--- Log opened Thu Oct 09 00:00:28 2014
20141009 00:08:01-!- ancestral [~ancestral@12.23.74.29] has joined #wesnoth-dev
20141009 00:08:51< irker820> wesnoth: Ignacio R. Morelle website:master 5b3392edad05 / start/1.12/ (12 files in 2 dirs): Progress with content and screenshots http://git.io/UBG2Pg
20141009 00:12:49-!- ancestral [~ancestral@12.23.74.29] has quit [Ping timeout: 272 seconds]
20141009 00:53:29< irker820> wesnoth: Chris Beck wesnoth:master 4576dd0d73da / data/core/units.cfg: add description for gryphon race http://git.io/K99t2g
20141009 01:08:09< irker820> wesnoth: gfgtdf wesnoth:master 41fa779b08d1 / src/serialization/ (ucs4_iterator_base.hpp unicode.cpp unicode.hpp): add utf16::iterator http://git.io/dYXMYg
20141009 01:08:11< irker820> wesnoth: gfgtdf wesnoth:master 692f9e2d4a34 / src/serialization/ (unicode.cpp unicode.hpp): add utf16string_to_ucs4string http://git.io/XN8XXg
20141009 01:08:13< irker820> wesnoth: gfgtdf wesnoth:master b037cbbfeeb9 / src/serialization/ (unicode.cpp unicode.hpp): add utf16string_to_string http://git.io/xQM-Ig
20141009 01:08:15< irker820> wesnoth: gfgtdf wesnoth:master 4a46ef1f68f6 / src/tests/test_serialization.cpp: add utf16 -> utf8 unittest http://git.io/eNXYiw
20141009 01:08:17< irker820> wesnoth: gfgtdf wesnoth:master d246baa56728 / src/ (4 files in 2 dirs): Merge pull request #293 from gfgtdf/utf16_to_utf32 http://git.io/2PXwOA
20141009 01:14:44-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20141009 01:15:09-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141009 01:15:36< mattsc> shadowm: agreed on the EI: Weldyn Besieged map
20141009 01:16:12< mattsc> bumbadadabum: haven’t you worked on some of the maps before?  Do you want to make this look a little less … bland?
20141009 01:16:27< bumbadadabum> oh right yes
20141009 01:16:39< bumbadadabum> I just got busy near the end
20141009 01:16:43< bumbadadabum> and never really finished it
20141009 01:17:03< bumbadadabum> university is taking up a lot of my time lately
20141009 01:17:13< bumbadadabum> I might try to get back to it
20141009 01:17:16< bumbadadabum> once I have the time
20141009 01:17:54< mattsc> bumbadadabum: no worries, I know how that goes.  On what timescale do you think you might have time for it?
20141009 01:18:02< bumbadadabum> uhhh
20141009 01:18:18< bumbadadabum> probably in december/january I have all the time
20141009 01:18:36< mattsc> ah, ok, that might be too late for 1.12.0  :)
20141009 01:18:45< bumbadadabum> oh shit
20141009 01:18:45< bumbadadabum> uhh
20141009 01:18:55< mattsc> I’ll ask zookeeper then and we can figure it out between the two of us, don’t worry about it
20141009 01:19:02< bumbadadabum> I could just try to do it next week
20141009 01:19:26< mattsc> how about I ping you again in a week, and we’ll take it from there?
20141009 01:19:36< bumbadadabum> ping me next monday or so
20141009 01:19:42< mattsc> will do, thanks
20141009 01:20:13< vultraz> mattsc: what's happening?
20141009 01:25:12< mattsc> vultraz: it’s getting dark on the west coast?
20141009 01:25:38< mattsc> what do you mean?
20141009 01:28:07< vultraz> mattsc: what is it you need to be done that you're talking to bumbadadabum about
20141009 01:28:17< bumbadadabum> finishing the map updates for EI
20141009 01:28:28< mattsc> vultraz: yes ^
20141009 01:28:49< shadowm> vultraz: Check the topic for a link to the logs.
20141009 01:31:43-!- prkc [~prkc@catv-80-99-60-5.catv.broadband.hu] has quit [Remote host closed the connection]
20141009 01:32:49< vultraz> ah
20141009 01:32:55< vultraz> I can also work on maps
20141009 01:34:10< shadowm> It's important to keep in mind that the map in question is a century or so after its version in HttT BfW.
20141009 01:40:04< mattsc> vultraz, bumbadadabum: personally, from my point of view it doesn’t matter who of you wants to do this.
20141009 01:40:34< mattsc> shadowm: is the announcement draft you committed up there ready for input from others?
20141009 01:41:16< mattsc> If so, we should point it out to people, esp. those mentioned by name/handle in it.  Most of them probably don’t know about that.
20141009 01:41:31< shadowm> Yeah, I mean, I don't think I have more writing juice left.
20141009 01:41:50< mattsc> Okay - do you want to do it or should I?
20141009 01:41:56< shadowm> Plus everything that says TODO is obviously beyond my ability to fill in right now.
20141009 01:42:06< mattsc> or mine
20141009 01:42:28< shadowm> Eh, you seem to be in contact with a few of the required people already, so why not?
20141009 01:42:31< vultraz> shadowm: proofread?
20141009 01:42:50< shadowm> You can proofread it but I warn you, it probably sucks.
20141009 01:42:51< mattsc> shadowm: okay, will do
20141009 01:43:07< mattsc> vultraz: it doesn’t suck, it’s just not done yet
20141009 01:43:11< shadowm> As I told mattsc in private before, I've been incredibly uninspired to write anything.
20141009 01:44:14-!- Necrosporus_ [~Necrospor@unaffiliated/necrosporus] has joined #wesnoth-dev
20141009 01:45:38-!- iceiceice [~chris@50.245.222.235] has joined #wesnoth-dev
20141009 01:45:38-!- iceiceice [~chris@50.245.222.235] has quit [Changing host]
20141009 01:45:38-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20141009 01:47:11-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has quit [Ping timeout: 258 seconds]
20141009 01:47:16< iceiceice> digging through the bug tracker just now... found this one, i wonder if/how it relates to our other unicode woes? https://gna.org/bugs/?20815
20141009 01:47:19< irker820> wesnoth: Ignacio R. Morelle website:master 8ac70199aff2 / start/1.12/template.html: Fix some HTML usage http://git.io/76xyiw
20141009 01:48:37< shadowm> It's unrelated to filesystem code, at least.
20141009 01:48:56< mattsc> shadowm: btw, is there a way to see the announcement as a webpage directly on the github website?  All I ever have ever managed to get so far is the html code.
20141009 01:49:13< shadowm> I wouldn't be able to tell whether it's an SDL problem, or GUI1 textboxes are failing to treat their contents as multibyte strings.
20141009 01:49:55< shadowm> (It's a common programmer mistake to assume that every std::string has 1-sized characters, and perform bytewise operations on them under that assumption.)
20141009 01:50:45< shadowm> (Or perhaps more accurately, assuming that std::string is actually designed for Unicode, which it is not.)
20141009 01:51:34< shadowm> mattsc: You can't on GitHub, but you can do it on your own system if you clone the repository.
20141009 01:52:10< shadowm> Just open template.html from your clone in your browser.
20141009 01:54:33< mattsc> shadowm: yeah, that’s what I’ve been doing, I was just wondering whether there was a lazier way
20141009 01:55:53-!- Gekko [~Gekko@dsl-tkubrasgw1-54fa1b-113.dhcp.inet.fi] has joined #wesnoth-dev
20141009 01:55:59< shadowm> mattsc But that's the lazy way. :p
20141009 01:56:07< Gekko> iceiceice: Thanks for fixing http://gna.org/bugs/?22251 !
20141009 01:56:25< shadowm> The non-lazy way is to treat is as a regular web page, which involves establishing a TCP connection to our http server, etc.
20141009 01:56:25< mattsc> shadowm: :D
20141009 01:56:51< Gekko> The labels is half the reason we play wesnoth http://plantmonster.net/pics/wesnoth_labels001.png
20141009 01:56:56< shadowm> Plus the lazy way actually reflects your changes without requiring me to git pull them on the remote or catch the half hourly cron job.
20141009 01:59:00< iceiceice> Gekko: thanks for the enthusiasm :)
20141009 01:59:30< iceiceice> we were sure to get them working again sooner or later i think...
20141009 02:00:07< iceiceice> i hope you guys will report any further problems, i'm only one man and i'm not sure if there were any other problems created in that code recently
20141009 02:00:20< Gekko> We found some but they were already fixed
20141009 02:03:03-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has joined #wesnoth-dev
20141009 02:05:13< vultraz> mattsc: I tried https://htmlpreview.github.io but it doesn't seem to work
20141009 02:08:24-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Ping timeout: 272 seconds]
20141009 02:15:49< shadowm> shadowm@nanacore:~/src/wesnoth-website% du -sh .git
20141009 02:15:49< shadowm> 30M     .git
20141009 02:19:05< mattsc> happygrue, thunderstruck, Ivanovic, fabi_, everybody who cares: shadowm has been working on the 1.12 announcement and produced a strong first draft, but we need input from you:  http://git.io/UBG2Pg
20141009 02:19:25< shadowm> iceiceice: Back to the unicode input bug, I'd say it's a blocker.
20141009 02:19:38< shadowm> Or even a security bug.
20141009 02:19:45< mattsc> shadowm: those are just the folks who are here right now, so that’s my lazy way of trying to get their attention
20141009 02:20:06< mattsc> I’ll follow up with others later, and do a post to the ML as well
20141009 02:20:13< shadowm> I seem to require zookeeper too.
20141009 02:20:22< shadowm> And lipkab.
20141009 02:20:28< vultraz> And me
20141009 02:20:31< shadowm> Also Ivanovic or Espreon.
20141009 02:20:45< mattsc> yes, but he’s asleep (and he often doesn’t read the logs, so I’ll chase him up tomorrow.
20141009 02:21:00< mattsc> Ivanovic is included.
20141009 02:21:11< shadowm> I don't see a single TODO mark mentioning vultraz.
20141009 02:21:14< mattsc> And as I said: these are the folks who are here right now and are specifically called out.
20141009 02:21:21< mattsc> I’ll chase up the rest later.
20141009 02:21:37< mattsc> vultraz is important enough he does not need to be called out by name  :)
20141009 02:21:49< mattsc> vultraz: see above ^  :)
20141009 02:22:02< shadowm> I filled in the UI redesign description first, so.
20141009 02:22:45< iceiceice> huh, wonder why this link renders as text and not a webpage: https://raw.githubusercontent.com/wesnoth/website/5b3392edad0542a3fad864c5932d5eee95655b51/start/1.12/template.html
20141009 02:23:09< mattsc> because that’s what github does
20141009 02:23:15< shadowm> Come on, guys.
20141009 02:23:21< shadowm> It's only 30 MiB.
20141009 02:23:32< mattsc> anyways, I need to do dishes …
20141009 02:23:44< mattsc> as I said, I’ll follow up with others later
20141009 02:24:12< mattsc> (oh, I’m set to afk already anyway; very convenient - but annoying)
20141009 02:25:22< vultraz> ged dhammit i guess ill just clone the website branch
20141009 02:25:33< shadowm> It's not a branch.
20141009 02:25:46< shadowm> I mean, yes, it is a branch. It's master.
20141009 02:25:55< vultraz> er,r epo
20141009 02:25:58< vultraz> repo
20141009 02:26:37< vultraz> shadowm: y u no show off the new bigmaps?
20141009 02:27:06< shadowm> 1) Let's be professional here and use full English sentences.
20141009 02:27:28< shadowm> 2) They don't make for good small screenshot material.
20141009 02:27:37-!- Ivanovic_ [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20141009 02:27:49< shadowm> (Small screenshots must be cropped, not scaled down.)
20141009 02:28:35< vultraz> Well, I'm off to lunch
20141009 02:28:42< vultraz> I'll look at this when I return
20141009 02:29:20< irker820> wesnoth: Ignacio R. Morelle wesnoth:master e0073a606418 / projectfiles/CodeBlocks/wesnoth.cbp: CodeBlocks project file update http://git.io/qrA5XQ
20141009 02:30:47< shadowm> iceiceice: I'm going to look into the Unicode char deletion crash.
20141009 02:30:52-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 240 seconds]
20141009 02:31:34-!- Ivanovic_ is now known as Ivanovic
20141009 02:31:40< shadowm> :|
20141009 02:32:14< shadowm> You'd think somebody would have done so already as it's been open since May 2013.
20141009 02:33:14< shadowm> mattsc: The bug I'm talking about: <https://gna.org/bugs/index.php?20815> Promoted to Blocker.
20141009 02:33:58< shadowm> I'm so glad we have a consistent test case for it in the form of an Internet meme.
20141009 02:41:05< iceiceice> ok, i'm glad it was useful to point this out...
20141009 02:41:11< iceiceice> do you know if it crashes other clients in mp?
20141009 02:41:14< iceiceice> or only your client
20141009 02:41:39< iceiceice> also, based on discussion, it sounds like some people wanted to view it as a gui1 vs gui2 issue and "postpone" until gui1 is gone
20141009 02:41:50< shadowm> The data isn't sent to other clients, so it can't crash other clients.
20141009 02:44:22< shadowm> iceiceice: See query.
20141009 02:44:38< shadowm> Also, yeah, that's bound to happen with every bug involving GUI1.
20141009 02:45:22< shadowm> In unrelated news, Code::Blocks 13.12 won't let me build Wesnoth 1.12 due to animated.i, yay.
20141009 02:45:31< shadowm> Can we rename it to animated.tpp in 1.12 too?
20141009 02:45:46< iceiceice> yeah that sounds like a good idea
20141009 02:45:58< shadowm> Okay, I'll get on it later.
20141009 02:47:03< shadowm> *** Error in `/home/shadowm/src/wesnoth-1.12/wesnoth-debug': free(): invalid size: 0x00000000069547c0 ***
20141009 02:47:15< shadowm> The fact that it crosses into SDL territory isn't too reassuring.
20141009 03:01:01< happygrue> okay, thanks mattsc:  I'll take a look tomorrow.
20141009 03:01:07< happygrue> busy these days.  :\
20141009 03:03:23-!- happygrue [~Laptop@wesnoth/developer/wintermute] has quit [Remote host closed the connection]
20141009 03:05:00< shadowm> I need to figure out how to tell SDL to not touch ALSA so I can debug without having to stop every other audio application first.
20141009 03:06:05< shadowm> SDL_AUDIODRIVER=disk SDL_DISKAUDIOFILE=/dev/null, hopefully.
20141009 03:09:51< mattsc> shadowm, iceiceice: thanks, I’ll add it to the list
20141009 03:19:50< shadowm> #12 0x000000000138e8ac in gui::textbox::draw_contents (this=0x7fffffff87d8) at src/widgets/textbox.cpp:235
20141009 03:19:54< shadowm> 235                             sdl_blit(text_image_, &src, surf, &dest);
20141009 03:20:05< shadowm> surf is the display framebuffer, which is obviously valid.
20141009 03:20:10< shadowm> text_image_ looks valid too. :\
20141009 03:22:50< shadowm> i.e. https://gist.githubusercontent.com/shikadilord/37ac16d917648a25ba38/raw/1c8f371d1dc532fcf9ee8acb972fc9ce6bdf5215/gistfile1.txt
20141009 03:24:54< shadowm> Although now that I think of it, that doesn't really guarantee anything down the pipeline since the malloc arena may have been corrupted.
20141009 03:25:42< iceiceice> yeah that sounds like i good guess
20141009 03:25:57< iceiceice> i would have thought that the crash would happen when it's tryign to generate the text_image_ for the malformed text
20141009 03:26:04< iceiceice> *a good guess
20141009 03:26:37< iceiceice> be back later
20141009 03:26:44-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20141009 03:31:44< shadowm> The result changes to an actual segmentation fault when performing an additional heap-based allocation near the fault point, so yeah.
20141009 03:35:08< shadowm> Though it's still deep in the same SDL path.
20141009 03:40:35< shadowm> There's a previous call to gui::textbox::update_text_cache  in charge of building that surface that somehow gets SDL to barf on the heap, let's see.
20141009 03:41:07< shadowm> The source is a ucs4::string.
20141009 03:42:22< shadowm> The field holding the text that is modified by SDLK_DELETE press events in gui::textbox::handle_event is also a ucs4::string.
20141009 03:42:57< shadowm>     text_.erase(text_.begin()+cursor_);
20141009 03:43:32< shadowm> typedef Uint32 char_t;
20141009 03:43:44< shadowm> typedef std::vector<char_t> string;
20141009 03:44:15< shadowm> So assumedly, it's a Unicode-safe operation and the problem stems from *something* at the rendering phase.
20141009 03:47:58-!- ancestral [~ancestral@12.23.74.29] has joined #wesnoth-dev
20141009 03:58:38< shadowm> 1. On valgrind, the crash doesn't happen.
20141009 03:59:30< shadowm> 2. valgrind found something, just as I expected: https://gist.githubusercontent.com/shikadilord/cce4628bde4ebe79bc8b/raw/1ab760263a4699e9f206efecf7ef41c0b3503e06/gistfile1.txt
20141009 04:00:21< shadowm> We might have a SDL_ttf bug in our hands.
20141009 04:00:48< shadowm> That'd be precisely what I'm afraid of. :\
20141009 04:05:31< shadowm> (As a side note, our test case involves combining characters.)
20141009 04:28:55-!- SZ_Bot [~SZ_Bot@27.83.235.80.dyn.estpak.ee] has joined #wesnoth-dev
20141009 04:29:08-!- Ravana_ [~SZ_Bot@27.83.235.80.dyn.estpak.ee] has quit [Disconnected by services]
20141009 04:34:06-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20141009 04:50:59< shadowm> In general, any string starting with U+0020 SPACE followed by U+0361 COMBINING DOUBLE INVERTED BREVE suffices.
20141009 04:51:17< shadowm> The 'starting' part is important.
20141009 04:52:32< iceiceice> shadowm: i tried a test of my own,
20141009 04:53:14< iceiceice> i made it so that when you delete a character from the text box, it dumps the text contents to console
20141009 04:53:16< iceiceice> https://github.com/cbeck88/wesnoth/compare/debug_unicode_textbox_20815
20141009 04:53:17< shadowm> This makes thins considerably easier: if (chunk.text.find("\315\241") != std::string::npos) { asm("int3"); }
20141009 04:53:20< iceiceice> this is what i got:
20141009 04:53:31< iceiceice> Deleting: text is initially '( ͡° ͜ʖ ͡°) '
20141009 04:53:31< iceiceice> Deleting: text is now ' ͡° ͜ʖ ͡°) '
20141009 04:53:31< iceiceice> *** Error in `./wesnoth/wesnoth': free(): invalid size: 0x0000000004f13170 ***
20141009 04:53:55< iceiceice> so there's not obviously a problem in the deleting step, like you suspected
20141009 04:54:18< shadowm> Yeah, I ruled out that first, then wrote a dumping probe: https://dl.dropboxusercontent.com/u/21371130/screenshots/chars.png
20141009 04:55:03< iceiceice> hmm so the problem is not really related to deleting?
20141009 04:55:11< iceiceice> its just if certain characters appear in some order?
20141009 04:55:19< shadowm> Nope, just rendering the two characters I mentioned above.
20141009 04:55:48< iceiceice> i wonder if the sdl people know about this
20141009 04:56:24< shadowm> "fooo       " + the above doesn't suffice to crash or make valgrind complain.
20141009 04:57:02< shadowm> So I suspect SDL_ttf is miscalculating something (and corrupting the malloc arena as a side effect).
20141009 04:57:23< shadowm> I'm still investigating, even got a local SDL_ttf build with full debug symbols.
20141009 05:00:58< shadowm> OK, broke into the debugger right before entering SDL_ttf.
20141009 05:02:09-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has quit [Remote host closed the connection]
20141009 05:06:58-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has joined #wesnoth-dev
20141009 05:10:45< shadowm> Hm.
20141009 05:12:38< shadowm> Uin32 *dst, *dst_check; /* ... */ *dst++ |= pixel | (alpha << 24); /* precondition: dst < dst_check */
20141009 05:12:48< shadowm> dst_check = (Uint32*)textbuf->pixels + textbuf->pitch/4 * textbuf->h;
20141009 05:14:12< shadowm> textbuf is allocated by SDL_ttf using SDL_AllocSurface() (an alias for SDL_CreateRGBSurface).
20141009 05:16:11< shadowm> But valgrind says there is an invalid read and consequence write at the *dst++ point.
20141009 05:16:17< shadowm> *consequent
20141009 05:16:29< shadowm> There's no way valgrind could be wrong about that...
20141009 05:18:05< shadowm> The only change done to textbuf beforehand is invoking SDL_FillRect on it .
20141009 05:18:21-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: Bye for now]
20141009 05:18:47< shadowm> Sure, the width and height passed to SDL_CreateRGBSurface() come from SDL_ttf.
20141009 05:19:02< shadowm> But they are valid: { w = 15, h = 19 }
20141009 05:22:21< shadowm> Just to double check, what SDL_ttf sees at this point is Uint16 text[] = { 0x0020, 0x0361, 0x0000 }
20141009 05:24:59< shadowm> (Side note: here's a simple fix: let's replace all SDL_ttf paths with ttext (Pango/Cairo wrapper) and we'd solve this and a hundred other bugs. :p)
20141009 06:07:53-!- Sulfur [~Miranda@p5B00887A.dip0.t-ipconnect.de] has joined #wesnoth-dev
20141009 06:12:41-!- Ayne [~Ayne@55d447d2.access.ecotel.net] has joined #wesnoth-dev
20141009 06:14:43< Ayne> mattsc, thunderstruck: Saw the bug report, I'm not really up to date but I can try to take a look. Maybe tonight, but probably not until Saturday, my dissertation and work are keeping me pretty busy  :/
20141009 06:17:34< shadowm> Oh god I failed to notice this is a nested loop.
20141009 06:17:40 * shadowm facepalms.
20141009 06:18:06< shadowm> How do I skip a breakpoint in gdb until a condition yields true?
20141009 06:19:26< shadowm> (Of course everything is fine atm because I'm probably looking at the U+0020 SPACE glyph rendering step, not the U+0361 COMBINING DOUBLE INVERTED BREVE of doom.)
20141009 06:20:04 * vultraz clones the website
20141009 06:20:15< shadowm> It's not the website.
20141009 06:21:17< shadowm> Okay, I can until-out of the nested loop.
20141009 06:22:12 * vultraz clones the website repo
20141009 06:26:53< shadowm> Either I'm failing at gdb or... :\
20141009 06:27:26< shadowm> Oh, SIGTRAPs generate permanent breakpoints.
20141009 06:27:48< shadowm> No wait, I'm actually at my breakpoint again.
20141009 06:33:59-!- Ayne [~Ayne@55d447d2.access.ecotel.net] has left #wesnoth-dev []
20141009 06:35:12< shadowm> Uint32* dst = 0x5174ad8   Uint32* dst_check = 0x5a98aa4
20141009 06:36:09< shadowm> That's a.... like 9.3 MiBs difference.
20141009 06:37:41< shadowm> That can't be right.
20141009 06:39:39< shadowm> Right, I had checked dst during evaluation of the rhs expression, that's why. :p
20141009 06:41:11< shadowm> Aha!
20141009 06:41:33< shadowm> glyph->minx is negative, SDL_ttf is writing *before* the allocated block, not past the allocation like I originally thought.
20141009 06:42:30< shadowm> Somebody should've added a check to make sure dst >= the base pointer.
20141009 06:44:39< shadowm> I'm guessing those two bytes preceding the allocation might belong to an essential bookkeeping structure used by either libc or SDL.
20141009 06:45:24< shadowm> Oops.
20141009 06:45:45< shadowm> Not two bytes. The difference is actually eight bytes.
20141009 06:54:47-!- boucman_work [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20141009 07:17:51< vultraz> shadowm: this sucks so far
20141009 07:37:42-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 245 seconds]
20141009 07:42:11-!- Sulfur [~Miranda@p5B00887A.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20141009 07:45:29-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Quit: Leaving]
20141009 07:59:25< shadowm> vultraz: Yeah, I'm sorry for wasting an hour of your time by spending two days writing that piece of crap.
20141009 07:59:39< shadowm> There was clearly no effort involved and it's my fault.
20141009 08:08:56-!- Necrosporus_ [~Necrospor@unaffiliated/necrosporus] has quit [Ping timeout: 244 seconds]
20141009 08:11:46-!- Sulfur [~Miranda@p5B00887A.dip0.t-ipconnect.de] has joined #wesnoth-dev
20141009 08:19:37-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20141009 08:20:41-!- molgrum [~molgrum@212.85.89.43] has joined #wesnoth-dev
20141009 08:22:25-!- Necrosporus_ [~Necrospor@unaffiliated/necrosporus] has joined #wesnoth-dev
20141009 08:28:55-!- ancestral [~ancestral@12.23.74.29] has quit [Quit: i go nstuf kthxbai]
20141009 08:31:01-!- ancestral [~ancestral@12.23.74.29] has joined #wesnoth-dev
20141009 08:31:01-!- ancestral [~ancestral@12.23.74.29] has quit [Client Quit]
20141009 08:35:37< vultraz> shadowm: it's alright. I'll improve it
20141009 08:37:00-!- mjs-de [~mjs-de@f048215180.adsl.alicedsl.de] has joined #wesnoth-dev
20141009 08:44:35-!- Coffee_irc [~david@ppp118-210-39-35.lns20.adl2.internode.on.net] has joined #wesnoth-dev
20141009 09:09:04-!- bumbadadabum [~bumbadada@wcw-stud-145-109-11-182.wireless.uva.nl] has joined #wesnoth-dev
20141009 09:16:52-!- Necrosporus_ is now known as Necrosporus
20141009 09:24:56-!- kex [~kex@77.29.199.94] has joined #wesnoth-dev
20141009 09:29:22-!- kex [~kex@77.29.199.94] has quit [Ping timeout: 240 seconds]
20141009 09:53:49-!- zookeeper [zookeeper@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20141009 09:57:34-!- Anakonda [Anakonda@87-92-217-228.bb.dnainternet.fi] has joined #wesnoth-dev
20141009 10:10:10-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20141009 10:27:55< vultraz> who manages the youtube channel?
20141009 10:28:01< zookeeper> shadowm, vultraz, bumbadadabum: WRT weldyn besieged, it's worth noting that the surrounding terrain should probably be made similar to the map of approaching weldyn (although the scales are different, so i'd omit for example the mountains).
20141009 10:28:11< zookeeper> we have a youtube channel?
20141009 10:28:14< bumbadadabum> ^
20141009 10:28:20< vultraz> Yes
20141009 10:28:26< vultraz> it's the one the trailer was posted on
20141009 10:28:33< vultraz> the old trailer from GCI2010
20141009 10:28:33< bumbadadabum> no idea who maintains it
20141009 10:28:35< bumbadadabum> I mean I could
20141009 10:28:37< bumbadadabum> but still
20141009 10:28:43< vultraz> https://www.youtube.com/channel/UCVS4swbN51HNp7j7NdZXvVA
20141009 10:28:52< vultraz> no one's bothered to do any maintence on it
20141009 10:29:23< bumbadadabum> it seems to be someone with the account name razvan
20141009 10:30:08< bumbadadabum> I mean, we could just make a new one
20141009 10:30:38< vultraz> To post what. We do 0 advertising/promoting or anything involving video
20141009 10:31:11< vultraz> I suppose the new trailer, but turuk is away
20141009 10:31:21< bumbadadabum> idk
20141009 10:31:22< vultraz> Without the trailer, we can't move forward on Steam
20141009 10:31:26< bumbadadabum> we could do more video things
20141009 10:31:44< bumbadadabum> showing off new features
20141009 10:31:46< bumbadadabum> that sort of thing
20141009 10:31:53< vultraz> We could
20141009 10:31:56< vultraz> but who would do it?
20141009 10:32:05< bumbadadabum> idk I could when I have more time
20141009 10:32:11< bumbadadabum> speaking of time and wesnoth
20141009 10:32:30< bumbadadabum> vultraz: when are you going to take some time to finish the animation update thingy for our add-on like you promised
20141009 10:32:33< bumbadadabum> it's still not on 1.12
20141009 10:32:44< vultraz> I thought I did that
20141009 10:32:52< bumbadadabum> did you?
20141009 10:33:01< bumbadadabum> well but you didn't fix all the stub animationwml
20141009 10:33:48-!- irker820 [~irker@fehu.ai0867.net] has quit [Quit: transmission timeout]
20141009 10:34:10< bumbadadabum> also, it's not on there
20141009 10:34:47-!- kex [~kex@77.29.199.94] has joined #wesnoth-dev
20141009 10:34:54< bumbadadabum> I have 3 assignments due on Monday
20141009 10:35:04< bumbadadabum> but I might try to get back next week
20141009 10:37:20< zookeeper> i wish that whoever does the next trailer will show more of the game instead of focusing on absurdly zoomed-in close-ups of individual fancy animations
20141009 10:37:44< zookeeper> you can't make an action-packed hollywood game trailer of wesnoth
20141009 10:37:48< bumbadadabum> ^
20141009 10:37:52< bumbadadabum> well yes you can
20141009 10:38:00< bumbadadabum> just show 50 seconds of TLU gameplay
20141009 10:40:22< vultraz> no, no
20141009 10:40:28< vultraz> 1 hour of LotI gameplay
20141009 10:40:41< bumbadadabum> do you want wesnoth to die
20141009 10:41:32< bumbadadabum> we also need to stress more on the community
20141009 10:41:36< bumbadadabum> which is what makes wesnoth fun
20141009 10:42:06< vultraz> Yeah
20141009 10:43:40< vultraz> Problem in, all this stuff, trailers, videos, website upgrades, need manpower, and right now everyone is doing this on their own time
20141009 10:44:05< vultraz> I wish there were some way to keep Wesnoth free and open source, but for us to still make a profit off it
20141009 10:50:00-!- bumbadadabum [~bumbadada@wcw-stud-145-109-11-182.wireless.uva.nl] has quit [Ping timeout: 246 seconds]
20141009 10:53:26-!- bumbadadabum [~bumbadada@wcw-stud-145-109-11-182.wireless.uva.nl] has joined #wesnoth-dev
20141009 11:21:43-!- Kexoth [~kex@77.28.5.99] has joined #wesnoth-dev
20141009 11:23:57-!- kex [~kex@77.29.199.94] has quit [Ping timeout: 246 seconds]
20141009 11:54:19-!- Coffee_irc [~david@ppp118-210-39-35.lns20.adl2.internode.on.net] has quit [Quit: Konversation terminated!]
20141009 12:26:41-!- Sulfur [~Miranda@p5B00887A.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20141009 12:36:16-!- Kexoth [~kex@77.28.5.99] has quit [Remote host closed the connection]
20141009 12:37:40-!- kex [~kex@77.28.5.99] has joined #wesnoth-dev
20141009 12:41:55-!- prkc [~prkc@catv-80-99-60-5.catv.broadband.hu] has joined #wesnoth-dev
20141009 12:42:10-!- Kexoth [~kex@77.28.5.99] has joined #wesnoth-dev
20141009 12:42:15-!- kex [~kex@77.28.5.99] has quit [Ping timeout: 244 seconds]
20141009 13:08:21-!- bumbadadabum [~bumbadada@wcw-stud-145-109-11-182.wireless.uva.nl] has quit [Ping timeout: 272 seconds]
20141009 13:27:51-!- Sulfur [~Miranda@p5B00887A.dip0.t-ipconnect.de] has joined #wesnoth-dev
20141009 13:41:02-!- cib0 [~cib@p5DC7598F.dip0.t-ipconnect.de] has joined #wesnoth-dev
20141009 13:44:39-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20141009 13:44:47-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20141009 13:48:24< mattsc> zookeeper: we need your help with the 1.12 announcement which shadowm is putting together?  http://git.io/UBG2Pg
20141009 13:48:35< mattsc> shikadibot: seen lipkab
20141009 13:48:35< shikadibot> mattsc: The person with the nick lipkab 1d ago they left with the message: Quit: .
20141009 13:49:25< zookeeper> yeah that's been sitting open in my browser for a while...
20141009 13:50:14< mattsc> okay, cool, just making sure you saw it
20141009 13:51:57< mattsc> thunderstruck, Ivanovic, fabi_, lipkab: did you see it too?  You are all called out by name for help with specific sections.   http://git.io/UBG2Pg
20141009 13:52:17< fabi_> hi mattsc
20141009 13:52:23< mattsc> hi fabi_ 
20141009 13:52:34< zookeeper> so i take it that i'll just edit the announcement text as i please rather than let the edits go through someone?
20141009 13:52:43-!- Kexoth [~kex@77.28.5.99] has quit [Remote host closed the connection]
20141009 13:53:08< mattsc> zookeeper: either way is fine; whatever is easier for you
20141009 13:53:29-!- kex [~kex@77.28.5.99] has joined #wesnoth-dev
20141009 13:53:44< zookeeper> well, text edits are easy enough to do via the web interface so i guess i'll just do them directly myself
20141009 13:54:08-!- kex [~kex@77.28.5.99] has quit [Read error: Connection reset by peer]
20141009 13:54:29-!- kex [~kex@77.28.5.99] has joined #wesnoth-dev
20141009 13:54:34< mattsc> okay
20141009 13:55:46< fabi_> mattsc: I have some time now. Let me see what is to be done.
20141009 13:55:48< mattsc> fabi_: anything I can help with with LoW?  Or anything else?
20141009 13:55:57< mattsc> fabi_: great
20141009 13:56:25< mattsc> For anybody who has not seen that yet: We’re trying to get 1.11.17 out on Saturday, make that 1.12-RC1 and then hopefully move to 1.12.0 quickly after that.
20141009 13:56:32< fabi_> mattsc: Is the resulting page anywhere to be seen?
20141009 13:57:04< mattsc> fabi_: no, you can either save a local copy which gives you the page without images, or clone the repo (30MB)
20141009 13:57:31< zookeeper> i have a shortish list of (not entirely up-to-date; missing for example the sighted change) potentially content-breaking WML changes between 1.10-1.12, i guess those should just go into a sticky thread upon release with a link from the announcement.
20141009 13:58:02< mattsc> zookeeper: yes, that make sense
20141009 13:58:11< mattsc> don’t we have that sticky thread already ?
20141009 14:01:40< mattsc> zookeeper: http://forums.wesnoth.org/viewtopic.php?f=21&t=36057
20141009 14:01:56< zookeeper> oh. yes, we do :P
20141009 14:02:06-!- riksteri [~riksteri@dsl-tkubrasgw3-54f96b-216.dhcp.inet.fi] has joined #wesnoth-dev
20141009 14:02:54-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141009 14:10:56-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: Computer's napping]
20141009 14:12:15< zookeeper> "You must be on a branch to make or propose changes to this file" oh what's this now
20141009 14:30:20-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20141009 14:47:38< fabi_> mattsc: A line about the possibility to change difficult levels during an ongoing campaign.
20141009 14:48:46< mattsc> fabi_: good point
20141009 14:55:03< fabi_> mattsc: It is a little difficult. There are so many changes that all won't make a good announcement anymore. Which of them are important?
20141009 14:56:11< mattsc> fabi_: I don’t think there’s a general rule to be applied here.  We just need to make a judgment call.
20141009 15:00:21< fabi_> rofl
20141009 15:00:36< fabi_> "<p>Something something editor. [TODO: fabi] Also, new terrains. [TODO: ?]</p>"
20141009 15:07:13< mattsc> iceiceice: do you know if this is still an issue:  https://gna.org/bugs/?21647
20141009 15:14:10< mattsc> There is no way of setting the layer for labels, right?
20141009 15:14:56< mattsc> As in, they will always be in front of units?  Meaning that this bug should be marked as “won’t fix”:  https://gna.org/bugs/?21936
20141009 15:18:30< zookeeper> "experience and life could not be seen" <- well, i guess labels could go in front of units but behind bars
20141009 15:18:43< zookeeper> doesn't sound like an important thing in any case
20141009 15:19:32< mattsc> zookeeper: agreed; and it is already marked as ‘2 - Minor’.  But I think it should be reclassified as a feature request, not a bug.
20141009 15:23:13< mattsc> fabi_: still around?
20141009 15:30:41-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20141009 15:31:35< fabi_> mattsc: yes
20141009 15:31:51< mattsc> fabi_: anything I can help with?  With LoW or otherwise?
20141009 15:33:00< fabi_> mattsc: Not yet. I need to catch some food and then start to work on LoW again. Maybe in 3 hours? Are you still around then?
20141009 15:33:40< mattsc> I will be generally around, although today’s going to be quite busy at work.  But I’ll stay online.
20141009 15:33:53< fabi_> Okay, I will contact you.
20141009 15:34:03< mattsc> Sounds good.  TTYL.
20141009 15:38:26-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has joined #wesnoth-dev
20141009 15:51:50< iceiceice> mattsc: as far as I know 21647 is still an issue
20141009 15:52:05< iceiceice> i think that was the one shadowm nominated as a blocker
20141009 15:52:52< iceiceice> 21936: i dont know enough about the display code to know if it has to be that way or not
20141009 15:53:08< mattsc> iceiceice: yes, he did, “assuming it hasn't been inadvertently fixed in the meantime”
20141009 15:53:23< iceiceice> it reminds me a bit of this bug report from tekelili
20141009 15:53:26< iceiceice> https://gna.org/bugs/?22373
20141009 15:54:51< mattsc> I don’t know if it has to be that way either, but I still think it’s a feature request rather than a bug.  It should also probably not be assigned to fabi_ (21936, I mean)
20141009 15:55:24< mattsc> because it’s not really a LoW issue
20141009 15:56:08< fabi_> yes
20141009 15:56:12< mattsc> iceiceice: do you know who’d be qualified to look into 21647 ?
20141009 15:56:14< fabi_> it is not really mine
20141009 15:57:18< iceiceice> mattsc: its hard to speculate
20141009 15:57:23< mattsc> fabi_: I’ll wait a little longer in case there are other comments, then I’ll change it to a FR and unassign it from you
20141009 15:57:50< iceiceice> mattsc: if you want it to get solved, what i would do is try to bisect it, then when you get to a commit / group of commits try to assign it to that person
20141009 15:58:02-!- irker524 [~irker@fehu.ai0867.net] has joined #wesnoth-dev
20141009 15:58:02< irker524> wesnoth: Lari Nieminen website:master 4606639267aa / start/1.12/template.html: Updates to art and content creation http://git.io/Pg1G4g
20141009 15:58:09< iceiceice> alternatively, try to figure out what part of code draws ellipses and look for anything squirrelly
20141009 15:58:30< fabi_> mattsc: I have already asked for some input for redesigning the map label system via the dev-ml some few years ago.
20141009 15:58:56< mattsc> iceiceice: I wouldn’t know how to recognize a squirrel in the replay code if it bit me in the nose …  :P
20141009 15:59:05< fabi_> But there seemed to be no interest since I got not a single response. 
20141009 15:59:32< iceiceice> fabi: maybe try a forum post?
20141009 15:59:43< iceiceice> actually i have a hunch you would get a lot of responses
20141009 15:59:52< mattsc> iceiceice: and bisecting is difficult unless there’s an easy way to replicate.  Even what you suggest still sounds pretty tedious.
20141009 16:00:07< iceiceice> yeah graphical stuff is a pita
20141009 16:00:24-!- ancestral [~ancestral@12.23.74.29] has joined #wesnoth-dev
20141009 16:00:27< iceiceice> ok, so i grepped for the word "ellipse",
20141009 16:01:32< iceiceice> i think the best candidate is the "redraw_unit" function, 
20141009 16:01:49< iceiceice> unless theres some evidence that the ellipse values in unit configs are actually getting messed up , which you should be able to see from save files
20141009 16:01:52< mattsc> fabi_: I don’t think that redoing the label system is a high priority for 1.12 (well, it would break the feature freeze anyway), but we could think about it for 1.13/1.14
20141009 16:02:06< fabi_> yes
20141009 16:02:34< fabi_> We have lived with the current one for ages. No need to take action and fix the bug for 1.12.
20141009 16:02:39< fabi_> I would just postpone it.
20141009 16:02:56< mattsc> ok
20141009 16:03:16< iceiceice> hmm
20141009 16:03:32< mattsc> iceiceice: well, I don’t think I have time to look into this for the next few days, but maybe next week
20141009 16:03:43< iceiceice> apparently there is a backgrund and foreground part of the ellipse?
20141009 16:03:59< fabi_> yes
20141009 16:05:11-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141009 16:06:27< fabi_> timotei: ping
20141009 16:09:34< iceiceice> mattsc: i think this bug has existed for a long time
20141009 16:09:52< mattsc> iceiceice: that wuodn’t surprise me
20141009 16:09:54< iceiceice> i'm not sure if its likely that the person who introduced is still active
20141009 16:10:05< mattsc> uh … *wouldn’t
20141009 16:10:56< iceiceice> boucman: do you have an opinion about this?
20141009 16:10:57< iceiceice> https://github.com/wesnoth/wesnoth/pull/285
20141009 16:12:18< irker524> wesnoth: gfgtdf wesnoth:master deb5b944e8b4 / src/game_preferences.cpp: fix default login on windows http://git.io/4DSsTA
20141009 16:12:20< irker524> wesnoth: gfgtdf wesnoth:master 65a07832e46e / src/game_preferences.cpp: Merge pull request #294 from gfgtdf/utf16_to_utf32 http://git.io/dWLPFw
20141009 16:13:48< irker524> wesnoth: gfgtdf wesnoth:master 51a0d49eafaf / src/game_preferences.cpp: add static to a local function. http://git.io/5LUJMA
20141009 16:15:51-!- ancestral [~ancestral@12.23.74.29] has quit [Quit: i go nstuf kthxbai]
20141009 16:17:31-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20141009 16:25:23-!- Ayne [~Ayne@55d447d2.access.ecotel.net] has joined #wesnoth-dev
20141009 16:27:42< irker524> wesnoth: fendrin wesnoth:master e98237b11c80 / src/ (8 files in 4 dirs): Moved the editor controller into a new subdirectory. http://git.io/pWW9CA
20141009 16:31:15-!- ancestral [~ancestral@17.114.45.53] has joined #wesnoth-dev
20141009 16:45:36-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20141009 16:48:07-!- boucman_work [~rosen@wesnoth/developer/boucman] has quit [Ping timeout: 245 seconds]
20141009 16:48:36< shadowm> zookeeper: I'd rather see that list of API changes made into a wiki page.
20141009 16:49:06< fabi_> hi shadowm
20141009 16:49:38< shadowm> fabi_: Ignore the new terrains TODO, I ended up doing it myself in the players section and forgetting I had left the TODO placeholder elsewhere.
20141009 16:49:58< fabi_> noted
20141009 16:52:12-!- kex [~kex@77.28.5.99] has quit [Remote host closed the connection]
20141009 16:52:38< shadowm> zookeeper: But HttT's portrait art for the protagonists was completely redone this cycle.
20141009 16:53:53< shadowm> As for why I didn't take a single screenshot involving portraits, it's because they are downscaled to be very tiny at 1024x768, get jagged edges, and in vultraz's words, would "make us look bad".
20141009 16:54:30< shadowm> (Pretty sure there's an open bug on the matter.)
20141009 16:55:03-!- cib0 [~cib@p5DC7598F.dip0.t-ipconnect.de] has quit [Ping timeout: 246 seconds]
20141009 16:55:18< fabi_> shadowm: Don't do screenshots in 1024.
20141009 16:55:25-!- cib0 [~cib@p5DC7598F.dip0.t-ipconnect.de] has joined #wesnoth-dev
20141009 16:55:50< shadowm> Ivanovic has always done screenshots in 1024x768 and I don't know what the next larger standard resolution is.
20141009 16:55:58< zookeeper> shadowm, eh, the new portraits were ported to 1.10.7
20141009 16:56:24< shadowm> Oh... right.
20141009 16:56:40< shadowm> Well, morale of the story, don't backport shiny new portraits to stable branches. :p
20141009 16:57:25< zookeeper> yeah i did wonder for a while which way i should phrase it...
20141009 16:57:49< zookeeper> feel free to change it if you want, i just felt this way is the most technically accurate description
20141009 16:58:09< shadowm> iceiceice: If my memory serves, https://gna.org/bugs/?22373 dates back to version 1.0. I'm pretty sure it was an issue with a TRoW scenario using [item] for custom "villages".
20141009 16:58:23< shadowm> (1.0, not 1.10.)
20141009 17:03:21-!- cib0 [~cib@p5DC7598F.dip0.t-ipconnect.de] has quit [Ping timeout: 258 seconds]
20141009 17:04:25-!- boucman_work [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20141009 17:09:05-!- kex [~kex@77.28.5.99] has joined #wesnoth-dev
20141009 17:13:57-!- kex [~kex@77.28.5.99] has quit [Ping timeout: 245 seconds]
20141009 17:15:39-!- boucman_work [~rosen@wesnoth/developer/boucman] has quit [Quit: Leaving.]
20141009 17:16:45-!- Ayne_ [~Ayne@55d41d0d.access.ecotel.net] has joined #wesnoth-dev
20141009 17:17:09-!- Ayne [~Ayne@55d447d2.access.ecotel.net] has quit [Ping timeout: 258 seconds]
20141009 17:18:35-!- Ayne_ [~Ayne@55d41d0d.access.ecotel.net] has quit [Client Quit]
20141009 17:18:52-!- Ayne_ [~Ayne@55d41d0d.access.ecotel.net] has joined #wesnoth-dev
20141009 17:23:43< shadowm> iceiceice: Okay, I confirm there is different behavior for #22373 between 1.10.7+dev and 1.13.0-dev.
20141009 17:25:31-!- Sulfur [~Miranda@p5B00887A.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20141009 17:27:42-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Read error: Connection reset by peer]
20141009 17:30:59-!- kex [~kex@77.28.5.99] has joined #wesnoth-dev
20141009 17:33:21-!- cib0 [~cib@p5DC7598F.dip0.t-ipconnect.de] has joined #wesnoth-dev
20141009 17:35:37-!- kex [~kex@77.28.5.99] has quit [Ping timeout: 245 seconds]
20141009 17:52:44-!- kex [~kex@77.28.5.99] has joined #wesnoth-dev
20141009 17:57:45-!- kex [~kex@77.28.5.99] has quit [Ping timeout: 260 seconds]
20141009 18:14:26-!- kex [~kex@77.28.5.99] has joined #wesnoth-dev
20141009 18:18:57-!- kex [~kex@77.28.5.99] has quit [Ping timeout: 245 seconds]
20141009 18:24:27-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20141009 18:27:37-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 260 seconds]
20141009 18:35:44-!- riksteri [~riksteri@dsl-tkubrasgw3-54f96b-216.dhcp.inet.fi] has quit [Read error: Connection reset by peer]
20141009 18:36:09-!- kex [~kex@77.28.5.99] has joined #wesnoth-dev
20141009 18:38:09-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141009 18:41:30-!- kex [~kex@77.28.5.99] has quit [Ping timeout: 272 seconds]
20141009 18:44:58-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20141009 18:49:29< shadowm> fabi_: You once said that src/editor/ files kept the "editor_" prefix to avoid clashes with MSVC++, but if that were the case then how would you explain the triplicate file names found in src/gui/{auxiliary/window_builder/,auxiliary/widget_definition/,widgets/}?
20141009 18:49:42< shadowm> s/with/in/
20141009 18:49:47-!- Ayne_ [~Ayne@55d41d0d.access.ecotel.net] has left #wesnoth-dev []
20141009 18:53:19< irker524> wesnoth: Ignacio R. Morelle wesnoth:master 395d15c46741 / projectfiles/CodeBlocks/wesnoth.cbp: CodeBlocks project file update http://git.io/PtT9bg
20141009 18:57:50-!- kex [~kex@77.28.5.99] has joined #wesnoth-dev
20141009 19:00:08-!- EdB [~edb@89-158-11-138.rev.numericable.fr] has joined #wesnoth-dev
20141009 19:03:02-!- kex [~kex@77.28.5.99] has quit [Ping timeout: 272 seconds]
20141009 19:04:17< thunderstruck> mattsc: Thanks for pointing that draft to me. I'll try to add my share soon. Most likely tomorrow.
20141009 19:08:37< mattsc> thunderstruck: thanks
20141009 19:11:16< irker524> wesnoth: mattsc wesnoth:master ead1916fe916 / projectfiles/Xcode/Wesnoth.xcodeproj/project.pbxproj: Xcode project file update http://git.io/3oMA8w
20141009 19:15:45-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20141009 19:19:34-!- kex [~kex@77.28.5.99] has joined #wesnoth-dev
20141009 19:20:57< Ivanovic> mattsc: is the start page available in the website repo?
20141009 19:21:04< Ivanovic> if it is, then there should be a working checkout on wesnoth.org
20141009 19:21:57< Ivanovic> at least in the past it was auto synced at wesnoth.org
20141009 19:22:52< mattsc> Ivanovic: I see.  Well, I just cloned the repo.  As shadowm said, it’s really not a big deal.
20141009 19:23:33-!- ancestral is now known as ancestral_lunch
20141009 19:24:09-!- kex [~kex@77.28.5.99] has quit [Ping timeout: 246 seconds]
20141009 19:24:24< Ivanovic> strangely my system does not want to "compile" the template right now...
20141009 19:28:08< shadowm> wfm
20141009 19:28:15-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141009 19:29:07< fabi_> shadowm: I can't explain it. No idea about msvc nor about window_builder stuff. I was just told to keep filenames unique. Maybe that restriction is lifted in the meanwhile.
20141009 19:29:49< shadowm> I think that restriction had already been lifted at the point we discussed this and we didn't notice.
20141009 19:30:20< shadowm> Perhaps it was an issue with some older version like MSVC++ 6.0, which would not compile Wesnoth nowadays for other reasons.
20141009 19:31:14< fabi_> I am fine with renaming the files below editor/
20141009 19:34:23-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20141009 19:40:57-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 246 seconds]
20141009 19:41:17-!- kex [~kex@77.28.5.99] has joined #wesnoth-dev
20141009 19:46:06-!- kex [~kex@77.28.5.99] has quit [Ping timeout: 272 seconds]
20141009 19:47:40< mattsc> Hi Ivanovic.  We still need a decision as to whether allow translation updates to be handled via PRs on github.  As I understand it the options are:
20141009 19:47:44< mattsc> 1. Don’t allow them
20141009 19:48:35< mattsc> 2. Give the translation maintainers commit permissions on github
20141009 19:49:04< mattsc> Any opinion?
20141009 19:49:22< timotei> fabi_: pong
20141009 19:51:03-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141009 19:55:29 * Ivanovic does not care as long as i don't have to handle this stuff!
20141009 19:55:31< Ivanovic> ;)
20141009 19:56:17< shadowm> Let me be clear about this: only someone who manages translations can make a decision about handling translations.
20141009 19:56:18< mattsc> Ivanovic: how are the .po updates usually handled?  People send them to the maintainers, and the maintainers send them to you?
20141009 19:56:26< Ivanovic> correct
20141009 19:56:37< Ivanovic> the maintainer goes over them, checks if it fits the canon and all that stuff
20141009 19:56:52< Ivanovic> then they send the files to me and i ensure that they are merged against the latest pot files and compile correctly
20141009 19:57:24< mattsc> Ivanovic: in that case, given that you don’t want to handle PRs (right?), I would suggest that we do not allow translation PRs.
20141009 19:57:27< irker524> wesnoth: Ignacio R. Morelle wesnoth:master c91604a71c9c / src/animated.tpp: Fix Doxygen comment http://git.io/vJ0iRA
20141009 19:57:42< irker524> wesnoth: Ignacio R. Morelle wesnoth:1.12 28a84edba310 / changelog src/gui/dialogs/gamestate_inspector.cpp: gui2/tgamestate_inspector: Include array subscripts for WML array elements http://git.io/67Rdug
20141009 19:57:45< irker524> wesnoth: Ignacio R. Morelle wesnoth:1.12 28f2bca11842 / src/gui/dialogs/gamestate_inspector.cpp: gui2/tgamestate_inspector: Only display N/M page indicators when M > 1 http://git.io/HcbnwQ
20141009 19:57:46< Ivanovic> you are right as in "i don't want to handle them"
20141009 19:57:48< irker524> wesnoth: Ignacio R. Morelle wesnoth:1.12 d67b710359f5 / / (6 files in 4 dirs): Rename animated.i to animated.tpp https://github.com/wesnoth/wesnoth/commit/d67b710359f582cff39d40aa8b9ddce144924bd2
20141009 19:57:53< Ivanovic> though i am not sure how we should proceed
20141009 19:58:08< timotei> Isn't "let the maintainer merge up to latest pot files" easier?
20141009 19:58:09< Ivanovic> since i am rather inactive these days and can't even tell how often I merge in po files
20141009 19:58:13< timotei> And then just merge them in?
20141009 19:58:39< Ivanovic> timotei: the idea is/was to remove the technical hurdle from the maintainer so that the maintainer could be a non-techie
20141009 19:58:45< shadowm> So who else is in charge of merging translations?
20141009 19:58:49< Ivanovic> since translations afford a different skillset
20141009 19:59:06< Ivanovic> in the past espreon and AI0867 have supported this
20141009 19:59:13< Ivanovic> both are inactive right now...
20141009 19:59:22< shadowm> Great...
20141009 20:00:01< Ivanovic> jepp
20141009 20:00:24< mattsc> Ivanovic: well, the old (pre-PR) method still works, right?
20141009 20:00:30-!- gfgtdf [~chatzilla@e177166190.adsl.alicedsl.de] has joined #wesnoth-dev
20141009 20:00:55< mattsc> There are exactly 2 people who have submitted translation PRs (and they’ve been sitting there since July & August)
20141009 20:01:12< Ivanovic> the old method still works, yes
20141009 20:01:14< mattsc> We tell them to send those directly to the maintainer.
20141009 20:01:29< mattsc> … and follow the old route.  Then we close those PRs.
20141009 20:01:31-!- kex [~kex@77.28.5.99] has joined #wesnoth-dev
20141009 20:01:45< mattsc> And we state somewhere that we can’t deal with translation PRs.
20141009 20:02:01< mattsc> At least not for the time being.
20141009 20:02:36< shadowm> And we instruct everybody else to reject and close translation PRs on sight.
20141009 20:02:46< mattsc> sure
20141009 20:02:52< shadowm> Everybody else = developers.
20141009 20:03:34< shadowm> Hopefully with an appropriate canned message such as "sorry we do not take translation pull requests, see this wiki page" or something better worded than that.
20141009 20:03:54< mattsc> Well, yeah, we don’t have to be nasty about it.
20141009 20:04:01< mattsc> Ivanovic: does that sound acceptable to you?  As shadowm says, the final decision on this needs to come from you (or Espreon or AI0867).
20141009 20:06:41< timotei> Ivanovic: Well... github already tells if a patch can merge cleanly or not, right?
20141009 20:06:43< shadowm> In general, you don't want to give people the impression that they completely wasted their time when rejecting a PR. There's nothing more demotivating than not seeing your changes merged upstream before the case is closed.
20141009 20:06:50< timotei> There might be a step which I don't see in the flow.
20141009 20:07:42< timotei> The one that modified the file (should/can?) just rebase it and if conflicts arise, well.. there's GUI for that and you're the best to know how to handle them no?
20141009 20:07:55< timotei> It's not like code conflicts which need so much more attention.
20141009 20:08:02< shadowm> Because of the pot-update line-wrapping, odds are that po files would become unmergeable pretty quickly.
20141009 20:08:22< shadowm> Especially if other people have been altering the pot or targeted po in the meantime.
20141009 20:08:57< fabi_> timotei: hello
20141009 20:09:14< shadowm> Resolving conflicts is a rather error-prone task when handling C++, I'd expect it to be even worse if you were inundated by context in a foreign language you don't speak.
20141009 20:09:46< fabi_> timotei: Is there any chance for a version of the eclipse wml plugin/standalone for 1.12?
20141009 20:16:04< Ivanovic> timotei: i will not handle hand merging po files from languages i don't speak
20141009 20:16:09< Ivanovic> timotei: to much chance for error
20141009 20:16:42< Ivanovic> even if i find files where the string is somehow broken, e.g. because of a different number of \n at the beginning or end of the string i will just mark the strings fuzzy
20141009 20:17:10< Ivanovic> though in probably 99% of all cases the wrong number of \n is just a tiny oversight by the translator
20141009 20:21:37< mattsc> Ivanovic: so I interpret that as you agreeing (or at least not disagreeing) with the process I outlined above?
20141009 20:21:57< Ivanovic> the process is fine by me
20141009 20:22:06< Ivanovic> translations by mail i can somehow handle
20141009 20:22:45-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20141009 20:22:48< mattsc> Okay - I’ll take care of letting the two authors of those PRs know then.
20141009 20:23:38< mattsc> And we should probably put a note onto the translation howto wiki page (or whatever it’s called).  I can do that too.
20141009 20:24:44< Ivanovic> there pull requests are not mentioned as option
20141009 20:24:53< Ivanovic> (though they are not mentioned as "not an option" either)
20141009 20:25:13< mattsc> yeah, we should just state that specifically
20141009 20:28:18-!- ancestral_lunch [~ancestral@17.114.45.53] has quit [Quit: ancestral_lunch]
20141009 20:33:00< fabi_> mattsc: Do you have some time to talk about LoW 14?
20141009 20:33:50< mattsc> fabi_: I have about 15 min right now; or I’ll have some time in ~1.5 hours.
20141009 20:33:52< timotei> [23:09] < shadowm> Resolving conflicts is a rather error-prone task when handling C++, I'd expect it to be even worse if you were inundated by context in a foreign language you don't        prkc speak.
20141009 20:34:16< timotei> Well, I expect(ed) that the ones that modified something and doesn't merge cleanly, to rebase its change against it :)
20141009 20:34:37< timotei> fabi_: Yes sure!
20141009 20:34:47< timotei> I can take some time this weekend. What exactly is missing?
20141009 20:34:57< fabi_> mattsc: Okay, let's talk later.
20141009 20:35:14< mattsc> fabi_: okay, I’ll ping you when I am back
20141009 20:35:31< fabi_> timotei: I am not sure. I thought you would know what is to be done :-)
20141009 20:35:57< mattsc> fabi_: in the meantime, you could commit the other changes/bugs you’ve dealt with  ;)
20141009 20:36:44< fabi_> yeah
20141009 20:37:53< timotei> fabi_: Welll... :D I need to find the TODO/ideas file then :P
20141009 20:40:27-!- travis-ci [~travis-ci@ec2-184-73-55-78.compute-1.amazonaws.com] has joined #wesnoth-dev
20141009 20:40:27< travis-ci> wesnoth/wesnoth#4154 (master - c91604a : Ignacio R. Morelle): The build was broken.
20141009 20:40:27< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/37536345
20141009 20:40:27-!- travis-ci [~travis-ci@ec2-184-73-55-78.compute-1.amazonaws.com] has left #wesnoth-dev []
20141009 20:41:44< shadowm> iceicice: That doesn't look like it's my fault. ^
20141009 20:42:13< shadowm> From what I understand, the characterize_pathfinding_one WML unit test failed.
20141009 20:47:31< gfgtdf> shadowm: that happens often due to timeout you have to restart it
20141009 20:47:56< gfgtdf> (or ignore it)
20141009 20:48:37< shadowm> Restarting the job means it'll have to compile everything again (and thus potentially time out again during the build or tests), so I'd rather leave it alone since I don't need to know the result.
20141009 21:00:00-!- ancestral [~ancestral@17.114.45.53] has joined #wesnoth-dev
20141009 21:01:21-!- trewe [~trewe@bl20-13-172.dsl.telepac.pt] has joined #wesnoth-dev
20141009 21:05:53-!- gfgtdf_ [~chatzilla@f054164233.adsl.alicedsl.de] has joined #wesnoth-dev
20141009 21:07:42-!- gfgtdf [~chatzilla@e177166190.adsl.alicedsl.de] has quit [Ping timeout: 245 seconds]
20141009 21:07:49-!- gfgtdf_ is now known as gfgtdf
20141009 21:09:36-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20141009 21:09:55< iceiceice> shadowm: yeah for whatever reason, sometimes our builds run on a slower machine, and then some of the tests time out
20141009 21:10:10< iceiceice> the "characterize_pathfinding_one" test is usually the culprit...
20141009 21:10:30< iceiceice> i think i'm just going to disable that test, i dont plan to change any of the pathfinding code for at least a few months
20141009 21:10:38< iceiceice> and i dont know that anyone else is doing this
20141009 21:11:02< iceiceice> in general, if you got a timeout on some test and it doesn't seem related to your patch, just restart the build
20141009 21:13:23-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 240 seconds]
20141009 21:16:38-!- EdB [~edb@89-158-11-138.rev.numericable.fr] has quit [Quit: Konversation terminated!]
20141009 21:28:17< iceiceice> how do we upload files to mediawiki?
20141009 21:28:25< iceiceice> i did it once but i cant figure out how anymore
20141009 21:29:20< iceiceice> did it get blocked or something?
20141009 21:35:02< Crendgrim> Are you sure you did it once? I remember it being blocked completely.
20141009 21:41:58< shadowm> It's never been a thing in wesnoth.org.
20141009 21:42:44< shadowm> Maybe before 2007, but then that would be long enough to count as 'never'.
20141009 21:44:55< iceiceice> how did all the comics and such get uploaded on wiki.wesnoth.org then
20141009 21:45:19< shadowm> Comics?
20141009 21:45:33< iceiceice> http://wiki.wesnoth.org/Category:Comics
20141009 21:46:28< shadowm> Firefox offers a handy View Image Info command in the context menu, you get the real URL amongst other things
20141009 21:46:34< Crendgrim> They appear to be hosted at exong.net.
20141009 21:46:52< Crendgrim> (Unrelated, but those pages are valid candidates for pre-2007 :P )
20141009 21:46:53< iceiceice> i see
20141009 21:46:59-!- molgrum [~molgrum@212.85.89.43] has quit [Quit: Lämnar]
20141009 21:47:09< iceiceice> i guess i must have uploaded to forums, and linked from wiki or something
20141009 21:47:14< shadowm> There are also some images hosted on wesnoth.org.
20141009 21:48:22< shadowm> If you absolutely need an image to be hosted on wesnoth.org, you have to sacrifice 12 sheep, 6 pigs, and 3 goats in a bloodied altar at midnight.
20141009 21:48:44< shadowm> Then hold a black candle and call my name three times.
20141009 21:49:07< Crendgrim> Would giving you cookies also be an option?
20141009 21:49:07< shadowm> *on
20141009 21:49:19-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has joined #wesnoth-dev
20141009 21:49:35< shadowm> Only if they have chocolate in some form.
20141009 21:52:18< gfgtdf> the wiki says that we shouldn't use wstring becasue some plattforms doesnt support it. does soneone know which plattforms that are ?
20141009 22:00:32< thunderstruck> gfgtdf: I think it could be that it's supported by all the platforms but they implement it differently. E.g. different platforms may have different sizes for wide strings.
20141009 22:00:57< gfgtdf> thunderstruck: y, but that also teh case for vector<wchar_t> which replaces it
20141009 22:01:01< gfgtdf> the*
20141009 22:01:35< thunderstruck> Nevermind then.
20141009 22:02:05-!- thunderstruck [~zaibotren@cpc3-sgyl29-2-0-cust613.sgyl.cable.virginm.net] has quit [Quit: leaving]
20141009 22:08:12< mattsc> fabi_: I would have some time right now if you’re still around  (will be busy again in about an hour or a bit more)
20141009 22:09:09-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20141009 22:27:36-!- zookeeper [zookeeper@wesnoth/developer/zookeeper] has quit [Ping timeout: 272 seconds]
20141009 22:32:52-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20141009 22:33:24-!- Anakonda [Anakonda@87-92-217-228.bb.dnainternet.fi] has quit [Read error: Connection reset by peer]
20141009 22:33:42< shadowm> gfgtdf: You probably didn't need confirmation, but I just confirmed your GetUserName PR works on Windows.
20141009 22:34:55< shadowm> C:\src\wesnoth\src\serialization\unicode.cpp:202:13: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
20141009 22:34:59< shadowm> C:\src\wesnoth\src\serialization\unicode.cpp:207:18: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
20141009 22:35:10< shadowm> I also got these with gcc 4.5 there.
20141009 22:36:16< gfgtdf> shadowm: ty
20141009 22:38:34< gfgtdf> shadowm: hm it looks like i typed int32_t instead of uint32_t i can fix that.
20141009 22:38:50< irker524> wesnoth: gfgtdf wesnoth:master 9f58cac6d890 / src/serialization/unicode.cpp: fix signed unsigned comparison http://git.io/fTUUWw
20141009 22:39:30< gfgtdf> done
20141009 22:40:09< shadowm> Yeah, the warnings are gone.
20141009 22:42:44< fabi_> mattsc: ping
20141009 22:43:24< mattsc> hi fabi_ 
20141009 22:58:04-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has quit [Quit: tomreyn]
20141009 23:00:52-!- SigurdFD [SigurdFD@dynamic-acs-72-23-113-185.zoominternet.net] has joined #wesnoth-dev
20141009 23:01:14-!- LovAFK is now known as LovCAPONE
20141009 23:10:20-!- cib0 [~cib@p5DC7598F.dip0.t-ipconnect.de] has quit [Ping timeout: 260 seconds]
20141009 23:13:36-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20141009 23:16:02< LovCAPONE> mattsc: Hi I'm on the recall list bug as of tonight and I'll do my best to have it fixed by saturday
20141009 23:18:23< mattsc> LovCAPONE: thanks and that would be great!
20141009 23:19:31< shadowm> LovCAPONE: Did you read my comment on the dialog text padding PR?
20141009 23:21:24< shadowm> iceiceice: So I can't reproduce my pet bug with a simple SDL_ttf test program I just wrote. :|
20141009 23:21:41< shadowm> At least valgrind doesn't say anything.
20141009 23:21:54< mattsc> LovCAPONE: a couple more minor things FYI
20141009 23:22:13-!- Kexoth [~kex@77.28.7.157] has joined #wesnoth-dev
20141009 23:22:25< mattsc> We are supposed to keep lines in the changelog to 80 characters.  (I fixed up the 1.12 CL, but haven’t done 1.13)
20141009 23:23:12< mattsc> Also, I think one or two of the bugs you fixed have forum threads about them going along with them.  It’s always nice to make a comment to that effect in those threads as well.
20141009 23:23:42< mattsc> LovCAPONE: but mostly: thanks for fixing all those bugs!  :)
20141009 23:24:48-!- kex [~kex@77.28.5.99] has quit [Ping timeout: 260 seconds]
20141009 23:26:18< LovCAPONE> mattsc: Ok no problem
20141009 23:26:43< LovCAPONE> shadowm: Yes I juste read it now, I'll get on it after fixing the 1.12.0 bug I'm assigned to
20141009 23:29:17< gfgtdf> LovCAPONE: the bug that you want to fix is this one https://gna.org/bugs/?15545 ?
20141009 23:30:07< shadowm> (Btw, you might want to keep in mind afk nicks and notifications are frowned upon around here and in many other channels. If we need to know you are around, just set /away when you aren't and we can find out on our own.)
20141009 23:30:23< gfgtdf> LovCAPONE: note that that bug was fixed by aa2cf7d635c49246e50011fe6f2d3e71fa1ce017 which was undone in 24eab27b52d0177f58be819b6190551cc586cb70 becasue it caused a visual bug
20141009 23:31:56< LovCAPONE> shadowm: Oh ok didn't know that. Thanks for the info
20141009 23:31:58< mattsc> gfgtdf: yes, and I bisected the origin of the problem to 24eab27b
20141009 23:33:02-!- trewe [~trewe@bl20-13-172.dsl.telepac.pt] has quit [Quit: quit]
20141009 23:37:43-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: Bye for now]
20141009 23:38:29-!- happygrue [~Laptop@wesnoth/developer/wintermute] has joined #wesnoth-dev
20141009 23:42:36< shadowm> The fact is that in my test program, the glyph rendering wraps around horizontally instead of writing before the allocation. :|
20141009 23:45:49< shadowm> I'm sure there's a sensible explanation for that having to do with how Wesnoth sets up the font, SDL_ttf, or SDL.
20141009 23:46:23< shadowm> For example, we use hardware surfaces on X11? I had no idea.
20141009 23:47:15< shadowm> But that's not it.
20141009 23:48:58< LovCAPONE> gfgtdf: Ok yes looking at it right now
20141009 23:52:38< shadowm> I'm suddenly reassured to know our usage of SDL_ttf isn't really exception safe at the font initialization stage.
20141009 23:53:44< shadowm> Granted, it's a very small time window and there's only a calls to *probably* non-throwing code in between.
20141009 23:53:51< shadowm> *a couple of calls
20141009 23:54:35-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Quit: Ik ga weg]
20141009 23:56:49-!- mjs-de [~mjs-de@f048215180.adsl.alicedsl.de] has quit [Remote host closed the connection]
20141009 23:57:16< shadowm> We have a text cache.
20141009 23:57:46< shadowm> :?
20141009 23:59:46< gfgtdf> LovCAPONE: did you find teh reason for the bug ?
--- Log closed Fri Oct 10 00:00:56 2014