--- Log opened Wed Feb 22 00:00:38 2012
20120222 00:14:02-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Quit: Leaving.]
20120222 00:31:39-!- ancestral [~ancestral@67-6-36-251.mpls.qwest.net] has joined #wesnoth-dev
20120222 00:36:54< CIA-61> ai0867 * r53150 /trunk/src/campaign_server/campaign_server.cpp: Make email field mandatory in PBL files
20120222 00:36:59< CIA-61> ai0867 * r53151 /trunk/src/campaign_server/campaign_server.cpp: Remove the 'master password' feature from campaignd
20120222 00:37:56-!- lansiir [~oldtopman@71-208-57-231.hlrn.qwest.net] has joined #wesnoth-dev
20120222 00:38:22-!- lansiir is now known as Guest54131
20120222 00:48:07-!- Guest54131 [~oldtopman@71-208-57-231.hlrn.qwest.net] has quit [Quit: Leaving]
20120222 00:49:21-!- cjhopman [~chris@24-183-40-128.dhcp.mdsn.wi.charter.com] has joined #wesnoth-dev
20120222 00:49:21-!- cjhopman [~chris@24-183-40-128.dhcp.mdsn.wi.charter.com] has quit [Changing host]
20120222 00:49:21-!- cjhopman [~chris@wesnoth/developer/cjhopman] has joined #wesnoth-dev
20120222 00:49:48-!- wesbot changed the topic of #wesnoth-dev to: 1.10.1 planned for 25.2.; hard stringfreeze on branches/1.10 | 164 bugs, 327 feature requests, 16 patches | Logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://pastebin.com | http://imagebin.org
20120222 00:51:23-!- oldtopman_ [~oldtopman@71-208-57-231.hlrn.qwest.net] has joined #wesnoth-dev
20120222 00:53:52-!- Johannes13 [~Johannes@unaffiliated/johannes13] has quit [Ping timeout: 260 seconds]
20120222 01:01:41-!- Elvish_Pillager [~eli@dhip-149.coburn.residences.colby.edu] has joined #wesnoth-dev
20120222 01:14:08-!- oldtopman_ [~oldtopman@71-208-57-231.hlrn.qwest.net] has quit [Quit: Leaving]
20120222 01:31:34-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20120222 01:58:04-!- vultraz [~chatzilla@124.109.10.221] has quit [Ping timeout: 245 seconds]
20120222 01:58:42-!- PolarPanda [~quassel@unaffiliated/peterporty] has quit [Read error: Network is unreachable]
20120222 01:59:00< fendrin_> esr: The troll cave in NR scenario01 is a kind of tomato surprise. Can we redesign that? 
20120222 01:59:13-!- PolarPanda [~quassel@unaffiliated/peterporty] has joined #wesnoth-dev
20120222 02:00:38< esr> fendrin_: Junk it entirely if you want.  It's not important to gameplay and I tend to forget it's there.  Or if you want to try to do something interesting with it my mind is open.
20120222 02:01:06-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20120222 02:04:30< fendrin_> esr: I think it is okay to stay in. Just tell the player that there are trolls in there. The sidekick could mention it. The player must understand that they are hostile to everyone and can be used against the trolls if released in the right situation. It would be nice to have a mechanism that allows the releasing unit to escape.
20120222 02:04:44< AI0867> it's foreshadowing for the next scenario
20120222 02:05:00< fendrin_> s/trolls/orcs 
20120222 02:05:28< AI0867> "there seem to be a large number of trolls in here, still open the doors? (Y/N)"
20120222 02:05:29< esr> Yeah, I'll add a warning.
20120222 02:06:34< esr> fendrin_: Or you do it.  Point me at the file.line number afterwards and I'll piolish dialog if required.
20120222 02:08:08< fendrin_> esr: Okay, shall I change anything to the gameplay? Like what AI0867 suggested? Maybe the trolls are leaving one round later, giving the unit some time to flee.
20120222 02:08:42< esr> That works for me.
20120222 02:10:20< fendrin_> That == old or That == new ?
20120222 02:12:38-!- Alarantalara [~Adium@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has joined #wesnoth-dev
20120222 02:13:45< shadowm> Alarantalara: how goes the implementation of the other hex status frames?
20120222 02:14:40< Alarantalara> I could finish them quickly if I knew what they were supposed to represent
20120222 02:14:40< shadowm> You might want to check this: https://gna.org/bugs/?19416
20120222 02:15:00< shadowm> hm, I thought eleazar explained that in his FR
20120222 02:16:09< fendrin_> Alarantalara: hi
20120222 02:16:25< Alarantalara> hello
20120222 02:16:58< shadowm> ah, he didn't specify *anything*
20120222 02:17:08< shadowm> well, you can always PM him in the forums :p
20120222 02:17:16< fendrin_> Alarantalara: Could you please mainline the underground water terrain from the "Roar of the Wose" campaign?
20120222 02:18:20< Alarantalara> fendrin_:  http://forums.wesnoth.org/viewtopic.php?f=9&t=29798&start=15#p425864
20120222 02:18:40< Alarantalara> some of that may no longer apply, but it represents the comments I got when I first drew it
20120222 02:18:59< Alarantalara> so while I could, I'm not sure I should
20120222 02:21:15< shadowm> Ivanovic: https://gna.org/bugs/?19482 perhaps someone should usggest the Russian translation to use shorter names for Preferences strings?
20120222 02:21:33< shadowm> And by someone I clearly mean someone with the Internationalization coordinator role.
20120222 02:22:19< fendrin_> Alarantalara: Hmmm, the current version seems perfectly fine for my needs and taste. But I am not an artist.
20120222 02:22:42< shadowm> Ivanovic: okay, never mind. My people are also doing the same shit.
20120222 02:22:48< shadowm> Espreon: please fix the Spanish translation. It breaks the Preferences dialog.
20120222 02:22:51< shadowm> :p
20120222 02:25:21< shadowm> The Advanced Preferences options spam that took place in 1.9.x doesn't help matters. I *could* truncate strings but that isn't a real solution for two reasons:
20120222 02:26:06< Alarantalara> fendrin_: There's also the issue that I've been using that terrain to experiment with the idea of sprite sheets for terrain
20120222 02:26:21< Alarantalara> though it wouldn't be hard to cut it up again
20120222 02:26:22< shadowm> 1) Truncating strings in the listboxes works on the assumption that letter width is more or less always the same. That assumption works with Latin alphabet users but not necessarily with everyone else.
20120222 02:28:45< Alarantalara> shadowm: I remember a guideline published by Microsoft that recommended leaving twice the space the original language took for a translation if the original was shorter than a sentence
20120222 02:28:53< fendrin_> Alarantalara: I think the current state of things is that you are the best canditate for being the terrain director. You could just accept the title and do whatever suits you.
20120222 02:29:49< shadowm> 2) It *is* possible to truncate strings in the listboxes, but not in the checkboxes/slider labels/whatever in Advanced Preferences. At least one instance must remain non-truncated so users don't need to read the tooltip to guess exactly what the option does. However, as I see things, it's still possible for this to cause problems with ven longer strings, so then there'd be a need to ditch usability in favor of not breaking the ...
20120222 02:29:55< shadowm> ... goeomrtry.
20120222 02:30:28< shadowm> Horrible connection here at home, makes it hell to write lines longer than a couple of characters through SSH.
20120222 02:31:07< shadowm> Alarantalara: the dialog almost fills the screen width at 8000 pixels already
20120222 02:31:26< shadowm> 800
20120222 02:32:24< shadowm> it looks like extending it would help for the SPanish translation, but I'm not sure about everyone else
20120222 02:33:05< shadowm> okay, funnily enough it tries to extend itself for Russian (?!)
20120222 02:33:49< Alarantalara> shadowm: is it possible to wrap the lines for the options?
20120222 02:34:39< shadowm> It's still quite messy: http://shadowm.rewound.net/f/preferences-ru.png
20120222 02:34:51< shadowm> Alarantalara: not in GUI1, no, and Preferences is still stuck with GUI1
20120222 02:38:02< shadowm> The translation for "Multiplayer" in Russian seems like comedy material. Unfortunately UI design is more like a drama.
20120222 02:38:27-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20120222 02:38:57-!- Gambit [~gambit@wesnoth/developer/grickit] has quit [Remote host closed the connection]
20120222 02:39:37< shadowm> It's even more hilarious that they managed to avoid the problem that affects the Spanish translation regarding the autosave options in the first page. That is, despite those contain the word "multiplayer" (albeit as an acronym in English)
20120222 02:45:47< CIA-61> ai0867 * r53153 /trunk/data/core/macros/abilities.cfg: Make the wmlxgettext workaround for feeding a bit less hacky
20120222 02:45:51< CIA-61> ai0867 * r53152 /trunk/utils/wmlxgettext: Allow # wmlxgettext: comments to influence wmlxgettext's tagstack
20120222 02:49:26< fendrin_> noy: Hi, why is the Elvish Shyde not able to fly over unwalkable terrain but the drakes are able to?
20120222 02:52:31< shadowm> I believe shydes aren't able to fly too high
20120222 02:52:52< shadowm> same goes for sylphs
20120222 02:52:53< noy> right
20120222 02:53:22< fendrin_> I see.
20120222 02:53:40< AI0867> they fly "in ground effect"
20120222 02:54:39-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Quit: Computer has gone to sleep.]
20120222 02:55:02< fendrin_> Then the shyde should be able to keep her mobility in caves more or less. She is more like a helicopter than an airplane. While the drake can't start in the cave because he is more the airplane type.
20120222 02:56:11< shadowm> terrible analogy
20120222 02:57:08< shadowm> forest elves don't seem to be able to navigate caves very well in general because they never get in caves unless forced to
20120222 02:58:18< fendrin_> Agreed.
20120222 02:59:10< fendrin_> The question is just, what should the shyde have for movement when it comes to underground water.
20120222 02:59:21< fendrin_> Shallow underground water.
20120222 02:59:32< AI0867> no better than normal shallow water
20120222 02:59:37< fendrin_> The shyde is very fast there, that assumes she is still flying.
20120222 03:00:00< shadowm> forest-bound shydes and sylphs don't need to be the exception, and if you need an exception you can always give her an object altering her movement costs and defenses
20120222 03:00:47< fendrin_> exception?
20120222 03:01:13< fendrin_> The general problem is that there is no value for flying units that are forced to swim.
20120222 03:01:39< shadowm> exception to the "forest elves live in forests and work better in forests" rule above, I mean
20120222 03:02:36< fendrin_> The shyde was just an example.
20120222 03:02:45< fendrin_> I am more worried about drakes.
20120222 03:03:03< shadowm> I personally see the underground water thing to be just a pipe dream.
20120222 03:03:23< fendrin_> pipe dream=
20120222 03:03:24< fendrin_> ?
20120222 03:03:37< shadowm> There are far many other breaks from reality scattered though the game (poison doesn't kill units and only leaves them at 1 HP for example)
20120222 03:04:02< fendrin_> It is not a reality problem.
20120222 03:04:14< fendrin_> It is a problem when I make maps for campaigns.
20120222 03:04:35< shadowm> It has never been a problem for me nor anyone else.
20120222 03:04:38< fendrin_> Underground scenarios can't feature water at all if you have flying units in the scenario.
20120222 03:04:51< shadowm> And I'm pretty sure I'm not the only one who makes huge underground scenarios with water and flooded areas.
20120222 03:06:31< fendrin_> Well, I have that problem when doing the drake campaign.
20120222 03:06:39< fendrin_> Nearly every drake is able to fly.
20120222 03:07:02-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20120222 03:07:22< fendrin_> shadowm: And how do you code flooded areas? 
20120222 03:07:28< shadowm> I don't code them.
20120222 03:07:51< fendrin_> So your flying unit can flydive through the flooded area?
20120222 03:08:05< fendrin_> Like it was air.
20120222 03:08:08< shadowm> Exactly.
20120222 03:08:12< fendrin_> crazy
20120222 03:08:19< fendrin_> That needs to be fixed.
20120222 03:08:38< shadowm> Sorry, but it's not a problem for me and since I'm the map designer it's my place to decide, not yours.
20120222 03:09:07< shadowm> As for your needs, there's nothing stopping you from applying an object altering your drakes' movement costs for the duration of a scenario.
20120222 03:09:28< fendrin_> Exactly, and I do need a better underground water terrain, usable for my suits.
20120222 03:10:00< shadowm> I thought we were discussing stats, not graphics.
20120222 03:10:17< fendrin_> Yes, stats
20120222 03:10:44< fendrin_> I think every unit should feature a value for water when it is forced to swim.
20120222 03:11:06< fendrin_> And terrains that use them.
20120222 03:11:30< shadowm> That sounds like the beginning of creeping featuritis.
20120222 03:11:53< shadowm> Not that terrain classification doesn't suffer from that already. mind you.
20120222 03:11:56< fendrin_> This makes underground terrain much easier to use, no more need to think about the two different unit types.
20120222 03:13:19< fendrin_> It's not creeping featuritis. Just one more underground terrain that works more predictable.
20120222 03:13:53< fendrin_> creeping featuritis is just a buzz word
20120222 03:14:09< shadowm> It's not a buzzword, it's a coding concern.
20120222 03:15:05< fendrin_> One more terrain is a coding concern? You are kidding.
20120222 03:15:14< shadowm> No, I am not.
20120222 03:15:21< shadowm> I could never understand the motivation behind the reef terrain and I still don't. I'm _really_ opposed to introducing new terrains for unit stats.
20120222 03:15:58< shadowm> For one, it means that every add-on out there needs to add a new terrain for every movement cost or defense table that could require it.
20120222 03:16:11< shadowm> Secondly, it means more shit for players to care about.
20120222 03:16:26< shadowm> Instead of having a graphical variation of a terrain, they get a variation with completely different stats to learn.
20120222 03:16:35< shadowm> That's even worse if the terrain is rare enough.
20120222 03:17:04< fendrin_> shadowm: Yes, Coastal Reef should be a simple mixed terrain. It should not be a base terrain. I can't understand why it was introduced as well. But this is another case.
20120222 03:17:52< shadowm> As for the actual CP that there currently is, was mostly referring to the terrain aliasing hell, which got started when arbitrary overlays got introduced.
20120222 03:17:57< fendrin_> shadowm: Let's get rid of Coastal Reef and introduce real swimming instead.
20120222 03:18:17< shadowm> And then some weirdos want infinite overlays.
20120222 03:18:38< shadowm> If you still think it's just a buzzword then you have certainly never tried to hack into the terrain report code.
20120222 03:18:50< ancestral> I just want a way to have simple hexes, no transitions, no variations
20120222 03:19:33< ancestral> And it's actually more work to undo all the… hacking? that's been added on over the years
20120222 03:19:55< shadowm> ancestral, different problem, different area
20120222 03:19:55< ancestral> (Maybe it's more elegant then I'm giving credit for)
20120222 03:20:13< fendrin_> Well, isn't that a good time to sit together and design the whole terrain thing more cleaner from scratch?
20120222 03:20:20< shadowm> I have not even mentioned terrain builder issues in this conversation so far
20120222 03:20:40< Alarantalara> ancestral: you can create an add-on that overrides every single terrain definition
20120222 03:20:41< shadowm> that would be a good idea, but you'll need more than just coders
20120222 03:20:59< shadowm> that is, coders don't necessarily make for good unit balancers or good UI designers
20120222 03:21:04< ancestral> Alarantalara: I was planning on doing that, but it's not very clear what I can overwrite and what I can't
20120222 03:21:13< Alarantalara> ancestral: everything
20120222 03:21:25< Alarantalara> except things from another add-on
20120222 03:22:46< ancestral> Okay, so parts of /core/terrain-graphics/base.cfg is required?
20120222 03:23:04< shadowm> in particular, as it's been stated multiple times, balancing isn't controlled by fancy formulas, but it's based instead on accumulated gameplay experience
20120222 03:23:32< shadowm> given that, completely changing how terrain aliasing works sounds, um, _dangerous_
20120222 03:23:32< Alarantalara> Leave all the macros alone - they don't matter
20120222 03:23:55< ancestral> Alarantalara: The problem is I don't want an add-on
20120222 03:24:09< Alarantalara> still leave them alone - it will break add-ons
20120222 03:25:24< Alarantalara> all add-ons get processed before any rules from core, so unless you plan to write C++ code for custom appearance switching, it's better to do it as an add-on
20120222 03:25:54< ancestral> I'd rather create transparent pngs for tons of files first
20120222 03:26:10< ancestral> At least I could use it in single player and multiplayer games I play with other people
20120222 03:26:51< Alarantalara> An add-on will let you do that
20120222 03:26:54< fendrin_> shadowm: Well, I think adding another base terrain is not bound to c++ coding at all, is it?
20120222 03:27:07< Alarantalara> fendrin_: it's not
20120222 03:27:17< ancestral> Alarantalara: But I can't use an add-on with another player in multiplayer unless they have it too, or I host
20120222 03:27:19< shadowm> fendrin_: no, but I'd like to not see that happen without a good reason because it creates more problems for me as an add-on maintainer
20120222 03:27:26< Alarantalara> ancestral: incorrect
20120222 03:27:41< Alarantalara> since it only affects appearance, it will not cause out of sync errors
20120222 03:28:05< shadowm> and so far I haven't seen good reasons beyond "it bothers me because I keep evading the possibility of using [object]"
20120222 03:28:11< ancestral> Right, but, if my friend hosts a game, I don't get to choose the add-on
20120222 03:28:20< fendrin_> shadowm: Well, I can't see how that would affect any of your addons, since the terrains and units used there are not changed in any way. And the reason is obvious and good enough.
20120222 03:28:30< shadowm> they are changed
20120222 03:28:34< Alarantalara> terrain graphics are not an era
20120222 03:28:49< Alarantalara> they will appear regardless of what your friend picks
20120222 03:28:57< fendrin_> shadowm: no
20120222 03:29:11< Alarantalara> if you want a demonstration, go get the high contrast water add-on from the 1.10 server and join any game
20120222 03:29:13< shadowm> if you introduce a new terrain archetype, I have to go through every single movement cost/resistance table in movetypes and unit-specific overrides to see whether using a sensible default would make less sense than using a custom value
20120222 03:29:21< Alarantalara> and enjoy the different coloured water
20120222 03:29:35< ancestral> I guess I'm not understanding how it loads the add-on
20120222 03:30:00< ancestral> But sure, I'll try it
20120222 03:30:20< Alarantalara> Because terrain_graphics is a top level tag, it goes outside of era, campaign and everything else
20120222 03:30:32< shadowm> furthermore, in this particular case I will inevitably have to do that and edit every single map
20120222 03:30:46< Alarantalara> as a result, the only thing controlling its appearance are preprocessor macros
20120222 03:30:50< shadowm> now that's just AtS. Think of the countless many other add-ons that are out there
20120222 03:31:06< Alarantalara> if they are not contained inside one, then they apply to everything and there is no way to stop it
20120222 03:31:15< fendrin_> shadowm: That is silly. You do not need to change any map at any of your campaigns just because of that. 
20120222 03:31:43< shadowm> then you don't need to introduce a new terrain archetype in mainline either
20120222 03:31:52< ancestral> Alarantalara: Okay, so I would modify the terrain-graphics.cfg file
20120222 03:31:56< Alarantalara> no
20120222 03:32:22< fendrin_> shadowm: Well, I feel like it.
20120222 03:32:32< shadowm> yes, and that's not a good reason
20120222 03:32:41< fendrin_> I have good reasons.
20120222 03:32:45< fendrin_> Yours are bad.
20120222 03:33:06< shadowm> I have not seen any good reasons
20120222 03:33:07< Alarantalara> ancestral: that file will get replaced every update, an add-on will not
20120222 03:33:20< fendrin_> And I have not seen any good reason against.
20120222 03:33:26< fendrin_> I need to change my maps.
20120222 03:33:30< fendrin_> You don't
20120222 03:33:31< shadowm> why can't you just use [object], for goodness sake?
20120222 03:33:47< fendrin_> Because it is a hack.
20120222 03:34:05< fendrin_> And I don't want a hacky campaign in mainline.
20120222 03:34:31< shadowm> *you* think it is a hack
20120222 03:34:42< fendrin_> It is a hack.
20120222 03:34:45< Alarantalara> fendrin_: it's what it was designed for
20120222 03:34:49< ancestral> Alarantalara: Oh, so when Wesnoth caches the add-ons it runs the pre-processor directives?
20120222 03:35:02< ancestral> For whatever reason I assumed it only happened when that particular add-on loaded
20120222 03:35:26< Alarantalara> there is no difference between an add-on and core
20120222 03:35:38< Alarantalara> all add-ons are always loaded
20120222 03:35:47< fendrin_> Using [object] is not the hack.
20120222 03:35:47< ancestral> I never realized that
20120222 03:36:00< Alarantalara> the preprocessor uses define to control which ones are active
20120222 03:36:35< shadowm> and yes, I do need to edit my maps afterwards, because you or someone else will change the map in the rest of the mainline campaigns, mainline MP scenarios, and then every other UMC author will follow suit
20120222 03:36:40< shadowm> *maps
20120222 03:36:58< Alarantalara> if your add-on has no #ifdef, then it is always active
20120222 03:36:59< shadowm> that's so painfully obvious I'm not sure why I have to explain myself
20120222 03:37:35< Alarantalara> It's why multiplayer is so fragile - every era is loaded and can overwrite each other depending on the order they are read
20120222 03:37:36< ancestral> Alarantalara: Thanks
20120222 03:37:39< fendrin_> shadowm: So because you have no time to follow all the new trends in wesnoth, wesnoth should develop a little bit slower, should it?
20120222 03:37:44< shadowm> now, care to explain what the hack is, then, since it's apparently something neither you nor I have mentioned?
20120222 03:38:14< shadowm> I have time to follow any reasonable trends. You still have not provided any good reasons in favor of your proposed change.
20120222 03:38:27< shadowm> That is, it's intended to accomplish something that CAN ALREADY BE ACCOMPLISHED.
20120222 03:38:46< fendrin_> Oh no, another buzz word. Can already be done.
20120222 03:39:11< fendrin_> It isn't KISS either, right?
20120222 03:39:20< shadowm> So you have adopted the "change for the sake of change" stance then?
20120222 03:40:07< fendrin_> That was always my point of view.
20120222 03:40:48< fendrin_> And I have explained several times that I am not with the KISS thing at all costs as well.
20120222 03:41:12< shadowm> I think what you want is to make your own game then.
20120222 03:41:29< Alarantalara> Personally, I'd like unlimited overlays, since it would make things like illuminated mushroom groves or bridges so much simpler to use, but until I have a way to clearly describe it in WML and in the editor UI, I'm not even considering starting it
20120222 03:41:46< shadowm> Alarantalara: also player UI
20120222 03:42:05< shadowm> because that'd inevitably be used for  more than just decorative purposes
20120222 03:42:25< fendrin_> Yeah right, my own game with a single terrain type extra, rofl. The game could be named : The unliked terrain type
20120222 03:42:46< shadowm> fendrin_: I am going to insist once more that you don't need a new terrain type for what you want.
20120222 03:43:00< Alarantalara> shadowm: exactly - the things I would want to use it for are more than decorative
20120222 03:43:02< shadowm> Or in particular, not one in mainline.
20120222 03:43:04< fendrin_> Yes, it could be hacked.
20120222 03:43:14< fendrin_> But I do not want to hack it.
20120222 03:43:24< shadowm> Okay, Mr. Buzz Word, could you quit it with the "hack" thing?
20120222 03:43:30< fendrin_> That terrain type does make sense.
20120222 03:43:34< shadowm> Because I'm sure you realize LoW is in mainline and it's full of hacks.
20120222 03:43:40< fendrin_> no way
20120222 03:43:44< fendrin_> it is not hacked any longer
20120222 03:43:53< fendrin_> I have done several fine cleanups
20120222 03:43:59< fendrin_> it's pretty hack fee
20120222 03:44:16< shadowm> #ifdef MULTIPLAYER [multiplayer] #else [scenario] #endif
20120222 03:44:26< shadowm> in every single scenario
20120222 03:44:52< fendrin_> That is obsolete and already removed to some degree. Crab_ coded support for mp campaigns at the fosdem.
20120222 03:44:56< shadowm> not hacky enough for you? Then I don't see how [object] is going to be worse.
20120222 03:45:06< fendrin_> It is not [object]
20120222 03:45:07< shadowm> I don't see that support in mainline yet.
20120222 03:45:27< fendrin_> Well, that is why the multiplayer port is disabled.
20120222 03:46:24< fendrin_> No hack visible to the user.
20120222 03:46:40< fendrin_> But the [object] hack would be visible to the user.
20120222 03:46:45< fendrin_> And confuse him.
20120222 03:47:08< fendrin_> There needs to be explained what happens a lot. And people miss it because they don't read story line. 
20120222 03:47:17< fendrin_> That is hacky.
20120222 03:47:37< shadowm> The user will probably see a reasonable explanation in the scenario's dialogue and objective notes, and they will also clearly see  that movement costs change in the infobar
20120222 03:47:45< shadowm> If people can't get a clue then they shouldn't be playing Wesnoth in the first place
20120222 03:47:46< fendrin_> A extra terrain is just fine. It will be understood from most players and easy to use. No scary value changing whatever.
20120222 03:48:16< fendrin_> Good point.
20120222 03:48:30< shadowm> There are many other "hacks" breaking from standard gameplay throughout mainline and user-made campaigns.
20120222 03:48:42< fendrin_> I will need it next when someone uses the "KISS" buzz word.
20120222 03:48:45< shadowm> Let's see, recall lists disappearing, different time of day schedules
20120222 03:48:46< Alarantalara> ancestral: http://pastebin.com/DW0fQUX0
20120222 03:49:03< Alarantalara> will disable almost all transitions (there's one water one left)
20120222 03:49:20< shadowm> [terrain_mask] or [replace_map] being used to change the supposedly-static map mid-scenario
20120222 03:49:31< Alarantalara> just draw your images at layer 0 (overlay macro) and everything else should vanish too
20120222 03:49:32< shadowm> AI being given gold mid-scenario to cheat
20120222 03:49:51< shadowm> (in one of the very first scenarios of the history of the game, no less!)
20120222 03:50:16< fendrin_> Nice to hear about all that rule breaking.
20120222 03:50:18< shadowm> units being able to swim in lava (lolwhat)
20120222 03:50:29< fendrin_> It inspires me.
20120222 03:50:30< shadowm> player controlling multiple leaders
20120222 03:50:54< shadowm> player controlling multiple sides
20120222 03:51:15< shadowm> player changing the storyline without being told so
20120222 03:51:24< fendrin_> ?
20120222 03:51:41< fendrin_> Where is that?
20120222 03:51:42< shadowm> a unit with a non-mainline ability that disrupts enemy ZoCs
20120222 03:52:31< fendrin_> Well, I think I can just go with extra terrain and special units for WoV.
20120222 03:53:29< fendrin_> The fact that the drakes are able to fly in chasm is just wonderful and fits to the story.
20120222 03:53:51< fendrin_> The ancient dwarven mine is deserted after an earth quake.
20120222 03:53:53< shadowm> so far I have mentioned "hacks" found in HttT, TSG, NR, TRoW, SoF, LoW, UtBS, in no particular order
20120222 03:54:21< fendrin_> It's full of chasms because of the quake and it's full of elementals.
20120222 03:54:24< shadowm> as you can see, [object] to apply custom movement costs to a predictable set of units for the duration of one or more individual scenarios is not a huge stretch compared to any of those
20120222 03:54:46< fendrin_> The drakes can use the chasms as highways through the dungeon, that is fine.
20120222 03:55:56< fendrin_> shadowm: You are right, and I guess I will use object for that scenario in which a strom forbids all drakes to fly. There I need a completely changed movement table. But I will use extra terrain and units for the underground water issue.
20120222 03:56:27< shadowm> and the extra terrain won't be in mainline
20120222 03:56:29< shadowm> sounds excellent!
20120222 03:56:44< fendrin_> If it will be any good it might find it's way into mainline like it deserves.
20120222 03:57:08< shadowm> we don't need things in mainline only because they "deserve" to be in mainline
20120222 03:57:28< fendrin_> I think we do.
20120222 03:57:41< shadowm> under that line of thought then perhaps every single UMC out there should be in trunk too
20120222 03:58:55< shadowm> just to be clear: I don't have a problem with worthless crap being in mainline as long as it isn't forced down people's throats
20120222 03:59:59< fendrin_> shadowm: you want worthless crap in mainline? 
20120222 04:00:13< shadowm> no
20120222 04:00:17< fendrin_> My standards are higher.
20120222 04:00:27< shadowm> want vs. "not have a problem with"
20120222 04:00:35< shadowm> now most UMC are much below "crap"
20120222 04:01:23< shadowm> and since I have just realized there's some terminology ambiguity here, whenever I say mainline I am not including mainline campaigns unless I'm implicitly or explicitly referring to campaigns in general
20120222 04:01:58< fendrin_> WoV will also have a season system.
20120222 04:01:58< shadowm> so when I speak of mainline terrains I'm not including terrains defined within mainline campaigns
20120222 04:02:21< fendrin_> Winter, Summer, Spring, Fall
20120222 04:02:27< fendrin_> All with different time tables.
20120222 04:02:55< shadowm> oh great, the next UtBS
20120222 04:03:21< fendrin_> UtBS++
20120222 04:03:29< shadowm> hopefully the plot will be just as good or better
20120222 04:03:53< fendrin_> Ah no.
20120222 04:04:04< shadowm> instead of just generic shit written with flowery speech
20120222 04:04:06< fendrin_> I am more the crazy game mechanics type of designer.
20120222 04:04:18< fendrin_> generic shit is fine for me.
20120222 04:04:21< shadowm> (because apparently some people think good writing == flowery speech)
20120222 04:04:50< shadowm> okay, then it's not the next UtBS :(
20120222 04:04:56< fendrin_> no
20120222 04:05:06< fendrin_> doesn't even try to be
20120222 04:05:06< shadowm> it's the next ANL!
20120222 04:05:07-!- Ivanovic_ [~ivanovic@dtmd-4db2c4bc.pool.mediaWays.net] has joined #wesnoth-dev
20120222 04:05:15< fendrin_> maybe
20120222 04:05:17< shadowm> the author will later be hit by a truck and absolutely no-one will know exactly how to fix the crapload of bugs in it
20120222 04:05:30< shadowm> and then it'll be considered candidate for deletion once or twice
20120222 04:06:15< shadowm> if you prefer, the author will later lose interest
20120222 04:06:42< Aethaeryn> (1+ shadowm)
20120222 04:07:20< Aethaeryn> Everyone uses C instead of Lisp when they increment for some reason.
20120222 04:07:37< ancestral> I prefer Reverse Polish Notation myself
20120222 04:07:39-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 245 seconds]
20120222 04:07:57< shadowm> I HAVE BEEN DISCUSSING SHIT WITH FENDRIN FOR ONE HOUR 8 MINUTES
20120222 04:08:06< shadowm> I'd better go
20120222 04:08:18< fendrin_> lol
20120222 04:08:27< shadowm> Now I wonder why we didn't have this excellent discussion in a mailing list.
20120222 04:08:43< Aethaeryn> ancestral: (1+ foo) isn't infix, it's actually prefix with the "1+" function
20120222 04:08:49< Aethaeryn> and is valid lisp, note the no space
20120222 04:08:50< fendrin_> I will write to the ml.
20120222 04:09:03-!- Ivanovic_ is now known as Ivanovic
20120222 04:09:17< shadowm> And hopefully you will wait for a few weeks before moving forward.
20120222 04:09:19< fendrin_> Let's see what people have to say about water and caves.
20120222 04:09:37< shadowm> Mind you, I don't consider the ML to be that effective for this kind of stuff.
20120222 04:09:51< ancestral> Aethaeryn: I see
20120222 04:10:11< shadowm> I sometimes suspect I could commit all campaigns to trunk and no-one would complain if the process was done slowly enough.
20120222 04:10:15-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20120222 04:10:26< Aethaeryn> ancestral: "1+" is roughly equivalent with ++, but I think it just returns incrment and doesn't set it, I'm not sure, I should test it.
20120222 04:10:38< fendrin_> shadowm: No, effective is not what I would have called the ml list either.
20120222 04:10:57< Aethaeryn> fendrin_: Either it's too detailed and gets ignored or it's a paint the bikeshed moment.
20120222 04:11:17< Aethaeryn> 99 ways to adjust campaign difficulty for instance.
20120222 04:11:46< ancestral> Aethaeryn: I like keeping things short and sweet. I've done my best to do that on forums these days
20120222 04:11:48< fendrin_> Indeed. And I never wanted to talk about difficulty.
20120222 04:12:05< shadowm> It's not so much the fault of the mailing list medium, but of the lack of effective leads.
20120222 04:12:12< ancestral> The more succinct, often the more powerful it is
20120222 04:12:19< shadowm> I recently noticed a poor user asked a licensing question and never got a response.
20120222 04:12:19< Aethaeryn> ancestral: I like being able to edit.
20120222 04:12:26< AI0867> 04:01 < shadowm> and since I have just realized there's some terminology ambiguity here, whenever I say mainline I am not including mainline campaigns unless I'm  implicitly or explicitly referring to campaigns in general <-- 'core'
20120222 04:12:39< shadowm> AI0867: oh yes! but that recently became a buzzword :p
20120222 04:12:40< ancestral> Aethaeryn: Editing is worthless when people have already responded
20120222 04:12:42< Aethaeryn> AI0867: thanks
20120222 04:12:49< shadowm> so I think I'm not supposed to use it
20120222 04:12:55< Aethaeryn> AI0867: that's the directory structure, right? data/core?
20120222 04:13:05< Aethaeryn> ancestral: Not true.
20120222 04:13:12< Aethaeryn> ancestral: Just use strikethrough instead of removing it
20120222 04:13:19< ancestral> Ugh. Annoying
20120222 04:13:35< Aethaeryn> Trust me, there are plenty of legit reasons to edit
20120222 04:13:37< ancestral> People aren't going to edit their own replies if they don't have to. It's too much hassle.
20120222 04:13:38< Aethaeryn> Plenty. Even typos.
20120222 04:13:44< ancestral> Oh sure, I don't disagree
20120222 04:13:46< fendrin_> Okay, I am going to change the wml syntax for the scenario maps soon.
20120222 04:13:48< ancestral> I'm just saying people don't care
20120222 04:13:55< Aethaeryn> they don't have to edit their stuff.
20120222 04:13:59< ancestral> (when you correct it)
20120222 04:14:04< Aethaeryn> But why leave something imperfect when you can fix it?
20120222 04:14:14< shadowm> fendrin_: elaborate please (in particular considering maps don't have WML)
20120222 04:14:16< ancestral> Sure, but other people don't care if your shit is broken
20120222 04:14:30< fendrin_> Meaning that map_data= is no longer the prefered way t o specify a scenario.
20120222 04:14:31< Aethaeryn> what is with everyone's dislike of capitalizing proper nouns these days, anyway?
20120222 04:14:44< shadowm> fendrin_: s/scenario/map/ yes?
20120222 04:14:48< Aethaeryn> (Everyone who spoke's nick, pretty much)
20120222 04:14:59-!- Aethaeryn is now known as aeth
20120222 04:15:02< fendrin_> shadowm: yes
20120222 04:15:12< shadowm> also, I had an idea that might be somewhat orthogonal to that, so I need to know further
20120222 04:15:37< shadowm> (replace_map.map being changed to replace_map.map_data)
20120222 04:15:48< shadowm> aeth: shadowm isn't a proper noun
20120222 04:15:50< fendrin_> [map] usage= border_size= data= [/map]
20120222 04:15:59< shadowm> it's my Unix username, which has to begin with a lowercase character
20120222 04:16:49< shadowm> fendrin_: uh, any particular reasons for that? I have never understood why usage and border_size exist, even
20120222 04:16:54< aeth> ugh
20120222 04:17:02< aeth> Damn you, case sensitivity!
20120222 04:17:06< shadowm> (yay, more use for #ifver...)
20120222 04:18:02< fendrin_> shadowm: Yes, that allows me to parse the map file with the editor.
20120222 04:18:22< Alarantalara> border_size has an effect, usage seems only a way to confirm what border_size says
20120222 04:18:53< fendrin_> And makes the syntax more consistent.
20120222 04:19:13< fendrin_> Having attributes inside an wml attribute isn't exactly very common.
20120222 04:19:17< shadowm> more generally, I've always thought that maps and terrain masks are defined by context rather than by themselves, with usage and border_size being useless syntactic sugar and .map and .mask being hints for front-ends
20120222 04:19:18< aeth> shadowm: My unix name tends to be "michael" on the machines I set up so no way can I get that on any IRC net
20120222 04:19:44< shadowm> *and the .map and .mask filename extensions
20120222 04:19:54< shadowm> fendrin_: the problem is that maps are not WML in the first place
20120222 04:20:00< shadowm> they were never intended to 
20120222 04:20:18< fendrin_> a map is just a string
20120222 04:20:33< fendrin_> an argument to a wml attribute
20120222 04:20:45< fendrin_> map_data
20120222 04:20:53< shadowm> so what do you want to do? replace maps with WML that can be directly included in a WML document, or remove all metadata from maps?
20120222 04:21:03< shadowm> and move it to the parent WML document?
20120222 04:21:42< fendrin_> maps can already be directly included in a wml document.
20120222 04:21:50< shadowm> no, they cannot
20120222 04:22:02< fendrin_> since when?
20120222 04:22:09< shadowm> the hex map itself breaks the WML parser
20120222 04:22:15-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20120222 04:22:16< shadowm> only the metadata is valid WML
20120222 04:23:13< fendrin_> Well, it is nevertheless working for the test scenario. 
20120222 04:23:23< shadowm> okay, when I said "directly included in a WML document" I meant in a non-string-literal context
20120222 04:23:26-!- PolarPanda is now known as ppanda
20120222 04:23:34< aeth> ppanda: hah
20120222 04:23:38< shadowm> thatis, {FOO} vs. attribute="{FOO}"
20120222 04:23:56< ppanda> aeth: Hmmm?
20120222 04:24:17< fendrin_> {FOO}
20120222 04:24:30< fendrin_> While FOO being the filename of the map.
20120222 04:24:53-!- ancestral is now known as ancstrl
20120222 04:25:12< aeth> ppanda, ancstrl, etc.: Now we all have web 2.0 ready names.
20120222 04:25:22< ppanda> ...
20120222 04:25:28< ancstrl> aeth: Wesnoth 2.0 ready names
20120222 04:25:55< aeth> westh 2.0
20120222 04:25:59< ppanda> Talking about that, I had this awesome idea; we should make Wesnoth 3D and RT and a shooter
20120222 04:26:15< aeth> westhther
20120222 04:26:30< ancstrl> ppanda: that's not 2.0
20120222 04:26:37< ancstrl> It's has to be social networking
20120222 04:26:47< ppanda> Ah...
20120222 04:26:50< aeth> ppanda: 2.0 is to make it a facebook app and remove 90% of the features but make it notify your friends every time you level a unit up
20120222 04:26:51< ancstrl> Like a fb game
20120222 04:26:59< ancstrl> Also, in app purchases
20120222 04:26:59< aeth> And the more friends you have, the more levels you can unlock 
20120222 04:27:02< ppanda> Sounds good
20120222 04:27:24< ancstrl> Gotta save these ideas for 2 months from now
20120222 04:27:26< ppanda> You can pay to unlock level 3 units too, right?
20120222 04:27:35< ppanda> Buying WesPoints
20120222 04:27:37< aeth> ppanda: leveling is strictly on a friends basis
20120222 04:27:41< aeth> ancstrl: One turn a day to play unless you pay 
20120222 04:27:46< aeth> So you can move multiple turns a day if oyu pay
20120222 04:27:58< ppanda> Ah
20120222 04:27:59< ancstrl> Nah, people won't play it then
20120222 04:28:08< aeth> ancstrl: well, most facebook apps are like that ;-)
20120222 04:28:12< ancstrl> Buy gold with money
20120222 04:28:19< aeth> Buy everything with money
20120222 04:28:23< aeth> $100 for a fire dragon 
20120222 04:28:25< ppanda> aeth: You can play several matches at the same time
20120222 04:28:26< ppanda> like... 3
20120222 04:28:34< ppanda> but if you pay, you can play up to 10
20120222 04:28:44-!- Blueblaze [~Blueblaze@adsl-76-202-23-45.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20120222 04:28:47< aeth> Also, you can pay to have the random number generator come out in your favor
20120222 04:28:50< ppanda> Blueblaze: <3
20120222 04:28:54< shadowm> fendrin_: yeah, we can't do that at the moment...
20120222 04:29:00< ppanda> aeth: fun
20120222 04:29:04< aeth> Blueblaze: quick set your nick to bblaze
20120222 04:29:15< fendrin_> shadowm: What can't we do?
20120222 04:29:16< shadowm> So the question stilll stands. Which is going to store the metadata: the map, or the parent WML document?
20120222 04:29:35< shadowm> including a map in a WML document not as part of an attribute's value
20120222 04:29:52-!- Gambit [~gambit@wesnoth/developer/grickit] has joined #wesnoth-dev
20120222 04:29:54< ppanda> Good night, people.
20120222 04:29:55< ppanda> <3
20120222 04:30:01-!- ppanda [~quassel@unaffiliated/peterporty] has quit [Remote host closed the connection]
20120222 04:30:02< fendrin_> The map file will feature the [map] tag and all attributes
20120222 04:30:21< shadowm> furthermore: is there going o be a compatibility path? because #ifver won't work in this case and it'll break with 1.10
20120222 04:30:38< shadowm> presumably not when preprocessing add-ons, but when loading/saving maps in the 1.10 editor
20120222 04:31:35< fendrin_> 1.10 editor?
20120222 04:32:25-!- Blueblaze [~Blueblaze@adsl-76-202-23-45.dsl.hstntx.sbcglobal.net] has quit [Client Quit]
20120222 04:33:10< fendrin_> The game will  be able to read the old format, Ivanovic meant it would take 3 releases from deprecated to no longer supported. The editor or game will always write in the new format.
20120222 04:39:44< shadowm> so I will be able to feed map.map_data the contents of a map created with the 1.10 editor?
20120222 04:39:56< shadowm> *without* receiving a noisy warning?
20120222 04:40:33< fendrin_> How would you do that?
20120222 04:40:35< shadowm> at a minimum, I don't want my users seeing the warning in the game UI; it can go to stderr if absolutely necessary
20120222 04:40:58< shadowm> why are you asking me if I'm not the one changing syntax here? :|
20120222 04:41:52< shadowm> I imagine that if map.map_data contains usage and border_size metadata and the map tag doesn't, then that metadata will be used instead
20120222 04:42:01< fendrin_> shadowm: Ivanovic said the policy would be to display that the syntax is deprecated. I can't change our policies.
20120222 04:42:11< shadowm> goddammit
20120222 04:42:28< shadowm> I begin to think everyone's set on making #ifver useless
20120222 04:42:48< fendrin_> ifver is useless in this case
20120222 04:42:58< shadowm> exactly, which is terrible
20120222 04:43:14< shadowm> I had implemented it hoping that it'd be *very* useful for maintaining dual compatibility with the same assets
20120222 04:43:38< shadowm> I had not expeced *language* changes to take place in 1.11.x
20120222 04:43:57< fendrin_> Well, [map] data= is the new syntax
20120222 04:44:10< shadowm> in particular, not for the map format, which is not WML and therefore can't even use #ifver because of the editor caveat
20120222 04:44:11< fendrin_> not [map] map_data=
20120222 04:44:34< shadowm> okay, let me think
20120222 04:44:43< shadowm> the warning will be shown when using scenario.map_data, yes?
20120222 04:44:50< fendrin_> right
20120222 04:45:05< shadowm> what will happen if I use scenario.map.data *but* the attribute value contains a 1.10 map with metadata?
20120222 04:45:12-!- ancstrl [~ancestral@67-6-36-251.mpls.qwest.net] has quit [Quit: i go sleeps kthxbai]
20120222 04:45:31< fendrin_> shadowm: It will format drive c:
20120222 04:46:08< shadowm> I think you've spent too much time with Gambit
20120222 04:46:11< shadowm> can we get back on track here?
20120222 04:46:20< fendrin_> That does not work right now.
20120222 04:46:30< fendrin_> And I do not know if I want to support it.
20120222 04:46:57< shadowm> I would *really* love to see it supported assuming it won't cause a flood of in-game warnings
20120222 04:47:07< shadowm> because otherwise my plans for 1.10/1.12 dual compatibility with #ifver are pretty much broken for everything that isn't an MP era
20120222 04:47:17< fendrin_> I think we should go without the deprecated warning.
20120222 04:50:38< shadowm> What I'd ideally use: https://pastebin.com/HGKXEwkm
20120222 04:51:30< shadowm> of course, it would be trivial to come up with a tool to maintain two sets of maps using one version's set as the source to generate the other set, but ideally I would not resort to that
20120222 04:53:22< shadowm> and I am indeed talking of the situation from my position as a UMC maintainer, not as a mainline backend developer
20120222 04:54:00< shadowm> (that said, I wonder whether mordante is aware of this?)
20120222 04:54:04< fendrin_> I think, the best solution is just to keep both syntaxes valid for 1.12
20120222 04:54:25< fendrin_> shadowm: mordante was fine with the proposed change.
20120222 04:54:46< shadowm> great! now why didn't anyone think of this in 1.3.x? :p
20120222 04:55:29< fendrin_> I think I have an idea why the syntax is.
20120222 04:55:35< fendrin_> like it is
20120222 04:56:35< shadowm> I imagine this means I don't need to bother with my plan of renaming replace_map.map (attr) to replace_map.map_data (attr) and writing a wmllint rule.
20120222 04:56:46< fendrin_> shadowm: But that will only delay the problem. Are you fine with dropping the support slowly during the first 1.13 dev releases?
20120222 04:57:18< shadowm> Presumably by then it'd be fine to drop it entirely in 1.13.0 .
20120222 04:57:28< fendrin_> Cool.
20120222 04:57:58< shadowm> I didn't intend #ifver to be used for oldstable/stable compatibility, but for stable/nextstable (a.k.a. stable/dev)
20120222 04:58:34< shadowm> by then 1.12 users will have converted their maps to the new format to avoid any deprecation warnings
20120222 05:00:16< shadowm> then again, maybe stable/nextstable is indeed a pipe dream, but since I have never seen people planning *before* nextstable during a beta or RC stage and map and WML syntax have stayed unchanged since late 1.3.x (2007), I didn't have much to work with at that point
20120222 05:01:03< shadowm> I'm just hoping nobody decides 1.13.x should use json instead of WML
20120222 05:01:40< fendrin_> The transformation of the map files can be done by the editor.
20120222 05:01:51< fendrin_> The scenarios need to be changed with wmllint.
20120222 05:02:30< fendrin_> wmllint could also handle the map file transformations.
20120222 05:02:54< fendrin_> I wonder if we should extract the starting positions from the terrain string as well.
20120222 05:03:42-!- ancestral [~ancestral@67-6-36-251.mpls.qwest.net] has joined #wesnoth-dev
20120222 05:03:52< Alarantalara> I think I would be in favour of that: it makes it a lot easier to place sides with numbers > 9
20120222 05:03:58< shadowm> I don't think that'd bring any benefits, but it might be a good idea to remove some of the constraints in the hexmap format; for starters, the need for - what Alarantalara said
20120222 05:04:22< shadowm> also the need for every tile string to be of a fixed length
20120222 05:04:40< Alarantalara> on second thought: how would the map editor place sides if they're not part of the file?
20120222 05:05:07< shadowm> exactly my concern, hence I said something about not bringing any benefits :p
20120222 05:05:29< fendrin_> Alarantalara: I think the starting positions would be wml tags inside of [map]
20120222 05:05:48< fendrin_> Easy to change by human.
20120222 05:05:56< fendrin_> with editor
20120222 05:06:07< shadowm> that's something you didn't clarify above: what will the editor produce? the string with the value of map.data, or the whole map node and its children?
20120222 05:06:24< fendrin_> [starting_position] x=12, y=23, player=3
20120222 05:06:55< fendrin_> The editor will produce the whole map node with its children.
20120222 05:07:21< Alarantalara> shadowm: the tile strings don't have to be a fixed length already, the editor just saves it that way
20120222 05:07:36< shadowm> Oh? 
20120222 05:07:49< Alarantalara> strip all spaces from any map, it will load fine
20120222 05:07:56< Alarantalara> only newlines are significant
20120222 05:07:59< shadowm> So that means the single-digit side numbers limitation became purely artificial in late 1.3.x.
20120222 05:08:24< shadowm> (Limitiation which is enforced by a fixed-size C array in the gamemap or a related class/struct)
20120222 05:09:22< shadowm> the map editor does waste a lot of storage space with blanks that way, too
20120222 05:09:33< fendrin_> Why is it a limitation?
20120222 05:10:03< Alarantalara> I think it was because people were still editing maps by hand in 1.3 and it's much easier to edit something that has a grid appearance
20120222 05:10:11< shadowm> fendrin_: Because maps are currently not allowed to have starting locations for side numbers > 10
20120222 05:10:31< fendrin_> Well, let's keep the grid.
20120222 05:10:39< shadowm> Alarantalara: I remember editing maps in a text editor in 0.9.x :p
20120222 05:10:40< fendrin_> And take the starting positions out.
20120222 05:11:08< shadowm> that was quite messy because you can't see how transitions play together, not to mention the odd/even column issue
20120222 05:11:57< Alarantalara> the current format could be described something like ([1-9] [a-zA-Z/\|]+^[a-zA-Z/\|]^,)*
20120222 05:12:12< Alarantalara> with several typoes
20120222 05:12:22< shadowm> in fact, we could probably try to do away with all the side number limitations in this cycle. As far that only involves dealing with the gamemap implementation details and wesnothd.
20120222 05:12:32< shadowm> *As far as I remember
20120222 05:13:20< fendrin_> Well, what speaks against the wml solution?
20120222 05:13:32< shadowm> Or any other code that was written using C arrays of size 9.
20120222 05:13:44< fendrin_> Let's not touch the terrain syntax itself, that will summon mordante.
20120222 05:13:56< shadowm> I haven't seen anything against the map.starting_position so far
20120222 05:14:10< fendrin_> Just keeping the starting positions outside will make the syntax of terrain easier, less complicated.
20120222 05:14:52< shadowm> and then we can remove the side number limitations, which is what I'm most interested in
20120222 05:15:51< fendrin_> But you now that this limitation is only for the editor presets? You can define more sides with wml [side] definitions.
20120222 05:16:23< fendrin_> s/now/know
20120222 05:16:48< shadowm> as I said, it's found both in the gamemap implementation details (there *is* a C array of size 9 somewhere, I have seen it before) and IIRC wesnothd, presumably for validation or synchronization purposes
20120222 05:17:20< shadowm> I know you can have more than 9 sides in WML, I have used that since 1.3.xin IftU :p
20120222 05:17:34< Alarantalara> if you want to get rid of all the trailing spaces generated by the editor: http://pastebin.com/zG1fHkp9
20120222 05:17:49< shadowm> the caveat being that those sides couldn't be given starting locations in the map
20120222 05:18:12< shadowm> so they had to be either leaderless, or the leader had to be placed manually with WML
20120222 05:19:55< shadowm> there it is, gamemap::startingPositions_ is a C array of size MAX_PLAYERS+1, and MAX_PLAYERS is 9
20120222 05:20:22< shadowm> the problem is why it is a C array in the first place instead of a variable-size container
20120222 05:21:02< Alarantalara> it's a map<int, coordinate> in read_game_map
20120222 05:21:58< fendrin_> shadowm: We will not touch it. Have the [starting_position] syntax being in place for more than 9 leaders. In early 1.13 it can get removed.
20120222 05:22:38< Alarantalara> and gets converted in gamemap, so the map format supports it but they get dropped by the game
20120222 05:23:01< shadowm> amazing
20120222 05:23:44< shadowm> server/game.cpp (in wesnothd) also refers several times to the gamemap::MAX_PLAYERS constant
20120222 05:25:57< shadowm> more generally: https://pastebin.com/egQtkDjA
20120222 05:26:46< Alarantalara> you know, that's low enough that it looks possible to just remove it entirely
20120222 05:27:05< shadowm> interestingly, I had heard someone (presumably a GSoC student) had wanted to remove this constraint during 1.9.x
20120222 05:27:20< Alarantalara> the constraint does not exist in the map format at all right now
20120222 05:27:39< shadowm> the problem is that most (all?) of the uses of the constant work with fixed-size C arrays, or STL containers expected to be fixed-size since construction
20120222 05:29:01< shadowm> so infinite (or "(size_t)(-1)") won't be feasible under that scheme unless the code in question gets further, potentially intrusive changes.
20120222 05:29:15< shadowm> it'd be easier to raise the limit to a definite value above 9 instead
20120222 05:32:01< Alarantalara> Actually, the real limit seems to be more the set of available colours
20120222 05:32:57< ancestral> Alarantalara: Could introduce secondary colors :-P
20120222 05:34:20< Alarantalara> and recolour every sprite in the game?!?
20120222 05:37:53< ancestral> Or just focus on the ellipses
20120222 05:38:19< Alarantalara> and the minimap
20120222 05:38:26< ancestral> Probably wouldn't be very effective
20120222 05:39:04< ancestral> You could still have a lot of colors, and just repeat them if necessary (if you have enough sides you probably don't care if you repeat them)
20120222 05:39:20< ancestral> Like over 50
20120222 05:45:11< aeth> 255 ^ 3 colors and drop the names and stick to 000000 to ffffff hexadecimal
20120222 05:45:57< aeth> 256
20120222 05:46:42< ancestral> aeth: Human eyes can barely distinguish each varying shade of blue
20120222 05:47:06< Alarantalara> not to mention that units use colour ramps, not single colours
20120222 05:47:13< aeth> so (2^8)^3 or 2^24 max sides
20120222 05:47:20< aeth> ;-)
20120222 05:47:39< aeth> Alarantalara: aww :-(
20120222 05:48:05< ancestral> Realistically, after about 20 you run into trouble
20120222 05:48:08< aeth> I want a 2^24 cap though :-P
20120222 05:48:20< ancestral> Actually
20120222 05:48:31< ancestral> Being able to define your own damn colors for a side is an attractive FR
20120222 05:48:37< aeth> Do C arrays even get that large?
20120222 05:48:42< ancestral> (potential FR)
20120222 05:53:07< shadowm> aeth: depending on the case and the target data type's storage size, it's constrained by available stack or heap
20120222 05:53:59< shadowm> a function-local array of size (size_t)(-1) should cause the process to run of stack space immediately, I think
20120222 05:55:36< Alarantalara> ancestral: feature request already present (except in multiplayer set-up)
20120222 05:57:16< shadowm> and I'm assuming the array type size is 1 byte
20120222 05:58:55< shadowm> (size_t)(-1) is 4294967295 for me. That's approximately 4 gigabytes
20120222 05:58:58< Alarantalara> I think this has been posted before, but this is what 99 teams looks like in the editor: http://imagebin.org/200139
20120222 05:59:39< ancestral> Being able to see the names of the sides would be beneficial
20120222 05:59:54< ancestral> Or, in this example, find better names :-P
20120222 06:00:06< Alarantalara> Map data doesn't know anything about the rest of the scenario, though
20120222 06:00:11< shadowm> Alarantalara: obviously that would call for a different UI
20120222 06:00:30< shadowm> for example, a number spin box
20120222 06:01:01< ancestral> shadowm: Still would be handy to know which is the rebels and which is the empire
20120222 06:01:02< Alarantalara> Actually, it wouldn't be difficult to replace the number with a text string
20120222 06:01:48< ancestral> Or which is the rebel sympathizers and the empire-backed bounty hunters and the…
20120222 06:02:11< Alarantalara> as long as it had no spaces or commas [a-zA-Z0-9_]* then it could replace the current side number with id and the map format wouldn't visible change
20120222 06:02:58< Alarantalara> since it would just increase the allowed range for sides
20120222 06:03:04< Alarantalara> then just type the side name you want
20120222 06:03:18< Alarantalara> and reference it in the scenario
20120222 06:03:54< shadowm> I think that approach overlaps with the Scenario Editor (as oppposed to Map Editor) dilemma
20120222 06:04:16< ancestral> Eh, perhaps
20120222 06:05:37< Alarantalara> I would be nice for bare maps in multiplayer - you could guess which sides should be allied
20120222 06:05:46< Alarantalara> even without a .cfg
20120222 06:06:45< shadowm> I really wish at times I had a scenario editor, but then I realize I have so much custom code, the editor wouldn't be very helpful
20120222 06:07:16< Alarantalara> I just considered it because the map loader is using a map anyway, so the source data really doesn't matter
20120222 06:07:35< shadowm> unless the editor itself was scriptable in some way so I could tell it "hey, this is how I define a hive spawn source in WML, use this when I ask you to place one in the UI"
20120222 06:08:36< shadowm> oh man, that would be so awesome
20120222 06:08:43-!- Elvish_Pillager [~eli@dhip-149.coburn.residences.colby.edu] has quit [Ping timeout: 244 seconds]
20120222 06:08:54< shadowm> and it'd take at least four years to complete
20120222 06:09:12< Alarantalara> You can sort of do that now by defining an invisible overlay
20120222 06:09:19< Alarantalara> it becomes placable in map
20120222 06:09:35< Alarantalara> and then your code just has to call store locations at scenario start
20120222 06:09:37< shadowm> yes, yes, but I can't define location-specific parameters unless I created as many overlays as I needed
20120222 06:09:49< Alarantalara> sadly yes
20120222 06:10:17< shadowm> in AtS in particular, that approach would require at least three dozens of overlays for the hive spawn locations
20120222 06:10:45< shadowm> and I'd be unable to tell them apart in the editor anyway since we don't currently have a way to define terrain graphics layout rules for the editor
20120222 06:10:55< shadowm> oh, actually, we can, but we don't have generic macros for that yet
20120222 06:11:01< Alarantalara> #ifdef EDITOR
20120222 06:11:13-!- ancestral [~ancestral@67-6-36-251.mpls.qwest.net] has quit [Read error: Connection reset by peer]
20120222 06:11:19< shadowm> nor a way to toggle that while in the editor in order to see through editor-specific overlays
20120222 06:11:46-!- ancestral [~ancestral@67-6-36-251.mpls.qwest.net] has joined #wesnoth-dev
20120222 06:12:22< shadowm> we don't have a way to toggle individual add-ons for the editor atm either, which is why I tend to avoid having IftU or TSL installed
20120222 06:12:26-!- ancestral [~ancestral@67-6-36-251.mpls.qwest.net] has quit [Remote host closed the connection]
20120222 06:12:38-!- ancestral [~ancestral@67-6-36-251.mpls.qwest.net] has joined #wesnoth-dev
20120222 06:12:57< shadowm> in case their graphics layout code could conflict with AtS' in some way
20120222 06:13:32< Alarantalara> we don't have a way to toggle individual add-ons anywhere
20120222 06:13:46< shadowm> Alarantalara: can we currently assign any kind of "tags" to terrain_graphics rules?
20120222 06:13:57< Alarantalara> we depend on people playing nice with the preprocessor
20120222 06:14:02< Alarantalara> what do you mean by tags?
20120222 06:14:24< shadowm> I'm thinking something like only_editor=yes, although that would only save the preprocessor rerun pass
20120222 06:15:00< shadowm> there'd still be a need to rebuild transitions when toggling editor-specific overlays
20120222 06:15:31< Alarantalara> within the rules themselves, no, just the global EDITOR define and conditional inclusion
20120222 06:16:02< shadowm> then again, that *is* an important optimization, because it takes longer to preprocess all WML than rebuilding transitions
20120222 06:17:32< shadowm> perhaps I could work on that. I often find myself accidentally removing overlays like the Impassable Overlay when doing cosmetic changes, and a visual aid beyond the noisy terrain strings display would be really helpful
20120222 06:18:51< shadowm> in fact, it might not even be necessary to change terrain_graphics if we keep it simple; a single overlay defined in terrain_type itself should suffice in most cases
20120222 06:20:37< Alarantalara> That would be bad separation of appearance and function, though
20120222 06:20:45< Alarantalara> hmm
20120222 06:21:54< shadowm> then again, I'm really supposed to be refactoring the add-ons client mess :p
20120222 06:24:24< shadowm> and oddly enough, creating new UI 
20120222 06:26:01-!- Upth [~ogmar@108-85-91-228.lightspeed.frokca.sbcglobal.net] has joined #wesnoth-dev
20120222 06:26:02-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20120222 06:28:40-!- Upthorn [~ogmar@108-85-91-228.lightspeed.frokca.sbcglobal.net] has quit [Ping timeout: 255 seconds]
20120222 06:30:35< Alarantalara> This isn't possible to toggle on and off, but is editor only: http://imagebin.org/200143
20120222 06:33:10< Alarantalara> If it looks useful enough, I can clean it up into a single macro and apply it to the rest of the invisible overlay terrains
20120222 06:33:58-!- oldtopman [~oldtopman@unaffiliated/oldtopman] has quit [Quit: oldtopman has left the house]
20120222 06:34:03< shadowm> hmm, seems good to me. Does it use the ~O() function?
20120222 06:34:59< Alarantalara> yes
20120222 06:36:54< Alarantalara> anyway, goodnight
20120222 06:37:04-!- Alarantalara [~Adium@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has quit [Quit: Leaving.]
20120222 06:41:11-!- Gambit [~gambit@wesnoth/developer/grickit] has quit [Remote host closed the connection]
20120222 06:43:13-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20120222 06:48:47-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20120222 06:58:26-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20120222 06:59:51-!- ancestral [~ancestral@67-6-36-251.mpls.qwest.net] has quit [Quit: And that’s the end of THAT chapter.]
20120222 07:01:50-!- Blueblaze [~Blueblaze@adsl-76-202-23-45.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20120222 07:24:36-!- Blueblaze [~Blueblaze@adsl-76-202-23-45.dsl.hstntx.sbcglobal.net] has quit [Quit: Blueblaze]
20120222 07:52:57-!- Gallaecio [~quassel@84.120.113.193.dyn.user.ono.com] has joined #wesnoth-dev
20120222 08:01:53-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20120222 08:07:54-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20120222 08:14:56-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: SMILE, SMILE, SMILE!]
20120222 08:24:47-!- vultraz [~chatzilla@124.109.10.221] has joined #wesnoth-dev
20120222 08:30:29-!- aeth is now known as Aethaeryn
20120222 08:39:14-!- Upthorn [~ogmar@108-85-91-228.lightspeed.frokca.sbcglobal.net] has joined #wesnoth-dev
20120222 08:41:01-!- Upth [~ogmar@108-85-91-228.lightspeed.frokca.sbcglobal.net] has quit [Ping timeout: 255 seconds]
20120222 08:44:31-!- Upth [~ogmar@108-85-91-228.lightspeed.frokca.sbcglobal.net] has joined #wesnoth-dev
20120222 08:46:20-!- Gallaecio [~quassel@84.120.113.193.dyn.user.ono.com] has quit [Read error: Connection reset by peer]
20120222 08:47:46-!- Upthorn [~ogmar@108-85-91-228.lightspeed.frokca.sbcglobal.net] has quit [Ping timeout: 255 seconds]
20120222 08:52:42-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20120222 08:52:56-!- Upthorn [~ogmar@108-85-91-228.lightspeed.frokca.sbcglobal.net] has joined #wesnoth-dev
20120222 08:53:10-!- Upth [~ogmar@108-85-91-228.lightspeed.frokca.sbcglobal.net] has quit [Ping timeout: 255 seconds]
20120222 09:06:02-!- Cookiee [~quassel@unaffiliated/cookiee] has joined #wesnoth-dev
20120222 09:31:13-!- loonybot [~loonybot@46.138.85.111] has joined #wesnoth-dev
20120222 09:31:13-!- loonybot [~loonybot@46.138.85.111] has quit [Changing host]
20120222 09:31:13-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20120222 09:36:59-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20120222 09:42:02-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Quit: Leaving]
20120222 09:46:24-!- Ivanovic [~ivanovic@dtmd-4db2c4bc.pool.mediaWays.net] has quit [Quit: Disconnecting from stoned server.]
20120222 09:46:40-!- Ivanovic [~ivanovic@dtmd-4db2c4bc.pool.mediaWays.net] has joined #wesnoth-dev
20120222 09:54:27-!- Ivanovic [~ivanovic@dtmd-4db2c4bc.pool.mediaWays.net] has quit [Changing host]
20120222 09:54:27-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20120222 09:55:02< Ivanovic> moin
20120222 10:01:41-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20120222 10:04:12< Ivanovic> shadowm: done
20120222 10:05:17< Ivanovic> sent a mail to the i18n list to raise awareness of the issues with the (advanced) preferences
20120222 10:17:33-!- horon [~horon@nttkyo220107.tkyo.nt.ngn2.ppp.infoweb.ne.jp] has joined #wesnoth-dev
20120222 10:43:36-!- enchilado [~enchilado@defocus/yummy/enchilado] has quit [Ping timeout: 260 seconds]
20120222 10:47:33-!- mjs-de [~mjs-de@d184052.adsl.hansenet.de] has joined #wesnoth-dev
20120222 10:50:49-!- Haldric [~delfador@AClermont-Ferrand-552-1-76-227.w109-208.abo.wanadoo.fr] has joined #wesnoth-dev
20120222 10:57:25-!- Haldric [~delfador@AClermont-Ferrand-552-1-76-227.w109-208.abo.wanadoo.fr] has quit [Read error: Connection reset by peer]
20120222 12:37:18-!- worduk [~jon@77.76.205.90] has joined #wesnoth-dev
20120222 12:37:39-!- Cookiee [~quassel@unaffiliated/cookiee] has quit [Ping timeout: 245 seconds]
20120222 13:13:46-!- oldtopman [~oldtopman@unaffiliated/oldtopman] has joined #wesnoth-dev
20120222 13:14:14< AI0867> there's still some talk of "queen mother" in HttT:21
20120222 13:17:06< zookeeper> what's the problem with that?
20120222 13:17:25< AI0867> that such a title belongs to someone who is the mother of the queen
20120222 13:17:33-!- vultraz [~chatzilla@124.109.10.221] has quit [Ping timeout: 244 seconds]
20120222 13:17:36< AI0867> which asheviere isn't until after her death
20120222 13:19:06< AI0867> Queen regnant is what she is, although this seems to be against tradition
20120222 13:20:24< AI0867> a queen consort typically is typically not an heir to that position
20120222 13:21:08< zookeeper> so it seems
20120222 13:21:11< AI0867> so, she *was* a queen consort, then became a queen dowager who seized the power of a queen regnant
20120222 13:21:54< AI0867> queen mother only applies if one of the queen dowager's children becomes the ruler
20120222 13:22:34< zookeeper> feel free to change it to queen regnant
20120222 13:22:59< AI0867> well, if she were that, there would not be a problem
20120222 13:23:15< AI0867> li'sar says "she is *only* the queen mother"
20120222 13:23:22< AI0867> which should be "she is *only* the queen dowager"
20120222 13:23:34< zookeeper> oh, right, two references.
20120222 13:24:03< AI0867> I'll finish merging some dutch translation updates and then change it
20120222 13:25:16-!- stikonas [~and@bcm-131-111-216-70.girton.cam.ac.uk] has joined #wesnoth-dev
20120222 13:25:16-!- stikonas [~and@bcm-131-111-216-70.girton.cam.ac.uk] has quit [Changing host]
20120222 13:25:16-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20120222 13:26:17< zookeeper> well, the thing is that surely almost no one knows what a queen dowager means in practise
20120222 13:28:08< zookeeper> so it'd be better to avoid such terminology altogether, if possible
20120222 13:28:19< zookeeper> but doesn't hurt to fix it as you suggest in the meantime, of course
20120222 13:33:25< AI0867> a number of references to "queen mother" were removed in the past
20120222 13:33:43< AI0867> the only reason these still exist is because they must refer to something *less* than a queen regnant
20120222 13:33:58< AI0867> and a queen mother is simply a special case of a queen dowager
20120222 13:34:42< AI0867> esr: do you have an opinion on this?
20120222 13:43:00< CIA-61> ai0867 * r53155 /trunk/data/campaigns/Heir_To_The_Throne/ (scenarios/21_Elven_Council.cfg units/Dark_Queen.cfg): Remove the last uses of 'queen mother', replacing some of them with the correct 'queen dowager'
20120222 13:43:11< CIA-61> ai0867 * r53154 /trunk/ (5 files in 4 dirs): Dutch translation update
20120222 13:43:30-!- PolarPanda [~quassel@unaffiliated/peterporty] has joined #wesnoth-dev
20120222 13:53:52< CIA-61> ai0867 * r53156 /branches/1.10/ (5 files in 4 dirs): Dutch translation update
20120222 14:19:05-!- Elvish_Pillager [~eli@dhip-149.coburn.residences.colby.edu] has joined #wesnoth-dev
20120222 14:52:54-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20120222 14:54:26-!- loonycyborg [loonycybor@wesnoth/developer/loonycyborg] has quit [Ping timeout: 260 seconds]
20120222 14:57:39-!- loonycyborg [loonycybor@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20120222 15:00:37-!- horon [~horon@nttkyo220107.tkyo.nt.ngn2.ppp.infoweb.ne.jp] has quit [Quit: Leaving...]
20120222 15:20:24-!- Gambit [~gambit@wesnoth/developer/grickit] has joined #wesnoth-dev
20120222 15:21:26-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20120222 15:26:07-!- Johannes13 [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20120222 15:28:13-!- negusnyul [~negusnyul@dsl4E5CCB0B.pool.t-online.hu] has joined #wesnoth-dev
20120222 15:44:31-!- sabayonuser2 [~sabayonus@77.76.205.90] has joined #wesnoth-dev
20120222 15:45:25-!- sabayonuser2 [~sabayonus@77.76.205.90] has quit [Client Quit]
20120222 15:45:46-!- worduk_ [~jon@77.76.205.90] has joined #wesnoth-dev
20120222 15:48:29-!- vultraz [~chatzilla@124.109.10.221] has joined #wesnoth-dev
20120222 15:55:57-!- negusnyul [~negusnyul@dsl4E5CCB0B.pool.t-online.hu] has left #wesnoth-dev ["Konversation terminated!"]
20120222 16:03:04-!- fendrin_ [~fabi@88-134-10-236-dynip.superkabel.de] has quit [Remote host closed the connection]
20120222 16:07:39-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20120222 16:34:46-!- s951 [~s951@dyn1064-176.hor.ic.ac.uk] has joined #wesnoth-dev
20120222 16:36:04< shadowm> Espreon: I dropped you a pull request addressing the Preferences issues
20120222 16:58:44< AI0867> 20120222 16:58:20 error config: config::error thrown during transaction with add-on server; "Unterminated quoted string, value '
20120222 16:58:48< AI0867> ' at <unknown>:8"
20120222 16:58:54< AI0867> trying to upload an add-on
20120222 16:58:57< AI0867> any ideas?
20120222 16:59:39< shadowm> 1.10 or trunk? what's the contents of the .pbl besides the passphrase?
20120222 17:01:22< Espreon> shadowm: Thanks. I'll take a look tonight.
20120222 17:02:33< AI0867> shadowm: ah, that might be it, trunk isn't up yet
20120222 17:02:57-!- Upthorn [~ogmar@108-85-91-228.lightspeed.frokca.sbcglobal.net] has quit [Ping timeout: 255 seconds]
20120222 17:03:19< AI0867> also, the python CampaignClient hasn't been updated
20120222 17:03:48< shadowm> AI0867: I was assuming you were using the in-game client. You should probably try both and report to elias if the external client is at fault.
20120222 17:03:55< AI0867> shadowm: I am
20120222 17:03:58< shadowm> Trunk clients currently connect to the 1.10 campaignd
20120222 17:04:01< AI0867> but using the trunk client
20120222 17:05:38< AI0867> the python client doesn't know about 1.10 yet
20120222 17:05:43< shadowm> 1.9.x
20120222 17:05:53< shadowm> I haven't tested uploading add-ons from trunk to 1.10, but there's been no changes in the 1.10 campaignd that haven't been also commmitted to trunk and I did test several times with a local campaignd uploading small garbage and the larger AtS, so it sounds a little surprising that it'd break while I wasn't looking.
20120222 17:06:04< shadowm> Furthermore, the live instance is outdated.
20120222 17:06:24< shadowm> But then again, there have been no protocol changes yet.
20120222 17:06:27< AI0867> so, 15002 or 15004?
20120222 17:06:46< shadowm> the python client can take a version string like 1.9.x, no?
20120222 17:06:48< AI0867> I could successfully look at 1.6 at 15003
20120222 17:07:06< AI0867> shadowm: yes, I'm updating it now, but I'm not sure with which vaue
20120222 17:07:13< AI0867> note: I was using the ingame client
20120222 17:07:19< shadowm> the value for 1.9.x should be the same for 1.10
20120222 17:07:22< AI0867> s/vaue/value
20120222 17:07:31< shadowm> that is, since the instance for 1.9.x was started for 1.9.0, the port number has never changed
20120222 17:07:31< AI0867> okay, so 15004 is not running?
20120222 17:07:46-!- Jozrael [~croselius@209.133.52.231] has joined #wesnoth-dev
20120222 17:08:01< shadowm> I'm trying to locate the server-side configuration atm.
20120222 17:08:20< shadowm> 15002 is 1.9.x
20120222 17:09:19< shadowm> 1.8.x: 15001, 1.6.x: 15003, and I can't read the configuration with 1.4.x with my current access level.
20120222 17:09:28< shadowm> s/with/for/
20120222 17:09:33< AI0867> and there is no trunk server atm?
20120222 17:09:57< shadowm> hm, actually, there is a campaignd-trunk process running
20120222 17:10:21< AI0867> heh
20120222 17:10:32< AI0867> I'll go poke random ports
20120222 17:10:35< AI0867> see what happens
20120222 17:11:08< shadowm> well, I'll check with netstat, although I can't see pid information with my account
20120222 17:11:28< AI0867> okay I can get a list of everything just fine
20120222 17:11:34< AI0867> I just can't upload
20120222 17:12:03< shadowm> we listen for crap on 14998 through 15005
20120222 17:12:23< shadowm> everything below 15001 is wesnothd IIRC
20120222 17:12:57-!- Crendgrim [~crend@77-22-112-160-dynip.superkabel.de] has joined #wesnoth-dev
20120222 17:14:59< shadowm> I'm not quite sure what 15004 is looking at the add-ons list.
20120222 17:15:46< shadowm> the response from 15005 makes the wesnoth client die due to a Boost.iostreams error, so it's most likely 1.4.x
20120222 17:19:42< elias> the python client port list was up-to-date a few days ago when i compared to the c++ sources
20120222 17:20:39< shadowm> The new add-ons server should have an identification command ;)
20120222 17:22:50-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20120222 17:23:00< AI0867> shadowm: it was an error on my end
20120222 17:23:23< shadowm> What kind of error?
20120222 17:23:25< AI0867> could you reset the TDH passphrase? Apparently it's already on the server 1.10 connects to by default, but I don't have it on this computer
20120222 17:23:36< AI0867> failing to fill out the PBL properly -_-'
20120222 17:25:03< shadowm> AI0867: the Python client supports asking the server to change the passphrase. You can PM me the new passphrase if you want.
20120222 17:25:11< shadowm> The alternative is dropping the add-on along with its stats.
20120222 17:26:41-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20120222 17:26:56< shadowm> AI0867: okay, it should be changed now
20120222 17:29:26< AI0867> thanks
20120222 17:42:12-!- faryshta [~faryshta@201.137.81.186] has joined #wesnoth-dev
20120222 17:48:09-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20120222 17:49:32-!- worduk [~jon@77.76.205.90] has quit [Remote host closed the connection]
20120222 17:49:32-!- worduk_ [~jon@77.76.205.90] has quit [Remote host closed the connection]
20120222 17:57:45-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has joined #wesnoth-dev
20120222 18:12:28< shadowm> amazing how Boost.foreach can turn a single mistake into a flood of identical errors
20120222 18:13:08< shadowm> a short line also appears to turn into a > 3000 characters long line in the translation chain
20120222 18:13:15-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Quit: Leaving]
20120222 18:16:56-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20120222 18:18:43-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Quit: Computer has gone to sleep.]
20120222 18:18:53< loonycyborg> shadowm: Preprocessor and template magic tends to cause such effects.
20120222 18:20:45< loonycyborg> boost.spirit and boost.python are the best generators of huge errors IME :P
20120222 18:22:13-!- Gambit [~gambit@wesnoth/developer/grickit] has quit [Remote host closed the connection]
20120222 18:22:40-!- faryshta [~faryshta@201.137.81.186] has quit [Quit: Saliendo]
20120222 18:23:56< CIA-61> shadowmaster * r53157 /trunk/src/ (6 files in 3 dirs): (log message trimmed)
20120222 18:23:56< CIA-61> Greatly extend addon_info struct to store more information about add-ons and teach it to read campaignd's WML responses
20120222 18:23:56< CIA-61> This will allow me later to get rid of some vectors that are passed
20120222 18:23:56< CIA-61> around a lot between the add-on management UI functions.
20120222 18:23:56< CIA-61> Additionally, the static format_file_size() function was moved from
20120222 18:23:57< CIA-61> addon/manager.cpp to addon/info.* because addon_info no longer stores
20120222 18:23:58< CIA-61> the add-on size display string and a GUI2 codepath needs it. I'm not
20120222 18:25:38< shadowm> oh yes, you non-scons non-cmake people will need to add a new object code target. I forgot to mention that.
20120222 18:30:09-!- Gallaecio [~quassel@84.120.113.193.dyn.user.ono.com] has joined #wesnoth-dev
20120222 18:35:59< shadowm> loonycyborg: would it be feasible to make network_asio throw errors with a translatable message?
20120222 18:36:24< shadowm> for you, that is :p
20120222 18:36:59< loonycyborg> iirc I'm not giving much useful info here.
20120222 18:37:08< loonycyborg> In the message that is.
20120222 18:37:21< shadowm> I see errors with the add-ons client are always shown as as "Remote host disconnected" in the frontend because it's a translatable alternative coded in it, but it's not at all accurate when trying to connect to say, 255.255.255.255.
20120222 18:37:37< shadowm> the actual error is shown in stderr instead:
20120222 18:37:38< shadowm> 20120222 14:33:14 error network: network_asio::error thrown during transaction with add-on server; "Network is unreachable"
20120222 18:38:08< shadowm> (that comes through the network_asio::error::what() method.)
20120222 18:38:08< loonycyborg> Yeah. Those errors are from system, that's why they can't be translated within wesnoth itself.
20120222 18:39:01< shadowm> if I had translations installed on my system and Wesnoth was using the correct locale, would I get the translated messages?
20120222 18:39:24< loonycyborg> Probably. This needs more experimentation.
20120222 18:39:26< shadowm> or does the system always report these in the C locale language?
20120222 18:40:17< loonycyborg> iirc they're affected by C locale but no idea how it will play with boost system and how it will behaveon different platforms.
20120222 18:40:55-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has joined #wesnoth-dev
20120222 18:42:20< loonycyborg> Moreover, they'll probably be in system encoding, not in utf8 on windows.
20120222 18:43:26< shadowm> hm, I wasn't counting on that problem. Stupid legacy systems. :p
20120222 18:44:49-!- Gallaecio [~quassel@84.120.113.193.dyn.user.ono.com] has quit [Remote host closed the connection]
20120222 18:46:57-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20120222 18:48:20-!- Gallaecio [~quassel@84.120.113.193.dyn.user.ono.com] has joined #wesnoth-dev
20120222 18:58:33-!- Gambit [~gambit@wesnoth/developer/grickit] has joined #wesnoth-dev
20120222 18:59:43-!- fendrin [~fabi@88-134-10-236-dynip.superkabel.de] has joined #wesnoth-dev
20120222 18:59:43-!- fendrin [~fabi@88-134-10-236-dynip.superkabel.de] has quit [Changing host]
20120222 18:59:43-!- fendrin [~fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20120222 19:00:27< fendrin> hi
20120222 19:01:17-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20120222 19:01:26< fendrin> hi boucman
20120222 19:01:33< boucman> hey
20120222 19:02:56< fendrin> boucman: You left so abrupt from the FOSDEM that I didn't had the chance to congratulate you to your speech. It was well done. 
20120222 19:03:07< boucman> thx :)
20120222 19:03:11< fendrin> :-)
20120222 19:03:33< boucman> yes, I came to FOSDEM with other people that were here mainly for things on saturday and were waiting for me for the car to get back to paris
20120222 19:03:37< boucman> so I had to run 
20120222 19:04:42< fendrin> boucman: We spoke about having a wesnoth coding event sometime during the summer, maybe where I live, at my university.
20120222 19:04:51< shadowm> is there any practical difference between size_t foo = size_t(bar) and size_t foo = static_cast<size_t>(bar) ?
20120222 19:04:53-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20120222 19:04:54< boucman> interesting...
20120222 19:05:20< shadowm> ("practical" here meaning "beyond pretending a POD type has a constructor")
20120222 19:07:37-!- stikonas [~gentoo@bcm-131-111-216-70.girton.cam.ac.uk] has joined #wesnoth-dev
20120222 19:07:37-!- stikonas [~gentoo@bcm-131-111-216-70.girton.cam.ac.uk] has quit [Changing host]
20120222 19:07:37-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20120222 19:22:24< shadowm> mordante, loonycyborg: is it safe to use Boost.bimap given our current requirement of Boost 1.36 or later?
20120222 19:23:19< vultraz> wesbot: seen Exasperation
20120222 19:23:20< wesbot> vultraz: The person with the nick Exasperation last spoke 29d 14h ago.  28d 18h ago they were seen quitting on the channel #wesnoth with the message: Client Quit
20120222 19:23:26< vultraz> :S
20120222 19:24:55-!- anonymissimus [~chatzilla@HSI-KBW-078-042-163-105.hsi3.kabel-badenwuerttemberg.de] has joined #wesnoth-dev
20120222 19:25:35< anonymissimus> shadowm: "new object code target" ?
20120222 19:25:45< loonycyborg> shadowm: It was added in 1.35
20120222 19:26:21< loonycyborg> I'm using it in mywesnothd branch
20120222 19:26:42< shadowm> anonymissimus: src/addon/info.o
20120222 19:27:00< shadowm> or src/addon/info.cpp if you prefer to call it that way
20120222 19:28:24< loonycyborg> shadowm: bimap was made as part of gsoc project btw
20120222 19:28:46< loonycyborg> It's mostly a framework over boost.multi_index
20120222 19:29:04< anonymissimus> well if that only means adding a file file to fix linker errors; people almost never say anything anyway ;)
20120222 19:30:03< shadowm> I prefer to see things from the point of view of the toolchain rather than its users
20120222 19:30:49< shadowm> users see a bunch of obscure error messages, but the poor object linker sees references to code it cannot find :p
20120222 19:31:01< loonycyborg> (and since bimap was part of a gsoc project it probably didn't get much changes or maintance for that matter since it was added :P)
20120222 19:31:29< shadowm> well, I guess. THe documentation appears to have not changed since 2007
20120222 19:32:13< vultraz> can someone test this, as it seems Exasperation has gone missing? http://pastebin.com/raw.php?i=rnEnLWxX
20120222 19:32:29< shadowm> as pretty much every other Boost library, though, there's a lot of terminology that makes my head spin around
20120222 19:33:37< vultraz> I know it compiles, but I can't actually test it as none of my builds work 
20120222 19:34:48< vultraz> (it's supposed to let get_widget_value read from labels and images)
20120222 19:38:14-!- Danceman [~Danceman@a94-132-158-220.cpe.netcabo.pt] has joined #wesnoth-dev
20120222 19:38:15< vultraz> Crendgrim: do you build wesnoth?
20120222 19:38:49< Crendgrim> vultraz: yes, why?
20120222 19:39:39< vultraz> could you test that, as you're playing around with your questlog gui? ^
20120222 19:40:07< Crendgrim> if I was playing with the questlog, I could.
20120222 19:40:10< Crendgrim> what is "that", btw?
20120222 19:40:14< Crendgrim> ah, I see.
20120222 19:40:53< Crendgrim> vultraz: to be applied to latest trunk?
20120222 19:41:28< Crendgrim> (to 1.11?)
20120222 19:41:34< vultraz> yeah
20120222 19:46:16< Crendgrim> what the.. openSUSE does not come with patch pre-installed?!
20120222 19:50:20< vultraz> woah o_O
20120222 19:50:48< Crendgrim> okay, and I see I'm too stupid to use patch.
20120222 19:51:52< Crendgrim> no, I'm not, I just can't read.
20120222 19:53:34< vultraz> isn't it just patch -OPTIONS < PATCHFILE?
20120222 19:53:39< Crendgrim> yes
20120222 19:53:47< Crendgrim> I mis-read and thought he did something wrong.
20120222 19:55:12< vultraz> ;)
20120222 19:58:00-!- Danceman [~Danceman@a94-132-158-220.cpe.netcabo.pt] has quit [Remote host closed the connection]
20120222 20:01:06-!- Danceman [~Danceman@a94-132-158-220.cpe.netcabo.pt] has joined #wesnoth-dev
20120222 20:07:52-!- lansiirserver is now known as oldtopmanserver
20120222 20:11:22-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20120222 20:12:38-!- Elvish_Pillager [~eli@dhip-149.coburn.residences.colby.edu] has quit [Ping timeout: 240 seconds]
20120222 20:34:07-!- mordante [~mordante@roadie.xs4all.nl] has joined #wesnoth-dev
20120222 20:34:07-!- mordante [~mordante@roadie.xs4all.nl] has quit [Changing host]
20120222 20:34:07-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20120222 20:34:22< mordante> servus
20120222 20:35:19< shadowm> served
20120222 20:36:23< anonymissimus> I relieve you from your state of slavery. You are now a free civilian.
20120222 20:36:23< Ivanovic> mordante: i saw that you uploaded some more work for the float alternative
20120222 20:36:23< Ivanovic> anything for me to test?
20120222 20:36:43< shadowm> slavery? I had turned mordante into stew
20120222 20:37:27< mordante> Ivanovic, not yet, working on more optimizations and numeric stability
20120222 20:38:08< mordante> in my ugly hacked version the mainscreen looks better, not yet acceptable but better
20120222 20:38:12< anonymissimus> yeah; and I into a Roman slave, since that's what servus is
20120222 20:38:35< Ivanovic> moriturus te salutat!
20120222 20:39:32< mordante> :-)
20120222 20:39:39< anonymissimus> vultraz: well, submit your code as a patch at gna if you wanna get it in
20120222 20:39:41< mordante> anonymissimus, I assume you know it's a German greeting as well
20120222 20:40:04< anonymissimus> yes, one I avoid to use due to the above
20120222 20:40:05< vultraz> Crendgrim: did it work?
20120222 20:40:35< mordante> anonymissimus, are you from the B or W part of BW?
20120222 20:40:58< Ivanovic> mordante: it ain't german
20120222 20:41:05< Ivanovic> it is bavarian, i beg to differ!
20120222 20:41:06< anonymissimus> the W one; but that doesn't matter
20120222 20:41:33< anonymissimus> some people say servus in RL as well yes
20120222 20:41:56< anonymissimus> vultraz: but you should be able to test it of course
20120222 20:42:32< mordante> Ivanovic, W is not Bavaria, I hear it in W quite a lot
20120222 20:42:43< vultraz> anonymissimus: my builds crash at start, remember?
20120222 20:42:44< anonymissimus> "servus" counts as a "Lehnwort" I think; similar like Mauer or Acker, Roman words as well
20120222 20:43:01< Ivanovic> mordante: those are not true people from wuertemberg
20120222 20:43:09< Ivanovic> those are bavarian invaders!
20120222 20:43:36< Ivanovic> the correct greeting for würtemberg is "grüß gottle"
20120222 20:43:52< anonymissimus> don't compare BW with Bavaria pls :|
20120222 20:43:52< mordante> guess then it might be time to drive back those invaders before they overtake the rest of Germany
20120222 20:44:54< anonymissimus> well, Bavaria should just get independent :P
20120222 20:45:08< mordante> I never heart the grüß Gottle, only grüß Gott
20120222 20:45:18< mordante> aren't they already ;-)
20120222 20:45:32< Ivanovic> mordante: than you never heard those with a real swabian dialect
20120222 20:45:42< Ivanovic> there the 'le' at the end is significant and important
20120222 20:46:11< mordante> I know about the 'le' part, just never heard it with grüß Gott
20120222 20:46:57< mordante> and I heard some rather some real Swabian dialect, but can't make too much of it
20120222 20:48:17< mordante> anonymissimus, can't find the picture but at the FOSDEM there was a poster of the open pandora and it stated "Made in Bavaria"
20120222 20:48:25< mordante> so not made in Germany
20120222 20:48:46< vultraz> Crendgrim?
20120222 20:49:19< mordante> Ivanovic, the floating point emulation is rather fun to do, forces me to look at compiler output and it's nice to see how smart it is with optimizations
20120222 20:49:54< Ivanovic> do you also want to have some arm compiler output to read?
20120222 20:50:15< Ivanovic> you could just install a pandora toolchain for yourself (easy to do) and build those binaries
20120222 20:50:19< mordante> maybe later, but I assume gcc will do equally smart things there
20120222 20:50:30< mordante> I considered that as well
20120222 20:50:33< Ivanovic> it is not as well optimised on arm as it is on x86
20120222 20:50:49< Ivanovic> http://boards.openpandora.org/index.php?/topic/7147-crosscompiler-toolchain-based-on-openpandoraorg-ipks/
20120222 20:51:03< Ivanovic> all info required to install such a (complete) toolchain is available in this thread
20120222 20:51:44< mordante> int r = 9; if(foo) { r = 0; } else if (bar) { r -= 4 ...
20120222 20:52:07< mordante> the initial assignment isn't done and the else if just assigns 5
20120222 20:52:31< mordante> so that part at least is platform independent
20120222 20:52:42< Ivanovic> yeah, that is some neat stuff
20120222 20:53:01< Ivanovic> so are you looking at some immediate representation or at the generated asm?
20120222 20:53:10< mordante> and several other constant optimizations
20120222 20:53:12< Ivanovic> s/immediate/intermediate
20120222 20:53:31< mordante> looked at the asm, my x86 asm is a bit rusty but good enough to follow simple code
20120222 20:53:58< Ivanovic> arm asm should be by far easier to read and follow
20120222 20:54:24< Ivanovic> and since the optimisation does explicitly target arm (since there is no floatingpoint unit there) this probably makes sense to do, too
20120222 20:54:41< Crendgrim> vultraz: not tested yet
20120222 20:55:06< mordante> yup, but this assembly is very basic, fetching memory, adding subtracting shifting, branching
20120222 20:55:13< mordante> all very basic
20120222 20:55:25< vultraz> Crendgrim: ah. well when you get a chance, ping me
20120222 20:55:41< mordante> haven't seen arm assembly, but I assume I can follow it rather fast
20120222 20:55:54< Crendgrim> vultraz: sure.
20120222 20:56:00< Crendgrim> not this evening though, I fear
20120222 20:56:34< vultraz> no hurry ;)
20120222 20:58:42< anonymissimus> vultraz: yes I know; which is why timotei and me suggested you downgrade your OS...
20120222 20:58:49< mordante> Ivanovic, ist there a open pandora emulator?
20120222 20:59:07< Ivanovic> mordante: (as far as i know) no
20120222 20:59:28< mordante> schade
20120222 20:59:33< anonymissimus> vultraz: there is this thread btw: http://forums.wesnoth.org/viewtopic.php?f=5&t=36208&p=521669#p521669
20120222 20:59:49< anonymissimus> which could be the same problem like yours
20120222 21:00:13< anonymissimus> or similar; no solution but at least you are united in despair
20120222 21:00:37< vultraz> ;)
20120222 21:00:50< vultraz> lol nice to know
20120222 21:01:14-!- Jozrael [~croselius@209.133.52.231] has quit []
20120222 21:01:54< vultraz> well, at least he was able to launch wesnoth
20120222 21:02:04< vultraz> instead of a loadscrren srash
20120222 21:02:41< vultraz> with either string disposal errors or the Xcode's build's error which I think might be similar
20120222 21:03:05< mordante> Ivanovic, thanks for the cross-compile guide, will try it another time
20120222 21:04:10< mordante> shadowm, fendrin the current map format is partially the result of design by comity 
20120222 21:04:35< mordante> the spaces for alignment are not required
20120222 21:04:37< Ivanovic> mordante: for compilation (after installing): have a look at the end of the file $PNDSDK/../sdk_utils/PandoraToolchain.cmake
20120222 21:04:56< mordante> I can't exactly remember why both usage and border_size were required
20120222 21:05:02< Ivanovic> it directly tells you the command to use for configuration with cmake to create the binaries which i use for the official package
20120222 21:05:42< mordante> the 10 starting location limit is due to the old single letter terrain format, where these numbers were the position
20120222 21:06:03< mordante> several people want to lift this limit after converting to the new map format
20120222 21:06:38< shadowm> mordante: who is comity?
20120222 21:06:38< mordante> but there are some non-coding issues as in team colours
20120222 21:06:56< shadowm> ah, never mind
20120222 21:07:22< shadowm> although the current map format was created entirely from scratch
20120222 21:07:50< mordante> several people wanted to work on allowing more starting locations, but thusfar nobody did so
20120222 21:07:56< shadowm> and as far as I remember there was no compatibility path built into the engine since it was too difficult to implement
20120222 21:08:09< mordante> there was a compatibility path
20120222 21:08:32< mordante> only it was planned from the start that the path would be short-lived
20120222 21:08:59< mordante> the only remain of it is the min_savegame_version (or something like that)
20120222 21:09:12< mordante> remaining thing*
20120222 21:09:27< mordante> Ivanovic, ok thanks
20120222 21:09:32< shadowm> that's not what I consider a compatibility path, though :p
20120222 21:09:46< mordante> shadowm, there was more
20120222 21:10:14< mordante> Wesnoth could load the old format and write the new format, but the code was a bit of a clutch
20120222 21:10:29< mordante> so it was planned to be removed after a while
20120222 21:10:34< shadowm> okay, it's been so long that my memory is really rusty on that kind of details
20120222 21:10:54< mordante> after which it was no longer possible to load savegames before the new format was introduced
20120222 21:11:11-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20120222 21:11:59-!- Upth [~ogmar@public-nat1.scc.losrios.edu] has joined #wesnoth-dev
20120222 21:11:59-!- Upth is now known as Upthorn
20120222 21:12:24< mordante> so I wouldn't be unhappy if the extra cruft is removed from the map_data
20120222 21:12:54< mordante> and I assume nobody hand-edits maps anymore so maybe also remove the extra spaces
20120222 21:13:05< shadowm> regarding the side number limit: As I said, it's far easier to raise the limit rather than remove it entirely, since the latter would require intrusive changes to code I am not familiarized with _at all_. I'm not keen on the former either because I for one dislike arbitrary limits, and while I'd be all right with a limit of 20 sides, someone could complain that we should really have 50 sides, and then other people would argue ...
20120222 21:13:11< shadowm> ... that 100 sides are a better limit, etc.
20120222 21:13:43< shadowm> so I'm not really going to work on that (sorry Gambit)
20120222 21:13:58< mordante> the point is if we allow 100 sides we also need 100 team colours
20120222 21:14:16< mordante> or a way to group them
20120222 21:14:27< Gambit> Not really
20120222 21:14:29< anonymissimus> zookeeper: do you understand this post: http://forums.wesnoth.org/viewtopic.php?p=521682#p521682 ?
20120222 21:14:39< shadowm> it's obvious that we already have the most team colors we can have without creating problems for people who aren't colorblind (colorblind people already have a lot of problems depending on their specific variation)
20120222 21:14:45-!- Elvish_Pillager [~eli@dhip-149.coburn.residences.colby.edu] has joined #wesnoth-dev
20120222 21:14:55< Gambit> Few games will use 100 teams
20120222 21:14:57< shadowm> the chromatic solution is repeating the same sequence over and over as necessary
20120222 21:15:19< shadowm> add-on authors can also decide best how to solve the problem and they have had the necessary tools for quite a while
20120222 21:15:25< Gambit> ^
20120222 21:16:27< zookeeper> anonymissimus, well, pending further information i'm assuming that the share_view=yes is the cause
20120222 21:17:22< shadowm> and actually, nine sides is already a bit too much as I mentioned in DD. For example, the difference between brown and orange isn't always obvious.
20120222 21:17:39< mordante> shadowm, Debian stable uses boost 1.42, so maybe it's an idea to raise the minimum version
20120222 21:18:04< mordante> Debian stable is already around for over a year, so I assume most distros use at least 1.42
20120222 21:18:26< shadowm> well, oldstable's security support was dropped last month I think
20120222 21:18:50< mordante> it was indeed dropped recently
20120222 21:19:26< mordante> I think there is still one oldstable release planned
20120222 21:20:35< shadowm> I don't see any need to raise the requirements at this very moment, but that might be something to consider depending on what crazy new dependencies people come up with this year
20120222 21:22:06< mordante> might be true, but I for boost people want to use newer features quite often
20120222 21:23:30< shadowm> There's a CRC library but nothing implementing SHA1 or MD5, huh.
20120222 21:39:16< anonymissimus> wesbot: bug 18138
20120222 21:39:17< wesbot> Bug #18138  Assigned to: Anonymissimus <anonymissimus>  Status: Fixed  Priority: 5 - Normal
20120222 21:39:20< wesbot> Summary: Wolves don't tip their heads in tropical & cave water
20120222 21:39:23< wesbot> Original submission: See png. The wolf-water and wolf-rider-water sprites are 
20120222 21:39:25< wesbot> not used in tropical and cave water.Terrain types used in attached png are:
20120222 21:39:28< wesbot> URL: https://gna.org/bugs/?18138
20120222 21:39:31< wesbot> Attached file (1st): https://gna.org/bugs/download.php?file_id=13006
20120222 21:41:21< shadowm> The fix for that topic's issue is trivial.
20120222 21:41:26< shadowm> I'm on it.
20120222 21:41:50< shadowm> technically fendrin is abusing the swamp village terrain, but whatever
20120222 21:46:59< shadowm> in fact, it might be better to fix the map
20120222 21:47:11< fendrin> shadowm: That is where?
20120222 21:48:02< shadowm> the Ka'lian map uses swamp village overlays on ground terrains, which makes the Wolf/Wolf Rider swimming animation match it
20120222 21:48:14< shadowm> http://forums.wesnoth.org/viewtopic.php?p=521694#p521694
20120222 21:49:05< shadowm> odds are the village you were looking for was the tropical village  (^Vht)
20120222 21:50:02< shadowm> there also happens to be a disjointed bridge tile at 8,10
20120222 21:50:48< CIA-61> anonymissimus * r53158 /trunk/projectfiles/VC9/wesnoth.vcproj: VC9 projectfile update
20120222 21:50:55< CIA-61> anonymissimus * r53159 /trunk/projectfiles/CodeBlocks/wesnoth.cbp: CB projectfile update
20120222 21:51:54< shadowm> okay, it looks like Alarantalara will take the choice I didn't want to; that is, extending the units' animation filters
20120222 21:53:36< anonymissimus> he will make them stricter again
20120222 21:53:42< fendrin> shadowm: Oh, I see. I have once visited Afrika, Kenia to be precisely. There they had such buildings with legs to be save from the animals. I didn't recognize that the terrain is a water only one.
20120222 21:54:15< anonymissimus> in the above bug I extended them since some terrains didn't get the water animation though should, but I overdid it
20120222 22:02:36-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 260 seconds]
20120222 22:04:20-!- fendrin [~fabi@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20120222 22:05:37-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20120222 22:06:06-!- fendrin [~fabi@88-134-10-236-dynip.superkabel.de] has joined #wesnoth-dev
20120222 22:06:06-!- fendrin [~fabi@88-134-10-236-dynip.superkabel.de] has quit [Changing host]
20120222 22:06:06-!- fendrin [~fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20120222 22:08:35-!- stikonas [~and@bcm-131-111-216-70.girton.cam.ac.uk] has joined #wesnoth-dev
20120222 22:08:35-!- stikonas [~and@bcm-131-111-216-70.girton.cam.ac.uk] has quit [Changing host]
20120222 22:08:35-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20120222 22:09:07< anonymissimus> fendrin: what do we gain from your [map] tag btw ?
20120222 22:09:50< Ivanovic> we can finally increase the version number in the savegame compatibility entry!
20120222 22:09:56< fendrin> anonymissimus: I can parse map files.
20120222 22:09:59< mordante> Yeah!
20120222 22:10:23< Ivanovic> shadowm, mordante: the boost version on the pandora is 1.41.0 btw
20120222 22:10:32< Ivanovic> so i speak in favor it not going for 1.42.0!
20120222 22:11:04< mordante> Ivanovic, ok, good reason
20120222 22:11:42< Ivanovic> mordante: besides i'd say that we should start discussing a boost version bump to 1.41.0 on the ML right now
20120222 22:12:03< Ivanovic> so that we basically have the OK done in case there are some features from one of the versions till 1.41 that a dev would like to use
20120222 22:12:20< Ivanovic> (once that is a case the bump to the minimum for this feature can be done without further discussion)
20120222 22:12:57< anonymissimus> fendrin: parsing where, for what ?
20120222 22:13:24< mordante> Ivanovic, if we discuss and agree, better bump the version directly the we can see whether it breaks things on the next dev release
20120222 22:14:01< fendrin> anonymissimus: The map editor can parse the map file with the wml parser. The current map files aren't parsable.
20120222 22:14:05< Ivanovic> i don't think it *can* break anything since with the minimum bump we don't change what the people got
20120222 22:14:25< Ivanovic> we just might force some people to somehow upgrade
20120222 22:14:34< mordante> well people could say that on their platform 1.41 is not supported
20120222 22:14:35< Ivanovic> and by now i think this number is rather small
20120222 22:14:52< Ivanovic> those that i care about most are the people participating on the ML
20120222 22:15:06< Ivanovic> if everyone there says that 1.41.0 is okay, then it should be fine
20120222 22:15:07< mordante> mainly yes
20120222 22:15:35< Ivanovic> (consider the amount of feedback we often get with dev releases and do you really think there will be relevant feedback before we say that this release is in beta/rc state?)
20120222 22:17:00< mordante> not really unless one of our packagers has an issue
20120222 22:17:16< mordante> like the Pandora packager who doesn't like boost 1.42 ;-P
20120222 22:17:58< Ivanovic> and those are likely to already state so on the ML
20120222 22:18:26< Ivanovic> in this case the people taking care of the oldest systems are probably Rhonda and me
20120222 22:18:48< Ivanovic> (since the pandora system is now 3 years old (basic software stack wise))
20120222 22:19:30< mordante> probably yes, but as long as Rhonda doesn't backport I already complain if it's too new
20120222 22:22:15< Ivanovic> :)
20120222 22:27:37< loonycyborg> Why do we need 1.42/1.41? I don't see a reason to bump requirements unless we need something particular.
20120222 22:28:00-!- PolarPanda [~quassel@unaffiliated/peterporty] has quit [Read error: Connection reset by peer]
20120222 22:30:18< loonycyborg> The only bump that I'd request would be to 1.44. That would allow us to use boost.filesystem v3.and add proper support for unicode filenames on windows.
20120222 22:33:32-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20120222 22:34:56< mordante> since bumping the minimum boost version is asked on a regular basis
20120222 22:35:25< mordante> IMO 1.44 is too new since it's not yet in Debian stable, but that might change before 1.12 is ready
20120222 22:35:34< mordante> however then we still have the Pandora
20120222 22:36:03< Ivanovic> there a version bump is unlikely to happen, at least not until the new firmware stream is ready which is still ages away
20120222 22:36:14< Ivanovic> (and even then at least those who have not updated so far can't run wesnoth)
20120222 22:38:29< shadowm> it would be nice if people could still compile in stable without using backports
20120222 22:38:53< mordante> exactly
20120222 22:38:59< shadowm> and assuming we don't go for yet another two-year cycle, 1.12 will be out before support for oldstable is dropped
20120222 22:39:18< shadowm> (that is, assuming oldstable = squeeze and stable = wheezy by that point, which sounds rather unlikely anyway)
20120222 22:40:06< mordante> well Debian's freeze is planned for the summer
20120222 22:40:33< Ivanovic> i could probably create and ship those boost libs as part of the pandora package
20120222 22:40:41< shadowm> if things go as with squeeze, then wheezy would become stable between February and March, and presumably 1.12 would be released before FOSDEM
20120222 22:40:52< Ivanovic> but i would prefer to have the system compile and work with the default setup available on the pandora
20120222 22:40:55< shadowm> so December or January
20120222 22:43:19< mordante> shadowm, I still have hopes Debian is able to release faster after the freeze
20120222 22:43:22< mordante> Ivanovic, agreed
20120222 22:44:14< shadowm> I also hope so. I'm not sure for how long I'll be able to live with KDE SC 4.7 (assuming it ever lands in testing) :p
20120222 22:47:55< shadowm> I just noticed this in the SDL 1.2.15 release notes: "Disabled MMX blitters since they don't compile on modern compilers"   . I wonder if that has any performance implications for Wesnoth?
20120222 22:48:11< shadowm> (in x86 and x86_64, of course)
20120222 22:50:04-!- Crendgrim [~crend@77-22-112-160-dynip.superkabel.de] has quit [Quit: ...]
20120222 22:52:30< mordante> I assume in x86_64 other instructions like SSE are used
20120222 22:53:44< shadowm> ah, right
20120222 22:54:21< mordante> but for older x86 it might be an issue, but IIRC SSE was introduced in 2000
20120222 22:54:54< mordante> but I'm off, night
20120222 22:55:15-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20120222 23:12:16-!- PolarPanda [~quassel@unaffiliated/peterporty] has joined #wesnoth-dev
20120222 23:19:35-!- Alarantalara [~Adium@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has joined #wesnoth-dev
20120222 23:21:08< CIA-61> alarantalara * r53160 /trunk/ (changelog data/core/units/goblins/Wolf_Rider.cfg): 
20120222 23:21:08< CIA-61> Make wolf raise head for terrain where unit is submerged only
20120222 23:21:08< CIA-61> Allows for all embellishments/overlays on submerged bases and
20120222 23:21:08< CIA-61> considers only overlays that affect submersion
20120222 23:21:08< CIA-61> http://forums.wesnoth.org/viewtopic.php?f=4&t=36212
20120222 23:24:49< CIA-61> alarantalara * r53161 /branches/1.10/ (changelog data/core/units/goblins/Wolf_Rider.cfg): 
20120222 23:24:49< CIA-61> backport r53160:
20120222 23:24:49< CIA-61> Make wolf raise head for terrain where unit is submerged only
20120222 23:24:49< CIA-61> Allows for all embellishments/overlays on submerged bases and
20120222 23:24:49< CIA-61> considers only overlays that affect submersion
20120222 23:24:49< CIA-61> http://forums.wesnoth.org/viewtopic.php?f=4&t=36212
20120222 23:27:11-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 260 seconds]
20120222 23:28:17-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Quit: Leaving.]
20120222 23:34:47-!- Blueblaze [~Blueblaze@adsl-76-202-23-45.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20120222 23:49:54-!- Upthorn [~ogmar@public-nat1.scc.losrios.edu] has quit [Ping timeout: 255 seconds]
20120222 23:50:09-!- Danceman [~Danceman@a94-132-158-220.cpe.netcabo.pt] has quit [Quit: Leaving]
--- Log closed Thu Feb 23 00:00:51 2012