Log for #openttdcoop.devzone on 7th July 2010:
Times are UTC Toggle Colours
00:03:33  *** KenjiE20 has quit IRC
05:50:06  *** andythenorth has joined #openttdcoop.devzone
07:50:46  *** ODM has joined #openttdcoop.devzone
07:53:20  *** TheODM has joined #openttdcoop.devzone
07:53:24  *** TheODM has quit IRC
09:31:35  *** FooBar has joined #openttdcoop.devzone
09:35:22  <Ammler> planetmaker: your nforenum -> renum choser seems not to work for every distro, do you like me to find which?
09:35:22  *** FooBar has quit IRC
09:36:48  *** FooBar has joined #openttdcoop.devzone
09:48:34  <Rubidium> oh... that's another advantage of "forking"... we can just call it nforenum and be done with it
09:51:42  <Rubidium> Ammler: ofcourse it's useful to know where something doesn't work as expected
09:52:00  <Rubidium> and I'm also wondering how long to wait for Dalestan before just pushing on
09:52:58  <Rubidium> and... what do you prefer? Continueing at the ttdpatch svn repository or just cloning it and moving "everything" to (so there'll be trackers and such)?
09:55:37  <FooBar> Has anyone tried contacting DaleStan?
09:55:49  <FooBar> That would be the least to do if you're unsure how long to wait
09:56:28  <Rubidium> FooBar: yes, I asked O. Rudge for his forum email address and did send a mail yesterday
09:56:33  <FooBar> ok good
09:56:44  <Rubidium> as a forum PM wouldn't do much good as he hasn't been there for over a month
09:57:01  <FooBar> I think he wouldn't object to the "fork-and-give-back-on-return" idea
09:57:45  <FooBar> He might have PM's forwarded to his email
09:58:50  <FooBar> But given that it's holiday season, you probably have to wait two weeks for email replies :(
09:59:48  <FooBar> But on the other hand: it's GPLed :)
10:02:17  <Rubidium> FooBar: ofcourse it's possible, but asking the previous stakeholders seems like the right/ethic thing to do
10:02:38  <FooBar> I know, hence the suggestion to try and contact Dale ;)
10:04:21  <FooBar> Is XChat any good?
10:04:29  <Rubidium> for Windows?
10:04:31  <FooBar> yes
10:05:11  <FooBar> I have ChatZilla now, but I have the idea that that boost the memory consumption of Firefox... firefox.exe is using 120 MB now!
10:05:44  <Ammler> [11:52] <Rubidium> and... what do you prefer? Continueing at the ttdpatch svn repository or just cloning it and moving "everything" to (so there'll be trackers and such)? <-- up to the devs, who would continue, like you and yexo :-)
10:05:52  <FooBar> So I'm looking for something lightweight and preferrably open source/free/of which no commercial version exists
10:06:23  <Rubidium> FooBar: I don't fancy the behaviour of the developer... with his paid for official version and the 3rd party free binary
10:06:56  <Ammler> I would prefer converting to HG of course ;-)
10:09:19  <FooBar> Rubidium: I see...
10:09:27  * FooBar carries on and looks for something else
10:10:25  <Ammler> I think, there is no big issue to fork it and when ttdpatcher returns, they pull the changes from the fork
10:11:34  <Rubidium> Ammler: problem is versioning... it'd basically mean we "need" to release a new release version "quickly" and then start with nightlies from r123 or something
10:11:59  <Ammler> or call it nforenum2 and grfcodec2
10:12:03  <Rubidium> as x.y.z-r2300 -> x.y.z-r1 isn't something that distributions like :)
10:12:12  <Rubidium> Ammler: that makes the upgrade path only harder
10:12:35  <Ammler> are you allowed to call a fork the same?
10:12:57  <Ammler> don't you need to make it obvious, it is a fork?
10:13:14  <Rubidium> that's why it isn't really "forking"
10:13:28  <Rubidium> and to prevent such a mess I rather have the okay of Dale
10:13:31  <Ammler> ah, ok :-)
10:14:07  <Ammler> en guete :-)
10:17:34  *** FooBar has quit IRC
10:23:36  *** FooBar has joined #openttdcoop.devzone
10:29:41  <FooBar> It seems Firefox' memory consumption is high just because I have a lot of unused ram sitting around
10:30:13  <FooBar> Doesn't have much to do with Chatzilla as it appears, so I'm going to stick with that
10:34:36  *** KenjiE20 has joined #openttdcoop.devzone
10:35:42  *** Seberoth has joined #openttdcoop.devzone
12:10:29  *** Seberoth has quit IRC
12:13:07  *** Seberoth has joined #openttdcoop.devzone
12:23:17  <Brot6> FIRS Industry Replacement Set - Bug #924 (Rejected): Warning (86): Offset 32: Testing nonexistant... (foobar) @
12:27:26  <Brot6> FIRS Industry Replacement Set - Bug #915 (Resolved): Incorrect strings (foobar) @
12:27:26  <Brot6> FIRS Industry Replacement Set - Bug #915 (Closed): Incorrect strings (foobar) @
12:44:25  <Brot6> FIRS Industry Replacement Set - Bug #713 (Closed): Existing language translations might be out of... (foobar) @
12:51:49  <Brot6> FIRS Industry Replacement Set - Bug #898 (Closed): Bank appearing in Towns/Cities and accepting Sand (foobar) @
14:00:58  <FooBar> Ammler: why does this not work: 'Feature #480; Feature #796; Feature #1062: Bring back fishing harbour as secondary industry.'?
14:01:08  <FooBar> as commit message
14:02:14  <Ammler> FooBar: that is ugly
14:02:23  <FooBar> suggestions?
14:02:42  <FooBar> There is very little differences between the issues to justify seperate commits
14:02:47  <Ammler> do "Feature: Bring back fishing harbour as secondary industry (issues #455, #455, #455)
14:02:57  <FooBar> ok, will do :)
14:03:09  <Ammler> or closes
14:03:36  <Ammler> hmm, would "Feature #345, #455, #45: ...." work?
14:04:33  <FooBar> already pushed, so you'll have to wait for next time something like this occurs before I test it ;)
14:04:39  <Brot6> FIRS Industry Replacement Set - Revision 1034:ddae1c5ae921: Feature: Bring back fishing harbour a... (foobar) @
14:05:06  <Ammler> LOL
14:05:16  <Ammler> you copied my suggestion :-P
14:05:27  <FooBar> well, obviously...
14:05:32  <Ammler> notice my dummy numbers
14:05:38  <Ammler> :-D
14:05:42  <FooBar> I noticed that something different is closed now...
14:05:54  <FooBar> how do we undo this?
14:06:13  <Ammler> not possible
14:07:00  <FooBar> ok, then I close manually...
14:07:44  <Brot6> FIRS Industry Replacement Set - Feature #480 (Closed): Fishing Harbour (foobar) @
14:08:59  <Brot6> FIRS Industry Replacement Set - Feature #796 (Closed): Fishing Harbour production code (foobar) @
14:08:59  <Brot6> FIRS Industry Replacement Set - Feature #1062 (Closed): Change Fishing Harbour to secondary industry (foobar) @
15:32:36  <DJNekkid> Go Thor !
15:35:28  <planetmaker> <Ammler> planetmaker: your nforenum -> renum choser seems not to work for every distro, do you like me to find which? <-- IIRC it only works for Debian(?). If you know more I'll happily add more.
15:35:55  <planetmaker> <Rubidium> and... what do you prefer? Continueing at the ttdpatch svn repository or just cloning it and moving "everything" to (so there'll be trackers and such)? <-- The latter would be more accessible from my POV
15:38:32  <Ammler> planetmaker: I thought, it is a general thing :-)
15:38:36  <DJNekkid> gratz Ammler :)
15:38:44  <Ammler> DJNekkid: ?
15:38:55  <planetmaker> But I do prefer the solution to officially continue those programmes w/o calling it a fork officially
15:39:07  <DJNekkid> tour de france victory :)
15:39:23  <Ammler> oh well :-)
15:42:28  <Ammler> if we add nforenum to the devzone, we can use nightlies like we use nml currently
15:46:17  <planetmaker> yes, indeed
15:46:36  <planetmaker> though we still could do that if hosted elsewhere, if we just pull and compile ourselves
15:46:44  <planetmaker> But yes, that'd make this thing much easier
15:47:00  <planetmaker> And it'd have the advantage that this would be THE central place for the newgrf tools
15:47:22  <planetmaker> which would make it easier for newcomers.
15:49:55  <Ammler> yes, of course, it "main" repo doesn't need to be on the devzone
15:50:00  <Ammler> the*
15:50:21  <Ammler> but currently, there is no need, as there is no development
15:50:39  <planetmaker> yes
15:50:41  <Ammler> I just mean, we don't need to make a release with every patch
15:50:51  <planetmaker> of course not.
15:51:20  <planetmaker> Rubidium, is certainly right, it'd make sense to fix the current open bugs (or which there solutions exist for) and then give that thing a release tag
15:51:26  <planetmaker> and then continue normally with nightlies
15:51:48  <planetmaker> And thinking of it... I'm in favour of moving the repo here to the devzone. Whatever VCS is preferred
15:52:05  <planetmaker> I don't mind that too much.
15:54:45  <Ammler> well, mercurial would have easiest access management
15:55:19  <Ammler> and easiest "backup"
16:13:32  <Ammler> planetmaker: how do you detect debian?
16:16:59  *** frosch123 has joined #openttdcoop.devzone
16:18:58  <Brot6> 2cctrainset: update from r562 to r563 done (1 errors) -
16:19:38  <Brot6> firs: update from r1031 to r1034 done -
16:20:23  <Brot6> nml: compile of r532 failed -
16:20:29  <Brot6> Following repos didn't need a nightlies update: 32bpp-extra (r36), airportsplus (r52), bros (r12), comic-houses (r70), fish (r386), heqs (r346), newgrf_makefile (r120), nmts (r16), nutracks (r82), ogfxplus (r39), opengfx (r464), openmsx (r80), opensfx (r96), snowlinemod (r15), swedishrails (r135), worldairlinersset (r648)
16:24:53  <Ammler> WHO played with railtypes? :-P
16:39:28  <Ammler> planetmaker: see the failed distros:
16:39:38  <Ammler> centos failed because of renum not found
16:40:07  <Ammler> 11.3 fails because of grfcodec
18:09:13  <andythenorth> evening
18:18:44  <FooBar> hello there
18:19:00  <FooBar> I've thrown myself in the pit of varaction2advanced
18:19:13  <FooBar> probably shouldn't have done that
18:19:57  <Hirundo> you're better off using NML for that :)
18:20:22  <andythenorth> FooBar: what are you trying?
18:20:50  <FooBar> I'm trying to use a registry to display the introduction year in the fund industry window
18:21:05  <andythenorth> that's nice, there was a ticket for that :)
18:21:12  <FooBar> I'm not even sure if that's even possible
18:21:17  <andythenorth> it is possible
18:21:25  <andythenorth> you'll need to understand the text stack
18:21:37  <FooBar> Well, the wiki doesn't list it as a possibility, but the text stack does display something
18:21:45  <FooBar> now I just have to set it to the proper value
18:22:01  <andythenorth> have you seen the debug text stuff in some of the templates?
18:22:36  <FooBar> not in particular, but I've looked at some other things in the templates
18:23:10  <andythenorth> have a look in template_secondary_action23_A.pnfo
18:23:30  <andythenorth> actions FE-F0
18:23:44  <andythenorth> they store stuff on the text stack
18:23:55  <andythenorth> the values are then printed in STR_INFO_DEBUG_1
18:24:08  <andythenorth> it took me a day of work to make sense of, but then it's quite easy
18:24:10  <andythenorth> :P
18:25:54  <FooBar> good to know, I might give up easier than you do. Then I leave the pieces to you to finish :P
18:26:09  <andythenorth> umm thanks
18:26:14  <FooBar> :)
18:26:40  <andythenorth> you have two routes, I don't know if both would work
18:26:47  <frosch123> FooBar: it took us several weeks to turn andy into an nfo monster :)
18:27:15  <frosch123> maybe it only worked after we offered him to become mac maintainer :p
18:27:16  <andythenorth> (1) use the CPP to write a value into a varaction2, then push that on the stack, then have a string that uses that
18:27:23  <FooBar> Then expect a lot of work: I'm just an nfo nibbler
18:27:27  <andythenorth> that's only one varact 2
18:27:30  <andythenorth> and one string
18:27:43  <andythenorth> (2) hmmm
18:27:44  <andythenorth> 2 is bad
18:27:52  <andythenorth> and involves defining lots of extra strings
18:28:08  <FooBar> I thought of (2) too, but didn't like it
18:28:24  <andythenorth> try for 1 then :)
18:28:25  <FooBar> Imagine 20 different introduction years
18:29:03  * andythenorth checks available string codes
18:29:24  <FooBar> great, now there's a "linter failure"
18:29:48  <andythenorth> happens
18:31:03  <FooBar> I probably missed something
18:32:06  <andythenorth> paste code if you get stuck
18:32:40  <FooBar> not yet
19:02:39  *** KenjiE20 has quit IRC
19:16:09  <Hirundo> What is wrong with this commit message: "Feature [#981]: Detect DOS/Win palette, and report an error upon encountering a palette that doesn't match either of them." ?
19:17:34  <FooBar> the [ and ]
19:17:50  <FooBar> use "Feature #123:"
19:18:33  <Hirundo> <- this commit has it too
19:21:13  <FooBar> andythenorth: I seem to fail to understand how to either read something from the stack or write something to it:
19:22:11  <FooBar> Hirundo: maybe it's some new check? I encountered something similar today too
19:26:10  <frosch123> INTRO_DATE? i thought build date.
19:26:43  <frosch123> for intro date you do not need the stack, just let the preprocessor include it in the text via #INTRO_DATE
19:27:20  <frosch123> hmm, ok, that would need separate texts for each industry, maybe a bad idea :)
19:29:11  <frosch123> likely you need a \d in front of INTRO_DATE, but you could also remove the add/divide part, and put it directly in the and-mask
19:30:33  <frosch123> and the result string likely needs some C0 or C4
19:30:46  <FooBar> no, intro date, it must be displayed in the fund industry window
19:31:22  <FooBar> and it has a \d in front of it in the define
19:31:42  <frosch123> and the string define?
19:32:28  <FooBar> the string definition is listed at the bottom
19:32:50  <FooBar> "Available: \UE07C"
19:33:07  <frosch123> well, but with \b or something like that?
19:33:32  <frosch123> or did you put the raw string in the sprite?
19:35:10  <FooBar> I updated pastebin a bit with some additional info
19:35:22  <FooBar> the STR INFO CB38 is a simple byte
19:35:40  <FooBar> That part works, as the string is displayed, but with the wrong value
19:35:51  <Ammler> Hirundo: yes, no [] around tickets
19:35:57  <Ammler> only () around revs
19:36:38  <Ammler>
19:37:30  <Hirundo> What's the [#NUM]-part in there, then?
19:37:43  <Ammler> []=optional :-)
19:38:05  <Ammler> the explaination could be better
19:38:14  <Ammler> please suggest something
19:38:41  <andythenorth> FooBar: making dinner
19:39:04  <FooBar> ok, enjoy your dinner then :)
19:39:39  <Ammler> watching football without sound is better
19:39:44  <frosch123> FooBar: anyway, you can simplify the whole stuff by using 1A 20 THIS_INTRO_DATE    0E 1A 20 \dx100
19:39:57  <andythenorth> that was my first thought
19:40:09  <frosch123> and putting the string code in the actual cases
19:40:13  <andythenorth> I was trying to make sense of the add-divide stuff, but I've not used it that way
19:40:25  <andythenorth> the code frosch123 wrote is what I would write
19:40:43  <FooBar> well, I needed to add something and I had no clue, so I just tried something.
19:40:51  <andythenorth> try the other thing instead :)
19:40:51  <FooBar> but thanks, I'll try this now
19:41:17  <Hirundo> Ammler: I still don't get, why it worked previously (e.g. nml r527)
19:41:42  <Ammler> because we enabled the hook since this weekend
19:42:04  <Ammler> and tickets enclosed in [] don't close
19:42:24  <Brot6> NewGRF Meta Language - Revision 533:f6bb803cf258: Add: palette data to (Hirundo) @
19:42:24  <Brot6> NewGRF Meta Language - Revision 534:87d758c4c666: Feature: Detect dos/win palette, and report an ... (Hirundo) @
19:43:26  <Hirundo> I didn't intend it to close, but nvm
19:44:08  <Ammler> hmm
19:44:50  <FooBar> which me luck :)
19:44:57  <FooBar> wish*
19:45:06  <frosch123> andythenorth: FooBar: that callback does not support the textstack at all
19:45:30  <FooBar> frosch123: I expected something like that because it wasn't documented, but it /does/ display output from something
19:45:40  <andythenorth> no text stack?  lame :P
19:45:48  <frosch123> let's add it :p
19:45:51  <FooBar> let that be a feature-request then
19:45:52  <Ammler> Hirundo: I do not like that Feature does autoclose
19:46:08  *** KenjiE20 has joined #openttdcoop.devzone
19:46:10  <FooBar> Ammler: I do :)
19:46:35  <Hirundo> then perhaps support Feature [closes #123]:
19:47:17  <FooBar> ah, luckily the new code also gives the same result
19:47:30  <FooBar> 1,242,376
19:48:27  <Ammler> Hirundo: if FooBar and planetmaker agrees, we could change that
19:48:41  <FooBar> do I have a veto?
19:48:42  <Ammler> or you add a custom config to nml
19:48:47  <andythenorth> FooBar I suspect we might run into problems because we are storing a string / dec date and reading it as a word
19:49:11  <andythenorth> or maybe not
19:49:22  <FooBar> well, it's all numbers...
19:49:31  <FooBar> I could live with the additional comma
19:51:07  <FooBar> Ammler: I quite like the autoclose, as I'm used to it
19:51:35  <Ammler> yes, but you could do that with an additional keyword close
19:51:51  <FooBar> plus, if a feature shouldn't close an issue, it's not a Feature, it an Add
19:51:52  <Ammler> ususally, Features have more than one commit
19:51:56  <Ammler> or need feedback
19:51:58  <FooBar> IMO
19:52:07  <frosch123> <- does that work?
19:53:08  <FooBar> if it works for you, it works for me ;) I cannot compile OpenTTD here, at least not within the next 30 hours because I have nothing set up to do so
19:53:29  <andythenorth> I can compile for Mac
19:53:30  <frosch123> i have no testcase :)
19:53:44  <frosch123> give me the diff for firs, and i can test :)
19:53:47  <andythenorth> FooBar: commit the test case?  Or diff it?
19:53:49  <FooBar> I can provide a testcase. I can just commit to FIRS, because it doesn't break anything
19:54:03  <FooBar> or provide the grf, as I have it compiled here...
19:54:13  <FooBar> that might work easiest
19:54:22  <frosch123> i also have a firs checkout
19:54:26  <andythenorth> does my intro date code even work anyway?
19:54:36  <andythenorth> how does /d1920 get used as a valid date?
19:54:43  <andythenorth> this must be broken surely?
19:55:04  <FooBar>
19:55:28  <andythenorth> hmm
19:55:30  <Brot6> FIRS Industry Replacement Set - firs.grf (foobar) @
19:55:58  <andythenorth> renum/grfcodec magically knows it's a date and not the value 1,920 apparently
19:56:16  <andythenorth> or perhaps not
19:56:20  <andythenorth> meh
19:56:20  <andythenorth>
19:56:29  <andythenorth> I should make dinner.
19:56:34  <FooBar> well, it shouldn't be a date in my case, just a year...
19:57:26  <Rubidium> Ammler: did you succeed with the newer gcc?
19:58:11  * andythenorth recalls testing intro dates, they seemed
19:58:31  <frosch123> cement plant: "available 1,902"
19:58:41  <FooBar> great!
19:58:44  <frosch123> you should use a stringcode without thousand separator :)
19:58:46  <andythenorth> great!
19:58:49  <andythenorth> apart from the comma
20:00:18  <FooBar> good to hear, then I can finish this feature
20:00:34  <FooBar> It does mean that some weird value is displayed in older versions (provided this gets committed)
20:01:07  <Ammler> :'-(
20:01:14  <andythenorth> FooBar: we could enforce a minimal version
20:01:33  * andythenorth looks for a string code without separator
20:02:00  <FooBar> don't know if there is any
20:02:07  <andythenorth> me neither
20:02:43  <andythenorth> what happens if we print it as a date (83)
20:03:48  <FooBar> I think date is expected as number of days since something (probably 1920)
20:04:11  <andythenorth> that's not impossible to provide
20:04:20  <andythenorth> it would mean changing the intro date varaction 2
20:04:26  <andythenorth> and doing some maths :P
20:04:51  <andythenorth> or we could multiply the year value by 365 plus some fiddle for leap years
20:04:55  <andythenorth> fricking leap years :P
20:04:59  <FooBar> I agree
20:06:18  <FooBar> multiply by 36525 and divide by 100 should get us close enough to display the correct month
20:06:28  <Ammler> 21:57] <Rubidium> Ammler: did you succeed with the newer gcc? <-- I guess, we just wait for 4.5.1 :-)
20:06:34  <andythenorth> or multiply by 1461 / 4
20:06:53  <FooBar> oh yes: (svn r20086) -Add: Textstack support for CB 38.
20:07:01  <Hirundo> year*365+year/4-year/100+year/400 is used by nml, though I agree that the above approximation probably suffices
20:07:04  <FooBar> I'll update the specification
20:07:32  <andythenorth> Hirundo: good luck doing that in a varact 2
20:07:43  <andythenorth> it's easy, but quite a lot of advanced varact 2 :)
20:07:58  <FooBar> oh, frosch already updated it
20:08:31  <frosch123> the datestuff only works from 1920 on
20:08:38  <andythenorth> oh poop
20:08:46  <frosch123> and there is no number without separator, except the hex numbers :p
20:08:49  <FooBar> have a napkin
20:08:53  <Rubidium> Ammler: gcc -dumpversion returns 4.5.0 for you?
20:09:12  <andythenorth> can we have a new string code?
20:09:21  <FooBar> well, then we split the year in two parts and use two registers
20:09:37  <andythenorth> FooBar: cunning
20:09:51  <andythenorth> or unpack it as two bytes
20:10:05  <FooBar> might give things like 0598 for years <1000, but I can live with that
20:10:12  <FooBar> also possible
20:10:33  <FooBar> might be even easier, although the year needs to be stored the other way round
20:10:42  <FooBar> 0219 for 1902
20:10:55  <andythenorth> once again we find nfo is a meta-game all of its own :P
20:10:56  <Rubidium> why not use the thing to print dates?
20:10:57  <FooBar> and then you might end up with 192
20:11:10  <frosch123> Rubidium: 1920 based
20:11:37  <FooBar> maybe only introduce industries after 1920 and have the rest from 0
20:11:56  <FooBar> or only introduce industries until 999 :P
20:11:57  <Hirundo> andythenorth:
20:12:07  <Rubidium> use HEX :)
20:12:21  <andythenorth> FooBar: I dislike the gameplay hack :)
20:12:28  <FooBar> :)
20:13:00  <frosch123> hex is defintely easiest
20:13:22  <frosch123> just put a \d INTRO_DATE in the action 0, and a \dx INTRO_DATE on the stack
20:13:28  <andythenorth> store the whole thing in register 1, nibble the second byte, nibble the first byte store, store them both in register 2
20:13:33  <frosch123> and hexnumbers have no leading 0 in ottd
20:13:35  <andythenorth> thereby reversing the bytes
20:13:49  <frosch123> what reversing?
20:14:10  <andythenorth> so we can then print the two bytes, avoiding the ,
20:14:24  <frosch123> damn, ottd prefixes hex numbers with 0x, so that does not work either :s
20:14:36  <Ammler> Rubidium:
20:14:37  <Ammler> > gcc -dumpversion
20:14:39  <Ammler> 4.4
20:14:40  <Ammler> :-o
20:14:47  <Rubidium> for gcc 4.5?!?
20:15:03  <Ammler> hehe, that is why I wonder too
20:15:11  <andythenorth> last time I tried something like this I conclude a slightly-pythonic thing
20:15:19  <andythenorth> writing out multiple values by hand may be bad, but it's better than writing arcane code
20:15:37  <Ammler> ah
20:16:37  <FooBar> I'll look into it tomorrow. There are several options at hand and I'm glad that the text stack works
20:16:52  <andythenorth> it's handy
20:17:02  <Ammler> maybe the bug is in opengfx, not grfcodec
20:17:15  <andythenorth> FooBar you know that the debug stuff shows you what's in storage now?
20:17:27  <andythenorth> so if you want to see a register value, push it to storage
20:17:57  <andythenorth> helps preserve sanity :)
20:18:33  <FooBar> yes, I saw the debug stuff. Very nice indeed!
20:18:53  <FooBar> but it's more that I dont' feel like looking into this anymore today :)
20:19:07  <FooBar> If you like to continue, I can push what I have so far...
20:19:26  <andythenorth> probably won't tonight in honesty
20:20:19  <FooBar> ok, then I just keep it here and finish it tomorrow
20:20:38  <andythenorth> there is an alternative route
20:20:40  <FooBar> Germany - Spain 0-1
20:21:18  <andythenorth> use a script to generate this stuff at compile time
20:21:24  <andythenorth> that already happens for cargos
20:22:04  <andythenorth> adds complexity though
20:22:18  <FooBar> well, "this stuff" just hooks into the template
20:22:26  <Rubidium> Ammler: okay... if you recompile grfcodec from scratch, does it create a broken opengfx?
20:23:31  <FooBar> if we're going to use more scripts, we better change to NML altogether ;)
20:26:42  * andythenorth has not objections, but isn't doing the code migration :o
20:27:35  <FooBar> well, me neither
20:27:47  <andythenorth> I guess we're stuck in nfo then
20:27:53  <andythenorth> nfo is good for the brain
20:28:14  <Rubidium> Ammler: did you create the grfcodec/nforenum repositories? If so, it's not that nice to put the "main" stuff already in a branch
20:28:17  <FooBar> Unless there will be an NFO importer for NML, I think the change for FIRS will never happen. Just too many lines of code
20:28:37  <andythenorth> most of it is just definitions
20:28:48  <andythenorth> there are only about 5 important production templates that are complex
20:28:54  <Ammler> Rubidium: I didn't care, just made hg convert :-)
20:29:19  <Ammler> I guess, it should be possible to change that with convert
20:29:36  * andythenorth wonders if action 0 could be converted to nml programatically 
20:29:52  <Brot6> NewGRF Meta Language - Revision 531:ca35bbd00bd4: Doc: Document Action0Property instance variables. (Alberth) @
20:30:56  <FooBar> Action 0 shouldn't be too hard. Especially the way we wrote it: every property on a new line
20:31:17  * Rubidium is considering starting the unofficial "port" with just some of the applied patches and then when it all becomes official really make the devzone stuff
20:31:19  <FooBar> well, not every...
20:31:35  <andythenorth> FooBar: layout definitions...
20:31:47  <FooBar> yes, I realized that (too late)
20:32:25  <Ammler> > gcc -dumpversion
20:32:27  <Ammler> 4.5
20:32:41  <Ammler> Rubidium: I was in wrong chroot last time :-$
20:32:43  <Rubidium> hmm... only major and minor :(
20:33:01  <Rubidium> what's gcc -v's last line
20:33:13  <Ammler> libgcc45-4.5.0_20100604-1.12.x86_64
20:33:24  <Ammler> gcc version 4.5.0 20100604 [gcc-4_5-branch revision 160292] (SUSE Linux)
20:33:37  <FooBar> anyways, I'm hitting the shower. Wash these dirty pixels off me...
20:33:42  <Rubidium> so grepping for "4\.5\.0" would match that :)
20:33:58  <Rubidium> or in other words: we can disable optimisations to work around the bug
20:33:59  <Ammler> ?
20:35:10  <Ammler> but if you disable optimations, you get another grf md5sum, don't you?
20:35:42  <Rubidium> Ammler: uhm... optimisations of gcc (I thought that was kinda obvious)
20:35:50  <Ammler> Rubidium: shall I make "clean" mercurial repos?
20:35:56  <Ammler> you like to work with it?
20:36:04  <Ammler> or will you? :-)
20:36:13  <Brot6> NewGRF Meta Language - Revision 535:c31036884a0e: Add: Toyland palettes. (#981) (Hirundo) @
20:36:29  <Rubidium> I don't have the convert plugin installed, so please
20:36:55  <Ammler> then I just need to know, how I disable branching
20:37:26  <Rubidium> you're the (relative to me) mercurial expert
20:37:37  <Brot6> NewGRF Meta Language - Bug #981 (Resolved): fail more gracefully for wrong palette (Hirundo) @
20:40:11  *** frosch123 has quit IRC
20:42:21  <Ammler> Rubidium:
20:44:25  <Rubidium> yay... that looks better
20:48:05  <Rubidium> thanks... I'll look into doing some actual work tomorrow or something
20:51:39  <Ammler> I setup a OTTD Dev group and added those as Manager, feel free to edit the projects :-)
20:54:54  <Ammler> hmm, group support isn't that good
20:57:12  <Rubidium> I'm not that sure the whole tracker and such should become public yet
20:58:33  <Ammler> and grfcodec
20:58:55  <Ammler> you can hide the project and add people as watcher
20:59:16  *** ODM has quit IRC
20:59:20  * FooBar hasn't seen anything
21:01:18  <Rubidium> okay... I've made them un-public for now
21:01:53  <Rubidium> as long as there's no reply from Dalestan (or a long time passes) it'll just stay a bit hidden
21:02:39  <Ammler> shall I add nightly build?
21:02:49  <Rubidium> nope
21:03:30  <Rubidium> also... is it wise for you to do nightly builds?
21:03:36  <Rubidium> can you build the Windows binaries"
21:03:40  <Rubidium> s/"/?
21:03:58  <Ammler> well, I guess, that would the official cf do
21:04:00  <Ammler> (yours)
21:04:18  <Ammler> I would just make it so our projects would use it
21:04:36  <Rubidium> which makes nightlies... especially when started at another time not that useful
21:04:49  <Rubidium> although building it to use it for your server should be fine
21:06:44  <Ammler> hmm
21:07:04  <Brot6> mode change on /home/ottdc/hg-repos/grfcodec
21:07:04  <Brot6> mode change on /home/ottdc/hg-repos/nforenum
21:51:16  <planetmaker> hello
21:54:12  <planetmaker> [21:51]	<FooBar>	plus, if a feature shouldn't close an issue, it's not a Feature, it an Add <-- I do disagree
21:54:25  <planetmaker> A feature can be cut down into several sub-features
21:55:02  <planetmaker> Like "new road vehicles". Which can be Feature: add baby buggy; Feature: Add ketcar
21:55:22  <planetmaker> as such I object to "Feature closes an issue"
21:55:42  <planetmaker> Actually I explicitly changed that to that (it did way back in time)
21:58:23  <planetmaker> [22:28]	<FooBar>	Unless there will be an NFO importer for NML, I think the change for FIRS will never happen. Just too many lines of code <-- it would take a bit.
21:58:33  <planetmaker> But it's probably less work than it looks
22:20:29  <FooBar> hi
22:21:06  <FooBar> ok, then feature not close issue, as there is more people against it than in favour :)
22:21:16  <FooBar> But let me know when it's changed ;)
22:21:58  <FooBar> But then again... "Feature: Add ketcar" is just "Add: ketcar" ;)
22:22:16  <planetmaker> no, why?
22:22:26  <planetmaker> If ketcar is a vehicle, it's IMHO a feature.
22:22:38  <planetmaker> or what would be in your eyes a feature for a vehicle grf?
22:22:56  <FooBar> Well, there is an /Add/ keyword...
22:23:05  <planetmaker> yes: Add: sprites for ketcar
22:23:09  <planetmaker> Feature: Implement ketcar
22:23:25  <FooBar> a feature is a feature if it's done, until then it's not yet a feature
22:23:27  <Brot6> NewGRF Meta Language - Bug #981: fail more gracefully for wrong palette (planetmaker) @
22:23:42  <FooBar> but then again, I'm not opposed to changing it
22:23:43  <planetmaker> FooBar: exactly. But an issue can have many features
22:23:49  <FooBar> ok
22:23:56  <Ammler> hehe, we had that discussion a year ago :-P
22:23:59  <planetmaker> Like the generic feature "add all Australiean cars"
22:24:12  <planetmaker> which could be many small ones.
22:24:21  <FooBar> I believe the outcome was different a year ago :P
22:24:25  <planetmaker> Or "revise all trains" like I have for OpenGFX
22:25:09  <FooBar> Maybe change it that only "close" actually closes an issue
22:25:19  <FooBar> then everyone can use keywords as they see most fit
22:27:25  <FooBar> Oh, and as for FIRS->NML: I'm open to porting it to NML, I'm just not open to porting it myself ;)
22:30:04  <planetmaker> Well, don't do it then.
22:30:12  <planetmaker> But you might gain
22:30:25  <planetmaker> But don't do it now
22:30:53  <planetmaker> wait a bit till NML matured a bit
22:30:57  <planetmaker> and at least 0.1 is out
22:31:04  <FooBar> might be smart
22:31:17  <planetmaker> FIRS is too complex to be 100% supported right now, I think
22:31:44  <FooBar> I will use it when I start recoding Dutch tramset and/or Transrapid track set
22:31:57  <planetmaker> That might be smart
22:32:20  <FooBar> dunno if andy will let me work on that any time soon though :P
22:32:41  <planetmaker> Well... anyone with GPL license is free to copy SwedishRails and just replace sprites. But I guess we talked about that ;-)
22:32:52  <planetmaker> FooBar: he won't. For sure
22:32:58  <FooBar> yes, we talked about that ;)
22:33:34  <planetmaker> Well. You'll understand the code probably immediately by reading it.
22:33:40  <planetmaker> Very much unlike NFO
22:33:41  <FooBar> I better work ahead on some features then and only commit one every 2 days :P
22:33:49  <planetmaker> :-D
22:34:49  <FooBar> I understand basic programming language syntaxes plus I understand a programming language (php), so that's probably enough to understand nml
22:34:57  <planetmaker> yes
22:35:24  <planetmaker> still valid though: read it backwards :-)
22:35:58  <planetmaker> the action3->action2->action1 is still present, though in words well disguised
22:36:18  <planetmaker> but you'll easily recognize that
22:36:55  * FooBar checks out swedish rails
22:40:46  <planetmaker> the code files are similarily structured as you'd expect it from nfo
22:41:22  <planetmaker> or rather from our other projects
22:41:33  <planetmaker> with single sings in separate files
22:41:41  <planetmaker> and even a few templates
22:41:59  <planetmaker> both NML and cpp ones
22:43:11  <planetmaker> I like the support for sprite templates :-)
22:44:16  <FooBar> yes
22:44:54  <FooBar> I think it's more understandable than NFO if you're new to it
22:45:05  <planetmaker> MUCH more
22:45:32  <FooBar> it uses words, not just numbers :D
22:45:48  <FooBar> (and I don't mean \w words :P)
22:46:40  <FooBar> good development, this
22:47:26  <FooBar> oh, by the way, too bad Germany lost tonight. I really hoped for a Netherlands - Germany final, that would've been epic
22:47:55  <Ammler> good night
22:48:02  <FooBar> good night Ammler
22:49:23  <planetmaker> good night, Ammler
22:49:34  <planetmaker> FooBar: yes, I'd have hoped for that match, too.
22:49:52  <planetmaker> But... the result of this evening's match is ok, given how teams played
22:50:47  <FooBar> yes, that's true, but given how Germany played in other matches, they did diserve to win
22:51:04  <FooBar> Seems that the octopus was right after all...
22:51:38  <planetmaker> what octopus?
22:51:47  <FooBar> you don't know?
22:51:54  <planetmaker> and yes, IF they had played like the two matches before... then they'd have won
22:52:00  <planetmaker> I don't
22:52:08  <FooBar> let me google it up...
22:52:57  <FooBar>
22:52:59  <Webster> Title: Octopus oracle shows Germany the red card - Yahoo! News (at
22:58:03  <planetmaker> omg...!
22:58:15  <planetmaker> so it was right :-)
22:58:24  <planetmaker> Makes it a nice thing ;-)
22:58:36  <planetmaker> and I ... think that bed will be nice now, too :-)
22:58:43  <planetmaker> So a good night from here, too
23:01:36  <FooBar> yes, same here. Good night!
23:09:39  *** FooBar has quit IRC
23:48:56  *** KenjiE20 has quit IRC

Powered by YARRSTE version: svn-trunk