Log for #openttdcoop.devzone on 18th July 2011:
Times are UTC Toggle Colours
00:23:57  *** LordAro has quit IRC
05:37:56  <Brot6> FIRS Industry Replacement Set - Bug #2864: Industries have wrong accept/produce cargos (andythenorth) @
06:01:35  <Brot6> FIRS Industry Replacement Set - Bug #2864: Industries have wrong accept/produce cargos (planetmaker) @
06:03:39  *** andythenorth has joined #openttdcoop.devzone
06:03:53  <Brot6> FIRS Industry Replacement Set - Bug #2864: Industries have wrong accept/produce cargos (andythenorth) @
06:07:31  * planetmaker has an idea and tests
06:08:28  * andythenorth had an idea, but was wrong :|
06:08:45  <andythenorth> some industries still have their cargos set numerically - but that's not the issue
06:10:01  <planetmaker> andythenorth: the cargo definitions in action0 industries: you must use the cargo label LVST etc instead of cargo_livestock
06:10:14  <planetmaker> that did the trick for sheep farm
06:10:26  <planetmaker> I know it was me who changed that :-)
06:10:38  <planetmaker> from numeric to the cargo_XXX identifier...
06:11:16  <planetmaker> I guess... it should be an easy fix
06:12:11  <andythenorth> planetmaker: works for me
06:12:36  <andythenorth> I guess I can adjust those
06:12:44  <andythenorth> and split industries at the same time
06:13:38  <andythenorth> so what is the term for these 'cargo_livestock' in nml?  Defines?  IDs?
06:13:41  <andythenorth> Constants?
06:15:24  <Brot6> FIRS Industry Replacement Set - Bug #2864: Industries have wrong accept/produce cargos (andythenorth) @
06:20:17  * planetmaker tests fixed build
06:21:02  <andythenorth> hmm
06:21:09  <andythenorth> we are about to have merge conflict in that case :P
06:22:57  <planetmaker> hm, did steel mill always accept sand?!
06:23:10  <andythenorth> nope :)
06:24:30  <planetmaker> and arable farm clay?
06:25:09  <planetmaker> hardly.
06:26:15  <andythenorth> nope
06:26:52  <andythenorth> planetmaker: you're fixing?  I started splitting-files-and-fixing....which will merge badly :D
06:27:39  <planetmaker> a simple global replace via sed
06:28:45  <planetmaker> but... where's steel mill defined?
06:29:04  <andythenorth> in god_object?
06:31:08  <planetmaker> doesn't seem so. where there?
06:31:23  <planetmaker> or it's not named steel mill
06:31:29  <planetmaker> or *ste*
06:33:08  <andythenorth> hmm
06:33:34  <planetmaker> however... I suggest I push this simple search&replace
06:33:43  <planetmaker> enough manual work, like for steel mill remains ;-)
06:34:07  <andythenorth> planetmaker: might be item_FEAT_INDUSTRIES_7
06:34:19  <andythenorth> anyway, push ;)
06:34:28  <andythenorth> I'll revert my local changes
06:36:57  <planetmaker> hm. still leaves me puzzled. Also with arable farm
06:37:33  <andythenorth> arable farm is fixed for me if I use cargo labels
06:39:44  <planetmaker> maybe I did something wrong
06:39:49  <planetmaker> anyway, pushed
06:40:09  <Brot6> FIRS Industry Replacement Set - Revision 2174:6ee0f7bf528c: Fix: Cargo references in industry and... (planetmaker) @
06:40:29  <Terkhen> how is FIRS going? today I have to tidy my home a bit but after that I should be able to start helping :)
06:40:43  <andythenorth> it had a bug :P
06:41:02  <andythenorth> seems to me best thing is to split industry files out
06:41:06  <Terkhen> caused by the nfo->nml conversion or by the fixes to nml code? :P
06:41:17  <planetmaker> hehe. This petrol pump is nice. Good for their revenue, as the single petrol pump in the town they're at the most distant piece of road in the smallest suburb :-P
06:41:26  <andythenorth> Terkhen failure of constants I think
06:41:36  <Terkhen> how are you doing it? moving action 0 only or action 0 along with all switches?
06:41:39  <planetmaker> Terkhen: probably some of my changes
06:41:56  <planetmaker> replacing numbers by wrong identifier's names
06:42:07  <planetmaker> cargos go via cargo label and not the cargo's identifier
06:42:11  <andythenorth> Terkhen: I'm moving everything for e.g. brewery to brewery.pnml
06:42:17  <andythenorth> we can sort the rest out later
06:43:06  <Terkhen> cool, so... 1) pick up some industry, 2) move the related item block to a file, 3) move all switch chained to it to the same file, 4) follow the style / format used in existing files
06:43:22  <planetmaker> yeah
06:43:40  <andythenorth> let me finish + commit some changes though
06:44:12  <andythenorth> this is one task not effective with multiple people doing it :P
06:44:12  <andythenorth> merging this sucks :P
06:44:22  <andythenorth> yet another reason to not have god_object
06:44:31  <Terkhen> don't worry, I won't start until this afternoon :)
06:44:40  <planetmaker> :-)
06:44:58  <planetmaker> you need a looong time to drink your morning cafe then, Terkhen ;-)
06:46:11  <Terkhen> no, I just have to clean up the house :P
06:47:08  <Terkhen> I have been neglecting it last week and yesterday we had a party so now you can barely move around here
06:48:18  <planetmaker> haha :-)
06:51:01  <planetmaker> a well-deserved party, though, I guess :-)
06:51:51  <andythenorth> only 36k lines left in god_object :P
06:51:59  <Brot6> FIRS Industry Replacement Set - Revision 2175:0a5b3c7c5768: Change: split more industries to sepa... (andythenorth) @
06:52:18  <Terkhen> yes :P
06:52:51  <andythenorth> planetmaker: I'm going to split more, unless you're planning any FIRS changes?
06:53:35  <andythenorth> hmm
06:53:42  <andythenorth> the spelling of 'diary' constant might cause some troubles :P
06:56:03  <planetmaker> andythenorth: in the next hours you won't see a commit from me. so go right ahead
06:56:10  <Brot6> FIRS Industry Replacement Set - Bug #2865 (Confirmed): Typo in 'diary' constant (andythenorth) @
06:56:39  <planetmaker> ^ in what way?
06:56:52  <andythenorth> should be 'dairy'
06:57:06  <planetmaker> the error report is not quite descriptive ;-)
06:57:07  <andythenorth> someone who types it correctly will spend time chasing a bug :)
06:57:14  <andythenorth> I'll fix the error later
06:57:32  <planetmaker> as it does not mention really what needs looking at, but only what the result should be
06:57:51  <planetmaker> quite the reverse of useful for searching ;-)
06:58:11  <planetmaker> oh... ai / ia ? :-P
06:58:24  <planetmaker> quite non-visible
07:00:41  <Brot6> FIRS Industry Replacement Set - Bug #2865 (Confirmed): Typo in 'diary' constant (andythenorth) @
07:05:11  <Brot6> FIRS Industry Replacement Set - Revision 2176:b6dd2394b204: Change: split more industries to pnml... (andythenorth) @
07:11:06  <Brot6> FIRS Industry Replacement Set - Revision 2177:2239f7bb1d6d: Change: split further industries to p... (andythenorth) @
07:29:07  *** ODM has joined #openttdcoop.devzone
08:02:17  *** ODM has quit IRC
08:26:46  *** andythenorth has quit IRC
09:46:51  *** andythenorth has joined #openttdcoop.devzone
10:07:29  *** ODM has joined #openttdcoop.devzone
10:48:28  *** andythenorth has quit IRC
11:57:58  <Brot6> Central European Train Set - Revision 46:4eb021f24441: add capacity and cargo class to tracking t... (Eddi) @
11:57:58  <Brot6> Central European Train Set - Revision 47:4620dd986c4e: very crude wagon scheme (Eddi) @
11:57:58  <Brot6> Central European Train Set - Revision 48:60efe113f80a: some formula changes resulting in more 0s (Eddi) @
12:17:06  <Brot6> Central European Train Set - Revision 49:7cb21febe1ee: some more or less random/unbalanced passen... (Eddi) @
12:22:18  <Brot6> test: update from r115 to r116 done -
13:32:37  *** FooBar has joined #openttdcoop.devzone
13:48:16  *** andythenorth has joined #openttdcoop.devzone
14:26:34  <Brot6> FIRS Industry Replacement Set - Revision 2178:a25c65bfcabd: Change: move further industries to pn... (andythenorth) @
14:32:16  <Brot6> FIRS Industry Replacement Set - Revision 2179:e23c27110162: Change: split further industries to p... (andythenorth) @
14:32:30  <andythenorth> god object is down to 12k lines from 54k :P
14:32:40  <planetmaker> :-)
14:32:49  <planetmaker> atheism rulez :-P
14:35:31  <Ammler> planetmaker: if you share sprites to different grfs, how does someone know a sprite he edits on ogfx is also used on ogfx-landscape?
14:35:55  <andythenorth> he doesn't
14:35:56  <Ammler> and probably break ogfx-landscape
14:36:04  <andythenorth> shared dependencies are a *bad* pattern
14:36:11  <andythenorth> for that reason
14:36:27  <andythenorth> it's safer to do manual reconciliation between them
14:36:42  <andythenorth> unless you've enforced true decoupling in your module - which you can't with pixels :P
14:36:53  <Ammler> well, you define the revision of it
14:37:00  <planetmaker> yes and no. It requires to not change the sprite layout within files
14:37:09  <Ammler> but it could be still a big mess, if you once update graphcis
14:37:24  <planetmaker> and... one might define the sprite templates which actually access the sprites also within that repo
14:37:24  <andythenorth> you would also maybe have a hard time diffing
14:37:31  <planetmaker> hm?
14:37:45  <andythenorth> diffing graphics is tricky
14:37:54  <planetmaker> yes. It sucks big time
14:38:05  <Ammler> well, if it just changes positions, you simply get white pixel errors
14:38:18  <planetmaker> Ammler, one must not move sprites. Sure enough
14:38:18  <andythenorth> and the buildbot tells you?
14:38:42  <Ammler> planetmaker: yes, but how do I know, which images I can "fix", which I shouldn't
14:38:52  <planetmaker> You can fix every image, can't you?
14:39:03  <planetmaker> Just don't move the sprite
14:39:12  <planetmaker> (or resize it )
14:39:16  <Ammler> well, move the sprites could be fixing :-)
14:39:23  <Ammler> like template it
14:39:36  <Ammler> s/fixing/cleanup/
14:39:42  <planetmaker> now I don't follow you
14:39:54  <planetmaker> you mean a cleanup operation will become somewhat impossible?
14:40:05  <Ammler> not impossible
14:40:28  <Ammler> but If I would change a sprite image, it could break another project
14:40:41  <Ammler> not break, but it could make it hard to update
14:41:17  <Ammler> so you should not just use the images, also the code
14:42:39  <Ammler> which would make sense, right?
14:43:14  <Ammler> so you rather set the whole ogfx as subrepo
14:43:18  <planetmaker> the sprite template would make sense within that project, too
14:44:32  <Ammler> then the question is why use subrepo, just make ogfx-source a dependency :-)
14:45:01  <Ammler> hmm, suprepo is good
14:45:05  <Ammler> to fix the revision
14:46:07  <Ammler> but that way, you don't have to think about source bundle
14:46:52  <planetmaker> well, the main idea to have a shared OpenGFX graphics repo is simple: OpenGFX and its NewGRF descendants should share graphics
14:47:04  <planetmaker> especially like landscape for ogfx+landscape and opengfx
14:47:15  <Ammler> yep, make ogfx a subrepo of ogfx-landscape
14:47:18  <planetmaker> or industries with ground sensitivity for opengfx+industries
14:47:27  <planetmaker> or vehicles with cargo / loading stage support
14:47:50  <planetmaker> thus it makes sense to use those graphics commonly in all opengfx projects
14:48:11  <Ammler> but there is no need to split ogfx in 2 repos
14:48:15  <planetmaker> which - in my interpretation - means to make the graphics a subrepo of each
14:48:47  *** andythenorth is now known as Guest2633
14:48:47  *** andythenorth has joined #openttdcoop.devzone
14:48:58  <Ammler> so basically ogfx is the graphics repo of the ogfx+ projects
14:49:25  *** andythenorth is now known as Guest2634
14:49:26  *** andythenorth has joined #openttdcoop.devzone
14:49:30  *** Guest2633 has quit IRC
14:49:38  *** Guest2634 has quit IRC
14:50:37  <planetmaker> how's that different then as I could then break OpenGFX just as well by modifying things for one of the NewGRFs
14:50:44  <Brot6> Central European Train Set - Revision 50:af06d11de8d4: modified callbacks for prospected vehicle ... (Eddi) @
14:50:44  <Brot6> Central European Train Set - Revision 51:4a3baf56c43d: add dummy vehicles to tracking table (Eddi) @
14:50:44  <Brot6> Central European Train Set - Revision 52:5d488a3f4a41: insert code for dummy vehicles (Eddi) @
14:50:49  <Ammler> how?
14:51:03  <Ammler> if you modify ogfx graphics, you would modify ogfx
14:51:25  <Ammler> you can't edit ogfx graphics without committing to ogfx
14:52:40  <planetmaker> err... ?
14:53:06  <Ammler> but if you have 2 repos for ogfx, I would not like to be the person, who does update the graphics repo
14:53:06  <planetmaker> thus you automatically break OpenGFX when modifying graphics for a OpenGFX+ project. As opposed to not breaking it when graphics are a separate repo
14:53:31  <Ammler> planetmaker: you might not break the current version
14:53:39  <Ammler> but you would break "a update"
14:53:48  <planetmaker> yes, I understand that
14:54:09  <Ammler> so you modify graphcis today and 2 months later you update ogfx and you have no clue anymore
14:54:17  <planetmaker> so you prefer breaking another project immediately over the other project choosing an update?
14:54:21  <Ammler> if ogfx is one repo, you would break ogfx today
14:54:30  <Ammler> planetmaker: yes
14:54:32  <planetmaker> yes, what's the advantage?
14:54:49  <Ammler> the one who updates can also fix ogfx
14:55:00  <Ammler> or do it right
14:55:18  <Ammler> well, it might be you anyway
14:55:35  <Ammler> so it is up to you, I just like to mention it
14:56:07  <Ammler> but it might also be easier for you to fix it when you change it not 2 months later, when the error arises on a update
14:57:05  <Ammler> the responsible person should be the one who changes the sprites, not the one who updates, imo
14:58:12  <Brot6> Nutracks - Bug #2862 (Rejected): "Fatal error: Attempt to use invalid ID" (oberhuemer) @
14:58:26  <Ammler> also the fix after a update could break the other project then :-P
14:58:40  <andythenorth> what / why do they share graphics?
14:59:11  <Ammler> andythenorth: for over/underlays mainly, I guess
15:00:01  <planetmaker> andythenorth, e.g. OpenGFX+ Industries. I might have the oil derriks from the base set. For OpenGFX+ Landscape and OpenGFX+ Industries I need the same graphics, but with other ground, for example
15:00:15  <planetmaker> Makes perfectly sense to not duplicate that, could all be in the same file
15:00:23  <planetmaker> just different layers
15:00:41  <Ammler> planetmaker: you see a disadvante on using ogfx as subrepo there?
15:00:47  <planetmaker> (both, OpenGFX+ Landscape and OpenGFX+ Industries will also need the normal sprites with unchanged ground)
15:01:08  <planetmaker> Ammler, yes: I immediately need to compile two projects.
15:01:25  <Ammler> why?
15:01:47  <planetmaker> when you change opengfx+ landscape you'll also then immediately need to compile and check opengfx as it's a dependency
15:02:16  <Ammler> don't get that, how is that differt to your current approach
15:02:42  <planetmaker> In the graphics-only sub-repo I don't change OpenGFX immediately
15:02:50  <Ammler> of course you need
15:02:50  <planetmaker> Thus I don't break it
15:03:10  <Ammler> else another guy who would like to update ogfx has to probably fix your update
15:03:30  <Ammler> planetmaker: how you know, you don't break it if you don't update it
15:03:30  <planetmaker> Yes, but it's a concious process: Updating the graphics (sub-)repo
15:03:47  <Brot6> Central European Train Set - Patch #2860 (Closed): Shorter standard length, added templates (oberhuemer) @
15:05:17  <Ammler> IMO, it is a good idea to use subrepo for the ogfx+ projects, but it is dangerous for ogfx, but all up to you, I have warned you :-P
15:05:51  <andythenorth> shared deps are booby traps for the unwary :D
15:05:57  <planetmaker> they are
15:06:13  <Ammler> andythenorth: they are versioned
15:06:32  <planetmaker> but every update of a subrepo has to be explicitly pulled into the main repo
15:06:41  <planetmaker> thus you can check the changelog
15:07:32  <planetmaker> but what I take from this discussion is: it needs a _very_ clear way to indicate which sprites are used where
15:07:57  <Ammler> not needed, if ogfx is the graphics repo
15:08:21  <planetmaker> no? Why not?
15:09:00  <Ammler> because then you only have to care on one side
15:09:04  <planetmaker> how would that work with the oil derrik sprites being used in ogfx+landscape, ogfx+industries and opengfx?
15:09:08  <Ammler> else you have to care on both
15:09:24  <Brot6> NewGRF Meta Language - Feature #2866 (New): allocate parameters more dynamically (Hirundo) @
15:10:11  <Ammler> well, we could install a hook on ogfx, so pushes to graphics repo do automatically update ogfx
15:10:46  <Ammler> but I don't see why in that case it couldn't be a full repo
15:11:09  <Brot6> Central European Train Set - Revision 53:ecf7d8bd6a69: turn nml assignments into #defines (Eddi) @
15:12:49  <planetmaker> well... both has IMHO pros and cons
15:13:21  <planetmaker> maybe your solution is way easier
15:15:20  <Brot6> Central European Train Set - Revision 54:af354f2212fa: forgot DRG dummy vehicles (Eddi) @
15:16:46  <Ammler> planetmaker: maybe not easiest, but safer
15:17:54  <Ammler> easier*
15:20:19  <Ammler> (at least for ogfx)
15:21:48  <Ammler> well, on the other side, you can also simply branch
15:29:09  <Brot6> Central European Train Set - Code Review #2868 (New): 64 px max length vs. 48 px max length (oberhuemer) @
15:59:14  *** Brot6 has quit IRC
15:59:57  *** Brot6 has joined #openttdcoop.devzone
16:48:14  *** Lakie has joined #openttdcoop.devzone
17:10:07  <Brot6> nml: update from r1514 to r1516 done -
17:19:23  <Brot6> firs: update from r2173 to r2179 done -
17:21:24  <Brot6> cets: update from r45 to r54 done (442 warnings) -
17:22:39  <Brot6> Following repos didn't need a nightlies update: ogfx-trains (r245), narvs (r37), bros (r52), ogfx-industries (r121), sub-landscape (r72), opengfx (r681), ailib-tile (r16), transrapidtrackset (r15), 2cctrainset (r750), ailib-list (r32), opensfx (r97), ttdviewer (r34), worldairlinersset (r672), heqs (r605), openmsx (r97), basecosts (r25), nutracks (r202), nml (r1516), 32bpp-extra (r40), manindu (r7), newgrf_makefile (r305),
17:22:39  <Brot6> ailib-direction (r17), ailib-common (r21), snowlinemod (r49), dutchtramset (r52), ai-admiralai (r75), swisstowns (r22), metrotrackset (r56), spanishtowns (r10), frenchtowns (r6), grfpack (r279), ogfx-rv (r107), fish (r655), ogfx-landscape (r70), ttrs (r36), source-test (r2), ogfx-trees (r51), swedishrails (r203), grfcodec (r832), ai-aroai (r39), german-townnames (r34), smts (r19), chips (r143), belarusiantowns (r8), indonesiantowns
17:22:41  <Brot6> (r41), ailib-string (r29), airportsplus (r107), comic-houses (r71)
17:23:59  <Brot6> narvs: compile of r37 still failed (#2789) -
17:24:38  <Brot6> ogfx-industries: compile of r121 still failed (#2792) -
17:25:56  <Ammler> silly, today grf2html worked
17:32:12  <Brot6> Dutch Tram Set - Revision 53:e8407fb4654a: Cleanup: docs (foobar) @
17:32:12  <Brot6> Dutch Tram Set - Revision 54:a7d3ff5f886a: Update: readme (foobar) @
17:33:45  *** frosch123 has joined #openttdcoop.devzone
17:35:13  <Hirundo> planetmaker: What do you think should happen with the 'old' callback documentation?
17:35:28  <Hirundo> Should it be kept for reference or removed?
17:35:37  <planetmaker> you mean the inclusion of callbacks in the action2 sequence?
17:36:01  <Hirundo> yes, VEH_CB_XXX and friends
17:37:12  <planetmaker> I'm still not convinced that one can do without that in all cases, also when the preferred method is the new one
17:37:34  <planetmaker> thus I'd keep it with a (strong) hint that the other method is the preferred way
17:37:54  <Brot6> Following repos rebuilds successful without any difference to earlier nightlies builds: ogfx-trains, sub-landscape, manindu (Diffsize: 1), newgrf_makefile, dutchtramset, swisstowns, spanishtowns (Diffsize: 1), frenchtowns, ogfx-rv (Diffsize: 4590), ogfx-landscape (1 warnings) (Diffsize: 5460), source-test (6 warnings), swedishrails, german-townnames, belarusiantowns (Diffsize: 30), indonesiantowns (1 warnings) (Diffsize: 1),
17:37:54  <Brot6> airportsplus (Diffsize: 15915)
17:43:41  <Hirundo> Everything in one list/table, or keep things separate?
17:50:09  <planetmaker> I'd keep the two methods separate, I guess. The old method maybe could be moved to an appendix?
17:50:34  <planetmaker> though.... maybe that's stupid
17:50:46  <planetmaker> it's just the same thing, differently named and at a slightly different place
17:52:39  <Hirundo> creating a separate page / appendix might make sense
17:54:40  <Hirundo> there are multiple cases where the mapping isn't really one-to-one (cb36, purchase stuff, production)
17:56:52  <planetmaker> ok
18:02:20  <Brot6> Central European Train Set - Revision 55:fa971d6af4cc: Added templates for lengths divisible by four (oberhuemer) @
18:06:30  <Brot6> Central European Train Set - Support #2784 (Assigned): graphics template (oberhuemer) @
18:23:07  *** bodis has joined #openttdcoop.devzone
19:16:15  * frosch123 wonders why someone would prefer .pcx over .png
19:16:27  <planetmaker> :-)
19:16:37  <planetmaker> you're not alone there
19:16:54  <V453000> what is the difference? png is just open-able friendly on the first sight
19:17:07  <planetmaker> see. That's it
19:17:50  <planetmaker> pcx is a rather ancient proprietary format albeit also documented. While png is by definition portable and well-documented
19:18:01  <planetmaker> png is meant for internet usage
19:18:21  <planetmaker> both use a compression... so... using one over the other is pointless
19:18:22  <V453000> :)
19:18:32  <planetmaker> and browsers showing png is _the_ argument IMHO
19:18:41  <V453000> yeah
19:18:57  <V453000> but not just browsers, I think simplier image viewers have trouble with pcx too, dont they
19:19:05  <planetmaker> additionally to all my file viewers also being able to handle png, but not pcx
19:19:09  <planetmaker> yeah
19:19:14  <V453000> :)
19:20:14  <frosch123> hmm, the norton command in my dosbox can preview .pcx files
19:20:21  <frosch123> but only 16 colour ones (or less)
19:20:55  <planetmaker> :-)
19:24:12  <V453000> 16 isnt ... too much :D
19:35:30  *** andythenorth has quit IRC
19:36:18  *** XeryusTC2 is now known as XeryusTC
19:43:04  <Brot6> NewGRF Meta Language - Revision 1517:4e775fa7e6df: Doc #1555: Move documentation on old-style cal... (Hirundo) @
19:43:04  <Brot6> NewGRF Meta Language - Revision 1518:42172c8d4087: Doc: Create a fixed order (properties - variab... (Hirundo) @
19:55:12  *** Lakie has quit IRC
20:43:58  *** frosch123 has quit IRC
20:57:30  *** FooBar has quit IRC
21:31:51  *** ODM has quit IRC
22:04:51  *** bodis has quit IRC

Powered by YARRSTE version: svn-trunk