Log for #openttdcoop.devzone on 7th March 2012:
Times are UTC Toggle Colours
04:29:31  *** andythenorth has joined #openttdcoop.devzone
04:36:43  *** andythenorth has quit IRC
06:40:33  *** JVassie has joined #openttdcoop.devzone
06:48:58  *** Zuu has joined #openttdcoop.devzone
07:05:28  *** Zuu has quit IRC
07:22:10  *** JVassie has quit IRC
07:59:56  *** andythenorth has joined #openttdcoop.devzone
09:16:20  <Brot6> BANDIT - Revision 370:faf477947a78: Codechange: default the bg colour of a spritesheet to white (andythenorth) @
09:16:20  <Brot6> BANDIT - Revision 371:afd16494471c: Codechange: improve coils template a little (andythenorth) @
09:16:20  <Brot6> BANDIT - Revision 372:28ce1d1105b9: Codechange: update existing Gestalts to current Pixa methods (andythenorth) @
09:19:23  *** welshdragon has quit IRC
09:19:25  *** welshdragon has joined #openttdcoop.devzone
09:33:12  <Brot6> BANDIT - Revision 373:0174c0454733: Codechange: move Gestalt IDs to top of Gestalt files (andythenorth) @
09:33:12  <Brot6> BANDIT - Revision 374:ce1085917738: Codechange: set correct hard-coded value for lengths (andythenorth) @
09:33:12  <Brot6> BANDIT - Revision 375:a6e96d8f9762: Codechange: comment in cargo generation (andythenorth) @
09:55:21  <Ammler> andythenorth: yes, entry_point broke the wrapper
09:56:19  <andythenorth> Ammler: did you try the previous rev of  (before entry_point)  - works for me
10:10:14  <Ammler> andythenorth: I guess, I already told multiple times that the nmlc wrapper before buildout was added works :-)
10:11:35  <Ammler>
10:11:42  <andythenorth> so revert / patch, add back any changes, problem solved :)
10:13:00  <Ammler> nah, I don't want to revert buildout, I didn't add it
10:13:52  <Ammler> right now, I siimply overwrite the nmlc from with the one from the source, which works too, might look ugly
10:14:33  <Ammler> but that way, I didn't have to touch the nml source, only build script
10:15:43  <Ammler> welshdragon: fix your irc bouncer :-P
10:20:56  <welshdragon> Ammler: should be fixed now :P
10:21:36  <andythenorth> planetmaker: Yexo any objections if I remove the buildout stuff from nml?
10:22:14  <Ammler> andythenorth: you could just comment on the ticket that you think buildout is  useless as it is now
10:22:20  <Ammler> let Yexo finally do it
10:22:56  <Ammler> according to the commit, it isn't just for osx, also MSYS was a reason
10:23:15  <Ammler> #3786
10:23:15  <Brot6> Ammler: #3786 is "NewGRF Meta Language - Bug #3786: runtime requirement for setuptools - #openttdcoop Development Zone"
10:23:38  <andythenorth> MSYS is solved by distributing a binary
10:23:52  <andythenorth> buildout is neither os x specific
10:24:16  <andythenorth> but it's just the wrong tool for this project (as none of us can maintain it)
10:26:22  <Brot6> NewGRF Meta Language - Bug #3786: runtime requirement for setuptools (Ammler) @
10:34:35  <andythenorth> Ammler: you're also going to move the install dependency from setuptools to Distribute?
10:34:37  <Brot6> NewGRF Meta Language - Feature #3581: Consider removing buildout (andythenorth) @
10:35:37  <Ammler> andythenorth: no, as older distros don't have that
10:37:39  <andythenorth> meh
10:37:43  <andythenorth> hysterical raisins :P
10:41:44  <Ammler> you should indeed not waste time with this issue, anyway, thank you :-)
10:42:17  * andythenorth hates cruft :)
10:42:20  <andythenorth> biab
10:42:26  *** andythenorth has quit IRC
12:05:39  *** andythenorth has joined #openttdcoop.devzone
12:27:55  <andythenorth> planetmaker: comic set :)
12:28:00  <andythenorth> guess what I'm going to suggest....
12:34:26  <planetmaker> machine-drawing?
12:36:21  <andythenorth> :)
12:36:41  <planetmaker> tbh, I find that intimidating ;-)
12:36:43  <andythenorth> it's very simple block colour
12:36:53  <andythenorth> wonder if we could teach it to draw black lines automatically
12:36:54  <planetmaker> but for comic it might fit best even
12:37:03  <andythenorth> needs edge detection :o
12:37:13  <planetmaker> a simple sobel filter and you're set
12:37:48  <planetmaker> just draw everything that filter returns in black (given a certain threshold)
12:37:53  <andythenorth> ho ho
12:37:59  <andythenorth> really might work
12:41:57  <andythenorth>
12:41:58  <Webster> Title: python - Applying the Sobel filter using scipy - Stack Overflow (at
12:43:13  <planetmaker> hm... maybe one wants to thin the result. But that will need judgement of what it gives in its pure form
12:53:59  <andythenorth> planetmaker: project for the future...after the other projects ;)
12:58:07  <planetmaker> :-)
14:14:49  *** ToniBabelony has joined #openttdcoop.devzone
14:15:03  <ToniBabelony> Hello all, I have a question regarding NML
14:16:34  <ToniBabelony> How can I make an assignment in the item/property block read another property, let's say 'speed'?
14:18:10  <planetmaker> not directly
14:18:56  <ToniBabelony> hmm, I already thought so
14:20:20  <planetmaker> you can do like
14:20:26  <planetmaker> somevar = 42
14:21:00  <ToniBabelony> I want to make a formulae that calculates the running_cost according to the given values in cargo_capacity, loading_speed, weight and speed. Is there another way, other than adjusting the values manually?
14:21:22  <planetmaker> item (FEAT_TRAINS, trainname) { properties { speed: somevar; weight: somevar * 2; } }
14:22:14  <planetmaker> you will want to re-use the variable 'somevar' for every train
14:22:55  <ToniBabelony> I see.
14:23:39  <ToniBabelony> So, I do have to adjust everything manually.
14:23:51  <planetmaker> re-use means you can of course assign different values between the different train definitions
14:23:59  <planetmaker> well... no?
14:24:22  <planetmaker> isn't this what you want?
14:26:06  <planetmaker> But people like Eddi, andy us a table with the values they want to define and then a script which takes these values and writes the proper nml from those in a scripted fashion
14:26:44  <planetmaker> s/us /use /
14:27:08  <ToniBabelony> So, I have to write something like this in the purchase_running_cost_factor callback: item(trainname) {properties {speed * weight} }
14:27:45  <ToniBabelony> I was thinking about a value table as well, but I want to try other options first
14:28:11  <planetmaker> ok, oh, so you want to access the properties in a callback?
14:28:31  <ToniBabelony> yep (sorry for being not clear enough)
14:29:25  <ToniBabelony> *i'm still an apprentice in NML and programming in general*
14:30:35  <planetmaker> Well, (unfortunately) it doesn't really change the answer...
14:31:23  <planetmaker> <-- you have these variables at your disposal. You'd need to define the rest hard-coded in a callback
14:32:27  <ToniBabelony> Ah ok, I thought so.
14:33:26  <andythenorth> ToniBabelony: you could template them in quite simply with python
14:33:42  <planetmaker> looking at the screens you posted, you have a whole fleet of different vehicles... thus some templating might make sense. But well :-)
14:33:57  <andythenorth> ToniBabelony: python templating is definitely one step up, you'd need to invest i
14:33:57  <andythenorth> n
14:34:03  <andythenorth> learning it
14:34:04  <planetmaker> If you use a structured way to order your file(s) putting it there manually is not a bad choice
14:34:05  <ToniBabelony> So, I have max_speed, cargo_capacity at my disposal
14:34:16  <planetmaker> It has the added advantage that you can also fine-tune values
14:34:31  <ToniBabelony> Python is maybe a next step in programming
14:35:32  <andythenorth> ToniBabelony: it's worth it, but you will go better if you're comfortable with NML first
14:36:24  <planetmaker> yep. It's good to get one NewGRF completely working IMHO before trying to make it easier to safe work
14:36:45  <ToniBabelony> Yeah, I noticed that, but it's a bit too late for that now.
14:37:20  <ToniBabelony> However, if I can manage to make it work somehow, you can find it in the source in the next release!
14:37:39  <ToniBabelony> Thanks so far :)
14:37:43  <planetmaker> well. Your NewGRF kinda works, doesn't it?
14:37:58  <ToniBabelony> Well, it works, but it's still unbalanced
14:38:21  <ToniBabelony> Plus, I want to have some kind of standard, since I'm planning more NewGRFs like TIAS
14:38:26  <planetmaker> yes. All I say is: don't try to completely re-write the code basis in order to fine-tune it
14:38:54  <planetmaker> Get out a balanced version you're somewhat comfortable with in the manual adjustment way
14:39:18  <planetmaker> If you're then still interested: then go for refactoring and making it such that the properties can be scripted from whatever comfortable table
14:39:42  <planetmaker> And yes, I also suggest that and especially when you try to make more NewGRFs :-)
14:39:51  <planetmaker> It's good to have something finished ;-)
14:40:06  <ToniBabelony> I know man :)
14:40:35  <ToniBabelony> I'll first try to learn all the ins and outs from NML and then maybe go on with scripting.
14:41:03  <ToniBabelony> NML is the first programming language I'm learning by the way...
14:41:36  <planetmaker> :-)
14:41:57  <planetmaker> Assuming that what you released with TIAS so far is all in NML you're doing quite fine :-)
14:42:29  <ToniBabelony> After this I want to learn JavaScript and Ruby (for commercial reasons)
14:42:43  <ToniBabelony> but firstly NML
14:43:51  <andythenorth> ToniBabelony: +1 to what planetmaker said wrt manual stuff
14:44:03  <andythenorth> HEQS now uses templating, but was for a long time, one very big nfo file
14:44:08  <andythenorth> and the grf is still fun :)
14:44:39  <andythenorth> ToniBabelony: also, I am offering python tutorialising should you be interested ;)
14:44:49  <ToniBabelony> HEQS is awesome still :)
14:45:00  <ToniBabelony> I love the cargo trams
14:46:26  <planetmaker> HEQS totally fills a niche no other NewGRF even comes near
14:47:14  <planetmaker> whatever scenario, it barely ever is wrong to add HEQS
14:47:54  <ToniBabelony> I now have something like this: purchase_running_cost_factor:	(((cargo_capacity + cargo_count)/2) * (1/max_speed) * (num_vehs_in_consist)); that should work :3
14:49:19  <planetmaker> running cost inversely proportional to speed? That sounds... wrong
14:49:55  <planetmaker> and what is cargo_capacity vs. cargo_count?
14:51:05  <ToniBabelony> Well, I'm not suer what I'm going to do with it, but the main point is that I now have callbacks to property values
14:51:08  <ToniBabelony> *sure
14:51:38  <ToniBabelony> plus, higher speeds = more expensive
14:52:12  <planetmaker> 1/max_speed means higher speeds = less expensive
14:52:24  <ToniBabelony> I know, my maths are very very rusty
14:52:38  <ToniBabelony> I'll have to check on that
14:54:07  <ToniBabelony> cargo_capacity + cargo_count (max. capacity + number of capacity at that moment of travel = value for when trains are used more = more expensive)
14:54:48  <ToniBabelony> rather or something like that :P
14:55:14  <ToniBabelony> When playing running cost become variable according to how much that train is used by passengers
14:56:31  <ToniBabelony> anyway, I'm going to experiment a bit with this
14:56:46  <ToniBabelony> thanks for helping me out on this! It has been bothering me for days
14:59:14  <planetmaker> ToniBabelony, for all those features, you should ask yourself though what they really add to the game :-)
15:00:09  <planetmaker> the added gameplay effect might be quite marginal compared to the invested time and effort
15:00:38  <planetmaker> those features = complicated running cost dependency
15:00:52  <ToniBabelony> true, but still... TIAS is an extreme niche NewGRF, so I suppose I could add those things to it (or make them parameter dependent)
15:01:51  <planetmaker> I don't mean nor say that it's bad. Just that you might think about spending rather time on more visible features ;-)
15:02:12  <planetmaker> Few people will notice nor care about running costs changing between loaded and unloaded or standing and running
15:02:32  <ToniBabelony> True that, but the visible part is already okay
15:02:35  <planetmaker> just my 2ct about complicated running costs :-)
15:03:07  <planetmaker> it's nothing I, as player, could change either. Nor actually decide upon on purchase
15:03:47  <ToniBabelony> That's why I could make a parameter for it ;)
15:04:02  <planetmaker> If no-one cares, a parameter won't change that either ;-)
15:04:13  <ToniBabelony> The costs are also dependent on the Japan Set's settings
15:04:25  <planetmaker> Yes, but that's a constant multiplicator, I assume
15:04:34  <ToniBabelony> Yep, a duplicator
15:05:27  <ToniBabelony> Very easy, but I don't want to think a lot about running costs in the future as well, so I'll rather have them automated according to weight and such things
15:05:53  <ToniBabelony> Something I can be lazy about in other projects :P
15:10:34  <planetmaker> yep
15:20:53  <andythenorth> ToniBabelony: L127 here is how I do run costs in BANDIT, this is the python script that builds the set
15:21:00  <andythenorth> it's a *really* stupid calculation :P
15:21:01  <andythenorth>
15:21:25  <andythenorth> but the point is I can improve it when I figure out what's better, and it will all just update everywhere used
15:23:20  <ToniBabelony> I see, but that's beyond my capacity for now
15:23:50  <andythenorth> sure
15:23:58  <andythenorth> one step at a time ;)
15:23:59  <ToniBabelony> for now, i'd like to keep it NML pure
15:24:13  <ToniBabelony> first master NML and then something else
15:24:24  <ToniBabelony> python isn't on my priority list as of yet
15:26:01  <andythenorth> template with Javascript :D
15:26:52  <andythenorth> use Jquery :o
15:27:44  * andythenorth is partly serious
15:28:08  <planetmaker> yet another template environment
15:28:10  <planetmaker> ;-)
15:28:24  <ToniBabelony> I was thinking about that as well
15:28:53  <ToniBabelony> I'm almost sure I'll use that in my next project
15:29:08  <ToniBabelony> my NML code is now a bit too long IMO
15:29:36  <ToniBabelony> Plus, JavaScript should be a bit easier to understand than python I reckon
15:31:27  * planetmaker is not convinced
15:32:01  * planetmaker would guess on "same level but different approach"
15:32:50  <ToniBabelony> fair enough ;)
15:33:24  <ToniBabelony> but JavaScript could prove a bit more useful for me in he end (web-applications, games and stuff)
15:37:06  <planetmaker> what makes you think python is not at least as well suited?
15:37:40  <planetmaker> (you know what bananas and the openttd pages are written in?) :-P
15:38:44  <planetmaker> I guess the difference is where the code runs: server or client-side :-)
15:39:50  <planetmaker> I really have no preference as I know little python and no JavaScript. Just honestly wondering :-)
15:53:57  <ToniBabelony> I see your point, but JavaScript is more mainstream, so I've chosen that as my next step in programming
15:54:48  <ToniBabelony> at least, mainstream in a sense that I see it all over the place (plus websites I use are mostly written with JavaScript included)
15:55:06  <ToniBabelony> anyway, I must be off now
15:55:21  <ToniBabelony> Cheers and thank you so much for helping me out!
15:55:22  <ToniBabelony> :)
15:55:26  *** ToniBabelony has quit IRC
15:56:14  <andythenorth> oh he's gone :o
15:56:19  <andythenorth> nvm
15:56:37  <andythenorth> he doesn't see python running the websites because it's backend :P
15:56:40  <andythenorth> but nvm
15:56:52  <planetmaker> yeah :-)
15:57:12  <andythenorth> I would award points to anyone who can sensibly template NML with javascript
15:57:17  <andythenorth> it's technically achieveable
15:57:20  <andythenorth> but might be no fun :P
16:03:08  *** frosch123 has joined #openttdcoop.devzone
16:09:46  *** orudge has quit IRC
16:12:43  *** orudge has joined #openttdcoop.devzone
17:15:44  <Brot6> nml: update from r1872 to r1876 done -
17:22:27  *** ODM has joined #openttdcoop.devzone
17:23:46  <Brot6> BANDIT - Bug #3787 (New): DevZone compile failed (compiler) @
17:25:46  <Brot6> dutchtrains: update from r290 to r291 done -
17:28:29  <andythenorth> hmm
17:28:38  <andythenorth> so...
17:28:41  <Brot6> ogfx-industries: rebuild of r132 done (Diffsize: 127099) (DiffDiffsize: 8) -
17:28:55  <andythenorth> Ammler: is setuptools present for the BANDIT repo?
17:29:01  <andythenorth> :)
17:29:38  <planetmaker> andythenorth, it was afaik only ever added to NML as that needs an installation routine
17:30:00  <planetmaker> the newgrf repos are not in a big need for that
17:30:15  <andythenorth> well....I have a new error for BANDIT (above) ;)
17:30:23  <andythenorth> and it's caused by a lack of setuptools
17:31:00  <andythenorth> or to be more precise, the lack of pkg_resources module distributed with setuptools
17:31:39  <planetmaker> yeah
17:31:52  <planetmaker> grf.spec in the .devzone dir needs adjustment
17:32:46  <planetmaker> or probably it needs creation
17:33:14  <planetmaker> you're using in your NewGRF modules which usually are not needed. And which then the CF does not install in the compile environment by default
17:33:21  <planetmaker> Or so I believe is the reason for this error
17:33:46  <planetmaker> OpenGFX has in .devzone/build/grf.spec a file which explicitly specifies the build requirements and steps
17:34:00  <planetmaker> with separate and "exclusive" build dependencies
17:34:29  <andythenorth> planetmaker: sounds 100% convincing to me
17:35:05  <planetmaker> you could try and copy that file and slightly adjust it
17:35:17  <planetmaker> and commit it to your repo
17:36:07  <andythenorth> I shall look ;)
17:36:33  <planetmaker> better ask Ammler about details, though :-)
17:41:12  <andythenorth> oh
17:41:16  <andythenorth> I see :)
17:45:23  <andythenorth> it's a custom rolled build script
17:46:46  <planetmaker> it's a build script for the OpenSuSE build service basically
17:53:05  <Brot6> BANDIT - Bug #3787: DevZone compile failed (andythenorth) @
18:03:59  <Brot6> Central European Train Set - Revision 653:cb38ae564aeb: Fix: Also use whistle sound for non-steam... (oberhuemer) @
18:17:40  <Brot6> cets: update from r652 to r653 done (187 warnings) -
18:19:04  <Brot6> foobarstramtracks: rebuild of r23 done (Diffsize: 28769) (DiffDiffsize: 11) -
18:20:09  <Brot6> bandit: compile of r375 still failed (#3787) -
18:28:28  <Brot6> cets: update from r652 to r653 done (187 warnings) -
18:29:23  <Brot6> bandit-withoutsetuptools: compile of r366 failed -
18:31:44  <Ammler> andythenorth: we should add setuptools as require to Chameleon?
18:32:05  <Ammler> or is Chamelon also fixeable so it doesn't need it?
18:33:26  <andythenorth> unlikely
18:33:42  <andythenorth> the issue will just migrate upwards to module xyz
18:34:04  <Ammler> nono, pil runs well without
18:34:07  <Ammler> as you see on nml
18:34:20  <andythenorth> hmm
18:34:21  <andythenorth> ok
18:34:32  <Brot6> dutchtramset: rebuild of r113 done (21 warnings) (Diffsize: 20613) (DiffDiffsize: 11) -
18:35:16  <andythenorth> you're installing PIL from a package, not with
18:36:52  <Ammler> I guess, you guys still don't get the difference :-P
18:37:07  <Brot6> make-nml: compile of r0 still failed (#3730) -
18:37:08  <andythenorth> not really no
18:37:14  <Ammler> andythenorth: how do you think, should a package be made?
18:37:18  <andythenorth> I've no idea :)
18:37:32  <andythenorth> I don't know what a package is :)
18:37:43  <Ammler> we install it and then zip the files to a bundle
18:38:00  <Ammler> packaging is basically very simple
18:38:21  <Ammler> it is not like macports or how those port distros do it
18:39:07  <Ammler> just check a spec, those are very easy to read, like a shell script
18:40:42  <Ammler> oh, new Version, 2.8
18:40:52  <Ammler> does bandit work with that?
18:41:35  <andythenorth> of which package?
18:42:05  <Ammler> Chameleon, what else?
18:42:19  <Brot6> dutchroadfurniture: rebuild of r145 done (Diffsize: 165910) (DiffDiffsize: 11) -
18:44:01  <andythenorth> Ammler: probably works with 2.8
18:44:08  <andythenorth> can't find a release log anywhere right now
18:44:27  <Brot6> opengfx-nosetuptools: compile of r960 failed -
18:44:43  <Ammler> I should remove those clones
18:45:11  <andythenorth> Ammler: I found the chameleon git page, 2.8 should be fine
18:46:16  <andythenorth> oh dear
18:46:24  <andythenorth> they added an intriguing feature :P
18:46:42  <andythenorth> this is not at all good :P
18:46:45  <andythenorth>
18:46:47  <Webster> Title: Language Reference Chameleon 2.8 documentation (at
18:47:02  <andythenorth> means could write python code directly into nml templates, and use chameleon to execute it :P
18:47:23  <Ammler> hmm, that sounds rather cool than bad
18:47:31  <andythenorth> indeed
18:47:40  <planetmaker> that sounds... nice
18:47:50  <planetmaker> work it out, then explain to me :-P
18:48:03  <andythenorth> it's just that....some of the people behind chameleon have spent 15 years saying 'separate code and templates'
18:48:09  <andythenorth> it's become almost religious
18:48:42  <andythenorth> planetmaker: if I've understood it correctly, it's a 4 line python script or so
18:48:49  <Ammler> well, Chameleon is not really spread, since I needed to make the suse package
18:49:00  <andythenorth> import chameleon, load a template, render template, save result
18:49:18  <andythenorth> ho you can even import pdb (the debugger)
18:49:49  <Ammler> hmm, build server maybe found the slowest mirror ever
18:50:08  <Ammler> it will need some time until I get Chameleon to work again...
18:50:17  <andythenorth> Ammler: no rush
18:51:09  <andythenorth> it's interestingly different to what I'm used to
18:51:26  <andythenorth> few python devs are going to get much done without setuptools present (or Distribute replacing it)
18:51:41  <andythenorth> dev versus build :P
18:52:20  <Ammler> well, that is like dev without mke
18:52:24  <Ammler> make
18:52:46  <Brot6> ogfx-landscape: rebuild of r110 done (2 warnings) (Diffsize: 31742) (DiffDiffsize: 8) -
18:53:52  <Brot6> swedishrails: rebuild of r237 done (Diffsize: 9075) (DiffDiffsize: 8) -
18:54:08  <andythenorth> what I wonder is why python community has rolled their own solution, instead of using packages
18:57:54  *** Revelator has joined #openttdcoop.devzone
18:59:30  <andythenorth> Ammler: chameleon 2.8 works locally
19:01:37  <Brot6> airportsplus: rebuild of r170 done (Diffsize: 1161) (DiffDiffsize: 8) -
19:01:41  <Brot6> Following repos rebuilds successful without any difference to earlier nightlies builds: ogfx-trains (48 warnings) (Diffsize: 6), firs (6 warnings) (Diffsize: 6), opengfx (Diffsize: 28), manindu (Diffsize: 2), newgrf_makefile (Diffsize: 1), rust (15 warnings) (Diffsize: 6813), swisstowns (Diffsize: 51), britrains (2 warnings), spanishtowns (Diffsize: 8), frenchtowns (Diffsize: 21), ogfx-rv (42 warnings) (Diffsize: 20398),
19:01:41  <Brot6> german-townnames (Diffsize: 5042), dach (104 warnings) (Diffsize: 25201), belarusiantowns (Diffsize: 72), indonesiantowns (1 warnings) (Diffsize: 350)
19:07:42  *** JVassie has joined #openttdcoop.devzone
19:08:09  <andythenorth> planetmaker:
19:08:14  <andythenorth> it's a dumb example :P
19:08:16  <andythenorth> but it works
19:08:28  <andythenorth> you'd need to 'easy_install Chameleon' (probably with sudo)
19:08:43  <Brot6> BANDIT - (andythenorth) @
19:09:31  <andythenorth> the actual code to call it:
19:10:04  <andythenorth> and the python-in-nml: L70 and L83 here:
19:11:10  <Brot6> BANDIT - Revision 376:5eab738ca594: Add: dumb example of using python directly in nml (andythenorth) @
19:11:52  <andythenorth> it's obviously stupid using python to replace an nml prop with a constant defined in local python
19:12:04  <andythenorth> it gets a lot more interesting when you think about what else might be in scope
19:12:12  <andythenorth> and when you understand the repeat var :)
19:12:39  <andythenorth> hmm let's see
19:14:28  <andythenorth>
19:14:43  <andythenorth> L70, L72, L73
19:14:47  <andythenorth> much more interesting :)
19:16:02  <planetmaker> hehe
19:16:27  <Brot6> bandit: compile of r376 still failed (#3787) -
19:16:27  <planetmaker> Nice actually
19:18:23  <andythenorth>
19:18:29  <andythenorth> there you go, properly interesting now ^
19:19:16  <andythenorth> personally I'd much prefer to construct the trucks as first class objects and put them in the template scope
19:19:24  <andythenorth> but it took me 7 years to get around to learning that :P
19:19:36  <andythenorth> previously I'd have done it with nested nested dicts :P
19:19:40  <andythenorth> as in example above
19:21:19  <andythenorth> Ammler: if you hadn't broken BANDIT build, I wouldn't have found this new chameleon feature :D
19:22:56  <andythenorth> planetmaker: in the dict example above, you still don't gain much except the repeat, as it's just substituting constants
19:23:11  <andythenorth> by the time a nested dict is built, it may as well be the nml properties block, which looks like a dict anyway
19:23:47  <andythenorth> however for cases where calculated properties are wanted, it could work well
19:25:36  <Brot6> Dutch Trains 2.0 - Revision 292:108b36fd8b23: Fix: retire years of DMUs (foobar) @
19:29:44  <Brot6> Japanese Landscape - Feature #3788 (New): Rice (PaulC) @
19:37:45  <planetmaker> indeed, andythenorth. :-)
19:50:09  <Brot6> Japanese Trains - Bug #3789 (New): HOKI Large Hopper (PaulC) @
20:36:28  <Ammler> andythenorth: Chameleon will now build with setuptools as require, next time bandit builds it should work again
20:37:00  <Ammler> if you don't have something to commit, you could just push a dummy commit like date in enable
20:37:54  <Ammler> but wait around an hour
20:56:15  <Brot6> Dutch Trains 2.0 - Revision 293:1b41873e2e08: Fix (r292): when fixing something, do it properly (foobar) @
20:56:15  <Brot6> Dutch Trains 2.0 - Revision 294:caf7453b0e53: Fix: retire years of electric engines (foobar) @
20:56:15  <Brot6> Dutch Trains 2.0 - Revision 295:52dc423840a5: Fix: retire years of diesel engines (foobar) @
20:59:42  <andythenorth> Ammler: thanks
21:00:10  <Brot6> BANDIT - Revision 377:586a7f6df97f: Codechange: use a dict for better example of dumb python-in-nml (andythenorth) @
21:04:53  <Brot6> bandit: compile of r377 still failed (#3787) -
21:11:34  <Brot6> Dutch Trains 2.0 - Revision 296:4f5cd6125215: Fix: retire years of steam engines (foobar) @
21:11:34  <Brot6> Dutch Trains 2.0 - Revision 297:9d4074b24de0: Fix: retire years of metros (foobar) @
21:11:34  <Brot6> Dutch Trains 2.0 - Revision 298:80443a0236e9: Fix: retire years of EMUs (foobar) @
21:27:51  <Ammler> this was no hour :-P
21:27:55  *** ODM has quit IRC
21:28:28  <Ammler> installing python-Chameleon-2.7.3-2.1
21:28:54  <Brot6> Dutch Trains 2.0 - Revision 299:187dd13158d3: Fix: retire years of coaches (foobar) @
21:28:54  <Brot6> Dutch Trains 2.0 - Revision 300:c42e906f187e: Fix: retire years of wagons (foobar) @
22:02:53  <andythenorth> Ammler: out of interest, what are/were the blockers for Chameleon?
22:14:27  <Ammler> your impatient
22:14:38  <Ammler> I said an hour :-)
22:16:10  <Ammler> oh
22:16:17  <Ammler> forgot to add the new source
22:16:23  <Ammler> sorry, another hour :-P
22:18:15  *** frosch123 has quit IRC
22:26:09  <andythenorth> Ammler: np :)  I'm in no rush.  Just wondered what the process is
22:35:50  <Ammler> next push works
23:06:52  *** Webster has joined #openttdcoop.devzone
23:09:59  *** Guest5429 has quit IRC
23:11:08  *** andythenorth has quit IRC
23:30:39  *** JVassie has quit IRC

Powered by YARRSTE version: svn-trunk