--- Log opened Thu Sep 22 00:00:32 2016
20160922 00:05:22-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev
20160922 00:12:21-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160922 00:18:56< EliDupree> LOL, I go to mess with EoHS after not looking at it for ages, and within 3 minutes of looking at the source code, I notice a typo in one of my descriptions that has been there for ages
20160922 00:19:28< celticminstrel> Heh.
20160922 00:20:35< EliDupree> "Your apprentices can any spell, not just ones you know."
20160922 00:25:42< EliDupree> I could really use an automated system for testing… Recently, I've been coding in Rust, which makes it REALLY easy to make unit tests, and now working with wesnoth is making me miss that a lot
20160922 00:26:59< EliDupree> I seem to remember that a lot of my releases had issues like "ERA IS COMPLETELY BROKEN IN MAPS WITH X" because my Lua just happened to not take into account X, and I just happened not to test on that map before releasing
20160922 00:35:26-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160922 00:36:04< EliDupree> I guess I could make an EoHS-test map that invokes a predetermined sequence of events when it starts (like casting every spell in various situations, etc.) and a shell script that runs wesnoth with that map, then make sure to run the script before every release
20160922 01:01:11-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds]
20160922 01:01:32-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160922 01:11:11-!- DeFender [~DeFender1@dsl217-132-173-106.bb.netvision.net.il] has quit [Ping timeout: 252 seconds]
20160922 01:12:22-!- DeFender [~DeFender1@89-138-245-35.bb.netvision.net.il] has joined #wesnoth-dev
20160922 01:22:25-!- gfgtdf_ [~chatzilla@x4e368599.dyn.telefonica.de] has joined #wesnoth-dev
20160922 01:25:35-!- gfgtdf [~chatzilla@x4e369e89.dyn.telefonica.de] has quit [Ping timeout: 260 seconds]
20160922 01:25:50-!- gfgtdf_ is now known as gfgtdf
20160922 01:27:18-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160922 01:31:32< EliDupree> What I really want, though, is a way to speed up my develop/test cycle. In order to reload the data and start a new game each time, I have to do something like 12 separate clicks with a wait in between each. I CAN skip the clicks by invoking wesnoth from the command line, but then I have to start the whole wesnoth instead of just reload the data, and there's extra bother switching between the text editor and command prompt as well.
20160922 01:32:07-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 265 seconds]
20160922 01:32:36< EliDupree> I have a script that automatically terminates and reruns a command whenever I change a file. This is good for a lot of things, but I'm a little worried that Wesnoth, with its idiosyncratic code base, might end up corrupting preferences files if I repeatedly terminated it at random times.
20160922 01:32:38-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160922 01:33:12< EliDupree> (Does anyone know if wesnoth is resilient against that? If it is, that would be convenient)
20160922 01:34:32-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 265 seconds]
20160922 01:37:05-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160922 01:42:12< celticminstrel> If you terminate it ungracefully, it won't save prefs.
20160922 01:42:56< celticminstrel> So I think it's probably resilient against that. Not 100% certain.
20160922 01:43:14-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 265 seconds]
20160922 01:43:55< EliDupree> That makes sense. And anyway right now I have a clean install, and my add-on is backed up and on GitHub, so even if it butchers everything, I can just reinstall and not try that the next time
20160922 01:44:44< shadowm> If something bad happens to the preferences file you can just delete it and it'll be recreated on the next run. You won't lose any add-ons.
20160922 01:44:46< celticminstrel> Would be nice to be able to press F5 in-game and have it reload the scenario and restart the game...
20160922 01:45:17< EliDupree> Yeah, that would pretty much be the best case scenario for me
20160922 01:45:44< shadowm> There is also no way to lose add-ons without clicking on things on the add-ons manager.
20160922 01:45:57< shadowm> Or the remove add-ons dialog.
20160922 01:46:54< EliDupree> Yeah, I wasn't actually worried about losing add-ons, just about the possibility of leaving a preferences file in a state that would make wesnoth have trouble starting
20160922 01:47:04-!- Bonobo [~Bonobo@2001:44b8:254:3200:94b7:4d7d:b28e:e338] has joined #wesnoth-dev
20160922 01:48:55-!- Jetrel_ [~Jetrel@2001:558:6014:1e:2422:435:dd84:bbf3] has quit [Quit: "The highest possible stage in moral culture is when we recognize that we ought to control our thoughts." - Charles Darwin]
20160922 01:49:31< shadowm> vultraz: I saw your screenshot on Facebook. Maybe instead of having a dedicated line for the leader's gender, just put the icon next to the unit type? (Also, what's the icon for random gender?)
20160922 01:50:16< vultraz> https://github.com/wesnoth/wesnoth/blob/master/images/icons/icon-random.png
20160922 01:50:29< tad_> EliDupree: I wipe out config/cache andprefs regularly.
20160922 01:50:51-!- Jetrel [~Jetrel@2001:558:6014:1e:2422:435:dd84:bbf3] has joined #wesnoth-dev
20160922 01:50:57< shadowm> I hope that won't parse as random unit type instead.
20160922 01:51:26< shadowm> tad_: Why?
20160922 01:52:06< tad_> shadowm: I get to points where I want to be sure I'm starting from a known point. Best known point I can think of is 'nothing known, at all."
20160922 01:52:15< vultraz> shadowm: it explicitly says random if it's a random type
20160922 01:52:20< vultraz> which is why i put gender on its own line
20160922 01:52:32< vultraz> since there are 3 randoms here
20160922 01:52:35< celticminstrel> shadowm: I imagine that combining that symbol with the actual name of a unit is unlikely to suggest "random unit type".
20160922 01:52:39< vultraz> random faction, random leader, random gender
20160922 01:53:12< shadowm> tad_: ... Okay.
20160922 01:53:42< shadowm> Just FYI, if you need to test something with the defaults for one session only and you are running Linux, you can use the utils/wesnoth-defaults wrapper I wrote.
20160922 01:54:04< shadowm> That way you don't need to mess around with your user config and user data directories.
20160922 01:54:47< shadowm> +Linux or any other POSIX platform
20160922 01:56:21< tad_> shadowm: I just rm the wesnoth directories. Fast. clean and sets me back to jsut like I new user
20160922 01:56:44< shadowm> And I'm telling you you don't need to do that if it's only for one time.
20160922 01:57:29< tad_> shadowm: What would I want to restore? That I completed a campaign?
20160922 01:57:39< shadowm> But if you insist, okay, sure, I'd just rather not see it being proposed as a general recommendation.
20160922 01:58:43< gfgtdf> EliDupree: i thought most of your codes are in lua? so you shouldnt need to reload wml if you want to save them
20160922 01:58:49< gfgtdf> change them*
20160922 01:59:34< EliDupree> They are, but they're embedded in WML so that they can be shared with other players without them having to download the add-on.
20160922 01:59:47< celticminstrel> I was about to bring up that possibility.
20160922 02:00:02< celticminstrel> I assume you're working on 1.12.
20160922 02:00:05< EliDupree> yes
20160922 02:00:39< celticminstrel> 1.13 has auto-downloading (no idea how well it works though).
20160922 02:01:07< EliDupree> I suppose I could make a scheme that allows me to use them through wesnoth.require when I'm testing locally, but I would still have to restart the scenario in in a lot of cases
20160922 02:02:21< gfgtdf> EliDupree: in one of my addons i use code liek this: http://pastebin.com/ryfdmzU8so that when developing my addon i can defeine PYR_NPT_REQUIRE_DOWNLOAD and the code gets loaded dynamically
20160922 02:02:43< gfgtdf> EliDupree: you need to make your fles begine with --<< and end with -->> though
20160922 02:02:55< gfgtdf> EliDupree: do that they work both as wml and as lua file
20160922 02:02:58< EliDupree> That link doesn't work
20160922 02:03:10< EliDupree> Yeah, I already have a scheme like that
20160922 02:03:31< gfgtdf> EliDupree: http://pastebin.com/raw/ryfdmzU8
20160922 02:03:40-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160922 02:04:21< gfgtdf> Eli
20160922 02:04:39< gfgtdf> EliDupree: and the like {LUAREQUIRE varname ~add-ons/... /filenamelua}
20160922 02:04:51< gfgtdf> filename.lua*
20160922 02:08:02< gfgtdf> then*
20160922 02:13:13< vultraz> celticminstrel: i tested the autodownloading, it works fine
20160922 02:14:05< celticminstrel> vultraz: So, does it download the full addon, or a diff?
20160922 02:14:15< EliDupree> I guess I can add that to the long list of wasteful hacks that I still have to use but might not have to at some point in the future
20160922 02:14:18< celticminstrel> (If you have an outdated version.)
20160922 02:14:57< vultraz> addons are always downloaded in full when updating
20160922 02:15:05< vultraz> they have never downloaded diffs
20160922 02:15:16< celticminstrel> This should be changed.
20160922 02:15:23< shadowm> Yes.
20160922 02:15:26< celticminstrel> But first, I need to make campaignd work without sendfile.
20160922 02:15:28< shadowm> Patches welcome. :p
20160922 02:15:32< celticminstrel> (Or wait for someone else to do it.)
20160922 02:15:52< shadowm> https://wiki.wesnoth.org/NotSoEasyCoding#Incremental_upgrades_and_uploads
20160922 02:15:56< vultraz> what is this 'sendfile' you people keep talking about
20160922 02:16:04< shadowm> I guess I should remove the section directly beneath since loonycyborg didn't.
20160922 02:16:53< tad_> SENDFILE(2)                                                                           Linux Programmer's Manual                                                                           SENDFILE(2)  NAME        sendfile - transfer data between file descriptors
20160922 02:17:50< vultraz> "As of 1.13.1+dev, GUI2's core and framework (including widgets and their APIs) are completely unmaintained." <- is false, as of now.
20160922 02:17:51< celticminstrel> I think a SHA256 hash is a valid WML key, so you could store the hashes of each file in a hashes.cfg as <hash>=<filename>.
20160922 02:18:11< shadowm> vultraz: Update the page then. This is supposed to be the active devs' responsibility.
20160922 02:18:14< celticminstrel> vultraz: I dunno, it's not exactly well-maintained?
20160922 02:18:28< vultraz> maybe not well, but not totally not.
20160922 02:18:29< celticminstrel> But it has gotten some significant updates, so I guess unmaintained is false.
20160922 02:19:07< celticminstrel> I think I know how to fix matrix placement in the presence of linked groups. Just need to actually get around to it.
20160922 02:19:32< celticminstrel> (Also maybe think about adding that num-cols hint thing.)
20160922 02:19:49-!- shadowm is now known as shadowltraz
20160922 02:19:57< shadowltraz> what is this 'matrix' you people keep talking about
20160922 02:20:01-!- shadowltraz is now known as shadowm
20160922 02:20:13< shadowm> (No, but seriously, I'm curious. What is it?)
20160922 02:20:28< celticminstrel> Theoretically, either a flow-layout or a table/spreadsheet-layout.
20160922 02:20:57< celticminstrel> flow-layout being how text normally is laid out (except with widgets).
20160922 02:21:13 * matthiaskrgr hands shadowm the red pill 
20160922 02:21:15< shadowm> GUI2 itself does table layout.
20160922 02:21:22< celticminstrel> I think the original purpose was the latter.
20160922 02:21:29< celticminstrel> Yes, grids are also a table layout.
20160922 02:21:43< gfgtdf> a flow layout panel might also be nice to have.
20160922 02:21:51< shadowm> So why theoretically? Is the flow layout option not implemented/finished yet?
20160922 02:21:52< celticminstrel> But grids don't like adding new elements after they're built.
20160922 02:22:04< celticminstrel> Actually, at the moment, it's a flow-layout.
20160922 02:22:07< vultraz> it's not finished, no
20160922 02:22:33< celticminstrel> And applying a linked group to the cells makes it a table layout, but it doesn't actually work correctly (gets cut off).
20160922 02:23:01< vultraz> and listbox::clear() doesn't work with it
20160922 02:23:08< vultraz> and there's no way to specify row length
20160922 02:23:11< celticminstrel> I'm guessing mordante intended it to be a table layout, but I implemented it as flow instead.
20160922 02:23:14< vultraz> or consistent cell size
20160922 02:23:23< shadowm> Sounds interesting nevertheless. Anywhere this is currently used in practice?
20160922 02:23:26< celticminstrel> vultraz: That's expected of a flow-layout. :P
20160922 02:23:33< celticminstrel> shadowm: Hotkey prefs, at the top.
20160922 02:23:37< vultraz> but I want table layout :|
20160922 02:23:45< celticminstrel> Yeah, I realize this.
20160922 02:23:52< shadowm> Ohhh, so that's how that weird thing was implemented.
20160922 02:24:09< shadowm> I was wondering why it seems to violate the axioms of GUI2 layout.
20160922 02:24:21< celticminstrel> Which axioms specifically?
20160922 02:24:29< celticminstrel> Things being aligned or something?
20160922 02:24:42< shadowm> Yes, the grid thing.
20160922 02:25:03< celticminstrel> I'm pretty sure it would be pretty easy to make a container widget that just lets you add things to specific locations.
20160922 02:25:58< celticminstrel> (There are certainly limited uses for such a thing, too. >_> Probably won't bother doing it, unless someone wants it for eg the storyscreen.)
20160922 02:26:01< shadowm> If you've made widgets before, sure. I only made a couple of changes to one and the framework seemed intimidating enough to me.
20160922 02:26:25< vultraz> oh, hey, I've done more than you than :D
20160922 02:26:33< shadowm> And eyeballed a PR involving another.
20160922 02:26:54< shadowm> Yes, vultraz, let's turn this into a contest of numbers.
20160922 02:27:07< vultraz> It's not a contest of numbers
20160922 02:27:24< shadowm> I'm aware that you've done a lot of tinkering with GUI2 in my absence and that's great.
20160922 02:29:36< shadowm> Maybe one of these days we should play a 1v1 on the trunk server to give me an excuse to look at the newer new lobby.
20160922 02:29:48< shadowm> And game setup UI.
20160922 02:29:55< vultraz> Not until after 1.13.6
20160922 02:29:59< shadowm> Which is part of the lobby I guess but still.
20160922 02:29:59< vultraz> It's not done yet.
20160922 02:30:02< shadowm> Yes, obviously.
20160922 02:33:26< shadowm> The help browser could probably use a flow layout container.
20160922 02:33:44< shadowm> Whenever you get around to porting that to GUI2, I mean.
20160922 02:33:55< vultraz> That's dead last along with the main game UI
20160922 02:34:55< shadowm> But it's a 1.14 target anyway, right?
20160922 02:35:11< vultraz> We'll see.
20160922 02:37:32-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Quit: wedge009]
20160922 02:37:52< vultraz> GUI2 is not easy to work with
20160922 02:38:06< vultraz> and I have been assigned to redesign the canvas implementation :|
20160922 02:44:28< celticminstrel> I have Ideas for the help browser.
20160922 02:44:30-!- jamibaraki [3d7d7260@gateway/web/freenode/ip.61.125.114.96] has quit [Quit: Page closed]
20160922 02:45:01< celticminstrel> I think GUI2 has nearly everything needed for it.
20160922 02:45:23< celticminstrel> What it needs is a way to have a floating image that text wraps around.
20160922 02:46:01< celticminstrel> Which apparently requires delving into Pango land. :/
20160922 02:46:18< vultraz> oh deer god
20160922 02:46:33< celticminstrel> Why are you invoking the god of deer?
20160922 02:55:29-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20160922 02:57:22< vultraz> it's my thing
20160922 02:57:53-!- gfgtdf [~chatzilla@x4e368599.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 48.0.2/20160823121617]]
20160922 02:57:54< celticminstrel> You should just change your name to Cernunnos then.
20160922 02:58:56< tad_> Just `s/ of d/ of b/` and you're fine
20160922 03:02:01-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 265 seconds]
20160922 03:04:58-!- vultraz [~chatzilla@124.109.10.167] has joined #wesnoth-dev
20160922 03:04:58-!- vultraz [~chatzilla@124.109.10.167] has quit [Changing host]
20160922 03:04:58-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160922 03:05:07-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Quit: Page closed]
20160922 03:06:27< vultraz> Artemis is a more interesting deer god
20160922 03:06:55< celticminstrel> Sure, if you'd rather be Artemis, I won't complain.
20160922 03:13:31-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160922 03:16:14< vultraz> not really :P
20160922 03:16:24< vultraz> I just like saying 'deer' instead of 'dear'
20160922 03:17:13< celticminstrel> :|
20160922 03:18:08< vultraz> now, you said you'd explain this diff_track_attribute today?
20160922 03:20:10< celticminstrel> Ah.
20160922 03:20:47< celticminstrel> I'm actually not sure how it works, but I guess it's related to the config diff system.
20160922 03:21:15< vultraz> I see
20160922 03:21:30< vultraz> are config diffs inefficient?
20160922 03:21:58< celticminstrel> Well, I assume they involve walking both configs in full, so probably not super-efficient.
20160922 03:22:11< vultraz> hm
20160922 03:22:40< vultraz> perhaps that's why gfgtdf said it's a problem that mp wait works solely with config diffs?
20160922 03:23:20< vultraz> i'm not sure why you'd need config diffs there anyway..
20160922 03:23:25< vultraz> or in the lobby, for that matter
20160922 03:23:56< celticminstrel> It may be inefficient in terms of CPU, but on average it's probably more efficient in terms of network traffic.
20160922 03:24:19< vultraz> instead of just... sending an attribute and appending it to the client config
20160922 03:24:47< vultraz> or tag
20160922 03:25:01< celticminstrel> Suppose there's 1000 games active on the server. Resending the full gamelist is obviously silly if the only thing that has happened is a new game being created, or a game ending, or one game changing settings.
20160922 03:25:18< vultraz> right
20160922 03:25:26< celticminstrel> [Sep 21@11:24:19pm] vultraz: instead of just... sending an attribute and appending it to the client config
20160922 03:25:27< celticminstrel> [Sep 21@11:24:47pm] vultraz: or tag
20160922 03:25:28< celticminstrel> Uhh. Exactly how is that different from sending a diff and applying it on the client?
20160922 03:26:12< vultraz> well, no diff would be involved... just say "change/remove/append this tag/key"
20160922 03:26:45< vultraz> you make it sound like it walks each config for the difference and then sends the diff
20160922 03:26:50< vultraz> instead of sending the change
20160922 03:27:07< vultraz> directly
20160922 03:27:43< celticminstrel> "change/remove/append this tag/key" seems to pretty concisely describe what a diff is.
20160922 03:27:55< vultraz> blah, nevermind
20160922 03:28:33< celticminstrel> I would suspect that applying a diff is faster than creating it.
20160922 03:29:03< celticminstrel> And it would be totally possible to build the diff as changes come in, rather than walking the config. No idea if this is done.
20160922 03:29:10< celticminstrel> (On the server side, that is.)
20160922 03:29:35-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160922 03:29:38< vultraz_iOS> Maybe I'm thinking of it wrong
20160922 04:15:02-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160922 04:19:32-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 255 seconds]
20160922 04:43:53-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160922 05:39:18-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has quit [Quit: Connection closed for inactivity]
20160922 05:44:23-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20160922 05:53:49-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The computer falls into a deep sleep, to awake again some other day!]
20160922 06:01:30-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has joined #wesnoth-dev
20160922 06:37:20-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160922 06:41:43-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 255 seconds]
20160922 06:54:52-!- Kwandulin [~Miranda@p200300760F2C717FF18C04240C29129C.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160922 07:01:17-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 250 seconds]
20160922 07:01:52-!- atarocch [~atarocch@93.56.160.28] has joined #wesnoth-dev
20160922 07:02:37-!- boucman_work [~boucman@bob75-2-81-56-46-209.fbx.proxad.net] has joined #wesnoth-dev
20160922 07:14:20-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160922 07:25:53-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20160922 07:33:26-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160922 07:38:02-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 265 seconds]
20160922 07:49:31-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160922 07:51:37-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 265 seconds]
20160922 07:51:38-!- wedge010 is now known as wedge009
20160922 08:03:56-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160922 08:09:01-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 265 seconds]
20160922 08:43:21-!- irker227 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160922 08:43:21< irker227> wesnoth: Charles Dang wesnoth:master 23106671f6cc / src/gui/ (7 files in 4 dirs): GUI2: cleaned up utils::string_map usage when appropriate https://github.com/wesnoth/wesnoth/commit/23106671f6cc0a0ea9d0ec18c1fd212dacc6b4b3
20160922 08:49:11-!- horrowind [~Icedove@2a02:810a:8380:10a8:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20160922 08:56:54-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20160922 08:58:45-!- Duthlet [~Duthlet@dslb-146-060-179-135.146.060.pools.vodafone-ip.de] has joined #wesnoth-dev
20160922 08:58:51-!- Nobun [~nobun@5.170.112.89] has joined #wesnoth-dev
20160922 09:14:53-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160922 09:19:42-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 264 seconds]
20160922 09:20:03-!- travis-ci [~travis-ci@ec2-54-80-234-64.compute-1.amazonaws.com] has joined #wesnoth-dev
20160922 09:20:04< travis-ci> wesnoth/wesnoth#11132 (master - 2310667 : Charles Dang): The build passed.
20160922 09:20:04< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/161853142
20160922 09:20:04-!- travis-ci [~travis-ci@ec2-54-80-234-64.compute-1.amazonaws.com] has left #wesnoth-dev []
20160922 09:35:34< Soliton> shadowm: there are a lot of quotes around variable expansions missing in wesnoth-defaults. it won't work with parameters with whitespace. also the EXIT trap is enough. it'll run no matter how the script exits.
20160922 09:55:01-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has quit [Quit: .]
20160922 09:57:16-!- Appleman1234_ [~Appleman1@KD119104043017.au-net.ne.jp] has joined #wesnoth-dev
20160922 10:00:41-!- Appleman1234 [~Appleman1@KD119104058103.au-net.ne.jp] has quit [Ping timeout: 248 seconds]
20160922 10:08:56-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has joined #wesnoth-dev
20160922 10:10:14< zookeeper> hmh. freeze on an AI turn.
20160922 10:11:53< JyrkiVesterinen> That sounds very similar to https://github.com/GregoryLundberg/wesnoth/issues/38 ...
20160922 10:12:12< JyrkiVesterinen> If you can reproduce the freeze, does it happen with Monte carlo simulation disabled?
20160922 10:12:42< zookeeper> and now a freeze just trying to attack a unit
20160922 10:14:08-!- Appleman1234_ is now known as Appleman1234
20160922 10:15:00< zookeeper> seems to reproduce
20160922 10:15:48< zookeeper> yeah, switching monte carlo off fixes the freeze on attack
20160922 10:16:43< JyrkiVesterinen> Oh. I must have broken something in https://github.com/wesnoth/wesnoth/commit/9f379b96e34236d087d1788add77698602dd78b9 .
20160922 10:17:07< JyrkiVesterinen> Please send a save file somewhere (e.g. the bug tracker).
20160922 10:18:01-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 265 seconds]
20160922 10:20:23< zookeeper> want just a download link?
20160922 10:21:28< JyrkiVesterinen> That's fine too.
20160922 10:21:46< zookeeper> https://dl.dropboxusercontent.com/u/63964618/wesnoth/HttT-The_Princess_of_Wesnoth_Turn_6.gz
20160922 10:22:05< zookeeper> it's from an edited scenario but i guess that doesn't matter
20160922 10:22:15< zookeeper> just try to attack the spearman with delfador
20160922 10:22:41< JyrkiVesterinen> File saved. I'll investigate the freeze later this week.
20160922 10:24:20< zookeeper> great
20160922 10:27:53-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160922 10:41:11< matthiaskrgr> mmh, running optipng script:   11973 binary files changed 168747548 b (160 Mb) -> 166376207 b (158 Mb) => [-2371341 b (-2 Mb)]      
20160922 10:42:08< matthiaskrgr> I did run it several times though with the optimizers isolated.  'problem' was if let's say optipng would change pixel colors, the script would revert the optimization, even if advdef and convert would reduce size 
20160922 10:42:36< matthiaskrgr> I worked around this by just running the script several times with only one crusher enabled at a time 
20160922 10:43:57< zookeeper> why is it a problem if the script doesn't allow pixel colors to be changed?
20160922 10:44:11< zookeeper> i have no idea what advdef is.
20160922 10:44:23< matthiaskrgr> it's one of the curshers used
20160922 10:44:50< matthiaskrgr> it tries different compression (zip, zopfli) methods
20160922 10:45:30-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160922 10:45:42< matthiaskrgr> the current implementation is:    use imagemagick,optipng, advdef, and if *any* of those change colors, revert
20160922 10:45:49< zookeeper> right
20160922 10:46:17< matthiaskrgr> my approch was: if only one of them changes colors, only don't apply that one, but still the other 2 
20160922 10:46:33< zookeeper> well, that's how it surely should work
20160922 10:46:41< matthiaskrgr> yes
20160922 10:47:08< matthiaskrgr> i tried to implement it tonight but it was PITA to do in bash :P 
20160922 10:47:29< zookeeper> anyway, what sort of filesize difference threshold does it have? i mean if it compresses a 200kb file to 199kb then that should be ignored.
20160922 10:47:44< matthiaskrgr> I don't think it has any threshold :) 
20160922 10:48:00< zookeeper> well then no one should commit the results until one has been added
20160922 10:48:47< matthiaskrgr>  https://github.com/wesnoth/wesnoth/blob/master/utils/wesnoth-optipng#L122 
20160922 10:49:41< matthiaskrgr> what should the threshold be? 
20160922 10:49:59< zookeeper> well, hard to say
20160922 10:50:48< zookeeper> but what absolutely should be avoided is a situation where someone runs it and it manages to compress most images by some tiny amount
20160922 10:51:39-!- horrowind [~Icedove@2a02:810a:8380:10a8:21b:fcff:fee3:c3ff] has quit [Quit: horrowind]
20160922 10:52:25< matthiaskrgr> 5% , 10% ? 
20160922 10:52:28< zookeeper> ideally i suppose one could just look at the output and pick a threshold that leaves pointless changes out, but i doubt that's easy :p
20160922 10:52:32-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160922 10:53:49< matthiaskrgr> I guess 10 % might be a good threshold 
20160922 10:54:01< zookeeper> seems like a reasonable default at least
20160922 10:56:10< zookeeper> or something like "revert if savings are <10% _or_ <1kb"
20160922 10:56:52< zookeeper> i mean it's kind of pointless to have an optimization run which shaves off a few dozen bytes off of hundreds of files
20160922 10:57:30-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 264 seconds]
20160922 10:59:45< zookeeper> it'd be really convenient if an optimized could be wired in directly to git/github :p it'd be a man-in-the-middle changing the actual files being pushed so i'd expect it to result in a terrible mess, but in principle it'd be great
20160922 10:59:50< zookeeper> optimizer, even
20160922 11:00:20< matthiaskrgr> er
20160922 11:00:38< matthiaskrgr> you mean a man-in-the-middle optimization? :P 
20160922 11:01:05< zookeeper> yes
20160922 11:01:21< matthiaskrgr> happy checksumming xD 
20160922 11:01:26< zookeeper> yeah
20160922 11:01:30< matthiaskrgr> it would not work anyway because git relys on checksums of files 
20160922 11:01:39< zookeeper> hence the "terrible mess" :p
20160922 11:01:44< matthiaskrgr> ^^
20160922 11:08:03-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160922 11:10:24< matthiaskrgr> I remember why I stopped coding in bash 
20160922 11:18:21< Nobun> matthiaskrgr: what do you try to develop in bash, exactly?
20160922 11:18:40< matthiaskrgr> currently wesnoth-optipng
20160922 11:19:17< matthiaskrgr> well, I obtained the concepts of if and else etc by using bash :P
20160922 11:19:23< matthiaskrgr> back then ...
20160922 11:19:39< matthiaskrgr> had I only known python... 
20160922 11:20:01< Nobun> and why you don't develop that kind of utility on python instead of bash?
20160922 11:20:17< matthiaskrgr> I don't want to have to rewrite it 
20160922 11:20:19< matthiaskrgr> too lazy
20160922 11:20:20< Nobun> python has also the advantage it can be run also on windows, unlike bash 
20160922 11:20:24< Nobun> lol
20160922 11:22:42-!- Kwandulin [~Miranda@p200300760F2C717FF18C04240C29129C.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160922 11:30:25< matthiaskrgr> ok
20160922 11:31:01< matthiaskrgr> it might work now  
20160922 11:32:45-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160922 11:35:51-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160922 11:37:42-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 264 seconds]
20160922 11:43:35-!- irker227 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160922 12:02:33-!- enchi [enchilado@defocus/yummy/enchilado] has quit [Ping timeout: 276 seconds]
20160922 12:03:46-!- enchi [enchilado@defocus/yummy/enchilado] has joined #wesnoth-dev
20160922 12:15:29-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20160922 12:36:25< bumbadadabum> matthiaskrgr: I would say rewriting it to python shouldn't be too hard
20160922 12:37:06< bumbadadabum> but idk I've never really looked at wesnoth-optipng
20160922 12:37:17< matthiaskrgr> yes
20160922 12:37:22< matthiaskrgr> but I have never done multiprocessing in python
20160922 12:37:36< matthiaskrgr> the rest should be fairly easy
20160922 12:38:14< bumbadadabum> it's python so it shouldn't be too hard
20160922 12:38:19< matthiaskrgr> :>
20160922 12:52:26-!- gfgtdf [~chatzilla@x4e368599.dyn.telefonica.de] has joined #wesnoth-dev
20160922 12:53:04-!- Nobun [~nobun@5.170.112.89] has quit [Quit: Salve a tutti]
20160922 13:13:42-!- hk238 [~kvirc@unaffiliated/hk238] has joined #wesnoth-dev
20160922 13:14:23-!- Kwandulin [~Miranda@p200300760F2C717FE46A8609D1184CF4.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160922 13:50:09-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20160922 14:00:56< mattsc> gfgtdf: Are you planning to add some sort of is_visible function for Lua?  I can do the required AI changes either way, I’d just like to know whether I should wait.
20160922 14:02:45< gfgtdf> mattsc: hmm not sure´.
20160922 14:03:23< gfgtdf> mattsc: i think it doesn't make sense to wait, specially besue updating the code to use that function shodul be quite easy
20160922 14:03:39-!- hk238 [~kvirc@unaffiliated/hk238] has quit [Quit: http://www.kvirc.net/ 4.9.2 Aria]
20160922 14:04:29< mattsc> gfgtdf: Sounds good, thanks.
20160922 14:05:51< vultraz> gfgtdf: anything on the network side of mp staging?
20160922 14:06:05< gfgtdf> vultraz: hmm ot yet
20160922 14:06:17-!- boucman_work [~boucman@bob75-2-81-56-46-209.fbx.proxad.net] has quit [Ping timeout: 240 seconds]
20160922 14:10:17-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160922 14:12:00-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160922 14:12:55-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Client Quit]
20160922 14:13:57-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160922 14:22:45-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160922 14:24:47-!- boucman_work [~boucman@193.56.60.161] has joined #wesnoth-dev
20160922 14:26:28-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has quit [Quit: .']
20160922 14:31:26< EliDupree> Following my questions from the other day, does anyone know a way to invoke wesnoth from the command line on Windows such that it DOES create a window, but the window is in the background rather than the foreground?
20160922 14:32:09< EliDupree> I've been searching for ways to do that on Windows, but it's hard to pick the right search terms because there's so much noise.
20160922 14:34:52-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160922 14:34:56-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20160922 14:35:45-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160922 14:38:42-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20160922 14:41:05-!- Kwandulin [~Miranda@p200300760F2C717FE46A8609D1184CF4.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160922 14:42:09< EliDupree> I tried:
20160922 14:42:16< EliDupree> start "westest" /d "C:\Users\Eli\AppData\Local\Battle for Wesnoth 1.12.6" /b /min "C:\Users\Eli\AppData\Local\Battle for Wesnoth 1.12.6\wesnoth.exe" --config-dir Wesnoth1.12
20160922 14:42:29< EliDupree> But it still appear in the foreground despite the /min
20160922 14:42:37-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160922 14:47:19< EliDupree> LOL, I may have run afoul of this 2.5-year-old wesnoth bug: http://gna.org/bugs/?21503
20160922 14:58:03-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 276 seconds]
20160922 15:02:23-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160922 15:03:27-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160922 15:04:37-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Client Quit]
20160922 15:05:03-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160922 15:05:25-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Client Quit]
20160922 15:05:54-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160922 15:08:05< matthiaskrgr> weird
20160922 15:08:18< matthiaskrgr> in multi player campaigns, allies are not moving their units
20160922 15:08:43< matthiaskrgr> er wait, that was not an allie
20160922 15:09:03< matthiaskrgr> ok maybe it switched sides
20160922 15:09:19< matthiaskrgr> but they just recuited and camped their base for a several turns
20160922 15:09:34-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: mattsc]
20160922 15:10:15-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Client Quit]
20160922 15:14:51-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds]
20160922 15:15:15-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160922 15:15:47-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160922 15:17:45-!- Kwandulin [~Miranda@p200300760F2C717FA91D2491176A9011.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160922 15:18:38-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Client Quit]
20160922 15:19:27-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160922 15:21:02< EliDupree> Well, I arranged a way to invoke a specific wesnoth scenario with a single click in my taskbar, so my testing problem is basically solved.
20160922 15:21:12-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Remote host closed the connection]
20160922 15:23:04-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20160922 15:23:23-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160922 15:25:12-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20160922 15:25:47-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160922 15:26:22-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20160922 15:26:49-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160922 15:27:08-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20160922 15:27:21-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160922 15:27:48-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160922 15:31:03-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 240 seconds]
20160922 15:35:22-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20160922 15:39:57-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160922 15:40:29-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160922 15:41:25-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160922 15:49:40-!- mkdroid [~null@unaffiliated/matthiaskrgr] has joined #wesnoth-dev
20160922 15:54:25-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160922 15:55:56-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev
20160922 16:00:00< vultraz> hm...
20160922 16:00:12< vultraz> ok, I actually don't... understand what's going on here
20160922 16:00:25< vultraz> http://pastebin.com/rwZCn8jg
20160922 16:00:30< vultraz> (top is my rewrite)
20160922 16:01:12< vultraz> but i just realized the whole premise seems to be taking every first character after a space...
20160922 16:01:18< gfgtdf> what was the intention of th rewrite ?
20160922 16:01:28< vultraz> so something like Ageless Era would become... "E"?
20160922 16:02:10< vultraz> (it's used in the lobby for the short era name in active games)
20160922 16:02:27< vultraz> but shouldn't it be 'AE'
20160922 16:02:34< vultraz> if it'd going for an abbreviation 
20160922 16:02:52< vultraz> gfgtdf: i was trying to make its function clearer
20160922 16:04:40< tad_> Was it broken?
20160922 16:04:51< vultraz> but unless im mistaken it looks like the old code was wrong :/
20160922 16:05:18< vultraz> then again,despite the code saying otherwise I never actually see the short era name in the lobby..
20160922 16:05:38< gfgtdf> vultraz: you you give me an example of an inptu that generated an invalid result on the old function?
20160922 16:06:19< tad_> I don't see any breakage
20160922 16:07:33< tad_> It builds a string containing the first non-space, non-? character following any space.
20160922 16:07:56< vultraz> right
20160922 16:08:02< vultraz> but the use cases don't make sense
20160922 16:08:04< vultraz> like, here's one:
20160922 16:08:10< vultraz> 			era = vgettext("Unknown era: $era_id", {{"era_id", game["mp_era"].str()}});
20160922 16:08:11< vultraz> 			era_short = "?" + make_short_name(era);
20160922 16:08:38< vultraz> is the result *supposed* to be "?e" followed by the first character of the name?
20160922 16:08:40< tad_> That explains the non-? but
20160922 16:08:42< tad_> bit
20160922 16:09:13< vultraz> or
20160922 16:09:16< vultraz> 			era = era_cfg["name"].str();
20160922 16:09:18< vultraz> 			era_short = era_cfg["short_name"].str();
20160922 16:09:19< vultraz> 			if(era_short.empty()) {
20160922 16:09:21< vultraz> 				era_short = make_short_name(era);
20160922 16:09:22< vultraz> 			}
20160922 16:09:24< vultraz> if the era was Ageless Era
20160922 16:09:30< vultraz> wouldn't the short string just be "E"
20160922 16:09:33< vultraz> that seems useless :/
20160922 16:10:47< tad_> Assume era_cfg has neither name nor short_name .. era == "Ageless Era" --> era_short should be "AE" by the original function.
20160922 16:11:35< vultraz> ohhh I see now
20160922 16:11:45< vultraz> because it had has_space as true initially
20160922 16:11:46 * tad_ smiles and nodes
20160922 16:11:49< tad_> Yep
20160922 16:12:21< vultraz> so it takes character ...  1? wait, shouldn't that be 0?
20160922 16:13:07< tad_> Now THAT I can't answer. In general, I'd have to test it to be sure it has an off-by-one.
20160922 16:14:24< tad_> If long_name[] is 0-based .. yes
20160922 16:14:36< tad_> But notice the use cases adding a "?" to the start
20160922 16:14:53< vultraz> that's true
20160922 16:15:00< vultraz> but that's not in the OTHER usecase
20160922 16:15:33< tad_> THAT is probably the lameness. "Huh? Ageless Era gave me 'gE'? Fuck it, I'll prepend '?' cuase I don't see why."
20160922 16:16:09< tad_> And the other use case was never checked.
20160922 16:17:18< vultraz> I'm going to remove the ? prepend and make this 0-based
20160922 16:17:32< tad_> Exactly what I was about to suggest.
20160922 16:18:17< tad_> And do a quick check .. send before/after to cout and hit both use cases .. then yank the debug lines.
20160922 16:18:44-!- boucman_work [~boucman@193.56.60.161] has quit [Remote host closed the connection]
20160922 16:19:49< matthiaskrgr> oooh 
20160922 16:20:04< matthiaskrgr> the testsuit needs to be run from the repo root ... 
20160922 16:20:51< Ravana_> should I add short_name= somewhere?
20160922 16:22:22< vultraz> don't think it's necessary, no
20160922 16:23:20< tad_> You mean for a campaign or era? It's not required but does avoid the '?' and off-by-one vultraz just ran across and looks to be fixin'
20160922 16:23:47-!- Kwandulin [~Miranda@p200300760F2C717FA91D2491176A9011.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160922 16:25:23-!- mjs-de [~mjs-de@x5ce4a398.dyn.telefonica.de] has joined #wesnoth-dev
20160922 16:28:02-!- Bonobo [~Bonobo@2001:44b8:254:3200:94b7:4d7d:b28e:e338] has quit [Quit: Leaving]
20160922 16:28:10-!- mkdroid [~null@unaffiliated/matthiaskrgr] has quit [Quit: I'll be back!]
20160922 16:31:05< tad_> That && long_name[i] != '?' .. yank it, too. It was there incase the prepend was done twice, I'll bet
20160922 16:35:17-!- JyrkiVesterinen [~JyrkiVest@78-27-121-191.bb.dnainternet.fi] has joined #wesnoth-dev
20160922 16:40:21-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 272 seconds]
20160922 16:52:00-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160922 16:56:42< JyrkiVesterinen> zookeeper: I can't reproduce the game freezing in combat simulation (with the save file you uploaded). The attack simulation dialog appears normally.
20160922 16:57:17< zookeeper> huh. my build is like... 2 days old, would that make a difference?
20160922 16:57:37< JyrkiVesterinen> That "Delfador versur spearman" simulation achieves just barely high enough complexity score for Monte carlo simulation to kick in (5040, the threshold is 5000).
20160922 16:58:16< JyrkiVesterinen> Two days old would indeed make a difference. I committed that fix two days ago.
20160922 16:58:24< JyrkiVesterinen> https://github.com/wesnoth/wesnoth/commit/9f379b96e34236d087d1788add77698602dd78b9
20160922 16:58:37< zookeeper> ah, i see. i didn't think to mention that earlier.
20160922 16:58:39< tad_> Entrely possible
20160922 16:58:47< JyrkiVesterinen> I guess you're hitting the same memory corruption problem.
20160922 16:59:07< tad_> Easy enough to check
20160922 16:59:10< zookeeper> well, i'll re-check once i download a new one
20160922 16:59:15< zookeeper> but yeah probably just that
20160922 17:00:27< JyrkiVesterinen> The fight has the conditions that caused the corruption to happen. The spearman is slowed. Delfador is a level 5 unit: if the spearman were to kill it, it would level up. (I think killing a unit at level 3 or above instantly levels up a level 1 unit.)
20160922 17:00:59< JyrkiVesterinen> And the complexity score is over 5000, triggering Monte Carlo mode.
20160922 17:01:29< tad_> zookeeper: I'm about done with THoT except for rebalancing the final scene. I think I'll make the PR when I'm at that point. Also, there is still that DM PR .. any issues left on it for me to look at?
20160922 17:02:48< zookeeper> tad_, i dunno, i haven't really even started yet :p
20160922 17:05:53< tad_> zookeeper: np. On THoT rebalance, I think the main problem is getting bounced by so many units from all the doors on the first hallway. I think I'll put in timers to have the doors break and spread out the attacks, sort of try to push the player down the path without overwhelming him. Other than that, don't know what to do.
20160922 17:06:39-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has joined #wesnoth-dev
20160922 17:18:36-!- Nobun [~nobun@5.170.113.239] has joined #wesnoth-dev
20160922 17:21:44-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20160922 17:25:27< Nobun> hi celticminstrel :P
20160922 17:27:52< celticminstrel> 'lo
20160922 17:29:11-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds]
20160922 17:32:48-!- hk238 [~kvirc@unaffiliated/hk238] has joined #wesnoth-dev
20160922 17:36:26-!- gfgtdf [~chatzilla@x4e368599.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 48.0.2/20160823121617]]
20160922 17:37:10< vultraz> did someone commit a wesnothd change..
20160922 17:37:18-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20160922 17:37:30< vultraz> no...?
20160922 17:37:31< mordante> servus
20160922 17:37:32< vultraz> :/
20160922 17:39:38< vultraz> why am i suddenly getting .objs-release\src\server\game.o:game.cpp|| undefined reference to `client_address[abi:cxx11](std::shared_ptr<boost::asio::basic_stream_socket<boost::asio::ip::tcp, boost::asio::stream_socket_service<boost::asio::ip::tcp> > >)'| errors when nothing in the server changed... 
20160922 17:41:11-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160922 17:41:49< tad_> Dunno. I sync'd and rebuilt just before coming into IRC and it was clean for me. There's been no commit messages since I arrived.
20160922 17:42:39< vultraz> it was literally working before i restarted my laptop :|
20160922 17:43:19 * vultraz attempts a clean rebuild
20160922 17:45:41-!- hk238 [~kvirc@unaffiliated/hk238] has quit [Quit: http://www.kvirc.net/ 4.9.2 Aria]
20160922 17:47:26 * tad_ wonders why one would go to all the trouble of creating a custom unit, placing it, then changing it to another customer unit before anyone can even look at the first variation .. and THEN select a variation of the second .. and for all this nothing changes on the screen (at all).
20160922 17:50:06< vultraz> bad coding? :)
20160922 17:50:14< tad_> There must have been something cut out, or someone was planning something they never got around to doing. It's the only thing which makes sense.
20160922 17:51:08< tad_> The only indication of all this was "My lord, what happened?" sort of comment. Seemed odd. I looked. Odder than it seemed once I looked.
20160922 17:51:16< vultraz> wait
20160922 17:51:21< vultraz> are you looking at thot?
20160922 17:51:27< tad_> Yep
20160922 17:51:48< vultraz> penultimate scenario?
20160922 17:51:50< tad_> THoT S11 start sequence
20160922 17:52:13< vultraz> oh
20160922 17:52:15< tad_> S12 is the last playable. 13 is the epilog
20160922 17:52:18< vultraz> well this is awkward 
20160922 17:52:25< vultraz> I completely overhauled that scenario :|
20160922 17:53:05< vultraz> well, uh...
20160922 17:53:10< tad_> Did it start with the Rune Lord, then after a fight expose him as an Undead .. and then umaks him?
20160922 17:53:13< vultraz> IIRC changing his variation changes his portrait..
20160922 17:53:47< vultraz> ah
20160922 17:53:54< vultraz> ok, so he starts as a DRL
20160922 17:53:57< vultraz> and becomes a UDRL
20160922 17:54:00< vultraz> because
20160922 17:54:04< vultraz> "# This is a customized Dwarvish Runemaster with a chill tempest
20160922 17:54:06< vultraz> # attack like the L3 Lich, and its own animations."
20160922 17:54:09< vultraz> that's the unit type description
20160922 17:54:36< vultraz> and that type has a "karrag-unmasked" variation with a different sprite
20160922 17:55:14< tad_> I'm reading it. I'll check the portraits. Maybe there's something there. This is all before anyone can move (start event) so it seems a lot of work for no effect
20160922 17:55:37< vultraz> so it gives him his new type
20160922 17:55:39< vultraz> unmasks him
20160922 17:55:42< vultraz> says more lines
20160922 17:55:46< vultraz> then re-masks him
20160922 17:55:49< vultraz> before turn 1
20160922 17:55:51-!- Duthlet [~Duthlet@dslb-146-060-179-135.146.060.pools.vodafone-ip.de] has quit [Quit: leaving]
20160922 17:56:07< tad_> None of that has a VISIBLE effect.  I'll check the portraits. Maybe it's using them wrong.
20160922 17:56:21< vultraz> no visible effect?
20160922 17:56:23< vultraz> well that's a bug
20160922 17:57:07< vultraz> pretty sure I was mistaken, and it's the sprites that change, not the portrait
20160922 17:57:08< vultraz> s
20160922 17:57:16< zookeeper> the visible effect is in the stats and the base frame
20160922 17:57:32< tad_> The portrait stays the masked w/ glowing eyes throughout. And I don't see the map icon changing (if it should) .. and the sidebar .. if it changes it's all in the ellipsis because the text is cut off
20160922 17:57:38< zookeeper> he looks normal, then he gets the skull and lich stats, then presumably gets re-masked as to avoid having animations for maskless version
20160922 17:57:52< zookeeper> at least that's what the WML looks like
20160922 17:58:15< vultraz> what zookeeper said
20160922 17:58:17 * tad_ nods. The WML looks like a lot of work. On screen, I'm not seeing any changes.
20160922 17:58:25< vultraz> if the sprite does not visibly change on-screen it's a bug
20160922 17:58:47< vultraz> because what zookeeper described is the behavior that occurred when I coded that
20160922 17:59:24< tad_> I'll run through it again. Maybe it's too subtle and I'm missing it.
20160922 17:59:42< vultraz> look at his face
20160922 17:59:44< zookeeper> hmh, yeah the sprite doesn't seem to be changin on-map, only in the sidebar
20160922 18:00:04< tad_> His face is like 5 pixels high .. not a lot of detail to see
20160922 18:00:23< vultraz> bug bug bug!
20160922 18:01:05< celticminstrel> Needs [redraw][/redraw]? Or engine bug?
20160922 18:01:13< zookeeper> 5 pixels, pff. it's 16 high :p
20160922 18:01:14< tad_> Has redraw
20160922 18:01:35< tad_> So his face is 3 pixels .. I still dont' see any changes.
20160922 18:02:51< zookeeper> well we are discussing a bug where the sprite doesn't change, except in the sidebar :p
20160922 18:02:55< tad_> Whatever the side, it's small and either to subtle for me to see or not making it to the map. The portrait is the same throughout
20160922 18:03:11< tad_> s/side/size/
20160922 18:03:25< vultraz> not making it to the map sounds right
20160922 18:03:28< vultraz> which is a bug
20160922 18:03:40< zookeeper> yep
20160922 18:04:05< tad_> I think I'll do a couple screen grabs and diff them to be sure
20160922 18:04:24< zookeeper> btw in cases like that where you're "transforming" a unit, i'd recommend going through the usual levelup animation. which might in this case function as a workaround to the bug as well.
20160922 18:04:41< vultraz> oh yeah
20160922 18:04:58< vultraz> good idea
20160922 18:05:13< zookeeper> it works as a good visual cue that the unit is becoming or revealed to be something else
20160922 18:05:28 * zookeeper needs to add that to the sea orc too
20160922 18:05:28< tad_> The glow does draw the eye
20160922 18:05:43-!- Nobun [~nobun@5.170.113.239] has quit [Ping timeout: 255 seconds]
20160922 18:06:12-!- prkc_ [~prkc@46.166.188.236] has joined #wesnoth-dev
20160922 18:06:42< tad_> For the sea orc, moving the camera to him helped a lot. The screen changed but was lost in the pan to his message just after. scrolling before helped a lot.
20160922 18:07:02< vultraz> nooo the undefined reference thing is back :(
20160922 18:07:02< zookeeper> checking him in unit help and clicking on the base unit produces "Reference to unknown topic: 'unit_Dwarvish Rune Lord.'"
20160922 18:07:10< zookeeper> haha
20160922 18:07:24< zookeeper> (i hope you were referring to the same thing)
20160922 18:07:25< vultraz> this is exactly what happened when i tried building with boost 1.60+
20160922 18:07:31< zookeeper> oh, apparently not.
20160922 18:07:47< vultraz> (not this specific error, but other undefined reference crap)
20160922 18:07:57< vultraz> looks like I won't be programming anytime soon.
20160922 18:08:54-!- prkc [~prkc@46.166.188.213] has quit [Ping timeout: 264 seconds]
20160922 18:09:13< zookeeper> someone didn't update vampire bat's shadow when they added its standing animation
20160922 18:09:40< Soliton> vultraz: https://gcc.gnu.org/onlinedocs/libstdc++/manual/using_dual_abi.html
20160922 18:09:42< tad_> vultraz: I'm using gcc 6.2.1 and boost 1.61.0 
20160922 18:11:02< vultraz> tad_: im the only one who seems to have problems with 1.60+
20160922 18:11:30< vultraz> anyway, reading that link..
20160922 18:12:03-!- irker884 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160922 18:12:03< irker884> wesnoth: ln-zookeeper wesnoth:master a624dca7d3e7 / data/core/images/units/undead/bat-se-3.png: Fixed a shadow https://github.com/wesnoth/wesnoth/commit/a624dca7d3e7461efa8feffc30122867bcfc48a1
20160922 18:12:29< vultraz> ok.. s
20160922 18:12:31< vultraz> o
20160922 18:12:39< vultraz> i have to rebuild boost?
20160922 18:13:15< vultraz> fun fun fun
20160922 18:14:26< Soliton> or set the _GLIBCXX_USE_CXX11_ABI macro to the old abi.
20160922 18:15:40< tad_> OK. Diff of images .. I see the weapon changes and the sprite in the sidebar changes but the pixles on the map are identical.
20160922 18:16:35< tad_> So are the stats/name/etc on the sidebar. Jsut the little picture and the weapons change.
20160922 18:16:38 * celticminstrel curious what you use for image diffing. Only one I know of is ImageMagick.
20160922 18:17:22< zookeeper> tad_, exactly. so everything's working as intended, except the pixels on the map don't change which is a bug.
20160922 18:17:24< tad_> I grab a before and after with The GIMP and diff them on screen. The flick the layer on and off so the eye catches subtle changes
20160922 18:17:38< celticminstrel> Oh, so not an automated diff program?
20160922 18:17:51< zookeeper> you can also set the upper layer to the "difference" blend mode to instantly see the areas that are different
20160922 18:18:03< zookeeper> (not very convenient with animated units and terrain though...)
20160922 18:18:05< celticminstrel> I see, paste them into the same doc and show/hide.
20160922 18:18:12< tad_> zookeeper: Yep and the flicker lets me see what
20160922 18:18:14< celticminstrel> Makes sense.
20160922 18:18:49< zookeeper> but yeah the flickering method is what i always use to compare different versions of the same image
20160922 18:20:35< tad_> The eye is optimized to see change.
20160922 18:21:37< tad_> Anyway .. so can I fix it by using modify_unit and showing the levelup anim? or do we have an engine bug here?
20160922 18:21:42< zookeeper> well, certain kinds of change anyway
20160922 18:21:52-!- Nobun [~nobun@5.170.107.135] has joined #wesnoth-dev
20160922 18:21:52< zookeeper> it's an engine bug
20160922 18:22:19< zookeeper> but as i said showing the levelup anim might work as a workaround, and it should probably be used anyway
20160922 18:23:06< tad_> The white glow effect does draw the eye to something happening even if you can't see what
20160922 18:23:47< zookeeper> yeah. the sprites are (or can be) so small on the map that small changes to them easily go entirely unnoticed unless there's something like the levelup anim to draw attention to it
20160922 18:24:22< tad_> even a facing change would hlep
20160922 18:29:20-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has quit [Quit: ChipmunkV]
20160922 18:30:17-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has joined #wesnoth-dev
20160922 18:30:21< tad_> celticminstrel: I'm getting some errors on redrawing the game window after resizing it. Areas which don't paint on right if I draw right wider. Converse: areas cut off if narrowed.  Redraws correctly if I drag by titlebar.
20160922 18:30:50-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160922 18:34:52-!- wedge009 [~Thunderbi@60.241.236.92] has joined #wesnoth-dev
20160922 18:41:36-!- gfgtdf [~chatzilla@x4e368599.dyn.telefonica.de] has joined #wesnoth-dev
20160922 18:42:25< celticminstrel> I don't really know anything about that.
20160922 18:45:03-!- wedge009 [~Thunderbi@60.241.236.92] has quit [Quit: wedge009]
20160922 18:45:06< tad_> celticminstrel: Who would? Anyone?
20160922 18:45:21< celticminstrel> Maybe Aginor. Not sure.
20160922 18:49:01-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160922 18:51:33-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160922 18:57:33< tad_> For a test I stored Karrag and inspected. The transform took place and the variation selected. So I killed the old unit and unstored it and it still shows the wrong image on the map.
20160922 18:58:06< tad_> But the sidebar shows the correct image when he next speaks.
20160922 19:04:37< matthiaskrgr> ok, I managed to run the test suite under asan and ubsan 
20160922 19:05:34< matthiaskrgr> who was interested in the ASAN results again? shadowm or zookeeper  ? (I don't remember)
20160922 19:05:48< zookeeper> i have no idea what that is so probably not me :P
20160922 19:06:07< zookeeper> or, i think i recognize the acronym but...
20160922 19:06:26 * zookeeper guesses address sanitizer
20160922 19:06:29< matthiaskrgr> ok than it was shadowmaster 
20160922 19:06:29< matthiaskrgr> yes
20160922 19:06:35< matthiaskrgr> ubsan is undefined behaviour sanitizer 
20160922 19:08:08< celticminstrel> What about threads sanitizer?
20160922 19:08:20< celticminstrel> I think there's still an occasional crash in the loading screen.
20160922 19:08:31< matthiaskrgr> when starting the game?
20160922 19:08:44< matthiaskrgr> its a use after free, I opened a ticket for that 2 days ago (with stacktrace) 
20160922 19:08:50< celticminstrel> I think so? Or it might've been when starting a game.
20160922 19:09:17< celticminstrel> So it wasn't threading related?
20160922 19:09:22< tad_> Oh, I can cause all sorts of problems when first starting the engine.
20160922 19:09:29< matthiaskrgr> well, it happens when the window is resized while loading
20160922 19:09:38< matthiaskrgr> not sure if that's related to your problem
20160922 19:09:40< tad_> That's one of 'em
20160922 19:09:58< tad_> Try closing it while loading, too
20160922 19:10:04< celticminstrel> If I recall correctly... one time I caused a crash is when starting a scenario that shows a dialog in pre-prestart (though that's currentl broken) and repeatedly clicking with the intent of dismissing it tht moment it appears.
20160922 19:10:35< celticminstrel> Possibly pressing Esc rather than clicking.
20160922 19:10:44< matthiaskrgr> mmh
20160922 19:10:56< celticminstrel> ^currently
20160922 19:11:23 * matthiaskrgr tries
20160922 19:11:49< celticminstrel> Might be harder since those dialogs now just error with something like "need video object".
20160922 19:11:56< vultraz> pushing that so i have one less thing sitting around
20160922 19:11:56< irker884> wesnoth: Charles Dang wesnoth:master d1053abe7195 / src/gui/dialogs/lobby/data.cpp: Fixed an issue with the short era name in the new lobby https://github.com/wesnoth/wesnoth/commit/d1053abe7195e717ca6eaebd88a5356819ee7575
20160922 19:12:52< vultraz> so, latest version of boost is 1.61
20160922 19:13:00< vultraz> which i already tried and failed to build with :|
20160922 19:13:32< tad_> which compiler and version?
20160922 19:14:02< vultraz> tdm gcc 5.0.1
20160922 19:14:27< vultraz> basically, any time ive build boost since 1.58 it's failed on linking with undefined reference errors
20160922 19:14:38< vultraz> when using it to build wesnoth, that is
20160922 19:14:42< vultraz> boost builds fine
20160922 19:15:25< vultraz> manage to plow past some by including various internal files from boost but i can't get rid of all of them because there seem to be a million of them :|
20160922 19:15:35< vultraz> which is why ive been stuck on 1.58
20160922 19:15:47< tad_> can you upgrade to gcc 6?
20160922 19:16:31< vultraz> s/5.0.1/5.1.0
20160922 19:16:34< vultraz> and no
20160922 19:16:39< vultraz> no tdm gcc package for 6 :/
20160922 19:17:06< tad_> Well, you could build from source if you have a couple-three days :P
20160922 19:17:21< vultraz> build gcc from source :|
20160922 19:17:22< vultraz> whut
20160922 19:17:29< celticminstrel> GCC or Boost?
20160922 19:17:33< celticminstrel> I was assuming Boost.
20160922 19:17:33< tad_> gcc
20160922 19:17:37< tad_> It's painful
20160922 19:17:42< vultraz> this is not windows user prerogative! :P
20160922 19:18:08< celticminstrel> Hey, I built Boost from source on Windows. (That was awhile ago, no longer using that version.)
20160922 19:18:19< celticminstrel> (Also built it from source on Mac.)
20160922 19:18:28< vultraz> oh, building boost is simple
20160922 19:18:32< vultraz> he's saying build gcc
20160922 19:18:41< celticminstrel> Yeah, takes awhile, but it's not that hard.
20160922 19:19:06< tad_> You're using mingw?
20160922 19:19:08< celticminstrel> This reminds me, I should update my clang for Windows.
20160922 19:19:20< matthiaskrgr> building gcc should only take a night if it doesn't fail in between
20160922 19:19:22< vultraz> I have no idea :/
20160922 19:19:56< tad_> tdm looks like a mingw thing. 
20160922 19:20:12< celticminstrel> Not sure whether I shouldo download 32-bit or 64-bit...
20160922 19:20:12< tad_> Are you on Windows 10 and updated to Aniversary Edition?
20160922 19:20:15< celticminstrel> ^should
20160922 19:20:20< vultraz> I follow the instructions here https://github.com/wesnoth/wesnoth/tree/master/projectfiles/CodeBlocks
20160922 19:20:23< vultraz> and it works
20160922 19:20:27< vultraz> so id on't question the internals :P
20160922 19:20:36< vultraz> and yes, I have Win 10 w/ Anniversary Update
20160922 19:20:42< tad_> Yep. That's a MinGW build
20160922 19:21:19-!- Nobun [~nobun@5.170.107.135] has quit [Ping timeout: 255 seconds]
20160922 19:21:30< tad_> Well, you could try building using native Linux from Ubuntu. I'd have to work out the steps because you'll want to upgrade things. I got 80% through it and decided it was a waste of time
20160922 19:22:04< vultraz> e_e
20160922 19:22:06< tad_> But that will give you a native Linux binary and I don't know if that would work at all for you. Probably not.
20160922 19:23:04< tad_> That codeblock stuff is a year old. Let me see if there's someplace more current you can get gcc for mingw
20160922 19:23:22< vultraz> maybe i should just build with --disable-libstdcxx-dual-abi
20160922 19:24:23< vultraz> tad_: this any good? https://sourceforge.net/projects/mingw-w64/files/Toolchains%20targetting%20Win32/Personal%20Builds/dongsheng-daily/6.x/
20160922 19:24:49< tad_> Worth a try. And the upstream mingw project on sourceforge does have a gcc 6.2.1 prebuilt.
20160922 19:25:29< tad_> Yep. That's what I found. Don't know if you drop it in on CodeBlocks, though. Never looked at codeblocks.
20160922 19:26:09< tad_> Worst case is you have to uninstall a lot of junk and re-install. But it might get you off dead-center for a clean compile without having to hack.
20160922 19:26:22< vultraz> I am a windows user, I like things to just work :P
20160922 19:26:59< tad_> Then you need to install Linux.
20160922 19:27:09 * tad_ chuckles.
20160922 19:27:54< tad_> Actually, I'd suggest using a VM at first so you can switch to Windows. That's how I'm set up although I almost never fall back to my Windows desktop any more.
20160922 19:28:07< vultraz> nah
20160922 19:28:18< vultraz> I'm a gamer too, which is another reason to use windows
20160922 19:28:42< tad_> VM solves that.
20160922 19:28:52< tad_> So does dual booting
20160922 19:29:03< vultraz> dual booting is a PITA
20160922 19:29:20< celticminstrel> Yeah.
20160922 19:29:31< celticminstrel> I have a Fedora install on the Windows computer but rarely use it.
20160922 19:29:40< vultraz> also there's this weird thing with this laptop where i can't seem to change the uefi boot order, so couldn't install it even if i wanted to
20160922 19:29:43< tad_> Other choice is to go to a local pawn shop and pick up another cheap laptop.
20160922 19:30:00< vultraz> > cannot build wesnoth
20160922 19:30:03< vultraz> > buys new laptop
20160922 19:30:07< vultraz> :P
20160922 19:30:52-!- JyrkiVesterinen [~JyrkiVest@78-27-121-191.bb.dnainternet.fi] has quit [Quit: Going to bed]
20160922 19:31:08< tad_> Sounds like a good plan. Gotta keep your priorities straight!
20160922 19:31:54< vultraz> im thinking of getting a new one soon anyway
20160922 19:32:01< vultraz> this one has given me so much grief :|
20160922 19:33:03< tad_> So you're using codeblocks and mingw to build a version you can run on windows?
20160922 19:33:04< vultraz> got it late 2013 but it's a late 2012 model and I've had to replace the hdd, reset the os, and reinstall the os once from a disk too
20160922 19:33:14< vultraz> a version of...what
20160922 19:33:21< tad_> wesnoth
20160922 19:33:23< vultraz> of course
20160922 19:34:52-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20160922 19:34:54< tad_> I wonder if it would be possible to get it working on the free Visual Studio and forget all that mingw junk. You really shouldn't need a unix-like layer and the game is supposed to make fine as a VS project, right?
20160922 19:34:57< vultraz> ive tried using MSVC but it's so dam slooowww
20160922 19:35:05< vultraz> damn*
20160922 19:35:11< vultraz> like, I dunno why
20160922 19:35:18< vultraz> but it takes 3x as long to build
20160922 19:35:40-!- mjs-de [~mjs-de@x5ce4a398.dyn.telefonica.de] has quit [Remote host closed the connection]
20160922 19:35:44< tad_> Well, isn't there a Microsoft logo on it? Need I ask more?
20160922 19:35:51 * tad_ chuckles.
20160922 19:35:52< vultraz> :P
20160922 19:36:39< tad_> I gave up on Microsoft C when they decided one floppy wasn't enough and I needed 8 just to install it.
20160922 19:36:59< vultraz> floppy?
20160922 19:37:01< vultraz> install?
20160922 19:37:25< vultraz> eh?
20160922 19:37:58< zookeeper> vultraz, have you actually ever used a floppy?
20160922 19:38:04< zookeeper> just wondering
20160922 19:38:05< vultraz> I have, actually
20160922 19:38:06< tad_> Well, it was '86 or so and it was on an IBM PC model 1.
20160922 19:38:09< vultraz> years and years ago
20160922 19:38:15< vultraz> in the dark ages of the early 2000s
20160922 19:38:37< zookeeper> wow, so hipstery
20160922 19:39:22< zookeeper> i still have a few floppies... just in case, i guess
20160922 19:39:38< vultraz> i remember when i was really young
20160922 19:39:57< vultraz> parents has VHSs...floppies... and cassettes e_e
20160922 19:40:12< matthiaskrgr> :D
20160922 19:40:14< tad_> Cassettes are making a comeback ...
20160922 19:40:29< vultraz> and actual *film* cameras 
20160922 19:40:39< tad_> I'm sure any floppies I'd find have all de-gaused by now.
20160922 19:41:04< celticminstrel> I might have a few floppies lying around somewhere.
20160922 19:41:06< zookeeper> tad_, aren't they still good for writing though?
20160922 19:41:13< vultraz> hmm
20160922 19:41:17< celticminstrel> I think I backed them all up as virtual images on the Windows computer though.
20160922 19:41:22< vultraz> ok, so i copied all the files from my download
20160922 19:41:31< vultraz> into the tdmgcc32 folder i had
20160922 19:41:34< tad_> zookeeper: Sure. But if I have 'em it's because 20 years ago I thought it was important I save something ...
20160922 19:41:35< celticminstrel> Or at least, all the ones that actually matter.
20160922 19:41:38-!- Nobun [~nobun@5.170.106.23] has joined #wesnoth-dev
20160922 19:41:49< vultraz> which is where cb looks
20160922 19:41:54< vultraz> and it seems to be building
20160922 19:42:07< vultraz> but i notice if i go to command prompt
20160922 19:42:10< vultraz> and use gcc --version
20160922 19:42:29< zookeeper> tad_, what, you wouldn't use them to transport files anymore? pfff.
20160922 19:42:31< vultraz> I get gcc (tdm64-1) 5.1.0
20160922 19:42:34< tad_> Now, it's all on my 4T NAS drive and I chuckle because when I started programming there wasn't 4T in the entire world if you added it all up and now I have one I got for $120.
20160922 19:42:40< vultraz> im thinking maybe
20160922 19:42:42< vultraz> when i build boost
20160922 19:42:49< vultraz> it uses the 64 version
20160922 19:42:52< vultraz> which doesn't work with wesnoth
20160922 19:42:56< vultraz> which use the 32 version
20160922 19:43:58< tad_> vultraz: that would cause a lot of problems, sure.
20160922 19:44:17< vultraz> let me uninstall 54
20160922 19:44:19< vultraz> 64*
20160922 19:44:51< vultraz> ahh
20160922 19:45:08< vultraz> and now when i do gcc --version I get gcc (GCC) 6.2.1
20160922 19:45:10< vultraz> :D
20160922 19:45:11< tad_> zookeeper: I tried just setting Karrag to Undead Dwarvish Rune Lord in [side] and he still does not show. So I'm thinking the problem is the varriation, not the unit_type itself
20160922 19:45:28< vultraz> tad_: i think celticminstrel was playing with side stuff
20160922 19:45:30< vultraz> er
20160922 19:45:32< vultraz> variation
20160922 19:45:34< vultraz> stuff
20160922 19:45:48< tad_> vultraz: Sounds like you should make clean / make all and try it now
20160922 19:46:02< vultraz> tad_: still need to rebuild boost
20160922 19:46:13< vultraz> CB has been correctly using the 32 bit gcc
20160922 19:46:17< tad_> HttT uses variation and I didn't notice any problems there
20160922 19:46:21< vultraz> but my attempts to build boost may not have been
20160922 19:46:28< celticminstrel> vultraz: What was I playing with now?
20160922 19:46:30< vultraz> since obviously, 64 was in PATH before 32..
20160922 19:46:31-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20160922 19:46:41< tad_> celticminstrel: variations for unit_type
20160922 19:46:53< vultraz> so passing --tolchain=gcc to b2 might have made it use 64
20160922 19:46:58< vultraz> if it looks in PATH
20160922 19:47:29< zookeeper> tad_, celticminstrel, the difference would be that in HttT they don't use inherit=yes
20160922 19:47:32< tad_> Building 6.2 lib and calling from 5.1 mainline would probably be a problem
20160922 19:47:48< vultraz> well yes
20160922 19:47:50< tad_> zookeeper: good point
20160922 19:47:53< vultraz> i need to rebuild wesnoth too :P
20160922 19:47:56< vultraz> which im doing first
20160922 19:48:09< vultraz> THEN boost
20160922 19:48:18< vultraz> then rebuild both
20160922 19:48:20< vultraz> i guess
20160922 19:48:37< celticminstrel> vultraz: What was I playing with now?
20160922 19:48:39< vultraz> first i just want to see if my drop-in solution actually gets through the building
20160922 19:48:42< vultraz> celticminstrel: unit variations
20160922 19:48:47< celticminstrel> What about them?
20160922 19:48:56< vultraz> idk
20160922 19:48:58< vultraz> ask tad_ 
20160922 19:49:00< celticminstrel> The only thing I remember doing involving variations was Lua API stuff.
20160922 19:49:05< tad_> celticminstrel: THoT S11 start event
20160922 19:49:31< celticminstrel> What's the problem?
20160922 19:49:56< celticminstrel> Oh, image not right.
20160922 19:50:03< tad_> celticminstrel: [side]2 uses a unit. During start it transforms to a unit type then chooses a variation.
20160922 19:50:23-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 244 seconds]
20160922 19:50:46< tad_> celticminstrel: I set the [side] to use the undead type stright off and removed the transform. still shows wrong image on map for variation
20160922 19:50:54< celticminstrel> Randomly: Can't it do that transformation in one step instead of two? (Unless it specifically wants to show the intermediate stage.)
20160922 19:51:21< celticminstrel> So the first question I would have is, is the variation unit_type correct.
20160922 19:51:45< tad_> celticminstrel: The issue is there is just the one image. So it wants to go back to the original variation so the anims are there after start event completes
20160922 19:51:51< celticminstrel> I'd go into the Lua console and type in, uhh... wesnoth.unit_types["type-name"].variations["variation-name"].image or something.
20160922 19:52:16< tad_> celticminstrel: The variation image appears on the sidebar when the unit speaks. But the map never shows it.
20160922 19:52:29< celticminstrel> Oh right, forgot about that.
20160922 19:52:50< tad_> celticminstrel: and I checked with inspect and the image= is correct
20160922 19:52:52< celticminstrel> And [redraw] does not work, right?
20160922 19:53:00< celticminstrel> Yeah, inspect could show that too.
20160922 19:53:04< tad_> celticminstrel: redraw is present and no effect
20160922 19:54:02< celticminstrel> I don't know then. Sorry.
20160922 19:54:57< tad_> So, istm the next thing I can try is eliminate the inherit=yes and do it like HttT 
20160922 19:55:38< celticminstrel> I think it would be preferred not to do that, though...
20160922 19:55:51< tad_> preferred vs works ...
20160922 19:55:58< celticminstrel> Did that issue in DM with the advancements get fixed? I can't remember.
20160922 19:56:07< tad_> celticminstrel: yes it is fixed
20160922 19:56:11< celticminstrel> Ah.
20160922 19:56:22< tad_> I did a PR and it merged a week or three ago.
20160922 19:56:28< celticminstrel> Oh, you fixed it. Okay.
20160922 19:57:06-!- Nobun [~nobun@5.170.106.23] has quit [Ping timeout: 244 seconds]
20160922 19:57:25-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 260 seconds]
20160922 19:57:48< tad_> Well, I'm going to try to get it working by 'fixing' the WML. Zookeeper will ding me for using a workaround instead of fixing the error but maybe it will help locate the error.
20160922 20:00:40-!- Nobun1 [~nobun@5.170.112.61] has joined #wesnoth-dev
20160922 20:05:51< tad_> zookeeper: You are correct. I copied the entire unit in instead of inherit=yes and the map image is now updating as desired.
20160922 20:06:16< tad_> celticminstrel: So the bug is in an alternative which uses inherit=yes
20160922 20:06:29-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160922 20:06:43< celticminstrel> What was the fix for the DM problem?
20160922 20:07:01< celticminstrel> (Asking because inherit=yes and [base_unit] likely use the same code.)
20160922 20:07:43< tad_> It's been a while. I'll have to check the commit log and refresh my memory. ISTM it was another one-liner to fix it.
20160922 20:08:14< celticminstrel> I was trying to find the commit, but couldn't.
20160922 20:12:01< tad_> celticminstrel: I can't find it either.
20160922 20:12:17< tad_> celticminstrel: and now I'm not even sure what the problem was
20160922 20:13:28-!- exciton [chuck-the-@89.208.170.132] has joined #wesnoth-dev
20160922 20:21:32< tad_> celticminstrel: grep for inherit. Looks like this is the only place with inherit=yes and image=<file>
20160922 20:26:39-!- exciton [chuck-the-@89.208.170.132] has quit [Remote host closed the connection]
20160922 20:31:42-!- Nobun1 [~nobun@5.170.112.61] has quit [Ping timeout: 264 seconds]
20160922 20:31:50-!- exciton [chuck-the-@89.208.170.132] has joined #wesnoth-dev
20160922 20:34:20-!- exciton [chuck-the-@89.208.170.132] has quit [Read error: Connection reset by peer]
20160922 20:36:44-!- exciton [chuck-the-@89.208.170.132] has joined #wesnoth-dev
20160922 20:50:17-!- Nobun1 [~nobun@5.170.111.190] has joined #wesnoth-dev
20160922 21:04:20-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Quit: Page closed]
20160922 21:07:41-!- noy [~Noy@24.244.32.190] has joined #wesnoth-dev
20160922 21:07:43-!- noy [~Noy@24.244.32.190] has quit [Changing host]
20160922 21:07:43-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160922 21:14:56-!- Nobun2 [~nobun@5.170.106.236] has joined #wesnoth-dev
20160922 21:17:01-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev
20160922 21:18:30-!- Nobun1 [~nobun@5.170.111.190] has quit [Ping timeout: 260 seconds]
20160922 21:28:38< tad_> Is it possible that the [standing_anim] or [idle_anim] is causing the map to not update for the new unit_type image= ???
20160922 21:36:35< celticminstrel> Dunno. A [standing_anim] is always implicitly defined though.
20160922 21:37:10-!- Netsplit *.net <-> *.split quits: Gambit, knotwork, Rhonda, stikonas, TheJJ, higgins, miniplane, aidanhs, Greg-Boggs, Polsaker,  (+21 more, use /NETSPLIT to show all of them)
20160922 21:37:25-!- Netsplit *.net <-> *.split quits: Sirp, wedge009, esr, loonycyborg, Aginor, EliDupree, AI0867, elias, timotei, _laco,  (+25 more, use /NETSPLIT to show all of them)
20160922 21:42:24-!- Nobun [~nobun@5.170.106.236] has joined #wesnoth-dev
20160922 21:42:30-!- Netsplit over, joins: noy, AI0867, Rhonda, Ivanovic, Soliton, zookeeper, shadowm, Ravana_, shikadibot, janebot (+50 more)
20160922 21:43:01< irker884> wesnoth: Celtic Minstrel wesnoth:menus_refactor 60fc1f639cf6 / src/ (18 files in 6 dirs): WIP Refactor of menus to remove old markup https://github.com/wesnoth/wesnoth/commit/60fc1f639cf6a63174806d7a95a654df4587bab6
20160922 21:43:01< zookeeper> oh duh, yes it has a standing anim
20160922 21:43:02< zookeeper> i thought of that when i was having a shower and intended to check whether it had one, but forgot
20160922 21:43:02-!- exciton_ [chuck-the-@89.208.170.132] has joined #wesnoth-dev
20160922 21:43:03< zookeeper> tad_, so yes of course the standing anim is the culprit
20160922 21:43:12-!- Netsplit over, joins: wedge009, Aginor, timotei, Yaiyan
20160922 21:43:12< tad_> So is it a bug or should I override it in the variation?
20160922 21:43:15-!- exciton [chuck-the-@89.208.170.132] has quit [Ping timeout: 240 seconds]
20160922 21:43:26-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160922 21:43:26-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20160922 21:44:48< zookeeper> not a bug, and yes you need to override it in the variation if you want the pixels to show up on the map
20160922 21:45:07< tad_> Testing that now, then.
20160922 21:46:57< tad_> That did it. Now I just need to hold the camera on it because it's changing and scrolling at the same time. I think a narrator comment about it is warranted.
20160922 21:47:34< celticminstrel> zookeeper: Do you know of any addons that use units with multiple [base_unit]?
20160922 21:47:54< celticminstrel> Because, I just noticed that it's supported.
20160922 21:47:56< zookeeper> celticminstrel, nope. i didn't think that was even possible..?
20160922 21:48:07< celticminstrel> Yeah, I didn't think so either!
20160922 21:48:16< zookeeper> sounds like madness, really
20160922 21:48:30< celticminstrel> Yeah, units isn't the sort of place I'd want to see multiple inheritance.
20160922 21:48:53-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has quit [Quit: ChipmunkV]
20160922 21:49:14< tad_> celticminstrel: "GAK!" is the correct comment, there. If you can prevent it easily, I'd say do it.
20160922 21:50:58< tad_> OK. I'm past this issue and about ready to take a swing at THoT S12. I'll check for obvious errors and such and put up a PR when I'm done. They I'll see about doing something to make it less of a rotten tomato scene.
20160922 21:51:40< tad_> First, I need to run to market and prepare dinner.  BBL.
20160922 21:51:40< celticminstrel> Might be useful to check if any addons discovered it, but yeah, I'm leaning towards making it just take the first.
20160922 21:52:19< tad_> celticminstrel: Could error and die if more than one but I'd be surprised if anyone depends upon it. It would be a f'ing mess.
20160922 21:52:23-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Quit: Page closed]
20160922 21:53:06-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 265 seconds]
20160922 21:53:06< zookeeper> well, no need to specifically prevent it i think. maybe someone discovers something fun that can be done with it.
20160922 21:53:23-!- Nobun [~nobun@5.170.106.236] has quit [Quit: Salve a tutti]
20160922 21:54:01< celticminstrel> It kinda interferes with the prototypes idea that I keep throwing around and doing nothing about. >_>
20160922 21:58:36-!- travis-ci [~travis-ci@ec2-54-166-125-71.compute-1.amazonaws.com] has joined #wesnoth-dev
20160922 21:58:37< travis-ci> wesnoth/wesnoth#11138 (menus_refactor - 60fc1f6 : Celtic Minstrel): The build failed.
20160922 21:58:37< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/162040029
20160922 21:58:37-!- travis-ci [~travis-ci@ec2-54-166-125-71.compute-1.amazonaws.com] has left #wesnoth-dev []
20160922 21:59:17-!- Lohengramm [sid1929@gateway/web/irccloud.com/x-ufthkmbheusqckgj] has joined #wesnoth-dev
20160922 22:02:51-!- exciton_ [chuck-the-@89.208.170.132] has quit [Quit: leaving]
20160922 22:03:35-!- exciton [chuck-the-@89.208.170.132] has joined #wesnoth-dev
20160922 22:04:13-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 265 seconds]
20160922 22:04:47-!- atarocch [~atarocch@93.56.160.28] has quit [Quit: Leaving]
20160922 22:54:25-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20160922 23:24:53-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 265 seconds]
20160922 23:27:00-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160922 23:35:01-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20160922 23:39:33-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160922 23:49:47-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev
--- Log closed Fri Sep 23 00:00:33 2016