--- Log opened Wed Jun 29 00:00:06 2011
--- Day changed Wed Jun 29 2011
20110629 00:00:06< tschmitz> gabba: which meant it needed knowledge of every subclass
20110629 00:00:16< gabba> tschmitz: oh yeah, its a static "factory method", so you can't override in subclasses, I see
20110629 00:01:25< tschmitz> gabba: Yeah factory
20110629 00:01:51< tschmitz> gabba: Do you think there could be a better way to deserialize them?
20110629 00:03:21-!- stikonas__ [~and@ctv-79-132-175-26.vinita.lt] has joined #wesnoth-dev
20110629 00:03:22< gabba> tschmitz: not at the moment
20110629 00:04:06< gabba> tschmitz: I'm just watchful of potential circular dependencies problems later on and/or slower builds because any file that imports action.hpp imports all this other stuff with it
20110629 00:04:31-!- Vorpal [~AnMaster@unaffiliated/anmaster] has quit [Ping timeout: 250 seconds]
20110629 00:04:46< tschmitz> gabba: Gotcha
20110629 00:04:48< gabba> but it should be fine, I'll let you know if I think of a better solution
20110629 00:06:09-!- stikonas_ [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 264 seconds]
20110629 00:06:34< tschmitz> gabba: wait
20110629 00:06:51< tschmitz> action.hpp doesn't include the subclasses' headers
20110629 00:06:58< tschmitz> gabba: Only action.cpp does that
20110629 00:07:28< gabba> tschmitz: true
20110629 00:07:33< tschmitz> gabba: So we're good, right?
20110629 00:07:47< gabba> tschmitz: ok it's not bad then
20110629 00:08:24< gabba> tschmitz: ok another thing
20110629 00:08:57< gabba> did you make sure you're correctly initializing the shared pointers at "return action_ptr(new move(cfg));" and similar lines?
20110629 00:09:18< gabba> I don't remember the correct syntax but there were some pitfall there
20110629 00:10:21< tschmitz> gabba: So you think I should compare it with the syntax you used in other parts of the code?
20110629 00:10:25< tschmitz> gabba: Let me look
20110629 00:11:17< gabba> tschmitz: at the minimum, and preferably go read up a bit on how boost wants you to initialize shared pointers
20110629 00:12:11< tschmitz> gabba: Yeah that seems easier, I'll Google it real quick
20110629 00:12:40< gabba> tschmitz: well, do both :)
20110629 00:12:50< gabba> I think we're limited to boost 1.35 syntax
20110629 00:13:25< tschmitz> gabba: OK; are you specifically worrying about the initialization from a subclass pointer?
20110629 00:13:44< gabba> (You probably know which version of boost you have in your project dependencies)
20110629 00:14:13< gabba> tschmitz: not particularly, no
20110629 00:14:13< tschmitz> gabba: The libraries that came in the "external" folder are (if memory serves?) 1.42
20110629 00:14:26< gabba> really? hmm
20110629 00:14:57< tschmitz> gabba: Actually, they appear to be 1_46
20110629 00:15:23< gabba> I think the server is compiled for a specific version of Debian, so we use the boost version that's on there
20110629 00:15:45< gabba> Ask mordante if you run into trouble there, I think it's him that told me that last year
20110629 00:16:09< gabba> We probably are aiming at a more recent Boost this year
20110629 00:16:59< gabba> It's usually backwards-compatible, but what you have to watch for is not to use newer syntax... you'll know it soon enough when other people can't compile
20110629 00:17:13< zaroth_> gabba: true enough
20110629 00:17:23< zaroth_> there was a change in program options 1.42
20110629 00:17:41< gabba> tschmitz: Anyways, the shared pointer syntax looks rather stable, I'm just telling you this as general advice
20110629 00:17:47-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Quit: Leaving.]
20110629 00:17:54< zaroth_> and I thought "well, in most recent ubuntu and the previous one it's already 1.42... it will be fine"
20110629 00:17:58< gabba> zaroth_: would you happen to know what's our target boost version?
20110629 00:18:24< zaroth_> turns out, somebody complained on forums about 5 days later - he was using Ubuntu LTS with boost 1.40 or something ;-)
20110629 00:19:04< zookeeper> shadowmaster, it was the thread you told him to create for his WML questions D:
20110629 00:19:15-!- deekay [~dk@wesnoth/developer/dragonking] has quit [Quit: deekay]
20110629 00:19:26< zaroth_> gabba: I would say that we should go for 1.42, since it's in debian stable
20110629 00:19:41< gabba> Ok that sounds right
20110629 00:19:42< zaroth_> I have no authority over that, though
20110629 00:20:37< gabba> Wow Ubuntu LTS behind Debian Stable? They should coordinate releases or something.
20110629 00:20:42< zaroth_> (and I did add an #ifdef for the poor user with Ubuntu LTS for boost <1.42 , since it was just one line with incompatible syntax)
20110629 00:21:03< gabba> awww that's kind of you :D
20110629 00:21:39< zaroth_> ;-)
20110629 00:22:56< tschmitz> gabba: OK so I just tested my new code for 5 minutes and it appears to be working fine
20110629 00:23:09< shadowmaster> zookeeper: I thought he already had a WML topic
20110629 00:23:27< gabba> tschmitz:  "@todo make this resistant to cheaters" :P
20110629 00:23:45< tschmitz> gabba: I tried making an undo-able move and then planning a subsequent action for that unit also
20110629 00:24:08< zaroth_> "todo" word count in wesnoth code is quite big already ;-)
20110629 00:24:10< zookeeper> shadowmaster, do you think that's relevant to him?
20110629 00:24:35< zaroth_> they should be somehow automatically transformed to gna task and assigned to commiter 
20110629 00:24:38< gabba> I'm not sure it's worth worrying about cheaters, since every client receives the whole gamestate I think -- but do discuss it with other devs, just in case
20110629 00:24:39< tschmitz> gabba: The current behavior is to display the arrow and ghost unit on allies' clients, but the unit itself doesn't appear to have moved to the "root" of the arrow yet
20110629 00:25:11< tschmitz> gabba: Hm, well assuming I did everything right (...), I have already addressed that todo
20110629 00:25:22< tschmitz> gabba: by modifying the server code.
20110629 00:25:42< tschmitz> gabba: Any luck with the compilation?
20110629 00:27:13< gabba> tschmitz: ok, I'm a bit surprised that the move is not caught as invalid and deleted, given it doesn't have a unit at its starting position
20110629 00:27:52< gabba> no luck with the compilation yet, I'm just looking at the patch to give you some feedback now
20110629 00:28:34< tschmitz> gabba: I think it's a result of the way I wrote side_actions::execute_net_cmd(), which doesn't do validity checks upon insertion into the side_actions
20110629 00:29:16< gabba> tschmitz: btw, did you have any trouble with the serialization? For instance I don't remember if I store any member variables with pointers, but those would be hard to deal with
20110629 00:30:07< tschmitz> gabba: heh, yes there are numerous pointers, but not any particularly annoying ones
20110629 00:30:11< gabba> tschmitz: ah ok, but whenever the player defines some planned actions of his own that'll trigger validation of actions received from others, or not even that?
20110629 00:31:27< tschmitz> gabba: I just tested that out and the invalid action didn't disappear
20110629 00:31:48< tschmitz> gabba: I suppose only the player's own action_queue is tested for validity, not the one owned by the other player
20110629 00:33:01< gabba> tschmitz: yeah, that's what I was gonna say... you'll have to see if it's desired behavior or not
20110629 00:33:03< tschmitz> gabba: The move class contains a pointer to a unit (the one that will be moving), but I serialized him by using his underlying_id_
20110629 00:33:47-!- stikonas__ [~and@ctv-79-132-175-26.vinita.lt] has quit [Ping timeout: 240 seconds]
20110629 00:35:26< gabba> tschmitz: did you duplicate a lot of code from the original constructor into move::move(config const& cfg) ?
20110629 00:35:48< tschmitz> gabba: Yes I believe I did
20110629 00:36:19< gabba> might be a good idea to move common code into a method, then
20110629 00:36:34< gabba> Duplicate code is hard to maintain
20110629 00:36:34< tschmitz> gabba: Sounds good, I'll do that now
20110629 00:38:08< gabba> tschmitz: same remark for constructors in other classes like recall, actually
20110629 00:39:17< tschmitz> gabba: What do you think about the particular case of the suppose_dead constructor?
20110629 00:39:49< tschmitz> gabba: They both have the same one-liner body
20110629 00:40:18< gabba> hmm
20110629 00:40:59< gabba> tschmitz: a method for one line does sound like overkill, but if you're doing a method for every other children of action, you might as well do the same there for consistency
20110629 00:41:07< gabba> more code might appear there later
20110629 00:41:25< tschmitz> gabba: Yeah
20110629 00:41:33< tschmitz> gabba: I'll go ahead and do it
20110629 00:41:50< tschmitz> gabba: Oh yeah, btw
20110629 00:42:00< tschmitz> gabba: How do you recommend I test the recall planned action?
20110629 00:42:28< gabba> can you be more specific?
20110629 00:44:20< gabba> tschmitz: (minor remark: in general, please space out the code correctly. In "pos>static_cast<int>" it's hard to distinguish the 'greater than' operator from the brackets around 'int'.)
20110629 00:44:23< tschmitz> gabba: I can't recall anybody in an ordinary multiplayer game
20110629 00:44:48< gabba> tschmitz: ok I see
20110629 00:45:01 * gabba wonders how he tested it
20110629 00:45:23< gabba> I bet that in the test scenario you can cheat and add units to that list
20110629 00:45:33< gabba> Actually, probably in any scenario
20110629 00:45:46< tschmitz> gabba: Is there a command for that?
20110629 00:45:52< gabba> Explore the cheat commands
20110629 00:46:40< gabba> tschmitz: also, there are multiplayer campaigns I believe
20110629 00:47:07< gabba> tschmitz: it may be easier to start one, recruit some units and cheat to win the scenario and move on to the next one
20110629 00:47:27< gabba> I hope you can enable cheats for testing purposes in multi, don't remember
20110629 00:48:05< tschmitz> gabba: Also, I went ahead and put spaces around the less-than and greater-than symbols, but now (to me) it makes it more difficult to see the lower-precedence || operator between the two comparisons
20110629 00:48:13< tschmitz> gabba: maybe that's just me
20110629 00:48:29< gabba> use parentheses then?
20110629 00:49:09< tschmitz> gabba: K did that; that's how I'll keep it, then
20110629 00:49:26< gabba> In any case, operator precedence does "get into your eyes" after a while
20110629 00:49:52< gabba> but if parentheses help you read the code, go ahead and use them
20110629 00:50:02< tschmitz> gabba: I think in general I am comfortable with fewer spaces in my expressions than most other people are
20110629 00:51:05< tschmitz> gabba: Oh yeah
20110629 00:51:27-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has quit [Remote host closed the connection]
20110629 00:51:46< tschmitz> gabba: As a matter of fact, some of the code in the move(config) constructor is borrowed from some functions in the manager class
20110629 00:51:48-!- negusnyul [~negusnyul@dsl4E5CC8DB.pool.t-online.hu] has quit [Remote host closed the connection]
20110629 00:52:18< gabba> you mean, as I wrote it?
20110629 00:52:35< tschmitz> gabba: Yeah, functions in the manager class that you wrote
20110629 00:53:12< tschmitz> gabba: Since the manager takes care of displaying the "temporary" arrow and fake unit while the user is trying to select their move,
20110629 00:53:52< tschmitz> gabba: when the user confirms the move, the newly constructed move object takes over the arrow and fake unit that were constructed by the manager
20110629 00:54:10< gabba> I do remember that now
20110629 00:56:10< tschmitz> gabba: I guess in some sense, the move object is partially constructed by the manager and then completely constructed later when the user confirms the move
20110629 00:58:23< gabba> It's kind of convoluted, I suppose the whole arrow and fake unit initial and final creation *could* be moved out of both move and config into a new class or in utility.cpp
20110629 00:58:54< gabba> If you find a better way to do it, go for it
20110629 01:00:09< tschmitz> gabba: All right, I'll think on it
20110629 01:01:04< tschmitz> gabba: Do you think your compilation is going to come through?
20110629 01:01:38< gabba> tschmitz: I'm finishing the patch review, and afterwards it's back to googling my problem
20110629 01:02:02< gabba> yeah I'll make it work tonight, somehow
20110629 01:02:51< tschmitz> gabba: I made some updates today to the code
20110629 01:03:33< tschmitz> gabba: I made another patch, although it will include a lot of the same stuff that's in the patch you're currently reviewing
20110629 01:04:00< tschmitz> gabba: Though perhaps you could diff the two patches, if you wanted to look at the stuff I did today
20110629 01:05:10< gabba> A diffdiff? gah, that sounds hard to read
20110629 01:05:38< gabba> You told me about the changes, so that's enough I think
20110629 01:06:02< gabba> tschmitz: Ok I'm done reading the code
20110629 01:06:30< tschmitz> eh, I edited the server code a little bit and I went back to change a few details in the client code
20110629 01:06:40< gabba> The serialization code looks fine overall
20110629 01:07:29< gabba> I still find the net_cmd appelation confusing, but I don't have a suggestion for a better name
20110629 01:07:43< tschmitz> lol dang
20110629 01:08:03< gabba> All in all: good job, looks like solid clean code to me
20110629 01:08:10< tschmitz> It's a command that came from across the network
20110629 01:08:12< tschmitz> I guess.
20110629 01:08:29< tschmitz> Thank you.
20110629 01:08:42-!- mjs-de [~mjs-de@d120235.adsl.hansenet.de] has quit [Quit: On the road again]
20110629 01:09:28< gabba> tschmitz: yeah... the thing is, you're using "execute_net_cmd" but I already used execute in another context, and stuff like that
20110629 01:09:44< tschmitz> gabba: Yes ... that is true.
20110629 01:09:45< gabba> it doesn't help form a clear mental picture of what's going on
20110629 01:17:49< tschmitz> gabba: Is there a reason why the arrow class doesn't really have any meaningful constructors?
20110629 01:18:21< gabba> tschmitz: it's a clever plan to confuse newcomers
20110629 01:18:32< gabba> err let me look at that code
20110629 01:19:43< tschmitz> gabba: Looks like the caller default-constructs the arrow and then calls several setters on the new object to set its fields
20110629 01:19:51< tschmitz> brb bathroom
20110629 01:22:45< tschmitz> back
20110629 01:23:05< gabba> tschmitz: yup, I guess I just never felt the need for constructors. But there's no reason why you couldn't add some.
20110629 01:23:46< tschmitz> gabba: OK
20110629 01:24:04< tschmitz> gabba: That'll probably be part of the refactoring duplicate code
20110629 01:25:17< gabba> I was trying to keep the arrows generic and separate from the whiteboard though
20110629 01:27:09< tschmitz> OK
20110629 01:36:53-!- happygrue [~quassel@rrcs-69-193-65-79.nys.biz.rr.com] has joined #wesnoth-dev
20110629 01:37:39< tschmitz> gabba: I think the main reason I wanted a constructor was because it could add itself to the screen automatically
20110629 01:38:21-!- fendrin [~fabi@88-134-20-121-dynip.superkabel.de] has joined #wesnoth-dev
20110629 01:38:29< tschmitz> gabba: kinda like how it removes itself from the screen in the destructor
20110629 01:39:30< gabba> sounds like a good idea
20110629 01:40:09< tschmitz> gabba: OK cool, just checking with you on that
20110629 01:44:28-!- happygrue_ [~quassel@rrcs-69-193-65-79.nys.biz.rr.com] has joined #wesnoth-dev
20110629 01:46:03-!- happygrue [~quassel@rrcs-69-193-65-79.nys.biz.rr.com] has quit [Ping timeout: 276 seconds]
20110629 01:51:15< tschmitz> gabba: by the way, MSVC Call Browser says that arrow objects are constructed in two places: one by you, one by me
20110629 01:51:28< tschmitz> gabba: Perhaps that's why you didn't need a constructor
20110629 01:51:31< tschmitz> hah
20110629 01:54:57-!- happygrue_ [~quassel@rrcs-69-193-65-79.nys.biz.rr.com] has quit [Ping timeout: 276 seconds]
20110629 01:59:16< gabba> tschmitz: True that. I guess you could create and destroy a temp arrow in the manager and then recreate it in the move constructor, if you prefer it like that.
20110629 01:59:19-!- happygrue [~quassel@rrcs-69-193-65-79.nys.biz.rr.com] has joined #wesnoth-dev
20110629 02:00:16< gabba> @all: any CMAKE expert around?
20110629 02:00:45-!- Bob_The_Mighty [~chatzilla@cpc5-brig16-2-0-cust70.3-3.cable.virginmedia.com] has joined #wesnoth-dev
20110629 02:01:22< gabba> "pkg-config --cflags glib-2.0" picks up the correct location for glibconfig.h, yet make doesn't find it when I build
20110629 02:01:44-!- anonymissimus [~chatzilla@HSI-KBW-095-208-215-104.hsi5.kabel-badenwuerttemberg.de] has quit [Quit: ChatZilla 0.9.86.1 [Firefox 3.6.2/20100316074819]]
20110629 02:01:49< tschmitz> gabba: Are you still building on Ubuntu?
20110629 02:01:54< gabba> tschmitz: yes
20110629 02:02:47< tschmitz> gabba: The Windows build went pretty smoothly for me once I decided it wasn't gonna work with VC10, and I switched to VC9, hah
20110629 02:03:29< gabba> tschmitz: Trying to drag me over to the dark side, eh? :P
20110629 02:04:11< tschmitz> gabba: Hey, whatever works ...
20110629 02:04:53< gabba> Not having to reboot to work on wesnoth would be convenient though, since I use Steam and other programs quite a bit
20110629 02:05:03< gabba> But I like my eclipse setup
20110629 02:06:01< tschmitz> Aye
20110629 02:06:35-!- Shakey [HydraIRC@c-24-1-105-76.hsd1.il.comcast.net] has joined #wesnoth-dev
20110629 02:06:45-!- happygrue [~quassel@rrcs-69-193-65-79.nys.biz.rr.com] has quit [Ping timeout: 264 seconds]
20110629 02:12:53< gabba> tschmitz: I think I found the problem, cmake caches the location of libs but apparently can't detect when you change OS and render those invalid
20110629 02:13:19< gabba> cleared my cmakecache.txt and it seems to be working fine - let's see if it finishes successfully
20110629 02:13:22< tschmitz> gabba: Hope it works
20110629 02:17:18< gabba> yup, compiled
20110629 02:18:48< gabba> tschmitz: ok, please send me your latest patch then
20110629 02:19:15< tschmitz> gabba: K hold on
20110629 02:20:14< tschmitz> gabba: Sent
20110629 02:21:13< tschmitz> gabba: Should I test it with you?
20110629 02:21:20< gabba> why not
20110629 02:22:10< gabba> one of use will have to host the server, we can't use the official one since you have custom server code isn't it
20110629 02:22:20< tschmitz> gabba: Yes
20110629 02:22:46< tschmitz> gabba: Hopefully my most recent successful build was the same one I just sent you ...
20110629 02:22:59< tschmitz> gabba: Are you compiling it now?
20110629 02:23:43< gabba> yes
20110629 02:24:15< tschmitz> gabba: Also, what port does wesnothd use again?
20110629 02:24:22< gabba> some errors
20110629 02:24:47< tschmitz> gabba: Seemingly significant?
20110629 02:25:08< gabba> well, significant enough to prevent compilation ;)
20110629 02:26:25< tschmitz> gabba: I've already modified my version of the code since the patch I sent you, but
20110629 02:26:37< tschmitz> gabba: So what are the errors, then?
20110629 02:26:47< gabba> http://pastebin.com/faj47fg3
20110629 02:27:18-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit []
20110629 02:27:34-!- happygrue [~quassel@rrcs-69-193-65-79.nys.biz.rr.com] has joined #wesnoth-dev
20110629 02:28:01< gabba> They're all basically "error: ambiguous overload for ‘operator=’"
20110629 02:28:13< tschmitz> gabba: K I can fix the ambiguous overloads with a static cast I'm pretty sure
20110629 02:28:25-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20110629 02:29:42< tschmitz> gabba: How about those font errors?
20110629 02:31:36< gabba> the font errors come up whenever there's another build error, from what I remember, so nothing to worry about
20110629 02:32:01< tschmitz> gabba: Can I just pastebin the fixed code?
20110629 02:32:40< tschmitz> The other option would be making a patch just for side_actions.cpp, but it's going to contain the changes from the patch that I already sent you also
20110629 02:33:37< gabba> just send me a fresh patch, I can easily revert the previous one
20110629 02:33:39< tschmitz> http://pastebin.com/EQmYuuPX contains the fixed functions
20110629 02:34:03< tschmitz> gabba: Well I already made other changes to my working copy
20110629 02:34:19< tschmitz> gabba: do you want a fresh patch of just side_actions.cpp?
20110629 02:34:25< gabba> Ok, guess I'll overwrite those functions then
20110629 02:34:56< tschmitz> OK
20110629 02:35:22< tschmitz> gabba: So what port is wesnothd on?
20110629 02:35:33< gabba> good question
20110629 02:36:08< tschmitz> 15000?
20110629 02:37:03< gabba> my search yields that as well
20110629 02:37:10< tschmitz> gabba: Thanks for testing the build on another compiler, btw
20110629 02:37:32< gabba> np, that'll save you some complaints when committing
20110629 02:38:04< tschmitz> gabba: So do you want to open the port and give me your IP address or something? What's the best way to do this?
20110629 02:39:01< gabba> we can try that, I'll have to configure my router first
20110629 02:39:28< tschmitz> gabba: Yeah ... that's why I didn't volunteer, heh
20110629 02:40:03< gabba> so you're letting your mentor do the hard work uh? :P
20110629 02:40:48< tschmitz> gabba: I'll invoke the excuse that my dad might not like that
20110629 02:41:03< tschmitz> gabba: I'm on his desktop computer for now
20110629 02:47:20-!- Blueblaze2 [~Blueblaze@adsl-99-182-53-71.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20110629 02:48:33-!- Blueblaze [~Blueblaze@adsl-99-171-160-99.dsl.hstntx.sbcglobal.net] has quit [Read error: Connection reset by peer]
20110629 02:51:50-!- Blueblaze [~Blueblaze@adsl-99-182-55-23.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20110629 02:54:30-!- Blueblaze2 [~Blueblaze@adsl-99-182-53-71.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 255 seconds]
20110629 02:56:03-!- Blueblaze [~Blueblaze@adsl-99-182-55-23.dsl.hstntx.sbcglobal.net] has quit [Read error: Connection reset by peer]
20110629 02:57:38-!- happygrue [~quassel@rrcs-69-193-65-79.nys.biz.rr.com] has quit [Ping timeout: 246 seconds]
20110629 03:02:20-!- Blueblaze [~Blueblaze@adsl-99-182-55-50.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20110629 03:03:45-!- monochromatic [~monochrom@pool-74-109-60-121.phlapa.east.verizon.net] has joined #wesnoth-dev
20110629 03:12:56-!- Upth [~ogmar@c-76-105-37-245.hsd1.ca.comcast.net] has joined #wesnoth-dev
20110629 03:12:56-!- Upth is now known as Upthorn
20110629 03:16:39< gabba> tschmitz: looks like the game crashed on me
20110629 03:16:55< tschmitz> gabba: Oh it did?
20110629 03:17:02< tschmitz> gabba: I assumed you'd quit, since you said you had to go
20110629 03:17:08< gabba> nice test nevertheless, it's pretty impressive to see that working
20110629 03:17:23< tschmitz> gabba: Yeah I liked seeing it for the first time
20110629 03:17:35< gabba> So you'll look into the recruit bugs and disabling undo while whiteboard is on?
20110629 03:17:38< tschmitz> gabba: I haven't done particularly comprehensive tests, unfortuately
20110629 03:18:09< tschmitz> gabba: I'll look into disabling undo
20110629 03:18:21< gabba> If only allies see plans I think this is sufficiently commit-ready
20110629 03:18:32< tschmitz> gabba: I'm not even sure how to reproduce those recruit bugs;
20110629 03:18:45< gabba> Maybe do a few more tests
20110629 03:18:55< tschmitz> gabba: Yeah hm, I haven't tested the new code for whether or not enemies see plans
20110629 03:19:01< tschmitz> gabba: but they didn't see plans in the old code
20110629 03:19:17< tschmitz> gabba: although a cheater who modified his client would be able to see the plans using the old code
20110629 03:19:32< gabba> to reproduce, just start the same game but with two clients on your own computer
20110629 03:19:45< tschmitz> oh hey yeah, wesnoth crashed on me as well
20110629 03:20:24< tschmitz> gabba: Actually, it's still running even though it claims there was an assertion failure
20110629 03:21:13< gabba> That's possible, if you're running it under Visual Studio maybe it's set up to continue on asserts
20110629 03:21:25< gabba> Anyways gotta go
20110629 03:21:29< tschmitz> K have fun
20110629 03:21:41< gabba> I'm pretty happy with how you're doing, keep it up!
20110629 03:21:50< tschmitz> Thanks!
20110629 03:21:57-!- gabba [~gabba@wesnoth/developer/gabba] has left #wesnoth-dev []
20110629 03:45:59< CIA-85> shadowmaster * r50075 /trunk/src/ (CMakeLists.txt SConscript): Move filechooser.o from libwesnoth to the main wesnoth proper since I keep hitting linking issues with scons otherwise when making it depend on more GUI2 code
20110629 03:46:28< CIA-85> shadowmaster * r50076 /trunk/ (7 files in 4 dirs): gui2: Convert the New Folder dialog to GUI2
20110629 03:47:52< CIA-85> shadowmaster * r50077 /trunk/src/gui/dialogs/folder_create.cpp: gui2: tfolder_create doesn't touch the title label, so undocument it
20110629 04:06:00-!- atomicbomb [~quassel@125.160.87.166] has joined #wesnoth-dev
20110629 04:08:16-!- LordNasty [~NaSTy@93-43-157-80.ip92.fastwebnet.it] has quit []
20110629 04:09:49-!- ancestral [~ancestral@67-6-61-230.mpls.qwest.net] has joined #wesnoth-dev
20110629 04:23:38-!- Bob_The_Mighty [~chatzilla@cpc5-brig16-2-0-cust70.3-3.cable.virginmedia.com] has quit [Quit: ChatZilla 0.9.87 [Firefox 5.0/20110615151330]]
20110629 04:23:44-!- Shakey [HydraIRC@c-24-1-105-76.hsd1.il.comcast.net] has quit [Quit:  HydraIRC -> http://www.hydrairc.com <- s0 d4Mn l33t |t'z 5c4rY!]
20110629 04:35:16-!- Ivanovic_ [~ivanovic@dtmd-4db2e954.pool.mediaWays.net] has joined #wesnoth-dev
20110629 04:35:16-!- Ivanovic_ [~ivanovic@dtmd-4db2e954.pool.mediaWays.net] has quit [Changing host]
20110629 04:35:16-!- Ivanovic_ [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20110629 04:38:57-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 240 seconds]
20110629 04:39:18-!- Ivanovic_ is now known as Ivanovic
20110629 04:41:23-!- tschmitz [626cdb95@gateway/web/freenode/ip.98.108.219.149] has quit [Quit: Page closed]
20110629 05:19:04-!- tschmitz [626cdb95@gateway/web/freenode/ip.98.108.219.149] has joined #wesnoth-dev
20110629 05:24:20< Exasperation> Ugh, my brain feels like it's been stirred with a stick, but I managed to make a multi_page widget with pages that have different layouts
20110629 05:24:55< Exasperation> wasn't someone on here trying to do that earlier?
20110629 05:27:06< shadowmaster> yes! :o
20110629 05:27:08< shadowmaster> me
20110629 05:29:20< Exasperation> It's not easy
20110629 05:31:13< Exasperation> the multi_page has a member function called set_page_builder that lets you change the page definition you use to add new pages to the multi_page
20110629 05:32:21< Exasperation> I'm making a pastebin, hang on
20110629 05:34:12< Exasperation> this is the function I wrote to change the page definition from Lua: http://pastebin.com/QNschPey
20110629 05:34:57< Exasperation> (I've been working on trying to expose more gui2 functionality to Lua)
20110629 05:36:25< Exasperation> does that help you at all?
20110629 05:36:40< shadowmaster> I'll have a look later, I'm currently in a game
20110629 05:37:05-!- alkenrinnstet [~alkenrinn@137.132.45.173] has joined #wesnoth-dev
20110629 05:37:09< Exasperation> hmm... maybe I should have set the pastebin to expire more than an hour from now
20110629 05:48:19-!- Gambit [~Gambit@wesnoth/developer/grickit] has quit [Remote host closed the connection]
20110629 05:56:02< shadowmaster> Exasperation: but that would not be easily manageable from non-dynamic WML (i.e. all of core) if I understand corrctly
20110629 05:56:17< shadowmaster> where whould I get a new grid from C++?
20110629 05:56:45< Exasperation> that's why I said it's not easy
20110629 05:57:37< shadowmaster> also, I came up with a working solution using more WML and stacked_widget as well
20110629 05:58:05< shadowmaster> the only problem with that solution is that invisible widgets seem to still capture events despite the code documentation says otherwise
20110629 05:58:36< Exasperation> as things stand now, you would have to hardcode it into the c++ side of the menu, I think
20110629 05:59:18< shadowmaster> yeah, that's not precisely optimal and kind of defeats the whole point of converting the dialog to GUI2
20110629 06:00:18< Exasperation> hmm, I haven't actually tested interaction with widgets in the multi_page, just display... I ought to test with a multipage that has sliders or toggle buttons or something
20110629 06:01:29-!- crimson_penguin [~ben@S0106002401f68dcd.vc.shawcable.net] has joined #wesnoth-dev
20110629 06:01:29-!- crimson_penguin [~ben@S0106002401f68dcd.vc.shawcable.net] has quit [Changing host]
20110629 06:01:29-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20110629 06:07:15< Exasperation> don't know what to tell you there... I guess there would need to be a change to how multi_pages are constructed from WML; maybe set the constructor up so that it can accept multiple page_definition and page_data subtags?
20110629 06:08:00< shadowmaster> I guess that would be good, yeah
20110629 06:23:24-!- alkenrinnstet [~alkenrinn@137.132.45.173] has quit [Quit: Leaving.]
20110629 06:31:10-!- ancestral [~ancestral@67-6-61-230.mpls.qwest.net] has quit [Quit: i go sleeps kthxbai]
20110629 06:34:25-!- Gallaecio [~quassel@92.39.broadband12.iol.cz] has quit [Remote host closed the connection]
20110629 06:39:59-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Quit: crimson_penguin]
20110629 07:15:30-!- timotei [~timotei@193.34.190.70] has joined #wesnoth-dev
20110629 07:15:30-!- timotei [~timotei@193.34.190.70] has quit [Changing host]
20110629 07:15:30-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20110629 07:26:07-!- Gallaecio [~quassel@92.39.broadband12.iol.cz] has joined #wesnoth-dev
20110629 07:26:52-!- crimson_penguin [~ben@S0106002401f68dcd.vc.shawcable.net] has joined #wesnoth-dev
20110629 07:26:52-!- crimson_penguin [~ben@S0106002401f68dcd.vc.shawcable.net] has quit [Changing host]
20110629 07:26:52-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20110629 07:36:53-!- fendrin [~fabi@88-134-20-121-dynip.superkabel.de] has quit [Remote host closed the connection]
20110629 07:49:53-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Quit: SIGKILL]
20110629 07:51:06-!- Blueblaze [~Blueblaze@adsl-99-182-55-50.dsl.hstntx.sbcglobal.net] has quit [Quit: Blueblaze]
20110629 07:58:13< Exasperation> argh, there seems to be a sizable bug in the part of the existing multi_page code labelled "This part is untested."
20110629 08:23:44-!- tschmitz [626cdb95@gateway/web/freenode/ip.98.108.219.149] has quit [Ping timeout: 252 seconds]
20110629 08:37:40-!- Exasperation [4a47319b@gateway/web/freenode/ip.74.71.49.155] has quit [Ping timeout: 252 seconds]
20110629 08:50:51-!- ancestral [~ancestral@67-6-61-230.mpls.qwest.net] has joined #wesnoth-dev
20110629 08:50:58-!- ancestral [~ancestral@67-6-61-230.mpls.qwest.net] has quit [Client Quit]
20110629 08:51:20-!- alkenrinnstet [~alkenrinn@bb121-6-177-142.singnet.com.sg] has joined #wesnoth-dev
20110629 08:53:40-!- Elvish_Pillager [~eli@71-10-224-192.dhcp.oxfr.ma.charter.com] has quit [Ping timeout: 260 seconds]
20110629 09:27:34-!- LordNasty [~NaSTy@93-43-157-80.ip92.fastwebnet.it] has joined #wesnoth-dev
20110629 09:35:49-!- covale [~covale@h55eb1ca0.selukra.dyn.perspektivbredband.net] has joined #wesnoth-dev
20110629 09:41:29-!- zookeeper [~l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20110629 09:41:45-!- negusnyul [~negusnyul@dsl4E5CC8DB.pool.t-online.hu] has joined #wesnoth-dev
20110629 09:54:10-!- Vorpal [~AnMaster@unaffiliated/anmaster] has joined #wesnoth-dev
20110629 10:00:02-!- covale [~covale@h55eb1ca0.selukra.dyn.perspektivbredband.net] has quit [Quit: Leaving.]
20110629 10:14:42-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has joined #wesnoth-dev
20110629 10:14:43-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has quit [Changing host]
20110629 10:14:43-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20110629 10:15:07-!- timotei [~timotei@193.34.190.70] has joined #wesnoth-dev
20110629 10:15:08-!- timotei [~timotei@193.34.190.70] has quit [Changing host]
20110629 10:15:08-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20110629 10:16:10-!- neph [~neph@80.233.231.12] has joined #wesnoth-dev
20110629 10:22:41-!- neph [~neph@80.233.231.12] has quit [Quit: Leaving]
20110629 10:27:23-!- Nephro [~Dmitry@80.233.231.12] has joined #wesnoth-dev
20110629 10:39:22-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: GPU failure]
20110629 10:40:34-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20110629 10:42:44-!- loonybot [~loonybot@46.138.87.135] has joined #wesnoth-dev
20110629 10:42:44-!- loonybot [~loonybot@46.138.87.135] has quit [Changing host]
20110629 10:42:44-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20110629 10:47:27-!- Appleman1234 [~Appleman1@CPE-60-226-179-130.qld.bigpond.net.au] has joined #wesnoth-dev
20110629 10:48:30-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: more GPU failure]
20110629 10:49:50-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20110629 10:56:02-!- MeccaGod [~majs@h146n3fls33o279.telia.com] has joined #wesnoth-dev
20110629 11:09:52-!- Appleman1234 [~Appleman1@CPE-60-226-179-130.qld.bigpond.net.au] has quit [Ping timeout: 252 seconds]
20110629 11:15:40-!- timotei21 [~timotei@193.34.190.70] has joined #wesnoth-dev
20110629 11:15:40-!- timotei21 [~timotei@193.34.190.70] has quit [Changing host]
20110629 11:15:40-!- timotei21 [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20110629 11:19:10-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 246 seconds]
20110629 11:22:28-!- Appleman1234 [~Appleman1@CPE-60-226-179-130.qld.bigpond.net.au] has joined #wesnoth-dev
20110629 11:30:51-!- Appleman1234 [~Appleman1@CPE-60-226-179-130.qld.bigpond.net.au] has quit [Ping timeout: 276 seconds]
20110629 11:31:23-!- Appleman1234 [~Appleman1@CPE-60-226-179-130.qld.bigpond.net.au] has joined #wesnoth-dev
20110629 11:37:09-!- atomicbomb [~quassel@125.160.87.166] has quit [Remote host closed the connection]
20110629 11:45:03-!- hhyloc [~hhyloc@113.166.158.218] has joined #wesnoth-dev
20110629 12:14:20< Nephro> Is it possible to attach a unit to the config? It's just that I have a method that only has an int and cfg as parameters, but I need to pass a unit too. Don't want to change the param list, since there is a long chain of calls and I would have to change every single one
20110629 12:24:10-!- Crab_ [~Crab@wesnoth/developer/crab] has joined #wesnoth-dev
20110629 12:27:36< Crab_> Nephro: put unit's id or location to the cfg
20110629 12:27:52< Nephro> Oh, hi Crab_ 
20110629 12:29:53< Nephro> ping
20110629 12:33:58< Crab_> pong
20110629 12:34:27-!- Vorpal [~AnMaster@unaffiliated/anmaster] has quit [Ping timeout: 240 seconds]
20110629 12:35:50< Nephro> Crab_, my last commit iirc was small and had only the counter. The next one is coming soon, probably, it's just that, to test all we talked about it has to be finished together. I don't know, but I bet it's no good to commit semifinished and untested code :)
20110629 12:36:52< Crab_> yes, it's better to commit something which works, even if it's unfinished.
20110629 12:38:59-!- mordante [~mordante@87.215.201.26] has joined #wesnoth-dev
20110629 12:39:02< mordante> servus
20110629 12:39:29< mordante> shadowm_laptop, I saw your comment before, but was only online shortly yesterday
20110629 12:40:01< mordante> need to check is later, hopefully I have time tomorrow
20110629 12:40:38< shadowmaster> k
20110629 12:40:54< mordante> gabba the minimum versions of a dependency are listed in INSTALL and updating to a newer version should be discussed
20110629 12:41:17< mordante> the server runs on Debian stable, but some users might user older versions of a dependency
20110629 12:42:58< Nephro> Wanted to ask something about the structure: (1) for the sticky behavior candidate actions - should I derive such a class from lua_candidate_action_wrapper or hack inside the existing one(I like the second choice better) (2) why would we need a Lua hook to disable CAs from inside Lua if we can just do this->disable(); after the lua_candidate_action_wrapper executes itself? 
20110629 12:43:15< Nephro> Crab_, ^
20110629 12:43:37< shadowmaster> Crab_: yay, I found you! what do you think about forward-porting the new chat log dialog to trunk?
20110629 12:49:28< Crab_> shadowmaster: yes, I've seen your message in the log. will do.
20110629 12:51:12< Crab_> for (2) -  we need to know if CA is done or not. so, we need to disable either from CA itself or conditionally based on it's return value, but in that case we'll still need a lua hook for setting that return value
20110629 12:53:14< Nephro> Crab_, oh, you must mean whether it's completely done. I thought we were talking about whether it's done in this turn(like patrol waypoint reached, if we don't disable it will just go to the next, so this turn it's done)
20110629 12:53:20-!- MeccaGod [~majs@h146n3fls33o279.telia.com] has quit []
20110629 12:53:24< Nephro> ok, got this
20110629 12:57:01< Nephro> oh and: I've set up the [add_ai_behavior] tag, wired it to a Lua function. How do I find out the ID of the stage with all the CAs(id of the loop stage) and/or side of the current AI without explicitly stating that?
20110629 13:00:08< Crab_> actually, I meant like 'done for this turn', but the 'done forever' is good, too.
20110629 13:00:25< Crab_> 'main_loop' it is
20110629 13:00:45< Crab_> and side= should be specified explicitly in [add_ai_behavior]
20110629 13:01:08< Nephro> Crab_, but even in our lua_ai arena the id of that stage wasn
20110629 13:01:12< Nephro> wan't main loop
20110629 13:01:35< Nephro> wasn't main_loop
20110629 13:01:50< Crab_> it is a test ai there
20110629 13:02:08< Crab_> by default, [add_ai_behavior] is supposed to influence the main loop of rca ai
20110629 13:02:15< Crab_> and it's id is main_loop
20110629 13:02:19< Crab_> for (1), do as you think is better.
20110629 13:02:28< Crab_> just avoid code duplication
20110629 13:03:20< Nephro> then I will make it use main_loop as default, but a different id if such is explicitly specified
20110629 13:03:52-!- Unnheulu [~ieuan@host86-147-248-76.range86-147.btcentralplus.com] has joined #wesnoth-dev
20110629 13:04:21-!- atomicbomb [~quassel@61.94.46.200] has joined #wesnoth-dev
20110629 13:05:01-!- Gambit [~Gambit@wesnoth/developer/grickit] has joined #wesnoth-dev
20110629 13:05:02-!- mjs-de [~mjs-de@f053188070.adsl.alicedsl.de] has joined #wesnoth-dev
20110629 13:09:44-!- Unnheulu_ [~ieuan@host86-147-248-76.range86-147.btcentralplus.com] has joined #wesnoth-dev
20110629 13:11:15-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit [Ping timeout: 258 seconds]
20110629 13:16:00-!- hhyloc [~hhyloc@113.166.158.218] has quit [Ping timeout: 252 seconds]
20110629 13:19:58< Crab_> about the ids - good, makes sense.
20110629 13:21:19-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 246 seconds]
20110629 13:26:25-!- Unnheulu [~ieuan@host86-147-248-76.range86-147.btcentralplus.com] has quit [Quit: Ex-Chat]
20110629 13:48:00-!- Nephro [~Dmitry@80.233.231.12] has quit [Remote host closed the connection]
20110629 13:50:59-!- AngelsJinx [~AngelsJin@119.224.78.213] has joined #wesnoth-dev
20110629 13:52:16-!- Sytyi [~sytyi@207-18-133-95.pool.ukrtel.net] has joined #wesnoth-dev
20110629 13:52:48< Sytyi> mordante: hi!
20110629 14:00:48-!- iwaim [~iwaim@rasteenie.alib.jp] has quit [Remote host closed the connection]
20110629 14:01:09-!- 30BAAGWMJ [~iwaim@rasteenie.alib.jp] has joined #wesnoth-dev
20110629 14:02:47< AngelsJinx> Hey guys, does anyone know of an active Android port in the works? Everywhere I look I find dead projects, would be really keen to help port it for android
20110629 14:07:45< timotei21> AngelsJinx: did you search in the forums? :)
20110629 14:08:37< AngelsJinx> I found three threads I think, one of which had been locked, and the other two didn't seem to be talking about anything currently in progress. Everyone seems to be waiting on KylePoole, who has vanished :P
20110629 14:09:05< AngelsJinx> Just wondered if there was a more secretive project in the works that only you irc dudes know about
20110629 14:09:29< timotei21> AngelsJinx: none that I know
20110629 14:09:42< AngelsJinx> Bummer. Might have to look into it myself then :D
20110629 14:10:15< Gambit> timotei21: None that you know of?
20110629 14:10:24< Gambit> Pfft I guess you don't subscribe to @WesnothOrg on twitter.
20110629 14:10:27< timotei21> AngelsJinx: well, looking on http://www.youtube.com/watch?v=Pl6Pkp2By8I it seem cjhopman is doing something :P
20110629 14:10:29< Gambit> [shameless plug]
20110629 14:10:35< Gambit> There you go.
20110629 14:10:40< Gambit> haha
20110629 14:10:42< timotei21> ?
20110629 14:11:17< AngelsJinx> I found that too, sent that guy a message via youtube to see if it's still being worked on, but have yet to recieve a reply
20110629 14:11:37< timotei21> AngelsJinx: he's here on channel :P
20110629 14:12:14< AngelsJinx> I did just see that. The big question, is he *really* here?
20110629 14:12:36< timotei21> AngelsJinx: idling on the channel is the best move you can do know :)
20110629 14:12:44< timotei21> I think he'll pop here sometime :P
20110629 14:13:33< AngelsJinx> Thats cool, can't say I've been on this irc channel before. Hopefully he swings by soonish :)
20110629 14:14:08< timotei21> and yeah, I hope we don't convert wesnoth to java 
20110629 14:14:13< timotei21> we won't*
20110629 14:15:47< AngelsJinx> I havn't looked to intensely at the NDK, but it looks like it'd be possible to handle graphics/UI stuff in java, and leave the majority as it is. Maybe swap networking to java 
20110629 14:16:08< timotei21> hmm
20110629 14:16:22< timotei21> doesn't Android support C/C++ ?
20110629 14:16:42< timotei21> IIRC, hedgewars, a game compiled in pascal
20110629 14:16:48< timotei21> had a GSOC project to make an Android port
20110629 14:16:51< AngelsJinx> I admit I havn't had a good look at the wesnoth source code either, but hopefully it's tidily seperated :D
20110629 14:16:53< AngelsJinx> Android supports C/C++ through the NDK, but recommends that everything is done through java
20110629 14:16:55< timotei21> maybe you could ask them what they did
20110629 14:17:04< timotei21>  #hedgewars
20110629 14:28:58< AngelsJinx> After reading this http://www.google-melange.com/gsoc/proposal/review/google/gsoc2011/xeli/1 looks like I might need to look more closely at JNI
20110629 14:29:31< timotei21> AngelsJinx: ah. you found it :P nice
20110629 14:29:56< timotei21> yeah, JNI should do
20110629 14:30:16< AngelsJinx> Yeah. It's 12:30am here, figured I shouldn't try contacting that guy on irc, lest I fail to go to bed at all. JNI research time :D
20110629 14:30:18-!- monochromatic [~monochrom@pool-74-109-60-121.phlapa.east.verizon.net] has quit [Read error: Operation timed out]
20110629 14:30:37< timotei21> AngelsJinx: have fun ;)
20110629 14:31:12< AngelsJinx> Have you got any experience in this stuff? You seem to know what you're on about
20110629 14:33:04-!- monochromatic [~monochrom@pool-173-59-67-167.phlapa.east.verizon.net] has joined #wesnoth-dev
20110629 14:33:28< timotei21> AngelsJinx: I haven't played with JNI yet :(
20110629 14:33:34< timotei21> nor with Android stuff
20110629 14:33:44< timotei21> maybe you could ask the #android guys for some pointers ;)
20110629 14:34:46< AngelsJinx> Looks like JNI could almost interface directly with the wesnoth functions, so it should only be necessary to write a wrapper to display UI and to call the wesnoth functions
20110629 14:34:49< timotei21> hm... vim doesn't cope well with embedded PHP code in HTML :(
20110629 14:35:06< timotei21> AngelsJinx: well, my first tought was: why not call directly the main method lol? :))
20110629 14:35:18< timotei21> the problem would be, you still need to be able to run natively C++ code
20110629 14:35:40< timotei21> or at least compile it into code  that is able to be executed by the Android OS
20110629 14:35:57< AngelsJinx> I'm pretty sure it's possible to run c++ (more or less) natively using the NDK
20110629 14:39:39< elias> i think it depends on the version number... 1.x you need to use JNI for some things
20110629 14:39:56< elias> then early 2.x you still needed java for a few things, like getting an opengl context
20110629 14:40:05< elias> 2.3 you basically don't need java
20110629 14:40:20-!- hhyloc [~hhyloc@113.166.158.218] has joined #wesnoth-dev
20110629 14:41:01-!- MeccaGod [~majs@h146n3fls33o279.telia.com] has joined #wesnoth-dev
20110629 14:41:05< elias> also SDL works on android, so chances are someone just needs to re-compile and fix performance/memory issues
20110629 14:41:15< AngelsJinx> Hm. First problem could be compiling wesnoth for ARM and as nice as using 2.3 without many changes would be, market share is still 2.2, so it'd be most useful to do it for that
20110629 14:41:58< elias> SDL should take care of it anyway
20110629 14:42:59< elias> i have dosbox and snes9x on my nexus s working perfectly, and i believe both use SDL
20110629 14:43:05< elias> (not 100% sure though)
20110629 14:43:51< AngelsJinx> yeah, looks like SDL can be compiled for android. Saves rewriting the UI from scratch in Java. Though I'm still not certain of where we need java, and where we could stick with C++.
20110629 14:44:21< AngelsJinx> Perhaps redo SDL in java (Since there are java bindings, so it shouldn't be a huge change)
20110629 14:44:24< elias> i'd say the only parts which need to be java are inside of SDL
20110629 14:45:50< AngelsJinx> Reckon wesnoth will compile to ARM easily enough?
20110629 14:46:27< elias> yes, no problem there
20110629 14:46:42< elias> and SDL 1.3 apparently supports android out of the box
20110629 14:46:58< AngelsJinx> Okay, so whats the huge problem that we haven't seen? Why hasn't someone done this already?
20110629 14:48:01< elias> still takes some time
20110629 14:48:26< elias> need someone who can dedicate 2 or 3 weeks of fulltime work i guess
20110629 14:49:12< AngelsJinx> well I'm happy to chip away at it, though I'm not as clued up as you guys seem to be
20110629 14:52:11-!- Alarantalara [~alarantal@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has joined #wesnoth-dev
20110629 14:52:41-!- hhyloc [~hhyloc@113.166.158.218] has quit [Quit: Leaving]
20110629 14:53:20< timotei21> AngelsJinx: you could ask around here if you bump in something unknown :P
20110629 14:53:47< AngelsJinx> Thats true, I expect I'll be on here alot the next few months. So, any advice on where to start?
20110629 14:54:24< timotei21> AngelsJinx: did you check about boost? :P
20110629 14:54:27< timotei21> or that doesn't matter
20110629 14:54:34< timotei21> you could create a simple SDL Game
20110629 14:54:39< timotei21> and check it on Android
20110629 14:54:40< timotei21> ;)
20110629 14:54:55< AngelsJinx> I saw that mentioned, but figured it wouldn't be major.
20110629 14:54:57< AngelsJinx> Yeah, I figure having a play with SDL is my starting point
20110629 14:55:05< elias> oh, that youtube video of the android port looks nice
20110629 14:55:17< elias> wesbot: seen cjhopman
20110629 14:55:17< wesbot> elias: Queried user last spoke 11d 4h ago.  cjhopman is currently here and on the channel #wesnoth.
20110629 14:56:15< elias> so i guess someone is working on a port then already :)
20110629 14:57:14< AngelsJinx> yeah, hoping to get in touch with him. Having just discovered wesbot courtesy of you, I might be able to hunt him down a bit quicker :P
20110629 14:57:48-!- Alarantalara [~alarantal@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has left #wesnoth-dev []
20110629 15:02:37-!- timotei [~timotei@193.34.190.70] has joined #wesnoth-dev
20110629 15:02:37-!- timotei [~timotei@193.34.190.70] has quit [Changing host]
20110629 15:02:37-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20110629 15:04:54-!- AngelsJinx [~AngelsJin@119.224.78.213] has quit [Read error: Connection reset by peer]
20110629 15:05:11-!- AngelsJinx [~AngelsJin@119.224.78.213] has joined #wesnoth-dev
20110629 15:05:42-!- Johannes13 [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20110629 15:06:00-!- timotei21 [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 276 seconds]
20110629 15:14:00-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Read error: Connection reset by peer]
20110629 15:14:22-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20110629 15:15:19-!- Crendgrim [~crend@77-22-254-233-dynip.superkabel.de] has joined #wesnoth-dev
20110629 15:48:38-!- AngelsJinx [~AngelsJin@119.224.78.213] has left #wesnoth-dev []
20110629 15:53:42-!- Cookiee [~quassel@unaffiliated/cookiee] has joined #wesnoth-dev
20110629 15:57:02-!- AngelsJinx [~AngelsJin@119.224.78.213] has joined #wesnoth-dev
20110629 15:57:27-!- AngelsJinx [~AngelsJin@119.224.78.213] has left #wesnoth-dev []
20110629 16:01:48-!- Crendgrim is now known as Crend
20110629 16:01:52-!- Crend is now known as Crendgrim
20110629 16:05:00-!- Gallaecio [~quassel@92.39.broadband12.iol.cz] has quit [Remote host closed the connection]
20110629 16:16:29-!- Shakey [~Shakey@12.190.80.225] has joined #wesnoth-dev
20110629 16:22:44-!- atomicbomb [~quassel@61.94.46.200] has quit [Quit: Look! Jack Sparrow!! Right there! In TV!]
20110629 16:32:39-!- Crab_ [~Crab@wesnoth/developer/crab] has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
20110629 16:36:02-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Read error: Operation timed out]
20110629 16:39:20-!- stikonas [~and@dynamic28.vpdn.csx.cam.ac.uk] has joined #wesnoth-dev
20110629 16:39:21-!- stikonas [~and@dynamic28.vpdn.csx.cam.ac.uk] has quit [Changing host]
20110629 16:39:21-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20110629 16:40:47-!- zookeeper [~l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20110629 16:41:12-!- stikonas_ [~and@ctv-79-132-175-26.vinita.lt] has joined #wesnoth-dev
20110629 16:41:12-!- stikonas_ [~and@ctv-79-132-175-26.vinita.lt] has quit [Changing host]
20110629 16:41:12-!- stikonas_ [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20110629 16:41:54-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Read error: Operation timed out]
20110629 17:02:37-!- stikonas_ [~and@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20110629 17:04:15-!- stikonas [~gentoo@ctv-79-132-175-26.vinita.lt] has joined #wesnoth-dev
20110629 17:11:31< zaroth_> awww, Crab_.
20110629 17:11:48< zaroth_> missed just by 40 minutes
20110629 17:15:20-!- tschmitz [626cdb95@gateway/web/freenode/ip.98.108.219.149] has joined #wesnoth-dev
20110629 17:20:12-!- vcap [~vcap@AReims-551-1-92-163.w92-155.abo.wanadoo.fr] has quit [Quit: leaving]
20110629 17:20:23< mordante> hi Sytyi 
20110629 17:20:52< mordante> Sytyi, I had a look at your patch and tested both the wiki_grabber and the schema_generator
20110629 17:20:58< mordante> Sytyi, some remarks:
20110629 17:21:09< mordante> there's still some /// doxygen comment left
20110629 17:21:44< mordante> the schema itself also has two oddities, the first item stats in column 2 instead of 1
20110629 17:22:21< mordante> there are some items like »mandatory=""""« which seems odd as well
20110629 17:22:21-!- alkenrinnstet [~alkenrinn@bb121-6-177-142.singnet.com.sg] has quit [Quit: Leaving.]
20110629 17:22:36-!- alkenrinnstet [~alkenrinn@bb121-6-177-142.singnet.com.sg] has joined #wesnoth-dev
20110629 17:22:40< Sytyi> mordante: ok. thanks. Will make that correct. 
20110629 17:22:54< mordante> there are also some minor style issues left with the patch, but I think I'll commit your patch
20110629 17:23:08< mordante> and then afterwards fix the style issues so you can see what I mean
20110629 17:23:20< mordante> I don't want to keep revising the patch over and over
20110629 17:23:48< mordante> I only think you should at least fix the »mandatory=""""« issue
20110629 17:23:59-!- Vorpal [~AnMaster@unaffiliated/anmaster] has joined #wesnoth-dev
20110629 17:24:36< mordante> I won't be able to commit today so if you can upload the revised patch before tomorrow evening I'll commit it
20110629 17:25:20< Sytyi> mordante: I need some help with annotating existing source. I can not understand if wiki macros in src/gui/auxilary/widget_definition.cpp and window_builder.cpp are about one tag resolution or different ?
20110629 17:26:16-!- fendrin [~fabi@88-134-20-121-dynip.superkabel.de] has joined #wesnoth-dev
20110629 17:26:18< Sytyi> mordante: OK. will polish that stuff ( case of """" is very easy detected. ). 
20110629 17:27:05< Sytyi> mordante: Can you give me a little advice how are the tags in gui/ organized ? 
20110629 17:27:15-!- alkenrinnstet [~alkenrinn@bb121-6-177-142.singnet.com.sg] has quit [Ping timeout: 276 seconds]
20110629 17:27:30-!- vcap [~vcap@AReims-551-1-92-163.w92-155.abo.wanadoo.fr] has joined #wesnoth-dev
20110629 17:27:53< mordante> Sytyi, in the widget definition several resolutions are defined
20110629 17:28:07< mordante> on every resolution a widget may look different
20110629 17:28:40< mordante> for example smaller decoration, or buttons at a smaller size, smaller font size etc etc
20110629 17:29:04< mordante> then one of these resolutions is used to build a widget
20110629 17:29:23< mordante> but that shouldn't actually matter for the schema stuff 
20110629 17:30:00< Sytyi> mordante: Can I annotate as I understand and then you will check it? 
20110629 17:30:28< mordante> since the WML is serialized only once, from that point on the resolution is transferred as a C struct
20110629 17:30:58< mordante> Sytyi, yes you can, but please keep that patch separate from the work you've done thusfar
20110629 17:31:48< mordante> I really like to commit the stuff you've done thusfar as a patch and then look separately at the new things
20110629 17:31:59< mordante> that way we can get a part of your code it
20110629 17:32:04< mordante> s/it/in/
20110629 17:33:38< Sytyi> mordante: ok. Maybe I will do an another checkout and apply patch there
20110629 17:34:17< mordante> do you use svn or git-svn?
20110629 17:34:29< Sytyi> svn
20110629 17:35:24< mordante> ok, with git-svn it would have been easy to create separate branches for these patches
20110629 17:35:37< mordante> do you have more questions?
20110629 17:36:09< Sytyi> mordante: How to read the cfg file to begin work on the next stage ? 
20110629 17:36:29< Sytyi> Can you show an example in source? 
20110629 17:38:45< mordante> Sytyi, load_settings() in src/gui/widgets/settings.cpp
20110629 17:39:10< mordante> the gui does its own cfg loading and it starts in that function
20110629 17:39:19< Sytyi> mordante: thanks. Tomorrow a patch or two will be ready. 
20110629 17:39:31< mordante> ok great
20110629 17:39:38< loonycyborg> mordante: So do you have something else to say about size_text?
20110629 17:39:49< Sytyi> mordante: no questions now, 
20110629 17:39:59< mordante> Sytyi, line 468 loads the pre processor setting
20110629 17:40:34< mordante> Sytyi, and line 470 does the pre-processing on the file
20110629 17:41:02< mordante> Sytyi, and then the file is read on line 471
20110629 17:42:03< loonycyborg> Currently the label's size seems to be only determined by its initial text.
20110629 17:43:04< loonycyborg> I need to display dynamic content in it, so I either need to resize it when its shown or have working size_text which is set to a generous value.
20110629 17:43:20< mordante> loonycyborg, sorry haven't had time for it yet :-(
20110629 17:43:45< mordante> but for example some dialogs that hit the same problem call invalidate_layout() as work-around for the issue
20110629 17:44:05< mordante> I want to fix it at some point... but need to find time for it
20110629 17:44:53< mordante> I'm off bye
20110629 17:45:04-!- stikonas [~gentoo@ctv-79-132-175-26.vinita.lt] has quit [Remote host closed the connection]
20110629 17:45:09-!- mordante [~mordante@87.215.201.26] has quit [Quit: Leaving]
20110629 17:51:04-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has joined #wesnoth-dev
20110629 17:51:04-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has quit [Changing host]
20110629 17:51:04-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20110629 17:57:05-!- timotei21 [~timotei@193.34.190.70] has joined #wesnoth-dev
20110629 17:57:05-!- timotei21 [~timotei@193.34.190.70] has quit [Changing host]
20110629 17:57:05-!- timotei21 [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20110629 17:58:40-!- Elvish_Pillager [~eli@71-10-224-192.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20110629 17:59:03-!- Cookiee [~quassel@unaffiliated/cookiee] has quit [Remote host closed the connection]
20110629 18:01:03-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 258 seconds]
20110629 18:19:37< CIA-85> anonymissimus * r50078 /trunk/projectfiles/VC9/ (4 files): 
20110629 18:19:37< CIA-85> enabled openmp for all configurations...
20110629 18:19:37< CIA-85> (forgot in r50057)
20110629 18:19:49< CIA-85> anonymissimus * r50079 /trunk/projectfiles/VC9/wesnoth.vcproj: VS9 projectfile update
20110629 18:20:07< CIA-85> anonymissimus * r50080 /trunk/projectfiles/CodeBlocks/wesnoth.cbp: CB projectfile update
20110629 18:21:40< timotei21> hmm
20110629 18:21:56< timotei21> weren't those commits from yesterday?
20110629 18:31:08-!- zaroth_ is now known as zaroth
20110629 18:39:59< CIA-85> loonycyborg * r50081 /trunk/ (2 files in 2 dirs): 
20110629 18:39:59< CIA-85> Network transmission dialog improvement:
20110629 18:39:59< CIA-85> Realign the numeric display label and use invalidate_layout()
20110629 18:39:59< CIA-85> to ensure that its size will stay correct as it's updated.
20110629 18:50:50-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20110629 18:59:00-!- anonymissimus [~chatzilla@HSI-KBW-095-208-215-104.hsi5.kabel-badenwuerttemberg.de] has joined #wesnoth-dev
20110629 18:59:17< anonymissimus> zookeeper: re: powershot
20110629 18:59:44< anonymissimus> either we form him into a productive member fitting into the wesnoth community :)
20110629 19:00:19< anonymissimus> or hes tricked into violating rules which would allow banning...
20110629 19:00:56< anonymissimus> but he seems careful enough to evade the ladder :P
20110629 19:01:37< anonymissimus> anyway that guy is the craziest I've seen since lurking or registering my account
20110629 19:07:09-!- LordNasty [~NaSTy@93-43-157-80.ip92.fastwebnet.it] has quit [Ping timeout: 276 seconds]
20110629 19:10:18-!- LordNasty [~NaSTy@93-43-157-80.ip92.fastwebnet.it] has joined #wesnoth-dev
20110629 19:13:27-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20110629 19:13:58< timotei21> fendrin: hi
20110629 19:17:05< timotei21> anonymissimus: or... you could...
20110629 19:17:27< timotei21> login in purpose with bad password, until it's locked:D
20110629 19:19:40< fendrin> timotei21: hi
20110629 19:20:13-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has joined #wesnoth-dev
20110629 19:20:13-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has quit [Changing host]
20110629 19:20:13-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20110629 19:20:14-!- negusnyul_ [~negusnyul@dsl4E5CC8DB.pool.t-online.hu] has joined #wesnoth-dev
20110629 19:21:30-!- negusnyul [~negusnyul@dsl4E5CC8DB.pool.t-online.hu] has quit [Remote host closed the connection]
20110629 19:24:01-!- LordNasty [~NaSTy@93-43-157-80.ip92.fastwebnet.it] has quit [Ping timeout: 240 seconds]
20110629 19:24:52< timotei21> fendrin: there is a small big problem
20110629 19:24:57< timotei21> since we have installs
20110629 19:25:01< timotei21> we can have different schema
20110629 19:25:11< timotei21> but there are points ..
20110629 19:25:24< timotei21> ah. damn me. I haven't thought enough (again)
20110629 19:25:27< timotei21> sorry :P
20110629 19:25:41< timotei21> the idea was: there are points, at which the install is not defined/available
20110629 19:25:47< timotei21> should I just use the default install's one?
20110629 19:25:53< timotei21> that seems for me the right choice
20110629 19:26:01-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 240 seconds]
20110629 19:27:57< fendrin> timotei21: the install is not defined? Sorry, I seem to miss the context.
20110629 19:28:02-!- LordNasty [~NaSTy@93-43-157-80.ip92.fastwebnet.it] has joined #wesnoth-dev
20110629 19:28:05< timotei21> fendrin: the wesnoth install.
20110629 19:28:14< timotei21> remember, we support multiple installs:D
20110629 19:28:28< fendrin> ya
20110629 19:28:50-!- timotei21 is now known as timotei
20110629 19:29:23< boucman> hey all
20110629 19:29:28< boucman> zaroth: what's up ?
20110629 19:29:30< fendrin> hi boucman
20110629 19:36:10-!- zaroth_ [~zaroth@wesnoth/developer/zaroth] has joined #wesnoth-dev
20110629 19:36:49-!- zaroth [~zaroth@wesnoth/developer/zaroth] has quit [Ping timeout: 240 seconds]
20110629 19:42:33-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has joined #wesnoth-dev
20110629 19:42:34-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has quit [Changing host]
20110629 19:42:34-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20110629 19:46:56< fendrin> timotei: Please explain your question in more detail. We have multiple wesnoth versions installed...
20110629 19:47:52< timotei> fendrin: well, we can have multiple version installed. But at times like: when creating a wizard... we don't know in advance the wesnoth install used
20110629 19:48:01< timotei> (so we can create the good tag)
20110629 19:48:13< timotei> but I forgot that by default, wizards use the default install (if any)
20110629 19:48:14< timotei> :P
20110629 19:48:51< fendrin> Determining for which version a newly created something is meant is the first thing the wizard should ask if it is not clear from the context.
20110629 19:57:32-!- KaKiLa [~juanpi@xdsl-188-154-131-148.adslplus.ch] has joined #wesnoth-dev
20110629 19:58:20< KaKiLa> Hi all, have you check the call for papers of the workshop on video games in Argetina? I think some of you may want to put something about Wesnoth there?
20110629 20:03:26< boucman> I don't know if we have any dev that could go there...
20110629 20:04:33< KaKiLa> well, shadowmaster lives in Chile..we can make a collect to pay for his travel :D
20110629 20:05:23< KaKiLa> but a paper is needed in order to participate... I am sure some devs are in academia, right?
20110629 20:05:28< boucman> KaKiLa: post on #wesnoth-dev, you'll have better chance to get an answer there...
20110629 20:05:53< KaKiLa> boucman: is not here #wesnoth-dev?
20110629 20:05:59< boucman> huh
20110629 20:06:05< boucman> Imeant on the dev mailing list
20110629 20:06:06< boucman> my bad
20110629 20:06:25< KaKiLa> oh, I do not know if I am subscribed... I will try
20110629 20:07:40< timotei> KaKiLa: iirc you can send anyway an e-mail
20110629 20:08:31< KaKiLa> timotei: Ok, working on it
20110629 20:08:50< timotei> KaKiLa: https://mail.gna.org/listinfo/wesnoth-dev
20110629 20:15:25< KaKiLa> ok, done, thanks timotei and boucman
20110629 20:15:54< KaKiLa> btw, have you seen the progress in the interactive timeline? http://www.angelgris.com/wesnoth/
20110629 20:16:22-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Read error: Operation timed out]
20110629 20:16:48-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has joined #wesnoth-dev
20110629 20:16:48-!- stikonas [~and@ctv-79-132-175-26.vinita.lt] has quit [Changing host]
20110629 20:16:48-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20110629 20:26:18-!- KaKiLa [~juanpi@xdsl-188-154-131-148.adslplus.ch] has left #wesnoth-dev []
20110629 20:28:23-!- Crab_ [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20110629 20:31:46-!- EdB [~edb@89.82.194.231] has joined #wesnoth-dev
20110629 20:47:10< CIA-85> timotei * r50082 /trunk/utils/umc_dev/org.wesnoth/build.properties: eclipse plugin: remove 'changelog' from build
20110629 20:47:17< CIA-85> timotei * r50083 /trunk/utils/umc_dev/ (12 files in 7 dirs): eclipse plugin: Move the Installs stuff to it's own package
20110629 20:47:29< CIA-85> timotei * r50084 /trunk/utils/umc_dev/ (22 files in 14 dirs): eclipse plugin: Move the Project stuff into its own package
20110629 20:47:43< CIA-85> timotei * r50085 /trunk/utils/umc_dev/org.wesnoth/src/org/wesnoth/ (10 files in 7 dirs): eclipse plugin: Move 'PreprocessorUtils' to a better package
20110629 20:47:54< CIA-85> timotei * r50086 /trunk/utils/umc_dev/org.wesnoth/src/org/wesnoth/ (5 files in 4 dirs): eclipse plugin: Change the plain fields to getters
20110629 20:48:04< CIA-85> timotei * r50087 /trunk/utils/umc_dev/org.wesnoth/src/org/wesnoth/ (2 files in 2 dirs): eclipse plugin: Add missing proper update code for the install
20110629 20:48:14< CIA-85> timotei * r50088 /trunk/utils/umc_dev/org.wesnoth/src/org/wesnoth/preferences/Preferences.java: 
20110629 20:48:14< CIA-85> eclipse plugin: Cache the 'Paths' object instead of creating
20110629 20:48:14< CIA-85> it each time is needed
20110629 20:48:24< CIA-85> timotei * r50089 /trunk/utils/umc_dev/org.wesnoth/src/org/wesnoth/installs/ (WesnothInstall.java WesnothInstallsUtils.java): 
20110629 20:48:25< CIA-85> eclipse plugin: Make the fields in WesnothInstall private
20110629 20:48:25< CIA-85> and clean some code
20110629 20:48:35< CIA-85> timotei * r50090 /trunk/utils/umc_dev/ (9 files in 8 dirs): 
20110629 20:48:35< CIA-85> eclipse plugin: Refactor the SchemaParser to take in
20110629 20:48:35< CIA-85> account the install
20110629 20:52:57-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Ping timeout: 264 seconds]
20110629 20:53:48-!- loonybot [~loonybot@46.138.86.250] has joined #wesnoth-dev
20110629 20:53:48-!- loonybot [~loonybot@46.138.86.250] has quit [Changing host]
20110629 20:53:48-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20110629 20:58:20< boucman> Espreon: around ?
20110629 20:58:45-!- Gallaecio [~quassel@92.39.broadband12.iol.cz] has joined #wesnoth-dev
20110629 20:59:13-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Quit: SIGKILL]
20110629 21:11:09-!- MeccaGod [~majs@h146n3fls33o279.telia.com] has quit [Ping timeout: 250 seconds]
20110629 21:15:17-!- MeccaGod [~majs@h146n3fls33o279.telia.com] has joined #wesnoth-dev
20110629 21:19:57-!- MeccaGod [~majs@h146n3fls33o279.telia.com] has quit [Ping timeout: 276 seconds]
20110629 21:20:10-!- EdB [~edb@89.82.194.231] has quit [Remote host closed the connection]
20110629 21:23:43-!- MeccaGod [~majs@h146n3fls33o279.telia.com] has joined #wesnoth-dev
20110629 21:31:20< fendrin> Ivanovic: Is there an extra compile flag for the Pandora?
20110629 21:31:26< Ivanovic> no
20110629 21:31:47< fendrin> That is no real port, is it?
20110629 21:31:50< Ivanovic> this is the line i use for cmake:
20110629 21:31:52< Ivanovic> PATH=/home/nils/pandora-dev/arm-2011.03/usr/bin:$PATH CFLAGS="-DPANDORA -O2 -pipe -march=armv7-a -mcpu=cortex-a8 -mtune=cortex-a8 -mfpu=neon -ftree-vectorize -mfloat-abi=soft -ffast-math -fsingle-precision-constant -fno-inline-functions" CXXFLAGS="-DPANDORA -O2 -pipe -march=armv7-a -mcpu=cortex-a8 -mtune=cortex-a8 -mfpu=neon -ftree-vectorize -mfloat-abi=soft -ffast-math -fsingle-precision-constant -fno-inline-functions" cmake -DCMAKE_
20110629 21:31:53< Ivanovic> BUILD_TYPE=release -DENABLE_STRICT_COMPILATION=off -DCMAKE_TOOLCHAIN_FILE=/home/nils/pandora-dev/sdk_utils/PandoraToolchain.cmake -DPKG_CONFIG_EXECUTABLE=/home/nils/pandora-dev/arm-2011.03/bin/arm-none-linux-gnueabi-pkg-config -DSDL_CONFIG=/home/nils/pandora-dev/arm-2011.03/usr/bin/sdl-config -DLIBINTL_INCLUDE_DIR=/home/nils/pandora-dev/arm-2011.03/arm-none-linux-gnueabi/libc/usr/include/ -DPREFERENCES_DIR=wesnoth-1.9_userdata ../
20110629 21:31:55< Ivanovic> wesnoth-1.9.x
20110629 21:32:12< fendrin> That is a cross compiling action?
20110629 21:32:31< Ivanovic> that is: i am using the crosscompiler toolchain definition file, the call to cmake so that it is used is at the end, too: http://pastebin.com/dQP33U1w
20110629 21:32:41< Ivanovic> yeah, it is cross compiling action
20110629 21:33:14< Ivanovic> i just change the default preference dir, disable strict compilation and yeah, set a shitload of compiler flags
20110629 21:33:51< fendrin> How can I contribute stuff that is Pandora only? Something that should get away with the binary. Like a different preferences defaults for example.
20110629 21:35:06< Ivanovic> there is no switch yet, though i use a -DPANDORA, so you could ifdef
20110629 21:35:45< fendrin> Okay, that is fine for c++ but what do I do with wml cfg files?
20110629 21:35:55< Ivanovic> there is no such way to do this
20110629 21:35:58< Ivanovic> simple as that
20110629 21:36:17< Ivanovic> you could of course define a wml define that is set if the game was compiled with -DPANDORA
20110629 21:37:38< fendrin> I think we should discuss a general procedure if such things are necessary, regarding portation to different platforms.
20110629 21:38:03< Ivanovic> yes, makes sense to provide some standard/expected behaviour
20110629 21:38:22< Ivanovic> possibly syncing with cjhopman who is atm working on an android port which is meant to somehow be merged with trunk
20110629 21:38:48< fendrin> Isn't the iphone porter also doing an android port?
20110629 21:39:12< Ivanovic> the iphone porter was freaking quiet and not really handing patches upstream in any usable way
20110629 21:40:44< fendrin> Ivanovic: I think that the device that tracks the orientation in space on some smartphone or nintendo wi like controllers might be accessible through the joystick interface.
20110629 21:41:47< fendrin> Could be used for scrolling.
20110629 21:44:10< Ivanovic> would be some cool stuff
20110629 21:44:23< Ivanovic> though i don't have a device with accelometer at hand
20110629 21:45:25< fendrin> Ivanovic: One last question, can you give me a link to the documentation how one can change the joystick mode of the pandora?
20110629 21:47:28< Ivanovic> http://www.pandorawiki.org/Nubs#Nub_Configuration
20110629 21:48:12< Ivanovic> the default is that nub0 is mouse, nub1 mbuttons
20110629 21:49:40< fendrin> Ivanovic: Is there a solution from out of a c++ program?
20110629 21:49:52< fendrin> Piping in devices is somewhat hacky.
20110629 21:49:54< Ivanovic> nope, this is a solution to be done in a shellscript
20110629 21:50:05< fendrin> Yes I know.
20110629 21:50:05< Ivanovic> in theory you switch the modes when starting the program
20110629 21:50:13< Ivanovic> no idea how to switch dynamically ingame
20110629 21:50:21< Ivanovic> you could ask if someone in #openpandora knows this
20110629 21:51:11-!- cjhopman [~chris@wesnoth/developer/cjhopman] has quit [Read error: Operation timed out]
20110629 21:53:28-!- cjhopman [~chris@wesnoth/developer/cjhopman] has joined #wesnoth-dev
20110629 21:56:29-!- MeccaGod [~majs@h146n3fls33o279.telia.com] has quit []
20110629 21:59:13-!- Johannes13 [~Johannes@unaffiliated/johannes13] has quit [Read error: Connection reset by peer]
20110629 22:18:23< zaroth_> boucman: Crab_: I'm here now
20110629 22:18:27-!- zaroth_ is now known as zaroth
20110629 22:18:44-!- Crendgrim [~crend@77-22-254-233-dynip.superkabel.de] has quit [Remote host closed the connection]
20110629 22:19:27< boucman> so, what's up today ?
20110629 22:23:00< zaroth> I started moving the statechart classes to trunk/src/statechart 
20110629 22:23:39< zaroth> but since now Crab_ is here, I'd like to hear his opinion about the state machine in there ;-)
20110629 22:24:21-!- Gallaecio [~quassel@92.39.broadband12.iol.cz] has quit [Remote host closed the connection]
20110629 22:25:08< zaroth> (I'm not very convinced as to its finalness, but as you said, we can find it out maybe after it's in trunk)
20110629 22:25:49< boucman> Crab_: comments ?
20110629 22:26:11< Crab_> boucman: will comment a bit later, looking at it atm
20110629 22:26:15< Crab_> (at the code)
20110629 22:27:12< boucman> sure
20110629 22:28:41-!- Sytyi [~sytyi@207-18-133-95.pool.ukrtel.net] has quit [Quit: Leaving.]
20110629 22:29:57< zaroth> there are not all the states implemented, but it should make the approach I took apparent
20110629 22:30:06< zaroth> you're very welcome to suggest another
20110629 22:30:39< zaroth> also please note, that the (possibly expensive) context<active>().game_config() call can be made once, and later referenced from the resource:: (const) reference
20110629 22:31:46< zaroth> (or maybe not put inside the "active" state at all, but then some additional mechanisms to restart its state should be possibly put in place)
20110629 22:33:47< zaroth> also, there is this restriction that if it's a composite state (such as game_instance, active) or a state machine, they have to be put together with their starting state in the same translation unit
20110629 22:33:54< zaroth> otherwise it won't compile at all
20110629 22:40:01< zaroth> I also tried to think of the states in terms of underlying config state
20110629 22:40:06< zaroth> not the screens or anything else
20110629 22:43:01-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20110629 22:52:17< Crab_> ok, taken a quick look through the code
20110629 22:53:49 * zaroth bites lips in anticipation
20110629 22:55:55< Crab_> I see that it's good, but it's only half of what we need
20110629 22:56:21< Crab_> i.e. the state transfers between adjacent states via config object parameter to events works, and it's good
20110629 22:57:00< Crab_> but, we somehow need to implement state transfers between non-adjacent states, i.e. let 'start sp campaign' affect multiple transitions at once
20110629 22:57:28< Crab_> do you think it'll be a good idea to test that mechanism out-of-wesnoth-source too ?
20110629 22:57:43< zaroth> wait a second
20110629 22:57:58< zaroth> I don't understand why should we want to skip a state
20110629 22:58:13< zaroth> can't we make the two transitions one after another implicitly?
20110629 22:58:21< zaroth> and automatically, just by sending two events at once?
20110629 22:58:30-!- deekay [~dk@89-79-35-236.dynamic.chello.pl] has joined #wesnoth-dev
20110629 22:58:30-!- deekay [~dk@89-79-35-236.dynamic.chello.pl] has quit [Changing host]
20110629 22:58:30-!- deekay [~dk@wesnoth/developer/dragonking] has joined #wesnoth-dev
20110629 22:58:41< Crab_> zaroth: what if we don't know the second transition ?
20110629 22:59:00< zaroth> ?
20110629 22:59:05< Crab_> zaroth: i.e.,we are in A, we want to do A->B, do all the stuff which is required in B, and then do B->C
20110629 22:59:25< Crab_> zaroth: and, while in A, we want to affect the configs for B and C
20110629 22:59:45-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20110629 23:00:05< Crab_> but, we don't want to influence B insofar the details for the transition are concerned - we want B to do it's work fully, and, only if we transition to C, we want our modifications to C to take place
20110629 23:00:06< zaroth> well, if we post two events in A and they get queued
20110629 23:00:32< Crab_> but what if the details/config of the second event are only known to the internals of state B ?
20110629 23:00:44< Crab_> i.e., some user input is required at B
20110629 23:01:10< Crab_> or, some checks and cleanup are required in B, and known only to B
20110629 23:01:10-!- Wurmish [46f6e342@gateway/web/freenode/ip.70.246.227.66] has joined #wesnoth-dev
20110629 23:01:11< zaroth> well, we could create events "provide_information_to_state_foo"
20110629 23:01:23< zaroth> which wouldn't trigger transitions, but would add new information
20110629 23:01:26< Crab_> how would they work ?
20110629 23:02:14< boucman> Crab_: do you have a practical case for that ?
20110629 23:02:20< zaroth> and they would be deferred to be processed later in all states but the destined one
20110629 23:02:46< zaroth> and would be removed from the queue entirely if we go back to main menu (i.e. main menu wouldn't defer them, as simple as that)
20110629 23:02:55< zaroth> oops, loaded_headers not main menu
20110629 23:04:14< Crab_> boucman: 'new sp campaign' button, for example
20110629 23:04:16< zaroth> to avoid issues with many queued events affecting one another, we could make that only the last received one takes effect, i.e. cancels the effects of all previous ones
20110629 23:04:30< zaroth> (although I'm not sure if it's good design)
20110629 23:04:35< Crab_> zaroth: ok, understood. this is quite similar to the 'stack of callbacks' approach I've tried to show earlier
20110629 23:05:03< Crab_> zaroth: what are the advantages/disadvantages of your method vs the 'callback stack' ?
20110629 23:05:41< zaroth> I think I should reread what you wrote about callback stack, because I'm afraid I don't remember it fully
20110629 23:05:49< zaroth> a moment please
20110629 23:05:56< boucman> Crab_: we had thought of that, it's one transition : <basic config> => <game details known>
20110629 23:06:12< boucman> the idea was to stay in basic config until we have all the info
20110629 23:07:24< Crab_> boucman: so, basically, skipping the intermediate stages if we are sure they aren't required ?
20110629 23:08:21< boucman> a state corresponds not a state of the config, not to a screen, so we move on when we have all the info, (whatever way we got the info doesn't matter)
20110629 23:08:26< boucman> so we don't really skip them
20110629 23:09:05-!- Appleman1234 [~Appleman1@CPE-60-226-179-130.qld.bigpond.net.au] has quit [Remote host closed the connection]
20110629 23:09:28< Crab_> zaroth: in my approach, I haven't used the deferred events (I haven't known about them) and maintained a std::stack< state_callback > callback_stack , and, on start of each state, I launched the top state_callback from the stack, letting it modify the config object of current state. if the callback decided to remove itself from the stack, calling the next top callback
20110629 23:09:59< Crab_> zaroth: this is quite similar to the deferred events you've described, only with managing them by hand and letting them hold their own state.
20110629 23:10:23< zaroth> yes, I'm repeating read through your http://pastebin.mozilla.org/1258253 now 
20110629 23:10:53< zaroth> one problem I'm seeing with this is the fact that we don't have the access to "upstream_transition_event" in a state constructor
20110629 23:11:06< zaroth> but it probably can be easily solved
20110629 23:11:25< zaroth> by putting it somewhere  externally and asking that entity
20110629 23:11:59-!- ancestral [~ancestral@67-6-61-230.mpls.qwest.net] has joined #wesnoth-dev
20110629 23:12:09-!- negusnyul_ [~negusnyul@dsl4E5CC8DB.pool.t-online.hu] has quit [Remote host closed the connection]
20110629 23:14:26-!- 30BAAGWMJ [~iwaim@rasteenie.alib.jp] has quit [Ping timeout: 244 seconds]
20110629 23:21:55< zaroth> so well, I think that by using callbacks we lose a bit flexibility concerning movement on the statechart graph, but we probably gain ease of implementing alternative solutions of traversing this graph (e.g. Lua)
20110629 23:22:20< Crab_> zaroth: what kind of flexibility is lost ?
20110629 23:22:36-!- Upthorn [~ogmar@c-76-105-37-245.hsd1.ca.comcast.net] has quit [Ping timeout: 276 seconds]
20110629 23:22:38< zaroth> before I answer, let me ask you one question
20110629 23:22:55< Crab_> zaroth: and do we need them at all ? i.e., as I understand, boucman says that stuff like 'new campaign' can be made without them
20110629 23:22:57< Crab_> ok
20110629 23:23:36< zaroth> do you consider moving one step back in the statechart graph important? (e.g. from map_settings_known to campaign_defines_known?)
20110629 23:24:23< zaroth> because right now I haven't seen other way of implementing this than going back to loaded_headers (which reloads config on enter) and going through all the previous states, which is terribly inefficient imho
20110629 23:24:42< zaroth> but I couldn't think of a better way of undoing things with one global config object, into everything is dumped
20110629 23:24:48< zaroth> but please correct me if I'm wrong
20110629 23:25:32< zaroth> especially undoing the step no_specifics_known->campaign_defines_known seems to be hard to do without reloading everything
20110629 23:26:10< Crab_> if a user in the MP lobby selects 'create new game', select map, starts to selects map-specific settings, then starts to wait for players, then presses cancel - where should he return ?
20110629 23:26:30< zaroth> that's my question
20110629 23:26:48< zaroth> in the current graph he can only return to loaded_headers, i.e. lobby
20110629 23:26:55< zaroth> and would see wesnoth reloading config on the way
20110629 23:27:14< zaroth> (though probably not very long, since every big include would be behind define guards)
20110629 23:27:51< zaroth> (I even think that WESNOTH_CORE should be hidden behind a define guard, and implicitly defined by a campaign header)
20110629 23:28:00< zaroth> (and maybe disabled by something like load_core=no)
20110629 23:28:21< zaroth> (that would speed up all these reload times prior to choosing the campaign)
20110629 23:29:00-!- Unnheulu_ [~ieuan@host86-147-248-76.range86-147.btcentralplus.com] has quit [Quit: Ex-Chat]
20110629 23:29:11< zaroth> it's certainly possible, since I checked and none of macros are needed in the lobby or something early, and even if they were, they could be put in a separate file
20110629 23:29:11-!- vultraz [~chatzilla@124.109.10.222] has quit [Ping timeout: 258 seconds]
20110629 23:29:22< zaroth> other than core/
20110629 23:30:18< Crab_> as of now, I don't understand what's the difference of going 'back' on the state chart vs going 'forward'
20110629 23:32:17< zaroth> no_specifics_known - you have barely anything in the global game_config object, user chooses campaign, an event is generated, and based upon the data in this event campaign defines are chosen and whole config reloaded, now you have lots of  data in the global game_config object
20110629 23:32:30< zaroth> how do you go back to no_specifics_known upon cancel?
20110629 23:33:05< zaroth> (you are in campaign_defines_known now)
20110629 23:34:05< Crab_> I'd change current state to 'no_specifics_known' and, as a side effect, let the global game_config object be replaced with a new one. let all interested observers know that the game_config object has changed.
20110629 23:34:31< Crab_> also, we'd need to know what to do with config object of the earlier state
20110629 23:35:32< Crab_> since we want to restore the earlier state somehow, yet we will probably want to keep some of the modifications done at the earlier state
20110629 23:35:44< zaroth> well, that can be done easy enough by defining a return event and reloading the config in reaction to this return event
20110629 23:36:23< Crab_> so, when you go from A to B, you'd pass it a parameter for the return event, as well ?
20110629 23:36:52< Crab_> i.e. go_from_no_specifics_known_to_next_stage(config for_them, config to_return_back_here)  ?
20110629 23:37:38< zaroth> that's getting a bit unKISSy maybe, but yes, I think so
20110629 23:37:51< zaroth> the go_back config would be probably just some defaults for the dialogs
20110629 23:38:10< Crab_> yet it's not enough as well, since there are still changes in yet earlier states
20110629 23:38:18< Crab_> i.e. if cancel is pressed 2 times :)
20110629 23:40:08-!- zaroth [~zaroth@wesnoth/developer/zaroth] has quit [Read error: Operation timed out]
20110629 23:41:50-!- zaroth [~zaroth@wesnoth/developer/zaroth] has joined #wesnoth-dev
20110629 23:42:29< zaroth> hmm... the Boost.Statechart has these two things: shallow_history and deep_history, which keep track of past transitions and their events afaik
20110629 23:42:38< zaroth> let me have a look if they suit our needs in this case
20110629 23:42:43< Crab_> ok
20110629 23:42:49< Crab_> meanwhlie, some thoughts:
20110629 23:43:46< zaroth> (by the way, I'm a big fan of KISS approach as well, so if you have a simpler idea for designing something at any moment, please bash me with it until I see its simplicity :P)
20110629 23:43:53< Crab_> each state, when we leave it, provides us with two things: (1) 'changes to global status which we have done with this state', and (2) 'ou input to next state'
20110629 23:44:09< Crab_> *our input
20110629 23:45:19< Crab_> if we go from A to B, and then go back from B to A with a cancel-type of transition, we should return at least to 'BASE+A1'
20110629 23:46:39< Crab_> our input to next state is nicely handled by that cfg we pass around
20110629 23:48:41-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20110629 23:51:10-!- ancestral [~ancestral@67-6-61-230.mpls.qwest.net] has quit [Quit: i go sleeps kthxbai]
20110629 23:52:09< Crab_> there's more trouble in interaction of cancels and auto-handled states
20110629 23:52:25-!- ancestral [~ancestral@67-6-61-230.mpls.qwest.net] has joined #wesnoth-dev
20110629 23:52:33-!- ancestral [~ancestral@67-6-61-230.mpls.qwest.net] has quit [Client Quit]
20110629 23:52:39< Crab_> if there's a state which is handled without user interaction, there's little point to skipping back to it, since we'll immediately auto-handle it and move forward again
20110629 23:53:56< Crab_> zaroth: what if we introduce a concept of savepoints, allowing to snapshot our status to cfg at some predefined moment and go back to that predefined moment via cancel button ?
20110629 23:54:03-!- mjs-de [~mjs-de@f053188070.adsl.alicedsl.de] has quit [Ping timeout: 255 seconds]
20110629 23:54:41-!- grzywacz [~grzywacz@89-78-52-133.dynamic.chello.pl] has joined #wesnoth-dev
20110629 23:54:41-!- grzywacz [~grzywacz@89-78-52-133.dynamic.chello.pl] has quit [Changing host]
20110629 23:54:41-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20110629 23:55:13< zaroth> Crab_: that's the kind of flexibility loss I was talking about
20110629 23:55:21-!- mjs-de [~mjs-de@f053188070.adsl.alicedsl.de] has joined #wesnoth-dev
20110629 23:55:21< zaroth> unless the callback function is keeping state
20110629 23:55:30< zaroth> it will dumbly press forward
20110629 23:55:39< Crab_> callback is keeping state
20110629 23:55:42< zaroth> while the event is executed once
20110629 23:55:50< zaroth> it is?
20110629 23:56:30< Crab_> zaroth: it should. you might remember one of my earlier examples - 'start campaign 1, cancel to main menu, start campaign 2,...'
20110629 23:56:45< Crab_> zaroth: it relied on doing different things in the same 'main menu' state
20110629 23:56:57< zaroth> oh. right.
20110629 23:57:13 * zaroth isn't happy that things just got even more complicated
20110629 23:57:41< Crab_> basically, the existence of 'cancel' button implies that we are in some sort of conversation/flow
20110629 23:57:54< Crab_> if it's not the 'go to the very start' kind of cancel
20110629 23:58:44< Crab_> so, the flow would need to know how to  handle cancellation
--- Log closed Thu Jun 30 00:00:52 2011