NoNNaN has quit [Remote host closed the connection]
avsm has quit [Quit: Leaving.]
NoNNaN has joined #ocaml
manud has quit [Quit: manud]
BitPuffin has joined #ocaml
parcs has left #ocaml ["WeeChat 1.1-dev"]
struktured has quit [Ping timeout: 244 seconds]
claudiuc has quit [Remote host closed the connection]
hausdorff has quit [Remote host closed the connection]
zpe has joined #ocaml
hausdorff has joined #ocaml
BitPuffin has quit [Ping timeout: 255 seconds]
zpe has quit [Ping timeout: 272 seconds]
jwatzman|work has quit [Quit: jwatzman|work]
manud has joined #ocaml
hausdorff has quit [Remote host closed the connection]
robink has quit [Ping timeout: 258 seconds]
jao has joined #ocaml
jao has quit [Changing host]
jao has joined #ocaml
robink has joined #ocaml
BitPuffin has joined #ocaml
struktured has joined #ocaml
ontologiae has joined #ocaml
manud has quit [Quit: manud]
BitPuffin has quit [Remote host closed the connection]
manud has joined #ocaml
BitPuffin has joined #ocaml
serge has joined #ocaml
thmslld has quit [Ping timeout: 260 seconds]
manud has quit [Quit: manud]
manud has joined #ocaml
serge has quit [Ping timeout: 265 seconds]
zpe has joined #ocaml
badkins has quit []
zpe has quit [Ping timeout: 244 seconds]
BitPuffin has quit [Ping timeout: 250 seconds]
araujo has quit [Read error: Connection reset by peer]
araujo has joined #ocaml
serge has joined #ocaml
zz_anildigital is now known as adgtl
adgtl is now known as anildigital
anildigital is now known as adgtl
zpe has joined #ocaml
zpe has quit [Ping timeout: 265 seconds]
lemongrabjo has joined #ocaml
taion809 has quit [Remote host closed the connection]
manud has quit [Quit: manud]
samrat has joined #ocaml
manud has joined #ocaml
siddharthv_away is now known as siddharthv
bytbox has quit [Remote host closed the connection]
manud has quit [Quit: manud]
ljs has quit [Quit: None]
adgtl is now known as zz_adgtl
ygrek has joined #ocaml
zpe has joined #ocaml
larhat has joined #ocaml
zpe has quit [Ping timeout: 244 seconds]
WraithM has joined #ocaml
zz_adgtl is now known as adgtl
adgtl is now known as anildigital
mcc has joined #ocaml
samrat has quit [Ping timeout: 258 seconds]
WraithM has quit [Ping timeout: 265 seconds]
dzhulk has joined #ocaml
larhat has quit [Quit: Leaving.]
WraithM has joined #ocaml
anildigital is now known as zz_anildigital
<mcc>
Hi… I wanna do some simple lexer / token-parser stuff, for reasons I think I wanna do it in ocaml, I haven't written any ML in years...
<whitequark>
sedlex as lexer and menhir as LALR(1) parser if you prefer those
<mcc>
Where is a sensible place to start with doing tokenization / parsing / string manip like this in ocaml? I'm looking at this tutorial in writing a language with LLVM on llvm.org and it is using camlp4/camlp5... that seems like a real big thing tho
<whitequark>
please don't use camlp4 and especially not camlp5
<whitequark>
I will be rewriting that tutorial in close future because it has really really bad advice
<whitequark>
(you're not the first one to be burned by that)
<mcc>
haha thank you for the advice
<mcc>
sedlex is a lexer or a lexer generator?
<whitequark>
it's a lexer syntax extension
<mcc>
ok.
<whitequark>
essentially, it's a lexer generator, but you don't see the generated code
<mcc>
heh i'm looking at this and going "i should show this to misty--" and then she's in the thread
<mcc>
hmm. i'm a little confused by this.
<mcc>
i'm installing opam right now via brew, actually
<mcc>
should i back it out and wait for this to go in?
dzhulk has quit [Quit: Leaving.]
<whitequark>
let me see
<mcc>
cuz this looks like something that will be resolved in short order.
<whitequark>
yes, probably best to wait for a bit.
<whitequark>
camlp4 is deprecated and the ecosystem is slowly migrating from it--I myself sent three or four PRs making camlp4 optional just previous week--but it's not happening quickly enough.
<whitequark>
so a lot of things still depend on it and break in sad ways.
<whitequark>
it's a bit complicated because camlp4 can refer to two things at once: 1) it is a generic extensible parser technology, 2) it also was/is used to extend OCaml syntax
<mcc>
i— oh, god *damn* it
<mcc>
==> Installing dependencies for opam: boost, cmake, re2c, scons, bison, g
<mcc>
i see.
<whitequark>
so sedlex is not a camlp4-based syntax extension
<whitequark>
but if you really want, you can still use it as a lexer for camlp4-based parsers
<whitequark>
(I'm not sure why would anyone want that)
<mcc>
what i mean is, you're suggesting i need this patch to support camlp4 in opam, but i don't see anything on the github page suggesting camlp4 is in fact a dependency?
<whitequark>
oh, it's not a dependency of sedlex
<mcc>
you just think i'll want it overall?
<whitequark>
it is a dependency of, for example, utop, which is a REPL which is usable
<mcc>
i see.
<whitequark>
transitively via a bunch of commonly used libraries.
<whitequark>
I've removed camlp4 dep from three of them, but for the fourth it's a 2kloc diff that's pending review (https://github.com/ocsigen/lwt/pull/95)
<whitequark>
mcc: btw, I'm not sure why opam repends on re2c, that sounds odd to me
ontologiae has quit [Ping timeout: 272 seconds]
<whitequark>
or rather, I look at its source and it doesn't, this seems like a formula bug to me
<mcc>
brew is often aggressive on dependencies. si it possible one of those other dependencies requires it? idk what re2c even is
<whitequark>
google's regexp library
<mcc>
also: i typed ocaml -help and saw this in one place:
<mcc>
-drawlambda (undocumented)
<mcc>
wtf
<whitequark>
-d* switches (from -debug) show the compiler's internal representaiton at various phases
<whitequark>
so, textual source, AST, typed AST, IR, uhhh some other form of IR, bytecode (instr)
<whitequark>
lambda is an untyped representation that looks sort of like lambda calculus, hence the name
<mcc>
i am so terrified about adding any dependencies to my project at this point
<whitequark>
why?
<mcc>
cuz everything gets these big trees and so i'm afraid i'll have a situation where i use sedlex so to build my thing people have to get re2c because it's a dependency of a dependency of the package manager you need to download one of my dependencies
tnguyen has joined #ocaml
<mcc>
i mean, i expect it'll be pretty hard to build anything without using opam, but this is a general problem
<whitequark>
I agree that this is an issue.
<mcc>
i tried playing with that clasp thing, i never even got it to build
<whitequark>
it's less pronounced on *nix systems, where you generally have binary packages for aspcud
<whitequark>
there was some work to provide aspcud-as-a-service
samrat has joined #ocaml
tnguyen has quit [Client Quit]
tnguyen has joined #ocaml
<mcc>
urrrg i'm really looking forward to getting the linux laptop i'm planning to get this winter
Submarine has quit [Remote host closed the connection]
raichoo has quit [Ping timeout: 265 seconds]
badon has joined #ocaml
pgomes has joined #ocaml
samrat has joined #ocaml
sgnb has joined #ocaml
cago has joined #ocaml
huza has joined #ocaml
raichoo has joined #ocaml
dsheets has joined #ocaml
Guest39355 is now known as Haudegen
ikaros has joined #ocaml
huza has quit [Ping timeout: 255 seconds]
AlexRussia has joined #ocaml
ddosia has quit [Quit: Leaving.]
AlexRussia has quit [Ping timeout: 256 seconds]
raichoo has quit [Ping timeout: 258 seconds]
AlexRussia has joined #ocaml
samrat has quit [Read error: Connection reset by peer]
hooplahoops has joined #ocaml
AlexRussia has quit [Ping timeout: 250 seconds]
ddosia has joined #ocaml
rand000 has joined #ocaml
_andre has joined #ocaml
avsm has joined #ocaml
huza has joined #ocaml
serge has quit [Remote host closed the connection]
serge has joined #ocaml
serge has quit [Ping timeout: 244 seconds]
zpe has quit [Remote host closed the connection]
raichoo has joined #ocaml
bezirg has joined #ocaml
huza has quit [Quit: WeeChat 0.3.8]
zpe has joined #ocaml
samrat has joined #ocaml
sinelaw has quit [Ping timeout: 245 seconds]
toolslive has joined #ocaml
<toolslive>
am I correct to observe that opam 1.2 can only use ${HOME}/.opam ? (no longer arbitrary directories)
hooplahoops has quit [Ping timeout: 265 seconds]
<dsheets>
toolslive, root?
<flux>
I have opam 1.2 (well some prerelease) and I don't have ~/.opam
<flux>
I do have environment variable OPAMROOT exported to point elsewhere
<toolslive>
right. I glanced over that option....
<toolslive>
I'm in a total mess: I have an opam 1.1 root and I cannot install ppx_deriving because of ppx_tools constraints.... upgrade only works half because some packages use the 1.2 definition format and my 1.1 cannot parse that.
<flux>
start from scratch, not that big a deal 8-)
<flux>
with some sed magic you can extract the list of installed packages and reinstall them with another opam..
<flux>
or maybe there's even an option allowing export/import
<MercurialAlchemi>
well, yeah, you can export your universe
<toolslive>
ok. starting from scratch...
nicoo has quit [Remote host closed the connection]
nicoo has joined #ocaml
jpdeplaix has quit [Ping timeout: 250 seconds]
ddosia has quit [Ping timeout: 256 seconds]
raichoo has quit [Quit: Leaving.]
sinelaw has joined #ocaml
serge has joined #ocaml
tupelolo is now known as shallow
jpdeplaix has joined #ocaml
<toolslive>
ok. most things work.... but ppx_deriving seems to falter somewhere...'Cannot locate deriver Show' prolly an API change...
_5kg has quit [Ping timeout: 265 seconds]
hooplahoops has joined #ocaml
robink has quit [Ping timeout: 244 seconds]
pharpend has quit [Quit: WeeChat 0.4.3]
robink has joined #ocaml
Thooms has quit [Quit: WeeChat 1.0.1]
ddosia has joined #ocaml
oscar_toro has quit [Ping timeout: 250 seconds]
pharpend has joined #ocaml
arj has quit [Quit: Leaving.]
kakadu has joined #ocaml
oscar_toro has joined #ocaml
_5kg has joined #ocaml
huza has joined #ocaml
bytbox has joined #ocaml
jonludlam has joined #ocaml
tnguyen has joined #ocaml
pgomes has left #ocaml [#ocaml]
hooplahoops has quit [Ping timeout: 245 seconds]
sinelaw has quit [Ping timeout: 245 seconds]
arj has joined #ocaml
tnguyen has quit [Quit: tnguyen]
tnguyen has joined #ocaml
oscar_toro has quit [Ping timeout: 264 seconds]
avsm has quit [Quit: Leaving.]
sepp2k has joined #ocaml
tnguyen has quit [Ping timeout: 260 seconds]
eizodo has joined #ocaml
<toolslive>
still some things completely wrong: "opam list | grep lwt yields": "lwt 2.4.6 A cooperative threads library for OCaml" but: "ocamlfind list | grep lwt" yields ""
Thooms has joined #ocaml
<def`>
toolslive: eval `opam config env`
struktured has joined #ocaml
<def`>
try evaluating this in your shell
<toolslive>
it's not that..... I installed lwt.2.4.5 and now ocamlfind finds the packages
<Drup>
flux: "opam switch export/import"
tnguyen has joined #ocaml
bytbox has quit [Remote host closed the connection]
badkins has joined #ocaml
tnguyen has quit [Ping timeout: 245 seconds]
<flux>
drup, nice
<flux>
(though I didn't expect it to be under switch)
<flux>
last time I did it with some opam list | sed
AlexRussia has joined #ocaml
shinnya has joined #ocaml
huza has quit [Quit: WeeChat 0.3.8]
sinelaw has joined #ocaml
eizodo has quit [Ping timeout: 246 seconds]
<toolslive>
"|xargs opam install "
mcc has quit [Quit: This computer has gone to sleep]
struktured has quit [Ping timeout: 265 seconds]
Thooms has quit [Ping timeout: 255 seconds]
siddharthv has quit [Ping timeout: 245 seconds]
siddharthv has joined #ocaml
sinelaw has quit [Ping timeout: 258 seconds]
siddharthv is now known as siddharthv_away
<flux>
toolslive, well, there was a file in between :)
darkf has quit [Quit: Leaving]
<toolslive>
yes, I realised that after I wrote it....
eizodo has joined #ocaml
tnguyen has joined #ocaml
toolslive has quit [Ping timeout: 260 seconds]
tnguyen has quit [Quit: tnguyen]
tnguyen has joined #ocaml
SethTIsue has joined #ocaml
dsheets has quit [Ping timeout: 258 seconds]
marynate has quit [Read error: Connection reset by peer]
<Drup>
whitequark: how do I run only the ocaml tests ?
dsheets has joined #ocaml
nlucaroni has joined #ocaml
sinelaw has joined #ocaml
rossberg has joined #ocaml
dsheets has quit [Ping timeout: 258 seconds]
lordkryss has joined #ocaml
avsm has joined #ocaml
toolslive has quit [Ping timeout: 260 seconds]
thmslld has joined #ocaml
BitPuffin has quit [Read error: Connection reset by peer]
dsheets has joined #ocaml
BitPuffin has joined #ocaml
toolslive has joined #ocaml
mort___ has quit [Quit: Leaving.]
chris2_ has joined #ocaml
chris2_ is now known as chris2
oscar_toro has joined #ocaml
<whitequark>
Drup: make lit TESTSUITE=Bindings/Ocaml
<whitequark>
the boolean is something like "zero-extend"
<Drup>
yeah, I noticed
<whitequark>
put a label on it and I will accept the patch
<Drup>
can I put an optional label on it ? =')
shinnya has quit [Ping timeout: 245 seconds]
<Drup>
(I'll have to move it in the arguments, so breaking compat)
SethTIsue has quit [Quit: SethTIsue]
<whitequark>
yes
<Drup>
fabulous.
<reynir>
Hm, I had this bug when doing "opam switch install 4.02.1" with -j 0 http://lpaste.net/113327
<reynir>
worked fine without
<whitequark>
please report it
yomimono has joined #ocaml
hooplahoops has joined #ocaml
ygrek has joined #ocaml
arj has quit [Quit: Leaving.]
ysz has joined #ocaml
<Drup>
whitequark: are you *sure* to test a llvmbool, it's "if (b)" and not supidly reversed or something ?
olauzon has joined #ocaml
ikaros has quit [Quit: Ex-Chat]
<whitequark>
it's not stupidly reversed, you just need to look at what exactly true or false means in that context
<Drup>
ARG, nevermind
mcc has joined #ocaml
<mcc>
whitequark: ok are you there?
sinelaw has quit [Ping timeout: 244 seconds]
<reynir>
Done, thanks
<reynir>
opam with -j 0 is supposed to use as many threads as there are cores, right?
<Drup>
whitequark: so, yes, it was reverse
<mcc>
whitequark: i'm gonna be honest i'm looking at ppx_deriving and i do not understand *what* it does.
<Drup>
or llvm is calling a boolean "losesInfo" which is true only if you don't lose any info.
<Drup>
=')
jako has joined #ocaml
jako has left #ocaml [#ocaml]
NoNNaN has quit [Remote host closed the connection]
<reynir>
opam help doesn't exactly say that, dunno how I got that idea...
<Drup>
mcc: to be honest, you shouldn't start by that
<Drup>
get back to ocaml slowly, you'll need it later, it's not important.
<toolslive>
I just upgraded to ppx_deriving 1.0 and I have some issues with this change: "Remove Findlib+dynlink integration. All derivers must now be explicitly required."
<whitequark>
mcc: here
<toolslive>
what exactly does "explicitly required" entail?
<Drup>
toolslive: you need to require ppx_deriving.show if you want the Show deriver
<whitequark>
toolslive: before 1.0, ppx_deriving would automagically locate and load deriver plugins based on the name you supply
<whitequark>
this was removed, so now you need to say beforehand which derivers to you want to use
<whitequark>
by requiring the corresponding package
<toolslive>
I understand that but I have troubles with making ocamlbuild understand
<whitequark>
what derivers do you use right now?
<toolslive>
well, show and enum
ysz has quit [Ping timeout: 245 seconds]
<whitequark>
ok, you need to replace package(ppx_deriving) with package(ppx_deriving.std)
<toolslive>
that's it?
<mcc>
drup: haha, for context, whitequark out of band suggested i bookmark it for later
<mcc>
but, yes, good advice
<whitequark>
toolslive: yep
<toolslive>
well, your project's ocamlbuild.ml suggests otherwise....
<whitequark>
toolslive: ppx_deriving's ocamlbuild itself? it cannot use -package, because it itself provides that package
<toolslive>
now I understand it's different for when you're inside the project....
<whitequark>
myocamlbuild.ml rather
<jpdeplaix>
whitequark: by the way, did you take a look at ollvm ?
<whitequark>
jpdeplaix: it's the thing that generates textual IR, right?
<whitequark>
this is explicitly discouraged by the LLVM docs.
<whitequark>
you can infer how high an opinion I have about it
<toolslive>
@whitequark you might want to add some line in the documentation somewhere as you will not be around to help out all the time....
<toolslive>
meanwhile, it works....
<whitequark>
toolslive: I wrote that in README and in the caml-list mail
ysz has joined #ocaml
<whitequark>
well, that being said, you're not the first with this issue. perhaps I was unclear
<toolslive>
"For every plugin, you need to require the corresponding package, e.g. ppx_deriving.show for the show deriver."
<toolslive>
problem is that "require" is not clear
<jpdeplaix>
whitequark: no, it aims to be « An higher-level LLVM binding for OCaml » It doesn't generates textual IR
<whitequark>
toolslive: oh I see! I will expand the README on that part
<toolslive>
people get wrongfooted with the other advise: ocamlfind c -ppx 'ocamlfind ppx_deriving/ppx_deriving src/ppx_deriving_foo.cma'
<toolslive>
I was already chaning the dispatch in myocamlbuild.ml
<whitequark>
it's broken, broken, broken. (* FIXME: support metadata strings and struct. Parsed as identifier here. *)
<whitequark>
the lexer is entirely wrong
<whitequark>
the parser is also broken and super outdated
<whitequark>
this is like the poster child for the reason why you should not parse textual IR
<whitequark>
it's true that existing bindings don't allow to read back the IR in its entirety, as Drup discovered, but the right approach is to fix the bindings
<jpdeplaix>
yes, but you can also only use the Llvm_ez module
<whitequark>
Llvm_ast and Llvm_ez should be imported inside the bindings
<whitequark>
well, an equivalent of
<Drup>
tbh, an ast is useless for llvm's IR
<Drup>
it's not a tree.
<whitequark>
Drup: not sure, haskell's LLVM-General uses it alright
<whitequark>
oh also, I have like a ton of patches that bring in-tree almost all functions that LLVM-General defines privately
<whitequark>
they would allow to read the entire IR
<Drup>
whitequark: they're building, not reading
<Drup>
to build, sure, use anything, it almost doesn't matter
<whitequark>
Drup: bidirectional
avsm has quit [Quit: Leaving.]
tane has joined #ocaml
<Drup>
anyway, I don't understand what's happening in APFloat.convert
<whitequark>
what's the issue?
<Drup>
it always returns the same boolean, both with 2 and 1e400
<hooplahoops>
flux: yeah, that's a really cool feature
<Drup>
yeah, it was that *and* something else
<Drup>
this is soo annoying
<hooplahoops>
flux: so a normal functor, when I instantiate it twice with the same argument module, it will return the same module?
<hooplahoops>
flux: does it have a cache for that at runtime?
<hooplahoops>
flux: because I can instantiate a functor F with module M from two different modules A and B
<hooplahoops>
flux: and I'd still get the same module returned at runtime, with the same reference cell (in the example)?
<whitequark>
I think generative functors are only about types
<hooplahoops>
Oh I see, so it'd have the same type, but different cells then
<Drup>
hooplahoops: it's not the same module, it's the same type
<hooplahoops>
Right, ok
<Drup>
(well, maybe it's the same module, but it doesn't really matter)
toolslive has quit [Ping timeout: 260 seconds]
<hooplahoops>
Well, it matters since I'd get different reference cells, or the same one
zpe has joined #ocaml
<Drup>
yes, that's my point, does it matter ? :)
toolslive has joined #ocaml
<Drup>
you're basically never using the physical equality on this things, and you shouldn't, it's subject to compiler optimisations
elfring has joined #ocaml
SethTIsue has joined #ocaml
koderok has joined #ocaml
<hooplahoops>
Hm, what I mean is that if I get the same module, it breaks the properties of the module since the cell would be shared unexpectedly
<Drup>
which properties ?
<hooplahoops>
Well, let's say that 't' was exposed, and the cell stores the last value I write into it, and I can query it
<hooplahoops>
I wouldn't want M1's writes to be shared with M2's writes
<Drup>
ah, you mean statefull modules
<hooplahoops>
Yeah :)
<Drup>
or modules with a state
<Drup>
yes, you should use generator functors precisely for those.
<Drup>
generative*
<hooplahoops>
As in, take an additional (dummy) value argument, right?
<Drup>
no, there is a special syntax
<Drup>
which is add () as argument
<Drup>
but it's not really an argument, it's just syntax
<hooplahoops>
Oh, you don't pass in () when you instantiate it?
<Drup>
I don't think so, need to check
<Drup>
but I'm sure it doesn't translate to anything at runtime anyway, it's really only typing stuff
Intensity has joined #ocaml
<hooplahoops>
Ah, ok
<hooplahoops>
Does that not surmount to a sort of whole-program compilation?
<hooplahoops>
E.g. I could make all modules a functor, that take a module argument Config or something, that's passed in all the way from the "main" part
<hooplahoops>
So now all modules will be specialized to this Config module?
avsm has quit [Quit: Leaving.]
<Drup>
yeah, except it has not much to do with compilation
<Drup>
it's more like dynamic linking, if you want
<hooplahoops>
Ah, so the type checking and compilation are entirely modular?
<Drup>
hum ?
<Drup>
what do you mean ?
<hooplahoops>
as in, I can compile my functor entirely independently of its users, and its users can be compiled independently of my functor
slash^ has joined #ocaml
<ggole>
Yep.
<hooplahoops>
Ah, cool :)
<ggole>
The only wrinkle is inlining, which breaks that a bit.
WraithM has joined #ocaml
<hooplahoops>
Yeah, but I suppose that's an optional feature
ygrek has quit [Ping timeout: 272 seconds]
<Drup>
whitequark: I'm done for now
<Drup>
whitequark: I'm doing a final rebuild after clean, just to be sure, but it's pushed, you can look at it.
<Drup>
I have enough C++ for at least a year.
WraithM has quit [Ping timeout: 265 seconds]
<whitequark>
lol k
larhat has quit [Quit: Leaving.]
ygrek has joined #ocaml
jonludlam has quit [Quit: Coyote finally caught me]
<reynir>
pattern matching on 3 nested records /o\
marynate has quit [Quit: Leaving]
bjorkintosh has quit [Ping timeout: 255 seconds]
hooplahoops has quit [Ping timeout: 256 seconds]
hooplahoops has joined #ocaml
* ggole
has been there
Hannibal_Smith has joined #ocaml
bezirg has joined #ocaml
dsheets has quit [Ping timeout: 265 seconds]
yomimono has quit [Quit: Leaving]
NoNNaN has joined #ocaml
ygrek has quit [Remote host closed the connection]
ygrek has joined #ocaml
artagnon has quit [Changing host]
artagnon has joined #ocaml
artagnon has joined #ocaml
avsm has joined #ocaml
hooplahoops has quit [Ping timeout: 260 seconds]
ygrek has quit [Remote host closed the connection]
ygrek has joined #ocaml
NoNNaN has quit [Ping timeout: 246 seconds]
malo has joined #ocaml
hausdorff has quit [Remote host closed the connection]
serge has joined #ocaml
serge has quit [Ping timeout: 245 seconds]
avsm has quit [Ping timeout: 265 seconds]
kakadu has quit [Quit: Page closed]
<Drup>
whitequark: re, it builds, test passes, you can merge if you're happy with it.
ygrek has quit [Ping timeout: 265 seconds]
lordkryss has quit [Quit: Connection closed for inactivity]
WraithM has joined #ocaml
<reynir>
I could use something like Lwt_stream.from, but with type (unit -> 'a list option Lwt.t) -> 'a t
<whitequark>
Drup: ugh.
<whitequark>
oh, nevermind
cody__ has joined #ocaml
* reynir
could use a rubber duck
<whitequark>
Drup: oh, haha, you missed a fun thing in the OCaml glue code
<whitequark>
it deliberately omits CAMLparam and friends when there are no OCaml objects on stack between allocations
<whitequark>
because speedz
bjorkintosh has joined #ocaml
<whitequark>
anyway, you don't need to change anything
rand000 has quit [Ping timeout: 255 seconds]
<ggole>
That'll never go wrong during maintainence.
jwatzman|work has joined #ocaml
<adrien>
:)
malo has quit [Remote host closed the connection]
jao has joined #ocaml
jao has quit [Changing host]
jao has joined #ocaml
kakadu has joined #ocaml
toolslive has quit [Ping timeout: 260 seconds]
zpe has quit [Remote host closed the connection]
zpe has joined #ocaml
kakadu has quit [Quit: Konversation terminated!]
kakadu has joined #ocaml
_andre has quit [Quit: leaving]
<whitequark>
fuck.
<whitequark>
LLVM's x86 buildbots use cmake.
<whitequark>
OCaml bindings can only be built with autoconf.
<whitequark>
yes, it has two mostly overlapping buildsystems.
<whitequark>
(╯°□°）╯︵ ┻━┻
zpe has quit [Ping timeout: 244 seconds]
rand000 has joined #ocaml
<adrien>
nah, they don't overlap
<adrien>
autoconf configures
<adrien>
cmake builds
sinelaw has joined #ocaml
<adrien>
it's definitely unable to do the configure step
<adrien>
(I wish that were a mere troll but seriously, cmake _sucks_ at the configure step)
bytbox has joined #ocaml
<adrien>
(by design)
<adrien>
(does it look like I don't feel like writing the code I need to write?)
<whitequark>
well, I dunno, it works for LLVM
<whitequark>
but now I have to somehow add OCaml support to cmake
<def`>
hehehe
<adrien>
what does llvm depend on? :]
<adrien>
and, good luck
<whitequark>
adrien: LLVM can be configured and built with either cmake or autoconf
<whitequark>
I don't know what godawful reasons led to this situation, honestly
samrat has quit [Quit: Computer has gone to sleep.]
toolslive has joined #ocaml
<adrien>
autoconf is more portable probably
<adrien>
so some people want to use something else
<adrien>
but it turns out auto* is still needed
<adrien>
so it's maintained in parallel
<adrien>
I've seen that several times
<whitequark>
I have a feeling like one of the main reasons autoconf is still there is that OCaml bindings depend on it
<Drup>
whitequark: why is this an issue only know ?
<def`>
adrien: I quite like ccmake to adjust config options that failed
zpe has joined #ocaml
<adrien>
and cmake's support for cross-compilation is in line with its ability to do the configure step
<whitequark>
Drup: I can't parse that sentence
<adrien>
basically for it to work, you provide each and every value that it's supposed to find by itself
<adrien>
whitequark: s/know/now/
<whitequark>
oh
<whitequark>
because I want OCaml bindings to be properly tested
<def`>
It's much easier to fix a broken cmake than a broken autoconf, and I must be unlucky because they are really often broken
<adrien>
yeah, horrible french people can't speak english
<Drup>
whitequark: oh, right
<whitequark>
right now they are only built on a hexagon buildbot
<adrien>
def`: :)
<whitequark>
hexagon is some obscure DSP architecture from TI
<adrien>
def`: my cmake invocations look like
<whitequark>
I have no clue why does it actually build the OCaml bindings
<whitequark>
it's completely useless, ocaml doesn't even run on hexagon
<whitequark>
I need to edit all your patches and add attribution to them, as they will be committed under my name
<Drup>
I know, hence the question :)
<whitequark>
and a few other issues as well
<Drup>
other issues ?
Sim_n has quit [Ping timeout: 260 seconds]
<reynir>
Not sure if it's ideal, though
<whitequark>
I fixed a few style issues as well
<Drup>
show me, so I can not do them next time
hausdorff has quit [Remote host closed the connection]
martintrojer has quit [Ping timeout: 272 seconds]
<whitequark>
space before ;, {} for single-line ifs, single-letter variable names
serge has joined #ocaml
<Drup>
oh, C++ syntax issues.
<whitequark>
indeed
martintrojer has joined #ocaml
<Drup>
yeah, I must say it's a language which I don't know the best practice
<Drup>
well, except "Don't."
<whitequark>
I also squashed some commits and edited commit messages and the like
<Drup>
k
<Drup>
you usually ask me to be very atomic for llvm, so I was :D
mcclurmc has quit [Remote host closed the connection]
<whitequark>
sure, it's not a problem. you don't have to know exact stylistic conventions, I'll edit them alright
<whitequark>
it's easier to squash than unsquash, too
<Drup>
indeed
serge has quit [Ping timeout: 258 seconds]
AlexRussia has quit [Quit: WeeChat 1.1-dev]
sinelaw has quit [Ping timeout: 256 seconds]
Sim_n has joined #ocaml
hausdorff has joined #ocaml
tnguyen has quit [Quit: tnguyen]
tnguyen has joined #ocaml
Simn has quit [Ping timeout: 260 seconds]
hooplahoops has joined #ocaml
AlexRussia has joined #ocaml
<whitequark>
Drup: ok, all committed
<Drup>
wooo
manud has joined #ocaml
Hannibal_Smith has quit [Quit: Sto andando via]
cody__ has quit [Quit: Leaving]
avsm has joined #ocaml
<nicoo>
whitequark: « I had to ship a patch to pkg-config » -> The baroque freedesktop implementation or pkgconf, the ANSI C reimplementation from GSoC 2011 ?
<whitequark>
it was pre-2011
<adrien>
.22 is definitely older than 2011
<adrien>
I'm never going to complain about pkg-config because...
<adrien>
% ls /usr/bin/*-config | wc -l
<adrien>
75
<adrien>
and these fail _hard_ for cross-compialtion
<adrien>
compilation*
tnguyen has quit [Ping timeout: 265 seconds]
tnguyen has joined #ocaml
Sim_n has quit [Ping timeout: 260 seconds]
hausdorff has quit [Remote host closed the connection]
<tinaj1234_>
I'm working on this file, which has both parts of ocaml and coccinelle script. I'm getting a error around line 61 of https://dpaste.de/sgmU . Seems like I've a syntax error. Can anyone please look into this?