Log for #openttdcoop.devzone on 30th December 2010:
Times are UTC Toggle Colours
01:03:40  *** KenjiE20 has quit IRC
01:25:18  *** fanioz has quit IRC
01:47:31  *** fanioz has joined #openttdcoop.devzone
02:11:42  <Brot6> 2cc train set - Feature #2081: More Graphical Improvements (EmperorJake) @
02:13:11  *** fanioz_ has joined #openttdcoop.devzone
02:16:19  *** fanioz has quit IRC
02:21:46  *** fanioz_ is now known as fanioz
02:56:24  <Brot6> 2cc train set - Feature #2081: More Graphical Improvements (EmperorJake) @
03:14:53  <Brot6> 2cc train set - Feature #1980: Pennsylvania Railroad GG1 (EmperorJake) @
03:26:23  *** thgergo has quit IRC
03:43:36  *** fanioz_ has joined #openttdcoop.devzone
03:49:19  *** fanioz has quit IRC
04:29:34  *** fanioz_ has quit IRC
08:23:21  <planetmaker> good morning
08:30:17  *** ODM has joined #openttdcoop.devzone
09:00:15  <Terkhen> <--- is this a too big hack?
09:00:33  <Terkhen> I need that file to exist before the dependency check
09:04:23  <Terkhen> generating it just for doing "make clean" is stupid, though
09:49:12  <Terkhen> hmmm... it is not possible to define a town_names block inside an if?
09:50:09  <planetmaker> Terkhen: it's for now a hack. There should be a better way and there is... but if it works for now, it's fine
09:50:52  <Yexo> <Terkhen> hmmm... it is not possible to define a town_names block inside an if? <- nfo supports it, but the nml python code does not
09:51:44  <Terkhen> oh, I see
09:52:53  <Brot6> Spanish Town Names - Revision 4:cf002b16458a: Fix: Generate the required town list file before de... (Terkhen) @
09:52:53  <Brot6> Spanish Town Names - Revision 5:423bcdd5a472: Feature: Assign town name probabilities based on th... (Terkhen) @
09:53:10  <Brot6> spanishtowns: update from r1 to r5 done -
10:02:56  <Terkhen> <--- this makes it work for me, I don't know if nml would need further changes elsewhere
10:03:12  <Ammler> Terkhen: use a variable for probability and set it to 0
10:03:32  <Terkhen> oh, that's a good idea too :D
10:07:34  <Ammler> he, your patch for nml is quite small
10:12:02  <Yexo> Terkhen: the main problems will arise when townname block A refers to block B where block B is inside an if-block
10:12:30  <Yexo> when the conditional turns out to be false, block B is never parsed by OpenTTD and I don't know how it'll handle the reference from block A
10:13:55  <Terkhen> then the issue is more complicated than what my patch tries to solve
10:14:03  <Yexo> also, you're not limited to one townname set per newgrf, you can have more than one and the user can chose inthe game options menu between them
10:14:30  <Terkhen> my idea is to let the user decide which spanish regions he wants to use
10:15:08  <Yexo> now that is not easily possible in nml
10:16:44  <Yexo> hmm, add some syntax to provide an optional conditional next to each probability. If the conditional is false nmlc will change the probability to 0
10:19:15  <Brot6> OpenGFX - Bug #2052: New Icon in Transparency Options (Yoshi) @
10:26:26  <Brot6> OpenGFX - Bug #2052: New Icon in Transparency Options (planetmaker) @
10:29:01  <Terkhen> I get a "expected a compile time constant" error; I guess it does not support adding that
10:57:18  <Brot6> Spanish Town Names - Revision 6:7fb0d26dcfa2: Add: Spanish translation (Terkhen) @
11:20:22  <Brot6> Spanish Town Names - Revision 7:1282265b70bd: Import: "Spanish" string from OpenTTD in all transl... (Terkhen) @
11:21:45  <Brot6> OpenGFX - Bug #2052: New Icon in Transparency Options (2006TTD) @
11:22:31  <Brot6> Spanish Town Names - Bug #2078 (Closed): DevZone compile failed (Terkhen) @
11:27:42  <Brot6> OpenGFX - Bug #2052: New Icon in Transparency Options (2006TTD) @
11:37:25  <Brot6> OpenGFX - Bug #2052: New Icon in Transparency Options (Ammler) @
11:45:58  <Terkhen> make: *** No rule to make target `spanish_town_names.src.dep', needed by `spanish_town_names.nml'.  Stop. <-- hmm... I get this error when compiling from a bundled source of spanish towns, but not when compiling from the hg repository
11:46:32  <Terkhen> I guess it is caused by my generation script, but I don't do anything with mercurial there
11:56:53  *** KenjiE20 has joined #openttdcoop.devzone
12:00:46  <Brot6> OpenGFX - Bug #2052: New Icon in Transparency Options (2006TTD) @
12:40:10  <Brot6> OpenGFX - Feature #885: Toyland House and Industry Ground tile (2006TTD) @
12:53:47  <planetmaker> Terkhen: try 'make depend'. Does it work then?
12:58:26  <Terkhen> no, it still fails
12:58:46  <Terkhen> <--- output of make depend in the mercurial repository; <--- output of make depend in the bundled src
13:00:25  <planetmaker> hm, right
13:07:32  <Ammler> why again do you get the files from the repo instead the "real" file tree?
13:14:41  <planetmaker> hm?
13:17:41  <Ammler> somehow you use the file list from the repo for the dep check, don't you?
13:18:38  <planetmaker> I use it for source bundles, I think
13:19:35  <Ammler> REPO_FILES and REPO_DIRS
13:20:11  <Ammler> why do you generate those lists from repo, instead from the real file tree
13:20:35  <planetmaker> my repos usually contain a lot of junk
13:20:44  <planetmaker> having that as dependency is not useful
13:21:09  <Ammler> how does openttd handle that?
13:21:15  <Ammler> with the source.lst?
13:21:52  <planetmaker> well. First: I have several different methods to get the deps
13:21:58  <planetmaker> the most-used one is using gcc
13:23:28  <planetmaker> that's used for the text files, if possible
13:24:02  <planetmaker> the graphics files are then just looked for in the files picked up by that. Repo_dirs and files has no issue there usually
13:25:40  <planetmaker> the repo_dirs and files are actually only used when creating source bundles
13:26:13  <planetmaker> as more files are part of the source than are an actual dependency.
13:26:28  <planetmaker> a good indicator for creating a source bundle is exactly that: whatever is part of the repo
13:26:47  <planetmaker> minus a bit stuff which source bundles don't need. like .hgtags etc
13:32:57  <Ammler> so those vars are in the Makefile.dist, if someone likes to make a source bundle from the source bunfle?
13:33:03  <Ammler> why should someone do that?
13:33:20  <planetmaker> just because ;-)
13:33:39  <planetmaker> maybe changed stuff
13:33:56  <Ammler> but that wouldn't work
13:35:13  <Ammler> I guess, there is something else, I just don't remember anymore
13:35:25  <Ammler> something the debian guys had issues with
13:35:37  <Ammler> maybe with cleaning...
13:36:48  <Ammler> maybe you should give up supporting base set and newgrfs with same makefile
13:39:58  <Ammler> also I recommend you once not to update newgrfs, where you don't develop yourself, they should update self or work with what they have
13:40:57  <planetmaker> That's what I do ;-)
14:04:16  <Brot6> Example NewGRF Project - Revision 250:19167e1de3e1: Fix: Keep the readme of the (new) NewGRF sepa... (planetmaker) @
14:04:22  <Brot6> newgrf_makefile: compile of r250 still failed (#2073) -
14:05:40  <planetmaker> hm, nice :-) but expected ;-)
14:08:54  <Ammler> the question, will you get it working again or shall I remove the test?
14:09:15  <planetmaker> <-- that log is currect, but those files it complains shall not be part of the repo
14:09:41  <planetmaker> a good test would be to build from the full nfo and nml package which result from running make bundle(_zip) there
14:09:58  <Ammler> it does that
14:10:07  <Ammler> cd example... && make
14:10:08  <planetmaker> I don't see the logs of that?
14:10:14  <planetmaker> hm?
14:10:31  <planetmaker> oh, this package is special. Not the source of this repo. That's in this case pointless
14:10:47  <Ammler>
14:10:47  <planetmaker> the resulting zip packages for nml and nfo would rather need testing here
14:10:54  <planetmaker> as they're the "product"
14:11:17  <planetmaker> ah
14:11:39  <Ammler> make works, make bundle_src doesn't
14:11:53  <planetmaker> it overwrites the first test ;-)
14:11:53  <Ammler> which also faild on Terkens projects
14:12:21  <Ammler> nah, it fails on first test and then exits
14:12:46  <Ammler> there is no need to make the 2nd test
14:13:34  <planetmaker> btw, it'd be better to give a link like (without the rev) in the bug report
14:13:45  <planetmaker> it will remain valid also after a version update ;-)
14:16:10  <Ammler> well, the current devzone now might also allow to modify a ticket
14:59:28  <Terkhen> <--- it seems that it always enters in the "Disallow construction" part of the switch... do you know what am I doing wrong?
15:18:38  <Hirundo> 0x8B is never in range 0..16
15:20:59  <Hirundo> Terkhen: 0x8B != variable 0x8B
15:21:13  <Terkhen> hmm...
15:22:18  <Hirundo> try var[0x8B]
15:23:38  <Terkhen> using nearby_tile_is_water causes a NML internal error
15:24:07  <Terkhen> let's see what happens with var[0x8B]
15:25:53  <Yexo> nearby_tile_is_water needs 2 arguments, the internal error without any arguments is bad
15:26:54  <Terkhen> oh, if that needs arguments then I'm not understanding this correctly
15:27:00  * Terkhen rereads
15:30:50  <Yexo> nearby_tile_is_water(x_offset, y_offset)
15:30:54  <Yexo> that is the correct syntax
15:34:17  <Terkhen> thanks, that removes the internal error
15:45:51  <Brot6> NewGRF Meta Language - Bug #2089 (New): name of function without parentheses triggers internal error (yexo) @
15:47:45  *** ODM has quit IRC
15:56:18  <Terkhen> hmm... I think I'm mixing different things here
16:06:49  <Brot6> Example NewGRF Project - Revision 251:d4337133ca51: Add: target 'remake' for the makefile of this... (planetmaker) @
16:07:30  <Brot6> newgrf_makefile: compile of r251 still failed (#2073) -
16:27:51  <Terkhen> hmmm... isn't nearby_tile_is_water supposed to contain the distance to the nearest water tile? it seems to return always 0 for me
16:29:10  <Terkhen> but in the newgrf specs it says "Distance to the closest water tile if property 1A bit 2 is clear"
16:30:17  <planetmaker> Terkhen: you say that r7 of spanishtowns doesn't build for you?
16:30:44  <Terkhen> it builds fine, but if I make a bundle_src and try to compile from it, the compilation fails
16:30:55  <planetmaker> ah
16:31:00  <Terkhen> actually it was a tagged release, not r7
16:34:16  <Hirundo> nearby_tile_is_water and var 8B (during CB 28) are two completely different things
16:34:50  <Hirundo> The former checks whether a given tile with offset (x, y) is water or not
16:37:59  <Yexo> Terkhen: to check var 8B during cb28 use: var[0x8B], nearby_tile_is_water is bit 1 from industry var 0x62
16:38:42  <Terkhen> thanks, it seems that I was mixing things :)
16:40:05  <Terkhen> great! now it is working :D
16:42:56  *** thgergo has joined #openttdcoop.devzone
16:46:12  <Terkhen> <--- IMO 16 tiles is enough distance from the coast, I'm not sure if it is worth the effort making this distance configurable
16:47:37  <Yexo> you're sure to get users who want to disable that limit completely
16:47:53  <Yexo> for example for all-land maps
16:48:21  * planetmaker would like to see an option to that end, yes ;-)
16:48:30  <Yexo> if you make a parameter for it you can use "0..parameter_name:" instead of "0..16:"
16:48:36  <planetmaker> refineries can be fed in central asia directly from pipelines
16:48:46  <Terkhen> hmm... completely disabling it makes sense
16:49:11  <Terkhen> okay, let's see :)
16:49:45  <Terkhen> given that OpenGFX+ Industries already allows industries to close, this is the only annoying thing missing for me
16:50:12  <planetmaker> :-D
17:16:01  <Terkhen> <--- done :)
17:18:00  <Brot6> spanishtowns: update from r5 to r7 done -
17:18:06  <Brot6> Following repos didn't need a nightlies update: 2cctrainset (r709), 32bpp-extra (r39), ai-admiralai (r75), ailib-common (r16), ailib-direction (r13), airportsplus (r70), basecosts (r22), belarusiantowns (r8), bros (r45), comic-houses (r71), firs (r1622), fish (r509), frenchtowns (r6), grfcodec (r818), heqs (r567), indonesiantowns (r41), manindu (r6), metrotrackset (r56), newgrf_makefile (ERROR r251), nml (r1118), nutracks (r121),
17:18:06  <Brot6> ogfx-nogrid (r11), ogfx-rv (r78), ogfx-trains (r201), ogfx-trees (r42), opengfx (r591), openmsx (r97), opensfx (r97), smts (r19), snowlinemod (r45), swedishrails (r193), swisstowns (r22), transrapidtrackset (r15), ttdviewer (r26), ttrs (r24), worldairlinersset (ERROR r670)
17:18:37  <Brot6> newgrf_makefile: compile of r251 still failed (#2073) -
17:18:39  <planetmaker> value=0 means no distance limit?
17:18:54  <planetmaker> yes... via CB
17:19:12  <Terkhen> with value=0, it will ignore the CB
17:19:38  <planetmaker> well. Go for it, if you like :-)
17:19:52  <Terkhen> okay, thanks everybody for your help :)
17:23:45  <Brot6> OpenGFX+ Industries - Revision 2:0ee4361c26d7: Feature: Configurable distance from water for oil ... (Terkhen) @
17:23:54  <Terkhen> do you have any further plans for OpenGFX+ Industries?
17:30:55  <Ammler> merge with manind
17:31:59  <Terkhen> I see :)
17:32:10  <Terkhen> hmmm... it does not seem to have nightlies enabled
17:32:40  <Ammler> I guess, there wasn't something in yet
17:33:17  <Ammler> I planned to add the lumber mill for every clima there too
17:33:21  <Ammler> or we*
17:33:47  <Terkhen> hmm... it would make sense to decide what chains to include, and what types
17:34:05  <Terkhen> for example: wood chain (none, standard, tropical)
17:34:40  <Ammler> well, basically everything should be optional, imo
17:34:46  <Terkhen> or oil chain (none, without oil rigs, with oil rigs)
17:35:24  <Terkhen> it might get complicated for farms and factories
17:36:03  <Ammler> also someone made a nml for snowaware forests
17:36:18  <Terkhen> oh, that would be really nice too :)
17:38:40  <Ammler> added you to the project
17:42:20  <Ammler> <-- Forest
17:42:22  <Webster> Title: Transport Tycoon Forums • View topic - NML - a Newgrf Meta Language (at
17:42:44  <Ammler> shall we ask Valance, if he likes to contribute to the ogfx+ind project?
17:47:22  <Brot6> OpenGFX+ Industries - Feature #2090 (New): Snowaware forests (Ammler) @
17:48:52  <Terkhen> thanks
17:49:06  <Terkhen> yes, that would be great :)
17:57:59  <Brot6> OpenGFX+ Industries - Feature #2090: Snowaware forests (Ammler) @
17:58:29  <planetmaker> well. I'd make the industries not per chain but per climate. For everything else we have FIRS :-P
17:58:51  <planetmaker> But I'd allow the same for climates as for the vehicles.
17:59:16  <planetmaker> with some exceptions maybe. Like the lumber mill
18:00:06  <planetmaker> though... with suggestions for chains I'm open to that, too as an alternative
18:00:13  <Ammler> IMO, such things can be decided, if there is much to chose
18:00:23  <Ammler> currently everything could be simply option for itself
18:00:37  <planetmaker> snow aware forrests need no option ;-)
18:00:43  <Ammler> is the closing disableable?
18:00:59  <planetmaker> what closing?
18:01:06  <planetmaker> general?
18:01:07  <Ammler> you added closing for some industries
18:01:23  <planetmaker> uhm... I forgot what I did here ;-)
18:02:46  <planetmaker> ah. No options yet
18:05:09  <DJNekkid> did "we" reach any conclution regarding nutracks? i became abit afk in the middle of it
18:07:47  <Brot6> 2cc train set - Feature #2091 (New): S-tog II (Voyager1) @
18:08:03  <planetmaker> Depends what conclusion _you_ draw from it...
18:08:45  <Brot6> OpenGFX+ Industries - Code Review #2092 (New): low max value (Ammler) @
18:10:53  *** LordAro has joined #openttdcoop.devzone
18:11:47  <DJNekkid> in not sure tbh
18:12:21  <DJNekkid> well, basicly that "we" need to implement some kind of introduction-system to openttd wrt railtype dates
18:12:39  <Terkhen> hmm... if you plan to introduce manual industres behaviour, closing of those industries should probably be included in manual industries parameters
18:12:41  <DJNekkid> and until then im not sure what to do
18:13:11  <planetmaker> the amount of railtypes is completely independent of the introduction date issue
18:13:30  <LordAro> Ammler: needs to be cleaned up - there are strange repos like "please-create-the-repo" and "maybe-now" and "please-now" and many others that are completely empty - unless there is some purpose for them ;)
18:14:33  <Brot6> 2cc train set - Feature #2091: S-tog II (DJNekkid) @
18:24:18  *** LordAro has quit IRC
18:27:31  <DJNekkid> planetmaker: as a temp. solution could i add parameters to remove the "more wierd" types, the subways and the planning tracks?
18:28:21  <planetmaker> why a temp solution?
18:30:11  <planetmaker> why not now the real solution? Whatever approach you favour?
18:39:28  <Ammler> hmm, do those empty repos hurt?
18:39:59  <planetmaker> empty?
18:43:25  <Ammler> mostly test respos from my autocreation tests
18:44:15  <planetmaker> oh. They didn't bother me until now. But they're pointless, too
18:44:22  <planetmaker> We have the sandbox for testing other stuff
18:44:38  <planetmaker> but... they're definitely at the end of any todo list
18:45:22  <Ammler>
18:45:58  <Terkhen> what should I do about Spanish Town Names then?
18:48:09  <planetmaker> between watchin movies I try to find the bug :-P
18:48:22  <Terkhen> ok :)
18:48:29  <planetmaker> but I keep finding other bugs in the makefile(s) :S
18:48:37  <planetmaker> but not this one
18:49:28  <Terkhen> I'd like to release it either today or tomorrow, as one of my new year resolutions is going to reduce my time for these projects
18:49:32  <planetmaker> but I'm using meanwhile the EST as my test newgrf
18:49:38  <planetmaker> ah
18:49:44  <planetmaker> hm
18:51:46  <DJNekkid> planetmaker: well, the set is still imho highly in a dev stage
18:52:58  <Terkhen> releasing would help me in avoiding temptation :P
18:55:05  <Ammler> is building the source bundle from the source bundle the only issue?
18:55:13  <Ammler> I would simply disable that then...
18:55:38  <Terkhen> building the grf from the source bundle
18:55:51  <Terkhen> it will complain about spanish_town_names.dep missing
18:56:27  <Ammler> use a earlier version
18:57:00  <Terkhen> of newgrf_makefile?
18:57:22  <Ammler> planetmaker: I also wondered why you added a new bash script to create a repo and still keep the
18:57:30  <Ammler> Terkhen: yes
18:57:52  <Ammler> I assume, this "bug" is introduced with last update
18:57:54  <Terkhen> for what revision should I look for?
18:58:01  <Ammler> last tag
18:58:25  <Terkhen> ok, let's see
19:00:06  <Ammler> is made to simply run it from newgrf_makefile with ./ ../<your-repo>
19:05:07  <planetmaker> I know. Which is quite convenient for me :-)
19:05:32  <planetmaker> actually without ../ before your-repo ;-)
19:06:06  <Ammler> well, ../ is easier for tab-completion
19:07:03  <Terkhen> hmm... there are no differences between example-newgrf-0.5.3-nml and example-newgrf-nightly-r246-nml; after overwriting hg diff only shows me the overwrite of my modifications to the changelog and other files unrelated to the Makefile system
19:08:07  <planetmaker> probably 0.5.3 is bugged, too
19:08:38  <Ammler> 0.5.3 is the last release
19:08:51  <planetmaker> yes, but it's like a few days old
19:09:00  <Ammler> you should update to somethign prior, if you downgrade :-P
19:09:32  <Ammler> 0.5.3 wouldn't build anymore with the new script :-)
19:12:08  <Terkhen> hmmm... okay, I'll try with 0.5.2
19:12:27  <Ammler> easiest is to clone newgrf_makefile and use
19:14:44  <planetmaker> swedishrails work
19:15:27  <planetmaker> he... it's at 0.4.5 :-P
19:17:56  <planetmaker> hm.
19:18:00  <Terkhen> hmm... after using it now gives me errors related to missing nfo things (sprites folder)
19:18:03  <planetmaker> it also works updated
19:18:04  <Terkhen> (0.5.2)
19:34:28  <Brot6> Java OpenTTD Admin Library - Revision 35:a7c97faba2fd: Change: support for cmd logging packet typ... (dih) @
19:35:20  <Ammler> Terkhen: you modified your makefile, didn't you?
19:35:27  <Ammler> you should bring that back to default
19:35:42  <Terkhen> only IIRC
19:35:48  *** Lakie has joined #openttdcoop.devzone
19:36:58  <Terkhen> yes, the rest of files are not modified
19:39:08  <Ammler> -# separate dirs for the nfo and pcx files (must be a subdir of
19:39:10  <Ammler> -# or the same as the SRC_DIR )
19:39:11  <Ammler> -NFO_DIR            := $(SRC_DIR)/nml
19:39:13  <Ammler> -PCX_DIR            := $(SRC_DIR)/gfx
19:39:14  <Ammler> you miss that :-)
19:39:26  <Ammler> well, the pcx dir might not be needed
19:39:41  <Terkhen> hmm... let me check again
19:39:52  <Ammler> or not
19:43:40  <Ammler> he, I updated the spanish set
19:43:48  <Ammler> now it does make bundles called "example..."
19:44:28  <Terkhen> using with newgrf_makefile tip breaks it for me
19:44:38  <Terkhen> I did not know that macros, I just followed the readme
19:44:45  <Terkhen> s/that/those/
19:45:41  <Ammler> but it works
19:45:51  <Ammler> the only thing it breaks here is the bundle name
19:46:26  <Ammler> make bundle_src does create a bundle called example-newgrf-nightly-r7M-source.tar.gz
19:46:38  <Ammler> then I unpacked and run make, which worked
19:47:31  <Ammler> so there is just a config var missing, I would guess
19:48:51  <Terkhen> <--- this is what I get after using ./update <spanish towns repo>
19:49:18  * Terkhen wonders if his VM is broken
19:49:24  <Ammler> I would most suspect your confit to the repo: REPO_NAME          := Spanish Town Names
19:49:27  <Ammler> spaces there
19:50:19  <Ammler> hmm, swedish rails does have that too
19:50:34  <Ammler> how does it make the name?
19:50:37  <planetmaker> spaces... are a killer
19:50:48  <Terkhen> ogfx-rv too, which is the project I used as an example
19:51:08  <planetmaker> but I'm not sure how repo_name is used right now.
19:51:13  <planetmaker> only as string possibly
19:51:40  <Terkhen> I'm getting "file is created in the future" errors all the time, so I'm going to see what's causing that first
19:51:54  <Terkhen> (although it still compiles without using
19:59:59  <Ammler> 	$(_V) rm -rf $(DIR_NAME_SRC) <-- this is in bundle_src, should be in clean
20:01:55  <planetmaker> no
20:02:29  <planetmaker> or I could not call bundle_src anew without cleaning first
20:02:53  <planetmaker> which is way more expensive than copying a few files
20:02:57  <planetmaker> (again)
20:03:25  <Ammler> why should you call it again?
20:03:34  <planetmaker> because I modified the code
20:03:50  <planetmaker> or I'd make it depend on everything in the repo
20:03:51  <Ammler> now I need to unzip it again
20:04:18  <planetmaker> another big dependency. And as it depends on the whole repo and it always is just a "copy all", I spare me that hassle
20:04:22  <Ammler> which is also a bit stupid as the makefile just made it
20:04:49  <planetmaker> Ammler: without that line this won't update the source repo:
20:05:02  <planetmaker> make bundle_src && edit some source file && make bundle_src
20:09:05  <Ammler> I removed that line and the dir does still disapear :-o
20:09:52  <Ammler> I guess, that is the reason, it is broken
20:10:03  <Ammler> it doesn't use those targets anymore
20:10:56  <Ammler> does copy the Makefile, which changed
20:11:32  <Ammler> you made custom targets to the Makefile directly, instead to
20:12:44  <Ammler> how does still work for you?
20:17:01  <planetmaker> that uncommited way ;-)
20:17:14  <planetmaker> (and now commited)
20:17:40  <Brot6> Example NewGRF Project - Revision 252:026f6cf78301: Fix: copied the wrong Makefile (planetmaker) @
20:17:40  <Brot6> newgrf_makefile: compile of r252 still failed (#2073) -
20:18:10  <Terkhen> hmm... should I try with that update?
20:18:13  <Ammler> no
20:18:23  <Ammler> as you see the test still fails :-)
20:18:27  <planetmaker> that doesn't fix your problem
20:18:33  <planetmaker> :-(
20:18:56  <Ammler> but it fixes the script
20:20:33  <Terkhen> :P
20:20:37  <Ammler> I also start to think, the test is not related to terkens issue
20:21:41  <planetmaker> I'm somewhat flabbergasted. SER works, EST not
20:22:03  <Ammler> planetmaker: why not start with the example project?
20:23:02  <planetmaker> well. If I solve it for Terkhen I might have solved that, too?
20:23:21  <planetmaker> and it's a tad easier to test ;-)
20:24:14  <Ammler> no
20:24:21  <Ammler> I know the issue of the test
20:24:31  <Ammler> it needs to be its own hg repo
20:24:46  <Ammler> shall I push the "fix"?
20:24:52  <planetmaker> can you show me?
20:25:03  <planetmaker> (instead of push)
20:25:14  <planetmaker> because it works for SER w/o hg repo
20:27:15  <Ammler>
20:27:32  <Ammler> SER is its own repo already
20:28:51  <Ammler> as the test runs in the newgrf_makefile repo, it does also depend on those files, whihc you need for bundle_src
20:28:53  <planetmaker> I'm testing the make command from the bundle_src. That's what fails for Terkhen
20:29:00  <planetmaker> and what fails the CF
20:29:05  <Ammler> no
20:29:19  <Ammler> my paste wouldn't make the cf fail anymore
20:29:36  <Ammler> and is valid
20:30:10  <Ammler> as the framework depends on the repo
20:30:25  <planetmaker> that patch is only for the newgrf_makefile, and no other repo, right?
20:30:28  <planetmaker> There it makes sense, yes
20:30:30  <Ammler> yes
20:30:37  <Ammler> it is for the test
20:31:35  <planetmaker> then please go for it :-)
20:33:55  <Ammler> he, authorization failed :-P
20:34:28  <Brot6> Example NewGRF Project - Revision 253:42c77d9d3a50: Fix (r247): the example newgrfs need to have ... (Ammler) @
20:34:33  <Ammler> using ssh backdoor
20:35:03  <Brot6> newgrf_makefile: compile of r253 still failed (#2073) -
20:35:07  <Ammler> mäh :-o
20:39:01  <planetmaker> I don't quite understand on what occasion it fails now...
20:40:00  <Ammler> I don't see it either
20:40:08  <Ammler> the nml test seems to have worked
20:40:39  <Ammler> hmm, I should test nfo locally...
20:41:02  <planetmaker> adding sprites/pcx/test1.pcx
20:41:03  <planetmaker> /var/tmp/rpm-tmp.DcB2IX: line 32: syntax error near unexpected token `&'
20:41:05  <planetmaker> error: Bad exit status from /var/tmp/rpm-tmp.DcB2IX (%check)
20:41:11  <planetmaker> ^ but I cannot make anything of those lines
20:41:58  <Ammler> local test of nfo worked too
20:44:49  <Ammler> I guess, my redirects are wrong...
20:45:04  <Ammler> 2>>&1 -> 2>&1
20:46:17  <Terkhen> bbl
20:46:20  <Brot6> Example NewGRF Project - Revision 254:fc1ecc19d3aa: Fix (rPREV): redirect is its own mistery (Ammler) @
20:46:36  <Brot6> newgrf_makefile: update from r246 to r254 done -
20:46:44  <planetmaker> nice :-)
20:47:58  <Brot6> OpenGFX - Bug #2093 (New): Monorail sprite (Yoshi) @
20:48:52  *** andythenorth has joined #openttdcoop.devzone
20:49:13  <Ammler> another target we should run there (all, bundle_src)
20:49:16  <Ammler> ?
20:49:48  <planetmaker> bundle_zip is enough for that project.
20:50:10  <planetmaker> (and the test of the correctness of the distributed newgrf stubs)
20:50:12  <Rubidium> bundle_dmg! :)
20:50:17  <andythenorth> evenings
20:50:22  <planetmaker> :-P
20:50:23  <planetmaker> hi andythenorth
20:50:30  <andythenorth> anything?
20:50:46  <planetmaker> something
20:50:51  <andythenorth> great
20:50:59  <andythenorth> I shall have some wine in that case
20:51:33  <planetmaker> no problem. White, rosé or red?
20:51:36  <Ammler> well, we still have not solved the Terken-issue
20:51:37  <andythenorth> I have white
20:51:42  <andythenorth> Terken is an issue?
20:51:43  <andythenorth> I like him
20:51:51  <andythenorth> is there a ticket about him?
20:51:51  <planetmaker> nah. He has issues :-P
20:51:55  <Ammler> yes, he lost the h
20:52:29  <planetmaker> he seems to have problems with names in general.... ;-)
20:52:35  <Ammler> andythenorth: don't you have a build system issue too?
20:52:41  <andythenorth> yes
20:52:49  <andythenorth> I have to use make clean && make install
20:52:59  <andythenorth> seems like changes are missed
20:53:06  <Ammler> in which repo?
20:53:36  <Ammler> planetmaker: bundle_zip instead all?
20:54:06  <planetmaker> it will call 'all' then automatically
20:54:27  <planetmaker> but the bundle_zip generates the 'output' which needs shipping
20:54:37  <planetmaker> or you need to zip the resulting dirs...
20:54:55  <andythenorth> Ammler: in FIRS and FISH
20:55:00  <andythenorth> I haven't tested HEQS
20:55:09  <andythenorth> it's not a drama as the build is relatively quick these days
20:55:46  <planetmaker> andythenorth: what kind of change is it typically that you need to clean first in order to install successfully?
20:56:22  <Brot6> 2cc train set - Feature #2091: S-tog II (Voyager1) @
20:58:24  <Ammler> planetmaker: I could add that to the test too
20:58:43  <Ammler> a simple touch sprites/mynewgrf.nfo should simulate a change, right?
20:58:52  <planetmaker> yes
20:59:01  <Ammler> does not rebuild on my local test
20:59:29  <planetmaker> :-(
20:59:30  <Ammler> marcel@inspiron:~/hg/newgrf_makefile/example-newgrf-nightly-r253-nfo> touch sprites/mynewgrf.nfo
20:59:32  <Ammler> marcel@inspiron:~/hg/newgrf_makefile/example-newgrf-nightly-r253-nfo> make bundle_zip
20:59:34  <Ammler> make: Nothing to be done for `bundle_zip'.
21:01:25  <Ammler> btw. feel free to add your own tests, newgrf_makefile should be a good example how to do that with rpm spec
21:02:54  <planetmaker> ok :-)
21:03:07  <Brot6> 2cc train set - Feature #2094 (New): S-tog IV (Voyager1) @
21:03:17  <andythenorth> planetmaker: I'm not sure without testing
21:03:36  <planetmaker> in order to add good tests, the newgrf would need to become much more complicated. And start to stretch the boundaries of the dep checks as andythenorth and DJNekkid are good at :-P
21:05:02  <Ammler> well, I guess, the simple things which fail now like no rebuild after change are fine enough
21:05:14  <planetmaker> yeah
21:05:20  <planetmaker> definitely
21:05:50  <Ammler> I just wonder, how I change "make nothing to do" to a exit 1
21:06:04  <planetmaker> :-D
21:07:00  <planetmaker> [ "`make all`" != "make nothing to do" ] or similar
21:08:47  <Ammler> you could make a target :-P
21:09:18  <planetmaker> that's possible...
21:09:41  <Ammler> make selftest :-P
21:10:02  <planetmaker> not a bad name ;-)
21:10:02  <Ammler> mäh, I use my tongue quite a lot today
21:10:13  <planetmaker> your tongue?
21:10:37  <Ammler> yes :-P
21:11:42  <Ammler> planetmaker: best would be you turn the makefile framework to a tool and it would make tests like grfcodec and nml
21:12:14  <Ammler> nmake
21:12:31  <planetmaker> well.... yes and no.
21:12:54  <planetmaker> what actually would be good might be a configure script. That'd get rid of much of the nasty dep stuff
21:13:14  <planetmaker> rid of as in call it less often
21:13:57  <Rubidium> I doubt that, or at least with OpenTTD it doesn't work that way
21:14:06  <Ammler> hmm, don't get what configure could do, what make can't
21:14:40  <Ammler> it would be nice, if nml could build without make :-)
21:14:59  <planetmaker> well. it can. But not a complicated project.
21:15:19  <planetmaker> Same with OpenTTD. It could. If you put everything into one source file. But it's insane. That what you have make for
21:15:28  <Ammler> it misses mainly the include
21:15:59  <Ammler> well, for openttd, you need hte build stuff anyway
21:16:25  <Ammler> Makefile is just not very beginner friednly
21:16:41  <planetmaker> what is beginner friendly?
21:16:53  <planetmaker> a shell script is neither
21:17:04  <Ammler> nmlc myproject.nml
21:17:17  <planetmaker> why is that better than make
21:17:21  <planetmaker> a simple 'make'
21:17:39  <Ammler> the tool make is hell to setup
21:17:46  <andythenorth> I found make easy & a blessing
21:17:57  <planetmaker> make difficult to setup?
21:18:05  <Ammler> yes, no windows support
21:18:10  <andythenorth> II was pleased with how easy my life got with a makefile
21:18:14  <planetmaker> Everywhere I needed it it was installed by default or installed by my package manager
21:18:40  <Hirundo> Some people out there use windows, you know :)
21:18:45  <planetmaker> Ammler: the windows-solution is to (re-)write this as bat file...
21:19:01  *** ODM has joined #openttdcoop.devzone
21:19:03  <Ammler> or include "include" in nml
21:19:07  <planetmaker> but I'm not quite ready to do that ;-)
21:19:39  <Ammler> maybe extend custom_tags a bit
21:20:14  <Ammler> basically you run nml twice, once to include the includes and replace the tags, 2nd to build like now
21:21:24  <andythenorth> maybe the solution should be remote
21:21:38  <Ammler> remote building is supported
21:21:39  <andythenorth> a dumb client app that gets a remote server to build the grf
21:21:44  <Ammler> with trigger push
21:21:56  <Ammler> belarusiantowns does have that
21:41:46  <Brot6> NewGRF Meta Language - Revision 1119:9e0da64a858c: Fix: An incorrect comment in the expression code. (Hirundo) @
21:55:27  <dihedral> @logs
21:55:27  <Webster> Logs:
22:09:18  <andythenorth> ach bedtime
22:09:20  <andythenorth> good night
22:09:21  *** andythenorth has left #openttdcoop.devzone
22:24:21  <Brot6> NewGRF Meta Language - Bug #2089 (Closed): name of function without parentheses triggers internal... (yexo) @
22:24:21  <Brot6> NewGRF Meta Language - Bug #2089 (Closed): name of function without parentheses triggers internal... (Hirundo) @
22:24:21  <Brot6> NewGRF Meta Language - Revision 1120:9794fc9a8944: Fix #2089: Use the same arguments for reducing... (Hirundo) @
22:42:42  <Terkhen> hello
22:44:28  <SmatZ> hello Terkhen
22:55:25  *** ODM has quit IRC
22:58:38  <Brot6> 2cc train set - Feature #2094: S-tog IV (DJNekkid) @

Powered by YARRSTE version: svn-trunk