Log for #openttdcoop.devzone on 11th September 2010:
Times are UTC Toggle Colours
02:46:06  *** Webster has joined #openttdcoop.devzone
03:06:07  *** Webster has joined #openttdcoop.devzone
06:36:38  <andythenorth> morning
06:52:12  <Brot6> NewGRF Meta Language - Revision 750:6090203eab93: Fix: Documentation mentioned old name of nmlc i... (planetmaker) @
06:52:27  <planetmaker> moin
07:36:59  <Brot6> NewGRF Meta Language - Revision 751:a1a8fd322a4e: Doc: Add a verbose railtypes example (planetmaker) @
07:37:59  <Brot6> NewGRF Meta Language - Feature #1465 (Feedback): Documentation for railtypes (yexo) @
07:38:00  <Brot6> NewGRF Meta Language - Feature #1465 (Feedback): Documentation for railtypes (planetmaker) @
07:47:40  <Brot6> NewGRF Meta Language - Feature #1465 (Closed): Documentation for railtypes (yexo) @
07:47:40  <Brot6> NewGRF Meta Language - Feature #1465 (Closed): Documentation for railtypes (yexo) @
07:54:55  *** Webster has joined #openttdcoop.devzone
08:12:47  *** Alberth has joined #openttdcoop.devzone
08:13:12  <Alberth> bad bad andythenorth, recommending hacking the code for partial solutions
08:13:27  <andythenorth> they might learn something :P
08:13:43  <andythenorth> one more checkout...
08:14:07  <andythenorth> I now hack things locally to get certain changes
08:15:11  * andythenorth thinks the arable farm could produce building materials....
08:15:12  <Alberth> a friendly yellow button with "WIN NOW" ?
08:15:20  <Terkhen> why?
08:15:43  <Alberth> all those bones... :)
08:16:38  <andythenorth> thatch (straw) :)
08:16:45  <andythenorth> not a serious suggestion :P
08:19:36  <Terkhen> :)
08:31:09  <Brot6> GRFCodec - Bug #1464: nforenum and grfcodec disagree on escapes format (Rubidium) @
08:33:29  *** ODM has joined #openttdcoop.devzone
09:42:43  *** Webster has joined #openttdcoop.devzone
09:56:14  *** thgergo has joined #openttdcoop.devzone
10:37:45  *** KenjiE20 has joined #openttdcoop.devzone
11:36:05  <Brot6> Example NewGRF Project - Bug #1449: Compiling under MinGW/MSYS (Terkhen) @
11:45:25  <Brot6> Example NewGRF Project - Bug #1449: Compiling under MinGW/MSYS (Ammler) @
12:01:21  *** thgergo has quit IRC
12:02:24  *** JackSwagger has joined #openttdcoop.devzone
12:02:34  <JackSwagger> anybody here?
12:02:35  *** JackSwagger has left #openttdcoop.devzone
12:20:37  *** Seberoth has joined #openttdcoop.devzone
12:42:12  *** thgergo has joined #openttdcoop.devzone
13:13:43  <Ammler> now, that would suck
13:14:03  <Ammler> I had all the install troubles because of broken disks :'-(
14:29:10  <andythenorth> Alberth: I tried to apply the timestamp patch but got rejects.  I manually changed the lines in the diff, but I'm not seeing timestamps
14:29:31  <andythenorth> could you stick it on and I'll apply it with curl?
14:29:51  <Alberth> you need to enable the show_date_in_logs setting too
14:30:31  <andythenorth> ah ok
14:30:32  <Alberth>   I have another one
14:30:44  <andythenorth> I'll apply that one
14:30:56  <Alberth> it inserts lines with game date in the debug output
14:31:10  <andythenorth> ok that applied
14:31:20  <andythenorth> do I need to enable anything to use it?
14:31:33  <Alberth> yes, the setting mentioned above
14:35:30  * andythenorth tries
14:42:54  <andythenorth> Alberth: it's giving me a timestamp....current timestamp in real world :)
14:44:24  <Alberth> you get additional lines "[2010-09-11 16:44:29] dbg: [grf] Game date: 2051/05/09.27"
14:44:56  <andythenorth> :( maybe I failed to apply the patch correctly
14:46:06  <andythenorth> for hg it should be -p1?
14:50:19  <Brot6> FIRS Industry Replacement Set - Feature #1466 (New): Consider if Machine Shop should accept MNSP (andythenorth) @
14:50:31  <Alberth> it;s a svn patch
14:50:51  <Alberth> lines "+++ src/debug.cpp" do not start with "a/" and "b/"
14:51:00  <andythenorth> Alberth: I've got it working :)
14:51:16  <andythenorth> dunno how I'd chart the data, but it's a start
14:51:25  * andythenorth wonders how to dump it to a file...
14:51:48  <Ammler> openttd | tee file
14:52:12  <andythenorth> cool
14:53:18  <andythenorth> apparently not :P
14:53:51  <andythenorth> Ammler: ./openttd -d industry=3 | tee ~/Documents/OpenTTD/industry_openings.txt
14:53:58  <andythenorth> creates the file, but doesn't write to it
14:54:14  <Ammler> hmm, only stdout?
14:54:36  <Ammler> do you need the output on the console too?
14:54:56  <Ammler> openttd 1>file 2>&1 would do it else
14:55:22  <Ammler> then you can watch the tile with "tail -f file"
14:55:28  <andythenorth> output on console not necessary
14:56:41  <Ammler> ./openttd -d industry=3 1>~/Documents/OpenTTD/industry_openings.txt 2>&1
14:56:45  <andythenorth> no idea how I'm going to chart this, but at least there's some data coming out :P
14:56:53  <Ammler> tail -f ~/Documents/OpenTTD/industry_openings.txt
14:57:33  <andythenorth> Alberth: incidentally, the industry patch appears to do what it's supposed to
14:57:33  <Ammler> if you want the debug data only you could pipe stderr only
14:57:54  <andythenorth> I only have limited data, but it does seem to build industry reasonably soon after first available
14:57:56  <Ammler> ./openttd -d industry=3 2>~/Documents/OpenTTD/industry_openings.txt
15:38:40  *** thgergo has quit IRC
16:20:42  <Brot6> firs: update from r1358 to r1361 done -
16:21:02  <Brot6> newgrf_makefile: update from r183 to r184 done -
16:21:52  <Brot6> nml: update from r742 to r751 done -
16:21:56  <Brot6> OpenGFX+ Airports - Bug #1467 (New): DevZone compile failed (compiler) @
16:22:07  <Brot6> Following repos didn't need a nightlies update: 2cctrainset (r613), 32bpp-extra (r39), basecosts (r20), belarusiantowns (r7), comic-houses (r71), fish (r390), frenchtowns (r4), grfcodec (r253), heqs (r372), metrotrackset (r56), nforenum (r502), (ERROR r486), nutracks (r115), ogfx-test (r529), ogfx-trees (r15), ogfxplus (r42), opengfx (r535), openmsx (r97), opensfx (r97), snowlinemod (r42), swedishrails (r156), swisstowns
16:22:07  <Brot6> (r14), transrapidtrackset (r15), ttdviewer (r25), ttrs (r18), worldairlinersset (r663)
16:24:35  <Brot6> Following repos rebuilds successful without any difference to earlier nightlies builds: belarusiantowns (3 errors) (Diffsize: 21), frenchtowns (4 errors) (Diffsize: 9), ogfxplus (Diffsize: 6), swedishrails (Diffsize: 6), swisstowns (4 errors) (Diffsize: 9)
16:52:48  <Ammler> Yexo: default build type is nfo
16:52:53  <Ammler> should we change that?
16:55:38  <Brot6> OpenGFX+ Airports - Bug #1467 (Closed): DevZone compile failed (compiler) @
16:55:38  <Brot6> OpenGFX+ Airports - Revision 59:b69876cf0101: Fix #1467: we need to set build type for non nfo pr... (Ammler) @
16:55:38  <Brot6> OpenGFX+ Airports - Bug #1467 (Closed): DevZone compile failed (Ammler) @
16:56:23  <Brot6> airportsplus: update from r54 to r59 done -
16:56:32  *** thgergo has joined #openttdcoop.devzone
18:00:48  <Brot6> 32bpp-ez-patches: update from r20782 to r20786 done -
18:01:38  <Alberth> wow, hg import is quite useful
18:02:37  <Brot6> NewGRF Meta Language - Revision 752:3f0115df5c8b: Doc: Docstring improvements. (Alberth) @
18:02:37  <Brot6> NewGRF Meta Language - Revision 753:51e4fe4f61ec: Codechange: More preprocessing (Alberth) @
18:09:20  <Brot6> serverpatches: update from r20782 to r20786 done (2 errors) -
18:30:25  <Brot6> NewGRF Meta Language - Feature Request #1419 (Reopened): Documentation for all vehicle properties... (Hirundo) @
20:02:41  <Brot6> NewGRF Meta Language - Revision 754:b980b4c47871: Codechange: Move code from TownNames to ActionF. (Alberth) @
20:47:31  <Brot6> NewGRF Meta Language - Revision 755:3460178bdda2: Fix (r754): debug_print() should not be moved. (Alberth) @
20:48:04  *** Alberth has left #openttdcoop.devzone
21:02:48  <Ammler> when we move the DevZone to the new server, shall change something? Except that I will put the gameservers to its own vm
21:43:56  *** GT has joined #openttdcoop.devzone
22:06:46  *** ODM has quit IRC
22:41:29  <Brot6> FIRS Industry Replacement Set - Feature #1466: Consider if Machine Shop should accept MNSP (andythenorth) @
22:49:56  <GT> Ammler: arent the 32bpp-ez binaries from the compile farm in the log.csv?
22:50:17  <Ammler> that is currently broken
22:50:32  <GT> I was kind of curious how often they got downloaded,
22:50:36  <GT> Oh, I see
22:50:41  <Ammler> :-(
22:51:03  <GT> No problem, seem to work very nice.
22:51:06  <Ammler> hmm, actually, we could enable it again
22:51:22  <GT> That is, interaction with the compile farm and producing bins
22:52:49  <GT> Btw how can I create a testing compile with the latest version of the patch without tagging it?
22:53:25  <GT> Since tagging seems to create a release rev.
22:54:31  <GT> The testing compile is nice though, shows when I need to update the patch against trunk
22:55:24  <GT> = very useful
22:57:34  <Ammler> that is the idea :-)
22:58:22  <Ammler> hmm, you like to trigger a test compile without triggering the compile farm?
23:02:41  <Ammler> GT, you like a svn patch on testing?
23:03:01  <Ammler> and a link to the diff without version?
23:03:54  <Ammler> log.csv should work again
23:05:00  <GT> well, personally I dont need the svn diff, as I have my stuff updated locally, but I do think an svn diff is useful for some, that dont have Mercurial.
23:05:20  <Ammler> yes, the release does make one
23:05:29  <Ammler> but on testing, it isn't enabled
23:05:58  <Ammler> and that is btw. a real svn patch again svn checkout :-)
23:06:00  <GT> Well, on testing, it's not really important.
23:06:15  <Ammler> not just sed
23:06:38  <Ammler> so fully working with tortoisesvn
23:07:05  <GT> But, to go back a cpl of lines, yes, when I update the patch, sometimes I'd like to test it against latest trunk before releasing.
23:07:33  <Ammler> and you can't do that locally?
23:08:06  <GT> I of course compile it locally, but as we've seen last week, that does not weed out all warnings, like 64<->32 bit stuff
23:08:56  <Ammler> well, there server here does compile it only again suse 11.3 64
23:09:09  <Ammler> the multiple compiles happen on the cf
23:09:14  <GT> 64 bit?
23:09:17  <Ammler> which need triggering
23:09:20  <Ammler> yes
23:09:54  <Ammler> but I see need for tiggering, you aren't the first who likes it
23:10:14  <Ammler> for newgrfs, the solution is to enable building on push
23:10:32  <GT> It's not essential, but I like to have my logs clean before tagging a release
23:10:38  <Ammler> but the openttd patch building needs around 9 mins
23:11:11  <GT> But I may be over-sensitive
23:11:12  <Ammler> so triggering a build on every push might not be optimal?
23:11:22  <Ammler> hmm
23:11:31  <Ammler> on the other side, there aren't many pushes
23:11:37  <GT> True
23:11:41  <Ammler> so just enable that :-)
23:11:47  <GT> How
23:11:59  <Ammler> touch .devzone/build/push/enable
23:12:29  <Ammler> and add that to your repo with your next push
23:12:46  <GT> Ok, did not find that on the wiki, but I may have missed it
23:13:37  <Ammler> yes, it is on the wiki but wrongly
23:13:39  <Rubidium> you could make a 64 bits cross compiler
23:13:41  <Ammler> "pushes"
23:13:50  <Rubidium> or maybe your distros comes with one?
23:14:09  <Ammler> Rubidium: doesn't that need special cpu?
23:14:21  <Ammler> something VT or so?
23:14:50  <Rubidium> why?
23:15:01  <Rubidium> oh... because you're thinking of doing it like our compile farm
23:15:10  <Rubidium> well, then yes-ish... but no
23:15:15  <Ammler> :-)
23:15:26  <Rubidium> we didn't need a powerpc CPU to build the Mac OS X binaries either, right?
23:15:30  <Ammler> yeah, I was thinking about setup a 64bit vm
23:15:36  <Ammler> on a 32bit maschine
23:15:40  <Rubidium> that's a cross compiler as well
23:15:59  <Rubidium> oh, that won't work... except in qemu (and alike)
23:16:13  <Ammler> that needs special setting on bios and cpu
23:16:30  <Ammler> at least what my vbox tells me :-)
23:16:36  <Rubidium> Ammler: 64bits VM on 32 bits hardware?
23:16:40  <Ammler> yep
23:16:57  <Rubidium> that's not going to work, except emulated through qmeu (and alike)
23:17:06  <Ammler> hmm
23:17:26  <Rubidium> 64 bits VM on 32 bits OS with 64 bits CPU, that *can* work with the appropriate settings and hardware
23:18:04  <Ammler> VT-x/AMD
23:18:15  <Rubidium> GT: what distro did you use?
23:18:33  <Ammler> GT=GenToo
23:18:50  <Ammler> or how is that distro called again?
23:19:27  <GT> Gentoo, switching to 64 bit compile is a matter of setting an arch flag in make.conf, but I tend to forget that, and have shit upon the next emerge
23:20:17  <GT> So, test compile on openttdcoop, before releasing to openttd CF would be my ideal scenario
23:20:33  <Ammler> GT, now since our devzone compiles you a complette diff, you can start to split, btw. :-)
23:21:07  <GT> BTW, thanks to both of you for setting up the CF, seems to work OK
23:21:43  <GT> complette must be kind of German
23:21:53  <GT> :)
23:22:22  <Ammler> -t :-)
23:22:25  <Ammler> omg
23:22:39  <Ammler> or total
23:22:45  <Ammler> how shall I call it?
23:23:17  <Ammler> :-P
23:23:19  <GT> full, complete, total,.. all fine by me
23:23:23  <Ammler> nobody said something yet
23:23:26  <Rubidium> coalesced?
23:23:36  <GT> overall
23:23:58  <Ammler> full sounds good too
23:24:04  <Rubidium> Ammler: yeah, I had to check the spelling of coalesced
23:24:24  <GT> pan
23:24:39  <Ammler> :-)
23:25:33  <GT> joined
23:25:48  <GT> Ok, Im overdoing it, I guess
23:25:49  <Ammler> and maybe I should remove the revision, so you can publish a direct link
23:26:15  <Ammler>
23:26:36  <Ammler>
23:28:57  <GT> Well, most users are interested only in the bins I guess, like the 32bpp-ez/LATEST, the most frequent user of the testing would be me, until hundreds of co-devs for the patch appear.
23:29:30  <GT> Which, unexpectedly, did not happen for the last two years.
23:29:34  <Ammler> you aren't
23:29:41  <Ammler> you have the patch locally already
23:29:44  <GT> Not?
23:30:18  <GT> Yeah, but I check the compile logs to see whether I need to make an update to trunk again.
23:30:43  <Ammler> what happen with Szvengar?
23:30:57  <GT> dunno, havent seen him for a while
23:31:21  <GT> also Ben Robbins has not been around for some time
23:32:27  <Ammler> hmm, my stupid XenServer Guest doesn't like to go to the internet :'-(
23:32:55  <GT> How right he it is, who would want to go there?
23:33:21  <Ammler> it is quite cool else
23:33:35  <Ammler> but running servers inside a host only is a bit boring
23:35:20  <GT> I think you're right there,
23:35:28  <Rubidium> Ammler: that's better than the issue we had (or maybe still have) with virtualbox
23:35:49  <Ammler> well, for the CF, this doesn't hurt :-)
23:36:10  <Ammler> but my guest should become the DevZone
23:36:15  <Rubidium> for some reason starting 3 vboxes at roughly the same time with networking enabled makes the closest router go haywire and drops the connection with the server
23:37:12  <GT> And the closest router of course wasnt yours?
23:37:41  <Ammler> Rubidium: the vboxes get public ips?
23:37:56  <Rubidium> Ammler: nope, still something happened
23:38:13  <Ammler> he, that sounds indeed quite strange
23:38:38  <Ammler> well, here the problem is that the ips are fixed to the macaddress
23:38:41  <Rubidium> you might remember the times when the dorpsgek would timeout when the nightly ran?
23:38:47  <Ammler> so I need to setup the host as router
23:39:02  <Rubidium> (and the website would become unreachable)
23:39:14  <Rubidium> well, that was the virtualbox issue
23:39:57  <Ammler> do you still use vbox?
23:40:02  <Rubidium> yes
23:40:16  <Ammler> so this qemu test are for seomthing else?
23:40:34  <Rubidium> yup
23:40:46  <Rubidium> seen my boasting about compile times a few days ago?
23:41:03  <Rubidium> 5 minutes for NFORenum, 28 minutes for NFORenum?
23:41:31  <Rubidium> the first is compilation within qemu-system-ppc (yes, emulated ppc), the second in hercules (emulated s390)
23:41:32  <Ammler> yeah, well lightly
23:41:56  <Rubidium> in other words, it's slow to extremely slow
23:42:11  <Ammler> that is because of the emu, I assume?
23:42:27  <Rubidium> e.g. in hercules compiling OpenTTD takes almost 200 minutes
23:42:30  <Ammler> if you could do it with native cpu, it wouldn't matter?
23:43:09  <Rubidium> if you've got a native CPU emulating would be stupid, but yes emulating the native CPU wouldn't be faster
23:43:13  <Ammler> did you ever consider xen as you setup the CF?
23:43:41  * GT wonders if Kenji is using a vbox
23:44:10  <Ammler> hmm?
23:44:43  <GT> ignore me, the logging server was out of service a couple of days, but its back again
23:44:52  <Ammler> ah :-)
23:45:04  <Ammler> yes, Webster is back
23:45:15  <Ammler> @services op
23:45:15  *** ChanServ sets mode: +o Webster
23:45:54  <Rubidium> Ammler: yes, but only after the current CF was built
23:46:43  <Rubidium> I didn't have a 64 bits CPU nor hardware virtualisation back then, so making a Xen image with Windows wouldn't be possible
23:47:07  <Ammler> hehe, same here, wasn't able to test xen
23:47:27  <Ammler> well, and we also had broken harddisk
23:47:30  <GT> what's the diff between brot6 and Webster?
23:47:51  <Ammler> brot6 is just announcer for devzone
23:48:01  <Ammler> on the same host
23:48:09  <Ammler> Webster: is our generic irc bot
23:48:43  <GT> @calc 1+1
23:48:44  <Webster> GT: 2
23:48:54  <Rubidium> and going to Xen seems to kind-of require a full reinstallation
23:49:00  <Ammler> !calc 1+1
23:49:19  <Rubidium> which requires a second server and loads of time. Not something I or TB are swimming in
23:49:27  <Ammler> :-)
23:50:55  <Ammler> hmm, I should test if the XenCenter works with wine
23:51:32  <GT> Well, TB doesnt do AI anymore, and also did not post on the forums since April, so he must have some spare time
23:51:57  <Ammler> hehe, tell him :-P
23:52:16  <GT> He went to sleep at 12 pm, he must be getting old
23:53:09  <Rubidium> nasty side-effect of the must-finish-masters-thesis-syndrome?
23:54:12  <GT> I thought he was already past that, but then he has a good reason, I wouldnt blame him for that.
23:55:12  *** KenjiE20 has quit IRC
23:55:27  <GT> And he needs to get his prios right, after all, what is more important, a title or be respected by the OTTD community.
23:58:32  <GT> Luckily I'm old enough to dont care about both. :-D

Powered by YARRSTE version: svn-trunk