Log for #openttdcoop.devzone on 7th June 2010:
Times are UTC Toggle Colours
00:05:30  *** KenjiE20 has quit IRC
00:49:28  *** desrik has quit IRC
04:55:43  *** Webster has joined #openttdcoop.devzone
05:00:49  *** Webster has joined #openttdcoop.devzone
06:24:21  <planetmaker> morning
06:26:29  *** welshdragon has quit IRC
07:07:48  *** ODM has joined #openttdcoop.devzone
07:48:00  *** welshdragon has joined #openttdcoop.devzone
08:09:13  *** Seberoth has quit IRC
08:11:15  *** Seberoth has joined #openttdcoop.devzone
08:13:43  *** ODM has quit IRC
08:15:02  *** ODM has joined #openttdcoop.devzone
09:40:25  *** KenjiE20 has joined #openttdcoop.devzone
11:39:58  *** Alberth has joined #openttdcoop.devzone
12:28:45  *** ODM has quit IRC
12:28:56  *** ODM has joined #openttdcoop.devzone
15:37:06  *** frosch123 has joined #openttdcoop.devzone
16:18:55  <Brot6> nml: update from r243 to r246 done -
16:19:01  <Brot6> Following repos didn't need a nightlies update: 2cctrainset (r545), 32bpp-extra (r36), airportsplus (r50), bros (r12), comic-houses (r70), firs (r895), fish (ERROR r374), heqs (r320), newgrf_makefile (ERROR r99), nmts (r16), nutracks (r69), ogfxplus (r23), opengfx (r459), openmsx (r57), opensfx (r94), snowlinemod (r12), swedishrails (r25), test (ERROR r39), worldairlinersset (r643)
16:45:52  <Brot6> NFO Meta Language - Patch #954: NFO lexer and basic parser (Alberth) @
17:18:57  * Hirundo won't restart the copyright discussion
17:21:08  <Alberth> good idea :)
17:23:13  <Alberth> do you have an opinion about the removal of 'import *', and do import of modules only, for example?
17:27:35  <Hirundo> I don't know much about the intricacies of importing in python, but importing as little as possible seems sensible to me
17:29:05  <Alberth> It is not about importing as little as possible, it is about polluting the name space, and not being able to tell where what comes from.
17:29:06  <Hirundo> The basestring -> Identifier cq StringLiteral change is already a 600+ line diff btw, and I haven't yet touched actions/*
17:29:41  <Hirundo> Alberth: Whatever, go ahead :)
17:30:00  <Alberth> with changing the handling of string literal included?
17:30:15  <Alberth> perhaps do it in smaller steps is better then
17:31:28  <Hirundo> In retrospect, yes
17:32:12  <Hirundo> Though when I started it, I didn't plan on changing all ID tokens as well
17:32:29  <Alberth> Yexo: about #995 (not raising strings, is the current list. I fail to see how these 4 cases are not ScriptErrors though)
17:32:55  <Alberth> Hirundo: it is often bigger than anticipated
17:33:18  <Hirundo> ^^ Changing the scope of the change half-way doesn't help
17:34:58  <Alberth> a good moment to reconsider :)
17:37:29  <Hirundo> I'm doing some rudimentary splitting-by-file now
17:52:19  <Alberth> Ammler: my checker seems to prevent file renaming :(
18:04:06  <Hirundo> Is there a log somewhere of the errors in the regression makefile ?
18:05:03  <Alberth> it does "diff -u output expected"
18:05:51  <Alberth> so there is no explicit log, but "make > mylog" would do :)"
18:06:08  <Hirundo> I mean, when there is an error inside nml
18:06:31  <Hirundo> I get 'make: *** [all] Error 1' <- not very helpful
18:07:05  <Alberth> replace the /dev/null  in the makefile by 'mylog' or so
18:07:28  <Alberth> I am not even sure 2>/dev/null  should be there
18:08:14  <Alberth> unless we want to test failures too, but then you want to compare errors with expected errors
18:08:58  <Alberth> ie redirting to /dev/null  is not useful then either
18:09:04  <Alberth> *redirecting
18:21:54  <Alberth> Ammler: new version :)
18:22:42  <Brot6> NFO Meta Language - Revision 247: Codechange: Rename to to prevent import name cla... (Alberth) @
18:26:18  <Brot6> FIRS Industry Replacement Set - Feature #977 (Closed): Rework Aluminium Plant (andythenorth) @
18:34:44  <Ammler> Alberth: I install it like that right away
18:34:59  <Ammler> let us see until the first does complain about something :-)
18:35:42  <Alberth> it doesn't, I could push that changeset. I don't understand why :(
18:36:21  <Alberth> ie r247
18:39:40  <Ammler> I just installed, so it should work from now on
18:40:37  <Ammler> Alberth: added you to
18:40:44  <Ammler> so you can make test runs :-)
18:41:06  <Alberth> ok :)
18:41:30  <Ammler> hmm, might it differ between ssh and https pushes?
18:42:09  <Alberth> perhaps commit and push are different
18:42:57  <Alberth> the commit aborted on trying to find the file-context of the old name, perhaps that does not happen with a push
18:43:10  <andythenorth> Ammler: fail
18:43:15  <andythenorth> on FIRS push
18:43:44  <Ammler> remote: abort: pretxnchangegroup.check hook exited with status 127
18:45:55  <Ammler> andythenorth: any error message?
18:46:00  <Ammler> could you push again?
18:46:36  <Ammler> Alberth: I didn't test it locally anymore
18:46:47  <Ammler> but the last worked :-)
18:49:01  <Alberth> andythenorth: what were you trying to commit?
18:49:18  <Ammler> the error I pasted was because I forgot pyhton:
18:50:52  <Brot6> test - Revision 40: 13 (Ammler) @
18:54:52  <Ammler> then I was able to push ^ :-)
18:59:23  <Alberth> oh, you create new versions every time
18:59:40  <Ammler> ?
18:59:49  <Ammler> how you mean=
18:59:51  <Ammler> ?
18:59:57  <Brot6> NFO Meta Language - Revision 248: Codechange: Import 'generic' module explicitly. (Alberth) @
19:00:07  <Alberth> I did not change anything other than the .py file
19:01:33  <andythenorth> Ammler: push worked
19:02:06  <Brot6> FIRS Industry Replacement Set - Revision 896: Fix: Aluminium Plant would have failed in arctic (andythenorth) @
19:02:27  <Ammler> then you pushed while I installed the hook
19:03:41  <Brot6> NFO Meta Language - Revision 249: Fix: Import fix, boolean literal fix. (Alberth) @
19:04:04  <Alberth> FIRS 896 seems fine, it should work
19:04:27  <Ammler> Alberth: yes, but my test commit message was just a number
19:26:33  <Brot6> 2cc train set - Bug #999 (New): Voyagers bugreport (DJNekkid) @
19:27:08  <Alberth> Hmm, I cannot add an expected .grf and .nfo any more :(
19:27:20  <Ammler> lol
19:27:28  <Ammler> so it works?
19:28:19  <Alberth> locally committing stuff does work very nicely
19:29:02  <planetmaker> he... adding grfs is needed for NML indeed
19:29:14  <Ammler> Alberth: what about my proposal about a custom re in .devzone per repo
19:29:26  <Alberth> push checking fails, obviously
19:29:59  <Ammler> like I do with the building system
19:30:18  <Alberth> I just found out what .devzone means yesterday :)
19:30:24  <planetmaker> :-)
19:30:26  <Brot6> NFO Meta Language - Revision 250: Add: Added 'sounds' regression test case. (Alberth) @
19:30:52  <Alberth> see?, that should not happen
19:31:23  <Ammler> he, so what, I thought, you weren't able to?
19:31:51  * Alberth also doesn;t understand this
19:32:05  <Alberth> apparently, remote pushing does things differently
19:35:35  <planetmaker> what is the difference between pushing and remote pushing?
19:35:44  <planetmaker> or whatever the two cases are?
19:35:53  <Ammler> don't think there is one
19:36:20  <Alberth> I don't know about .devzone. I'd need a path to the repo working copy. But hg can have a situation where you don't have a working copy at all
19:36:44  <planetmaker> like the devzone
19:36:48  <planetmaker> it only has the repo
19:36:50  <planetmaker> no checkout
19:36:59  <planetmaker> (usually that is)
19:37:23  <Ammler> hg cat ...
19:37:34  <Alberth> planetmaker: I tested the with pushing to a local 'upstream' repo. Then I was stopped trying to commit bad things. With remote pushing, that does not seem to happen
19:38:52  <Alberth> planetmaker: may be that is the significant difference here, although I don't see how.
19:40:17  <Ammler> might it be possible because we run 2 hooks on the server?
19:40:24  <planetmaker> hm. I have to admit I didn't play much with commit hooks yet.
19:41:15  <planetmaker> but in principle there should be no difference. It's after all what a _distributed_ vcs is about
19:42:45  <Alberth> euhm, for push checking you use a push hook, not a commit hook, of course :)
19:43:19  <Alberth> my primary reason was to have a local check of my commits, as I tend to mess them up.
19:43:30  <Ammler> pretxnchangegroup.check = python:/home/ottdc/hg-repos/misc/mercurial/hooks/
19:44:14  <Ammler> changegroup.uprepos = /home/ottdc/compiler/ &
19:45:15  <Alberth> I think you can have several hooks active
19:45:39  <Alberth> why a "&" ?
19:45:58  <Alberth> seems weird
19:46:25  <Ammler> so you don't need to wait until redmine fetched the push
19:46:54  <Alberth> hmm yeah, makes sense.
19:47:11  <Hirundo> argh... rejects when updating mqueue
19:51:18  <Hirundo> 'ScriptError' -> 'generic.ScriptError' fixes them all, though :)
19:52:04  <Alberth> sorry
19:53:29  <Alberth> I have another one in (merge of builtin_functions/py)
19:55:14  <Hirundo> You want to merge those two files?
19:55:47  <Alberth> they are a cyclic import, merging is the simplest solution
19:56:04  <Alberth> and builtin_functions is very small
19:57:29  <Alberth> do you have a problem with merging those files?
19:59:19  <Hirundo> merging now, only to split it off again later might be a waste, I don't know how many functions will be added there though
20:01:06  <Alberth> well, moving the import to the end does not help at all
20:01:25  <Hirundo> why not?
20:03:55  <Hirundo> I got the suggestion from here, btw:
20:03:56  <Webster> Title: Circular dependency in Python - Stack Overflow (at
20:06:43  <Alberth> if you apply the import * removal patch, while not merging, you will notice it fails to import
20:07:42  <Alberth> the builtin_* funcs seem to use Binop and other expression classes, and reduce_expr() seems to use the builtin_funcs.
20:08:28  <Alberth> one solution would be to move the reduce_expr() to a new file, but that is in contradiction with your desire to merge reduce() into those classes
20:10:02  <Alberth> the builtin_* funcs seem so close to the expression nodes, that keeping them together would be best imho
20:10:26  <Hirundo> If there is no easy other solution, then merging may be the best
20:11:12  * Hirundo wonders what OpenTTD would look like without circular dependencies
20:12:40  <Alberth> very different :)
20:13:05  <Alberth> trouble with Python is that you don't have declarations, only definitions
20:14:51  <Alberth> for now, is about 1/3 or, so growth is still allowed :)
20:18:53  <Hirundo> could use some pruning as well
20:20:53  <Alberth> it does
20:25:41  <Alberth> let's merge it now, and worry about a good split up later. A lot of the code is going to change heavily any way, so perhaps the issue gets resolved by itself in time
20:26:31  <Hirundo> ^^ refactoring should at this stage be done only when it is required for proper implementation of new features
20:27:19  <Hirundo> NML now fails for me to run, with import-related errors in Is that to be expected?
20:28:21  <Alberth> hmm, it works for me
20:28:48  <Alberth> parser,py had an import line added I think
20:29:39  <planetmaker> I can confirm the failrue
20:29:42  <planetmaker> *failure
20:30:52  <Hirundo> Found it
20:30:57  <Hirundo> delete nml.pyc
20:31:16  <planetmaker> ImportError: cannot import name generic
20:31:45  <Alberth> ah, yes, I had to do that too
20:32:19  <Alberth> the joy of using the same name for the package and for a module in that package :)
20:32:23  <planetmaker> yes, that helps
20:32:50  <planetmaker> :-)
20:34:02  <Alberth> another merge
20:34:11  <Alberth> please remove :)
20:34:17  <Brot6> NFO Meta Language - Revision 251: Codechange: Merge into expression. (Alberth) @
20:34:36  <Alberth> although it does not matter in this case, since the module is not referred any more
20:41:20  <Brot6> NFO Meta Language - Revision 252: Change: Consistency in constant naming (planetmaker) @
20:45:06  <Brot6> OpenGFX+ - Revision 24: Change: Climate-specific constants changed in NML (planetmaker) @
20:46:56  *** welshdragon has quit IRC
20:51:57  <Alberth> planetmaker: you missed the reference.html file
20:52:09  <planetmaker> damn. yes
20:54:21  <Alberth> good night
20:56:26  <Brot6> NFO Meta Language - Revision 253: Fix (r252): Also document what you change (planetmaker) @
20:56:34  <planetmaker> good night, Alberth
20:57:03  *** Alberth has left #openttdcoop.devzone
21:08:12  *** Seberoth has quit IRC
21:09:50  <planetmaker> do we want something like
21:09:51  *** Seberoth has joined #openttdcoop.devzone
21:09:52  <planetmaker> 	'INDUSTRYTYPE_COAL_MINE'         : 00,
21:09:54  <planetmaker> 	'INDUSTRYTYPE_POWER_PLANT'       : 01,
21:09:56  <planetmaker> defined?
21:10:05  *** frosch123 has quit IRC
21:10:07  <planetmaker> (the default IDs)?
21:13:32  <Hirundo> Why not?
21:14:37  <Hirundo> I've got the string -> Identifier change done to the extent that my test code compiles
21:14:56  *** welshdragon has joined #openttdcoop.devzone
21:16:30  <planetmaker> 	'INDUSTRYTYPE_TROPIC_BANK'           : 0x10,
21:16:31  <planetmaker> 	'INDUSTRYTYPE_ARCTIC_BANK'           : 0x10,
21:16:33  <planetmaker> ^^ would that be confusing?
21:16:52  <planetmaker> and should it be only one identifier?
21:17:06  <planetmaker> I tend to say yes. And no ;-)
21:17:13  <planetmaker> probably yes
21:17:32  <planetmaker> what does the string->Identifier change imply / provide?
21:21:08  <Hirundo> A wrapper around IDs (currently unicode strings), just like ConstantNumeric around numbers
21:23:14  <Hirundo> Yexo: Shall I commit it, knowing almost certainly that I'll break stuff, or shall I do some more testing first?
22:04:38  <Brot6> test: abort: push creates new remote heads!
22:06:48  <Brot6> feed devactivity had 12 updates, showing the latest 10
22:06:48  <Brot6> NFO Meta Language - Revision 247:8c5caaa73c31: Codechange: Rename to to prevent im... (Alberth) @
22:06:48  <Brot6> FIRS Industry Replacement Set - Revision 896:03613847e207: Fix: Aluminium Plant would have failed... (andythenorth) @
22:06:49  <Brot6> test - Revision 40:8b8cb7d762f4: 13 (Ammler) @
22:06:52  <Brot6> NFO Meta Language - Revision 248:2286c6274c40: Codechange: Import 'generic' module explicitly. (Alberth) @
22:06:56  <Brot6> NFO Meta Language - Revision 249:a0efd2d19857: Fix: Import fix, boolean literal fix. (Alberth) @
22:07:00  <Brot6> NFO Meta Language - Revision 250:733dc73b674a: Add: Added 'sounds' regression test case. (Alberth) @
22:07:04  <Brot6> NFO Meta Language - Revision 251:fcc180327507: Codechange: Merge into expression. (Alberth) @
22:07:08  <Brot6> NFO Meta Language - Revision 252:2d8b5164d0c2: Change: Consistency in constant naming (planetmaker) @
22:07:12  <Brot6> OpenGFX+ - Revision 24:065504984771: Change: Climate-specific constants changed in NML (planetmaker) @
22:07:16  <Brot6> NFO Meta Language - Revision 253:2f9377e184f3: Fix (r252): Also document what you change (planetmaker) @
22:07:20  <Ammler> hmm
22:08:43  <planetmaker> item(FEAT_INDUSTRIES, temperatebank, INDUSTRYTYPE_TEMPERATE_BANK) { <-- Hirundo, why can't I use a numeric constant there as INDUSTRYTYPE_TEMPERATE_BANK?
22:08:57  <planetmaker> or is it something else I miss? For example your patch?
22:09:48  <Brot6> test - Revision 41:fd2f46a54564: 14 (Ammler) @
22:11:51  <Brot6> NFO Meta Language - Revision 254:b883077847b0: Add: Default industry IDs as constants (planetmaker) @
22:17:00  <Brot6> test - Revision 42:ac7aca259a79: 14.2 (Ammler) @
22:17:01  <Brot6> test - Revision 43:ed72d8bd6ab3: 14.2 (Ammler) @
22:17:18  <Ammler> ok, should work now
22:17:22  <Ammler> (I hope :-P
22:18:34  <planetmaker> you can check now ;-)
22:19:07  <Brot6> NFO Meta Language - Patch #954: NFO lexer and basic parser (Alberth) @
22:19:07  <Brot6> NFO Meta Language - Revision 247:8c5caaa73c31: Codechange: Rename to to prevent im... (Alberth) @
22:19:07  <Brot6> OpenGFX+ - Revision 25:bf570f9eec24: Change: [Makefile] Update to r99 (planetmaker) @
22:19:07  <Brot6> OpenGFX+ - Revision 26:6384afdb49e3: Feature: Allow banks, power plants and water towers to close... (planetmaker) @
22:25:36  <Brot6> 32bpp-ez-patches - Revision 40:1f21ba1bc750: Cleanup: correct some whitespace (svn 19894) (GeekToo) @
22:26:19  <Hirundo> planetmaker: Consider it a bug
22:27:15  <planetmaker> consider it reported :-P
22:27:49  *** ODM has quit IRC
22:33:38  <Ammler> what's a bug?
22:34:02  <Ammler> ah, you don't speak about my change
22:38:10  *** DJNekkid has quit IRC
22:50:49  <Brot6> OpenGFX+ - Revision 27:9c939795bbfa: Doc: Add proper credits to the readme (planetmaker) @
22:58:53  <Yexo> Hirundo: please commit it when it passes all your own tests
23:21:19  <Brot6> NFO Meta Language - Revision 255:63bf0043cc8c: DevZone: if the build fails, also report it as fai... (Ammler) @
23:28:25  <PeterT> Ammler: you guys update to 0.090? ;-)
23:45:22  <Ammler> PeterT: why?
23:45:39  <PeterT> just wondering what you thought of the new webadmin :-)
23:55:39  <Ammler> last time, I had to update
23:55:48  <Ammler> currently I am fine

Powered by YARRSTE version: svn-trunk