- Haskell-Cafe - Haskell.org

Glade Gtk2Hs Tutorial
by Hans van Thiel 27 Nov '06

27 Nov '06

Hello, The Haskell Gtk2Hs adaptation of the 'Developing Gnome Apps with Glade' tutorial for beginners is now also on http://eddy.writelinux.com/ Note there are translations there of the original version for C in French, Dutch, Spanish, Turkish and Korean. Native speakers of those languages who are interested could use these and check out the Haskell Gtk2Hs listings later. Thanks, Hans van Thiel

2 1

Priority Queue?
by Ken Takusagawa 27 Nov '06

27 Nov '06

Is there a Haskell implementation of an efficient priority queue (probably heap-based) out there that I can use? I do not see it in the GHC libraries.

5 4

Implementations of bit vectors
by Dominic Steinitz 26 Nov '06

26 Nov '06

You could also look at http://darcs.haskell.org/crypto/Data/LargeWord.hs but it would need modifying to produce a 33 bit word. Dominic.

1 0

re: Improving library documentation
by brad clawsie 26 Nov '06

26 Nov '06

don, i am glad you raised this point, i was going to write a note to this list soon with a similar request i would suggest comparing perldoc to haddock and other haskell documentation tools. generally speaking, i find that documentation for perl libraries is written as if the author is enthusiastic about users understanding it. there are typically detailed function descriptions, and often inlined examples. on the other hand i have found most haddock documentation to consist merely of function signatures. as a result i have to employ spotty supplemental references like the haskell ref at zvon.org (which i am sure is a copy of a well-known reference set floating around out there), which includes some examples, or google's code search with haskell support. furthermore, perldoc has querying capabilities (-f etc) which preclude having to surf around references, another great feature. regardless of format, well-written documentation will go a long way to forwarding haskell adoption. brad

1 0

Haddock question
by Luis Cabellos 26 Nov '06

26 Nov '06

Hello, I follow the How to Write a Haskell program from http://www.haskell.org/haskellwiki/How_to_write_a_Haskell_program, but I have a problem with Haddock. I use Windows XP, I when I install GHC + Haddock I have this warning output C:\code\haskell\test> runhaskell Setup.hs haddock Preprocessing executables for Test-0.0... Running Haddock for Test-0.0... Warning: cannot use package Test-0.0: ghc-pkg failed Warning: cannot use package base-2.0: does not exist.y $topdir\html\libraries\base There is something generated in dist\doc\html\test path but only the main function page but it don't has the comments from *.hs file. I use --prefix=c:\programs instead of --prefix=$HOME in configure. Because I am working on Windows. I don't use darcs, but I think that it's not the problem. -- Thanks a lot, Luis Cabellos

2 2

Improving library documentation
by dons＠cse.unsw.edu.au 26 Nov '06

26 Nov '06

People sometimes complain that the standard libraries need more documentation. Since we have so many eyes reading the haddock source, a low barrier way for people to improve the docs is needed. I've created a new wiki page, linked from the libraries page: http://haskell.org/haskellwiki/Improving_library_documentation This page is to track documentation improvments suggested by *you*. So if you're reading the haddocks, and notice something's not documented, or missing details, please record it on this page, so it can be improved and committed. -- Don

1 0

Implementations of bit vectors
by Maurício 25 Nov '06

25 Nov '06

Hi, Does anyone knows of a nice implementation of bit vectors? I like the functions I see in Data.Bits (rotate, shift etc.), but that only works with predefined sizes and I need a bit vector of exactly 33 bits. Thanks, Maurício

4 3

ghc 6.6 crash
by Tim Newsham 25 Nov '06

25 Nov '06

I just built GHC 6.6 on my linux Fedora Core box (athlon 700mhz). When I run it, it crashes. The 6.4 version works fine (which I also built). # ghc-6.6 Segmentation fault # gdb /usr/local/lib/ghc-6.6/ghc-6.6 GNU gdb Red Hat Linux (6.0post-0.20040223.19rh) Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "i386-redhat-linux-gnu"...(no debugging symbols found)...Using host libthread_db library "/lib/libthread_db.so.1". (gdb) run -B/usr/local/lib/ghc-6.6 Starting program: /usr/local/lib/ghc-6.6/ghc-6.6 -B/usr/local/lib/ghc-6.6 Error while mapping shared library sections: : Success. Error while reading shared library symbols: : No such file or directory. (no debugging symbols found)...(no debugging symbols found)...[Thread debugging using libthread_db enabled] [New Thread 16384 (LWP 20880)] Error while reading shared library symbols: : No such file or directory. Error while reading shared library symbols: : No such file or directory. Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 16384 (LWP 20880)] 0x00421d3b in __pthread_clock_gettime () from /lib/i686/libpthread.so.0 (gdb) where #0 0x00421d3b in __pthread_clock_gettime () from /lib/i686/libpthread.so.0 #1 0x00dc0331 in clock_gettime () from /lib/i686/librt.so.1 #2 0x0888fb4d in ?? () #3 0x00000003 in ?? () #4 0xfef22228 in ?? () #5 0xfef222e8 in ?? () #6 0x0888fe26 in ?? () #7 0x09462dd4 in ?? () #8 0x00000000 in ?? () (gdb) x/64 $esp 0xfef221a4: 0x13ea7d62 0x00000000 0x00000003 0x29a79300 0xfef221b4: 0x006ec840 0x00421cf0 0x00dc3ff4 0x00000003 0xfef221c4: 0x00000000 0xfef22208 0x00dc0331 0x00000003 0xfef221d4: 0x29a79300 0x00000000 0xfef22228 0x00dbc7ec 0xfef221e4: 0x09462d88 0x29a79300 0x00000000 0xffffffff 0xfef221f4: 0x006ec840 0x00000000 0x09462d88 0xfef222d0 0xfef22204: 0x00000002 0xfef222e8 0x0888fb4d 0x00000003 0xfef22214: 0xfef22228 0xfef222e8 0x0888fe26 0x09462dd4 0xfef22224: 0x00000000 0x0945c280 0x00000000 0x00000000 0xfef22234: 0x09462d88 0xfef222d0 0x0888f127 0x09462dd4 0xfef22244: 0x08962469 0xfef22268 0x0041d525 0xfef222d4 0xfef22254: 0xfef222d0 0x00000002 0x0888f23d 0xfef222d0 0xfef22264: 0x00000002 0xfef222d4 0x08888c73 0x089e2aa0 0xfef22274: 0x00000000 0x00000000 0x00000000 0x00000000 0xfef22284: 0x00000000 0xfef222d4 0x08890019 0xfef222d4 0xfef22294: 0xfef222d0 0xfef222d4 0x08889bde 0xfef222d0 # uname -a Linux box 2.6.5-1.358 #1 Sat May 8 09:04:50 EDT 2004 i686 athlon i386 GNU/Linux # ls -ld /lib/*pthread* -rwxr-xr-x 1 root root 95012 May 11 2004 /lib/libpthread-0.10.so* lrwxrwxrwx 1 root root 18 Jun 21 15:00 /lib/libpthread.so.0 -> libpthread-0.10.so* Tim Newsham http://www.thenewsh.com/~newsham/

1 1

Starting your own Haskell project: part 2
by dons＠cse.unsw.edu.au 25 Nov '06

25 Nov '06

We've expanded the wiki page on 'How to write a Haskell project', to include a complete walkthrough creating: * Darcs * Cabal * QuickCheck infrastructure. http://haskell.org/haskellwiki/How_to_write_a_Haskell_program Feedback welcome! -- Don P.S. It might even be useful to have a tool, haskell-project, which sets up all these files automatically.

8 16

Foldl
by Marco Comini 25 Nov '06

25 Nov '06

Donald Bruce Stewart: > zacara: > > i'd like to write a function that given a list like [1,2,3,4...] > > returns a list of couple where the first element is the corresponding > > element of the string, and the second is the sum of the previous > > elements. > Looks like a scanl (a prefix fold) -- the "of the previous > elements" is a clue that you're describing a scan: > > Prelude> :t scanl > scanl :: (a -> b -> a) -> a -> [b] -> [a] > > Prelude> scanl (+) 0 [1..4] > [0,1,3,6,10] > > You should be able to work out the rest from here (or rewrite it as a > fold). I agree Don, certainly with scanl one can solve the problem elegantly. [For ClareZako: it could be something like f (x:xs) = scanl acc (x,0) xs where acc (x,s) y = (y, s+x) ] On the contrary a solution in terms of a fold is indeed more tricky. Probably ClareZako was actually looking for a solution in term of a scan instead of a fold... otherwise would be interesting to know why he's willing to solve this problem in such a twisted way... don't you agree? Cheers, -- Marco.

2 1