#apache #archlinux #asterisk #centos #debian #gentoo #haskell #kde #kubuntu #lisp #math #mysql #perl #python #ruby-lang #rubyonrails #suse #ubuntu #vim #wikipedia 0 1 2 3 4 5 6 7 8 9 10 11 12 13
Top Prev 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425 3426 3427 Next
#haskell
<dons> :-)
<dons> if we can get jhc to pass, say, 1/4 of the tests
<dons> then i'll include it
<thorat> iron32: according to web page it supports emacs as well
<sorear> dons: if we include it now JohnMeacham might just fix the assertion failure
<dons> true. i'd need to build it first. 'll try to do that tonight
<sorear> ah.
<LoganCapaldo> refactor haskell code?
<sorear> in that case there is yet hope, I gave up halfway throug.
<iron32> Well I have had trouble with the Haskell mode for Xemacs under both windows and cygwin
<SamB> LoganCapaldo: is it such a strange concept?
<LoganCapaldo> My brain can barely handle a non-messy haskell function
<LoganCapaldo> much less one so messy that it needed to be refactored
<nmessenger> I usually just change stuff and let the compiler point out all the dependencies
<dons> some of these runtime numbers look quite small sorear
<iron32> Well refactoring can be simple as renaming things
<dons> we're just measuring only cpu time for what now?
<iron32> or more complex
<sorear> CPU time
<sorear> cumulative cpu time
<SamB> LoganCapaldo: heh
<nmessenger> a common "refactoring" you learn is giving names to confusing sub-expressions in a where clause
<iron32> :t ++
<allbery_b> iron32: haskell-mode is nt well tested under xemacs, and under some versions xemacs deals *very* poorly. I updated my xemacs and got something that mostly works
<lambdabot> parse error on input `++'
<tautologico> has anyone here head "the algebra of programming" by bird and de moor?
<allbery_b> :t (++)
<lambdabot> forall a. [a] -> [a] -> [a]
<iron32> allery_b : well I hacked on it a while got something mostly working but decdided that the eclipse FP plugin gave me more out of the box for less hassle
<monochrom> I have read "the algebra of programming" by bird and de moor.
<iron32> monochrom any good
<tautologico> monochrom: is it good? I can't find it anywhere, only copies priced over a hundred bucks
<allbery_b> eclipse is painfully slow on my g4 macmini :)
<dufflebunk> s/on my g4 macmini :)/:(/
<dufflebunk> fixed that for you
<emu> someone needs to write shim-vim
<sorear> dons: how much do I need to edit to add a new benchmark? just imaginary/Makefile?
<iron32> allery_b : Well runs great on my Pentium 3 GHZ
<monochrom> It is deep. It is about deriving implementation from specification. To get an idea whether you will want it or not, try reading some papers of Bird first. Look for those on "greedy algorithms" and "dynamic programming", the book is an organized version of those topics.
<tautologico> monochrom: I know, squiggol, bird-meertens formalism, isn't it? I wanted to know exactly if the book offers something for someone who has read some of the papers
<dons> sorear: right.
<dons> sorear: what kind of benchmark? hello world?
<sorear> dons: ok, than my patch should be valid
<sorear> yeah, pretty much
<sorear> it highlights the 0.65s for ghci ... slowest impl by a longshot
<sorear> guess which of my compilers is fastest for nop :)
<sorear> jhc 0.000 seconds (NaN x)
<sorear> followed by a three-way 0.020 tie yhc/ghc/ghc-old
<dons> ah, i see why nhc98 is winning integer. no output, nhc98 does nothing
<dons> the others are all stricter ;)
Previous Page Next Page