--- Log opened Mon Jun 28 00:00:08 2010
--- Day changed Mon Jun 28 2010
20100628 00:00:08-!- Bob_The_Mighty [~chatzilla@cpc8-brig15-2-0-cust40.know.cable.virginmedia.com] has quit [Ping timeout: 265 seconds]
20100628 00:00:22< freim> making the mountains was indeed a lot of work
20100628 00:01:15< freim> I think it was some of the first multihex terrain, so we was exploring the "wml mechanics" of it also at the same thime
20100628 00:01:24< freim> where*
20100628 00:01:35-!- happygrue [~George@wesnoth/developer/wintermute] has quit [Ping timeout: 265 seconds]
20100628 00:01:37< zookeeper> freim, yeah, i just meant better than the current horrible desert mountains
20100628 00:02:04< zookeeper> that is, something which wouldn't make every mapmaker avoid the terrain like plague
20100628 00:02:10< eleazar> those desert mountains definitely need to go away
20100628 00:02:40< freim> I'll see if I can make something that doesn't look like beehives :o
20100628 00:09:25< eleazar> does the human, mountain, snow villiage work for anybody else?
20100628 00:13:06-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit []
20100628 00:15:39-!- happygrue [~George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20100628 00:18:39-!- happygrue_ [~George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20100628 00:19:28-!- happygrue__ [~George@c-67-175-144-211.hsd1.in.comcast.net] has quit [Ping timeout: 265 seconds]
20100628 00:19:28-!- Johannes13_ [~Johannes@unaffiliated/johannes13] has quit [Ping timeout: 265 seconds]
20100628 00:20:37-!- knotwork__ [~markm@142.177.178.67] has quit [Ping timeout: 260 seconds]
20100628 00:22:22-!- happygrue [~George@wesnoth/developer/wintermute] has quit [Ping timeout: 265 seconds]
20100628 00:26:56-!- happygrue [~George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20100628 00:30:35-!- happygrue_ [~George@wesnoth/developer/wintermute] has quit [Ping timeout: 265 seconds]
20100628 00:33:16-!- knotwork__ [~markm@142.177.235.95] has joined #wesnoth-dev
20100628 00:36:34-!- phlaem [~a@e178087037.adsl.alicedsl.de] has quit [Quit: Leaving]
20100628 00:44:49-!- happygrue_ [~George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20100628 00:48:04-!- happygrue [~George@wesnoth/developer/wintermute] has quit [Ping timeout: 245 seconds]
20100628 00:49:43-!- Tigge_ is now known as Tigge
20100628 00:50:02< CIA-87> eleazar * r43757 /trunk/data/core/images/terrain/village/ (72 files): altered image diminsions on most villages, since there is no longer a reason for them to have very large pixel dimensions. Also cleaned up a few glitches. This should help avoid useless image cacheing.
20100628 00:51:41-!- un214 [~quassel@75.45.4.19] has joined #wesnoth-dev
20100628 00:52:14-!- happygrue [~George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20100628 00:55:34-!- happygrue_ [~George@wesnoth/developer/wintermute] has quit [Ping timeout: 245 seconds]
20100628 00:57:34-!- happygrue_ [~George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20100628 00:58:12-!- eleazar [~eleazar@ppp-70-226-197-86.dsl.spfdil.ameritech.net] has left #wesnoth-dev []
20100628 01:00:59-!- happygrue [~George@wesnoth/developer/wintermute] has quit [Ping timeout: 245 seconds]
20100628 01:10:46-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Quit: ...]
20100628 01:11:14-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20100628 01:13:19-!- Gambit [~quassel@pa-67-234-73-7.dhcp.embarqhsd.net] has quit [Read error: Connection reset by peer]
20100628 01:13:42-!- Gambit [~quassel@pa-67-234-73-7.dhcp.embarqhsd.net] has joined #wesnoth-dev
20100628 01:17:51-!- Bob_The_Mighty [~chatzilla@cpc8-brig15-2-0-cust40.know.cable.virginmedia.com] has joined #wesnoth-dev
20100628 01:23:23-!- Bob_The_Mighty [~chatzilla@cpc8-brig15-2-0-cust40.know.cable.virginmedia.com] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.4/20100611143157]]
20100628 01:27:19 * Upth is now awake and compiling alink's changes.
20100628 01:28:17-!- Upth is now known as orn
20100628 01:28:18-!- Upthorn is now known as Upth
20100628 01:28:19-!- orn is now known as Upthorn
20100628 01:31:28-!- Bob_The_Mighty [~chatzilla@cpc8-brig15-2-0-cust40.know.cable.virginmedia.com] has joined #wesnoth-dev
20100628 01:34:42< Bob_The_Mighty> Anyone know why a certain png file might not show in game? The code is right, the images are not.
20100628 01:34:54< Bob_The_Mighty> I mean, something to do with palette or file format?
20100628 01:37:17-!- un214 [~quassel@75.45.4.19] has quit [Remote host closed the connection]
20100628 01:48:30< Bob_The_Mighty> okay, what about this: can you meddle with a unit's profile via storing to set it's portrait to the right with ~RIGHT() ?
20100628 01:51:38-!- orn [ogmar@adsl-75-26-161-110.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100628 01:51:38-!- Upthorn [~ogmar@adsl-75-26-178-38.dsl.scrm01.sbcglobal.net] has quit [Disconnected by services]
20100628 01:51:43-!- orn is now known as Upthorn
20100628 01:52:23-!- Upth [~ogmar@adsl-75-26-178-38.dsl.scrm01.sbcglobal.net] has quit [Ping timeout: 240 seconds]
20100628 01:53:00-!- Upth [~ogmar@75.26.161.110] has joined #wesnoth-dev
20100628 02:04:55< Upthorn> now here's something odd in the log
20100628 02:05:30< Upthorn> okay I just closed it because I am dumb so this might not be an exact quote, but
20100628 02:05:36< Upthorn> "Invalid WML found: image terrain/ missing .png extension."
20100628 02:06:47< alink> Upthorn: yes that's me. Not the error, but the fact that it's now reported
20100628 02:07:37< alink> Upthorn: note that silene also changed image.cpp to include the vector::resize thing, plus some other changes
20100628 02:08:21< alink> Bob_The_Mighty: If you think it comes from the png file, just replace it by a mainline one (with same dimensions, just to be sure)
20100628 02:08:46< alink> Bob_The_Mighty: and to check the code, perhaps use a mainline url in the WML key
20100628 02:08:53< Bob_The_Mighty> Actually, I think it might be my file structure 
20100628 02:09:18< alink> ah the binary path stuff ?
20100628 02:10:05< Bob_The_Mighty> trying to get images to show in a campaign. i thought it worked the same as for scenarios, seems not to. not sure.
20100628 02:11:04< alink> you mean story images, profile images or units/terrain sprites ?
20100628 02:11:53< alink> eleazar: thanks for the village cleaning, this waste of pixels annoyed me since a looong time
20100628 02:13:15< alink> freim: no, sprite sheets would not help for this specific problem (waste of space by empty areas)
20100628 02:13:31< Bob_The_Mighty> just items and scenery, its odd
20100628 02:13:55< alink> in fact, it would be even more complex: a bin packing problem instead of simple framing
20100628 02:14:27< alink> Bob_The_Mighty: ah scenery, do you use "../" in your file paths?
20100628 02:14:49< alink> Bob_The_Mighty: also check about the auto addition of "terrain/" prefix for terrain images
20100628 02:16:18< Bob_The_Mighty> I am using images exactly like I have in my add-ons before, only difference is this is a campaign, rather than a single scenario or era. Why is it different?
20100628 02:17:11< alink> not sure but the use [binary_path] is different from MP I think
20100628 02:17:25< alink> http://wiki.wesnoth.org/BinaryPathWML
20100628 02:17:52< Bob_The_Mighty> this is an mp campaign
20100628 02:18:21< alink> ah, ok sorry, then i don't know if my remark is still valid :-/
20100628 02:18:52< alink> Bob_The_Mighty: when using debug, do you see the "working peasant" instead of your image
20100628 02:19:17< Bob_The_Mighty> i don't use debug really, i will try
20100628 02:19:44< alink> also check log output to see if a file is missing
20100628 02:19:59-!- Bocom [~Bocom@c-b7cfe255.013-31-6b736412.cust.bredbandsbolaget.se] has quit [Quit: Lämnar]
20100628 02:20:02< alink> if you see an incorrect url there, you will see what is wrong
20100628 02:20:27< alink> --log-debug:filesystem is also useful to debug binary path issues
20100628 02:20:29< Bob_The_Mighty> I see no peasant sign. And the log says: error display: could not open image 'items/ambusher-statue-e.png'
20100628 02:20:49< alink> and this is the correct filename ?
20100628 02:20:53< Bob_The_Mighty> yeah
20100628 02:21:19< alink> in images/ directory
20100628 02:21:26< Bob_The_Mighty> It is not the only custom image to not show.
20100628 02:21:56< Bob_The_Mighty> the ambusher statue image is in: images/items/
20100628 02:22:15< alink> with debug, the peasant sign should be visible where the item is supposed to be
20100628 02:23:32< Bob_The_Mighty> It isn't
20100628 02:24:12< alink> If this work when using a mainline url like "items/altar-evil.png" then it's surely a binary path problem
20100628 02:26:29< alink> shadowmaster: btw about 64bpp bit depth PNG file. This is what my file manager says, but I don't find this info in Gimp to verify
20100628 02:26:48< Bob_The_Mighty> mainline images show ok
20100628 02:26:59< alink> shadowm: and it says 32bpp for most of them
20100628 02:28:30< shadowm_laptop> alink, Bob_The_Mighty : no, [binary_path] is the same fo rmultiplayer
20100628 02:29:18< shadowm_laptop> make sure that which ever images/ dir that has your image path belongs to a registered [binary_path]
20100628 02:32:22-!- happygrue_ is now known as happygrue
20100628 02:33:13< Bob_The_Mighty> I have tried both {PLACE_IMAGE items/clay.png 26 19} and {PLACE_IMAGE "~add-ons/The_Altaz_Mariners/images/items/clay.png" 26 20} but niether show. Both generate: error display: could not open image 'items/clay.png'
20100628 02:33:43< alink> ah it's --log-debug=filesystem  and it show in which directory the game search your images
20100628 02:34:27< Bob_The_Mighty> I'm noot sure what you mean
20100628 02:35:10< alink> Bob_The_Mighty: it's a parameter to use when launching wesnoth from the command line, but if you are on windows, forget it
20100628 02:35:22< Bob_The_Mighty> I am on windows
20100628 02:36:18 * Upthorn is having inconfidence in his timing method, but preliminary results show that recent image caching related changes have decreased his test-case worktime by 2 minutes +
20100628 02:36:42< alink> Upthorn: good
20100628 02:37:10< Upthorn> I'm running it again to make sure that I'm not accidentally counting "reading files and creating cache" on one but not both.
20100628 02:38:49< alink> mmh you can change the log level in game by using ":log debug filesystem" and then ":refresh". but still need to explain where to fins the stderr.txt file :-/
20100628 02:39:02< alink> s/fins/find
20100628 02:40:54< Bob_The_Mighty> i use the stderr file
20100628 02:41:47< alink> then, in game  try ":log debug filesystem", then "refresh", then search clay.png and look which directory are scanned
20100628 02:42:30< alink> even if the answer is probably obvious: not the good ones
20100628 02:43:35< Bob_The_Mighty> hmm, i think I've realised what the problem is. I was testing the scenario seperately (via the use of allow_new_game=yes) not part of the campaign. I guess that will be it.
20100628 02:44:42 * alink doesn't know allow_new_game, but yes using the simplest/cleanest way to verify something weird is often advised
20100628 02:45:15< Bob_The_Mighty> sorry about that
20100628 02:45:37-!- ancestral [~ancestral@97-116-120-9.mpls.qwest.net] has joined #wesnoth-dev
20100628 02:45:48< alink> and the "working peasant" seems to work here with a broken PLACE_IMAGE + debug
20100628 02:47:39< Bob_The_Mighty> sorry for wasting your time, alink. but thanks for trying to help. i didn't see the peasant at work image at all.
20100628 02:47:47< alink> ah I see, you must start with debug
20100628 02:47:49< Bob_The_Mighty> i guess cos the path was correct
20100628 02:48:03< alink> or use ":debug" then ":refresh" and you will see the peasant
20100628 02:48:22< alink> Bob_The_Mighty: np, I learned few things :-)
20100628 02:52:32< alink> mmh using refresh seems to break unit animations :-/
20100628 02:52:40< alink> possibly new
20100628 02:52:50-!- happygrue_ [~George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20100628 02:56:33-!- happygrue [~George@wesnoth/developer/wintermute] has quit [Ping timeout: 265 seconds]
20100628 02:58:49< alink> silene: I suspect r43754 to causes memory corruption after using :refresh in-game
20100628 02:59:37< alink> symptoms: UI element replacing idle frames, fight anims triggering assert with impossible SDL surfaces info
20100628 03:00:15< alink> like urface_lock::surface_lock(const surface&): Assertion `(surface_->flags & 0x00004000) == 0 && surface_->format->BytesPerPixel == 4 && surface_->format->Rmask == 0xFF0000u && (surface_->format->Amask | 0xFF000000u) == 0xFF000000u'
20100628 03:01:25< alink> also had some idle frames switched between 2 units, which gives funny results :-)
20100628 03:03:11-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz]
20100628 03:03:44-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20100628 03:11:19< Upthorn> okay yeah
20100628 03:11:36< Upthorn> I'm seeing a difference of 4 minutes 55 seconds +/- 5 seconds
20100628 03:12:31< Upthorn> this is 2minutes 40seconds on the new version, 7 minutes 35 seconds on the old
20100628 03:12:51< Upthorn> rounded to the nearest 5s mark to account for my slow reaction time
20100628 03:19:23< alink> Upthorn: still abnormally too long. Does other savegames take so much time ? 
20100628 03:20:10< alink> perhaps there is something special about your test case
20100628 03:22:10-!- happygrue_ [~George@wesnoth/developer/wintermute] has quit [Ping timeout: 265 seconds]
20100628 03:22:49< Upthorn> alink: all my savegames on trunk have always taken about this long
20100628 03:23:05< Upthorn> well 2-3 minutes
20100628 03:23:32< Upthorn> but it was only recently that they got up to 7 minutes +
20100628 03:24:36< alink> with relatively modern hardware ? weird
20100628 03:26:11< alink> If on windows and build by yourself, maybe compare with the official binary 
20100628 03:26:39< Upthorn> is there an official binary of trunk?
20100628 03:27:19< alink> yes somewhere, zookeeper uses it, ask him
20100628 03:27:53< Upthorn> I'm building debug (fast), rather than release, so my times are probably a lot slower than the official binary will be
20100628 03:28:49< alink> ah yes you need to use optimized build to have meaningfull timing
20100628 03:29:26< alink> otherwise you risk optimizing things that the compiler would have done by itself
20100628 03:30:31< alink> even if a very slow area often stay (relatively) slow in optimized build
20100628 03:47:40-!- akeenanr_ [~quassel@cpe-74-74-148-190.rochester.res.rr.com] has joined #wesnoth-dev
20100628 03:48:59< shadowm_laptop> alink, Upthorn : http://files.wesnoth.org/hidden/win
20100628 03:50:02< alink> shadowm: thanks, I assume it's optimized build ?
20100628 03:50:17< Upthorn> that should be pre-change
20100628 03:50:18< shadowm_laptop> no idea
20100628 03:50:29< Upthorn> judging by the timestamp
20100628 03:50:49< alink> the revision is visible in titlescreen too
20100628 03:50:57< Upthorn> which is actually helpful if it is optimized
20100628 03:51:32< Upthorn> because that means I only have to build one release version
20100628 03:51:50< Upthorn> important, because full rebuild takes hours.
20100628 03:51:58< alink> mmmh or not, there is a flag to set somewhere to show the revision ?
20100628 03:52:11< alink> or we stopped doing that
20100628 03:52:16< Upthorn> and modifying image.hpp triggers basically a full rebuild
20100628 04:05:53-!- Elvish_Pillager [~eli@71-10-224-192.dhcp.oxfr.ma.charter.com] has quit [Quit: Hi! I'm a quit message virus vaccine. If you see a quit message virus, don't replace your quit message with it!]
20100628 04:13:29< CIA-87> eleazar * r43758 /trunk/data/core/images/terrain/ (8 files): new grid transitions to properly finish the edge of the outside edges of the grid. Probably needs to be wired into the code.
20100628 04:24:07-!- meric [~Eric@124-170-33-254.dyn.iinet.net.au] has joined #wesnoth-dev
20100628 04:30:59< alink> hum, eleazar should have waited the new wiring before removing old images(which broke grid)
20100628 04:31:20< alink> also i fails to see the point of grid's transitions
20100628 04:31:53< alink> and there is a small cost to use 6 images instead of 1 (or 2)
20100628 04:32:56< alink> and he's probably not aware that the top/bottom thing was a new feature fixing "grid on top of foreground terrain" 
20100628 04:36:33< alink> I guess we could use normal grid everywhere except at "the edge of the outside edges"
20100628 04:37:41< alink> but that means checking where we are for each hex and generate transitions etc and/or find a place to cache that
20100628 04:38:11< alink> pfff seems to be a lot of work for a minor detail :-/
20100628 04:39:04< alink> I propose to use grid everywhere, even on the outside hexes. That's an easy fix for the aesthetic problem :-) 
20100628 04:45:45-!- Ivanovic_ [~ivanovic@dtmd-4db2b7be.pool.mediaWays.net] has joined #wesnoth-dev
20100628 04:46:23-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 240 seconds]
20100628 04:47:43-!- Ivanovic_ is now known as Ivanovic
20100628 04:52:55< Upthorn> okay, so when I compiled as release, (which the thing shadowmaster linked clearly was not, based on the filesize), everything took a ton less time
20100628 04:53:23< Upthorn> but the work was not in vein, because I need to run it under debug configuration when I am actively working
20100628 04:53:45< Upthorn> and I haven't tested to see how the previous revision performs in release
20100628 04:55:19< Upthorn> err in vain
20100628 04:55:40 * Upthorn compiles 43749 as release for comparison
20100628 04:56:41< alink> so it's now 7 minutes 35 seconds - 1 ton ? That gives less than 50 degrees Kelvin ;-p 
20100628 04:58:14< alink> but it could be interesting to verify if the slowdown that zookeeper mentioned on his build are on an optimized or debug build
20100628 05:08:37-!- eleazar_ [~eleazar@ppp-70-226-197-86.dsl.spfdil.ameritech.net] has joined #wesnoth-dev
20100628 05:08:54< eleazar_> alink: around?
20100628 05:09:03< alink> yes
20100628 05:09:12< eleazar_> apparently i missunderstood how the grid works
20100628 05:09:34< alink> eleazar_: well, it was a fresh new feature
20100628 05:09:43< eleazar_> what's it supposed to do?
20100628 05:10:10< alink> it's allow to display grid on top of foreground terrain but under units
20100628 05:10:28< eleazar_> ah
20100628 05:10:36< alink> because foreground terrain was often hiding grid a bit
20100628 05:10:59< alink> the unit pass trough the grid using the same trick as top/bottom ellipse
20100628 05:11:35< eleazar_> shouldn't it have been spit at a different point then?
20100628 05:12:18< alink> um, maybe, have you a unit where it matters ?
20100628 05:12:27< eleazar_> let me see if i can put it back...
20100628 05:12:44< alink> units rarely spread to south hexes
20100628 05:13:13< alink> and cutting to the corners may look less obvious if/when an artifact is visible
20100628 05:13:31< eleazar_> so it's the plain "grid.png" that isn't used
20100628 05:13:39< alink> (because of the exact hexagonal shape of grid)
20100628 05:14:03< alink> eleazar_: yes I should have removed grid.png
20100628 05:14:30< alink> I let it there to allow easy test by changing game_config.cfg
20100628 05:15:17< alink> this grid feature is still a bit experimental 
20100628 05:15:49< alink> eleazar_: btw what do you think about simply extend the grid outside the map ?
20100628 05:16:16< eleazar_> you mean cover everthing with the basic grid?
20100628 05:16:33< eleazar_> even the off map areas
20100628 05:16:35< eleazar_> ?
20100628 05:17:07< alink> yes, to fix your missing grid edge problem
20100628 05:17:36< alink> hum now I remember that some other dev didn't want that
20100628 05:17:44< alink> like mordante, I think
20100628 05:18:01< alink> but I don't remember the logic behind it
20100628 05:18:18< CIA-87> eleazar * r43759 /trunk/data/core/images/terrain/ (grid-bottom.png grid-top.png): replaceing actually functional grid pieces i shouldn't have deleted.
20100628 05:18:34< eleazar_> i think that would be a little better than the status quo,
20100628 05:18:56< eleazar_> as it it the grid makes a distinction that's useful
20100628 05:19:23< eleazar_> but it doesn't necessarily make the distinction is a visible way, i.e. it's unrelyable
20100628 05:20:36< eleazar_> but even better would be to use my transitions:  useful distiction, always visible
20100628 05:20:37< alink> indeed, except that we must admit "it doesn't necessarily make the distinction in a *very* visible way "
20100628 05:21:29< alink> I suppose that I could look about using these transitions for the off-map tiles
20100628 05:21:52< alink> if used only for the off-map tiles, it fixes my problems with it
20100628 05:22:18< eleazar_> yeah, it would only need to be used on the outside edges
20100628 05:22:52< alink> that is the unit/foreground layer and the possible performance cost (off-map hexes are relatively rare)
20100628 05:23:47< eleazar_> it's certainly not the most urgent problem
20100628 05:23:47< alink> mmmh unless maybe some rare case when there is big unit and foreground terrain at the edge of the map, but that shouldn't be a problem
20100628 05:28:19< eleazar_> is this off-map thing still under contruction?
20100628 05:28:57< alink> I don't think so. Everyone seems happy about it now (= no more complains)
20100628 05:29:25< eleazar_> oh, i see how those pieces fit together now
20100628 05:29:44< alink> yeah I have nice debug images showing it
20100628 05:30:52< alink> oh I failed to save the imagebin :-/
20100628 05:31:18< eleazar_> i actively dislike how offmap works now, but i have more constructive things i can turn my attention to
20100628 05:31:30-!- Bob_The_Mighty [~chatzilla@cpc8-brig15-2-0-cust40.know.cable.virginmedia.com] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.4/20100611143157]]
20100628 05:31:50< eleazar_> anyway, i'm done destroying things tonight.  sorry about that.
20100628 05:32:00< eleazar_> good night all
20100628 05:32:04< alink> np, esp. if you restore it :-)
20100628 05:32:09< alink> eleazar_: gn
20100628 05:35:35-!- Gambit [~quassel@pa-67-234-73-7.dhcp.embarqhsd.net] has quit [Quit: Though to our health we drank a thousand times, it's time to ramble on.]
20100628 05:36:06< alink> it's true that it may be not obvious that half-hexes are off-map, but the "black stripes" clearly mark them as unreachable
20100628 05:36:39< alink> also they don't highlight when the mouse move on them
20100628 05:36:53< shadowm_laptop> I think it's obvious enough
20100628 05:37:15< alink> so just looking a static screenshot is not fair
20100628 05:37:44< alink> shadowm_laptop: yes me too, but I am trying to look with a noob eye 
20100628 05:37:59< alink> (hard exercise to do right)
20100628 05:38:31< shadowm_laptop> I remember I came here around the time there was a hot discussion about how to do the off-map hexes right
20100628 05:39:40< alink> yes, that was an old long war. Perhaps, we shouldn't reopen that topic
20100628 05:40:19< shadowm_laptop> good idea ;)
20100628 05:43:59< Upthorn> okay, optimized it is 21 seconds vs 23
20100628 05:44:45< alink> Upthorn: I had about the same difference, maybe a bit less
20100628 05:46:05< alink> still a bit long
20100628 05:46:09< Upthorn> but in debug configuration it makes a huge difference, which is really important for me because I can't use the debugging functionality in release mode
20100628 05:46:51< alink> Upthorn: I use 2 builds tree (also for other reasons)
20100628 05:46:53< Upthorn> specifically breakpoints and variable watches don't work when I compile the optimized version
20100628 05:51:53< alink> eleazar_: found the image that I searched :  http://img59.imageshack.us/img59/8966/grid1x.png
20100628 05:52:32-!- elvish_sovereign [~Zoltan@pool-108-2-88-55.phlapa.east.verizon.net] has quit [Quit: elvish_sovereign]
20100628 05:52:49< alink> also shows some weirdness with cave wall, but that's less visible with grid image, and even less compared to the old grid system
20100628 05:53:40< alink> (blue is bottom and green is top)
20100628 05:54:39< alink> old system : http://img69.imageshack.us/img69/107/grid2.png
20100628 05:56:20-!- silene [~plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20100628 06:07:03< CIA-87> silene * r43760 /trunk/src/image.cpp: Partly reverted r43754 as it causes glitches when using :refresh.
20100628 06:32:20-!- Espreon [~espreon@wesnoth/developer/espreon] has quit [Ping timeout: 248 seconds]
20100628 06:32:37-!- Espreon [~espreon@83.149.110.34] has joined #wesnoth-dev
20100628 06:33:05-!- Espreon is now known as Guest93829
20100628 06:33:13-!- shadowmaster [~ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 258 seconds]
20100628 06:33:48-!- shikadibot [~shikadi@wesnoth/umc-dev/bot/shikadibot] has quit [Ping timeout: 260 seconds]
20100628 06:34:30-!- shadowm_laptop is now known as shadowmaster_
20100628 06:35:20-!- shadowmaster [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100628 06:36:16-!- shadowmaster is now known as shadowm
20100628 06:36:28-!- knotwork__ [~markm@142.177.235.95] has quit [Ping timeout: 265 seconds]
20100628 06:36:47-!- shadowm is now known as shadowmaster
20100628 06:37:36-!- knotwork [~markm@142.177.235.95] has joined #wesnoth-dev
20100628 06:41:15-!- Guest93829 [~espreon@83.149.110.34] has quit [Ping timeout: 240 seconds]
20100628 06:41:26-!- AI0867 [~ai@wesnoth/developer/ai0867] has quit [Ping timeout: 276 seconds]
20100628 06:41:43-!- AI0867 [~ai@wesnoth/developer/ai0867] has joined #wesnoth-dev
20100628 06:41:46-!- shadowmaster [~ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 264 seconds]
20100628 06:41:48-!- Espreon_ [~espreon@83.149.110.34] has joined #wesnoth-dev
20100628 06:42:18-!- shadowmaster [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100628 06:45:05-!- knotwork [~markm@142.177.235.95] has quit [Ping timeout: 240 seconds]
20100628 06:45:08-!- shadowmaster_ is now known as shadowm_laptop
20100628 06:47:02-!- mjs-de [~mjs-de@vpw.wh.Uni-Dortmund.DE] has joined #wesnoth-dev
20100628 06:50:09-!- shikadibot [~shikadi@wesnoth/umc-dev/bot/shikadibot] has joined #wesnoth-dev
20100628 06:58:08-!- knotwork [~markm@142.177.233.55] has joined #wesnoth-dev
20100628 06:59:09-!- Upthorn [ogmar@adsl-75-26-161-110.dsl.scrm01.sbcglobal.net] has quit [Ping timeout: 240 seconds]
20100628 06:59:15-!- Upthorn [~ogmar@adsl-75-26-161-54.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100628 06:59:25-!- Upth [~ogmar@75.26.161.110] has quit [Ping timeout: 240 seconds]
20100628 06:59:43-!- Upth [~ogmar@adsl-75-26-161-54.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100628 07:04:02-!- ancestral [~ancestral@97-116-120-9.mpls.qwest.net] has quit [Quit: i go sleeps kthxbai]
20100628 07:06:18-!- ancestral [~ancestral@97-116-120-9.mpls.qwest.net] has joined #wesnoth-dev
20100628 07:07:11-!- meric [~Eric@124-170-33-254.dyn.iinet.net.au] has quit [Read error: Connection reset by peer]
20100628 07:07:46-!- meric [~Eric@124-170-179-10.dyn.iinet.net.au] has joined #wesnoth-dev
20100628 07:22:30< crimson_penguin> is that whiteboard thing just in trunk, or somewhere else?
20100628 07:28:59-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: extreme lag]
20100628 07:32:23-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100628 07:41:23-!- meric [~Eric@124-170-179-10.dyn.iinet.net.au] has quit [Read error: Connection reset by peer]
20100628 07:41:44-!- meric [~Eric@124-170-158-179.dyn.iinet.net.au] has joined #wesnoth-dev
20100628 07:50:29-!- ancestral [~ancestral@97-116-120-9.mpls.qwest.net] has quit [Quit: And that’s the end of THAT chapter.]
20100628 07:53:12-!- meric [~Eric@124-170-158-179.dyn.iinet.net.au] has quit [Read error: Connection reset by peer]
20100628 07:54:03-!- meric [~Eric@124-170-184-187.dyn.iinet.net.au] has joined #wesnoth-dev
20100628 08:01:15-!- zookeeper [~l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20100628 08:04:27-!- fendrin [~fabi@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20100628 08:04:45-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Quit: crimson_penguin]
20100628 08:24:40-!- Appleman1234 [~Appleman1@131.181.100.229] has joined #wesnoth-dev
20100628 08:25:16-!- silene [~plouf@wesnoth/developer/silene] has quit [Quit: Leaving.]
20100628 08:26:50< CIA-87> jetryl * r43761 /trunk/data/core/ (5 files in 2 dirs): Added spearman idle anim.
20100628 08:31:16-!- knotwork [~markm@142.177.233.55] has quit [Ping timeout: 248 seconds]
20100628 08:31:42-!- stikonas [~and@ctv-79-132-162-160.vinita.lt] has joined #wesnoth-dev
20100628 08:31:42-!- stikonas [~and@ctv-79-132-162-160.vinita.lt] has quit [Changing host]
20100628 08:31:42-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100628 08:32:33-!- wesbot changed the topic of #wesnoth-dev to: 133 bugs, 281 feature requests, 14 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100628 08:36:34< CIA-87> billynux * r43762 /trunk/src/ (7 files in 4 dirs): More work on the stats part of ana and some changes to network_ana.cpp.
20100628 08:38:06-!- Netsplit *.net <-> *.split quits: elias, meric
20100628 08:39:28-!- knotwork [~markm@142.177.233.55] has joined #wesnoth-dev
20100628 08:47:05-!- Crab_ [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20100628 08:57:21-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: Lost terminal]
20100628 08:58:43-!- Netsplit over, joins: meric, elias
20100628 09:08:02-!- shadowm_tty [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100628 09:14:28-!- Upth [~ogmar@adsl-75-26-161-54.dsl.scrm01.sbcglobal.net] has quit [Ping timeout: 240 seconds]
20100628 09:14:51-!- Upthorn [~ogmar@adsl-75-26-161-54.dsl.scrm01.sbcglobal.net] has quit [Ping timeout: 252 seconds]
20100628 09:14:53-!- Upth [ogmar@adsl-75-26-182-85.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100628 09:14:53-!- Upth is now known as Upthorn
20100628 09:15:15-!- Upth [ogmar@adsl-75-26-182-85.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100628 09:36:43-!- Upth [ogmar@adsl-75-26-182-85.dsl.scrm01.sbcglobal.net] has quit [Quit: The problem is choice! / trapped in cunnilingus webs / Mod your PSP]
20100628 09:46:52-!- Ivanovic [~ivanovic@dtmd-4db2b7be.pool.mediaWays.net] has quit [Changing host]
20100628 09:46:52-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20100628 09:50:12-!- Blueblaze [~nick@adsl-99-188-135-67.dsl.hstntx.sbcglobal.net] has quit [Remote host closed the connection]
20100628 09:51:05-!- shadowm_tty [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: Lost terminal]
20100628 09:52:11< Ivanovic> moin
20100628 10:08:38< Aethaeryn> eleazar_: are you aware of transition issues between the newer wall (not cave wall) and the Dwarf castle?
20100628 10:37:17< Aethaeryn> In 1.8, if you put a dwarf castle directly below the mine/building wall (with nothing to the southwest edge of the wall) the southwest edge looks like the cavewall transition.
20100628 10:37:39< Aethaeryn> If a 1.8 trunk thing could fix it without causing OOS? Does it break the stable rules?
20100628 10:37:45< Aethaeryn> *1.8 SVN thing
20100628 10:37:51< Aethaeryn> 1.8 trunk doesn't exist ;-)
20100628 10:38:08-!- shadowm_tty [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100628 10:39:32< Aethaeryn> Just a really annoying transition glitch that has shown up in almost all of my recent maps (underground ones)
20100628 10:42:06-!- euschn [~eugen@wesnoth/developer/euschn] has joined #wesnoth-dev
20100628 10:50:58< zookeeper> Aethaeryn, yes, such a fix can be put in 1.8
20100628 10:51:24-!- akeenanr_ [~quassel@cpe-74-74-148-190.rochester.res.rr.com] has quit [Ping timeout: 245 seconds]
20100628 10:51:34< Aethaeryn> should in theory be possible if there's anyone who wants to touch 1.8 terrain WML
20100628 10:54:11 * Aethaeryn is hoping it's an easy fix. :-)
20100628 10:57:31-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100628 11:10:10< zookeeper> eh, anyone familiar with lua around?
20100628 11:11:12 * zookeeper sits and waits for silene
20100628 11:16:30-!- shadowm_tty [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: night]
20100628 11:19:48-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Quit: ...]
20100628 11:24:50< Crab_> zookeeper: I might be able to help with lua, depending on what's needed
20100628 11:26:17< zookeeper> Crab_, all right...i'm trying to make a new tag which allows me to easily show different messages to the current side, his teammates and his enemies. this is what i have so far: http://wesnoth.pastebin.com/8SBS5nga
20100628 11:26:38< zookeeper> so the main thing i don
20100628 11:26:59< zookeeper> i don't know how to do is how do i insert the for_side=i into the teammate and enemy messages
20100628 11:27:23< zookeeper> (and for_side=cfg.side for the first message)
20100628 11:28:21< zookeeper> the second thing is that i get an odd error "scripting/lua: [string "..."]:17: attempt to index a nil value" error when i move, and i don't know what causes it and why
20100628 11:28:41< zookeeper> (it happens when the enemy message should be displayed)
20100628 11:30:00< Crab_> zookeeper: you mean 'side_for' attribute ?
20100628 11:30:12< zookeeper> err, yes
20100628 11:31:53< Crab_> try something like     wesnoth.fire("message",    { side_for = i , message = message_for_enemies })
20100628 11:32:00< zookeeper> eh, i guess the error i get is because i don't check if i is a valid side number...
20100628 11:32:13 * zookeeper tries
20100628 11:32:43< Crab_> and yes, wesnoth.get_side(i).team_name  can give you 'attempt to index a nil value"' if i is not valid
20100628 11:34:08< zookeeper> hmh :/ error scripting/lua: [string "..."]:20: bad argument #2 to 'fire' (WML table expected, got table)
20100628 11:34:17< zookeeper> wesnoth.fire("message", { side_for = i , message = message_for_teammates })
20100628 11:36:46< zookeeper> i'll try the same using helper.set_wml_tag_metatable 
20100628 11:37:07-!- Espreon_ is now known as Espreon
20100628 11:37:37-!- Espreon is now known as Guest77295
20100628 11:37:38-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100628 11:37:43< zookeeper> well, if i can figure out how..
20100628 11:37:47< Crab_> zookeeper: try wesnoth.fire("message", { side_for = i , message = "test" })
20100628 11:38:46-!- Guest77295 is now known as Espreon
20100628 11:39:00-!- Espreon [~espreon@83.149.110.34] has quit [Changing host]
20100628 11:39:00-!- Espreon [~espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20100628 11:39:03-!- loonybot [~loonybot@ppp79-139-137-245.pppoe.spdop.ru] has joined #wesnoth-dev
20100628 11:39:03-!- loonybot [~loonybot@ppp79-139-137-245.pppoe.spdop.ru] has quit [Changing host]
20100628 11:39:03-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20100628 11:40:25-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100628 11:40:30< zookeeper> Crab_, right, that works
20100628 11:40:32< Crab_> zookeeper: If it works, try replacing  like this :     local message_for_teammates = helper.get_child(cfg, "message_for_teammates")  with  local message_for_teammates = cfg.message_for_teammates
20100628 11:40:47< Crab_> because get_child gives you a [subtag], not an attribute
20100628 11:41:15-!- phlaem [~a@e178082161.adsl.alicedsl.de] has joined #wesnoth-dev
20100628 11:41:26< Crab_> (the same with message_for_enemies)
20100628 11:41:39< zookeeper> Crab_, message_for_teammates is supposed to be the whole tag though, not just the message key
20100628 11:42:36< Crab_> zookeeper: ok, understood. then, yes, you need to use get_child()
20100628 11:43:26< Crab_> then, what about        message_for_teammates.side_for = i          before wesnoth.fire      ?
20100628 11:43:27-!- Upth [~ogmar@adsl-75-26-182-85.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100628 11:43:47< Crab_> (and using the   wesnoth.fire("message", message_for_teammates)  syntax)
20100628 11:43:51< zookeeper> i think i tried it already, but i'll check again..
20100628 11:45:55< zookeeper> Crab_, that just gets me a "attempt to index local 'message_for_enemies' (a userdata value)" error
20100628 11:47:08< zookeeper> i wonder if i should try something like table.insert(message_for_enemies, ???)
20100628 11:53:53< CIA-87> espreon * r43763 /trunk/data/core/images/terrain/village/ (69 files): 
20100628 11:53:53< CIA-87> Ran wesnoth-optipng:
20100628 11:53:53< CIA-87> Overall statistics (only for files with a smaller recompressed size):
20100628 11:53:53< CIA-87> Original size: 963 KiB on 69 files
20100628 11:53:53< CIA-87> Optimized size: 780 KiB
20100628 11:53:53< CIA-87> Total saving: 182 KiB = 18% decrease
20100628 11:54:55< Crab_> zookeeper: what if we try to do      local message_for_teammates = helper.get_child(cfg, "message_for_teammates").__parsed   ?
20100628 11:55:38< zookeeper> "Functions registered by wesnoth.register_wml_action receive their data in a userdata object which has the exact same structure as above. It is read-only however."
20100628 11:55:48< zookeeper> yeah, i was about to try something like that :P
20100628 11:56:05< Espreon> alink: Uh, Windows users can just create batch files that would launch Wesnoth with their favorite parameters.
20100628 11:58:14< zookeeper> Crab_, ok, i get no errors now so it probably works
20100628 11:58:24< Crab_> great
20100628 11:58:30< zookeeper> gotta test it online though to make sure, just a moment..
20100628 12:01:37< CIA-87> espreon * r43764 /trunk/data/core/images/terrain/ (6 files): 
20100628 12:01:37< CIA-87> Ran wesnoth-optipng:
20100628 12:01:37< CIA-87> Overall statistics (only for files with a smaller recompressed size):
20100628 12:01:37< CIA-87> Original size: 17 KiB on 6 files
20100628 12:01:37< CIA-87> Optimized size: 2 KiB
20100628 12:01:37< CIA-87> Total saving: 14 KiB = 84% decrease
20100628 12:03:32< zookeeper> Crab_, yeah, it works now. thanks
20100628 12:04:30< Crab_> good. sorry that I've, initially. misunderstood the way you've wanted the tag to work (I thought that you've wanted to only customize the message= part, not all the parameters)
20100628 12:04:58< CIA-87> espreon * r43765 /trunk/data/core/images/units/human-loyalists/ (4 files): Ran umcpropfix.
20100628 12:06:09< CIA-87> espreon * r43766 /trunk/data/core/images/units/human-loyalists/ (4 files): 
20100628 12:06:09< CIA-87> Ran wesnoth-optipng:
20100628 12:06:09< CIA-87> Overall statistics (only for files with a smaller recompressed size):
20100628 12:06:09< CIA-87> Original size: 9 KiB on 4 files
20100628 12:06:09< CIA-87> Optimized size: 5 KiB
20100628 12:06:10< CIA-87> Total saving: 4 KiB = 44% decrease
20100628 12:09:10-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100628 12:12:40< zookeeper> in wml-tags.lua i see things like "wesnoth.eval_conditional(show_if)" being used, but i can't find eval_conditional explained anywhere
20100628 12:12:56< zookeeper> maybe it's accidentally missing
20100628 12:14:22< Crab_> zookeeper: it's a c++ function usable from lua. it does a 'game_events::conditional_passed' call and returns true or false. it checks a conditional in a config exactly like [if] and [while] conditions are checked
20100628 12:14:45< zookeeper> i figured that's how it works
20100628 12:17:46< Crab_> I think you're right about the 'accidentally missing' part
20100628 12:32:37< zookeeper> added the check for a valid side too (might be a bit inefficient, perhaps): http://wesnoth.pastebin.com/VNS8Hdbz i guess that's my first working lua tag \o/
20100628 12:34:51< Crab_> zookeeper: wesnoth.get_side_count() might be easier
20100628 12:35:13< zookeeper> Crab_, right. that's missing from the wiki too then
20100628 12:38:40-!- silene [~plouf@lri4-236.lri.fr] has joined #wesnoth-dev
20100628 12:41:12-!- meric [~Eric@124-170-184-187.dyn.iinet.net.au] has quit [Read error: Connection reset by peer]
20100628 12:41:25< silene> zookeeper: your code is not quite correct, don't access __parsed before testing if it is nil (it will crash if it is nil, and it always pass if it isn't)
20100628 12:42:08< silene> also, you should use __literal instead of __parsed, otherwise variable substitution will happen twice, which would be bad for nested [command]s for instance
20100628 12:42:24< zookeeper> mmh, okay
20100628 12:43:12< silene> also, in order to know if a team exists, just store the result of wesnoth.get_side(i) in a variable and test this variable
20100628 12:43:13-!- meric [~Eric@124-170-198-191.dyn.iinet.net.au] has joined #wesnoth-dev
20100628 12:43:15< zookeeper> right, i actually want it to be possible to omit any of the message tags
20100628 12:44:26< silene> your inner condition is wrong too, you should only call [message] on message_for_teammates if message_for_teammates actually exists
20100628 12:44:29< zookeeper> so, does "if message then" pass as a check for whether message is nil, or do i have to change that to "if message ~= nil then"?
20100628 12:44:40< silene> "if message then" is fine
20100628 12:47:48< silene> you should display a wml error message if cfg.side doesn't exist, otherwise you will get lua error messages when doing wesnoth.get_side(cfg.side)
20100628 12:49:20< silene> the usual way for querying and testing at the same time is: local side = cfg.side or helper.wml_error "[multi_message] missing a side= attribute"
20100628 12:50:51-!- Appleman1234 [~Appleman1@131.181.100.229] has quit [Remote host closed the connection]
20100628 12:51:28< silene> and finally, i would suggest always putting lua code inside separate files; that way you can do remote debugging of the lua scripts (step-by-step execution, display of lua variables, and so on) while playing wesnoth
20100628 12:51:50-!- Upth [~ogmar@adsl-75-26-182-85.dsl.scrm01.sbcglobal.net] has quit [Quit: We're the knights who say: / it is my job to freeze you / 10 second delay]
20100628 12:51:57< zookeeper> silene, ok, how's this? http://wesnoth.pastebin.com/wXLNE5VV
20100628 12:53:03< silene> zookeeper: you have to first call get_child, then test the return value, then access __literal
20100628 12:53:16< zookeeper> oh
20100628 12:53:22< zookeeper> got it
20100628 12:54:00< silene> does wesnoth.get_side_count even exist?
20100628 12:54:10< zookeeper> no idea, it was Crab_'s idea :P
20100628 12:54:12< Crab_> silene: in the source, yes
20100628 12:54:29< silene> Crab_: who added it? and why isn't it commented?
20100628 12:55:16< zookeeper> so is it valid to do message_for_enemies.__literal.side_for = i   ?
20100628 12:55:25< zookeeper> or do i really need to put it in a temp variable first?
20100628 12:55:30< silene> zookeeper: yes, but it won't do what you expect ;-)
20100628 12:55:31< Crab_> silene: I think that it might be one of gsoc candidates. in that case it's my fault that it's not in the wiki.
20100628 12:55:47< silene> is it in 1.8?
20100628 12:56:35< silene> (if not, don't document it, i will remove it, as i have a better system in draft for accessing teams: a plain table)
20100628 12:57:10< Crab_> silene: no, it's not in 1.8
20100628 12:57:14< silene> zookeeper: just do: message_for_enemies = message_for_enemies.__literal, then you can use your current code
20100628 12:57:38< zookeeper> cool, that's what i already did, just didn't test yet
20100628 12:58:22< Crab_> zookeeper: also, note that the team name can be a comma-separated string (e.g. team_name=player,trolls )
20100628 12:59:02< Crab_> zookeeper: so, 'wesnoth.get_side(i).team_name == wesnoth.get_side(side).team_name' checks are not always correct.
20100628 12:59:18< zookeeper> Crab_, i know, i'm intentionally forgetting about that right now, unless someone wants to tell me the code for checking for that ;)
20100628 13:00:04< zookeeper> (i can't just use string.find, i'd have to split the string first and i didn't want to bother with that)
20100628 13:00:20< Crab_> zookeeper: trunk has wesnoth.is_enemy(side_number_1, side_number_2)   (with the same origin). maybe silene will want to replace it with a better solution, too.
20100628 13:02:34< silene> hmm... is_enemy seems sensible, it will probably stay as is
20100628 13:02:42< Crab_> ok, then I'll document it
20100628 13:03:56-!- meric [~Eric@124-170-198-191.dyn.iinet.net.au] has quit [Ping timeout: 240 seconds]
20100628 13:03:58-!- meric_ [~Eric@124-171-26-143.dyn.iinet.net.au] has joined #wesnoth-dev
20100628 13:04:23-!- Ken_Oh [~briang@static-71-178-174-220.washdc.fios.verizon.net] has joined #wesnoth-dev
20100628 13:16:31-!- stikonas_ [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100628 13:16:36-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 248 seconds]
20100628 13:31:47-!- stikonas_ [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 240 seconds]
20100628 13:50:28-!- stikonas_ [~and@ctv-79-132-162-160.vinita.lt] has joined #wesnoth-dev
20100628 13:50:28-!- stikonas_ [~and@ctv-79-132-162-160.vinita.lt] has quit [Changing host]
20100628 13:50:28-!- stikonas_ [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100628 13:51:30-!- stikonas_ is now known as stikonas
20100628 14:11:40-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20100628 14:41:06-!- Gambit [~quassel@pa-67-234-73-7.dhcp.embarqhsd.net] has joined #wesnoth-dev
20100628 14:44:05-!- Gambit [~quassel@pa-67-234-73-7.dhcp.embarqhsd.net] has quit [Remote host closed the connection]
20100628 14:56:36-!- Elvish_Pillager [~eli@71-10-224-192.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20100628 15:01:37-!- alink [~alink@wesnoth/developer/alink] has quit [Remote host closed the connection]
20100628 15:31:11-!- Gambit [~quassel@pa-67-234-73-7.dhcp.embarqhsd.net] has joined #wesnoth-dev
20100628 15:32:39-!- happygrue [~George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20100628 15:40:39< CIA-87> eleazar * r43767 /trunk/data/core/terrain.cfg: 
20100628 15:40:39< CIA-87> Several times the most expedient way to collaborate with boucman has been to
20100628 15:40:39< CIA-87> commit odd things to terrain for testing purposes. Now i'm adding a couple fake
20100628 15:40:39< CIA-87> test terrains for that purpose, so i don't have to mess up real terrains. Also
20100628 15:40:39< CIA-87> the current terrain interactions are so complex, that it's muc easier to figure
20100628 15:40:39< CIA-87> stuff out with these dummy terrains. That's how i did the latest grass. These
20100628 15:40:40< CIA-87> can be removed before the next stable release.
20100628 15:45:37< eleazar_> Aethaeryn, zookeeper: the new wall transitions badly with pretty much everything.  I'm pretty sure it's not a quick WML fix
20100628 15:59:25-!- silene [~plouf@lri4-236.lri.fr] has quit [Ping timeout: 260 seconds]
20100628 16:01:35-!- silene [~plouf@lri4-236.lri.fr] has joined #wesnoth-dev
20100628 16:06:37-!- silene [~plouf@lri4-236.lri.fr] has quit [Changing host]
20100628 16:06:37-!- silene [~plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20100628 16:11:22-!- Upth [ogmar@adsl-75-26-195-109.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100628 16:13:43-!- Valkier [~IceChat7@c-174-55-104-2.hsd1.pa.comcast.net] has joined #wesnoth-dev
20100628 16:14:07-!- Upthorn [ogmar@adsl-75-26-182-85.dsl.scrm01.sbcglobal.net] has quit [Ping timeout: 240 seconds]
20100628 16:16:50-!- phlaem [~a@e178082161.adsl.alicedsl.de] has quit [Read error: Connection reset by peer]
20100628 16:17:19-!- phlaem [~a@e178082161.adsl.alicedsl.de] has joined #wesnoth-dev
20100628 16:18:45-!- eleazar_ [~eleazar@ppp-70-226-197-86.dsl.spfdil.ameritech.net] has quit [Ping timeout: 240 seconds]
20100628 16:21:38-!- ahdont [~ahdont@124-171-74-102.dyn.iinet.net.au] has joined #wesnoth-dev
20100628 16:24:11-!- eleazar [~eleazar@ppp-70-226-197-86.dsl.spfdil.ameritech.net] has joined #wesnoth-dev
20100628 17:01:01-!- Crab_ [~Crab_@wesnoth/developer/crab] has quit [Quit: Leaving.]
20100628 17:14:38-!- ahdont [~ahdont@124-171-74-102.dyn.iinet.net.au] has left #wesnoth-dev []
20100628 17:26:03-!- euschn [~eugen@wesnoth/developer/euschn] has quit [Quit: Leaving.]
20100628 17:42:55-!- billynux [~billy@wesnoth/developer/billynux] has joined #wesnoth-dev
20100628 18:01:51-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20100628 18:04:57-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20100628 18:10:00-!- meric_ [~Eric@124-171-26-143.dyn.iinet.net.au] has quit [Read error: Connection reset by peer]
20100628 18:10:22-!- meric [~Eric@124-170-66-85.dyn.iinet.net.au] has joined #wesnoth-dev
20100628 18:12:46-!- silene [~plouf@wesnoth/developer/silene] has quit [Quit: Leaving.]
20100628 18:15:18< CIA-87> eleazar * r43768 /trunk/data/core/ (45 files in 3 dirs): The images for the test terrain, various grass transition updates that synchronize my terrain with trunk. There are plenty of missing transitions, etc.
20100628 18:15:36-!- meric [~Eric@124-170-66-85.dyn.iinet.net.au] has quit [Quit: meric]
20100628 18:40:56-!- billynux [~billy@wesnoth/developer/billynux] has quit [Quit: Leaving]
20100628 19:01:44-!- silene [~plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20100628 19:31:15< Elvish_Pillager> If the same key appears twice in a tag, is there a defined order of precedence for which one is used?
20100628 19:33:05-!- EdB [~edb@4.103.89-79.rev.gaoland.net] has joined #wesnoth-dev
20100628 19:34:25< silene> Elvish_Pillager: the second one overwrites the first one
20100628 19:34:42< Elvish_Pillager> That's convenient :)
20100628 19:39:00-!- Blueblaze [~nick@adsl-99-188-135-67.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100628 19:42:30-!- kevg [~kevg@94.232.5.226] has joined #wesnoth-dev
20100628 19:46:38-!- EdB [~edb@4.103.89-79.rev.gaoland.net] has quit [Remote host closed the connection]
20100628 19:53:01-!- Ivanovic_ [~ivanovic@dtmd-4db2b7be.pool.mediaWays.net] has joined #wesnoth-dev
20100628 19:57:46-!- timotei [~timo@193.34.191.4] has joined #wesnoth-dev
20100628 20:00:00-!- Netsplit *.net <-> *.split quits: Rhonda, AnMaster, Tycale, Greywhind, chris|, CIA-87, happygrue, Smar, loonybot, Blueblaze,  (+14 more, use /NETSPLIT to show all of them)
20100628 20:00:00-!- Ivanovic_ is now known as Ivanovic
20100628 20:00:59-!- Rh0nda [~rhonda@anguilla.debian.or.at] has joined #wesnoth-dev
20100628 20:00:59-!- Vetinari_ [~lukjad@unaffiliated/lukjad] has joined #wesnoth-dev
20100628 20:00:59-!- Tycale_ [~Thibault@viki-network.net] has joined #wesnoth-dev
20100628 20:01:07-!- Netsplit over, joins: happygrue, Blueblaze, Gambit, wesbot, Espreon, phlaem, Upth, loonybot, knotwork, AI0867 (+10 more)
20100628 20:01:27-!- AnMaster [~AnMaster@unaffiliated/anmaster] has quit [Quit: ZNC - http://znc.sourceforge.net]
20100628 20:03:56-!- Tycale [~Thibault@viki-network.net] has joined #wesnoth-dev
20100628 20:04:53-!- AnMaster [~AnMaster@unaffiliated/anmaster] has joined #wesnoth-dev
20100628 20:05:35-!- ancestral [~ancestral@12.145.225.25] has joined #wesnoth-dev
20100628 20:05:40-!- Tycale_ [~Thibault@viki-network.net] has quit [Write error: Broken pipe]
20100628 20:06:30-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100628 20:22:02-!- ancestral [~ancestral@12.145.225.25] has quit [Quit: i go sleeps kthxbai]
20100628 20:37:55-!- billynux [d073d2d2@gateway/web/freenode/ip.208.115.210.210] has joined #wesnoth-dev
20100628 20:37:55-!- billynux [d073d2d2@gateway/web/freenode/ip.208.115.210.210] has quit [Changing host]
20100628 20:37:55-!- billynux [d073d2d2@wesnoth/developer/billynux] has joined #wesnoth-dev
20100628 20:39:05< billynux> !@$#@!#$!@#$ -> to the sys admin here at the university, blocking IRC and almost every web proxy out there! (sry, catharsis)
20100628 20:39:05< shikadibot> billynux: I don't see $# around...
20100628 20:40:38< billynux> wesbot: seen mordante
20100628 20:40:39< wesbot> billynux: The person with the nick mordante last spoke 8d ago.  8d ago was here and on the channel #wesnoth-de with the message: Quit: Leaving
20100628 20:43:52-!- YogiHH [YogiHH@c147009.adsl.hansenet.de] has joined #wesnoth-dev
20100628 20:44:05-!- YogiHH [YogiHH@c147009.adsl.hansenet.de] has quit [Changing host]
20100628 20:44:05-!- YogiHH [YogiHH@wesnoth/developer/yogihh] has joined #wesnoth-dev
20100628 20:50:17-!- YogiHH [YogiHH@wesnoth/developer/yogihh] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539]]
20100628 20:51:33-!- timotei [~timo@193.34.191.4] has quit [Remote host closed the connection]
20100628 20:53:56-!- Rh0nda is now known as Rhonda
20100628 20:54:15-!- Rhonda [~rhonda@anguilla.debian.or.at] has quit [Changing host]
20100628 20:54:15-!- Rhonda [~rhonda@wesnoth/developer/rhonda] has joined #wesnoth-dev
20100628 20:56:11-!- timotei [~timo@wesnoth/developer/timotei] has joined #wesnoth-dev
20100628 20:58:10-!- billynux [d073d2d2@wesnoth/developer/billynux] has quit [Ping timeout: 252 seconds]
20100628 20:59:23-!- timotei [~timo@wesnoth/developer/timotei] has quit [Remote host closed the connection]
20100628 21:08:35-!- happygrue_ [~George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20100628 21:11:45-!- happygrue [~George@wesnoth/developer/wintermute] has quit [Ping timeout: 240 seconds]
20100628 21:13:28-!- timotei [~timo@wesnoth/developer/timotei] has joined #wesnoth-dev
20100628 21:24:01< timotei> wesbot: seen esr ?
20100628 21:24:01< wesbot> timotei: The person with the nick esr 4d ago was here and on the channel #wesnoth-umc-dev with the message: Ping timeout: 240 seconds
20100628 21:24:17-!- happygrue [~George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20100628 21:27:50-!- happygrue_ [~George@wesnoth/developer/wintermute] has quit [Ping timeout: 260 seconds]
20100628 21:33:02-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100628 21:36:12-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: GO, GET TO THE CHOPPAH!!!]
20100628 21:36:28-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100628 21:38:10-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100628 21:39:05-!- billynux [d073d2d2@gateway/web/freenode/ip.208.115.210.210] has joined #wesnoth-dev
20100628 21:39:27-!- akeenanr [~quassel@cpe-74-74-148-190.rochester.res.rr.com] has joined #wesnoth-dev
20100628 21:43:38-!- billynux [d073d2d2@gateway/web/freenode/ip.208.115.210.210] has quit [Ping timeout: 252 seconds]
20100628 21:44:30-!- Ken_Oh [~briang@static-71-178-174-220.washdc.fios.verizon.net] has quit [Read error: Connection reset by peer]
20100628 21:46:05-!- akeenanr [~quassel@cpe-74-74-148-190.rochester.res.rr.com] has quit [Ping timeout: 240 seconds]
20100628 21:48:30-!- billynux [d073d2d2@gateway/web/freenode/ip.208.115.210.210] has joined #wesnoth-dev
20100628 21:48:30-!- billynux [d073d2d2@gateway/web/freenode/ip.208.115.210.210] has quit [Changing host]
20100628 21:48:31-!- billynux [d073d2d2@wesnoth/developer/billynux] has joined #wesnoth-dev
20100628 21:54:11< shadowm_laptop> billynux: do they block all of the ports used by freenode?
20100628 21:54:48< billynux> shadowm_laptop: I tried a few... how many are there?
20100628 21:55:09< shadowm_laptop> freenode servers listen on ports 6665, 6666, 6667, 8000, 8001 and 8002 for plain-text connections; ports 7000 and 7070 are available for SSL connections 
20100628 21:55:09< billynux> I've tried using a VPN, but they block that too
20100628 21:56:15-!- Crab_ [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20100628 21:57:31-!- akeenanr__ [~quassel@cpe-74-74-148-190.rochester.res.rr.com] has joined #wesnoth-dev
20100628 21:58:05-!- billynux_on_ssl [~billy@felipe.dc.exa.unrc.edu.ar] has joined #wesnoth-dev
20100628 21:58:07< billynux_on_ssl> :)
20100628 21:58:24< timotei> ;))
20100628 21:58:27< timotei> gg billynux_on_ssl
20100628 21:58:31< timotei> hi Crab_ :)
20100628 21:58:35< Crab_> hi, timotei
20100628 21:58:46< billynux_on_ssl> hi timotei, Crab_ 
20100628 21:58:47< timotei> billynux_on_ssl: using some "private"/university ssl/ssh?
20100628 21:58:56< billynux_on_ssl> and thanks a bunch shadowm
20100628 21:59:01< timotei> cause I need to try that too, mine blocking all non 80 ports too
20100628 21:59:05< timotei> but ssh works afair
20100628 21:59:12< billynux_on_ssl> timotei, yes... university... they have blocked everything to kingdom come!
20100628 21:59:13< shadowm_laptop> no problem
20100628 21:59:22< Crab_> if almost all things get blocked, 'outgoing ssh/openvpn to server listening on port 443' is the usual way to bypass :)
20100628 21:59:24-!- billynux [d073d2d2@wesnoth/developer/billynux] has quit [Ping timeout: 252 seconds]
20100628 21:59:28< billynux_on_ssl> timotei, you can use webchat.freenode.net on 80
20100628 21:59:44< timotei> billynux_on_ssl: yeah, but don't like that way
20100628 21:59:48< timotei> better ssh/vpn
20100628 21:59:55< timotei> so i can Wow too :))
20100628 22:00:13< billynux_on_ssl> Well... I work here, so gaming is morally wrong for me
20100628 22:00:23< billynux_on_ssl> but not GSoCing
20100628 22:00:26< timotei> :P
20100628 22:00:43< timotei> I just said wow, in the way of "needing" ports
20100628 22:01:24< billynux_on_ssl> I'm afraid to try WOW (the game), due to addictive tendencies of mine, I'll stick to working
20100628 22:01:52< billynux_on_ssl> Crab_, mordante is a pretty hard guy to reach, so I'm a bit lost ATM
20100628 22:02:13< billynux_on_ssl> We've discussed mantaining the current network API (network.hpp)
20100628 22:02:36-!- billynux_on_ssl is now known as billynux
20100628 22:02:46-!- billynux [~billy@felipe.dc.exa.unrc.edu.ar] has quit [Changing host]
20100628 22:02:46-!- billynux [~billy@wesnoth/developer/billynux] has joined #wesnoth-dev
20100628 22:03:05< Crab_> billynux: I can help ? describe things, then, I'll try 
20100628 22:03:15-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100628 22:03:45< billynux> Crab_, well... I can do that (keep the current API), but it will involve some ugly programming
20100628 22:04:07< Crab_> what kind of ugly programming ? why ?
20100628 22:04:17< billynux> ana (the lib) is natively asynchronous
20100628 22:04:56< billynux> while the API is not
20100628 22:05:06< billynux> the way to circumvent this is (a bit) ugly:
20100628 22:05:17< Crab_> wrap async api into a sync api ?
20100628 22:05:22< billynux> yep
20100628 22:05:41< Crab_> but 'direct usage' of async api would be better, right ?
20100628 22:05:53< billynux> 1- create a mutex before sending, 2- create a handler object that locks the mutex until called, 3-start the send op. with this handler, 4-reacquire the mutex
20100628 22:06:01< billynux> Crab_, definitely
20100628 22:06:05< Crab_> (except in places where we need to sync anyway...)
20100628 22:06:41< billynux> I can redesign the API to make it better, but mordante has a point, the current API "works"
20100628 22:06:54< billynux> meaning: its used the right way
20100628 22:07:12< Crab_> well, then, the details can be hidden from the user (e.g., by constructing the object which will do your 1-4 )
20100628 22:08:14< billynux> yes... I want to finish this part (implementing the network.hpp)... this week
20100628 22:08:32< billynux> s/.hpp/.hpp methods/
20100628 22:08:54< billynux> But I'm having trouble at every corner
20100628 22:09:02< Crab_> what kind of trouble ?
20100628 22:09:43-!- yann [~dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has quit [Read error: No route to host]
20100628 22:09:55< billynux> you can check my last commit, in network_ana.cpp:140 -> looks awful
20100628 22:10:09-!- yann [~dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has joined #wesnoth-dev
20100628 22:11:16< billynux> the API allows one to build many servers and many clients
20100628 22:11:38< billynux> there is a violation of ISP there, but thats beyond the point
20100628 22:12:02< billynux> most likely, the different apps. use the API in different ways
20100628 22:12:33< billynux> i.e. I don't think a campaign server needs to create a client
20100628 22:13:20-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has quit [Remote host closed the connection]
20100628 22:13:30< billynux> Crab_, my (preliminary) idea is having either a servers_, and clients_ map and a set<net_id> of the ids of every connected client
20100628 22:15:11< billynux> or just a set of ana components (pointers to servers + clients)
20100628 22:16:18< Crab_> billynux: well, looked at the network_ana.cpp code of your last commit. looks good enough even now (maybe my tolerance for bad code is quite high due to my current work ;) )
20100628 22:16:32< Crab_> what part specifically looks awful to you ?
20100628 22:16:53-!- alink [~alink@wesnoth/developer/alink] has joined #wesnoth-dev
20100628 22:17:40< billynux> Well... I tolerate about 10 to 20 LOCs per method, but that's just on the stringent side of things
20100628 22:17:52< billynux> Crab_, what I don't like is having to rummage through 3 containers
20100628 22:18:01< billynux> to find the id I'm looking for
20100628 22:19:07< billynux> but I believe most use cases will simple involve either one server or one (or a few) clients created
20100628 22:19:20< timotei> billynux: what's LOC?:P
20100628 22:19:21< Crab_> well, if your client and server would both extend a common interface...
20100628 22:19:25< Crab_> timotei: Line-of-code
20100628 22:19:25< billynux> (there may be many client_proxy elements connected to one server
20100628 22:19:34< billynux> yep
20100628 22:19:35< timotei> Crab_: thanks
20100628 22:20:49< billynux> Crab_, I'm thinking about adding a struct that inherits from both of them (ana::server & ana::client)
20100628 22:21:01< billynux> and then having the manager hold a set of it
20100628 22:21:07< Crab_> billynux: or see http://wesnoth.pastebin.com/APCzxcsF
20100628 22:21:36< billynux> with some basic functionality (as get_id)
20100628 22:22:05< billynux> Crab_, yes, that looks good
20100628 22:22:16< Crab_> the details may vary, but the idea is to use a wrapper around a boost::variant to store both server and client in a same c++ object
20100628 22:22:28< Crab_> (you can use a custom solution in place of boost::variant)
20100628 22:22:59< billynux> good, since we are at it, what do the attributes of a network::statistics struct mean?
20100628 22:23:08< billynux> total/current/current_max ?
20100628 22:23:19< Crab_> (all what is needed is that the object needs to do the lookups in the ctor, searching in servers_ and clients_ for this id, cache the object that is found, and set the is_server flag internally )
20100628 22:23:59< billynux> network.hpp:229
20100628 22:24:10< Crab_> (the wrapper object in there would be, most likely, not a ana::net_id but something else, but this doesn't matter)
20100628 22:24:18< Crab_> ok, looking at network.hpp:229 ...
20100628 22:24:32-!- phlaem [~a@e178082161.adsl.alicedsl.de] has quit [Read error: Connection reset by peer]
20100628 22:24:37< billynux> Crab_, agreed, that was pretty much my approach too
20100628 22:24:58-!- phlaem [~a@e178082161.adsl.alicedsl.de] has joined #wesnoth-dev
20100628 22:26:00< Crab_> billynux: note that if both server and client implement a common interface that has get_stats(), the code can be simplified to http://wesnoth.pastebin.com/HzkmpFXC
20100628 22:27:00< billynux> yes, they have that in common
20100628 22:27:28< billynux> and I'll approach it to get minimum code
20100628 22:28:06< billynux> Crab_, my net_id's will only go up (i.e. every new one will be greater than all previous). A set stores sortable keys in a bintree right? any better sorted container out there?
20100628 22:28:24< Crab_> set uses a tree, yes
20100628 22:28:40< billynux> if I have a sorted_list I only need to push_back elements
20100628 22:29:10< billynux> and find should be implemented with a binary search
20100628 22:29:32< billynux> (which will improve a tree's performance)
20100628 22:29:43< Crab_> I'd say 'use a set and solve other issues, then profile it and see if it's a problem'
20100628 22:30:03< billynux> ok
20100628 22:30:40< billynux> so... an ana_network_manager will hold a set< common_struct >
20100628 22:30:59< billynux> where common_struct will have things as get_id, is_server, etc..
20100628 22:30:59< Crab_> not a set of pointers to common structs ?
20100628 22:31:13< billynux> I was confusing there
20100628 22:31:23< Crab_> or you were thinking about a union-like object ?
20100628 22:31:24< billynux> a set of ana_wesnoth_components
20100628 22:31:39< billynux> which holds a pointer to a common_ana_something
20100628 22:32:12< Crab_> about the stats...
20100628 22:32:19< billynux> yes?
20100628 22:33:25< billynux> I added automatic stat collecting to ana (which holds stats accumulated stats and the last second/minute/hour/day)
20100628 22:33:44< Crab_> judging from the code.. 'total' is total count of bytes sent (received) during a session.  'current' is the number of bytes sent/received from a current buffer that we need to send/receive, 'current_max' is the maximum 'current' value that was ever obtained.
20100628 22:34:41< Crab_> e.g., if we tell the program to send 10000 bytes, it'll do a SDLNet_TCP_Send which will, say, send 4096 bytes. then, total will be 4096, current = 4096 and current_max = 0.
20100628 22:35:03< Crab_> after all 10000 bytes are sent, in a few SDLNet_TCP_Send calls, total will be 10000, current will be 0 and current_max will be 10000
20100628 22:35:23< billynux> mm
20100628 22:35:49< billynux> how do you know which one is the "current buffer" ?
20100628 22:36:04< Crab_> then, if we tell the program to send another 8000 bytes, and after the first few SDLNet_TCP_Send calls the program will send 6072 bytes, total will be 16072, current will be 6072, and current_max will be 10000
20100628 22:36:23< Crab_> the one which is passed to `static SOCKET_STATE send_buffer(TCPsocket sock, std::vector<char>& buf, int in_size = -1)`
20100628 22:39:18< billynux> Crab_, got it, weird, I don't think I need this
20100628 22:39:31< billynux> boost::asio is in charge of sending everything behind my back
20100628 22:40:11< Crab_> yes, but you can trace total and current_max just fine
20100628 22:40:15< billynux> yes
20100628 22:41:00< Crab_> it might be important for the network server - if you have the info about the distribution of buffer sizes, you can adjust OS settings trying to get optimal performance
20100628 22:41:23< billynux> point taken
20100628 22:41:35< CIA-87> timotei * r43769 /trunk/utils/java/eclipse_plugin/changelog: eclipse plugin: update changelog
20100628 22:41:38< CIA-87> timotei * r43770 /trunk/utils/java/eclipse_plugin/ (4 files in 3 dirs): eclipse plugin: add menu for wmlindent
20100628 22:41:44< CIA-87> timotei * r43771 /trunk/utils/java/eclipse_plugin/ (6 files in 2 dirs): eclipse plugin: some name changes
20100628 22:42:32< Crab_> e.g., if you know '80% of the time the data that we want to send fits in 16kb', '95% of the time the data that we want to send fits in 32kb', etc, you can make informed decisions. of course, you can play&guess even without such info :)
20100628 22:42:50< Crab_> and, the number of clients of wesnoth's network server is not that high :)
20100628 22:44:02-!- gabba [~gabba@wesnoth/developer/gabba] has joined #wesnoth-dev
20100628 22:44:10< gabba> bonjour
20100628 22:44:25< billynux> bonjour gabba
20100628 22:44:36< gabba> salut billynux
20100628 22:45:04< billynux> Crab_, if all, it will be a great testing ground for ana!
20100628 22:45:51< timotei> hi gabba:D
20100628 22:46:02< gabba> hello timotei
20100628 22:49:59< billynux> I'm off to class people, bye & good luck. Crab_: thanks!
20100628 22:50:14< Crab_> bye :)
20100628 22:50:33-!- billynux [~billy@wesnoth/developer/billynux] has quit [Quit: Leaving]
20100628 22:50:54< timotei> bye guys :D
20100628 22:51:09-!- timotei [~timo@wesnoth/developer/timotei] has quit [Remote host closed the connection]
20100628 22:55:37-!- Crab_ [~Crab_@wesnoth/developer/crab] has left #wesnoth-dev []
20100628 22:56:28-!- elvish_sovereign [~Zoltan@pool-108-2-88-55.phlapa.east.verizon.net] has joined #wesnoth-dev
20100628 22:57:55-!- silene [~plouf@wesnoth/developer/silene] has quit [Quit: Leaving.]
20100628 22:59:37-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit []
20100628 23:16:40-!- Bob_The_Mighty [~chatzilla@cpc8-brig15-2-0-cust40.know.cable.virginmedia.com] has joined #wesnoth-dev
20100628 23:29:39-!- Upthorn [ogmar@adsl-75-26-190-247.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100628 23:32:13-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100628 23:32:46-!- Upth [ogmar@adsl-75-26-195-109.dsl.scrm01.sbcglobal.net] has quit [Ping timeout: 240 seconds]
20100628 23:36:22-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20100628 23:36:30< boucman> hey all
20100628 23:36:58< Espreon> Hello.
20100628 23:38:47-!- stikonas [~and@ctv-79-132-162-160.vinita.lt] has joined #wesnoth-dev
20100628 23:38:47-!- stikonas [~and@ctv-79-132-162-160.vinita.lt] has quit [Changing host]
20100628 23:38:47-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100628 23:40:29< boucman> hey gabba
20100628 23:40:33< boucman> how is it going ?
20100628 23:40:51< gabba> hello boucman, fine and you?
20100628 23:41:23< boucman> fine, fine...
20100628 23:41:24< gabba> I'm going to work on the switch between prototype A and B on move execution today
20100628 23:41:40< boucman> great, do you anticipate any difficulty ?
20100628 23:42:22< gabba> No real obstacle, just some structural decisions that might prove hard to make
20100628 23:42:36< boucman> hehe :)
20100628 23:42:50< boucman> isn't that more or less our problem since the beginning ?
20100628 23:43:04< gabba> pretty much, yes :)
20100628 23:43:07< boucman> more seriously, what decision, and do you need to discuss with me ?
20100628 23:44:51< gabba> I need to decide how I change the visuals: directly affect fake units in the manager, or store some state in the action objects
20100628 23:44:58< gabba> the second is more oo of course
20100628 23:46:18< gabba> Also when giving the illusion of the real unit being at its future position, I need some way of letting a move know it's the last move of its target unit, so it can change it's fake unit visuals accordingly
20100628 23:47:24< gabba> For this one I was trying to think up some kind of handle/semaphor object that you pass around and can be owned by a single move at a time might do the trick
20100628 23:47:49< boucman> gabba: how about storing the display state ("current_is_active vs future_is_active") in the manager and have the actions refer to it on redraw...
20100628 23:48:50< boucman> this doesn't solve how to find if an action is the first/last, though
20100628 23:49:09< gabba> boucman: I was trying to keep actions self-contained, and not introduce a dependency of actions on the manager. Bidirectionnal dependencies are usually a source of trouble...
20100628 23:49:54< boucman> indeed... though in that case it would be simpler and you can get the manager via ressources:: so no circular dependencies in the header
20100628 23:50:55< gabba> the header dependency is one thing, but what about OO principles :P ?
20100628 23:51:33< boucman> hehe
20100628 23:52:18< boucman> well, in that case I think respecting OO would have an extra cost (i.e make code actually more complicated) for no benefits. 
20100628 23:53:29< boucman> gabba: your find_visitor could be expanded to have a find_last action, to find the last action involving a unit... or you could keep a list of units for whom you already draw an actions, and thus further actions are not the first (the last when you go backward)
20100628 23:53:34-!- Khezar [~Greywhind@pool-96-238-43-241.prvdri.fios.verizon.net] has joined #wesnoth-dev
20100628 23:53:50-!- Khezar [~Greywhind@pool-96-238-43-241.prvdri.fios.verizon.net] has quit [Client Quit]
20100628 23:54:15< boucman> but whatever way you go, it's probably either the manager that tells the display that pass the info back to the actions, or the actions that look for it in the manager at redraw time...
20100628 23:54:31< gabba> ^You got a point. So, you've been the tie-breaker for this one, since I was considering both solutions.
20100628 23:54:48< boucman> the info is in the manager and needs to go into the actions, and having a copy of that info in all actions seems error prone
20100628 23:54:48< gabba> boucman: ok, two interesting ideas, I'll keep'em in mind
20100628 23:56:18< gabba> I have the manager/side_actions/action separation which is very clean and all, but sometimes figuring out what goes in which gets a bit tough
20100628 23:58:27< gabba> The third thing I need to address is the nasty flickering/redraw problems:
20100628 23:58:33< boucman> if you want to keep the clean separation you could make "is_future_drawing_mode" be a static member of the action class.... since it's a common parameter for all members of the class
20100628 23:59:37-!- happygrue_ [~George@wesnoth/developer/wintermute] has joined #wesnoth-dev
--- Log closed Tue Jun 29 00:00:01 2010