IRC logs for #openttd on OFTC at 2011-10-20
⏴ go to previous day
02:13:36 *** supermop__- has joined #openttd
03:42:27 *** pjpenispump has joined #openttd
04:56:39 *** Eddi|zuHause2 has joined #openttd
06:01:43 <peter1138> there's a bug in welsh
06:02:20 *** Cybertinus has joined #openttd
06:02:22 <peter1138> for STR_SETTINGS_MENU_GAME_OPTIONS it has "Dewisiadau Ge^m" instead of "Dewisiadau Gêm"
06:02:51 <peter1138> there's some more like that actually
06:09:43 *** devilsadvocate_ has quit IRC
06:09:46 *** devilsadvocate_ has joined #openttd
06:12:05 *** Maarten has joined #openttd
06:19:46 *** andythenorth has joined #openttd
06:20:52 *** sla_ro|master has joined #openttd
06:21:57 <Pikka> good morning andy a north
06:32:44 * Pikka listens to chameleon and codes all the things
06:34:26 *** valhallasw has joined #openttd
06:41:55 <CIA-6> OpenTTD: peter1138 * r23039 /trunk/src/lang/welsh.txt: -Fix: Replace e^ and w^ with ?\195?\170 and ?\197?\181 respectively.
06:43:29 <planetmaker> ehm... that's what wt3 is for?
06:44:32 <peter1138> it's part of our source code
06:52:04 <peter1138> i'm with tron there
06:52:44 <peter1138> if the translators don't care to notice something is wrong for 2 years...
06:54:22 *** erik1984 has joined #openttd
06:57:22 * andythenorth has been talking to peopl in australia
06:57:30 <andythenorth> and now australian peopl are talking here
07:03:38 <Pikka> there's a lot of it (ie, australians) about, you know.
07:07:01 *** andythenorth has joined #openttd
07:17:23 *** DayDreamer has joined #openttd
07:46:10 <andythenorth> new planes from Pikka
07:46:51 <Pikka> was meant to be 64 but I didn't quite make it :P
07:47:29 <Pikka> actually I've done nothing for the last three weeks, then today I modelled, rendered, cleaned and coded the last two planes that made it in. :]
07:48:09 <Pikka> and yes, 62 vs 29 in the original av8. :o
07:48:13 * andythenorth should do similar for FISH :P
07:48:50 <andythenorth> I am currently at 29
07:49:03 <planetmaker> what you were lazy?!
07:49:07 <andythenorth> drawing boats is rather a chore...
07:49:12 <Pikka> you've got 5 years to make another 33! :D
07:49:20 <planetmaker> incredible... ;-)
07:50:06 <planetmaker> sweet thing, Pikka :-)
07:51:30 <planetmaker> or should I say aluminum thing(s)? :-P
07:51:52 <planetmaker> and yes: "happy birthday" :-)
07:52:13 <Pikka> happy birthday to av8 :]
07:52:24 <Pikka> and to me last week :o I'm old now!
07:52:28 <planetmaker> yes, I was sure you could relay that ;-)
07:52:34 <planetmaker> oh... then belated to you, too :-)
07:52:48 * andythenorth will have to start new game for playing game purposes
07:52:54 <andythenorth> but now - other things
07:55:56 <z-MaTRiX> oh i have accidentally added opera's and google's spyware server domains to blocklist using a wildcard
07:56:47 <planetmaker> ah, sweet. The seaplanes are there but need a parameter. Sounds like exactly the right thing for that, Pikka :-)
07:58:45 <planetmaker> I should try at some stage to get an experimental seaport going. Possibly also requiring a parameter ;-)
07:59:05 <planetmaker> ... or (alternatively, also) checking that parameter? ;-)
07:59:31 <Pikka> the parameter is just there because there's no standard way of checking if an airport is a seaport yet
07:59:48 <Pikka> if the parameter is on, the plane can operate from any airport, if it's off, it can't operate from any
08:00:00 <Pikka> so it's really just a temporary thing
08:00:21 <Pikka> once we get a standard in for aircraft and airports to communicate with each other, I'll update the grf
08:00:57 <planetmaker> yes, I understand that
08:02:44 <planetmaker> that's why the parameter thing is IMHO the best solution to this. With default "not available"
08:04:21 <Pikka> it's always buildable, it just won't leave the hangar. :)
08:04:46 <Pikka> but I don't see why the airport would need any such parameter
08:05:44 <planetmaker> Pikka, for the same reason: any plane could land on it and it would incur possibly bug reports if a A380 lands in the sea and all is fine
08:06:51 <Pikka> although by the time it's a widely released grf hopefully the spec will have moved to the point that it can be handled "properly". :)
08:32:26 *** Progman has joined #openttd
08:50:52 *** Eddi|zuHause2 is now known as Eddi|zuHause
10:40:55 *** michi_cc has joined #openttd
10:40:55 *** ChanServ sets mode: +v michi_cc
10:51:26 *** Biolunar has joined #openttd
11:05:25 <Eddi|zuHause> no, google, i do not want to give you my phone number...
11:06:38 <Rubidium> Eddi|zuHause: just use +1.6502530000
11:07:06 <Eddi|zuHause> Rubidium: what's that?
11:07:43 <Rubidium> Eddi|zuHause: (obviously) one of Google's phone numbers ;)
11:09:36 <__ln__> Eddi|zuHause: every good citizen shares not only their own phone number but their whole contact lists with google.
11:16:43 <Elukka> every citizen will have their contact lists recreated by facebook from data taken from your acquaintances who let them use theirs, whether you want it or not or have anything to do with facebook
11:18:14 <Eddi|zuHause> none of my acquaintances really uses facebook
11:19:28 <__ln__> facebook also collects data on people who are not members.
11:21:04 <Eddi|zuHause> but the more removed the next person using facebook is from me, the less data they get
11:24:22 <Elukka> i don't use facebook, don't want anything to do with it, still got a LOOK, ALL THESE FRIENDS OF YOURS USE IT TOO email
11:24:51 <Eddi|zuHause> i never got one of those
11:25:15 <__ln__> and spotify made the ingenious decision that new users can only register through facebook.
11:31:42 <Eddi|zuHause> luckily, spotify is not available in my country, then.
12:04:54 *** Brianetta has joined #openttd
12:05:33 <Yexo> Eddi|zuHause: random_switch works if you apply that simple patch?
12:08:08 *** Adambean has joined #openttd
12:10:45 <Eddi|zuHause> next i need the cb23 patch commited :)
12:11:49 <Eddi|zuHause> when is devzone's nml updated so i know when it's safe to push my changes?
12:17:43 <Yexo> the nml nightly is build about 10 minutes before the grf nightlies are build
12:18:22 <Eddi|zuHause> oberhümer enabled build on push for some reason
12:21:54 <planetmaker> Then... it'll fail till the nightly is built
12:25:34 <Eddi|zuHause> exactly. and that's why i asked when that is
12:30:35 <planetmaker> it's in cet 18:10 iirc
12:36:30 <planetmaker> yes. NML is compiled at 19:10h CEST
13:10:39 *** blathijs_ is now known as blathijs
13:13:40 *** Kurimus has joined #openttd
13:30:42 <Eddi|zuHause> yesterday was WEDNESDAY!!!
13:35:14 <Belugas> wednesday... what a boring day...
13:42:26 *** Wolfsherz has joined #openttd
13:44:42 *** HerzogDeXtEr has joined #openttd
14:17:01 *** TWerkhoven has joined #openttd
14:34:55 *** z-MaTRiX has joined #openttd
14:39:39 <z-MaTRiX> 10 seconds of recording took 6MB on my "new" webcam
14:41:29 <z-MaTRiX> recompressed using lavc and got 1.1MB
14:42:16 <z-MaTRiX> maybe if i get the lighting right there will be less noise that must be compressed
14:43:23 <planetmaker> what format was the 6MB in?
14:43:53 <z-MaTRiX> 640x480x30fps i guess
14:44:21 <planetmaker> do you care about lossless or not lossless?
14:44:38 <Eddi|zuHause> i'd care about latency
14:44:57 <z-MaTRiX> i have realtime kernel
14:45:05 <Eddi|zuHause> but better compression is usually taking longer
14:45:32 <planetmaker> @calc 30*640*480*3
14:45:32 <DorpsGek> planetmaker: 27648000
14:45:37 <planetmaker> @calc 30*640*480*3 / 1024**2
14:45:37 <DorpsGek> planetmaker: 26.3671875
14:45:46 <z-MaTRiX> dont know havent tried any special programs yet
14:46:04 <planetmaker> any computer should be able to stream that to disk w/o compression
14:46:37 <planetmaker> thus you can use the strongest one which just doesn't pile up memory (if the campera writes images into a queue)
14:47:14 <Eddi|zuHause> that is really unimportant...
14:47:53 <z-MaTRiX> yes well was thinking about bandwidth will not cause images to queue
14:48:19 <z-MaTRiX> usb 1.1 would be 12Mbit/s and should be enough for 640x480x30fps anyway
14:48:35 <Eddi|zuHause> but my personal experience is that a/v asynch gets noticeable at around 100ms
14:48:45 <Eddi|zuHause> that's up to 3 frames
14:49:03 <Eddi|zuHause> so that's around the time it should take at maximum to compress one frame
14:49:04 <z-MaTRiX> dont know what would cause that
14:49:20 <planetmaker> it *really* depends on what you want to do and record with the webcam, Eddi|zuHause
14:49:37 <Eddi|zuHause> (at least if you want to transmit stuff)
14:49:56 <z-MaTRiX> ahm you thinking about streaming it through net right?
14:50:02 <planetmaker> even then. Who cares about 100ms latency on the status of the coffee machine?
14:50:39 <Eddi|zuHause> depends on whether the communication is one-way or two-way
14:51:36 <z-MaTRiX> btw you can get very cheap webcams nowadays, and they are better than security cameras
14:51:48 <z-MaTRiX> $6-$10 for a webcam...
14:52:04 <Eddi|zuHause> latency is the time between initiating an action and seeing the result of that action
14:52:13 <planetmaker> so you "misuse" on as security cam?
14:52:31 <z-MaTRiX> a camera is a camera no?
14:53:10 <z-MaTRiX> well actually thinking about warranty-vioding it
14:53:15 <planetmaker> but I'd not do everything with every camera and some are suited better for one thing or the other
14:53:23 <Eddi|zuHause> occasionally people disagree whether a "Quellen-TKÜ"-trojan is a "Onlinedurchsuchung"-trojan :p
14:54:11 <planetmaker> which is by rule of the constitutional court (and also otherwise) very different things
14:54:22 <planetmaker> If only our minister for internal affairs would see that...
14:54:49 <planetmaker> he's supposed to protect the constitution and not advocate violation of it despite contrary rulings...
14:54:50 <z-MaTRiX> its funny this cam doesnt have new windows drivers, but works instantly on linux
14:56:40 <z-MaTRiX> btw did you know opera and google has spyware servers?
15:09:55 *** LordAro has joined #openttd
15:10:05 <z-MaTRiX> lpp01m01-in-f120.1e100.net:http ; fx-in-f147.1e100.net:http
15:10:15 <z-MaTRiX> sitecheck2.opera.com
15:20:49 *** sla_ro|master has joined #openttd
15:31:22 <LordAro> i'm getting bored of ubuntu's little problems
15:31:30 <LordAro> how does changing to debian sound?
15:32:38 <Terkhen> that's was my reason for trying debian after years of ubuntu being the only distro I knew how to use
15:38:15 <Eddi|zuHause> so... how do i set train property 1A (sort purchase list) in nml?
15:38:45 <Terkhen> debian is fine, but I ended up changing to arch because I wanted to experiment a bit
15:38:50 <Terkhen> next time, I'll probably install debian agian
15:40:57 <Yexo> Eddi|zuHause: you'd have to wait for #2856 to be implemented
15:42:54 <Eddi|zuHause> i'd do that right away, but i don't know enough about the internals of nml
15:43:27 <Yexo> sort_vehicles(veha, vehb, vehc, ...); <- is that syntax ok for you?
15:44:36 <Eddi|zuHause> it should be a fairly trivial thing, something like: for vehicle in vehicle_list[:-1]: write(prop_1A = vehicle_list[-1])
15:44:48 <Yexo> that was my plan exactly
15:45:13 <Eddi|zuHause> syntax looks alright
15:47:56 <LordAro> Terkhen/whoever: cool, i think i'll probably do that in the near future. just 1 thing to check - (most) programs that run under ubuntu will also work with debian, yes?
15:48:08 <Yexo> I'll make it sort_vehicles(FEAT_TRAINS, veha, vehb...); for easier coding
15:49:02 <Terkhen> LordAro: debian and ubuntu are very similar (ubuntu is still using a lot of stuff from debian IIRC) so it should feel similar
15:49:07 <Eddi|zuHause> or make it sort(FEAT, [a, b, ...])?
15:49:18 <Eddi|zuHause> may make it more obvious
15:49:22 <Terkhen> also, a program that works in a linux distribution should work in all of them
15:49:35 <Terkhen> unless something strange happens or you are not running the same DE
15:50:13 <LordAro> DE? (development environment?)
15:50:52 <Terkhen> even then they should work too
15:51:02 <Terkhen> but you can get some strange things :P
15:52:06 <Eddi|zuHause> for KDE apps it should usually only require to have KDE installed, they usually run under all other DEs as well
15:52:53 <LordAro> that leads onto the next question: which DE?
15:53:08 <LordAro> (do you use/like most)
15:53:19 <Terkhen> gnome 2, which is the one I'm used to work with
15:53:21 <Eddi|zuHause> don't ever ask such questions
15:54:34 <Eddi|zuHause> with 100 people in the channel, chances are you'll get 120 different answers and 3 flamewars
15:54:55 <Terkhen> don't teach everyone how to troll easily either :P
16:08:01 *** Bluelight has joined #openttd
16:24:09 *** Prof_Frink has joined #openttd
16:33:15 *** joho^_^ is now known as joho
16:35:29 <Eddi|zuHause> Yexo: actually this would be enough for me right now: www.informatik.uni-halle.de/~krause/vehicle_list_sort.diff
16:38:23 <Yexo> I'll have an implementation of "sort(feat, [veh_a, veh_b]);" within an hour
16:38:51 <Eddi|zuHause> well, you have only half an hour until nml is built :)
16:39:09 <Eddi|zuHause> and i'm sure it makes sense to provide both methods
16:39:49 <peter1138> the original music sounds weird with fluidsynth
16:40:03 <peter1138> (with the fluid sf2)
16:47:41 <LordAro> within an hour? within 10 mins more like :)
16:47:44 <Yexo> Eddi|zuHause: sort-statement is in
16:47:55 <Yexo> I've attached your patch to the issue tracker but won't commit it now
16:48:03 <peter1138> midi on the sb live sounds a bit flat
16:48:37 <peter1138> hmm, the end of the intro theme seems to have the start of the next track
16:49:17 <Eddi|zuHause> now i need to ponder sorting algorithms
16:49:32 <CIA-6> OpenTTD: yexo * r23040 /trunk/src/build_vehicle_gui.cpp: -Add: [NewGRF] allow use of newgrf textstack during callback 23
16:49:35 <Eddi|zuHause> not how-to-sort :)
16:51:01 <CIA-6> OpenTTD: yexo * r23041 /trunk/src/strings.cpp: -Codechange: rename slightly unclear parameter name
16:51:33 <peter1138> hurr, the 4-6-0 king arthur is broked :S
16:51:55 <CIA-6> OpenTTD: yexo * r23042 /trunk/src/ (strings.cpp strings_func.h): -Codechange: make case_index an explicit parameter of GetStringWithArgs instead of hiding it in the stringid parameter
16:52:09 *** valhallasw has joined #openttd
16:52:59 <CIA-6> OpenTTD: yexo * r23043 /trunk/src/strings.cpp: -Fix: allow the use of cases only for those cases where strgen allows them and reset the case_index properly after {DATE} and {DATE_LONG} codes
16:53:35 <CIA-6> OpenTTD: yexo * r23044 /trunk/src/strings.cpp: -Codechange: reorder parameter order so we can make use of the default=0 more often
16:57:09 *** valhalla1w has joined #openttd
16:59:08 <Eddi|zuHause> without ; i get "unexpected end of file"
16:59:55 <planetmaker> remove the last ,
17:00:12 <Eddi|zuHause> planetmaker: that should really not be a problem
17:03:30 <Eddi|zuHause> planetmaker: also happens without that ,
17:05:03 <Eddi|zuHause> + raise generic.ScriptError("Second parameter is not an array of one of the items in it could not be reduced to a constnat numer", self.pos) <-- typo "of"->"or"
17:05:06 <Yexo> Eddi|zuHause: the function is called "sort" as you suggested, not "sort_vehicles"
17:06:07 <Yexo> the error should be improved, but that's more work
17:06:33 <Eddi|zuHause> right, getting new error now. this time a real one :)
17:06:39 <Eddi|zuHause> i can solve that one :)
17:08:00 *** Brianetta has joined #openttd
17:08:21 <Eddi|zuHause> i need another refactoring, i think...
17:10:52 *** Alberth has joined #openttd
17:10:52 *** ChanServ sets mode: +o Alberth
17:28:02 *** TWerkhoven2 has joined #openttd
17:42:37 <z-MaTRiX> voided warrant of my webcam
17:42:52 <z-MaTRiX> now it can see infrared
17:44:47 <z-MaTRiX> sounds like you dont like the idea
17:45:09 <peter1138> my video camera has a switch to enable "night mode", heh
17:45:27 <z-MaTRiX> and i have remoed the IR filter
17:45:34 <peter1138> though it goes all green, heh
17:46:38 <z-MaTRiX> my mobile phone had night mode too
17:46:52 <z-MaTRiX> but that only adds blur
17:47:10 <z-MaTRiX> (more exposure time)
17:47:44 *** supermop has joined #openttd
17:47:51 *** mahmoud has joined #openttd
17:48:29 * SpComb sets aperture to f/0.8
17:49:08 * planetmaker just bought 70mm @ f/2.8 ;-)
17:49:35 <z-MaTRiX> do you make IR photos?
17:49:58 <z-MaTRiX> i saw some interesting ones
17:50:01 <planetmaker> at least not normally
17:50:14 <planetmaker> making IR photos usually requires also some hardware modification
17:50:19 <LordAro> hai Alberth, get my pm?
17:50:20 <planetmaker> and specific lenses
17:50:36 <Alberth> yep, you got mine? :)
17:50:56 <planetmaker> z-MaTRiX, but... if I need it... I have an IR camera 4 doors further down the hallway :-P
17:51:05 <LordAro> (i.e. haven't looked)
17:51:56 <Alberth> phew! (was afraid I didn't press 'send') :)
17:52:53 <Eddi|zuHause> anyone have python magic to shorten expressions like "x if f(x) else y", to avoid repeating the x?
17:53:39 <peter1138> "f(x) ? x : y" ? :p
17:54:08 <peter1138> or i don't quite understand you
17:54:21 <b_jonas> Eddi|zuHause: make a function for it
17:54:34 <b_jonas> so you just write foo(x,f,lambda: y)
17:54:44 <b_jonas> use a better name instead of foo
17:55:50 <Eddi|zuHause> SpComb: that cannot be used inside an expression
17:56:23 <Eddi|zuHause> peter1138: that's exactly what this already is
17:56:28 <SpComb> use a statement then :p
17:56:54 <Eddi|zuHause> SpComb: but functional programming is way more beautiful :)
18:00:34 <Alberth> you're using the wrong language :)
18:15:25 <Alberth> Eddi|zuHause: x or y :)
18:17:02 <Alberth> it is only a few miles outside 'pythonic'
18:20:38 <Eddi|zuHause> Alberth: that requires lazy evaluation on "bitwise" or :)
18:20:45 <blathijs> Alberth: (filter(f, [x]) + [y])[0] ?
18:21:02 <blathijs> uh, s/Albert/Eddi|zuHause/
18:21:15 <Alberth> Eddi|zuHause: luckily python does that :)
18:21:33 <blathijs> Alberth: Isn't the f() gone in your "x or y" ?
18:22:19 <Alberth> good point, forgot about f() :(
18:22:28 <blathijs> Eddi|zuHause: (I know the filter thing is quite ugly, but it doesn't repeat the x, like you asked, and 'filter' was the first thing that came to my mind when you said something about "return x if f(x) is true)
18:22:49 <Yexo> (lambda a,b: a if a else b)(x, y) <- even more ugly
18:22:56 <Eddi|zuHause> blathijs: hm, that might be interesting
18:24:16 <Alberth> Eddi|zuHause: write a function like the one by Yexo, and use that
18:24:49 <Eddi|zuHause> basically i have a sequence of operations, and i need the first non-zero result
18:25:08 <Eddi|zuHause> so filter may do exactly that
18:26:06 <blathijs> Eddi|zuHause: sequence of operations, or sequence of operands?
18:26:16 <blathijs> filter does the latter, but a smart lambda might of course fix that
18:26:17 <Eddi|zuHause> sequence of operations
18:29:09 <Alberth> print next(x for x in [0,0,0,12,13] if x)
18:29:28 <Alberth> oh 'if f(x)' of course
18:29:28 <SpComb> for x in l: if x: return x
18:29:48 <Yexo> spcomb: that's a statement, no longer an expression
18:30:00 <SpComb> it's a function definition body
18:30:23 <Eddi|zuHause> that's the same thing
18:30:59 <Eddi|zuHause> i don't think SpComb has quite grasped the meaning of functional programming :)
18:32:02 <SpComb> I don't think Eddi has grasped the point in writing Python code :p
18:32:07 <Yexo> def getval(list): return list[0] if f(list[0]) else getval(list[1:])
18:32:22 <SpComb> unles this is all esoteric something
18:32:49 <SpComb> what concrete state/code do you have now? Pastebin?
18:32:50 <Eddi|zuHause> the filter thing works well, thanks blathijs
18:33:44 <Eddi|zuHause> still room for improvement :)
18:34:30 <Eddi|zuHause> SpComb: it's a simple sorting function :)
18:37:22 *** andythenorth has joined #openttd
18:37:23 <SpComb> Eddi|zuHause: you're trying to be too clever or something
18:37:47 <Alberth> except filter filters over all values instead of upto the first non-zero one :)
18:37:50 <Eddi|zuHause> SpComb: i'm trying to remove redundancy
18:37:57 <SpComb> then write helper functions
18:38:08 <Eddi|zuHause> Alberth: small detail
18:38:12 <SpComb> first(cmp(table[key1][item], table[key2][item]) for item in ("...", ...))
18:38:30 <SpComb> and define those string keys somewhere not-inline
18:39:09 <SpComb> takes way too much seeking to read that code
18:39:34 <Alberth> and it has too many lambdas :)
18:41:12 <michi_cc> Eddi|zuHause: There are quite some "pure white" warnings in CETS, is that the code or the sprites?
18:41:27 *** frosch123 has joined #openttd
18:43:21 <Eddi|zuHause> Alberth: ooh, i was about to add more lambdas :p
18:46:15 <Alberth> you should use 'def' instead :)
18:55:32 <Eddi|zuHause> should use more "map", i think...
19:02:08 <valhalla1w> Eddi|zuHause: I prefer the (x["allowed" for x in (table[key1], table[key2])) syntax
19:02:14 <CIA-6> OpenTTD: yexo * r23045 /trunk/src/build_vehicle_gui.cpp: -Change: copy 6 registers to newgrf textstack after cb23 instead of 4
19:02:44 <Eddi|zuHause> valhalla1w: maybe
19:02:48 <valhalla1w> and besides, it would be more readable to extract a cmp(a,b) method
19:03:23 <planetmaker> hello andythenorth
19:03:31 <Eddi|zuHause> valhalla1w: how do you mean?
19:03:44 <valhalla1w> Eddi|zuHause: sec, let me get creative
19:04:07 * Alberth ponders about 'metal fabrication plant'
19:04:10 <Eddi|zuHause> valhalla1w: most of these things are strings
19:04:47 <andythenorth> Alberth: ponders about the translation?
19:05:11 <Alberth> especially the 'metal' part :)
19:05:15 <andythenorth> it's a bit of a cheat industry
19:05:39 <Alberth> I have now a 'parts factory'
19:06:15 <Eddi|zuHause> does that fabricate metal, or fabricate stuff from metal?
19:06:25 <valhalla1w> oh, it's slightly more difficult than I thought
19:06:34 <andythenorth> it fabricates and paints metal parts
19:07:16 <andythenorth> depending how you interpret it, they could be anything from large custom parts for construction (building materials), to 10,000 car doors per week
19:07:29 <planetmaker> Alberth: "Metallwaren-Fabrik"
19:07:30 <andythenorth> presses, stamps, cuts, welds metal
19:07:38 <planetmaker> maybe you can Dutch-y-fy that ;-)
19:07:40 <andythenorth> then paints it (hence chemicals)
19:07:40 <Alberth> like metal fabrication plant parts :p
19:08:30 <Alberth> it needs the 'metal' bit too in the name, I guess
19:08:59 <andythenorth> doesn't have to if it translates better another way
19:09:13 <andythenorth> it's only there to make it easy to figure out the chain + what it does
19:09:32 <valhalla1w> Eddi|zuHause: argh, and it's actually pretty elaborate -_-
19:10:10 <Eddi|zuHause> valhalla1w: unless you have a really genious idea, i'll leave it at that now
19:10:32 <valhalla1w> Eddi|zuHause: using 'or' instead of filter might be a start ;-)
19:10:48 <valhalla1w> oh, no, that doesn't work either
19:11:10 <valhalla1w> but adding two lines of comments 'this does this' might be a good idea
19:11:34 <valhalla1w> '# wasted time on improvements: 15 minutes'
19:11:44 <andythenorth> Alberth: is 'metal workshop' clear?
19:12:36 *** Brianetta has joined #openttd
19:14:00 <valhalla1w> Eddi|zuHause: yes, that helps
19:14:36 <valhalla1w> Eddi|zuHause: actually: it does the same as 'or'
19:14:47 <valhalla1w> 0 or -1 or 1 or 0 == -1, 0 or 0 or 1 or -1 or 0 == 1
19:15:08 <Eddi|zuHause> maybe, but the list syntax is cleaner, i think
19:15:19 <Eddi|zuHause> could fold on "or"
19:15:45 <valhalla1w> yeah, the or syntax is not really clear either
19:16:03 <Eddi|zuHause> but that again involves iterating over all items, so it's no improvement
19:17:20 <Alberth> andythenorth: oh, you changed that too, and I really liked my translation :p
19:18:34 <Alberth> I am working on the list provided by check_language from top to bottom :p
19:20:46 <Yexo> the check_language script should ignore strings with cases instead of reporting them as superfluous
19:22:06 <Alberth> it should also allow for noting 'current translation is correct'
19:24:05 *** mahmoud has joined #openttd
19:24:54 <Yexo> that needs either external storage or modifying the language file without changing anything, ie by adding/removing a space before the colon
19:27:55 <Alberth> it needs a proper translation web service :p
19:28:02 <Eddi|zuHause> who wanted to extend webtranslator for grfs again? :p
19:28:23 <Yexo> TrueBrain, where are you? :p
19:30:14 <Alberth> hmm, why do you always see errors after a commit? :p
19:30:49 <Eddi|zuHause> that's some kind of natural law :)
19:31:17 <TrueBrain> Yexo: in Leiden, why?
19:58:58 <blathijs> Hmm, can you iterate a list-of-lists in a list comprehension in Python?
20:00:12 <andythenorth> if you're smart enough
20:00:23 <andythenorth> list comprehensions are a bit unpythonic
20:00:36 <Eddi|zuHause> fold the list of lists?
20:00:38 <andythenorth> they're very cool, but can be baffling
20:00:44 <Alberth> blathijs: yep, [x for xl in L for x in xl]
20:01:19 <andythenorth> what's wrong with lots of 'for i in foo:' stuff
20:01:26 <blathijs> andythenorth: Are they? I found the pretty functional and neat, having worked with Haskell for a year or so :-)
20:01:27 <Eddi|zuHause> i did something like that once, but don't remember it
20:01:46 <blathijs> Alberth: I don't think that worked
20:02:03 <andythenorth> blathijs: simple ones are very neat, they can get complex
20:02:15 <blathijs> Alberth: Oh wait, it does
20:02:22 <blathijs> Alberth: I had the for's reversed
20:02:50 * andythenorth is not a great python coder
20:03:13 <blathijs> I had expected the fors in a list comprehension to be processed in reverse order, but apparently they aren't
20:03:34 <Eddi|zuHause> andythenorth: if you never properly learned functional languages, you'll find no benefits of list comprehensions
20:03:35 <Alberth> no, just from left to right, except for the final expression
20:03:51 <andythenorth> Eddi|zuHause: I never properly learned functional languages :)
20:04:12 <andythenorth> it's when you nest list comprehensions that it can be baffling
20:04:16 *** Biolunar_ has joined #openttd
20:04:34 <andythenorth> I would prefer the verbose version
20:04:38 <blathijs> Alberth: I guess that's the reason I expected the reverse; the final expression is at the left :-)
20:05:02 <andythenorth> complex nested list comprehension smells of one coder showing off, leaving the next guy thinking 'wtf?'
20:05:29 <Alberth> blathijs: just plain mathematics: { z | x\in X, y\in Y, z = x+y }
20:05:50 <blathijs> Alberth: Good point :-)
20:06:25 <blathijs> andythenorth: In this case, it's finding out the maximum label length from a list of widgets, so that's still pretty clear with a comprehension
20:06:36 <blathijs> it's just that the list got nested just now, so I needed to adapt it :-)
20:06:49 <blathijs> It works now, though. Thanks guys :-)
20:07:29 <Eddi|zuHause> blathijs: it may be better to use a recursive funtion
20:07:44 <Eddi|zuHause> or you'll rewrite this again for the next level of lists
20:08:00 * andythenorth doesn't really grok functional programming
20:08:03 <blathijs> Eddi|zuHause: It's not nested arbitrarily
20:09:12 <Alberth> andythenorth: it is mostly about expressing some algorithm in a neat and tidy way
20:09:54 <Alberth> ie you don't need extra variables, initialization, and other clutter around it
20:12:50 * andythenorth always thinks functional programming looks like lego :P
20:12:52 <Eddi|zuHause> non-functional functions like list.sort are tricky
20:13:03 <andythenorth> map | reduce is usually functional?
20:13:26 <Yexo> Eddi|zuHause: there is also the function "sorted"
20:13:45 <Eddi|zuHause> Yexo: i never remember these things
20:14:20 <blathijs> Strong typesystems associated with some functional languages are also a blessing, IMHO
20:14:34 <blathijs> The typechecker usually found 99% of the bugs I coded into my Haskell programs :-)
20:15:05 <Eddi|zuHause> i could use with some static typechecking in python
20:18:33 <Eddi|zuHause> what also might be useful is a syntax like "def func(param1:type1, param2:type2) that automatically generates an "assert isinstance(param1,type1)" etc.
20:19:54 <blathijs> I guess that's not very duck-typy
20:20:53 <Eddi|zuHause> blathijs: yes, isinstance kinda contradicts this paradigm
20:21:12 *** FFLaguna has joined #openttd
20:21:29 <Eddi|zuHause> so there also needs to be an isduck(var, type)
20:21:39 <Eddi|zuHause> where type is an interface i need
20:22:41 <FFLaguna> Hello, is there a command or setting to let everyone start with $1,000,000 in a multiplayer game? :)
20:23:04 *** mahmoud has joined #openttd
20:23:26 <Eddi|zuHause> FFLaguna: everyone can just go to the finance window, and ctrl+click on lend money
20:24:15 <Eddi|zuHause> (or is it "borrow"?)
20:24:51 <FFLaguna> Can I change how much they can borrow, initially?
20:25:03 <Yexo> anyway, start the multiplayer game, create all companies, save game, open in singleplayer, cheat money for all companies, save game, load in multiplayer
20:25:12 <Eddi|zuHause> FFLaguna: yes, in the difficulty settings
20:25:34 <FFLaguna> Eddi - Does that work on multiplayer? I seem to recall that even though I could loan $600,000 in a game I was hosting, everyone else could only loan $300k
20:25:46 <andythenorth> Eddi|zuHause: just try casting it
20:25:53 <andythenorth> with try: except:
20:26:00 <Yexo> FFLaguna: yes, it works in a multiplayer game
20:26:06 <andythenorth> first try, then ask forgiveness
20:26:08 <Yexo> as for your example, are you sure everyone was using the same currency?
20:26:20 <Yexo> perhaps you could loan 600k dollar and the other players 300k pounds
20:26:36 <andythenorth> Eddi|zuHause: enough variations on str(var), int(var) etc. will tell you what it is :P
20:27:09 <andythenorth> although you have to pass errors silently, which is considered to be A Very Bad Thing
20:27:11 <FFLaguna> Yexo - I'll try starting multiplayer and editing in singleplayer
20:27:16 <Eddi|zuHause> andythenorth: exceptions don't mix well with functional programming
20:28:04 <planetmaker> FFLaguna: different currency might fool you into believing that different players have available different amounts of loan
20:28:18 <planetmaker> but... yexo said so already :-)
20:44:57 <Eddi|zuHause> planetmaker: any ideas about improving the sort order? i'm not finding a good way to ensure passenger and mail wagons stick together
20:45:50 <andythenorth> Eddi|zuHause: patch ottd?
20:46:23 <Eddi|zuHause> andythenorth: not sure how that's going to help
20:46:25 <andythenorth> what's the issue?
20:46:34 <andythenorth> nml provides the IDs for you?
20:46:51 * planetmaker hasn't yet looked at that.
20:47:11 <planetmaker> What stops you to script it such that passengers are put there first, mail 2nd etc?
20:47:21 <planetmaker> or how is it done currently?
20:47:35 <Eddi|zuHause> currently i sort by refittable cargo classes
20:48:04 <andythenorth> leave it to players?
20:48:12 <andythenorth> the list is sortable in game...
20:48:25 <Eddi|zuHause> and that is a lexicographic sort
20:48:36 <Eddi|zuHause> so it depends on the names of the cargo classes
20:49:36 <Eddi|zuHause> so i need some exception for the passenger and mail cargo classes
20:51:57 * andythenorth ponders FIRS station rating algo.
20:52:07 <andythenorth> might need tweaking
20:52:13 <andythenorth> or might be near-perfect
20:52:21 <FFLaguna> Does the dev version of OTTD have a lot of important changes not found in the latest stable?
20:53:12 <andythenorth> you can use dev versions of of several newgrfs for starters
20:53:18 <FFLaguna> Ah, found the changelog
20:54:03 <FFLaguna> Oh, doesn't really look like that many changes, to be honest
20:59:15 <Rubidium> there not being a lot of important changes in the dev version means that we release often enough ;)
20:59:37 <Rubidium> also the number of important changes that can be done diminishes over time
21:01:13 * planetmaker wouldn't want to do without the newgrf changes, though ;-)
21:02:54 * andythenorth ponders changing FIRS grfid
21:03:53 * planetmaker brought up that idea but currently thinks "wait till 1.2 beta and don't change it" ;-)
21:04:38 <FFLaguna> Exactly, seems the last release was in September :)
21:05:34 <planetmaker> FFLaguna: releases which bring new features are only once per year really ;)
21:05:53 <planetmaker> so to see what is added you have to go back to everything since about March
21:06:18 <planetmaker> around the year we add to the stable branch only bug fixes
21:21:25 * Yexo wonders what changelog FFLaguna found
21:25:56 <FFLaguna> Seems it's just for one build, though
21:26:46 <Yexo> that one seems to be fore amore than a single build
21:29:10 <Eddi|zuHause> it's by far not the complete list of changes since last major release
21:29:11 <planetmaker> yeah... that seems like one nightly's changelog
21:30:55 <Yexo> r22082 was the 1.1 branch, so almost all features introduces after that are not in 1.1
21:31:29 <planetmaker> did we port a single _feature_?
21:32:13 <Yexo> - Feature: [NewGRF] Allow to filter by town of the current industry when using industry variable 0x68 [FS#4591] (r22434)
21:32:34 <Yexo> and a few more in 1.1.0-RC1, but nothing significant
21:35:38 *** KenjiE20 has joined #openttd
21:42:51 *** Chruker has joined #openttd
21:46:20 *** Chruker has joined #openttd
22:04:57 *** mahmoud has joined #openttd
22:21:35 <FFLaguna> Can anyone give me the main new features of the latest dev builds, compared to the stables?
22:21:51 <FFLaguna> What are the big changes that come to mind first?
22:23:02 <Eddi|zuHause> lots of new newgrf features
22:23:42 <FFLaguna> Anything that interfaces with 2cc Trainset and NuTracks?
22:24:20 <Eddi|zuHause> not that i know of
22:24:54 <FFLaguna> Subways on NuTracks... you still have to bulldoze the houses and roads in cities to build subways?
22:25:14 <KenjiE20> nightlies generally you'll get the new features as they get made, and all of their associated bugs
22:25:34 <KenjiE20> sometimes, that's server breaking :p
22:25:49 <Eddi|zuHause> FIRS uses some of the new newgrf features
22:26:15 <Eddi|zuHause> and CETS, but that doesn't really count yet :)
22:26:49 <KenjiE20> you could probably ask in #openttdcoop.devzone for a more direct/verbose answer
22:26:58 <FFLaguna> Hmm... how does FIRS integrate with ECS? Are they compatible? Nothing on the wiki page about it
22:27:40 <Eddi|zuHause> no. it's an either-or thing
22:27:40 <FFLaguna> Or are they mutually exclusive?
22:27:41 <planetmaker> 100% incompatible
22:27:59 <KenjiE20> they both break industires equally? :P
22:27:59 <FFLaguna> So which is better? ;)
22:28:40 <KenjiE20> FFLaguna: have a dig about in the coop archive's I think we've had both ecs and firs games in the past
22:28:51 <Elukka> my opinion is FIRS is better designed and it has more consistently good graphics
22:29:02 <Elukka> but fundamentally which is better is entirely subjective
22:29:15 <Yexo> heh, I just went over the changelog looking for commits starting with -Feature. From the last 6 months, we have some minor things to do with transparancy, some gui tweaks, and besides that only newgrf features
22:29:46 <FFLaguna> Yexo - Oh, thank you very much
22:29:50 <KenjiE20> Yexo: don't forget undocumented features :)
22:30:01 <Eddi|zuHause> Yexo: you miss the "Add YACD" and "Add more height levels" :p
22:31:33 <Yexo> hmm, I did miss that one
22:31:56 <planetmaker> -Add: ... probably
22:32:21 <Yexo> (svn r22767) -Add: river generation
22:32:56 <Eddi|zuHause> blame Rubidium for being too good at hiding things :)
22:33:00 *** DayDreamer has joined #openttd
22:34:35 <Elukka> looks much better to me
22:34:40 <Elukka> might even make me use fences
22:38:09 <planetmaker> Eddi|zuHause: was it ever posted to FS?
22:38:49 <planetmaker> it looks better, yes... but now is really sleeping time :-) Good night
continue to next day ⏵
