Back to Home / #openttd / 2009 / 08 / Prev Day | Next Day
#openttd IRC Logs for 2009-08-06

---Logopened Thu Aug 06 00:00:39 2009
00:05-!-DPyroo [] has joined #openttd
00:07-!-Svish [~Svish@] has joined #openttd
00:08-!-Svish [~Svish@] has quit []
00:08-!-Svish [~Svish@] has joined #openttd
00:14-!-DPyro [] has quit [Quit: ajax IRC Client]
00:31-!-Fuco [] has joined #openttd
00:38-!-`Fuco`` [] has quit [Ping timeout: 480 seconds]
00:39-!-`Fuco`` [] has joined #openttd
00:42-!-Fuco [] has quit [Ping timeout: 480 seconds]
01:20-!-Lakie [~Lakie@] has quit [Quit: Sleep.]
01:35-!-R0b0t1 [] has joined #openttd
01:38-!-Wolle [] has quit [Quit: - das Wiki rund um's Thema Lager und Logistik]
01:49-!-keoz [] has joined #openttd
02:02<R0b0t1>For like, small local circuits, is a maglev actually any better than a monorail?
02:02<R0b0t1>It seems that the extra speed doesn't help too much
02:02<R0b0t1>And it just costs more (I mean, the speed could help you transport more stuff, but the resource generally doesn't "recharge")
02:18<Xaroth>Long circuits generally earn more tho :P
02:23-!-Pygma [~quassel@] has joined #openttd
02:39-!-Zorn [] has joined #openttd
02:41-!-Cybertinus [] has joined #openttd
02:42-!-stuffcorpse [~rick@] has quit [Ping timeout: 480 seconds]
02:44-!-stuffcorpse [~rick@] has joined #openttd
02:48-!-`Fuco`` [] has quit [Ping timeout: 480 seconds]
02:50-!-Pygma [~quassel@] has quit [Remote host closed the connection]
02:51-!-Pygma [~quassel@] has joined #openttd
03:04<R0b0t1>"Road Vehicle 9's profit last year was -$666"
03:04<R0b0t1>Uh oh, my game is haunted...
03:18-!-|Jeroen| [] has joined #openttd
03:51<TrueBrain>good morning all
03:59-!-reldred [~reldred@] has joined #openttd
04:18<R0b0t1>Does anyone have a save of a pretty complex train network they wouldn't mind sharing?
04:18<R0b0t1>I need something to compare mine too :D
04:18<TrueBrain>try the forums
04:36-!-DJNekkid [] has quit []
04:36-!-fonsinchen [] has joined #openttd
04:41-!-ecke_ [] has joined #openttd
04:45<CIA-1>OpenTTD: truebrain * r17082 /extra/website/ (noai/ templates/header.html): [website] -Remove: NoAI is no longer a Django website, but is now managed by RedMine
04:46<CIA-1>OpenTTD: truebrain * r17083 /extra/website/ [website] -Fix r17082: to stay traditional: forgot to remove one reference
04:51<@Rubidium>R0b0t1: take a look at the save games/scenarios section on the forum or at
04:54<R0b0t1>didn't know there was a section
04:57-!-DPyro [] has joined #openttd
05:00<@Rubidium>petern: please tell me what string would be the best of the ones in
05:00<TrueBrain>2) {VEHICLE} stopped due to ordered refit failure <- that is more readable :)
05:01<TrueBrain>1) is nice too :)
05:01<TrueBrain>6) says more than others :p
05:01<planetmaker>2) read like "vehicle stopped as we ordered a failure"
05:02<planetmaker>1) needs an addition "t" in refitting.
05:03-!-DPyroo [] has quit [Ping timeout: 480 seconds]
05:03<planetmaker>I propse 7) {VEHICLE} stopped due to failure of an ordered refit
05:03-!-Yrol [] has joined #openttd
05:04<planetmaker>and +o :-P
05:04<@Rubidium>where +o?
05:04<planetmaker>in propose :-)
05:04<TrueBrain>hi SmatZ!! :)
05:04<SmatZ>hello TrueBrain! :-)
05:05<SmatZ>what about "-an" ?
05:05<@Rubidium>is it a cloud? is it a plane? no it's SmatZ :)
05:05<SmatZ>hello hello :-)
05:05<TrueBrain>how is all with SmatZ?
05:05<SmatZ>I would like to say "I coming to save you", but I'm not :(
05:05<@Rubidium>I'll just add more suggestions
05:05<TrueBrain>you are not? Then what are you here for?!
05:06<TrueBrain>to scare us?
05:06<@Rubidium>it's BOFH!
05:06*SmatZ tries to scare you
05:06*TrueBrain is happy SmatZ is back
05:10-!-C-Otto [~cotto@ip2-169.halifax.RWTH-Aachen.DE] has joined #openttd
05:10<C-Otto>hi there
05:10<C-Otto>i'd like to use openttd to demonstrate some aspects of software verification to young students
05:10<planetmaker>go right ahead :-)
05:10<C-Otto>(by using trains, crashes and signals)
05:11<C-Otto>i found the "ignore signal" button, which is quite useful
05:11<@Rubidium>that would be funky ;)
05:11<C-Otto>is there some related option that trains do not wait until the segment is free?
05:11<C-Otto>i want them to go around without stopping (and crashing) as long as nothing disturbs them
05:11<planetmaker>ignore signal button maybe?
05:11*Rubidium would like to see that course
05:11<C-Otto>i need to press that button way more often than i can :)
05:11-!-OwenS [] has joined #openttd
05:12<planetmaker>there's no mass-ignore signal afaik
05:12<@Rubidium>C-Otto: either don't place signals at all or modify the code a bit so all signals are green
05:12<C-Otto>Rubidium: i like the first idea :)
05:13<@Rubidium>you 'just' need to ignore-signal the trains out of their depot
05:13<planetmaker>C-Otto, that's easily done. Just a bit dangerous :-)
05:13<C-Otto>yep, had my first crash...
05:13<C-Otto>okay, the depot is a signal
05:14*planetmaker wonders how train crashes relate to software engineering
05:14<@Rubidium>road crossings are a nice way to demonstrate stuff too; if the train is slow it won't crash through the road vehicle, but if the train is fast it will crash
05:14<@Rubidium>planetmaker: I'm thinking about verifying in relation with threaded software
05:14<C-Otto>ah. the train reversed direction!
05:14<C-Otto>bad, crash.
05:14<planetmaker>each train a single thread... he.
05:14<C-Otto>planetmaker: this is part of some very basic introduction to theoretic computer science - model checking etc.
05:15<C-Otto>at a later stage we will model the signals using synchrinized automata
05:15<@Rubidium>planetmaker: yes, and each track section is a different piece of 'data', which can't be concurrently used
05:15<C-Otto>now all trains died :(
05:17<Yrol>how about setting the penalties for red signals to a real low ( zero? ) value? wouldnt that kinda be a "nonstop ignore red signal"-switch?
05:17<Yrol>or some other pathfindermagic?
05:17<TrueBrain>you clearly don't understand the concept of penalties and how it is related to the game :) Hihi :)
05:17<C-Otto>i don't know the pathfinder options
05:17<Yrol>(TrueBrain) no, i dont °giggles°
05:18<OwenS>Yrol: Setting the penalty to zero will just mean YaPF causes trains to drive up to them rather than finding better routes :p
05:18<TrueBrain>C-Otto: just ignore that suggestion; red lights are red lights
05:18<@Rubidium>C-Otto: and neither does Yrol understand the way signals work, so disregard everything he said about signals
05:18<Yrol>uhm... it was not a suggestion. but a question.
05:18<TrueBrain>in that case the asnwer is a hard: NO
05:19<OwenS>Heck; Signals are such it's impossible, at present, for any new kind of block signal to override the "red if train in block" logic
05:19<Yrol>thank you, TrueBrain.
05:19<TrueBrain>C-Otto: can you send us a copy of your class? :p
05:22<OwenS>Writing recursive descent parsers is fun...
05:23<TrueBrain>doing it once, yes
05:23<TrueBrain>after that, it is boring
05:23<@Rubidium>TrueBrain: a (small) paper would be nice too; Using OpenTTD to teach software verification ;)
05:23<C-Otto>TrueBrain: of the save games?
05:23<TrueBrain>C-Otto: I agree with Rubidium :)
05:24<TrueBrain>C-Otto: no, I wonder how OpenTTD can ever be set to use in a class, so I would love to see that :)
05:24<C-Otto>easy concept
05:24<C-Otto>1) play a bit
05:24<C-Otto>2) let them get the idea that "two trains on the same track is bad" (by crashing them!)
05:24<OwenS>TrueBrain: The "..." was supposed to indicate the "not really" part :P
05:24<C-Otto>3) introduce signals
05:24<C-Otto>4) let them figure out, by testing, that the signals actually work
05:24<TrueBrain>OwenS: well, the first time I thought it was fun :)
05:24<C-Otto>5) explain the underlying model (red -> green -> red etc.)
05:25<@Rubidium>TrueBrain: we ought to follow that course ;) Got some time in S2 2009/2010?
05:25<C-Otto>6) use that model to formalize the system using some automata
05:25-!-Coco-Banana-Man [] has joined #openttd
05:25<OwenS>I'd much rather be using Bison if it didn't generate crappy code :p
05:25<C-Otto>7) synchronize the automata and see that everything works out
05:25<TrueBrain>Rubidium: you think C-Otto would pay for our travel?
05:25<C-Otto>this is no real class, but some introduction for pupils
05:25<@Rubidium>TrueBrain: you've got OV, right?
05:25<TrueBrain>OwenS: you lose that idea soon enough, Bison is not as bad as you might think :p
05:25<TrueBrain>C-Otto: still, sounds rather interesting :)
05:25<OwenS>TrueBrain: It is when you want multi-threadedness
05:25<@Rubidium>Aachen is like... not far over the border, so it can't be that expensive
05:25<TrueBrain>Rubidium: not valid in germany :p
05:25-!-Mucht [] has quit [Ping timeout: 480 seconds]
05:26<TrueBrain>Rubidium: true true
05:26<TrueBrain>C-Otto: so, we will join the introduction, okay? :)
05:26<OwenS>TrueBrain: Fortunately my grammar is rather simple. And I have the awesome re2c building the lexer :p
05:27<C-Otto>TrueBrain: how old are you?
05:27<TrueBrain>I build 2 lexers myself, used flex a few more times .. in the end it really doesn't matter :p
05:27<OwenS>Blindingly fast, needs a bit more glue work than flex/bison, but generates nicer code
05:27<TrueBrain>C-Otto: too old for an introduction class, I guess :p
05:27<@Rubidium>TrueBrain: OV to Heerlen and then two busses
05:28<OwenS>And it's much nicer to work with IMO :P
05:28<TrueBrain>OwenS: never tried it
05:29<TrueBrain>Rubidium: it takes me 3hours to get in Heerlen :p
05:29<@Rubidium>or the train from heerlen
05:29-!-Azrael- [] has joined #openttd
05:30<OwenS>Hmm, to support threading my "VM"'s gonna have to switch to a different stack
05:30<TrueBrain>reminds me that we never finished NAIL :(
05:30<@Rubidium>TrueBrain: still less that what I need
05:31<OwenS>TrueBrain: As I said, LLVM makes it stupidly easy (And very fast too)
05:31<TrueBrain>OwenS: NAIL has to be lightweighted ... LLVM doesn't fit that bill :p
05:32<OwenS>I'm working on it for an app that has two important conditions: Firstly, very, very fast; Secondly, pre-emptible
05:35<Yrol>i am sure to be bashed again, but... would it not be visually easier to understand, if one uses the blocksignals, which do not revert to red the sec the locomotive has passed them? °puts on helmet°
05:35<C-Otto>btw, openttd rocks on a 30" with 2560x1600 :P
05:36<OwenS>Oh yeah... I have another requirement: Security :p
05:37<TrueBrain>C-Otto: or put a webcam in the class, and send us the result :p
05:37<C-Otto>TrueBrain: the students will make some video presentation on the second day, where they should explain their topic to the other groups
05:38<C-Otto>maybe we get some openttd-video instead of a real one :)
05:38<TrueBrain>either way, we would love to see such videos :) Even more if you allow us to post them on our webpage
05:39-!-Exl [] has joined #openttd
05:39<OwenS>TrueBrain: Anyway, my language is somewhere between Python and C :p
05:40<TrueBrain>which says absolutely nothing :)
05:40<TrueBrain>more important are things like: strong typed? weak typed? pointers? string-autocasts?
05:40<OwenS>Weak typed; No pointers; Auto coercion when no data loss
05:40<OwenS>Object oriented, but in the funny "Primitive types are virtual objects" way :p
05:41<TrueBrain>I hate weak typed languages :p
05:41<TrueBrain>(but easier to implement :p)
05:41<OwenS>Well attach type tags to your variables when I implement them :P
05:41<OwenS>var [Int32] myvar;
05:42<TrueBrain>the biggest evil about weak types is that you can't predict function return types (the rest you can determ in most of the cases). I mean like:
05:42<TrueBrain>function () { if (b) return "a"; return 1; }
05:42<TrueBrain>same goes of course for statements like: a = b ? 1 : "a"
05:42<OwenS>def [Int32] x() will probably be the syntax for that :p
05:42<TrueBrain>kills many optimizations, if you don't know the type
05:43<@Rubidium>simple solution; only return ints
05:43<TrueBrain>Rubidium: when doing NAIL we already found out that is not what is happening :p
05:44<OwenS>But the compiler will attempt to infer types when it can from given information
05:44<TrueBrain>Rubidium: ah, yes :p
05:45<TrueBrain>OwenS: either way, if you really want speed (very very fast, as you said), you will need to make it strong typed
05:45<TrueBrain>else the runtime needs to check if c = a + b, is valid
05:45<TrueBrain>(needs to check a is integer, b is integer, ...)
05:45<TrueBrain>just telling you now, so you don't have to find that out later :) We did ... :p
05:46<OwenS>TrueBrain: I need speed to an extent. However, I also need user friendlyness as it's not really going in an app targetted at programmers :p
05:46<@Rubidium>pre-emptible, fast and noob-friendly?
05:46<OwenS>It's gotta find a medium between them :p
05:47-!-Mucht [] has joined #openttd
05:47<@Rubidium>try Char
05:47<TrueBrain>OwenS: so you don't need very very fast .. you just want speed :)
05:47<OwenS>Yeah. I need to be able to run quite a few of these in between rendering 3D frames :p
05:48<TrueBrain>depends on the framerate you expect :p
05:48<TrueBrain>create a JIT, and you might make it :)
05:48<OwenS>LLVM JIT ;-)
05:48<OwenS>But luckily most of the time these will be running when frame rates will be low anyway (I.E. before the scene optimizer has run)
05:49<TrueBrain>good luck ;)
05:49<blathijs>OwenS: Working with LLVM? Nice :-)
05:50<TrueBrain>always a future for NoAI: JIT the code :) :p
05:51<Yrol>i do not get it. why are you so agressive against people who ask a question? a simple "no" or "it doesnt work that way" would have been totally fine. instead of not only implying the asking person is to be totally ignored and not credible, like a slap in the face. sure, you probably get many of weird or silly questions here, but, is this a channel for openttd or for openttd-developers with a masters degree only? it would be nice to have some kind of expl
05:51<OwenS>Well, LLVM does that, at a cost of a big lib :P
05:51<TrueBrain>OwenS: too big, if you ask me :(
05:52<TrueBrain>somehow HelenOS got my attention .. dunno why :)
05:52<blathijs>Yrol: Your message was croppend at "have some kind of expl..."
05:54<@Rubidium>Yrol: 1) it sounded very much like you "knew" what you were talking about, basically giving him an idea that for sure is a waste of time to investigate.
05:55<@Rubidium>TrueBrain: because of Brianetta?
05:55<TrueBrain>Rubidium: no?
05:55<@Rubidium>Yrol: 2) the people bashing you don't have a masters degree (yet)
05:55<@Rubidium>Yrol: 3) I dislike people telling stuff that isn't true at all
05:56-!-R0b0t1 [] has quit [Read error: Operation timed out]
05:57<@Rubidium>Yrol: 4) it is more clear to tell someone you don't have a clue about the signals than sugar coating it with all kinds of stuff. Especially because both are (likely) non-native English speakers
05:57<OwenS>assert(&Parser::parseUnaryExpr && 0); /me likes such expressions to say "I was called in a stupid way"
05:57<TrueBrain>OwenS: euh .. ieuw?
05:57<SmatZ>&& 0?
05:58<OwenS>Make the assertion always fail because it's in a branch that shoiuld be unreachable
05:58<Yrol>( rest of message ) " it would be nice to have some kind of explanation for such behaviour."
05:58<TrueBrain>then use: assert(!"I should never come here")
05:58<OwenS>I should probably use that :P
05:58<TrueBrain>OwenS: but remember asserts can be disabled
05:58<TrueBrain>that is the idea of an assert ;)
05:59<OwenS>Yes. The code should never reach it :P I'm following it with an abort for runtime
05:59<@Rubidium>TrueBrain: I rather use NOT_REACHED() ;)
05:59<OwenS>release builds**
05:59<TrueBrain>Rubidium: I agree :p
06:00<TrueBrain>OwenS: and given you use C++, why not use 'false' over '0'?
06:00<OwenS>Nice thing about LLVM is that I can make my code generator generate some really unoptimized code... and the optimizer passes will do a brilliant job of cleaning it up :p
06:00<TrueBrain>so much more clear :)
06:00<OwenS>Also a good point on the false =/
06:00<TrueBrain>isn't that the general idea of an optimizer? :)
06:01<OwenS>I mean I can generate some quite pathologially lazy code :P
06:01<OwenS>Other than type inference (Something beyond LLVM's low level remit :P) I'm not doing any other optimizations
06:01<@Rubidium>you mean like int x; while (--x);
06:01<TrueBrain>it btw is nice to at least once in your life write your own optimizer :)
06:03<blathijs>Or look at LLVM's? :-)
06:03<TrueBrain>Yrol: 5) we told you 'no', and we had to make sure you were ignored by C-Otto, as you were confusing C-Otto. You might consider stating your questions in a different form, to avoid such confusion :) (just my 2 pennies)
06:03*blathijs did some work on LLVM's optimizer passes during his internship last year
06:03<TrueBrain>blathijs: looking at it doesn't cut it :) Doing it .. that is nice :p
06:04<blathijs>TrueBrain: Agreed, when you look at it, it looks so simple :-p
06:04<TrueBrain>when I had to write my first optimizer, it really took some time to get my head around it :p
06:04<TrueBrain>Eddi|zuHause: will you be posting civ4 specs? :p
06:07<Yrol>(Rubidium) aha. i guess, i totally not understand the meaning of language then either, if a clearly not "suggestive" sentence ( i did not quite said "do that, do this" ) is misunderstood in such a way. oh well... may ESR be praised. ( and thanks, OwenS, for the kind explanation why it wont work )
06:08<TrueBrain>Yrol: funny, how a 'clearly' sentence was misinterperted by at least 3 people from different nations and inside OpenTTD knowledge. But okay, take it how ever you want :)
06:09<OwenS>I suppose 16bytes/variant isn't so bad for a 64-bit arch (struct { Type* type; union { Object* obj; int64_t int; double dbl; etc }};
06:09<TrueBrain>blegh @ codestyle of OwenS
06:10-!-ecke_ [] has quit [Quit: ecke_]
06:10<OwenS>lol, in what way? :p
06:11<TrueBrain>Type* type, type2
06:11<TrueBrain>what type is 'type2'?
06:11<Eddi|zuHause>TrueBrain: yes, but not right now.
06:11<TrueBrain>Type* or Type
06:11<TrueBrain>Eddi|zuHause: yippie :)
06:11<TrueBrain>OwenS: so * belongs to type .. write it there :p
06:11<OwenS>I know; thats why I don't define multiple pointer types like that :P
06:11<TrueBrain>(I really dislike that coding style :))
06:11<TrueBrain>makes code obscure :p
06:11<OwenS>It's my pet peeve with C++ :P
06:11<TrueBrain>but okay :)
06:12<OwenS>I should be saying "Type*" is the type... now define all these vars as "Type*"s :p
06:12<TrueBrain>I think Pascal has a nice thing for that, that you define Type_P and stuff
06:12<TrueBrain>then you have: Type_P type
06:12<TrueBrain>immediatly removes any unclearness about the type :)
06:13<OwenS>I suppose I could typedef Type* TypePtr but thats just ugly :P
06:13<TrueBrain>for C(++) it is 'ugly', yes :)
06:13<TrueBrain>for Pascal it is the most normal thing there is :)
06:13<@petern>this is why you write it was Type *foo
06:13-!-tux_mark_5 [] has joined #openttd
06:14<TrueBrain>but okay, my likes/dislikes about C++ :)
06:14*OwenS starts defining & building the AST
06:15-!-R0b0t1 [] has joined #openttd
06:19-!-Mucht [] has quit [Remote host closed the connection]
06:27-!-Mucht [] has joined #openttd
06:31<TrueBrain>hello dihedral :)
06:32<dihedral>hey ho :-) how are you sir?
06:32<TrueBrain>good good :) Slowly working on WT3.1 again ...
06:32<TrueBrain>but it is so hot here ...
06:34-!-Polygon [] has joined #openttd
06:38-!-glx [glx@2a01:e35:2f59:c7c0:a95b:37a2:a8cf:4a17] has joined #openttd
06:38-!-mode/#openttd [+v glx] by ChanServ
06:49<dihedral> <- dihedral having a creative mood :-P
06:51*TrueBrain laughs hard at post of dihedral (in a positive way :p)
06:56<OwenS>I do wonder how one would implement signals in tunnels :p
06:56<OwenS>(As in, the matter of placing them :P )
06:56<TrueBrain>not many people know tunnels are blackholes
06:56<TrueBrain>most think that somehow the rails is really on all those tiles
06:56<OwenS>Wouldn't wormholes be a better term? :p
06:57<TrueBrain>fair enough
06:57<TrueBrain>although I once read that every blackhole has a whitehole
06:57<OwenS>Well thats not accepted scientific theory :P
06:57<TrueBrain>blackholes are neither
06:57<OwenS>Blackholes are commonly accepted to exist and be holding the galaxy together
06:57<TrueBrain>still not proven
06:58<TrueBrain>You got to love astronomy :)
06:58<OwenS>They're accepted theory though and probably will be for a while (I.E. until we develop FTL travel or go to the galactic core and back :P )
06:59<TrueBrain>Rubidium: we need to think about branching strgen I think? (you wanted to keep a version for 0.7 branch not?)
06:59<@petern>strgen inside the tree...
06:59<TrueBrain>OwenS: well, define 'accepted' :) But okay, I will give you thisone, in general it is assumed there is a blackhole
07:00<OwenS>Accepted: Assumed to be and apparently correct but unproven
07:00<@Rubidium>not quite branching, just compiling an ancient enough strgen and marking that with e.g. -0.7 instead of -r12346
07:00<TrueBrain>Rubidium: hmm .. tricky
07:00<TrueBrain>(as the strgen branch can only compile HEAD)
07:01<TrueBrain>strgen branch being: /extra/strgen
07:01<@petern>... /tags/0.7.2/src/strgen
07:02<TrueBrain>running 'make' there doesn't really do anything
07:02<TrueBrain>except of course giving an error
07:02<@petern>no but that's where the correct source is
07:02<TrueBrain>yes, thank you for that. Now back to the issue .. ;)
07:02<TrueBrain>(it btw also needs a file from table/)
07:02<@petern>so the issue is the build system doesn't package up the strgen binary
07:02<TrueBrain>the issue is that /extra/strgen only compiles HEAD
07:03<@petern>which should be done at the same time as building openttd
07:03<@Rubidium>petern: unless Apple donates us a machine we're not natively compiling strgen on all platforms, making packing it quite tricky
07:03<TrueBrain>Rubidium: maybe add some small magic to that Makefile script, which allows: make TAG=0.7.0 or what ever, which fetches the right revision for the svn:external?
07:04<@petern>if you can compile openttd for other platforms, compiling the much simpler strgen shouldn't be a problem
07:04<TrueBrain>Rubidium: I think we need to give petern our job :)
07:05<@petern>if you decided to remove strgen from the source tree, and make /extra/strgen the official and only place to get it, then i would agree with your viewpoint
07:05<@petern>but currently /extra/strgen is just extra confusing unnecessary cruft
07:06<TrueBrain>for you maybe; for us, not so much
07:06<@Rubidium>petern: compiling not, automating it without much extra effort is something else
07:07<@Rubidium>and we've learned that using svn:externals isn't that favourable in /trunk
07:07<TrueBrain>euhm, no, that was a bad move :)
07:08<Eddi|zuHause>what was actually the problem there?
07:08<@Rubidium>TrueBrain: what about REV=XYZ, if unset use HEAD and svn up the externals to that revision? For the 0.7 branch version we just need the right revision
07:08<@Rubidium>Eddi|zuHause: hg/git people failing to compile
07:08<TrueBrain>Rubidium: hmm .. I now realise a problem: 0.7 is in a branch
07:08<TrueBrain>and svn:external fetches trunk
07:08<TrueBrain>so that only works for 0.7.0
07:09<TrueBrain>or we need to modify 'make' to do the fetching
07:09<@Rubidium>Eddi|zuHause: svn up -r foo doesn't update the externals to the same revision, making binary searches difficult
07:09<@petern>well, if you need a branch
07:09<@Rubidium>TrueBrain: strgen hasn't changed in /0.7 and is unlikely to significantly change
07:09<@petern>just make a normal copy
07:10<TrueBrain>Rubidium: 'significantly' ;)
07:10<TrueBrain>what if in 0.8 you decide it is needed ? :)
07:11<@Rubidium>well, add an svn switch option too?
07:11<TrueBrain>hmm .. MSVC of course never calls make :p
07:11<@Rubidium>we're not using MSVC to compile the thing, right?
07:12<TrueBrain>no? mingw for windows?
07:13<TrueBrain>so that would be possible
07:13<TrueBrain>although a bit ugly, and consumes unneeded time on the VMs :p
07:14<TrueBrain>hmm .. do the VMs have networking? :p
07:14<@Rubidium>needs to be done before it's send to the VMs (no network)
07:14<TrueBrain>hehe :)
07:16-!-KenjiE20 [~KenjiE20@] has joined #openttd
07:17-!-R0b0t1 [] has quit [Remote host closed the connection]
07:19<TrueBrain>k, let me try to call apple first now :)
07:19<blathijs>Call apple?
07:20<TrueBrain>or you know a better way in?
07:24<TrueBrain>redirected 3 times, whoho
07:24<TrueBrain>finally you get an email address ... whoho!
07:26<@Rubidium>TrueBrain: (got to go now though, ciao)
07:26<TrueBrain>Rubidium: have fun!!
07:27<TrueBrain>problem with your patch is that it needs to run BEFORE the VMs get the data .. which is tricky :(
07:31<blathijs>TrueBrain: Dunno, what do you want from apple? :-)
07:31<TrueBrain>blathijs: a server
07:31<blathijs>Ah, but probably not to buy one :-)
07:31<TrueBrain>we don't have 3000 euros :p
07:33<planetmaker>2,3k€ are sufficient :-P
07:33<blathijs>I'll ask with my study association, IIRC they have had contact with Apple on sponsoring related business before
07:49<TrueBrain>I am out of the office, returning on Monday, August 10th. I will respond to your email as soon as possible upon my return. For urgent matters, please call me on my cell at
07:51<@petern>TrueBrain, hey loo, planetmaker offered :D
07:54<fonsinchen>did anyone take a look at my smallmap-zoom-out and zoomlevels patches?
07:55<fonsinchen>FS#54 and FS#3094
08:09-!-Yrol [] has quit [Quit: MOOOOOOOOOO?????]
08:20-!-HerzogDeXtEr1 [~Flex@] has joined #openttd
08:20-!-oskari89 [oskari89@] has quit [Quit: Utm A½ - Aja 35]
08:27-!-HerzogDeXtEr [~Flex@] has quit [Ping timeout: 480 seconds]
08:29-!-Baffage [] has joined #openttd
08:29<Baffage>Hello fellow openTTD-players! Baffage is back, and this time he needs some advice!
08:30<TrueBrain>what a nice greeting :)
08:30<TrueBrain>what can we do for you?
08:30<Baffage>I'm playing coop with a friend (he joined my company etc), but in the new rail veichles meny, there are no locomotives
08:30<TrueBrain>what year are you at?
08:30<Baffage>2034. started in 2030
08:31<TrueBrain>you have newgrfs loaded, where engines expire?
08:31<Baffage> .. probably :p how do i find out?
08:31<TrueBrain>there my knowledge stops :) I never play with newgrfs :)
08:33<Baffage>There's... nothing in the newgrf settings
08:33<Baffage>should I download a preset or something?
08:33<CIA-1>OpenTTD: smatz * r17084 /trunk/src/ (6 files): -Codechange: _age_cargo_skip_counter is common to all vehicle types
08:35<Baffage>.. was that answer for me? :o
08:35<TrueBrain>Baffage: haha, no :)
08:35<Baffage>.. great :]
08:35<TrueBrain>CIA-1 informs us about changes other developers made to the current development code
08:35<Baffage>i see
08:36<+glx>Baffage: what kind of rail depot?
08:36<Baffage>just a regular one i guess?
08:36<+glx>try monorail or maglev
08:38<Baffage>how? i can only find a regular train depot in the train meny
08:38<TrueBrain>meny = menu I guess? :)
08:38<TrueBrain>at least a consistant typo :)
08:38<Baffage>oh yeah, menu :]
08:39<Baffage>menu is meny in norwegian, it's kindof programmed into my brain to press Y i guess :]
08:39<TrueBrain>haha :) Well, I was wondering, because you did it so consistantly :)
08:40<Baffage>woah, didnt notice
08:40<OwenS>I think I'm gonna ignore nested functions for v0.1; continuations can come later :P
08:41<+glx>Baffage: hold mouse button when clicking on "open rail toolbox" button
08:41<+glx>you'll see a drop down to select rail type
08:42<+glx>same goes for road toolbox
08:43<Baffage>aaaaaaah thanks =D
08:43<TrueBrain>glx: you are GOOOOODDD :)
08:43<TrueBrain>I would never expect that a user doesn't know that :)
08:44<Baffage>Anyway, while we're at it, my friends joined my company through the multiplayer lobby and so on, but when I abandon the game, is there any way to continue where we dropped off?
08:44<+glx>TrueBrain: common unknown thing
08:44<CIA-1>OpenTTD: smatz * r17085 /trunk/src/ (5 files): -Codechange: unify the place where vehicle's cargo ages
08:45<OwenS>Baffage: Save and load the game as normal
08:45<Baffage>Then how does he join me afterwards?
08:45<TrueBrain>when you start a multiplayer game, you can also load a game
08:46<+glx> <-- I like the idea
08:46<Baffage>aaaah, well that wasn't too hard :]
08:48<TrueBrain>glx: and yet we fully support IE6 :p
08:49<+glx>but supporting IE6 is not fun ;)
08:49<TrueBrain>that is true :)
09:44-!-reldred [~reldred@] has quit [Read error: Connection reset by peer]
09:51<OwenS>TrueBrain: Hmm.. in a language designed to be easy to use... is making assignment an expression rather than a statement a good idea?
09:52<CIA-1>OpenTTD: smatz * r17086 /trunk/src/economy.cpp: -Cleanup (r16654): remove code used only for debugging
09:53<TrueBrain>OwenS: well, depends I guess :)
09:53<TrueBrain>it always has to be partly an expression if you want to allow: a = b = 1
09:53<TrueBrain>or while (a = func()) {
09:53<OwenS>It's just how many people are gonna go "if(a = b)" and get horribly confused? :p
09:54<TrueBrain>for those things you generate a warning
09:54<TrueBrain>but yes, a statement would reduse that confusion, at the cost of flexibility
09:55<@petern>livejournal is down
09:55<@petern>facebook is down
09:55<@petern>twitter is down
09:55<TrueBrain>MY LIFE IS GONE!
09:55<OwenS>" Assignment inside a conditional statement - this is normally unintentional. If this is your intention, enclose the assignment in parentheses to supress this warning"
09:55*TrueBrain starts screaming and running around naked
09:55<@petern>naked? woohoo!
09:56<TrueBrain>OwenS: exactly
09:56<OwenS>I then just need to put validateCondition() calls everywhere which check if the expr is a BinaryOpNode, and if so, what the operator is :p
09:57<blathijs>Not making assignment an expression might also make the compiler simpler :-)
09:58<blathijs>OwenS: Your warning is quite better that GCC's warning ("Suggest parentheses around truth value")
10:00<OwenS>Isn't it "Suggest parentheses around assignment used as truth value"?
10:00<@Rubidium>TrueBrain: to what extent can we manipulate/script the stuff *before* source.tar is made? (CF)
10:01<TrueBrain>Rubidium: hmm
10:01<TrueBrain>Rubidium: there isn't really a prehook
10:02<TrueBrain>Rubidium: as nice hack, strgen/ in projects might be able to take care of it :p
10:02<TrueBrain>Rubidium: as can the extra/strgen, but that is a REALLY ugly hack :
10:07<TrueBrain>clearly, the CF needs a rewrite :p Ghehe :)
10:09<@Rubidium>looks like might do the trick
10:09-!-Progman [] has joined #openttd
10:09<TrueBrain>for now, that looks like a nice working hack, yes :)
10:10<@Rubidium>though I think it needs manual 'work'
10:11<TrueBrain>I will modify the scripts so knows what the 'release' value is from the project
10:11<TrueBrain>should do the trick, I guess
10:11<TrueBrain>project = schedule
10:11<@Rubidium>the schedule doesn't want to run :(
10:11<TrueBrain>what did you do? :p
10:12<@Rubidium>nothing, though... it looks like it hasn't attempted to run it for ages
10:13<TrueBrain>haha, again :p
10:13<TrueBrain>it tends to get stuck :p
10:13<TrueBrain>failures! WHOHO! :)
10:14<@Rubidium>woepsie ;)
10:14<TrueBrain>and it looks like one of the VMs now hangs
10:14<TrueBrain>ah, no, it finished
10:16<TrueBrain>it is still running :)
10:16<TrueBrain>version updating :)
10:17<blathijs>OwenS: I don't think so, but perhaps it is
10:20<TrueBrain>Rubidium: seems to work!
10:20<@Rubidium>yup, though... needs manual work for releases. Not that it really matters much
10:20<TrueBrain>now we just need to compile one for the recent nightly, as finger now gives a wrong version :p
10:22<@Rubidium>will do ;)
10:22<@Rubidium>16886 is new enough :)
10:22<TrueBrain>it is the latest, I guess :p
10:22<TrueBrain>now I need to fight strgen with WT3.1 ..
10:23<TrueBrain>I wonder if it isn't easier to put the data I need on or whatever :p
10:23<OwenS>blathijs: "warning: suggest brackets around assignment used as truth value" (And thanks for reminding me to turn on -Wall :P )
10:24<@Rubidium>TrueBrain: what data do you need?
10:29<OwenS>Hmm, I suppose I should now build an AST printer :p
10:29<TrueBrain>Rubidium: the -export stuff
10:29<TrueBrain>as it turns out to be pretty tricky to compile stuff from WT3.1 (and keep it up-to-date)
10:30-!-HerzogDeXtEr1 [~Flex@] has quit [Read error: Connection reset by peer]
10:31-!-PhoenixII [] has joined #openttd
10:31-!-Phoenix_the_II [] has quit [Read error: Connection reset by peer]
10:32<@Rubidium>TrueBrain: true-ish, but if you're going to support branches it'll become even trickier
10:32<TrueBrain>so I was thinking, what if we publish the exports on<release>/
10:32<@Rubidium>also with finger.openttd
10:33<TrueBrain>a post-commit hook in subversion can take care of the updates, I guess
10:33<@Rubidium>TrueBrain: release would mean 0.7, trunk etc?
10:33<TrueBrain>still leaves my problem that if you commit a strgen update and a language update which uses that at the same time
10:34<TrueBrain>I somehow need to synchronize such data
10:34<@Rubidium>that'd mean that the strgen update needs to be done first in the post commit hook
10:34<TrueBrain>post-commit is async
10:34<TrueBrain>(on your request :p)
10:34<@Rubidium>and that only after that is succesfully done can the language thingy be started
10:35<TrueBrain>but I guess it is possible to make those 2 synchronized :)
10:35<@Rubidium>what if we make a new post commit script that sequentially does the strgen update and then language update and call that async from the postcommit hook
10:35<@Rubidium>prrrroblem solved ;)
10:36<TrueBrain>that is what I meant too, so yes
10:36<@Rubidium>though, for each language update it needs to check finger
10:36<TrueBrain>well, in the call to WT3.1 we can then also indicate if there was a strgen update
10:36<TrueBrain>not really pretty though
10:37<@Rubidium>separate call that some the strgen data changed?
10:37<TrueBrain>should work
10:37<@Rubidium>the only issue: how to cheaply compile them
10:38<TrueBrain>branches are here again an issue, yes
10:46<OwenS>w00t! My AST looks reasonable!
10:46-!-Fuco [] has joined #openttd
10:47-!-bb10 [] has joined #openttd
10:51<Eddi|zuHause>TrueBrain: if it said "account successfully created", is it normal that i can't login?
10:59<+glx>Eddi|zuHause: on WT3 ?
10:59<Eddi|zuHause>on noaddedsugar
11:01-!-worldemar [~tsukimiya@] has quit [Remote host closed the connection]
11:04<@Belugas>two sugar and no milk for me, please
11:11<Eddi|zuHause>i am not your coffee maid
11:12<@Belugas>hey... I said Please!
11:13<@Belugas>what can one man do to have a decent service in here???
11:14-!-Lakie [~Lakie@] has joined #openttd
11:14<TrueBrain>Eddi|zuHause: you should have received an email
11:14<TrueBrain>but I enabled your account
11:14<Eddi|zuHause>yeah, i think the email took a while
11:14-!-frosch123 [] has joined #openttd
11:20-!-Pygma [~quassel@] has quit [Remote host closed the connection]
11:22-!-Biolunar [] has joined #openttd
11:23-!-Pygma [~quassel@] has joined #openttd
11:25-!-Pygma [~quassel@] has quit [Remote host closed the connection]
11:27-!-PhoenixII [] has quit [Read error: Connection reset by peer]
11:28-!-Phoenix_the_II [] has joined #openttd
11:53-!-Mucht [] has quit [Remote host closed the connection]
11:55-!-Wolle [] has joined #openttd
11:57<CIA-1>OpenTTD: rubidium * r17087 /extra/strgen/ (Makefile Makefile.sub): [strgen] -Change: rework the makefile a bit so it can be reused more efficiently
12:01<CIA-1>OpenTTD: smatz * r17088 /trunk/src/date.cpp: -Codechange: split IncreaseDate() to more procedures
12:01-!-ecke [~ecke@] has quit [Read error: Connection reset by peer]
12:04<fonsinchen>Smatz, are you still interested in smallmap zoom?
12:04<SmatZ>fonsinchen: yeah :)
12:04<fonsinchen>I have split the patch in FS#54 in three parts, as I had announces
12:05<SmatZ>ah, great :)
12:05<fonsinchen>smallmap-zoom-out doesn't contain any code unnecessary for zooming out.
12:05<SmatZ>I got the notice, but haven't read that yet
12:06<fonsinchen>ok, I just wanted to mention that.
12:14*valhallasw sees TrueBrain doing the kabouterdans
12:15<TrueBrain>valhallasw: yesterday I tried to find a Windows computer at HL which was XP or newer ... I couldn't :(
12:15<TrueBrain>do you know how much that annoyed me? :)
12:16<valhallasw>TrueBrain: the new computers in the PHYSICS network will be XP
12:16<TrueBrain>WILL be
12:16<valhallasw>and most non-education computers are XP afaik :P
12:16<TrueBrain>all strw run linux :p
12:16<valhallasw>or win95/98 for measurement computers *grin*
12:17<TrueBrain>and because of the holidays all people who could help me, were out :p
12:17<TrueBrain>I was so annoyed!
12:17<valhallasw>what do you want to do? change your password? :P
12:18<TrueBrain>no, a friend of mine wanted to print her thesis
12:18-!-Zahl [] has joined #openttd
12:18<TrueBrain>with a non-normal font
12:18-!-C-Otto [~cotto@ip2-169.halifax.RWTH-Aachen.DE] has left #openttd []
12:18-!-|Jeroen| [] has quit [Quit: oO]
12:18<TrueBrain>fun fact about passwords: I always use my ssh key to login to strw ... turns out my password is invalidated long ago :p
12:18<TrueBrain>so I couldn't login at the systems via tty :p
12:19<TrueBrain>downside of always working from home :)
12:20<TrueBrain>passwords .. reminds me I was going to look into ldap authentication for all OpenTTD services
12:22<TrueBrain>"flyspray openttd" gives 10k results in google ... it is one of the suggestions .. lol
12:26-!-fonsinchen [] has quit [Remote host closed the connection]
12:34<blathijs>TrueBrain: I've recently struggled with LDAP for my server auth, so I might be able to provide some support when you try
12:39<OwenS>Hmm... time to learn how to include some LLVM ASM in my binary :p
12:50<OwenS>I should probably join #llvm :p
12:52<TrueBrain>blathijs: well, I need to know how to connect all the software to it :)
12:52<TrueBrain>and how to create accounts in it :p
12:54<blathijs>TrueBrain: I've been using phpldapadmin for the latter, but it is not quite optimal
12:54-!-TheMask96 [] has quit [Ping timeout: 480 seconds]
12:54<TrueBrain>Eddi|zuHause: very clear story, tnx; will see wha tI can do for you :)
12:54<TrueBrain>blathijs: yeah, it has to be automated :)
12:55<TrueBrain>flyspray for sure can used ldap
12:56<blathijs>TrueBrain: Ah, automating is easier :-)
12:56<blathijs>Just generate an LDIF and cram it into ldap_modify
12:56<blathijs>Or use ldapvi, which I do as well (which supports a LDIF-like format that is slightly more friendly)
12:56<blathijs>TrueBrain: That's the default text mode format for LDAP changes
12:57<blathijs>or LDAP output, for that matter
12:57<TrueBrain>ah :)
12:57<TrueBrain>well, first things first ... I need to know how Django supports LDAP
12:57<TrueBrain>and I wonder if I can link PAM accounts to it
12:57<blathijs>I don't have that setup yet, most of my webapps authenicate against phpbb for me :-)
12:58<blathijs>PAM to LDAP, or Django to PAM?
12:58<TrueBrain>PAM to LDAP
12:58<TrueBrain>for ssh logins
12:58<TrueBrain>that the developers passwords are synced too
12:58<blathijs>Ah, that's not so hard
12:59<blathijs>If you limit to just LDAP, it's really easy, if you want both LDAP and /etc/passwd, it requires slightly more work for graceful fallback
13:00-!-TheMask96 [] has joined #openttd
13:00<TrueBrain>I guess LDAP groups should work too
13:02-!-OwenSX48BD [] has joined #openttd
13:02<TrueBrain>k, by the looks integrating LDAP with Django is easy enough
13:02-!-OwenS is now known as Guest719
13:02-!-OwenSX48BD is now known as owens
13:02-!-owens is now known as OwenS
13:03<CIA-1>OpenTTD: smatz * r17089 /trunk/src/ (date.cpp vehicle.cpp vehicle_base.h): -Codechange: move RunVehicleDayProc() to vehicle.cpp
13:03<TrueBrain>Redmine can use LDAP just fine .. then it is just a matter of doing this without breaking any system :p
13:04-!-Biolunar [] has quit [Quit: reboot]
13:04<TrueBrain>(and setting up a LDAP .. which I have never done before :p)
13:05<blathijs>TrueBrain: Have a look at;a=tree;f=etc;h=882814f3a3f00
13:05<blathijs>TrueBrain: Especially /etc/pam.d/common-* and /etc/*nss*
13:05-!-Guest719 [] has quit [Ping timeout: 480 seconds]
13:06<TrueBrain>blathijs: looks useful, tnx
13:07-!-Biolunar [] has joined #openttd
13:07-!-TinoDidriksen [] has quit [Read error: Connection reset by peer]
13:07<TrueBrain>blathijs: I will bug you some more next week or something
13:08-!-TinoDidriksen [] has joined #openttd
13:10<Ammler>hmm, would the 2nd company color work for base graphics?
13:10<+glx>it's a newgrf feature
13:13-!-Svish|eee [] has joined #openttd
13:14-!-Pygma [~quassel@] has joined #openttd
13:14<Ammler>so you need to code a sprite, just using the 2cc color in the sprite isn't enouh?
13:17<Ammler>just asking because Born Acorn used orange for the 2nd color in his opengfx trains, if the company color is orange too, it looks a bit monotone.
13:18<blathijs>TrueBrain: Feel free :-)
13:18<blathijs>Though I'll be away until tuesday, probably
13:18<TrueBrain>I don't expect to start earlier ;)
13:22<+glx>Ammler: 2cc usage is set in Action 0
13:22<+glx>misc flags
13:24<CIA-1>OpenTTD: smatz * r17090 /trunk/bin/ai/regression/regression.txt: -Fix (r17089): regression test failed (frosch123)
13:24-!-Belugas [~belugas@] has quit [Read error: No route to host]
13:28-!-Yexo [] has joined #openttd
13:28-!-Belugas [~belugas@] has joined #openttd
13:28-!-mode/#openttd [+o Belugas] by ChanServ
13:28<Yexo>good evening
13:29<Eddi|zuHause>Ammler: if the 2cc flag is not set, only the 1st cc recolour map is applied
13:33-!-Mark [~Mark@] has quit [Read error: Connection reset by peer]
13:35-!-Baffage [] has quit []
13:35<CIA-1>OpenTTD: smatz * r17091 /trunk/ (bin/ai/regression/regression.txt src/vehicle.cpp): -Fix (r17089)(r17090): regression test was right
13:35<SmatZ>enough shame for today
13:36*Belugas knows the feeling
13:36<@petern>scotland's shame?
13:36<@Belugas>hooo.... so tempting!
13:36<@Belugas>so.. would it be M or keep on NIN?
13:37<@petern># keep on ninning
13:37<@Belugas>but... that's becasue no M around
13:37<@Belugas>#Keep on Rocking in a free world
13:43<TrueBrain>howdie Yexo
13:43<TrueBrain> is all set up
13:44<TrueBrain>Subversion repos are created every hour
13:44<TrueBrain>I believe at :17
13:44<Yexo>hello TrueBrain
13:44<Yexo>thanks, nice work :)
13:44<TrueBrain>so after creating a project you have to wait for a bit :)
13:44<Yexo>that's no problem
13:44<TrueBrain>stupid Redmine requires root priveleges to do so ...
13:44<Yexo>but doesn't load currently
13:45<CIA-1>OpenTTD: translators * r17092 /trunk/src/lang/ (4 files in 2 dirs):
13:45<CIA-1>OpenTTD: -Update from WebTranslator v3.0:
13:45<CIA-1>OpenTTD: simplified_chinese - 2 changes by Gavin
13:45<CIA-1>OpenTTD: frisian - 36 changes by huddekul
13:45<CIA-1>OpenTTD: german - 57 changes by Roujin
13:45<CIA-1>OpenTTD: spanish - 48 changes by Terkhen
13:45<TrueBrain>hmm .. long live apache :(
13:46-!-fjb_ [] has joined #openttd
13:47<Ammler>TrueBrain: we run redmine with passenger, then it runs with the user the redmine web belongs too.
13:47<OwenS>I'm gonna need to create a series of macros for writing these intrinsics...
13:47<TrueBrain>Ammler: your point being?
13:48<@Rubidium>petern: can I get your preference on ?
13:48<Ammler>in our case it is the user redmine, you could make sudoers for it, e.g. dunno, what exactly root needs :-)
13:49<TrueBrain>yeah, and for that you really need passenger
13:49<TrueBrain>lol :)
13:49<TrueBrain>but okay, tnx for your 2 cents :)
13:49-!-Pygma_ [~quassel@] has joined #openttd
13:50<Ammler>oh, nvm, it doesn't run with the web user, it is like suexec or how that is called.
13:52<CIA-1>OpenTTD: truebrain * r17093 /extra/website/general/ [website] -Fix: wrong render_to_response caused 500 errors
13:52<TrueBrain>anyway ... noai still doesn't load ... no idae why not :(
13:52<@Rubidium>try booting lighty?
13:52<TrueBrain> - - [06/Aug/2009:17:12:14 +0000] "GET / HTTP/1.1" 400 375 "-" "-"
13:53-!-Pygma [~quassel@] has quit [Remote host closed the connection]
13:53-!-Pygma_ [~quassel@] has quit [Remote host closed the connection]
13:53-!-fjb [] has quit [Ping timeout: 480 seconds]
13:55<TrueBrain>there we go
13:55<TrueBrain>passenger was hanging
13:55<@Rubidium>suicidal application?
13:55-!-Pygma [~quassel@] has joined #openttd
13:56<TrueBrain>dunno what went wrong :) I do know it runs again
13:56<frosch123>do we need a forward to yexo? :)
13:56<TrueBrain>Yexo: let me know if you need anything else :)
13:56<TrueBrain>frosch123: there already is :p
13:56*OwenS wonders how well LLVM will optimize his binary operation on variant code :p
13:59-!-Progman [] has quit [Remote host closed the connection]
14:01<Yexo>I'd like some advice on how to set up the svn repos, since I've never done that before
14:01<TrueBrain>create the dirs 'tags', 'branches' and 'trunk' ;)
14:01<Yexo>should I just create a directory trunk/ and put main.nut / library.nut in there?
14:01<TrueBrain>I always make r1 the layout commit, and after that the content :)
14:02<frosch123>@openttd commit 1
14:02<@DorpsGek>frosch123: Commit by truelight :: r1 / (202 files in 13 dirs) (2004-08-09 17:04:08 UTC)
14:02<@DorpsGek>frosch123: Import of revision 975 of old (crashed) SVN
14:02<frosch123>^^ always :)
14:02<TrueBrain>exceptions there ;)
14:03<+glx>doesn't count
14:03<TrueBrain>and damn, that was long ago :)
14:03<frosch123>true, it is by truelight
14:03<TrueBrain>I am getting OLD! :(
14:05*TrueBrain hugs Belugas
14:05<Yexo>TrueBrain: what username / password is valid for svn?
14:05<TrueBrain>Yexo: the one you login with to RedMine :)
14:06<TrueBrain>RedMine managed the SVN access
14:06-!-Mark [] has joined #openttd
14:06<Yexo>either tha tone doesn't work, or I'm doing something wrong
14:06<TrueBrain>so if you assign a random person as Developer to it, he can commit
14:06<TrueBrain>what URL do you use?
14:07-!-Biolunar [] has quit [Quit: saufen \o/]
14:07<Yexo>the URL* field in Redmine says: "file:///var/repos/svn/lib-binaryheap", but that doesn't help me at all
14:07<TrueBrain>haha, no :p
14:07<TrueBrain>the URL is correct
14:08<TrueBrain>usernames appear case-sensitive
14:08-!-thisismynick [~chatzilla@] has joined #openttd
14:08<Yexo>I'm using "Yexo", as is my loginname in Redmine
14:08<TrueBrain>I get an error with MKACTIVITY
14:08<Yexo>same here, but only after some attempts
14:09<TrueBrain>appears to be https problem
14:10<TrueBrain>and doing https is tricky
14:10-!-ecke [~ecke@] has joined #openttd
14:10-!-Svish|eee_ [] has joined #openttd
14:17-!-Svish|eee__ [] has joined #openttd
14:18-!-Svish|eee [] has quit [Ping timeout: 480 seconds]
14:19-!-ecke [~ecke@] has quit [Ping timeout: 480 seconds]
14:20-!-Elton06131 [~Delphi@] has joined #openttd
14:22-!-Polygon [] has quit [Quit: Flieht, ihr Narren!]
14:22<TrueBrain>Yexo: hmm, it looks like authentication fails in general for the SVN dir .. and I have no clue why :p
14:23<Yexo>if I can help in any way, please let me know
14:23<TrueBrain>fix my problem? :P
14:24-!-Svish|eee_ [] has quit [Ping timeout: 480 seconds]
14:25<@Belugas>yes, you can Yexo: two sugar and no milk for me, please
14:25<@Belugas>thanks ;)
14:25<Yexo>as soon as you arrive here your coffee will be ready :)
14:26-!-Elton06131 [~Delphi@] has quit [Read error: Connection reset by peer]
14:26<TrueBrain>nowhere it records WHY it fails :(
14:32-!-Splex [~splex@] has quit [Remote host closed the connection]
14:38-!-TheMask96 [] has quit [Read error: Operation timed out]
14:39-!-TheMask96 [] has joined #openttd
14:41<TrueBrain>hmm .. the URL is a bit fuzzy
14:41<OwenS>OMG YAY
14:42<OwenS>I got my intrinsics to assemble :-D
14:42<TrueBrain>it keeps on thinking it is accessing /svn
14:43<OwenS>It only took 60 lines of LLVM ASM to perform type coercion... :p
14:43-!-Svish|eee__ [] has quit [Ping timeout: 480 seconds]
14:45<TrueBrain>Yexo: try commiting now
14:45<Yexo>it worked :)
14:45<Yexo>thanks :)
14:45<TrueBrain>very simple problem:
14:46<TrueBrain>you were not a manager of the project
14:46<TrueBrain>I assumed Administrators would always be allowed .. but they are not :)
14:46-!-forconin [] has joined #openttd
14:46<TrueBrain>Yexo: so for next time: add yourself as manager or developer :p
14:46<TrueBrain>Yexo: can I suggest to follow OpenTTD commit style?
14:46<forconin>Hey, can anyone tell me where to config sub-tropical climate in openttd.cfg ?
14:46<TrueBrain>-Add / -Fix / ...
14:46<forconin>On a dedicated server that is.
14:46<TrueBrain>took me 30 minutes to figure this out :s bah ..
14:47-!-Jeej [] has joined #openttd
14:47-!-forconin is now known as Forconin
14:47<Jeej>Hello, how replace i simple all the cars i have?
14:47<Forconin>Noone? I would think it is the "landscape"-section.
14:47<frosch123>TrueBrain: can I suggest to use the same pre-commit-hook as for ottd, esp. for whitespace :)
14:47-!-thisismynick [~chatzilla@] has quit [Quit: ChatZilla 0.9.85 [Firefox 3.5.2/20090729225027]]
14:47<Forconin>But the wiki isn't helping, it just describes the different climates.
14:48<TrueBrain>frosch123: hmm .. good point, but it is not that easy to insert pre/post hooks in this automated situation ...
14:48<OwenS>Hang on... crap... my intrinsics function assumes that a binary operator returns the same type as it takes in... it probably doesn't...
14:48<Yexo><Forconin> Noone? <- give it some more time, irc isn't "Ask and always get an answer in 2 minutes"
14:48-!-Markk [] has quit [Ping timeout: 480 seconds]
14:49<@Belugas>Forconin : easy. remove openttd.cfg, start a game in sub trop, save it, close the game. check waht openttd.cfg now contains ^_^
14:49<Yexo>Forconin: it's 'landscape' under [game_creation]
14:49<Forconin>dedicated linux-server
14:49<Forconin>no GUI
14:49<Forconin>Yexo: Thought so, any idea what it should say?
14:49<Forconin>tried with "tropical" and "sub-tropical"
14:49<Forconin>no go
14:50<TrueBrain>clearly you can test it, which means you have a GUI somewhere
14:50<TrueBrain>so Belugas is the most safe way to approach your problem
14:50<Forconin>Thank you very much :)
14:51<TrueBrain>now edit the wiki page and add that :p
14:51<TrueBrain>Jeej: auto replace?
14:51<Yexo>since you have searched the wiki and haven't found this information, can you please add it to the wiki page you'd expect it?
14:51<Forconin>Give me a minute =)
14:51<Forconin>and yes of course.
14:51<Jeej>TrueBrain: with auto replace you are not able to replace a car of the same type
14:52<TrueBrain>then I wouldn't know :)
14:52<Yexo>Jeej: use autorenew
14:52<Yexo>but then cars will only be replaced when they get old
14:52<@Belugas>why running on a dedicated server withouth better experience??? this is runing for trouble
14:52<Yexo>it's either that or autoreplace
14:52<Jeej>Where can i find autorenew?
14:52*Belugas shakes his head, a bit disapointed
14:53<Yexo>Jeej: somewhere under "Advanced settings"
14:54<Jeej>Yexo: thank you, found it
14:54-!-mode/#openttd [+v Belugas] by ChanServ
14:54-!-mode/#openttd [+v petern] by ChanServ
14:54-!-mode/#openttd [+v Rubidium] by ChanServ
14:54-!-mode/#openttd [+v orudge] by ChanServ
14:55-!-Jeej [] has quit [Quit: Jeej]
14:55<Forconin>There, added to wiki.
14:55<Forconin>Thanks for the help. :)
14:56<Yexo>Forconin: where did you add it? no change of you show up on Special:Recentchanges
14:58-!-fonsinchen [] has joined #openttd
14:58<Forconin>Oh [[Climate]]
14:58<Forconin>Didnt notice the CAPTCHA
14:59<Forconin>Registered a account at the wiki for next time.
15:04-!-Brianetta [] has joined #openttd
15:10-!-fjb_ is now known as fjb
15:11<OwenS>Hmm.. Typeof, operator or builtin function? :p
15:12<frosch123>how could typeof and siizeof be functions?
15:12<OwenS>In the language I'm developing :p
15:12-!-Azrael- [] has quit [Ping timeout: 480 seconds]
15:14<frosch123>in that case what is the difference of functions and operators?
15:14-!-andythenorth [~andy@] has joined #openttd
15:14-!-andythenorth [~andy@] has quit [Remote host closed the connection]
15:15<OwenS>Syntax mainly: typeof(x) vs typeof x
15:15-!-andythenorth [~andy@] has joined #openttd
15:17<blathijs>OwenS: I'd go for operator, since a function is really something else (mostly runtime vs compiletime)
15:18<OwenS>Not in a scripting language :p
15:18<TrueBrain>even in a scripting language
15:18<TrueBrain>runtime means that even after JIT it needs to be evaluated every time it comes past that function
15:18<TrueBrain>where compiletime means it is done just once
15:19<OwenS>Aah true, I can in some cases optimize away typeofs :p
15:19<blathijs>TrueBrain: For an dynamically typed language, typeof might still be runtime
15:19<TrueBrain>blathijs: true
15:19<TrueBrain>but yeah, it allows optimizing
15:19-!-Azrael- [] has joined #openttd
15:19<TrueBrain>typeof, instanceof, ... you want that to be part of your language
15:20<blathijs>OwenS: OTOH, in LLVM the typeof thing will probably become a function again :-)
15:20<Prof_Frink>Sacro: Fail.
15:20<OwenS>blathijs: Depends :p
15:20<Sacro>Prof_Frink: wha?
15:21<blathijs>OwenS: Are there any other options?
15:21<Prof_Frink>See over there <--
15:21<OwenS>TrueBrain: instanceof? Nah. if typeof(x) == Integer is better :P
15:21<Sacro>Prof_Frink: wha?
15:21<Prof_Frink>window 2.
15:21<blathijs>OwenS: Intrinsics perhaps, but those are really functions as well
15:21<OwenS>blathijs: Yes. If my front end can deduce that the variable is of a type at compile time :p
15:21<Sacro>Prof_Frink: window to is #freeside on
15:21<blathijs>OwenS: How will that handle inheritance?
15:21<Sacro>2 i mean
15:21<OwenS>I.E. if you do "typeof 2" I know it's an Integer :p
15:22<Sacro>nope, truth
15:22<Prof_Frink>2 is a bad place.
15:22<blathijs>OwenS: Right, but in that case, the typeof will disappear alltogether :-)
15:22<OwenS>And if it doesn't dissapear it's just a load of the first member of the variant struct :p
15:22<Sacro>Prof_Frink: oh?
15:23<OwenS>blathijs: Good point. Need instanceof :p
15:24<blathijs>OwenS: Any particular occasion for building this language, or just for sport?
15:25<OwenS>blathijs: To go in an app I'm developing. Nothing matches the mix of speed, ease of use and preemptibility I need
15:25<blathijs>What kind of app?
15:26-!-Splex [~splex@] has joined #openttd
15:26<OwenS>3D virtual world type thing :p
15:26<blathijs>Ah, nice
15:26<blathijs>Are there any other script language with an LLVM backend that you know of?
15:26<OwenS>No. I know of compiled languages, but not script ones. Other than the "Kaleidoscope" tutorial language :p
15:27<blathijs>And kaleidoscope didn't match your requirements? Weird :-p
15:27<blathijs>Planning on releasing your language separately
15:27<OwenS>Yeah, I'm developing it seperately as AlterScript (The program it's going in is called AlterVerse, so it seems logical :P )
15:28<blathijs>Both with a free license?
15:28<OwenS>I'm not quite sure about AlterVerse being free (Most probably though), but AlterScript will be at least LGPL
15:30-!-Polygon [] has joined #openttd
15:30<OwenS>LLVM does have a function inlining optimizer right? :P
15:30<blathijs>You might want to take the hint from LLVM and use a BSD license :-)
15:31<blathijs>Yup, it has
15:31<OwenS>:-). I'm gonna need it for my "_as_binop" 'intrinsic'
15:32<blathijs>It should also still have some cool constant propagation optimizers, on which I hacked a year back :-P
15:32<OwenS>(It's the big block of code which vectors binary operators correctly)
15:36<CIA-1>OpenTTD: rubidium * r17094 /trunk/src/ai/api/ai_changelog.hpp: -Update: AI changelog
15:37<blathijs>I wonder if my commit access still works :-)
15:38<OwenS>I wish C++ protected was more like Java protected though (I.E. accessible from members of your namespace)
15:39<@Rubidium>blathijs: quite likely
15:40<OwenS>(Wait crap, wrong window)
15:47-!-KritiK [] has joined #openttd
15:49*OwenS prepares to embed the LLVM Assembler in his language (To assemble the intrinsics)
15:50<@petern>blathijs, it probably will, i don't think you're on "i don't like this person so i'll pretend we disable access after a while" list
15:51<blathijs>OwenS: Can't you just ship with bitcode?
15:52<blathijs>Should be easy to embed the assembler, though :-)
15:52<OwenS>That would probably be better, just involve more prodding :p
15:52<OwenS>Class LLParser :p
15:52<blathijs>more prodding?
15:52<OwenS>To find the class to use to load bitcode :p
15:52<blathijs>ah :-)
15:52<Yexo>Any suggestions for a directory structure for the ai libraries? See for my current thoughts
15:53<blathijs>OwenS: You might want to have a look at the opt source I think
15:53<blathijs>I can't remember how this stuff worked exactly anymore :-)
15:53*blathijs is off for a bit
15:53<OwenS>Linker.h... hmm! :p
15:54<@Rubidium>Yexo: what about making bananas scrub license.txt if you don't specify 'custom license'?
15:54<Yexo>that solves one of the problems
15:55<OwenS>Aah! llvm::BitcodeReader
15:56<@Rubidium>place the tar in the directory of the makefile, or make a bundle/bundles directory, but that doesn't make much sense as basically everything gets into the tar
15:56<Yexo>thanks, I'll do that
15:58<blathijs>OwenS: Yeah, that was the one :-)
16:03-!-tux_mark_5 [] has quit [Quit: KVIrc Insomnia 4.0.0, revision: , sources date: 20090115, built on: 2009/03/07 00:45:02 UTC]
16:04-!-tux_mark_5 [] has joined #openttd
16:05<Yexo>TrueBrain: can you please implement fs#2954 ?
16:11<TrueBrain>OwenS: instanceof !== typeof, I hope you do realise that
16:11<TrueBrain>typeof(instance) == class
16:11<TrueBrain>instanceof(instance) == "name-of-class"
16:11<TrueBrain>Yexo: tomorrow, if I don't forget
16:11<TrueBrain>(isn't the easiest thing to do :( )
16:11<Yexo>TrueBrain: bananas misses 0.7.2 as version
16:11<TrueBrain>that too, I will do tomorrow :)
16:12<Yexo>ok ;)
16:12<Yexo>and what's your opinion about removing license.txt from uploaded tar files if you don't specify 'custom license'?
16:12<OwenS>TrueBrain: In my case, a instanceof t = typeof a == t or typeof a == t.base or typeof a == t.base.base ...
16:14<Yexo>OwenS: that looks very wrong, don't you mean "subclass of t" instead of "t.base"?
16:14<OwenS>Sorry, by bases are on the wrong side :p
16:16-!-Markk [] has joined #openttd
16:17<TrueBrain>Yexo: explicit adding license.txt never hurts anyone
16:17<TrueBrain>the idea is simple: if you downloaded the file, you don't know the license
16:17<TrueBrain>so a license.txt is needed to specify that after downloading :)
16:17<Yexo>TrueBrain: that's completely besides the point
16:17<TrueBrain>oh, uploading
16:17<TrueBrain>sorry, read downloading
16:17<Yexo>yes :)
16:17<TrueBrain>at uploading it isn't needed
16:18<TrueBrain>or at least, not the last time I touched the code
16:18<TrueBrain>you only need to supply license.txt when you do a custom license
16:18<Yexo>last time I tested, license.txt if forbidden when uploading when chosing a non-custom license
16:18<TrueBrain>okay, so I guess I didn't understand you again :p
16:19<TrueBrain>this time it is more because of the way you wrote your question ;)
16:19<TrueBrain>I think you mean, that in case someone does supply a license.txt, it should silently be removed?
16:19<Yexo>the problem: I'd like to generate ai library tar files from a makefile. That makefile should include license.txt in the tar file. Now I have to make another tile file to upload to bananas.
16:19<frosch123>pick the language based on an white-space ignoring compare of the uploaded license with the default ones :)
16:19<@Rubidium>he wants that if license_selected != custom && license_added != None: license_added = None
16:19<TrueBrain>okay, then I understand what you mean .. took me long enough :p
16:20<TrueBrain>that said, I think it is good that you can't have a license.txt when selecting a non-custom
16:20<TrueBrain>say you have a custom license, you upload, select a license not knowing what it does
16:20<TrueBrain>poef, other license
16:20<Yexo>yes, but see my case above, there are some problems with it
16:20<TrueBrain>so what frosch123 says is an option
16:20<Yexo>that'd work fine :)
16:21<TrueBrain>k .. will put it on my list
16:21<TrueBrain>don't expect it any time soon, as BaNaNaS code is ugly!!! :p
16:21<TrueBrain>haha :)
16:21<TrueBrain>gone for the night
16:21<TrueBrain>bye all :)
16:21<Yexo>gn TrueBrain
16:21<@Rubidium>night TrueBrain
16:22<CIA-1>OpenTTD: rubidium * r17095 /trunk/src/ (strgen/strgen.cpp strings.cpp): -Codechange: make ParseStringChoice a bit safer
16:24-!-Progman [] has joined #openttd
16:29<CIA-1>OpenTTD: rubidium * r17096 /trunk/src/lang/english.txt: -Fix [FS#3091]: news message about ordered refits failing wasn't very clear
16:31-!-tdev [] has joined #openttd
16:37<@petern>hey tdev, fixed the scripting error yet? :D
16:37-!-lws1984 [] has joined #openttd
16:37<tdev>presentation at work today, was not able to work on anything else :/
16:38<@petern>oh, you mean you normally work on ror at work? ;)
16:42<@Belugas>Is it THAT surprising? heheh
16:46-!-frosch123 [] has quit [Remote host closed the connection]
16:48<Eddi|zuHause>what's weird about my railway is that i am missing (at least) two steam engines and a set of passenger+mail wagons
16:49-!-bb10 [] has quit [Ping timeout: 480 seconds]
16:49<Eddi|zuHause>i searched all accessible parts of the house, and couldn't find them
16:50<@Belugas>the cat swallowed them
16:50<Eddi|zuHause>i don't think so...
16:51<@Rubidium>they have been misplaced
16:51<@Rubidium>either by you, but you can't remember it, or by someone else, possibly without your consent
16:52<Eddi|zuHause>most likely, the haven't been touched since we moved
16:52<Eddi|zuHause>that was 12 years ago
16:52-!-Zr40 [] has joined #openttd
16:56<Eddi|zuHause>they could possibly be in places that are not accessible
16:58-!-Zr40_ [] has quit [Ping timeout: 480 seconds]
17:02-!-andythenorth [~andy@] has quit [Quit: andythenorth]
17:16<@Belugas>damned... ninjam is TOTALLY empty
17:17<@petern>i'm on our server
17:17<@petern>although not playing, hehe
17:17*petern moves
17:18<@Belugas>they'll be coming like flies over... hem...
17:18<@Belugas>like kids on a bag of candies ;)
17:21<Eddi|zuHause>and apparently my desk is not fully horizontal...
17:21<Eddi|zuHause>the wagons roll downhill
17:24<OwenS>Eddi|zuHause: Isn't that realistic? =P
17:25<Eddi|zuHause>err... yes, but here shouldn't be a "down"
17:26<OwenS>Our fish tank is on a decided slant. It's very obvious when the water level comes lower than the rim arround the top :p
17:26<OwenS>I think the slant is tank induced though (I.E. 0.3tonnes of water on some wood...)
17:27<@Belugas>told you, petern :) already 2 flies hem... kids
17:34<@petern>not working though
17:35-!-Svish [~Svish@] has quit [Quit: Quit]
17:37<Eddi|zuHause>who stole my screwdriver?
17:37-!-keoz [] has quit [Quit: Leaving]
17:37<@Rubidium>you put it with the steam engines
17:38-!-Elton04627 [] has quit [Read error: Connection reset by peer]
17:38<@Belugas>i'm not there... what can you expect ^_^
17:42<@Belugas>Eddi|zuHause !
17:42<@Belugas>[17:27] <Shinaku> oh btw
17:42<@Belugas>[17:27] <Shinaku> found your screwdriver
17:42<Eddi|zuHause>haha :P
17:43<Eddi|zuHause>where is this guy so i can sue him for stealing?
17:43<@Belugas>#tycoon :D
17:44-!-DPyro [] has quit []
17:44-!-Zahl [] has quit [Quit: *schiel*]
17:46<Eddi|zuHause>where do i get tiny replacement light bulbs from?
17:46<Eddi|zuHause>i have at least three that are broken
17:48-!-Cybertinus [] has quit [Remote host closed the connection]
17:48<@Belugas>Reno-Depot.. They carray a lot of bulb types
17:48<@Belugas>Yui could try Canadian Tires too, but they tend to be expensive
17:49<OwenS>Hmm... Perhaps the reason my compiler isn't compiling is that my front end is only calling the parser...
17:49<+glx>seems a valid reason OwenS
17:50<OwenS>Error loading intrinsic bitcode file: Invalid bitcode signature O_o
17:50<OwenS>src/AlterScript/Intrinsics.llo: ELF 64-bit LSB relocatable, x86-64, version 1 (SYSV), not stripped <-- May explain something
17:52<OwenS>I'm passing -emit-llvm; why are you emitting an elf? O_o
17:54<Eddi|zuHause>Belugas: i should mention that those are east-german made vehicles, so the companies which made those might not be around anymore
17:55<@Belugas>mmh... right
17:55<@Belugas>well... if you are looking for originals, might be better to go for ebay
17:56<@Belugas>otherwise, all the hobby shops, i'd say
17:56<@Belugas>hobby = home, right? let's go home!
17:57-!-Yexo [] has quit [Quit: bye]
18:00<CIA-1>OpenTTD: rubidium * r17097 /trunk/src/ (7 files in 2 dirs): -Fix [FS#3092] (r13256): make restart command work again and make the help show how it works and how it doesn't work
18:05<Eddi|zuHause>apparently, they are available on ebay 10 pieces for 5,90€ + 4,90€ shipping...
18:09-!-Pygma [~quassel@] has quit [Remote host closed the connection]
18:17<+glx>not expensive it seems
18:18<Eddi|zuHause>"function untested"
18:18<Eddi|zuHause>this steam engine is weird... the motor is mounted in the tender, not in the engine itself :p
18:19<Eddi|zuHause>my test track is too short...
18:21-!-MapperOG [] has joined #openttd
18:22<MapperOG>what do you think about simutrans?
18:22<Eddi|zuHause>we absolutely loathe and depise it... what did you think?
18:23-!-DPyro [] has joined #openttd
18:23<DPyro>but can your compiler compile a compiler?
18:23<MapperOG>nah, I was interested in a real answer, no experience with simutrans, just saw a video and noticed that it's quite similar to openttd
18:23-!-random345 [] has joined #openttd
18:23<OwenS>DPyro: Said function just wastes cycles. It calculates 2 * 4 then discards it :p
18:24<+glx>so it's a useful function :)
18:25-!-random345 [] has quit []
18:25<OwenS>Yeah. I haven't implemented function calls or return yet :p
18:27-!-PeterT [] has joined #openttd
18:27<PeterT>is TrueBrain here?
18:27<OwenS>No. Hes sleeping
18:28<PeterT>sorry to burden you with this, but when/if he wakes up, could you tell him i ordered a book about c++
18:29<PeterT>and also, where can i check what revision 0.6.3 was?
18:30<+glx>0.6.3 is 0.6.3
18:31<PeterT>so what is the revision closest to it (lower)
18:31<+glx>won't help
18:31<+glx>revisions are shared between branches
18:31<Ammler>buy a svn book
18:31<+glx>if you want 0.6.3 get tags/0.6.3
18:32<PeterT>no, im trying to figure out if r11047 savegame is compatible with 0.6.3
18:33<Ammler>afaik, stable branches don't make safe bumps
18:33<+glx>doesn't matter for savegames :)
18:33<OwenS>tags/0.6.3 will tell you which revision 0.6.3 was
18:33<OwenS>But, you need to check 0.6.0 for savegame versions :p
18:33<OwenS>Well, 0.6.0rcX? :P
18:33<+glx>btw just try to load it in 0.6.3 and you'll know
18:34<@Rubidium>tags/0.6.3 would say that it's current revision in 17097, which isn't quite useful
18:34<PeterT>i must not be following
18:36<OwenS>OK. Why is casting one double to a pointer working and another not? O_o
18:37<OwenS>Oh, unary operators :P
18:37<Ammler>PeterT: might be useful too
18:39<TrueBrain>the sleeping ghost speaks to you
18:40-!-Exl [] has quit []
18:40<PeterT>do i have to repeat myself or did you read it?
18:41<TrueBrain>I see you double
18:41<TrueBrain>so I think anything you say will be useless
18:41<TrueBrain>night all
18:48<OwenS>Yay! I just compiled a slightly longer useless function
18:49<OwenS>(Not executable yet because I haven't filled the Type structures with operators :P )
18:52<PeterT>i will know what that means when i finish my c++ book
18:52<PeterT>is it possible at ALL to patch IS and cargodest together?
18:52<OwenS>I wasn't talking C++ :p
18:52<+glx>everything is possible
18:54<Eddi|zuHause>*mental note* do not touch both tracks when 12V DC are attached to it...
18:54-!-nicfer [~Usuario@] has joined #openttd
18:55-!-Brianetta [] has quit [Quit: Tschüß]
18:55<OwenS>lol. Hurt? :p
18:56<fonsinchen>PeterT: it has been done with earlier revisions; multiple time and with cargodist as well as with cargodest
18:56<fonsinchen>So I guess it's not that hard.
18:57<PeterT>its been done? where can i download it?
18:58<Eddi|zuHause>OwenS: not hurt in that sense, but it's not exaclty a pleasant feeling
18:58<Eddi|zuHause>OwenS: like touching a cow fence
18:58<fonsinchen>Oh, sorry ... that was timetable separation, not infrastructure sharing.
18:59<Aali>How could you even *feel* 12V?
18:59<OwenS>I wish I could say I knew what a cow fence feels like... I think?
18:59<Aali>unless you were licking the rails :P
18:59<fonsinchen>so I guess it hasn't been done. Which suggests it's exceptionally hard.
18:59<Eddi|zuHause>fonsinchen: IS was done, too, afaik
19:00<PeterT>too bad
19:00<fonsinchen>Was it? Perhaps sometime earlier. Just look in the cargodist thread and you might find something.
19:00<PeterT>cargodEst thread
19:01<fonsinchen>well, if you want cargodEst I can't help you
19:01<PeterT>whats the difference, i could never tell?
19:01<fonsinchen>in short: cargodist is still developed, cargodest isn't
19:02<fonsinchen>also they have different concepts on what demand is and how multiple parallel flows are to be handled
19:02<fonsinchen>read it up in the Wiki
19:02<Eddi|zuHause>i'm quite sure that cargodEst and IS was tested
19:02<PeterT>and did it work?
19:03<OwenS>Probably. A long time ago. With a long obsolete nightly
19:03<PeterT>this is really annoying
19:04<Aali>merging IS2.0 and cargodEst is dead simple
19:04<Aali>cargodist I have never tried but I can't imagine it being much worse
19:04<PeterT>how is it dead simple?
19:04<Aali>are you a programmer?
19:04-!-KritiK [] has quit [Quit: Leaving]
19:04<Aali>do you have some insight into the openttd codebase?
19:04<Aali>then it's not simple
19:05<PeterT>i have ordered a book on c++
19:05<PeterT>but tell me your thoughts
19:05<Aali>so you have a few years to go then
19:05<Aali>just pick one of the two patches for the time being
19:05<Eddi|zuHause>did i ever tell you watching a 1'D1' engine making its rounds is awesome :)
19:05<fonsinchen>merging two patches can probably be done with less than a few years of experience
19:06<PeterT>how many dollars (euros) would i need to pay you to do it for me?
19:06<Aali>fonsinchen: yes, doing it properly however is another matter :P
19:06<Aali>PeterT: I'm on vacation
19:07<fonsinchen>Forget it. I'm not going to create more diverging versions of cargodist. If you want those, you have to do it yourself.
19:07<OwenS>Aali: The proper way is merge one into OpenTTD trunk, let the second's developer find all the subtle incompatibilities, then merge the second :p
19:07<PeterT>ok, well, im off
19:07-!-PeterT [] has quit [Quit: Leaving]
19:08<Eddi|zuHause>OwenS: the OpenTTD way is "let B merge A's patch, and then watch A throwing a hissy fit and revert B's commit"
19:08<Aali>it's not the subtle incompatibilities I'm worried about
19:09<Aali>everyone breaks the saveload code
19:09<OwenS>Eddi|zuHause: lol
19:09<Eddi|zuHause>OwenS: well, that actually happened ;)
19:09<OwenS>Aali: Aah true. Everything touches it :P
19:09<Aali>even if they by divine intervention manage to stitch the rest together
19:10<fonsinchen>It's probably impossible not to break the saveload code when merging two patches with savegame versions different from the trunk savegame version.
19:10<Aali>the saveload code simply cannot be stitched together without breaking
19:10<OwenS>Well, it can on rare occasion
19:11<Aali>fonsinchen: you can't make it load everything, obviously
19:11<Eddi|zuHause>i once managed to break my saveload code so far that i could load old versions of a patched savegame into trunk ;)
19:11<fonsinchen>We should make the savegame version an extendable vector, each entry telling about the version of some feature.
19:11<Aali>fonsinchen: no we should not :)
19:11<Eddi|zuHause>there is a minor savegame version that could be used for patches, but nobody had the courage to actually use that one
19:12<fonsinchen>Then I could add a linkgraph feature to the savegame and it wouldn't break compatibility with trunk if something is changed in trunk.
19:12<Eddi|zuHause>(it says "do not use!"
19:12<fonsinchen>Furthermore, it's not obvious how the minor version is intended to be used.
19:12<Aali>the minor version was killed for a good reason
19:13<Aali>it should stay dead
19:13<fonsinchen>Aali, do you actually like all patches breaking with each trunk savegame bump?
19:14<Aali>I couldn't care less
19:14<Aali>want to play an old patched game? use the old version that created it
19:15<fonsinchen>On the other hand you do like patches to be well tested before they are accepted, don't you?
19:15<Eddi|zuHause>Aali: that is not a good philosophy
19:16<fonsinchen>Being able to load old savegames encourages people to run tests with new versions of the patch.
19:16<fonsinchen>If they aren't able to load their savegames, they'll stick with some old version for playing them. Which is exactly what no one wants.
19:16<Eddi|zuHause>exactly. savegame incompatibility is the main reason why people stay away from patches
19:17<Aali>great, I'll just put down your names on the "Will provide support for all old versions of everything forever"-list then, shall I? :)
19:18<fonsinchen>I will only provide support for old versions of cargodist, provided the infrastructure is there to make that less of a pain than it is now.
19:19<fonsinchen>Other people could do the same with their patches, though.
19:22<Eddi|zuHause>Aali: nobody says trunk should be able to load such savegames
19:22-!-Elton03575 [] has joined #openttd
19:23-!-lws1984 [] has quit [Quit: leaving]
19:24-!-fonsinchen1 [] has joined #openttd
19:24-!-fonsinchen1 [] has quit [Remote host closed the connection]
19:25<Aali>you're missing the point
19:25<Eddi|zuHause>yes, because you did not actually mention the point
19:25<Aali>old savegames are as obsolete as the software that created them, _in my opinion_
19:26<Eddi|zuHause>all you splurt out are subtle implications of a great evil
19:26<Aali>therefore, time spent on making them compatible is time wasted
19:26<OwenS>Aali: The majority of people disagree with you
19:26<Eddi|zuHause>Aali: then why does OpenTTD support loading of 15 year old TTO savegames?
19:26<Aali>I'm aware of that
19:28<@Rubidium>then why have loading old savegames at all?
19:28<Eddi|zuHause>and why do people insist on using open standards when they notice that they can't open 10 year old Word documents with their current software?
19:30-!-fonsinchen [] has quit [Ping timeout: 480 seconds]
19:30<Eddi|zuHause>what if i get a new operation system and r12160 does not compile anymore?
19:30<Aali>because word documents are usually more important than your little collection of imaginary trains
19:31<Aali>this discussion has been completely derailed
19:31<Eddi|zuHause>maybe i steganographed secrets into my imaginary trains?
19:32<Eddi|zuHause>and in 5 years i urgently need to have access to these blueprints of the nuclear facility?
19:32<Eddi|zuHause>how can you say my savegames are not important?
19:32-!-Eddi|zuHause [] has quit []
19:33<OwenS>I think he revealed too much information :P
19:33-!-Eddi|zuHause [] has joined #openttd
19:33<OwenS>Men in black suits trying to get you? :P
19:34<Eddi|zuHause>they're called "BKA" over here...
19:35<Eddi|zuHause>and they're after filesharemurderterrorists
19:37<Eddi|zuHause>the funny thing is, i had such a nicely timed disconnection in the past already ;)
19:41<OwenS>Even better: My code generator is generating valid cod! :D
19:41<OwenS>(I've just been informed that an invalid operator was invoked :P )
19:53-!-tokai|mdlx [] has joined #openttd
19:54-!-tdev [] has quit [Quit: free open source vehicle simulator:]
19:56-!-tokai|mdlx [] has quit []
19:57-!-tokai|mdlx [] has joined #openttd
19:57-!-tokai [] has quit [Ping timeout: 480 seconds]
20:00-!-Progman [] has quit [Remote host closed the connection]
20:02-!-Zorni [] has joined #openttd
20:03-!-Coco-Banana-Man [] has quit [Quit: Raubgut ist vom Umtausch ausgeschlossen!]
20:09-!-tux_mark_5 [] has quit [Remote host closed the connection]
20:09-!-Zorn [] has quit [Ping timeout: 480 seconds]
20:32-!-Polygon [] has quit [Quit: Flieht, ihr Narren!]
20:40-!-KenjiE20|LT [] has joined #openttd
20:41-!-KenjiE20 [~KenjiE20@] has quit [Quit: WeeChat 0.3.0-rc2]
20:52-!-Chris_Booth [] has joined #openttd
21:05-!-OwenS [] has quit [Remote host closed the connection]
21:07-!-KenjiE20 [] has joined #openttd
21:07-!-KenjiE20 is "KenjiE20" on #openttd #tycoon @#openttdcoop #merctest #simsig
21:10-!-KenjiE20|LT is now known as Guest38
21:10-!-KenjiE20 is now known as KenjiE20|LT
21:13-!-Guest38 [] has quit [Ping timeout: 480 seconds]
21:19-!-Azrael- [] has quit [Quit: <3]
21:23-!-Eddi|zuHause2 [] has joined #openttd
21:23-!-Eddi|zuHause [] has quit [Read error: Connection reset by peer]
21:23-!-Eddi|zuHause2 is now known as Eddi|zuHause
21:29-!-Elton03575 [] has quit [Read error: Connection reset by peer]
21:52-!-MapperOG_ [] has joined #openttd
21:59-!-MapperOG [] has quit [Ping timeout: 480 seconds]
22:28-!-glx [glx@2a01:e35:2f59:c7c0:a95b:37a2:a8cf:4a17] has quit [Quit: bye]
22:31-!-KenjiE20|LT [] has quit [Ping timeout: 480 seconds]
22:53-!-SirSquid1ess [] has joined #openttd
22:57-!-SirSquidness [] has quit [Ping timeout: 480 seconds]
23:04-!-SirSquid1ess is now known as SirSquidness
23:04-!-nicfer [~Usuario@] has quit [Read error: Connection reset by peer]
23:08-!-TinoDidriksen [] has quit [Ping timeout: 480 seconds]
23:12-!-TinoDidriksen [] has joined #openttd
23:30-!-SirSquid1ess [] has joined #openttd
23:31-!-SirSquidness [] has quit [Ping timeout: 480 seconds]
23:34-!-SirSquid1ess is now known as SirSquidness
23:38-!-TinoDidriksen [] has quit [Ping timeout: 480 seconds]
23:42-!-TinoDidriksen [] has joined #openttd
23:49-!-Lakie [~Lakie@] has quit [Quit: Sleep.]
---Logclosed Fri Aug 07 00:00:44 2009