IRC logs for #openttd on OFTC at 2009-08-07
⏴ go to previous day
00:03:37 *** Coco-Banana-Man has quit IRC
00:40:22 *** KenjiE20|LT has joined #openttd
00:52:01 *** Chris_Booth has joined #openttd
01:07:07 *** KenjiE20 has joined #openttd
01:10:29 *** KenjiE20|LT is now known as Guest38
01:10:29 *** KenjiE20 is now known as KenjiE20|LT
01:23:09 *** Eddi|zuHause2 has joined #openttd
01:23:21 *** Eddi|zuHause2 is now known as Eddi|zuHause
01:52:49 *** MapperOG_ has joined #openttd
02:53:19 *** SirSquid1ess has joined #openttd
03:04:30 *** SirSquid1ess is now known as SirSquidness
03:12:31 *** TinoDidriksen has joined #openttd
03:30:14 *** SirSquid1ess has joined #openttd
03:34:05 *** SirSquid1ess is now known as SirSquidness
03:42:54 *** TinoDidriksen has joined #openttd
05:07:55 <Aali> woah, japanset EMUs are completely unreasonable
05:09:38 <Aali> I'm paying 100k in running costs for a 400 passenger EMU, while a faster, more powerful engine/wagon combination that takes 720 passengers costs me 27k
05:15:37 <Aali> right, I'm supposed to use the wagons for EMUs too
06:36:23 *** Cybertinus has joined #openttd
07:24:20 *** |Jeroen| has joined #openttd
07:45:45 *** fonsinchen has joined #openttd
07:56:41 *** reldred has joined #openttd
08:15:01 *** [1]Mark has joined #openttd
08:15:34 *** Cybert1nus has joined #openttd
08:15:56 *** Chris_Booth_ has joined #openttd
08:17:44 *** TrueBrain_ has joined #openttd
08:18:58 *** Xaroth_ has joined #openttd
08:19:29 *** valhalla1w has joined #openttd
08:19:48 *** Chris_Booth_ is now known as Chris_Booth
08:23:22 *** FloSoft has joined #openttd
08:24:27 *** fonsinchen has joined #openttd
08:24:37 *** michi_cc has joined #openttd
08:24:37 *** ChanServ sets mode: +v michi_cc
08:25:35 *** stuffcorpse has joined #openttd
08:28:59 *** Cybert1nus is now known as Cybertinus
08:38:00 *** Progman has joined #openttd
08:45:58 *** Cybertinus has joined #openttd
08:49:31 *** lewymati has joined #openttd
08:59:04 *** Coco-Banana-Man has joined #openttd
09:42:31 *** Pygmalion has joined #openttd
10:01:47 *** tux_mark_5 has joined #openttd
10:08:11 <dihedral> file_put_contents() allows overwriting the currently running php script!!
10:08:22 <dihedral> stupid piece of !%$#*@
10:12:06 *** SmatZ is now known as Guest80
10:26:18 <CIA-1> OpenTTD: truebrain * r17098 /extra/website/bananas/base.sql: [Website] -Add: add LGPLv2 as valid BaNaNaS license
10:26:58 <CIA-1> OpenTTD: truebrain * r17099 /extra/website/bananas/widgets.py: [Website] -Add: 0.7.2 is released, add it to BaNaNaS too
10:29:06 *** Azrael- has joined #openttd
10:31:38 <CIA-1> OpenTTD: truebrain * r17100 /extra/website/bananas/views.py: [Website] -Fix: allow to supply a license.txt which is equal to the license you selected, so you can upload the package you downloaded
10:54:21 *** KenjiE20 has joined #openttd
10:56:35 <planetmaker> uhm... I guess because they were drawn that way. But are there?
10:57:01 *** SmatZ is now known as Guest89
10:57:02 *** Guest80 is now known as SmatZ
10:58:02 <planetmaker> the houses which I coded should be < 32 pixels - iirc
10:58:18 <planetmaker> but, sure, there might be others. Do you have an example?
10:58:20 <Rubidium> planetmaker: everything with "66 -31" is suspect
10:59:28 <Rubidium> e.g. sprite 384 of building.pcx
11:00:55 <Rubidium> or shall we call it OGFX#77?
11:01:37 <planetmaker> issue numbers a unique throughout the whole devzone, so #77 is fine :-)
11:03:51 <planetmaker> hm... but why is everything with 66 -31 suspect? It's 63 pixels across from left to right, isn't it?
11:04:44 <planetmaker> And #384... well, has a transparent line on the left and right. Though, admittedly, there are two non transparent pixels on the left of that building. Dunno why.
11:04:53 <planetmaker> *sprite #384 in buildings.pcx
11:05:36 <Rubidium> planetmaker: because (town) building sprites must be at most 64 pixels wide
11:05:56 <Rubidium> and with 66 pixels wide, the 31 is likely wrong too
11:06:34 <planetmaker> well. But that sprite is from x=674 to x=739, thus with a delta of 26+39 = 65 pixels. And one of those lines is completely transparent.
11:07:10 *** Polygon has joined #openttd
11:07:19 <Rubidium> thus, as described in the NewGRF specs/TTDP wiki it will glitch
11:09:15 <planetmaker> indeed with the two left pixels it may.
11:09:22 <Rubidium> still, if it's 2 wider and still centered, the xoffs needs to be 1 less
11:10:18 <planetmaker> I'll add this conversation to the ticket :-)
11:11:43 <planetmaker> and those two pixels could be shaved off w/o significant loss.
11:13:02 *** roboboy has joined #openttd
11:18:44 *** GregVernon has joined #openttd
11:18:55 <OwenS> TrueBrain_: AlterScript now executes something somewhat useless :P
11:19:35 <TrueBrain_> sounds very useful
11:19:39 *** TrueBrain_ is now known as TrueBrain
11:20:03 <OwenS> I suppose the next step is to actually implement variables :p
11:29:50 *** thingwath has joined #openttd
11:45:37 *** OwenSX48BD has joined #openttd
11:45:46 *** Eddi|zuHause has joined #openttd
11:47:35 *** OwenS is now known as Guest93
11:47:36 *** OwenSX48BD is now known as OwenS
11:47:42 <OwenS> dns1.llvm.org returned (SERVFAIL) <-- ANNOYING
11:51:37 <TrueBrain> not being able to fix your DOS emulator, now THAT is annoying
11:51:41 <OwenS> apt-get install llvm-doc <-- Workarround :P
11:52:41 *** Eddi|zuHause2 has joined #openttd
11:53:34 <OwenS> libAlterScript.so: 19.9MB. I SO hope thats all debugging info
11:53:50 <TrueBrain> strip libAlterScript.so
11:54:01 *** OsteHovel^PDA has joined #openttd
11:54:06 <TrueBrain> btw, using uppercase in linux shared libraries is not FHS (or what is the name)
11:54:47 *** Biolunar has joined #openttd
11:54:57 <OwenS> How have 1.5MB of objects produced 8.7MB of binary?!
11:56:18 <Rubidium> linking with other libs
11:56:54 <OwenS> Aah woops had the LLVM link in the wrong place
11:57:26 <OwenS> 768.2kb interpreter, 8mb llvm...
11:58:15 <OwenS> No... 170kb interpreter (Forgot to strip)
12:01:12 <OwenS> I have big dependencies = very yes
12:05:10 *** OsteHovel has joined #openttd
12:05:38 <OsteHovel> Why do my pda always run out of battery
12:10:54 *** Biolunar has joined #openttd
12:11:31 *** OsteHovel is now known as OsteHovel^PDA
12:26:46 *** Chruker has joined #openttd
12:27:14 <TrueBrain> woei, WT3.1 is starting to do something :)
12:28:05 <TrueBrain> I think I like my API now :)
12:28:07 *** lobster has joined #openttd
12:28:40 <OwenS> OK, I'm now alloca'ing local variables =)
12:31:32 <TrueBrain> stupiest thing about WT3.1, most of the time is in reading and processing a format to the database .. after that it is a walk in the park .. say, 60% goes into that job, 20% into making the frontend API, and 20% into the frontend itself .. stupid stuff :p
12:37:22 <OwenS> I should learn Dot syntax so I can generate pretty graphs of my AST :p
12:46:23 * TrueBrain gets the party heads
13:04:23 *** Svish|eee has joined #openttd
13:21:50 <OwenS> OK, LLVM Optimizer, I'd like to ask... WHAT GAVE YOU THE MORONIC IDEA THAT THE ENTRY POINT OF A FUNCTION IS UNREACHABLE?!
13:24:16 *** colde_ is now known as colde
13:26:50 <OwenS> What appears to be happening is the instruction combining pass vomits then one of the later passes has a fit...
13:34:09 <CIA-1> OpenTTD: smatz * r17101 /trunk/src/misc/hashtable.hpp: -Revert [FS#3065](r16546): large table on stack caused stack overflow on some architectures
14:00:19 <TrueBrain> hmm .. I want a sed to only give me the first match ... but it gives me the latest :(
14:01:45 *** Svish|eee has joined #openttd
14:02:34 *** Eddi|zuHause2 is now known as Eddi|zuHause
14:02:44 <Eddi|zuHause> i'm pretty sure there was a command for that
14:03:11 <TrueBrain> sed -r 's/(.*)distance:"(.*) km"(.*)/\2/'
14:03:14 <TrueBrain> is the best I could come up with
14:13:07 <Rubidium> TrueBrain: any more clues about the input file format?
14:13:20 <TrueBrain> distance:"12 km",distance:"14 km"
14:13:25 <Rubidium> like e.g. no ':' before distance
14:13:31 <TrueBrain> (blabla),distance:"12 km",(blabla),distance:"14 km",(blabla)
14:14:11 <Rubidium> s/^[^:]*distance:"(.*) km.*/\1/
14:14:51 <Rubidium> s/^[^:]*distance:"([^,]*) km.*/\1/ <- maybe even that
14:15:31 *** [1]Mark is now known as Mark
14:16:48 <Eddi|zuHause> i thought it was \( and \)?
14:17:04 <TrueBrain> I simply don't get why it would only return the last entry .. :(
14:17:55 *** z-MaTRiX has joined #openttd
14:18:18 <Rubidium> because of it's lazy behaviour; oh, the '.' still matches, ...
14:18:26 <Eddi|zuHause> what if you pipe it through "tr ',' '\n'" first?
14:18:51 <Rubidium> hmm too far, now I need to backtrack, which results in the last to be matched
14:19:08 <TrueBrain> Eddi|zuHause: lol, in reality, the values are like: 12,6 km
14:19:28 <Eddi|zuHause> makes it more difficult ;)
14:20:28 <Rubidium> echo "(blabla),distance:"12 km",(blabla),distance:"14 km",(blabla)" | sed 's/^[^:]*distance:\([^ ]*\) km.*$/\1/'
14:20:30 <TrueBrain> grep "distance\:" | tr , '\n' | grep distance -A1 | tr '\n' , | cut -b 11- | cut -d\" -f1
14:20:45 <Eddi|zuHause> what about s/[^d]*distance:"([0-9.,]*) km".*/\1/?
14:21:01 <TrueBrain> Rubidium: for some reason, it returns ALL data on my stream
14:21:10 <Rubidium> TrueBrain: it doesn't for me
14:21:18 <TrueBrain> Rubidium: on MY stream
14:21:30 <Rubidium> unless you've not given a representable section of the stream
14:21:37 <Eddi|zuHause> it doesn't change lines that do not contain "distance"
14:22:03 <TrueBrain> Eddi|zuHause: yours leaves the result on the end of the string, that is something :p
14:22:31 <z-MaTRiX> depends on what would you like to do
14:22:38 <TrueBrain> Rubidium: the string contains 19k chars, sorry if it is not represenstable enough :p
14:23:24 <TrueBrain> but okay, tnx, I now got something working :)
14:23:42 <Eddi|zuHause> regexp are not the ultimate language ;)
14:24:32 <TrueBrain> I needed to calculate the distance to and from my roommates parents home :p
14:25:16 <TrueBrain> so now I fetch an url from google maps, and I needed to strip everything which didn't indicate the distance :p
14:25:33 <OwenS> OK WTF is up with llvm.org? =(
14:25:56 <Eddi|zuHause> TrueBrain: maybe yacc would have been a more fitting tool ;)
14:26:14 <OwenS> Eddi|zuHause: Yacc/Bison and Lex/Flex/Re2c :p
14:26:14 <TrueBrain> Eddi|zuHause: yeah ... it is stupid that with tr, grep, and cut, I have the result faster than sed :(
14:26:51 * OwenS generates doxyfile for llvm headers...
14:27:16 <Eddi|zuHause> ply and a decent tutorial, and you have the result in no time...
14:27:52 <Rubidium> OwenS: I can only say what's not up with llvm.org
14:28:07 <OwenS> How do both DNS servers go down at once? O_o
14:28:44 <OwenS> No... How do both DNS servers go SERVFAIL at once I should say. Both going down would imply possible DDOS (Though why is a question :P )
14:28:47 <Rubidium> someone with a fancy script that uses an already fixed exploit in the DNS server?
14:29:00 <OwenS> Probably if it's BIND...
14:29:11 <Rubidium> or some incorrect setting that got mirrored and screwed it up
14:29:26 <Rubidium> why has firefox become so suicidal lately?
14:30:51 <Rubidium> didn't that waste even more memory?
14:31:21 <OwenS> I've never had Opera go suicidal on me
14:33:10 <TrueBrain> hiih, google, was thinking I was a bot :p
14:34:55 <OwenS> LLVM... I'm calling PassManager::run on a module... why aren't you actually running the passes?
14:35:04 <OwenS> I'd like my functions inlining, y'know? :p
14:35:41 <OwenS> I so love Opera's undo close tab feature
14:37:47 <z-MaTRiX> TrueBrain<< įd say fucking grep
14:42:42 <z-MaTRiX> $ printf "(blabla),distance:"12 km",(blabla),distance:"14 km",(blabla)\n" | grep -io 'distance:"12 km"'
14:43:06 <OwenS> Hmm... perhaps the reason my functions were dissapearing was because I was declaring them with private linkage and LLVM was discarding them :p
14:47:08 <Eddi|zuHause> z-MaTRiX: i'm not sure you actually understood the problem
14:47:25 <TrueBrain> Eddi|zuHause: I thought it would be easier to ignore him
14:47:31 <z-MaTRiX> just realized your echo is not proper
14:48:09 <Eddi|zuHause> TrueBrain: yeah, i tend to screw that up... i'm too nice of a person
14:48:21 <z-MaTRiX> printf '(blabla),distance:"12 km",(blabla),distance:"14 km",(blabla)\n' |
14:48:23 <TrueBrain> btw, civ4 has some nasty language file
14:48:45 <Eddi|zuHause> i can send you the files, if you want to have a live example
14:48:51 *** Progman has joined #openttd
14:49:10 <z-MaTRiX> anyway here is your solution
14:49:11 <z-MaTRiX> printf '(blabla),distance:"12 km",(blabla),distance:"14 km",(blabla)\n' | grep -Eio 'distance[ ]*:["]*[0-9.,][0-9.,]*[ ]*(m|km)' | sed -re 's;.*distance:[ ]*["]*([0-9]+[ ]*(km|m)).*;\1;Ig'
14:49:11 <TrueBrain> I expected it to be really easy, but it will require more time
14:49:31 <TrueBrain> so I will first finish OpenTTD, and make that to work completely, then work on civ4 :)
14:50:16 <Eddi|zuHause> the "simple" part might already be helping me
14:51:00 <Eddi|zuHause> i.e. the <Language>string</Language> part and the automatic generation of all missing strings
14:51:27 <Eddi|zuHause> without the <Gender> stuff
14:51:28 <TrueBrain> that would be relative easy :)
14:51:59 <Eddi|zuHause> just make sure the complex stuff is not destroyed
14:54:23 <Eddi|zuHause> well, you could convert the complex stuff as raw string, and convert it back on saving
14:55:37 <TrueBrain> gimme some time, I will make you something pretty :)
14:58:19 *** Polygon has joined #openttd
15:01:21 *** oskari89 has joined #openttd
15:07:10 * Eddi|zuHause imagines TB's smile to more resemble an evil grin :p
15:07:56 <z-MaTRiX> Belugas<< dont forget the printed circuit board <,
15:13:35 <Ammler> Rubidium: opengfx download from my server is around 10% as from BaNaNaS
15:14:10 <Ammler> (so, not really noticeable :-)
15:15:12 <Rubidium> Ammler: I was more thinking about what happens if it gets downloaded 100 times more from your server
15:15:30 <Ammler> I have around 2.5 TB free
15:15:55 <Ammler> the whole openttd binareis need around 100 GB
15:17:01 <Rubidium> then you got a whole lot of spare ;)
15:17:44 <Ammler> I use around 500 GB for Torrents
15:18:21 <Ammler> well, you are free to use it, if you want ;-)
15:18:38 <Ammler> your key is already there.
15:18:39 <TrueBrain> I thought the conversation was about bandwidth?
15:19:04 <Ammler> my harddisk is very limited
15:19:24 <TrueBrain> so you have 5 GB diskspace free and 10 GB bandwidth?
15:19:50 <Ammler> but I need only around 0.5
15:20:01 <TrueBrain> I think I should stop trying to follow this conversation :)
15:20:23 <z-MaTRiX> TrueBrain<< did i solve the problem?
15:20:29 <TrueBrain> [17:17] <Rubidium> then you got a whole lot of spare ;)
15:20:31 <TrueBrain> [17:17] <Ammler> no, only 10 GB
15:20:37 <TrueBrain> I think that was just a but timing ;)
15:20:57 <Ammler> that was refering to "[17:16] <Ammler> the whole openttd binareis need around 100 GB"
15:21:14 <TrueBrain> okay :) I really was wondering why you would only have 10 GB of free bandwidth :p
15:21:45 <Ammler> there is also the traffic for the is2 and cargodist binaries
15:22:55 <TrueBrain> we use just 750 GB of traffic a month :p (openttd.org only)
15:23:28 <Ammler> I need that too, with the Torrents :-P
15:23:47 <TrueBrain> so our legal bandwidth is higher than your illegal :)
15:24:18 <OwenS> TrueBrain: He may be distributing Linux ISOs ;-)
15:24:22 <Ammler> hmm, I share only legal stuff from there :P
15:24:49 <TrueBrain> I do not believe you!! :P
15:24:54 <OwenS> Terrible choice of distro :p
15:25:24 <Ammler> dunno, here it is, but where my server is located, it might be not.
15:25:39 <Ammler> well, the ISP knows about it.
15:25:54 <TrueBrain> in which country is it legal to upload illegal material?
15:26:04 <OwenS> Flaah... How am I supposed to GEP this? O_o
15:26:30 <Ammler> well, we quit that discussion :P
15:26:47 <TrueBrain> maybe I should ask: where is here?
15:27:35 <TrueBrain> clearly no laws there :p Hihi :)
15:28:03 <Ammler> but the server is in germany
15:28:52 <TrueBrain> okay, I need to solve this graph ... I have 16 points with the distance between each point, I want the fastest route to all those points :p
15:28:54 <OwenS> As far as I'm aware uploading copyrighted material without permission is illegal anywhere
15:29:01 <TrueBrain> (classes flash before my eyes)
15:29:15 <TrueBrain> well, for sure in Germany :p Retarted country :)
15:29:17 <OwenS> TrueBrain: Isn't that travelling salesperson? :P
15:29:21 <TrueBrain> OwenS: yes it is :)
15:29:41 <OwenS> I had an exam in May where I had to solve a travelling salesperson by hand... fun :p
15:29:50 <TrueBrain> for small graphs, not a real issue
15:29:55 <TrueBrain> even more if there are missing legs
15:30:38 <Ammler> sharing something you own is legal.
15:30:50 <OwenS> You don't own copyrighted stuff :p
15:31:09 <OwenS> And it's illegal under the Berne Convention, which I'm sure you follow :p
15:31:36 <TrueBrain> how can you own a tv serie?
15:31:39 <orudge> speaking of binaries and so on
15:31:43 <OwenS> You own a license to posess it :p
15:31:48 <Ammler> I don't own the serie, but the copy
15:31:49 <orudge> is the US mirror ever going to be linked from the openttd web site? ;)
15:32:05 <TrueBrain> orudge: you only care about the US mirror?
15:32:10 <TrueBrain> the UK, CZ and DE are not important to you?
15:32:25 <TrueBrain> well, as with everything: when we have the time :)
15:32:32 <OwenS> Ammler: Therefore you don't own the data you're copying. You own a medium on which the data has been placed.
15:32:37 <TrueBrain> most likely at the time we launch translation of our website :)
15:33:35 <Ammler> OwenS: maybe you speak about software.
15:33:50 <OwenS> It's the same for all copyrighted material
15:33:55 <Ammler> but anyway, I am fine, if you follow your law, I follow mine :P
15:34:42 *** LaSeandre has joined #openttd
15:34:48 <OwenS> In any case... I'm positive it's illegal outside of a few backwards countries with one or more of "People's", "Democratic", "Republic" and/or "Free" in their names
15:35:05 *** Brianetta has joined #openttd
15:35:41 <OwenS> (Because pretty much all said countries are actually dictatorships :P)
15:36:10 <TrueBrain> OwenS: copyright on software is vastly different from copyright on movies/music
15:36:24 <LaSeandre> and actually have nothing to do with being "People's", "Democratic", "Free" or "Republic".
15:37:04 <OwenS> TrueBrain: Yes, but I'm still certain nowhere is it legal to upload copyrighted material you don't have a license to do so with :p
15:37:32 <TrueBrain> OwenS: I myself don't know such country too, but if Ammler says it is legal in his country, I can do nothing else but beileve him
15:37:38 <TrueBrain> or check the laws in his country on this claim
15:37:48 <TrueBrain> so I suggest you do the latter, or say to Ammler he is right :)
15:37:48 <Ammler> TrueBrain: I didn't say, it's legal
15:38:10 <Ammler> I said, it isn't illegal
15:38:47 <OwenS> Something can't be neither legal or illegal :p
15:38:53 <Ammler> I have no idea, it is just my guess :P
15:39:09 <TrueBrain> Ammler: then you should shut up too :)
15:39:15 <TrueBrain> OwenS: yes, something can be neither legal nor illegal
15:39:28 <TrueBrain> in other words: there is no law preventing it, and no court allowing it
15:40:10 <Ammler> and I am quite sure, there is difference between software and media
15:40:46 <Ammler> (like mp3 and tv shows)
15:41:17 <OwenS> TrueBrain: It's still technically one or the two. Can we just say it's in a schroedinger state, I.E. simultaniously neither until somebody goes to court over it? :p
15:41:54 <OwenS> Though I'd say under most systems something is technically legal until said otherwise
15:42:12 <Ammler> that is mostly the case in our country.
15:42:17 <TrueBrain> OwenS: it technically can be both; this is not quantum physics
15:53:59 *** valhalla1w is now known as valhallasw
15:55:46 <Eddi|zuHause> <TrueBrain> okay, I need to solve this graph ... I have 16 points with the distance between each point, I want the fastest route to all those points :p <-- you mean the "travelling salesman problem"?
15:56:14 <TrueBrain> like OwenS said, yes
15:57:35 <Eddi|zuHause> yeah... the discussion kinda drowned in the ... other discussion
15:58:48 <Eddi|zuHause> it gets even more fun if you can use steiner trees instead of spanning trees ;)
16:00:40 <Eddi|zuHause> bäh... google maps still does not work in konqueror 3.5
16:01:15 <OwenS> Eddi|zuHause: Perhaps thats because Konq 3.5 is a couple of years old? :P
16:01:22 <TrueBrain> hmm .. bruteforce should give a nice answer I guess ..
16:01:32 <Eddi|zuHause> so? IE6 is way older
16:01:55 <Eddi|zuHause> TrueBrain: a friend of mine worked on heuristics for TSP
16:02:29 <OwenS> Eddi|zuHause: Supporting an obsolete browser is different from supporting an obsolete browser that nobody uses. When a replacement is available.
16:03:47 <Ammler> Eddi|zuHause: time for KDE4, since 4.3, I don't miss something from 3.5
16:04:14 <Eddi|zuHause> i only ever used konqueror 4 to access google maps...
16:04:34 <Ammler> 4.3 release is around a week old
16:04:53 <Eddi|zuHause> yeah, i heard about that
16:04:59 <Eddi|zuHause> let's check the repo...
16:05:22 <Eddi|zuHause> i don't use factory...
16:05:41 <Ammler> there is factory KDE for 11.1
16:06:10 <OwenS> I want to know why LLVM's optimizers feel the inability to check the equality of two constants...
16:06:31 <OwenS> Oh sorry lol they don't! =)
16:06:41 <Ammler> if you don't use factory, then you should stay with 3.5, as the official stable for 11.1 is 4.1 and that is ugly.
16:07:07 <OwenS> I was looking at the code for the stuff after the operation :p
16:08:40 <OwenS> It's time to make some more ops intrinsics
16:13:36 *** Cybertinus has joined #openttd
16:13:36 <Eddi|zuHause> i have 11.0, i think
16:13:43 <Eddi|zuHause> and i added some kde4 repo for it
16:14:17 <Eddi|zuHause> question: where did i get konversation-1.2alpha2 from?
16:14:47 <Eddi|zuHause> it conflicts, and i can't find an upgrade...
16:18:04 <Eddi|zuHause> so... hoplessly screwing up my system now...
16:26:26 <Belugas> mmhh... roasted almonds
16:27:05 <TrueBrain> hmmm ... solving a 17x17 graph kind of kills my CPU :p
16:28:52 *** einKarl has joined #openttd
16:32:13 *** OwenSX48BD has joined #openttd
16:33:03 *** OwenS is now known as Guest115
16:33:03 *** OwenSX48BD is now known as OwenS
16:33:23 <TrueBrain> Python for sure can't resolve the question in reasonable amount of time :p
16:33:44 *** Guest115 was kicked by Belugas (not needed anymore. SHOOOOO RUNAWAY)
16:36:12 <valhallasw> TrueBrain: brute force and python... no, probably not :P
16:36:30 <OwenS> Well... perhaps with Psyco
16:36:50 <TrueBrain> problem is that a C variant won't be easy
16:38:19 *** frosch123 has joined #openttd
16:41:08 <OwenS> Oh hehe... Some CPP files in my project are compiled to LLVM .bc files; some are compiled to native code :p
16:45:38 *** OwenSX48BD has joined #openttd
16:45:50 *** OwenS is now known as Guest116
16:45:50 *** OwenSX48BD is now known as OwenS
16:47:27 <Eddi|zuHause> TrueBrain: i'm sure hundreds of people implemented reasonably fast algorithms for TSP already, why not reuse some of that?
16:47:47 <TrueBrain> mostly because I don't know TSP?
16:48:39 <Eddi|zuHause> shall i redirect you to someone who does? ;)
16:48:57 <TrueBrain> haha :) No need :) Just toying a bit :)
16:51:14 <TrueBrain> depth-first doesn't allow early cuts.. hehe :)
16:57:37 <TrueBrain> 20 seconds to do 1 depth5 iterration
17:02:13 <TrueBrain> 3:30 for 1 depth4 iterration .. who can do the math? :)
17:03:04 *** fonsinchen has joined #openttd
17:04:13 <TrueBrain> Time to complete: 109 hours
17:12:21 * OwenS has moved a bunch of the builtin code inside the VM =)
17:12:50 <Eddi|zuHause> i told you, heuristics are the way to go
17:13:18 <TrueBrain> hmm .. why don't I just solve it with A* :p Haha :) Would be too easy :p
17:13:46 <Eddi|zuHause> wrong solution...
17:14:03 <Eddi|zuHause> or rather: solution to a different problem
17:14:13 <TrueBrain> you are absolutely right :)
17:14:47 <Eddi|zuHause> damn, i do not have enough balancer tracks...
17:21:20 <Eddi|zuHause> each switch needs one balancer track to cancel out the length difference of the straight path and the branching path
17:21:33 <Eddi|zuHause> i have 18 switches and 7 balancer tracks
17:22:04 <Eddi|zuHause> someone do the math? :p
17:22:26 <OwenS> You need 11 more balancer tracks :P
17:24:16 <Eddi|zuHause> next problem: nobody produces these tracks anymore :p
17:31:01 *** Alberth has joined #openttd
17:45:23 <CIA-1> OpenTTD: translators * r17102 /trunk/src/lang/ (polish.txt spanish.txt):
17:45:23 <CIA-1> OpenTTD: -Update from WebTranslator v3.0:
17:45:23 <CIA-1> OpenTTD: polish - 69 changes by Adam123
17:45:23 <CIA-1> OpenTTD: spanish - 1 changes by Terkhen
17:49:21 <OwenS> Eddi|zuHause: They're not a standard size?
17:52:42 <Eddi|zuHause> OwenS: it's not about the size
17:52:59 <Eddi|zuHause> it's about the track geometry, and the connectors
17:53:14 <OwenS> I thought all track of a guage was compatible?
17:53:15 <Eddi|zuHause> every company does these differently
17:58:01 <Eddi|zuHause> they usually sell connector tracks between the different companies, but that makes only sense when you have two large sets of tracks and want to combine them
18:01:52 <Eddi|zuHause> hm... this layout may work...
18:02:27 *** thingwath has joined #openttd
18:05:07 <OwenS> Question: When's a Variant not a variant? Answer: When LLVM thinks it's not -_-
18:06:29 <OwenS> I have a Variant pointer and a Variant. The difference is significantly more than just a *
18:10:46 <TrueBrain> hmm .. I could also solve my graph problem via a GP
18:10:50 <TrueBrain> that might be of interest :p Haha :)
18:10:54 <TrueBrain> oh, I should get back to WT3.1 :)
18:11:28 <OwenS> Hmm it mainly just seems to be a difference of opinion of how I and LLVM-GCC handle unions
18:13:34 <OwenS> OMG YAY OUR TYPES AGREE!
18:13:56 <Alberth> something must be wrong! :p
18:13:59 <TrueBrain> (as in: very happy :p)
18:14:11 <OwenS> I was thinking awkward typo :p
18:19:41 <OwenS> Hmm... Does GCC pass the address to store outsize return values in as the first or last parameter?
18:21:12 <SmatZ> of course it applies only to structs
18:21:33 <OwenS> First, I just checked LLVM's output
18:22:02 <OwenS> Last wouldn't work in the case of mismatched prototypes, first does
18:22:09 <frosch123> when you use vararg lists, it shoudl be first :)
18:22:29 <SmatZ> it can be passed in a special, designated place
18:22:49 <SmatZ> like, special register, first parameter on stack, ...
18:24:40 <OwenS> Mean. GCC has type-pruned my enum when it generates the LLVM output...
18:24:50 *** xmakina has joined #openttd
18:27:49 <SmatZ> If the type has class MEMORY, then the caller provides space for the return
18:27:50 <SmatZ> value and passes the address of this storage in %rdi as if it were the first
18:27:52 <SmatZ> argument to the function
18:28:04 <SmatZ> OwenS: ^^^ at least for x86_64 linux, it's passed as the first argument ;)
18:28:29 *** Lakie` is now known as Lakie
18:28:30 <OwenS> I think thats pretty much common across all C calling conventions
18:45:54 <TrueBrain> OwenS: I don't thin k %rdi exists on all platforms :p :p :p
18:47:30 <OwenS> TrueBrain: I was refering to first arg :P
18:47:43 <OwenS> LLVM's optimizer is doing quite a bit of the work for me =)
18:47:59 <OwenS> It's diving inside structures and going "Yep, those members are equal"
18:49:10 <OwenS> I think I need to tell it that as_floatType is constant though
18:49:49 <OwenS> So it can dive inside that and inline through it :p
18:50:56 <OwenS> Question: How do I define something as constant in C and still have it's symbol emitted? :p
18:50:58 <Belugas> petern, you did not know? ith's the new sport around, to see who has the bigger one ^_^
18:52:05 <TrueBrain> [19:52] <Eddi|zuHause> OwenS: it's not about the size
19:03:13 *** lewymati has joined #openttd
19:09:31 <Belugas> grrrrr... what's the fucking point of doing a certification when the code required to be certified WAS NOT FIXED
19:10:10 <TrueBrain> no, not my testicles!
19:10:29 <Belugas> what the hell are they doing in my way ???
19:14:35 <xmakina> TrueBrain: Is it possible to save Paths? I want to store the pathfinder.path result into a seperate structure. Can I do that?
19:14:59 <TrueBrain> why you ask me? I have absolutely no idea :)
19:15:06 <TrueBrain> well, everything is possible, so ...
19:15:42 <xmakina> i asked you because your name is at the top of the main.nut :S
19:15:58 <TrueBrain> well, for start, glad you say it is about AIs
19:16:08 <TrueBrain> I was more thinking about OpenTTD
19:16:16 <TrueBrain> second, please ask your question on the forums, I am sure someone can help you out there
19:16:50 <TrueBrain> (I haven't done any mainstream development for ages, so I really have no idea, can't even remember how it looks like :p)
19:17:23 <xmakina> lol - well the very top of the main.nut file reads: /* $Id: main.nut 15101 2009-01-16 00:05:26Z truebrain $ */
19:17:40 <frosch123> so, you mean the noai library pathfinder :)
19:17:41 <TrueBrain> ah, means I was the las tperson who commited to that file .. doesn't really mean anything :)
19:17:50 <TrueBrain> here, this person is the one you might want to ask :p
19:18:01 <Yexo> thanks for the fixes to bananas TrueBrain :)
19:18:03 <xmakina> hi yexo - just the man i was looking for
19:18:23 <TrueBrain> Yexo: about the COPY problem: lighttpd picks up the COPY instead of proxying him to the WebDAV of Apache
19:18:34 <TrueBrain> no idea yet how I am going to fix that ... crazy ass lighttpd doing something he shouldn't be doing ...
19:18:42 <TrueBrain> (even more weird it returns 200 on it ...)
19:18:55 <Yexo> is there any workaround? Rubidium mentioned that a local commit worked fine
19:18:56 <xmakina> yexo: I was wondering if I can save Pathfinder paths into another structure. I'd like to store them in an information class along with other useful bits of information regarding the path.
19:19:18 <TrueBrain> Yexo: for you, not really. Rubidium accessed the internal VPS and did the commit via SVN direct (over file://)
19:19:21 <TrueBrain> but this is not general available
19:20:47 <Yexo> xmakina: I think the easiest way is to include the aystar library in your AI, then change the class AyStar.Path to whatever you want
19:21:35 <Eddi|zuHause> ok, these tracks need serious cleaning done...
19:22:16 <Yexo> another option would be to extend the AyStar.Path class after loading the AyStar library, no idea if that would work
19:25:17 *** KenjiE20 is now known as Guest125
19:25:20 *** KenjiE20 has joined #openttd
19:31:15 *** Brianetta has joined #openttd
19:34:50 *** The_Fun_Warden has joined #openttd
19:36:13 <Belugas> Eddi|zuHause, we DESERVE some shots of it, once finished. Or even a free visit!
19:36:20 <The_Fun_Warden> hi everyone i was wondering if someone could help me, i've just upgraded my version of the PPCTTD and was wondering how you access the cheats
19:36:28 <petern> # FAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADE
19:36:31 <petern> # OUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUU
19:36:34 <petern> # AGAIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIN
19:37:19 <glx> The_Fun_Warden: ctrl-alt-C (ie ask the author of this version as we don't support it)
19:37:31 <Eddi|zuHause> Belugas: it isn't really that much of a deal...
19:37:48 <The_Fun_Warden> it's the pocket pc version
19:38:01 <Belugas> you make it sound as if it was ^_^
19:38:02 <glx> yes and we don't support it
19:39:58 <The_Fun_Warden> just thought i'd ask here just incase
19:41:38 <Belugas> well.. maybe one day... who knows
19:44:04 <The_Fun_Warden> whats the different between the PPC and Windows version surely they are virtually the same
19:44:49 <Yexo> yes, but the PPC version is not created by an of the openttd devs, and because we don't create it we don't give support on it
19:45:09 <glx> and they are not the same
19:46:11 <The_Fun_Warden> my mistake then
19:47:08 <Yexo> TrueBrain: I've just created another project @noai.openttd.org, now how did the repository creation work again? Do I need to click "Create" under the Repository tab uder Settings, or will it be automatically created?
19:48:20 <The_Fun_Warden> before i look on the forum, can you new pictures for the company owners, i'm sick of the same old faces
19:48:36 <Yexo> you can create a newgrf for that
19:48:43 <CIA-1> OpenTTD: rubidium * r17103 /trunk/src/lang/unfinished/ (9 files): -Update: pretranslate a bunch of strings that can't be translated
19:48:53 <Yexo> opengfx has at least some new faces, but I'm not sure if they're complete already
19:49:36 <Rubidium> about half of OpenGFX's faces is done
19:50:05 * Belugas imagines a face without a nose
19:52:00 <TrueBrain> Yexo: :17 repos are created
19:52:38 <TrueBrain> so you will have to wait 25 more minutes for it
19:52:47 <TrueBrain> I know, stupid, but RedMine doesn't support anything else :(
19:52:51 <Yexo> I'm creating a bunch of projects now :)
19:53:10 *** The_Fun_Warden has quit IRC
19:56:25 <TrueBrain> Yexo: don't forget to put the SVN address somewhere :) RedMine doesn't show it by default
19:56:50 <Yexo> ok, but I can onl do that after :17 :)
19:57:00 <TrueBrain> you already know the address it will get ;)
20:01:22 *** andythenorth has joined #openttd
20:12:53 <Eddi|zuHause> Belugas: well, the main problem is getting along with the limited set of tracks i have...
20:13:06 <Eddi|zuHause> it's only going to be a temporary setup anyway...
20:24:45 <CIA-1> OpenTTD: smatz * r17104 /trunk/src/ (subsidy.cpp subsidy_base.h): -Codechange: move code related to subsidy awarding to separate procedure
20:30:50 <CIA-1> OpenTTD: smatz * r17105 /trunk/src/station.cpp: -Codechange: no need to reset list of industries nearby that often, clearing is enough
20:34:31 <Yexo> TrueBrain: "<m:human-readable errcode="2">
20:34:31 <Yexo> Could not open the requested SVN filesystem
20:34:31 <Yexo> </m:human-readable>" <- the svn repos are not generated yet
20:36:18 *** Nite_Owl has joined #openttd
20:36:35 <TrueBrain> sticky bit failed .. bah
20:37:24 <Belugas> yurk... sticky bit... watch where you drop your bit, man!
20:37:32 *** Baffage has joined #openttd
20:40:24 <TrueBrain> nginx configuration is a bitch
20:44:59 <TrueBrain> Yexo: can you let me know when you need to do a copy again?
20:45:08 <TrueBrain> (rather sooner than later :p)
20:45:17 <TrueBrain> I have an alternative URL for you to try, which hopefully does allow the forwarding of COPY :p
20:45:35 <Yexo> ok, give me a few minutes
20:49:36 *** fonsinchen has joined #openttd
20:52:54 <TrueBrain> FUCKING RETARTED FIREFOX! It complains to me about an invalid certificate .. and I can't tell him ANYWHERE to go ahead anyway :(
20:53:32 <TrueBrain> I get a popup about sec_error_unknown_issuer
20:54:03 *** Cybertinus has joined #openttd
20:55:02 *** andythenorth has joined #openttd
20:55:15 <frosch123> obviously "3.something" != "3.5" :)
20:55:23 *** andythenorth has left #openttd
20:55:29 <TrueBrain> retarted nginx .. it seems that I have to tell for both http and https where to proxy requests ..
20:55:33 <TrueBrain> I don't want that .. I want to do it at once!
20:55:38 <TrueBrain> at once = in one time
20:55:55 <Rubidium> proxy all https to the http?
20:56:20 <Yexo> TrueBrain: I'm ready to try a copy again
20:56:30 <TrueBrain> Yexo: okay, please try it on the url with port :84
20:56:34 <TrueBrain> (the rest stays the same)
20:58:10 <TrueBrain> COPY returned a 502 ..
20:58:12 <Yexo> svn: Server sent unexpected return value (502 Bad Gateway) in response to COPY r <- this time
20:59:17 <TrueBrain> okay, it seems to be a problem at the WebDAV backend after all
20:59:54 <TrueBrain> but I can't find anything back about it in the access.log or error.log
21:01:44 <TrueBrain> the proxy tells me the connection was closed, and apache tells me nothing was going on
21:03:04 *** Progman has joined #openttd
21:05:06 <TrueBrain> hahahaha :) Oh, this problem is nasty :) the COPY command has 2 URLs in them
21:05:11 <TrueBrain> because we proxy, one of the two is rewritten
21:05:20 <TrueBrain> Apache validates if both are equal, else it disallows it
21:06:21 <Belugas> i don't want to be here
21:06:30 <TrueBrain> you don't like us? :(
21:06:57 <Belugas> here, is TOTALLY related to this fucking chair/desk i'm glued to
21:07:05 * TrueBrain gives Belugas a hug
21:07:13 <Belugas> you know i love you, sweety
21:07:26 <Belugas> can't get awaay from your virtual arms
21:07:39 <Belugas> by the way, have you cleaned your sticky bit?
21:07:54 *** Progman has joined #openttd
21:09:20 <TrueBrain> Yexo: I will think a moment about a solution .. now first a bit of food :) Bbl!
21:12:11 <CIA-1> OpenTTD: smatz * r17106 /trunk/src/ (station.cpp station_base.h): -Codechange: move computation of station's catchment rectagle to separate function
21:16:48 <TrueBrain> Yexo: can you try now?
21:23:05 *** Polygon has joined #openttd
21:32:08 <TrueBrain> bah, tcpdump can't see the internal traffic ...
21:38:53 <TrueBrain> Yexo: mind hitting it again?
21:39:34 <TrueBrain> now it is just annoying me
21:40:19 <TrueBrain> lol, as soon as you printed those links
21:40:25 <TrueBrain> things started to hit the server looking at them
21:41:03 <TrueBrain> Yexo: I assume it is, yes
21:41:48 <TrueBrain> hit it again Yexo :p
21:42:27 <TrueBrain> it sucks that apache doesn't tell me why it drops
21:42:40 <Eddi|zuHause> TrueBrain: shows you how many people click on random links in a chat :p
21:42:51 <TrueBrain> Eddi|zuHause: this was not clicking
21:43:39 <TrueBrain> Yexo: and now? (what will be insane if it would work)
21:45:07 <Eddi|zuHause> TrueBrain: url catchers looking at headlines and somesuch?
21:45:21 <TrueBrain> darn, had tcpdump configured wrong
21:45:53 <TrueBrain> okay, got the headers now ..
21:46:32 <TrueBrain> what the proxy receives all looks very sane
21:47:43 <TrueBrain> I can see the results ;)
21:48:20 <TrueBrain> ARGH FUCKING SHITTING STUPID ....
21:48:48 <TrueBrain> I hate Apache, I hate Apache, I hate Apache, I hate Apache
21:48:51 <OwenS> Time to factorise out my runtime cast code...
21:49:37 *** xmakina has joined #openttd
21:50:36 *** Brianett1 has joined #openttd
21:55:28 <TrueBrain> Yexo: I am afraid to ask .. but hit it again?
21:55:46 <Yexo> just ask as many times as it's needed
21:57:09 <TrueBrain> possible the problem is the different port I guess ...
21:57:23 <SmatZ> _ln: should I see anything there?
21:58:10 <Belugas> #Hit me with yur best shot!
21:58:50 <SmatZ> oh no, what have I done!
21:59:33 *** Belugas has joined #openttd
21:59:34 *** ChanServ sets mode: +o Belugas
21:59:50 <Belugas> FEAR THE MIGHTY ZOMBIE!
22:00:48 <Nite_Owl> would you like to buy some brains ?
22:01:13 <Belugas> no need, just have to lay over TB
22:01:26 * TrueBrain gives a piece of brain to Belugas
22:02:42 <Belugas> and now, LadyHawk and gentlemen...
22:03:13 <Belugas> that i'll be in vacattions for 2 weeks, starting
22:03:17 <Belugas> NOWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWWW
22:03:25 <Nite_Owl> SYNAPSE SYNAPSE SYNAPSE !!!!
22:03:26 <TrueBrain> we will miss you I guss :)
22:03:43 <Belugas> i'llbe lurking around, don't worry
22:03:46 <SmatZ> enjoy as much as you can :)
22:04:11 <SmatZ> have nice holidays, Belugas :)
22:04:44 <Yexo> enjoy our holidays Belugas
22:04:55 <Yexo> s/our/your/ stupid keyboard
22:05:48 <TrueBrain> completely different fix .. Yexo?:)
22:06:03 <Yexo> different error this time
22:06:13 <TrueBrain> I never saw a COPY :p
22:06:34 <Yexo> svn: COPY of '/svn/lib-fibonacciheap/!svn/bc/2/trunk': Could not read status lin
22:07:32 <TrueBrain> I should learn how to write destination :p Yexo? :)
22:07:45 <TrueBrain> I now never see a COPY :(
22:08:15 <TrueBrain> now the old error again I hope?
22:10:41 <TrueBrain> even in debug Apache doesn't tell about the COPY
22:12:43 <Yexo> "could not read status line" again
22:14:42 <TrueBrain> Yexo: can you try over https please?
22:15:29 <Yexo> svn: COPY of '/svn/lib-fibonacciheap/!svn/bc/2/trunk': Could not read status lin
22:16:48 *** Xaroth_ is now known as Xaroth
22:17:40 <TrueBrain> back to the 502 ...
22:20:47 <TrueBrain> I really need to see what Apache receives if I want to continue this ..
22:20:53 <TrueBrain> this is useless .. running blind :s
22:21:00 <Yexo> is there any way I can help with that?
22:21:27 <TrueBrain> I can see what you send
22:21:30 <TrueBrain> but I can't see what the proxy sends
22:23:37 *** Eddi|zuHause2 has joined #openttd
22:23:48 <CIA-1> OpenTTD: smatz * r17107 /trunk/src/ (13 files in 3 dirs): -Codechange: store type of subsidy source and destination in the Subsidy struct instead of determining it every time it's needed
22:26:19 <Eddi|zuHause2> maybe i should use a less dark theme for kde4
22:27:05 <TrueBrain> whoho, I have tcpdump working
22:28:36 <TrueBrain> the only thing 'wrong' I can find, is that it is a http/1.0 request, the proxy sends
22:30:46 <TrueBrain> Yexo: one more time :)
22:31:26 <TrueBrain> yup, only difference left is http/1.0 instead of http/1.1 ...
22:31:35 *** Eddi|zuHause has joined #openttd
22:32:00 <OwenS> Isn't SVN done over WebDav?
22:32:19 <OwenS> WebDav didn't exist in HTTP/1.1 - apache being paranoid?
22:34:10 <Yexo> svn: REPORT of '/svn/lib-fibonacciheap/!svn/bc/0': Unknown transfer-coding in re
22:34:38 <TrueBrain> so modifying nginx to send http/1.1 instead of http/1.0 doesn't work like that :p
22:36:51 <OwenS> Whew! After a lot of effort I have implemented implicit typecasts between tagless variables (Tagless means that they're not in a Variant because the compiler knows their type)
22:37:45 <Yexo> still using port 84 over http
22:38:58 <Yexo> can you see those errors?
22:39:36 <TrueBrain> hmm ... very weird. ..
22:41:43 <TrueBrain> okay, it turned out my tries at the beginning of this day were wrong because nginx configuration sucks
22:41:48 <TrueBrain> it never did anything with the configuration :(
22:42:45 <TrueBrain> Yexo: try it again please :)
22:45:41 <TrueBrain> is there a way to avoid chunked transfer-codings
22:47:13 <TrueBrain> how does it like it now Yexo? :)
22:48:15 <TrueBrain> seems http/1.1 always comes chunked
22:48:56 <TrueBrain> back to 1.0 ... last try for the night?
22:49:24 *** xander1987 has joined #openttd
22:50:13 <TrueBrain> stupid how httpds can fail on something so simple
22:50:30 <OwenS> TrueBrain: I'd help, but I don't have a clue about nginx http proxying. FastCGI yes, http no...
22:51:32 <TrueBrain> I don't know who is to blame .. nginx for doing 1.0, or Apache for failing on it
22:53:49 <Rubidium> OwenS: Happily it was a one-line fix, but discovering that correct line ... that was obnoxious. <- I guess that's TB's feeling too
22:54:09 <TrueBrain> sadly enough, he doesn't say what the line is
22:54:28 <TrueBrain> or he thinks the https / http is the problem too ..
22:55:04 <Rubidium> TrueBrain: I guess the RequestHeader edit Destination ^https http early
22:55:27 <TrueBrain> Yexo: so he says it should work like this now
22:55:55 <Yexo> well, it clearly doesn't work
22:56:05 <TrueBrain> mostly because we don't use https :p
22:56:51 <OwenS> It going across as HTTP/1.1?
22:56:59 <TrueBrain> nginx is http/1.0 proxy
22:57:10 <OwenS> So it can't generate 1.1 queries? =(
22:59:24 <TrueBrain> a few more thingies on that page .. Yexo?
23:00:02 <TrueBrain> it all just looks fine
23:00:10 <TrueBrain> the only difference I can see, is the HTTP/1.1 :(
23:00:25 <TrueBrain> so how can others make it to work?
23:00:34 <TrueBrain> lets see what happens when he tries it directly on the apache
23:03:37 <TrueBrain> this time, I can't see the result :p
23:03:38 <Yexo> 500 internal server error
23:03:50 <Yexo> svn: Server sent unexpected return value (500 Internal Server Error) in response
23:04:34 <TrueBrain> mysql access error :p
23:04:55 <Yexo> (the noai site is down currently)
23:05:04 <TrueBrain> had to redirect it :)
23:07:52 <TrueBrain> OwenS: I don't have time for obvious questions, sorry
23:08:00 <Rubidium> OwenS: because that's like... already in fracking use?
23:08:02 <TrueBrain> hmm .. mysql priveleges .. how does that work
23:08:14 <OwenS> Rubidium: I suppose you have a very valid point :p
23:10:19 <Yexo> svn: COPY of '/svn/lib-fibonacciheap/!svn/bc/2/trunk': Could not read status lin
23:10:31 <TrueBrain> LOL! So Apache WebDAV is FUCKED UP
23:11:20 <OwenS> Checking for the obvious error: DAV svn is enabled? :p
23:11:22 *** thingwath has joined #openttd
23:11:41 <TrueBrain> retarded fucking bitching Apache
23:11:51 <TrueBrain> here I am fighting all night with the proxy, while YOU ARE TO BLAME
23:12:20 <TrueBrain> of course NOTHING about the request
23:13:37 <Yexo> now "svn log" doesn't work anymore
23:15:00 <TrueBrain> Yexo: and if you use host: 85.17.162.188:84
23:16:23 <Yexo> same error (connection closed by server)
23:17:11 <TrueBrain> now I wonder if it is a problem of the port ...
23:20:56 <TrueBrain> what a piece of crap
23:22:09 <OwenS> Put the conf file somewhere? :p
23:22:43 <TrueBrain> SVNParentPath "/var/repos/svn"
23:23:22 <OwenS> ... Which should work...
23:27:08 <OwenS> Yexo: Have you tried "svn relocating" your repo before doing the copy?
23:27:19 <TrueBrain> copy is not related to a WC
23:27:22 <OwenS> (To make sure it's pointing at the right URL)
23:27:29 <Yexo> OwenS: that's not relavant, since it's a server -> server copy
23:29:02 <OwenS> Every article on the error seems to be related to people with sigsegving httpds...
23:29:18 <TrueBrain> k, rolled back everything, noai should now be again normally reachable
23:29:32 <TrueBrain> Yexo: what I will be doing tomorrow, is replacing apache2 with nginx, as it seems to have DAV support too
23:29:38 <TrueBrain> maybe, just maybe, it does know how to handle COPY
23:29:59 <OwenS> SVN requires special DAV features which the SVN developers have only implemented for Apache2
23:30:06 <Yexo> just ping me again when you need me to try anything
23:30:58 <TrueBrain> also, I will setup an apache locally, see if it does the same ..
23:31:02 <TrueBrain> as this is just plain WEIRD!
23:31:16 <TrueBrain> then again .. maybe it will never work
23:31:21 <TrueBrain> Yexo: if you make a copy inside a WC, does tha twork?
23:31:35 <Yexo> same errors (the few times I tried it)
23:31:44 <TrueBrain> did you try it over :84?
23:31:48 <Eddi|zuHause> the task bar in kde4 still seems to lack a lot of settings...
23:31:54 <Yexo> TrueBrain: I don't think so
23:33:26 *** Eddi|zuHause has joined #openttd
23:34:15 <Yexo> TrueBrain: copy within wc:
23:34:17 <Yexo> svn: Commit failed (details follow):
23:34:17 <Yexo> svn: Server sent unexpected return value (502 Bad Gateway) in response to COPY r
23:34:17 <Yexo> equest for '/svn/lib-fibonacciheap/!svn/bc/1/trunk'
23:35:24 <Eddi|zuHause> and there it crashed...
23:41:20 <TrueBrain> how lovely ... lighttpd fucks up the proxy request :(
23:41:31 <TrueBrain> [07/Aug/2009:23:40:45 +0000] " " 400 0 "-" "-"
23:41:34 <TrueBrain> double spaces page?
23:42:00 <Eddi|zuHause> and it crashed again...
23:42:03 <Eddi|zuHause> what is this shit?
23:42:43 <OwenS> TrueBrain: Wait... you have/had apache behind nginx behind lighty?
23:43:02 <OwenS> Good because that would be silly :p
23:43:03 <Yexo> TrueBrain: good luck if you're going to do more testing, but I'm going to sleep
23:44:28 <TrueBrain> okay, the problem is the forward fields
23:44:33 <TrueBrain> ::ffff: seems to be the problem
23:44:36 <Eddi|zuHause> how can people use the kde4 panel when you can't set it up properly and it crashes all the time?
23:45:28 <OwenS> Eddi|zuHause: I've never had the 4.2 panel crash on me
23:46:36 <Eddi|zuHause> well, it did twice within a few minutes here
23:46:49 <Eddi|zuHause> and now all programs from the task bar have gone into oblivion
23:46:55 <Eddi|zuHause> how can i get them back?
23:47:08 <Ammler> Eddi|zuHause: with >=4.2?
23:47:12 <OwenS> You have a Task Manager plasmoid?
23:47:43 <Eddi|zuHause> if that's what it is called
23:47:51 <Eddi|zuHause> Ammler: 4.3, i believe
23:47:54 <OwenS> Then they should be listed =/
23:48:08 <Ammler> the only issue I had yet is with FF
23:48:39 <Ammler> something there reaches to crash the whole KDE and needs restart of X
23:48:54 <Ammler> happens around once per week.
23:49:10 <Eddi|zuHause> Ammler: might be related to a wine application registering to the task bar
23:49:22 <Eddi|zuHause> but the weird thing is, it seems to work, and then crash a while later
23:49:38 <Eddi|zuHause> now, how do i get all the icons back?
23:50:02 <Eddi|zuHause> like amarok, kinternet, kpowersave
23:50:08 <Ammler> well, "safest" might be to restart KDE
23:50:18 <Eddi|zuHause> that would be rediculous
23:50:34 <Eddi|zuHause> that means going back to kde3.5, be promised
23:50:46 <Eddi|zuHause> i'm furious right now...
23:51:56 <Eddi|zuHause> i'm going to rant in freenode/#kde...
23:51:58 <Ammler> well, I asked at #opensuse before I upgraded to 4.3, nobody mentioned problems.
23:52:13 <Ammler> and I am really happy until now.
23:52:33 <OwenS> Does 4.3 use ~/.kde to store settings?
23:52:49 <TrueBrain> good luck joining freenode
23:52:49 <OwenS> And if so.. Is it possible it's loading the KDE 3.5 settings and getting majorly confused?
23:52:52 <TrueBrain> make sure to put on your helmet!
23:52:54 <OwenS> KDE4 on my system uses ~/.kde
23:53:20 <Eddi|zuHause> it's set up so you can have both in parallel
23:53:25 <Ammler> OwenS: then you don't have 3.5 anymore?
23:53:39 <OwenS> Ammler: KUbuntu hasn't shipped 3 for a couple of versions now :p
23:53:58 <OwenS> I think they went entirely 4 with 8.10
23:54:13 <Ammler> but then somethin glike 4.1 or such?
23:54:16 <Eddi|zuHause> silly distribution...
23:54:47 <OwenS> 4.1 worked well mostly - and they shipped lots of KDE3 apps along side in a side by side config to fill in gaps
23:55:44 <OwenS> Yeah; 8.10 shipped 4.1.2
23:56:14 <OwenS> 4.1 acheived stability and most features; 4.2 acheived 3.5 feature parity
23:56:19 <Ammler> I went back to 3.5 quite fast and waited until 4.2
23:57:15 <Eddi|zuHause> "feature parity" is still not reached, apparently
23:57:18 <Ammler> 4.2 doesn't have e.g. konversation
23:57:31 <OwenS> Ammler: Feature parity in KDEBase
23:59:28 *** tokai|mdlx has joined #openttd
continue to next day ⏵
