search of the:   for: 
code logs -> 2007 -> Mon, 14 May 2007< code.20070513.log - code.20070515.log >
--- Log opened Mon May 14 00:00:18 2007
00:24 Mahal [~Mahal@Nightstar-12512.worldnet.co.nz] has joined #Code
00:24 mode/#code [+o Mahal] by ChanServ
00:48 McMartin [~mcmartin@Nightstar-2079.dsl.snfc21.pacbell.net] has quit [Quit: Homeward bound]
00:51 Thaqui [~Thaqui@Nightstar-25913.jetstream.xtra.co.nz] has joined #code
00:51 mode/#code [+o Thaqui] by ChanServ
01:34 Takyoji [~Takyoji@Nightstar-25812.dhcp.roch.mn.charter.com] has joined #code
01:35
< Takyoji>
I hate to keep coming back and asking for RegExs.. but I can't figure out how to make a statement that means "Find anything EXCEPT for 'text1' "
01:37
< Takyoji>
Which is used in <\!-- START (.*) --\> to find anything with '<!-- START ' before it and ' -->' after it excluding the ases that match "<!-- START text1 -->" for example
01:38
< Takyoji>
cases* not ases
01:54 gnolam [lenin@Nightstar-13557.8.5.253.se.wasadata.net] has quit [Quit: ]
02:05
< Takyoji>
So I take it that nobody knows?
02:07
<@Vornicus>
What particular regex language are you using?
02:12
< MyCatVerbs>
<\!-- START ([^t][^e][^x][^t][^1]|.{0,4}|.{6,}) --\>
02:13
< MyCatVerbs>
...god that's fucking ugly.
02:13
<@Vornicus>
Uh, no.
02:13
<@Vornicus>
That's... not the way to do it.
02:13
< MyCatVerbs>
Takyoji: you should look for a language feature that says "lines that don't match"
02:13
<@Vornicus>
"negative lookahead" will likely work.
02:14
< MyCatVerbs>
I'd test for lines that do match <\!-- START (.*) --\>, then text for those that *don't* match <\!-- START text1 --\>. Lazy me. ^^
02:14
<@Vornicus>
MCV's works too, and is how I'd likely do it.
02:14
< Takyoji>
Actually I'm trying to work on a work-around now
02:16
< Takyoji>
So anything that's put in a group is returned as a value, correct?
02:16
< Takyoji>
Meaning, anything inbetween a ( and ) is return in the match, right?
02:17
<@Vornicus>
Unless you're using it for a specific thing, like negative lookahead.
02:17
<@Vornicus>
What regex language are you using?
02:17
< Takyoji>
The Perl-ish version I think..
02:18
< Takyoji>
in PHP
02:19
<@Vornicus>
Okay. PCRE uses (?!pattern) for negative lookahead.
02:20
<@Vornicus>
so your thing looks like <!-- START (?!text1).* -->
02:24
< Takyoji>
ahh k
02:41
< Takyoji>
Sorry for my VERY late reply, but I've got it to work now from your help
03:16 Vornicus is now known as Vornicus-Latens
04:06 Takyoji [~Takyoji@Nightstar-25812.dhcp.roch.mn.charter.com] has quit [Quit: Leaving]
05:42 ReivZzz is now known as Reiver
06:42 You're now known as TheWatcher
06:53
< jerith>
I've had to use negative lookahead. I don't like it very much.
06:55
< Reiver>
?
06:56
< jerith>
(?!pattern)
06:56
< jerith>
It strikes me as a horrible hack.
07:15 GeekSoldier is now known as GeekSoldier|wr0k
07:18
< jerith>
The time I used it, though, I had no control over anything but the regexp.
08:13 You're now known as TheWatcher[afk]
08:51 KBot [~karma.bot@Nightstar-29204.neoplus.adsl.tpnet.pl] has joined #Code
08:52 KarmaBot [~karma.bot@Nightstar-29603.neoplus.adsl.tpnet.pl] has quit [Ping Timeout]
08:52 KBot is now known as KarmaBot
08:52 AnnoDomini [~farkoff@Nightstar-29603.neoplus.adsl.tpnet.pl] has quit [Ping Timeout]
08:56 Chalcy [~Chalcedon@Nightstar-1216.ue.woosh.co.nz] has joined #code
08:56 mode/#code [+o Chalcy] by ChanServ
08:58 AnnoDomini [~farkoff@Nightstar-29204.neoplus.adsl.tpnet.pl] has joined #Code
08:58 mode/#code [+o AnnoDomini] by ChanServ
08:58 Chalcedon [~Chalcedon@Nightstar-1216.ue.woosh.co.nz] has quit [Ping Timeout]
09:34 ChalcyAFK [~Chalcedon@Nightstar-1216.ue.woosh.co.nz] has joined #code
09:35 Chalcy [~Chalcedon@Nightstar-1216.ue.woosh.co.nz] has quit [Ping Timeout]
09:48 You're now known as TheWatcher[wr0k]
10:32 ChalcyAFK is now known as Chalcedon
10:33 Chalcedon is now known as NSGuest-1922
10:34 NSGuest-1922 is now known as Chalcedon
10:35 Chalcedon [~Chalcedon@Nightstar-1216.ue.woosh.co.nz] has quit [Quit: slep]
11:29 ToxicFrog [~ToxicFrog@Admin.Nightstar.Net] has quit [Ping Timeout]
11:35 Mahal is now known as mahalbed
11:38 GeekSoldier_ [~Rob@Nightstar-5397.pools.arcor-ip.net] has joined #code
11:39 GeekSoldier|wr0k [~Rob@Nightstar-4056.pools.arcor-ip.net] has quit [Ping Timeout]
12:43 Raif [~corvusign@Nightstar-22484.hsd1.ca.comcast.net] has quit [Connection reset by peer]
12:43 Raif [~corvusign@Nightstar-22484.hsd1.ca.comcast.net] has joined #Code
12:44 mode/#code [+o Raif] by ChanServ
13:04 Vornicus-Latens is now known as Vornicus
13:54 gnolam [lenin@Nightstar-13557.8.5.253.se.wasadata.net] has joined #Code
13:54 mode/#code [+o gnolam] by ChanServ
14:01 Thaqui [~Thaqui@Nightstar-25913.jetstream.xtra.co.nz] has left #code [Leaving]
14:48 ToxicFrog [~ToxicFrog@Admin.Nightstar.Net] has joined #code
14:48 mode/#code [+o ToxicFrog] by ChanServ
15:28 GeekSoldier_ is now known as GeekSoldier
15:43 You're now known as TheWatcher[afk]
15:52 ToxicFrog is now known as ToxicFrog|wr0k
16:21 * ToxicFrog|wr0k blinks at this driver
16:21
<@ToxicFrog|wr0k>
/* send command to the chip */
16:21
<@ToxicFrog|wr0k>
... code snipped ..
16:21
< GeekSoldier>
Is Miss Daisy not pleased?
16:21
<@ToxicFrog|wr0k>
/* coalesce doorbells */
16:22
< GeekSoldier>
coalesce doorbells?
16:22
< GeekSoldier>
is there great magic in that snip?
16:23
<@ToxicFrog|wr0k>
No, it's just lock_command_queue, memcpy, inc_command_queue.
16:24
< GeekSoldier>
hmm.
16:24
< GeekSoldier>
silly doorbells.
16:24
<@ToxicFrog|wr0k>
More precisely:
16:25
<@ToxicFrog|wr0k>
/* coalesce doorbells */
16:25
<@ToxicFrog|wr0k>
n1_dev->doorbell_count[queue]++;
16:25
< GeekSoldier>
heh.
16:52 Serah [~Z@87.72.36.ns-26407] has quit [Ping Timeout]
17:18 You're now known as TheWatcher
17:27 Reiver is now known as ReivZzz
17:37
<@ToxicFrog|wr0k>
aughskldfjfjgisdhjgh
17:37
<@ToxicFrog|wr0k>
This code keeps using types that aren't defined anywhere!
17:41
<@Vornicus>
Cool.
17:41
<@ToxicFrog|wr0k>
Like n1_request_buffer.
17:42
< TheWatcher>
...
17:42 * TheWatcher recommends the application of copious amounts of fire
17:42
<@ToxicFrog|wr0k>
Not declared anywhere in the driver sources, the kernel sources, or /usr/include.
17:42
<@ToxicFrog|wr0k>
And yet, this code builds and runs.
17:42 You're now known as TheWatcher[afk]
17:42
<@ToxicFrog|wr0k>
Indeed it is running right now.
17:43
<@ToxicFrog|wr0k>
This is rather vexing, because I kind of want the type signature so that I can add stat gathering to the driver.
17:43
<@ToxicFrog|wr0k>
And I'm having to reverse engineer it from do_request().
17:44
< TheWatcher[afk]>
which driver?
17:44
<@ToxicFrog|wr0k>
Cavium CN1xxx SSL accelerator.
18:26
<@Vornicus>
...arg, remind me: I can make an instance var const and still assign to it in a constructor, yes?
18:26
<@ToxicFrog|wr0k>
It has to go in the var(value) part.
18:26
<@Vornicus>
I think I missed it. what?
18:27
<@ToxicFrog|wr0k>
Eg, ClassName(...): foo(bar), baz(moby) { ... }
18:27
<@Vornicus>
ah. I didn't know I could do that.
18:27
<@ToxicFrog|wr0k>
You are allowed to put a sequence of initializers after the signature and before the actual code.
18:29 * ToxicFrog|wr0k ponders Lua, decides that you can't implement const in it without proxytables.
18:41
<@Vornicus>
...I wish I had some sense of how other numeric types are dealt with, I could write this one more easily.
18:42
<@ToxicFrog|wr0k>
?
18:42 * ToxicFrog|wr0k memotoselves: use W`rkn, wr0k is deprecated.
18:43
<@Vornicus>
Oh, I'm trying to write a vector class, because all the ones I can find don't build, and all the ones I've written in the past suck balls.
18:43
<@ToxicFrog|wr0k>
...isn't that just three floats and some operator overloads?
18:44
<@ToxicFrog|wr0k>
(for that matter, aren't Vector and Point basically the same thing?)
18:45
<@Vornicus>
Yes, it's three floats and some operator overloads.
18:45
<@Vornicus>
My problem is that I don't know how to do the += etc overloads without making for insanity
18:45
<@Vornicus>
...
18:46
<@Vornicus>
...this is because I'm assuming stuff like dictionaries go by reference, which they do in referency languages like Python, and they do /not/ in C++.
18:46
<@ToxicFrog|wr0k>
You could just make them immutable and have the operators return new vectors.
18:46
<@Vornicus>
...wait, += lets me return new vectors?
18:46
<@ToxicFrog|wr0k>
And thus make .= a build error.
18:46
<@ToxicFrog|wr0k>
No.
18:46
<@ToxicFrog|wr0k>
That's the point.
18:46
<@ToxicFrog|wr0k>
You define +, which takes two Vector/Points and returns a new one, and not +=.
18:47
<@Vornicus>
Thing is that otherwise they act like numbers a lot - and the numbers allow +=
18:47
<@ToxicFrog|wr0k>
(I'm not sure if += lets you return a new object. It might.)
18:48
<@ToxicFrog|wr0k>
(or, rather, if it does, I don't know if 'foo += bar' becomes equal to 'foo = foo + bar', or if you would need 'foo = (foo += bar)'
18:48
<@Vornicus>
I've figured it out now, though - dictionaries and the like in C++ don't do by reference but by copy, and constify, so I don't have to worry about it.
18:48
<@ToxicFrog|wr0k>
Aah.
18:49
<@Vornicus>
I've been working in decent languages for too long.
18:51
<@ToxicFrog|wr0k>
Heh. I know that feeling.
18:51 * ToxicFrog|wr0k is currently glaring at C for not supporting sparse arrays
19:00 You're now known as TheWatcher
19:18
< MyCatVerbs>
Interesting definition.
19:20
<@Vornicus>
Interesting definition of what?
19:20
< MyCatVerbs>
"Scripting language" as any language which has no implementation written in itself.
19:21
<@Vornicus>
A bad one, too.
19:21
< MyCatVerbs>
Pity there are so many counterexamples, but oh well.
19:23 * ToxicFrog|wr0k would personally define it as 'any language that can load additional parts of itself, from source, at runtime'
19:23
< MyCatVerbs>
Oooh, that is a good definition.
19:24
< MyCatVerbs>
Addenum, "without resorting to an external compiler"? You could always call out to gcc to built a shared object at runtime...
19:25
<@Vornicus>
that doesn't work in and of itself though
19:26
< MyCatVerbs>
Oh yes, fair point. You have to perform horrible hacks to actually load the code in and link against it.
19:26
<@Vornicus>
So, essentially: anything with eval/exec
19:31
<@ToxicFrog|wr0k>
(you know what I would like to have in
19:32
<@ToxicFrog|wr0k>
(you know what I would like to have in Lua? The ability to load a function from a string, and say 'treat this function as though it were at the same lexical scope as myself')
19:32
<@ToxicFrog|wr0k>
(you can do setfenv(), but that doesn't give you access to locals, etc)
19:32
<@ToxicFrog|wr0k>
(I wouldn't want to be the one who has to implement that, though ??)
19:32
<@Vornicus>
(heh)
19:33
< MyCatVerbs>
http://antirez.com/articoli/tclmisunderstood.html <-- fun.
19:35
<@Vornicus>
Okay, concept 5?
19:35
<@Vornicus>
Scares The Fuck Out Of Me.
19:37
<@ToxicFrog|wr0k>
I like #5, although I've already Been There and Done That in bash.
19:37
<@ToxicFrog|wr0k>
A lot.
19:37
<@ToxicFrog|wr0k>
#7 irritates me greatly.
19:37
<@ToxicFrog|wr0k>
Although at least you can work around it.
19:40
<@ToxicFrog|wr0k>
uplevel is made of awesome and is in fact exactly what I just described.
19:43
<@ToxicFrog|wr0k>
Vornicus: #5 really isn't that far a step from what bash already does, and I know you've used bash a fair bit.
19:47 * MyCatVerbs ponders.
19:48 * TheWatcher blinks
19:48 * TheWatcher has just tried A="ec"; B="ho"; $A$B "foo";
19:48
< MyCatVerbs>
set a {$bu}; set b puts; set bu "Hello World!";$b $a
19:49
<@Vornicus>
I know I've done it to compose paths.
19:49
< TheWatcher>
I didn't even know that was possible in bash
19:49
<@ToxicFrog|wr0k>
TheWatcher: you didn't?
19:49
< TheWatcher>
Nope
19:49
<@ToxicFrog|wr0k>
You've never seen code like:
19:49
<@ToxicFrog|wr0k>
CC="gcc -Wall"
19:49
<@ToxicFrog|wr0k>
$CC foo
19:49
< TheWatcher>
True
19:49
< TheWatcher>
O just.. didn't think of it that way
19:50
< TheWatcher>
*I
19:50
< TheWatcher>
I should have realised, but...
19:51
< MyCatVerbs>
Fascinating.
19:51
< MyCatVerbs>
So that's why TCL is so goddamn slow. All strings, no native integers or symbols. Not to mention the whole language is based around eval.
19:52
<@ToxicFrog|wr0k>
It does cacheing, though.
19:52
<@ToxicFrog|wr0k>
Which is presumably what gives it the speed advantage over Ruby, not that that's particularly hard~
19:52
< MyCatVerbs>
Oooh. Cacheing is spiffy.
20:17 Netsplit DeepThought.NY.US.Nightstar.Net <-> Blargh.CA.US.Nightstar.Net quits: @mahalbed, ReivZzz, @AnnoDomini
20:18 Netsplit over, joins: ReivZzz
20:18 Netsplit over, joins: mahalbed
20:18 Netsplit over, joins: AnnoDomini
20:18 mode/#code [+o AnnoDomini] by ChanServ
20:31 You're now known as TheWatcher[afk]
20:48 Chalcedon [~Chalcedon@Nightstar-1216.ue.woosh.co.nz] has joined #code
20:48 mode/#code [+o Chalcedon] by ChanServ
21:53 You're now known as TheWatcher
22:09 AnnoDomini [~farkoff@Nightstar-29204.neoplus.adsl.tpnet.pl] has quit [Ping Timeout]
22:14 AnnoDomini [~farkoff@Nightstar-29204.neoplus.adsl.tpnet.pl] has joined #Code
22:16 AnnoDomini [~farkoff@Nightstar-29204.neoplus.adsl.tpnet.pl] has quit [Quit: Kill the interloper! Rip out its life!]
22:19 AnnoDomini [~farkoff@Nightstar-29204.neoplus.adsl.tpnet.pl] has joined #Code
22:19 mode/#code [+o AnnoDomini] by ChanServ
22:20 mahalbed is now known as Mahal
22:22 You're now known as TheWatcher[T-2]
22:42 You're now known as TheWatcher[zZzZ]
23:23 Mahal is now known as MahalErrand
23:58 Takyoji [~Takyoji@Nightstar-25812.dhcp.roch.mn.charter.com] has joined #code
23:59
< Takyoji>
What's the escape code for a backslash? IE \n creates a new line
--- Log closed Tue May 15 00:00:16 2007
code logs -> 2007 -> Mon, 14 May 2007< code.20070513.log - code.20070515.log >
 search of the:   for: