Log for #openttd on 29th October 2015:
Times are UTC Toggle Colours
00:02:58  *** Wormnest [] has quit [Quit: Leaving]
00:04:47  *** Pikka [] has joined #openttd
00:12:08  *** OsteHovel [] has quit [Ping timeout: 480 seconds]
00:14:57  *** OsteHovel [] has joined #openttd
00:22:40  <Wolf01> 'night
00:22:43  *** Wolf01 [] has quit [Quit: Once again the world is quick to bury me.]
00:26:47  *** Snail [] has joined #openttd
00:39:09  *** HerzogDeXtEr [] has quit [Quit: Leaving.]
00:54:45  *** sim-al2 [] has joined #openttd
01:07:16  <berndj> how close does a station need to be to a thing before that thing starts sulking that it isn't being serviced regularly enough? i have a nice big coal mine i want to service, but an iron ore mine is nearby too, that i don't want to annoy
01:08:20  <Eddi|zuHause> if you never pick up any iron ore, the mine doesn't mind
01:11:53  <Eddi|zuHause> ("pick up" in this context means sending a vehicle could carry that cargo)
01:12:25  <berndj> so as long as i don't start, they don't care? but towns/cities do care if you start building stuff nearby?
01:12:39  <Eddi|zuHause> yes
01:30:45  *** Pokka [] has joined #openttd
01:34:52  *** cha0tics [] has joined #openttd
01:36:14  *** fjb [] has quit [Ping timeout: 480 seconds]
01:37:33  *** Pikka [] has quit [Ping timeout: 480 seconds]
01:47:40  *** GenevaMode [~Adium@] has quit [Quit: Leaving.]
02:25:55  *** Pokka [] has quit [Read error: Connection reset by peer]
02:26:20  *** Pokka [] has joined #openttd
03:08:47  *** tipsyTentacle [~tipsy@] has joined #openttd
03:45:30  *** Snail [] has quit [Quit: Snail]
03:52:41  *** glx [] has quit [Quit: Bye]
03:53:21  *** Pokka [] has quit [Read error: Connection reset by peer]
03:53:46  *** Pokka [] has joined #openttd
04:01:42  *** smoke_fumus [~smoke_fum@] has quit [Quit: KVIrc 4.2.0 Equilibrium]
04:02:49  *** Pokka [] has quit [Read error: Connection reset by peer]
04:03:13  *** Pokka [] has joined #openttd
05:03:25  *** Pukka [] has joined #openttd
05:10:04  *** Pokka [] has quit [Ping timeout: 480 seconds]
05:20:46  *** Flygon [] has quit [Read error: Connection reset by peer]
05:20:59  *** Flygon [] has joined #openttd
05:23:55  *** joho [] has quit [Remote host closed the connection]
05:24:07  *** joho [] has joined #openttd
05:25:08  *** kirjs_______ [] has quit [Read error: Connection reset by peer]
05:25:20  *** kirjs_______ [] has joined #openttd
05:26:04  *** strohalm [~smoofi@] has quit [Remote host closed the connection]
05:26:37  *** strohalm [~smoofi@] has joined #openttd
05:56:02  *** Eddi|zuHause [] has quit []
05:56:16  *** Eddi|zuHause [] has joined #openttd
06:39:31  *** cha0tics [] has quit [Ping timeout: 480 seconds]
06:54:01  *** liq3 [] has joined #openttd
07:11:38  *** andythenorth [] has joined #openttd
07:11:44  *** andythenorth [] has left #openttd []
07:12:24  *** Pukka is now known as Pikka
07:36:02  *** shirish [] has quit [Quit: No Ping reply in 180 seconds.]
07:37:28  *** shirish [~quassel@] has joined #openttd
07:42:32  *** tipsyTentacle [~tipsy@] has quit [Read error: Connection reset by peer]
07:58:06  *** JezK [~jez@2407:7800:400:107f:3db5:daca:8457:e66a] has quit [Quit: :q!]
08:09:47  *** sim-al2 [] has quit [Ping timeout: 480 seconds]
08:21:02  *** andythenorth [] has joined #openttd
08:24:53  <Pikka> o/
08:27:12  *** shirish [] has quit [Ping timeout: 480 seconds]
08:27:13  <andythenorth> lo pikka bob
08:28:30  <Pikka> lo
08:28:43  <Pikka> brbrb
09:30:36  <Eddi|zuHause> blob
09:40:34  <Pikka> blab
09:44:19  <andythenorth> Pikka how is bob?
09:44:39  <Pikka> hmm not bad. uni all done, at last
09:44:44  <andythenorth> \o/
09:44:48  <Pikka> et tu brute?
09:44:50  <andythenorth> do you have a certificate?
09:44:56  <Pikka> oh
09:45:01  <Pikka> uni all done for the semester at last :)
09:45:11  <andythenorth> it is raining here
09:45:12  <Pikka> halfway through the degree...
09:45:24  <andythenorth> and I receive daily pictures by sms from our guys who are in Perth
09:45:30  <andythenorth> at the beach
09:45:39  <Pikka> it's been threatening to rain here all day. It's going to wait until I'm out letterboxing tonight before it actually starts.
09:45:42  <Eddi|zuHause> hasn't the semester just started?
09:45:57  <Pikka> it's the beginning of the summer break here, Eddi
09:46:02  <andythenorth> are you involved in politics again? o_O
09:46:42  <Pikka> yep, local government this time
09:47:06  <Pikka> wot larks
09:47:46  <Pikka> I'm involved in two campaigns, so twice as much chance of winning \o/
09:48:26  * andythenorth has been a leafleter
09:48:34  <andythenorth> my dad was a local councillor
09:48:43  <Pikka> it's good fun
09:48:43  <andythenorth> much pounding of streets, opening of gates, avoiding of dogs
09:49:05  <andythenorth> 50p per night I think he paid us
09:49:10  <Pikka> oh
09:49:19  <Pikka> I don't get paid.. tch.
09:49:23  <andythenorth> I was about 9
09:49:27  <andythenorth> child labour
09:49:31  <Pikka> terrible
09:50:31  * andythenorth has been on a newgrf spree
09:50:35  <andythenorth> hmmm
09:50:38  <Pikka> making, or playing?
09:50:39  <andythenorth> I should forums post some stuff
09:50:41  <andythenorth> both
09:50:57  <Pikka> fancy
09:51:00  <Pikka> you should
09:51:05  <andythenorth> I am bored of being in a nearly-inactive community :P
09:51:16  <andythenorth> but maybe I should do something instead of complaining :P
09:51:28  <Pikka> we'll just have to make a new game, with a new community ;)
09:52:27  <andythenorth> I think that’s called factorio?
09:52:37  <Pikka> hmm
09:52:50  <andythenorth> haz you other plans?
09:53:23  <Pikka> I have pipe dreams :) but seriously, probably no time soon.
09:53:49  <V453000> o/
09:55:00  <Pikka> yikes o/
09:55:18  *** Biolunar [] has joined #openttd
09:56:29  <V453000> I think players are still very active andythenorth
09:56:40  <V453000> idk about newgrf and development :) dont follow anymore
09:56:55  <andythenorth> V453000: without you, I’m the only one left writing monologues in irc :)
09:56:56  <andythenorth> nvm
09:56:57  <V453000> I wouldnt take forums as a good indicator of activity
09:57:02  <V453000> XD
09:57:05  <V453000> right
09:57:44  <Eddi|zuHause> maybe irc is dying, and you should start a facesnapgram?
09:58:43  <andythenorth> irc died years ago
09:58:49  <andythenorth> now it’s an unkillable walking zombie
10:00:31  <V453000> andythenorth, come make roadtypes mod for factorio :P
10:00:55  <andythenorth> nah
10:01:03  <andythenorth> I tried factorio, it doesn’t click for me :)
10:01:05  <andythenorth> most games don't
10:01:45  <V453000> but it has trains? :D
10:02:38  <andythenorth> I know :D
10:03:19  * Pikka dindins
10:03:27  <andythenorth> enjoy pie
10:03:31  <Pikka> mm pie
10:24:53  <Pikka> shhh
10:25:18  <andythenorth> I didn’t say anything :(
10:25:30  <Pikka> why not?
10:25:54  <andythenorth> well
10:26:03  <V453000> enjoy pipe?
10:26:04  <andythenorth> because I get mocked for being the only one who talks here :P
10:26:09  <andythenorth> and even andythenorth has feelings
10:26:31  <Pikka> how can you get mocked if no-one else talks, hmm? :D
10:26:42  <V453000> strong argument
10:27:03  <andythenorth> logical
10:27:05  <andythenorth> flawless
10:27:13  * andythenorth has made a copper cargo
10:27:15  <andythenorth> what larks
10:27:24  <Pikka> I did that once...
10:27:37  <andythenorth> I wonder why I didn’t just use metal?
10:27:40  <andythenorth> but it seemed wrong
10:28:10  <Pikka> it's all just stuff
10:28:18  * Pikka made a stuff cargo once, too...
10:28:41  * andythenorth has been making south american trains
10:28:43  <andythenorth> all the realisms
10:28:51  <V453000> omfg
10:28:54  <Pikka> what did you decide about gauges?
10:29:12  <andythenorth> ‘RAIL’ = broad gauge
10:29:30  <andythenorth> then narrow gauge for all the others, 75cm, meter gauge etc
10:29:48  <Pikka> hmm
10:29:58  <andythenorth> got 6 railtypes in this game, including PIPE
10:30:01  <andythenorth> seems ok though
10:30:02  * Pikka feels meter gauge isn't "narrow", but that might just be what I'm used to.
10:30:11  <andythenorth> relative :P
10:30:27  <Pikka> sure, but if they're "real trains" they're not narrow gauge, or something.
10:30:52  <andythenorth> when we do Pineapple stuff, NG can be RAIL :P
10:31:21  <andythenorth> FIRS queensland economy
10:31:51  <Pikka> rum, bananas and coal
10:32:14  <andythenorth> sugar
10:32:55  *** Pokka [] has joined #openttd
10:33:06  <Pokka> ouch
10:33:38  <Pokka> anyway, sugar makes rum
10:34:03  <andythenorth> sugar refinery -> food, alcohol
10:35:18  <Pokka> fair enough
10:35:24  <andythenorth> maybe
10:35:28  <V453000> slug makes food
10:35:52  <andythenorth> Road Hog is not bad now
10:36:00  <andythenorth> “At last, a non-shit vehicle set”
10:36:05  <andythenorth> * besides eGRVTS
10:36:15  <V453000> nice
10:36:27  *** Wolf01 [] has joined #openttd
10:36:35  <Wolf01> o/
10:36:42  <Pokka> double nice
10:39:52  *** Pikka [] has quit [Ping timeout: 480 seconds]
10:39:58  *** Pokka is now known as Pikka
10:42:28  <Eddi|zuHause> what's next? monster nice?
10:42:44  <Pikka> monster rice?
10:43:07  <Eddi|zuHause> mustard spice?
10:43:30  <Pikka> mango spline.
10:43:34  <Wolf01> cookies
10:43:41  <Pikka> easily
11:09:50  *** fjb [] has joined #openttd
11:25:06  *** TheMask96 [] has quit [Ping timeout: 480 seconds]
11:28:57  *** TheMask96 [] has joined #openttd
11:33:27  *** Pokka [] has joined #openttd
11:33:31  <Pokka> oops
11:40:34  *** Pikka [] has quit [Ping timeout: 480 seconds]
11:40:39  *** Pokka is now known as Pikka
11:55:34  *** Pikka [] has quit [Read error: Connection reset by peer]
12:09:06  *** smoke_fumus [~smoke_fum@] has joined #openttd
12:35:11  *** Supercheese [] has quit [Read error: Connection reset by peer]
12:35:45  *** Supercheese [] has joined #openttd
12:36:16  <LadyHawk> i just broke a setting in the config file, and i think gave me what i wanted.. *digs*
12:38:31  *** Snail [] has joined #openttd
12:51:34  *** Hiddenfunstuff [] has joined #openttd
12:56:36  *** luaduck [] has joined #openttd
12:58:22  *** Snail [] has quit [Quit: Snail]
13:04:02  *** shirish [~quassel@] has joined #openttd
13:33:39  <LadyHawk> and i've run into a little issue which i dunno how to solve.. a depot block with heavy traffic... traffic seems to pick a depot and then decide to reserve a track to it.. and they will wait at a blocked path for centuries
13:34:19  <LadyHawk> they should first reserve a free track and then decide to pick said depot on the reserved track
13:35:06  <LadyHawk> i'm finding it strange since it's a forced service, rather than a depot order
13:35:22  <LadyHawk> didn't think they'd actually go and pick a preferred depot and get all stubborn
13:36:01  <LadyHawk> (pbs signals since that's much easier to split the track than entry/combo/exits
13:36:02  <LadyHawk> )
13:37:20  <LadyHawk> maybe i can put a waypoint behind the pbs signals and from the waypoint give them a depot order, would that make them pick the depot on their reserved track?
13:37:49  <LadyHawk> so they'd be clueless about any depot untill it's too late
13:38:00  <LadyHawk> for them to get all stubborn about it
13:44:26  <V453000> if I understand this correctly you have 2 depots and want trains to use kind of both
13:48:24  <LadyHawk> 1 track 2 depots on a single tile split by pbs signals is how it started, traffic is clever enough then since 1 depot is always occupied with a train entering or exiting so they filter between the two perfectly
13:49:11  <LadyHawk> now it's advanced to 3 tracks entering a 6 depot block so with a bigger junction, they're starting to show a preferred depot and wait instead of picking a free track (which there always IS)
13:49:29  <LadyHawk> im hoping i just solved the problem but i have to wait for the trains to get there to see
13:50:04  <LadyHawk> choice cannot be part of the equasion
13:51:49  <LadyHawk> yeah looks like the problem is fixed.. at the point they have to reserve a track to a depot, they still have the waypoint order which is right behidn the pbs signals
13:52:15  <LadyHawk> they dont concern themselves with depots until after they reach  the waypoint, but they already are stuck on a reserved track so they cant choose which one they go to
13:52:16  <LadyHawk> <3
13:55:15  <LadyHawk>
14:05:12  *** FLHerne [] has joined #openttd
14:10:02  <V453000> looking at people using 2way PBS makes me want to throw up
14:10:32  <andythenorth> what is that?
14:10:38  * andythenorth just makes simple trains
14:11:03  <LadyHawk> lol im well aware of the popular opinion on pbs
14:12:11  <LadyHawk> 2 way pbs? the whole track is 1 way, just with the ability for trains to filter over to a different track (3 tracks side by side to handle traffic)
14:12:34  <LadyHawk> so they devide better
14:12:42  <V453000> sure, whole track is whole way so you will use 2way PBS there instead of 1-way
14:12:46  <Wolf01> uhm, V453000, does the auto update of factorio work? i'm using the 0.12 and i just found i was behind of about 3 releases
14:12:47  <V453000> makes perfect sense ;)
14:12:49  <LadyHawk> wrong signal?
14:12:58  <V453000> I am not sure Wolf01, maybe not atm
14:13:04  <LadyHawk> im using the wrong signal then, but it's 1 way
14:13:05  <LadyHawk> lol
14:13:25  <V453000> yes LadyHawk that is the bullshit of 2way PBS
14:13:27  <LadyHawk> it works so never bothered to use a different signal
14:13:28  <V453000> it looks like 1way
14:13:29  <V453000> but is 2way
14:13:36  <V453000> yeah great approach :D
14:13:41  <V453000> just perfect :D
14:13:45  <LadyHawk> ahahahaha
14:14:02  <LadyHawk> i'll do a bit of digging when ig et back from work
14:14:04  <LadyHawk> no time now gtg
14:14:06  <LadyHawk> :p
14:14:23  <LadyHawk> see if i can figure out this 'one way pbs coz im using the wrong signal' thing
14:14:24  <V453000> Wolf01: I know guys at the office did some work on it, but whether it works now, I havent tested
14:14:24  <LadyHawk> afk
14:15:13  <LadyHawk> signals are a bit hard to understand from reading the description.. i just went and stuck 1 down and see what the trains do with it.. if they do what i want them to do, good enough for me
14:15:18  <Wolf01> btw, i really like the new smoke, i'm hoping to see a lot of new graphics changes in the future ;)
14:15:45  <V453000> :D in .14 the updater even tells me that there is new version but it cant be obtained automatically atm
14:15:50  <V453000> thanks :) more stuff coming
14:18:33  <V453000> the smoke itself might get some small additions, but for now is is nice enough :) making it universally work for everything in the game is not very easy :) but rewarding, now it can be controlled by wind etc, like before but nicer
15:23:11  *** liq3 [] has quit []
15:37:15  *** Wormnest [] has joined #openttd
15:53:04  *** fjb [] has quit [Ping timeout: 480 seconds]
15:56:53  *** Supercheese [] has quit [Ping timeout: 480 seconds]
15:58:59  *** Alberth [~alberth@2001:981:c6c5:1:be5f:f4ff:feac:e11] has joined #openttd
15:59:02  *** mode/#openttd [+o Alberth] by ChanServ
16:09:46  <argoneus> does ottd have unit tests?
16:09:55  <argoneus> or is that not supported / not needed / redundant
16:21:00  <Wolf01> ottd seem to only have asserts
16:25:17  <Alberth> not sure you could feasibly write unit tests, given that the source code is 300K lines
16:30:00  <andythenorth> nor that you should
16:31:56  <argoneus> well
16:32:00  <argoneus> there's nothing wrong with unit tests
16:32:06  <argoneus> they should be separated from the game code anyway
16:32:17  <argoneus> they can only help fix things, not break them
16:32:49  <argoneus> then again
16:32:56  <argoneus> it's C, so there probably aren't any "units" to test
16:32:58  <argoneus> other than random functions
16:33:02  <Alberth> the biggest problem of unit tests is that they take effort to create adn maintain
16:34:01  <Alberth> programming language has nothing to do with how to setup the program
16:35:22  <Alberth> but you have performance requirements that make adding arbitrary cuts for testing non-feasible
16:35:49  <argoneus> well
16:35:53  <argoneus> tests should be separated from the code
16:36:00  <argoneus> code should have no idea there are any
16:37:14  <Alberth> we got that covered already :p
16:44:24  <argoneus> :<
16:44:27  <argoneus> ok sorry
16:44:42  <Alberth> no problem :)
16:44:44  <argoneus> I'm still trying to figure out why and when should people write unit tests
16:45:00  <argoneus> and I thought large scale projects is exactly where you need them
16:45:17  <andythenorth> you need them when it costs a lot if you get it wrong
16:45:38  <argoneus> so if someone paid you to make openttd from scratch
16:45:42  <argoneus> you'd test things?
16:46:19  <Alberth> I don't think there is universal agreement on it
16:47:11  <andythenorth> I universally agree with myself on it
16:47:11  <Alberth> I seldomly write unit tests, pretty much only for complicated code buried deep down which will be difficult to check in program context
16:47:37  <V453000> yes.
16:47:52  <Alberth> on the other hand, there is the "test everything methodology", which says only to write code if a test fails
16:48:33  *** Progman [] has joined #openttd
16:49:05  <andythenorth> unless you really are writing safety critical systems, that methodology smokes crack
16:49:26  <argoneus> we're working on a "large" project for uni, with 6 people
16:49:34  <argoneus> and I set up all sorts of linting and code coverage testing and stuff
16:49:38  <argoneus> and I'm wondering if it's a waste of time
16:49:40  <andythenorth> no
16:50:10  <andythenorth> there is a current debate around a specific attitude towards TDD
16:50:31  <andythenorth> because some of the original proponents of TDD are now concerned that it leads to test-driven design
16:50:36  <andythenorth> and test-driven architecture
16:50:50  <andythenorth> which can be nonsensical, but fits is philosophically pure
16:50:55  <andythenorth> -fits
16:51:22  *** TheMask96 [] has quit [Ping timeout: 480 seconds]
16:51:27  <argoneus> idk
16:51:30  <argoneus> what we are doing is
16:51:35  <argoneus> write a class, implement it to do something
16:51:36  <andythenorth> testing, for example by linting, is non-controversial as long as (1) the test results aren’t spurious or noisy (2) they run often and fast (3) they are looked at and fixed when failing
16:51:43  <argoneus> then write a test to make sure the class is doing what we want
16:51:52  <andythenorth> that’s not TDD :)
16:51:55  <argoneus> also
16:51:59  <argoneus> isn't linting static analysis?
16:52:06  <argoneus> I thought they were synonymous
16:52:13  <argoneus> since all pylint does is tell me if my code is shit
16:52:19  <andythenorth> potato / potato yes
16:52:29  <andythenorth> it catches common errors that you might have missed
16:52:35  <argoneus> it's also annoying
16:52:37  <argoneus> it says stuff like
16:52:42  <andythenorth> especially ones that weren’t triggered when you ran your code locally
16:52:44  <argoneus> "class only has one public function"
16:52:50  <argoneus> and that's a problem apparently
16:52:50  <andythenorth> well that’s political
16:52:52  <argoneus> because it should be a dict
16:52:54  <argoneus> because reasons
16:52:55  <Alberth> you can configure pylint :)
16:53:00  <andythenorth> all linters have a degree of politics embedded
16:53:02  <andythenorth> imo
16:53:13  <andythenorth> you have to choose what to turn off
16:53:24  <Alberth> by definition, I'd say, or you don't have a base to check against
16:53:40  <argoneus> so basically
16:54:03  <argoneus> the general consensus is write code instead of TDD, and if there's a part that might break, write a test for it
16:54:06  <argoneus> ?
16:54:12  <andythenorth> that’s a common practice
16:54:20  <argoneus> though
16:54:23  <andythenorth> when it breaks in production write a guard
16:54:27  <argoneus> I can see why people enforce linting and tests in big projects
16:54:35  <andythenorth> if you can predict it will break in production, write a guard before you upset the customer :P
16:54:39  <argoneus> we are only 6, but I'm happy when people write a test for code they wrote
16:54:45  <argoneus> because then I just run the test and I know it works
16:54:50  <Alberth> nope
16:54:51  <andythenorth> no you know the test works
16:54:53  <argoneus> or I glance at the test quickly to check it's testing the right stuff
16:54:57  <andythenorth> the test might be useless
16:54:58  <argoneus> and then I know it's fine
16:55:00  <argoneus> yeah
16:55:03  <argoneus> but if the test looks reasonable
16:55:08  <andythenorth> we have an app that had ‘good coverage’
16:55:09  <argoneus> I don't need to worry about the implementation
16:55:17  <argoneus> unless it's slow, which will be profiled by another tool
16:55:22  <Alberth> if the author of code writes the test, both code will contain the same errors
16:55:28  <andythenorth> but one of the tests was something like: foo.x = 1, assert(foo.x == 1)
16:55:32  <argoneus> that's where static analysis comes in
16:55:33  <argoneus> :D
16:55:54  <argoneus> but yeah I don't know
16:56:00  <argoneus> we could probably do just fine with committing and pushing raw code
16:56:01  * andythenorth finds the DHH blog post
16:56:03  <Alberth> errors = wrong assumptions/ideas on how things are supposed to work
16:56:04  <argoneus> with random debug strings in the code
16:56:08  *** TheMask96 [] has joined #openttd
16:56:15  <argoneus> there's also the thing about feeling good
16:56:23  <argoneus> you feel like your code is good when pylint reports nothing and tests work
16:56:25  <andythenorth> here you go
16:56:28  <argoneus> and then the bugs come flowing in
16:56:45  <andythenorth> tests will save your ass
16:56:50  <Alberth> if you have someone else write the tests, you have a check from a secondary source
16:57:24  <Alberth> and imho, a test that doesn't fail is a waste of time
16:57:35  <Alberth> since you didn't catch anything
16:57:41  <andythenorth> also
16:57:49  <andythenorth> argoneus: what does the app do?
16:58:20  <argoneus> a web crawler
16:58:23  <argoneus> that downloads and parses stuff
16:58:30  <andythenorth> ok so validate the output and input
16:58:36  <andythenorth> probably with a jenkins job or such
16:58:44  <andythenorth> which CI server are you using?
16:58:47  <argoneus> jenkins
16:58:50  <andythenorth> ok
16:58:53  <Alberth>
16:59:01  <argoneus> though
16:59:02  <andythenorth> so do you have it run test cases on real input data?
16:59:06  <argoneus> we don't really have many unit tests
16:59:10  <argoneus> it's mostly tests
16:59:13  <argoneus> that actually download stuff
16:59:17  <argoneus> and actually communicate with db
16:59:18  <andythenorth> you don’t need many unit tests for that
16:59:23  <argoneus> because none of us are proficient in mocking
16:59:25  <argoneus> and it seems like a hassle
16:59:27  <andythenorth> you aren’t writing a financial app
16:59:45  <andythenorth> you don’t need to check that the pence are rounding correctly, or the decimal point isn’t transposed
16:59:49  <andythenorth> you have a simple in-out app
17:00:02  <argoneus> well it's not very simple
17:00:08  <argoneus> it also has a database and webfront part
17:00:10  <andythenorth> feed it pages with multiple encodings, malformed tags, multiple html doctypes etc
17:00:13  <argoneus> with a priority queue and proxies and stuff
17:00:25  <argoneus> it's a 12 month project for 6 people keep in mind
17:01:25  <andythenorth> you can use selenium to test the web front end
17:02:01  <andythenorth> have you written your own parser, or used off the shelf lxml or something?
17:02:12  <argoneus> requests+lxml
17:02:26  <andythenorth> parser is a classic thing to throw unit tests at
17:02:36  <andythenorth> but only if you’re not using standard library code
17:02:37  <argoneus> but yeah
17:02:41  <argoneus> from what I've seen
17:02:48  <argoneus> the best test is one that actually tests the whole application
17:02:53  <argoneus> like, gives it some inputs
17:02:56  <argoneus> goes through the whole thing
17:02:58  <argoneus> and compares outputs
17:03:09  <andythenorth> yes
17:03:11  <argoneus> + unit tests for some tricky parts that you need to debug quickly
17:03:13  <andythenorth> it’s slower though
17:03:15  <argoneus> like parsers, calculations
17:03:19  <argoneus> and stuff like that
17:03:22  <argoneus> is that a good methodology?
17:03:34  <andythenorth> it’s pragmatic
17:03:54  <andythenorth> there is a jenkins multijob plugin, if you can parallelise your test cases you get much faster cycle time
17:04:13  <argoneus> i still havent figured out how to properly virtualenv with jenkins
17:04:19  <argoneus> I found something called shiningpanda but it's weird
17:04:52  <argoneus> by the way
17:05:02  <argoneus> do you think the reason why many applications today use bad approaches for things and such
17:05:07  <argoneus> is because most developers are incompetent idiots?
17:05:13  <andythenorth> only partly
17:05:16  <argoneus> undergrads, dropouts
17:05:27  <andythenorth> somewhat yes
17:05:38  <Alberth> pretty much nobody knows how to write software
17:05:48  <argoneus> like
17:05:52  <argoneus> im in the final year of my bachelor's year
17:06:00  <Alberth> which is not surprising, given the big complexity, and the constant change of technology
17:06:02  <argoneus> and there's still people who can't solve problems
17:06:04  <argoneus> like they get told
17:06:09  <argoneus> "write a program that does this"
17:06:16  <andythenorth> this is interesting reading
17:06:18  <argoneus> they look up some ways to solve it
17:06:23  <argoneus> and instead of trying to pick the best one
17:06:25  *** HerzogDeXtEr [] has joined #openttd
17:06:29  <andythenorth> I bailed from Civil Engineering partly because too much calculus
17:06:30  <argoneus> they just pick a random lib and try to force it on the program
17:06:32  <argoneus> argh
17:06:38  <andythenorth> which working engineers told me strongly I’d never need in my job
17:06:54  <andythenorth> of the developers I employ 3 out of 4 hated calculus
17:07:10  <argoneus> anyway
17:07:14  <argoneus> so this theory of things
17:07:21  <argoneus> "all code should be tested" "coverage should be 100%"
17:07:27  <argoneus> are these stupid opinions or not?
17:07:35  <andythenorth> yes
17:07:37  <andythenorth> honestly
17:07:38  <argoneus> like, if you have a large system test, you probably get 100% coverage, no?
17:07:43  <andythenorth> no
17:07:44  <argoneus> and you should have an overall test
17:07:53  <argoneus> for proper software
17:07:54  <argoneus> or idk
17:08:00  <andythenorth> these people are stupid, or work in a very unique environment
17:08:02  <Alberth> not in a million years you can do a 100% test
17:08:28  <Alberth> a single "if" means 2 tests
17:08:29  <andythenorth> you can do 100% test if you write an academic project to demonstrate 100% testing
17:08:37  <Alberth> 2 "if" means 2*2 = 4 tests
17:08:40  <argoneus> though you are right
17:08:43  <argoneus> there's code you don't need to test
17:08:46  <argoneus> because it's e.g. a library
17:08:51  <Alberth> 64 "if" means 2**64 tests
17:08:51  <argoneus> and it's up to the maintainer to test that code
17:08:53  <andythenorth> libraries have bugs
17:08:58  <andythenorth> but those are Not Your Monkeys
17:09:09  <andythenorth> we find library bugs all the time
17:09:27  <andythenorth> if you are using lxml, you will also find library bugs :P
17:09:30  <argoneus> so basically
17:09:34  <argoneus> tests are good
17:09:37  <argoneus> but don't test obvious things
17:09:42  <andythenorth> tests are excellent
17:09:42  <argoneus> tldr don't be an idiot
17:10:02  <argoneus> if I have an in-out app, I have a test that compares in-out
17:10:12  <argoneus> if I have a parser, I have unit tests for the subparts
17:10:17  <argoneus> + a big test for in-out
17:10:22  <argoneus> things like that?
17:10:24  <andythenorth> yes
17:10:39  <Mazur> Actual users _always_ find something you'd never expect in eternity.
17:10:52  <argoneus> I suppose you are right
17:10:58  <argoneus> there are libraries that are 10 years old, do simple things
17:11:03  <Mazur> And so never tested.
17:11:10  <argoneus> and they still get patched every month
17:11:15  <andythenorth> practically, you’re trying to catch stuff like ‘Bob made a load of parser changes, but he only fed it ascii / latin, and it barfs on UTF-8 in production’
17:12:09  <argoneus> this kinda triggers me in my soft. eng. class though
17:12:12  <andythenorth> you want that tested on the feature branch, before it goes anywhere near a release
17:12:18  <argoneus> we learn all these buzzwords that people in the "real world" use
17:12:28  *** frosch123 [] has joined #openttd
17:12:33  <argoneus> and then people on the internet say it's retarded
17:12:38  <andythenorth> oh we learn those all the time too
17:12:54  <andythenorth> mostly from videos of keynote speeches, or more often, vendor PDFs
17:12:58  <andythenorth> or some other crap
17:13:11  <andythenorth> then someone turns up and says ‘now we must all do x'
17:13:16  <andythenorth> [laughter]
17:13:34  <argoneus> I guess it's because software development is a very new thing
17:13:44  <argoneus> it took a while for people to figure out how to make iron into swords as well
17:13:47  <argoneus> I guess
17:13:53  <Alberth> it only completely changes every X years
17:14:26  <Alberth> no change to build experience
17:14:41  <argoneus> well
17:14:47  <argoneus> I'll just keep doing things that make sense I guess
17:14:52  <argoneus> and test things I deem test-worthy
17:14:55  <argoneus> no need to overdo things
17:15:11  <argoneus> time spent trying to force myself to unittest a simple class could be spent fixing actual problems
17:15:33  <argoneus> thanks lads
17:16:16  <Alberth> doing what seems to make sense, and meanwhile try to improve, is the best you can do :)
17:27:23  <frosch123> argoneus: not sure whether anyone mentioned it, but yes ottd does have unittests
17:27:51  <frosch123> there is a gs and ai script based unittest that runs several commands
17:28:03  <frosch123> the tests are run on the compile farm on each commit
17:29:32  <frosch123> they are mostly for the script api, but also test some ottd internals, but nothing complex (well, "unit"-test)
17:36:54  <Terkhen> hello
17:40:53  *** Biolunar [] has quit [Quit: leaving]
17:41:29  <andythenorth> hi Terkhen :)
17:42:22  <Terkhen> :)
17:43:01  <andythenorth> life good?
17:43:22  <Terkhen> crazy, but yes :P
17:45:12  <DorpsGek> Commit by translators :: r27416 trunk/src/lang/welsh.txt (2015-10-29 18:45:08 +0100 )
17:45:13  <DorpsGek> -Update from WebTranslator v3.0:
17:45:14  <DorpsGek> welsh - 2 changes by kazzie
17:52:45  *** glx [] has joined #openttd
17:52:48  *** mode/#openttd [+v glx] by ChanServ
18:03:57  *** sim-al2 [] has joined #openttd
18:07:36  *** Zuu [] has joined #openttd
18:07:43  <Alberth> moin
18:09:00  <frosch123> heffer: thanks for the nml/pillow patch :) i pushed it to our repo
18:15:27  *** FLHerne [] has quit [Ping timeout: 480 seconds]
18:16:23  *** FLHerne [] has joined #openttd
18:24:07  * andythenorth needs to draw a vineyard
18:25:42  <frosch123> do you make a "vineyard -> winery -> gold mine" chain ?
18:26:05  <frosch123> hmm, i haven't played widelands in a while
18:26:13  <andythenorth> now I have to google that :P
18:26:23  <andythenorth> I am supposed to be doing openttd stuff :P
18:26:47  <andythenorth> vineyard probably looks quite like
18:26:56  <frosch123> it is settlers 2.... well, as much as ottd is ttd
18:26:56  <andythenorth> but I already have that in the same economy
18:27:57  <andythenorth> widelands eh
18:28:09  <andythenorth> does it have fog of war? o_O
18:28:31  <frosch123> it can, but i think i disable it usually
18:29:15  <frosch123> it lacks the catapult from s2
18:29:27  <andythenorth> I would really like a game that is similar to warcraft 1
18:29:33  <andythenorth> get resources, then battle
18:29:35  <frosch123> but when there was a catapult you usually ignored the rest of the game
18:29:35  <andythenorth> then next map
18:30:28  <frosch123> i never played anything of the wc series
18:30:40  <frosch123> only starcraft
18:31:06  <frosch123> i think i backseated some c&c games though
18:31:59  <Zuu> Hello
18:32:21  <frosch123> hey zuu :)
18:32:49  <andythenorth> are we releasing a new OpenTTD in April ish?
18:33:47  <frosch123> i had some plans for my own game in 2007, but then joined openttd
18:34:16  <frosch123> i noticed that some elements are quite similar to factorio :p
18:34:33  <andythenorth> :P
18:35:09  <frosch123> even though my starting point was colonization
18:35:48  <frosch123> but by removing things i dislike and extending on things i like, it somehow is similar to factorio
18:36:06  <frosch123> hmm, though only if i remove some parts of factorio which i dislike
18:36:16  * Zuu made a dumbed down version of Sim Tower for last ludum dare:
18:36:28  <frosch123> so... maybe it is not such big of a surprise that stuff that i like is similar
18:37:04  *** FLHerne [] has quit [Read error: No route to host]
18:37:16  <andythenorth> hmm
18:37:18  *** FLHerne [] has joined #openttd
18:37:21  <andythenorth> looking at pictures
18:37:26  <andythenorth> vineyards don’t have trees
18:37:48  <frosch123> they have a lot of hill though
18:38:04  <frosch123> i think "hill" is the main part of a vineyard
18:38:26  <Alberth> at the sunny side :)
18:39:09  <frosch123> exactly, so you can postpone drawing, and instead discuss the direction of light in ottd :)
18:39:26  <andythenorth> but that’s solved :|
18:39:32  <andythenorth> no interesting debate left
18:39:35  <andythenorth> meh
18:39:36  <frosch123> really?
18:39:43  <andythenorth> ‘solved’
18:40:01  <andythenorth> I am right, everyone else is wrong
18:40:04  <andythenorth> done
18:40:41  * andythenorth finds a small tree sprite in base set
18:52:03  <andythenorth> hmm
18:52:10  <andythenorth> soldier trains
18:52:30  * andythenorth ponders a GS that replicates warcraft / settlers kind of games, but with vehicles
18:52:45  <andythenorth> logistics instead of people or units
18:52:54  <andythenorth> needs fog of war :P
18:53:20  <andythenorth> also programmable ‘disasters'
18:53:32  <andythenorth> “you lost a battle, 3 ships destroyed”
18:54:42  <frosch123> if it was settlers, it would be "vast area of infrastructure and industries destroyed"
18:55:12  <andythenorth> trigger factory explosion
18:55:42  <frosch123> settlers is one of those games where i have difficulties to imagine people playing it competitively in multiplayer
18:56:24  * andythenorth has never played it
18:56:31  <andythenorth> I saw populous played a lot
18:56:34  <andythenorth> and black and white
18:57:14  <frosch123> well, in settlers as i know it (settlers 1, 2 and widelands) you spend hours to build a nice economy before even thinking about going to battle
18:58:28  *** cha0tics [] has joined #openttd
18:58:42  <andythenorth> that’s also how I play Silcon Valley :P
18:58:49  * andythenorth is playing SV right now
19:02:24  *** sim-al2 is now known as Guest7075
19:02:24  *** sim-al2 [] has joined #openttd
19:06:58  *** Guest7075 [] has quit [Ping timeout: 480 seconds]
19:08:55  *** GenevaMode [~Adium@] has joined #openttd
19:23:02  *** shirish [] has quit [Remote host closed the connection]
19:23:23  <DorpsGek> Commit by frosch :: r27417 trunk/src/zoom_func.h (2015-10-29 20:23:21 +0100 )
19:23:24  <DorpsGek> -Doc: ScaleGUITrad (adf88)
19:29:20  <andythenorth> what does it do?
19:29:51  <Eddi|zuHause> if there only were a way to find out...
19:30:10  * andythenorth reading the diff
19:32:16  <Eddi|zuHause> <frosch123> settlers is one of those games where i have difficulties to imagine people playing it competitively in multiplayer <-- the way i heard it (never done this myself) was like "rule: 1 hour no attacking", and then it's over pretty quicky
19:33:45  <andythenorth> warcraft 1 a head-head (null modem) 2 player mode
19:33:55  <andythenorth> basically it was ‘build a fuckton of catapults’
19:33:58  <andythenorth> then win
19:34:04  <andythenorth> quickly
19:34:12  <andythenorth> unless your opponent had more catapults
19:34:17  <andythenorth> good game
19:34:25  <andythenorth> hmm
19:36:15  <Eddi|zuHause> the only times i played warcraft (3) on lanparties, it was usually some adventure-ish type maps
19:36:44  <frosch123> andythenorth: must be an american... sugar in bread :/
19:36:45  <Eddi|zuHause> so everybody gets a hero, and then coop-style
19:37:19  <andythenorth> sugar in your peas
19:37:26  <andythenorth> a reference none of you will understand :)
19:37:29  <andythenorth> not even most people in the uk
19:37:38  <Eddi|zuHause> better than pee in your sugar.
19:38:10  <andythenorth> +1
19:46:10  <andythenorth> I _think_ he might run up against the cargo limit ;)
19:46:15  <andythenorth> somewhat
19:46:39  <Eddi|zuHause> that's exactly what i just posted
19:46:51  <andythenorth> also isn’t he just describing ECS?
19:46:58  <andythenorth> I think george has him covered :)
19:47:16  <andythenorth> there’s nothing wrong with the suggestions...
19:47:20  <Eddi|zuHause> ecs has a lot of other oddities
19:55:24  * andythenorth makes Horse
19:55:30  <andythenorth> much bugs to fix
19:56:55  * Zuu rebalance stairs :-)
19:58:59  <andythenorth> :)
19:59:01  <andythenorth> hurgh FIRS
20:00:13  <Zuu> Some users told that they were to bright/dominating:
20:02:07  <andythenorth> they are right :)
20:02:14  <andythenorth> nice shape though
20:02:47  <Zuu> I worked a bit on the shape today. And dimmed them down a bit.
20:08:55  <Eddi|zuHause> Zuu: the stairs (on the left) are definitely too primary-colour, whereas all the other stuff is dull shaded
20:09:38  <Eddi|zuHause> the distinct black framing doesn't help either
20:10:05  <Zuu> Yeah
20:11:57  <Zuu> It was a quick draw to get the game done in 48 hours. But I do agree with the user feedback that the stairs is not really fitting well with the other graphics.
20:14:23  *** Alberth [~alberth@2001:981:c6c5:1:be5f:f4ff:feac:e11] has left #openttd []
20:17:15  <andythenorth> well
20:17:37  <andythenorth> FIRS Basic + Silicon Valley or NoCarGoal
20:17:43  <andythenorth> FIRS Full + Busy Bee
20:17:48  <andythenorth> seems to be a sweet spot
20:24:28  *** sim-al2 [] has quit [Ping timeout: 480 seconds]
20:42:46  *** sim-al2 [] has joined #openttd
20:46:26  <Zuu> New stairs:
20:47:20  <Zuu> I dimmed them down a bit more at first, but then decided that I want them not to dimed down because they are a key element in the game. Still better shape and a bit dimmed down compare to the old ones.
20:48:45  <Zuu> Oh, and the image shows off a limitation with the current code. Rooms only care about having accecss, not the quality of it. And stairs cannot be built above floor 15. So stair to floor 15 and then elevator works fine for the game. :-p
20:49:50  <frosch123> i never played simtower
20:49:56  <frosch123> is it better to build high than broad?
20:50:00  <Zuu> I played it a lot as kid
20:50:25  <Eddi|zuHause> i never played it either
20:51:14  <Zuu> In SimTower, it was quite vague if the sims cared if they had to walk sidewides or not. You could have a quite wide building shareing a central elevator. The problem was more that the capacity of a single elevator was not enough for a too wide building.
20:51:30  <Zuu> But you could of couse have more elevators and build wide.
20:51:53  <Zuu> There was a room you got at 5:th star that could only be built at floor 100.
20:55:32  <andythenorth> hmm that’s an idea
20:55:35  <andythenorth> industry ladder
20:55:48  <andythenorth> start out with just a couple of industry types available
20:55:58  <andythenorth> GS unlocks more as you achieve goals
21:00:09  <Zuu> Well, if the GS -> NewGRF thing is gonig to get done, it needs a use case. The GS I tried to write I end up realizing it was just FIRS supplies implemented as a GS. :-)
21:02:21  <frosch123> hmm, the main difficulty seems to be that there is no undo knob :)
21:03:20  <Zuu> difficulty in what?
21:03:51  <frosch123> if i misclick and build an elevator one step to the left/right, it will look ugly forever :)
21:04:18  <Zuu> You can click on it (while not building) to open a window where you can delete it.
21:04:27  <frosch123> oh
21:04:40  <Zuu> It will not only look ugly, but also not work. Because the code check if the parts are properly connected. :-)
21:04:44  <frosch123> where do i need to "click to continue play"?
21:04:53  <Zuu> Anywhere
21:05:14  <frosch123> doesn't work
21:05:27  <Zuu> Oh, then it is broken :-)
21:05:32  <andythenorth> Zuu: iirc, we were looking at handling closure with the GS->NewGRF patch?
21:05:40  <andythenorth> I can’t remember what my idea was at the time :)
21:06:39  <frosch123> oi, second try, i am bankrupting :p
21:06:52  <frosch123> i built too many offices, and could only afford one stair
21:07:04  <frosch123> but apparantly the offices cost when empty :p
21:08:08  <Zuu> Yeah, it is a bit harsh in that if you get into the red then it is hard to get out. No loans to take.
21:11:03  <Zuu> Oh btw, you can "click to continue" using the enter key as the continue link gets focus but cannot be clicked until I solve the bug. :-)
21:12:31  *** hawking [] has joined #openttd
21:12:48  *** nadavs [~nadsnir@] has joined #openttd
21:13:03  *** hawking is now known as Guest7086
21:13:45  *** Guest7086 [] has quit []
21:13:59  *** hawking_ [] has joined #openttd
21:14:34  *** hawking_ is now known as Guest7087
21:18:11  <Zuu> Ah.. the continue link got no width or height. :-)
21:19:01  <Eddi|zuHause> that probably makes it difficult to click on :p
21:20:31  <Zuu> I use a hybrid between HTML5 canvas for graphics and transparent DOM elemets for user input. Not sure if it is the best solution, but it is an interesting one. :-)
21:24:03  *** gelignite [] has joined #openttd
21:33:33  <andythenorth> bedtime :D
21:33:35  *** andythenorth [] has left #openttd []
21:41:49  *** Zuu [] has left #openttd [Leaving]
21:41:56  *** Zuu [] has joined #openttd
21:55:32  *** frosch123 [] has quit [Quit: be yourself, except: if you have the opportunity to be a unicorn, then be a unicorn]
21:55:47  *** Zuu [] has quit [Quit: Leaving]
22:17:24  *** Progman [] has quit [Remote host closed the connection]
22:27:31  *** gelignite [] has quit [Quit:]
23:04:54  *** Wormnest [] has quit [Quit: Leaving]
23:12:45  *** cha0tics_ [] has joined #openttd
23:20:09  *** cha0tics [] has quit [Ping timeout: 480 seconds]
23:23:42  *** HerzogDeXtEr [] has quit [Quit: Leaving.]
23:26:47  *** sim-al2 [] has quit [Ping timeout: 480 seconds]
23:29:19  *** Supercheese [] has joined #openttd
23:43:27  *** Hiddenfunstuff [] has quit [Quit:  HydraIRC -> <- Chicks dig it]
23:55:52  *** Pikka [] has joined #openttd

Powered by YARRSTE version: svn-trunk