2009-03-13

<hcarty> brendan: You need to load the dynlink module first ... camlp4 3.11 requires Dynlink

2009-03-12

<hcarty> munga: http://www.gnu.org/software/gsl/manual/html_node/Statistics.html has the documentation for the C library
<hcarty> munga: I have a patch to add the correlation function if you have a need
<munga> thx hcarty. exactly what I was looking for !
<Yoric[DT]> hcarty: thanks for submitting it
<hcarty> munga: The GSL bindings have a number of statistics functions
<hcarty> Yoric[DT]: Thanks for reviewing and applying the Bigarray patch

2009-03-10

<hcarty> Yoric[DT]: Enjoy your meal
<hcarty> I am happy to do so
<hcarty> Thank you for taking the time to look at it
<hcarty> Not at all a problem. They have been here as well.
<Yoric[DT]> hcarty: arf, my bad.
<hcarty> Yoric[DT]: Do you think you will have time to review the Bigarray patch I submitted before the beta 1 deadline?
<hcarty> thelema: Could you look at the Bigarray patch I submitted for Batteries?
<alexyk> since hcarty is in, repeat! :) how do I refer to the non-option arguments, such as file names, after Arg.parse?

2009-03-06

<hcarty> yziquel: The build process takes about 5 minutes if you are interested
<hcarty> If you have deb-design experience, contributions are greatly appreciated :-)
<hcarty> yziquel: It is setup to build the entire library, bindings and all, together
<yziquel> hcarty: what a pain... thanks.
<hcarty> yziquel: I'm not sure. I install from source since the OCaml packages are not included for Debian yet
<yziquel> hcarty: so i have to remove the debian package to rebuild all of it?
<yziquel> hcarty: 5.9.2
<hcarty> yziquel: You need to build the whole library
<hcarty> yziquel: Which version?
<yziquel> hcarty: got it from apt-get source plplot
<yziquel> hcarty: confused. the readme says 'type make', but no makefile... http://paste.lisp.org/display/76562
<hcarty> yziquel: Please let me know how it goes
<yziquel> hcarty: yes, i just noted your binding. trying to compile it.
<hcarty> yziquel: http://plplot.sf.net
<hcarty> yziquel: plplot

2009-03-04

<hcarty> mfp: Thank you! I thought ocamlnet had something
<hcarty> If that isn't wrapped already
<hcarty> Camarade_Tux: Ah. With that much work, I may as well use Perl4OCaml and Net::FTP
<Camarade_Tux> hcarty, by hand I thought opening a connection, checking the first few messages from the server, sending three messages and that's it, it's a plain-text protocol after all
<hcarty> Camarade_Tux: Yes, I'm thinking I may just use ftp, wget or similar directly
<hcarty> Any suggestions on a simple OCaml FTP library? I only need to log in (usually anonymous), list files in a directory and download them
<Yoric[DT]> hcarty: yeah, I know about this project, and we might eventually integrate it into Batteries.
<hcarty> Yoric[DT]: http://www.pps.jussieu.fr/~li/software/index.html#enhtop -- would this be useful?
<hcarty> Yoric[DT]: #listmodule
<hcarty> gcc version 4.3.2 (Ubuntu 4.3.2-1ubuntu12)
<hcarty> ehird: No such issues here, binprot 1.2.9
<ehird> hcarty: do you not get errors building bin-prot?
<hcarty> ehird: Batteries works here with GODI + 3.11

2009-03-03

<hcarty> http://ocaml-tutorial.org/ has a set of other intro material listed
<hcarty> bjorkintosh: http://ocaml-tutorial.org/ and the first chapter of the OCaml manual
<hcarty> mfp: Thanks, I'll take a look
<mfp> hcarty: see man findlib.conf
<mfp> hcarty: might work if the META and the .cm(x)a + .cmi are in a directory in OCAMLPATH
<hcarty> Assuming that question is worded sanely
<hcarty> Will findlib pick up a library not installed with findlib if it is installed without findlib, assuming an appropriate META file?

2009-02-28

<hcarty> bluestorm: codepad.org or ocaml.pastebin.com

2009-02-26

<hcarty> Yoric[DT]: http://www.ocaml.info/vim/ has some (I think) update OCaml vim support. OMLet is also a nice vim mode, though its indentation is SLOW on large files: http://www.lix.polytechnique.fr/~dbaelde/productions/omlet.html

2009-02-23

<xevz> A quick test shows your code works (with the modification hcarty mentioned).
<hcarty> paul424: I think you need a ; after the "if ... begin ... end"

2009-02-22

<hcarty> So discovering OCaml and the surrounding community has been quite fun
<hcarty> No, definitely not CS :-) But physical sciences are sorely lacking when it comes to using recent (last 30 years??) CS developments
<hcarty> It is a study of variations in rain structure in the tropics
<hcarty> Yoric[DT]: PhD proposal/topic defense
<Yoric[DT]> hcarty: what kind of deadline is that?
<hcarty> _zack, thelema: Care to take a look at a Batteries patch for ExtBigarray? It adds map, mapi, to_array to 1D, 2D and 3D Bigarrays
<hcarty> Sometimes work has to come first :-) I'm coming up on a grad school deadline, so the issue is at the front of my mind
<hcarty> Yoric[DT]: Understood. Thanks for information
<Yoric[DT]> hcarty: not yet, sorry.
<hcarty> Yoric[DT]: Have you had a chance to look at the Bigarray map/mapi/to_array patch?

2009-02-21

<hcarty> tsuyoshi: What is the camlzip bug?

2009-02-20

<hcarty> munga: You're welcome - even after finding that, it frustrated the heck out of me because I didn't realize I had to change my myocamlbuild.ml to take proper advantage of it
<munga> hcarty: thanks !!
<hcarty> munga: It may require some extra steps in myocamlbuild.ml or _tags, depending on the build process you use
<hcarty> munga: ocamlbuild foo.inferred.mli -- "man ocamlbuild" should give you that information under the Target Names section

2009-02-19

<Yoric[DT]> hcarty: I'll check this out tomorrow
<hcarty> It seems to work properly here
<hcarty> Yoric[DT], thelema_: Bigarray map, mapi, to_array patch submitted -- https://forge.ocamlcore.org/tracker/index.php?func=detail&aid=167&group_id=17&atid=153
<hcarty> Yoric[DT]: Darn... I was just about to submit a patch to address the missing Bigarray unsafe functions
<kaustuv> hcarty: according to src/core/extlib/_tags, all .mli files have use_optcomp
<Yoric[DT]> hcarty: you need a change in the _tags file.
<hcarty> Yoric[DT]: Do I need to change anything in the build system to have extBigarray.mli preprocessed by optcomp? (I think that's the name)
<hcarty> thelema_: Do I need to make any changes to the build system to have extBigarray.mli preprocessed with whatever handles the "#if ocaml_version..." directives?
<hcarty> kaustuv: Thanks. I think I may stick with git-diff in this case, as it really is just a simple patch
<thelema_> hcarty: I can also take git-diff patches, if you're really just submitting one set of changes (no history)
<kaustuv> hcarty: What I do is rebase a topic branch onto whatever (master, post-1.0, etc.), then run git-bundle create whatever.bdl master..HEAD
<hcarty> kaustuv: I've never submitted a patch using git before, aside from using git-diff to make a patch
<kaustuv> hcarty: Do you know how to use git-format-patch? If so, you already know everything you need to use git-bundle.
<hcarty> kaustuv: Any tips on creation?
<thelema_> hcarty: I can't tell you how to create them, but kaustuv probably can
<thelema_> hcarty: as of today, I know what to do with git bundles
<Alpounet> hcarty, probably reporting the bug and explaining + attaching the patch, using the tracker
<hcarty> thelema_: What is the proper way to create a git patch for submission to Batteries?
<hcarty> Ok, cool. I have the functions written for my own ExtBigarray library. I just need to move the code over to Batteries and add some documentation.
<Yoric[DT]> hcarty: definitely
<thelema_> hcarty: I'm quite inclusive in what can/should go into batteries. I lean towards yes.
<hcarty> Yoric[DT], thelema_: Would map, mapi(j|jk) and to_array functions be accepted for ExtBigarray in Batteries?
<kaustuv> hcarty: ah, good idea. I could just change the configure.ac.
<hcarty> alias ocamlbuild="ocamlbuild -j 4" ... though I have no idea if that would actually work or not
<hcarty> kaustuv: -j 4 or -j 5 according to "ocamlbuild --help"

2009-02-18

<hcarty> Poor OCaml. Being lulz-d at like that.
<hcarty> "val sum : float array array -> float" or something similar?
<hcarty> jeromeg: Sum a list/array of vectors?

2009-02-17

<adrian_2002ca> hcarty: kk, thanks!
<hcarty> adrian_2002ca: If someone here has time and can help they will likely offer their assistance

2009-02-16

<hcarty> That's why I thought "with" would work well, as it reads appropriately and I think it should avoid ambiguities
<hcarty> Yoric[DT]: Does the Batteries "module Foo = Foo open Bar" syntax have the same ambiguity problem as the previous "include" form?

2009-02-15

<hcarty> Though they still require some inline C to properly wrap the functions
<hcarty> oleg|log: There are some interesting examples in the ciml darcs repo
<hcarty> At least for functions with simple interfaces
<hcarty> oleg|log: I haven't used it, but it seems very promising
<oleg|log> hcarty: thanks, looks nice
<flux> hcarty, might be it. I doubt there are two :)
<Yoric[DT]> hcarty: good idea.
<hcarty> The Fedora source is, from what I can tell, a clean CVS checkout
<hcarty> What I will/should do is make my changes in to a patch against the source as provided by Fedora (what I used as a base)
<hcarty> And I'm not sure if that is even maintained
<hcarty> Just CVS availability
<hcarty> Yoric[DT]: There is no official .tgz. To my knowledge, not official release was ever made
<Yoric[DT]> hcarty: Why do you mirror the .tgz, though?
<hcarty> Unfortunately, the section of the GODI manual related to conf-* packages seems to be unwritten
<hcarty> Yoric[DT]: Thanks for testing
<Yoric[DT]> hcarty: installation works.
<Yoric[DT]> hcarty: testing now
<Yoric[DT]> hcarty: sorry, I didn't have time to test your package yesterday.

2009-02-14

<hcarty> Yoric[DT]: Thanks. No rush if you have to wait. It works here, but I've only tested on one system
<hcarty> But I don't know how to do that process yet, and I'd like to test this portion first
<hcarty> A conf-cairo would be ideal
<hcarty> autoconf
<hcarty> Yoric[DT]: It uses a configure script
<Yoric[DT]> hcarty: it doesn't require any conf?
<hcarty> Yoric[DT]: When/if you have a chance, http://0ok.org/ocaml/cairo.godiva is the file for cairo-ocaml
<Yoric[DT]> hcarty: ok
<hcarty> Yoric[DT]: I should be around then - just let me know
<Yoric[DT]> hcarty: not right at this moment but in 1 or 2 hours, with pleasure.
<hcarty> "hacked/patched" meaning that I made some changes to the code to make it GODI+godiva friendly
<hcarty> If this works, I'll try the process a bit cleaner
<hcarty> Yoric[DT]: I have a godiva file for a (hacked/patched) cairo-ocaml which seems to work properly on my system. Would you be interested in/willing to test?

2009-02-13

<hcarty> Wow... that's pretty nifty
<hcarty> Ah, I'm unfamiliar with colors in the toplevel
<hcarty> And the set output is ... long
<hcarty> Just an underline
<hcarty> Camarade_Tux: No colors
<Camarade_Tux> hcarty, could you paste the output of 'set' from your computer ? I can't get the colours
<hcarty> Camarade_Tux: It works in an xterm here
<hcarty> It's been a long while since I've looked though
<hcarty> but the last time I looked the documentation wasn't very clear
<hcarty> Yoric[DT]: What are the steps to make a GODI package using godiva? The Batteries .godiva looks fairly straightforward...
<hcarty> Take a look at the META file and the .spec file in the F-10 directory
<hcarty> AxleLonghorn: You can also get the source package directly
<hcarty> They each have a META file and you can (roughly) follow their methods for building and installation
<hcarty> AxleLonghorn: I'd recommend looking at the Debian and/of Fedora packages for cairo-ocaml

2009-02-12

<hcarty> brendan: I tried hg, and it seemed quite similar to git, but without the in-place branches
<hcarty> I need to leave for a seminar. I'm happy to catch up on comments here and discuss this later though.
<hcarty> I'm not sure
<hcarty> Would ~step:-1.0 work?
<hcarty> I may be outvoted on this
<hcarty> thelema_: I think that is harder to read, and still needs () or an extra ~ for negative values
<hcarty> At least that's what I have in the feature request
<hcarty> thelema_: It's (start, step) --. end
<hcarty> I would personally prefer float in Pervasives (op-plus-. generally refers to floating point operators) but versions for other types could be added to the other numeric modules
<hcarty> I'm not sure what the default in Pervasives should be, then
<hcarty> Yoric[DT]: True
<hcarty> Where are the other enum operators placed?
<hcarty> Yoric[DT]: Pervasives would be my preference. But I don't know how often others would use it.
<Yoric[DT]> hcarty: about your RfF on --.
<hcarty> Regarding the use of with - do "open X with init ()" and "module X = Y with A, B" interfere with one another?
<hcarty> From the definition of Bar, that is
<hcarty> I'm not sure how useful "include foo_t from Bar" would be for a type foo_t unless camlp4 could find and pick out the definition of foo_t from Bar
<hcarty> Well, I'm thinking specifically of values in the case of "include foo from Bar"
<hcarty> Oh, right... UIDENT and IDENT or something along those lines
<hcarty> Yoric[DT]: True
<Yoric[DT]> hcarty: mmmmhhhhh....
<hcarty> Yoric[DT]: Could it be used for both? IIRC, camlp4 can distinguish between modules and values
<hcarty> So it's not really type inference as much as "I expect this to be the default type here"
<hcarty> flux: Well, "your self" is the programmer here
<Yoric[DT]> hcarty: "with" doesn't work, we already use keyword "with" for initialisation of modules.
<hcarty> Though I'm quite happy with the results of pa-do. I have found it wonderfully useful.
<hcarty> pa-do uses Int.(), Float.(), etc to do things along those lines. But this is, as you said, effectively doing the inference your self
<hcarty> Yes, it would be interesting to witness what people come up with given, for example, type information
<hcarty> Definitely :-) As long as the processing terminates
<hcarty> programmer -> camlp4 -> compiler -> camlp4 -> compiler -> ...
<hcarty> It would be cool if it could though
<hcarty> I don't know if camlp4 can get without a lot of extra work
<hcarty> Would that require the full type definition though?
<hcarty> I think modules may not save any typing or clarity. I'm not sure about types though
<hcarty> Does "let function_a = Module.function_a" introduce any extra runtime overhead if one then uses function_a rather than Module.function_a?
<hcarty> Yoric[DT]: Will do!
<Yoric[DT]> hcarty: not yet
<hcarty> Does Batteries currently have an extension to provide syntax like "include printf, eprintf from Printf" --> "let printf = Printf.printf let eprintf = Printf.eprintf"? I remember some discussion of this in the past but I don't know if it happened or not
<hcarty> module List = List with Labels, ExceptionLess
<hcarty> Yoric[DT]: Would "module List = List with Labels" be ok?
<hcarty> Submitted. And I am glad to help.
<Yoric[DT]> hcarty: please do
<flux> hcarty, well thank you, or perhaps the Batteries community should thank you :)
<hcarty> flux, Yoric[DT]: The "module Foo = Foo\ninclude Bar" concern with Batteries seems to be valid, at least when using the toplevel and OCaml 3.11 -- I can submit a bug if you'd like

2009-02-10

<hcarty> s/much//
<hcarty> Or perhaps the problem runs much deeper... I'm just toying with a simple file here
<hcarty> Would -sep " " work any better?
<hcarty> Ah, ok. How do you get it to keep comments after preprocessing?
<hcarty> Yoric[DT]: Why, out of curiosity, -sep?

2009-02-09

<hcarty> jado: I'm far from an emacs expert... I just added that to my .emacs (or .emacs.d/init.el) and (re)start emacs
<jado> hcarty: how can i test it ?
<hcarty> You're welcome. I hope it works for you as well
<jado> hcarty: thanks : )
<hcarty> jado: That's the hope :-)
<jado> hcarty: that will re-highlight the code correctly on C-x C-x ?
<hcarty> jado: (define-key global-map (read-kbd-macro "C-x C-x") 'font-lock-fontify-buffer)
<hcarty> jado: I think I ran in to the same problem when I was using emacs. I had to force it to re-highlight the code every now and then

2009-02-06

<jonasb> hcarty: will do :-) read the batteries presentation from the ocaml meeting, and it all sounded very good
<hcarty> Have fun with it :-)
<hcarty> The proper package name is under "Comprehension expressions"
<hcarty> jonasb: I don't think you included the comprehension extension
<hcarty> Without omake installed
<hcarty> Not omake, but Batteries
<hcarty> I was able to get it build here on godi + 3.11
<hcarty> Does it?
<hcarty> It looks like it is trying and failing to build omake, perhaps?
<hcarty> jonasb: I don't think that's a Batteries error
<hcarty> ttamttam: I imagine that does (approximately?) the same thing
<hcarty> After that the install seemed to work properly
<hcarty> And batteries_nothreads
<hcarty> Then repeated for the packages batteries_threads
<hcarty> I used: "ocamlfind remove batteries -destdir $GODI_DIR/lib/ocaml/pkg-lib"
<hcarty> ttamttam: I had the same problem
<hcarty> Yoric[DT]: Submitted.
<Yoric[DT]> hcarty: yes, please, could you submit this as a bug?
<hcarty> Yoric[DT]: Or, if it would be better, I can submit that as a bug report on the forge instead
<hcarty> Yoric[DT]: It may be worth noting in the Batteries syntax extension documentation that comprehesions do work in the 3.11 toplevel
<hcarty> I use Calendar in a lot of my code
<hcarty> Yoric[DT]: Ok, will do
<Yoric[DT]> hcarty: it will depend on whether it's not too hard to get Netdate and Calendar to interoperate.
<hcarty> It may be too large or outside of the intended scope of Batteries
<hcarty> Yoric[DT]: Are there any plans to use/integrate the OCaml Calendar library in Batteries for time/date functions? I haven't used Netdate, so I'm not sure how they compare...
<hcarty> Anything new or exciting in the INRIA OCaml talk? That seems to be the only one which doesn't have slides available on cocan
<rwmjones> hcarty, yes, it was good
<hcarty> rwmjones: The ping was just to ask how the OCaml meeting went, since your name was here in IRC
<rwmjones> hcarty, pong

2009-02-05

<Yoric[DT]> hcarty: thanks
<Snark> hcarty, that was pretty interesting
<hcarty> Snark: How were the talks overall? The slides look interesting all around
<Snark> hcarty, the talk was good too :-)
<hcarty> Yoric[DT]: Very nice Batteries slides
<hcarty> rwmjones: ping

2009-02-01

<hcarty> Yoric[DT]: I need to go for a bit, but if there is extra testing or information I can provide please leave a message here and I'll be happy to do so
<hcarty> Yoric[DT]: Yes
<Yoric[DT]> hcarty: mmmhh....
<Yoric[DT]> hcarty: ok
<hcarty> As does "the tutorial"
<hcarty> #man "for programmers";; does work
<hcarty> Same error
<hcarty> And using the godi package for section 3.11
<hcarty> Oh, yes
<hcarty> godi, 3.11
<hcarty> Everything else seems to work properly lthough
<hcarty> # Error showing url: Operation not supported
<hcarty> Yoric[DT]: # #man "print_endline";;
<hcarty> This is mainly for plotting purposes though, so appropriate structures may be different
<hcarty> I'm not sure of a appropriate data structures. I'm working on reading and interpreting GMT's coast and country outline data, though I haven't had much time to spend on it
<Camarade_Tux> hcarty, I'd like to store openstreetmap data and I've just started looking at the different data structures (I currently know fast nothing about them)
<hcarty> Camarade_Tux: Of what sort?
<hcarty> That's reasonable, it would just be nice to have while testing from git
<hcarty> Or add a "make uninstall-docs" target if that seems better
<hcarty> Yoric[DT]: One request (godi may handle this): Could you alter "make uninstall" so that it also removes the docs?
<hcarty> Yoric[DT]: Ah, ok. I'll give it a shot
<Yoric[DT]> hcarty: I was looking for testers for the new GODI package.
<hcarty> Yoric[DT]: pong
<Yoric[DT]> hcarty: ping
<hcarty> If you use estring or some other wide character-aware string library and write a special toplevel printer for it then you could probably get what you want
<hcarty> If you print it, everything comes out fine
<hcarty> There was talk at last year's meeting about adding support. But I have no idea if this is happening or not.
<hcarty> xah_lee: Not that I know of. Camomile may have some sort of support, but vanilla OCaml does not have >8bit character support as far as I know
<xah_lee> hcarty: is there a way to get it to print a star instead?
<hcarty> - : string = "\226\152\133"
<hcarty> # "★";;
<hcarty> xah_lee: That's what happens here

2009-01-31

<hcarty> That's unfortunate
<hcarty> bluestorm: I'm not sure if there is built-in support for something other than Module.() syntax. There is support in the API for arbitrary rewriting within the () though.
<bluestorm> hcarty: i considered pa_do for other things recently
<hcarty> JoCaml uses &, so I don't know if that would cause long-term problems
<hcarty> I've never done much with Haskell
<hcarty> Ok, I thought it was something like that
<hcarty> No, $ would be (I think?) the oppposite of Haskell
<hcarty> ( & ) and ( $ ) work, but they don't indicate direction as nicely and they clash with other things
<hcarty> bluestorm: pa-do/examples/pa_compos.ml in pa-do
<hcarty> bluestorm: ocaml-3.11.0/camlp4/examples/apply_operator.ml in the main distribution
<hcarty> More an extra optimization pass than anything else, I suppose
<hcarty> Yes
<hcarty> Yes, it would probably be best as a separate and user-specified extension if it were included
<hcarty> There is an example in the ocaml sources, and in pa-do (two different methods, somewhat obviously)
<bluestorm> hcarty: actually
<hcarty> bluestorm: Would you use pa-do or "raw" camlp4 for the <| and |> implementation and optimization in Batteries?

2009-01-30

<hcarty> There is a doc/godi-batteries/html/ but not doc/godi-batteries/doc/
<hcarty> Oh, sorry... had that on the pastebin
<hcarty> Then "ocamlfind batteries/ocaml"
<hcarty> Yoric[DT]: make all opt install install-doc
<Yoric[DT]> hcarty: how did you install?
<Yoric[DT]> hcarty: weird.
<hcarty> mfp: You're quite welcome... good luck! :-)