search of the:   for: 
code logs -> 2007 -> Mon, 08 Oct 2007< code.20071007.log - code.20071009.log >
--- Log opened Mon Oct 08 00:00:05 2007
00:10 Syloq [~Syloq@NetAdmin.Nightstar.Net] has joined #code
00:10 Syloqs-AFH [Coder@NetAdmin.Nightstar.Net] has quit [Ping Timeout]
00:11 Syloq is now known as Syloqs-AFH
00:17 You're now known as TheWatcher[T-2]
00:21 You're now known as TheWatcher[zZzZ]
01:25 gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has quit [Quit: Z?]
04:52 Forj [~Forj@Nightstar-10836.ue.woosh.co.nz] has joined #code
04:52 mode/#code [+o Forj] by ChanServ
06:20 Vornicus [~vorn@ServicesOp.Nightstar.Net] has quit [Connection reset by peer]
06:24 Vornicus [~vorn@Nightstar-28547.dsl.mrdnct.sbcglobal.net] has joined #code
06:25 Vornicus is now known as NSGuest-1441
06:26 NSGuest-1441 is now known as Vornicus
07:27 mode/#code [+o AnnoDomini] by ChanServ
07:29 EvilDarkLord [~jjlehto3@Nightstar-2194.vipunen.hut.fi] has quit [Ping Timeout]
07:30 EvilDarkLord [~jjlehto3@Nightstar-2194.vipunen.hut.fi] has joined #code
07:31 Vornicus is now known as Vornicus-Latens
07:31 EvilDarkLord is now known as NSGuest-1445
08:06 You're now known as TheWatcher
08:16 GeekSoldier|bed is now known as GeekSoldier
09:51 Chalcedon [~Chalcedon@Nightstar-10836.ue.woosh.co.nz] has quit [Quit: Gone]
10:36 You're now known as TheWatcher[afk]
10:42 Forj [~Forj@Nightstar-10836.ue.woosh.co.nz] has quit [Quit: Gone]
11:03 MyCatSchemes [~rb6822@Nightstar-23512.cs.bris.ac.uk] has joined #code
11:06 You're now known as TheWatcher
11:21 MyCatSchemes [~rb6822@Nightstar-23512.cs.bris.ac.uk] has quit [Quit: Swim, swim, hungry!]
11:27 MyCatSchemes [~rb6822@Nightstar-23512.cs.bris.ac.uk] has joined #code
11:55 MyCatSchemes [~rb6822@Nightstar-23512.cs.bris.ac.uk] has quit [Quit: Swim, swim, hungry!]
12:03 Vornotron [~vorn@64.252.71.ns-12695] has joined #code
12:04 Vornicus-Latens [~vorn@Admin.Nightstar.Net] has quit [Ping Timeout]
12:14 Kethal is now known as Raif
12:26 Vornucopia [~vorn@64.252.76.ns-12762] has joined #code
12:28 Vornotron [~vorn@64.252.71.ns-12695] has quit [Ping Timeout]
12:29 NSGuest-1445 is now known as EvilDarkLord
12:32 Vornotron [~vorn@Nightstar-8447.dsl.mrdnct.sbcglobal.net] has joined #code
12:34 Vornucopia [~vorn@64.252.76.ns-12762] has quit [Ping Timeout]
12:38 Vornucopia [~vorn@64.252.67.ns-5172] has joined #code
12:40 Vornotron [~vorn@Nightstar-8447.dsl.mrdnct.sbcglobal.net] has quit [Ping Timeout]
12:41 Vornotron [~vorn@64.252.80.ns-3719] has joined #code
12:42 Vornucopia [~vorn@64.252.67.ns-5172] has quit [Ping Timeout]
12:43 gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has joined #Code
12:43 mode/#code [+o gnolam] by ChanServ
12:48 Vornotron [~vorn@64.252.80.ns-3719] has quit [Ping Timeout]
12:48 Vornotron [~vorn@64.252.86.ns-13496] has joined #code
12:53 Vornucopia [~vorn@64.252.83.ns-3939] has joined #code
12:54 Vornotron [~vorn@64.252.86.ns-13496] has quit [Ping Timeout]
12:56 Vornotron [~vorn@64.252.109.ns-13130] has joined #code
12:57 Vornucopia [~vorn@64.252.83.ns-3939] has quit [Ping Timeout]
12:59 Vornotron [~vorn@64.252.109.ns-13130] has quit [Ping Timeout]
13:07 Thaqui [~Thaqui@Nightstar-26433.jetstream.xtra.co.nz] has quit [Quit: This computer has gone to sleep]
13:40 MyCatSchemes [~rb6822@Nightstar-18127.cs.bris.ac.uk] has joined #code
13:49 Reiver is now known as ReivZzz
14:59 Vornotron [~vorn@64.252.8.ns-3016] has joined #code
16:11 MyCatSchemes [~rb6822@Nightstar-18127.cs.bris.ac.uk] has quit [Quit: Swim, swim, hungry!]
16:41
<@ToxicFrog>
Aaaaaaaagh
16:41
<@ToxicFrog>
Fuck you, Java
16:41
<@ToxicFrog>
FUUUUUUUCK YOUUUUUUUUU
16:42
<@McMartin>
What now?
16:42
<@ToxicFrog>
I have a Map<String, String>
16:42
<@ToxicFrog>
headers.
16:43
<@ToxicFrog>
I want to see if Cache-control: no-cache.
16:43
<@ToxicFrog>
Intuitively, this is headers.get("Cache-control") == "no-cache"
16:43
<@ToxicFrog>
However, == is actually is, so instead I need headers.get("Cache-control").equals("no-cache"), which is ugly, but not that bad...
16:44
<@ToxicFrog>
Except that, unlike ==, if there isn't a Cache-control header, it throws a NullPointerException instead of evaluating to false.
16:45
<@McMartin>
Throw it in a try block.
16:45
<@McMartin>
It's not Approved Style (tm) and I wouldn't do it in an inner loop, but what the Hell, it's a homework assignment.
16:45
<@ToxicFrog>
Yeah, except that breaks the expression, which is: (foo || headers(bar) == baz || headers(moby) == frotz)
16:45
<@ToxicFrog>
If headers(bar) throws an exception, it never gets to headers(moby)
16:46
<@McMartin>
Oh, right, Java is not ML
16:46
<@ToxicFrog>
So then the exception handler needs to figure out which on threw, and reevaluate, and it's just goatfucking.
16:46
<@McMartin>
Object TFUtil.mapGet (Map map, Object defaultValue), imo.
16:46
<@ToxicFrog>
Oh, right.
16:46
<@ToxicFrog>
Or just subclass HashMap, even.
16:46
<@McMartin>
Screw subclassing HashMap.
16:47
<@McMartin>
Just implement Map to begin with.
16:47
<@McMartin>
Back it with a field.
16:47
<@McMartin>
Though I suppose it might be easier to do the subclass, unless HashMap turns out to be final.
16:47
<@ToxicFrog>
...hmm. I can't actually see how to solve the problem in this way, though.
16:47
<@ToxicFrog>
I mean, I can solve the specific case of Map<String,String> by having it return ""
16:48
<@ToxicFrog>
But I can't solve the general case of Map<Object,Object>.
16:48
<@McMartin>
Two possibilities.
16:48
<@McMartin>
(1) I'm here saying "Be like Python and let the user pick the default value"
16:49
<@McMartin>
(2) In general you could also have static Object TFUtil.NOT_FOUND = new Object();, which is guaranteed to not be equals() to any other object.
16:49
<@McMartin>
Er. static final Object.
16:49
<@ToxicFrog>
Former is less likely to cause the type checker to flip out, I think, so let's go with that.
16:50
<@McMartin>
Oh, right.
16:50
<@McMartin>
Without type erasure you can still get ClassCastExceptions for anything besides equals().
16:52
<@ToxicFrog>
Alright, it's working, thank you.
16:52 * ToxicFrog adds some accurate, but snarky, comments to ProxyUtil.getFromMap()
16:53
<@McMartin>
It's what they get for exposing NULL.
16:54 Vornotron is now known as Vornicus
16:56
<@ToxicFrog>
...
16:56
<@ToxicFrog>
Alternately:
16:56
<@ToxicFrog>
"no-cache".equals(headers.get("Cache-control"))
16:56
<@McMartin>
True enough!
16:56
<@ToxicFrog>
I don't think that's a NullPointerException...
16:56
<@McMartin>
Nope. Constant Strings are neven NULL.
16:56
<@McMartin>
never
16:57
<@McMartin>
Having a Free Default is still a handy utility function, though.
16:57
<@ToxicFrog>
Yes, I know that, what I'm unsure about is whether String.equals(null) is one.
16:58
<@McMartin>
Oh. No.
16:58
<@McMartin>
Or it leeging better not be.
16:58
<@McMartin>
equals() is not allowed to do so by its contract.
17:00
<@ToxicFrog>
It's not. Excellent.
17:10 MyCatSchemes [~rb6822@Nightstar-18492.cs.bris.ac.uk] has joined #code
17:11 You're now known as TheWatcher[afk]
17:34 McMartin is now known as McMartin[travels]
18:13 AnnoDomini [AnnoDomini@Nightstar-29341.neoplus.adsl.tpnet.pl] has quit [Ping Timeout]
18:19 AD [AnnoDomini@Nightstar-29699.neoplus.adsl.tpnet.pl] has joined #Code
18:21 AD is now known as AnnoDomini
18:24 You're now known as TheWatcher
18:25 Kyrre [~Z@87.72.35.ns-26506] has quit [Connection reset by peer]
18:30 Forj [~Forj@Nightstar-10836.ue.woosh.co.nz] has joined #code
18:30 mode/#code [+o Forj] by ChanServ
18:49 Kyrre [~Z@87.72.35.ns-26506] has joined #Code
18:51 Forj [~Forj@Nightstar-10836.ue.woosh.co.nz] has quit [Quit: Gone]
18:52 GeekSoldier [~Rob@Nightstar-4481.pools.arcor-ip.net] has quit [Ping Timeout]
18:54 GeekSoldier [~Rob@Nightstar-6005.pools.arcor-ip.net] has joined #code
19:10 GeekSoldier is now known as GeekSoldier|bed
19:17 Vornotron [~vorn@64.252.124.ns-22028] has joined #code
19:17 Vornicus [~vorn@ServicesOp.Nightstar.Net] has quit [Ping Timeout]
19:21 Vornotron [~vorn@64.252.124.ns-22028] has quit [Ping Timeout]
20:08 MyCatSchemes [~rb6822@Nightstar-18492.cs.bris.ac.uk] has quit [Quit: Swim, swim, hungry!]
20:14 * ToxicFrog eyes
20:14
<@ToxicFrog>
NoSuchMethodError shouldn't be a runtime error unless I'm doing reflective or casting stuff, surely
20:17 Vornotron [~vorn@64.148.10.ns-11316] has joined #code
20:19 * Kyrre reflects ToxicFrog's spell.
20:23 Vornotron is now known as Vornicus
20:29
<@ToxicFrog>
...oh, screw you, HTTP
20:32
< Vornicus>
??
20:35
<@ToxicFrog>
Transfer-encoding: chunked
20:41
< Vornicus>
o...kay?
20:41
<@ToxicFrog>
TE: chunked divides the data into "chunks", each one of the form SIZE CRLF DATA CRFL
20:41
<@ToxicFrog>
Rather than just having a convenient Content-Length in the header.
20:42
<@ToxicFrog>
(and of course, if you specify Connection: close, it's allowed to do neither and instead signal EOF by close()ing the socket)
20:43
<@ToxicFrog>
And, you know, you'd think that Socket or InputStream or Reader or something would have a "read all data until EOF" method, but fucked if I can find it.
20:43
< Vornicus>
This sounds like something you want a library to handle.
20:44
<@ToxicFrog>
Yeah, except at least in theory I should already have a library to handle it, to wit, the entirety of java.io.* and java.net.*
20:44
<@ToxicFrog>
I can't use the convenient HTTP parsing parts of java.net (screw you, Xiang), but I can use the rest.
20:45
< Vornicus>
so, in short, this professor is forcing onto you all the weaknesses of Java without any of the strengths.
20:45
<@ToxicFrog>
Yeah, basically.
20:46
<@ToxicFrog>
I've sent mail to the TA (who is the one actually marking it) asking if we can at least use java.net.URL (RFC-mostly-compliant URL parsing) and java.util.SortedMap.
20:46
<@ToxicFrog>
If the answer is no, I can drop the course from the comfort of my home~
20:47
<@ToxicFrog>
That said, there should be something, somewhere, that lets me take an InputStream or similar and turn it into a String or a byte[] or the like by reading until EOF.
20:48
<@ToxicFrog>
I have a hard time believing that Java, which has eighteen different kinds of automatically resizing array, does not have this.
20:51 * Vornicus eyes. that is odd.
20:52
<@ToxicFrog>
...there's java.io.Reader::read(java.nio.CharBuffer target), but CharBuffer is abstract and nothing implements it!
20:52
< Vornicus>
eew
20:53
< Vornicus>
Unfortunately even the "strength" of Java is pretty damn weak.
20:53 Chalcedon [~Chalcedon@Nightstar-10836.ue.woosh.co.nz] has joined #code
20:53 mode/#code [+o Chalcedon] by ChanServ
20:57 * Kyrre hates Java.
20:57 * ToxicFrog ponders.
20:57
<@ToxicFrog>
So.
20:57 * Kyrre also wants to stab the coders of firefox in their thighs for coding their "update available" screen so damn badly.
20:57
< Vornicus>
So?
20:57
<@ToxicFrog>
I need to deal with bytes here, not chars.
20:58
< Vornicus>
Right
20:59
<@ToxicFrog>
This basically limits me to, hrm...InputStream::read(), read(byte[]) and read(byte[],int,int)
20:59
< Vornicus>
The middle one is probably what you want.
21:00
<@ToxicFrog>
It won't resize b[], though
21:00
<@ToxicFrog>
None of them will.
21:00
<@ToxicFrog>
So I need to wrap it in something that allocates storage as needed.
21:00
<@ToxicFrog>
ArrayList<byte>, possibly.
21:01
< Vornicus>
uh.
21:01 * Vornicus examines.
21:01
< Vornicus>
Have you looked at java.io.ByteArrayInputStream.buf ?
21:02
< Vornicus>
oh, it's protected, nemmind
21:02
<@ToxicFrog>
ByteArrayInputStream is for reading from ByteArrays.
21:02
<@ToxicFrog>
It lets you turn a byte[] into an InputStream.
21:02
< Vornicus>
ah
21:06
<@ToxicFrog>
Why do Collections use add(), but Maps use put()?
21:06
<@ToxicFrog>
..
21:06
<@ToxicFrog>
...
21:06
<@ToxicFrog>
The fuck
21:06
< Vornicus>
??
21:06
<@ToxicFrog>
You can't use primitive types in generics?!
21:06
<@ToxicFrog>
Apparently, ArrayList<byte> is illegal
21:06
< Vornicus>
Agh
21:07
< Vornicus>
Try ArrayList<Byte>; I think it should box and unbox primitives automatically.
21:09
<@ToxicFrog>
Only sometimes!
21:09
<@ToxicFrog>
But I'll give that a go.
21:09 * Vornicus wonders what the Java library would look like if they actually remade it to work properly.
21:09
<@ToxicFrog>
No, it won't unbox from Byte[] to byte[]
21:10
<@ToxicFrog>
Well, for one thing, the language wouldn't have primitives.
21:10
<@ToxicFrog>
A lot of library ugliness arises from Java's schizophrenia over what is, and isn't, an object.
21:11
<@ToxicFrog>
Such as, say, the nine different versions of java.util.Arrays::copyOf(), one for Object[] and one for each primitive type.
21:11
< Vornicus>
and by doing that you force operator overloads into the open... which they specifically said they didn't want to do, but I still think the argument is specious; misnamed methods are everywhere already.
21:12
<@ToxicFrog>
Not to mention that Java already has operator overloads; they just aren't exposed to anyone outside Sun.
21:12
<@ToxicFrog>
Really, they have the worst of all possible worlds; misuse of operator overloads built into the language, the inability for developers to use them, and it's still trivial to write totally unreadable code.
21:12
< Vornicus>
Then you'll probably end up with One True List, One True Set, and One True Map.
21:15
< Vornicus>
io and util probably see the most improvement from objectification.
21:16
<@ToxicFrog>
Yeah.
21:24 * ToxicFrog eyes these headers coming back from Google
21:24
<@ToxicFrog>
[read-headers] K=nnCoection V=close
21:25 Vornicus [~vorn@64.148.10.ns-11316] has quit [Ping Timeout]
21:29 Vornicus [~vorn@64.148.10.ns-11316] has joined #code
21:29
<@ToxicFrog>
<ToxicFrog> [read-headers] K=nnCoection V=close <-- from Google
21:30 Vornicus is now known as NSGuest-1459
21:31 NSGuest-1459 is now known as Vornicus
21:31 * ToxicFrog asks his proxy for the U of G website and the world goes mad
21:34
< Vornicus>
Saw that.
21:37 Vornicus [~vorn@64.148.10.ns-11316] has quit [Quit: Leaving]
21:37 Vornicus [~vorn@64.148.10.ns-11316] has joined #code
21:37 mode/#code [+o Vornicus] by ChanServ
21:40
<@Vornicus>
Apparently not even limiting the number of lines per window prevents a memory leak in this stupid thing.
21:49 ReivZzz is now known as ReivClass
21:54
<@jerith>
A lot of load balancers rewrite Connection: headers.
21:55
<@jerith>
[jerith@jerith-lap1 ~]$ curl -I http://ec2.amazonaws.com/
21:55
<@jerith>
nnCoection: close
21:56
<@jerith>
(The rest ommitted for space.)
22:01 You're now known as TheWatcher[T-2]
22:05 You're now known as TheWatcher[zZzZ]
22:05
<@ToxicFrog>
Why do they do that
22:05
<@ToxicFrog>
?
22:06
<@jerith>
They want to keep it, but they don't want it to be interpreted or something.
22:13 * ToxicFrog decides to work on his OO homework for a while instead...we get to write a roguelike. In Java.
22:14
<@jerith>
The pain.
22:14
<@Vornicus>
DO they at least let you use all the libraries you want?
22:14
<@ToxicFrog>
Yes.
22:15
<@ToxicFrog>
With the caveat that the final result must match the interface specified by the RogueTest package.
22:15
<@Vornicus>
That's not bad.
22:51
<@ToxicFrog>
The documentation on the existing framework we have to work in appears to be...sketchy.
22:53
<@gnolam>
For some reason, it always is.
23:05
<@ToxicFrog>
It would be nice, for example, to know what you can do with a DungeonView.Direction
23:06
<@ToxicFrog>
(trial and error says that "express diagonal directions" is not on the board, though)
23:06
<@ToxicFrog>
It would also be nice if the sample code weren't broken.
23:07
<@ToxicFrog>
By which I mean "even if you implement all the missing classes and methods correctly, the sample test code makes incorrect assumptions about the world model that lead it to generate a geometrically impossible map"
23:07
<@jerith>
Broken sample code is a standard feature of this kind of thing.
23:08
<@ToxicFrog>
Which would be fine if the prof hadn't said "after implementing Room and Hallway, you should be able to run RogueTest and have it work"
23:25 * gnolam remembers the Java lab framework that assumed that arrays were 1-based.
23:25 * jerith remembers the Java framework that... oh, yeah, that was this morning.
23:26
<@ToxicFrog>
...
23:26
<@ToxicFrog>
Cannot find method: getContentsOfAdjacentRoom
23:26
<@ToxicFrog>
Why? Because it's actually getContentsofAdjacentRoom
23:28
<@Vornicus>
Sigh
23:30
<@ToxicFrog>
Aaargh
23:30
<@ToxicFrog>
So, we need to have player movement respect room and hallway boundaries
23:32
<@ToxicFrog>
At the moment, we do this by (KILL ME) asking the render layer what character is on screen in the cell we're trying to move to.
23:32
<@ToxicFrog>
If it's | or -, it's a wall, so we don't move.
23:32
<@ToxicFrog>
However!
23:33 Thaqui [~Thaqui@Nightstar-26433.jetstream.xtra.co.nz] has joined #code
23:33 mode/#code [+o Thaqui] by ChanServ
23:33
<@ToxicFrog>
The way we do this is with dungeon_view.getContentsofAdjacentSquare(player, dungeon_view.getElementRoom(player), direction)
23:33
<@ToxicFrog>
Note: getElementRoom
23:33
<@ToxicFrog>
If you're in a hall, this method throws a ClassCastException
23:33
<@Vornicus>
Cool.
23:34
<@ToxicFrog>
So what do you do? Well, the way I've found is to keep a Hallway current_hall around, and watch to see when the player goes through a door, and when they do, current_hall = (very ugly set of calls that retrieves the Hallway they just moved into) and it uses that instead.
23:34
<@Vornicus>
;_;
23:58
<@ToxicFrog>
...
23:58
<@ToxicFrog>
...
23:58
<@ToxicFrog>
gCOAS never uses the room parameter.
23:58
<@ToxicFrog>
You can just go dv.gCOAS(player, null, direction)
23:59
<@Vornicus>
...one would think that compilation would catch unused parameters.
--- Log closed Tue Oct 09 00:00:12 2007
code logs -> 2007 -> Mon, 08 Oct 2007< code.20071007.log - code.20071009.log >
 search of the:   for: