Log for #openttdcoop.devzone on 31st October 2010:
Times are UTC Toggle Colours
06:20:42  *** Webster has joined #openttdcoop.devzone
06:20:51  *** andythenorth_ has joined #openttdcoop.devzone
07:19:08  *** ODM has joined #openttdcoop.devzone
07:28:46  *** andythenorth_ has quit IRC
07:43:33  *** Alberth has joined #openttdcoop.devzone
08:09:59  <Brot6> NewGRF Meta Language - Revision 966:29f25ad3e758: Codechange: Aligning values of the property dic... (Alberth) @
08:16:23  *** andythenorth_ has joined #openttdcoop.devzone
08:34:42  <Brot6> NewGRF Meta Language - Revision 967:2b11206b0f14: Codechange: Adding some main markers in the file. (Alberth) @
08:34:42  <Brot6> NewGRF Meta Language - Revision 968:74dc7040a4b5: Doc: Created a placeholder for properties expla... (Alberth) @
09:09:10  <Brot6> NewGRF Meta Language - Revision 956:5ea5a898d5bc: Change: Register all sprite groups and sets dur... (Hirundo) @
09:09:10  <Brot6> NFORenum - Feature #1746 (New): improve warning text (George) @
09:13:52  <Alberth> you're a little late with r956, Brot6
09:13:52  <Brot6> hey Alberth
09:14:59  <Alberth> Brot6: don't you ever get bored of repeating those stupid messages to a channel that nobody reads?
09:24:32  <Ammler> :'-(
09:27:38  <Alberth> he's not so fast :)
09:28:12  <Alberth> good morning
09:29:19  <Ammler> I have no idea, how you manged to announce r956 again
09:31:01  <Alberth> the aftermath of the temporal shift that happened early this morning.
09:42:31  <andythenorth_> if there was no brot, I'd code less
09:44:31  *** thgergo has joined #openttdcoop.devzone
09:45:43  <Alberth> making bot that sends random messages about commits from your repos is doable :p
10:35:12  <Alberth> what does 'unit_conversion' do in the action 0 properties?
10:40:10  *** KenjiE20 has joined #openttdcoop.devzone
10:52:52  <Brot6> NewGRF Meta Language - Revision 957:e1b217f0a156: Codechange: Make TRANSPARANT a normal constant. (Hirundo) @
10:52:52  <Brot6> NewGRF Meta Language - Revision 955:6f12b0c7592b: Codechange: Generalize the register / resolve s... (Hirundo) @
10:52:52  <Brot6> NFORenum - Feature #1746: improve warning text (Rubidium) @
11:01:07  <Brot6> Unable to connect to execution expired
11:16:18  <Hirundo> Alberth: 'unit_conversion' is used to handle the difference between e.g. RV and train speed units
11:17:19  <Alberth> I have a conversion of weight for a cargo:     'weight'                    : {'num' : 0x0F, 'size' : 1, 'unit_type' : 'weight'}, # @TODO: Needs a unit_conversion 1/16 or 16 or 62.5 or 1/62.5 depending on the meaning of the 'unit_conversion'.
11:17:52  <Alberth> ie 16 means 1 ton
11:22:16  <Alberth> but never mind, I'll wait until someone adds some documentation about it
11:23:23  <Hirundo> That's the cargo property?
11:25:02  * Alberth nods
11:25:17  <Alberth>
11:28:01  <Hirundo> There is a certain reference unit (m/s, hp, ton) defined in
11:29:09  <Hirundo> The value of unit_conversion is equal to 1 such reference unit in nfo terms
11:29:24  <Hirundo> In this case, the value is 16
11:30:41  <Alberth> thank you for your time
11:34:22  <Alberth> do we use sets somewhere?  013_train_callback.nfo and grf have changed
11:36:01  <Alberth> r968, with some cargo property changes, not related to trains
11:38:07  <Alberth> pushed the cargo props
11:38:51  <Brot6> NewGRF Meta Language - Revision 963:44c8e3409574: Feature #1695: Make it possible to combine spri... (Hirundo) @
11:38:51  <Brot6> NewGRF Meta Language - Revision 960:07f0c57e0934: Add: Store references to all used sprite sets i... (Hirundo) @
11:38:51  <Brot6> NewGRF Meta Language - Revision 969:62ebd6624970: Add: Cargo properties (mostly). (Alberth) @
11:43:16  <Brot6> NewGRF Meta Language - Revision 957:e1b217f0a156: Codechange: Make TRANSPARANT a normal constant. (Hirundo) @
11:43:17  <Brot6> NewGRF Meta Language - Feature Request #1747 (New): cargo property changes (Alberth) @
11:45:50  <Alberth> Brot is really lost today, I have seen r957 about 40 minutes ago as well
11:48:28  * Hirundo sees the light
11:48:51  <Hirundo> Indeed my recent sprite set code uses sets
11:49:33  <Hirundo> I was confused already yesterday, when regression test seemed to break in some strange manner
11:59:24  <Hirundo> Alberth: Sorting the list by should fix it, right?
11:59:49  <Alberth> if the name is unique, yes
12:04:04  <Brot6> NewGRF Meta Language - Revision 963:44c8e3409574: Feature #1695: Make it possible to combine spri... (Hirundo) @
12:04:04  <Brot6> NewGRF Meta Language - Revision 960:07f0c57e0934: Add: Store references to all used sprite sets i... (Hirundo) @
12:04:04  <Brot6> NewGRF Meta Language - Revision 959:340d0ed2b83d: Add: Store references to all referring sprite g... (Hirundo) @
12:04:06  <Brot6> NewGRF Meta Language - Revision 970:a7430f7050f1: Fix: Don't rely on unordered sets to provide a ... (Hirundo) @
12:05:19  <Hirundo> Brot has definitely seen better days....
12:07:04  <Hirundo> Alberth: Thanks for your pointer about sets
12:07:41  <Alberth> sort of standard problem when output changes order seemingly randomly :)
12:08:47  <Alberth> problem is that some python dev considered it useful to use id() as compare criterium, so you sort on memory address
12:10:03  <Alberth> It also makes that     max(2.5, "hai")    does not give an error :(
12:43:47  <Alberth> andythenorth_: quaryy and builders yard are conviently together to deliver to/from a cement plant
12:44:11  <andythenorth_> Alberth: screenshot?
12:45:12  <Alberth>
12:45:49  <Alberth> clay pit is also nicely positioned, but probably by accident
12:48:39  <Alberth> hmm, elsewhere I have a builders yard and clay pit close together
14:28:17  <Alberth> a Builders yard never stops threatening to close down in 30 months?
14:41:18  <Hirundo> Alberth: How urgent is your need for *someone* to document action0properties?
14:42:23  <andythenorth_> Alberth: looks like a bug :P
14:43:23  *** andythenorth_ has quit IRC
14:48:52  *** pm has joined #openttdcoop.devzone
14:49:06  *** pm is now known as planetmaker
14:50:17  <planetmaker> good evening
14:57:30  <Hirundo> hello pm
15:11:05  *** thgergo has quit IRC
15:29:48  <Alberth> hello planetmaker
15:34:51  *** Levi has joined #openttdcoop.devzone
15:35:12  *** thgergo has joined #openttdcoop.devzone
15:40:58  <Alberth> Hirundo: look at it this way, due to not documenting (in this case the properties list data structure) I am guessing what things mean, searching in code, asking questions, waiting for answers, you helping me, spending time on different things than what you work on.
15:40:58  <Alberth> I asked twice already now, and people cannot be bothered apparently. So should I be bothered to waste my time looking for clues in random pieces of code?  I think not.
15:41:18  <Alberth> (sorry, I am a bit pissed. I know you mean well.)
16:31:28  <Ammler> Feature: [Makefile] build from source bundle and make rpm <-- what does "make rpm"?
16:32:06  <Ammler> in opengfx changelog 0.3.1
16:34:16  <planetmaker> is that a question for me?
16:34:23  <Ammler> I assume so
16:34:36  * planetmaker feels like reading that commit message the first time
16:34:47  <Ammler> I read it from the changelog
16:34:51  <Ammler> the top line
16:34:57  <planetmaker> :-D
16:35:15  <Ammler> but there is no target rpm
16:36:30  <planetmaker> I guess it's ambiguously re-written r488
16:36:40  <Alberth> rpm is package format of red-hat
16:36:42  <planetmaker> Zusammenfassung: Cleanup: those specs are outdated, find up2date specs in .devzone
16:36:49  <planetmaker> and of suse :-)
16:37:11  <Alberth> they just 'stole' it :p
16:37:56  <Alberth> which is probably better than inventing a new format, or using plain tar.gz :)
16:38:04  <planetmaker> :-) quite so
16:56:32  <Ammler> Alberth: I don't think "they stole" it
16:56:39  <Ammler> I would assume, it was opensource
16:56:49  <Alberth> it was, and still is :)
16:56:52  <Ammler> or do you have any sources?
16:57:14  <Ammler> Rubidium: catcodec installs the docs to /usr/share/doc/readme.txt
16:57:57  <Ammler> (using make install DESTDIR=%{buildroot} prefix=%{_prefix})
16:59:01  <Ammler> should be /usr/share/doc/catcodec/*
16:59:34  <Ammler> as it is with grfcodec
17:01:56  <Ammler> docdir         ?= $(datarootdir)/doc/$(PACKAGE_NAME) <-- looks like PACKAGE_NAME is empty
17:06:24  <Ammler> does catcodec have a tracker?
17:09:25  <Rubidium> Ammler: not really
17:09:40  <Ammler> well, then use this ^ as bugreport :-P
17:10:21  <Ammler> my workaround is: make install DESTDIR=%{buildroot} prefix=%{_prefix} PACKAGE_NAME=catcodec
17:10:37  <Brot6> nml: update from r953 to r970 done -
17:11:10  <planetmaker> wow. Quite some progress
17:11:44  <Ammler> oh, version 1.0.2 should be mentioned maybe
17:13:09  <Rubidium> Ammler: in what way should 1.0.2 be mentioned?
17:14:52  <Rubidium> Ammler: is enough?
17:18:47  <Brot6> Following repos didn't need a nightlies update: 2cctrainset (r635), 32bpp-extra (r39), ai-admiralai (r71), airportsplus (r64), basecosts (r22), belarusiantowns (r7), comic-houses (r71), firs (r1483), fish (r415), frenchtowns (r4), grfcodec (r785), heqs (r479), indonesiantowns (r37), manindu (r5), metrotrackset (r56), newgrf_makefile (r220), nml (r970), nutracks (r117), ogfx-trains (r86), ogfx-trees (r41), opengfx (r554), openmsx (r97),
17:18:47  <Brot6> opensfx (r97), smts (r19), snowlinemod (r45), swedishrails (r187), swisstowns (r20), transrapidtrackset (r15), ttdviewer (r26), ttrs (r23), worldairlinersset (r667)
17:19:03  <Ammler> Rubidium: patch works
17:20:19  <Brot6> airportsplus: compile of r64 still failed (#1741) -
17:21:43  <Brot6> indonesiantowns: compile of r37 still failed (#1704) -
17:22:06  <Brot6> manindu: compile of r5 still failed (#1742) -
17:23:01  <Ammler> at least the town names failure isn't needed
17:23:22  <Ammler> does nml have a option, so it is a bit less pedantic?
17:23:41  <Brot6> swedishrails: rebuild of r187 done (Diffsize: 6) (DiffDiffsize: 8) -
17:24:02  <Brot6> swisstowns: compile of r20 still failed (#1705) -
17:24:12  <Brot6> Following repos rebuilds successful without any difference to earlier nightlies builds: belarusiantowns (3 errors) (Diffsize: 21), frenchtowns (4 errors) (Diffsize: 9), ogfx-trains (1 errors) (Diffsize: 12)
17:28:50  <Ammler> It is also not very helpful as regression test if you make those fail just because of new introduced non-required vars
17:29:45  <Ammler> and why belarusain and french doesn't fail is also quite confusing
17:33:05  <Ammler> ah, I see, they don't have the "grf check" :-)
18:10:23  <Rubidium> Ammler: tonight a new source tarball will be made for catcodec
18:10:38  <Ammler> okidoki :-)
18:11:17  <Ammler> suse will soon also make independent packages and build from source
18:23:31  *** avdg has joined #openttdcoop.devzone
18:35:07  *** V453000 is now known as Guest1374
18:39:11  *** michi_cc has quit IRC
18:47:48  *** michi_cc has joined #openttdcoop.devzone
18:58:50  <Hirundo> Ammler: wrt NML backwards compatibility, IMO we should, when possible, (as a transition) support both 'old' and 'new' behaviour and raise a warning if 'old' stuff is encountered
18:59:43  <Hirundo> Alberth: Your comments included '@todo: Define meaning of C{None}' <- what do you mean by that?
19:00:34  <Alberth> properties[x] may contain None. What does that mean?
19:00:49  <Brot6> NewGRF Meta Language - Revision 971:86b8249db08a: Doc: Action 0 property features. (Hirundo) @
19:00:49  <Brot6> NewGRF Meta Language - Revision 972:a15333b04fa7: Doc: The meaning of the action0 property charac... (Hirundo) @
19:01:19  *** andythenorth_ has joined #openttdcoop.devzone
19:01:34  <Alberth> hello andy
19:03:45  <Hirundo> None is just a placeholder, since sparse lists are not possible
19:03:58  <Hirundo> It currently triggers an internal error, though, which is Bad (tm)
19:04:06  * Hirundo fixes
19:04:52  <Alberth> I appreciate what you are doing, but it is a more general problem, I am afraid
19:05:25  <andythenorth_> evening
19:06:50  <Alberth> andythenorth_: I have a similar case with forest, sawmill, and a Lumber yard
19:08:05  <andythenorth_> is this good or bad? :)
19:08:52  <Alberth> it is effective use of tracks :)
19:09:58  <Brot6> 32bpp-ez-patches: update from r21060 to r21068 done -
19:10:03  * andythenorth_ ponders revamping FIRS cargos
19:11:12  <planetmaker> hi andythenorth_
19:11:30  <andythenorth_> hi
19:11:35  <planetmaker> did mb talk to you long enough that you start fulfilling his cargo wishes?
19:11:56  <andythenorth_> well it's not much good if they're ill-conceived
19:12:16  <andythenorth_> the problem is that ECS has already defined everything correctly, and FIRS isn't following that
19:12:31  <planetmaker> what makes you think that ECS is correct?
19:13:07  <andythenorth_> perhaps it's not
19:13:10  <Alberth>
19:13:11  <planetmaker> why is ECS more correct than the default industries?
19:13:14  <planetmaker> It breaks them
19:13:20  <planetmaker> as well as cargos
19:13:39  <andythenorth_> ECS defines real cargos
19:13:56  <andythenorth_> which means that accurate vehicle set graphic support can be provided
19:14:18  <andythenorth_> Alberth: that looks very convenient
19:14:24  <planetmaker> only because some people lack imagination and phantasy to interpret 'supplies' by different cargos doesn't mean it's bad
19:14:32  <andythenorth_> Alberth: there's no code to help with that, you got lucky :)
19:14:51  <planetmaker> supplies allows to have vehicles transporting it, showing a vehicle part or a container or whatever
19:14:55  <planetmaker> it's not wrong by any means
19:15:07  <andythenorth_> but it is wrong to transport glass in a bulk hopper :P
19:15:16  <planetmaker> it's a lack of phantasy and the fear for flexibility which drives people to reject 'supplies' as a good idea
19:15:33  <planetmaker> andythenorth_: that's a problem of the vehicle set. Not of the industry set
19:15:40  <andythenorth_> building materials should be: glass, timber, cement, bricks, and metalwork
19:15:48  <planetmaker> and transporting glass in a hopper _can_ make sense
19:16:02  <andythenorth_> which means it needs 5 cargo slots not 1.  Which cargos shall we eliminate to fit the extra ones in?
19:16:03  <Alberth> straight to the recycling center :p
19:16:16  <planetmaker> I don't follow you, andythenorth_
19:16:21  <andythenorth_> and we now need 2 builders yards, because there are only three accepted cargo slots
19:16:23  <andythenorth_> planetmaker: best not
19:16:32  <andythenorth_> do something useful instead :P
19:16:40  <planetmaker> ok
19:16:55  * andythenorth_ is grumbling
19:17:32  <andythenorth_> I've learnt from GUI design that you don't get to be correct with GUI problems.  You only get to be more or less wrong
19:17:39  <andythenorth_> gameplay design is similar
19:17:53  <Alberth> probably more so
19:17:54  <planetmaker> my fear is: you'll end up in just another boring ECS clone.
19:18:17  <planetmaker> now FIRS is unique and a different concept. And that's great
19:18:32  <andythenorth_> yes, but if I clone ECS, the vehicle support will be beautiful
19:18:33  <planetmaker> not everyone has to like it. Especially not the core designer(s) of ECS
19:18:35  <andythenorth_> a work of art :P
19:18:52  <andythenorth_> the problem is probably that there's no design :P
19:19:00  <Brot6> clientpatches: update from r21060 to r21068 done -
19:19:12  <planetmaker> partially yes
19:19:18  <Alberth> won't you get vehicle support, given some time?
19:19:26  <andythenorth_> except there is a design, because I designed it :P
19:19:39  <planetmaker> besides the vehicle support for FIRS - you'll get it
19:19:47  * andythenorth_ has had an inappropriate amount to drink
19:19:52  <andythenorth_> i.e. not enough
19:19:56  <planetmaker> As long as you don't re-define the cargo classes (too much)
19:20:00  <Hirundo> my 2cc: I think the concept of generic cargos (Metal, supplies, building materials) is great, don't change away from that
19:20:01  <planetmaker> of existing cargos
19:20:03  <andythenorth_> I'm not going to touch them
19:20:16  <andythenorth_> they are done for at least a few versions
19:20:21  <planetmaker> I fully agree, Hirundo
19:20:23  <Brot6> serverpatches: compile of r21068 still failed (#1658) -
19:20:28  <andythenorth_> I think FIRS 1.0 will ship with these cargos
19:20:38  <andythenorth_> any changes would be in 2.0 or economies
19:20:49  <Hirundo> <Alberth>	I appreciate what you are doing, but it is a more general problem, I am afraid <- You mean, that currently no checks on feature validity are done?
19:21:07  * andythenorth_ unponders revamping FIRS cargos :D
19:21:51  <Alberth> good idea, stop changing it for a while, and see what happens
19:22:07  <andythenorth_> one thing that happens is mb complains it's all wrong :P
19:22:29  <planetmaker> yes. Unless you write the grf for him and him alone
19:22:55  <planetmaker> and make it virtually the ecs which george didn't write him
19:22:59  <andythenorth_> he can write his own
19:23:28  <Hirundo> MB's ECS is one of his many vapourwarez...
19:28:53  <Alberth> Hirundo: no, program crashes are not a problem neither are bugs or feature requests. It it this continuously having to wait for you or yexo to give me the information that should be simply available in the code documentation. It is not that documenting code is anything new. I get nowhere this way.
19:29:50  <Hirundo> My current change includes documentation for what 'None' constitutes
19:30:39  <Hirundo> Alberth: I fully understand your problem
19:30:42  <Alberth> Appreciated. I know it is not your fault, but I have had enough of it.
19:31:19  <Hirundo> I'm trying to stay on a thin line between writing code and writing documentation without overdoing either
19:31:49  <Hirundo> My main problem is that I can't see what's 'obvious' c.q. self-documenting and what isn't
19:31:56  <Alberth> errorring at the documentation side usually doesn't hurt :)
19:32:49  <Hirundo> If I want to document all code I'd have to spend months. I don't have that much time before we want to do a release
19:33:21  <Alberth> that's why you don't do that afterwards, you do it at the same time
19:33:26  <Hirundo> Me and Yexo having started this, neither of us having much (if any) python experience doesn't help either
19:34:01  <Hirundo> I'm trying to document all new code I write, but that still leaves a pretty large, often messy, codebase
19:34:24  <Alberth> imho a line describing what a function does, has very little to do with any programming language
19:34:44  <Alberth> and even documenting what goes in and what comes out
19:34:59  <andythenorth_> no docstrings?
19:35:09  <Alberth> nope
19:35:20  <Hirundo> At times, our code isn't as self-documenting as python can be
19:35:37  <Alberth> self-documenting code does not exist, it is a myth
19:35:56  <andythenorth_> docstrings are middlingly useful
19:36:07  <andythenorth_> quite a lot of them I've seen consist of ""
19:36:21  <andythenorth_> or more accurately: """ """
19:36:24  <Alberth> if documentation is that easy, spending the time to write the text is also short
19:36:29  <Hirundo> a docstring is only as good as the one who wrote it
19:37:35  <Alberth> sure, and it is incomplete, unclear, buggy, and some other bad properties. However, it beats nothing by any standard
19:37:36  <Hirundo> There's no point arguing whether you're right, since you are
19:37:55  <Alberth> I completely agree :)
19:44:59  <Brot6> NewGRF Meta Language - Revision 958:f16e28197259: Change: Resolve sprite groups during pre-proces... (Hirundo) @
19:44:59  <Brot6> NewGRF Meta Language - Revision 962:35ed440dc196: Codechange: Don't push the feature of the sprit... (Hirundo) @
19:44:59  <Brot6> NewGRF Meta Language - Revision 973:611ecbc235e9: Fix: Correctly handle invalid features and feat... (Hirundo) @
20:00:01  *** avdg has quit IRC
20:05:16  *** Alberth has left #openttdcoop.devzone
20:12:16  *** Lakie has joined #openttdcoop.devzone
20:47:25  *** andythenorth_ has quit IRC
20:47:31  *** andythenorth_ has joined #openttdcoop.devzone
20:58:52  *** andythenorth_ has quit IRC
21:07:21  *** Yexo has joined #openttdcoop.devzone
21:13:44  *** andythenorth_ has joined #openttdcoop.devzone
21:59:25  *** ODM has quit IRC
22:48:20  *** andythenorth_ has quit IRC
22:54:57  *** andythenorth_ has joined #openttdcoop.devzone
23:00:03  *** andythenorth_ has quit IRC
23:01:04  <Brot6> NewGRF Meta Language - Revision 961:8779dd9ba998: Codechange: Don't use a dictionary to map sprit... (Hirundo) @
23:01:04  <Brot6> NewGRF Meta Language - Revision 964:95c0c8e90208: Fix: Regression test. (Hirundo) @
23:16:00  *** Lakie` has joined #openttdcoop.devzone
23:23:05  *** Lakie has quit IRC
23:44:35  *** planetmaker has quit IRC

Powered by YARRSTE version: svn-trunk