Browse files


  • Loading branch information...
zli committed Jul 22, 2011
1 parent 724a52a commit ef84f31a8a2c88f18cb3ec0e159d3ab83b917760
Showing with 53 additions and 0 deletions.
  1. +53 −0 2011-07-22-irc.log
@@ -0,0 +1,53 @@
+08:53:19 <adrien> C++/g++ are making me think that ocaml's error message are perfect
+09:41:11 <sgnb> +1
+09:47:29 <adrien> well, at least ocaml doesn't throw some kind of syntax error when you're actually missing a declaration! (typed the name of another variable)
+09:48:28 <flux> on the other hand, c(++) compilers are able to spit out more than one error at a time :)
+09:55:55 <adrien> but several are usually related
+09:56:06 <adrien> I agree that's nice but maybe not always doable for ocaml
+09:56:32 <adrien> maybe it could do it independantly for different functions (but something more precise would be tough)
+11:39:53 <menace> is there besides the xemacs or the cli-client any other interactive interpreter of ocaml? i don't use emacs and the cli-client is somewhat uncomfortable :(
+11:40:31 <flux> enhtop and lwt's enhanced toplevel come to mind
+11:40:36 <flux> (disclaimer: I haven't tried these)
+11:40:38 <rwmjones> menace: not exactly sure what you mean, but how about 'rlwrap ocaml'
+11:41:57 <adrien> rlwrap and lwt's toplevel will provide the usual editing functionnalities and history
+11:42:32 <olasd> lwt's toplevel works quite fine in my experience
+11:48:09 <menace> thx
+11:49:08 <flux> what does lwt's toplevel bring into the game, btw?
+11:56:55 <olasd> context-sensitive tab-completion, history
+11:56:58 <olasd> colors
+11:57:24 <olasd> even though it may or may not be a feature ;)
+12:02:20 <adrien> I need to try it again: when I first tried it (quite a long time ago), it didn't seem to work well
+12:02:26 <adrien> might be pebkac too
+12:08:52 <flux> does it work under emacs in a useful way?-)
+12:09:04 <flux> context-sensitive tab completion seems useful
+12:09:09 <flux> although emacs' dynamic completion comes close
+12:10:41 <olasd> flux: well, under emacs i only tested it in the context of tuareg eval region
+12:11:18 <olasd> I'm not sure it will be that helpful
+15:10:23 <NaCl> adrien: now... to add ocamlfind
+15:11:15 <adrien> bit by bit =)
+15:59:03 <NaCl> ocaml-3.12.1-52.1.x86_64.rpm
+15:59:11 <NaCl> 52 recompiles of this thing
+16:13:02 <adrien> it's ok, you still haven't beaten my kernels: Linux yoda 3.0.0-rc7+ #79 SMP Fri Jul 22 07:27:20 CEST
+16:13:11 <adrien> (on a much longer timer period however)
+20:08:28 <whallz> hi
+20:08:34 <whallz> a quick question
+20:08:40 <whallz> i'm just starting
+20:09:02 <whallz> how can i return to the last command written in the cli?
+20:09:16 <whallz> the toplevel sys
+20:09:29 <whallz> up arrow does not work
+20:09:34 <whallz> is there a way to do so?
+20:10:09 <dsheets> rlwrap ocaml
+20:12:01 <whallz> dsheets: thanks
+20:12:09 <dsheets> np
+20:14:28 <whallz> i'm reading "Think OCaml, How to think like a computer scientist"
+20:14:38 <whallz> is there a better book?
+20:14:41 <dsheets> url?
+20:14:59 <whallz>
+20:15:20 <dsheets> i like this book:
+20:15:24 <whallz> i come from a more web-based background
+20:15:32 <whallz> php, javascript
+20:16:12 <whallz> great, i'll check it out, thanks
+20:16:53 <dsheets> i think the biggest philosophical change from those language to ocaml is the understanding of state and side-effects
+20:18:51 <dsheets> if you can change your perspective to think in terms of data structural transformations (functions) instead of state transformations (side-effects), you'll be 80% of the way there
+20:20:06 <whallz> cool, thank you dsheets
+20:20:12 <whallz> see ya

0 comments on commit ef84f31

Please sign in to comment.