flux changed the topic of #ocaml to: Discussions about the OCaml programming language | http://caml.inria.fr/ | Grab OCaml 3.10.2 from http://caml.inria.fr/ocaml/release.html (featuring new camlp4 and more!)
middayc has joined #ocaml
longh has quit [Read error: 104 (Connection reset by peer)]
middayc has quit []
seafood has quit [Read error: 60 (Operation timed out)]
seafood has joined #ocaml
pango has quit [Remote closed the connection]
pango has joined #ocaml
struktured_ has joined #ocaml
struktured has quit [Connection timed out]
struktured_ is now known as struktured
netx has joined #ocaml
struktured_ has joined #ocaml
seafood has quit [Read error: 60 (Operation timed out)]
seafood has joined #ocaml
struktured has quit [Read error: 110 (Connection timed out)]
Associat0r has quit []
davidm123 has quit ["ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]"]
netx has quit [Read error: 60 (Operation timed out)]
struktured_ is now known as struktured
seafood_ has joined #ocaml
seafood has quit [Read error: 110 (Connection timed out)]
seafood_ has quit [Read error: 60 (Operation timed out)]
seafood has joined #ocaml
struktured has quit [Connection timed out]
seafood has quit [Read error: 60 (Operation timed out)]
seafood has joined #ocaml
seafood_ has joined #ocaml
ygrek has joined #ocaml
seafood_ has quit []
seafood has quit [Read error: 110 (Connection timed out)]
Camarade_Tux has joined #ocaml
Palace_Chan has quit ["Palace goes to sleep"]
Snark has joined #ocaml
seafood has joined #ocaml
GustNG has joined #ocaml
filp has joined #ocaml
<Camarade_Tux> hum
<Camarade_Tux> godi-bin-prot is required for godi-bin-prot.
<Camarade_Tux> by godi ;p
<Camarade_Tux> I'm wondering if my godi install is sane
<Camarade_Tux> s/if/whether
<palomer> haha
<Camarade_Tux> findlib-1.2.3 doesn't install, I have 1.2.2 currently
<Camarade_Tux> palomer, on the other hand it will only take 1 minute of my time to recompile everything (and three hours of compile time ;p )
<Camarade_Tux> of course, findlib-1.2.3 is selected because of the dependencies =/
mishok13 has quit [Read error: 104 (Connection reset by peer)]
_zack has joined #ocaml
mattam has joined #ocaml
filp has quit ["Bye"]
mishok13 has joined #ocaml
ygrek has quit [Remote closed the connection]
GustNG1 has joined #ocaml
mattam has quit [Remote closed the connection]
GustNG has quit [Read error: 110 (Connection timed out)]
<flux> hm, batteries has rangeiter and bulk_iter in the same module, a bit inconsistent?
Linktim has joined #ocaml
itewsh has joined #ocaml
vbmithr has joined #ocaml
asmanur has joined #ocaml
marmotine has joined #ocaml
<flux> hmph, apparently there's a game written in ocaml using cairo bindings by jpakaste, but his home pages have disappeared (http://koti.welho.com/jpakaste)
<Camarade_Tux> flux, here it is : http://juripakaste.fi/caboodle/
<Camarade_Tux> ;)
<Camarade_Tux> welho seems to be a finnish isp or something like that so it's not surprising the regular page is not available anymore
<Camarade_Tux> (* ils ont de la pub pour canal+ en finlande... *)
mattam has joined #ocaml
<flux> yes, it is.. but I didn't think of finding out the guy's real name, jpakaste was useless for googling :)
itewsh has quit ["KTHXBYE"]
<Camarade_Tux> in fact I found the game name and googled for caboodle+cairo ;)
<flux> bah, you have superior google skills ;)
<Camarade_Tux> sometimes it seems so but there's something else !
<Camarade_Tux> instead of going to google.com and put my query, I use yaxm.org/!query
<Camarade_Tux> there's an apache redirection to google.com/lots-of-parameters
<Camarade_Tux> this usually gives better results (especially when you're looking for something in english and are not in an english-speaking country and/or you're computer is not set to use english)
<Camarade_Tux> plus google somehow learned this was mostly used to do caml-related searches and now it ranks caml-related results better
<Camarade_Tux> so if I search for something related to C, it gives me results about ocaml bindings rather ;)
<flux> naah, it's just because ocaml bindings are going nplaces ;)
<grirgz> Camarade_Tux: you are afabulating, the yaxm search is just a standard search, nobody learned nothing
<grirgz> =)
<Camarade_Tux> grirgz, I never said it was more ;)
<Camarade_Tux> it's just muuuuuuch faster to write :p
<grirgz> i will use it now ;)
<Camarade_Tux> and it used to enable keyboard shortcuts but I've disabled it the day the cursor got automatically placed on the 5 or sixth result which was a youtube one
<Camarade_Tux> in fact it would put you on google results first, no matter their rank !
mattam has quit [Remote closed the connection]
<Camarade_Tux> and really the good thing is it sets google to english, otherwise it's usually for french and I never get the results I want as they are not in french usually
<Camarade_Tux> grirgz, go on ;)
<grirgz> yes, why put a "search in france / search in worldwide", the results are biased
<grirgz> yes, and when you configure your browser to "fr", google.com always redirect to google.fr
<grirgz> i hate google more and more each day
<Camarade_Tux> grirgz, that's exactly what I and Asmadeus experienced, he had the idea, I had the domain ;)
<grirgz> yesterday i was looking for "uninstall flash firefox linux" and this f***ing google show me results for with keyword install, it think it is smart, but it's not, i hate it
jlouis has joined #ocaml
<Camarade_Tux> rm /u/l/m/p/<TAB> (zsh inside) which expands to rm /usr/lib/mozilla/plugins/libflashplayer.so, but I'm not sure every distributions appreciate that...
<grirgz> zsh rox ;)
<grirgz> yes i have found now
Linktim_ has joined #ocaml
<Camarade_Tux> I used to use bash, zsh being on my todo list (had no time to configure it !), then I tried and I couldn't go back to bash now ;p
Yoric[DT] has joined #ocaml
<Camarade_Tux> now I need to make something to have ocaml completion
<Camarade_Tux> and findlib, and, ocamlmklib, and, and, and ... :p
<grirgz> but it's strange, i've installed gnash, and iceweasel dont know it until i make a symlink from the system plugin to my plugin home dir
GustNG1 has quit [Read error: 60 (Operation timed out)]
<grirgz> ocaml completion ? with the google summer of code project ?
<grirgz> (i've forgoten the name)
<Camarade_Tux> you mean it's installed in the system one and it can't find it ?
<grirgz> yes
<Camarade_Tux> hmm, I had troubles with detection too when making something to make npapi plugins in ocaml
<Camarade_Tux> I eventually stopped and discovered somebody had already completed that ;p
<grirgz> npapi plugins ?
<Camarade_Tux> as for the completion, nah, only when running ocaml, for instance completing the .cma names according to the folders already included and so
<Camarade_Tux> npapi, basically browser plugins
Associat0r has joined #ocaml
GustNG1 has joined #ocaml
<Camarade_Tux> with browser being any browser except internet explorer
<grirgz> of course =)
asmanur has quit ["Lost terminal"]
<Camarade_Tux> there's a npplugin to what it wants though
Linktim has quit [Read error: 110 (Connection timed out)]
<grirgz> i am wondering the proportion of french in this channel =)
<Camarade_Tux> at least 11 out of 76 but then if I only count people I know of it's rather 11 out of 25 ;)
<Camarade_Tux> (approximately of course)
<Yoric[DT]> hi
<Camarade_Tux> hi Yoric[DT], everything alright ?
<Yoric[DT]> Well, a few bug reports.
<Yoric[DT]> But that's ok.
<Yoric[DT]> That's actually what we hoped :)
<Camarade_Tux> perfect then (and it seems the internet is a bit dead this week-end)
<Yoric[DT]> We've had 70 downloads since yesterday:)
<Camarade_Tux> that's quite a lot indeed \o/
<Camarade_Tux> it seems my godi died though
<Camarade_Tux> but I'll just wipe everything soon
<Yoric[DT]> Did I kill your godi?
pango has quit [Remote closed the connection]
<Camarade_Tux> I don't really know, binprot won't install, findlibe won't upgrade, some (most ?) packages work though
pango has joined #ocaml
Linktim_ has quit [Read error: 110 (Connection timed out)]
mattam has joined #ocaml
kig has quit [Read error: 110 (Connection timed out)]
_zack has quit [Read error: 110 (Connection timed out)]
_zack has joined #ocaml
seafood has quit []
seafood has joined #ocaml
kig has joined #ocaml
ygrek has joined #ocaml
<flux> yoric[dt], I guess you prefer actual bug reports over irc messages?-)
<flux> (nothing more than a couple things in the documentation yet, though..)
Linktim has joined #ocaml
mattam has quit [Remote closed the connection]
<Yoric[DT]> flux: yes, preferably.
mattam has joined #ocaml
Linktim_ has joined #ocaml
Camarade_Tux has quit []
Camarade_Tux has joined #ocaml
Linktim has quit [Read error: 110 (Connection timed out)]
Linktim has joined #ocaml
Camarade_Tux has quit [Remote closed the connection]
Camarade_Tux has joined #ocaml
Linktim_ has quit [Read error: 110 (Connection timed out)]
<thelema> flux: rangeiter and bulkiter are different.
zbrown has joined #ocaml
okagawa has joined #ocaml
Linktim has quit [Read error: 110 (Connection timed out)]
ofaurax has joined #ocaml
mishok13 has quit [Read error: 104 (Connection reset by peer)]
<flux> thelema, I know, but one of them is with underscore and the other without
<flux> I think usually the words are separated with underscores, right?
<thelema> ah. okay.
mishok13 has joined #ocaml
<zbrown> hrm
<thelema> flux: fixed in svn
okagawa has quit []
<Camarade_Tux> metaocaml seems to work nicely, I just hope it works under windows too...
<Camarade_Tux> I think I'll need at least a short tutorial...
<flux> camarade_tux, can you also write an example that shows how much more convenient it is for expressing certain progreams, as that hasn't become clear for me, although I have an idea what it can do :)
<flux> thelema, great
<olegfink> Camarade_Tux: which version do you use?
<Camarade_Tux> flux, currently I only need to eval user-provided code, nothing more but if I make more complicated things, I will
<Camarade_Tux> olegfink, 3.09.1 alpha 030
<Yoric[DT]> Great...
<Yoric[DT]> I've just had a paper rejected.
<thelema> Yoric[DT]: that sucks
filp has joined #ocaml
<Yoric[DT]> Out of the three reviewers, two are familiar with the field, and the last one isn't.
<Yoric[DT]> The two who are give very good marks to the paper.
<Yoric[DT]> The third one obviously doesn't understand anything.
filp has quit [Client Quit]
<Yoric[DT]> Which means that I can't even improve the paper based on the feedback, it's already close to maximal marks by the two reviewers who could actually give some feedback.
<thelema> Yoric[DT]: or you didn't give enough background for him to catch on.
<Yoric[DT]> thelema: in 5 pages?
<Yoric[DT]> The first sentence of his review is him actually complaining about too much background and the last sentence is him complaining that there's not enough.
<thelema> I admit there's limits, but just about every paper I read has a decent explanation of the problem space
<olegfink> Yoric[DT]: yeah, I know how it feels.
<thelema> maybe your background was both too long and insufficient for him.
<Yoric[DT]> thelema: well, obviously.
<Yoric[DT]> thelema: but that's not feedback I can use.
<Yoric[DT]> olegfink: :/
<olegfink> my paper was rejected just because reviewers just didn't want to try to think in functional terms
<Yoric[DT]> Well, my paper to last year's ML workshop was rejected because of the premise: I didn't want to modify the compiler.
<thelema> there's plenty of politics in publishing. :(
longh has joined #ocaml
<Yoric[DT]> In that case, the two good reviews give me marks of respectively 8.2 and 8.6/10, with (quoting the second one) "the best account of dependent types and effects I have ever encountered."
<Yoric[DT]> It's terribly frustrating.
mattam has quit [Remote closed the connection]
<Yoric[DT]> ertai: ping
<_zack> hi all guys, channel newbie speaking, actually I wasn't aware of the popularity of this channel before reading about it in batteries release announcement
<_zack> pleased to meet you all
<Yoric[DT]> Nice to meet you.
hkBst has joined #ocaml
<thelema> hi _zack
<_zack> btw, congratulations for batteries release, I hope to have a Debian package ready soon
<Yoric[DT]> _zack: Hi Stefano :)
<_zack> Yoric[DT]: ;)
<Yoric[DT]> Thanks for the patch.
<Yoric[DT]> I think thelema just committed it.
<_zack> you're welcome, in fact I think I'll come up with a more generic patch implementing the various packaging needs
<Yoric[DT]> That'll be nice.
<_zack> only let me know if you have a clue about the ocamlopt need when building bytecode
<_zack> I hope it's not something deep inside ocamlbuild ...
<_zack> usually it's just surface, but in these ocamlbuild days one never know ;)
<Yoric[DT]> That's actually surprising.
<Yoric[DT]> We're not issuing any ocamlopt specific command.
<Yoric[DT]> More precisely, we're letting ocamlfind choose the compiler.
<_zack> uh? ocamlfind needs to be invoked with either opt or ocamlc?
<_zack> s/\?$//
<_zack> you mean ocamlbuild?
<Yoric[DT]> No, I mean ocamlfind.
<Yoric[DT]> In ocamlbuild, we replace "ocamlc" with "ocamlfind ocamlc" and "ocamlopt" with "ocamlfind ocamlopt".
<Yoric[DT]> And that's it.
<Yoric[DT]> So yes, ocamlbuild has also a role in the choice.
<_zack> yup, ocamlfind does not decide whether to invoke ocamlc (whatever it is) or ocamlopt, indeed
<thelema> iirc, it's in compiling the myocamlbuild.ml that it wants ocamlopt
<_zack> thelema: argh :(
<Yoric[DT]> Ah.
<Yoric[DT]> That may be a valid reason, too.
<Yoric[DT]> Yep.
* Yoric[DT] is currently compiling with verbosity.
<Yoric[DT]> The only occurrence of ocamlopt is when building myocamlbuild.ml .
<thelema> sounds like an ocamlbuild bug - but I guess it doesn't happen for other projects... because they don't use myocamlbuild?
<_zack> actually, ocambuild is still very *unpopular*
* Yoric[DT] likes it.
<Yoric[DT]> It's very powerful.
<_zack> I'm using it for my project, but from the POV of packaging batteries is the first project using it
<Yoric[DT]> Too bad the code is undocumented.
<_zack> I like it too
<_zack> yes, don't get me started on that ;)
<_zack> so it is very possible that there is a bug like that, not yet discovered
<Yoric[DT]> ertai: ping
<Yoric[DT]> _zack: sometimes, Nicolas Pouillard lurks on this channel.
<_zack> k
<Yoric[DT]> (as ertai)
* _zack is re-reading the bugreport about that
<_zack> yup, is definitely building myocamlbuild
<Yoric[DT]> mmmhh...
<Yoric[DT]> -byte-plugin Don't use a native plugin but bytecode
<_zack> if that's enough I'm fine with that, I've already patches your Makefile to pass externally flags to ocamlbuid
<_zack> but that's something probably worth to be support by your makefile per se, as there can be other people out there without ocamlopt
<_zack> yes, it works
<_zack> it is enough to add -byte-plugin to the 2 ocamlbuild invocation of the all: target apparently
<_zack> (maybe even to install)
<_zack> nope, install is fine as it is, because obviously it just uses ocamlfind and not ocamlbuild
<Yoric[DT]> Yep, that's enough.
<Yoric[DT]> I've just patched our Makefile.
<_zack> cool
<Yoric[DT]> Mmmmhhh....
<Yoric[DT]> Actually, what do you prefer?
hkBst has quit [Read error: 104 (Connection reset by peer)]
<Yoric[DT]> An environment variable?
<_zack> I'll show my current lines
<_zack> byte:
<_zack> ocamlbuild -byte-plugin $(OBFLAGS) src/main/threads/batteries.cma
<_zack> ocamlbuild -byte-plugin $(OBFLAGS) src/main/nothreads/batteries.cma
<_zack> OBFLAGS is defined as empty upper in the Makefile
<_zack> this way, during debian package build, I can also invoke make as: make byte OBFLAGS=-classic-disply
<_zack> (which is nicer for the build logs)
GustNG has joined #ocaml
hkBst has joined #ocaml
<Yoric[DT]> ok
mattam has joined #ocaml
<Yoric[DT]> Done, here too.
<Yoric[DT]> (not committed yet)
<_zack> kthx, no problem, I'm not in a hurry :)
<_zack> actually I think I should indeed start following your version control system ...
<Yoric[DT]> Great:)
<Yoric[DT]> For the moment, we're still using the trunk, although we will probably move bugfixing to a branch at some point.
<_zack> Yoric[DT]: if you want git or some other vcs let us know, on ocamlcore we can set them up as well
<Yoric[DT]> Now, regarding your ITP discussion, what do you think of "candidate-for-standard development platform"?
<Yoric[DT]> git would be nice
<_zack> when I heard saying "branch" in svn context I'm always scared :)
<Yoric[DT]> :)
* thelema likes git
<_zack> and actually for the package I'm using git, so having that will help sharing patches
<Yoric[DT]> Many people around me advocate git, so I'm willing to give it a try.
<_zack> Yoric[DT]: for the description, I've no strong opinions, but I understand Eric's point
<Yoric[DT]> So do I.
<_zack> maybe you can have a look at similar batteries included efforts and see what they say about what they are?
<_zack> in fact, the current description of batteries is wonderful, but it is indeed missing a short statement of what it is
<Yoric[DT]> Well, for Haskell, they speak of their "standardization process".
<Yoric[DT]> Wonderful?
* Yoric[DT] blushes.
<_zack> yup, I've liked reading it
<_zack> (but maybe because it is permeated by the very same frustration I've always had about OCaml ;) )
<Yoric[DT]> Thanks.
<Yoric[DT]> :)
<Yoric[DT]> Do you mean that I should add something to the README or to about.html?
* thelema would be just happy with a de-facto standard development environment
<_zack> Yoric[DT]: I mean that you should have a sort of "tagline" describing what batteries is
<_zack> short and to the point
<_zack> maybe, before it is standard, we can go for something like "comprehensive" ?
<_zack> or something which recall uniformity ...
<_zack> then you should add it at the beinning of both README and about.html
<thelema> Batteries: a complete development environment for OCaml
<_zack> why not, though I like "platform", which you have been used here and there
<thelema> I'm unsatisfied with "development environment" too.
<thelema> but platform says very little
<olegfink> it's not a development environment
<olegfink> my development environment is unix, rc, vim, acme and ocaml
<thelema> olegfink: it's not a development environment in terms of IDE, but it does form a good chunk of the context in which development takes place
<olegfink> I refuse to call the libraries I use an "environment."
<thelema> maybe the big difference is that it's an extended, richer OCaml ... it's not a language...
<thelema> the intent is that it's more than just a library.
<thelema> it's an extension to OCaml.
<thelema> it extends the syntax and the builtins
<flux> how is it not a platform?
<flux> upon which you build things
<flux> environment is just something you work in
<flux> not something you end up delivering..
<olegfink> it's either a platform or a framework
<flux> and it's not a framework, unless it is extensible in a fashion that you can write more Data.*-modules for instance (and I don't think you can)
<thelema> flux: that's not the intent
<flux> well, I suppose it doesn't deliver all the parts that build a pltaform
<flux> only parts of it
<thelema> Batteries: extending OCaml into a rich language
<thelema> awkward, but the meaning's not bad.
<flux> it does very little to the language :)
GustNG1 has quit [Read error: 110 (Connection timed out)]
<thelema> it extends the standard libraries (part of the language as much as the syntax) to a large degree.
<Yoric[DT]> Well, I personally like Platform.
<Yoric[DT]> I think it's too early for "de-facto standard".
<Yoric[DT]> Perhaps the word "community" should appear in the tagline.
* thelema has "de-facto standard" as the goal
<Yoric[DT]> Goal, yes.
<Yoric[DT]> But that's not achieved.
<thelema> Batteries: a community-built rich platform for OCaml programming
<Yoric[DT]> I concur that it's not exactly a framework.
<Yoric[DT]> That's one possibility.
* thelema brainstorms
<Yoric[DT]> :)
<Yoric[DT]> Batteries: your one-stop library for OCaml programming :)
<_zack> we are drifting about too much towards commercials, aren't we? ;)
<Yoric[DT]> Naaaah.
* Yoric[DT] fires synaptic
pango has quit [Remote closed the connection]
<thelema> _zack: you wanted a tagline, a sound-bite
<_zack> ok, I'll take it back :-P, I just want a "short description"
<hcarty> Batteries: Powered by OCaml
<Yoric[DT]> hcarty: or "Empowering OCaml"?
<Yoric[DT]> Or does that sound too much like Oprah?
<hcarty> Yoric[DT]: I suppose it depends on the attention you are trying to attract
<Camarade_Tux> I'd use the word "comprehensive", I'll let you decide how and where
<Yoric[DT]> "Batteries: Standardization libraries and language extensions for OCaml, maintained by the Community"
<Yoric[DT]> s/Community/community
<olegfink> why don't you like Foundation?
<Yoric[DT]> olegfink: how would you use it>
<Yoric[DT]> olegfink: how would you use it?
<Camarade_Tux> Yoric[DT], seems good
<olegfink> Batteries is an ocaml library foundation
<Yoric[DT]> mmmhh....
* Yoric[DT] was deeply scarred by the JFC, a few years ago.
<Yoric[DT]> "Batteries: Library Foundation for OCaml, maintained by the community -- every library you need to start working with OCaml."
<olegfink> good, just do s/ --.*//
<thelema> "maintained by the community" -> "the community's"
<Yoric[DT]> "Batteries: the community's Library Foundation for OCaml. Every library you need to start working with OCaml."
Linktim has joined #ocaml
<thelema> library foundation ... every library -- cen re polish out one of the two "library"
<Yoric[DT]> "Batteries: The community is back with a vengeance. Drop your Python, your Java, surrender your Ruby, and start here." :)
<olegfink> ehm, the second sentence isn't true for me personally.
<Yoric[DT]> olegfink: Yeah, I agree it's a bit too ambitious.
<thelema> I like this last one.
<olegfink> why "every", why "start"?
<Yoric[DT]> Because people who don't "start" probably know if they need Batteries.
<Yoric[DT]> People who do start may require gentle nudging :)
<thelema> "start" is appropriate. If you've already got a big project in ocaml, adding batteries won't do too much.
<thelema> you've already built all the infrastructure you need for your project
<Yoric[DT]> "Batteries: Library Foundation for OCaml. A community-maintained base for your OCaml projects."
<olegfink> for me "start" means that when I master my ocaml skills I will have to find another library, as Batteries is just for the starters.
<Yoric[DT]> Fair enough.
<olegfink> Yoric[DT]: excellent!
<thelema> olegfink: start your project with batteries. If you outgrow it, that's okay.
<Yoric[DT]> Thanks.
<Yoric[DT]> thelema: yes, but it may suggest that if you're experienced, you'll never need Batteries.
<thelema> how about just the second sentence: "Batteries: a community-maintained base for your OCaml projecs."
<Yoric[DT]> Why not.
<olegfink> do you still maintain "Batteries Included" name?
<flux> that tells very little about the (envisioned) breadth of the libraries
<olegfink> if not, then the first sentence is effectively a fully quialified name
<olegfink> that's why it is probably still needed while it doesn't make much sense.
<ertai> Yoric[DT]: pong
<flux> maybe it doesn't need to be decided right now, let the idea linger around for a while ;)
<Yoric[DT]> ertai: hi, how do you do?
<Yoric[DT]> ertai: I had a Camlp4-related question.
<Yoric[DT]> ertai: how do I access comments?
<ertai> Yoric[DT]: fine, and you ?
<Yoric[DT]> ertai: fine, thanks.
<olegfink> as in 'batteries, library foundation for ocaml, is a community-maintained base for your OCaml projects'
<olegfink> short titile, full title, description
<Yoric[DT]> ertai: Disappointed at having a paper rejected despite very good comments.
<Yoric[DT]> olegfink: well, for the moment, the full title is "OCaml Batteries Included"
DRMacIver has quit [Remote closed the connection]
<ertai> Yoric[DT]: that's sad
<olegfink> ah
<ertai> Yoric[DT]: you can access comments by setting up a token filter
<Yoric[DT]> ertai: can't I do that from the Ast?
pango has joined #ocaml
<ertai> no they're not in the AST
<ertai> (that's far from easy to put them in the AST)
DRMacIver has joined #ocaml
<Yoric[DT]> ertai: ok.
<Yoric[DT]> In that case, Camlp4 may not be the right tool for what I have in mind.
<ertai> what do you have in mind?
<Yoric[DT]> Essentially, I have things such as [module Foo : sig (**Some ocamldoc comment*) ... end] and I want to turn that into [(**Some ocamldoc comment*)module Foo : sig ... end] .
<Yoric[DT]> Still as part of my quest to get ocamldoc, ocamlbuild and mlpack to play nicely together :)
<ertai> As a source-to-source translation
<Yoric[DT]> Yes.
<ertai> to be done once or in the build system
<Yoric[DT]> In the build system.
<Yoric[DT]> Essentially, I'm generating a .mli from a .mlpack .
<Yoric[DT]> To do that, I'm copying the contents of each .mli referenced by the .mlpack .
<ertai> yes I was guessing it
<Yoric[DT]> But the first ocamldoc comment inside a .mli should actually appear before [module MyFileName : sig].
<hcarty> ertai: Is there anything someone relatively ignorant of the OCaml internals, such as myself, can do to help resolve OCaml bugs #4495 and/or #4593? (camlp4 + toplevel)
<ertai> yes but you put the [module MyFileName : sig] part
<Yoric[DT]> ertai: indeed
<Yoric[DT]> I just need to move that part to after the first ocamldoc comment.
* ertai look at the BTS
<_zack> Yoric[DT]: in fact ,it seems you already have a git repo :-) http://git.ocamlcore.org/?p=batteries/batteries.git;a=summary
<ertai> hcarty: thanks but too late :)
<Yoric[DT]> _zack: do I?
<Yoric[DT]> _zack: gasp.
<Yoric[DT]> Well, now I only need to find out how to move :)
<Yoric[DT]> _zack: thanks.
jeddhaberstro has joined #ocaml
<_zack> Yoric[DT]: it's pretty easy, have a look at the git-svn doc
<ertai> _zack: wow, thanks
<_zack> Yoric[DT], ertai: in fact it's munga which did the work a while ago, but we all forgot about it ;)
<Yoric[DT]> :)
<Yoric[DT]> Thanks.
<ertai> hcarty: trying to identify the bug by looking at the code could help
<ertai> #4495 seems easier though
<hcarty> #4495 is the one I'm more concerned about, personally
<thelema> _zack: I made that batteries repo from the public SVN -- when I did another checkout with the developer SVN, I got a completely different git tree.
<hcarty> ertai: Any suggestions on where to look? I tried digging through the code myself a while ago, but didn't have any luck at the time.
<_zack> thelema: you sure you are talking about the repository I've mentioned? that one is empty with no commits at all
hkBst has quit [Read error: 104 (Connection reset by peer)]
* Yoric[DT] is considering splitting IO in two parts, one with just the definition of IO.input and IO.output and the other one with all the utilities.
<thelema> _zack: oops, I thought I saw a github url.
<Yoric[DT]> That will let us write [print] functions for each data structure, including those which already have some [write] function in [IO].
hkBst has joined #ocaml
<Yoric[DT]> _zack: so should I initialize it with git-svn?
<ertai> hcarty: I'm looking at camlp4/Camlp4Top/Top.ml and toplevel/toploop.ml
<ertai> hcarty: But not sure to find it
<_zack> Yoric[DT]: I presume the best way to do that would be to use git-svn on a local repo of yours and check that everything is ok, then git push everything to the repo on the forge
<thelema> Yoric[DT]: ok, we can remove UTF8.t = string
<ertai> Yoric[DT]: have you removed the no-op syntax extension of batteries (aka pa_mainfun)
<ertai> ?
<Yoric[DT]> thelema: well, I put you in charge of the item.
<thelema> Yoric[DT]: may I upload my git tree?
<Yoric[DT]> thelema: were to?
<Yoric[DT]> That new git?
<thelema> yes
hkBst has quit [Read error: 104 (Connection reset by peer)]
<thelema> I already use git-svn locally
hkBst has joined #ocaml
<Yoric[DT]> I suppose so, you're the one who knows about git and git-svn.
<Yoric[DT]> Just make sure that you put the whole repository, not just the trunk.
<thelema> yup, that's what I have.
<Yoric[DT]> Just give me 5 minutes to commit the changes to the Makefile.
<thelema> ok.
<Yoric[DT]> Ok, done.
<Yoric[DT]> ertai: no, why?
<thelema> hmm, no push URL...
<olegfink> does anyone recall what's the name of the simple imperative language that was proposed to be implemented in functional languages?
* thelema extrapolates from repo.or.cz
<Yoric[DT]> olegfink: perl?
<Yoric[DT]> (no, I don't recall)
<olegfink> it was sort of a challenge
<olegfink> it's called minim.
<olegfink> google to the rescue
<Yoric[DT]> :)
<olegfink> it has a camlp4 implementation, if anyone cares.
<Yoric[DT]> ertai: ?
<Yoric[DT]> olegfink: interesting
<thelema> http://forge.ocamlcore.org/tracker/index.php?func=detail&aid=22&group_id=1&atid=101 <- it looks like batteries git on ocamlcore doesn't work yet
<Yoric[DT]> :/
pierre- has joined #ocaml
<hcarty> ertai: As far as I can tell, the #4495 problem comes from the wrap function in camlp4/Camlp4Top/Top.ml. Something is generating a Sys_error exception and (I think?) gen_print in camlp4/Camlp4/ErrorHandler.ml is printing the error
<hcarty> ertai: I don't know if this helps at all
bluestorm has joined #ocaml
<ertai> hcarty: I came to the same conclusion and have changed something to this wrap function
* ertai have to recompile the whole thing and see the diff
<ertai> Yoric[DT]: pong
<Yoric[DT]> ertai: the answer to your question was "no, wy?"
<Yoric[DT]> ertai: the answer to your question was "no, why?"
<hcarty> ertai: I am happy to test here as well, if it would help
<ertai> your mainfun extension is equal to no-op
<ertai> you replace a grammar rule by almost the same one (but equivalent)
<Yoric[DT]> ertai: yes, but I add one operation in the meantime.
<bluestorm> (i was interested too in the signification of that mainfun thing)
<ertai> it's already legal OCaml to put an expression in a structre item
<bluestorm> (i assumed it was revised-syntax related, but did not read the thing carefully)
<ertai> Yoric[DT]: what operation let _ =?
<Yoric[DT]> ertai: in revised syntax, you can just write "5" as a structure item.
<Yoric[DT]> In original syntax, you can't.
<ertai> in original too :)
<Yoric[DT]> You can?
<_zack> Yoric[DT]: ehm, bad news in generating batteries docs here: http://paste.debian.net/19086/
<bluestorm> yes you can
<_zack> does it happen only to me?
<bluestorm> Yoric[DT]: that may require some ;; before, though
<bluestorm> (obvious parsing issues)
<Yoric[DT]> Gasp.
* Yoric[DT] feels foolish.
<ertai> yes in original syntax one recommend either ;; to the end the sentence or let _ = to begin it
<_zack> Yoric[DT]: it happened after 10 minutes or so of doc building
<Yoric[DT]> _zack: I'll take a look.
<Yoric[DT]> _zack: that's surprising.
<ertai> but that's just to avoid let x = e1;; e2 to be read as let x = e1 e2
<Yoric[DT]> Ok, I'll remove that extension.
<ertai> Yoric[DT]: thanks
<Yoric[DT]> Well, thank you.
<Yoric[DT]> Come to think about it, I had forgotten to install it.
<thelema> Yoric[DT]: I removed the installation code from SVN.
<Yoric[DT]> thelema: ah, ok.
<Yoric[DT]> thelema: please don't forget to update the ChangeLog when you do that kind of thing.
<thelema> Yes, that's something I need to get in the habit of.
<Yoric[DT]> Well, I've uploaded the ChangeLog for you.
<Yoric[DT]> _zack: building now...
<Yoric[DT]> _zack: we've had issues with ocamldoc since day 1.
<_zack> yup, it is not very "flexible"
<_zack> Yoric[DT]: i'm trying again building the doc as well, but even if it is transient I would be worried, because the buildds are for sure gonna hit that, thanks to Murphy
<Yoric[DT]> Yeah.
<Yoric[DT]> Perhaps by extending the stack?
* Yoric[DT] doesn't quite feel like rewriting ocamldoc for now.
<_zack> eh :), well, if you feel it is in ocamldoc sure, I was just wondering whether you felt that the non tail recursive code was in your code
marmotine has quit [Excess Flood]
GustNG1 has joined #ocaml
marmotine has joined #ocaml
<Yoric[DT]> Well, when my code actually starts working, it first displays message "[Final stage, generating html pages]".
<Yoric[DT]> s/starts working/starts its work/, that is
* Yoric[DT] just added a [flush_all ()] to be sure.
GustNG2 has joined #ocaml
<Yoric[DT]> _zack: can you try adding [ flush_all (); ] at line 703 for a test?
<_zack> line 703 of what?
<Yoric[DT]> build/odoc_generator_batlib.ml
<_zack> k
Mr_Awesome has quit [Read error: 110 (Connection timed out)]
* _zack is retarting the doc build again :)
* Yoric[DT] did that 6:43 minutes ago.
GustNG1 has quit [Read error: 60 (Operation timed out)]
GustNG has quit [Read error: 60 (Operation timed out)]
GustNG2 is now known as GustNG
middayc has joined #ocaml
<flux> what makes ocamldoc so slow?
<Yoric[DT]> I'm not sure.
<thelema> Yoric[DT]: what's the right way to add changelog entries? (I assume there's a way to generate the header automatically)
<Yoric[DT]> thelema: I'm doing it manually.
<Yoric[DT]> _zack: worked for me
<thelema> really...
<_zack> still going here ..
<ertai> thelema: emacs has a macro for it
* ertai don't use emacs
<thelema> great. I use emacs.
<_zack> finished here ... again with Stack overflow :(
<_zack> I'll try increasing the stack
GustNG1 has joined #ocaml
<Yoric[DT]> _zack: :/
<Yoric[DT]> And did it reach the stage where our code was used?
mattam has quit [Remote closed the connection]
<_zack> nope
<_zack> so ocamlbuild's fault
<Yoric[DT]> I think so.
<_zack> ocamldoc's fault
<Yoric[DT]> Yeah, that's what I understood :)
<Yoric[DT]> There's one List.fold_right in OCamlDoc's odoc_dep.ml
<ertai> and the list is too long ?
<Yoric[DT]> but I truly doubt it's the cause of the problem.
<Yoric[DT]> I don't know.
ygrek has quit [Remote closed the connection]
mishok13 has quit [Read error: 110 (Connection timed out)]
mishok13 has joined #ocaml
ygrek has joined #ocaml
GustNG has quit [Read error: 60 (Operation timed out)]
jeddhaberstro_ has joined #ocaml
jeddhaberstro has quit [Read error: 104 (Connection reset by peer)]
<hcarty> ertai: Any luck with the wrap function update?
<ertai> hcarty: not yet, it was not the right fix
<ertai> but I have another idea
<ertai> hcarty: I've made progress
<ertai> Now I have to convince me that I've not broken something worse
<hcarty> ertai: Thank you for working on this
<_zack> Yoric[DT]: with stack at 16M it worked
<flux> ah, if it only were as simple as grepping for fold_right, to find scaling problems due to non-tail-recursion :)
<flux> (but I suppose compiling ocamldoc with -g and getting the exception dump would reveal the problem)
<thelema> flux: on major platforms
<flux> thelema, so there are pltaforms where that's not supported, not even for byte code?
<ertai> hcarty: I have to go and so pause this camlp4 fix don't hesitate to recall me about it
<ertai> it would be nice to have it in 3.11 (not sure though)
<thelema> hmm, maybe everywhere for bytecode.
<hcarty> ertai: Thanks. I agree that it would be nice to have it for 3.11 if at all possible.
<Yoric[DT]> _zack: ok.
<Yoric[DT]> flux: well, grepping is always a start :)
<_zack> Yoric[DT]: ok, next issue for you :) http://paste.debian.net/19091/
<_zack> it's a top-level log
<thelema> hmm, more problems with Ref...
longh has quit [Client Quit]
DRMacIver has quit [Read error: 110 (Connection timed out)]
pierre- has quit [Read error: 104 (Connection reset by peer)]
bluestorm has quit ["Konversation terminated!"]
bluestorm has joined #ocaml
Palace_Chan has joined #ocaml
pierre- has joined #ocaml
<Yoric[DT]> _zack: I see.
* Yoric[DT] will try and deal with that after dinner.
sporkmonger has quit []
GustNG has joined #ocaml
mattam has joined #ocaml
GustNG1 has quit [Read error: 60 (Operation timed out)]
* thelema wrestles with make
* ygrek bets on make
<thelema> http://pastebin.ca/1225266 <- any help? the error is "make: *** No rule to make target `nth.png'. Stop.
GustNG1 has joined #ocaml
<olegfink> wow, metaocaml is a bless
<olegfink> the best way to code useless things elegantly
<thelema> aha, fixed
* Yoric[DT] is back.
<Yoric[DT]> olegfink: :)
<olegfink> my latest masterpiece generates a function which is an identity on first n natural numbers and raises an exception otherwise.
<olegfink> very elegant and absolutely useless.
<thelema> if x<n then x else failwith "too big"
<olegfink> hey, that's too simple
<olegfink> and doesn't use runtime code generation
<olegfink> s n = fun x -> if x = 1 then 1 else if x = 2 then 2 else ... else failwith "too big"
Snark has quit ["Ex-Chat"]
<thelema> yours beats mine for inefficiency
<Yoric[DT]> Much more elegant, of course :)
<olegfink> sure it is inefficient
<olegfink> but I still think metaocaml is a nice way to mask the following way of solving math problems: http://thedailywtf.com/Articles/Out-of-All-the-Possible-Answers.aspx
jeddhaberstro_ has quit []
<olegfink> actually, I wonder if a pa_memo-like thing can be written in metaocaml?
<olegfink> I tried but couldn't manage to get it right.
<Yoric[DT]> _zack: I've created a bug.
<_zack> Yoric[DT]: noted, thanks
GustNG has quit [Read error: 110 (Connection timed out)]
<_zack> Yoric[DT]: tiny bug report on the api doc, the main page has a link to Data.Containers.html, which does not exist
<_zack> I think it should be Data.html
<_zack> let me know if you want a bug report on this
<Yoric[DT]> Indeed.
<Yoric[DT]> No, I'll just fix that directly, thanks.
<_zack> k
<Yoric[DT]> Ok, I think we've found the problem for Ref.
<Yoric[DT]> We'll need a Findlib pro to give us a hand with that.
<Yoric[DT]> Would you be that person, _zack, by any chance?
<_zack> Yoric[DT]: I don't think I'm a findlib pro, but I can give a look
<Yoric[DT]> The interesting comments are in the last few lines.
<_zack> the problem is in the syntax subpackage?
<Yoric[DT]> Yes.
<Yoric[DT]> From Pierre Magistry experiment's it seems that [open Batteries] does the trick.
<Yoric[DT]> Which probably means that [Ref] can't be found because it's not opened.
<Yoric[DT]> Which probably means that [Ref] can't be found because [Batteries] is not opened.
<Yoric[DT]> (I don't know why)
<Yoric[DT]> In turn, [Batteries] is not opened because that particular part of the META file doesn't work.
<Yoric[DT]> (currently, we're using myocamlbuild.ml as a workaround)
<_zack> ah ok, so the problem is not specific of the toplevel, but you have an external workaround
<_zack> anyhow I'm running out of time for tonight, I'm leaving really soon now, I'll have a look tomorrow
<_zack> see you then
<Yoric[DT]> Ok, thanks.
<Yoric[DT]> See you then.
GustNG has joined #ocaml
<_zack> (BTW: the preliminary debian packages are *ready* I just need to fill in the proper descriptions!)
<Camarade_Tux> Yoric[DT], this morning I said I would probably be reinstalling godi and everything, guess what ?
<Yoric[DT]> What?
<Camarade_Tux> I wiped half of my hard drive :)
<Yoric[DT]> :/
<Camarade_Tux> /bin, most of /usr, ...
<Yoric[DT]> With GODI?
jeddhaberstro has joined #ocaml
<Camarade_Tux> no, with a shell script I was *debugging*
<Yoric[DT]> :/
<Camarade_Tux> the funny part is that music keeps on playing, vim is still working perfectly, xterm is just a bit useless as I can't run anything but that was fun :)
_zack has left #ocaml []
Camarade_Tux has quit []
Palace_Chan has quit ["Palace goes to sleep"]
Submarine has joined #ocaml
GustNG2 has joined #ocaml
GustNG1 has quit [Read error: 110 (Connection timed out)]
pierre- has quit [Success]
Mr_Awesome has joined #ocaml
GustNG has quit [Read error: 110 (Connection timed out)]
<det> What is the point of "val" and "method private" in OCaml objects ? Can't both of these be replaces with a let statement before the object declaration ?
<bluestorm> hm
<bluestorm> val are specific to the object instance
<bluestorm> while the let-declared values are not
<bluestorm> (same for method private, wich can depend on the val-declared values)
<det> specific to the object instance ?
<det> "let f () = let a = 0 in object method a = a end" "let f () = object val a = 0 method a = a end"
Amorphous has quit [Read error: 104 (Connection reset by peer)]
<det> I don't see the difference
<bluestorm> hm
<bluestorm> actually, let-declared values are specific to the instance too
<bluestorm> i was wrong
<flux> det, you can find a difference by playing with this class: class a = object(self) val a = ref 0 method private b = ref 0 method access_a = a method access_b = self#b end;;
<det> Ok, so private and val only make sense in the context of inheritence
<det> That makes sense, thanks.
<flux> hm, actually, my hint was based on faulty testing :)
<Yoric[DT]> grmph...
<flux> however, a#access_b is created each time
<Yoric[DT]> That Ref problem is complex.
<flux> what is the Ref-problem? some library building related issue with batteries?
<Yoric[DT]> Well, one of the modules of Batteries is [Ref].
<flux> atleast the documentation for it looks absurd
<Yoric[DT]> Now, if we don't put [Ref] in the list of modules which must be linked into the .cmxa, then linking examples against the .cmxa works but loading the .cma into the toplevel fails.
<bluestorm> (what's the difference between Ref and any other module ?)
<Yoric[DT]> If we put [Ref] in that list, linking examples against the .cmxa fails and something else fails in the toplevel.
<Yoric[DT]> bluestorm: probably just a matter of dependencies.
<Yoric[DT]> This module depends on nothing.
<Yoric[DT]> It's actually slightly more complex.
<Yoric[DT]> Essentially, both the .cma and the .cmxa work if I don't link Ref manually (because it has already been linked by some -pack) but when loading the .cma, the toplevel complains that it can't find Ref.
<flux> have you tried building a version of Batteries with nothing else than Ref?
<flux> and, assuming that works, continue with binary search ;)
<Yoric[DT]> :)
<Yoric[DT]> Well, it won't work.
<Yoric[DT]> I mean, to link only Ref, I need to rip apart large chunks of Batteries.
<flux> yes, but as you said, Ref depends on nothing, so atleast it should be relatively painless.. but I understand, there can be build-related issues making that difficult.
<flux> but, I'm off to sleep
<Yoric[DT]> 'night
<flux> happy hacking - or good night
Palace_Chan has joined #ocaml
Amorphous has joined #ocaml
middayc has left #ocaml []
hkBst has quit [Read error: 104 (Connection reset by peer)]
tar_ has joined #ocaml
root has joined #ocaml
root is now known as Guest38653
Guest38653 is now known as Camarade_Tux
thelema has quit [Read error: 110 (Connection timed out)]
<Camarade_Tux> Yoric[DT], it was not that terrible after all :)
bluestorm has quit [Remote closed the connection]
Linktim_ has joined #ocaml
thelema has joined #ocaml
Camarade_Tux has quit ["Leaving"]
ofaurax has quit ["Leaving"]
Linktim has quit [Read error: 113 (No route to host)]
ygrek has quit [Remote closed the connection]
root has joined #ocaml
root is now known as Guest74405
Guest74405 is now known as Camarade_Tux
jlouis has quit ["Leaving"]
_zack has joined #ocaml
GustNG2 has quit ["Leaving."]
Submarine has quit [Read error: 110 (Connection timed out)]
tar_ has quit ["byebye"]
kevas has joined #ocaml
<Yoric[DT]> Well, bugfixing will wait for tomorrow.
<Yoric[DT]> Good night everyone.
Yoric[DT] has quit ["Ex-Chat"]
Linktim_ has quit [Read error: 110 (Connection timed out)]
kevas has quit [Remote closed the connection]
ludyso has joined #ocaml
Camarade_Tux has quit ["Leaving"]
ludyso has quit [Remote closed the connection]
marmotine has quit ["mv marmotine Laurie"]
dobblego has joined #ocaml
lorof has joined #ocaml
lorof has quit [K-lined]
vbmithr has quit [Read error: 110 (Connection timed out)]
_zack has quit ["Leaving."]
pkl has joined #ocaml
Associat0r has quit []
pkl has quit [Remote closed the connection]
middayc has joined #ocaml
threeve has joined #ocaml