--- Log opened Sat Jan 03 00:00:20 2009
--- Day changed Sat Jan 03 2009
20090103 00:00:20-!- Gnutoo [n=gnutoo@host14-134-dynamic.40-79-r.retail.telecomitalia.it] has quit ["Leaving"]
20090103 00:01:59< zookeeper> yo sapient
20090103 00:02:06< zookeeper> i had a little something for you in the log
20090103 00:02:14< Sapient> yo zookeeper
20090103 00:02:24< Sapient> I must have missed it
20090103 00:03:09< Sapient> 13:20 <zookeeper> Sapient, whenever you got the time, Eastern_Invasion/scenarios/05.Northern_Outpost.cfg:261-266 triggers quite an interesting preprocessor bug/feature. neither VARIABLE_OP ever gets used, presumably because of having that #ifdef inside the macro parameter.
20090103 00:03:39< Sapient> sounds nasty
20090103 00:03:39< zookeeper> yup
20090103 00:03:45< zookeeper> yeah, it was weird.
20090103 00:11:19-!- freim [n=hogne@feathers.os-tr.net] has quit [Read error: 54 (Connection reset by peer)]
20090103 00:15:20< Sapient> I'll take a look at it
20090103 00:15:58< Sapient> we probably need a new log level for "parser"
20090103 00:16:18< Sapient> and lots of print outs
20090103 00:23:23-!- loonycyborg [n=sergey@79.139.246.35] has quit ["Zzzzzzzzzzzzzzzzzzzzzzz"]
20090103 00:24:03-!- loonybot [n=loonybot@79.139.246.35] has quit [Remote closed the connection]
20090103 00:27:15-!- ivan_i [n=chatzill@ppp91-76-34-135.pppoe.mtu-net.ru] has joined #wesnoth-dev
20090103 00:30:17-!- fendrin [n=fabi@g227034255.adsl.alicedsl.de] has joined #wesnoth-dev
20090103 00:31:24< zookeeper> Sapient, but you think it should be fixed? great!
20090103 00:31:44< Sapient> well, I didn't say that much :)
20090103 00:31:55< Sapient> it may turn out that the WML is wrong when I get into it
20090103 00:32:12< Sapient> for now I got to go eat stuff though
20090103 00:33:19< zookeeper> ok
20090103 00:35:19 * Sapient afk
20090103 00:39:19< ivan_i> Great Python shift is pretty controversial topic... reading email... :)
20090103 00:46:03< Ivanovic> time to head off to bed, n8
20090103 00:58:50< ivan_i> can someone tell me about "old widgets" and "new widgets"?
20090103 01:01:02< Soliton> old widgets are what we use now mostly and new widget is what is getting implemented under src/gui/.
20090103 01:02:01< ivan_i> Thanks, what are the plans for that new widgets?
20090103 01:02:26< ilor> ivan_i: the plans are to replace the old widgets and the old theme system 
20090103 01:02:37< ilor> right now it's mainly some of the dialogs
20090103 01:02:58< ivan_i> I just want to know more details about it
20090103 01:03:19< ilor> in general, Mordante is *the* new widgets person
20090103 01:04:17< ilor> as far as I can tell the idea is to have a more modern design compared to the old one which was fairly clunky, and have more stuff WML-configurable
20090103 01:04:38-!- kthakore2 [n=kthakore@CPE001310a1899c-CM001868e2ad12.cpe.net.cable.rogers.com] has joined #wesnoth-dev
20090103 01:08:45< ivan_i> I must that old widgets are more Python friendly, unfortunately
20090103 01:14:14< ilor> as far as I know the old widgets were very inflexible, layout-wise
20090103 01:20:48< ivan_i> I can see this, I mean we need to re-think the way it is done implementation-wise
20090103 01:22:00< ilor> I think the actual _use_ of, say, a new-widgets dialog might not be all that python-unfriendly. See how the editor dialogs are used in editor_controller.cpp for instance.
20090103 01:25:02< ivan_i> I may be wrong though, just because I did old-widgets based preferences dialog in Python, it wasn't easy but I was able to do it look at py__gui.cpp
20090103 01:25:53< ilor> ivan_i: I haven't had time to look at your code, sorry
20090103 01:26:52< ivan_i> No need to sorry, it's work-in-progress and quite experimantal
20090103 01:27:54< Sirp_> ivan_i: as I think I suggested to loonycyborg on the mailing list, I think efforts would be better spent on making a way to program scenarios and events in Python by hooking in game_events.cpp to allow Python code to respond to events.
20090103 01:31:04-!- Noyga [n=lame-z@AVelizy-151-1-83-225.w86-205.abo.wanadoo.fr] has quit [Read error: 113 (No route to host)]
20090103 01:31:11< ivan_i> I will work on this
20090103 01:32:26< Sirp_> ivan_i: as I'm sure you can appreciate, the only 'experiences' Wesnoth has had so far with Python in the main codebase have been rather negative, or neutral at best, so I think our developers will remain skeptical of Python at least until they see a positive example of how it can benefit the project.
20090103 01:33:02< zookeeper> boucman, random thought: it'd be useful if animations had two (or an arbitrary number?) of different x,y keys for offsets. for example, if you're making a ballistic trajectory for a projectile you need one set of x,y offsets to position the projectile on the unit's sling or whatever, and another set of x,y offsets to create the vertical arc.
20090103 01:33:24-!- freim [n=hogne@feathers.os-tr.net] has joined #wesnoth-dev
20090103 01:33:44< zookeeper> i've found myself having to use both x,y and halo_x,halo_y atm
20090103 01:34:36< ivan_i> I see this and I spend a lot of time making stuff very reliable (as opposed to previous Python efforts)
20090103 01:39:33-!- Elvish_Pillage2 [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has quit ["Hi! I'm a quit message virus vaccine. If you see a quit message virus, don't replace your quit message with it!"]
20090103 01:47:01-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev
20090103 01:47:46-!- Rrenys [n=rrenys@81-20-159-197.levira.ee] has quit [Read error: 104 (Connection reset by peer)]
20090103 01:48:35-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has quit [Client Quit]
20090103 01:49:44-!- zookeeper [n=l@dsl-tkubrasgw1-fe9ff800-215.dhcp.inet.fi] has quit []
20090103 01:50:30< Sapient> ilor: in fact, the old widgets were not inflexible layoutwise
20090103 01:51:02< Sapient> I introduced transitional code which allowed you to completely tweak widget layout through two different methods
20090103 01:51:31< Sapient> in one method, you didn't have to make a subclass
20090103 01:51:58< Sapient> just take the result of dialog.layout() then tweak it
20090103 01:52:45< Sapient> the other strategy was to override dialog.layout in a subclass, call the parent class, then tweak the result before returning it
20090103 01:53:36< Sapient> finally, things which were commonly meant to be aligned in different configurations could have a constant defined such as GUI::BUTTON_LEFT, etc
20090103 01:54:34< Sapient> there was at least one example of overriding the layout method (see the editor's file_chooser from the last time I was working on it)
20090103 01:54:42-!- kthakore2 [n=kthakore@CPE001310a1899c-CM001868e2ad12.cpe.net.cable.rogers.com] has quit ["Leaving"]
20090103 01:55:55< Sapient> however, none of these strategies were integrated with the theme code, or WML configurable
20090103 01:56:14< Sapient> that was the main deficiency
20090103 01:56:52-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090103 01:57:31< Sapient> I think the main reason it was abandoned was because mordante preferred a fresh start rather than building on someone else's work, and I don't blame him for that decision
20090103 01:58:04< ilor> Sapient: thanks for a thorough explanation ;)
20090103 01:58:26< Sirp_> ...especially since the Wesnoth widget system had evolved over five years from something that was meant to be 'just good enough' to scrape to Wesnoth 1.0
20090103 02:01:32< Sapient> that may be the very reason I stopped working on it, because I wasn't eager to totally rewrite some large components (such as menu.cpp) and it was becoming apparent that they did need it
20090103 02:09:21-!- boucman [n=rosen@159.83.102-84.rev.gaoland.net] has quit [Remote closed the connection]
20090103 02:09:48-!- Turuk [n=JDiSab@pool-71-242-102-166.phlapa.east.verizon.net] has quit ["To err is human.. but to really foul things up requires a computer"]
20090103 02:11:06-!- BenUrban [n=benurban@c-68-49-10-243.hsd1.md.comcast.net] has joined #wesnoth-dev
20090103 02:15:50< ivan_i> Old widget stuff may not be well organized but it's very simple and straightforward
20090103 02:20:23-!- ivan_i [n=chatzill@ppp91-76-34-135.pppoe.mtu-net.ru] has quit [Read error: 60 (Operation timed out)]
20090103 02:34:38-!- BenUrban_ [n=benurban@c-68-49-10-243.hsd1.md.comcast.net] has joined #wesnoth-dev
20090103 02:34:38-!- BenUrban [n=benurban@unaffiliated/benurban] has quit [Read error: 54 (Connection reset by peer)]
20090103 02:36:07-!- ivan_i [n=chatzill@ppp91-76-35-150.pppoe.mtu-net.ru] has joined #wesnoth-dev
20090103 02:36:45-!- BenUrban_ is now known as BenUrban
20090103 02:40:03-!- [Relic] [n=[Relic]@adsl-76-229-202-137.dsl.milwwi.sbcglobal.net] has joined #wesnoth-dev
20090103 02:41:06< [Relic]> Hello :)
20090103 02:41:18< Sapient> hi [Relic]
20090103 02:42:13< [Relic]> :)
20090103 02:42:52< ivan_i> Sirp_: can you tell me please what do you think about my work and do you accept my ideas?
20090103 02:43:54< Sapient> well, he did express his opinion on the mailing list already
20090103 02:44:22< ivan_i> It wasn't clear, that's why I ask
20090103 02:44:24< Sirp_> ivan_i: well as I said, I think the best approach to win buy-in from developers would be to do something like make the events system have Python support
20090103 02:44:48< Sirp_> and then demonstrate that a large number of Python developers would join and produce things that would move Wesnoth forward.
20090103 02:45:03< Sirp_> ...or random map generation code would also be a prime candidate for moving to Python....
20090103 02:49:22< Sirp_> to be honest, the idea of rewriting large portions of currently working code in Python I'm very dubious about. I'm highly dubious that Python is a better language for the job, despite the claims of its advocates, and I'm doubtful that the kind of programmers attracted to the project would be more desirable than the programmers likely displaced and alienated by having the code they maintain removed.
20090103 02:54:04-!- cib0 [n=cib@p4FD0BDB0.dip.t-dialin.net] has quit ["Leaving."]
20090103 03:01:20< ivan_i> I don't remove anything, if you look at GameController class for example you can see that it's very similar to the original
20090103 03:02:01-!- cib0 [n=cib@p4FD0BDB0.dip.t-dialin.net] has joined #wesnoth-dev
20090103 03:02:48< Sapient> I think Wesnoth development should be focused on improving Wesnoth, not changing its language to suit someone's preference
20090103 03:12:53-!- JW [n=X@c-71-57-85-168.hsd1.il.comcast.net] has joined #wesnoth-dev
20090103 03:13:08-!- JW is now known as Guest62627
20090103 03:13:25< ivan_i> It's not just a preference: the shift to the higher-level language opens new possibilities that was not possible before
20090103 03:14:07< Sapient> the proof is in the pudding as they say
20090103 03:14:16-!- Guest62627 is now known as JW1
20090103 03:15:18< Sapient> if you can offer _real_ improvements in an _incremental_ fashion that fits _cleanly_ with the existing code, then your work will be accepted for what it is
20090103 03:15:20< Sirp_> ivan_i: if we get into a discussion of what is 'better' for Wesnoth out of C++ and Python, no-one will win. (meaning you will lose)
20090103 03:16:26-!- cib0 [n=cib@p4FD0BDB0.dip.t-dialin.net] has quit ["Leaving."]
20090103 03:16:40-!- JW1 is now known as JW
20090103 03:17:13-!- JW is now known as Guest452
20090103 03:17:38< Guest452> okay, I can't figure out this registered nickname stuff...
20090103 03:17:58< ivan_i> I don't anything against C++, I'm all for combination of both, low level engine in C++, everythin else in Python
20090103 03:18:19< Sapient> C++ is an object oriented language
20090103 03:18:29< Sapient> it's not like we're using assembler
20090103 03:18:37-!- Guest452 is now known as JW1
20090103 03:19:03< Sirp_> but why not everything in C++?
20090103 03:21:36< Sirp_> that is the question we can argue about forever. Personally I think that everything in C++ is a reasonable option. I'm willing to accept trying Python if it can bring immediate benefit. What I'm not willing to do is just blindly accept re-writing tested, maintained code for Python, with some "on faith" attitude that the Python will be better and more maintainable and there will be more maintainers for it.
20090103 03:22:58-!- crimson_penguin [n=ben@64.201.60.218] has quit [No route to host]
20090103 03:23:45< Sirp_> so this is why I suggest concentrating on an area where it can be shown that Python actually makes the project *better*. :)
20090103 03:24:31< Sirp_> rather than just have our code stripped of all its valuable type checking. :)
20090103 03:24:57< Sapient> and constness ;)
20090103 03:30:42 * Sapient afk
20090103 03:37:14< ilor> gn
20090103 03:37:16-!- ilor [n=user@unaffiliated/ilor] has quit []
20090103 03:45:32-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev
20090103 03:46:23< Sirp_> Sapient: exactly. :)
20090103 03:46:49< ivan_i> Unit tests are more reliable then type checkong
20090103 03:47:00< ivan_i> *checking
20090103 03:48:20< Sapient> are spoons more reliable than forks? ;)
20090103 03:49:14< Sirp_> ivan_i: well, I downloaded your branch. Where are the unit tests located?
20090103 03:49:53< Sapient> ivan_i: I'm not trying to be rude but you are making some odd claims
20090103 03:50:46< Sirp_> also, if you have a function, that takes a unit and returns its hitpoints, like int get_hitpoints(const unit& u); in C++ this guarantees it won't modify the unit.
20090103 03:51:07< Sirp_> in Python this would be def get_hitpoints(u): ...  -- in Python how would you unit test this function to make sure it doesn't modify the unit?
20090103 03:51:14< Sapient> ivan_i: I think you've gotten your answers and there is a tangible way forward if you want to take it
20090103 03:53:17< Sirp_> basically I just refuse to believe that *anyone* does the kind of unit testing that would test all the type checking that C++ does. If you did, it would certainly make the Python code size dwarf the C++ code size to do so.
20090103 03:54:07< Sirp_> unit testing is good at a completely different level. It should test a function's overall goals, not the tiny errata that C++'s type checking so much more elegantly and reliably guarantees.
20090103 03:55:04< ivan_i> I would make it a "hitpoints" property, and I know that it won't modify the unit because Python is not C++ and it works differently
20090103 03:55:40-!- voris [n=voris@c-24-16-107-149.hsd1.wa.comcast.net] has joined #wesnoth-dev
20090103 03:55:41< Sirp_> but how do you know that it won't modify the unit?
20090103 03:55:52< ivan_i> http://effbot.org/zone/python-objects.htm
20090103 03:56:02< ivan_i> please read
20090103 03:56:33< Sirp_> I am well aware of how objects work in Python.
20090103 03:56:57< Sirp_> that function could still modify the unit.
20090103 03:57:13< ivan_i> http://effbot.org/zone/call-by-object.htm
20090103 03:57:15< ivan_i> and this
20090103 03:58:06< ivan_i> it *could* but normally it doen't happen in Python
20090103 03:59:28< Sirp_> ivan_i: like I said, I am well aware of how Python's object model works.
20090103 04:00:31< Sirp_> it is simply that in C++, one can easily obtain information from an object or variable's signature of what it modifies and what it doesn't. In Python you don't, and IMO, this loses you very valuable information.
20090103 04:01:25< Sirp_> this can be demonstrated in something as simple as the 'remove' function of the standard 'list' object
20090103 04:01:39< Sirp_> if you look at the function definition it will look something like this:
20090103 04:01:44< Sirp_> def remove(x): ...
20090103 04:02:03< Sirp_> from reading that, you don't know how it operates? Does it return the list with the item removed, or does it operate 'in-place' on the list and return nothing?
20090103 04:03:51< Sapient> dang that second article is confusing
20090103 04:04:07< Sirp_> personally, I think this is a huge disadvantage of Python. The code is not self-explanatory. You have to document and unit test everything.
20090103 04:04:19< Sapient> what is the distinction they are making by saying that "call by reference" is different than "call by object reference"?
20090103 04:04:42< Sirp_> Sapient: yeah no kidding. Which is yet another problem imo. Python (and Java for that matter) claim to be 'simple' but they actually have subtly confusing object and memory models.
20090103 04:05:01< Sapient> Are they saying that an "object reference" is not a type of "reference"? or what?
20090103 04:05:24-!- crimson_penguin [n=ben@64.201.60.218] has joined #wesnoth-dev
20090103 04:06:19< Sapient> "IT IS NOT CALL BY REFERENCE because access is not given to the variables of the caller, but merely to certain objects."
20090103 04:06:52< Sapient> that makes it about clear as mud
20090103 04:07:02< Sirp_> I *think* what it means is basically that you can mutate the object referred to (unless it's immutable) but you can't actually mutate what object is referred to.
20090103 04:07:38< Sapient> ah
20090103 04:07:58< Sirp_> Sapient: I agree it's incredibly poorly written and if I didn't know better I would say it was written purely so that anyone who criticizes python in any way can be smugly referred to the article and told "ahhh you're just saying this because you don't understand how Python works properly!" ;)
20090103 04:08:00< Sapient> thanks for translating into english
20090103 04:08:12< Sirp_> Sapient: that's just what I *think* it means. I don't really have a clue.
20090103 04:09:07< Sapient> <mysterious voice> You have given access to... "certain objects"...
20090103 04:09:38< Sirp_> but I looked at this in some detail back when I read the code to the cpython interpreter.
20090103 04:11:37< ivan_i> I think that http://www.haskell.org/~pairwise/intro/section1.html could be helpful for both of you
20090103 04:12:08< Sapient> yes, clearly I need help. thank you.
20090103 04:13:23< Sirp_> ivan_i: I really don't think we need help. :) I know C++, Python, and Haskell all very well, thank you.
20090103 04:13:25< ivan_i> I just want to say that there are diffrent ways to think about programming
20090103 04:13:26< Sapient> this humble Haskell tutorial has opened my eyes to the majesty of a total Wesnoth conversion into Python
20090103 04:13:45 * Sapient ommm
20090103 04:13:58< Sirp_> ivan_i: absolutely. The question is, WHY should we think in these different ways? WHY should we take this different approach?
20090103 04:14:20< Sirp_> we are NOT convinced that such a conversion gives us any benefit at all.
20090103 04:14:59< Sapient> heretic! step into the light
20090103 04:15:22 * Sapient levitates
20090103 04:16:04< Sirp_> ivan_i: perhaps it is you who doesn't see how C++ is a powerful high and low level language that is ideal for games projects, rather than us who don't understand how powerful Python is?
20090103 04:17:54-!- Ivanovic_ [n=ivanovic@dtmd-4db2b748.pool.einsundeins.de] has joined #wesnoth-dev
20090103 04:17:57< Sirp_> after all, we are one of the most successful open source games around, while other games that have been programmed in Python, such as GalaxyMage, have failed quite badly. I think that is quite strong evidence we are doing something right.
20090103 04:27:37-!- Turuk [n=JDiSab@pool-71-242-102-166.phlapa.east.verizon.net] has joined #wesnoth-dev
20090103 04:33:36-!- Ivanovic [n=ivanovic@dtmd-4db2a3f2.pool.einsundeins.de] has quit [Read error: 110 (Connection timed out)]
20090103 04:33:52-!- Ivanovic_ is now known as Ivanovic
20090103 04:35:47< ivan_i> And all I want is to make this game even better. And Python is a right way to go in my opinion.
20090103 04:36:56< Sapient> opinions are plentiful but results are far more compelling
20090103 04:38:12< Sirp_> ivan_i: The thing is, all anyone has to show for it so far is their opinion. You have to realize that our opinions are just as valid as yours. So, we want to see something compelling: I gave examples of things that can actually give substantial benefits -- if we could write scenarios using Python, or could write random map generators in Python, I think that would be a huge and immediate benefit to users.
20090103 04:38:26-!- Turuk [n=JDiSab@pool-71-242-102-166.phlapa.east.verizon.net] has left #wesnoth-dev []
20090103 04:38:32< ivan_i> I have a public repo, you may see the results
20090103 04:38:34< Sirp_> later on, I think it would be awesome if we could write custom dialogs in Python -- imagine if a scenario designer could make their own custom dialog with custom placed controls on it.
20090103 04:38:57-!- Turuk [n=JDiSab@pool-71-242-102-166.phlapa.east.verizon.net] has joined #wesnoth-dev
20090103 04:39:42< Sapient> I'll point you to my earlier statement: " if you can offer _real_ improvements in an _incremental_ fashion that fits _cleanly_ with the existing code, then your work will be accepted for what it is"
20090103 04:41:01< Sapient> so there is a chance for more Python to be welcomed into Wesnoth but it is dependent upon certain conditions
20090103 04:42:28< Sapient> cya later, have fun.
20090103 04:42:28-!- Sapient [n=sapien-x@c-24-126-133-155.hsd1.ga.comcast.net] has left #wesnoth-dev []
20090103 04:42:32< Sirp_> personally I think rather than looking at "C++ vs Python" people should just look at "hey I want to add xyz functionality to Wesnoth, and hey, I'll do it in Python because I happen to think this is the best tool for the job."
20090103 04:43:24< ivan_i> Dave this is *exacly* what I think and doo
20090103 04:43:42< Sirp_> okay, so what functionality are you adding?
20090103 04:44:48< ivan_i> People can write new authoring tools using the same Wesnoth GUI instead of pygtk
20090103 04:45:15< ivan_i> I made it possible already
20090103 04:46:13< ivan_i> no tests and no documentation -- i just didn't have enough time
20090103 04:47:51< Sirp_> okay; that would be valuable, I would like to see that.
20090103 04:55:42< ivan_i> run it as --mod testmod and you will see
20090103 05:04:29-!- happygrue_ [n=happygru@c-98-223-188-175.hsd1.in.comcast.net] has joined #wesnoth-dev
20090103 05:06:03-!- [Relic] [n=[Relic]@adsl-76-229-202-137.dsl.milwwi.sbcglobal.net] has quit ["Leaving"]
20090103 05:46:09-!- happygrue_ [n=happygru@c-98-223-188-175.hsd1.in.comcast.net] has quit []
20090103 06:03:37-!- crimson_penguin [n=ben@64.201.60.218] has quit ["Leaving"]
20090103 06:36:27-!- Sirp_ [n=me@c-76-102-104-121.hsd1.ca.comcast.net] has quit [Read error: 60 (Operation timed out)]
20090103 06:42:09-!- Sirp_ [n=me@c-76-102-104-121.hsd1.ca.comcast.net] has joined #wesnoth-dev
20090103 06:57:22-!- JW1 [n=X@c-71-57-85-168.hsd1.il.comcast.net] has quit []
20090103 07:23:14-!- ivan_i [n=chatzill@ppp91-76-35-150.pppoe.mtu-net.ru] has quit ["ChatZilla 0.9.84 [Firefox 3.0.5/2008121621]"]
20090103 08:06:19-!- Sapient [n=sapien-x@c-24-126-133-155.hsd1.ga.comcast.net] has joined #wesnoth-dev
20090103 08:07:13< Sapient> zookeeper: I've been staring at preprocessor.cpp for the past few hours...
20090103 08:07:45< Sapient> whoever wrote this thing (I think silene) apparently didn't believe in the idea of comments
20090103 08:08:20< Sapient> so it's still not clear what the heck certain things are doing in the relevant code sections
20090103 08:09:44< Sapient> if anyone intimate with the preprocessor can delineate the purpose of the different token types and how exactly temporary variables such as "strings_" are being used, then I'll take a look at it
20090103 08:10:42< Sapient> for now, let's just say you can't use #ifdef inside a macro argument
20090103 08:11:12-!- Sapient [n=sapien-x@c-24-126-133-155.hsd1.ga.comcast.net] has quit [Client Quit]
20090103 08:12:39< Sirp_> Sapient: Yeah, I find preprocessor.cpp very confusing too
20090103 08:26:58-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev
20090103 08:40:29-!- mordante [n=chatzill@roadie.xs4all.nl] has joined #wesnoth-dev
20090103 08:40:40< Espreon> Hello mordante.
20090103 08:40:56< mordante> morning
20090103 08:41:14< mordante> hi Espreon 
20090103 08:41:57< Sirp_> hi mordante 
20090103 08:42:05< mordante> hi Sirp_ 
20090103 09:12:59-!- loonybot [n=loonybot@79.139.246.35] has joined #wesnoth-dev
20090103 09:14:44-!- Bosshoss [n=chris@c-69-181-182-116.hsd1.ca.comcast.net] has joined #wesnoth-dev
20090103 09:14:51< Bosshoss> how do I go about donating?
20090103 09:15:04-!- loonycyborg [n=sergey@79.139.246.35] has joined #wesnoth-dev
20090103 09:15:52< Bosshoss> I cant find anything on the site about donating to the devs?
20090103 09:15:56< mordante> Sirp_ ^
20090103 09:17:24< mordante> there is/was a wiki page about it, but can't find it :-/
20090103 09:20:41< Bosshoss> i will keep looking
20090103 09:22:06< mordante> Bosshoss http://www.wesnoth.org/wiki/Donate
20090103 09:23:11< Bosshoss> i want to help beyond bug reporting
20090103 09:23:19< Bosshoss> i lack the skills otherwise lol
20090103 09:23:33< mordante> thanks :-)
20090103 09:26:41< Sirp_> Bosshoss: please don't donate.
20090103 09:26:50< Sirp_> and we need to take that wiki page down.
20090103 09:27:14< Sirp_> Bosshoss: we appreciate that you enjoy Wesnoth, and we would love to have feedback, and bug reports, and other help, but we don't need any donations, though we greatly appreciate the thought.
20090103 09:29:32< Bosshoss> well ok
20090103 09:29:34< Sirp_> Bosshoss: alternatively, if there is a particular feature or set of images you enjoy, we could tell you which developer was responsible for it...
20090103 09:29:38< Sirp_> and you could send them a gift... :)
20090103 09:29:39< Bosshoss> your work is appreciated
20090103 09:29:54< Bosshoss> maybe I can look into that
20090103 09:30:16< Sirp_> Bosshoss: for instance, chocolates. :)
20090103 09:30:23< mordante> :-)
20090103 09:30:28< Bosshoss> I guess Mammon proves to be more of a problem in the project than its worth huh? 
20090103 09:30:31< Bosshoss> hah
20090103 09:30:36< Bosshoss> Chocolates for EVERYONE!
20090103 09:30:46< Sirp_> Bosshoss: not really. We simply have a pretty good funding source in ads on the website, as well as some support from Google.
20090103 09:31:00< Espreon> Yay for chocolate.
20090103 09:31:03< Bosshoss> how about a 12 pack of Natural Light for everone (above 21 of course)
20090103 09:31:12< Espreon> ...
20090103 09:31:22< Sirp_> Bosshoss: so, we simply think there are better causes. We have a decent enough funding source.
20090103 09:31:23< mordante> is that something alcoholic?
20090103 09:31:23< Bosshoss> the Cheapist beer money can buy!
20090103 09:31:27< Espreon> How 'bout something even the younger devs may enjoy?
20090103 09:31:46< Sirp_> I have been meaning to set up something for sending developers gifts.
20090103 09:33:01< mordante> Bosshoss we had a summit and some project suggested chocolate since it offends nobody, booze can offend people due to age or religion  ;-)
20090103 09:33:36< Bosshoss> fair enough Sirp_ 
20090103 09:33:38< Sirp_> Bosshoss: yes, mordante, Ivanovic and I were recently at a summit where there was an entire session dedicated to discussing the concept of sending gifts of chocolate as an incentive. :)
20090103 09:34:03< mordante> the session was good and the choclate also ;-)
20090103 09:34:13< Bosshoss> Special Amsterdam Chocolate for ALL
20090103 09:35:11< mordante> I never heard of it :-/ (and I can be in Amsterdam in less than one hour)
20090103 09:35:51< Espreon> How 'bout Skittles?
20090103 09:38:05< Espreon> Fourty-one ounce bags of Skittles for everyone.
20090103 09:39:48-!- EdB [n=EdB@79.88.117.169] has joined #wesnoth-dev
20090103 09:42:22-!- Bosshoss [n=chris@c-69-181-182-116.hsd1.ca.comcast.net] has quit []
20090103 09:49:46-!- Bosshoss [n=chris@c-69-181-182-116.hsd1.ca.comcast.net] has joined #wesnoth-dev
20090103 09:49:53-!- Bosshoss [n=chris@c-69-181-182-116.hsd1.ca.comcast.net] has quit [Client Quit]
20090103 09:58:52-!- Sirp_ [n=me@c-76-102-104-121.hsd1.ca.comcast.net] has quit ["leaving"]
20090103 10:20:07-!- boucman [n=rosen@159.83.102-84.rev.gaoland.net] has joined #wesnoth-dev
20090103 10:38:41< Ivanovic> moin
20090103 10:38:47< boucman> moin moin
20090103 10:39:08-!- BenUrban [n=benurban@unaffiliated/benurban] has quit [Read error: 104 (Connection reset by peer)]
20090103 10:39:21-!- BenUrban [n=benurban@c-68-49-10-243.hsd1.md.comcast.net] has joined #wesnoth-dev
20090103 10:39:39< mordante> moin moin moin
20090103 10:40:42 * Ivanovic heard of chocoloate
20090103 10:40:53< Ivanovic> where is chocolate?
20090103 10:41:19< mordante> too late
20090103 10:42:21< Ivanovic> :(
20090103 11:01:46< CIA-52> ivanovic * r31915 /trunk/data/core/about.cfg: updated Credits for Hungarian translation
20090103 11:03:29-!- Noyga [n=lame-z@AVelizy-151-1-83-225.w86-205.abo.wanadoo.fr] has joined #wesnoth-dev
20090103 11:03:36-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has quit [Remote closed the connection]
20090103 11:03:52< noy> Ivanovic: where is chocolate?
20090103 11:04:08< Ivanovic> not here anymore
20090103 11:04:14< Ivanovic> looks like mordante ate all himself
20090103 11:04:16< Ivanovic> :(
20090103 11:05:35< boucman> ivan_i is not a very talkative guy... it's hard to get info from him...
20090103 11:05:55 * mordante has tommy ache :-/
20090103 11:06:17< CIA-52> boucman * r31916 /trunk/ (4 files in 4 dirs): add new HI portrait by Emilien Rotival
20090103 11:06:28< mordante> boucman indeed
20090103 11:06:52 * boucman still has no opinion on the propposal
20090103 11:06:58-!- zookeeper [n=l@dsl-tkubrasgw1-fe9ff800-215.dhcp.inet.fi] has joined #wesnoth-dev
20090103 11:07:54 * Ivanovic is rather sceptical regarding the impact
20090103 11:09:01< boucman> I personally think that the playsingle controller is one of these part that is old and clunky, and a rewrite (whatever the language) is probably worth it
20090103 11:10:46 * mordante is also sceptical, like with all rewrite projects, especially in a different language as if a different language can magically fix a problem
20090103 11:11:05< boucman> hmm
20090103 11:11:11< mordante> (and I did see enough rewrite ideas due to buzzword foo)
20090103 11:12:08< boucman> I'd tend to agree that c++ is not a good language, you can do good things with it, but the level of C++knowledge you need to have just to avoit the most common traps is a bit frightening
20090103 11:13:20< mordante> I agree playcontroller is in a rather bad shape and can use a rewrite, but the current idea seems to rewrite quite a bit more
20090103 11:13:43< mordante> (of course we still don't know what it exactly rewrites)
20090103 11:14:19< boucman> we already did big changes and most of them have turn well so far (the only one I see that didn't work was the opengl branch)
20090103 11:14:40< boucman> when you think of it, wirting a new gui engine is a way bigger change than switching language :)
20090103 11:14:53< mordante> I know
20090103 11:15:47< boucman> I seem to be the only dev taking part in the discussion that don't know python, and I'm not afraid of learning...
20090103 11:16:52< mordante> I know some Python and will need to learn quite a bit more to work on a project the size of Wesnoth
20090103 11:17:04< mordante> but my lack of C++ knowledge never stopped me ;-)
20090103 11:17:21< CIA-52> zookeeper * r31917 /trunk/data/core/units/drakes/ (Enforcer.cfg Gladiator.cfg Slasher.cfg Warden.cfg): The clasher line doesn't fly.
20090103 11:18:20< boucman> zookeeper: oups
20090103 11:18:22< mordante> but what bothers me is that Ivan still didn't answer the questions about what the changes exactly are and what benefit Python brings us
20090103 11:18:34< boucman> ESR_: around ?
20090103 11:18:43< mordante> and we also don't know how experimental the code is
20090103 11:18:50< boucman> mordante: seeing how talkative he is, we'd better ask ESR_ :P
20090103 11:19:33< loonycyborg> mordante: The chief benefit of python that it doesn't require recompilation and somewhat easier to modify than C++.
20090103 11:19:36< mordante> boucman to be honest if the original author can't answer those questions I really start to fear to use it
20090103 11:20:02-!- cib01 [n=cib@p4FD0F24A.dip.t-dialin.net] has joined #wesnoth-dev
20090103 11:20:14< boucman> mordante: some people are great coders but are not good talkers, he's just shy
20090103 11:20:22< mordante> loonycyborg easier is in the eye of the beholder
20090103 11:20:31< boucman> and it seems there is a fair amount of job in what he did, so it's really worth having a look at it
20090103 11:21:04< boucman> now, if it has to be at least a transpartent replacement to get in
20090103 11:21:10< boucman> and I don't think it's there yet
20090103 11:21:13< mordante> boucman well he was quite talkative in the log, but not about the details of his project
20090103 11:21:18< loonycyborg> mordante: E.g. in C++ in order to add a function you need to modify both the source and header while in python you just add it.
20090103 11:21:25< boucman> maybe do a python branch and merge later, we had some success with tha
20090103 11:22:29< cib01> here the C++/Python discussion, too?
20090103 11:22:54< mordante> and another problem is I don't have time to look at it now, I'm quite busy with 1.6
20090103 11:23:06< mordante> IMO this is a rather bad timing to look into it
20090103 11:23:27< boucman> mordante: I disagree on that last point
20090103 11:23:32< boucman> announce it now,
20090103 11:23:40< loonycyborg> mordante: Well.. Then don't waste your time talking and do something productive instead :)
20090103 11:23:45< boucman> implements the missing bits in febuary 
20090103 11:23:53< boucman> merge, first thing in 1.7
20090103 11:23:57-!- JW1 [n=X@c-71-57-85-168.hsd1.il.comcast.net] has joined #wesnoth-dev
20090103 11:25:42< mordante> loonycyborg I try do do productive things, but if I ignore the discussion now, people will claim I never objected to the idea :-|
20090103 11:26:14< boucman> mordante: you can be productive now, neither ESR_ nor ivan_i are around :)
20090103 11:26:15< loonycyborg> mordante: Nobody is going to merge it before 1.6 anyway.
20090103 11:26:38< mordante> boucman and that's exactly my problem I want to see whether it's mergable before attempting to merge
20090103 11:27:09< boucman> mordante: thatn you have at least till febuary to look at it, so it's fine
20090103 11:27:26< mordante> afaik nobody even reviewed the code and we're already talking about merging it :-|
20090103 11:27:33< loonycyborg> As a branch it's mergeable, but it's not ready for trunk yet :)
20090103 11:27:50< mordante> boucman and I probably will be busy until 1.6 with the widgets
20090103 11:29:17< boucman> mordante: we're not talking about merging it NOW
20090103 11:29:37< boucman> it's a big disruptive change, and it needs early planning, and that's what we're doing
20090103 11:30:01< boucman> basically we are saying that we are not against the idea, but it won't happen until it's ready anyway...
20090103 11:30:29< mordante> yes and that's my problem, I like to discuss whether we _want_ to merge it before discussion _how_ to merge it
20090103 11:31:50< boucman> mordante: except the last paragraph in ESR_,s mail that nobody really discussed, that's what we're doing
20090103 11:32:50< mordante> which paragraph?
20090103 11:33:50< boucman> the last one of his original email, but everybody forgot it, so don't resurect it or we might actually start discussing how to merge :P
20090103 11:35:46 * mordante is really confused
20090103 11:36:05 * boucman checks the mail
20090103 11:36:23< boucman> paragraph titled == Getting There From Here ==
20090103 11:37:41 * mordante has another idea what paragraph means (so was looking at the last 3 lines)
20090103 11:39:10< mordante> I did reply to that part 
20090103 11:40:54< boucman> mordante: yes, but you were the only one...
20090103 11:41:09< boucman> ok, bbiam
20090103 11:47:04-!- ilor [n=user@unaffiliated/ilor] has joined #wesnoth-dev
20090103 11:47:16< cib01> this python suggestion seems to have disturbed the whole wesnoth community
20090103 11:48:13-!- cib01 is now known as cib0
20090103 11:50:15-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090103 11:59:00< CIA-52> zookeeper * r31918 /trunk/data/campaigns/Eastern_Invasion/scenarios/05.Northern_Outpost.cfg: Worked around the preprocessor bug.
20090103 11:59:11< loonycyborg> It's nice that the ML is so active for a change :)
20090103 12:00:04-!- cib0 [n=cib@p4FD0F24A.dip.t-dialin.net] has quit ["Leaving."]
20090103 12:04:26< CIA-52> zookeeper * r31919 /trunk/data/core/about.cfg: Someone broke about.cfg.
20090103 12:04:41 * boucman not guilty
20090103 12:12:03-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev
20090103 12:12:49-!- Netsplit kubrick.freenode.net <-> irc.freenode.net quits: Smar, isaac
20090103 12:14:19-!- Netsplit over, joins: Smar, isaac
20090103 12:18:49-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has quit [Remote closed the connection]
20090103 12:19:40 * Ivanovic is guilty of breaking about.cfg
20090103 12:19:53< Ivanovic> hey, i wanted to break something in the game, too!
20090103 12:20:58< loonycyborg> Ivanovic: Try inserting backspaces in C++ code to break something in game :)
20090103 12:21:11< loonycyborg> Make sure that it's not in comments :)
20090103 12:21:46< mordante> why not deleting the */ also breaks stuff
20090103 12:22:58< loonycyborg> Better do "dd if=/dev/random of=/dev/hda". That's a much more reliable way of breaking stuff.
20090103 12:23:24< Ivanovic> loonycyborg: i do not want to break *my* system
20090103 12:23:32< Ivanovic> i wanted to break wesnoth trunk a little
20090103 12:24:10< loonycyborg> Ivanovic: Place wesnoth trunk on partition of its own and dd it :)
20090103 12:28:50< Ivanovic> loonycyborg: but then only i got this fun
20090103 12:29:01< Ivanovic> since uploading won't work nicely...
20090103 12:32:15 * loonycyborg is out of ideas for breaking things :(
20090103 12:42:46-!- ivan_i [n=chatzill@ppp85-140-59-128.pppoe.mtu-net.ru] has joined #wesnoth-dev
20090103 12:44:07< ivan_i> High
20090103 12:48:05< mordante> hi ivan_i 
20090103 12:48:25< ivan_i> hi mordante
20090103 12:48:28< loonycyborg> ПРЕВЕД :)
20090103 12:48:43< ivan_i> Здарова Серёга
20090103 12:49:00< boucman> hey ivan_i
20090103 12:49:15< boucman> ivan_i: could you detail a little more what you did and how far you are
20090103 12:49:37< boucman> i.e what works and what doesn't in your branch, when you think you will have a fully functional prototype etc...
20090103 12:49:42< ivan_i> it's hard
20090103 12:49:50< boucman> yeah, but it's important
20090103 12:50:14< boucman> we don't really know where we're going, and "it's in python" is not an argument by itself...
20090103 12:50:31< boucman> how many LOC do you replace, what functionalities do you reimplement etc...
20090103 12:51:06< ivan_i> It is a long strory...
20090103 12:51:51< loonycyborg> Then be sure to post it on the ML..
20090103 12:52:07< mordante> yes but I would like to know the answers, without them we can't make an informed decission 
20090103 12:52:24< ivan_i> Yes I amg going to do it
20090103 12:52:29< mordante> thanks
20090103 12:54:09< boucman> and don't worry if your english is not correct, we have lot of non-native contributors
20090103 13:00:56-!- Rrenys [n=rrenys@81-20-159-197.levira.ee] has joined #wesnoth-dev
20090103 13:06:29< ivan_i> I think it's very important to make a decision about the direction we are moving, it's impossible to move forward at this point unless we have some kind of agreement
20090103 13:07:38< mordante> ivan_i it's not possible to decide if we don't know what we decide :-|
20090103 13:08:31-!- Rrenys [n=rrenys@81-20-159-197.levira.ee] has quit [Success]
20090103 13:13:20-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has quit []
20090103 13:26:40< ivan_i> This could be useful http://github.com/ii/wesnoth/commits/wespython/ commit messages have some details you may be interested in
20090103 13:31:57< mordante> I hope you don't mean we should look through 56 pages of history to find your commits and read those log in order to gain some insight 
20090103 13:34:01< freim> we obviously have a language barrier here, so please lets try to be a bit patient
20090103 13:34:18< mordante> I'm not in a hurry
20090103 13:34:40< ivan_i> I will write about it and post it to the mailing list, but it requires some time
20090103 13:34:59< mordante> no problem
20090103 13:35:01< zookeeper> it's not going to get seriously worked on in probably at least a couple of months since we're working on 1.6, so take your time to write out all the details these folks want to know.
20090103 13:35:41< ivan_i> thanks, zookeeper
20090103 13:35:42< mordante> only don't expect us to decide a direction before we know a little bit more about what your plans are
20090103 13:37:53< mordante> freim IMO this is also not a decision to make in a hurry
20090103 13:48:38< freim> mordante: of course not. I think the barrier for a branch should be low, but for merging to main we need to carefully evaluate it
20090103 13:49:21< mordante> freim it's already in a git branch and branching/merging in git is much easier
20090103 13:49:40< mordante> at least so I've been told, but I know merging with svn is hard
20090103 13:49:47< mordante> never tried with git
20090103 13:52:11< ivan_i> I have just rebased my repo against the latest SVN head and pushed the changes to github
20090103 13:56:15-!- Turuk_ [n=JDiSab@pool-71-242-102-166.phlapa.east.verizon.net] has joined #wesnoth-dev
20090103 14:06:04< ivan_i> Running the game with "--mod testmod" switch is a good way to see some advantages of my approach
20090103 14:13:08-!- Turuk [n=JDiSab@pool-71-242-102-166.phlapa.east.verizon.net] has quit [Read error: 110 (Connection timed out)]
20090103 14:17:54< mordante> I rather read your email first
20090103 14:21:22< boucman> ivan_i: assume we can't test your change
20090103 14:28:53< ivan_i> "If your ideas are any good, you’ll have to ram them down people’s throats"
20090103 14:29:16< mordante> what's that supposed to mean :-|
20090103 14:31:54< boucman> ivan_i: I'm not sure you're in a position to have that sort of attitude
20090103 14:32:23< boucman> and we've been rather receptive to your ideas, but we need explanations.
20090103 14:32:40< ivan_i> why not?
20090103 14:32:46< boucman> dumping code on us and going "accept it or else" will probably get your code rejected
20090103 14:33:03< boucman> because we are the ones who have the veto power here...
20090103 14:37:04< mordante> and to be honest I think a lot of projects wouldn't even consider to switch languages when somebody reimplementes a part of their codebase
20090103 14:39:00< ilor> ivan_i: you have to write *something* about you have done if only because not all devs know python and certainly not all of them have the time to go through your code and see what it does
20090103 14:39:21< ilor> s/about you/about what you/
20090103 14:39:53< mordante> and also what your ideas for the future regarding your project
20090103 14:40:26< mordante> I assume you have some kind of masterplan, which we also want to know
20090103 14:41:26< ivan_i> the future depends on whether it is accepted or rejected, in fact I don't worry about being rejected at all
20090103 14:41:29< ilor> right now those who have not dug into your code know that a) it is in Python b) it reiplements something and c) it apparently does something faster.
20090103 14:42:00< boucman> and d) it's not usefull yet since you can't play a game with it
20090103 14:42:25< mordante> ivan_i of course it depends on whether it's accepted, but I assume you have some ideas in case it gets accepted
20090103 14:43:33< boucman> ivan_i: right now you don't guve us any good reason to accept it except "it exists" I'm pretty sure you've put a lot of effort in it and it really have good things in it
20090103 14:43:55< boucman> actually I would be very happy to accept it, but you don't tell us anything about it
20090103 14:44:43< boucman> you give us no reasons, you don't explain your work, so I have no way to decide if it's good or not, and if I don't have a convincing argument I tend to go the safe way i.e reject it
20090103 14:45:34< ivan_i> thanks I will write in detail about it..
20090103 14:47:22< mordante> it's quite simple you'll need to convince us that merging your branch is a good idea
20090103 14:47:56< mordante> and there's no need to hurry, I rather read a good email later as a hurried one sooner
20090103 14:49:51-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090103 14:58:53-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev
20090103 15:05:07-!- ettin_ [n=jorda@205.206.219.87.dynamic.jazztel.es] has joined #wesnoth-dev
20090103 15:11:19< ivan_i> mordante: what is *your* ideas and masterplan about the new widgets?
20090103 15:15:56< mordante> to make sure that the various parts of widget code get replaced with one code base
20090103 15:16:07< mordante> this includes theme and dialog
20090103 15:16:21< mordante> make them fully configurable from wml
20090103 15:17:21< mordante> make them easy adaptable for various screensizes from the small handheld devices to the 30" screens which some people might use
20090103 15:17:43 * Ivanovic will be using a 40" TV soon
20090103 15:17:44< mordante> at least 1680x1050 is used by some people
20090103 15:17:45< Ivanovic> ^^
20090103 15:17:52< Ivanovic> (okay, just 1920x1080)
20090103 15:18:38-!- ettin [n=jorda@145.207.219.87.dynamic.jazztel.es] has quit [Read error: 110 (Connection timed out)]
20090103 15:19:01 * mordante hopes Ivanovic will play on a small screen at the FOSDEM
20090103 15:20:02 * Ivanovic hopes to get his pandora till then
20090103 15:20:16< mordante> that's what I meant ;-)
20090103 15:20:32< Ivanovic> two days ago one of the pandora devs told me that they hope to have shipped all in the term of the next two months
20090103 15:20:56< Ivanovic> no idea if i will get mine in time for fosdem, somehow i don't believe so...
20090103 15:21:25< mordante> with the new widgets it's also easier to draw in layers, which might open new interesting features, but I want to look at that after phase I (as described above) has been finished
20090103 15:21:26< Ivanovic> but maybe EvilDragon comes to fosdem and will hand me mine in a huge ceremony...
20090103 15:21:39< Ivanovic> ^^
20090103 15:22:03< mordante> two months will be after the FOSDEM
20090103 15:22:18< mordante> does the ceremony involve chocolate?
20090103 15:22:51< mordante> I really would like to see how Wesnoth runs on the Pandora
20090103 15:23:16< ivan_i> thanks, the details about new widgets are very important,  I think it would be great if we could coordinate our work somehow, the current implementation is written in a way that is quite hard to wrap and expost to Python
20090103 15:23:44< mordante> well I didn't write it with Python in mind ;-)
20090103 15:23:58< mordante> what makes it hard to wrap to Python?
20090103 15:24:49< ivan_i> C++ idioms :)
20090103 15:25:25< mordante> well it's written in C++ :-)
20090103 15:25:46< loonycyborg> So is boost::python ;)
20090103 15:26:20< mordante> I really don't know too much about Python and nothing about your bindings so you really need to be more specific
20090103 15:27:05< mordante> and I assume boost::python uses even more C++ idioms ;-)
20090103 15:30:25< loonycyborg> mordante: I think that weird copy semantics can be rather hard to wrap..
20090103 15:30:44< mordante> loonycyborg which copy semantics?
20090103 15:32:06< ivan_i> new widgets is the only thin in curretn code base that made me seriously consider using boost:python
20090103 15:32:07< loonycyborg> mordante: IIRC you had some problems with regards to what copy should mean for ttext class.
20090103 15:32:28< loonycyborg> Though I didn't research this in any detail.
20090103 15:32:30< mordante> had indeed, noncopyable now
20090103 15:33:28-!- elias [n=elias@allegro/developer/allefant] has joined #wesnoth-dev
20090103 15:35:02< mordante> ivan_i what's then the exact problem with the new widgets?
20090103 15:35:18-!- kitty_ [n=kitty@e180213063.adsl.alicedsl.de] has joined #wesnoth-dev
20090103 15:37:24-!- stikonas [n=and@ctv-79-132-188-113.vinita.lt] has joined #wesnoth-dev
20090103 15:38:26< loonycyborg> mordante: I think that even single inheritance can be a pain to wrap without boost::python
20090103 15:39:25< mordante> playcontroller also uses inheritance and the old widgets as well
20090103 15:40:21< mordante> that's why I asked ivan_i what the problem is ;-)
20090103 15:40:35< Soliton> what are "true variable-extent types"?
20090103 15:40:42< ivan_i> I did wrap the old widgets system in a way that it looks as though Python objects inherit from C++ objects, and that wasn't easy
20090103 15:40:47< Soliton> gogle doesn't seem to know anything about it.
20090103 15:43:56< mordante> no idea Soliton 
20090103 15:44:42< mordante> I'm confused are the old or the new widgets hard to wrap?
20090103 15:46:38-!- Turuk_ [n=JDiSab@pool-71-242-102-166.phlapa.east.verizon.net] has quit ["To err is human.. but to really foul things up requires a computer"]
20090103 15:47:21< ivan_i> Old widgets are already wrapped :)
20090103 15:47:39< elias> didn't think about coming here before already asking on the mailing list - but why are you using the Python-C-API at all?
20090103 15:47:46< zookeeper> kitty_, am i imagining or aren't those drake portraits somehow a lot brighter than before? http://www.wesnoth.org/forum/viewtopic.php?p=329424#p329424
20090103 15:48:01< elias> boost::python would seem like a more logical choice
20090103 15:49:36< Espreon> zookeeper: I do not think so.
20090103 15:49:43< ivan_i> maybe boost::python looks very attractive
20090103 15:49:57 * mordante still has no clue why wrapping the new widgets is hard :-(
20090103 15:50:05< kitty_> zookeeper: i don't think so, i took them from the image library and changed nothing about the colours  - are you sure?
20090103 15:50:11< loonycyborg> elias: boost::python is an additional dependency. If he's a masochist let him do it, and save some work for packagers :)
20090103 15:50:35< elias> better to save work for code-maintainers than packers though
20090103 15:50:41< elias> packagers
20090103 15:50:56< mordante> loonycyborg software needs to be maintained, which I prefer to make easier
20090103 15:52:13< zookeeper> kitty_, well, compared to the ones currently in the game they are brighter and lack some texturing. i'm not sure if you're intentionally using some other version.
20090103 15:52:16< elias> i've never used boost::python myself since I don't use C++, but for C it's much easier to maintain cython wrappers instead of python-c-api
20090103 15:52:18-!- busfahrer [n=busfahre@ulmg-5f70cdf7.pool.einsundeins.de] has joined #wesnoth-dev
20090103 15:53:25< zookeeper> kitty_, someone was making some altered versions of them at some point presumably to make them fit in better style-wise, but i don't really remember the details.
20090103 15:53:26< kitty_> zookeeper: surely not intentional! i'll look into it - does that only affect the drakes or the orcs also?
20090103 15:53:32< zookeeper> drakes only
20090103 15:53:54< kitty_> zookeeper: that was doofus-01, but they didnt look very finished.
20090103 15:54:47< loonycyborg> There are even more convenient solutions than using boost::python directly: http://www.language-binding.net/pyplusplus/pyplusplus.html
20090103 15:55:08< Soliton> elias: just curious, are you using C over C++ by choice or existing projects require it?
20090103 15:55:19< elias> by choice
20090103 15:55:35< Soliton> why's that?
20090103 15:55:37< elias> I think it's a much superior language :)
20090103 15:55:41< elias> simpler
20090103 15:55:54< elias> you can describe all the syntax/semantics of C on far less space than C++
20090103 15:56:20< Soliton> so the added possibilities of C++ just make things more complicated/confusing?
20090103 15:56:31< zookeeper> kitty_, wait, actually the orcs are a little bit brighter too.
20090103 15:57:00< zookeeper> not that i would noticed if i hadn't taken a closer look, seems you also cleaned up a couple of blurry spots and defined some lines better, right?
20090103 15:57:04< elias> to me they do. but nowadays I prefer Python over anything - just using C where I need the extra performance
20090103 15:57:18< Soliton> ok, i see.
20090103 15:58:08< ivan_i> mordante: Python wrapping is hard by itself that's why boost::python was written, in general wrapping any new widget class means: 1. create/inherit a C++ class with PyObject pointer with a sole purpose of redirecting C++ methods to Python methods 2. create the Python class that understands the things written in 1. and can be subclassed from Python. May be it's not *hard*. Just too much boilerplate
20090103 16:00:50< mordante> ivan_i and how does the new widgets differ from the old widgets in that regard, the number of widgets?
20090103 16:01:52< ivan_i> old widgets are more plain old C, new widgets are idiomatic C++ that's the only problem
20090103 16:02:08< kitty_> zookeeper: hmm, i just checked the drakes and orcs with the colourpicker - and the colours are identical or very, very similar. the only drake which is a lot brighter is the burner - i'll upload a new version for him in a minute
20090103 16:02:23< mordante> which idiomatic part of C++ causes the problem?
20090103 16:03:24< ivan_i> everything that is not plain old C causes problems, Python is written in C, not C++
20090103 16:04:11< zookeeper> kitty_, at least in warlord2 and 3 the difference is quite visible
20090103 16:04:20< elias> what Python itself is written in should have nothing to do with it
20090103 16:04:32< elias> but the Python-C-API is for wrapping C, not C++
20090103 16:04:47< zookeeper> not that it's a bad thing necessarily. it's not a big difference and probably no one notices.
20090103 16:08:08< kitty_> zookeeper: i've just changed the burner - but i don't see it with the orcs. i'm confused... i'd like to understand what caused it but on the other hand it's not really important...
20090103 16:08:58< mordante> a lot the Wesnoth code uses C++ specific features :-/
20090103 16:09:26< zookeeper> kitty_, http://imagebin.org/34927
20090103 16:09:28< mordante> but what do you suggest to do so the new widgets can be wrapped in Python easier?
20090103 16:09:35< zookeeper> old ones on the left. the texturing is completely different.
20090103 16:09:40-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has quit [Read error: 54 (Connection reset by peer)]
20090103 16:10:08< kitty_> the burner is changed to that version - but the fighter is not the one in the image library!
20090103 16:11:11< zookeeper> the image library just contains whatever images people have posted on the forums, so you probably took some funny tweak WIP someone had made
20090103 16:11:46< kitty_> damn, the wing was the part i invested most work in... where is the place to find the right ones?
20090103 16:12:12< ivan_i> elias: Python/C API works perfectly with C++ and Objective-C, but everything that is not plain C can cause problems
20090103 16:12:23< zookeeper> well...from the game, i suppose. didn't someone tell you how to find them inside the app?`
20090103 16:12:40< kitty_> uhm, no...
20090103 16:13:07< zookeeper> ok, well, you can find all of them through http://svn.gna.org/viewcvs/wesnoth/trunk/data/core/images/portraits/ too
20090103 16:13:33< kitty_> thanks
20090103 16:14:01< zookeeper> but porting the wing you made to the proper version isn't probably too hard for you, i'd imagine
20090103 16:14:21< kitty_> i still have to texture it...
20090103 16:15:37< elias> ivan_i: i know. i maintained the python AI for some time - that's what made me tell myself if I had written it, I wouldn't have used the Python-C-API
20090103 16:15:58< mordante> kitty_ did you already update the male mage of the light?
20090103 16:15:58< elias> of course I was too lazy to ever invest the time to change it to something else :)
20090103 16:16:03-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev
20090103 16:16:41< kitty_> mordant: sorry, no - i hope to get the MoLs done later today
20090103 16:17:16< mordante> ok, no hurry was just curious, I don't read the forum that often ;-)
20090103 16:18:48< mordante> ivan_i just curious, but why did you decide to use the Python-C-API, when you know it might be hard to use for C++?
20090103 16:18:53< zookeeper> kitty_, i'll wait until you've dealt with the drakes before committing, the warlords are just fine as it is.
20090103 16:19:16< kitty_> thanks, zoopkeeper
20090103 16:23:33< ivan_i> mordante: pencils are harder to use then crayons... but we still use them to draw
20090103 16:25:26< mordante> ivan_i sorry but you speak in riddles, no idea what you mean with that
20090103 16:25:40< mordante> I normally try to pick the best tool for the job
20090103 16:26:48< elias> or the one you know best how to use, which is a valid reason as well
20090103 16:27:33< elias> although in this case, where we should factor in years of maintenance by different people, it wouldn't be a good reason
20090103 16:29:07-!- stikonas [n=and@ctv-79-132-188-113.vinita.lt] has quit [Read error: 110 (Connection timed out)]
20090103 16:30:04< mordante> is the tool can be used, yes but in this case the tool doesn't seem to be too well fit for the job
20090103 16:30:48< mordante> when I first needed to write some text parsing tools, I started to learn Perl instead of trying to use the languages I already knew
20090103 16:31:25< Soliton> mordante: i think he just doesn't know because he hasn't tried. it's just a suspicion.
20090103 16:32:11< mordante> that might be
20090103 16:32:44< mordante> I'm just cursious why ivan_i picked the Python-C-API
20090103 16:32:44< ivan_i> Does anybody hew know Haskell?
20090103 16:33:14< elias> don't mention Haskell as long as Ivanovic is here
20090103 16:33:19< elias> or he will kill you :)
20090103 16:33:25< ivan_i> :)
20090103 16:34:05< mordante> elias and after that he'll kill you for pinging him ;-)
20090103 16:34:40< elias> heh
20090103 16:35:47< ivan_i> I really want to understand monads and arrows...
20090103 16:36:14< ivan_i> I can't get it still...
20090103 16:38:06 * Ivanovic goes to the shed in the garden to fetch the large ax...
20090103 16:38:22 * Espreon gives Ivanovic a flamethrower.
20090103 16:38:39< Ivanovic> i don't need such strange tools
20090103 16:38:49< kitty_> zookeeper: i just updated the drakes.
20090103 16:38:56< Ivanovic> i am strong enough to get problems "solved" with an ax...
20090103 16:39:06< Espreon> But fire is fun!
20090103 16:39:45< ivan_i> You want ti kill me just because I mentioned monads and Haskell? I want to hear the whole story
20090103 16:40:28< loonycyborg> ivan_i: I've learned some haskell.
20090103 16:40:28< mordante> I'm sure Ivanovic will explain, while using the axe on you ;-)
20090103 16:41:56< loonycyborg> Still don't understand monads as good I would like, though I use type system as guide :)
20090103 16:42:48< mordante> ivan_i any suggestions what we can do to make the new widgets more Python friendly?
20090103 16:45:31< CIA-52> mordante * r31920 /trunk/src/gui/widgets/event_handler.cpp: Fix a typo in my last commit.
20090103 16:45:37< CIA-52> mordante * r31921 /trunk/src/gui/widgets/ (13 files): 
20090103 16:45:37< CIA-52> Code cleanup step 1.
20090103 16:45:37< CIA-52> At the moment widget have an coordinate in the parent window and on the
20090103 16:45:37< CIA-52> screen. After the changes to the drawing engine the coordinate in the
20090103 16:45:38< CIA-52> parent window is not really needed anymore. So we can do several
20090103 16:45:40< CIA-52> refactoring and cleanup steps.
20090103 16:45:42< CIA-52> Remove the old coordinates and use the sceen coordinates everywhere.
20090103 16:45:44< CIA-52> mordante * r31922 /trunk/src/gui/widgets/ (12 files): 
20090103 16:45:46< CIA-52> Cleanup part 2.
20090103 16:45:48< CIA-52> Rename the screen_foo to foo (partly reverts part 1).
20090103 16:45:50< CIA-52> mordante * r31923 /trunk/src/gui/widgets/ (window.cpp window.hpp): 
20090103 16:45:54< CIA-52> Cleanup part 3.
20090103 16:45:56< CIA-52> twindow::get_client_rect() can now use the version in tpanel and thus
20090103 16:45:58< CIA-52> can be removed from window.?pp.
20090103 16:46:02< CIA-52> mordante * r31924 /trunk/src/gui/widgets/ (7 files): 
20090103 16:46:04< CIA-52> Cleanup part 4.
20090103 16:46:06< CIA-52> find_widget2 now acts the same as find_widget so remove the duplicate
20090103 16:46:08< CIA-52> function and change it's callers to use find_widget.
20090103 16:46:12< CIA-52> mordante * r31925 /trunk/ (20 files in 2 dirs): 
20090103 16:46:14< CIA-52> Cleanup part 5.
20090103 16:46:16< CIA-52> The full redraw flag is no longer needed since widgets are always fully
20090103 16:46:18< CIA-52> redrawn now.
20090103 16:46:20< CIA-52> mordante * r31926 /trunk/src/gui/widgets/container.hpp: 
20090103 16:46:22< CIA-52> Cleanup part 6 (the last part).
20090103 16:46:24< CIA-52> Remove the now obsolete background_changed_ member and it's setter and
20090103 16:46:28< CIA-52> getter.
20090103 16:46:39< loonycyborg> Ivanovic: http://learnyouahaskell.com/ :)
20090103 16:46:41-!- stikonas [n=and@ctv-79-132-188-113.vinita.lt] has joined #wesnoth-dev
20090103 16:47:20< ivan_i> loonycyborg: it looks that we live in the same city, it could be a good idea to meet in person :)
20090103 16:47:30-!- stikonas [n=and@ctv-79-132-188-113.vinita.lt] has quit [Remote closed the connection]
20090103 16:47:59 * loonycyborg doesn't like meeting in person :)
20090103 16:48:07< loonycyborg> *meetings
20090103 16:48:18< loonycyborg> IRC is so much nicer..
20090103 16:48:48< ivan_i> Do you drink beer?
20090103 16:48:56< loonycyborg> Nope.
20090103 16:51:43< loonycyborg> I also don't smoke and don't watch soccer. In other words I'm totally anti-social person :)
20090103 16:52:32< Espreon> Heh.
20090103 16:52:37< ivan_i> I dislike soccer too, but I do have some unhealthy habbits and I do like to sicialize somtime
20090103 16:53:36< busfahrer> Mmmm beer
20090103 16:53:39< ivan_i> *socialize
20090103 16:54:07 * mordante also likes beer :-)
20090103 16:54:34 * Espreon does not like the thought of consuming beer.
20090103 16:57:37< busfahrer> I'm so glad to be German ;-)
20090103 16:59:38< AI0867> Espreon: that's good, since there is very little beer in the US ;)
20090103 17:00:29 * mordante 'll get new German beer soon :-)
20090103 17:00:42< busfahrer> mordante: What are you going to get?
20090103 17:01:22< mordante> Tannenzäpfle
20090103 17:01:40< busfahrer> Very nice, one of the best Pilseners there is.
20090103 17:01:53< mordante> and as Wesnoth player I love their bottles, I always TC them red in my mind ;-)
20090103 17:02:02< Ivanovic> hehehe
20090103 17:02:06 * Espreon wants alcohol extracted from Skittles.
20090103 17:02:12< Espreon> It would be very potent.
20090103 17:02:46< mordante> yes but I haven't found it in the Netherlands
20090103 17:02:55< mordante> only bad beers get exported 
20090103 17:03:02< mordante> ;-)
20090103 17:03:54< busfahrer> True. Also, export beers are usually brewed seperately for the foreign market, so you basically get a different beer with the same label.
20090103 17:04:10< busfahrer> At least it's the way it works for German beer, since other countries don't have the law of purity.
20090103 17:04:12< AI0867> Espreon: depends, what kind of sugars does it contain?
20090103 17:04:49< ivan_i> The best beer is a fresh beer -- something that was made right there
20090103 17:04:50< AI0867> and, with fermentation, you won't get past 15% or so, you'll need a still if you want stronger stuff =P
20090103 17:05:06< busfahrer> Skittleschnaps!
20090103 17:05:41< kitty_> mordante: i just finished cleaning both MoLs and some minor edits to some other portraits - is it more comfortable for you if i mail them to you or should i post them on the forum?  
20090103 17:05:41< busfahrer> bbl
20090103 17:05:57< mordante> kitty_ I prefer email
20090103 17:08:51-!- crimson_penguin [n=ben@64.201.60.218] has joined #wesnoth-dev
20090103 17:08:52< kitty_> mordante: sent
20090103 17:10:56< mordante> kitty_ thanks, got it
20090103 17:11:02< kitty_> :)
20090103 17:12:31-!- ilor_ [n=user@unaffiliated/ilor] has joined #wesnoth-dev
20090103 17:15:02-!- ilor [n=user@unaffiliated/ilor] has quit [Read error: 60 (Operation timed out)]
20090103 17:15:51< CIA-52> mordante * r31927 /trunk/data/core/images/portraits/Kitty/transparent/ (6 files): Updated version of Kitty's portraits.
20090103 17:16:36< mordante> kitty_ uploaded ^
20090103 17:16:56< mordante> Rhonda did you manage to test my patch for your PPC problem?
20090103 17:30:39< CIA-52> mordante * r31928 /trunk/src/gui/widgets/ (generator.cpp scrollbar_container.cpp): 
20090103 17:30:39< CIA-52> Another small cleanup.
20090103 17:30:39< CIA-52> Now that we only have screen coordinates the set_size hacks in the
20090103 17:30:39< CIA-52> generator and scroll_container are no longer needed and thus removed.
20090103 17:41:14-!- Espreon is now known as Espreon_
20090103 17:50:10< CIA-52> ai0867 * r31929 /trunk/data/core/units/drakes/Sky.cfg: Fix a typoed image path.
20090103 17:51:11< Rhonda> mordante: No, because it's weekend. It's even strange that I'm responding right now. :)
20090103 17:52:03< mordante> :-)
20090103 17:52:28< mordante> the question now it does the world get even stranger :-
20090103 17:52:37< mordante> is*
20090103 17:54:03< mordante> but then commit I it as is
20090103 17:55:15-!- ivan_i [n=chatzill@ppp85-140-59-128.pppoe.mtu-net.ru] has quit ["ChatZilla 0.9.84 [Firefox 3.0.5/2008121621]"]
20090103 17:55:29< Rhonda> Now _that_ sentense was strangely set.
20090103 17:55:31-!- Elvish_Pillager [n=eli@24-177-37-183.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090103 17:55:35< Rhonda> yoda speak you do
20090103 17:56:52< mordante> then it seems the world gets stranger ;-)
20090103 17:58:08-!- elias [n=elias@allegro/developer/allefant] has quit ["Leaving"]
20090103 18:05:55-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit [Read error: 110 (Connection timed out)]
20090103 18:11:59-!- Espreon_ [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has quit [Remote closed the connection]
20090103 18:14:33-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090103 18:14:43-!- kitty_ [n=kitty@e180213063.adsl.alicedsl.de] has quit []
20090103 18:17:21-!- ettin [n=jorda@103.207.219.87.dynamic.jazztel.es] has joined #wesnoth-dev
20090103 18:28:43-!- ettin_ [n=jorda@205.206.219.87.dynamic.jazztel.es] has quit [Read error: 110 (Connection timed out)]
20090103 18:35:20< Rhonda> wtf - wesnoth.org mail is handled by 10 65.18.193.12.wesnoth.org.
20090103 18:35:37< Rhonda> Ivanovic, you fumbled seriously. Even though 65.18.193.12.wesnoth.org has address 65.18.193.12
20090103 18:36:08< Ivanovic> uhm, okay
20090103 18:37:01< Ivanovic> then let me just point the entry to wesnoth.org
20090103 18:37:09< Rhonda> Or just remove it. ;)
20090103 18:37:16< Rhonda> that will have the same result
20090103 18:37:24< Rhonda> no MX means mails go to the A record
20090103 18:37:30< Ivanovic> okay, will be difficult, jexiste currently does nto like me...
20090103 18:40:21< Rhonda> Leave it like it is.
20090103 18:42:30< Ivanovic> "does not like me" as in "the site ain't working atm
20090103 18:52:15-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["Changing server"]
20090103 18:53:07-!- busfahrer [n=busfahre@ulmg-5f70cdf7.pool.einsundeins.de] has joined #wesnoth-dev
20090103 18:53:38-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit [Client Quit]
20090103 18:54:00-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090103 19:02:05< CIA-52> mordante * r31930 /trunk/ (changelog src/text.cpp): Fix text render issue on big endian machines.
20090103 19:10:04< zookeeper> oh dear. EI has some serious schedule breakage.
20090103 19:13:17-!- Elvish_Pillage2 [n=eli@24-177-37-183.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090103 19:14:43< zookeeper> ESR_, pling pling. r30297 removed {DEFAULT_SCHEDULE}s from (at least some) EI scenarios and thus naturally breaking the schedules.
20090103 19:15:24< zookeeper> whatever method you used to do those playlist tweaks, i suggest you go over your other similar commits and check them..
20090103 19:18:54< zookeeper> currently at least the fifth scenario is missing the schedule entirely.
20090103 19:19:31-!- kitty_ [n=kitty@e180200017.adsl.alicedsl.de] has joined #wesnoth-dev
20090103 19:20:57-!- Psyche^ [n=Psyche@e177114241.adsl.alicedsl.de] has joined #wesnoth-dev
20090103 19:28:47-!- Patterner [n=Psyche@f054000073.adsl.alicedsl.de] has quit [Read error: 145 (Connection timed out)]
20090103 19:28:48-!- Psyche^ is now known as Patterner
20090103 19:29:09-!- Elvish_Pillager [n=eli@24-177-37-183.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090103 19:29:16< zookeeper> it's pretty infuriating when psp x can do batch processing, but you just can't run the png optimizer that way -.-
20090103 19:38:41-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090103 19:47:06< CIA-52> zookeeper * r31931 /trunk/data/core/images/portraits/ (21 files in 4 dirs): A bunch of background removals.
20090103 19:51:12-!- Baufo [n=thomas@62-47-152-202.adsl.highway.telekom.at] has joined #wesnoth-dev
20090103 19:54:15-!- Sirp_ [n=me@c-76-102-104-121.hsd1.ca.comcast.net] has joined #wesnoth-dev
20090103 19:54:36< Sirp_> Greetings.
20090103 19:54:46< mordante> hi Sirp_ 
20090103 19:57:20< zookeeper> kitty_, is the cavalryman fine? http://www.wesnoth.org/forum/download/file.php?id=26452
20090103 19:57:35< zookeeper> (or do you want him flipped)
20090103 19:58:13< kitty_> zookeeper_ everything posted by doofus in this thread is fine with me.
20090103 19:58:29< boucman> hey Sirp_
20090103 20:00:04< zookeeper> kitty_, ok
20090103 20:01:17< CIA-52> zookeeper * r31932 /trunk/data/core/images/portraits/ (5 files in 3 dirs): A couple more background removals.
20090103 20:11:28< zookeeper> kitty_, i'll edit your post's list of what's left, ok?
20090103 20:11:42< kitty_> ok, is it much?
20090103 20:12:18< zookeeper> err, actually only royal guard, spearman, swordsman, but those weren't on your list to begin with.
20090103 20:12:32< kitty_> huh, misunderstanding - you mean the forum post, right?
20090103 20:12:38< zookeeper> yes
20090103 20:12:53< kitty_> and the ones you mentioned have new lordbob portraits...
20090103 20:13:12< zookeeper> yeah, true enough, those won't need to be converted.
20090103 20:13:46< kitty_> then you have done all others? sounds like you did a lot of work!
20090103 20:14:09< zookeeper> i guess :p
20090103 20:14:30< zookeeper> but as i said, it's not really hard to do, so..
20090103 20:14:32< kitty_> what about the mainline campains? 
20090103 20:15:15< zookeeper> i've done a few already, but there's of course several still left
20090103 20:16:21< kitty_> which? i could do a bit right now...
20090103 20:17:13< zookeeper> eastern invasion, only has 5 portraits, but mal-ravanal's one might be a bit tricky so maybe you should do it ;)
20090103 20:17:57< zookeeper> gweddry's portrait needs a _little_ bit drawing for his helmet...thing
20090103 20:18:04< kitty_> and where is the official place to find them? don't want to make the same mistake as with the drakes...
20090103 20:18:41< zookeeper> http://svn.gna.org/viewcvs/wesnoth/trunk/data/campaigns/
20090103 20:19:20< zookeeper> i've done liberty and the south guard, but haven't committed either (i'll go over them again before i do)
20090103 20:19:59< zookeeper> UtBS is another one in need of doing, but i'm not sure if someone should just contact the artist and ask him for transparent versions since he might have some
20090103 20:20:11< kitty_> ok, so i'll do eastern invasion and which campaign is gweddry from?
20090103 20:20:18< zookeeper> that one
20090103 20:20:27< kitty_> oh.
20090103 20:21:03< zookeeper> should i PM jetryl for big transparent versions of the HttT portraits or will you?
20090103 20:22:48< kitty_> i can ask jetryl, and do you have transparent versions for SotbE, or should i ask spaceinvader?
20090103 20:22:55< zookeeper> i have them
20090103 20:23:30< zookeeper> hmm, i can handle sceptre of fire too, except for khrakrahs: http://svn.gna.org/viewcvs/*checkout*/wesnoth/trunk/data/campaigns/Sceptre_of_Fire/images/portraits/khrakrahs.png?rev=30533
20090103 20:23:56< kitty_> i'll do khrakahs
20090103 20:24:20-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev
20090103 20:24:20< zookeeper> thanks
20090103 20:24:21< kitty_> and the "someone" contacting the UtbS artist, is that you or me (if it's me - who is the artist?)
20090103 20:30:43< zookeeper> pickslide is
20090103 20:30:51< zookeeper> eh, i'll try to handle that
20090103 20:31:09< kitty_> is he still around? and thank you!
20090103 20:32:07< mordante> IIRC Pickslide also had larger versions of some of his images on his own site
20090103 20:32:15< mordante> (don't ask me the url ;-) )
20090103 20:34:08< kitty_> that would be really, really usefull... 
20090103 20:34:23< boucman> google might know
20090103 20:34:44< kitty_> not under pickslide
20090103 20:35:21< boucman> i think he used his nickname on his site..
20090103 20:35:42< zookeeper> https://webspace.utexas.edu/woojs1/Wesnoth/Pickslide.html
20090103 20:36:12< zookeeper> so yeah, actually using those would be a great idea.
20090103 20:36:20< kitty_> perfect!
20090103 20:36:31< mordante> https://webspace.utexas.edu/woojs1/Wesnoth/Pickslide.html
20090103 20:36:43< mordante> see I'm too late
20090103 20:37:03< zookeeper> bookmarks beat googling any day ;)
20090103 20:37:20< zookeeper> anyways, i'll go have some dinner ->
20090103 20:37:21< mordante> i didn't google ;-)
20090103 20:37:25< zookeeper> oh, ok
20090103 20:50:56< fendrin> hi
20090103 20:51:53< mordante> hi fendrin 
20090103 20:52:36-!- fnaek [n=fnaek@athedsl-4495008.home.otenet.gr] has joined #wesnoth-dev
20090103 20:53:26< fnaek> hi fendrin, I get a strange Dm error in 1.5.7 wesnoth: unit_map.cpp:68: std::pair<map_location, unit>* unit_map::unit_iterator::operator->() const: ? ?????????? valid()' ???????.
20090103 20:54:06-!- Smar [n=smar@freenet/translator/finnish/Smar] has quit [Read error: 104 (Connection reset by peer)]
20090103 20:55:27-!- Smar [n=smar@a88-113-60-192.elisa-laajakaista.fi] has joined #wesnoth-dev
20090103 20:55:42-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has quit []
20090103 20:57:35-!- fendrin_ [n=fabi@e176251131.adsl.alicedsl.de] has joined #wesnoth-dev
20090103 20:59:07< fnaek> hi fendrin
20090103 20:59:49-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev
20090103 21:00:08< zookeeper> kitty_, so maybe you should post in your thread that all of the listed ones are done now?
20090103 21:00:33< mordante> zookeeper kaleh is already done
20090103 21:01:05< kitty_> zookeeper: ok i'll do that. i really hate all those little shines and glows on these portraits - arg!
20090103 21:01:35< kitty_> mordante: by whom? you?
20090103 21:01:39< zookeeper> kitty_, yeah they're tricky
20090103 21:01:40-!- Jetrel [n=Jetrel@mn-10k-dhcp1-151.dsl.hickorytech.net] has joined #wesnoth-dev
20090103 21:01:50< zookeeper> ah, jetrel, just the guy we need
20090103 21:01:53< mordante> kitty_ now you're joking :-P
20090103 21:02:02< kitty_> mordante: then who did it?
20090103 21:02:04< Jetrel> zookeeper:  so I heard you're working on converting pickslide's portraits
20090103 21:02:07< mordante> no Jetrel did them
20090103 21:02:13< Jetrel> Did what?
20090103 21:02:17< boucman> hey Jetrel
20090103 21:02:18< mordante> kaleh
20090103 21:02:19< zookeeper> Jetrel, yeah
20090103 21:02:38< kitty_> hi jetrel
20090103 21:02:56< Jetrel> Yeah, where "did" means "stripped off the black matte".
20090103 21:03:21< kitty_> Jetrel: all of them or just kaleh?
20090103 21:03:28< Jetrel> Just kaleh
20090103 21:03:44< zookeeper> Jetrel, whenever you get the time, could you commit and replace the HttT portraits with bigger and transparent ones?
20090103 21:04:13< Jetrel> Sure.
20090103 21:04:35< Jetrel> I hope I still have them somewhere, but I'm pretty sure I do.
20090103 21:04:49< Jetrel> TBH I really want to replace those sometime soon.
20090103 21:04:58< zookeeper> you've been saying that for a long time ;)
20090103 21:05:26< boucman> Jetrel: http://www.wesnoth.org/forum/viewtopic.php?p=317755#p317755 <= this guy seems a decent spriter, do you think it's worth trying to get him to do some mainline stuff ?
20090103 21:05:28< Jetrel> :/  Yeah.
20090103 21:06:31< Jetrel> Boucman: possibly.  There's not much left to do in terms of new base sprites, but help with animations would always be nice.
20090103 21:06:57< Jetrel> (I am in close contact with neoriceisgood, and I can summon him up to finish the rest of the new drakes in short order.
20090103 21:06:57< boucman> thought it was woth pointing that to you...
20090103 21:07:00< noy> Jetrel: there is always a certain middle east faction that needs sprites
20090103 21:07:13< boucman> Jetrel: seriously ??? that's great !!!
20090103 21:07:20< Jetrel> noy:  if you want to ask him for help with that, go right ahead.
20090103 21:07:40< zookeeper> (there's still a lot of campaign units with crappy sprites too)
20090103 21:08:11< noy> Well I was secretly hoping you'd find some time to maybe start on some?
20090103 21:08:19< Jetrel> noy:  mainline first.
20090103 21:09:08< noy> ... I guess I'll never see them then
20090103 21:09:14< Jetrel> noy:  don't get me wrong, I am interested in having those, but I'm not touching them, or any of the other little additions I'd like to do, myself, till I'm done with the mainline sprites.
20090103 21:09:26< noy> think this guy will be any good?
20090103 21:09:37< boucman> Jetrel: maybe revamping campaign unit could be a good way to pull that guy toward mainline
20090103 21:09:40< ESR_> zookeeper: Re r30297, my apologies. I thought I was removing {VICTORY_AND_DEFEAT_MUSIC} instances only.  Do you want to fix it or shall I?
20090103 21:10:00< noy> I also don't want to ask him to do the art... then have you do them later
20090103 21:10:05< noy> you know?
20090103 21:10:58< Jetrel> noy: I do, but honestly, on a project like this, that's not a bad thing.  I probably wouldn't go around replacing his art if it was any good - I'd just be touching things up.
20090103 21:11:05-!- fnaek [n=fnaek@athedsl-4495008.home.otenet.gr] has quit ["using sirc version 2.211+KSIRC/1.3.12"]
20090103 21:11:15< noy> Okay... thats great
20090103 21:11:33< Jetrel> If it wasn't any good, then the worst that's happened to him is that he got some good practice.
20090103 21:11:57< zookeeper> ESR_, rather you fix it :]
20090103 21:12:16< ESR_> OK.
20090103 21:13:02-!- fendrin [n=fabi@g227034255.adsl.alicedsl.de] has quit [Read error: 110 (Connection timed out)]
20090103 21:13:06< ESR_> wesbot: revision 30297
20090103 21:13:08< zookeeper> and please make sure there aren't other cases like that, unless it was just some manual copypaste error you did
20090103 21:13:40< noy> Jetrel: would it be possible if you could critique his work if he starts doing this?
20090103 21:13:42< Ivanovic> wesbot: log r30297
20090103 21:13:45< wesbot> esr * r30297 : Custom playlists, and one geographical continuity tweak, for Eastern Invasion. 
20090103 21:13:48< wesbot> URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=30297
20090103 21:13:52< Jetrel> noy: sure
20090103 21:13:59< noy> K, thanks alot
20090103 21:14:00< ESR_> Ivanovic: Thanks.
20090103 21:14:14< mordante> now that the states are awake; in the MP dialog there's an option "Connect to Server" shouldn't that be "Connect to a Server"
20090103 21:14:25< Jetrel> zookeeper/kitty:  so if we're currently working on portrait conversions, assign me some of the ones from pickslide, and I'll get to work on them today.
20090103 21:14:47< kitty_> if you continue UtbS would be great
20090103 21:15:06< kitty_> i hate to do those little shines he adds everywhere...
20090103 21:15:36< mordante> kitty_ Jetrel zookeeper now that we're all here I would like to discuss the new dialogs a bit more, especially the scrollbars
20090103 21:16:06< kitty_> as i said yesterday - i think we need at least mockups of both options...
20090103 21:16:41< mordante> I'm making a screenshot of how it looks at the moment
20090103 21:17:36< Jetrel> kitty_:  Yeah, I know.  Doing Eloh is gonna get ugly..  :)
20090103 21:18:02< kitty_> Jetrel_ just completely repainting all fx...
20090103 21:20:55-!- fnaek [n=fnaek@athedsl-4495008.home.otenet.gr] has joined #wesnoth-dev
20090103 21:21:03-!- Sirp__ [n=me@c-76-102-104-121.hsd1.ca.comcast.net] has joined #wesnoth-dev
20090103 21:21:50< Ivanovic> mordante: since you are working on this stuff: have your gui changes also affected the story screens?
20090103 21:22:02< mordante> Ivanovic not yet
20090103 21:22:06< fnaek> anyone seen an error in 1.5.7:unit_map.cpp:68: std::pair<map_location, unit>* unit_map::unit_iterator::operator->() const
20090103 21:22:13< Ivanovic> mordante: that is: in 800x600 story screens really look ugly
20090103 21:22:30< Ivanovic> the border on the left and right without content is just too big and thus the text box gets huge
20090103 21:22:41-!- EdB [n=EdB@79.88.117.169] has quit ["Konversation terminated!"]
20090103 21:23:36 * Ivanovic wonders how this might look on even smaller resolutions
20090103 21:23:46< mordante> uglier ;-)
20090103 21:23:58< Ivanovic> starting two brothers in german in 800x600 is already "interesting"
20090103 21:24:14< Ivanovic> and you know what i mean when i say "interesting"...
20090103 21:24:15< Ivanovic> ;)
20090103 21:24:25< mordante> only the text is to short for a scrollbar :-(
20090103 21:24:41< mordante> I just ran it on 800x480 so I know how it looks ;-)
20090103 21:24:50< Ivanovic> :)
20090103 21:25:17< loonycyborg> fnaek: seems like noone did :) How do you repro it?
20090103 21:25:53< fnaek> loonycyborg: I was playin a DM scenario that played fine under 1.5.6 and got this crash
20090103 21:26:55< mordante> zookeeper kitty_ Jetrel this is how it looks if the text doesn't fit http://imagebin.ca/view/pOpbEd.html
20090103 21:29:02< Jetrel> Looks like it has a lot of buttons. :/
20090103 21:30:13< mordante> 1 button to close the dialog
20090103 21:30:17< kitty_> i'm not really convinced, cluttered...
20090103 21:30:31< boucman> instinctively I would expect the close button packed on the right side...
20090103 21:30:41< kitty_> the buuton plus the scrollbar
20090103 21:30:42< kitty_> and what if the text would just extend higher?
20090103 21:31:27< mordante> that would give us more space, but it still might not be enough
20090103 21:31:31< kitty_> boucman: if we have to have a button i would like it 	left-aligned with the textblock
20090103 21:31:43< mordante> UTBS has some really long texts
20090103 21:34:40< loonycyborg> fnaek: Does it happen if you restart the scenario?
20090103 21:39:51< fnaek> same thing, look at http://www.wesnoth.org/forum/viewtopic.php?f=4&t=23526
20090103 21:40:46< fnaek> I did not play this scenario from a replay; I played from 5 or so scenaria back
20090103 21:42:04< Jetrel> Ultimately, we'd need some sort of fail-over if there is too much text to display.
20090103 21:42:25< mordante> yes, that's my problem
20090103 21:42:27< Jetrel> The options I can see are:
20090103 21:42:39< mordante> and I like the way I did it, but I'm not artist
20090103 21:43:02< Jetrel> 1]  Just don't display the offending text.  Some campaign authors are liable to break this, though, and we'd be playing a constant game of cat and mouse.
20090103 21:44:09-!- Sirp_ [n=me@c-76-102-104-121.hsd1.ca.comcast.net] has quit [Read error: 110 (Connection timed out)]
20090103 21:44:27< Jetrel> 2] Automatically send "too much text" over to a second text screen.  Has the possible but unlikely problem of someone putting reams of text into a single dialogue, and creating hundreds of dialogues as a result.
20090103 21:44:31< Ivanovic> Jetrel: 1) is not possible when it comes to translations
20090103 21:44:40< Ivanovic> translations often take *lots* of extra space
20090103 21:44:55< Jetrel> 2]  it's assumed in the above that if it were too much for two screens, it would spill over to 3, 4, etc.
20090103 21:44:55< boucman> I like 2
20090103 21:45:06< kitty_> i don't understnd 2]
20090103 21:45:07< Ivanovic> that is: there you *have* to use this much space to say the same stuff, less is not possible since you have no option to break parts apart
20090103 21:45:11< boucman> we only have to pack a "next" and "previous" button
20090103 21:45:11< mordante> I see 1 also not as an option, I've been hit by it in UtbS and hated it
20090103 21:45:59< Jetrel> 3]  Do what we're currently doing, in having some sort of scrolling system on the individual dialogues.  Messy because it introduces buttons and scrollbars to an otherwise buttonless interface.
20090103 21:46:12< mordante> boucman and a close button
20090103 21:46:19< boucman> yeah,
20090103 21:46:30< boucman> well that's still three buttons, and probably on a single row
20090103 21:46:38< Ivanovic> 2) is not nice since it will be interesting for translators
20090103 21:46:46< Jetrel> kitty_:  Basically, if the text is too long, it automatically creates a second dialogue to display the extra text.
20090103 21:46:55< Ivanovic> some content creators would work on strings that are like whole novels
20090103 21:46:55< boucman> Ivanovic: not sure what you mean
20090103 21:47:04< Ivanovic> those strings are *hell* to work on when they get fuzzy
20090103 21:47:06< kitty_> ok, thanks for the explanation - sounds best to me
20090103 21:47:17< boucman> Ivanovic: I'm not too worried about that, 
20090103 21:47:23< mordante> also the text might also be in an option list and be too long, which need a scrollbar
20090103 21:47:29< boucman> most campaigns tend to be rather dry
20090103 21:47:30< mordante> needs*
20090103 21:47:44< Jetrel> Ivanovic:  it wouldn't break anything with their translations.  It's much like 
20090103 21:47:56< mordante> boucman did you play UtbS ;-)
20090103 21:47:56< Jetrel> Ivanovic:  having an irc sentence split across two lines.
20090103 21:48:04< boucman> nope
20090103 21:48:07< Ivanovic> boucman: have you had a look at the amount of strings in iftu?
20090103 21:48:12< boucman> I'm waaaay not good enough to play that
20090103 21:48:29< Jetrel> Ivanovic:  slightly awkward, but would only be invoked when there was a huge amount of text.
20090103 21:48:42< mordante> I also managed UtbS ;-)
20090103 21:48:42< Jetrel> Mostly because I'd also suggest doing it in conjuction with:
20090103 21:48:50< Ivanovic> Jetrel: it just depends on when it occurs
20090103 21:49:08< Ivanovic> and even more interesting: when can things be split apart?
20090103 21:49:20< Ivanovic> do we have our own rules or is it just "x chars"?
20090103 21:49:32< Jetrel> 4]  Grow the dialogues vertically to house a substantial amount of additional text.  Does have a ceiling on how far this can go.
20090103 21:49:39< Ivanovic> with "own rules" like "only complete sentences" are problematic for translations
20090103 21:49:54< boucman> maybe by introducing a "paragraph mark" and cutting between paragraphs when possible
20090103 21:50:03< boucman> paragraphs are kept by translators
20090103 21:50:10< Jetrel> Ivanovic:  Normally, you shoot for the nearest space which is underneath the max char count.
20090103 21:50:12< mordante> 4] is not an option, that's hoping the problem won't occur 
20090103 21:50:30< Jetrel> mordante:  I'm suggesting a combination of #4 and #2.
20090103 21:50:39< mordante> also I think a dialog over the entrire screen looks bad
20090103 21:50:39< Ivanovic> and the problem *will* occur on really small resolutions
20090103 21:50:46 * Ivanovic is thinking handheld again...
20090103 21:50:52< Jetrel> Mordante:  do #4 first, and if that's not enough, do #2 as well.
20090103 21:50:57< boucman> 2+4+recognise paragraphs might be a sound solution
20090103 21:51:36< mordante> I like to try 4 first to see whether or not it gets ugly
20090103 21:51:37< boucman> on handheld/low res, I think covering 90% of screen is actually a better idea than having 3 words per page
20090103 21:51:47< mordante> (I fear it will)
20090103 21:51:56< boucman> the the maxe % of screen a dialog should use might be a theme/resolution parameter
20090103 21:52:17< Ivanovic> yes, that is correct
20090103 21:54:18< Jetrel> kitty_:  btw, I just realized my "quick and dirty" method of doing the transparency, although it looked fine on most backgrounds, looked bad on a white background, so I'm cleaning up the transparency on the kaleh portrait.  Should be done in a little bit.
20090103 21:55:08< noy> I messaged him Jetrel. Thanks for your help.
20090103 21:55:20< kitty_> Jetrel: but they won't ever be used on white... just on wesnoth terrain. or?
20090103 21:55:34< Jetrel> kitty_:  unfortunately, I usually do my photoshop work with the transparency set to about a neutral gray, to keep colors/brightnesses from it from interfering.  However, white seems to be better for doing this kind of masking - there were a lot of things I missed on the previous one.
20090103 21:56:03< Jetrel> kitty_:  -probably-.  However, these could easily get used on some webpage.
20090103 21:56:03< kitty_> bad luck...
20090103 21:57:07< Jetrel> Jagged, ugly job that took 5min:  http://www.allacrost.org/staff/user/jetryl/Kaleh.png
20090103 21:57:12< kitty_> zookeeper: i did the EI portraits and the dragon http://www.wesnoth.org/forum/viewtopic.php?f=9&p=329700#p329700
20090103 21:57:31< Jetrel> I'm cleaning it up by hand right now... almost done.
20090103 21:57:50< kitty_> there is a line between the hand and the body at the bottom, and perhaps the hair?
20090103 21:57:54< fnaek> loonycyborg, do you have any idea what thic could be due?
20090103 21:58:14< Jetrel> Yeah.
20090103 22:02:36< kitty_> and Jetrel, do you have a comment on the troll whelp i did? http://www.wesnoth.org/forum/viewtopic.php?f=9&t=23420&start=15 there was quite a bit of discussion if he was wenothian...
20090103 22:03:57< Jetrel> kitty_:  other than somewhat strong/black linework, he looks exactly like I'd imagined wesnoth's trolls.
20090103 22:04:23< Jetrel> And I really like how you make him look "rock-like".
20090103 22:04:34< kitty_> ok, i'll tweak the lines a bit! thanks :)
20090103 22:05:03< Jetrel> In fact, that comment was coming from one of him higher-up on the page...
20090103 22:05:57-!- ettin [n=jorda@103.207.219.87.dynamic.jazztel.es] has quit [Read error: 110 (Connection timed out)]
20090103 22:06:11< Jetrel> It's probably not as necessary for the one down below. :)
20090103 22:06:41< kitty_> i'll still do the tweak to the shoulder girgistan suggested. 
20090103 22:06:51 * Jetrel nods
20090103 22:07:47< Jetrel> kitty_:  reload to see the cleaned-up version:  http://www.allacrost.org/staff/user/jetryl/Kaleh.png
20090103 22:08:54-!- fnaek [n=fnaek@athedsl-4495008.home.otenet.gr] has quit ["using sirc version 2.211+KSIRC/1.3.12"]
20090103 22:08:54< kitty_> looks good - but there is still the black line at the bottom between his right hand and the body
20090103 22:10:05< Jetrel> No, that isn't there in the redone version.
20090103 22:10:24< Jetrel> Unless it's something different from what I'd seen.
20090103 22:10:35< Jetrel> Try holding down shift, and clicking the reload button.
20090103 22:10:51< Jetrel> (In most browsers, this does a "hard reload")
20090103 22:11:07< Shadow_Master> hi
20090103 22:11:07< kitty_> sorry, you're right - he looks great
20090103 22:11:23< kitty_> hi Shadow_Master
20090103 22:11:28< Jetrel> mordante:  so, uh, where should we be committing these?
20090103 22:13:05< kitty_> Jetrel: not that i know anything about commiting, but as far as i understood zookeeper wanted to replace the old ones directly.
20090103 22:13:08< Shadow_Master> argh... I'0ll switch to 16 bpp :(
20090103 22:13:41< Shadow_Master> brb
20090103 22:13:53< boucman> Jetrel: currently we have a "transparent" subdirectory within the portrait directory and we put them there...
20090103 22:14:06< boucman> however you also need a bit on the WML side to have the unit use it
20090103 22:14:07< Shadow_Master> back
20090103 22:14:11< mordante> yes I think zookeeper replaced the old images, not sure whether he also made modifications to the wml
20090103 22:14:17< boucman> look at any drake to see how it's done...
20090103 22:14:42< Jetrel> >_> hah.
20090103 22:15:40< Jetrel> Zookeeper's suggestion does make sense, since we should, programmatically, be able to generate equivalents of the old "on black" portraits from scratch.   E.g. we no longer need them at all.
20090103 22:16:44< boucman> Jetrel: it's not worth writing the code to do that, simpler to keep both portraits for the time being
20090103 22:17:19< Shadow_Master> ah, for onrce nobody needs anything for me
20090103 22:17:22< Shadow_Master> *from
20090103 22:17:30< boucman> Shadow_Master: hehe
20090103 22:18:06 * mordante thinks Shadow_Master tempts fate ;-)
20090103 22:18:28< Ivanovic> Shadow_Master: uhm, you could...
20090103 22:18:30< Ivanovic> ;)
20090103 22:19:49< Jetrel> laff.
20090103 22:20:25< Shadow_Master> I'd recommend you to not try to ask me anything until March...
20090103 22:20:33< Shadow_Master> at leaswt not anything that is critical for 1.6
20090103 22:20:39< boucman> exams ?
20090103 22:20:50< Ivanovic> boucman: summer in chile
20090103 22:21:44< Jetrel> Ivanovic:  oh we can fix that real fast.  All we have to do is alter the earth's axial tilt and ....  you know. :)
20090103 22:22:41< boucman> if we all jump at the same time :)
20090103 22:22:47< Shadow_Master> yeah, and the stray wifi signal is basically unavailable to me now
20090103 22:22:47< Shadow_Master> so I can only wait for march when i'll be in university
20090103 22:23:11< boucman> Shadow_Master: you can code without wifi ;)
20090103 22:23:22< loonycyborg> Shadow_Master: Does your new laptop have a good graphics card? :)
20090103 22:23:29< Shadow_Master> loonycyborg: an ATI Radeon HD.
20090103 22:23:37< Shadow_Master> it doesn't work wiell with the free "radeon"
20090103 22:24:08< Ivanovic> nice (since there are some 3d specs for the latests cards now, too and open drivers are already available in an extra branch in the freedesktop git)
20090103 22:24:14< Ivanovic> Shadow_Master: try radeonhd
20090103 22:24:22< loonycyborg> Still it's better than most intel cards which are used in laptops often..
20090103 22:24:24< Shadow_Master> and doesn't worth at all with the free "radeonhd". The free "radeon" shows me things at 16 bpp eeven on 32 bpp mode, and the non-free fglrx requires a kernel module that won't compile on 2.6.28
20090103 22:24:37< Ivanovic> though it will take some time until the 3d support and acceleration stuff is available in master for your card
20090103 22:24:39< Shadow_Master> and the free "radeonhd" just crashes
20090103 22:24:40< Ivanovic> might be until match
20090103 22:24:44< Ivanovic> s/match/march
20090103 22:24:51< Shadow_Master> and it's a kernel ccrash, unfortunately
20090103 22:25:13< Ivanovic> which version of radeonhd?
20090103 22:25:14< loonycyborg> Though intel cards have relatively stable drivers..
20090103 22:25:25< Shadow_Master> lenny's latest? :)
20090103 22:25:32< Ivanovic> i was using radeonhd for a while and it was working really well, even better than plain radeon
20090103 22:25:42< Ivanovic> that is: i was of course using git head
20090103 22:27:10< Ivanovic> loonycyborg: yeah, the drivers for intel are stable and well working
20090103 22:27:16< Ivanovic> (got such a card in my tablet)
20090103 22:27:23< Ivanovic> but they are not too fast
20090103 22:30:03< Shadow_Master> lenny's is radeonhd "1.2.1"
20090103 22:30:22< Shadow_Master> but I've got no idea is that is the real numbering used by the x.org project for it?
20090103 22:30:24< Ivanovic> Shadow_Master: that is *ancient*
20090103 22:30:28< Shadow_Master> ooh
20090103 22:30:38< Ivanovic> 1.2.1 is 9 month old
20090103 22:30:52< Ivanovic> the last tag was 1.2.4 - 3 weeks ago
20090103 22:31:04< Shadow_Master> ow,I'll pull from git then
20090103 22:31:18< Ivanovic> yes, do so, this one really works well at least for me
20090103 22:31:22< Shadow_Master> (assuming it is possible to get it as a separate component?)
20090103 22:31:50< Ivanovic> git://anongit.freedesktop.org/git/xorg/driver/xf86-video-radeonhd
20090103 22:31:54< Ivanovic> branch to use is master
20090103 22:31:59< Shadow_Master> sure
20090103 22:33:26< Shadow_Master> um, I have figured a new image path that would be nice to have
20090103 22:33:32< Shadow_Master> * image path function
20090103 22:34:53< kitty_> good night
20090103 22:34:55-!- kitty_ [n=kitty@e180200017.adsl.alicedsl.de] has quit []
20090103 22:35:18< Shadow_Master> oh my god, compiling wesnoth on this laptop is *nice*
20090103 22:35:28< Shadow_Master> less than 10 minutes at -O3 if I use two tasks at a time
20090103 22:35:40< Shadow_Master> and that for a full recompile..
20090103 22:36:47< Ivanovic> Shadow_Master: use "-j" to make it be really fast
20090103 22:36:58< Ivanovic> but better use some nicelevel...
20090103 22:37:00< Shadow_Master> I said "if I use two tasks at a time"
20090103 22:37:04< Shadow_Master> e.g. -j2
20090103 22:37:18< Ivanovic> yes, use "-j" without parameter
20090103 22:37:24< Ivanovic> then it will use as many threads as it can
20090103 22:37:43< Ivanovic> sure to produce full load on probably any machine when using this for the kernel
20090103 22:37:47< boucman> Shadow_Master: submerge
20090103 22:37:56< Shadow_Master> boucman: no.
20090103 22:38:00< boucman> why not ?
20090103 22:38:13< Ivanovic> beside this: general rule: use "#CPUs +1" threads for compiling
20090103 22:38:21< Shadow_Master> boucman: because it isn't what I was thinking of
20090103 22:38:26< Shadow_Master> ~HUE()
20090103 22:38:27< boucman> k
20090103 22:38:39< Shadow_Master> or ~TINT(), depends
20090103 22:39:14< boucman> do we already have an ~ALPHA() or similar
20090103 22:39:15< boucman> ?
20090103 22:39:58< Shadow_Master> boucman:  ~O()
20090103 22:40:03< Shadow_Master> I implemented that and it is in 1.5.6
20090103 22:40:03< boucman> k
20090103 22:40:10< CIA-52> mordante * r31933 /trunk/src/gui/widgets/text.hpp: Fix a comment copy paste error.
20090103 22:40:14< CIA-52> mordante * r31934 /trunk/src/gui/widgets/ (scrollbar_container.cpp scrollbar_container.hpp): 
20090103 22:40:14< CIA-52> Restored basic container scrolling behaviour.
20090103 22:40:14< CIA-52> All containers (scroll label and listboxes can scroll again). The arrows
20090103 22:40:14< CIA-52> don't select yet in a listbox. Also with the new scrollbars the up and
20090103 22:40:14< CIA-52> down arrow scroll one pixel at the time, which might need to be
20090103 22:40:15< CIA-52> increased a little bit.
20090103 22:40:19< CIA-52> mordante * r31935 /trunk/src/gui/dialogs/message.cpp: 
20090103 22:40:21< CIA-52> Give the label in the message the default focus.
20090103 22:40:23< CIA-52> This enables scrolling when the message is shown.
20090103 22:40:50< boucman> mordante: can you scroll with the mouse wheel ?
20090103 22:41:31< mordante> no and I don't have the hardware to test, but I think it would be easy to add
20090103 22:41:49< Jetrel> zookeeper: there?
20090103 22:42:04< mordante> does the current implementation allow the scrollwheel ?
20090103 22:46:25< Ivanovic> when currently using wesnoth the mousewheel moves the map
20090103 22:46:36< Ivanovic> so that you scroll in vertical direction
20090103 22:46:44< mordante> and in listboxes?
20090103 22:47:17< Ivanovic> no idea
20090103 22:47:21< Ivanovic> where are list boxes?
20090103 22:47:25< Ivanovic> addon server dialog?
20090103 22:47:56< mordante> the addon download list
20090103 22:48:20< mordante> that uses the old widgets, I know the new ones don't support it
20090103 22:48:26< Ivanovic> when with the mouse in the addonlist (really the cursor above the addon entries, not above the buttons outside the list or outside the widget
20090103 22:48:37< Ivanovic> then you can scroll up/down with the mouse wheel
20090103 22:49:18-!- fnaek [n=fnaek@athedsl-4506563.home.otenet.gr] has joined #wesnoth-dev
20090103 22:49:25< mordante> oke then it's a regression and more important
20090103 22:49:41< Ivanovic> :)
20090103 22:53:18< Ivanovic> mordante: so you got no external mouse with a scrollwheel somewhere in your house?
20090103 22:54:07< mordante> I might have one, but I don't expect it
20090103 22:54:14 * Ivanovic can even "mouse scroll" on his tablet since he got three buttons there and when clicking the middle button and using the strange "pin" you mouse scroll...
20090103 22:54:43< mordante> of course I have a lot of (ancient) hardware so there might be one
20090103 22:55:14< Ivanovic> :)
20090103 22:55:26< Ivanovic> you should search a little and maybe you find one
20090103 22:55:43< mordante> when I scroll my wrist starts to hurt pretty soon, so rather have a normal three button mouse
20090103 22:55:47< Ivanovic> this would help a lot when working on new elements since i'd guess that many users got a "mouse wheel" somewhere
20090103 22:56:28< Ivanovic> uhm, you still find mice without scroll wheel?
20090103 22:56:33< Ivanovic> wow, this is suprising
20090103 22:56:59< mordante> I fear that I'm in bad luck when I buy my next computer :-(
20090103 22:57:22< mordante> especially since my current mouse is PS/2
20090103 22:57:38-!- fnaek [n=fnaek@athedsl-4506563.home.otenet.gr] has left #wesnoth-dev []
20090103 22:58:04< mordante> of course I can buy one with scrollwheel and disable the wheel in X (so I won't scroll with it ;-) )
20090103 22:58:08-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090103 22:58:23< Ivanovic> or you can just not use it for scrolling while still having it active
20090103 22:58:27< Ivanovic> ;)
20090103 22:58:34< Ivanovic> [22:56:59] <mordante> I fear that I'm in bad luck when I buy my next computer :-(
20090103 22:58:35< Ivanovic> [22:57:22] <mordante> especially since my current mouse is PS/2
20090103 22:58:40< Ivanovic> ups...
20090103 22:58:52< mordante> ?
20090103 22:59:03< Shadow_Master> what does the mouse protocol have to do with scrolling?
20090103 22:59:11< Shadow_Master> or you just don't own a mouse with scroll button?
20090103 22:59:42< mordante> nothing just that I think it will be hard to find a new computer which still has a PS/2 connection
20090103 22:59:47< zookeeper> Jetrel, yeah, here now
20090103 22:59:51 * zookeeper is reading the log
20090103 22:59:52< Shadow_Master> um, desktop boxes do have AFAIK
20090103 23:00:07< Shadow_Master> laptops don't, but mouse are really cheap nowadays so it shouldn'be an issue
20090103 23:00:14< Shadow_Master> (I even got a free one with the new laptop)
20090103 23:00:35< mordante> try to find one without a scrollwheel ;-)
20090103 23:00:36< Jetrel> zookeeper:  So basically, I'm working on making transparent versions of the UTBS portraits, and I'm wondering where to commit them.
20090103 23:01:26< Shadow_Master> Jetrel: probably the resources branch
20090103 23:02:23< Jetrel> Shadow_Master:  Might not be a bad place to stow them till other folks get their shit together.  :]
20090103 23:03:14< Ivanovic> Jetrel: all other portraits seem to be in some transparent/ folder from the path of the normal portraits
20090103 23:03:25< zookeeper> my suggestion on the scrollbars etc: make the dialogs simply expand vertically to include all the text just like the old ones did, and let's see how it works out. my guess is that it'll work out just fine.
20090103 23:04:28< zookeeper> after all, it has worked just fine with the current ones, and it's exceedingly rare for anyone to have so much text in one dialog that the dialog would expand to be higher than a 205x205 portrait.
20090103 23:04:46< zookeeper> i mean, that the lower edge would expand more than what the portrait makes it expand.
20090103 23:05:08< CIA-52> mordante * r31937 /trunk/src/gui/widgets/ (5 files): 
20090103 23:05:08< CIA-52> Let the arrow key select items in the listbox.
20090103 23:05:08< CIA-52> Note when an item outside the current view is selected the listbox
20090103 23:05:08< CIA-52> doesn't scroll yet.
20090103 23:05:12< CIA-52> mordante * r31936 /trunk/src/gui/dialogs/message.cpp: Remove an obsolete todo item.
20090103 23:05:55< mordante> I just looked at over 200 mice all with at least one wheel :-(
20090103 23:06:24< Jetrel> zookeeper:  thanks.  I totally agree with you.  We're having a bit of analysis paralysis with this stuff.
20090103 23:06:31< mordante> zookeeper how high do you want to expand?
20090103 23:06:55< Ivanovic> mordante: i got a (now ancient) logitech mx500 and it is really nice if you are right handed
20090103 23:06:58< zookeeper> and yes, just replace the old portraits with the new ones.
20090103 23:08:22< mordante> Jetrel I think that the case where the text doesn't fit is rare but we should be able to handle it
20090103 23:08:24< Ivanovic> that is: the form is really great, i love the side buttons and the "normal" buttons (left/right as well as whell(inkl button)) work nicely, too
20090103 23:08:47< zookeeper> mordante, as high as necessary?
20090103 23:08:54< zookeeper> i don't see any point in putting a maximum to that
20090103 23:09:02< zookeeper> (besides the screen height obviously)
20090103 23:09:35< mordante> it currently does best height with a maximum, but that can be removed
20090103 23:09:56< Jetrel> mordante:  then I'd do the "spill over to additional window" if that ever happened, but tbh, if it does, it's almost something that whoever was responsible for making too much text should have their hands swatted for.
20090103 23:10:06-!- IAmNotShadowMast [n=ignacior@190.196.28.74] has joined #wesnoth-dev
20090103 23:10:21< mordante> I can change the maximum to be the screenheight minus the menubar
20090103 23:10:32-!- IAmNotShadowMast [n=ignacior@190.196.28.74] has quit [Client Quit]
20090103 23:10:35< zookeeper> currently people make dialogs that cover half the screen. you know, a stats/help/inventory screen in a special scenario or whatever.
20090103 23:11:28< zookeeper> supporting scrollbars is great, but i don't see a reason for using those except when the dialog would really go off the screen
20090103 23:12:05< mordante> the idea is also to only use the scrollbars if needed
20090103 23:12:14-!- NotShadowMaster [n=ignacior@190.196.28.74] has joined #wesnoth-dev
20090103 23:12:17-!- Shadow_Master is now known as irmorelle
20090103 23:12:28-!- NotShadowMaster is now known as Shadow_Master
20090103 23:12:43-!- Shadow_Master [n=ignacior@190.196.28.74] has quit [Client Quit]
20090103 23:13:11-!- irmorelle is now known as Shadow_Master
20090103 23:13:20< mordante> Jetrel if we allow the height to be the entire dialog, I expect the overflow to be not that often, which makes me wonder whether it's worth the effort to split it in multiple screens
20090103 23:13:42< Ivanovic> no, make it just like in your example
20090103 23:13:43< mordante> I see a can of worms with splitting the text in multiple windows
20090103 23:13:46< Jetrel> mordante:  my thoughts exactly.
20090103 23:14:20< mordante> so in that case I prefer to keep the current version with scrollbars if needed and increase the maximum height
20090103 23:14:31 * Jetrel nods
20090103 23:15:02< zookeeper> ok, so...since i had to skip a bit of that discussion of yours, what does this "scrollbars if needed" mean exactly?
20090103 23:15:06< zookeeper> needed when..?
20090103 23:15:14-!- ettin [n=jorda@174.206.219.87.dynamic.jazztel.es] has joined #wesnoth-dev
20090103 23:15:17< mordante> when the text doesn't fit in the dialog
20090103 23:15:29< Ivanovic> zookeeper: when even more like the full screen would be used
20090103 23:15:37 * Jetrel nods
20090103 23:15:37< zookeeper> meaning that the dialog fills basically the whole screen? ok, great in that case
20090103 23:15:53< Ivanovic> like it might happen on really small res like 400x300 (something handheld like or such)
20090103 23:16:06 * Shadow_Master killing the X server to try radeonhd
20090103 23:16:12< Jetrel> But yeah, I strongly side with zookeeper that we should just go ahead and let dialogues go all the way up to filling the screen.
20090103 23:16:26< mordante> yes I'll change the code that the fills the screen until the menubar and if that doesn't fit it falls back to scrollbars
20090103 23:17:02< mordante> and the button to dismiss the dialog
20090103 23:17:31 * crimson_penguin thinks using the whole screen when there's that much text is fine
20090103 23:18:14< Ivanovic> mordante: i'd say that at the top the same space between bar and dialog box should be left as on the buttom to the window border
20090103 23:18:30< Ivanovic> i think there are 4px or something like this of the map shown below the dialog, right?
20090103 23:18:31< Shadow_Master> Ivanovic: pah, latest radeonhd has the same issue
20090103 23:18:44 * Jetrel is going afk for about 30min.
20090103 23:18:47< Shadow_Master> er, I mean. I tdoesn't crash. But it doesn't display at 32bpp either
20090103 23:19:10< Ivanovic> no idea what your problems might be then
20090103 23:19:14< mordante> Ivanovic yes might be, I'll have to see, will need to modify the code first
20090103 23:19:14< Shadow_Master> er, I mean... it crashes
20090103 23:19:16< Ivanovic> beside this: uhm, why 32bit?
20090103 23:19:28< Shadow_Master> the x.org server wasnot reloading the config
20090103 23:19:28< Ivanovic> just use 24bit, that is the correct value!
20090103 23:19:39< Shadow_Master> Ivanovic: 24bit is what I meant all thje time
20090103 23:19:57< Ivanovic> should be possible
20090103 23:20:05< Shadow_Master> xdpyinfo says the server is at 24bpp with the free "radeon", although for my eyes it isn't
20090103 23:20:08< Ivanovic> though i never forced a default, had a rather bare xorg.conf file
20090103 23:20:22< Ivanovic> and things did look well
20090103 23:20:23< Shadow_Master> the current "radeonhd" just crashes. And yeah, I do have a bare xorg.conf mostly
20090103 23:20:52< Shadow_Master> uh,. it seems it just "isn't there"
20090103 23:21:27< Ivanovic> this is basically my config: http://pastebin.com/m7bd74b35
20090103 23:21:36< Ivanovic> and this was what i used for radeonhd, too
20090103 23:21:58< Ivanovic> though Option "XVideo" "Enable" and Option "Composite" "Disable" got no effect with it
20090103 23:23:20< Ivanovic> and yeah, everything looked really nicely with this though, due to missing video/2d/3d acceleration, it was not too fast
20090103 23:23:37< Ivanovic> normal browsing and working was no problem, but watching a video in fullscreen was impossible
20090103 23:23:52< Ivanovic> (all rendering via cpu really kills the system...)
20090103 23:23:57< zookeeper> away for 20 minutes..
20090103 23:25:01< Shadow_Master> Ivanovic: ah solved it
20090103 23:25:15< Ivanovic> what was the problem?
20090103 23:25:34< Shadow_Master> the server wasn't finidng to .so 
20090103 23:25:37< Shadow_Master> *finding the
20090103 23:25:50< Shadow_Master> it works like a charm now
20090103 23:25:53< Shadow_Master> bye..
20090103 23:25:54< Ivanovic> :)
20090103 23:28:40-!- Elvish_Pillager [n=eli@24-177-37-183.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090103 23:30:21-!- Sapient [n=sapien-x@c-24-126-133-155.hsd1.ga.comcast.net] has joined #wesnoth-dev
20090103 23:43:11< Sapient> hmm... for al my .gz savefiles, the Load Game dialog shows "(Invalid)"
20090103 23:43:40< Sapient> but then if I extract it, the uncompressed version works
20090103 23:44:06-!- Elvish_Pillage2 [n=eli@24-177-37-183.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090103 23:46:02< mordante> odd and do you use wesnoth to extract the .gz files?
20090103 23:46:54-!- Sapien-X99 [n=sapien-x@c-24-126-133-155.hsd1.ga.comcast.net] has joined #wesnoth-dev
20090103 23:46:58-!- Sapient [n=sapien-x@c-24-126-133-155.hsd1.ga.comcast.net] has quit [Nick collision from services.]
20090103 23:47:00-!- Sapien-X99 is now known as Sapient
20090103 23:47:16< mordante> odd and do you use wesnoth to extract the .gz files?
20090103 23:47:27< Sapient> no I used 7zip
20090103 23:48:07< Sapient> and whenever I press "Get addons" it says "An error occurred while communicating with the server"
20090103 23:48:14< mordante> can you try with wesnoth, maybe the decoder has a problem
20090103 23:49:17< Sapient> I am getting a lot of handled exceptions in boost::iostreams::basic_gzip_decompressor
20090103 23:50:00< mordante> that explains the savegame issue
20090103 23:50:02< Sapient> #1  0x00599461 in read_save_file (name=@0x22f188, cfg=@0x22f0c8,
20090103 23:50:02< Sapient>     error_log=0x22f184) at C:\projects\wesnoth\wesnothd\src\gamestatus.cpp:853
20090103 23:50:16< mordante> since when is it broken?
20090103 23:50:25< Sapient> The file you have tried to load is corrupt: "
20090103 23:50:41-!- boucman [n=rosen@159.83.102-84.rev.gaoland.net] has quit [Remote closed the connection]
20090103 23:50:59< Sapient> dunno. I have been getting these exceptions since I first started using gdb
20090103 23:51:31< Sapient> I was just ignoring them
20090103 23:51:44< mordante> I meant since when do savegames no longer load?
20090103 23:52:05< Sapient> dunno. today was the first time I tried
20090103 23:52:20< Sapient> I was trying to test santi's bug
20090103 23:52:51< Sapient> this is probably a problem with my boost setup
20090103 23:52:54< loonycyborg> Sapient: I remember getting similar errors when loading other people's saves.
20090103 23:53:24< Sapient> I haven't been able to connect with trunk to the MP server as long as we have been using boost
20090103 23:53:29< loonycyborg> I resorted to manual unpacking too.
20090103 23:54:15< loonycyborg> Sapient: Does it happen with games you save yourself?
20090103 23:54:26< Sapient> let me check
20090103 23:54:48< mordante> does this problem also occur with the 'offical' windows version?
20090103 23:55:10< Sapient> I haven't downloaded it
20090103 23:55:22< loonycyborg> I don't remember whether I've got this on linux or on windows..
20090103 23:55:28< Sapient> probably we would be getting a lot of reports if so.
20090103 23:57:17< Sapient> yeah, I can't load my own saves either
20090103 23:58:04< Sapient> I am really starting to hate boost
20090103 23:58:47< loonycyborg> Sapient: Your hate is misplaced. You should hate windows :)
20090103 23:59:19< Sapient> I don't think you can blame Bill Gates for this one
20090103 23:59:37< Sapient> boost is so freaking hard to configure correctly
--- Log closed Sun Jan 04 00:00:32 2009