00:01
<
struktured >
I was wondering what all the focus about fixed points in ocaml was all of a sudden..
00:01
keen_ has quit [Read error: Connection reset by peer]
00:04
keen_ has joined #ocaml
00:16
seanmcl has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
00:18
AltGr has joined #ocaml
00:26
badkins has joined #ocaml
00:26
swgillespie has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
00:27
kdef has quit [Quit: Leaving]
00:38
<
Drup >
rgrinberg: the issue is earlier in the toolchain
00:38
<
Drup >
when parsing the header
00:39
<
Drup >
well, the frame really
00:39
<
Drup >
in ocsigen branch, the url is truncated and contains only the path and after, it seems
00:40
<
Drup >
master branch*
00:40
<
Drup >
I'm not sure what to fix, at this point.
00:40
nojb has quit [Quit: nojb]
00:41
<
Drup >
the cohttp migration break ocsigenserver because cohttp is correct :>
00:42
Nijikokun has joined #ocaml
00:43
nojb has joined #ocaml
00:47
<
Drup >
the funny part is that I'm pretty sure some value are garanted to be None in the master branch because of that, I wonder how it works down the line
00:48
nojb has quit [Quit: nojb]
00:54
jwatzman|work has quit [Quit: jwatzman|work]
00:58
<
rgrinberg >
Drup: i'll have a look later tonight. can you update the PR with your findings
00:58
<
rgrinberg >
e.g. link to the truncation
00:58
nojb has joined #ocaml
00:59
_pakchoi_ has joined #ocaml
00:59
<
Drup >
I'm fixing it
01:00
<
Drup >
by removing the old pipeline and iterating until it produces consistent metadata between old and new version
01:01
<
rgrinberg >
go go Drup
01:01
_pakchoi_ has quit [Client Quit]
01:01
_pakchoi_ has joined #ocaml
01:02
<
Drup >
and recompiling eliom inbetween è_é
01:02
_pakchoi_ has quit [Client Quit]
01:03
claudiuc has quit [Remote host closed the connection]
01:03
swgillespie has joined #ocaml
01:05
<
Drup >
it's working
01:06
<
Drup >
well, not completely
01:06
<
Drup >
it's working
*better*
01:07
djellemah_ has joined #ocaml
01:09
<
Drup >
oops, wrong one
01:09
<
rgrinberg >
how does the first link not have a scheme? O_O
01:09
<
rgrinberg >
i thuoght i added it
01:09
<
Drup >
rgrinberg: I may not have updated cohttp
01:09
<
rgrinberg >
no that was an ocsigenserver#cohttp work around
01:10
<
Drup >
I don't remember such patch
01:10
djellemah has quit [Ping timeout: 256 seconds]
01:13
nojb has quit [Quit: nojb]
01:14
<
rgrinberg >
wtf O_O trying to find it
01:15
<
Drup >
I remember your other patch
01:15
<
rgrinberg >
Drup: wtf you killed my patch :P
01:15
<
rgrinberg >
haha i don't know but it's gone
01:16
<
Drup >
did you do a pull request for it ?
01:17
<
rgrinberg >
yes that one
01:17
<
Drup >
ok, adding it back
01:18
<
rgrinberg >
Drup: what happened to it?
01:18
<
Drup >
I have no idea
01:18
<
rgrinberg >
lost in some rebase haze
01:18
<
Drup >
I tried to be carefull when rebasing
01:19
<
rgrinberg >
anyway i don't think this will fix anything but for the poor sod out there relying on it...
01:19
<
rgrinberg >
btw this is why the url parsing was failing befor O_O
01:19
<
Drup >
hum, I probably picked your commit, tried to rebase it into a previous commit, show some conflicts, decided it was too complicated, aborted and forgot to pick again
01:19
The_Mad_Pirate has joined #ocaml
01:19
<
rgrinberg >
like i said it relies on ^https? being present
01:20
<
Drup >
rgrinberg: as I said
01:20
<
Drup >
even before parsing
01:20
<
Drup >
the url are different
01:21
<
Drup >
after parsing the headers
01:21
<
Drup >
(but yes, it's totally possible)
01:21
natrium1970 has quit [Quit: natrium1970]
01:21
nojb has joined #ocaml
01:22
moei has joined #ocaml
01:22
_pakchoi_ has joined #ocaml
01:24
<
Drup >
(why can't git do partial stashes :x)
01:26
nojb has quit [Client Quit]
01:34
<
rgrinberg >
Drup: the patch doesn't apply cleanly?
01:34
<
rgrinberg >
it's trivial
01:34
<
Drup >
rgrinberg: it works fine with your patch
01:35
<
Drup >
I feel so sad now that I spend time debuging this é_è
01:35
<
rgrinberg >
now you learned some uri though
01:35
<
Drup >
and I added loging
01:35
<
rgrinberg >
hopefully all the patches are there
01:35
<
Drup >
pretty sure that's the only I missed
01:36
<
rgrinberg >
yeah, dinos stuff looks like it's there
01:36
<
rgrinberg >
Drup: you also cleaned up that ugly re too right?
01:36
<
rgrinberg >
relying on Uri accessors
01:36
<
rgrinberg >
instead
01:37
<
Drup >
No, I stashed the patch
01:37
<
Drup >
it doesn't work for everything directly
01:37
<
rgrinberg >
ok, can you make a PR for it somewhere though?
01:37
<
Drup >
so gotta take the safe solution for now and bring it back later
01:37
<
rgrinberg >
i'd like to have a look
01:42
pyon has quit [Ping timeout: 264 seconds]
01:44
cesar_ has joined #ocaml
01:44
cesar_ is now known as Guest868
01:46
<
Drup >
rgrinberg: pushed all my current patches now, you can start again testing :)
01:48
Guest868 has quit [Ping timeout: 265 seconds]
01:52
<
struktured >
what is a good "is_nan" in ocaml? compare nan x = 0 ? or is it.. nan == x ? or something else entirely?
01:54
<
Drup >
that's (almost) the specification of nan :p
01:54
<
struktured >
yeah that was my first solution but I feel like I'm abusing ==
01:55
<
struktured >
nan is so weird..I'll never come to terms with it
01:55
<
Drup >
this is probably the least weird things among FP numbers
01:55
<
struktured >
so true
01:55
<
Drup >
denormalized numbers are ... X_x
01:55
<
struktured >
I wasted 4 hours yesteday thinking my algorithm was numerically unstable. turns out I didn't understand the range of my function
01:56
<
Drup >
you think it's clever, then you try to prove properties about them, and you cry
01:57
<
struktured >
related to it, is how they don't know how make iterations to approximate a continued fraction. fabulous
01:57
<
struktured >
*how many iteraitons
01:57
rgrinberg has quit [Ping timeout: 256 seconds]
02:00
<
Drup >
Jean Michel Muller was my teacher about FP things :D
02:01
BitPuffin has quit [Ping timeout: 245 seconds]
02:02
<
Drup >
(proof about FP numbers, on blackboard at 9 in the morning, he literally killed us)
02:04
swgillespie has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
02:04
ygrek has joined #ocaml
02:07
badkins has quit []
02:09
rgrinberg has joined #ocaml
02:10
swgillespie has joined #ocaml
02:10
<
rgrinberg >
Drup: why is the ip different :O
02:11
<
Drup >
I have no idea
02:11
<
Drup >
I'm doing all the tests right now
02:11
<
Drup >
can you try too
02:12
<
rgrinberg >
kind of curious - how much work would it take to rip out string urls out of eliom?
02:13
<
Drup >
for tyxml, basically none
02:14
<
Drup >
it's in the base functor
02:14
<
Drup >
so, for the html stuff, it's pretty much free
02:14
struktured has quit [Ping timeout: 245 seconds]
02:15
<
Drup >
once you have done ocsigenserver, eliom will follow, I think
02:15
<
Drup >
the tricky part being, obviously, to produce the exact same behavior as current ocsigenserver
02:17
<
rgrinberg >
hopefully not bug for bug
02:18
pyon has joined #ocaml
02:22
<
rgrinberg >
"broken on both" ^_^
02:23
<
Drup >
so many tests
02:23
<
Drup >
so little automation
02:24
<
rgrinberg >
Drup: would it be possible to have ppx_deriving show piggy back on sexp when there's no deriving instance?
02:24
<
Drup >
you need access to the Env for that, and you don't have it at ppx time
02:25
manizzle has quit [Ping timeout: 246 seconds]
02:25
<
Drup >
however, [@printer (fun fmt x -> Sexp.pp fmt (to_sexp x)] should work well
02:26
<
Drup >
note that for ocsigenserver, dinosaure wrote a printing function (and removed all usages, for whatever reason)
02:26
<
Drup >
(or Cohttp.Request.t)
02:29
<
rgrinberg >
what kind of a test suite is this T_T
02:30
<
Drup >
don't ask me
02:31
<
rgrinberg >
Drup: i guess we assume that double failure is success
02:31
<
rgrinberg >
or file a bug on eliom
02:32
<
Drup >
yeah, I'm just posting them there by lazyness
02:33
<
rgrinberg >
ps 0.16.0 has been released so you don't need to pin cohttp
02:34
<
rgrinberg >
although i run everything against master
02:34
<
Drup >
comet is broken
02:34
<
rgrinberg >
which test is that
02:35
<
Drup >
nevermind, it's only broken if you use 127.0.0.1 instead of localhost, and for both
02:36
<
Drup >
(and work otherwise)
02:37
cesar_ has joined #ocaml
02:37
cesar_ is now known as Guest7023
02:38
struktured has joined #ocaml
02:39
<
rgrinberg >
Drup: you brought back lwt.extra T_T
02:39
<
Drup >
I forgot to remove it :D
02:39
nullcat_ has joined #ocaml
02:41
shinnya_ has quit [Ping timeout: 264 seconds]
02:41
<
Drup >
didn't now this eliom monitor thing
02:42
<
Drup >
rgrinberg: at the /monitor path
02:42
<
Drup >
the cohttp branch has twice less minor gc
02:44
<
rgrinberg >
any other differences?
02:45
<
Drup >
twice less active lwt thread, probably the cause
02:46
<
Drup >
twice more words in the major heap
02:47
<
Drup >
a bit less file descriptor too
02:47
<
Drup >
nothing highly conclusive, anyway, the server is not doing much
02:48
<
Drup >
rgrinberg: are you motivated to investigate the IP address bug ?
02:49
<
Drup >
I can tell you it's the remote_ip_addr_* fields in the request_info stuff
02:50
Sorella has quit [Quit: Connection closed for inactivity]
02:50
<
rgrinberg >
Drup: sure but imo it's low priority
02:51
<
Drup >
well, I don't see any other failure at this point
02:51
<
rgrinberg >
what was the conclusion with cumulus?
02:51
<
Drup >
dinosaure said it was working
02:52
<
Drup >
AltGr: ping ?
02:52
<
rgrinberg >
can he convince jpdeplaix to run it in production?
02:52
<
rgrinberg >
if it's "working"
02:52
<
Drup >
rgrinberg: they did at some point :D
02:53
<
rgrinberg >
ok what about ocsigen.org itself
02:53
<
rgrinberg >
can we run that in production :P
02:53
<
Drup >
ok, trying to compile ocsimore :>
02:53
* Drup
expects explosions
02:54
<
Drup >
I don't even have it cloned :D
02:59
<
rgrinberg >
Drup: about make run.opt.local, why doesn't it serve the directory content for /ocsigenstuff/
02:59
<
rgrinberg >
Drup: it doesn't seem like a cohttp thing but i'd like to test it
03:00
<
Drup >
you mean list the files
03:00
<
Drup >
probably because it's an option disabled by default
03:00
<
Drup >
you can enable it, I guess
03:00
<
rgrinberg >
Drup: how do i?
03:00
<
Drup >
I linked you the config doc the other day
03:00
<
Drup >
give me a sec
03:01
<
rgrinberg >
it talks about "listdir"
03:01
<
rgrinberg >
but i'm not sure where to shove it
03:02
_pakchoi_ has quit [Remote host closed the connection]
03:03
<
AltGr >
Drup, hm, so you need specific apt repo to be installed for the package tests to run ?
03:03
nullcat_ has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
03:04
<
Drup >
AltGr: well, for the package to compile period
03:04
<
Drup >
I need llvm 3.6 and that's not in ubuntu
03:04
<
Drup >
rgrinberg: add <extension findlib-package="ocsigenserver.ext.extendconfiguration"/> under the other extension
03:04
<
AltGr >
hm, there is currently a way, but I really don't like it
03:05
<
Drup >
and add <listdirs value="true" /> inside the <host> tag
03:05
<
AltGr >
(it's totally insecure, and will be removed in 1.3; although I don't have a solution yet to fill that gap)
03:05
<
Drup >
(god that would be better with option arguments)
03:05
nullcat has joined #ocaml
03:05
<
AltGr >
also, what about other OSes ?
03:06
<
Drup >
what about it ? my thing is pure ocaml, once llvm is installed
03:07
<
AltGr >
well, you can host a shell-script and put the URL in depexts, with flag "source"
03:07
<
Drup >
AltGr: also if you look at the logs, I think there may be some aspcud issues. The chocies are not always optimal
03:07
<
Drup >
AltGr: shouldn't that be done in llvm ?
03:07
<
Drup >
wait, isn't that already done in llvm ?
03:07
<
AltGr >
then opam-depext will download and run it
03:07
<
AltGr >
ah, possibly
03:07
<
rgrinberg >
Drup: insane, it works
03:08
<
Drup >
rgrinberg: xml magic
03:08
<
Drup >
AltGr: wouldn't help the "recent-llvm" issue
03:08
<
rgrinberg >
let's see if ocsigen can do something useful with some pom.xml i have around...
03:09
<
Drup >
AltGr: then the opam CI script should use that ?
03:09
<
Drup >
(it clearly doesn't at the moment)
03:09
yomimono has quit [Ping timeout: 250 seconds]
03:09
nullcat has quit [Client Quit]
03:10
<
Drup >
(I'm not sure how I feel about the script adding a ppa, but that's not really my concern)
03:10
Sorella has joined #ocaml
03:13
Guest7023 has quit [Remote host closed the connection]
03:16
<
AltGr >
hm, indeed, I thought opam-repo's travis script used opam-depext, but not yet ; it does the checks and installs manually from the shell
03:18
<
Drup >
So many questions in this file :|
03:20
<
Drup >
oh, rgrinberg, it's even in latin1 !
03:20
<
AltGr >
and for some reason `opam install llvmgraph -e source,linux` doesn't return the expected results (bad versions ?). It should use `opam list --rec --required-by llvmgraph -e source,linux -s`, which works properly, anyway.
03:20
<
rgrinberg >
Drup: there was this news that the French government posted their civil laws on github
03:20
<
rgrinberg >
i wanted to check if it was latin-1 :P
03:21
<
Drup >
oh, I didn't look
03:21
<
Drup >
(and it was not the french gov, just a french enthusiast)
03:22
<
rgrinberg >
so does ocsimore compile?
03:24
<
Drup >
nope, but for something unreleated
03:24
<
Drup >
(we changed the logging system not too long ago)
03:27
<
Drup >
that's going to be tricky
03:27
<
Drup >
I'm not sure on which branch of eliom it relies now
03:28
<
Drup >
will ask vincent tomorrow
03:29
<
rgrinberg >
We'll need some longer running tests in production
03:29
<
rgrinberg >
to test for things like memory leaks
03:31
darkf has joined #ocaml
03:32
<
rgrinberg >
Drup: so what is an ocsigen http frame conceptually?
03:34
<
rgrinberg >
just a response i guess..
03:34
<
Drup >
well, look at the types
03:34
<
Drup >
I don't really have a better answer than that
03:34
<
rgrinberg >
yeah the name thuogh
03:35
<
rgrinberg >
not really sure what it's supposed to hint at
03:35
<
Drup >
I always though it was, you know, an http frame :D
03:35
<
Drup >
but never really wondered, since it doesn't interest me
03:35
<
Drup >
there is a ldap client in ocsimore
03:35
<
Drup >
and a wikicreole parser
03:35
<
AltGr >
Drup, ok: it's just because the llvm.3.2 package is broken
03:36
<
AltGr >
and the CI scripts run without aspcud, and that version gets chosen
03:37
<
Drup >
AltGr: hum, is it the case for all bots ?
03:38
<
AltGr >
I guess; I have plans for large improvements of opam repo's CI scripts
03:39
<
Drup >
rgrinberg: also some pam stuff in there
03:40
MisaMisa has joined #ocaml
03:42
<
Drup >
(fabulous adventures in the land of unsplitted ocsigen libraries)
03:45
<
rgrinberg >
any smtp implementations?
03:45
<
rgrinberg >
an odd proof of fermat's last theorem lying around?
03:45
<
rgrinberg >
perhaps
03:45
<
Drup >
I wouldn't exclude the smtp implementation somwhere
03:52
<
Drup >
rgrinberg: do you have a script for latin1 to utf8 ?
03:56
<
Drup >
AltGr: what should I do for the llvmgraph PR ?
03:58
<
AltGr >
either depend on a higher version of llvm, or add depext scripts to the older 3.2, I guess
03:59
<
Drup >
I already depend on llvm 3.6 for the new version
03:59
<
Drup >
the old version doesn't, because it has no reason to
03:59
rgrinberg has quit [Ping timeout: 244 seconds]
04:04
rgrinberg has joined #ocaml
04:05
ggole has joined #ocaml
04:06
<
Drup >
rgrinberg: ocsimore fixed, didn't tried to run it, as it needs a postgres db and stuff and I'm going to bed
04:06
<
rgrinberg >
Drup: no but i do it pretty quickly if it's the same kind of change in vim
04:06
<
rgrinberg >
or i just delete the comment :P
04:06
<
rgrinberg >
Drup: so early?
04:10
<
Drup >
yes, so early :p
04:16
s1n4 has joined #ocaml
04:18
swgillespie has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
04:32
rgrinberg has quit [Ping timeout: 244 seconds]
04:34
WraithM has joined #ocaml
04:34
swgillespie has joined #ocaml
04:35
MercurialAlchemi has joined #ocaml
04:35
swgillespie has quit [Client Quit]
04:35
<
SGrondin >
I'm benchmarking different platforms at work for a new backend that needs to serve over 100,000 req/s
04:36
<
SGrondin >
I did Node, OCaml, Python and now I'm doing Vert.x (Java)
04:36
<
SGrondin >
OCaml blows everything out of the water in terms of speed and code quality
04:37
<
SGrondin >
Vert.x is nice but sooooooooooo verbose, all the callbacks are done by hand in anonymous inner classes and the API is both complex and incredibly undocumented
04:37
<
SGrondin >
Not sure about its performance yet, but it usually wins all the http benchmarks... stay tuned
04:38
rgrinberg has joined #ocaml
04:38
<
Drup >
SGrondin: which lib for ocaml ?
04:38
<
SGrondin >
cohttp of course!
04:38
<
SGrondin >
with Core and Lwt
04:41
<
rgrinberg >
SGrondin: i'd actually async to do better than lwt
04:41
<
rgrinberg >
it's much more optimized
04:41
<
SGrondin >
hmm, good to know
04:41
<
SGrondin >
I just like using Lwt so much more
04:42
<
rgrinberg >
SGrondin: i had a blog post about it :P
04:43
<
rgrinberg >
lwt vs async that is
04:43
<
rgrinberg >
async also wraps epoll directly which i imagine is pretty impossible to beat on linux
04:43
<
SGrondin >
Yeah I remember it
04:43
<
Drup >
last time I asked about async vs lwt benches, the results were mixed
04:43
<
Drup >
it was basically "depends of what you are doing"
04:43
<
Drup >
you are saying async is faster ?
04:44
<
rgrinberg >
Drup: when was that? I keep seeing more and more optimizations done to async
04:44
<
rgrinberg >
while lwt is kind of in maintenance mode
04:44
<
Drup >
a few month ago
04:44
<
rgrinberg >
Drup: can i see it?
04:45
<
Drup >
not really, it was IRL :p
04:45
swgillespie has joined #ocaml
04:45
<
rgrinberg >
anyway cohttp is pretty good for testing both of these
04:45
<
rgrinberg >
but yes there's too many variables
04:46
<
rgrinberg >
does not matter though, using kills kittens and the community
04:46
<
rgrinberg >
only use it when you must
04:47
<
SGrondin >
I agree, I just wish Lwt was more maintained
04:47
<
SGrondin >
In the future I'd like to contribute to it
04:47
<
rgrinberg >
SGrondin: that's one advantage to lwt
04:47
<
rgrinberg >
you can contribute to its development effectively
04:48
<
rgrinberg >
async is developed behind closed doors and tarball dropped on us
04:49
<
SGrondin >
oh yeah!
05:02
cesar_ has joined #ocaml
05:02
cesar_ is now known as Guest87813
05:06
Guest87813 has quit [Ping timeout: 245 seconds]
05:23
larhat has quit [Quit: Leaving.]
05:37
MisaMisa has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
05:39
olibjerd has joined #ocaml
05:42
Submarine has joined #ocaml
05:49
manizzle has joined #ocaml
05:52
ygrek has quit [Ping timeout: 255 seconds]
05:53
psy_ has quit [Ping timeout: 246 seconds]
05:58
Submarine has quit [Quit: Leaving]
05:58
SGrondin has left #ocaml [#ocaml]
05:59
ericwa has quit [Quit: Leaving...]
06:04
MisaMisa has joined #ocaml
06:12
swgillespie has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
06:13
psy_ has joined #ocaml
06:23
<
jpdeplaix >
rgrinberg: mmmh ? What did I do ?
06:24
<
rgrinberg >
jpdeplaix: we want to guinea pig cumulus to test the cohttp ocsigenserver branch
06:24
<
rgrinberg >
in production :D
06:24
nullcat has joined #ocaml
06:24
swgillespie has joined #ocaml
06:24
<
rgrinberg >
as it seems to work locally
06:26
<
jpdeplaix >
ahhh euh yes. Which branches of eliom and ocsigenserver do I have to use ?
06:28
<
rgrinberg >
eliom/cohttp only has 1 tiny commit though
06:30
Sorella has quit [Quit: Connection closed for inactivity]
06:31
<
jpdeplaix >
rgrinberg: does it still relies on the dev version of cohttp and conduit ?
06:31
<
rgrinberg >
jpdeplaix: nah, the 2 latest versions of cohttp should work (0.15.2 and 0.16.0)
06:31
<
rgrinberg >
prefer the latest though
06:40
mjboa has joined #ocaml
06:49
ygrek has joined #ocaml
06:50
swgillespie has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
06:50
cesar_ has joined #ocaml
06:51
cesar_ is now known as Guest25481
06:55
Guest25481 has quit [Ping timeout: 245 seconds]
07:08
nojb has joined #ocaml
07:09
WraithM has quit [Quit: leaving]
07:09
nojb has quit [Client Quit]
07:15
Nijikokun has quit [Ping timeout: 252 seconds]
07:25
tane has joined #ocaml
07:27
djellemah_ has quit [Quit: Leaving]
07:27
<
jpdeplaix >
Drup: I got « ocamlfind: Package `ppx_deriving.std' not found » with the 4.01.0
07:27
djellemah has joined #ocaml
07:27
<
jpdeplaix >
rgrinberg: ^
07:28
<
rgrinberg >
jpdeplaix: looks like you need to install ppx_deriving too Drup added it for debugging :P
07:29
<
jpdeplaix >
rgrinberg: but I'm not using the 4.02 :/
07:29
<
jpdeplaix >
so I can't install it
07:31
<
jpdeplaix >
mmh ok I'm gonna do another branch without this one
07:33
darkf_ has joined #ocaml
07:36
darkf has quit [Ping timeout: 245 seconds]
07:42
Simn has joined #ocaml
07:42
<
jpdeplaix >
rgrinberg: done. It's running
07:42
<
jpdeplaix >
Drup: ^
07:42
<
jpdeplaix >
dinosaure: ^
07:43
<
rgrinberg >
jpdeplaix: awesome! just checked still running for me :D
07:52
darkf_ is now known as darkf
07:54
ocp has joined #ocaml
08:03
<
companion_cube >
o/
08:08
rwmjones_hol is now known as rwmjones
08:10
johnelse is now known as johnel_away
08:10
lanii has joined #ocaml
08:14
johnel_away is now known as johnelse
08:24
axiles has joined #ocaml
08:27
testcocoon has quit [Quit: Coyote finally caught me]
08:27
rgrinberg has quit [Ping timeout: 256 seconds]
08:32
larhat has joined #ocaml
08:34
nullcat has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
08:35
freling has joined #ocaml
08:39
cesar_ has joined #ocaml
08:40
cesar_ is now known as Guest98443
08:44
Guest98443 has quit [Ping timeout: 245 seconds]
08:45
jbrown has joined #ocaml
08:49
testcocoon has joined #ocaml
08:50
dsheets has quit [Ping timeout: 265 seconds]
08:58
milosn has quit [Ping timeout: 250 seconds]
09:00
rossberg has joined #ocaml
09:03
lanius777 has joined #ocaml
09:06
lanii has quit [Ping timeout: 245 seconds]
09:11
manizzle has quit [Ping timeout: 245 seconds]
09:23
rgrinberg has joined #ocaml
09:28
rgrinberg has quit [Ping timeout: 250 seconds]
09:29
kakadu has joined #ocaml
09:33
dsheets has joined #ocaml
09:45
nojb has joined #ocaml
09:45
oscar_toro has joined #ocaml
09:49
Haudegen has quit [Ping timeout: 244 seconds]
10:05
Haudegen has joined #ocaml
10:12
milosn has joined #ocaml
10:21
AltGr has left #ocaml [#ocaml]
10:22
ollehar has joined #ocaml
10:28
mjboa has quit [Quit: leaving]
10:28
cesar_ has joined #ocaml
10:29
cesar_ is now known as Guest41639
10:33
Guest41639 has quit [Ping timeout: 245 seconds]
10:37
nojb has quit [Quit: nojb]
10:42
milosn has quit [Ping timeout: 246 seconds]
10:44
lanius777 has quit [Remote host closed the connection]
10:45
lanius777 has joined #ocaml
11:00
milosn has joined #ocaml
11:00
lanii has joined #ocaml
11:02
lanius777 has quit [Ping timeout: 272 seconds]
11:12
rgrinberg has joined #ocaml
11:17
rgrinberg has quit [Ping timeout: 250 seconds]
11:18
octachron has joined #ocaml
11:19
nojb has joined #ocaml
11:21
milosn has quit [Ping timeout: 245 seconds]
11:25
ygrek has quit [Ping timeout: 245 seconds]
11:25
_andre has joined #ocaml
11:33
flx has joined #ocaml
11:33
rand000 has joined #ocaml
11:34
flxx has quit [Ping timeout: 246 seconds]
11:38
ski has quit [Ping timeout: 251 seconds]
11:39
Drup has quit [Quit: Tatayoyo]
11:40
ski has joined #ocaml
11:41
Drup has joined #ocaml
11:46
jpdeplaix has quit [Ping timeout: 250 seconds]
11:47
araujo has quit [Ping timeout: 255 seconds]
11:47
araujo has joined #ocaml
11:47
milosn has joined #ocaml
11:53
kushal has joined #ocaml
12:08
sdothum has joined #ocaml
12:13
jpdeplaix has joined #ocaml
12:17
cesar_ has joined #ocaml
12:17
cesar_ is now known as Guest29253
12:19
Gama11 has joined #ocaml
12:21
BitPuffin has joined #ocaml
12:22
Guest29253 has quit [Ping timeout: 245 seconds]
12:28
seanmcl has joined #ocaml
12:32
kushal has quit [Ping timeout: 256 seconds]
12:37
mengu has joined #ocaml
12:38
yomimono has joined #ocaml
12:41
s1n4 has quit [Ping timeout: 272 seconds]
12:43
milosn has quit [Ping timeout: 264 seconds]
12:48
nojb has quit [Quit: nojb]
12:54
nojb has joined #ocaml
12:54
klaymen has joined #ocaml
12:55
milosn has joined #ocaml
12:58
rgrinberg has joined #ocaml
12:59
klaymen has left #ocaml [#ocaml]
13:02
rgrinberg has quit [Ping timeout: 240 seconds]
13:02
kushal has joined #ocaml
13:04
nojb has quit [Ping timeout: 264 seconds]
13:07
milosn has quit [Read error: Connection reset by peer]
13:09
kushal has quit [Ping timeout: 245 seconds]
13:09
idegen has joined #ocaml
13:13
milosn has joined #ocaml
13:19
nojb has joined #ocaml
13:33
Sorella has joined #ocaml
13:40
mengu has quit [Remote host closed the connection]
13:41
rand000 has quit [Quit: leaving]
13:43
larhat has quit [Quit: Leaving.]
13:45
larhat has joined #ocaml
13:48
mengu has joined #ocaml
13:49
agarie has joined #ocaml
13:51
mengu has quit [Client Quit]
13:56
milosn has quit [Ping timeout: 244 seconds]
14:01
yomimono has quit [Ping timeout: 248 seconds]
14:06
cesar_ has joined #ocaml
14:06
lanii has quit [Quit: Leaving]
14:06
cesar_ is now known as Guest10725
14:10
Guest10725 has quit [Ping timeout: 245 seconds]
14:15
nojb has quit [Quit: nojb]
14:15
rgrinberg has joined #ocaml
14:22
nojb has joined #ocaml
14:22
sdothum has joined #ocaml
14:24
nojb_ has joined #ocaml
14:26
nojb has quit [Ping timeout: 240 seconds]
14:26
nojb_ is now known as nojb
14:29
s1n4 has joined #ocaml
14:30
Haudegen has quit [Ping timeout: 272 seconds]
14:39
photex is now known as photex[away]
14:45
Haudegen has joined #ocaml
14:49
pyon is now known as combinatory-pyon
14:50
combinatory-pyon is now known as pyon-binator
14:51
kakadu has quit [Ping timeout: 246 seconds]
14:57
klaymen_ has joined #ocaml
14:59
kushal has joined #ocaml
15:02
ocp has quit [Quit: Leaving.]
15:06
kakadu has joined #ocaml
15:10
kdef has joined #ocaml
15:13
rand000 has joined #ocaml
15:16
milosn has joined #ocaml
15:22
Gama11_ has joined #ocaml
15:23
Gama11 has quit [Ping timeout: 245 seconds]
15:26
larhat has quit [Quit: Leaving.]
15:26
<
haesbaert >
is there a way to match on a variant _type_, not the value ?
15:26
<
haesbaert >
suppose: type dhcp_option = | Subnet_mask of Ipaddr.V4.t (* code 1
*) | Time_offset of Int32.t (* code 2 *)
15:28
<
haesbaert >
let me formulate this better, 1s
15:30
<
octachron >
GADT can do that
15:30
<
pippijn >
octachron: what can they do?
15:32
<
octachron >
oups, you are right, bad answer. More precisely, with a parameterized type 'a t, you can use GADT to "match" on the type of "'a"
15:32
<
dsheets >
Ipaddr.V4.Prefix.t may be a better type for a subnet mask...
15:33
<
haesbaert >
dsheets: indeed, I had not noticed, I'll use it
15:35
ygrek has joined #ocaml
15:35
ollehar1 has joined #ocaml
15:36
Algebr has joined #ocaml
15:37
<
haesbaert >
the code I wrote more or less does what I want, but I'd like to have it in a generic way
15:37
<
haesbaert >
since I have around 60-70 options
15:38
<
haesbaert >
basically I need to join the given dhcp_option with the pre-existing option, but only if the subtype is a list of things
15:39
seanmcl has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
15:39
<
Drup >
are these options alternatives or C-like flags ?
15:40
darkf has quit [Quit: Leaving]
15:40
s1n4 has quit [Quit: leaving]
15:40
<
haesbaert >
no, basically dhcp sends a bunch of options, each option has a format
15:41
<
haesbaert >
for instance, routers is variable sized multiple of 4, and each 4 bytes has a router
15:41
<
haesbaert >
so I encode it as a Routers of Ip.V4.t list
15:41
<
Drup >
shouldn't it be a record then ?
15:42
<
haesbaert >
I could have a record type for Routers, but the variant seemed enough
15:44
<
Drup >
oh, you want first class constructors
15:44
<
Algebr >
Drup: I have a problem in eliom where if I go to a service directly my js in the script tag is executed but if I open the html of the services reached via a tyxml a tag, the js code doesn't work
15:44
<
Drup >
no, you can't have that, sorry
15:45
<
Drup >
haesbaert: eta expand
15:45
<
Drup >
(fun x -> Result x)
15:45
agarie has quit [Remote host closed the connection]
15:45
<
Drup >
if you want a function
15:45
<
Drup >
but you can't match on it then
15:45
<
Drup >
Algebr: I need a code example
15:45
<
haesbaert >
Drup: not sure I follow
15:46
<
haesbaert >
you mean pass a function instead of an "opt"
15:46
<
haesbaert >
which basically does the "is_this_the_dude_I_want"
15:46
<
Drup >
I mean that "Result" by itself, without it's argument, is not a thing
15:46
<
haesbaert >
ah, indeed
15:46
<
haesbaert >
that's what I've just learned :/
15:47
<
Drup >
you may be able to do a gadt thing to do what you want
15:47
<
Algebr >
Drup: go to edgar.haus (Yes its a real domain name haha) and then to the different link. you'll see the twitter box doesn't load unless you refresh the page itself, as if you went to it initally rather than via the a tag.
15:47
<
Drup >
but it's not really more terse
15:48
<
Drup >
what is it ? :O
15:48
<
Algebr >
just a website to play around with
15:49
<
Algebr >
haus means house, so its my "house"?
15:49
<
haesbaert >
ack, I'll try to implement this in another way then
15:49
<
Drup >
I know it means house, I'm wondering how it managed to be a tld x)
15:50
<
Drup >
ok, the germans managed to get it among the new tlds
15:50
<
Algebr >
I think it said it was for german real estate
15:50
<
Drup >
.maison is available too!
15:51
<
Drup >
sorry, I got diverted :D
15:52
<
Drup >
(drup.maison is available but expansive)
15:53
<
rgrinberg >
Drup: submit something on cumulus :D
15:53
seanmcl has joined #ocaml
15:54
<
rgrinberg >
it's running cohttp now
15:54
<
Drup >
I actually don't use cumulus and don't have an account, do it :p
15:55
cesar_ has joined #ocaml
15:55
cesar_ is now known as Guest46624
15:55
<
adrien_znc >
slacker
15:56
uris77 has joined #ocaml
15:56
<
rgrinberg >
made an account but don't really have anything to submit
15:56
<
rgrinberg >
upvoting works
15:57
nullcat has joined #ocaml
15:58
kushal has quit [Quit: Leaving]
15:59
<
Algebr >
Can merlin have compiler flags turned on, like -safe-string
15:59
<
Drup >
FLG -safe-string
15:59
ygrek has quit [Ping timeout: 244 seconds]
15:59
Guest46624 has quit [Ping timeout: 245 seconds]
16:06
<
olibjerd >
Labltk supports ttk themes?
16:07
<
Algebr >
Drup: does utop have a corresponding compiler flag turn on mechanism?
16:07
<
Drup >
"utop -safe-string"
16:07
<
Drup >
(like the ocaml top level)
16:07
<
adrien_znc >
olibjerd: not sure at all
16:07
<
adrien_znc >
try to check the caml-list
16:07
<
adrien_znc >
I think the question has already been raised
16:07
<
Drup >
Algebr: are you loading the script element only on this specific page ?
16:07
<
adrien_znc >
but I haven't followed labltk recently
16:07
<
Drup >
and using an App thingy for registration ?
16:08
<
Algebr >
Drup: every page is getting it, the script
16:09
<
adrien_znc >
0 upgraded, 45 newly installed, 261 downgraded, 44 to remove and 0 not upgraded.
16:09
<
adrien_znc >
Need to get 13.5 MB/166 MB of archives.
16:09
<
Drup >
all of them ?
16:09
<
Algebr >
What do you mean app thingy for registartion? Do you mean using the functor for the app instead of Eliom_registration?
16:09
<
adrien_znc >
woops, ECHAN
16:10
<
Drup >
yes, I mean that
16:10
<
Drup >
(well, it's not really "instead")
16:10
<
Algebr >
what do you mean
16:10
milosn has quit [Quit: Lost terminal]
16:11
milosn has joined #ocaml
16:11
<
Drup >
are you using ELiom_registration.App ?
16:12
MrScout has quit [Ping timeout: 245 seconds]
16:16
icicled` has quit [Read error: Connection reset by peer]
16:16
octachron has quit [Quit: Leaving]
16:17
nullcat has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
16:19
nojb has quit [Quit: nojb]
16:25
shinnya has joined #ocaml
16:25
johnelse is now known as johnel_away
16:27
johnel_away is now known as johnelse
16:33
kushal has joined #ocaml
16:33
kushal has quit [Changing host]
16:33
kushal has joined #ocaml
16:34
badkins has joined #ocaml
16:36
ollehar1 has quit [Ping timeout: 252 seconds]
16:36
lordkryss has joined #ocaml
16:41
freling has quit [Quit: Leaving.]
16:42
agarie has joined #ocaml
16:44
fraggle_ has quit [Remote host closed the connection]
16:44
cmtptr has quit [Ping timeout: 248 seconds]
16:47
shinnya has quit [Ping timeout: 250 seconds]
16:48
fraggle_ has joined #ocaml
16:51
cmtptr has joined #ocaml
16:54
Submarine has joined #ocaml
16:56
kakadu has quit [Quit: Page closed]
17:03
struk|work has joined #ocaml
17:03
dsheets has quit [Ping timeout: 256 seconds]
17:13
MrScout has joined #ocaml
17:21
martintrojer has joined #ocaml
17:27
<
Drup >
Algebr: ok looking
17:29
<
Drup >
load all the scripts everywhere.
17:29
freling has joined #ocaml
17:33
dsheets has joined #ocaml
17:36
<
Drup >
hum, a script in the middle of the page like that
17:36
<
Drup >
did you tried to put it back at the head level ?
17:37
<
Algebr >
no, but that shouldn't matter.
17:37
<
Algebr >
because the code does load when refreshing the page but not when coming to it via a service.
17:37
<
Drup >
emphasis on "shouldn't" :D
17:37
seanmcl has quit [Ping timeout: 272 seconds]
17:37
<
Drup >
yes, but that's ... to be expected
17:37
<
Algebr >
so this is a real bug?
17:37
<
Drup >
eliom doesn't really refresh the page when you are navigating inside the same app
17:38
<
Algebr >
what? its not like its changing the dom either, its providing a new html document
17:38
<
Drup >
it AJAXs the content of the new page and replace the page without restarting the javascript
17:38
seanmcl has joined #ocaml
17:40
<
Drup >
and the behavior when there is a script element on some pages but not others can lead to the bug you are describing.
17:40
<
Algebr >
this is undesirable
17:41
<
Algebr >
So what can I do?
17:41
<
Drup >
There are two workarounds: make separated applications or load the scripts everywhere
17:41
<
Drup >
I would personally load everywhere with the async argument
17:42
<
Algebr >
async argument of what
17:43
yomimono has joined #ocaml
17:43
<
Drup >
oh, nevermind, it's only for non async scripts
17:44
cesar_ has joined #ocaml
17:44
cesar_ is now known as Guest31346
17:44
<
Algebr >
So that is not a solution?
17:46
pyon-binator is now known as dysfunc-pyon-al
17:47
<
Drup >
Algebr: try to put the script at the head, tell me how it goes
17:48
Guest31346 has quit [Ping timeout: 245 seconds]
17:48
<
Algebr >
I don't think I can put the script in the head because it depends on the html element to exist
17:49
<
Algebr >
which won't exist yet...
17:49
<
Drup >
then put it under the body.
17:50
<
Algebr >
I don't see how that is any different than I'm doing now?
17:51
<
Drup >
scripts are treated differently when really inside the html or at the root
17:52
<
Algebr >
so do you mean put it after the body or right before?
17:52
<
Algebr >
after right?
17:53
<
Drup >
(eliom should really manage this things for you)
17:54
<
Algebr >
Drup: Does eliom have soemthing that can take a .markdown and produce the appropriate list of divs/ps, etc
17:54
<
Drup >
eliom doesn't
17:54
<
Drup >
but lambdoc can do that
17:56
jwatzman|work has joined #ocaml
17:57
<
Drup >
it takes various input and returns tyxml stuff
17:57
<
Algebr >
right, so I can just call it and dump the results right into a page, right
17:57
<
Algebr >
I want to do that for like blog posts
17:58
<
Drup >
there are good examples in lambdoc that do exactly that
17:58
<
Drup >
the boilerplate is a bit involved, but after that, it works mostly by itself
17:58
<
Drup >
(it's involved because multiple functor applications :p)
18:00
<
Algebr >
I'm fine with investing time for something worthwhile.
18:01
<
Drup >
note that lambdoc doesn't fully support markdown just yet. It has an internal better supported format that is a bit tex-like
18:01
<
Drup >
(but less insane)
18:01
<
Drup >
(that's not difficult)
18:01
<
Algebr >
actually the best thing would be taking a .org file
18:02
<
Drup >
you can contribute that, I guess
18:04
yomimono has quit [Ping timeout: 244 seconds]
18:04
MrScout has quit [Remote host closed the connection]
18:04
yomimono has joined #ocaml
18:04
MrScout has joined #ocaml
18:05
Haudegen has quit [Ping timeout: 244 seconds]
18:09
yomimono has quit [Ping timeout: 250 seconds]
18:09
yomimono has joined #ocaml
18:18
Haudegen has joined #ocaml
18:23
Anarchos has joined #ocaml
18:25
yomimono has quit [Ping timeout: 250 seconds]
18:34
mort___ has joined #ocaml
18:35
mcclurmc_ has joined #ocaml
18:37
mcclurmc has quit [Ping timeout: 250 seconds]
18:38
idegen has quit [Ping timeout: 250 seconds]
18:44
c355E3B has joined #ocaml
18:47
Algebr has quit [Ping timeout: 250 seconds]
18:53
idegen has joined #ocaml
18:57
agarie has quit [Remote host closed the connection]
19:02
Algebr has joined #ocaml
19:02
agarie has joined #ocaml
19:05
mort___ has quit [Ping timeout: 250 seconds]
19:12
Denommus has joined #ocaml
19:12
BitPuffin has quit [Ping timeout: 265 seconds]
19:14
seanmcl has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
19:14
uris77 has quit [Quit: leaving]
19:15
agarie has quit [Remote host closed the connection]
19:16
agarie has joined #ocaml
19:17
larhat has joined #ocaml
19:18
MisaMisa has joined #ocaml
19:25
ericwa has joined #ocaml
19:25
Anarchos has quit [Quit: Vision[0.9.7-H-20140108]: i've been blurred!]
19:26
Denommus has quit [Read error: Connection reset by peer]
19:27
Denommus has joined #ocaml
19:29
mearnsh has quit [Ping timeout: 272 seconds]
19:31
mearnsh has joined #ocaml
19:31
Denommus has quit [Read error: Connection reset by peer]
19:32
Denommus` has joined #ocaml
19:33
cesar_ has joined #ocaml
19:33
cesar_ is now known as Guest17363
19:34
MisaMisa has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
19:35
MisaMisa has joined #ocaml
19:35
rgrinberg has quit [Ping timeout: 256 seconds]
19:37
Guest17363 has quit [Ping timeout: 245 seconds]
19:39
yomimono has joined #ocaml
19:42
rgrinberg has joined #ocaml
19:45
Submarine has quit [Quit: Leaving]
19:50
claudiuc has joined #ocaml
19:56
Denommus` has quit [Read error: Connection reset by peer]
19:56
Denommus` has joined #ocaml
20:01
Denommus` is now known as Denommus
20:03
kakadu has joined #ocaml
20:14
kushal has quit [Ping timeout: 244 seconds]
20:15
c355E3B has quit [Quit: Leaving]
20:17
seanmcl has joined #ocaml
20:19
WraithM has joined #ocaml
20:20
mengu has joined #ocaml
20:27
Denommus` has joined #ocaml
20:30
Denommus has quit [Ping timeout: 264 seconds]
20:32
pgomes has joined #ocaml
20:32
kushal has joined #ocaml
20:37
axiles has quit [Remote host closed the connection]
20:38
ontologiae has joined #ocaml
20:40
Nijikokun has joined #ocaml
20:42
Denommus` is now known as Denommus
20:45
agarie has quit [Remote host closed the connection]
20:48
oscar_toro has quit [Quit: Lost terminal]
20:56
Denommus` has joined #ocaml
20:58
agarie has joined #ocaml
21:00
Denommus has quit [Ping timeout: 248 seconds]
21:03
freling has quit [Quit: Leaving.]
21:05
Algebr has quit [Ping timeout: 250 seconds]
21:08
pgomes has quit [Quit: Leaving]
21:13
WraithM has quit [Quit: leaving]
21:16
mengu has quit [Remote host closed the connection]
21:19
MercurialAlchemi has quit [Ping timeout: 250 seconds]
21:19
c355E3B has joined #ocaml
21:21
cesar_ has joined #ocaml
21:22
cesar_ is now known as Guest87200
21:25
ollehar1 has joined #ocaml
21:26
Guest87200 has quit [Ping timeout: 245 seconds]
21:26
swgillespie has joined #ocaml
21:27
<
ollehar1 >
strip core binary?
21:27
<
ollehar1 >
google fail me
21:28
rand000 has quit [Quit: leaving]
21:28
lvgx has joined #ocaml
21:29
Denommus` has quit [Ping timeout: 246 seconds]
21:35
Algebr has joined #ocaml
21:41
dysfunc-pyon-al is now known as idem-pyon-tent
21:42
Gama11_ has quit [Read error: Connection reset by peer]
21:43
Jefffrey has joined #ocaml
21:45
seanmcl has quit [Ping timeout: 248 seconds]
21:48
justinjaffray has quit [Remote host closed the connection]
21:49
shum has joined #ocaml
21:53
Haudegen has quit [Ping timeout: 240 seconds]
22:04
Simn has quit [Quit: Leaving]
22:10
ollehar has quit [Ping timeout: 240 seconds]
22:11
ollehar has joined #ocaml
22:14
yomimono_ has joined #ocaml
22:15
Haudegen has joined #ocaml
22:17
yomimono has quit [Ping timeout: 246 seconds]
22:20
shinnya has joined #ocaml
22:23
kushal has quit [Ping timeout: 256 seconds]
22:25
oscar_toro has joined #ocaml
22:26
mengu has joined #ocaml
22:29
tane has quit [Remote host closed the connection]
22:31
mengu has quit [Ping timeout: 248 seconds]
22:32
Algebr has quit [Ping timeout: 245 seconds]
22:34
kakadu has quit [Remote host closed the connection]
22:34
nojb has joined #ocaml
22:43
mxv has joined #ocaml
22:46
ousado_ has joined #ocaml
22:47
ousado has quit [Ping timeout: 250 seconds]
22:47
ousado_ is now known as ousado
22:48
ousado has quit [Changing host]
22:48
ousado has joined #ocaml
22:49
WraithM has joined #ocaml
22:49
WraithM has quit [Client Quit]
23:04
mxv has quit [Remote host closed the connection]
23:08
Nijikokun has quit [Ping timeout: 245 seconds]
23:10
ontologiae has quit [Ping timeout: 264 seconds]
23:10
cesar_ has joined #ocaml
23:11
cesar_ is now known as Guest57597
23:13
ollehar1 has quit [Remote host closed the connection]
23:15
Guest57597 has quit [Ping timeout: 245 seconds]
23:15
Nijikokun has joined #ocaml
23:16
Jefffrey has quit [Quit: Leaving...]
23:19
BitPuffin has joined #ocaml
23:26
c355E3B has quit [Quit: Leaving]
23:28
yaewa has joined #ocaml
23:29
moei has quit [Ping timeout: 252 seconds]
23:30
olibjerd has quit [Ping timeout: 246 seconds]
23:33
pancake has joined #ocaml
23:33
<
pancake >
how can i get the Unix.file_descr from a string?
23:33
<
nojb >
pancake: what do you mean ?
23:33
<
pancake >
i'm doing external file_descr_of_string : string -> Unix.file_descr = "%identity";; but then Unix.write fd_out "foo" 0 3 fails
23:33
<
pancake >
with this error: Exception: Unix.Unix_error (EBADF, "write", "").
23:34
<
Drup >
that seems like a terrible idea
23:34
madroach has quit [Ping timeout: 264 seconds]
23:34
<
pancake >
im passing two filedescriptor numbers by env var to a child process
23:35
<
pancake >
i want to read/write on those fd, taking the file_descr from the Sys.getenv
23:35
<
Drup >
then why string ?
23:35
<
Drup >
(at least, int would make half sense, in a hacky kind of way)
23:35
<
pancake >
env vars are strings
23:35
madroach has joined #ocaml
23:36
<
pancake >
i can make another intermediate step
23:37
<
Drup >
(note: you shouldn't use external and %identity if you don't know what they do ;))
23:37
<
pancake >
im trying to learn by trying random statements
23:38
<
nojb >
I doubt you will learn much this way :(
23:39
<
Drup >
stop trying to transform a string in a file_descr, it will not work. I'm surprised it's not segfaulting x)
23:40
<
nojb >
it is not because the string points to valid memory, the first few bytes are interpreted as an invalid file_descr and the '
23:40
<
nojb >
write' call fails
23:42
<
pancake >
how can i convert int ->Unix.file_descr ?
23:42
<
pancake >
or fdopen()
23:43
<
nojb >
there is no way to do it in OCaml
23:43
<
Drup >
with the stdlib, you can't
23:44
<
dmbaturin >
That's the point of abstract types.
23:46
kdef has quit [Quit: Leaving]
23:47
shinnya has quit [Ping timeout: 252 seconds]
23:48
<
mrvn >
You also can't get an open fd as int. So there is no need for that conversion.
23:49
<
dmbaturin >
The idea is that clients shouldn't call functions that work with file descriptors with random stuff instead of file descriptors.
23:49
badon has joined #ocaml
23:52
ollehar has quit [Read error: Connection reset by peer]
23:52
ollehar1 has joined #ocaml
23:55
nojb has quit [Quit: nojb]
23:57
rgrinberg has quit [Ping timeout: 265 seconds]