Log for #openttdcoop.devzone on 31st July 2010:
Times are UTC Toggle Colours
00:19:37  *** Seberoth2 has joined #openttdcoop.devzone
00:27:22  *** Seberoth has quit IRC
00:34:02  *** KenjiE20 has quit IRC
00:49:12  *** Seberoth2 has quit IRC
01:02:59  *** frosch has joined #openttdcoop.devzone
01:05:28  *** frosch123 is now known as Guest356
01:05:29  *** frosch is now known as frosch123
01:09:57  *** Guest356 has quit IRC
01:34:27  *** frosch123 has quit IRC
04:43:26  *** XeryusTC2 has joined #openttdcoop.devzone
04:44:26  *** SmatZ- has joined #openttdcoop.devzone
04:45:02  *** planetmaker has quit IRC
04:45:02  *** SmatZ has quit IRC
04:45:02  *** XeryusTC has quit IRC
04:45:26  *** planetmaker has joined #openttdcoop.devzone
04:46:46  *** SmatZ has joined #openttdcoop.devzone
04:46:46  *** XeryusTC has joined #openttdcoop.devzone
04:46:53  *** XeryusTC has quit IRC
04:47:08  *** SmatZ has quit IRC
05:13:56  *** XeryusTC has joined #openttdcoop.devzone
05:15:56  *** XeryusTC2 has quit IRC
07:55:43  *** Alberth has joined #openttdcoop.devzone
08:06:08  *** ODM has joined #openttdcoop.devzone
08:14:51  *** ODM has quit IRC
08:16:10  *** ODM has joined #openttdcoop.devzone
08:24:57  *** ODM has quit IRC
08:25:35  *** ODM has joined #openttdcoop.devzone
08:29:41  <Brot6> NewGRF Meta Language - Revision 612:9b403e6e3356: Codechange: Add name of the class that needs to... (Alberth) @
08:54:41  *** ODM has quit IRC
08:55:16  *** ODM has joined #openttdcoop.devzone
08:58:15  *** FooBar has joined #openttdcoop.devzone
08:59:23  <Brot6> NewGRF Meta Language - Revision 613:8fd8d3357119: Codechange: Added abstract output methods to Ou... (Alberth) @
09:15:29  <Brot6> NewGRF Meta Language - Revision 614:45c8d0098a49: Add: Added missing str() methods in expression ... (Alberth) @
09:18:22  *** ODM has quit IRC
09:29:10  *** ODM has joined #openttdcoop.devzone
09:32:46  <Alberth> the 'write' method is messy, in some cases there is an additional 'size' parameter.
09:36:36  <Yexo> not in any "action->write" methods, but some other classes have a write method as well
09:37:52  *** ODM has quit IRC
09:37:57  <Alberth>   is what pylint reports about write, the last line seems to contradict your last statement
09:37:59  <Webster> Title: Paste #243731 | LodgeIt! (at
09:40:12  <Alberth> is there something I can do without getting in your way?
09:41:05  <Yexo> in my way? just do whatever you want
09:42:18  <Yexo> Alberth: the only "wrong' one that causes all those warnings is the write function in
09:42:25  <Yexo> all action2 base classes don't have a size parameter
09:42:32  <Yexo> so perhaps rename the write-function in action2
09:43:11  <Alberth> ok :)
09:43:45  <Yexo> about your last changes, the __str__ function for Not should read "!%s"
09:44:06  <Yexo> for Boolean I'm not sure, perhaps we should also add "(bool)" to the parser
09:50:07  <Alberth> there is no parse rule for it, it seems, it gets added in the three with the boolean operators
09:58:28  *** ODM has joined #openttdcoop.devzone
10:03:22  <Brot6> NewGRF Meta Language - Revision 615:7e0b71dc8de2: Fix (r614): Use correct syntax for NOT operator. (Alberth) @
10:03:22  <Brot6> NewGRF Meta Language - Revision 616:c3fb33999783: Codechange: Rename duplicate write() method. (Alberth) @
10:09:46  *** frosch123 has joined #openttdcoop.devzone
10:15:16  <Brot6> NewGRF Meta Language - Revision 617:9432ec0e50fe: Codechange: Document abstract Action2, remove u... (Alberth) @
10:20:56  <Brot6> NFORenum - Revision 414:e94aaccc13bf: Add: initial support for action 14. (Rubidium) @
10:22:45  *** KenjiE20 has joined #openttdcoop.devzone
10:25:22  <Yexo> Alberth: @ivar id: Unique number of this action2.
10:25:22  <Yexo>  <- action2 id's are not unique within the grf file, they get can (and do get) reused after the last reference to an action2
10:25:59  <Alberth> argh
10:27:17  <Brot6> NewGRF Meta Language - Revision 618:6a4bc3dd74e4: Fix (r617): Fix doc-string of 'id'. (Alberth) @
10:28:31  <FooBar> is action 14 supposed to be something that delivers information about the grf, like palette?
10:29:18  <Yexo> yes (and more importantly, information about it's parameters)
10:29:38  <Yexo> and it's makes the name/description (normally action8) translateable
10:30:08  <FooBar> sweet :)
10:30:40  <Yexo> <- initial design by frosch123, although the current implementation doesn't follow that design 100%, it's still approximately the same
10:32:35  <FooBar> thanks for that link
11:12:32  *** ODM has quit IRC
11:13:08  *** ODM has joined #openttdcoop.devzone
11:13:38  <frosch123> hmm, wouldn't a language-id be an essential part of action 13?
11:17:38  <Yexo> yes (which is why action13 is completely useless)
11:18:06  <frosch123> it might work in ttdp, where one can check the language using action 7 :)
11:18:49  <Hirundo> And use that, to change [insert desync prone stuff here]
11:19:18  <frosch123> yup, luckily the locale information is not available in ottd :)
11:21:16  <Brot6> NewGRF Meta Language - Feature #1174 (New): Action 14 support (Hirundo) @
11:22:06  <Hirundo> NML has to follow suit, of course ^^
11:24:51  <Ammler> basically Action14 are data loaded together with Action8, I assume?
11:28:31  <Alberth> Hirundo: I think it would be useful to initialize all variables of a class in its constructor.
11:29:22  <Brot6> NewGRF Meta Language - Feature #1174: Action 14 support (yexo) @
11:29:48  <Hirundo> Alberth: I must say I have my doubts as well
11:31:31  <Alberth> also, we seem to have two different methods called 'start_sprite', with a different number of parameters? Perhaps rewrite one of them?
11:31:36  * FooBar thinks action 13 is intended to have each translation in separate grf file
11:32:24  <Alberth> Hirundo: the rationale of initializing in the constructor is that you de-couple methods from each other.
11:32:28  * FooBar also thinks that's quite useless
11:34:11  <Hirundo> I may have followed the nml wiki a bit too rigorously :)
11:34:49  <Alberth> or perhaps the wiki was not complete enough? :)
11:36:42  <Alberth> I found "resolve_tmp_storage()" in nml/ before the prepare_output. Is that included in the new structure?
11:36:48  <Hirundo> So, basically r608 and r611 can be reverted
11:37:35  <Hirundo> I'm not really familiar with the details of the action2 code, so I can't tell for sure
11:38:11  *** ODM has quit IRC
11:39:43  <Alberth> I like the idea of doing nothing in the constructor
11:41:09  <Hirundo> Basically, you're incorporating a certain implicit 'state' in the AST classes; whether they've been pre-processed or not
11:41:09  *** ODM has joined #openttdcoop.devzone
11:42:10  <Alberth> we have state anyway, namely, did they create actions or not?
11:42:43  <Alberth> doing nothing in the constructor means that the parser will not fail because of errors in that processing
11:45:11  <Yexo> <Alberth> doing nothing in the constructor means that the parser will not fail because of errors in that processing <- that is imo a very good argument for doing nothing in the constructor
11:45:25  <Yexo> but then we shouldn not even do reduce() there
11:50:11  <Alberth> indeed
12:04:43  <Brot6> NFORenum - Revision 415:8b69e78708ad: Add: Checking of types and strings in an Action 14 (Rubidium) @
12:22:36  *** Seberoth has joined #openttdcoop.devzone
12:31:25  *** Seberoth2 has joined #openttdcoop.devzone
12:35:44  *** Seberoth has quit IRC
12:37:51  <FooBar> Is it possible that the makefile doesn't work when there are no commits in a repo yet?
12:38:15  <FooBar> Otherwise I broke it now on linux as well :(
12:38:53  <Alberth> don't know, what does the error say?
12:39:30  <Alberth> I would expect that you get revision 0 or so
12:42:06  <FooBar> I get revision -1, but it says "nothing to be done for file.grf"
12:42:23  <Ammler> FooBar: simply commit
12:42:29  <FooBar> well, more particular: make[1]: *** No rule to make target `traratrk.grf'.  Stop.
12:42:34  <Ammler> and try, you can rollback after
12:42:46  * FooBar doesn't want to commit yet :P
12:42:48  <FooBar> but ok
12:42:53  <Ammler> hg rollback
12:43:09  <Ammler> and commit is not push
12:44:33  <FooBar> I committed the makefile, but it's still the same
12:44:48  <FooBar> guess I have to try another commit and then strip...
12:45:14  <Ammler> strip does also remove the files, so be patient
12:45:48  <FooBar> hmmm...still the same
12:45:50  <Alberth> or does the makefile query the revision of the source files?
12:46:07  <Ammler> careful*
12:46:28  <FooBar> I think it gets the revision of the repo
12:46:30  <Yexo> makefile uses hg to find out which files to build / or for the depencencies or somethign like that
12:47:43  <FooBar> now I committed all files
12:47:57  <FooBar> I'll just remove the .hg dir and make it a new repo later
12:48:20  <FooBar> now it works :S
12:48:39  <Alberth> I usually do it the other way around; make a test clone first to mess up :p
12:48:45  <FooBar> it probably needs the initial pnfo file committed
12:49:07  <Ammler> what yexo said, imo
12:49:17  <Alberth> yeah "no rule to make X" usually means it cannot find a source file
12:50:39  <FooBar> that's probably the reason why it doesn't work on Windows then
12:51:21  <FooBar> or at least my Windows
12:51:59  <FooBar> It appears (did another test) that the initial pnfo file listed in makefile.config needs to be committed
12:52:11  <FooBar> as it works if I only commit that
12:54:24  <FooBar> another mystery solved then
12:58:58  <FooBar> On a related note, I still think there should be an option to not do the dependency check and just start with compiling, as either gcc or grfcodec will tell me if there's a file missing anyways
13:02:14  <Alberth> make does that implicitly
13:02:38  <Alberth> the idea is that rebuilding takes more time than checking whether something is up-to-date
13:02:49  <Alberth> if that does not hold, you should not use make
13:03:01  <Yexo> which is not true on cygwin with grf projects
13:03:33  <Rubidium> with only one output file that is not linked from multiple intermediate files make is usually pointless
13:04:15  <Alberth> the culprit is most likely not even the time-stamp checking that make does, but the computation of which file needs which other file.
13:04:30  <Yexo> exactly
13:04:38  <Alberth> a simple way around that is to make it dependent on all files
13:05:03  <Yexo> in that case you still need a list of "all flies"
13:05:11  <Rubidium> problem is that that "dependency generation" happens whenever a .nfo file is edited
13:05:14  <Yexo> with the current makefile that is not needed
13:05:31  <Rubidium> which means the dependency stuff *only* makes sense when only changing the pcx files
13:05:54  <FooBar> for a compilefarm it is fine to do such check. While developing a grf, I /know/ when there's a change and only then build the grf, so I don't need a check to see if something changed as I already know that
13:06:36  <Yexo> a "make build" that skips the dependency check woukld already be enough
13:07:20  <FooBar> yes, that would work for me
13:08:01  <FooBar> and maybe for Windows as well. If not, it still saves a lot of time
13:09:23  <FooBar> unfortunately I don't understand much of the make syntax, otherwise I'd already attempted to add something like that myself
13:09:51  <Alberth> do you have an url for it?
13:10:26  <FooBar> ?
13:10:37  <Alberth> so I can have a look
13:10:54  <FooBar> of the makefile source?
13:11:00  <FooBar> let me check
13:11:01  <Alberth> yes
13:11:07  <Ammler> imo, the dep check is also useless for the compile farm
13:11:20  <Ammler> (for newgrfs)
13:11:24  <FooBar>
13:11:32  <Alberth> you need to test the dep calculation too :p
13:11:48  <FooBar> Alberth: ^^
13:12:00  <Alberth> I noticed, thanks
13:12:12  <FooBar> the main stuff is in scripts/makefile.common
13:12:29  <Ammler> since building the grf is faster then the whole dep checks
13:13:10  <FooBar> true, I can appreciate the idea behind it, but it's not much beneficial
13:14:11  <FooBar> In Make, does $@ by any chance hold the name of the current target?
13:14:13  <Brot6> NFORenum - Revision 416:ffdefca07d27: Change: don't read data to check the IDs just yet (i.e. sil... (Rubidium) @
13:14:43  <FooBar> If so, then I might start to understand something of it
13:14:57  *** V453000 is now known as Guest443
13:16:55  <Alberth> yes, afaik, but I need to look it up or test it each time :)
13:18:13  <Alberth> a test: In Makefile, line 28 (containing "all: depend") + 29    copy those 2 lines to above line 27
13:18:30  <Alberth> then remove " depend" and rename "all" to "quick" or so
13:18:45  <FooBar> will try
13:18:58  <Alberth> hmm, perhaps better copy it to below the 'all' target
13:19:07  <Alberth> then 'make quick'  :)
13:19:44  <Alberth> ie leave out the 'depend' sub-target
13:20:35  <Alberth> touch somefile.nfo   to trigger updating
13:21:00  <Alberth> (touch is like loading the file in the editor, and writing it back to disk)
13:22:25  <FooBar> it still says "[generating] makefile.dep"
13:22:40  <FooBar> I must now time it to see if it's any quicker...
13:22:56  <FooBar> also, do your line numbers start with 0?
13:23:11  <FooBar> as they seemed to be off-by-1 in your description :)
13:24:15  <FooBar> time make quick: 12.4; time make: 12.6 :(
13:24:48  <Alberth> no, it should not generate a makefile.dep
13:25:10  <Alberth> I used line numbers as provided by Redmine
13:25:36  <FooBar> I might have made a bad, so I'll try again with the lines in some different order
13:26:22  <FooBar> hmmm...still does generate makefile.dep with time 12.5
13:26:45  <Alberth> -include ${MAKEFILE_DEP}    <-- add a "#" in front of that
13:26:46  <FooBar> note that this is not a very big grf, but with firs make times are in the order of a minute
13:27:20  <Alberth> yeah, dependency calculation is done in the shell, which is horrible to say the least
13:27:32  <FooBar> time 6.5 :)
13:28:07  <frosch123>
13:28:33  <frosch123> i guess it will me maximal tedious to find someone who can add it to the table of contents :)
13:28:57  <FooBar> thanks Alberth, I will now try with FIRS to see the real time savings
13:29:01  <Yexo> maybe lakie can?
13:29:26  <frosch123> good point, he is the only one around regulary :)
13:29:27  <Alberth> well, it is a hack, it should be done in a more clean way
13:29:56  <FooBar> maybe leave that to the founder of the makefile?
13:30:05  <FooBar> unless you want to write a patch for it or something
13:30:52  <Alberth> I was pondering such an action :)
13:31:27  <FooBar> still doesn't work on Windows, but then I already learned to live with that. The time saving is more important and wether I use a native Windows commandline or a virtual linux one doesn't make much difference either
13:36:22  <Alberth> it does some pretty scary shell stuff, so I am not surprised it breaks at windows
13:36:58  <FooBar> example of what's scary?
13:38:13  <Alberth> have a look in scripts/Makefile.common :)
13:38:35  <FooBar> FIRS    original: 1m29.114s    new: 0m24.359s
13:38:50  <Alberth> else # NFO and NML projects   <-- below that line
13:39:27  <Alberth>   if you care to test it, please don't hesitate :)
13:40:24  <Alberth> with the patch, it should do dep checking. DO_DEP_CHECK       := false   should disable it
13:40:48  <Ammler> the palette will be written by grfcodec?
13:41:15  <Alberth> is that related to dep checking?
13:41:20  <Ammler> or is that simple info which the author needs to be care?
13:41:32  <FooBar> I wonder if it's really the dep check that's the problem on Windows, as without it, it still doesn't work. It generally also does the dep check on windows, just takes ages and creates a million of grep instances :)
13:41:45  <Yexo> Ammler: author has to care
13:41:56  <Yexo> nml will automatically write it though (in the future)
13:42:56  <Ammler> well, support for an other palette is quite obsolete, isn't?
13:43:16  <Alberth> FooBar: ok, I have no clue, haven't touched any windows machine in about a decade or so
13:43:19  <Yexo> not really, grfs with the dos palette still work fine
13:43:28  <Yexo> but nobody creates such grfs anymore
13:43:36  <Ammler> yes, that what I mean
13:43:48  <Ammler> since you can't save it in the preset
13:44:17  <FooBar> Alberth: no problem. I appreciate the time savings much more :)
13:44:19  <Rubidium> Ammler: it will make it easier to go with your base pack to the dos palette :)
13:44:59  <Ammler> but now, with Action14, someone can again make dos grfs :-)
13:45:18  * FooBar installs patch
13:45:48  <FooBar> I'm surprised something elementary as patch doesn't come with ubuntu
13:46:23  <Ammler> aren't there patterns like "Development"?
13:47:17  <FooBar> as with most of linux: I have no clue :)
13:47:49  <Rubidium> FooBar: "lusers" don't need patch, as such Ubuntu doesn't come with it installed by default
13:48:28  <FooBar> there's a point to that
13:48:45  <Rubidium> although apt-get install build-essentials should get you the most essential stuff
13:49:32  * FooBar checks how much room is left in virtual hard drive
13:50:15  <Ammler> didn't you define a "floating" drive
13:50:19  <Alberth> Ubuntu aims for the average window user, not for people doing actual development work :)
13:50:54  <FooBar> it also seemed a good start for the more advanced windows user
13:51:28  <FooBar> but then what do I use of it... the terminal window mostly
13:51:33  <Alberth> yeah, it attracts a lot of such users
13:51:50  <Alberth> that's so very non-ubuntu :p
13:52:17  <Alberth> you may be better off with a different linux distro
13:52:33  <FooBar> for even basic copy/paste I have to use the terminal, as every time it tells me that I'm not allowed and no User Account Control window pops up :P
13:52:53  <Alberth> ?
13:53:11  <FooBar> oh that's right, you don't use Windows...
13:53:12  <Alberth> you know that you don't have to use CTL+C ?
13:53:31  <FooBar> well, every time is a big word
13:53:56  <FooBar> but every time I need to upgrade grfcodec and nforenum, which both need to be in some bin dir that is root-only
13:54:05  <Alberth> ie select by dragging with left mouse button, paste with right
13:54:24  <Alberth> that sounds wrong
13:54:38  <Ammler> grfcodec and nforenum works also quite nice in userspace
13:54:46  <Alberth> you are doing a regular install?
13:54:48  <FooBar> I decided to put grfcodec in /usr/local/bin
13:55:03  <Alberth> use ~/bin instead
13:55:06  <FooBar> no install, just dumping the executable there
13:55:14  <FooBar> hmmm...
13:55:22  <Ammler> yes, for things, which you build regulary, you shouldn't use root
13:56:02  <FooBar> Alberth: there's no such thing in ubuntu, as "lusers" wouldn't need that
13:56:11  <Ammler> installation tests are made by our compile farm :-D
13:56:51  <Alberth> type 'echo $PATH'  and 'echo $HOME/bin'    is there in the former something that looks like the latter?
13:57:03  <Alberth> most likely, yes
13:57:21  <FooBar>  /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games
13:57:39  <Alberth> ouch
13:57:55  <FooBar> and the other one: /home/jasper/bin
13:58:01  <Alberth> are you sure you want to stay with ubuntu?  :p
13:58:19  <Alberth> echo "$SHELL"    gives you something 'bash' right?
13:58:36  <FooBar> if you know how much effort it cost me to get the virtual link to a dir in windows working: yes :D
13:58:53  <Hirundo> Alberth: I missed some hours of the discussion, what were the conclusions regarding e.g. ?
13:58:56  <FooBar>  /bin/bash
13:59:06  <Hirundo> Should we continue in that way or not?
13:59:17  <Alberth> I can imagine, it is one of the reasons I stay away from win* :)
13:59:29  <Ammler> FooBar:  you run echo $PATH as user?
13:59:39  <FooBar> Ammler: yes
13:59:46  <Ammler> then it is indeed crap
13:59:47  <Alberth> Hirundo: I think so
14:00:13  <Ammler> Alberth: but that is debian, I guess
14:00:27  <Ammler> on our debian server, there is also no ~/bin in path
14:00:40  <Ammler> suse of course has it :-P
14:00:42  <Alberth> does  ls ~/.bash_profile ~/.bashrc   give some files
14:01:44  <FooBar> first dir does not exist, second just gives the location
14:02:30  <Alberth> ok, open that file, add   "PATH=$PATH:$HOME/bin"  at the end probably
14:02:50  <Alberth> then create a new terminal, and ~/bin should be in your path
14:03:07  <Alberth> (for each new terminal that you open from then on)
14:03:14  <FooBar> will try
14:04:03  <Alberth> don't close the current terminal until you are sure it works :)
14:04:17  <Alberth> and if you are paranoid, make a copy of ~/.bashrc first :)
14:04:31  <Ammler> Alberth: what distro do you use, if I may ask?
14:04:45  <Alberth> fedora
14:04:54  <FooBar>  /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/home/jasper/bin
14:04:54  <Alberth> ie quite bleeding edge
14:05:00  <FooBar> so I think it works :)
14:05:12  <Alberth> nice
14:05:56  <Alberth> now you need a new executable to put in that directory :)
14:06:13  <Alberth> gtg now, bbl
14:06:20  <FooBar> ok, thanks!
14:06:35  <Ammler> or just symlink to the repo
14:06:52  <Alberth> that's what I do a lot yeah
14:06:53  <Ammler> ln -s ~/hg-repos/nforenum/nforenum ~/bin/
14:07:03  *** Alberth has left #openttdcoop.devzone
14:07:14  <FooBar> I'll try that then
14:07:38  <FooBar> again something new, althoug I've used junctions on NTFS before
14:09:39  <FooBar> now to empty /usr/local/bin
14:13:06  <FooBar> darn, "The program 'grfcodec' is currently not installed."
14:14:32  <Ammler> ls ~/bin
14:14:58  <Ammler> the shell you are in does already have ~/bin in $PATH?
14:17:24  <FooBar> echo $PATH gives me /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/home/jasper/bin, and ls /home/jasper/bin/grfcodec/grfcodec gives me /home/jasper/bin/grfcodec/grfcodec
14:17:30  <FooBar> is that readable?
14:17:39  <Ammler> oh
14:17:44  <Ammler> you symlinked the wrong
14:17:49  <Ammler> only symlink the binary
14:18:02  <FooBar> ah
14:18:08  <FooBar> I must've misread then
14:18:13  * FooBar tries again
14:18:18  <Ammler> my example should be right
14:18:41  <FooBar> yes, I should've read more carefully
14:20:09  <FooBar> hmm...i need to unsymlink first. Appearantly on linux a dir and a file is the same thing
14:20:35  <FooBar> can I just rm the symlink?
14:21:24  <FooBar> it appears so
14:21:44  <FooBar> hooray for zoidberg!
14:22:58  <Ammler> yeah, symlink is a file with a link
14:23:15  <Brot6> NewGRF Meta Language - Revision 619:2ebd8e9af6f9: Add: Action14 class and some helper classes for... (yexo) @
14:23:15  <Brot6> NewGRF Meta Language - Revision 620:bd7a6e20e7eb: Feature: write action14 with grf name/descripti... (yexo) @
14:24:19  <FooBar> still, I cannot have a normal folder and a file with the same name next to each other...
14:24:31  <FooBar> I must remember that for when developing grfs
14:25:07  <Ammler> hmm, really?
14:25:12  <Ammler> well, ugly anway
14:25:15  <FooBar> interesting... i cannot do that on Windows either...
14:25:37  <FooBar> I figured I could, but I guess I never actually came across that before
14:25:52  <FooBar> well, then I don't have to remember that
14:25:56  <FooBar> :)
14:26:56  <FooBar> even dos prompt cannot :)
14:27:08  <FooBar> so just ignore what I said then
14:27:20  <FooBar> but I agree, it would be a bit ugly
14:27:36  <Ammler> which palette is default, when I don't define it with Action14?
14:27:56  <Ammler> s/used/default/
14:27:57  <Yexo> I think the same as the baseset
14:30:11  <Brot6> NewGRF Meta Language - Revision 621:f6960a7379a7: Doc: Document replace-block. (Hirundo) @
14:32:03  <Brot6> NewGRF Meta Language - Revision 622:3e119ede7673: Codechange: Remove useless variable from ActionA (Hirundo) @
14:32:28  <Ammler> do I need to skip Action14 for older openttd?
14:32:58  <Ammler> it says, I should for ttdp
14:33:13  <Yexo> not sure, I don't think so
14:33:29  <Yexo> you don't have to skip it for older openttd
14:34:18  <Hirundo> For TTDP, one'd have to check the asm source
14:34:34  <Yexo> I tried, but I couldn't find the relevant source
14:35:28  <Ammler> I can try for you, if you want
14:35:39  <Ammler> but first, I need to complete my baseset :-)
14:36:17  <FooBar> what baseset are you making? Hopefully not the same that I intend to make...
14:36:33  <Ammler> @wiki Basesets
14:36:35  <Webster> Search results for "Basesets" - #openttdcoop Wiki -
14:36:38  <Ammler> not a basesets
14:36:45  <Ammler> @wiki BaseCosts
14:36:48  <Webster> BaseCosts - #openttdcoop Wiki -
14:37:07  <FooBar> ah, good. That's something totally different then
14:37:14  <Ammler> yeah :-P
14:38:53  <Ammler> FooBar: if you mention BaseSets, OpenGFX is still not finished
14:39:00  <Ammler> maybe you like to review there a bit
14:39:26  <Ammler> I hope, we didn't screw up all since you paused your development
14:39:41  <FooBar> I'm not drawing any new graphics, and am actually using OpenGFX as source for mine :)
14:40:04  <FooBar> And no, I didn't see screwups; you guys took over actually very well, I appreciate that
14:40:06  <Ammler> ah, so just a Fork?
14:40:38  <Ammler> I hope, it won't become a competitior set :-P
14:40:52  <FooBar> basically, yes, but not just any fork. The first test showed that it was 300 kB in total size, but I don't want to say much more at this point :D
14:41:44  <Ammler> I was thinking about making a minigfx, which has only gui, so it could be included to the openttd source like nosound
14:42:17  <FooBar> ok, you got me. That's basically what I am attempting. But don't tell anyone yet!
14:42:26  <Ammler> :-D
14:43:05  <FooBar> I have the idea for like 6 months now and this week finally attempted something
14:43:10  <Ammler> hmm, maybe simple "branch" opengfx?
14:43:52  <FooBar> I could do that
14:44:04  <Hirundo> Yexo / Ammler: It seems skipping is indeed required for TTDP
14:44:06  <Ammler> the Makefile does support hg branches
14:44:58  <FooBar> I have no repo yet; for the test I just decoded ogfx, made all graphics except some gui blue and then recoded it
14:45:23  <Ammler> well, you might have seen, I splitted the opengfx pnfos
14:45:30  <FooBar> For a branch, i think it's more useful to change the nfo instead
14:45:48  <Ammler> that way, it could be easy to keep a parallel mini set
14:46:11  <FooBar> exactly, because as graphics change, they change in mini set as well
14:47:51  <FooBar> that is, it's more easy to do the change
14:49:03  <Ammler> the same set could be used by dedi
14:49:05  <Brot6> NewGRF Meta Language - Revision 623:4536e85b8ad7: Codechange: check the types of the elements in ... (yexo) @
14:49:05  <Brot6> repository /home/ottdc/hg-repos/basecosts registered in Redmine with url /home/ottdc/hg-repos/basecosts
14:49:05  <Brot6> repository /home/ottdc/hg-repos/basecosts created
14:50:56  <Brot6> NewGRF Meta Language - Revision 624:0abdfe0a91e0: Doc: Document ActionA. (Hirundo) @
14:52:46  <Hirundo> hmm... reduce_constant can also result in a float?
14:53:06  <Hirundo> I think that would be an unexpected result in around 90% of the cases
14:53:12  <Yexo> yes, indeed
14:53:20  <Yexo> I noticed it before but haven't yet done anything about it
14:54:14  <Hirundo> Are floats useful anywhere outside action0?
14:55:06  <Yexo> not sure, but I don't think so
14:59:42  <Hirundo> reduce_constant should return only integer constants, as changing all non-action0 code would be a lot of work
15:04:14  <Yexo> I agree
15:06:12  <Brot6> NewGRF Meta Language - Revision 625:0bdf71d48a5c: Fix: protect the action14 by an action9 for ttd... (yexo) @
15:24:44  <Brot6> NewGRF Meta Language - Revision 626:bd4a657972b3: Codechange: Shuffle some ActionA / Action5 code... (Hirundo) @
15:34:45  <Ammler> the whole basecosts part isn't supported by NML yet?
15:36:57  <Hirundo> no
15:37:15  <Ammler> ok, then I don't need to convert that minigrf anyway :-)
15:42:05  <Brot6> repository /home/ottdc/hg-repos/basecosts-presets registered in Redmine with url /home/ottdc/hg-repos/basecosts-presets
15:42:05  <Brot6> repository /home/ottdc/hg-repos/basecosts-presets created
15:44:33  <Hirundo> basecosts is 'just' one action0 property, right?
15:44:43  <Yexo> yes
15:47:20  <Brot6> NFORenum - Bug #1175 (New): Bridge IDs 0B and 0C not accepted (foobar) @
15:48:23  <Ammler>
15:49:10  <Ammler> prop 08
15:50:53  <Hirundo> That shouldn't be too hard
15:55:47  <Ammler> I guess, I still don't know, when I need "export FOO=bar" and when "FOO=bar" is fine
15:57:24  <Rubidium> FooBar: do you have a piece of nfo for me to test with?
15:57:41  <FooBar> the bridge stuff?
15:58:17  <Rubidium> yes
15:59:13  <FooBar> here's a nice one, without new graphics:
15:59:38  <FooBar> If you want something smaller, I can attempt to create something
16:00:10  <Rubidium> thanks
16:00:30  <FooBar> you can also try this: -1 * 0 00 06 01 01 09 0C 03
16:00:46  <FooBar> no wait...that's not good
16:01:08  <FooBar> this: -1 * 0 00 06 01 01 0C 0A 03
16:02:37  <Rubidium> you can try this:
16:02:54  * FooBar waits
16:02:59  * Rubidium waits as well
16:03:02  <FooBar> :)
16:03:33  <FooBar> I already received the email for issue closed :P
16:03:43  <Brot6> NFORenum - Bug #1175 (Closed): Bridge IDs 0B and 0C not accepted (foobar) @
16:03:43  <Brot6> NFORenum - Revision 417:4d5eb7a59426: Fix: bridge IDs 0B and 0C were not accepted, even though th... (Rubidium) @
16:03:43  <Brot6> NFORenum - Bug #1175 (Closed): Bridge IDs 0B and 0C not accepted (Rubidium) @
16:03:49  <FooBar> ah, there we go :)
16:06:26  <FooBar> I built the same thing I already had :S
16:08:10  <FooBar> crap, post-pull had not update specified
16:09:51  <FooBar> Rubidium: works!
16:10:16  <FooBar> and thanks :)
16:11:05  * Ammler awaits again rebuild boom in 10 mins :-)
16:11:19  <Rubidium> with a lot of extra errors today
16:13:47  <Brot6> FIRS Industry Replacement Set - Bug #1176 (Confirmed): Faulty tile on Aluminium Plant (foobar) @
16:19:11  <Brot6> 32bpp-extra: update from r36 to r38 done (1 errors) -
16:19:32  <Brot6> basecosts: update from  to r6 done -
16:20:10  <Rubidium> 32bpp-extra Makefile doesn't have md5, should the rpm spec then call make md5?
16:20:38  <Ammler> I assume, the 32bpp doesn't have a custom spec
16:20:40  *** PeterT has joined #openttdcoop.devzone
16:21:08  <Rubidium> dunno
16:21:32  <Brot6> nforenum: update from r412 to r417 done -
16:22:12  <Brot6> nml: compile of r626 failed -
16:22:25  <Brot6> Following repos didn't need a nightlies update: 2cctrainset (r573), airportsplus (r52), comic-houses (r71), firs (r1076), fish (r386), grfcodec (r188), heqs (r371), metrotrackset (r40), newgrf_makefile (r124), nutracks (r86), ogfxplus (r41), opengfx (r469), openmsx (r91), opensfx (r97), snowlinemod (r15), swedishrails (r140), worldairlinersset (r659)
16:23:09  <Rubidium> nml regression "failure"
16:23:52  <Ammler> then we have a bit less rebuild spam :-)
16:28:08  <Ammler> none?
16:28:24  <Rubidium> then something went wrongk
16:28:35  <Ammler> yep :-(
16:28:58  <Ammler> osc is running...
16:29:04  <Brot6> NewGRF Meta Language - Revision 627:129528db1ad9: Fix (r626): regression test failed due to diffe... (yexo) @
16:29:40  <Brot6> nutracks: rebuild of r86 done (180 errors) (Diffsize: 243) -
16:30:16  <Ammler> now a diff of the diff :-)
16:30:23  <Ammler> s
16:32:27  <Brot6> Following repos rebuilds successful without any difference to earlier nightlies builds: 2cctrainset (6 errors), 32bpp-extra (1 errors), airportsplus, basecosts, comic-houses (3 errors), firs (6 errors), fish (6 errors), heqs, metrotrackset (6 errors), newgrf_makefile, opengfx, snowlinemod, worldairlinersset
16:32:43  *** Seberoth2 is now known as Seberoth
16:33:03  *** Alberth has joined #openttdcoop.devzone
16:34:48  <Ammler> is the nutracks now simply lack of update
16:34:59  <Ammler> or are there still missing things in nforenum?=
16:35:22  <Rubidium> most NFO! errors are due to action2 with railtypes
16:35:52  <Rubidium> the first 4 NFO warnings are entirely valid, but the majority (say some 160+ errors) is pure whiteness
16:37:46  <Rubidium> the diff is completely due to still that got implemented/fixed in nforenum
16:40:53  <Ammler> is it possible make a diff without dates?
16:41:13  <Ammler> that is the only output of a diffdiff
16:42:06  <Rubidium> you could use touch to set some time stamp on the files
16:43:07  <Alberth> replace the dates with fixed text using sed
16:43:10  <Yexo> for hg you can set "nodates = True" in .hgrc (in the [diff] section), but that might not be relevant in this case
16:43:47  <Ammler> yeah, using diff the tool, seems not have such a option
16:44:47  <Alberth> diff doesn't know what it is diffing
16:45:04  * Rubidium ponders doing some fixes of nutracks, but I'm not sure how DJNekkid handles his graphics
16:45:20  <Ammler> in what case?
16:45:33  <Rubidium> nutracks
16:45:39  <Rubidium> whether I can just touch the pcx files
16:45:59  <Rubidium> maybe he has some pngs or something at home
16:46:22  <Ammler> they are usually int he repo as well then
16:46:28  <Brot6> Base Costs Mod - Revision 2:ef39b0296c1d: Add: version detection (Ammler) @
16:46:28  <Brot6> Base Costs Mod - Revision 3:38c6f457c77c: Rename makegrf to (Ammler) @
16:46:28  <Brot6> Base Costs Mod - Revision 4:d6892cd9e564: Change: dedicated version detection, renum -> nforenum (Ammler) @
16:46:28  <Brot6> Base Costs Mod - Revision 5:0eaecd83e3b7: DevZone: enable compiler with type bash (Ammler) @
16:46:32  <Brot6> Base Costs Mod - Revision 6:a885e917e24c: Feature: generate md5sum of the grf (Ammler) @
16:46:35  <Brot6> Base Costs Mod - Presets - Revision 5:f7cb5b1b12cb: Change: rename script and replace with dedica... (Ammler) @
16:46:39  <Brot6> Base Costs Mod - Presets - Revision 6:324bdd89d15b: [lib] Add: version detection, build grf (Ammler) @
16:46:43  <Brot6> Base Costs Mod - Revision 7:3cd5681638f7: Added tag 3.0 for changeset a885e917e24c (Ammler) @
16:47:00  <Ammler> autsch...
16:49:16  <Ammler> DJNekkid: is happier about someone doing it than just blaming him not doing it :-)
16:49:55  <Ammler> or blaming him, doing it wrongly
16:53:32  <Brot6> Nutracks - Revision 87:00bb0925faa3: Fix: cpp redefinition warning (Rubidium) @
16:53:32  <Brot6> Nutracks - Revision 88:def5c69828e7: Fix: when defining 18h props, also tell you've got 18h props... (Rubidium) @
16:55:28  <Yexo> 18, not 18h (fix is correct, msg is not)
16:55:59  <Rubidium> oh.. too bad :)
17:02:22  *** ODM has quit IRC
17:07:57  <Brot6> basecosts: update from  to 3.0 done -
17:08:22  <Brot6> Base Costs Mod - Revision 9:1f8f2b69edc0: Added tag 3.0 for changeset f75866807e00 (Ammler) @
17:08:22  <Brot6> Base Costs Mod - Revision 10:5406eee24a05: DevZone: s (Ammler) @
17:11:58  <Brot6> basecosts: rebuild of 3.0 done (Diffsize: 3) -
17:15:27  <Rubidium> Yexo: do we have a clue how to handle the action3 default ID stuff?
17:15:37  <Rubidium> for railtypes
17:16:30  <Brot6> Nutracks - Revision 89:e4ae6d58c3e3: Fix: lots of pure white pixel warnings (Rubidium) @
17:16:30  <Brot6> Nutracks - Revision 90:e79aaf940d73: Change: the new NFORenum handles railtypes better, so remove... (Rubidium) @
17:18:13  <Yexo> Rubidium: there are two options:1. change nforenum to completely ignore the action3 default 2. do nothing because the default will in the future be used for railtype callbacks (eg railtype availability)
17:19:28  <Rubidium> okay, lets go for #2: NFOrenum is right
17:27:43  <Ammler> hmm, the openmsx theme music is a bit too short
17:30:15  <Rubidium> Yexo: so the default ID has to "point" to some action2 callback handling chain, right?
17:30:35  <Yexo> yes
17:32:51  <Brot6> basecosts-presets: update from  to 1.4.1 done -
17:33:05  <Rubidium> why so many releases?
17:34:19  <Brot6> Base Costs Mod - Presets - Revision 7:acebf8996074: [lib] Update hg.version with commit date (Ammler) @
17:34:19  <Brot6> Base Costs Mod - Presets - Revision 8:df92cf19a4f7: Change: use commit date instead current date (Ammler) @
17:34:19  <Brot6> Base Costs Mod - Presets - Revision 10:a14cf3060e73: Added tag 1.4.1 for changeset e960481fc769 (Ammler) @
17:35:11  <Ammler> different grfs
17:37:15  <Brot6> NFORenum - Support #1165: action3 railtypes default ID (Rubidium) @
17:37:22  <Ammler> hmm, md5sum doesn't match with bananas here
17:47:11  <Yexo> <- nml yntax proposal for action14 parameters
17:51:58  <Ammler> can't you indent it ot the first and then trash param:0
17:53:54  <Yexo> indent it ot the first?
17:54:11  <Yexo> and no, param:0; can't be trashed because it is the param number where to store the value
17:57:15  <Brot6> Nutracks - Patch #1177 (New): Action 14 support (Rubidium) @
17:57:45  <Ammler> Yexo: isn't first setting the defination of param0 and the 2nd setting a part of that?
17:58:17  * Ammler thinks, you do that too much nfo related :-)
18:00:40  <Yexo> Ammler: you mean like this?
18:02:19  <Yexo> <- new version again
18:02:40  <Rubidium> parameter description?
18:02:59  <Yexo> yes
18:03:37  <Rubidium> am I missing that, or haven't you added it yet?
18:04:08  <Yexo> you're missing it
18:04:12  <Ammler> isn't setting_bool part of setting_num?
18:04:24  <Yexo> <- see there
18:04:34  <Yexo> the "when finished ..." string is the parameter description
18:04:40  <Yexo> the name is "Economy"
18:04:49  <Yexo> and the name for value 0 is "Test Economy"
18:06:21  <Yexo> Ammler: what do you mean?
18:06:49  <Yexo> 3 bits starting at bit 0 means bit 0, 1 and 2
18:06:53  <Ammler> don't you define the bitset with setting_num?
18:07:04  <Ammler> and then with setting_bool one of this bits?
18:07:37  <Yexo> the setting_num defines a setting that is stored in parameter0 bits 0,1,2
18:07:47  <Yexo> the settign_bool defines another setting that is stored in parameter0 bit 3
18:08:40  <Yexo> <- bigger example with the same syntax
18:08:47  <Ammler> then it ok, sorry for my confusion :-)
18:53:46  *** ODM has joined #openttdcoop.devzone
19:11:54  <Brot6> FIRS Industry Replacement Set - Bug #1176 (Confirmed): Faulty tile on Aluminium Plant (foobar) @
19:15:51  <FooBar> Hickup ^? As it already announced that at 18:13...
19:16:17  <Alberth> yeah, Brot6 likes to repeat messages every now and then
19:16:50  <Alberth> FooBar: does the 'real' patch work (both for 'true' and 'false' for DO_DEP_CHECK) ?
19:17:06  <FooBar> actually it does work very nicely
19:17:19  <FooBar> so many thanks for that!
19:17:39  <Alberth> ok, I'll submit the patch then (if I am allowed to do that)
19:17:42  <Alberth> thanks for testing
19:17:55  <FooBar> for "true" I haven't really tested to be honest
19:18:00  <FooBar> but false works
19:18:12  <FooBar> I can test "true" now if you like
19:19:05  <FooBar> that works as well ;)
19:20:32  <Alberth> true should revert to the old behaviour of course :)
19:21:41  <FooBar> obviously :)
19:22:19  <FooBar> I had to wait longer and it said that it was generating dependency file ;)
19:23:26  <Alberth> ok, I broke nothing thus :)
19:24:37  <Brot6> Example NewGRF Project - Code Review #1178 (New): Make dependency checking optional (Alberth) @
19:24:52  <Alberth> done :)
19:26:50  <Alberth> hmm, I may have broken the revision file
19:27:11  <FooBar> I still get one
19:27:17  <FooBar> with correct number
19:27:43  <Alberth> remove the file, and set the DO_DEP_CHECK to false
19:28:14  <Alberth> although if it is only needed for dependency calculations, it is not a problem
19:29:09  <FooBar> I got a new one
19:29:26  <Alberth> ok, so it gets computed correctly. good
19:29:38  <Alberth> I really didn't break anything :p
19:30:58  <FooBar> I'll add the modified makefile to the repo once I've finished replacing all sprites and am able to actually commit and push something to transrapidtrackset; that way tommorows nightly will be ultimate test :)
19:31:12  <Ammler> what's wrong here:
19:32:09  <Rubidium> ran it through nforenum?
19:34:18  <Ammler> yep, that is after nforenum
19:35:35  <Ammler> as soon as I add "C" "PARA"
19:36:45  <Rubidium> hmm... something's fishy there
19:37:29  <Rubidium> oh, you say you have 1 parameter, but you declare the 5th parameter
19:38:10  <Rubidium> "C" \d4
19:38:18  <Rubidium> -> "C" \d0 and it works
19:39:23  <Ammler> yes, thanks, thought that is the amount of different values
19:39:51  <Ammler> hmm, there is a glitch
19:40:01  <Ammler> maybe not :-)
19:40:44  <Rubidium> it say setting number, not number of settings
19:44:38  <Brot6> Base Costs Mod - Presets - Revision 10:b7290e241758: DevZone: enable release compiler (Ammler) @
19:44:58  <Ammler> works nice :-)
19:45:30  <Ammler> awesome :-)
19:46:04  <Rubidium> don't forget to skip the action if TTDP is used
19:46:43  <Ammler> not sure, if the set supports ttdp anyway
19:56:39  <Ammler> default value should be somehow possible
19:56:47  <Ammler> I set those with ActionD yet
20:20:11  <Brot6> TransRapid Track Set - Bug #1179 (New): railtypes construction buttons (foobar) @
20:29:20  <Brot6> TransRapid Track Set - Bug #1180 (New): fences not removed by default if railtypes (foobar) @
20:34:20  <Brot6> TransRapid Track Set - Bug #1181 (New): stations missing (foobar) @
20:34:20  <Brot6> TransRapid Track Set - Bug #1182 (New): parameters don't do what they should do (foobar) @
20:36:05  <Brot6> TransRapid Track Set - Feature #1183 (New): snowy depot for railtypes (foobar) @
20:37:29  <Rubidium> looks like someone is really busy todat with filing reports
20:37:50  <FooBar> only what comes to mind :)
20:38:02  <FooBar> it's like a sticky note
20:43:55  <Brot6> TransRapid Track Set - Revision 0:ff135ab310de: Populate repo: with code based on r40 of metrotra... (foobar) @
20:45:08  <Ammler> hmm, I get an error with TTDP also with skip of Action14
20:45:27  <Rubidium> ask Yexo :)
20:45:36  <Rubidium> how did you skip it?
20:46:48  <Yexo> <- if TTDPatch gives an error during the "Load" stage there is nothing you can do about it
20:47:16  <frosch123> you cannot even skip it with action 9 :o
20:47:50  <Yexo> if ttdpatch works as it says on that page and really _ignores_ all actions other than those listed it shouldn't give an error at all
20:50:05  <Ammler> # Datei "newgrf/basecosts-presets.grf" hat ungţâ╝ltiges Sprite #2 (code 2/1).
20:51:40  <frosch123> hmm, pseudospriteaction checks it everytime the grf is processed
20:51:43  <frosch123> so in every stage
20:51:44  <Rubidium> so file a bug report in ttdpatch's tracker
20:52:23  <Yexo> I thought that every program loading a newgrf should ignore all unknown actions, but I can't find that anywhere in the ttdpatch wiki
21:10:21  <Brot6> basecosts-presets: compile of  failed -
21:10:49  <Brot6> Base Costs Mod - Presets - Revision 11:1b1c3df6d4d6: [lib] Feature: new var REVISION for dev and ... (Ammler) @
21:10:49  <Brot6> Base Costs Mod - Presets - Revision 12:57a9c7f50e7f: Feature: Action14 (Ammler) @
21:11:11  <Brot6> basecosts-presets: update from 1.4.1 to 1.4.2 done -
21:12:42  <FooBar> good night everyone
21:12:49  *** FooBar has quit IRC
21:13:01  <Brot6> Base Costs Mod - Presets - Revision 13:584d6efc13b8: Added tag 1.4.2 for changeset 57a9c7f50e7f (Ammler) @
21:22:12  <Ammler> if you need a demo for Action14 on banans ^ :-)
21:33:58  <frosch123> he, the default 0 is no valid value :p
21:34:28  <frosch123> what does the grf do? disable with error "you need to select a profile!" ?
21:46:03  *** Alberth has left #openttdcoop.devzone
21:54:11  <Ammler> frosch123: default 0?
21:54:34  <Ammler> [21:56] <Ammler> default value should be somehow possible
21:54:38  <frosch123> if you freshly add the cost preset grf to the list
21:55:13  <Ammler> workaround would be to use 0 as default :-)
21:55:15  <Yexo> a problem with setting a default is that is breaks the "this parameter was set by the user" feature
21:57:36  <Ammler>
21:57:37  <Webster> Title: plasma-desktopSa3974.jpg (at
21:58:01  <Ammler> why does it show "0" there?
21:58:07  <Ammler> also min is 1
22:00:43  <Yexo> because the current value is 0
22:00:50  <Yexo> it'll only be changed once the user changes it
22:01:05  <Ammler> but ActionD does set it
22:01:13  <Yexo> not yet
22:01:24  <Yexo> the default is 0 for all parameters
22:01:33  <Yexo> you'll never see the result of ActionD in this window
22:02:01  <Yexo> there are actually two lists of parameters, the list as set by the user and the list as changed by ActionD code
22:02:24  <Yexo> that is, in OpenTTD, in TTDPatch there is afaik only a single list
22:03:14  <Ammler> <-- sprite 3
22:03:45  <Ammler> well, it might not matter
22:05:21  <Ammler> is "unset" the same as 0 then?
22:07:55  <Yexo> no
22:08:12  <Yexo> there is one variable that records the maximum parameter that was set
22:08:30  <Yexo> but when unset, it's always 0
22:11:39  <Brot6> Backup transrapidtrackset: warning: repository is unrelated
22:17:24  <Ammler> hmm
22:18:33  <Rubidium> oh... the first #openttdcoop.devzone divorce!
22:22:14  <Brot6> TransRapid Track Set - Code Review #1184 (New): backup.push to the same repo :-P (Ammler) @
22:56:40  <ODM> divorce? any lawyers?

Powered by YARRSTE version: svn-trunk