--- Log opened Wed Oct 26 00:00:35 2016
20161026 00:01:46-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has joined #wesnoth-dev
20161026 00:06:42< tad_carlucci> vultraz, Conditionally correct. https://blogs.msdn.microsoft.com/oldnewthing/20131011-00/?p=2953  or  stackoverflow.com/questions/2089083/pure-virtual-function-with-implementation
20161026 00:18:33-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 250 seconds]
20161026 00:20:26-!- Porusaka [~Polsaker@donger/wielder/Polsaker] has quit [Ping timeout: 250 seconds]
20161026 00:20:57-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20161026 00:24:32-!- atarocch [~atarocch@193.92.163.210] has quit [Remote host closed the connection]
20161026 00:25:02-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 250 seconds]
20161026 00:27:51-!- tad_carlucci [~lundberg@173.217.65.103] has quit [Quit: Off to resolve a merge conflict between the wife and husband branches of my real life.]
20161026 00:42:31< celticminstrel> [Oct 25@3:39:37pm] vultraz: DeFender1031: you could ask celticminstrel to add a lua wrapper for invalidate_layout
20161026 00:42:32< celticminstrel> Why would I want to do that? It seems kinda hacky.
20161026 00:42:34< celticminstrel> [Oct 25@3:56:17pm] DeFender1031: heck, the :inspect dialog changes size all the time depending on the content of what's being inspected
20161026 00:42:35< celticminstrel> [Oct 25@3:56:26pm] vultraz: not so much in master
20161026 00:42:37< celticminstrel> The dialog doesn't, but the vertical divider still moves all over the place. You could avoid it by initializing the whole treeview from the start, I suppose (rather than initializing subnodes on first click), but I'm not sure that's a good idea.
20161026 00:42:38< celticminstrel> [Oct 25@7:42:52pm] vultraz: celticminstrel: pure virtual functions have no implementation in the base class, correct?
20161026 00:42:38< celticminstrel> That's not exactly the right way of looking at it, but 99.999% of pure virtual functions that are not destructors indeed have no implementation. (Mind you, I don't generally recommend making your destructors pure virtual. I'd only do it if you have no other virtual methods but need the class to be polymorphic.) It's possible to provide an implementation anyway, though not inside the class declaration.
20161026 00:46:21 * celticminstrel notes that if the size_fixater is merged after 1.13.6 (unless it's directly after) then it would have to be updated to reflect my planned renames.
20161026 00:46:38< celticminstrel> ...also, I don't much like the name size_fixater, but I can't think of anything better either, so whatever.
20161026 00:47:57-!- irker764 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20161026 01:00:33< shadowm> size_lock?
20161026 01:06:48< Aginor> hmm
20161026 01:10:09-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has quit [Quit: ChipmunkV]
20161026 01:39:01-!- gfgtdf_ [~chatzilla@x4e36a5f1.dyn.telefonica.de] has joined #wesnoth-dev
20161026 01:41:02-!- gfgtdf [~chatzilla@x4e36315c.dyn.telefonica.de] has quit [Ping timeout: 250 seconds]
20161026 01:41:11-!- gfgtdf_ is now known as gfgtdf
20161026 01:54:41< mattsc> size_infatuator?  ;)
20161026 02:02:41< Aginor> size_stopgap
20161026 02:02:53-!- SpaceRobot [49527b0b@gateway/web/cgi-irc/kiwiirc.com/ip.73.82.123.11] has joined #wesnoth-dev
20161026 02:03:31< SpaceRobot> hey - I get a mismatched SHA256 on the macOS stable version 1.12.6
20161026 02:04:00< SpaceRobot> mine is c94fe7880b1ed0f52e31ad570d3f082412ffbeb9d7ddadf0c64d501dfc1f0589
20161026 02:04:56< SpaceRobot> is this the right place to mention this, or the mailing list (or another)?
20161026 02:05:59< Aginor> Hi SpaceRobot, have you tried downloading more than once to make sure that it's not a download error?
20161026 02:06:15< SpaceRobot> no, I will do so now
20161026 02:07:23< SpaceRobot> while it's downloading, I wanted to ask also: have alternatives to sourceforge been considered for distribution of the binaries?
20161026 02:07:32< shadowm> I think in this case it'd be easier to PM ancestral on the forums than post to the mailing list.
20161026 02:07:52< ancestral> Yeah that is known
20161026 02:08:01< mattsc> SpaceRobot: I seem to be getting the same as you on the DMG I downloaded quite some time ago.
20161026 02:08:02< shadowm> What is?
20161026 02:08:30< ancestral> The mismatch. We should probably update the sha256 we have on the website
20161026 02:08:43< ancestral> The moment I upload it arrives and receives a different SHA
20161026 02:08:49< SpaceRobot> hm. I am checking the SHASUM because I don't trust sourceforge :D
20161026 02:08:50< shadowm> Well, it wasn't known to me. This is the first time I hear about it.
20161026 02:09:01< ancestral> I emailed sf.net and they scratched their heads and seemed to show everything was okay on their end
20161026 02:09:15< shadowm> It's not good that the upload on SF.net is being modified. I wouldn't trust it.
20161026 02:09:18< Aginor> the fact that it gets a different sha is a bit suspect
20161026 02:09:29< Aginor> (quite a bit)
20161026 02:09:32< ancestral> I tried uploading it via sftp, ssh, web, exact same thing every time
20161026 02:09:45< ancestral> Made a silent change to the project, re-uploaded
20161026 02:09:49< ancestral> Still the same issue
20161026 02:10:06< ancestral> I *believe* I tried another computer, though I can’t remember now
20161026 02:10:12< pydsigner> It's also not super surprising unfortunately
20161026 02:10:18< Aginor> ancestral: do you have the original file around?
20161026 02:10:41< shadowm> It is super surprising. The new administration supposedly halted their bullcrap program.
20161026 02:10:46< SpaceRobot> my concern is via this and the like: http://www.makeuseof.com/tag/rise-fall-slashdot-media/
20161026 02:10:47< ancestral> (There are chat logs of this btw in #wesnoth-dev)
20161026 02:10:57< ancestral> Aginor: I can look around
20161026 02:11:26< Aginor> ancestral: if you could find it, I'd appreciate it if you could share it wth me so I can compare it with what I'm downloading from sf right now
20161026 02:11:57< ancestral> Aginor, pydsigner, SpaceRobot, shadowm: I’m banking on whatever it is that I could be doing wrong or whatever is changing the SHA is probably innocuous, probably a side-effect or artifact rather than malicious intent
20161026 02:12:18< shadowm> What method are you using to upload? rsync?
20161026 02:12:31< SpaceRobot> ok. but please keep the SHA256 you have from your machine pre-upload as the one listed on the website
20161026 02:13:06 * Aginor agrees, don't change the webite sha256 until we've gotten to the bottom of this
20161026 02:13:14< SpaceRobot> you may also consider using PGP to sign the binary, in addition, if suspicion intensifies
20161026 02:13:20< SpaceRobot> or GPG
20161026 02:14:35< ancestral> I’m going to have to look into my backups
20161026 02:15:10< shadowm> Also very important: noting if this affects all mirrors or just one or some. It wouldn't be surprising if one or more mirrors had been compromised by an external attacker.
20161026 02:15:28< SpaceRobot> i've been going by the standard clicking from the main download page
20161026 02:15:31< shadowm> Nor it would be surprising if SF.net didn't notice.
20161026 02:15:34< SpaceRobot> i'll try a mirror, too...
20161026 02:15:58< shadowm> SF.net would autoselect a mirror based on your geographical location in that case, I believe.
20161026 02:16:46< shadowm> Easiest way to tell where the download actually comes from would be to use curl or wget or your browser's network inspector functionality to see where the download is actually being fetched from.
20161026 02:16:48< Aginor> hmm
20161026 02:16:54< shadowm> (i.e. after following HTTP redirects.)
20161026 02:17:09< Aginor> it's using https (for me) so it's unlikely to be a proxy mangling it on the way
20161026 02:17:41< shadowm> The machine serving via https can be comprimised anyway. :)
20161026 02:17:49< Aginor> definitively
20161026 02:18:09< shadowm> *compromised
20161026 02:18:13< Aginor> although that's less likely than other crap going on
20161026 02:18:41< Aginor> normally occam's razor would point the other way, but with SF I'm less trusting :D
20161026 02:19:44< SpaceRobot> how many downloads do you measure every month, across all binaries?
20161026 02:19:50< SpaceRobot> just curious
20161026 02:20:32< Aginor> https://sourceforge.net/projects/wesnoth/files/?source=navbar for some stats
20161026 02:21:32< Aginor> with very few people checking the hashes
20161026 02:21:41< SpaceRobot> 2,687 for the exe of 1.12.6
20161026 02:22:01< SpaceRobot> that's roughy 1TiB of upload
20161026 02:22:28< SpaceRobot> i get the feeling SF is throttling
20161026 02:22:55< SpaceRobot> Aginor: yea hashes for games.. i'm not sure that would be common, either :)
20161026 02:22:57< shadowm> Aginor: Since version 1.12.5 and 1.13.2 we include the SHA256 digests in the announcement posts and the downloads list on the wiki, in plain text.
20161026 02:23:38< SpaceRobot> ok i get the same SHA
20161026 02:23:48< SpaceRobot> downloaded again, and another concurrently from germany
20161026 02:23:59< shadowm> (And the downloads list on the wiki is sourced from pages that are protected so they can only be edited by admins.)
20161026 02:24:05< SpaceRobot> both are c94fe7880b1ed0f52e31ad570d3f082412ffbeb9d7ddadf0c64d501dfc1f0589
20161026 02:26:09< Aginor> I'm still downloading, I seem to have gotten a slow morror
20161026 02:26:16< Aginor> but I expect to get the same result now
20161026 02:27:44-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20161026 02:28:13< shadowm> http://pastebin.com/raw/wEYra4ex
20161026 02:29:35< SpaceRobot> i'm not surprised we all seem to be getting the same hash (though originally it could have been a failed download for me)
20161026 02:29:52< SpaceRobot> rather why is the hash mismatching with the original file
20161026 02:29:57< SpaceRobot> wherever that is
20161026 02:30:12< Aginor> how about unpacking it and doing a full sha256 sum of the files as well? - if ancestral finds the original we should be able to see if there's any contents that mismatches
20161026 02:32:54-!- gfgtdf [~chatzilla@x4e36a5f1.dyn.telefonica.de] has quit [Ping timeout: 252 seconds]
20161026 02:33:34< shadowm> shikadilord@uruz:~/Downloads$ file Wesnoth_1.12.6.dmg
20161026 02:33:40< shadowm> Wesnoth_1.12.6.dmg: bzip2 compressed data, block size = 100k
20161026 02:33:43< shadowm> shikadilord@uruz:~/Downloads$ bunzip2 Wesnoth_1.12.6.bz2
20161026 02:33:45< shadowm> bunzip2: Wesnoth_1.12.6.bz2: trailing garbage after EOF ignored
20161026 02:33:49< ancestral> I’ve done this before
20161026 02:34:10< ancestral> I’m a little preoccupied at the moment but I will discuss this shortly
20161026 02:34:21< shadowm> I guess dmgs are some kind of hybrid thing that starts with a bzip2 section.
20161026 02:34:32< SpaceRobot> i have some DMG that are not bzip
20161026 02:36:52< SpaceRobot> i'm running sha256 on all the interior files now
20161026 02:40:26< SpaceRobot> ok .. pastebin won't let me post a 2.2MiB file
20161026 02:40:59< pydsigner> dropbox it?
20161026 02:41:04< ancestral> a50f384cead15f68f31cfa1a311e76a12098428702cb674d3521eb169eb92e4e  wesnoth-1.12.6.tar.bz2 is the sha I show
20161026 02:41:20< ancestral> From when I attempted to upload
20161026 02:41:30< SpaceRobot> pydsigner: yea i'm doing that
20161026 02:41:49< SpaceRobot> ancestral: for the dmg? .. or the sources
20161026 02:42:12< ancestral> Wait, that might be the wrong one
20161026 02:42:18< shadowm> tar.bz2 must be the source tarball from us.
20161026 02:42:21< shadowm> (from me)
20161026 02:42:27 * ancestral nods
20161026 02:45:35< SpaceRobot> https://dl.dropboxusercontent.com/u/16676715/wesnoth-1.12.6-macos-dmg.sha256
20161026 02:45:49-!- tad_carlucci [~lundberg@173.217.65.103] has joined #wesnoth-dev
20161026 02:45:51< SpaceRobot> i mounted the DMG then went into the .app and hashed all files
20161026 02:48:29< Aginor> ok, I'm a moron, I downloaded the source
20161026 02:49:18< SpaceRobot> it's ok. check the hash on that too :D
20161026 02:49:45< Aginor> a50f384cead15f68f31cfa1a311e76a12098428702cb674d3521eb169eb92e4e  wesnoth-1.12.6.tar.bz2
20161026 02:49:50< Aginor> that matches
20161026 02:49:56< SpaceRobot> ok good
20161026 02:50:08< SpaceRobot> but source balls aren't installer code
20161026 02:50:18< SpaceRobot> so i'd want to check the exe dmg and others
20161026 02:50:26< SpaceRobot> too
20161026 02:50:47< Aginor> indeed
20161026 02:51:23< Aginor> I shall ponder alpha blending for a while since SF hates me living at the edge of the world
20161026 02:53:45< SpaceRobot> i've used digitalocean for a few months, and they seem to be decent.. have you considered alternative locations for hosting the files?
20161026 02:54:31< SpaceRobot> though that'd be a manual solution
20161026 02:54:52< SpaceRobot> or does github allow large release files?
20161026 02:56:24< SpaceRobot> oh cool -- https://help.github.com/articles/about-releases/
20161026 02:56:29< SpaceRobot> "We don't limit the total size of your binary release files, nor the bandwidth used to deliver them. However, each individual file must be under 2 GB in size."
20161026 02:59:03-!- tad_carlucci [~lundberg@173.217.65.103] has quit [Quit: Switching to Unix to get some real work done.]
20161026 03:01:54< Aginor> I think they removed that feature again
20161026 03:02:07< SpaceRobot> :-\
20161026 03:02:33< Aginor> they discovered they hosted a lot of malware when random people could upload and host executables
20161026 03:03:50< Aginor> if I look at one of my projects I don't see any trace of the releases
20161026 03:04:22< shadowm> Huh?
20161026 03:04:33< shadowm> https://github.com/shikadilord/After_the_Storm/releases
20161026 03:04:45< Aginor> hmm
20161026 03:05:00< shadowm> Although you can't see the draft with custom attachments I have.
20161026 03:05:19< SpaceRobot> but can releases contain DMG or exe?
20161026 03:05:28< SpaceRobot> i think that was Aginor's comment
20161026 03:06:01< shadowm> https://github.com/shikadilord/After_the_Storm/releases/tag/TEST
20161026 03:06:21< shadowm> Surely they wouldn't rely on file extensions to determine what can be uploaded.
20161026 03:06:43< SpaceRobot> you can never be sure....
20161026 03:06:45< Aginor> I was talking about the "downloads section"
20161026 03:06:46< shadowm> The two first downloads are custom attachments.
20161026 03:06:52< shadowm> first two
20161026 03:07:01< Aginor> but it seems that the releases are simply based off tags, so that might be an option
20161026 03:07:09< shadowm> I don't think there was ever a downloads section other than the one in each release's page.
20161026 03:07:30< Aginor> I usually don't bother with tags for my little projects, so that'd explain why I can't see it
20161026 03:07:40< SpaceRobot> so would putting some of these files there be useful?
20161026 03:07:46< Aginor> shadowm: years ago there was
20161026 03:08:04< shadowm> You can create your own release by hand and GH will automatically generate a tag for you like it just did here.
20161026 03:08:17< shadowm> Although normally you'd create it from an existing tag instead, yep.
20161026 03:10:03< Aginor> ah, yes
20161026 03:10:05< Aginor> found it
20161026 03:10:30< Aginor> that could work, if someone could be bothered to re-release the last two releases on github
20161026 03:12:14< shadowm> I didn't bother trying since GitHub didn't allow release file uploads via ssh at the time.
20161026 03:13:10< shadowm> Uploading large files via http(s) isn't only a recipe for disaster, it's also impossible for me.
20161026 03:25:27< Aginor> hmm
20161026 03:25:54< Aginor> blit_surface is doing something interesting when blending the alpha channel
20161026 03:33:04 * Aginor wonders if someone got their src and destination mixed up when writing that line
20161026 03:33:14< Aginor> although it's admittedly looking better than if it's fixed
20161026 03:33:21-!- tad_carlucci [~lundberg@173.217.65.103] has joined #wesnoth-dev
20161026 03:33:36-!- Polsaker [~Polsaker@donger/wielder/Polsaker] has joined #wesnoth-dev
20161026 03:33:53< Aginor> src colour is blended into the destination, taking src and destination alpha into account nicely
20161026 03:34:18< tad_carlucci> celticminstrel, PR 843 for your request
20161026 03:34:32< Aginor> dst alpha is kept over src alpha
20161026 03:35:03< Aginor> I think this is only really noticable in the pango text, which is the most heavy alpha blending we have
20161026 03:35:46< Aginor> I've gone over the code where it's demultiplied, and while we could certainly do with a version that is more optimisable, that's looking correct
20161026 03:35:53-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 245 seconds]
20161026 03:36:32< Aginor> but I see no convenient way out of the alpha blending problem
20161026 03:36:42 * Aginor goes to ponder
20161026 03:36:47< Aginor> ponderously
20161026 03:51:14< ancestral> I’m struggling to find the conversation we had
20161026 03:51:17< ancestral> I know we had it
20161026 03:55:32< ancestral> Maybe I had a PM with vultraz
20161026 03:56:04< ancestral> Hmm, nah, not in there
20161026 03:57:33< ancestral> Hmm, dates are all wrong on my log files
20161026 04:00:48< ancestral> https://www.wesnoth.org/irclogs/2016/08/%23wesnoth-dev.2016-08-09.log
20161026 04:02:36< ancestral> “Hello,
20161026 04:02:36< ancestral> I am unable to reproduce this issue on my test project. I checked the files that are on the site and verified that the downloaded version is the same as the one on the site. Is it possible you are not generating the correct md5? If you check the file after uploading SFTP or SCP are the final md5 and sha1 the same between those?”
20161026 04:02:42< ancestral> That’s the response I got from sf
20161026 04:04:21-!- astrelyon [~astrelyon@dh207-109-100.xnet.hr] has joined #wesnoth-dev
20161026 04:05:32< shadowm> When you find the original .dmg try uploading it to files.wesnoth.org.
20161026 04:09:09< SpaceRobot> > md5 and sha1
20161026 04:09:16< SpaceRobot> the download page does not have those?
20161026 04:09:26< SpaceRobot> s/?//
20161026 04:09:29< shadowm> SF.net's does.
20161026 04:09:49< SpaceRobot> ah i see.. but the issue is the sha256 from wesnoth.org's page, right?
20161026 04:09:59< shadowm> Yes, and the SHA256 is safer.
20161026 04:10:12< SpaceRobot> right, not disagreeing there
20161026 04:11:20< SpaceRobot> wouldn't SF also need to check those, then, too?
20161026 04:11:45< shadowm> https://sourceforge.net/projects/wesnoth/files/wesnoth-1.12/wesnoth-1.12.6/ -- If you click on the "i" icon, you'll see the SHA1 and MD5 digest for each file as SF.net sees it.
20161026 04:12:07< shadowm> That means that SF.net is generating those from its own authoritative copy of each file prior to propagating it to mirrors.
20161026 04:12:16< shadowm> That authoritative copy might not match the uploader's original.
20161026 04:12:37< SpaceRobot> i understand
20161026 04:12:41< shadowm> That's why I started asking people to use SHA256 and to upload the digests of the originals to files.wesnoth.org.
20161026 04:12:59< SpaceRobot> but then what is the protocol when there is a mismatch? :)
20161026 04:13:27< shadowm> There isn't one besides reuploading because that hasn't happened before.
20161026 04:13:48< SpaceRobot> or perhaps people haven't been checking the hashes? maybe?
20161026 04:14:08< shadowm> I think we had just one or two incidents where the downloaded file was just corrupted and turned out to be a transient issue or an issue on the client end.
20161026 04:14:28< SpaceRobot> honestly i don't much.. hosting on SF made me do it for wesnoth this time, though, since I guess I was expecting another hosting provider to have been used by this point :-x
20161026 04:14:32< SpaceRobot> ok
20161026 04:14:53< shadowm> Sorry, as I said above I had to rule out GitHub for practical reasons.
20161026 04:15:11< SpaceRobot> gitlab or bitbucket?
20161026 04:15:21< shadowm> We don't use either.
20161026 04:15:49< SpaceRobot> but couldn't one just be a mirror and host the release files, if the GH reasons don't exist with these?
20161026 04:15:59< SpaceRobot> i'm just trying to brainstorm :)
20161026 04:16:19< shadowm> I wouldn't have considered this at the time because the project's technical platform is already too fragmented.
20161026 04:16:25< SpaceRobot> ah
20161026 04:16:27< shadowm> (GitHub + SF.net + Gna)
20161026 04:16:57< shadowm> It's now out of my control but I don't think the new guy would consider it either -- he'd probably just move everything to GitHub instead.
20161026 04:16:59< SpaceRobot> what if.. the files were hosted directly from the website?
20161026 04:17:08< SpaceRobot> hm ok
20161026 04:17:13< shadowm> There were bandwidth and disk space concerns.
20161026 04:17:50< shadowm> I tried asking if people could help me move files.wesnoth.org to a separate host like a CDN of some sort and no-one really did much, so that didn't happen.
20161026 04:18:13< SpaceRobot> ok :)
20161026 04:18:22< shadowm> (By which I mean I wasn't even given a certain answer as to what to ask our financial organization to pay for.)
20161026 04:18:40< SpaceRobot> VMs seem to be quite cheap these days
20161026 04:18:44< SpaceRobot> +bandwidth
20161026 04:19:56< shadowm> Maybe, but about 12 hours a day we have less disk I/O bandwidth than we have network bandwidth, plus the latter is already shared with too many services (website and all it implies, plus wesnothd and campaignd instances).
20161026 04:21:37< SpaceRobot> ah, yea that sounds non-trivial to organize
20161026 04:32:06-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20161026 04:33:39< vultraz> tad_carlucci: is an int guaranteed to be whole with no decimals?
20161026 04:34:20< tad_carlucci> Huh?
20161026 04:34:24< tad_carlucci> You mean in C++?
20161026 04:35:02< vultraz> yes
20161026 04:36:13< tad_carlucci> Unless you're imagining a decimal point, yes.  For example, I could say "integer pennies" and 123 -- > $1.23 but it's still an int value inside the program.
20161026 04:36:18< shadowm> Integer.
20161026 04:37:14< vultraz> hm
20161026 04:37:28< shadowm> If you want decimals you must use rational or irrational numbers.
20161026 04:37:57< shadowm> Or more generally, non-integer real or complex numbers.
20161026 04:38:04< tad_carlucci> Or scale, if you know where you want the decimal point, as I just did.
20161026 04:38:06-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20161026 04:40:31< shadowm> At that point you're discussing semantics.
20161026 04:40:48< shadowm> Like, "is an int guaranteed to represent an integer and not a bitfield".
20161026 04:41:24< vultraz> So if one divided (or cast to) two ints (say, 12 / 5), would the result be a whole 2 and not 2.4?
20161026 04:41:36< shadowm> Or "can an int represent an opaque handle to an object".
20161026 04:42:12< vultraz> er, divided or cast the result of the division of*
20161026 04:42:20-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 252 seconds]
20161026 04:45:56-!- SpaceRobot [49527b0b@gateway/web/cgi-irc/kiwiirc.com/ip.73.82.123.11] has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client]
20161026 04:47:23-!- ancestral [~ancestral@75-168-80-79.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20161026 04:48:42< celticminstrel> vultraz: Integer division is truncated, that is, rounded towards zero.
20161026 04:48:56< celticminstrel> aka, drop the fractional portion
20161026 04:49:18-!- iceiceice [~chris@unaffiliated/iceiceice] has joined #wesnoth-dev
20161026 04:49:57< celticminstrel> tad_carlucci: I assume you tested it, yes?
20161026 04:50:36< tad_carlucci> vultraz, What we you trying to achieve?  int i = 12 / 5 ---> 2, yes.  int = -12 / 5 --> -2
20161026 04:50:54< vultraz> yes, that
20161026 04:51:23< tad_carlucci> celticminstrel, OF course. Did not effect normal messages and tested [message][text_input][/text_input][/message] to ensure even degenerate case worked.
20161026 04:51:40-!- irker921 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20161026 04:51:41< irker921> wesnoth: Gregory A Lundberg wesnoth:master 36ab1f9a5cd5 / data/lua/wml/message.lua src/scripting/lua_gui2.cpp: wesnoth.show_message_dialog: Don't expose has_input flag to Lua https://github.com/wesnoth/wesnoth/commit/36ab1f9a5cd59f0eb7ea3669fb0d29d98fc3863a
20161026 04:52:23< celticminstrel> \o/
20161026 04:52:40< celticminstrel> Maybe I should look at the [for] issue quickly before bed.
20161026 04:53:17< celticminstrel> I just remembered I had been working on the cave generator...
20161026 04:53:53< celticminstrel> Okay so basically loop_lim.step is probably nil...
20161026 04:54:41< celticminstrel> Which seems impossible though...
20161026 04:55:16< celticminstrel> Oh wait.
20161026 04:56:21-!- ancestral [~ancestral@75-168-80-79.mpls.qwest.net] has joined #wesnoth-dev
20161026 04:59:04< tad_carlucci> celticminstrel, vultraz Bug 25186 "[message] lag" with PR 842 in I cannot reproduce but Pentarctagon says he still see. I tried using is OP example with errors corrected. Perhaps his issue is the {EQ} macro which HttT throws an error on.
20161026 04:59:18< celticminstrel> So I have a fix for zookeeper now, I think.
20161026 04:59:59< celticminstrel> I want to add a test, so I'll commit tomorrow probably (feel free to remind me).
20161026 05:00:15< vultraz> alright
20161026 05:00:21< vultraz> tad_carlucci: EQ?
20161026 05:00:25 * tad_carlucci nods.
20161026 05:00:32 * celticminstrel wonders what vultraz is alrighting.
20161026 05:00:43< vultraz> celticminstrel: you
20161026 05:00:49< tad_carlucci> probably reminding you
20161026 05:01:01< vultraz> tad_carlucci: what is this EQ macro?
20161026 05:02:09< tad_carlucci> vultraz, Dunno. HttT throws an error because it's not known there. I replaced it with [true][/true] to infinite loop and I cannot reproduce the hang. It's just constantly slow.
20161026 05:03:30< tad_carlucci> NB that before PR 842 I **could** reproduce the hang.
20161026 05:14:27< vultraz> hmm
20161026 05:15:25< vultraz> well at least you fixed the hang...
20161026 05:15:40< vultraz> step in the right direction
20161026 05:16:02< tad_carlucci> But he claims I didn't. So either I'm missing something or he is. Sorta want to get that worked out.
20161026 05:17:24< vultraz> blagh
20161026 05:17:28< vultraz> why is my code not working
20161026 05:19:19 * vultraz is trying to make gui2 sliders snap to items
20161026 05:20:56< vultraz> slightly harder task than it looks :|
20161026 05:21:17 * tad_carlucci smiles.
20161026 05:22:31< tad_carlucci> I've not seen your code but I remember something like that years and years ago. I think I solved it by scaling the mouse units down to items instead of trying to scale the item units up to mickeys.
20161026 05:23:20< vultraz> 	if(std::abs(distance_moved) >= get_pixels_per_step()) {
20161026 05:23:22< vultraz> 		int offset = distance_moved / get_pixels_per_step();
20161026 05:23:23< vultraz> 		set_item_position(get_item_position() + offset);
20161026 05:23:25< vultraz> is essentially what im trying
20161026 05:23:32< vultraz> but it's going haywire!
20161026 05:24:54< tad_carlucci> Rounding error?  Try: int offest = (disance_moved + (get_pixles_per_step()/2)) / get_pixels_per_step(); ???
20161026 05:25:36< tad_carlucci> Fixing my typos, of cours.
20161026 05:26:10< vultraz> er, why would that be needed?
20161026 05:26:19< vultraz> anyway, I think the cause is more mundane..
20161026 05:27:07< vultraz> it works up to the length of one step, so I think the conditional is the problem
20161026 05:27:16< vultraz> need to think of a better one..
20161026 05:28:32< tad_carlucci> is item position in pixels?
20161026 05:28:50< tad_carlucci> Or item slots?
20161026 05:30:01< vultraz> slots
20161026 05:30:07< vultraz> that's the point of the division 
20161026 05:30:16< vultraz> see how many slots could be covered by distance
20161026 05:30:24< vultraz> distance_moved*
20161026 05:30:33< vultraz> distance_moved is only set once, on mouse down..
20161026 05:30:47< vultraz> I think it needs to be reset on each new successful jump..
20161026 05:32:01< celticminstrel> vultraz: What are you doing?
20161026 05:32:07< celticminstrel> Oh.
20161026 05:32:19< celticminstrel> I just saw your explanation right at the top of the visible scrollback.
20161026 05:32:41-!- travis-ci [~travis-ci@ec2-54-158-65-122.compute-1.amazonaws.com] has joined #wesnoth-dev
20161026 05:32:41< celticminstrel> So for sliders it's reasonable to consider every item to be of the same size, certainly.
20161026 05:32:42< travis-ci> wesnoth/wesnoth#11757 (master - 36ab1f9 : Gregory A Lundberg): The build passed.
20161026 05:32:42< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/170669595
20161026 05:32:42-!- travis-ci [~travis-ci@ec2-54-158-65-122.compute-1.amazonaws.com] has left #wesnoth-dev []
20161026 05:33:35< vultraz> point being?
20161026 05:33:54< celticminstrel> Nothing really, I guess.
20161026 05:34:16< celticminstrel> For a minute I thought you were talking about listboxes.
20161026 05:34:21< vultraz> (is std::abs different from abs)
20161026 05:34:49< celticminstrel> No.
20161026 05:37:07< iceiceice> celticminstrel, are you sure?
20161026 05:37:14< celticminstrel> Yes.
20161026 05:37:16< iceiceice> i thought there were some crappy issues we had to fix once
20161026 05:37:27< iceiceice> where like, abs was only defined for like float and int or something
20161026 05:37:35< iceiceice> but std::abs was a template so it always is the same type you pass it
20161026 05:37:41< iceiceice> i dont remember the details any more
20161026 05:37:52< iceiceice> and it might have been libc++ only or something
20161026 05:37:55< celticminstrel> Well, if you include <stdlib.h>, then abs() might be a macro.
20161026 05:38:22< celticminstrel> You should include <cstdlib> instead.
20161026 05:38:22< iceiceice> it might have been std::min, it was a long time ago
20161026 05:38:48< celticminstrel> Theoretically it might be the case that <cstdlib> only defines std::abs and not abs on some platforms, though in practice I think that's not usually the case?
20161026 05:39:01< celticminstrel> What I mean is it might be allowed by the standard.
20161026 05:39:23< iceiceice> yah i always have to look at the docu to remember stuff like this
20161026 05:39:39< celticminstrel> In any case, they're identical in meaning, so it doesn't really matter that much IMO unless you actually get errors with one way.
20161026 05:41:09< iceiceice> i mean i think the error it can cause is like
20161026 05:41:17< iceiceice> if you have a double, and you take abs of it, you may lose precision
20161026 05:41:25< iceiceice> because it may get casted to float and then back to double or osmething dumb
20161026 05:41:48< celticminstrel> I thought the C lib usually defines abs as a macro though? Maybe I'm thinking of max/min.
20161026 05:41:50< vultraz> http://stackoverflow.com/questions/21392627/abs-vs-stdabs-what-does-the-reference-say
20161026 05:42:59< vultraz> also, yay, my code is working!
20161026 05:43:15< vultraz> not perfectly, but at least it's not haywire.
20161026 05:43:28< vultraz> need to make a few improvements and clean up the impl
20161026 05:47:52-!- JyrkiVesterinen [~JyrkiVest@87-100-141-59.bb.dnainternet.fi] has joined #wesnoth-dev
20161026 05:50:44< vultraz> it's unfortunate the the positioner will often lab behind
20161026 05:51:18< vultraz> lag
20161026 05:55:37< vultraz> but that's not my fault and more the fault of gui2 not updating fast enough
20161026 05:55:45-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20161026 05:55:51< vultraz> Aginor's branch did improve that quite a bit
20161026 05:56:01< vultraz> hopefully he revisits it sometime.
20161026 05:56:06< Aginor> Aginor's branch also broke a lot of things
20161026 05:56:16< vultraz> Indeed.
20161026 05:56:29< vultraz> But the reduction of lag in GUI2 was amazing.
20161026 05:57:13< vultraz> I wonder if that's something that could be cherrypicked...
20161026 05:57:19< Aginor> I don't think that you should implement it in such a way that different parts of the same component may be out sync
20161026 05:57:36< Aginor> no, it was a happy coincidense of how I was restructuring things
20161026 05:57:56< Aginor> I was expecting it to be slower at that stage to be honest
20161026 05:58:09< vultraz> I'm not deliberately making anything slower here
20161026 05:58:13< Aginor> although I was trying to eliminate all of the gratoitus flip's
20161026 05:58:17< Aginor> flips even
20161026 05:58:25< vultraz> it's just a side affect of the rate at which the canvas is updated.
20161026 05:59:00< vultraz> is there any chance of your revisiting that before 1.14?
20161026 05:59:43< Aginor> no, it will break too much and I won't have the time to dedicate to fixing it all
20161026 06:00:00< vultraz> blah
20161026 06:00:11< vultraz> sadness :(
20161026 06:00:34< Aginor> I'm currently trying to get rid of blit_surface, but it's proving to be annoying
20161026 06:00:55< vultraz> I can imagine
20161026 06:00:58< Aginor> I've gone through the math, and I am wondering if someone got something wrong or did it they way they did it on purpose
20161026 06:01:25< Aginor> but the alpha component in blit_surface is calculated backwards from how it should be
20161026 06:01:44< Aginor> or at least how I expect it to be
20161026 06:03:16< vultraz> huh
20161026 06:05:32< Aginor> can you give me an example of where a gui2 multimenu is used?
20161026 06:06:33< vultraz> multimenu?
20161026 06:06:38< vultraz> what is a multimenu
20161026 06:07:26< celticminstrel> It's GUI1.
20161026 06:07:33< celticminstrel> Isn't it?
20161026 06:07:36< vultraz> I know of no such widget in GUI2.
20161026 06:07:39< celticminstrel> It's used in the MP Create dialog.
20161026 06:07:42< vultraz> pretty sure that's gui1
20161026 06:07:44< celticminstrel> For modifications.
20161026 06:07:45< vultraz> as celticminstrel says
20161026 06:07:49< vultraz> yeah
20161026 06:07:51< Aginor> oh, sorry, it is
20161026 06:07:55< vultraz> and that screen is slated for removal post-release
20161026 06:08:00< vultraz> along with the widget
20161026 06:14:20< irker921> wesnoth: Andreas Löf wesnoth:tidyup_blitting 319abf417a2b / src/ (7 files in 5 dirs): This removes references to blit_surface and introduces masks for colours https://github.com/wesnoth/wesnoth/commit/319abf417a2b02f7b15a2804343d320132a23235
20161026 06:15:29< vultraz> Aginor: oh, yeah, the reason i restored the use of blit_surface in the multimenu was because of a compilation error on clang
20161026 06:15:40< Aginor> https://github.com/wesnoth/wesnoth/pull/844 <-- please help out and test
20161026 06:16:04< Aginor> vultraz: fixing the error would have been the better option :)
20161026 06:16:13< vultraz> i was too lazy :P
20161026 06:16:21< vultraz> especially since i do not use clang 
20161026 06:16:31< Aginor> that's no excuse
20161026 06:16:56< vultraz> it was something about the fact that i made a parameter non-const which clang didn't like
20161026 06:16:58< Aginor> stand up a VM if you can't be bothered to compile things strictly enough locally
20161026 06:17:01< vultraz> i suppose it could have been a copy..
20161026 06:17:54< celticminstrel> vultraz: Is that the one where you made something non-const in the signature of an overridden function?
20161026 06:17:59< vultraz> yes
20161026 06:18:06< Aginor> https://github.com/wesnoth/wesnoth/pull/844/files#diff-548aa9ce2daa1f0c65bdeddc667187dbR29
20161026 06:18:26< Aginor> any rectangles passed to sdl_blit should not be const
20161026 06:18:31< Aginor> well
20161026 06:18:44< Aginor> they can be const pointers, but we have to drop the constedness of them
20161026 06:18:54< Aginor> but the values may be updated by SDL
20161026 06:19:18< Aginor> and a const pointer != const value
20161026 06:20:35-!- ancestral [~ancestral@75-168-80-79.mpls.qwest.net] has quit [Quit: End Transmission.]
20161026 06:24:03-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20161026 06:26:22-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20161026 06:30:48-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 256 seconds]
20161026 06:31:55-!- JyrkiVesterinen [~JyrkiVest@87-100-141-59.bb.dnainternet.fi] has quit [Quit: .]
20161026 06:40:59-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20161026 06:44:37-!- atarocch [~atarocch@37.176.161.188] has joined #wesnoth-dev
20161026 06:50:41-!- astrelyon [~astrelyon@dh207-109-100.xnet.hr] has quit [Quit: WeeChat 0.4.2]
20161026 06:55:31-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The computer falls into a deep sleep, to awake again some other day!]
20161026 06:57:20-!- travis-ci [~travis-ci@ec2-54-147-180-243.compute-1.amazonaws.com] has joined #wesnoth-dev
20161026 06:57:21< travis-ci> wesnoth/wesnoth#11758 (tidyup_blitting - 319abf4 : Andreas Löf): The build passed.
20161026 06:57:21< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/170679325
20161026 06:57:21-!- travis-ci [~travis-ci@ec2-54-147-180-243.compute-1.amazonaws.com] has left #wesnoth-dev []
20161026 06:57:56-!- tad_carlucci [~lundberg@173.217.65.103] has quit [Quit: Off to resolve a merge conflict between the wife and husband branches of my real life.]
20161026 07:00:18< irker921> wesnoth: Charles Dang wesnoth:master e2d3d585be48 / src/gui/widgets/ (scrollbar.hpp slider.cpp slider.hpp): Slider: made slider positioner snap to values on drag https://github.com/wesnoth/wesnoth/commit/e2d3d585be485adc954c74e863b6a404107df7dd
20161026 07:16:48< vultraz> Aginor: for the record, since you worked on renderpath_redo the number of flip calls has been reduced.
20161026 07:16:57< vultraz> mostly via gui2 conversion
20161026 07:39:24< irker921> wesnoth: Andreas Löf wesnoth:renderpath_redo_new 48bcb277d420 / src/ (13 files in 8 dirs): Refactor calls to CVideo::flip() https://github.com/wesnoth/wesnoth/commit/48bcb277d420309e792aa30e0d782b81dd5cced7
20161026 07:39:57< vultraz> Aginor: ^ rebased the commit and resolved conflicts and removed obsolete changes
20161026 07:40:25< vultraz> even if you don't intend to work on it now it's certainly useful work and i'd rather it didn't get lost behind a million merge conflicts
20161026 07:42:11< vultraz> I
20161026 07:42:26< vultraz> ve also deleted the old renderpath_redo branch since it was 2000 commits behind :P
20161026 07:45:56< vultraz> Aginor: for the record, I can confirm that change produces beautiful smooth scrolling in new credits 
20161026 07:46:00-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has joined #wesnoth-dev
20161026 07:49:45< shadowm> Hang on, are the credits supposed to display on master without any command line switches?
20161026 07:50:07< vultraz> yes
20161026 07:50:16< shadowm> Well, I don't see any credits.
20161026 07:51:00< vultraz> There's a weird bug where if you have too many UMC with credits installed, they won't show at all. Default mainline without any UMC shows fine.
20161026 07:51:23< vultraz> I believe it's something to do with surface size, and I've been trying to get mordante to look at it, but to no avail :/
20161026 07:51:43< shadowm> No comment.
20161026 07:51:54< vultraz> Start wesnoth with a new userdata path and you should be able to see it.
20161026 07:51:59< vultraz> If not, then we really have a problem.
20161026 07:52:14< vultraz> (it = the credits)
20161026 07:52:28< shadowm> I see them, they are incredibly tiny.
20161026 07:52:46< shadowm> There's also a long pause at the start before they start scrolling unless I press down first.
20161026 07:53:04< vultraz> long pause is deliberate
20161026 07:53:23< shadowm> What a weird design decision.
20161026 07:53:35< shadowm> Is the smaller font size deliberate too?
20161026 07:53:53< vultraz> Yes
20161026 07:54:27< shadowm> So the one place that has long runs of text gets smaller fonts while everything else is humongous in comparison? Okay.
20161026 07:54:57< vultraz> It's needed for contrast with the xxl headers
20161026 07:55:04< vultraz> xl and xxl*
20161026 07:55:20< shadowm> Not really.
20161026 07:55:33< shadowm> But whatever.
20161026 07:56:27< vultraz> I hope you don't hate them, at least
20161026 07:56:41< vultraz> I know you expressed disinterest in centered credits text in the past.
20161026 07:57:43< shadowm> I can't not hate anything that has that gaudy-looking font on it, but sure, I don't hate the credits text any moe than the rest.
20161026 07:58:20< shadowm> But with the smaller font size I can't really see this as an improvement over the old credits, rather it's just a lateral change.
20161026 08:01:03-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20161026 08:07:08< zookeeper> i see that the credits screen is still incredibly unresponsive
20161026 08:07:24< zookeeper> also the text starts at the top
20161026 08:08:29< vultraz> ..what?
20161026 08:08:34< vultraz> why would it not?
20161026 08:09:17< zookeeper> think
20161026 08:09:18< zookeeper> for a second
20161026 08:09:20< zookeeper> look at it
20161026 08:09:29< zookeeper> try to imagine you're a user trying to check out the credits
20161026 08:13:09< zookeeper> it's not a coincidence that you've never seen scrolling credits which start like ours do right now
20161026 08:14:47-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20161026 08:16:48< vultraz> I fail to see your point
20161026 08:18:21-!- travis-ci [~travis-ci@ec2-54-147-180-243.compute-1.amazonaws.com] has joined #wesnoth-dev
20161026 08:18:22< travis-ci> wesnoth/wesnoth#11761 (renderpath_redo_new - 48bcb27 : Andreas Löf): The build passed.
20161026 08:18:22< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/170692583
20161026 08:18:22-!- travis-ci [~travis-ci@ec2-54-147-180-243.compute-1.amazonaws.com] has left #wesnoth-dev []
20161026 08:18:28-!- iceiceice [~chris@unaffiliated/iceiceice] has quit [Ping timeout: 256 seconds]
20161026 08:18:48-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 245 seconds]
20161026 08:19:32< zookeeper> sigh. the credits appear, and after like 2-3 seconds the top names have already scrolled off-screen. that is not good, because it's bad, because you're supposed to be able to read them.
20161026 08:20:37< vultraz> yeah, well, what would you do instead?
20161026 08:21:04< zookeeper> make them scroll from the bottom like every other scrolling credits in the universe?
20161026 08:22:07< vultraz> scroll from the *bottom*? you mean the last entry first?
20161026 08:22:31< zookeeper> ...
20161026 08:22:35< zookeeper> are you trolling me?
20161026 08:22:47< vultraz> no :|
20161026 08:23:16-!- iceiceice [~chris@pool-173-61-153-221.cmdnnj.fios.verizon.net] has joined #wesnoth-dev
20161026 08:23:17< vultraz> or do you mean reverse the order of scrolling?
20161026 08:23:25< vultraz> so they go down instead of up?
20161026 08:23:35< zookeeper> good grief
20161026 08:23:59< vultraz> seriously *what are you talking about* :|
20161026 08:24:27< vultraz> start with the thing scrolled to the bottom?
20161026 08:24:45< zookeeper> <zookeeper> sigh. the credits appear, and after like 2-3 seconds the top names have already scrolled off-screen. that is not good, because it's bad, because you're supposed to be able to read them.
20161026 08:24:46< zookeeper> <vultraz> yeah, well, what would you do instead?
20161026 08:24:49< zookeeper> <zookeeper> make them scroll from the bottom like every other scrolling credits in the universe?
20161026 08:24:58< zookeeper> if you can't parse that then you've suffered brain damage during the night
20161026 08:25:13< vultraz> I'm pretty damn sure credits don't start from the bottom
20161026 08:25:26< vultraz> do you even movie.
20161026 08:26:24< vultraz> I am almost *certain* I have never seen credits starting from the bottom
20161026 08:26:27< zookeeper> https://youtu.be/vd0stUFsZ-o?t=110
20161026 08:26:36< zookeeper> here, your favorite movie
20161026 08:26:52< vultraz> . . .
20161026 08:26:56< vultraz> god dammit :|
20161026 08:27:11< vultraz> that's not what one thinks when you say "scroll from the bottom" 
20161026 08:27:50< vultraz> if you had said "scroll IN from the bottom" or "enter the screen from the bottom" FINE I would have understood. :|
20161026 08:27:53< zookeeper> it's what you should think when i say "like every other scrolling credits in the universe"
20161026 08:28:33< vultraz> no!
20161026 08:28:35< zookeeper> same thing when i say "the credits appear, and after like 2-3 seconds the top names have already scrolled off-screen"
20161026 08:28:58< vultraz> you have to be more damn specific :|
20161026 08:29:10< iceiceice> english is a poor language
20161026 08:29:37< iceiceice> fwiw my thought about the credits screen is that the button does the opposite of what i expect it to
20161026 08:30:00< iceiceice> i press up to make it go down faster?
20161026 08:30:27< iceiceice> i never played any game where thats how it works
20161026 08:30:42< iceiceice> and im trying really hard to think of one
20161026 08:30:44< vultraz> that can be changed
20161026 08:32:56< zookeeper> also if there's speed controls then you should be able to at least make it go in reverse too, if we're not going to have an actual scrollbar that one can use to backtrack
20161026 08:33:17< vultraz> who the hell has a scrollbar in credits!
20161026 08:33:21< vultraz> unless it's non-animated
20161026 08:37:18< zookeeper> yeah, a regular scrollbar would be very ugly there
20161026 08:37:48< zookeeper> speed buttons are fine although as said they should be able to reverse direction too
20161026 08:39:29< vultraz> what kind of credits allow you to reverse either!
20161026 08:44:25< zookeeper> the rare kind which actually cares about usability? :p
20161026 08:44:54< vultraz> credits aren't about usability!
20161026 08:45:00< vultraz> they're about aesthetics and showing off
20161026 08:45:07< zookeeper> people aren't going to use the speed controls to set a comfortable pace at which they're able to read every one of the >800 names
20161026 08:45:34< zookeeper> they want to use it to make it go faster until they spot something they want to actually read (say, a given subsection) and then slow down
20161026 08:45:41< shadowm> The new credits aren't much better than the old ones in the aesthetics department.
20161026 08:45:42< zookeeper> and if they miss it, they want to scroll back up a bit
20161026 08:45:57< vultraz> no one reads credits!
20161026 08:46:58-!- horrowind [~Icedove@2a02:810a:8380:10a8:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20161026 08:47:03< shadowm> The small font size seems more of a regression than anything, because there's just this narrow column of tiny text surrounded by a massive amount of empty space.
20161026 08:47:22< vultraz> I'm not going to go larger until I can get thin text working
20161026 08:48:08< shadowm> I don't think this is what light fonts are intended for.
20161026 08:48:21< shadowm> If that's what you mean by 'thin'.
20161026 08:49:13< shadowm> On a 96 DPI screen light fonts tend to look bad anyway, especially with the amount of hinting Windows uses.
20161026 08:50:16< vultraz> I'd do two-column text if I could, too.
20161026 08:51:22< shadowm> Also, I wonder, have people (particularly those of an artistic inclination) posted proposals for a new credits screen before? That'd be an interesting thing to see.
20161026 08:51:55< shadowm> s/proposals/ideas/
20161026 08:51:58< vultraz> not that I know of
20161026 08:52:01< vultraz> I could ask LB...
20161026 08:52:22< vultraz> I've had some ideas, but they wouldn't work with our tech
20161026 08:52:26-!- horrowind [~Icedove@2a02:810a:8380:10a8:21b:fcff:fee3:c3ff] has quit [Quit: horrowind]
20161026 08:52:42< zookeeper> multiple columns for long sections would be my primary suggestion for improvement
20161026 08:52:45< shadowm> He's not the only visually-oriented person that's ever been around here, mind you.
20161026 08:53:08< shadowm> He's not the only one present these days either.
20161026 08:53:15< vultraz> yes, but he's my primary aesthetic consultant.
20161026 08:53:27< vultraz> jet usually just greenlights everything i throw at him 
20161026 08:53:47< shadowm> Hm, yeah, I can see why he'd do that when he's obviously busy with other projects.
20161026 08:55:29< vultraz> LB will at least provide criticism and suggestions 
20161026 08:55:58< vultraz> and he's readily available.
20161026 08:57:15< vultraz> and regarding jet, im afraid he won't finish the desert elves in time
20161026 08:57:58 * shadowm just barely remembers the times when developers consulted the community for ideas.
20161026 08:58:01< zookeeper> which reminds me that i've been waffling about whether to merge my desert elves PR or not for the past few weeks, and if i do i guess i better get it done now
20161026 08:58:06< vultraz> i can't force anyone to do anything and there's just so god damn much to do and im starting to feel like im taking on too much in the goal of getting shit done.
20161026 08:58:20< zookeeper> since it's revertable i guess there's no harm done even if it fails
20161026 08:59:41< vultraz> just merge it
20161026 08:59:54< vultraz> we don't have time to dawdle 
20161026 09:00:39< zookeeper> indeed
20161026 09:01:39< vultraz> utbs 2.0 IS a 1.14 target and 1.14 is a pretty set target
20161026 09:05:39< zookeeper> i have a bad feeling about this, but here goes
20161026 09:06:53< zookeeper> oh wait, i wanted to make the diff a bit easier first
20161026 09:10:22-!- Duthlet [~Duthlet@dslb-188-106-041-180.188.106.pools.vodafone-ip.de] has joined #wesnoth-dev
20161026 09:12:43< irker921> wesnoth: Lari Nieminen wesnoth:master 07c94876aa9a / data/campaigns/Under_the_Burning_Suns/scenarios_old/ (15 files): Split off old scenario files https://github.com/wesnoth/wesnoth/commit/07c94876aa9ad1370b15f3da624b8e43c1012b0e
20161026 09:12:45< irker921> wesnoth: Lari Nieminen wesnoth:master fffb5f0e4fc3 / data/campaigns/Under_the_Burning_Suns/ (78 files in 9 dirs): UtBS: Integrate the redesigned Quenoth elves https://github.com/wesnoth/wesnoth/commit/fffb5f0e4fc34c30cb147faeff0dd1d5874b8e53
20161026 09:12:47< irker921> wesnoth: Lari Nieminen wesnoth:master 83e6150a791d / changelog players_changelog: Updated changelogs https://github.com/wesnoth/wesnoth/commit/83e6150a791d11ad2fd36b143b98093b31a3df6b
20161026 09:12:49< irker921> wesnoth: Lari Nieminen wesnoth:master 0b98a78d7e93 / / (95 files in 11 dirs): Merge pull request #810 from ln-zookeeper/quenoth_integration https://github.com/wesnoth/wesnoth/commit/0b98a78d7e932c19a7c379c698646e8177b7d0d2
20161026 09:16:57-!- Duthlet [~Duthlet@dslb-188-106-041-180.188.106.pools.vodafone-ip.de] has quit [Ping timeout: 260 seconds]
20161026 09:21:16< zookeeper> as a disclaimer, i'm perfectly fine with pulling the whole thing before 1.14 if it doesn't pan out
20161026 09:28:08< vultraz> keep reminding Jetrel of the need for the sprites
20161026 09:29:49-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has quit [Quit: .]
20161026 09:51:07-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has joined #wesnoth-dev
20161026 10:03:01-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20161026 10:06:27-!- Shiki [~Shiki@141.57.57.72] has joined #wesnoth-dev
20161026 10:07:49-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 260 seconds]
20161026 10:23:58< Shiki> Hello zookeeper, I made PR824, with another username.
20161026 10:24:11< Shiki> Do you have some questions about it?
20161026 10:26:06< zookeeper> no, i just haven't tested it yet
20161026 10:26:20-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20161026 10:26:23< zookeeper> the commit message is a bit misleading though seeing how it isn't only about the drakes
20161026 10:28:56< Shiki> ok.
20161026 10:34:29< Shiki> I'm not sure if it's the best idea to change the terrains to sth. like Chw*^* instead of Chw^*. When somebody makes an UMC terrain, they would still use the same drakes, so they _would_ fly on e.g. Chwd. Though, it would maybe lead  to different people to chose more likely the same terrain codes.... 
20161026 10:48:42< zookeeper> if you want to make testing easier and thus faster, then you can upload whichever map you used for testing
20161026 10:53:09-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has quit [Quit: .]
20161026 11:03:28-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has joined #wesnoth-dev
20161026 11:09:52< Shiki> ok, will do
20161026 11:21:15-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20161026 11:24:44-!- gfgtdf [~chatzilla@x4e36a5f1.dyn.telefonica.de] has joined #wesnoth-dev
20161026 11:25:51< JyrkiVesterinen> Heh. I mentioned earlier in this chat that I maxed out memory in my desktop PC because I thought that DDR3 prices will eventually start going up.
20161026 11:25:58< JyrkiVesterinen> Guess what, it's happening. http://www.digitimes.com/news/a20161024PD201.html
20161026 11:29:50-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20161026 11:29:53-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20161026 11:32:10< zookeeper> i've maxed out my memory too, sadly it's still DDR2... :p
20161026 11:37:28-!- stikonas_ is now known as stikonas
20161026 11:44:20< Shiki> the editor crashes from time to time with  src/variable.cpp:479: virtual scoped_wml_variable::~scoped_wml_variable(): Assertion `resources::gamedata' failed.
20161026 11:44:50< Shiki> https://bpaste.net/show/de9bff1241c8 https://bpaste.net/show/4ad39fa0f354
20161026 11:46:17< Shiki> in the first case is in the .cfg an error, but it crashed anyway
20161026 11:47:32< Shiki> maybe it has sth to do with the placed wolfes, it crashed in the first case while i placed them. cfg files are in the paste too
20161026 11:50:30-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20161026 11:55:22-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 276 seconds]
20161026 12:01:29-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20161026 12:03:01-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20161026 12:08:07-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has quit [Quit: Rebooting]
20161026 12:09:06-!- iceiceice [~chris@pool-173-61-153-221.cmdnnj.fios.verizon.net] has quit [Ping timeout: 256 seconds]
20161026 12:09:13-!- atarocch [~atarocch@37.176.161.188] has quit [Ping timeout: 256 seconds]
20161026 12:10:19-!- Shiki [~Shiki@141.57.57.72] has quit [Remote host closed the connection]
20161026 12:11:31-!- Shiki [~Shiki@141.57.57.72] has joined #wesnoth-dev
20161026 12:11:52-!- atarocch [~atarocch@37.176.161.188] has joined #wesnoth-dev
20161026 12:12:59-!- irker921 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20161026 12:21:08< gfgtdf> Aginor: are your working on the faction_select not showing bug?
20161026 12:37:23-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has joined #wesnoth-dev
20161026 12:44:42-!- Kwandulin [~Miranda@p5DDD2B8F.dip0.t-ipconnect.de] has joined #wesnoth-dev
20161026 12:59:06-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: mattsc]
20161026 13:02:35-!- SpaceRobot [8fd7cf04@gateway/web/cgi-irc/kiwiirc.com/ip.143.215.207.4] has joined #wesnoth-dev
20161026 13:03:39-!- SpaceRobot [8fd7cf04@gateway/web/cgi-irc/kiwiirc.com/ip.143.215.207.4] has left #wesnoth-dev []
20161026 13:10:17-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20161026 13:14:35-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20161026 13:30:03< vultraz> gfgtdf: i dont think he is no
20161026 13:37:21-!- iceiceice [~chris@50-254-88-91-static.hfc.comcastbusiness.net] has joined #wesnoth-dev
20161026 13:52:14-!- iceiceice [~chris@50-254-88-91-static.hfc.comcastbusiness.net] has quit [Ping timeout: 256 seconds]
20161026 13:56:11-!- Shiki [~Shiki@141.57.57.72] has quit [Remote host closed the connection]
20161026 14:04:52-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has quit [Quit: .]
20161026 14:42:42-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: mattsc]
20161026 15:03:29-!- Gambit is now known as GamBorisJohnson
20161026 15:08:09-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20161026 15:09:37-!- Shiki [~Shiki@141.39.226.226] has joined #wesnoth-dev
20161026 15:14:03-!- GamBorisJohnson is now known as Gambit
20161026 15:16:57< vultraz> zookeeper: i need you to take a look at and merge 824 before 1.13.6
20161026 15:17:15< zookeeper> yes
20161026 15:25:32< Shiki> zookeeper, that's the map I use https://bpaste.net/show/aad49e45c2b1 I'm thinking about if the wolf-riders should maybe use the normal sprite on terrain with 0.3 submerge.
20161026 15:29:52< zookeeper> on water castles, it seems more appropriate to stand
20161026 15:31:15< zookeeper> and with your change they fly on the swamp-village-on-sand
20161026 15:32:43< Shiki> the swamp village I changed back, already commited
20161026 15:33:27< Shiki> should we let the wolves then use the normal image on coast and ford too ?
20161026 15:34:24< zookeeper> only where it visually makes sense
20161026 15:52:52-!- atarocch [~atarocch@37.176.161.188] has quit [Ping timeout: 260 seconds]
20161026 15:58:46-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20161026 16:00:00< vultraz> Aginor: i was looking further at your flip refactor post-rebase, and I have a question: is the problem that the main game screen constant redraws and has no knowledge of when to not draw? Obviously any change should be rendered but it seems like the main game gets stuck in a drawdrawdrawdrawdraw loop and there's no way to mark specific changes as 'need to draw'.
20161026 16:00:06< vultraz> Aginor: is that accurate?
20161026 16:09:48-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Read error: Connection reset by peer]
20161026 16:11:16-!- vultraz [~chatzilla@124.109.10.167] has joined #wesnoth-dev
20161026 16:11:16-!- vultraz [~chatzilla@124.109.10.167] has quit [Changing host]
20161026 16:11:16-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20161026 16:13:48< Shiki> If I click in the addon server on the Description of an addon, the window doesn't redraw anymore
20161026 16:14:16< vultraz> known
20161026 16:14:19< vultraz> also affects Options
20161026 16:14:27< vultraz> will be fixed once we implement the new addon manager
20161026 16:14:39< vultraz> if I ever get there
20161026 16:15:08-!- JyrkiVesterinen [~JyrkiVest@87-92-57-246.bb.dnainternet.fi] has joined #wesnoth-dev
20161026 16:15:20< Shiki> ok
20161026 16:16:01< Shiki> new addon manager ... sounds interesting 
20161026 16:16:53-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20161026 16:18:31< vultraz> Shiki: https://drive.google.com/file/d/0B-mR9s8FduLLNTZ1ZUhQLW9aQm8/view?usp=sharing
20161026 16:27:12< Shiki> Looks cool.  And it has a voting system?
20161026 16:31:35< vultraz> no
20161026 16:31:37< vultraz> not yet :/
20161026 16:31:43< vultraz> it's not even fully functional
20161026 16:34:48-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20161026 16:38:29< Shiki> will come ^^
20161026 16:39:07-!- midzer_ [~quassel@p57B454D1.dip0.t-ipconnect.de] has joined #wesnoth-dev
20161026 16:39:42< Shiki> Sth I was missing on the new style  .. I didn't really figure it out, maybe it's the yellow frame, which is sligthly different
20161026 16:39:53-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20161026 16:40:06< gfgtdf> vultraz: what is the renderpath todo new branch  ?
20161026 16:40:41< vultraz> gfgtdf: i rebased Aginor's old renderpath_redo branch and pushed it to a new one since i wasn't going to spam the channel with 2000 commit updates (even if i did mute irker)
20161026 16:41:00< gfgtdf> vultraz: so it has onyl one commit that is not merged yet?
20161026 16:41:13< gfgtdf> vultraz: do you know why that commit didnt made it to master?
20161026 16:41:22-!- midzer [~quassel@p57B45F9C.dip0.t-ipconnect.de] has quit [Ping timeout: 276 seconds]
20161026 16:41:39< vultraz> gfgtdf: because it breaks the main game
20161026 16:41:57< gfgtdf> vultraz: like in crashing/segaulting ?
20161026 16:42:23< vultraz> gfgtdf: essentially, it puts draw events on a constant timer which makes GUI2 100% more responsive, but the main game cannot handle the constant draw events so everything gets stuck in a drawdrawdrawdrawdrawdraw loop
20161026 16:42:25-!- tad_carlucci [~lundberg@173.217.65.103] has joined #wesnoth-dev
20161026 16:44:01< vultraz> gfgtdf: it makes even selecting a unit not work
20161026 16:44:25< vultraz> gfgtdf: basically, the game would need to be made to only actually execute a draw if there IS something to draw
20161026 16:44:26< gfgtdf> vultraz: so it feezes because it alyways rendering ?
20161026 16:45:04< JyrkiVesterinen> Only drawing when there is something to draw is the current behavior.
20161026 16:45:16< vultraz> gfgtdf: not completely freezes but as i said it makes stuff like hotkeys to bring up the Message or COmmand lines not work, Esc not bring up the quit popup, selecting a unit not work, and map labels contant flash
20161026 16:45:24< vultraz> constantly
20161026 16:46:11< vultraz> JyrkiVesterinen: as i said, the commit makes drawing happen on a timer
20161026 16:46:47< JyrkiVesterinen> Right. I was saying that what needs to be done is make the game handle that drawing happens on a timer.
20161026 16:47:36< vultraz> yes
20161026 16:47:40< vultraz> if that happens we could merge it
20161026 16:47:44< vultraz> and i really would like to
20161026 16:47:50< vultraz> since it's an incredibly performance boost for GUI2
20161026 16:48:07< vultraz> one gripe i've had with gui2 is it's always seemed a little less responsive than gui1
20161026 16:48:11< vultraz> but this fixes that
20161026 16:50:02< vultraz> incredible*
20161026 16:50:29< gfgtdf> vultraz: i wonder how it can be a performance boost in gui2. From looking at the commit it seems like he just took the gui2 drawing timer stuff and moved it to events.cpp so that it is used by the whole game.
20161026 16:50:30< tad_carlucci> I think "a little less responsive" is "a little understatement"
20161026 16:50:44< vultraz> gfgtdf: idk either but try it out and everything is much smoother
20161026 16:50:57< vultraz> gfgtdf: maybe it's the reduction of delay from 20 to 10?
20161026 16:52:13< vultraz> actually i should try that in master..
20161026 16:52:54< vultraz> waitn o
20161026 16:52:56< vultraz> that's in gui1
20161026 16:52:58< vultraz> lvl
20161026 16:53:00< vultraz> nvm*
20161026 16:53:32< vultraz> gfgtdf: i really have no idea then
20161026 16:53:41< vultraz> actually i think Aginor doesn't either
20161026 16:53:44-!- Kwandulin [~Miranda@p5DDD2B8F.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20161026 16:54:07< vultraz> oh
20161026 16:54:09< vultraz> hmmm
20161026 16:54:16< vultraz> I see  draw_interval = 30
20161026 16:54:23< vultraz> and in the change it's 20
20161026 16:55:11-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has joined #wesnoth-dev
20161026 16:55:38< vultraz> yeah actually I think that must be it
20161026 16:55:52< vultraz> GUI2 is certainly smoother if i make it match 20
20161026 16:57:33< gfgtdf> vultraz: does that branch also fix the other ui issiesisues like the faction select now showing and addons menu?
20161026 16:57:41< vultraz> i haven't tried
20161026 16:57:44< vultraz> but likelynot
20161026 16:58:25< tad_carlucci> What is that interval used for? If it's the timer for flags flapping and water waving, making it shorter might help with overall game smoothness but if going to make those elements seem frantic.
20161026 17:00:27-!- irker441 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20161026 17:00:27< irker441> wesnoth: Charles Dang wesnoth:master c70da182b176 / src/gui/widgets/window.cpp: GUI2: reduce draw interval from 30 to 20 https://github.com/wesnoth/wesnoth/commit/c70da182b176f607e9411e69d20b451eea683061
20161026 17:00:36 * vultraz kicks himself for thinking it was some arcane magic for 7 months when in reality it was one friggin number :|
20161026 17:01:32< vultraz> tad_carlucci: draw event timer. but terrain graphics like waves are separate, I think
20161026 17:01:58< tad_carlucci> vultraz, Building it now.
20161026 17:02:00< vultraz> tad_carlucci: even though the game is so must redrawing the game screen the water displays fine
20161026 17:03:20-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20161026 17:03:27< vultraz> Aginor: ^ so it seems i've found the cause of the gui2 boost :P
20161026 17:03:52< vultraz> JyrkiVesterinen: for the record, one other way we can boost gui2 performance is making ttext (the pango text stuff) use the text cache.
20161026 17:04:35< JyrkiVesterinen> You may want to mention the improvement in changelog and maybe players_changelog. :)
20161026 17:04:54< JyrkiVesterinen> Why did you ping me in the message above?
20161026 17:05:18< vultraz> since you were talking about performance
20161026 17:06:05< JyrkiVesterinen> Well, I simply responded to a message that seemed wrong.
20161026 17:06:24< JyrkiVesterinen> [19:44:27] <vultraz> gfgtdf: basically, the game would need to be made to only actually execute a draw if there IS something to draw
20161026 17:06:48< irker441> wesnoth: Charles Dang wesnoth:master 707621ebb469 / changelog players_changelog: Updated changelogs https://github.com/wesnoth/wesnoth/commit/707621ebb469b3f559878a86147cbae7e34d45af
20161026 17:06:49< JyrkiVesterinen> It looked like you wanted to undo the whole "draw on timer" change which was the whole point of the commit.
20161026 17:07:25< vultraz> JyrkiVesterinen: ah. perhaps i worded myself badly
20161026 17:07:56< vultraz> JyrkiVesterinen: what I meant was fire a draw event on a timer, but handle it in a way that doesn't cause the game to be so busy redrawing everything it can't do much else
20161026 17:08:44< JyrkiVesterinen> The normal way to do that is, of course, hardware rendering.
20161026 17:09:11< JyrkiVesterinen> With HW acceleration rendering dozens of units per frame wouldn't cause any trouble at all.
20161026 17:09:47< vultraz> right
20161026 17:09:55< vultraz> but this is supposed to be a precursor to Hw rendering
20161026 17:10:58< JyrkiVesterinen> I think "handling" higher number of redraw events by presumably skipping drawing of stuff that hasn't changed isn't really a precursor to HW rendering.
20161026 17:11:19< vultraz> no, removing the manual calls to CVideo::flip is
20161026 17:12:05< vultraz> flip calls SDL_RenderPresent
20161026 17:12:13< JyrkiVesterinen> I mean that the way towards HW rendering is a more simple rendering pipeline that eventually relies on raw power of GPUs alone.
20161026 17:12:15< vultraz> on the SDL_Window object
20161026 17:12:26< vultraz> we shouldn't need to call that manually
20161026 17:12:40< vultraz> we've, of course, been reducing the cases by transitioning to GUI2
20161026 17:12:52-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20161026 17:12:52-!- Kwandulin [~Miranda@p5DDD2B8F.dip0.t-ipconnect.de] has joined #wesnoth-dev
20161026 17:13:04< JyrkiVesterinen> The reason why HW rendering is not possible right now is precisely the existing optimizations (such as undrawing) which are necessary because we don't have HW rendering.
20161026 17:13:56< vultraz> Well, Aginor *did* say he was slightly unhappy with the design of the commit
20161026 17:14:23< vultraz> Since we now know the GUI2 performance increase was a draw_interval change, I would not be against a new method
20161026 17:14:30< vultraz> But the question is
20161026 17:14:35< vultraz> *what is that method*
20161026 17:15:28< JyrkiVesterinen> What I'm saying that optimizing handling of draw events wouldn't really make any progress towards achieving hardware rendering. That kind of optimizations would make the rendering pipeline more complex and likely harder to port to HW rendering.
20161026 17:15:44< vultraz> Ok, sure.
20161026 17:15:51< vultraz> But then what path should we take
20161026 17:16:05< vultraz> The end goal is HW rendering, great. How do we get there
20161026 17:16:17< JyrkiVesterinen> It's up to whoever wants to spend the effort.
20161026 17:16:54< vultraz> If you say that this path is counterproductive, then perhaps we should abandon it.
20161026 17:16:58< JyrkiVesterinen> Myself I have been focusing on fixing bugs that I keep running into.
20161026 17:17:16< vultraz> Sadly, as of right now we have no one fully dedicated to the HW project.
20161026 17:17:30< vultraz> Aginor is too busy to do it.
20161026 17:17:34< vultraz> It's outside my skill
20161026 17:17:41< vultraz> And no one else is really interested.
20161026 17:17:52< vultraz> So instead, we stick to micro-optimizations
20161026 17:18:00< JyrkiVesterinen> In that case, we probably shouldn't do anything about the HW project right now. Just put it on hold.
20161026 17:18:44< zookeeper> only the hypothetical person who would be willing to do most of the work can do anything about it, i'd imagine, so if there is no such person then hard to do anything but wait and hope for the best.
20161026 17:20:37< vultraz> I at least *hope* the stuff we've been doing (reducing custom blit calls, phasing out GUI1) will help in the long run
20161026 17:21:06< vultraz> I've been told the former will and the design of GUI1 seems to indicate it will help, so I'm hopeful
20161026 17:21:55< zookeeper> fair enough, i'm sure there's plenty of preparation work like that which will end up being helpful
20161026 17:22:26< vultraz> To be really honest, I'm starting to feel burned out after the past few months of gui2 work
20161026 17:22:39< vultraz> It's not easy :|
20161026 17:23:07< vultraz> And there seems to be no end of stuff to do.
20161026 17:23:15-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20161026 17:23:20< zookeeper> take a break, play some UtBS or something :p
20161026 17:23:22< vultraz> Not to mention the million administrative duties I have.
20161026 17:23:33< mordante> servus
20161026 17:24:02< JyrkiVesterinen> vultraz: I intentionally throttle my contributions, precisely to ensure that I don't even come close to burning out.
20161026 17:24:18< JyrkiVesterinen> You might want to consider something similar.
20161026 17:24:55< vultraz> JyrkiVesterinen: well... problem is, when that happens I feel like I need to make up the deficit or no one will :|
20161026 17:25:03< vultraz> ie, if someone works less, I feel like I need to work more
20161026 17:25:48< vultraz> Which is perhaps a bad outlook since then you end up doing too much..
20161026 17:26:11-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20161026 17:26:56< vultraz> But it's hard not to when I see the bug 1.14 deadline looming and there's *so much* left to do :(
20161026 17:27:02< vultraz> big*
20161026 17:27:17< gfgtdf> mordante: maybe you have an idea why the faction select dialog doesnt show in the gui2 wait dialog
20161026 17:27:25-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20161026 17:27:39< vultraz> if i take a break, who will finish the gui2 conversion?
20161026 17:27:42< vultraz> for example..
20161026 17:27:45< vultraz> :(
20161026 17:28:07< vultraz> gfgtdf, mordante: specifically, it's attempting to show a dialog from the post_build event of another
20161026 17:28:10< vultraz> keep thatin mind
20161026 17:28:41< gfgtdf> vultraz: didn't you say it alspo happened when you moved it out of of post_build?
20161026 17:29:39-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20161026 17:29:49< tad_carlucci> If you burn out and take a permanent break, what's to happen, then.  At some point we all learn to pace ourselves. If 1.14 slips it slips. It's not like we're under contract and don't get paid if we miss a deadline.
20161026 17:30:17< vultraz> tad_carlucci: we've made a public commitment for an "early 2017" release on Steam :/
20161026 17:30:52< vultraz> gfgtdf: yes it also did not work in pre_show
20161026 17:31:36< vultraz> tad_carlucci: we've been Greenlit on Steam Greenlight, and the page publicly says "early 2017".
20161026 17:32:01< vultraz> While it's not a hard deadline, I do not want to miss it 
20161026 17:32:23< vultraz> Look at Frogatto - they promised a October 2015 release and it's not out on Steam yet.
20161026 17:32:37< vultraz> when their greenlight campaign was in December *2013* :|
20161026 17:33:52< vultraz> "early 2017" gives us until March, maybe.
20161026 17:34:11< vultraz> It's the end of October, and we still have not:
20161026 17:34:18< vultraz> * finished the gui2 conversion
20161026 17:34:21< vultraz> * moved the bug tracker
20161026 17:34:26< vultraz> * finished the new website
20161026 17:34:37< vultraz> * drafted the announcemenet page
20161026 17:34:46< vultraz> * finished the forum phpbb update
20161026 17:34:50< vultraz> * finished UTBS v2
20161026 17:35:31< zookeeper> * finished all unit animations
20161026 17:35:33 * zookeeper ducks
20161026 17:35:42 * vultraz throws something at zookeeper
20161026 17:35:50< zookeeper> oww
20161026 17:36:38< vultraz> not to mention finishing the Steamworks registration and figuring out the entire Steam release process and schooling our packages and stuff on using it!
20161026 17:37:06< vultraz> and in the midst of that, fix bugs and improve performance.
20161026 17:37:52< zookeeper> so is there some reason why GUI1 _needs_ to be phased out ASAP?
20161026 17:37:55< vultraz> granted, I'm not solely responsible for all of these things, but I have the responsibility as project admin, release manager, and Board member to ensure they get done and done right!
20161026 17:38:03< zookeeper> like, a technical reason
20161026 17:38:16< vultraz> zookeeper: because gui1 sucks and it's evil and it doesn't work well
20161026 17:38:22< vultraz> and it looks worse too
20161026 17:38:48< tad_carlucci> From my perspective, the issue with pushing a Steam release is that gui2 step. Forget personal feelings and be honest, is there any reason 1.12.6 should/could not be the version on Steam if all the other customer-facing work you mentioned got done?
20161026 17:39:41< vultraz> Yes, because 1.12 is *so far behind* master in terms of improvements that we would *literally* be rewinding time 2 years.
20161026 17:39:42< zookeeper> vultraz, sounds like a lot of those other things would be more important so you could probably at least take a break from the GUI2 work if you wanted
20161026 17:40:32< JyrkiVesterinen> I wouldn't talk about rewinding time. Players use stable releases. Even existing players, let alone new players, don't have any idea about the improvements in 1.13.
20161026 17:40:46< JyrkiVesterinen> They won't feel like 1.12.6 is outdated.
20161026 17:40:57< vultraz> I think it's important to have 1.14 as the steam release
20161026 17:41:01< zookeeper> the problem would be getting people to do any serious work on 1.12 again, i'd think
20161026 17:41:08< vultraz> and that too
20161026 17:41:27< vultraz> no one wants to go back to working with gui1 and sdl 1.2 :|
20161026 17:42:41< zookeeper> if 1.12 wasn't quite that old then maybe we could make it the steam version, but i think at this point it's better to delay the steam release and have 1.14 than to use 1.12 for it
20161026 17:42:53< zookeeper> if one had to choose, that is
20161026 17:43:13< vultraz> Honestly, the thing that bothers me most right now is it feels like no one else is really as invested in the steam release as I am
20161026 17:43:39< JyrkiVesterinen> You may well be right. We are all volunteers after all.
20161026 17:43:53< zookeeper> well surely that doesn't come as a surprise? you're the person who's been pushing for steam after all.
20161026 17:44:15< vultraz> yeah
20161026 17:44:25-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20161026 17:44:29< vultraz> but someone to split the responsibilities with would be nice :(
20161026 17:44:48< vultraz> even the rest of the Board is rather removed. Dave is doing more, yes, but he has other projects under its purview to focus on.
20161026 17:45:27< vultraz> ice is doing some pr stuff related to setting up a webpage, and jet has essentially done nothing.
20161026 17:45:31< vultraz> but focus on frogatto
20161026 17:46:36< vultraz> and it's so frustrating when half of a group of people who are *supposed* to give a shit really feel like they don't :|
20161026 17:47:57 * JyrkiVesterinen feels kind of glad for not voting in the board election - at least the situation above isn't his fault 
20161026 17:49:38< gfgtdf> JyrkiVesterinen: by not voting you agreee with the voting of the others.
20161026 17:49:52< vultraz> dave took the lead on the iOS problem, yes (and im grateful he did), but that's sitting now while we wait back to hear from ElGallo.
20161026 17:49:52< JyrkiVesterinen> I know.
20161026 17:49:58< zookeeper> but that's different from the result being your fault :p
20161026 17:50:02< vultraz> ice is supposed to be our treasurer but i have no idea what the official state of that is
20161026 17:50:21< vultraz> and he's working on setting up a website so we can post info about the Board
20161026 17:50:31< zookeeper> oh, nice
20161026 17:50:35< vultraz> but we haven't even *decided how or where to make minutes/logs whatever public*!
20161026 17:50:45-!- Appleman1234 [~Appleman1@KD106181179234.au-net.ne.jp] has quit [Ping timeout: 265 seconds]
20161026 17:50:57< vultraz> I can't even get an answer out of jet whether *he will finish the desert elves or not*!
20161026 17:51:29< vultraz> and I just *know* it will be me who has to remind dave or whatever that we need to deal with Steamworks soon
20161026 17:52:20< vultraz> I don't know if shadowm plans to finish the new Website soon
20161026 17:52:41< vultraz> I don't know how to migrate our bug tracker and Aginor's over there in the corner saying we should look at however many different options
20161026 17:53:21< vultraz> and the bottom line is I have no idea what anyone is working on or is planning to do or when or how or even how to make this whole thing work!!!
20161026 17:54:11 * vultraz takes deep breaths
20161026 17:57:32< vultraz> oh and let's not forget I ALSO HAVE TO MANAGE ALL OUT SOCIAL MEDIA OUTREACH
20161026 17:57:36< vultraz> our*
20161026 17:58:11< vultraz> AND coordinate releases
20161026 17:58:12< tad_carlucci> So far it sounds like any new manager realizing, for the first time, he can't do it all and has to depend upon the team.
20161026 17:58:35< shadowm> I'm as disappointed as you that I haven't been able to focus on wesmere.
20161026 17:58:46< zookeeper> volunteer collaboration tends to get frustrating and slow indeed. that's why i've learned to try to avoid it in most cases, i just fail at the attempt more often than i'd like :p
20161026 17:59:19< vultraz> tad_carlucci: the problem is i feel like the team is way over *there* while I'm alone over here
20161026 18:00:46< vultraz> and i thought the Board would bring some semblance of 'team over here' and structure and order (at least to the more distribution-oriented problems) but it didn't.
20161026 18:02:35< shadowm> Hopefully you've communicated your concerns to them as well.
20161026 18:03:36< tad_carlucci> Its like the stages of grief, you're going through the normal process and will get either get through it or not. As to the board, specifically, well, it was a rush job so problems should be expected.
20161026 18:05:08< vultraz> I'm trying my best
20161026 18:06:25-!- Kwandulin [~Miranda@p5DDD2B8F.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20161026 18:08:37< tad_carlucci> One has to adjust ones expectations to ones capabilities. The trick being a project manager is learning to do that for the team as a whole and aligning each individual's capabiliries with the overall goals. Sometimes you can't get there and either need to change the team makup or change the goals.
20161026 18:09:47< vultraz> It's slightly easier when you can actually assign people tasks. Not so when everyone's a volunteer.
20161026 18:10:18< vultraz> I can't make any of you do anything 
20161026 18:10:56 * celticminstrel offers hugs
20161026 18:11:15< vultraz> thanks
20161026 18:11:18< vultraz> appreciated 
20161026 18:11:49< zookeeper> also, try to not get personally attached to things which are at all up to other people. if you're only attached to solo projects then you might still not get them done as planned, but at least the path is more straightforward and you don't get frustrated by other people :p
20161026 18:18:33 * tad_carlucci mutters something about water and horses drinking.
20161026 18:18:47< tad_carlucci> Seems to me one approach would be to make the water more appealing.
20161026 18:24:59< vultraz> what?
20161026 18:26:52< tad_carlucci> Getting people to work on things. In industry we typically raise the wages to entice someone to work on something. But there are other ways. People want some reward or recognition for their efforts. So if you need someone to do something, try to come up with a way they'll feel they get that.
20161026 18:28:07< vultraz> not so easy, that is
20161026 18:28:44< vultraz> financial compensation is wholly off the table.
20161026 18:29:41< zookeeper> speaking of which, i take it that lordbob isn't getting paid for the TRoW story art?
20161026 18:30:07< vultraz> no
20161026 18:30:33< zookeeper> okay, great
20161026 18:31:04< zookeeper> (because he's making a lot more than strictly needed :p)
20161026 18:34:11< tad_carlucci> Take Lord Bob, then, as an example. He said he's doing the work to experiment with new techniques. But he also put in a plug for his personal site. So, maybe, what that tells us is that some people will see enough reward if there is a way to link what they contribute to Wesnoth to other projects, or even products, they might have.
20161026 18:38:13< Elvish_Hunter> Hi all. Long time no see :)
20161026 18:38:31< irker441> wesnoth: Elvish_Hunter wesnoth:master d3516a1d3596 / data/tools/wmlscope: wmlscope: do not read whole files at once while calculating MD5 hashes https://github.com/wesnoth/wesnoth/commit/d3516a1d3596cb388421b3cd354b856c3c1a0d61
20161026 18:38:33< irker441> wesnoth: Elvish_Hunter wesnoth:master 33e03a5b8b8d / data/tools/wmllint: wmllint: added base_names and help_text to the list of translatable keys https://github.com/wesnoth/wesnoth/commit/33e03a5b8b8d01ec7cfc823f8f025672269268b8
20161026 18:38:35< irker441> wesnoth: Elvish_Hunter wesnoth:master 63ac16702d76 / data/lua/wml-tags.lua: [set_variable]: added support for abs=, root= and power= keys https://github.com/wesnoth/wesnoth/commit/63ac16702d76600d8bfed210ae8fb981724126cf
20161026 18:38:59< Elvish_Hunter> celticminstrel: ^
20161026 18:41:00 * celticminstrel opens PR 845.
20161026 18:41:17< gfgtdf> i wonder whether its perfroamcne wise a diference whether to write root=quare or power=0.5
20161026 18:42:14< Elvish_Hunter> Probably yes, otherwise I guess that Lua wouldn't have a math.sqrt() function at all.
20161026 18:42:40< celticminstrel> Not sure whether it's a matter of performance or precision...
20161026 18:43:25< tad_carlucci> Not sure anyone needs to do 10,000 sqrt() calls in an event so not sure it really matters.
20161026 18:43:39< celticminstrel> Yeah, performance definitely isn't important there.
20161026 18:47:16-!- mjs-de [~mjs-de@x4e301580.dyn.telefonica.de] has joined #wesnoth-dev
20161026 18:47:26-!- Appleman1234 [~Appleman1@KD106181169138.au-net.ne.jp] has joined #wesnoth-dev
20161026 18:49:18< irker441> wesnoth: Elvish_Hunter wesnoth:master 2ab52fdb3638 / changelog: changelog entry https://github.com/wesnoth/wesnoth/commit/2ab52fdb3638d6daae7e17057c345a3c45a742d9
20161026 18:49:44< celticminstrel> I feel like I forgot a changelog entry for something...
20161026 18:50:50< celticminstrel> Oh, maybe the deprecated AI functions? Though I think those were listed as deprecated in the AI refactor...?
20161026 18:51:36< celticminstrel> Hm, seems there's a few other things too.
20161026 18:53:39< celticminstrel> wedge009: Did PR 496 and 800 make it into the changelog?
20161026 18:53:58< celticminstrel> Oh. One of them included a changelog entry.
20161026 18:54:14< celticminstrel> Both did. Never mind, sorry to bother you.
20161026 18:57:44< celticminstrel> mattsc: I think the AI section of the changelog shouldn't be at the top...
20161026 18:59:54< mattsc> celticminstrel: I really don’t care one way or the other, but the changelog sections have traditionally (mostly, but not stricly) been ordered alphabetically.  That’s why it is where it is.
20161026 19:03:45< tad_carlucci> I was at 800x600 and started HttT .. oops I wanted full screen so click maximize .. crashes back to title screen. Add a debugger and catch all throw() .. nope, nothing throws. So why does maximizing while the black loading this-or-that is running abort to the titlescreen instead of finishing its work?
20161026 19:22:56< celticminstrel> mattsc: ...they have? I never noticed such a thing.
20161026 19:23:41< mattsc> celticminstrel: well, I trust that you can verify this yourself …  ;)
20161026 19:24:05< celticminstrel> gfgtdf: You removed fire_event_by_name, right?
20161026 19:24:14< gfgtdf> celticminstrel: yes i did
20161026 19:24:57< gfgtdf> celticminstrel: note that it wasn't in any release.
20161026 19:25:33< celticminstrel> gfgtdf: Right, but there was a changelog entry noting its addition, so I'm fixing that.
20161026 19:25:48< gfgtdf> celticminstrel: ok thx
20161026 19:26:00< celticminstrel> Changelog entry is needed for pentarctagon's [message] update...
20161026 19:26:52< vultraz> celticminstrel: thoughts on the slider snap behavior?
20161026 19:27:03< vultraz> celticminstrel: do you think it's acceptable?
20161026 19:27:05< celticminstrel> Haven't looked at it yet.
20161026 19:27:13< celticminstrel> I haven't built since pulling.
20161026 19:27:49< vultraz> ah, fuck. i just found a case where it doesn't work
20161026 19:28:33< vultraz> or maybe it's something elese..
20161026 19:29:43< celticminstrel> mattsc: Do you think I should mention the ai.aspects.attacks thing under Lua API or under AI?
20161026 19:29:44< vultraz> the dark forecasts lider isn't working :(
20161026 19:30:31< vultraz> slider
20161026 19:31:01< mattsc> celticminstrel: good question; I’d probably put it under AI, but not because of a real preference for it over the other
20161026 19:32:15< Shiki> Which wolves do you like more? https://i.imgbox.com/Dkc5JPCg.png I tend to leave them like it is now
20161026 19:32:41< irker441> wesnoth: Celtic Minstrel wesnoth:master 293151687cdc / data/lua/wml-flow.lua: Fix number-nil comparison in [for] https://github.com/wesnoth/wesnoth/commit/293151687cdc1651fb373a9f361ef136e51c100f
20161026 19:32:43< irker441> wesnoth: Celtic Minstrel wesnoth:master 8e992d04d18b / RELEASE_NOTES changelog: Update changelog and release notes https://github.com/wesnoth/wesnoth/commit/8e992d04d18b9771879328a4a6f782e185098efe
20161026 19:35:07< irker441> wesnoth: Celtic Minstrel wesnoth:lua_mapgen_stuff 264b5b50d756 / src/ (game_launcher.cpp play_controller.cpp): Improve handling of map generator errors https://github.com/wesnoth/wesnoth/commit/264b5b50d756a61395c2aa7aa6532b468ca0dc7b
20161026 19:35:09< irker441> wesnoth: Celtic Minstrel wesnoth:lua_mapgen_stuff 1b31646d6d6e / / (6 files in 4 dirs): Refactor Lua cave map generator https://github.com/wesnoth/wesnoth/commit/1b31646d6d6efc06aba3f639a9b2e0e0b8454ee8
20161026 19:35:11< irker441> wesnoth: Celtic Minstrel wesnoth:lua_mapgen_stuff 024b4f420059 / data/lua/cave_map_generator.lua: Lua Cave Mapgen: Support [chamber]chance= https://github.com/wesnoth/wesnoth/commit/024b4f420059ff1dc370d492a9ef3f54f06b2933
20161026 19:35:13< irker441> wesnoth: Celtic Minstrel wesnoth:lua_mapgen_stuff 7a11df76559c / data/ (4 files in 3 dirs): Lua Cave Mapgen: Support random chance of flipping the map https://github.com/wesnoth/wesnoth/commit/7a11df76559c73fd5e2107d66d7362d3fa1ca97a
20161026 19:35:15< irker441> wesnoth: Celtic Minstrel wesnoth:lua_mapgen_stuff 7c9f1a2968fc / data/lua/cave_map_generator.lua: Lua Cave Mapgen: Support scenario generation https://github.com/wesnoth/wesnoth/commit/7c9f1a2968fcabe201acc25403675cc3ec69055d
20161026 19:35:18< celticminstrel> Oh right, I forgot to look at your answer before pushing. Well, you could move it I guess.
20161026 19:36:06< celticminstrel> So anyone have any comments on 845?
20161026 19:36:26< vultraz> not i
20161026 19:37:03< mattsc> celticminstrel: Nah, that’s fine.  I think a very brief explanation as to why ai.get_attacks() is not deprectaed (or at least a reference to the changelog) would be good in the RN though.  Otherwise it’s one of those WTF? things …
20161026 19:38:32-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has quit [Quit: ChipmunkV]
20161026 19:43:27< vultraz> ok, I'm commenting out the slider snap code
20161026 19:43:46< vultraz> seems to take issue with this slider configuration in particular
20161026 19:44:22< vultraz> I'm thinking of reimplementing it differently.
20161026 19:44:32< vultraz> ie, smooth drag, but snap to the nearest value on release.
20161026 19:46:32< irker441> wesnoth: Charles Dang wesnoth:master d55265a86c9d / src/gui/widgets/ (slider.cpp slider.hpp): Slider: comment out slider snap-to-value code (e2d3d585be48) https://github.com/wesnoth/wesnoth/commit/d55265a86c9d6f78c84ab51b0241223b0c356db9
20161026 19:46:35< tad_carlucci> celticminstrel, one comment
20161026 19:46:49< vultraz> it seems that not moving the positioner except to snap to a value leads to too much distance between the mouse and itself, especially on large steps
20161026 19:47:12< celticminstrel> It should move when the mouse is more than halfway to the next point.
20161026 19:47:20< celticminstrel> Or to put it another way...
20161026 19:47:24-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20161026 19:47:30< celticminstrel> Snap to the point that's closest to the mouse position.
20161026 19:47:52< vultraz> perhaps
20161026 19:48:02< celticminstrel> That's not how you comment out large code blocks! The correct way is #if 0
20161026 19:48:04< vultraz> but that would put it ahead of the mouse
20161026 19:48:11< celticminstrel> So?
20161026 19:48:52< vultraz> id rather keep the positioner with the mouse
20161026 19:48:53< tad_carlucci> I've not looked at the code but I'm envisioning the issue being you have the mouse down and dragged N units and the item jumped .. but the mouse is still thinking it's not jumped. Perhaps a synthetic 'I just moves things, let's fake a mouse up mouse down to adjust'
20161026 19:49:11< vultraz> which is why im considering smooth drag, but snap on release
20161026 19:49:30< celticminstrel> I think I prefer non-smooth drag.
20161026 19:49:35< vultraz> also, seems like there's a bug with left/right arrows in the dark forecast slider...
20161026 19:49:37< vultraz> blagh
20161026 19:49:46< vultraz> (and it never reaches max, but that's another bug :| )
20161026 19:49:50< celticminstrel> But snap on release isn't terrible.
20161026 19:50:02< vultraz> (ie, max is 30 but you can't set it higher than 20)
20161026 19:50:11< celticminstrel> What's the step?
20161026 19:50:14< tad_carlucci> The problem with snap on release is you don't know where it'll snap to until you do it.
20161026 19:50:54< vultraz> 10
20161026 19:51:05< celticminstrel> Sounds like an off-by-one error.
20161026 19:51:19< vultraz> celticminstrel: https://gna.org/bugs/index.php?24531
20161026 19:51:23< tad_carlucci> "I dragged a bit but it snapped back!" "I dragged too far and it snapped to far!" Why can't it show me what it's proposing before I have to start over?
20161026 19:51:44< celticminstrel> Yeah, I prefer it snapping while it moves.
20161026 19:54:36< tad_carlucci> Basically, when you drag, after some distance it snaps. But at that point you have to adjust your idea of what the drag distance was. I drag >1/2 left and it snaps. Now I have to drag >1/2 right to move it back, or >1 more left to snap another left.
20161026 19:54:54< vultraz> right now it's > 1
20161026 19:54:59< vultraz> well, was
20161026 19:56:15< tad_carlucci> Whatever it is, when I release the mouse what it was just before release is what it should be after release.
20161026 19:58:02-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20161026 20:02:40< Shiki> zookeeper, pushed it. I have nothing on my list anymore.
20161026 20:05:45-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20161026 20:06:06< vultraz> ok, the arrow button thing is something to do with step..
20161026 20:14:39-!- atarocch [~atarocch@193.92.163.210] has joined #wesnoth-dev
20161026 20:16:37< gfgtdf> vultraz: i actually think the code for snapping should already be in the scrollbar widget since it has all those varaibles like step_size etc in ti.
20161026 20:23:14-!- matthiaskgr is now known as matthiaskrgr
20161026 20:24:06-!- horrowind [~Icedove@2a02:810a:8380:10a8:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20161026 20:25:21< celticminstrel> gfgtdf: I'd be satisfied with that, but vultraz seems to believe that smooth scrolling is extremely important for some reason.
20161026 20:31:18< gfgtdf> celticminstrel: i think the tscrollbar_container whcih handles the 'normal' scrollbars is completeley unarare of a the'number of items' or the real widget (lisbox/rewwview etc), so mostlikley is uses just scrollbar_items = pixle size, so it shodunt really ahv an effect on that
20161026 20:31:29< gfgtdf> unaware*
20161026 20:32:31-!- louis94 [~~louis94@91.178.240.45] has joined #wesnoth-dev
20161026 20:32:54< celticminstrel> Well, that's more or less right...
20161026 20:35:10-!- travis-ci [~travis-ci@ec2-54-159-13-111.compute-1.amazonaws.com] has joined #wesnoth-dev
20161026 20:35:11< travis-ci> wesnoth/wesnoth#11772 (master - 2ab52fd : Elvish_Hunter): The build has errored.
20161026 20:35:11< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/170868343
20161026 20:35:11-!- travis-ci [~travis-ci@ec2-54-159-13-111.compute-1.amazonaws.com] has left #wesnoth-dev []
20161026 20:36:51< zookeeper> Shiki, i don't see why you chose to change the submerge values
20161026 20:38:15< Shiki> Of the castle? I looked at it again and felt it looks better. The swamp and sunken castles have the same.
20161026 20:40:38< zookeeper> mmkay
20161026 20:41:52< vultraz> celticminstrel: smooth scrolling is very important
20161026 20:41:56< irker441> wesnoth: Severin Glöckner wesnoth:master 1606bb72cef2 / data/core/ (4 files in 4 dirs): Made Drake Flying Animations working on more terrains. https://github.com/wesnoth/wesnoth/commit/1606bb72cef2c2bde02ec3f2227c7b5d138ada91
20161026 20:41:58< irker441> wesnoth: Severin Glöckner wesnoth:master a8d63e063f8d / data/core/ (macros/animation-utils.cfg units/goblins/Wolf_Rider.cfg units/monsters/Wolf.cfg): reverted addition of ^Vhs https://github.com/wesnoth/wesnoth/commit/a8d63e063f8d31d12a1b3110cd193cf5dfa3179a
20161026 20:42:00< irker441> wesnoth: Severin Glöckner wesnoth:master 07aea4009188 / data/core/ (terrain.cfg units/goblins/Wolf_Rider.cfg units/monsters/Wolf.cfg): changed submerge to 0.4 in aquatic castles https://github.com/wesnoth/wesnoth/commit/07aea40091888a1a1196444c58a45eb926752f35
20161026 20:42:02< irker441> wesnoth: Lari Nieminen wesnoth:master 7950cdadd586 / data/core/ (5 files in 5 dirs): Merge pull request #824 from sevu/master https://github.com/wesnoth/wesnoth/commit/7950cdadd5867de5b9b8e9e671326fc37300bde0
20161026 20:43:02< celticminstrel> vultraz: I disagree.
20161026 20:43:09< vultraz> snapping sliders aren't widely used, I think
20161026 20:43:26< vultraz> at least not that i can see in windows, android, and ios
20161026 20:45:11< celticminstrel> The energy saver slider in System Preferences does, but it's a little different; it snaps to certain exact values but also permits in-between values.
20161026 20:45:50< celticminstrel> But that's a bad example, because time is continuous.
20161026 20:46:05< celticminstrel> The sliders in Wesnoth are discrete values.
20161026 20:47:16< gfgtdf> the firefox srcollbar of the hispory/bookmaks is discrete, the scollbar of the uselist of my irc client is discrete
20161026 20:47:23< gfgtdf> just of examples i see right now
20161026 20:47:28< gfgtdf> s/of/2
20161026 20:48:00< gfgtdf> explorers scollbar on the left side list si also discrete
20161026 20:48:07< celticminstrel> ...
20161026 20:48:18< celticminstrel> Why are you talking about scrollbars? We want sliders here.
20161026 20:49:02< celticminstrel> Ah, you're right, Firefox's history view scrolls one element at a time.
20161026 20:49:24< gfgtdf> wasn't  your point that chanign this in  the tscrollbar class woudl alos effect scrollbars (as opposed to putting extra stuff in tslider class liek vultraz did)
20161026 20:50:16-!- mjs-de [~mjs-de@x4e301580.dyn.telefonica.de] has quit [Remote host closed the connection]
20161026 20:50:58< Aginor> gfgtdf: no
20161026 20:51:57< celticminstrel> I found a snapping slider in System Preferences - the key repeat/delay sliders snap to fixed locations.
20161026 20:52:06< celticminstrel> No idea if the Windows equivalent do the same though.
20161026 20:53:01< Aginor> JyrkiVesterinen, vultraz: my plan for the hardware transitioning was as follows:
20161026 20:53:35< Aginor> remove as much of the the custom blitting as possible 
20161026 20:58:06< vultraz> and...?
20161026 20:58:33< Shiki> We should add the new terrains from Doofus to the players changelog
20161026 20:59:06< Aginor> enable hardware rendering and transition sdl_blit to ender textures from the surfaces (by virtue of converting each surface to a texture)
20161026 20:59:23< Aginor> introduce a render argument for a texture directly
20161026 20:59:32< Aginor> s/argument/command
20161026 21:00:10< Aginor> and start holding onto textures directly instead of surfaces, especially for things that are just re-used
20161026 21:00:55< Aginor> other things that's needed to make things better: better handling of windows or obscured data so that we know what to re-render instead of taking a copy of the framebuffer and restoring it
20161026 21:01:18< Aginor> as that'll be incredibly slow compared to just re-rendering the textures in the video memory
20161026 21:02:20< Aginor> but all of these are things that will introduce a lot of breakage, and aren't good to do when there's a lot of other churn in the same areas, so I'm waiting for 1.14 and the gui churn to settle down
20161026 21:02:39< Aginor> I also have a bunch of life stuff going on at the moment :)
20161026 21:03:17-!- travis-ci [~travis-ci@ec2-54-159-13-111.compute-1.amazonaws.com] has joined #wesnoth-dev
20161026 21:03:18< travis-ci> wesnoth/wesnoth#11773 (master - 8e992d0 : Celtic Minstrel): The build has errored.
20161026 21:03:18< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/170883453
20161026 21:03:18-!- travis-ci [~travis-ci@ec2-54-159-13-111.compute-1.amazonaws.com] has left #wesnoth-dev []
20161026 21:04:04< Aginor> vultraz: I also see that you picked some of my changes for that timer; You also need to pick the changes to ensure that there's never more than one draw-event on the queue
20161026 21:04:30< vultraz> oh?
20161026 21:04:35< vultraz> which part of that?
20161026 21:04:49< Aginor> that will stop the game from becoming unresponsive because it's not processing other events
20161026 21:05:06< Aginor> I don't remember, and I'm at work so I won't look them up
20161026 21:05:16< Aginor> I just look here when things are compiling :)
20161026 21:07:05-!- Waste [~Cracker@blk-138-73-98.eastlink.ca] has joined #wesnoth-dev
20161026 21:07:43< Aginor> but just having pulled in the timing issue will make some issues worse unless you also manage the event stack properly
20161026 21:08:37< vultraz> gahhhhhhhhhhhhhhhhhhhhhh
20161026 21:08:39 * vultraz curses
20161026 21:09:02< Aginor> and that's a terribly vague statement :D
20161026 21:09:05-!- Herk_ [b00af9e0@gateway/web/freenode/ip.176.10.249.224] has joined #wesnoth-dev
20161026 21:10:17< celticminstrel> I did comment on that commit about something similar.
20161026 21:11:04< Aginor> celticminstrel: you're completely correct
20161026 21:11:21< celticminstrel> The comment?
20161026 21:11:25-!- Waste [~Cracker@blk-138-73-98.eastlink.ca] has quit [Client Quit]
20161026 21:11:29< Aginor> but it also introduces even more complexity in an area when we need to simplify
20161026 21:11:46 * vultraz curses again
20161026 21:11:58< Aginor> celticminstrel: I haven't read the comment, I was just agreeing with you about having spotted the same thing
20161026 21:12:03< celticminstrel> Ah.
20161026 21:12:33< Aginor> vultraz: branches and PRs are good, it means that things can be checked out before they are comitted into the main branch
20161026 21:12:55< Aginor> yes, it's slower, but that's not a bad thing if it means that there
20161026 21:12:56< vultraz> as you have said before
20161026 21:13:00< Aginor> s less breakage
20161026 21:13:32< celticminstrel> gfgtdf should probably use more branches+PRs too. (Perhaps I should too.)
20161026 21:14:15-!- louis94 [~~louis94@91.178.240.45] has quit [Quit: Konversation terminated!]
20161026 21:14:18< celticminstrel> Speaking of PRs, what about 845?
20161026 21:14:45< vultraz> no comment
20161026 21:14:45< gfgtdf> vultraz: did you try moving the tmp_join_game::post_buidl stuf completeley out of tmp_join_game? to game_init/multiplayer.cpp 
20161026 21:14:59< vultraz> gfgtdf:  i could but thatwould mean copying some functions like get_scenarios
20161026 21:15:03< vultraz> get_scenario
20161026 21:15:06< vultraz> and it would be messy
20161026 21:15:12< vultraz> celticminstrel: feel free to merge
20161026 21:15:44< celticminstrel> gfgtdf: Do you want to look at it before merging?
20161026 21:15:47< gfgtdf> vultraz: but itz currently seems liek this issue block the next relese and we have night a plan on how to handle it nor someone who intends to work on it.
20161026 21:15:55< celticminstrel> Like testing or something.
20161026 21:16:20< celticminstrel> gfgtdf definitely has a point on the flg issue.
20161026 21:17:10< gfgtdf> celticminstrel: it looks liek with the new flip syntex its not possible to have both flipx and flipy
20161026 21:18:05< celticminstrel> Yeah, I was thinking of adding something like flip_xy but hadn't gotten that far.
20161026 21:20:05< gfgtdf> celticminstrel: i wonder whats the reson for those changes in  src/play_controller.cpp
20161026 21:21:01< gfgtdf> celticminstrel: the commit mesage says "Improve handling of map generator errors " but the radnom map generator is called before that.
20161026 21:21:29< gfgtdf> celticminstrel: so i don't see how it is related to map geration erros.
20161026 21:22:39-!- Herk_ [b00af9e0@gateway/web/freenode/ip.176.10.249.224] has quit [Quit: Page closed]
20161026 21:23:03< gfgtdf> celticminstrel: also i still think [settings] shdoul be renamed to [scenario]
20161026 21:24:51< celticminstrel> gfgtdf: Huh? Are you sure? That change turned a crash into an error message dialog.
20161026 21:25:06-!- JyrkiVesterinen [~JyrkiVest@87-92-57-246.bb.dnainternet.fi] has quit [Quit: Going to bed]
20161026 21:25:14< celticminstrel> A crash due to a Lua error in the map generator.
20161026 21:26:16< gfgtdf> celticminstrel: yes i am sure that the lua generator is not called there, but the code that actuall parses the returned string into a c++ map object is there so mmostlikeley thats the reason.
20161026 21:26:24< celticminstrel> Ah, okay.
20161026 21:26:36< celticminstrel> That makes sense, since the specific error was that the string was missing the last tile.
20161026 21:26:47< celticminstrel> I'd forgotten that.
20161026 21:27:00< celticminstrel> So it was due to a logic error in the Lua, not a syntax or runtime error.
20161026 21:28:27< irker441> wesnoth: Celtic Minstrel wesnoth:lua_mapgen_stuff f50ed4b747f6 / data/lua/cave_map_generator.lua: Lua Cave Mapgen: Support scenario generation https://github.com/wesnoth/wesnoth/commit/f50ed4b747f6a8fc1e3143951b8fe88ffc48acda
20161026 21:29:21-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20161026 21:34:06< gfgtdf> celticminstrel: added seom comment in the pr
20161026 21:34:36< gfgtdf> celticminstrel: in the particluar i think changign https://github.com/wesnoth/wesnoth/blob/master/src/gui/dialogs/loadscreen.cpp#L121 to catch(...) i a better fix for the crash in play_cotnroller.cpp
20161026 21:34:46< gfgtdf> s/i/is
20161026 21:37:43-!- Shiki [~Shiki@141.39.226.226] has quit [Quit: Verlassend]
20161026 21:42:51< celticminstrel> Hmm, that's a good point, since it's only storing the exception for rethrow. I guess I could try it.
20161026 21:54:42-!- travis-ci [~travis-ci@ec2-54-158-65-122.compute-1.amazonaws.com] has joined #wesnoth-dev
20161026 21:54:43< travis-ci> wesnoth/wesnoth#11775 (lua_mapgen_stuff - 7c9f1a2 : Celtic Minstrel): The build has errored.
20161026 21:54:43< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/170884659
20161026 21:54:43-!- travis-ci [~travis-ci@ec2-54-158-65-122.compute-1.amazonaws.com] has left #wesnoth-dev []
20161026 21:58:26-!- horrowind [~Icedove@2a02:810a:8380:10a8:21b:fcff:fee3:c3ff] has quit [Quit: horrowind]
20161026 22:16:19-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20161026 22:43:16-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 252 seconds]
20161026 22:45:41-!- travis-ci [~travis-ci@ec2-54-158-65-122.compute-1.amazonaws.com] has joined #wesnoth-dev
20161026 22:45:42< travis-ci> wesnoth/wesnoth#11778 (master - 7950cda : Lari Nieminen): The build passed.
20161026 22:45:42< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/170905097
20161026 22:45:42-!- travis-ci [~travis-ci@ec2-54-158-65-122.compute-1.amazonaws.com] has left #wesnoth-dev []
20161026 22:53:54-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20161026 23:04:25-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 250 seconds]
20161026 23:26:49-!- tad_carlucci [~lundberg@173.217.65.103] has quit [Quit: Off to resolve a merge conflict between the wife and husband branches of my real life.]
20161026 23:40:28-!- atarocch [~atarocch@193.92.163.210] has quit [Remote host closed the connection]
--- Log closed Thu Oct 27 00:00:27 2016