--- Log opened Mon Dec 22 00:00:10 2008
20081222 00:00:10< Soliton> ilor: maybe something for you: https://gna.org/bugs/index.php?12769
20081222 00:01:28< ilor> Soliton: looks like it, I'll assign it to myself
20081222 00:01:54< Soliton> great. :-)
20081222 00:05:34< ilor> sheesh, my VC9 build doesn't create a stderr.txt file
20081222 00:06:31< Ivanovic> Soliton: okay, done regarding backups for the moment
20081222 00:06:42< Ivanovic> should get all of /home as well as all four current databases
20081222 00:06:55< Ivanovic> plus the website
20081222 00:06:58< Soliton> nice.
20081222 00:07:54< Soliton> even updated /root/crontab. :-)
20081222 00:08:04< Ivanovic> jupp
20081222 00:08:22< Ivanovic> it should be identical to the one in /backup/scripts/root-crontab
20081222 00:08:38-!- boucman [n=rosen@159.83.102-84.rev.gaoland.net] has quit [Remote closed the connection]
20081222 00:08:50< Soliton> is there anything to this FR? http://gna.org/bugs/?12772
20081222 00:09:17< Soliton> seems a little naive to me to assume that binary saves will compress as well as text saves.
20081222 00:10:00< Ivanovic> yes, it really sounds so
20081222 00:11:20< Ivanovic> okay, now that we should have some at least partly relyable backups, i will head off to bed
20081222 00:11:22< Ivanovic> n8
20081222 00:11:48< Espreon> Good night.
20081222 00:17:23< CIA-8> torangan * r31613 /trunk/po/wesnoth-anl/nl.po: dutch translation update
20081222 00:19:50< Sirp> so is there anyone who uses Windows who'd like to investigate the best way for an application to send data in a firewall friendly way?
20081222 00:44:40< loonycyborg> Sirp: Perhaps you'd better ask virus writers about that :)
20081222 00:49:26< Sirp> loonycyborg: would it be rude of me to suggest that it's possible that you're the most likely to be located near many virus writers? ;)
20081222 00:50:45 * loonycyborg never communicated with any virus writer in any way :)
20081222 00:50:54< Sirp> well, neither have I. :)
20081222 00:51:10< Sirp> though they seem to try to communicate with me. ;)
20081222 00:53:38-!- grzywacz [n=grzywacz@moinmoin/developer/karol] has quit [":wq"]
20081222 00:54:28-!- Shadow_Master [n=ignacior@unaffiliated/shadowmaster/x-275924] has joined #wesnoth-dev
20081222 00:55:28< Shadow_Master> hello?
20081222 00:55:41< loonycyborg> hi
20081222 00:55:52< Shadow_Master> am I still in time to introduce a few image-path changes I wrote a week ago?
20081222 00:56:07< Espreon> Hello Shadow_Master.
20081222 00:56:50< loonycyborg> Shadow_Master: Big surprises await you in svn. Like code chagnes causing full recompile and a big, nice pot-update :)
20081222 00:57:04-!- Shadow_Master [n=ignacior@unaffiliated/shadowmaster/x-275924] has quit [Client Quit]
20081222 00:57:26< Espreon> loonycyborg: I think you scared him away...
20081222 00:58:49-!- cib0 [n=cib@p4FD0FE2C.dip.t-dialin.net] has left #wesnoth-dev []
20081222 00:58:50< loonycyborg> Oops! :)
20081222 00:58:51-!- Shadow_Master [n=shadowm@unaffiliated/shadowmaster/x-275924] has joined #wesnoth-dev
20081222 00:58:59< Espreon> Welcome back.
20081222 00:59:55< Shadow_Master> loonycyborg: I do not fear full recompiles ...
20081222 01:00:21< Shadow_Master> not after eperiencing some repeated HDD crashes on the desktop box
20081222 01:00:24< Espreon> But you hate translation updates.
20081222 01:02:01< loonycyborg> Shadow_Master: About your image changes: right in time :)
20081222 01:02:28< loonycyborg> Release just got delayed due to said full recompile causing changes.
20081222 01:03:38< Shadow_Master> oh :
20081222 01:05:17< loonycyborg> Does "right in time" make sense or it's blatant example of Runglish? :)
20081222 01:06:17< Shadow_Master> it translates well into Spanglish :o) 
20081222 01:06:33< Shadow_Master> although I think the real English should be "just in time" 
20081222 01:17:23-!- Espreon [n=Espreon@c-76-117-239-33.hsd1.pa.comcast.net] has quit ["ChatZilla 0.9.84 [Firefox 3.0.3/2008092417]"]
20081222 01:20:57-!- Dragonking [n=dk@dedikerad/dragonking] has quit []
20081222 01:25:22-!- loonycyborg [n=sergey@79.139.247.111] has quit ["Zzzzzzzzzzzzzzzzzzzzz"]
20081222 01:26:52-!- loonycyborg [n=sergey@79.139.247.111] has joined #wesnoth-dev
20081222 01:27:57< loonycyborg> /usr/bin/msgfmt -c --statistics -o translations/nl/LC_MESSAGES/wesnoth-anl.mo po/wesnoth-anl/nl.po
20081222 01:27:57< loonycyborg> \po/wesnoth-anl/nl.po:558: `msgid' and `msgstr' entries do not both end with '\n'
20081222 01:28:04< loonycyborg> Ivanovic: ^^
20081222 01:28:38-!- loonycyborg [n=sergey@79.139.247.111] has quit [Client Quit]
20081222 01:29:14-!- loonybot [n=loonybot@79.139.247.111] has quit [Remote closed the connection]
20081222 01:40:04< ilor> night all
20081222 01:40:13-!- ilor [n=user@unaffiliated/ilor] has quit []
20081222 02:05:31< thespaceinvader> Shadow_Master: there?
20081222 02:15:48< CIA-8> esr * r31614 /trunk/doc/manual/manual.txt: Edit pass before freeze - only needed minor changes.
20081222 02:26:16-!- thespaceinvader [n=chatzill@91.108.174.23] has quit ["ChatZilla 0.9.84 [Firefox 3.0.5/2008120122]"]
20081222 02:26:53-!- ESR__ [n=chatzill@static-71-162-243-5.phlapa.fios.verizon.net] has joined #wesnoth-dev
20081222 02:27:23< Sirp> hi ESR__ 
20081222 02:27:33< ESR__> Hi.
20081222 02:28:15< ESR__> Ivanovic, I;'ve done a pass over the msanual, looking particularly at the health sections.  It needed only trivial changes.
20081222 02:32:28< Sirp> ESR__: I'm guessing there's no chance that you know how firewalls work on $@$@#%@$ Windows? :)
20081222 02:32:43< ESR__> Sirp: None at all, sorry.
20081222 02:59:37 * ESR__ thinks putting a firewall on Wundows is a lot like making sure you have installed a deadbolt lock on a screen door.  On a submarine.
20081222 03:01:57< Sirp> ESR__: well the problem is that the Windows firewall installed by default on XP and Vista seems to (silently?) block outgoing connections a lot.
20081222 03:02:03< Sirp> and do it on an application basis.
20081222 03:03:03< Sirp> so basically it's like designing a prison and making it very easy for prisoners to escape or have visitors who can smuggle in items and so forth, but have incredibly heavy restrictions on prisoners sending mail from the prison.
20081222 03:03:31< ESR__> Yeah, that sounds like Microsoft, allright.
20081222 03:04:35< Sirp> in other news, my mother's new laptop running Vista can't play video DVD's, because "the required decoder is not installed". Why is it that Windows is more desktop-ready than Linux again if it can't even do something simple like that without displaying a bizarre error message?
20081222 03:38:04-!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has quit ["Hi! I'm a quit message virus vaccine. If you see a quit message virus, don't replace your quit message with it!"]
20081222 03:56:29-!- crimson_penguin [n=ben@64.201.60.217] has joined #wesnoth-dev
20081222 04:30:21-!- Ivanovic_ [n=ivanovic@dtmd-4db2b490.pool.einsundeins.de] has joined #wesnoth-dev
20081222 04:46:22-!- Ivanovic [n=ivanovic@dtmd-4db2a9e9.pool.einsundeins.de] has quit [Read error: 110 (Connection timed out)]
20081222 04:48:19-!- Ivanovic_ is now known as Ivanovic
20081222 05:14:48-!- BenUrban [n=benurban@unaffiliated/benurban] has quit [Read error: 104 (Connection reset by peer)]
20081222 05:15:20-!- BenUrban [n=benurban@unaffiliated/benurban] has joined #wesnoth-dev
20081222 05:27:29-!- dlr365 [n=doug@h44-199.reznet.ucalgary.ca] has joined #wesnoth-dev
20081222 05:41:15-!- crimson_penguin [n=ben@64.201.60.217] has quit ["Leaving"]
20081222 06:12:56-!- [Relic] [n=[Relic]@76.229.202.137] has quit ["Leaving"]
20081222 06:27:32< AI0867> Ivanovic: latest dutch translation update (wesnoth-anl) broke something
20081222 06:31:26-!- Sapient [n=sapien-x@c-24-126-133-155.hsd1.ga.comcast.net] has joined #wesnoth-dev
20081222 06:31:49< Sapient> nemo: check the Game Development forum, in off topic
20081222 06:32:08< Sapient> mt
20081222 06:43:31< Sapient> cya later, have fun.
20081222 06:43:31-!- Sapient [n=sapien-x@c-24-126-133-155.hsd1.ga.comcast.net] has left #wesnoth-dev []
20081222 07:32:32-!- kthakore2 [n=kthakore@CPE001310a1899c-CM001868e2ad12.cpe.net.cable.rogers.com] has joined #wesnoth-dev
20081222 08:30:19< CIA-8> zookeeper * r31615 /trunk/ (changelog players_changelog): Updated changelogs.
20081222 08:38:18-!- mordante [n=chatzill@roadie.xs4all.nl] has joined #wesnoth-dev
20081222 08:40:17< mordante> morning
20081222 08:48:50< Rhonda> Sirp: Restored, hope everything is fine in there.
20081222 08:50:13< Rhonda> uh oh
20081222 08:51:21< Rhonda> Soliton: Did you remove /backup/mysql_dbdump/forum yesterday? Thanks if you did. :)
20081222 09:10:19< Sirp> Rhonda: what'd you restore? I think people restored things earlier. :)
20081222 09:13:22< Rhonda> Sirp: There was no stats database, at least not from what I saw?
20081222 09:15:53< Sirp> Rhonda: well when did you restore it? Just now, or several hours ago?
20081222 09:17:33< Rhonda> Just now before I wrote it.
20081222 09:19:18< Sirp> Rhonda: there was a stats database before; someone (not sure who) restored it some hours ago. You seem to have reverted it back to that state (it had some updates since then; but no big deal)
20081222 09:20:51< Rhonda> erm ...
20081222 09:21:25< Sirp> well at least the stats database existed some hours ago, before I went off to dinner. :)
20081222 09:21:33< Rhonda> By Soliton's explicit hilight I though he didn't do it so ...  %-/
20081222 09:21:41< Rhonda> Sorry.
20081222 09:21:49< Sirp> the stats page at http://www.wesnoth.org/cgi-bin/stats/stats.pl which uses it was working, and when I reloaded it just now, the stats went backwards.
20081222 09:22:12< Sirp> Rhonda: it's not a problem at all, and thank you anyhow.
20081222 09:22:19< Rhonda> Hope I didn't destroy valueable hours of work. :(
20081222 09:23:20< Rhonda> Didn't know we primarily have apple players? Or ... is the stats sending part enabled there by default? ;)
20081222 09:26:50< Sirp> Rhonda: no, I think the problem is that XP and Vista have some application specific firewall thing that is blocking most of our data sending.
20081222 09:27:02< Sirp> Rhonda: do you know anything about how the firewalls work on XP/Vista? :)
20081222 09:28:21< mordante> Sapient regarding r31606 why not use lexical_cast_default instead?
20081222 09:35:50< Rhonda> Sirp: Erm, I deny to confess that I ever used that beats
20081222 09:42:02< Sirp> Rhonda: :) is okay
20081222 09:43:05< Rhonda> beasts*
20081222 09:49:52-!- ilor [n=user@unaffiliated/ilor] has joined #wesnoth-dev
20081222 09:50:03< mordante> hi ilor 
20081222 09:50:25< ilor> hi mordante
20081222 09:51:26< ilor> any thoughts on what could be causing the new dialogs to die in my build? ;)
20081222 09:51:56< mordante> no do you have a callstack?
20081222 09:52:14< ilor> right now I have it go down in gui2::tfield<int,class gui2::tinteger_selector_,int const >::validate_widget
20081222 09:53:55< ilor> so it doesnt actually crash, but gets a wml exception and closes
20081222 09:58:07< mordante> does it give info which widget is missing? 
20081222 09:58:16< mordante> what dialog does it happen?
20081222 09:59:24< ilor> all dialogs I tried made it close, with differents info in the exception
20081222 09:59:25-!- Sirp_ [n=me@203.167.141.18] has joined #wesnoth-dev
20081222 09:59:35< ilor> User message: Mandatory widget 'width' hasn't been defined.
20081222 09:59:43< ilor> for example, in the editor new map dialog
20081222 09:59:50-!- Sirp [n=me@203.167.141.18] has quit [Read error: 104 (Connection reset by peer)]
20081222 09:59:58< ilor> in the language dialog it was some other widget name
20081222 10:00:29< ilor> 'language_list' there
20081222 10:01:21< mordante> odd, works for me 
20081222 10:01:38-!- kthakore2 [n=kthakore@CPE001310a1899c-CM001868e2ad12.cpe.net.cable.rogers.com] has quit [Read error: 104 (Connection reset by peer)]
20081222 10:02:11< ilor> well I suppose it does work for everyone who isn't compiling in MSVC
20081222 10:02:50< mordante> you use MSVC now? which version?
20081222 10:02:55< ilor> 2008
20081222 10:03:04< ilor> got a working build just yesterday
20081222 10:03:24< mordante> no problem with the dummy unit?
20081222 10:03:41< ilor> well, the tutorial works
20081222 10:04:12< mordante> I fear that either MSVC does something wrong or there's some UB which works properly on g++
20081222 10:04:43< mordante> guess the latter might be more likely ;-)
20081222 10:04:57< ilor> there's quite a few warnings which would suggest that, but also the compiler behaves very strangely at times
20081222 10:05:19< ilor> where did people get the dummy unit bug?
20081222 10:06:11< mordante> wesbot bug 12524
20081222 10:06:12< wesbot> Bug #12524  Assigned to: Mark de Wever <mordante>  Status: None  Priority: 5 - Normal
20081222 10:06:15< wesbot> Summary: exception in dialogs.cpp
20081222 10:06:17< wesbot> Original submission: (rev 30375)
20081222 10:06:20< wesbot> URL: https://gna.org/bugs/?12524
20081222 10:06:57< mordante> what warnings does MSVC give which might indicate that?
20081222 10:07:20< ilor> yeah the load dialog crashes :/
20081222 10:08:37< ilor> there's a warning C4273: '_putenv' : inconsistent dll linkage which could mean anything, for instance
20081222 10:10:23 * Rhonda wants 1.5.7!
20081222 10:12:16< mordante> trunk first needs to be repaired before that ;-)
20081222 10:12:56< Rhonda> You mean, your part? :P
20081222 10:13:17< CIA-8> ivanovic * r31616 /trunk/ (8 files in 2 dirs): updated Turkish translation (manual including images)
20081222 10:13:46< Ivanovic> moin
20081222 10:14:11 * Rhonda . o O ( commit before moin in channel. baaaaaad Ivanovic )
20081222 10:14:13< mordante> no it wasn't me :-P
20081222 10:14:22< mordante> hi Ivanovic 
20081222 10:14:47< mordante> Ivanovic the Dutch translation of anl is broken
20081222 10:15:16< mordante> Rhonda feel free to blame the Netherlands :-P
20081222 10:17:53< mordante> Ivanovic I've a beautiful commit tonight, got email from Kitty :-)
20081222 10:18:25< CIA-8> ivanovic * r31617 /trunk/data/multiplayer/scenarios/ANL_utils/ANL_research_options.cfg: remove ~500 spaces from a line only consisting of those (because those are not trival to see in translations and easily are forgotten leading to probs with the translation due to *not* ending with a newline)...
20081222 10:18:56< Rhonda> mordante: broken translations break the compile all together?!
20081222 10:19:13< Rhonda> hmmm
20081222 10:19:26< Rhonda> Did the graphic change much between 1.5.6 and 1.5.7?
20081222 10:19:43< Rhonda> I am asking because of the requirement for the main-menu picture.
20081222 10:19:51< CIA-8> ivanovic * r31618 /trunk/po/wesnoth-anl/nl.po: fix dutch translation file not compiling
20081222 10:19:52< Rhonda> Needs to get redone for the manual.
20081222 10:19:58< Ivanovic> Rhonda: no, they did not change much
20081222 10:20:10< Rhonda> And I don't want to pull svn just to be able to do proper screenshots.
20081222 10:20:16< Ivanovic> just make sure that you got a picture with the new editor in the menu
20081222 10:20:46< Ivanovic> and wait 2 more mins before translating, i am about to commit an update to the po files for the manual since esr fixed some stuff last night
20081222 10:21:21< mordante> Rhonda no but it breaks the make of the translations
20081222 10:21:40< mordante> but looks like Ivanovic fixed the blocker already :-)
20081222 10:22:02< Ivanovic> mordante: i fixen fixed the cause, that is a string which is *difficult* to see
20081222 10:22:16< CIA-8> ivanovic * r31619 /trunk/ (64 files in 2 dirs): 
20081222 10:22:16< CIA-8> regenerate doc files
20081222 10:22:16< CIA-8> po update for the manual (fixes by esr)
20081222 10:22:36-!- loonybot [n=loonybot@79.139.247.111] has joined #wesnoth-dev
20081222 10:23:33-!- loonycyborg [n=sergey@79.139.247.111] has joined #wesnoth-dev
20081222 10:24:47< mordante> you did fix the problem, works now ;-)
20081222 10:25:01< mordante> loonycyborg thanks for your fix as well, it works again :-)
20081222 10:25:21< loonycyborg> np
20081222 10:25:39< Ivanovic> Rhonda: i removed the folder /backup/mysql_dbdump/forum
20081222 10:25:49-!- kthakore2 [n=kthakore@CPE001310a1899c-CM001868e2ad12.cpe.net.cable.rogers.com] has joined #wesnoth-dev
20081222 10:26:02-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20081222 10:26:02< Rhonda> Ivanovic: Ah, then the thanks have to go to you. Sorry, Soliton :P
20081222 10:26:29< Rhonda> Ivanovic: Maybe we should add all databases to that backup hook?
20081222 10:26:35< Ivanovic> and i had restored the stats database, too
20081222 10:26:41< Ivanovic> and yes, they are now all in the hook
20081222 10:26:46< Rhonda> Good. :)
20081222 10:26:55< Ivanovic> at least all databases that were on the old drive should now be on the new one, too
20081222 10:27:00< Ivanovic> and listed in the backup hook
20081222 10:27:06< Ivanovic> and now all of /home is backed up, too
20081222 10:27:26< Ivanovic> and a copy of the olddrive /home is in /backup, too, just to be sure
20081222 10:27:39< Rhonda> Hmm, mysql itself is missing, that is, the access informations and users stuff ...
20081222 10:28:04< Rhonda> I think that was more valueable than pushing 1.5.7 out, thanks for your efforts.
20081222 10:28:07< Ivanovic> uhm, then add this one, too
20081222 10:28:38< Rhonda> added
20081222 10:28:39< Ivanovic> and for not pushing 1.5.7 out you can blame mordante
20081222 10:28:46< Ivanovic> since his changes are not ready yet
20081222 10:28:55< Ivanovic> mordante: and yesterday you missed kitty in here
20081222 10:29:11< Rhonda> Ah, different kitty
20081222 10:30:06< Ivanovic> uhm, no, our portrait kitty
20081222 10:32:03< Ivanovic> Rhonda: is there anything else on the server you would like to see backed up, maybe /etc?
20081222 10:32:45< mordante> Ivanovic I know I'm too blame :-(
20081222 10:33:18< mordante> yes I know I missed her, I read the logs
20081222 10:34:23< mordante> when do you want to release now, you want to wait for my changes or not?
20081222 10:35:05< Ivanovic> mordante: get them in ASAP and i will release once they are in and i find the time to do so
20081222 10:35:21< Ivanovic> (no, this will not be on wednesday at exactly 7pm)
20081222 10:35:36< mordante> Ivanovic well I still need to debug some things
20081222 10:35:43< Ivanovic> but due to me having some holidays now i can basically release every day of the week
20081222 10:35:57< Ivanovic> what do you think of just doing so then?
20081222 10:35:58< Ivanovic> ;)
20081222 10:36:03< Rhonda> Ivanovic: Not too sure currently, sorry.
20081222 10:36:22 * mordante . o O (what's with Wednesday 7 pm?)
20081222 10:36:47< Ivanovic> mordante: you know, celebrating christmas, dinner with the family, stuff like this
20081222 10:37:25< mordante> oh yes of course you celebrate the day before
20081222 10:37:41< mordante> we normally celebrate the 25th and 26th
20081222 10:37:52< mordante> we == in the Netherlands
20081222 10:38:24< Ivanovic> we celebrate on 25th, too
20081222 10:38:37< Ivanovic> that is: the main stuff is on 24th (presents and such)
20081222 10:38:42< mordante> yes but we don't celebrate on the 24th
20081222 10:40:15< mordante> and traditionally presents happen at Nicolaas but now also with Christmas, but on the 25th
20081222 10:43:36< busfahrer> What a nice present, that day's my birthday :-D
20081222 10:44:36< Ivanovic> ahhh, ain't vista lovely? http://www.wesnoth.org/forum/viewtopic.php?p=327239#p327239
20081222 10:46:42< busfahrer> Who uses a 1.5 GHz box with 3 GB of RAM?
20081222 10:47:18< loonycyborg> People who upgrade their pcs :)
20081222 10:48:02< loonycyborg> RAM is cheaper and matters much more sometimes.
20081222 10:52:28-!- zookeeper [n=l@dsl-tkubrasgw1-fe9ff800-215.dhcp.inet.fi] has joined #wesnoth-dev
20081222 10:54:21< ilor> mordante: the current approach with the dummy unit is UB
20081222 10:55:28< loonycyborg> shikadibot: UB
20081222 10:55:28< Ivanovic> busfahrer: someone who wants to use vista
20081222 10:55:29 * shikadibot is offended
20081222 10:55:35< loonycyborg> shikadibot: UB?
20081222 10:55:35< shikadibot> Sorry, I don't know what 'UB' means...
20081222 10:55:44< Ivanovic> ;)
20081222 10:56:03< mordante> ilor UB or unspecified behaviour?
20081222 10:56:10< ilor> UB
20081222 10:56:38< ilor> some versions of gcc catch simple testcases with "error: attempt to compare iterators from different sequences"
20081222 10:57:01< loonycyborg> Last time I encountered UB it meant "Universal Binary" :)
20081222 10:57:18< mordante> it's undefined behaviour 
20081222 10:57:35< ilor> I guess it passes compilation if it's veiled in function that returns some iterator that may or may not be from the same sequence
20081222 10:57:46< ilor> difference is MSVC checks it at runtime
20081222 10:57:48< mordante> yes that happens
20081222 10:58:29< ilor> who wrote that bit anyway?
20081222 10:58:31< mordante> well I read a discussion on usenet about it not that long ago and can't remember whether it's UB or unspecified or badly worded in the standard
20081222 10:58:35< mordante> YogiHH
20081222 10:58:53< ilor> has he been around lately?
20081222 10:59:08< mordante> I already had a short look at it, but too busy to have a longer look
20081222 10:59:22< loonycyborg> ilor: gcc checks iterators too with -DGLIBCXX_DEBUG
20081222 10:59:32< mordante> he pops in at times, but I already discussed it with him a while ago
20081222 10:59:36< ilor> and?
20081222 11:00:13< mordante> he also wasn't sure about the best solution
20081222 11:01:27< ilor> I'd say have unit_map::find() return the end() iterator if the unit wasn't found and make the calling code deal with it
20081222 11:04:08< mordante> yes but I'm not sure whether there's a difference between not found and dummy
20081222 11:05:34< ilor> the offending bit in find() returns the dummy if the unit wasn't found
20081222 11:05:40< ilor> ok now for a different issue
20081222 11:06:00< ilor> why the hell doesn't MSVC want to compile src/gui/widgets/image.cpp?
20081222 11:07:11< mordante> ilor found the discussion http://groups.google.com/group/comp.std.c++/browse_thread/thread/50d0e06743308049?fwc=1&pli=1
20081222 11:07:30< mordante> seems indeed unclear but UB was intended
20081222 11:07:54< mordante> I'm off now
20081222 11:07:55< ilor> yeah I found the same thing and also asked in ##c++
20081222 11:07:56< mordante> bye
20081222 11:07:58< ilor> bye mordante
20081222 11:08:18< mordante> maybe I should join ##c++ as well
20081222 11:08:28-!- mordante [n=chatzill@roadie.xs4all.nl] has quit ["ChatZilla 0.9.84 [Firefox 3.0.4/2008102920]"]
20081222 11:11:22 * loonycyborg got codeblocks projectfiles that use scons to build :)
20081222 11:16:39< CIA-8> loonycyborg * r31620 /trunk/projectfiles/CodeBlocks-SCons/ (. wesnoth.cbp wesnoth.workspace wesnothd.cbp): Added CodeBlocks projectfiles which use SCons to build.
20081222 11:21:13-!- Shadow_Master_ [n=ignacior@unaffiliated/shadowmaster/x-275924] has joined #wesnoth-dev
20081222 11:21:50< Shadow_Master_> three days awake and still alive... :(
20081222 11:22:38< loonycyborg> Three days awake is enough to go bonkers!
20081222 11:25:41-!- Shadow_Master_ [n=ignacior@unaffiliated/shadowmaster/x-275924] has quit [Client Quit]
20081222 11:32:06-!- Unnheulu [n=ieuan@cpc1-pnth1-0-0-cust903.cdif.cable.ntl.com] has joined #wesnoth-dev
20081222 11:32:23 * Unnheulu heheh
20081222 11:33:00< Unnheulu> after chrsitmas i might try my hand at deving wesnoth
20081222 11:33:08< Unnheulu> where would be a good palce to start?
20081222 11:33:52< loonycyborg> Unnheulu: Whatever interests you most :)
20081222 11:34:15< Unnheulu> im not great a c++ though, only spent about 6 months learning it between nnow and 2004 ;)
20081222 11:34:22< loonycyborg> Whatever itch you want to scratch :)
20081222 11:36:34< Ivanovic> loonycyborg: no need for extra cmake specific code::blocks files
20081222 11:36:43< Ivanovic> since cmake is able to output those project files
20081222 11:36:44< Ivanovic> ;)
20081222 11:38:35< loonycyborg> Ivanovic: But cmake generated project files cannot run scons ;)
20081222 11:39:56< loonycyborg> And they're hard to change since they're supposed to be constatnly regenerated.
20081222 11:39:58< Ivanovic> loonycyborg: that is correct
20081222 11:40:05< Ivanovic> this is correct, too
20081222 11:40:13< Ivanovic> oh, and cmake can output scons files
20081222 11:40:14< Ivanovic> ;)
20081222 11:40:27< loonycyborg> Since when? :)
20081222 11:40:30< Ivanovic> and yes, those files are not as neat as the handtined ones
20081222 11:42:19< Ivanovic> hmm, thought this was possible somehow
20081222 11:42:24< Ivanovic> no idea where i read it
20081222 11:42:54< loonycyborg> That's impossible until someone writes scons backend.
20081222 11:43:10< loonycyborg> Unless someone already did that.
20081222 11:48:51-!- dlr365 [n=doug@h44-199.reznet.ucalgary.ca] has quit ["rawr I'm an emu!"]
20081222 11:58:25< CIA-8> ivanovic * r31621 /trunk/data/gui/default/window/mp_method_selection.cfg: add a missing ' in string
20081222 12:04:12< CIA-8> ivanovic * r31622 /trunk/po/ (wesnoth/de.po wesnoth-lib/de.po): updated German translation (tiny update for screenshots for the manual)
20081222 12:19:47-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has quit []
20081222 12:27:24< CIA-8> ivanovic * r31623 /trunk/ (10 files in 5 dirs): 
20081222 12:27:24< CIA-8> updated German translation
20081222 12:27:24< CIA-8> regenerated doc files
20081222 12:28:31< Ivanovic> Rhonda: it would be really great if you could have a look at our php settings to make sure that we are not sooo vulberable to bugs in php
20081222 12:29:03< Ivanovic> that is: the german gp2x forum (and with it the pandora forum) were hacked some days ago probably due to a bug within phpbb)
20081222 12:29:31< Ivanovic> and maybe you could have a look at updating our phpbb version, since it is rather outdated (and I failed miserably when trying to upgrade it right after our crash)
20081222 12:29:36-!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20081222 12:29:47< Ivanovic> ilor: the editor seems to have a real problem with translations
20081222 12:30:14< Ivanovic> ilor: that is several buttons like "Quit Editor" are not translated, even though they should be (since the strings are in the po files)
20081222 12:30:23< Ivanovic> it would be great if you could investigate in this
20081222 12:30:54< Ivanovic> just select rg the turkish translation, there most stuff should be translated and appear differently than in english
20081222 12:31:05< Ivanovic> strings you just find in english are most likely broken
20081222 12:32:02 * loonycyborg wonders why turkish has most progress in translation
20081222 12:32:04< Ivanovic> afk for a while
20081222 12:32:07 * zookeeper is making good progress on the ai controller again
20081222 12:38:17-!- Sirp_ [n=me@203.167.141.18] has left #wesnoth-dev []
20081222 13:01:18-!- kthakore2 [n=kthakore@CPE001310a1899c-CM001868e2ad12.cpe.net.cable.rogers.com] has quit ["Leaving"]
20081222 13:03:34< Rhonda> Ivanovic: We are not sooo vulnerable to bugs in php if we stop using php. :)
20081222 13:04:38< loonycyborg> Rhonda: Do you know any good alternatives to phpBB?
20081222 13:05:06< Rhonda> Mailinglists, Newsgroups? :P
20081222 13:06:01 * loonycyborg hates mailing lists and newsgroups
20081222 13:06:29 * loonycyborg thinks that e-mail should be only for personal communication
20081222 13:14:55-!- Elvish_Pillage2 [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20081222 13:27:22< CIA-8> ivanovic * r31624 /trunk/doc/manual/images/de/recruit-1.5.7.jpg: add missing image to german manual
20081222 13:31:26-!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20081222 13:39:51< CIA-8> fendrin * r31625 /trunk/data/campaigns/Legend_of_Wesmere/maps/Kalian.map: LoW 3+7 map: minor cosmetic changes
20081222 13:50:40-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20081222 13:59:27 * Rhonda hates forums
20081222 13:59:36 * Rhonda thinks that forums should be banned from the net
20081222 14:00:40< Rhonda> Inconvenient to the end, not adaptable to ones workflow, one has to actively look around instead of having any convenience with properly getting presented of new things.
20081222 14:01:30< loonycyborg> But they look so pretty :P
20081222 14:02:34< Rhonda> Yeah, and one person decides the way others are able to use it, and if they don't like they can go someplace else. Quite interesting approach to keeping the userbase to only a specific person group.
20081222 14:03:05< Rhonda> And ... erm ... pretty. Yeah, sure.
20081222 14:05:54< loonycyborg> And they work in any browser, don't require you to configure mail clients etc..
20081222 14:06:51< zookeeper> err, i suppose you don't really use the forums much if you think mailing lists and newsgroups would somehow be more convenient.
20081222 14:16:00< CIA-8> fendrin * r31626 /trunk/data/campaigns/Legend_of_Wesmere/utils/ai_controller.cfg: disabled the ai_controller popup.
20081222 14:16:09< Rhonda> loonycyborg: That's an overstatement that they work in _any_ browser.
20081222 14:16:24 * Shadow_Master simply doesn't pay attention to new topics
20081222 14:17:20< Rhonda> zookeeper: I haven't yet found in _any_ forum the posibility to only get displayed not-yet read entries. Or limiting the displayed entries to ones that match specific patterns.
20081222 14:18:10< Rhonda> In all forums I had to either log in within a specific timeframe so that it can track my last-visited timestamp properly, and even then none of them were able to conveniently display me unread items. They _try_, but they fail miserably.
20081222 14:18:51< loonycyborg> Rhonda: Wesnoth's forum works in lynx. I just tried :)
20081222 14:19:07< Rhonda> loonycyborg: Completely? The way it's "meant to be"?
20081222 14:19:43< Rhonda> The last-read thing is alread hilarious in firefox, but it's just tireing in lynx.
20081222 14:19:51< loonycyborg> Well.. I can browse posts, so..
20081222 14:20:11< Rhonda> Besides that lynx is an extremely bad example because it can't cope with tables or frames.
20081222 14:20:20< Ivanovic> Rhonda: uhm, i am using firefox and it works nicely over here
20081222 14:20:41< Ivanovic> that is, both "view unread posts" and "view new posts" works
20081222 14:21:11< Rhonda> Ivanovic: What part? I didn't say that it wouldn't work in firefox - if you don't care about missing something because there is no real read/unread tracking but assumptions of when you open a thread that you read it completely.
20081222 14:21:22< Ivanovic> and the "view unread posts" basically shows all posts not marked red
20081222 14:21:33< Ivanovic> and when you click on the orange image, you directly start at the first new post
20081222 14:21:55< Ivanovic> ah, so you want to set markers how far you were
20081222 14:21:59< Rhonda> Ivanovic: That's the point. What gets marked read? I don't see anything to be able to mark something as unread or read later. The forum just thinks I read everything it presents me with.
20081222 14:22:01< Ivanovic> you are right, this is not possible
20081222 14:22:34< Rhonda> And the "view unread posts" is just an extremely uncomfortable assumption of a single timestamp, not even one for each thread.
20081222 14:22:40< Ivanovic> Rhonda: and for every post there is a link "mark post as unread"
20081222 14:22:47< Rhonda> Which makes it even more of a laugh.
20081222 14:22:54< Ivanovic> and yes, this one works perfectly
20081222 14:23:08< Ivanovic> at least over here setting this does work
20081222 14:23:21< Rhonda> That's the wrong way round. That's putting buckets under a leaking roof instead of fixing the roof.
20081222 14:23:32< Ivanovic> uhm, is it?
20081222 14:23:55< Ivanovic> the asumption is "when you join a thread, you read it or don't care about the rest you don't read" so it is marked as read
20081222 14:24:15< Ivanovic> and when you not completely read it but want to continue later on, you just set the "mark as unread" and continue there later on
20081222 14:24:22< Rhonda> The forum definies how I am allowed to use it instead of giving me the posibility to choose a client that would adapt to the way a convenient workflow works for me.
20081222 14:24:27< Ivanovic> sounds as the perfectly sane asumption to me
20081222 14:24:38< Rhonda> It's just that it doesn't really work.
20081222 14:25:09< Rhonda> Tried exactly that approaches in several different forums, including phpbb, never worked. They always forgot these things when I didn't login just the next day again or such.
20081222 14:43:44-!- EdB [n=EdB@57.117.88-79.rev.gaoland.net] has joined #wesnoth-dev
20081222 14:58:27-!- Behl [n=tobizaz@3510ds4-od.0.fullrate.dk] has quit []
20081222 15:14:59< zookeeper> Rhonda, so, uh, how _should_ it work then?
20081222 15:24:34< Rhonda> zookeeper: Like in any usenet client or mail client. Not claiming one has read everything up to a specific timestamp but only the messages they have read.
20081222 15:25:25< Rhonda> Hell, even within a thread one doesn't usually read all the entries one after another but might want to jump either following a discussion branch (which isn't possible in the forum at all) or skip a whole branch but stay interested in the others.
20081222 15:26:54< Rhonda> And breaking a discussion thread is highly inconvenient, too. When topic drifts off eith either loses its connection when being taken to a different place, or it annoys other people reading the thread but not interested in the way that subpart of it went.
20081222 15:27:21< zookeeper> well, sounds fine assuming that i wouldn't have to constantly click around to select which posts i'd like to read, instead of just quickly scrolling through the whole thing to see if there's anything interesting.
20081222 15:28:28< CIA-8> shadowmaster * r31627 /trunk/src/image.cpp: 
20081222 15:28:28< CIA-8> Tiny possible optimization: load_image_sub_file()
20081222 15:28:28< CIA-8> Eliminates calls to recolor_image() when the ~RC() or ~TC() functions
20081222 15:28:28< CIA-8> haven't been actually used in the path. Although recolor_image() knows
20081222 15:28:28< CIA-8> that it should do nothing in such a situation, the extra calls
20081222 15:28:31< CIA-8> (recolor_image() -> std::map<Uint32,Uint32>::size()) which the old
20081222 15:28:33< CIA-8> method implied worry me slightly.
20081222 15:28:39< CIA-8> shadowmaster * r31628 /trunk/src/image.cpp: Remove a potential loose reference-to-const
20081222 15:28:56< CIA-8> shadowmaster * r31629 /trunk/src/image.cpp: 
20081222 15:28:56< CIA-8> Expose adjust_surface_color() as the ~CS(<R>,<G>,<B>) image-path function
20081222 15:28:56< CIA-8> The default value for R, G, B is zero.
20081222 15:29:14< CIA-8> shadowmaster * r31630 /trunk/src/image.cpp: Apply the ~GS() (greyscale) modification _after_ ~CS() (color-shift)
20081222 15:29:20< Rhonda> zookeeper: That's why a forum doesn't work, it depends on one either clicking around like hell or have no comfort at all. The browser interface doesn't really allow for covenient (or even, personal) shortcuts.
20081222 15:29:28< CIA-8> shadowmaster * r31631 /trunk/src/image.cpp: (log message trimmed)
20081222 15:29:28< CIA-8> Implement ~R(), ~G() and ~B() as subsets of the ~CS() functionality.
20081222 15:29:28< CIA-8> These are merely syntactic sugar. ~CS() should be, in general, faster
20081222 15:29:28< CIA-8> than a ~R()~G()~B() sequence since it doesn't have to iterate over 2
20081222 15:29:30< CIA-8> additional image mods in the sequence.
20081222 15:29:32< CIA-8> However, for single channel changes, one of them is faster than ~CS()
20081222 15:29:34< CIA-8> since no argument-splitting is made (which also implies that ~R(2,2)
20081222 15:29:42< CIA-8> shadowmaster * r31632 /trunk/src/image.cpp: 
20081222 15:29:42< CIA-8> Expose blur_alpha_surface() as the image-path function ~BL()
20081222 15:29:42< CIA-8> ~BL() takes only one argument, which seems to be the radius of blur
20081222 15:29:44< CIA-8> (blur_alpha_surface(), 2nd argument a.k.a. 'depth').
20081222 15:29:54< CIA-8> shadowmaster * r31633 /trunk/src/image.cpp: (log message trimmed)
20081222 15:29:54< CIA-8> Expose adjust_surface_alpha() as the ~O() image-path function
20081222 15:29:54< CIA-8> (O for Opacity).
20081222 15:29:54< CIA-8> Unlike ~CS() and sons, this is a "pretty" function which may be given an
20081222 15:29:56< CIA-8> argument in either a multiplier form (i.e. 0.5 for half-opaque, 1.0 for
20081222 15:29:58< CIA-8> original opacity), or as a percentage (50% and 100% for the
20081222 15:30:00< CIA-8> aforementioned examples). If the argument exceeds 100% (1.0), things get
20081222 15:30:13< CIA-8> shadowmaster * r31634 /trunk/src/image.cpp: Added some code comments
20081222 15:30:53< Rhonda> Shadow_Master: Don't destroy our discussion.  :P
20081222 15:34:19< Shadow_Master> blame git svn
20081222 15:35:41< Rhonda> ah, dcommit :)
20081222 15:36:16< CIA-8> shadowmaster * r31635 /trunk/changelog: Update changelog, mentioning new image-path functions
20081222 15:38:33< zookeeper> Shadow_Master, nice O.o
20081222 15:38:46< zookeeper> wanna make a scale image path function too :P?
20081222 15:42:59< Shadow_Master> yeah, later
20081222 15:56:03-!- wp_ [n=chatzill@85.24.79.72.generic-hostname.arrownet.dk] has joined #wesnoth-dev
20081222 15:58:35< wp_> Hello there... I tried to make my own map in the map editor... I have added players, villiges and castles... but I get a error syaing that "the scenario is invalid becuse there is no sides"... so where do I add these sides? can't find anything in the editor about that... pls help
20081222 16:00:28< zookeeper> use the side starting position placement tool
20081222 16:00:32< zookeeper> the little blue flag IIRC
20081222 16:03:39< wp_> yep... placed 4 players, nothing changes
20081222 16:04:33-!- cib0 [n=cib@p4FD0BF34.dip.t-dialin.net] has joined #wesnoth-dev
20081222 16:08:33< wp_> ohhh... had placced 2 of the players in the outer boarder region of the game... I moved them one square in and that helped.... it is a great game ;)
20081222 16:08:40-!- wp_ [n=chatzill@85.24.79.72.generic-hostname.arrownet.dk] has quit ["ChatZilla 0.9.84 [Firefox 3.0.5/2008121621]"]
20081222 16:37:24-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20081222 16:38:12-!- mordante [n=chatzill@roadie.xs4all.nl] has joined #wesnoth-dev
20081222 16:38:42< mordante> hi
20081222 16:38:52< loonycyborg> hello
20081222 16:38:59< mordante> hi loonycyborg 
20081222 16:42:43< Ivanovic> welcome back mordante
20081222 16:42:54< mordante> hi Ivanovic 
20081222 16:48:13< mordante> Shadow_Master what's r31628 about?
20081222 16:51:26< loonycyborg> mordante: Does utils::split return by value?
20081222 16:51:34< mordante> yes
20081222 16:51:40-!- crimson_penguin [n=ben@64.201.60.217] has joined #wesnoth-dev
20081222 16:52:15< loonycyborg> So, what assigning that return value to a reference do?
20081222 16:52:55< mordante> make a reference to a temporary, but since it's a const reference it extends its livetime
20081222 16:53:11 * loonycyborg thinks that his last phrase is botched
20081222 16:53:19< mordante> ?
20081222 16:53:55< loonycyborg> Perhaps it should be "what *does* assigning that return value to a reference do"
20081222 16:54:33< mordante> it's a const reference so it binds itself to the temporary and extends it's livetime to the livetime of the reference
20081222 16:54:55< mordante> (welcome to the joyful world of reference binding rules ;-) )
20081222 16:55:26< loonycyborg> Interesting..
20081222 16:55:52 * loonycyborg really needs to read another C++ book
20081222 16:55:53 * mordante hands ESR_ a new clip of ammunition ;-)
20081222 16:56:10< ilor> hi mordante
20081222 16:56:14< mordante> hi ilor 
20081222 16:56:22< mordante> any luck with finding the problem?
20081222 16:56:45< cib0> i wonder whether C++ will still be "in" in 10-20 years
20081222 16:56:58< loonycyborg> In what? :)
20081222 16:57:15< cib0> popular among programmers
20081222 16:57:48< Ivanovic> ilor: please have a look at the editor and translations
20081222 16:58:02< Ivanovic> ilor: something is really fishy there since they ain't working for several menu entries
20081222 16:58:34< Ivanovic> that is: eg in sorting the hotkeys, the translated value is considered for *ordering*, but it is not used at displaying
20081222 16:59:06< loonycyborg> cib0: Yes. Unless they manage to learn Haskell :P
20081222 16:59:29< Ivanovic> (when using the german translation, just look for "Quit Editor", the german string would be "Editor beenden" and thus be by far closer to the top, but displayed (at the correct position for "Editor beenden" is "Quit Editor")
20081222 16:59:31< mordante> cib0 I think that's hard to predict, guess K&R never expected C to be along that long
20081222 17:00:00< cib0> yeah, i guess
20081222 17:00:00< mordante> but I think it will be at least around, but whether it's still a sexy language no idea
20081222 17:01:40< Shadow_Master> mordante: const references... remember your memory corruption issues with that story screen?
20081222 17:02:07< loonycyborg> It'll likely take place of C in low-level system programming :)
20081222 17:02:19< cib0> i don't really feel like learning all the special things about C++, but i guess if they dont throw out a great, much better language the next years, i will get more familiar with it whether i want or not
20081222 17:02:49< mordante> Shadow_Master yes, but we use const references a lot and this one should be save
20081222 17:03:36< Shadow_Master> why mine wasn't safe? ,.,
20081222 17:03:38< loonycyborg> cib0: All those special things represent inherent complexity of the problem. You won't run away from them no matter what language you use.
20081222 17:03:59< mordante> cib0 well I think it won't look bad on your resume ;-) and besides when you master one language the second is easier to learn
20081222 17:04:35< loonycyborg> C/C++ just don't hide all those ugly details so much as other languages.
20081222 17:04:39< mordante> Shadow_Master I still think that bug lately was rather fishy, haven't looked close into it, but it was weird
20081222 17:04:45< mordante> assembly :-P
20081222 17:06:54< loonycyborg> Only python manages to successfully hide all this reference/pointer stuff :P
20081222 17:07:04< loonycyborg> At cost of being slow.
20081222 17:07:05< cib0> C/C++ dont hide anything ugly in particular
20081222 17:07:31< mordante> Pascal also does a nice job on it
20081222 17:08:06< Shadow_Master> mordante: well, I'll revert that tiny hange then
20081222 17:08:27< cib0> i actually like stuff like pointers and that.. what i hate is having to do type conversion without knowing the right tools
20081222 17:09:01< mordante> Shadow_Master ok, most of the time a const reference binding is save
20081222 17:09:09< loonycyborg> "right tools"?
20081222 17:09:26< mordante> our tstring class is one of the exceptions
20081222 17:09:28< cib0> tools, functions, whatever you call them
20081222 17:09:49< Shadow_Master> mordante: aha!
20081222 17:10:00< CIA-8> shadowmaster * r31636 /trunk/src/image.cpp: 
20081222 17:10:00< CIA-8> Revert "Remove a potential loose reference-to-const"
20081222 17:10:00< CIA-8> This reverts commit 662b881a5918ab88ca6e576fc561e48cddbd57c4.
20081222 17:10:00< CIA-8> (SVN changeset of revision r31628).
20081222 17:10:10< Shadow_Master> that caused the massive leak that invoked the awesome powers of the kernel's OOM killer during my debugging session :(
20081222 17:10:15< Shadow_Master> (before 1.5.6 release)
20081222 17:10:28< mordante> oke
20081222 17:10:39< fendrin__> cib0: maybe casts?
20081222 17:11:00< cib0> luckily wesnoth usually provides converters between the used types already, that's why i find it much easier to modify wesnoth than to start a program from scratch
20081222 17:11:02< ilor> where do I manually change the game language?
20081222 17:11:02< mordante> I can't remember the exact details, I had a lot of problems due to it
20081222 17:11:05< Shadow_Master> mordante: would be nice if a comment on that regard was added to the t_string interface :)
20081222 17:11:44< cib0> yeah, i guess it is called casts
20081222 17:12:02< mordante> Shadow_Master I thought it was already done, but appears not to be the case
20081222 17:13:07< mordante> meh :-( and I can't remember the exact details anymore
20081222 17:13:11< Shadow_Master> "this class' methods, if used incorrectly, can fry your CPU and consume the entire virtual memory space"
20081222 17:13:16-!- Unnheulu [n=ieuan@cpc1-pnth1-0-0-cust903.cdif.cable.ntl.com] has left #wesnoth-dev ["Ex-Chat"]
20081222 17:14:10< mordante> I think the only problem is config::operator[] which when cast to a string causes some UB which works properly on most compilers (except some ancient gcc)
20081222 17:14:31< mordante> (I guess the UB is removed when the compiler uses RVO)
20081222 17:15:19 * loonycyborg *does* remember that RVO is return value optimization :)
20081222 17:15:29< mordante> jup
20081222 17:15:42< CIA-8> ilor * r31637 /trunk/src/ (floating_textbox.cpp language.cpp log.cpp): add some missing includes VC9 complained about
20081222 17:15:46< Shadow_Master> ugh, you guys did a lot of stuff for a week...
20081222 17:16:10< ilor> anyone? manually change the language if I don't have the dialog workin atm?
20081222 17:16:49< mordante> set the language in the preference fiel
20081222 17:16:53< mordante> file*
20081222 17:16:54< loonycyborg> IIRC there's some config file in userdata
20081222 17:27:41< Shadow_Master> mordante: where did the new transarent portraits go?
20081222 17:28:00< Shadow_Master> e.g. I am still examining the git log for WML-impacting changes, so...
20081222 17:28:00< mordante> portraits/artist/transparent
20081222 17:28:27< Shadow_Master> I am blind...
20081222 17:28:39< mordante> and I'll add the latest ones today :-)
20081222 17:28:40< Shadow_Master> or I didn't get them with the git svn rebase
20081222 17:28:48< mordante> locate Kitty ;-)
20081222 17:28:56< Shadow_Master> I still see just the arch mage (f), lich, shyde and druid
20081222 17:29:12< mordante> that's all there is at the moment
20081222 17:29:29< Shadow_Master> what was that commit about then? mmmm
20081222 17:29:55< mordante> of another artist, the peasant, swordsman and spearman
20081222 17:30:31< ilor> okay dumb question time
20081222 17:30:37< Shadow_Master> um, I like all of them except the peasant :(
20081222 17:30:42< ilor> what do I do to get translations working in my working copy?
20081222 17:31:10< loonycyborg> Provide .mo files in translations/ dir
20081222 17:31:14< mordante> I like the peasant too
20081222 17:31:26< Shadow_Master> _except_
20081222 17:31:50< mordante> I like them all, the peasant too ;-)
20081222 17:32:20< Shadow_Master> your first sentence parsed as mangled spanish in my head :( sorry
20081222 17:33:53< mordante> well the sentence was a bit mangled ;-)
20081222 17:34:23< CIA-8> ilor * r31638 /trunk/src/dialogs.cpp: add another missing #include VC9 complained about
20081222 17:35:53< ilor> how do I create the current mo files?
20081222 17:36:05< loonycyborg> "scons translations" :)
20081222 17:36:15< Ivanovic> ilor: on which OS?
20081222 17:36:21< ilor> windows
20081222 17:36:32< Ivanovic> uhm, loud swearing might help...
20081222 17:37:10< Shadow_Master> gettext.
20081222 17:37:17< Shadow_Master> erm, get gettext tools
20081222 17:37:55< Ivanovic> or i could just upload a recent tarball with the files to files.wesnoth.org
20081222 17:38:22 * loonycyborg would make scons support msvc if he weren't microsoft hater :)
20081222 17:38:26< Ivanovic> ilor: want me to do so?
20081222 17:38:51< ilor> Ivanovic: that would probably be faster
20081222 17:38:59< loonycyborg> I also lack space on windows partition to install MSVC
20081222 17:39:23< CIA-8> fendrin * r31639 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/02_Hostile_mountains.cfg: LoW 2: Redone Olurf's team switching code.
20081222 17:41:18< Shadow_Master> I didn't notice I had a build running in the background. there's so much noise here that I didn't hear the fans ...  thought it'd take me longer anyway
20081222 17:41:43< Ivanovic> ilor: uploading the tarball, has ~15MB
20081222 17:41:46< Ivanovic> will take some mins
20081222 17:41:48< ilor> ok
20081222 17:43:57< Ivanovic> http://www.wesnoth.org/files/wesnoth-translations-trunk-r31638.tar.bz2
20081222 17:44:09< Ivanovic> the tarball just includes the translations/ folder with all the stuff that belongs into it
20081222 17:44:30< Ivanovic> no idea if this is called "translations" under windows nowaddays, or if it is still po/ or something like this
20081222 17:44:40< Ivanovic> loonycyborg should know this
20081222 17:44:50< Ivanovic> afk for some hours
20081222 17:44:53< loonycyborg> I've renamed it to translations.
20081222 17:44:58< Shadow_Master> it is itranslations IIRC
20081222 17:44:59< ilor> I had a look at an old 1.5 install and it's called translations
20081222 17:45:01< Shadow_Master> eek.
20081222 17:46:41< loonycyborg> Shadow_Master: ?
20081222 17:50:32< Shadow_Master> ESR__: please mention the _init(ial)?\.cfg thing in the changelog
20081222 17:50:46< Shadow_Master> otherwise it doesn't make sense to have a 'full' changelog at all
20081222 17:51:07< ESR__> Shadow_Master: I'll do so.
20081222 17:51:31-!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20081222 17:52:19< CIA-8> shadowmaster * r31640 /trunk/ (changelog players_changelog): Mention Naga Warrior's graphic updates in changelogs.
20081222 17:52:51< fendrin__> ESR__: http://www.wesnoth.org/forum/viewtopic.php?f=4&t=23353
20081222 17:53:52< ESR__> fendrin__: Heh.  That's an entertaining bug.
20081222 17:54:38< zookeeper> fendrin__, i figured out a little design obstacle i was having with my ai controller. i split the orders into two groups, "objectives" and "behaviour". for example, focusing on a certain enemy or protecting a certain side are objectives and aggressiveness and caution etc are behaviours. an ally can only have one objective and one behaviour active at any given time.
20081222 17:55:37< zookeeper> that way it's probably more clear to the player, since he doesn't need to wonder what multiple overlapping orders will result in, yet he can make some combinations
20081222 17:56:26-!- Espreon [n=Espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev
20081222 17:56:55< Espreon> Hello people.
20081222 17:57:17-!- Behl [n=tobizaz@0x5550f4db.adsl.cybercity.dk] has joined #wesnoth-dev
20081222 17:58:08< CIA-8> esr * r31641 /trunk/changelog: Note addition of _initial.cfg.
20081222 18:06:47< Shadow_Master> wesbot: seen thespaceinvader
20081222 18:06:48< wesbot> Shadow_Master: The person with the nick thespaceinvader last spoke 16h 1m ago.  15h 40m ago was here and on the channel #wesnoth with the message: "ChatZilla 0.9.84 [Firefox 3.0.5/2008120122]"
20081222 18:07:48-!- Elvish_Pillage2 [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20081222 18:13:49-!- JonJon [n=chatzill@217-211-27-121-no23.tbcn.telia.com] has joined #wesnoth-dev
20081222 18:13:53-!- JonJon [n=chatzill@217-211-27-121-no23.tbcn.telia.com] has quit [Client Quit]
20081222 18:14:00-!- Behl_ [n=tobizaz@0x5550f4db.adsl.cybercity.dk] has joined #wesnoth-dev
20081222 18:14:42-!- Behl [n=tobizaz@0x5550f4db.adsl.cybercity.dk] has quit [Read error: 104 (Connection reset by peer)]
20081222 18:22:57-!- mjs-de [n=mjs-de@p3EE2634B.dip.t-dialin.net] has joined #wesnoth-dev
20081222 18:23:28-!- Espreon_ [n=Espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev
20081222 18:24:31< fendrin__> Espreon, ESR__ : You talked about the map?
20081222 18:24:41< Espreon_> No.
20081222 18:24:57-!- Rrenys [n=rrenys@81-20-159-197.levira.ee] has joined #wesnoth-dev
20081222 18:27:57-!- Espreon [n=Espreon@c-76-117-239-33.hsd1.pa.comcast.net] has quit [Read error: 60 (Operation timed out)]
20081222 18:28:07-!- Espreon_ is now known as Espreon
20081222 18:36:31< ilor> mordante: seems like msvc doesn't like having two .cpp files with the same name in one project, and it barfs of image.cpp
20081222 18:37:50< ilor> or there's some trickery required that I don't know about
20081222 18:38:54< loonycyborg> That would be silly even by microsoft standards :)
20081222 18:38:56< mordante> that would be really odd, also others have been able to compile wesnoth on MSVC
20081222 18:39:15< ilor> well
20081222 18:39:21< ilor> it does behave in a retarded way
20081222 18:39:25< mordante> and also the files in widget/ have a lot of double names with gui/widgets
20081222 18:39:29< loonycyborg> mordante: When did you add image.cpp in widgets/ ?
20081222 18:39:47< loonycyborg> IIRC that's all recent additions.
20081222 18:39:49< ilor> I even went as far as changing the gui version to contain "#error 1"
20081222 18:40:06< ilor> rightclick on it in the msvc gui, compile -> the src/ one compiles
20081222 18:40:29< ilor> I really admire the MSVC team for breaking something as simple as single-file compilation
20081222 18:40:49< mordante> added it on the 22th of September
20081222 18:41:26< ilor> okay I worked around it by manually editing the project file
20081222 18:41:52< loonycyborg> ilor: If you're lucky, making scons using msvc will be one little change to one file ;)
20081222 18:41:57< ilor> other files that had duplicate names had some extra info in the .vcproj
20081222 18:42:44< loonycyborg> SConstruct line 107
20081222 18:42:55< ilor> it seems that due to the unique way in which new versions of microsoft products work, adding a file with a duplicate name no longer adds these extra entries as it apparently did in some older version
20081222 18:43:35< mordante> and that's a kind of progress?
20081222 18:44:17< ilor> I can't say I'm terribly surprised
20081222 18:44:40 * mordante wonders why ilor uses MSVC ;-)
20081222 18:44:55< ilor> it's the same company that spent half a year designing a new shutdown menu for vista that makes it more difficult to actually shut the system down
20081222 18:45:14< loonycyborg> Maybe because msvc compiler runs faster?
20081222 18:45:35< loonycyborg> Though I'm not sure about that.
20081222 18:45:37< Shadow_Master> welll, peopole call "Advanced Programming" a course that teaches you how to program in C# .NET using VC# .NET ? :)
20081222 18:45:48< ilor> the ide and debugger is a bit better than the alternatives
20081222 18:46:09< loonycyborg> I tried compiling wesnoth with icc but compile wasn't much faster.
20081222 18:47:28< loonycyborg> I once used MSVC ide but totally dropped using any sort of ide after moving to linux.
20081222 18:47:30-!- voris [n=voris@c-24-16-107-149.hsd1.wa.comcast.net] has joined #wesnoth-dev
20081222 18:47:32< Shadow_Master> ilor: gdb is good enough when you get used to oit :)
20081222 18:47:32-!- Behl_ [n=tobizaz@0x5550f4db.adsl.cybercity.dk] has quit ["So long suckers"]
20081222 18:47:48< loonycyborg> gdb is nice :)
20081222 18:47:58< Shadow_Master> loonycyborg: funny, I did the same with IDEs
20081222 18:48:22< loonycyborg> Though simutrans somehow manages to crash gdb :?
20081222 18:50:01 * loonycyborg though that crashing debuggers are horror stories from unix hater's handbook
20081222 18:50:03< ilor> could anyone take a look at sound.cpp:657?
20081222 18:50:07< loonycyborg> *thought
20081222 18:50:13< Shadow_Master> I read of changes to the parser recently?
20081222 18:50:17< Shadow_Master> (WML parser)
20081222 18:50:52< mordante> what's with it?
20081222 18:50:56< loonycyborg> ilor: "audio_lock lock()"?
20081222 18:51:08< ilor> yeah
20081222 18:51:14< ilor> it's not a variable definition
20081222 18:51:17< Shadow_Master> I am getting a fatal WML Parser rerror in my campaign...
20081222 18:51:27< ilor> it's a function declaration that does nothing
20081222 18:51:38< Shadow_Master> the cuplrit line seems to be a  _"sword" #enddef . Error being "nested quote ... blahblah"
20081222 18:52:02< loonycyborg> ilor: IIRC function declaration aren't allowed within other functions.
20081222 18:52:11< ilor> loonybot:  struct a { a() {cout << 1;}}; int main() { a c(); }
20081222 18:52:17< loonybot>  
20081222 18:52:21< ilor> loonybot:  struct a { a() {cout << 1;}}; int main() { a c; }
20081222 18:52:22< loonybot> 1
20081222 18:52:54< Shadow_Master> Ah, never mind. It was a "domino effect" bug introduced by some WML newbie
20081222 18:53:20< Shadow_Master> I'll go back to mad-admin mode and not let anyone touch my add-ons in the repo...
20081222 18:53:31< ilor> loonycyborg: see the problem? ;)
20081222 18:53:53< loonycyborg> Indeed. Nuke () then :)
20081222 18:55:18 * loonycyborg really needs to read some more C++ books. 5(or so) is clearly not enough :)
20081222 18:55:37 * mordante has less C++ books ;-)
20081222 18:56:10< mordante> but good catch 
20081222 18:56:12< ilor> it all fairness, the MSVC compiler caught that with a warning ;)
20081222 18:56:34 * loonycyborg doesn't need to have books in order to read them :)
20081222 18:56:41< mordante> what warning did it give ?
20081222 18:57:00< CIA-8> ilor * r31642 /trunk/src/sound.cpp: fix a C++ typo: audio_lock lock(); does not create a variable, it's a function declaration that does nothing. What was most likely intended was audio_lock lock;
20081222 18:57:03< ilor> Warning	12	warning C4930: '`anonymous-namespace'::audio_lock lock(void)': prototyped function not called (was a variable definition intended?)	\src\sound.cpp	657
20081222 18:57:36< mordante> it's even a nice warning :-)
20081222 18:57:52< mordante> maybe it also fixed some hard to catch bugs 
20081222 18:58:43< ilor> well, we didn't use the sdl audio locking and someone thought we did
20081222 18:58:49< ilor> so maybe ;)
20081222 18:59:25 * mordante has a deja vu :-P http://groups.google.com/group/comp.lang.c++/browse_thread/thread/07f7e110243a6a4d#
20081222 19:00:43< loonycyborg> :)
20081222 19:02:35< ilor> mordante: any ideas as to what could be causing the find_widget() call to fail?
20081222 19:04:24< mordante> ilor no not really, the problem was with the templated version?
20081222 19:04:29< ilor> yes
20081222 19:05:14< mordante> maybe change the find_widget to this->find_widget
20081222 19:05:20< ilor> I'm doing a debug build so I should be able to single-step through it and see what's wrong
20081222 19:06:09< mordante> that might be the easiest, I've to admit MSVC has a really nice debugger
20081222 19:12:03< ilor> another fun fact
20081222 19:12:23< ilor> to make a debug build compile and link, I either have to have the debug python libs (which are not distributed in binary form)
20081222 19:12:23< loonycyborg> Ah! You can declare functions within other functions, but not define them.
20081222 19:12:40< ilor> or #undef _DEBUG before #including <Python.h>
20081222 19:13:45< ilor> mordante: the very nice debugger and the different warnings that might uncover some bugs are one of the reasons why I'm trying MSVC
20081222 19:14:32 * loonycyborg wonders whether WinDbg is compatible with mingw's debug info
20081222 19:14:35< ilor> also, I'll have my revenge on the gcc 4.3-using people that insisted on me fixing warnings I didn't even get in 4.2 :P
20081222 19:14:49< mordante> :-)
20081222 19:15:14< mordante> we already fixed quite some MSVC 2008 problems ;-)
20081222 19:15:37< ilor> yes, apart from the image.cpp thing I think I only had one bizarre error
20081222 19:15:50< ilor> when compiling, that is
20081222 19:17:04< ilor> soundsource.cpp:224 somehow MSVC barfs unless I remove the "this->" part
20081222 19:18:18< loonycyborg> ilor: Did you try hand-coded loop but with this-> ?
20081222 19:18:18< ilor> it looks like a boost-msvc thing
20081222 19:18:35< loonycyborg> Maybe it's foreach issue?
20081222 19:18:39< ilor> looks like it
20081222 19:18:53< ilor> I created a reference to this->locations_ just above and it worked
20081222 19:19:08< ilor> it works just as well if I just remove the "this->" ;)
20081222 19:19:58< mordante> not sure why the this is needed?
20081222 19:20:01 * loonycyborg wonders why there are this-> used there.
20081222 19:20:54< ilor> it's not needed and it doesn't compile on "some" compilers if it is there, so...
20081222 19:21:13< CIA-8> ilor * r31643 /trunk/src/soundsource.cpp: fix a MSVC build error with boost's foreach using this->
20081222 19:21:34< loonycyborg> Perhaps also remove other instances of this->? :)
20081222 19:22:23< ilor> if it builds, why touch it? ;P
20081222 19:22:47< loonycyborg> To prevent it from building, hopefully :)
20081222 19:23:03< ilor> some of the msvc warnings are quite retarded though
20081222 19:23:10< ilor> http://msdn.microsoft.com/en-us/library/695x5bes.aspx
20081222 19:23:47< ilor> like it means anything if a struct is forward declared as a class or the other way around
20081222 19:28:45< ilor> mordante: you'll like the next one
20081222 19:29:01< ilor> Warning	17	warning C4146: unary minus operator applied to unsigned type, result still unsigned   \src\widgets\scrollbar.cpp	314
20081222 19:30:27< ilor> loonybot: {unsigned i = 1; cout << -i; }
20081222 19:30:32< loonybot> 4294967295
20081222 19:30:37-!- boucman [n=rosen@159.83.102-84.rev.gaoland.net] has joined #wesnoth-dev
20081222 19:32:14< mordante> loonybot: foo(int bar) { cout << bar; } int main(){unsigned a = 1; foo(-a); }
20081222 19:32:15< loonybot> error: ISO C++ forbids declaration of 'foo' with no type
20081222 19:32:24< mordante> loonybot: void foo(int bar) { cout << bar; } int main(){unsigned a = 1; foo(-a); }
20081222 19:32:25< loonybot> -1
20081222 19:33:10< mordante> now the question is who's right gcc or msvc?
20081222 19:33:23< ilor> mordante: the function converts it to an int
20081222 19:33:27< Shadow_Master> the sunos compiler.
20081222 19:33:30< Shadow_Master> :o)
20081222 19:33:39< loonycyborg> No icc :)
20081222 19:33:52< ilor> loonybot: void foo(double bar) { cout << bar; } int main(){unsigned a = 1; foo(-a); }
20081222 19:33:53< loonybot> 4.29497e+09
20081222 19:33:55< mordante> well it's like what happens in the code you pointed out as well ;-
20081222 19:34:06< ilor> it's still wrong ;)
20081222 19:34:14< mordante> ;-)
20081222 19:34:27-!- Psyche^ [n=Psyche@e177234099.adsl.alicedsl.de] has joined #wesnoth-dev
20081222 19:34:50< mordante> loonybot: {unsigned i = 1; cout << static_cast<int>(-i); }
20081222 19:34:51< loonybot> -1
20081222 19:35:04< mordante> maybe that makes it safer 
20081222 19:35:40< ilor> I don't mind an explicit static_cast here or there to make sure everybody sees what's going on (hey, it's unsigned, need a cast)
20081222 19:35:41< loonycyborg> loonybot: {unsigned i = 1; cout << -static_cast<int>(i); }
20081222 19:35:42< loonybot> -1
20081222 19:37:30< CIA-8> ilor * r31644 /trunk/src/ (gui/widgets/text.cpp widgets/scrollbar.cpp): add static_cast<int> in two instances of taking the negative of a unsigned number which may not work as expected in some cases (and msvc complains with a warning)
20081222 19:39:53< boucman> Shadow_Master: around ?
20081222 19:40:06< Shadow_Master> yes
20081222 19:40:25< Shadow_Master> dealing with administrative mistakes, kinda busy
20081222 19:40:46< boucman> Shadow_Master: could you note in your todo list to add the new imagePath macros in the wiki ?
20081222 19:40:55< boucman> these sort of things are useless if undocumented...
20081222 19:41:00< Shadow_Master> boucman: I know
20081222 19:41:05< boucman> :)
20081222 19:41:37< Shadow_Master> ESR__: did you document _init(ial?) in the iwki?
20081222 19:42:58< ESR__> Shadow_Master: Not yet, that's top of my to-do list.
20081222 19:44:13< Espreon> ESR__: I want to discuss the map with you.
20081222 19:45:15< ESR__> Espreon: What about it?
20081222 19:45:50< Espreon> Yeah, fendrin__ said that the map in LoW times... is... empty.
20081222 19:46:10< Espreon> And I was wondering if there were any towns/forts/cities/etc. that were settled at the time that were omitted.
20081222 19:46:29< Espreon> *omitted from the map.
20081222 19:47:17< ESR__> No.  It's empty for a reason.  Elves don't build cities where men can see them, and there hasn't been enough time since First Landing for the great cities of Wesnoth to be founded yrt.
20081222 19:48:06< loonycyborg> ESR__: Do elves have cities at all?
20081222 19:48:31< Espreon> Also, fendrin__wants the Saurian capitol to be named... (If named, I assume that he would want the label on the map)
20081222 19:48:40< Espreon> Let me brainstorm.
20081222 19:48:50< Espreon> ... after I eat.
20081222 19:49:00< ESR__> If you look at the main timeline, you'll see that Haldric's people have not even really settled the great central plain yet.  That won't happen until orc raids drive rhem inland from the coast.
20081222 19:51:06< Shadow_Master> zookeeper: were you serious when you suggested an image-scaling image function?
20081222 19:51:53-!- Patterner [n=Psyche@d137109.adsl.hansenet.de] has quit [Connection timed out]
20081222 19:51:53-!- Psyche^ is now known as Patterner
20081222 19:55:05< CIA-8> ilor * r31645 /trunk/src/ (10 files in 3 dirs): clear up some struct/class inconsistencies MSVC complains about
20081222 19:59:03< CIA-8> ilor * r31646 /trunk/src/wml_exception.hpp: define __func__ as __FUNCTION__ if the former isn't found and the latter is. This also allows removing the sunos special VALIDATE version
20081222 19:59:09< ESR__> Shadow_Master: _initial.cfg is now documented in WMLPreprocessorRef
20081222 20:00:19< cib0> is it bad style to use // coments in the C++?
20081222 20:00:48< loonycyborg> cib0: No :)
20081222 20:00:58< cib0> when commenting a block of code in particular
20081222 20:01:00< cib0> ok
20081222 20:01:23< cib0> i just see one-line comments with /* so often and i don't really get it
20081222 20:01:34< loonycyborg> When commenting blocks #ifdef 0 is often used.
20081222 20:02:07< CIA-8> mordante * r31647 /trunk/data/core/ (12 files in 4 dirs): 
20081222 20:02:07< CIA-8> Remove Kitty's large transparent portraits.
20081222 20:02:07< CIA-8> Kitty send the final version of all current portraits which will be added in
20081222 20:02:07< CIA-8> the next commit.
20081222 20:02:16< loonycyborg> So often that editors recognize #ifdef 0 ass comments and highlight accordingly.
20081222 20:02:25< loonycyborg> *as
20081222 20:02:42< cib0> yay o.o
20081222 20:03:36< Shadow_Master> cib0: /** are doxygen comments.
20081222 20:03:41< Shadow_Master> they are treated specially
20081222 20:03:51< Shadow_Master> mordante: is our blurring function a Gaussian Blur implementation?
20081222 20:05:21< mordante> Shadow_Master no idea, iirc ESR_ implemented it
20081222 20:05:44< Shadow_Master> ESR__: ^
20081222 20:05:49< cib0> ah, i see
20081222 20:06:20< Shadow_Master> cib0: they are used to generate things such as devdocs.wesnoth.org
20081222 20:07:02< Shadow_Master> mordante: how long for that next commit? :)
20081222 20:07:33< mordante> Shadow_Master might take a bit, need to add all and test
20081222 20:07:44< Shadow_Master> oka,y I'll awit for tomorrow then.
20081222 20:07:46< Shadow_Master> good afternoon
20081222 20:07:52< mordante> that's why I prefer to do that without the old cruft
20081222 20:07:55< mordante> bye Shadow_Master 
20081222 20:08:14-!- Shadow_Master_ [n=ignacior@unaffiliated/shadowmaster/x-275924] has joined #wesnoth-dev
20081222 20:08:39< Shadow_Master_> boucman: documented (http://www.wesnoth.org/wiki/ImagePathFunctionWML)
20081222 20:08:39-!- Shadow_Master_ [n=ignacior@unaffiliated/shadowmaster/x-275924] has quit [Client Quit]
20081222 20:08:45< boucman> thx
20081222 20:08:58< ESR__> Shadow_Master: Actually, I didn't.  I think I found a canned blurring function in the SDL layer.
20081222 20:12:27< loonycyborg> ESR__: What's the point of copying it since wesnoth is using SDL anyway?
20081222 20:13:17< loonycyborg> SDL doesn't expose blurring algorithms in convenient way? :)
20081222 20:14:30< ESR__> loonycyborg: Bad choice of words. I mean I copied a pre-existing call to an SDL blur function, IIRC.  I don't really understand what SDL is doing underneath but it seems to work.
20081222 20:19:03-!- mjs-de [n=mjs-de@p3EE2634B.dip.t-dialin.net] has quit ["On the road again"]
20081222 20:20:45-!- Espreon [n=Espreon@c-76-117-239-33.hsd1.pa.comcast.net] has quit ["ChatZilla 0.9.84 [Firefox 3.0.3/2008092417]"]
20081222 20:22:08-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev
20081222 20:31:19< Ivanovic> ESR__: http://www.wesnoth.org/forum/viewtopic.php?p=327301#p327301
20081222 20:31:35< ESR__> Ivanovic: Seen it.
20081222 20:31:45< Ivanovic> okay
20081222 20:32:22< ESR__> Manual is in good shape, BTW.  I looked over the health descriptions.
20081222 20:33:01< Ivanovic> oaky
20081222 20:33:15< Ivanovic> i already asked translators to have a look at it and explained how to create the screenshots
20081222 20:33:54< CIA-8> ilor * r31648 /trunk/src/ (terrain_filter.cpp terrain_filter.hpp): explain existing vc9 hack and disable warnings there
20081222 20:35:10< Espreon> Yay... more stuff to co.
20081222 20:35:10-!- fendrin__ [n=fabi@g227025147.adsl.alicedsl.de] has quit [Read error: 104 (Connection reset by peer)]
20081222 20:38:12-!- fendrin__ [n=fabi@g226141206.adsl.alicedsl.de] has joined #wesnoth-dev
20081222 20:39:47< ilor> mordante: around?
20081222 20:40:35< mordante> ilor in a few minutes, on the phone
20081222 20:41:07< ilor> okay
20081222 20:41:54< ilor> so, now I'm getting the new-dialogs crash in a different place, here's a backtrace: http://rafb.net/p/6c8xGL31.html
20081222 20:42:43< ilor> the error is an access violation and in the debugger gui I see the "widget" var on line 709 as being invalid
20081222 20:44:58< ilor> I get the same crash regardless of the dialog I try to use, here's a backtrace from a different one: http://rafb.net/p/IRJZwS89.html
20081222 20:45:41< Ivanovic> ilor: so do you see the problems with translations on windows, too?
20081222 20:46:00< ilor> Ivanovic: I had issues switching the language
20081222 20:46:14< Ivanovic> just enter the new language in the config file
20081222 20:46:31< ilor> right now I have the game in Polish and entering something else in the config file doesn't change anything
20081222 20:46:41< Ivanovic> locale="de_DE"
20081222 20:46:56< Ivanovic> and this *should* change things
20081222 20:47:10< ilor> should is the critical word here
20081222 20:47:29< Ivanovic> looks like it is only using the system locale, right?
20081222 20:47:39< ilor> yes
20081222 20:47:51-!- stikonas [n=and@ctv-79-132-180-249.vinita.lt] has joined #wesnoth-dev
20081222 20:47:52< ilor> in a old mingw compiled binary where I actually have the dialog
20081222 20:48:09< ilor> the language is shown as "German" but the strings are clearly Polish
20081222 20:48:21< ilor> (I've set de_DE in preferences)
20081222 20:48:29< Ivanovic> uhm, you could just change the files
20081222 20:48:45< Ivanovic> that is: remove the pl folder and rename the de folder to it
20081222 20:49:33< ilor> Ivanovic: good idea ;)
20081222 20:49:58-!- JonJon [n=chatzill@217-211-27-121-no23.tbcn.telia.com] has joined #wesnoth-dev
20081222 20:50:08-!- JonJon [n=chatzill@217-211-27-121-no23.tbcn.telia.com] has left #wesnoth-dev []
20081222 20:53:35< ilor> Ivanovic: I see the editor hotkey list sorted normally although some entries are not translated
20081222 20:56:10< ilor> there are some display artifacts though, with the hot key getting off the list to the right
20081222 21:00:30< Ivanovic> ilor: in the german translatio "quit editor" sould be translated
20081222 21:00:41< Ivanovic> that one is in the po file
20081222 21:00:55< Ivanovic> basically all the hotkeys should probably be translated
20081222 21:01:06 * loonycyborg doesn't like wesnoth's code littered with msvc workarounds like boost :((((
20081222 21:01:32 * Ivanovic does not like msvc...
20081222 21:02:00 * ilor hides
20081222 21:03:16-!- EdB [n=EdB@57.117.88-79.rev.gaoland.net] has quit ["Konversation terminated!"]
20081222 21:05:14< loonycyborg> But I think that depending on RVO is a hack.
20081222 21:08:27< ilor> yeah I'm not particularly thrilled about that either
20081222 21:09:15< ilor> also, I found a hotfix for that workaround I touched just a while ago, but it's a 300MB download and I haven't tested it yet
20081222 21:10:01< loonycyborg> Hm.. Perhaps I'm confusing this with something else and this has nothing to do with RVO
20081222 21:12:36< ilor> this was a VC9SP1 bug requiring a default constructor for when the class doesn't have to have one
20081222 21:13:07 * loonycyborg wonders what msvc will do if this default constructor is made private
20081222 21:13:50< ilor> probably barf
20081222 21:15:07< loonycyborg> Anyway, how does microsoft manage to stuff 300Mb in a hotfix?
20081222 21:15:41< Espreon> Satanic powers?
20081222 21:15:55< loonycyborg> I'm sure that installing MSVC will solve my problem with too much free space :)
20081222 21:19:01< mordante> ilor what error message do you get by that backtrace?
20081222 21:19:09< ilor> access violation
20081222 21:19:14< ilor> i.e. segfault, I guess
20081222 21:20:38< mordante> hmm odd, I'll have to look into it later, really need to fix some other problems first
20081222 21:21:25< ilor> Ivanovic: I think there's something I don't know about how gettext works :/
20081222 21:22:01< ilor> mordante: okay then, ping me when you have time to investigate. I guess I'll have to set up a mingw dev enviroment after all..
20081222 21:22:52< zookeeper> Shadow_Master, yeah, i was kinda serious. not useful very _often_, but maybe for some special effects, or add-on icons or whatever.
20081222 21:23:06< mordante> ilor I have an idea you might try
20081222 21:23:50< mordante> remove the destructor of the grid (and let the memory leak)
20081222 21:24:32< mordante> I already discovered that that's a bug, but didn't cause problems thusfar
20081222 21:24:40< mordante> guess the good old RVO ;-)
20081222 21:25:10< Ivanovic> ilor: hey, i don't know anything about the part of "getting it to work" either
20081222 21:25:24< Ivanovic> that is: i know how to make use of the strings and make them appear in the po files
20081222 21:25:35< Ivanovic> but not how it works the other way around
20081222 21:25:52< ilor> Ivanovic: do you by any chance know someone who does? ;)
20081222 21:25:59< Ivanovic> no
20081222 21:26:03-!- yann [n=dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has joined #wesnoth-dev
20081222 21:26:17< ilor> mordante: different crash
20081222 21:26:36< mordante> how different?
20081222 21:27:48< ilor> mordante: http://rafb.net/p/WKyykK54.html
20081222 21:29:18< ilor> it does a few iterations in the widget.cpp loop and ends up with a value of 0xcccc... as result, which crashes.
20081222 21:29:55< mordante> 8 c's in a row? 
20081222 21:30:03< ilor> yes
20081222 21:30:07< mordante> IIRC that's a magic value 
20081222 21:31:09< ilor> the iteration before the 0xccc it's a bad value that manages not to crash with an access violation
20081222 21:32:13< ilor> by bad value I mean x_ = 1237444, y_ = 22759806, vptr = 0xccc..
20081222 21:32:15< mordante> can you hunt for more destructors?
20081222 21:32:36< mordante> vptr looks indeed rather wrong
20081222 21:33:25< ilor> it looks like the bogus value that would crash if the grid dtor fired
20081222 21:35:05< mordante> well google indeed confirmed a magic value "0xcccccccc is the value MSVC's compiler uses to fill variables in debug mode."
20081222 21:35:26< ilor> Ivanovic: also here's how the hotkey window looks for me: http://img362.imageshack.us/img362/9256/wesnothyr0.png , not very nice :/
20081222 21:35:50< Ivanovic> yeah, i think i got something like this, too
20081222 21:36:00< Ivanovic> do not ask me why it happens, maybe mordante can fix it
20081222 21:36:02< Ivanovic> ^^
20081222 21:36:42< mordante> if the image loads... I can see what you mean ;-)
20081222 21:37:26< mordante> hmm yes that looks odd, it uses the old dialogs but I still might have a look
20081222 21:37:42< mordante> but please file a bug, I will forget otherwise
20081222 21:37:46< ilor> ok on it
20081222 21:38:01< mordante> and I really want to get the ready for 1.5.7
20081222 21:38:35< Ivanovic> mordante: yes, please get ready for 1.5.7
20081222 21:39:03< mordante> I'm busy, wiring in Kitty's new images :-)
20081222 21:39:12 * Ivanovic cheers
20081222 21:39:30< loonycyborg> I'm still getting errors with -D_GLIBCXX_DEBUG. Perhaps it's another of them is screwing msvc :)
20081222 21:39:42< mordante> well I had a nice surprise in my inbox today :-)
20081222 21:40:55< Espreon> fendrin__: Here?
20081222 21:43:05< ilor> mordante: seems 0xccccccc is used to initialize *local* variables
20081222 21:43:22-!- [Relic] [n=[Relic]@adsl-76-229-202-137.dsl.milwwi.sbcglobal.net] has joined #wesnoth-dev
20081222 21:44:03< mordante> ok, but at least it means I probably have some UB, which gives problems on MSVC 
20081222 21:44:19< [Relic]> Hello :)
20081222 21:44:22< mordante> hi [Relic] 
20081222 21:44:29< mordante> loonycyborg what errors?
20081222 21:44:56< loonycyborg> Those caused by comparing iterators from different sequences.
20081222 21:44:57< [Relic]> :)
20081222 21:45:42< loonycyborg> E.g. dummy unit type one.
20081222 21:45:59< mordante> ah yes that one
20081222 21:46:08< loonycyborg> Also, I'm getting some other error if I run unit test.
20081222 21:47:09< loonycyborg> "attempt to access an element in an empty container."
20081222 21:48:31< loonycyborg> http://rafb.net/p/TFxiI424.html
20081222 21:50:15< mordante> seems it would be a good idea to test wesnoth a bit more in debug mode
20081222 21:53:06< ilor> mordante: I'm going to try adding a check for invalid values in widget::set_parent
20081222 21:53:22< mordante> ok
20081222 21:54:16< mordante> well I still want to twidget from boost::non_copyable but that needs some other (minor) changes
20081222 21:54:28< mordante> to inherit
20081222 21:57:04< ilor> mordante: didn't work
20081222 21:57:19< mordante> :-(
20081222 21:58:23< ilor> one more idea...
20081222 22:02:47-!- lynxlynxlynx [n=navaden@sourcemage/warlock/lynxlynxlynx] has joined #wesnoth-dev
20081222 22:05:40< lynxlynxlynx> cmake/FindFriBiDi.cmake has a syntax error inducing typo; sed 's,_FRIBIDI_INCL,FRIBIDI_INCL,' # to fix it
20081222 22:07:45< ilor> mordante: got a bit more info
20081222 22:08:12< ilor> in set_parent I have a call that sets a parent whose grandpa is 0xccc
20081222 22:08:59< ilor> it's called from tlistbox::add_row
20081222 22:09:09< Ivanovic> lynxlynxlynx: uhm, where have you found this?
20081222 22:09:15< Ivanovic> lynxlynxlynx: i don't see this in current trunk
20081222 22:10:09< lynxlynxlynx> oh crap, viewcvs fooled me
20081222 22:10:10< Ivanovic> that is: the file was not changed for ages and this part is not in there
20081222 22:10:25< lynxlynxlynx> i'm not looking at the latest revision
20081222 22:10:31< Espreon> Ivanovic: Is it too late to make changes to the Desert Hunter line's images?
20081222 22:10:39< Ivanovic> to images?
20081222 22:10:46< Ivanovic> no, images can be altered all the time
20081222 22:10:51< Espreon> All right.
20081222 22:11:01< Espreon> I will modify them momentarily.
20081222 22:11:30< lynxlynxlynx> Ivanovic: yeah, it was fixed already
20081222 22:12:52< Espreon> ESR__: Still here?
20081222 22:13:11< ESR__> Yes, what's up?
20081222 22:13:23< mordante> ilor that file is being rewritten at the moment, so I rather wait until the next revision is in
20081222 22:13:37< mordante> it's a complete rewrite using the new generator class
20081222 22:13:49< ilor> mordante: what file?
20081222 22:13:59< mordante> generator*
20081222 22:14:21< mordante> which is not copyable
20081222 22:15:39< Espreon> The desert in the Far North map looks cheap to me... Is it OK if I change it to look like the desert in this map: http://i125.photobucket.com/albums/p69/CupidGodofLove/EOS2.png?t=1229980517
20081222 22:18:56< Espreon> ESR__: Would that style be all right?
20081222 22:31:41< mordante> it seems I miss a few portraits, so I'll ask Kitty, I'll commit when I have all of them (I want to test them all at the same time)
20081222 22:33:02-!- loonybot [n=loonybot@79.139.247.111] has quit [Read error: 104 (Connection reset by peer)]
20081222 22:33:13< ilor> mordante: it looks like a copy issue that RVO can possibly hide, in window_builder.cpp's build()
20081222 22:33:30< ilor> line 207 and so on
20081222 22:33:42< mordante> yes, which is the reason twidget needs to be copyable
20081222 22:34:11< mordante> I already did a short test to disable copying, I spotted the problem while working on the generator
20081222 22:34:13< CIA-8> espreon * r31649 /branches/resources/cartography-tools/brushes/wesnoth_tree.gbr: * Removed unecessary pixels from the tree brush.
20081222 22:34:37< ilor> either that or have the function new and return via pointer
20081222 22:34:41< mordante> but since it needed some more rewriting I postponed it (nobody had a problem)
20081222 22:34:53< mordante> pointer
20081222 22:35:02< ilor> seems like msvc is less RVO-happy in general
20081222 22:35:06< mordante> return was intended
20081222 22:35:23< mordante> well it seems sun as well
20081222 22:36:44< mordante> the problem when you mainly test with only one compiler ;-)
20081222 22:37:06< mordante> I think at the moment most developers use some gcc version
20081222 22:37:35< ilor> I hope you get round to fixing it soon ;)
20081222 22:37:58< ilor> why isn't the window object new()ed and passed by pointer anyway?
20081222 22:38:04< mordante> yes it's high on my list, but I try to avoid mixing 4 changes at the same time
20081222 22:38:08-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev
20081222 22:38:49-!- lynxlynxlynx [n=navaden@sourcemage/warlock/lynxlynxlynx] has left #wesnoth-dev ["Sayonara, not to be confused with cyanide, which is, of course, goodbye in any language."]
20081222 22:39:15< ilor> btw hey, you were supposed to publish your todo ;)
20081222 22:39:28< mordante> since the current method works for me (tm)
20081222 22:40:00< mordante> yes I know I just want to have things done before the feature freeze 
20081222 22:40:09< ilor> I see, that's a very good reason as long as it doesn't crash for someone else ;)
20081222 22:41:03< mordante> well I overlooked the destructor problem and the RVO in gcc hides the bug
20081222 22:41:56< mordante> it was intended to be fixed, even if you didn't discover the problem ;-)
20081222 22:42:33< ilor> as a workaround I'll try heap-allocating the window (leaking it) and returning by value anywa, maybe it'll work
20081222 22:43:11< mordante> only odd that the other MSVC users didn't run into those problems :-/
20081222 22:44:23< ilor> I don't mind catching bugs as long as we can actually fix them (in contrast to the boost filesystem thing from a while ago)
20081222 22:44:31< CIA-8> espreon * r31650 /branches/resources/cartography-tools/brushes/wesnoth_tree.gbr: * Corrected the tree brush's spacing.
20081222 22:45:25 * mordante fully agrees, he still has nightmares of bugs in third party tools
20081222 22:48:56< mordante> night
20081222 22:49:18< Espreon> ESR__: I also notice that the Far North map uses black towns/forts/etc. while the main map uses brownish ones. Would you like me to make the Far North map use brownish ones?
20081222 22:49:22< Espreon> Good night mordante.
20081222 22:49:26-!- mordante [n=chatzill@roadie.xs4all.nl] has quit ["ChatZilla 0.9.84 [Firefox 3.0.4/2008102920]"]
20081222 22:50:20< Ivanovic> ilor: do you know which "other MSVC users" mordante meant?
20081222 22:50:22< Ivanovic> are there any?
20081222 22:50:34< Ivanovic> at least for wesnoth
20081222 22:50:39< ilor> ;)
20081222 22:50:39< ESR__> Espreon: It's exactly the same set pgf ivcon widgets.  It don't know why they look different -- perhaps the effect of background contrast?
20081222 22:50:54< ilor> Ivanovic: there was this one guy in the forums
20081222 22:51:00< ESR__> s/pgf ivcon/of icon/
20081222 22:52:26< ilor> mordante, okay, my hack-fix made it hang instead of crash, so I guess I'll better wait till you get to it :) 
20081222 22:56:22< Espreon> ESR__: Maybe not...
20081222 22:56:43< cib0> i think i broke something, im getting errors in files that i havent modified o.o
20081222 22:56:44< Espreon> I also notice this with Parthyn in the DiD layer.
20081222 22:58:52< Espreon> ESR__: I am going to change the Parthyn piece, since all the other city/fort/etc. pieces are brownish, and the Parthyn piece is black.
20081222 22:59:21< ESR__> Thst's OK.
20081222 23:00:00< ilor> what's supposed to happen if I build wesnoth without fribidi?
20081222 23:01:50< Soliton> you don't get rtl support.
20081222 23:02:10< Soliton> i.e. nothing.
20081222 23:02:51< ilor> thanks
20081222 23:04:03< CIA-8> ilor * r31651 /trunk/src/color_range.cpp: silence a warning
20081222 23:04:40< loonycyborg> You *will* get rtl support in new widgets even without fribidi :)
20081222 23:04:57< CIA-8> ilor * r31652 /trunk/src/network.cpp: silence a warning
20081222 23:05:09< loonycyborg> Since pango includes fribidi of its own
20081222 23:05:11< ilor> when I get new widgets not crashing ;)
20081222 23:06:30< ilor> BTW, I think we should have some form of fatal_error(reason) function to use instead of assert(false) in impossible checks
20081222 23:07:35< Soliton> there is ASSERT_LOG and ERROR_LOG.
20081222 23:07:44< CIA-8> espreon * r31653 /branches/resources/cartography-tools/great-continent.xcf: 
20081222 23:07:44< CIA-8> * Tweaked the forests on the Far North map.
20081222 23:07:44< CIA-8> * Removed some debris from the Far North map.
20081222 23:08:54< Soliton> ilor: src/assert.hpp has a bunch of new assert macros.
20081222 23:09:10< CIA-8> espreon * r31654 /branches/resources/cartography-tools/great-continent.xcf: * Made the Parthyn piece brownish instead of black.
20081222 23:11:16< ilor> I recently found a good article on assert macros - http://powerof2games.com/node/10
20081222 23:13:25< Espreon> ESR__: I notice even more black pieces... so I am going to correct those.
20081222 23:14:09< Espreon> Also, the trees in the Liberty layer look unnatural.
20081222 23:19:19-!- [Relic] [n=[Relic]@adsl-76-229-202-137.dsl.milwwi.sbcglobal.net] has quit ["Leaving"]
20081222 23:20:12< CIA-8> ilor * r31655 /trunk/src/ (ai_python.hpp filechooser.cpp gui/dialogs/field.hpp):
20081222 23:21:09< Soliton> ilor: sounds like we should improve those assert macros then. :-)
20081222 23:22:56< CIA-8> ilor * r31656 /trunk/src/ (ai_python.hpp filechooser.cpp gui/dialogs/field.hpp): revert misclick commit, sorry for that.
20081222 23:23:12< ilor> Soliton: I'll play with it but not today, clearly I should get some rest now
20081222 23:23:33< Soliton> "misclick" :-P
20081222 23:23:55 * Soliton suggests git-svn.
20081222 23:24:10< CIA-8> espreon * r31657 /branches/resources/cartography-tools/great-continent.xcf: 
20081222 23:24:10< CIA-8> * Fixed the trees in the Liberty layer.
20081222 23:24:10< CIA-8> * Converted all the black pieces in the Liberty layer to brownish pieces.
20081222 23:24:54< ilor> at least I remembered how to do a undo-merge without looking at the docs ;)
20081222 23:25:30< loonycyborg> Git revert will even write commit message for you :)
20081222 23:25:52-!- yann [n=dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has quit [Remote closed the connection]
20081222 23:26:41< ilor> hmm, nice, will have to try it someday
20081222 23:26:59< fendrin__> ESR__: Have you had a look into the map of LoW? I am sure that most wesnothian cities weren't foundet at that time but the elves/dwarves/whatever would at least have named some of the rivers.
20081222 23:28:07< CIA-8> ilor * r31658 /trunk/projectfiles/VC9/ (editor.vcproj wesnoth.sln wesnoth.vcproj): 
20081222 23:28:07< CIA-8> MSVC projectfile update:
20081222 23:28:07< CIA-8> * removed the editor project
20081222 23:28:07< CIA-8> * the server project is unloaded and not built at the moment
20081222 23:28:07< CIA-8> * added missing files to the wesnoth project and modified some options (works for me [tm])
20081222 23:29:02< ESR__> fendrin: I left tyhe map empty because at this point in history most of the Great Continent was still unknown to humans.  No one had yet seen the River Listra, for example, or named the Estmarks.
20081222 23:29:41< ESR__> So yes, it should look like a big exchoing blank thing :-)
20081222 23:30:17< ESR__> I stripped all those layers off the main map deliberately.
20081222 23:31:41< CIA-8> espreon * r31659 /branches/resources/cartography-tools/great-continent.xcf: * Converted all the black pieces on the Eastern Invasion layer to brownish pieces.
20081222 23:31:50< fendrin__> ESR__: I see your point but can't the map have at least some more locations named in ancient languages?
20081222 23:33:07< CIA-8> zookeeper * r31660 /trunk/data/campaigns/The_South_Guard/scenarios/07a_Into_The_Depths.cfg: Dropped M'Brin's gold a little bit on the easiest difficulties.
20081222 23:33:45< ESR__> If you want to break the names layers into "ancient" and "modern" and show the abcient ones on that map, feel free.  I don't think it's necessary.
20081222 23:34:22< Espreon> fendrin__: If you want to do that... tell me what to name... and I will do it.
20081222 23:34:59< ESR__> About the only two names I'm sure should go in the ancient layer are the Great River and the Arkan-thoria.
20081222 23:36:55-!- Elvish_Pillage2 [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20081222 23:38:37< Espreon> fendrin__: What would you like them to be named as?
20081222 23:40:44< fendrin__> Espreon: Saurgrath the saurion capital could go in that layer. Together with the 2 locations named by ESR__ and maybe one or two cities of the dwarves. Let me see if I can find more names in the campaign.
20081222 23:41:02< cib0> http://wesnoth.pastebin.com/m4d3b15d1 can anyone tell me what im doing wrong here?
20081222 23:41:57< Espreon> fendrin__: Ummm... is Saurgrath featured in any scenarios? And if so, which one(s)?
20081222 23:41:59< ESR__> About that Saurian capitol....I'm thinking maybe it should move.
20081222 23:42:42< fendrin__> Espreon: It's featured in scenario 5 and in 19.
20081222 23:43:13< fendrin__> ESR__: Agreed, I had that thought before.
20081222 23:43:31< ESR__> Everwhere else in-game Saurians are swamp creatures.  Saurgrath is in mountain country in the far norrth; this seems wrong to me.  I think it might make more senxe to relocate it into the Swamp of Dread.  That would make the journey tracks less weird, too.
20081222 23:43:32< Soliton> cib0: put custom_endlevel_music(endlevel_music_list) last in the initialization list.
20081222 23:43:42-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has quit []
20081222 23:44:14< ESR__> All this would take is a trackplacer edit, I think.
20081222 23:47:29< Espreon> ESR__: I would be happy to relocate the Saurian capitol.
20081222 23:48:48< Espreon> ESR__: So you want it in the Far North map?
20081222 23:48:56< ESR__> Before tyou do that, check the story text in scenarios 4, 5, 6, 18, 19, and 20 to make sure relocating it won't falsify any travel descriptions.
20081222 23:49:01< ESR__> Oh hell no.
20081222 23:49:13< Espreon> You did mention Far North.
20081222 23:49:31< Espreon> I was not sure if you meant Far North or far north.
20081222 23:49:46< ESR__> The Swamp of Dread is on the main map, just north of the Great River.  I meant it's presently in the far north of the main map.
20081222 23:50:13< Espreon> So, I should label it as Saurgrath in the LoW layer?
20081222 23:50:50< ESR__> Yes.  
20081222 23:52:25< fendrin__> The saurgrath map contains mountains to the north and hills to the west, That would fit to the North East corner of that swamp.
20081222 23:53:26-!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20081222 23:57:50< fendrin__> ESR__ , Espreon: I am not sure about the language. Does "To evade the saurians blocking the eastern approaches to Wesmere ..." fit to the new location? The swamps are in the east.
20081222 23:59:20< ESR__> I don't think it's a problem -- it's the approaches *to Wesmere*, and if the Saurians were coming from the Swamp of Dead that's exactly where they'd be concentrated.
--- Log closed Tue Dec 23 00:00:18 2008