00:08 (quit) bjz: Quit: Leaving...
00:14 (join) vkz
00:17 (quit) vkz: Client Quit
00:42 (join) bjz
00:42 (join) mithos28
01:06 (join) cdidd
01:11 (quit) mceier: Quit: leaving
01:12 (quit) pcavs: Quit: Leaving.
01:23 (join) gciolli
01:25 (join) jeapostrophe
01:25 (quit) jeapostrophe: Changing host
01:25 (join) jeapostrophe
01:27 (quit) getpwnam: Ping timeout: 256 seconds
01:27 (quit) hash_table: Ping timeout: 256 seconds
01:34 (quit) ynniv: Quit: ynniv
01:36 (join) pcavs
01:38 (quit) pcavs: Client Quit
01:40 (join) noelw
01:42 (quit) noelw: Client Quit
01:45 (quit) jeapostrophe: Ping timeout: 256 seconds
01:45 (join) noelw
01:49 (quit) noelw: Client Quit
01:50 (join) __rahul__
01:51 __rahul__: what's the best way to check if a mapping exists for a key in a hash table?
01:54 jonrafkind: hash-has-key?
01:54 jonrafkind: http://docs.racket-lang.org/reference/hashtables.html?q=hash#(def._((lib._racket/private/more-scheme..rkt)._hash-has-key~3f))
01:55 (join) RacketCommitBot
01:55 RacketCommitBot: [racket] plt pushed 7 new commits to master: http://git.io/Y8Zy_A
01:55 RacketCommitBot: [racket/master] [honu] completely expand function arguments - Jon Rafkind
01:55 RacketCommitBot: [racket/master] [honu] add phase{ ... }, which is like begin-for-syntax - Jon Rafkind
01:55 RacketCommitBot: [racket/master] [honu] add macro user api - Jon Rafkind
01:55 (part) RacketCommitBot
01:55 __rahul__: thanks, i just found it too
02:18 (quit) __rahul__: Quit: Page closed
02:19 (quit) jonrafkind: Ping timeout: 240 seconds
02:24 (join) mceier
02:32 (quit) gciolli: Quit: Leaving.
02:41 (join) neilv
02:51 (quit) wtetzner: Ping timeout: 245 seconds
02:53 (join) hkBst
02:53 (quit) hkBst: Changing host
02:53 (join) hkBst
02:56 (join) djcoin
02:58 (join) wtetzner
03:09 (join) jesyspa
03:16 (join) soegaard
03:20 (quit) asumu: Ping timeout: 260 seconds
03:32 (join) chturne
03:34 (join) bitonic
03:34 (join) vkz
03:44 (join) asumu
03:56 (quit) mithos28: Quit: mithos28
03:56 (quit) chturne: Ping timeout: 255 seconds
04:03 (quit) dsantiago: Read error: Connection reset by peer
04:03 (join) dsantiago
04:04 (quit) bitonic: Ping timeout: 260 seconds
04:19 (quit) soegaard: Quit: soegaard
04:21 (quit) mrcarrot: Remote host closed the connection
04:23 (quit) hkBst: Remote host closed the connection
04:23 (join) hkBst
04:23 (quit) hkBst: Changing host
04:23 (join) hkBst
04:25 (quit) micaeked: Quit: WeeChat 0.3.9
04:32 (join) chturne
04:47 (join) noelw
04:51 (join) bitonic
05:00 (quit) bitonic: Ping timeout: 246 seconds
05:04 (join) bitonic
05:14 Cryovat: I assume most of you read Hacker News, but found this interesting: https://google-styleguide.googlecode.com/svn/trunk/lispguide.xml
05:15 Cryovat: The interesting implication is that Google must be using Common Lisp
05:20 neilv: perhaps because they bought itasoftware
05:20 neilv: and i suspect only a minority of people here read that site
05:21 Cryovat: I\m not familiar with ITA >*
05:21 Cryovat: :(
05:29 (quit) DT``: Read error: Connection reset by peer
05:35 (quit) chturne: Ping timeout: 240 seconds
05:48 (quit) mceier: Quit: leaving
05:54 Gertm: ITA is even explicitly mentioned in the guide, so it's almost certain this is where it comes from.
05:58 (quit) bitonic: Ping timeout: 252 seconds
05:58 (join) masm
06:00 Cryovat: I always feel a bit uncomfortable looking at Common Lisp docs
06:00 Cryovat: And seeing every function name in uppercase
06:12 (quit) vkz: Quit: vkz
06:25 (join) vkz
06:55 (quit) Nisstyre: Ping timeout: 276 seconds
06:58 (join) bitonic
07:08 (join) Nisstyre
07:28 (quit) Nisstyre: Read error: Operation timed out
07:44 (join) Nisstyre
07:58 (join) dnolen
08:01 (quit) bitonic: Ping timeout: 245 seconds
08:02 (quit) dnolen: Client Quit
08:03 (join) dnolen
08:03 (join) bitonic
08:06 (quit) dnolen: Remote host closed the connection
08:08 (join) dnolen
08:31 (quit) bitonic: Ping timeout: 246 seconds
08:35 (quit) snorble_: Read error: Connection reset by peer
08:49 (join) chturne
09:00 (join) mye
09:01 (join) anRch
09:01 (join) mizu_no_oto
09:03 (join) getpwnam
09:03 (join) hash_table
09:03 (quit) jrslepak: Quit: What happened to Systems A through E?
09:12 (quit) mario-goulart: Remote host closed the connection
09:13 (join) mario-goulart
09:17 (quit) mizu_no_oto: Ping timeout: 245 seconds
09:19 (quit) jamessan: Ping timeout: 245 seconds
09:25 (join) jamessan
09:32 (quit) hash_table: Ping timeout: 244 seconds
09:32 (quit) getpwnam: Ping timeout: 244 seconds
09:35 (join) jrslepak
09:38 (join) jeapostrophe
09:38 (quit) jeapostrophe: Changing host
09:38 (join) jeapostrophe
09:38 (join) mizu_no_oto
09:40 (join) bitonic
09:42 (quit) vkz: Ping timeout: 246 seconds
09:45 (quit) chturne: Ping timeout: 260 seconds
09:46 (quit) neilv: Quit: Leaving
09:46 Gertm: I've installed racket with ./configure --futures-enabled, but when I do (futures-enabled?) I get #f
09:47 Gertm: How can I make sure these work?
09:55 (quit) Demosthenes: Ping timeout: 248 seconds
09:55 (quit) anRch: Quit: anRch
09:56 (quit) mye: Quit: mye
09:56 (quit) bitonic: Ping timeout: 252 seconds
09:57 (join) Demosthenes
09:57 (join) mceier
09:59 (join) bitonic
10:10 (quit) bitonic: Ping timeout: 245 seconds
10:25 (join) pcavs
10:28 (quit) jamessan: Ping timeout: 240 seconds
10:28 (quit) jeapostrophe: Ping timeout: 246 seconds
10:30 (join) bitonic
10:34 (join) jamessan
10:38 (join) DT`
10:42 (quit) dnolen: Ping timeout: 255 seconds
10:44 (join) getpwnam
10:44 (join) hash_table
10:45 (quit) bitonic: Ping timeout: 265 seconds
10:58 (join) bitonic
11:01 stamourv: Gertm: IIRC, in 5.3 futures are enabled by default.
11:01 stamourv: As for making sure they work, you can try the futures visualizer.
11:02 stamourv: docs.racket-lang.org/reference/futures-visualizer.html
11:03 stamourv: The guide has some examples you can try:
11:03 stamourv: http://docs.racket-lang.org/guide/performance.html#%28part._effective-futures%29
11:03 rudybot: http://tinyurl.com/6f2v62y
11:13 (quit) noelw: Quit: noelw
11:20 (join) MayDaniel
11:20 Gertm: stamourv: I installed the normal binary package but (futures-enabled?) still reports #f
11:22 stamourv: Which version?
11:23 Gertm: Latest. 5.3
11:25 samth: Gertm: what OS/architecture are you on?
11:26 Gertm: Linux Aramaki 3.5.6-1.fc17.x86_64 #1 SMP
11:27 stamourv: Gertm: Just checked, `futures-enabled?' is broken in 5.3. But futures still worked.
11:27 stamourv: It got fixed since.
11:27 (quit) hkBst: Quit: Konversation terminated!
11:27 Gertm: Aha, so if I'd build from the git repo, that should work?
11:27 stamourv: If you don't need `futures-enabled?' itself, you can use futures anyway.
11:28 Gertm: I'll try that later then. But is it generally a good idea or a bad idea to build from the git directly?
11:28 stamourv: Either that, or use the nightly builds.
11:28 stamourv: Git HEAD is usually pretty stabler.
11:28 Gertm: I don't mind building it myself.
11:28 stamourv: s/ler/le/
11:29 stamourv: You can check drdr.racket-lang.org to see whether a given state is broken.
11:29 stamourv: DrDr builds and runs the tests after each push.
11:29 Gertm: Cool, good to know.
11:30 Gertm: While I have people answering me here,..
11:30 Gertm: Will racket support #lang r7rs when it arrives?
11:31 stamourv: I don't know.
11:31 Gertm: Just curious.
11:31 stamourv: It's not due for a couple more years though, isn't it?
11:31 Gertm: Yah, won't be for the immediate future.
11:34 (quit) mizu_no_oto: Ping timeout: 276 seconds
11:38 (join) nilyaK
11:41 (quit) nilyaK: Read error: Connection reset by peer
11:48 (join) mizu_no_oto_
11:48 (nick) mizu_no_oto_ -> mizu_no_oto
11:52 (join) jao
11:52 (quit) jao: Changing host
11:52 (join) jao
11:53 (quit) bitonic: Read error: Operation timed out
11:54 (join) anRch
11:59 (join) bitonic
12:02 (quit) anRch: Ping timeout: 260 seconds
12:03 (join) mizu_no_oto
12:04 (join) jeapostrophe
12:04 (quit) jeapostrophe: Changing host
12:04 (join) jeapostrophe
12:06 (quit) pcavs: Quit: Leaving.
12:08 (join) mithos28
12:10 (join) anRch
12:12 (quit) mizu_no_oto: Ping timeout: 248 seconds
12:17 Gertm: Cool, now (futures-enabled?) returns #t.
12:20 (join) bitonic`
12:20 danl-ndi: jeapostrophe: where can I find the source for drdr?
12:21 jeapostrophe: danl-ndi: collects/meta/drdr
12:21 jeapostrophe: only in the "full" or git distro though
12:21 danl-ndi: ah, thanks
12:22 jeapostrophe: https://github.com/plt/racket/tree/master/collects/meta/drdr
12:22 (quit) bitonic: Ping timeout: 240 seconds
12:23 SamB_MacG5 didn't know there were partial distros
12:24 jeapostrophe: SamB_MacG5: ya, there's "text" and then "normal"
12:24 (join) pcavs
12:24 SamB_MacG5: what OSes are those for?
12:24 jeapostrophe: SamB_MacG5: and "normal" doesn't have stuff like tests or meta (iirc)
12:24 jeapostrophe: all
12:25 SamB_MacG5: I don't remember seeing OS X binaries for those ... then again, there are a lot of things I don't remember ...
12:25 jeapostrophe: http://download.racket-lang.org/all-versions.html
12:25 jeapostrophe: The second link for each release is the textual
12:26 jeapostrophe: and the only way to get binaries of everything in the dev repo is to use the nightly builds (where it is called "full", i think)
12:26 jeapostrophe: this page describes the versions pretty simply: http://pre.racket-lang.org/installers/
12:30 SamB_MacG5 knows another way ;-)
12:32 (quit) carleastlund: Quit: carleastlund
12:32 (quit) Nisstyre: Ping timeout: 240 seconds
12:33 SamB_MacG5 wonders why the OS X binaries are single-arch, and why there is a special source distribution for OS X
12:35 SamB_MacG5: For shame! text/html is NOT supposed to have an XML declaration!
12:35 (join) zhenya
12:36 zhenya: anyone know about rackunit?
12:36 jeapostrophe: zhenya: what do you need to know?
12:36 SamB_MacG5: http://validator.nu/?doc=http%3A%2F%2Fracket-lang.org%2Fdownload%2F
12:37 jeapostrophe: SamB_MacG5: eli maintains that page
12:37 SamB_MacG5: or, well, http://validator.nu/?doc=http%3A%2F%2Fracket-lang.org%2F for that matter
12:37 zhenya: jeapostrophe: when a check/test throws an exception, I would like to preserve the various check-info structs that with-check-info* puts "on the stack"
12:38 SamB_MacG5: hmm, the "div" errors are questionable though
12:38 zhenya: I have spent some time browsing the code and experimenting, and it seems that exception is caught "too far up the stack"
12:39 (quit) ssbr_: Ping timeout: 246 seconds
12:39 jeapostrophe: zhenya: can you use current-check-handler? can you gist a small example of what you'd like?
12:41 SamB_MacG5: jeapostrophe: do you know where the source is ?
12:41 jeapostrophe: SamB_MacG5: collects/meta/web -- https://github.com/plt/racket/tree/master/collects/meta/web
12:42 SamB_MacG5: okay
12:42 zhenya: silly example: I write: "(define-test-suite mySuite [test-equal? "" (/ 1 0) 1])"
12:42 (quit) anRch: Quit: anRch
12:43 SamB_MacG5: its probably a scribble thing anyway
12:43 jeapostrophe: SamB_MacG5: actually no, it doesn't use the scribble renderer
12:44 zhenya: I then use fold-test-results to run my suite, and collect test-failures and test-errors
12:45 SamB_MacG5: well, I'll take a crack at it after I do this weeding ...
12:45 zhenya: the test-failure struct has a stack field, which tells me about the various check-info bits that were added before the test was run
12:46 zhenya: However, test-error doesn't appear to have any of the check-info bits
12:47 (join) Nisstyre
12:47 jeapostrophe: test-error is inside an exn:test:check and the example you gave won't throw a rackunit error anyways
12:47 jeapostrophe: err scratch that first thing
12:48 zhenya: the point is test-error doesn't have a stack field which contains a list of check-infos
12:49 (quit) mithos28: Quit: mithos28
12:50 zhenya: the documentation for current-check-handler says it, "handles exceptions raised by check failures"
12:51 zhenya: the exception I am interested in, OTOH, is not "raised by check failure": it is raised by executing the test expression
12:51 jeapostrophe: yes
12:52 zhenya: and therein lies my difficulty :-)
12:53 jeapostrophe: i think we can and should change the test-result struct to have the check-info-stack in it
12:54 (join) snorble_
12:54 zhenya: it would need to be test-error, I think in order to help my specific case...
12:55 zhenya: basically, I would like some way to say, "please convert any exceptions you see to test-failures", or similar
12:58 jeapostrophe: zhenya: test-error is a subtype of test-result
12:59 jeapostrophe: and test-failure doesn't have the stack either
12:59 (join) mithos28
12:59 zhenya: test-failure does have the check-info stack
12:59 (join) jonrafkind
13:00 jeapostrophe: i don't want to make the change with talking with noel and ryanc though... could you start a thread on dev@ or user@ about this?
13:00 jeapostrophe: are you sure? i know exn:test:check does but that's not test-failure
13:01 zhenya: the result in test-failure _is_ 	exn:test:check
13:02 danl-ndi: jeapostrophe: I'm enjoying your post about running out of inodes.
13:02 danl-ndi: jeapostrophe: for comparison we have a 30 Tb file server and we are using 99% of the inodes right now
13:02 (join) ssbr_
13:02 jeapostrophe: zhenya: ah yes. i think it would be equally wise to just store the stack in all results, because you may want it for successes
13:03 jeapostrophe: danl-ndi: yikes!
13:03 danl-ndi: tell me about it. it's a disaster
13:03 jeapostrophe: danl-ndi: i considered using a loopback fs as well, but decided that it would be annoying to maintain. but it would be more orthogonal
13:04 danl-ndi: we using lots and lots of symlinks to version control large binary files
13:04 (quit) pcavs: Quit: Leaving.
13:05 danl-ndi: loopback... like a ramdisk?
13:05 danl-ndi: oh ... in a file. I see.
13:05 zhenya: run-test-case catches exn:test:check, and stores it in a test-failure as result
13:06 jeapostrophe: loopback: in linux you can create a big file, create a filesystem on it, and then mount that file as a new mount point
13:06 jeapostrophe: so you could do "mount /opt/tree.fake-disk /opt/tree"
13:07 jeapostrophe: and the root fs has 1 inode and a big file where the tree is an independent FS
13:07 jeapostrophe: the big pain is that you have to decide how big of a slice of the root fs you want to give it
13:07 danl-ndi: gotcha. that would have certain advantages. the inode problem exists, but now just in your virtual file system
13:07 zhenya: and fold-test-results uses run-test-case by default to run the tests...
13:07 (quit) ssbr_: Ping timeout: 252 seconds
13:07 danl-ndi: and the fs tree is totally isolated, which could be good or bad
13:07 jeapostrophe: ya, in my case it would have been pretty good because they're just "archives" and nothing will ever change
13:08 jeapostrophe: ya
13:08 jeapostrophe: zhenya: can you send an email to dev@ or user@ and i'll bring noel and ryan in on the change? we try not to change struct definitions willy nilly
13:08 jeapostrophe: danl-ndi: this is the sort of thing where ZFS would be great
13:09 jeapostrophe: or the plan 9 FS (for your symlink problem)
13:09 (quit) djcoin: Quit: WeeChat 0.3.7
13:09 zhenya: understood, about not changing struct definitions, and maybe I do have an unusual need...
13:09 (join) pcavs
13:09 (quit) mithos28: Quit: mithos28
13:10 danl-ndi: btw we have 200 users and each has 80k to 150k symlinks. hence inode madness.
13:10 (join) mithos28
13:10 jeapostrophe: zhenya: in the worst-case, the definition of the code that you're running is under 200 lines, so copying it is not crazy. BUT, i think we will be willing to change the struct def
13:10 (quit) bitonic`: Read error: Operation timed out
13:11 zhenya: yeah, true; I am not overly-enthusiastic about having "my private version of rackunit", however :-)
13:11 zhenya: anyway, II'll send an email (I can also explain the motivation better in an email)
13:12 jeapostrophe: zhenya: great
13:12 zhenya: thank you for walking through this with me!
13:14 jeapostrophe: np, better than talking to students ;)
13:15 (quit) mithos28: Quit: mithos28
13:19 danl-ndi: heh... yeah I went to college, but I'm not sure I earned the title "student" at any point
13:20 jeapostrophe: the best and worst part of professing is the students. when they are whining about having to work or grade grubbing, it is the worst. when they are enlightened and passionate because of your teaching, its the best
13:20 (quit) zhenya: 
13:21 jonrafkind: 80/20?
13:22 jeapostrophe: jonrafkind: the worst take up about 90% of your memories and dreads. but they are probably more like 40% of the students
13:25 danl-ndi: I certainly made some professors' days difficult
13:26 (join) noelw
13:29 (join) mizu_no_oto
13:32 danl-ndi: I'd love to find the professor who taught scheme at UCLA in 1992-1993....
13:35 jeapostrophe: he's made a big impact :)
13:37 danl-ndi: jeapostrophe: you're right about that. you guys do good work :)
13:38 (quit) noelw: Quit: noelw
13:39 danl-ndi: heh, found him. David Jefferson. He's at LLNL now...
13:41 danl-ndi: he taught a lower div class in scheme, and it started at 8am. the class was always full. he was clear, direct, and jumped right into the challenging stuff. it was great.
13:42 danl-ndi: i was really motivated for that class....
13:44 jeapostrophe: did he use EOPL? or SICP?
13:45 danl-ndi: neither... I totally forgot the textbook.
13:45 jeapostrophe: well, it was good anyways :)
13:46 danl-ndi: yup, it was.
13:46 (join) rsw
13:46 (quit) rsw: Client Quit
13:48 danl-ndi: he was working on internet-related election security back in 1994!
13:54 (quit) jeapostrophe: Ping timeout: 246 seconds
13:59 SamB_MacG5: if I want to invoke something in collections as a command-line program, how would I do that?
14:00 SamB_MacG5: (Or is racket(1) a proper manpage?)
14:00 jonrafkind: you mean without giving the full path to the file?
14:00 SamB_MacG5: well, I'm thinking of Python's -m flag
14:00 jonrafkind: actually racket -l foobar might work
14:00 jonrafkind: its like (require foobar) which should look in the collects tree by default
14:01 jonrafkind: hm now i have to read about python..
14:01 SamB_MacG5: does that do the right thing with the remaining args?
14:01 jonrafkind: not sure..
14:01 jonrafkind: maybe you need extra options to make it handle args properly
14:02 SamB_MacG5: actually, I think it does ...
14:02 (join) anRch
14:02 SamB_MacG5 seems to remember trying something of that sort before and it doing what he wanted, even if he didn't actually know precisely how it worked
14:03 SamB_MacG5: (-- might be needed)
14:04 SamB_MacG5: Soo ... any idea why racket *doesn't* have proper manpages?
14:04 SamB_MacG5: only stubs saying to try --help?
14:09 danl-ndi: I use -l <module> -m -- <remaing args to pass to module main>
14:09 danl-ndi: works a treat
14:09 SamB_MacG5: hmm, it might be working for me even without -m
14:10 danl-ndi: in your module do a (define (main . args) ...) and use command-line
14:10 danl-ndi: yeah -m just calls your main function. I use it as a consistent entry point to command line tools
14:11 SamB_MacG5: as it happens, the module is actually meta/web/build
14:12 SamB_MacG5: I don't think I've really seen any programs in racket's own tree follow that pattern, even though I'd rather expect them to do it whether or not there was special support for it
14:13 SamB_MacG5: (Though there might be slight variations in the arguments expected without the special support, of course.)
14:15 SamB_MacG5: I mean, Python hasn't got such support either, but it's more-or-less the standard except in really simple scripts ...
14:16 SamB_MacG5: er. scratch the "either"
14:16 SamB_MacG5 is goofy ...
14:27 danl-ndi: yeah, I'm not sure the reasons why say 'raco' isn't something that follows that pattern. I'm guessing it's used in setup a certain way...?
14:30 (join) noelw
14:31 (quit) masm: Ping timeout: 260 seconds
14:31 (join) bitonic`
14:34 SamB_MacG5: danl-ndi: well, raco/main and setup/main have a reasonable excuse for being a bit odd -- avoiding stale .zo files ...
14:36 SamB_MacG5: but given that they are also not supposed to be `require'd by normal stuff, the actual code from setup/main should probably be broken out into a `main' in a different module
14:37 (quit) noelw: Quit: noelw
14:38 SamB_MacG5: unless all that code is just there for the avoidance of stale .zo files ...
14:41 (join) jeapostrophe
14:41 (quit) jeapostrophe: Changing host
14:41 (join) jeapostrophe
14:41 danl-ndi: how do they avoid stale .zo files?
14:44 (quit) jonrafkind: Ping timeout: 260 seconds
14:46 (quit) jeapostrophe: Ping timeout: 256 seconds
14:46 (join) masm
14:54 samth: danl-ndi: the easy way -- they ignore zo files
15:03 danl-ndi: is that the compile-omit-paths in the info.rkt/
15:03 danl-ndi: ?
15:04 (join) jeapostrophe
15:04 (quit) jeapostrophe: Changing host
15:04 (join) jeapostrophe
15:06 (quit) mizu_no_oto: Quit: ["Textual IRC Client: www.textualapp.com"]
15:06 (quit) anRch: Quit: anRch
15:09 (join) jonrafkind
15:13 (join) noelw
15:13 samth: danl-ndi: no, that's just files that are ignored by the compilation process for that collection
15:14 samth: raco/main and such genuinely don't every use zo files, even if they exist
15:15 (quit) noelw: Client Quit
15:29 (join) uselpa
15:30 (quit) jeapostrophe: Read error: Operation timed out
15:30 (join) jeapostrophe
15:30 (quit) jeapostrophe: Changing host
15:30 (join) jeapostrophe
15:32 (join) chturne
15:55 (quit) uselpa: Remote host closed the connection
16:22 (join) broject_manager
16:27 (quit) cdidd: Remote host closed the connection
16:56 jrslepak: is there a simple way to get redex's trace/ps to produce output that's formatted according to one of traces' layout menu options? (dot, neato, etc.)
16:59 (quit) bitonic`: Ping timeout: 246 seconds
17:01 (quit) hash_table: Ping timeout: 260 seconds
17:01 (quit) getpwnam: Ping timeout: 260 seconds
17:16 (quit) Demosthenes: Ping timeout: 276 seconds
17:22 (join) bitonic`
17:25 (join) ssbr_
17:28 (join) Demosthenes
17:44 (join) noelw
17:45 (join) RacketCommitBot
17:45 RacketCommitBot: [racket] plt pushed 2 new commits to master: http://git.io/WENymA
17:45 RacketCommitBot: [racket/master] fixed template tests - Ryan Culpepper
17:45 RacketCommitBot: [racket/master] make macro stepper recover from jumps within expansion - Ryan Culpepper
17:45 (part) RacketCommitBot
17:46 (quit) MayDaniel: Read error: Connection reset by peer
17:52 (quit) noelw: Quit: noelw
17:54 (quit) jrslepak: Quit: What happened to Systems A through E?
17:59 (join) getpwnam
17:59 (join) hash_table
18:16 (quit) jeapostrophe: Ping timeout: 260 seconds
18:21 eli: SamB_MacG5: That xml line is a result of some obscure problems that were happenning at the time.
18:21 eli: I don't remember the details, but it was all a very tricky issue to deal with.
18:21 eli: Things like IE obscure behavior etc.
18:21 (quit) bitonic`: Ping timeout: 265 seconds
18:21 SamB_MacG5: hmm
18:21 eli: You can see that header line at the top of "collects/scribble/html/html.rkt"
18:21 eli: so changing it is very easy.
18:22 eli: I don't want to touch it too much ATM, since my understanding is that HTML5 fixes the whole mess.
18:22 eli: (By dumping the whole proper xml thing.)
18:23 SamB_MacG5: HTML5 doesn't dump XML ... but they do scrap appendix C of the XHTML 1.0 spec, as I understand it
18:23 SamB_MacG5: not that that was ever normative anyway
18:23 eli: It doesn't?
18:24 SamB_MacG5: HTML5 does undeprecate the non-XML syntax, and also stops pretending it's this syntax is an SGML application
18:24 eli: I'm 80% sure that I saw somewhere that they'll allow stuff like <p>...<p>...<p> which is not valid xml.
18:24 SamB_MacG5: anyway, probably for the site you would want to use the "HTML syntax"
18:24 samth: right
18:25 samth: i would be shocked if taking out that XML line had any bad effect
18:25 samth: we really don't want IE quirks mode
18:25 eli: samth: If you're willing to test things out thoroughly, I'll be happy to take it out.
18:26 eli: (thoroughly = not just using that snapshots site...)
18:26 SamB_MacG5: eli: well, yes
18:26 samth: meaning what then?
18:26 SamB_MacG5: obviously you'd want to run the validator over a larger corpus ;-)
18:26 eli: (It might also be related to the apache configuration and some need to keep the .html suffix, I don't remember.)
18:26 eli: samth: Actually using it.
18:27 SamB_MacG5: eli: I don't think IE supports the XML syntax ...
18:27 samth: wait, what do you mean by "snapshots site"?
18:27 eli: samth: Some site where you give it a url and it gives you a bunch of snapshots of how it looks like on N browsers.
18:28 samth: there are sites that let you actually browse it with N browsers
18:28 SamB_MacG5: eli: oh, I thought you meant not just using that download page
18:28 samth: would that be sufficient for you?
18:28 SamB_MacG5: samth: how in the world do they do that?
18:28 samth: SamB_MacG5: virtual machines
18:28 eli: samth: No -- whetever the problems were, they ran deeper than just looking at the rendering.
18:29 SamB_MacG5: samth: how does the input get back to the browsers?
18:29 samth: eli: what other problems could there be?
18:29 SamB_MacG5: samth: behaviour
18:29 samth: and how would i discover them?
18:29 (join) jrslepak
18:29 eli: samth: Like I said, I don't really remember...
18:29 samth: i would do the same thing either way
18:30 eli: The bottom line is that I am in no way trying to keep that xml line -- my only concern is to avoid breaking things.
18:30 samth: then i don't see how i could possibly test to your statisfaction
18:30 eli: Yes, you'll have to use Windows + IE.
18:30 eli: Preferably an older one.
18:30 eli: Opera too.
18:30 SamB_MacG5: good thing MS provides VM images for that
18:30 samth: but what would I *do* with it?
18:30 samth: just click around?
18:31 eli: And Links/Lynx/whatever.
18:31 SamB_MacG5: try browsing the docs and clicking on gadgetry
18:31 SamB_MacG5: links is not going to be hard ;-)
18:31 eli: SamB_MacG5: No, not the docs -- that's a completely different renderer/
18:31 eli: s/\//./
18:31 SamB_MacG5: eli: oh, really?
18:31 SamB_MacG5 is confused by the "scribble" in that URL
18:31 SamB_MacG5: er. directory name
18:32 eli: SamB_MacG5: "Scribble" first referred to the syntax, then it referred to the syntax + the documentation system, now it's more associated with just the documentation system.
18:32 (quit) pcavs: Quit: Leaving.
18:32 (quit) bjz: Quit: Leaving...
18:32 eli: The `scribble/html' code is a library to render html content, using the scribble syntax.
18:34 SamB_MacG5: ah
18:34 SamB_MacG5: perhaps it should be renamed
18:35 eli: We've talked about this, but no major problem yet to force it.
18:35 SamB_MacG5: speaking of which, it seems like there ought to be a way to indicate that a library is being migrated to a new name, so that e.g. DrRacket could give gentle nudges about it ...
18:35 eli: There are some more generic tools in there.  Like `scribble/text' and using `scribble/reader' directly for alternative concrete syntax.
18:36 SamB_MacG5: yeah, scribble/syntax makes perfect sense
18:36 SamB_MacG5: er, /reader
18:36 SamB_MacG5: because that's the reader described in the scribble docs
18:39 SamB_MacG5: ... anyways, I would have expected the webpages to share more infrastructure with the documentation, though obviously the manual/ stuff wouldn't be too helpful...
18:42 eli: Exactly -- the amount of sharing would be tiny.
18:42 eli: It should happen at some point -- currently the docs are rendered via xexprs, instead of the newer `scribble/html'.
18:42 eli: (Mostly my fault -- it's still not documented.)
18:42 eli disappears
18:44 SamB_MacG5: eli: ah. that would be a valid reason to include an xml declaration, at least ...
18:44 SamB_MacG5: assuming xexprs doesn't have an HTML output mode
18:45 (quit) jonrafkind: Ping timeout: 252 seconds
18:45 (join) bjz
18:59 (join) RacketCommitBot
18:59 RacketCommitBot: [racket] plt pushed 10 new commits to release: http://git.io/Yxz9ew
18:59 RacketCommitBot: [racket/release] adjust the code that initializes the pre-"first Run" repl - Robby Findler
18:59 RacketCommitBot: [racket/release] Allow `#:opaque` and `#:struct` in `require/typed`. - Sam Tobin-Hochstadt
18:59 RacketCommitBot: [racket/release] Handle contract generation for recursive class types. - Sam Tobin-Hochstadt
18:59 (part) RacketCommitBot
19:18 (join) RacketCommitBot
19:18 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/rJ6ZZA
19:18 RacketCommitBot: [racket/master] fix expander bug introduced by my last commit - Ryan Culpepper
19:18 (part) RacketCommitBot
19:19 (quit) chturne: Read error: Operation timed out
19:21 (join) jeapostrophe
19:21 (quit) jeapostrophe: Changing host
19:21 (join) jeapostrophe
19:28 (quit) jeapostrophe: Ping timeout: 252 seconds
19:28 (quit) bjz: Quit: Leaving...
19:30 (join) RacketCommitBot
19:30 RacketCommitBot: [racket] plt pushed 1 new commit to release: http://git.io/PSWsWQ
19:30 RacketCommitBot: [racket/release] New Racket version 5.3.0.900. - Eli Barzilay
19:30 (part) RacketCommitBot
19:40 (quit) ssbr_: Remote host closed the connection
19:50 SamB_MacG5: eli: I don't suppose you could tell me an easy way to generate a copy of the site for testing on some other domain than racket-lang.org
19:57 (join) jeapostrophe
19:59 (quit) dsantiago: Ping timeout: 245 seconds
20:01 (join) dsantiago
20:02 (quit) jeapostrophe: Ping timeout: 240 seconds
20:06 (quit) dsantiago: Ping timeout: 245 seconds
20:06 SamB_MacG5: possibly an apache configuration template ?
20:08 (join) dsantiago
20:17 (quit) dsantiago: Ping timeout: 260 seconds
20:20 (join) dsantiago
20:50 (quit) masm: Quit: Leaving.
21:03 (join) jeapostrophe
21:03 (quit) jeapostrophe: Changing host
21:03 (join) jeapostrophe
21:03 (quit) jesyspa: Quit: leaving
21:29 SamB_MacG5: hmm, "#lang meta/web" has funky syntax highlighting ...
21:30 ianjneu: can ya dig it
21:30 (join) pcavs
21:32 SamB_MacG5: I was thinking it might be accidental
21:32 (join) bjz
21:35 SamB_MacG5: something to do with using #:info of scribble/base with a different reader, maybe?
21:39 ianjneu: No idea.
21:51 (quit) pcavs: Quit: Leaving.
21:52 (join) pcavs
21:56 (join) ioa
22:45 (quit) pcavs: Quit: Leaving.
22:48 (join) pcavs
22:57 (quit) bjz: Ping timeout: 245 seconds
23:00 (join) bjz
23:01 (join) jonrafkind
23:22 (quit) Shviller: Ping timeout: 245 seconds
23:22 (join) Shviller
23:27 SamB_MacG5: Looks like my guess was right: that #:info function uses the -inside version of the lexer, and it gets paired up with the non -inside version of the reader ...
23:30 (join) Fare
23:46 (join) vu3rdd
23:46 (quit) vu3rdd: Changing host
23:46 (join) vu3rdd
23:51 (join) RacketCommitBot
23:51 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/XcBXtg
23:51 RacketCommitBot: [racket/master] [honu] remove debugging lines - Jon Rafkind
23:51 (part) RacketCommitBot