05:48:24< shikadibot> shadowm: Web interface URL: https://github.com/wesnoth/wesnoth/commit/7abdbceb1844
20141011 05:49:09< vultraz> blegh. I forgot to y enforce the logo div
20141011 05:49:17< vultraz> s/logo/title
20141011 05:53:21< vultraz> shadowm: what do you think of perhaps putting the "wesnoth 1.12" text in the header, or should it be kept in its title position
20141011 05:53:46< shadowm> mattsc: I've killed https://gna.org/bugs/index.php?22354, which was the only other blocker assigned to me.
20141011 05:53:54< shadowm> I think.
20141011 05:54:19< shadowm> Yeah.
20141011 05:54:27< shadowm> I'm only left with two server-side security issues.
20141011 05:55:28< shadowm> Sadly, I still can't find a way to fix #20477 without changing the campaignd protocol and breaking compatibility with 1.5.x - 1.12.x clients.
20141011 05:56:18< shadowm> vultraz: Header = navbar?
20141011 05:56:56< shadowm> I'd like to not see it in the navbar so as to keep the layout consistent with the rest of everything glamdrol out there.
20141011 05:58:43< shadowm> (Navbar = #header + #nav)
20141011 06:01:26< shadowm> mattsc: Uh, I think we all are forgetting something important.
20141011 06:02:01< shadowm> mattsc: http://forums.wesnoth.org/viewtopic.php?p=573753#p573753 (context in previous page)
20141011 06:02:23< shadowm> Well, I technically volunteered myself for this task, so...
20141011 06:12:01< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 76ba22b2b346 / data/core/terrain.cfg: Revert "Fix bug #21971: visual glitch of terrains on 2p - Hornshark Island" http://git.io/hFDh0w
20141011 06:12:04< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 827a8bc2710f / changelog data/core/terrain.cfg players_changelog: Revert "Make Lit Mushroom Grove stats identical to Mushroom Grove" https://github.com/wesnoth/wesnoth/commit/827a8bc2710f049f8dae54e447b59bca8d742fe9
20141011 06:12:07< irker456> wesnoth: Ignacio R. Morelle wesnoth:1.12 cba8414f5a54 / changelog players_changelog: Changelog entries for the reversal of the fix for bug #21971 http://git.io/2KUuCQ
20141011 06:16:00< irker456> wesnoth: Ignacio R. Morelle wesnoth:master 7b708514c7be / changelog data/core/terrain.cfg players_changelog: Revert "Make Mushroom Grove/Lit Mushroom Grove behave like forests, move-wise" http://git.io/jd098Q
20141011 06:19:36< shadowm> mattsc: Done too now.
20141011 06:31:55< vultraz> dammit, these divs won't positionr ight
20141011 06:31:57< vultraz> position
20141011 06:32:49< vultraz> AH
20141011 06:32:52< vultraz> missing quote
20141011 06:32:54< vultraz> of course
20141011 06:33:09< vultraz> syntax highlighting is so useful
20141011 06:39:42< irker456> wesnoth: Charles Dang website:Vultraz_template_revisions ac3f8a3542c1 / start/1.12/ (images/title.png template.html): Re-rendered the title text par shadowm's suggestions http://git.io/MnDBMw
20141011 06:39:54< vultraz> shadowm: ^ please inspect the button and the new title
20141011 06:40:40< shadowm> The title's awful typesetting hasn't improved.
20141011 06:41:20< shadowm> Something just occurred to me. Why am I straining my eyes to read yellow text on a pale pink/orange background?
20141011 06:42:14< shadowm> Also, why does the button get pressed down when I mouse over it?
20141011 06:42:21< shadowm> Buttons don't do that.
20141011 06:43:06< shadowm> The lead's font size and family combination is still wrong.
20141011 06:50:20-!- Ivanovic [~ivanovic@frnk-4d016be1.pool.mediaWays.net] has quit [Changing host]
20141011 06:50:20-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20141011 06:51:58< Ivanovic> mattsc, shadowm: i can try to do a release every 2 weeks, at least the technical parts
20141011 06:52:26< Ivanovic> it might not be exactly 2 weeks apart (e.g. in 2 weeks at this time i will by flying to san francisco and not around for a week, but i could tag on friday)
20141011 06:53:45< Ivanovic> shadowm: and i really hope to have the release out in november
20141011 06:54:08< Ivanovic> shadowm: because i will not be available for the first three weeks of december, there i will most likely be on vacation and completely offline
20141011 06:55:12< shadowm> If you want I could also try my hand at the technical part of the process, but I would have to generate the tarball for upload from baldras since I can't afford uploading large files from home, and I'm not sure that'd be okay with baldras.
20141011 06:55:57< shadowm> (And generate a separate tarball with identical steps for testing on my own machine.)
20141011 06:56:26< Ivanovic> the general generating part is very easy
20141011 06:56:34< Ivanovic> in general all the parts are easy
20141011 06:56:36< Ivanovic> ;)
20141011 06:57:15-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Quit: Leaving]
20141011 06:57:39< shadowm> Easy, right, but the testing part clearly looks like a tedious chore.
20141011 06:57:48< Ivanovic> shadowm: this is my basic list of things to do for a release: http://pastebin.com/2FwTxnbY
20141011 06:58:37< shadowm> Which is the authoritative list, the pastebin, or the wiki page?
20141011 06:58:40< Ivanovic> and first i run a "configure" and pot-update using this script: http://pastebin.com/umWM338C
20141011 06:58:51< Ivanovic> i think they are both basically in sync
20141011 06:59:05< Ivanovic> but i am using this paste since i don't connect to the wiki for doing the release
20141011 06:59:10< Ivanovic> and the testing:
20141011 06:59:22< Ivanovic> 1) start every campaign in at least one difficulty level
20141011 06:59:28< Ivanovic> 2) open the tutorial
20141011 06:59:51< Ivanovic> 3) open the ingame help and click through some items, (has to include the credits)
20141011 06:59:55< shadowm> Hm, now that I remember, I once had concerns about something in the pot-update procedure yielding different results for me when I needed to test something.
20141011 07:00:03< Ivanovic> 4) connect to the addon server
20141011 07:00:14< Ivanovic> 5) connect to the MP server
20141011 07:00:34< Ivanovic> 6) play at least one campaign scenario (can also be done by AI), i usually use the first scenario of two brothers on easy for this
20141011 07:00:58< Ivanovic> shadowm: if you want to i can run the pot-update today and you try the rest?
20141011 07:01:22< shadowm> Ehhh, this soon?
20141011 07:01:26< Ivanovic> :)
20141011 07:01:37< Ivanovic> sure, directly into the deep part
20141011 07:01:38< shadowm> Maybe I'd rather let you do it and I can do it separately and see if my results match yours.
20141011 07:01:41< Ivanovic> since i am around today
20141011 07:04:25< shadowm> Oh, the script assumes a specific directory layout starting from /src.
20141011 07:05:11< shadowm> Also it says GITBRANCH=master, I assume that's not what you've been using for the last few releases.
20141011 07:07:54< shadowm> I'm getting a *lot* of warnings involving po/CMakeLists.txt.
20141011 07:08:24< shadowm> http://pastebin.com/2WNbCVaX
20141011 07:11:59-!- ancestral [~ancestral@12.23.74.29] has quit [Quit: i go nstuf kthxbai]
20141011 07:13:20-!- EdB [~edb@89-158-11-138.rev.numericable.fr] has joined #wesnoth-dev
20141011 07:14:46< shadowm> > warning: failed to load external entity "/usr/share/asciidoc/docbook-xsl/xhtml.xsl"
20141011 07:14:56< shadowm> > doc/manual/CMakeFiles/update-po4a-manual.dir/build.make:1082: recipe for target '/home/shadowm/src/wesnoth-1.12/doc/manual/manual.en.xml' failed
20141011 07:15:15< shadowm> I can't find a single Debian package that provides the file needed here.
20141011 07:16:00< shadowm> Ivanovic: What things do I need for a successful pot-update, again?
20141011 07:17:53< Ivanovic> docbook-xml-dtd and po4a
20141011 07:18:14-!- ancestral [~ancestral@12.23.74.29] has joined #wesnoth-dev
20141011 07:19:11< shadowm> Ah, I think I see what's going on here. The cmake recipe makes an assumption about the file's location that doesn't hold true on my system.
20141011 07:19:38< shadowm> Probably. It wants /usr/share/asciidoc/docbook-xsl/xhtml.xsl, but I only have /etc/asciidoc/docbook-xsl/xhtml.xsl.
20141011 07:20:35< Ivanovic> right, i do have a checkout of 1.12
20141011 07:20:49< Ivanovic> so there are some slight adjustments to be done for a stable release
20141011 07:21:04< Ivanovic> and when running a pot-update you have to watch out for the manual.*.html files
20141011 07:21:18< Ivanovic> if someone screwed up the syntax in translations from there on all other files will fail
20141011 07:21:22< shadowm> Seems to be working after replacing /usr/share with /etc above.
20141011 07:21:27< Ivanovic> so if you have a broken one, you need to fix that
20141011 07:21:58< shadowm> Ivanovic: How would I know if one of the files is bad?
20141011 07:22:00< Ivanovic> usually it is a case of the pipe character being replaced bz a capital i or small l or the other way around....
20141011 07:22:10< Ivanovic> it breaks at creation of the html file with an error
20141011 07:22:13< shadowm> I mean, shouldn't that cause the whole process to abort?
20141011 07:22:37< shadowm> Your average makefile requires all *.c files involved to compile, for example.
20141011 07:22:43< Ivanovic> back in 30, got to have some breakfast
20141011 07:23:34-!- zookeeper [zookeeper@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20141011 07:24:06< shadowm> ../../doc/man/wesnoth.6:54: (po4a::man) Unknown '<' or '>' sequence. Faulty message: \fB\-\-campaign\-difficulty [E<lt>obtížnost<gt>]\fP
20141011 07:25:41< shadowm> The update-po4a-manual target's log: http://pastebin.com/dkLn1atA
20141011 07:26:36< shadowm> There are a few error messages around the .html files generation.
20141011 07:27:09< shadowm> Uh, and for the main pot-update step:
20141011 07:27:11< shadowm> [ 27%] mo-update [wesnoth-fr]: Creating mo file.
20141011 07:27:11< shadowm> fr.po:7567: 'msgid' and 'msgstr' entries do not both begin with '\n'
20141011 07:27:47-!- Guest2509 [matthiaskr@I.Eat.Babies.PanicBNC.net] has quit [Changing host]
20141011 07:27:47-!- Guest2509 [matthiaskr@unaffiliated/matthiaskrgr] has joined #wesnoth-dev
20141011 07:27:57-!- Guest2509 is now known as matthiaskrgr
20141011 07:28:41< shadowm> I'm not sure what's going on there besides the msgstr having newlines that the msgid doesn't.
20141011 07:30:02< shadowm> Wow, why is most of the pot-update output such a jumbled mess?
20141011 07:32:28< shadowm> Most of the output consists of `#define GETTEXT_DOMAIN "foo"`, there are several wmlxgettext errors involving the test scenario and a GUI2 dialog, and although there aren't any errors or program output, the part where it updates po files is unreadable with lines clobbering each other.
20141011 07:32:53< shadowm> With dots everywhere.
20141011 07:33:16< shadowm> I hope it doesn't usually look like this for you. <.<
20141011 07:34:20< shadowm> OK, so the msgstr at po/wesnoth/fr.po:7567 is clearly wrong, but why does this abort the whole process? I didn't think msgfmt kept an eye on how much punctuation the msgstr has compared to its msgid.
20141011 07:35:09 * shadowm is clearly not cut out for this. -.-
20141011 07:36:44< shadowm> Ivanovic: Anyway, to be honest, I cannot really guarantee I'll be around today without having planned for this. I'd still like to learn how to do it so we can have a backup plan in the future if a release is urgently needed (think one of those delicious security bugs I've fixed before) and you aren't available.
20141011 07:37:15< shadowm> (Or if we need more than 1 RC and you wind up unable to commit to the 2 weeks schedule.)
20141011 07:39:34< shadowm> (Feel free to replace "security bugs" with "too many important things are broken and nobody noticed in time" like it happened with 1.11.3 and 1.11.11.)
20141011 07:42:35-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20141011 07:42:56-!- molgrum [~molgrum@212.85.89.43] has joined #wesnoth-dev
20141011 07:52:29< Ivanovic> shadowm: okay, to explain things:
20141011 07:52:46< Ivanovic> 1) the manual error you see is one of the cases where someone used the wrong characters
20141011 07:53:09< Ivanovic> 2) yes, this is a problem which should not happen if the updates are done as normal, since it should fail already before the upload
20141011 07:53:35< Ivanovic> in those cases ('2') i just "fix" the po files by marking the respective string as fuzzy
20141011 07:53:48< Ivanovic> in case of '1' I tend to just fix the broken markup
20141011 07:54:37< Ivanovic> that is: you can also release without doing the pot-update
20141011 07:54:42< Ivanovic> at least in really urgent cases
20141011 07:54:59< Ivanovic> let me run a pot-update to check what happens
20141011 08:01:41< Ivanovic> AI0867, boucman, Coffee_irc, elias, fabi_, iceiceice, loonycyborg, shadowm, Soliton, thunderstruck, timotei, vultraz, zookeeper, everyone else who cares: ready for 1.12 RC1?
20141011 08:02:03< iceiceice> yes
20141011 08:02:05< Ivanovic> shadowm: ehm, the logfile you pasted for me looks fine
20141011 08:02:50< Coffee_irc> yes (although I haven't been that active of late)
20141011 08:03:07< Ivanovic> Coffee_irc: neither have I, so no idea who was active
20141011 08:03:09< Ivanovic> ;)
20141011 08:05:15< iceiceice> gfgtdf should be on the above list ;)
20141011 08:05:31< Ivanovic> iceiceice: he is not online in irc right now
20141011 08:05:36< iceiceice> that's true
20141011 08:05:38< Ivanovic> otherwise mattsc would be on the list, too
20141011 08:06:34< Ivanovic> but right now i need to fix my system since i can't run the po update...
20141011 08:06:44< Coffee_irc> I'd like to get a tentacle of the deep animation in, but it can wait for the next point release
20141011 08:06:47< Ivanovic> (perl woes)
20141011 08:07:37< zookeeper> eh, i guess i could add a few small changelog entries...
20141011 08:08:12< Ivanovic> zookeeper: yes, you should
20141011 08:08:19< Ivanovic> ideally directly when commiting the changes
20141011 08:08:20< Ivanovic> ;)
20141011 08:08:31< zookeeper> if github for windows wasn't being stupid, and allowed me to add the 1.12 clone i have to it
20141011 08:08:50< Ivanovic> zookeeper: have you considered using sourcetree on windows?
20141011 08:09:01< Ivanovic> i use that tool at work for working with git and it seems to be okayish
20141011 08:09:13< zookeeper> i did, but the trial expired before i had a chance to try it even once :p
20141011 08:09:23< Ivanovic> trail?
20141011 08:09:28< Ivanovic> this is not a free version?
20141011 08:09:29< Ivanovic> wow...
20141011 08:09:44< zookeeper> i have no idea, i just had a 30 day trial or whatever
20141011 08:09:50< Coffee_irc> lol
20141011 08:10:33< Ivanovic> http://blog.sourcetreeapp.com/2014/08/19/sourcetree-for-windows-1-6-now-available/
20141011 08:10:36< Ivanovic> should be freeware
20141011 08:11:07< zookeeper> hmh, right
20141011 08:11:10< zookeeper> maybe i'll try it
20141011 08:11:36< Coffee_irc> that's the way to do it though, find a free version of something for one system and package it as a trial (only $30) with a flashy GUI for another system
20141011 08:11:39< Ivanovic> Coffee_irc: since we are using some atlassian product at work it might have been that we get a full version based on the IP range we are in
20141011 08:12:01< Coffee_irc> I wouldn't have to work again :P
20141011 08:17:50< irker456> wesnoth: ln-zookeeper wesnoth:1.12 0c4a93fa5608 / changelog: Updated changelog http://git.io/RXMMfg
20141011 08:18:08< shadowm> The changelog needed line-wrapping IIRC.
20141011 08:19:27< irker456> wesnoth: ln-zookeeper wesnoth:master ec08205c5e18 / changelog: Updated changelog http://git.io/6ioLkg
20141011 08:19:28< zookeeper> there we go, someone had already changelogged the important stuff
20141011 08:19:43< shadowm> Ivanovic: The log has things like "ERROR: xref linking to 戰役 has no generated link text." in it during the manual build stage, are you sure that's fine?
20141011 08:19:55< irker456> wesnoth: Nils Kneuper wesnoth:1.12 fea45838f802 / po/wesnoth-manpages/cs.po: fixed syntax error in Czech manpages http://git.io/3QMrXg
20141011 08:20:14< Ivanovic> shadowm: i know that i told the chinese guys several times to fix this and they ignored it
20141011 08:20:21< Ivanovic> this is just a broken link inside the html page
20141011 08:22:26< shadowm> Who updated RELEASE_NOTES with the mushroom terrain stuff before I even committed the revert?
20141011 08:22:35< shadowm> :\
20141011 08:23:15< shadowm> iceiceice did it in commit dcba73e0d83299adbc261a3297aad76a73e65c52, hm.
20141011 08:23:36< iceiceice> i think i was confused about whether it had happened on 1.12 or master
20141011 08:24:05< shadowm> Yeah, it hadn't happened in either branch at that point yet, I did it just a few hours ago.
20141011 08:24:08< iceiceice> apparently it didn't happen on either though...
20141011 08:24:10< iceiceice> yeah i'm not sure
20141011 08:24:18< shadowm> I had already agreed to revert it, though.
20141011 08:24:22< iceiceice> for some reason i thought it had happened
20141011 08:25:17 * shadowm got worried for a moment he had reverted a previous revert.
20141011 08:25:32< shadowm> Either that, or unconsciously traveled in time.
20141011 08:26:05< iceiceice> yeah if theres no evidence in bug tracker or forum post then i think i just jumped the gun
20141011 08:26:13< iceiceice> and i dont see any such evidence
20141011 08:41:33< irker456> wesnoth: Nils Kneuper wesnoth:1.12 fc26d99823d8 / po/wesnoth-manpages/cs.po: fixed even more syntax issues in the Czech manpages http://git.io/kcDPlw
20141011 08:41:52< shadowm> iceiceice: #22609 looks like a thing I'd like to try implementing in the future (along with debug command autocomplete), but I'm not taking it in case I later forget about it/decide it's beyond my skill level.
20141011 08:42:32< shadowm> I mean, in theory it should be easy, but that's what I said as well back when I decided to <insert nightmarish task here>.
20141011 08:44:07< shadowm> (OK, concrete example: embedding a GUI1 combobox in the add-ons manager dialog. Didn't pan out because I couldn't figure out how to get past an interclass dependency issue.)
20141011 08:44:09< irker456> wesnoth: Nils Kneuper wesnoth:master 9a1d5a6b9cda / po/wesnoth-manpages/cs.po: fixed syntax errors in Czech manpages in master, too http://git.io/6tIfaA
20141011 08:44:26< iceiceice> yeah so i've been messing around with lua api recently, i think i know how to do this... but i dont have time to try for the foreseeable future
20141011 08:45:44< shadowm> Another thing I'd like to look into at some later point is defining ConditionalWML (think [if] conditions) in Lua. I've found myself wanting a [if][this_convoluted_recurrent_pattern_yields_a_positive_result] tag way too often.
20141011 08:46:05-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has joined #wesnoth-dev
20141011 08:49:03-!- mjs-de [~mjs-de@f048239245.adsl.alicedsl.de] has joined #wesnoth-dev
20141011 08:49:53< Ivanovic> shadowm: okay, pot-update is done
20141011 08:50:07< Ivanovic> shadowm: you were right with those manpages failing, but the french po files were fine over here
20141011 08:50:38< shadowm> Maybe my version of msgfmt (0.19.2) is more paranoid/strict somehow?
20141011 08:50:48< irker456> wesnoth: Nils Kneuper wesnoth:1.12 f50ca190cdd0 / / (854 files in 30 dirs): pot-update and regenerated doc files http://git.io/kcGf1g
20141011 08:51:02< Ivanovic> shadowm: it should be the same for the string you mentioned
20141011 08:51:08< Ivanovic> since this one is fatal to gettext
20141011 08:51:37< Ivanovic> can you do a pull of the latest version and try to compile the game with internationalization active?
20141011 08:51:47< shadowm> Eh, yes, the string I mentioned is from po/wesnoth/fr.po.
20141011 08:51:51< Ivanovic> if this error exists in the files i uploaded, it should break for you at compiling the po files
20141011 08:52:12< Ivanovic> i hand checked the file prior running the pot-update and there the string did not exist in the uncommented area
20141011 08:52:55< shadowm> Huh.
20141011 08:53:48< Ivanovic> meaning: there was no string where the number of \n at beginning/end of the strings did not match
20141011 08:54:05< shadowm> Compiling the game (scons nls=true) works.
20141011 08:54:33< shadowm> And manually invoking msgfmt  on that file does work for me too. (I did try doing it by hand before, it gave the same error.)
20141011 08:54:52< Ivanovic> hmm, that is a strange one
20141011 08:54:57< shadowm> So I guess somehow the extraneous \ns were introducd during my pot-update.
20141011 08:59:46< shadowm> It did it again during my new attempt to pot-update.
20141011 09:00:48< shadowm> Oh, I've also realized I'm experiencing the same issue last time I tried to pot-update: po/wesnoth-ai/POTFILES.in gets filled with files from every other textdomain.
20141011 09:01:28< shadowm> I should've looked at the po diff first. So I'm unable to pot-update for some reason. :\
20141011 09:02:04< Ivanovic> :(
20141011 09:02:26< Ivanovic> honestly, i have no idea why it might break
20141011 09:02:32< Ivanovic> for me it just works (tm)
20141011 09:03:27< shadowm> Yeah, just like doc/manual/CMakeLists.txt has a hardcoded path that works for you but not for me (and other Debian users). Perhaps there's a similarly hardcoded thing in po/CMakeLists.txt.
20141011 09:03:42< Ivanovic> could be
20141011 09:03:59< Ivanovic> though those changes surprise me somehow, considering that mordante, a debian user, created the scripts
20141011 09:04:18< Ivanovic> so i would guess it works for debian and i would have experienced problems on gentoo
20141011 09:04:20< Ivanovic> ;)
20141011 09:04:35< shadowm> Alternatively it's my asciidoc version that changed the file location.
20141011 09:07:25< shadowm> I see references to POTFILES.in in po/CMakeList.txt but nothing that generates it, so I guess I'm missing something obvious and cmakey.
20141011 09:11:49-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20141011 09:11:52< shadowm> I can't say scons pot-update works, because even though it runs to completion, it seems to shuffle strings around within po files for no obvious reason.
20141011 09:12:35< shadowm> Still better than corrupting the whole textdomain hierarchy.
20141011 09:14:23< shadowm> Anyway, I'm leaving now, later.
20141011 09:15:40< irker456> wesnoth: Nils Kneuper website:master 45034fe634b2 / start/1.12/ (44 files in 2 dirs): ran make to update po files and generate html files http://git.io/LV1Ljg
20141011 09:15:59< shadowm> Oh yeah, that.
20141011 09:16:04< shadowm> The template isn't ready yet. :p
20141011 09:16:42< Ivanovic> shadowm: do you know which script handles the git updates on the new server?
20141011 09:16:54< Ivanovic> shadowm: i got the impression that the website repo is not synced / checked out
20141011 09:17:27< shadowm> That's a thing I have to do by hand unless the gettext.w.o update cronjob beats me to it.
20141011 09:17:32-!- ancestral [~ancestral@12.23.74.29] has quit [Quit: i go nstuf kthxbai]
20141011 09:17:39< Ivanovic> i think in the past those updates were handled with the g.w.o update script
20141011 09:17:48< shadowm> Yeah, and that's not the sanest thing to do.
20141011 09:17:53< Ivanovic> but looking at the latest script it seems as if this one only updates the gettext.wesnoth.org subdir
20141011 09:18:06< shadowm> Does it? Hm.
20141011 09:18:13< Ivanovic> #update our codebase
20141011 09:18:15< Ivanovic> cd $WEBSITE_PATH/gettext.wesnoth.org
20141011 09:18:16< Ivanovic> git pull -q
20141011 09:18:31< Ivanovic> let me remove the gettext.wesnoth.org part there to update all...
20141011 09:18:34< shadowm> It's part of the same repository, you can git pull from any subdir and it'll work.
20141011 09:18:57< shadowm> All that said, surely you aren't trying to read http://wesnoth.org/start/1.12/template.html ?
20141011 09:19:21< irker456> wesnoth: Nils Kneuper website:master a8bde8264be0 / gettext.wesnoth.org/bin/update-gettext-stats: add the whole website repo to the regular update cycle http://git.io/n3VbgA
20141011 09:19:26< shadowm> Because the start dir is not a symlink to the repository itself. Its subdirs are, instead, and there's no symlink for 1.12.
20141011 09:19:45< Ivanovic> ahhh, okay, that is also a good reason...
20141011 09:19:59< Ivanovic> i thought the start dir was sym linked to
20141011 09:20:07< Ivanovic> can you add the 1.12 link?
20141011 09:20:08< shadowm> See, that commit isn't needed and will in fact break things.
20141011 09:20:08< Ivanovic> thanks!
20141011 09:20:30< shadowm> As the current working dir doesn't have a grab-stats.php to run from line 38 that way.
20141011 09:20:41-!- ancestral [~ancestral@12.23.74.29] has joined #wesnoth-dev
20141011 09:20:49< irker456> wesnoth: Nils Kneuper website:master 3bb5f99f847f / gettext.wesnoth.org/bin/update-gettext-stats: get rid of last commit, should not be required http://git.io/CJ1NDA
20141011 09:20:57< Ivanovic> okay, reverted
20141011 09:21:05< shadowm> http://www.wesnoth.org/start/1.12/template.html
20141011 09:21:19< Ivanovic> perfect!
20141011 09:21:21< shadowm> I can't wait to see how many lemmings will drop into that hole.
20141011 09:21:39< shadowm> "why are there so many TODOs this is clearly a bug with the website"
20141011 09:21:47< Ivanovic> i also ran an update on the files, so the po files already match this first version
20141011 09:23:09< shadowm> (Why I am against this whole autopull thing that gettext.w.o likes so much: get a developer's computer hijacked and then upload malware to the website without even needing to get past me and into the web ct's shell itself.)
20141011 09:24:13< shadowm> (Or find some exploit to trick GitHub into giving you push access when you aren't supposed to have such.)
20141011 09:30:07< Ivanovic> okay, working on tagging 1.11.17 now
20141011 09:30:44< irker456> wesnoth: Nils Kneuper wesnoth:1.12 d16a32d38e3f / Doxyfile changelog players_changelog src/wesconfig.h: bump version to 1.11.17 http://git.io/cG10WQ
20141011 09:33:13-!- Ivanovic changed the topic of #wesnoth-dev to: releasing 1.12 RC1 (aka 1.11.17), using d16a32d38e3f | string+feature freeze active on 1.12 | 256 bugs, 338 feature requests, 29 patches | Logs: http://irclogs.wesnoth.org | Alternate logs: http://wesnoth.debian.net | Don't paste on IRC! Use a pastebin: http://pastebin.com | http://imagebin.org
20141011 09:37:29< shadowm> I just realized there's a misplaced changelog entry in the 1.12 changelog.
20141011 09:38:32< shadowm> Two, in fact.
20141011 09:38:51< Ivanovic> bad ones or "ignore them"?
20141011 09:39:22< shadowm> They can always be fixed after the fact, they just will be missing (more accurately, found under the 1.11.16 section instead of 1.11.17) from the tagged version.
20141011 09:40:35< shadowm> I do have the fix ready, may I commit it now or should I wait?
20141011 09:48:34< Ivanovic> please wait, i already created the tarball and am about to test it
20141011 09:50:11< Ivanovic> the upload will take me about one hour over here (since at home with my parents right now, not in my frankfurt flat)
20141011 09:51:48< Ivanovic> shadowm: btw i use the -test versions especially for the userdata folder to more easily test if things work nicely with a fresh config
20141011 09:51:58< Ivanovic> in fact i tend to not test with old configs but just with plain and fresh ones
20141011 09:52:15< shadowm> "-test" versions?
20141011 09:52:16< Ivanovic> and another test which makes sense to do is create a save and load it
20141011 09:52:36< Ivanovic> in my command list you will see that cmake is configured to store userdata into ~/.wesnoth-test
20141011 09:52:38< shadowm> I have a script that runs Wesnoth with a little temporary sandbox user config and data dir.
20141011 09:52:43< Ivanovic> instead of the shared config folder
20141011 09:52:56< shadowm> i.e. without requiring compile-time configuration.
20141011 09:53:49< shadowm> I committed it a while ago as utils/wesnoth-defaults, in fact, but I produced it long in advance so aid me with UI testing.
20141011 09:54:00< shadowm> *to
20141011 10:01:03< irker456> wesnoth: ivanovic wesnoth: d16a32d38e3f tagged as 1.11.17
20141011 10:01:17-!- Ivanovic changed the topic of #wesnoth-dev to: tagged 1.12 RC1 (aka 1.11.17), announcing "soon" | string+feature freeze active on 1.12 | 256 bugs, 338 feature requests, 29 patches | Logs: http://irclogs.wesnoth.org | Alternate logs: http://wesnoth.debian.net | Don't paste on IRC! Use a pastebin: http://pastebin.com | http://imagebin.org
20141011 10:01:32< Ivanovic> loonycyborg, Rhonda: 1.11.17 aka 1.12.0 RC1 tagged
20141011 10:01:45< Ivanovic> currently uploading the tarball to wesnoth.org (ETA: 50min)
20141011 10:03:57< shadowm> Your instructions in that pastebin for tagging are slightly off.
20141011 10:04:06< Ivanovic> they are?
20141011 10:04:22< Ivanovic> i just go to the 1.12 checkout folder and run exactly those 2 commands
20141011 10:04:23< shadowm> Oh wait, -m implies -a, nvm.
20141011 10:22:32-!- Ayne [~Ayne@55d41d5a.access.ecotel.net] has joined #wesnoth-dev
20141011 10:26:01-!- travis-ci [~travis-ci@ec2-54-87-134-110.compute-1.amazonaws.com] has joined #wesnoth-dev
20141011 10:26:01< travis-ci> wesnoth/wesnoth#4186 (1.11.17 - d16a32d : Nils Kneuper): The build passed.
20141011 10:26:01< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/37673971
20141011 10:26:01-!- travis-ci [~travis-ci@ec2-54-87-134-110.compute-1.amazonaws.com] has left #wesnoth-dev []
20141011 10:31:57-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Ping timeout: 244 seconds]
20141011 10:39:06-!- ancestral [~ancestral@12.23.74.29] has quit [Quit: i go nstuf kthxbai]
20141011 10:42:00-!- Anakonda [Anakonda@87-92-223-209.bb.dnainternet.fi] has joined #wesnoth-dev
20141011 10:43:52-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 240 seconds]
20141011 10:45:34-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20141011 10:58:56< Ivanovic> shadowm, mattsc: the "technical" part of the release is done now
20141011 10:59:00< Ivanovic> what is still missing is:
20141011 10:59:05< Ivanovic> 1) announcement forum post
20141011 10:59:15< Ivanovic> 2) update of wiki download pages ("normal" and xdelta page)
20141011 10:59:36< Ivanovic> 3) update frontpage
20141011 10:59:51< Ivanovic> i already updated the credits with the latest version and am currently uploading the pandora package for RC1
20141011 11:01:28< irker456> wesnoth: Nils Kneuper wesnoth:1.12 5bac74360bbd / Doxyfile changelog players_changelog src/wesconfig.h: post release version bump to 1.11.17+dev http://git.io/HTK_lw
20141011 11:05:46< Ivanovic> and of course the 1.12 announcement needs some work
20141011 11:05:59< Ivanovic> here i ran the first creation of files and it can now be viewed on the website, too
20141011 11:06:20< Ivanovic> template: http://www.wesnoth.org/start/1.12/template.html
20141011 11:06:47< Ivanovic> "real" page including translations: http://www.wesnoth.org/start/1.12/
20141011 11:07:02< Ivanovic> to be more precise the english version: http://www.wesnoth.org/start/1.12/index.en.html
20141011 11:07:23< Ivanovic> and yes, people should *NOT* update this index.en.html file but the template file
20141011 11:07:30< Ivanovic> the other files are generated!
20141011 11:09:11-!- Sulfur [~Miranda@p5B009391.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20141011 11:14:01-!- Guest63353 is now known as Haudegen
20141011 11:34:54-!- prkc [~prkc@catv-80-99-60-5.catv.broadband.hu] has joined #wesnoth-dev
20141011 11:39:22-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Ping timeout: 240 seconds]
20141011 11:56:38-!- trewe [~trewe@188.251.214.137] has joined #wesnoth-dev
20141011 12:16:54< vultraz> Ivanovic: I'm working on an updated version of the template
20141011 12:20:06< vultraz> shadowm: I don't know what the problem with the rendering is
20141011 12:20:13< vultraz> I followed the instructions
20141011 12:21:36< vultraz> zookeeper: do you have a tilable version of the wooden BG?
20141011 12:22:47-!- Gallaecio [~quassel@105.153.165.83.dynamic.mundo-r.com] has quit [Remote host closed the connection]
20141011 12:24:21< vultraz> shadowm: Reducing the intro's size to default makes it look, like, well, not an intro. suggestions?
20141011 12:32:05< vultraz> ...why do the letters look so...flat
20141011 12:40:45-!- Ayne [~Ayne@55d41d5a.access.ecotel.net] has quit [Ping timeout: 246 seconds]
20141011 13:11:01-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20141011 13:19:03< vultraz> hey mattsc 
20141011 13:23:31-!- Ayne [~Ayne@55d41d5a.access.ecotel.net] has joined #wesnoth-dev
20141011 13:24:10< mattsc> hi vultraz 
20141011 13:24:31< vultraz> Have any suggestions while I'm working on the announcement template update?
20141011 13:26:19< mattsc> vultraz: no, not really; I’m not very good with those layout sort of things, I didn’t even notice lots of the things mentioned yesterday  :|
20141011 13:27:12< mattsc> Ivanovic: great, thanks!  I just got up and have a lot of catching up to do, it appears
20141011 13:28:01< vultraz> Alright
20141011 13:28:14< vultraz> It's really frustrating me >_> I'm not very good with elegance
20141011 13:28:43< mattsc> vultraz: I’m afraid I’m no help with that, sorry
20141011 13:29:27< mattsc> Ivanovic, shadowm: I’ll work on the announcement forum post; as usual though, I have a lot of other things going on on weekends, so I cannot tell yet when today I will get it done, other than that it will be today (my timezone)
20141011 13:29:38< mattsc> I’ll also do the OS X package on the side …
20141011 14:01:41-!- irker456 [~irker@fehu.ai0867.net] has quit [Quit: transmission timeout]
20141011 14:04:43-!- gfgtdf [~chatzilla@f054159222.adsl.alicedsl.de] has joined #wesnoth-dev
20141011 14:16:42< Ayne> thunderstruck, are you around?
20141011 14:24:13-!- irker562 [~irker@fehu.ai0867.net] has joined #wesnoth-dev
20141011 14:24:13< irker562> wesnoth: Charles Dang website:Vultraz_template_revisions 4b06609294ba / start/1.12/template.html: Some progress http://git.io/QYtQGQ
20141011 14:33:40< irker562> wesnoth: Charles Dang website:Vultraz_template_revisions f3c19056cd52 / start/1.12/ (styles/intl.css template.html): Moved the .dlbutton class to the external stylesheet http://git.io/l2QZjw
20141011 14:36:28< zookeeper> vultraz, nope, it's not tilable due to perspective
20141011 14:36:39< vultraz> Alright
20141011 14:36:45< zookeeper> tilable? tileable? dunno which, i guess the latter
20141011 14:37:31< zookeeper> what would you have wanted it for?
20141011 14:37:53< vultraz> considering it as either a page bg or a replacement for the stonework header
20141011 14:38:06< vultraz> But as a bg it would be a bit unwieldy because of size
20141011 14:39:17< zookeeper> yeah, the stonework is better in the header, and the wood bg can't really be adapted for webpage bg use
20141011 14:41:34< vultraz> I just can't figure out how to rectify the stonework with my header
20141011 14:41:44< vultraz> I could just not use the header, but it looks nice ._. /
20141011 14:45:23< vultraz> zookeeper: do you perhaps know why GIMP seems to be rendering the wesnoth font...short?
20141011 14:45:38< vultraz> compare what I have on my template to the mm text, and the latter is taller
20141011 14:45:40< vultraz> I'm not sure why.
20141011 14:47:25< zookeeper> no idea
20141011 14:52:53< vultraz> Is it short for you?
20141011 14:53:59< gfgtdf> does someone know why wesnoth has commandline parmaeters (--gunzip, --gzip) that turns it into a very simple zip/unzip programm ?
20141011 14:56:56< vultraz> addon uploads?
20141011 14:58:02< zookeeper> i don't want to fire up GIMP, then start interrogating you about what the "wesnoth font" is and looking into whether i have it (probably not) and where i can get it etc etc just to check that
20141011 15:00:22< thunderstruck> Ayne: Yes.
20141011 15:00:49< vultraz> The wiki said Celtic Garamond the 2nd, which is what I used. If that's wrong, then that's the problem right there
20141011 15:05:53< Ayne> thunderstruck: I've been looking into the issue with the multiplayer saves. What puzzles me is that singleplayer sos saves seems to load just fine despite containing the same data. Any idea why that might be? Do you have any insights from when you tried to fix it before that you can share?
20141011 15:08:12< thunderstruck> Ayne: SP and MP uses different code paths to load these sos saves. First of all, I think that MP codepath tries to access the wrong part of the save file.
20141011 15:08:26< thunderstruck> Take a look to my fix attempt which is linked in that gna ticket.
20141011 15:09:30< thunderstruck> I managed to change things so that the right part of save file is used. But then some things were still missing and if I remember correctly, sides in the game were messed up.
20141011 15:09:45< thunderstruck> But the sos saves were loading in MP.
20141011 15:14:43< Ayne> I figured as much, I just can't seem to find the relevant differences. The code has changed a lot from when I was working on it two years ago, so I'm a bit lost ;)
20141011 15:22:34< mattsc> Ayne: btw, thank you very much for “coming back” and looking into this!
20141011 15:27:09< Ayne> mattsc: You're welcome, though I'm not sure I'm the "expert" you were looking for ;)
20141011 15:27:56< mattsc> Ayne: that’s okay, you’re more expert than most or all of the rest of us.
20141011 15:28:43< thunderstruck> Ayne: I have little understanding of that area myself. I have something else to finish now so I can't start debugging wesnoth's code. However, in a couple of hours I'll try fire up the debugger to check what's going on there.
20141011 15:31:29< thunderstruck> Ayne: Just looking at my fix attempt I'm thinking that maybe I was supposed to update game state too..
20141011 15:31:40< thunderstruck> Maybe that's causing problems in my fix.
20141011 15:32:21< Ayne> thunderstruck: That's alright, I just thought you might know a little more since you've had a look at it before
20141011 15:33:16< Ayne> hm, could be
20141011 15:42:27< mattsc> shadowm, everybody else who cares: I have produced a first very very rough and messy version of the 1.11.17 release announcement:  http://pastebin.com/G9Diadrz
20141011 15:42:54< mattsc> I apologize for sending this around in this state, but I need to do other things for several hours now.
20141011 15:43:10< mattsc> Obviously, there is lots of work to be done, the formatting is all messed up etc.
20141011 15:44:00< mattsc> The only thing I would like people to look at (if you have time and care) is whether the list of bugs fixed looks correct and whether anything should be pulled out of the “misc.” section and given its own section.
20141011 15:44:18< mattsc> Of course, if anybody actually wants to work on the wording, I won’t complain either ...
20141011 15:45:38< mattsc> Again, sorry for this mess, I just have not had much time for this yet, and might not be able to get back to it until late afternoon/evening in my timezone (UTC-7).
20141011 16:01:29-!- iceiceice [~chris@192.16.204.74] has joined #wesnoth-dev
20141011 16:01:29-!- iceiceice [~chris@192.16.204.74] has quit [Changing host]
20141011 16:01:29-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20141011 16:16:52-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20141011 16:42:37< gfgtdf> iceiceice: is there anything i have to watch out for when i try to fixup a merge commit by rebasing ?
20141011 16:43:20< iceiceice> i dont know, i never did that
20141011 16:43:56< iceiceice> in the past if i had to redo a merge i would just do reset --hard HEAD^ and try again
20141011 16:45:42< iceiceice> there's apparently some option --preserve-merges for git rebase?
20141011 16:45:43< iceiceice> http://marketblog.envato.com/general/rebasing-merge-commits-in-git/
20141011 16:45:52< iceiceice> it sounds like it will actually just drop your merge commit if you try to rebase it
20141011 16:53:45-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has joined #wesnoth-dev
20141011 16:56:36-!- cib0 [~cib@p508BCA6D.dip0.t-ipconnect.de] has joined #wesnoth-dev
20141011 17:03:22< gfgtdf> iceiceice: ok i make a merge from boostfs into master teht compiles at lest on my local machine https://github.com/gfgtdf/wesnoth-old/tree/AI-boost_filesystem_2
20141011 17:04:25< iceiceice> ok i'll try to build
20141011 17:07:45-!- travis-ci [~travis-ci@ec2-54-196-148-65.compute-1.amazonaws.com] has joined #wesnoth-dev
20141011 17:07:45< travis-ci> gfgtdf/wesnoth-old#262 (AI-boost_filesystem_2 - ba20b37 : gfgtdf): The build failed.
20141011 17:07:45< travis-ci> Build details : http://travis-ci.org/gfgtdf/wesnoth-old/builds/37691398
20141011 17:07:45-!- travis-ci [~travis-ci@ec2-54-196-148-65.compute-1.amazonaws.com] has left #wesnoth-dev []
20141011 17:11:44< iceiceice> hmm ok
20141011 17:11:48< iceiceice> it builds and seems to run fine
20141011 17:11:52< iceiceice> the only thing thats a bit wierd,
20141011 17:12:07< iceiceice> now i'm getting an error in an addon, from it trying to include a file with . in front of its name that i was hiding
20141011 17:13:10< gfgtdf> hmm can you give me a stacktrace or a way to reproduce it ? 
20141011 17:13:56< gfgtdf> iceiceice: did you upload the addon to the server ?
20141011 17:14:10< iceiceice> no, its just some test thing
20141011 17:14:34< iceiceice> if i remove that add-on i dont get any more erors
20141011 17:14:54< iceiceice> i can upload
20141011 17:19:05< iceiceice> gfgtdf: https://drive.google.com/file/d/0B54Q-3UjP5fieW9SMmFMNkpzUGs/view?usp=sharing
20141011 17:19:21< iceiceice> thats actually 2 addons, one uses functions from the other
20141011 17:19:21-!- ancestral [~ancestral@12.23.74.29] has joined #wesnoth-dev
20141011 17:22:21< gfgtdf> iceiceice: ok i think i found
20141011 17:24:42< gfgtdf> it checks for "./" at teh beginning but teh new implementation is a bit lazy and onyl checks for '.'
20141011 17:24:44< gfgtdf> iceiceice: ^
20141011 17:25:22< iceiceice> ok i will test a bit
20141011 17:25:34< gfgtdf> iceiceice: https://github.com/gfgtdf/wesnoth-old/blob/AI-boost_filesystem_2/src/filesystem_boost.cpp#L902
20141011 17:33:43-!- irker562 [~irker@fehu.ai0867.net] has quit [Quit: transmission timeout]
20141011 17:35:24-!- travis-ci [~travis-ci@ec2-54-196-148-65.compute-1.amazonaws.com] has joined #wesnoth-dev
20141011 17:35:24< travis-ci> gfgtdf/wesnoth-old#263 (AI-boost_filesystem_2 - 1be271f : gfgtdf): The build failed.
20141011 17:35:24< travis-ci> Build details : http://travis-ci.org/gfgtdf/wesnoth-old/builds/37692849
20141011 17:35:24-!- travis-ci [~travis-ci@ec2-54-196-148-65.compute-1.amazonaws.com] has left #wesnoth-dev []
20141011 17:37:35< gfgtdf> iceiceice: actualy i think what i just said migth be wrong
20141011 17:38:36< gfgtdf> iceiceice: i just tried yur 2 campaign and with the lua test  didnt notices somethign liek that and the enlevel i couldnt play becaue it didn't find the [scenario] with id=endlevel_test_scen
20141011 17:39:35< loonycyborg> Ivanovic: I've uploaded 1.11.7 windows release
20141011 17:40:54< iceiceice> hmmm
20141011 17:41:08< iceiceice> maybe i my test scenarios are linked more than i thought :/
20141011 17:41:09< iceiceice> one sec
20141011 17:43:06< iceiceice> ok, when i load with only those two addons in the folder, i get just this error:
20141011 17:43:07< iceiceice> http://hastebin.com/qivogesobo.vhdl
20141011 17:43:16< iceiceice> it comes up during config caching
20141011 17:43:30< iceiceice> and it blocks the lua_test
20141011 17:45:17< iceiceice> i've got the endlevel_test_scen in this file: "./add-ons/End_Level_Test/mp_scenario/end_level_test.cfg"
20141011 17:45:39-!- happygrue [~Laptop@wesnoth/developer/wintermute] has joined #wesnoth-dev
20141011 17:48:23< iceiceice> in general it doesnt seem to have any problems loading core stuff, i tried a few campaigns now
20141011 17:50:28< zookeeper> i'm downloading the windows installer, will try to give it a try in the next few hours...
20141011 18:03:16< gfgtdf> iceiceice: sp teh expected bahviour is files beginning with a . not beeing included ?
20141011 18:03:18< gfgtdf> so the*
20141011 18:05:19< iceiceice> yeah i can't find any notes on wiki,
20141011 18:05:29< iceiceice> but in my experience wesnoth is supposed to ignore files beginning with .
20141011 18:05:38< iceiceice> also directories maybe
20141011 18:05:51< iceiceice> so if your addon folder is a git repo, its not supposed to upload the .git part
20141011 18:06:45< gfgtdf> iceiceice: maybe this is the cause: https://github.com/wesnoth/wesnoth/blob/master/src/filesystem.cpp#L138
20141011 18:08:41< iceiceice> gfgtdf: in current master that code looks like this i guess:
20141011 18:08:42< iceiceice> https://github.com/wesnoth/wesnoth/blob/master/src/filesystem.cpp#L138
20141011 18:09:21< gfgtdf> iceiceice: yes thats teh link i just posted
20141011 18:09:23< iceiceice> oh
20141011 18:09:27< iceiceice> i thought your link was to your branch
20141011 18:11:46-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20141011 18:12:08-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20141011 18:12:37< iceiceice> gfgtdf: i wonder if there should be a "if (filename[0] == '.') result.clear();" at this line: https://github.com/gfgtdf/wesnoth-old/blob/AI-boost_filesystem_2/src/filesystem_boost.cpp#L897
20141011 18:13:23< gfgtdf> iceiceice: teh original code was in get_files_in_dir
20141011 18:17:30< iceiceice> maybe here:
20141011 18:17:32< iceiceice> https://github.com/gfgtdf/wesnoth-old/blob/AI-boost_filesystem_2/src/filesystem_boost.cpp#L206
20141011 18:18:25< iceiceice> something like `if(di->path().filename().string()[0] == '.') continue;` ?
20141011 18:20:23< gfgtdf> iceiceice: ye thats somilat to what i thought, but does taht also work for directories?
20141011 18:21:02< iceiceice> i think maybe the `if (st.type() == bfs::regular_file)` rules out that case?
20141011 18:21:57< gfgtdf> iceiceice: ye right and in teh directorycase we already have std::string basename = di->path().filename().string(); 
20141011 18:24:38< gfgtdf> iceiceice: ok i oushed a commit on https://github.com/gfgtdf/wesnoth-old/tree/AI-boost_filesystem_2
20141011 18:27:36< iceiceice> is there a difference between "di->path().string()" and "di->path().filename().string()"
20141011 18:27:38< iceiceice> ?
20141011 18:28:11< iceiceice> i think this commit is going to make `looks_like_pbl` be called on the second one instead of the first
20141011 18:28:14< iceiceice> https://github.com/gfgtdf/wesnoth-old/commit/d280260f342964c86d94395f878db9cd2962d85e
20141011 18:29:20< gfgtdf> what you mean by 'second one' ?
20141011 18:30:09< iceiceice> so just looking at the diff,
20141011 18:30:20< iceiceice> we used to be calling `looks_like_pbl(di->path().string())`
20141011 18:30:29< iceiceice> now we are calling `looks_like_pbl(basename)`
20141011 18:30:44< iceiceice> i dont know if its actually different, its why i'm asking
20141011 18:31:36< gfgtdf> iceiceice: looks_like_pbl checks whether teh strign ends with *.pbl, filename() trings everything before the last '\' so it should be ok 
20141011 18:31:46< gfgtdf> trimms
20141011 18:31:52< iceiceice> oh ok
20141011 18:38:28< iceiceice> ok it fixes it for me
20141011 18:43:06< thunderstruck> What's the preferred way to build Wesnoth on OSX? Command line or XCode?
20141011 18:43:47< thunderstruck> I guess, I should say not 'preferred', but the 'easiest' way.
20141011 18:46:42< iceiceice> thunderstruck: my guess is Xcode but i dont really know
20141011 18:46:59< iceiceice> idk if the project file is up to date right now but it often has been in the past
20141011 18:47:21< iceiceice> i'm out for now, be back later
20141011 18:53:15-!- travis-ci [~travis-ci@ec2-54-196-148-65.compute-1.amazonaws.com] has joined #wesnoth-dev
20141011 18:53:16< travis-ci> gfgtdf/wesnoth-old#264 (AI-boost_filesystem_2 - d280260 : gfgtdf): The build is still failing.
20141011 18:53:16< travis-ci> Build details : http://travis-ci.org/gfgtdf/wesnoth-old/builds/37697472
20141011 18:53:16-!- travis-ci [~travis-ci@ec2-54-196-148-65.compute-1.amazonaws.com] has left #wesnoth-dev []
20141011 18:56:14< thunderstruck> Ayne: Did you look at that bug again?
20141011 18:58:28< Ayne> thunderstruck: Yeah, still looking, still not getting anywhere unfortunately
20141011 19:00:16< thunderstruck> Ayne: Okay. I'll look into it too.
20141011 19:06:27-!- EdB [~edb@89-158-11-138.rev.numericable.fr] has quit [Quit: Konversation terminated!]
20141011 19:08:55-!- zookeeper [zookeeper@wesnoth/developer/zookeeper] has quit []
20141011 19:13:05-!- travis-ci [~travis-ci@ec2-23-23-46-129.compute-1.amazonaws.com] has joined #wesnoth-dev
20141011 19:13:05< travis-ci> gfgtdf/wesnoth-old#265 (AI-boost_filesystem_2 - 9212518 : gfgtdf): The build was fixed.
20141011 19:13:05< travis-ci> Build details : http://travis-ci.org/gfgtdf/wesnoth-old/builds/37697719
20141011 19:13:05-!- travis-ci [~travis-ci@ec2-23-23-46-129.compute-1.amazonaws.com] has left #wesnoth-dev []
20141011 19:21:16-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20141011 19:22:08< vultraz> hey shadowm
20141011 19:22:46-!- irker021 [~irker@fehu.ai0867.net] has joined #wesnoth-dev
20141011 19:22:46< irker021> wesnoth: Charles Dang website:Vultraz_template_revisions 4f47c0c4b1c7 / start/1.12/template.html: Font tweaks to intro http://git.io/6RmRfw
20141011 19:24:16< mattsc> thunderstruck: Xcode
20141011 19:28:11< thunderstruck> mattsc: Thanks. I'll try Xcode then.
20141011 19:31:03-!- zookeeper [zookeeper@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20141011 19:31:14< mattsc> thunderstruck: the Debug builds *should* build right out of the box if you have Xcode 5 or later
20141011 19:31:56< mattsc> The Release builds probably won’t ...
20141011 19:32:44< shadowm> Hi happygrue.
20141011 19:34:43< irker021> wesnoth: Charles Dang website:Vultraz_template_revisions 5d7802d2a9a4 / start/1.12/template.html: Removed Downloads section since we have a nice Downloads button http://git.io/_1SDRg
20141011 19:34:59< shadowm> happygrue: Would you be able to write a short blurb on the Khalifate for us for the 1.12 press release-type announcement, or at least point us to someone who can? Right now we only have a placeholder: http://www.wesnoth.org/start/1.12/template.html#multiplayer
20141011 19:36:28< vultraz> We also need blubs on the Editor
20141011 19:36:46< shadowm> That's for fabi_, not happygrue.
20141011 19:37:15< vultraz> And "other changes at PAGE THAT DOESN’T EXIST"
20141011 19:37:24< irker021> wesnoth: Ignacio R. Morelle wesnoth:1.12 c4b9f452e796 / changelog: Fix two changelog entries I misplaced somehow http://git.io/__xH_A
20141011 19:37:27< irker021> wesnoth: Ignacio R. Morelle wesnoth:1.12 bef9bc482467 / RELEASE_NOTES: Update RELEASE_NOTES http://git.io/SR_C2A
20141011 19:37:57< vultraz> shadowm: I think I'm done with my template, save the issues with the header and title text
20141011 19:38:30< vultraz> I can't figure out the title text issue, perhaps you could do a rendering to see if it looks different
20141011 19:38:43< vultraz> As for the header...not sure
20141011 19:38:49< vultraz> I'll figure something out
20141011 19:39:42< shadowm> I would replace it with a text header.
20141011 19:40:36< shadowm> I can't be bothered to do all that manual typesetting again, it wasn't fun back then, it won't be fun now, and using text images in web pages goes directly against my principles.
20141011 19:41:42< shadowm> mattsc: I just added a bit to R_N I should've added before.
20141011 19:41:49< vultraz> You don't need the kerning, just see if it types any different, and if the bumpmap looks the same
20141011 19:42:15< shadowm> You ignored the last part of my sentence, vultraz.
20141011 19:43:05< vultraz> I compromised with the download button, I think one little textimage isn't going to hurt anyone
20141011 19:44:02< irker021> wesnoth: Ignacio R. Morelle website:master 8130451a2960 / start/1.12/template.html: Move the terrain changes description to where it belongs http://git.io/x9VDRQ
20141011 19:44:38< mattsc> shadowm: thanks - it’ll still be a couple hours before I can get back to this, but I will get there, eventually
20141011 19:44:42< Ayne> thunderstruck: I have to go now. I'm afraid I'll be away all day tomorrow, so I won't be able to take another look until Monday evening.
20141011 19:53:06< thunderstruck> Ayne: Okay.
20141011 19:54:36< Ayne> Good luck if you're going to keep looking ;)
20141011 19:54:43-!- Ayne [~Ayne@55d41d5a.access.ecotel.net] has quit [Quit: Leaving]
20141011 20:00:04< happygrue> yes shadowm, I can do that.  Is tomorrow night okay?  I have company this weekend.
20141011 20:00:35< vultraz> Blegh
20141011 20:00:38< shadowm> happygrue: I don't know when Ivanovic plans to freeze the announcement, but it's clearly still very incomplete so it probably won't be tomorrow.
20141011 20:00:38< vultraz> I need to lrn 2 git
20141011 20:01:39< happygrue> okay.
20141011 20:02:41< shadowm> vultraz: Explain your choice of font family for the lead.
20141011 20:03:04< vultraz> shadowm: I believed it looked simpler and more elegant than just  120% default
20141011 20:03:21< shadowm> Explain your choice of font family for the lead.
20141011 20:03:54< irker021> wesnoth: mattsc wesnoth:1.12 c8f5768002d4 / projectfiles/Xcode/ (3 files in 2 dirs): Xcode project update for 1.11.17 http://git.io/EjKklA
20141011 20:05:11< vultraz> Oh, ah, it's just what http://cssfontstack.com/ gave me
20141011 20:05:19< shadowm> Uh-huh.
20141011 20:05:38 * vultraz kicks git
20141011 20:09:34< mattsc> Ivanovic, shadowm: I just uploaded the 1.11.17 OS X package to SF  (not that that really matters until I have the announcement done…)
20141011 20:10:10< shadowm> mattsc: I can take care of the announcement if you want but I can't promise it'll be ready before tomorrow in the evening.
20141011 20:11:12-!- ancestral [~ancestral@12.23.74.29] has quit [Quit: i go nstuf kthxbai]
20141011 20:11:43< shadowm> vultraz: This "grab your friends" phrase sounds so cliché it's almost painful.
20141011 20:12:02< mattsc> shadowm: if you want take over, that would really be great today.  I’m struggling with everything I need to do today (non-Wesnoth stuff, I mean).
20141011 20:12:04< shadowm> Especially for single-player people.
20141011 20:12:07< vultraz> Alternative?
20141011 20:12:12< shadowm> Not using it.
20141011 20:12:19< mattsc> shadowm: Did you see the pastebin in which I just thew everything together from the changelogs?
20141011 20:12:27< mattsc> *threw
20141011 20:12:28< shadowm> mattsc: I saw it, didn't read it.
20141011 20:12:57< shadowm> vultraz: That floating <br />.
20141011 20:13:13< shadowm> That invalid floating <br> I mean.
20141011 20:13:20< vultraz> Wasn't mine
20141011 20:13:21< mattsc> shadowm: okay (I didn’t expect you to).  Well, I’ll ping you tonight my time and we can decide what we do then.  In the meantime, if you want to work on it, please do so.
20141011 20:13:36< shadowm> vultraz: It is yours.
20141011 20:13:44< vultraz> Oh
20141011 20:14:52< shadowm> mattsc: I'll start working on it now.
20141011 20:15:20< vultraz> I'm fucking up my local copy, yay!
20141011 20:17:33< vultraz> Rebased but somehow didn't marge 8130451a2960 correctly and now I have no idea what to do
20141011 20:18:33< shadowm> mattsc: Protip: [wiki=Foo]Blah[/wiki] in the forums is equivalent to [url=http://wiki.wesnoth.org/Foo]Blah[/url].
20141011 20:19:04< shadowm> And [wiki]Foo[/wiki] is equivalent to [wiki=Foo]Foo[/wiki].
20141011 20:19:15< mattsc> shadowm: thanks!  (you don’t know how big a relief that is for me.  :)
20141011 20:19:39< mattsc> shadowm: yeah, I know that in principle, I just keep forgetting about it.  :P
20141011 20:19:40< shadowm> It may also select the https protocol some day when we get our https setup sorted.
20141011 20:20:28< mattsc> shadowm: and I am sure there are all kinds of double spaces behind stops etc. in it.  I really didn’t have time to look into formatting yet and I do the double spacing automatically without even thinking about it.
20141011 20:20:46< mattsc> (I did have it written down to search for doubles spaces when I was done)
20141011 20:21:09< shadowm> The forums (well, people's web browsers) collapse space sequences into one character.
20141011 20:21:25< mattsc> right - but still
20141011 20:21:45< mattsc> shadowm: anyways, I usually wouldn’t have sent you something in such a miserable state, but it was clear that there would be at least an 8h break before I could get back to it, probably more.
20141011 20:22:27< shadowm> You killed the stricken ETAs. :(
20141011 20:22:47< mattsc> stricken ETAs?
20141011 20:23:00< mattsc> oh ...
20141011 20:23:07< mattsc> yes …
20141011 20:23:19< mattsc> feel free to add them back in  :)
20141011 20:24:45< mattsc> I did not know that ‘stricken’ was a valid past participle of ‘to strike’ !
20141011 20:25:58< mattsc> anyways, I need to be afk again, talk to you later and thank you!
20141011 20:26:57< vultraz> agh
20141011 20:27:48< vultraz> I dunno what I'm doing anymore here
20141011 20:29:09< vultraz> shadowm: ok, besides the friends line and the floating <br>, anything else?
20141011 20:29:25< shadowm> Yes, a lot of stuff. It's very unpolished.
20141011 20:29:36< vultraz> The design or the code
20141011 20:29:48< shadowm> I don't even want to look at the code yet.
20141011 20:30:16< shadowm> I'll try to clean it up a bit later to illustrate some points, since otherwise I'll be waiting forever for you to learn HTML and CSS.
20141011 20:34:14< vultraz> ok this shit seems to be resolved
20141011 20:34:22< irker021> wesnoth: Chris Beck website:Vultraz_template_revisions a30d7888e8a6 / README.md: Create README.md http://git.io/qHdxfQ
20141011 20:34:24< irker021> wesnoth: Nils Kneuper website:Vultraz_template_revisions 45034fe634b2 / start/1.12/ (44 files in 2 dirs): ran make to update po files and generate html files http://git.io/LV1Ljg
20141011 20:34:26< irker021> wesnoth: Nils Kneuper website:Vultraz_template_revisions a8bde8264be0 / gettext.wesnoth.org/bin/update-gettext-stats: add the whole website repo to the regular update cycle http://git.io/n3VbgA
20141011 20:34:28< irker021> wesnoth: Nils Kneuper website:Vultraz_template_revisions 3bb5f99f847f / gettext.wesnoth.org/bin/update-gettext-stats: get rid of last commit, should not be required http://git.io/CJ1NDA
20141011 20:34:30< irker021> wesnoth: Ignacio R. Morelle website:Vultraz_template_revisions 8130451a2960 / start/1.12/template.html: Move the terrain changes description to where it belongs http://git.io/x9VDRQ
20141011 20:34:33< irker021> wesnoth: Charles Dang website:Vultraz_template_revisions 021ad8e2f99c / start/1.12/ (6 files in 3 dirs): Merge branch 'Vultraz_template_revisions' of github.com:wesnoth/website into Vul http://git.io/DKYjxg
20141011 20:36:38< shadowm> I just realized it's time for me to setup the 1.12 MP server.
20141011 20:36:53< shadowm> On server., server2. and server3.
20141011 20:37:27< irker021> wesnoth: Charles Dang website:Vultraz_template_revisions 5ddeb1d73bc8 / start/1.12/template.html: Fixed an invalid break tag http://git.io/3o0JZw
20141011 20:37:29< irker021> wesnoth: Charles Dang website:Vultraz_template_revisions 715aaf8dab19 / start/1.12/template.html: Removed cheesy line from the intro http://git.io/YGMG_A
20141011 20:37:39< vultraz> Alright!
20141011 20:37:52< vultraz> Make whatever edits you want
20141011 20:43:59< shadowm> gfgtdf: You (?) say in R_N "This one aditional check was removed." regarding defeat_condition and [endlevel]. Does this mean all sides are always carried over, or something else?
20141011 20:45:50< shadowm> iceiceice: "Version 1.11.7 from just before feature freeze"... four months difference to 1.11.10.
20141011 20:46:00-!- happygrue [~Laptop@wesnoth/developer/wintermute] has quit [Remote host closed the connection]
20141011 20:46:02< shadowm> Just a small observation.
20141011 20:46:24< gfgtdf> iceiceice: no that's not teh case
20141011 20:46:56< gfgtdf> shadowm: no that's not teh case
20141011 20:47:25< shadowm> Okay, so how does it impact side carryover?
20141011 20:49:09< gfgtdf> shadowm: teh current state is: the 'defeated check' is always applied after any user action (move, attack, menuclick, ...). A side is not carries over if it was defeated during the last 'defeat check'
20141011 20:49:19< gfgtdf> carried*
20141011 20:50:05< gfgtdf> this can be disabled by eigher settign removbe_from_carryover_on_defeat=no, or my manualy selecting the sides that should be carriedover in a victory event
20141011 20:50:13< shadowm> So who is the first grammatical person in "Buggy Carryover WML feature removed", I suddenly need to know because...
20141011 20:50:20< shadowm> "the main use I had in mind"
20141011 20:51:12< shadowm> It's iceiceice, so I guess I'll just say "the author" unless iceiceice wants me to name him specifically.
20141011 20:52:13< shadowm> gfgtdf: Let's say the leader of a side is dead and I endlevel in his die event, what happens to that side?
20141011 20:52:56< shadowm> Or if you prefer, is the defeat check performed before or after 'die' events?
20141011 20:53:29< shadowm> (I assume it's always after preceding 'last breath' events at least.)
20141011 20:54:56< gfgtdf> shadowm: teh curretn behaviour is: there is no 'defeat check' in this user action, becasue  before there was an [endlevel] before the during the user action. So whether a side carries over depends on whether it was defeated on teh previous 'defeat check'
20141011 21:00:02< gfgtdf> shadowm: current means 1.12 becasue soem of this is still not ported to master
20141011 21:03:12< shadowm> Best hurry up and port it if it applies to master too, otherwise it's too easy to forget patches.
20141011 21:04:56< shadowm> gfgtdf: My question is whether the defeat check happens after or before 'die' events.
20141011 21:06:12< shadowm> Also, I guess that all this means that killing a side's units in an event and calling [endlevel] before that event finishes processing no longer causes the side to be defeated if it wasn't already?
20141011 21:06:35< gfgtdf> shadowm: it woudl happen after the die event, but since tehre was an [endlevel] before it wint happen at all, so what counts then is the last 'defeat check'  before the user action that invokes the die event (usualy teh 'attack' user action)
20141011 21:07:00< gfgtdf> s/wint/won's
20141011 21:07:36< gfgtdf> which is most likeley the 'defeat check' after the 'movement' to attack
20141011 21:07:53< gfgtdf> from teh engine persopective move+attack are 2 unrelated actions
20141011 21:07:54< shadowm> But the game doesn't know at that point that the leader is going to die, does it?
20141011 21:08:01< gfgtdf> no
20141011 21:08:04< gfgtdf> it doesnt
20141011 21:28:53< gfgtdf> in a  1 human vs 1 ai game this will ususaly never remove the human sides carryover, but for example in a 2 vs 2 game with p1 & p2 vs p3 & p4, if the player die in this order p1,p3,p4,p2    side 1s team will win but his units wonr carryover to the next scenario because he got defeated.
20141011 21:28:56< gfgtdf> shadowm^
20141011 21:31:08< zookeeper> anything in particular that i should test in 1.11.17?
20141011 21:31:14-!- gfgtdf_ [~chatzilla@e176185190.adsl.alicedsl.de] has joined #wesnoth-dev
20141011 21:33:07-!- gfgtdf [~chatzilla@f054159222.adsl.alicedsl.de] has quit [Ping timeout: 245 seconds]
20141011 21:33:10-!- gfgtdf_ is now known as gfgtdf
20141011 21:35:05< gfgtdf> iceiceice: can i use travis to test whether code works in a lower version of boost ?
20141011 21:45:05< gfgtdf> iceiceice: for example if i wont to test whether a by branch also compiles with 1.44 or 1.40 
20141011 21:48:31-!- Gekko [~Gekko@dsl-tkubrasgw1-54fa1b-113.dhcp.inet.fi] has left #wesnoth-dev []
20141011 21:48:59< mattsc> zookeeper: we need somebody to play through all of LoW  ;)
20141011 21:49:13< fabi_> mattsc: hi
20141011 21:49:21< mattsc> hi fabi_ 
20141011 21:49:40< fabi_> I did not commit anything new into LoW.
20141011 21:50:16< zookeeper> mattsc, ahah.. ha... funny
20141011 21:51:02< mattsc> fabi_: yeah, I saw.  I was going to tell zookeeper that in case he was willing to do so, but for some reason I knew his answer in advance.
20141011 21:51:25< fabi_> Yeah, zookeeper is quite predictable :-P
20141011 21:52:03< zookeeper> predictability is a virtue :>
20141011 21:52:41< mattsc> fabi_: current plan is to get out the next version (which theoretically could be 1.12.0) in less than two weeks (Friday the week after next).
20141011 21:53:07< mattsc> do you think you/we can get LoW done and tested by then?
20141011 21:53:19< mattsc> zookeeper: nobody was complaining  :)
20141011 21:53:21< fabi_> Why don't we aim at Xmas like we did already some times before?
20141011 21:53:51< fabi_> It seems more reachable and is also a nice date for a release.
20141011 21:53:56< mattsc> because it would be really nice to get 1.12.0 into the next debian stable
20141011 21:54:05< fabi_> oh
20141011 21:54:15< vultraz> Also 1.12 was supposed to be out in April
20141011 21:54:20< vultraz> That was the original projection
20141011 21:54:20< mattsc> also, because it’s been so long already ...
20141011 21:54:29< vultraz> It's now October
20141011 21:54:31< mattsc> right, as vultraz says
20141011 21:54:37< fabi_> Well, when is debian stable ready?
20141011 21:54:47< mattsc> feature freeze on Nov 5
20141011 21:54:54< fabi_> I think it is not a good idea to have an early 1.12.x in debian.
20141011 21:55:21< vultraz> Better than a late 1.10
20141011 21:55:34< fabi_> 1.12.0 won't work, I promise.
20141011 21:56:07< vultraz> We've been testing and bugfixing for months, now
20141011 21:56:22< fabi_> I bet it will be Wesnoth's KDE 4.0 release.
20141011 21:57:39< mattsc> fabi_: independent of this, we should try to get 1.12.0 released as soon as possible.  Waiting another 2.5 months because christmas is a nice date for a release is not a good idea IMO
20141011 21:57:52< fabi_> okay
20141011 21:58:34< fabi_> I am glad you are pushing this forward. So all hail to the new chief.
20141011 21:58:44< vultraz> Funny, this is the reverse Skyrim problem...they made sure they would release on 11.11.11...we're making sure we release before a special date
20141011 21:58:51< Ivanovic> fabi_: because i seem to be the only one who really knows the pot-update system and what to fix where to make the files function
20141011 21:59:09< Ivanovic> fabi_: and i will be on vacation (and 100% offline!) starting in the beginning of december until basically christmas
20141011 21:59:35< mattsc> fabi_: thanks, but no hailing please :)
20141011 21:59:51< fabi_> Okay, so let's get that thing into debian stable.
20141011 21:59:56< Ivanovic> fabi_: shadowm tried to run a pot-update but something seems to not work correctly for him since the files he created included some errors
20141011 22:00:19< Ivanovic> that is even more errors than there were in the files themselves already
20141011 22:00:21< Ivanovic> ;)
20141011 22:01:10< fabi_> mattsc: What I am doing to LoW14 is mostly making everything much more simple than the previous solution. Thus I bet it is not a big problem. Regarding that the current thing is at least broken for MP anyway.
20141011 22:01:44< fabi_> Need to go ...
20141011 22:03:30< mattsc> fabi_: ok
20141011 22:03:39< mattsc> let me know if you need any help with anything
20141011 22:03:43< mattsc> and zookeeper ;)
20141011 22:04:34< shadowm> Ivanovic: "Some errors" would be an understatement. For all intents and purposes, the results for me are corrupted and unusable.
20141011 22:05:25< irker021> wesnoth: Nils Kneuper wesnoth:1.12 899bd3889f8f / / (11 files in 10 dirs): updated Lithuanian translation http://git.io/_6YMgA
20141011 22:05:27< irker021> wesnoth: Nils Kneuper wesnoth:master 43b572d948fd / / (10 files in 10 dirs): updated Lithuanian translation http://git.io/XuBpBA
20141011 22:05:46< Ivanovic> shadowm: ahhhh, it is just some errors for the moment and with some debugging it will probably be possible to find the cause
20141011 22:05:52< Ivanovic> at least the programs themselves start and run
20141011 22:05:54< Ivanovic> ;)
20141011 22:06:37< stikonas> thanks, Ivanovic
20141011 22:07:18< shadowm> I'd like to revisit the issue at a later point because I've often found myself wanting to see what some code change would look like for translators.
20141011 22:07:36< mattsc> Ivanovic: btw, I have contacted the two people who had submitted translation PRs.  One of them got back to me and closed the PR.  Still waiting to hear back from the other.
20141011 22:07:36< shadowm> You know, since I do UI stuff.
20141011 22:08:03< Ivanovic> shadowm: yeah, that is why it would be nice to have a look at what is really going on there
20141011 22:08:15< mattsc> I also added a statement on the wiki page that translations should not be submitted by PR.
20141011 22:09:20< stikonas> what about Image translations?
20141011 22:09:39< stikonas> they require more changes, i.e.update l10n-track file...
20141011 22:09:46< Ivanovic> stikonas: the problem is that someone needs to handle those
20141011 22:09:57< stikonas> yeah, it used to be Espreon...
20141011 22:09:59< Ivanovic> and honestly, i don't handle the image commits since i just got no clue what needs to be done
20141011 22:10:03< stikonas> now he is absent
20141011 22:10:29< stikonas> I roughly know them...
20141011 22:10:51< vultraz> I am speaking to espreon right now
20141011 22:11:34< shadowm> "Marked the report private due to concerns of the reporter about privacy when people connect account names." pfffft.
20141011 22:11:48< shadowm> Now I can't use bug #20089 in the announcement.
20141011 22:12:21< shadowm> Bah, I'll mention it anyway.
20141011 22:14:34-!- mjs-de [~mjs-de@f048239245.adsl.alicedsl.de] has quit [Remote host closed the connection]
20141011 22:16:15< vultraz> Ivanovic: espreon reports he lost the l10n script anyway
20141011 22:16:32< vultraz> or AI0867 might have stuck it in utils/
20141011 22:17:24< vultraz> "[9:17:13 AM] espreon: All I know is that it was a script called lbundle_check.py or something like that"
20141011 22:17:26< Ivanovic> at least not in 1.12
20141011 22:18:16< vultraz> He recommends asking some Serbian translator by the name of Chusslove
20141011 22:18:29< Ivanovic> okay
20141011 22:18:46< stikonas> I can check for lbundle_check in some KDE utils
20141011 22:18:58< stikonas> Chusllove did the original localization of images
20141011 22:19:15< stikonas> and it is highly likely that he used some KDE scripts
20141011 22:19:55< stikonas> ok, it is here: https://techbase.kde.org/Localization/Tools/Lbundle_Check
20141011 22:20:50-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20141011 22:21:28< stikonas> I actually have that lbundle_check.py script in my kde localization folder...
20141011 22:22:38< vultraz> *sigh* "[9:21:44 AM] espreon: Give them this: [06:18:37 PM] espreon: http://wiki.wesnoth.org/Translation_Maintainance_Commands#Image_Synchronization"
20141011 22:24:38< shadowm> mattsc: Is it time to include the OS X-specific known bugs now?
20141011 22:25:38< mattsc> shadowm: I’m not sure.  It’s not like we’re going to do anything about them for 1.12 anyway.
20141011 22:26:21< mattsc> So I’d probably put them somewhere on the wiki and include a link to it somewhere once 1.12.0 is out.  But I really don’t know how this _should_ be handled.
20141011 22:28:49< shadowm> I'm going to add them to the 1.11.17 RA.
20141011 22:29:09< stikonas> looks like Chusslove is still listed as image maintainer...
20141011 22:29:37< stikonas> last commit was in March. But maybe nobody translated any images since...
20141011 22:35:37< mattsc> shadowm: ok
20141011 22:37:46< thunderstruck> mattsc: I've been looking into MP sos bug and I think I know what's the problem. I'm going to sleep now and tomorrow I'll try to implement a fix I have in mind.
20141011 22:38:07< thunderstruck> sos saves*
20141011 22:38:24< thunderstruck> Ayne ^
20141011 22:44:27< mattsc> thunderstruck: great, thank you!
20141011 22:47:18< shadowm> mattsc: One for your list: https://gna.org/bugs/?21943 (already added it to my draft).
20141011 22:48:00< shadowm> mattsc: IIRC you already fixed https://gna.org/bugs/?21649 by upgrading pango+cairo in your bundles?
20141011 22:54:21< Ivanovic> n8
20141011 22:54:30< shadowm> Draft's up: http://forums.wesnoth.org/viewtopic.php?f=16&t=41071
20141011 22:56:20-!- ancestral [~ancestral@12.23.74.29] has joined #wesnoth-dev
20141011 22:59:14< vultraz> "really close" -> "quite close" and then make it "you find so we can fix them"
20141011 22:59:18< vultraz> rest is fine
20141011 22:59:50< shadowm> Why that change?
20141011 23:00:00< shadowm> The really->quite change, that is.
20141011 23:05:17< vultraz> "really" is more informal
20141011 23:05:27< shadowm> So is most of the intro.
20141011 23:05:37< vultraz> *shrug*
20141011 23:05:39< vultraz> up to you
20141011 23:05:42< iceiceice> shadowm: i will check the 1.11.7 remark, its probably inaccurate
20141011 23:06:02< iceiceice> re: buggy carryover wml feature removed, i like what you did with that
20141011 23:06:30< shadowm> And I'm going to paint it yellow now along with the mushroom change.
20141011 23:06:41< shadowm> I mean orange.
20141011 23:07:33< shadowm> That's a call to attention to people who might have made use of the carryover feature and rebalancd maps in the meantime.
20141011 23:15:34-!- gfgtdf [~chatzilla@e176185190.adsl.alicedsl.de] has quit [Ping timeout: 255 seconds]
20141011 23:18:35-!- fabi_ [~fabi@wesnoth/developer/fendrin] has quit [Quit: Konversation terminated!]
20141011 23:19:46< iceiceice> yeah ok
20141011 23:19:59< iceiceice> gfgtdf: i dont know about reducing boost version on travis
20141011 23:20:06< iceiceice> its probably possible
20141011 23:20:33< iceiceice> the thing is, old boost versions are generally not available from linux repositories,
20141011 23:20:44< iceiceice> maybe only the three most recent versions
20141011 23:21:03< iceiceice> so you'd have to find some PPA or something that carries it i guess, if that exists
20141011 23:24:01< iceiceice> shadowm, mattsc: here's another bug report i forgot about
20141011 23:24:01< iceiceice> http://forums.wesnoth.org/viewtopic.php?f=5&t=40745&start=15#p575565
20141011 23:24:05< iceiceice> #7 from bob
20141011 23:24:44< iceiceice> i was a bit clueless about this... i dont know if any of the other player name patches that are going on will impact this?
20141011 23:26:38-!- ancestral [~ancestral@12.23.74.29] has quit [Quit: ancestral]
20141011 23:26:40< iceiceice> im not sure why i bring this up now, i dont know if this, or #6, or #9 should be considered critical
20141011 23:27:02< iceiceice> its unfortunate that these dont have proper bug reports in the bug tracker...
20141011 23:28:17-!- ancestral [~ancestral@12.23.74.29] has joined #wesnoth-dev
20141011 23:32:35-!- kex [~kex@78.157.29.160] has joined #wesnoth-dev
20141011 23:35:19-!- ancestral [~ancestral@12.23.74.29] has quit [Read error: Connection reset by peer]
20141011 23:38:52-!- ancestral [~ancestral@12.23.74.29] has joined #wesnoth-dev
20141011 23:39:30< mattsc> shadowm: okay on 21943
20141011 23:40:15< mattsc> I updgraded the pango/cairo libs for 1.13 only, not 1.12.  And I don’t remember is that fixes #21649, let me check quickly.
20141011 23:43:17< mattsc> It does.  So yes, it’s fixed in master, not in 1.12.
20141011 23:43:35-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20141011 23:44:36< mattsc> iceiceice: I don’t think that #6 is important.
20141011 23:45:08< mattsc> #7 and #9, I don’t know, I almost never play MP, so somebody else will have to say.
20141011 23:49:46< shadowm> mattsc: I'll add #21649 then.
20141011 23:51:52-!- kex [~kex@78.157.29.160] has quit [Ping timeout: 245 seconds]
20141011 23:53:07-!- zookeeper [zookeeper@wesnoth/developer/zookeeper] has quit [Ping timeout: 245 seconds]
20141011 23:53:25< mattsc> shadowm: I discussed this with Alaranatalara and crimson_penguin back when we upgraded the libs and we decided against putting it into 1.12.0 (too dangerous; plus there are a bunch of font issues that I have not figured out yet), but thought that we could try in 1.12.1.
20141011 23:54:02< shadowm> Okay.
20141011 23:54:04-!- kex [~kex@78.157.29.160] has joined #wesnoth-dev
20141011 23:54:15< shadowm> I added a comment to the bug in the tracker too.
20141011 23:55:13< mattsc> shadowm: I’m currently reading the announcement draft.  I don’t like the opening paragraph.
20141011 23:55:24< mattsc> I know that I wrote it, which is probably why I don’t like it.  :P
20141011 23:55:49< mattsc> It feels like it is *almost* right, but not quite, but I cannot put my finger on what is bothering me.
20141011 23:58:56< shadowm> iceiceice: As a total MP noob I can't offer an opinion on Bob's #7, sorry.
20141011 23:59:17< Coffee_irc> does anyone need help testing MP bugs?
20141011 23:59:25< shadowm> My gut feeling is that it's wrong and AI leaders should have whichever name the scenario coder specified just like in SP.
--- Log closed Sun Oct 12 00:00:45 2014