August 2002 - Haskell

RE: 3 `\x y -> x + y` 4 `\x y -> x + y` 5: Wrong version pasted
by Rijk J. C. van Haaften 22 Aug '02

22 Aug '02

I pasted the wrong version into the mail. This is the final one: module InfixFunction where infixl 0 $- infixr 0 $+ -- for left-associative functions ($<) :: a -> (a -> b) -> b ($<) x f = f x -- for right-associative functions ($>) :: a -> (a -> b) -> b ($>) x f = f x example1 :: Int example1 = -- 3 `\x y -> x + y` 4 3 $< (\y x -> x + y) 4 example2 :: Int example2 = -- 1 `\x y -> x + y` 2 `\x y -> x` 3 1 $> (\y x -> x + y) 2 $> (\y x -> x + y) 3 Rijk-Jan van Haaften

2 1

RE: 3 `\x y -> x + y` 4 `\x y -> x + y` 5
by Rijk J. C. van Haaften 22 Aug '02

22 Aug '02

>module InfixFunction where > >-- This is a flipped $ >infixl 0 $- >($-) :: a -> (a -> b) -> b >($-) x f = f x > >example1 :: Int >example1 = > -- 3 `\x y -> x + y` 4 > 3 $- (\y x -> x + y) 4 > >example2 :: Int >example2 = > -- 1 `\x y -> x + y` 2 `\x y -> x` 3 > 1 $- (\y x -> x + y) 2 $- (\y x -> x + y) 3 Note the swap in argument order in the lambda abstractions! Rijk-Jan van Haaften

1 0

3 `\x y -> x + y` 4 `\x y -> x + y` 5
by Rijk J. C. van Haaften 22 Aug '02

22 Aug '02

Hello everybody, Every now and then I hear people complain that writing expressions like in the subject line is impossible, though desirable in some cases. Today, I ran into a situation where I really needed that feature, because I were working with a code generator (UU_AG) that allows just infix functions at some point. So I thought about it for a minute and then I solved it this way. Maybe it's useful to others as well, so I post it here. module InfixFunction where -- This is a flipped $ infixl 0 $- ($-) :: a -> (a -> b) -> b ($-) x f = f x example1 :: Int example1 = -- 3 `\x y -> x + y` 4 3 $- (\y x -> x + y) 4 example2 :: Int example2 = -- 1 `\x y -> x + y` 2 `\x y -> x` 3 1 $- (\y x -> x + y) 2 $- (\y x -> x + y) 3 Rijk-Jan van Haaften

1 0

Re: Word8-Based IO
by Ashley Yakeley 22 Aug '02

22 Aug '02

At 2002-08-21 13:03, Alastair Reid wrote: >Can this discussion be moved to the i18n list please? It was originally on Libraries, as it didn't seem to concern internationalisation per se... -- Ashley Yakeley, Seattle WA

3 2

Re: Word8-Based IO
by Wolfgang Jeltsch 21 Aug '02

21 Aug '02

On Wednesday, 2002-08-21, 06:55, CEST, Ashley Yakeley wrote: > [...] > * System.IO: addition of new Word8-based functions > > hGetOctet :: Handle -> IO Word8 > hLookAheadOctet :: Handle -> IO Word8 > hPutOctet :: Handle -> Word8 -> IO () > hPutArray :: Handle -> [Word8] -> IO () > hLazyGetArray :: Handle -> IO [Word8] I wouldn't speak of arrays here because the functions don't operate with arrays but with Word8 lists. > ...as per hGetChar, hLookAhead, hPutChar, hPutStr and hGetContents. To achieve consistency, it would be nice to rename some of the Char-based functions according to their Word8 counterparts and provide the same functions under their old names for compatibility. > hGetArrayBlock :: Handle -> Int -> IO [Word8] > hGetArrayReady :: Handle -> Int -> IO [Word8] See above for the problem with "array". > [...] > hSetFileSize :: Handle -> Integer -> IO () > > There's currently no way to set the size of a file. hSetFileSize would > truncate the file or pad with zero octets. Maybe it would be good to split this into two functions according to the two functionalities: hEnlargeFile :: Handle -> Integer -> IO () hCutFile :: Handle -> Integer -> IO () These functions would have to check if the Integer argument is less than the current file size (hEnlargeFile) or greater than the current file size (hCutFile) and throw an exception in this case. > [...] > sendTo :: Socket -> [Word8] -> SockAddr -> IO Int > recvFrom :: Socket -> Int -> IO ([Word8], Int, SockAddr) > send :: Socket -> [Word8] -> IO Int > recv :: Socket -> Int -> IO [Word8] Isn't "recv" an ugly abbreviation? > -- > Ashley Yakeley, Seattle WA Wolfgang

2 1

Re: Word8-Based IO
by Wolfgang Jeltsch 21 Aug '02

21 Aug '02

On Wednesday, 2002-08-21, 09:52, CEST, Ketil Z. Malde wrote: > [...] > I don't particularly like 'octet' (probably because I find it a bit > pedantic, and it reminds me too much of committee decisions), and > would prefer hGetWord8 or even hGetByte/hGetWord. hGetWord8 sounds better to me than hGetOctet, at least as long as we have nothing like type Octet = Word8. > Also, I find 'Array' a bit uncomfortable -- it's a list, isn't it? As I also said in my earlier mail. > But why not simply: > > hPutOctet (or Word, Word8, Byte) :: Handle -> Word8 -> IO () > hPutOctets (hPutWords, Bytes) :: Handle -> [Word8] -> IO () > and > hGetOctet :: Handle -> IO Word8 > hGetOctets :: Handle -> IO [Word8] -- hGetContents-alike It's nice to use just the plural of octet. But what I dislike a bit is that hPutOctets and hGetOctets don't really correspond to each other because hGetOctets isn't the "inversion" of hPutOctets. I mean, hGetOctets won't generally read exactly those octets an earlier invocation of hPutOctets put but occasionally more than these. > (I'd rather annotate the strict versions, lazy is usually the default) I think so, too. > I'd also like a 'readFile'-like function for octets, I find my > programs do a lot of their IO by lazily reading files, and being able > to bury all the handle-stuff is a good thing, IMHO. Good point. > [...] Wolfgang

1 0

Papers accepted for 'Functional and Declarative Languages in Education 2002'
by S.J.Thompson 21 Aug '02

21 Aug '02

Details of the papers accepted for the workshop on Functional and Declarative Languages in Education 2002 October 7, PLI, Pittsburgh, USA are available at http://www.cs.ukc.ac.uk/people/staff/sjt/fdpe02/accepted.html Simon Thompson

1 0

GpH/GdH Parallel & Distributed Haskell Research
by Phil Trinder 21 Aug '02

21 Aug '02

Prompted by Simon PJ, here's a bulletin on GpH/GdH parallel and distributed Haskell research at Heriot-Watt and St Andrews Universities. Kevin Hammond has recently had an EPSRC project funded to use GpH as parallel coordination layer for large GAP (symbolic algebra) computations. We're also part of an EU framework VI Network bid on parallel symbolic computation. Jan Nystrom will join us at Heriot-Watt in November from Upsala as the RA on a new EPSRC/Motorola Research Labs project investigating Erlang & GdH for distributed telecoms applications. We're maintaining collaboration with the Eden (another parallel Haskell) groups at Marburg in Germany and Madrid, funded by Britsh Council ARC and Accion Integradas respectively. Hans Wolfgang Loidl has just returned to Germany after completing his 3-year Post-Doc. He's been working on improving the architecture independence of GpH, and participates part time. There are 5 research students working on GpH and GdH: - Robert Pointon (Finishing PhD) has designed & implemented GdH and is now evaluating it. - Alvaro Rebon (Finishing PhD) is implementing the granularity anaylsis proposed in Hans Wolfgang Loidl's thesis as a free standing tool. - Andre Du Bois (Entering 2nd year PhD) is designing and planning to construct a mobile Haskell by extending GdH. - Abyd Al Zain (Starting PhD) is designing and planning to construct a GRID implementation for GpH on networks of HPCs. The key problems are to extend GpH's existing dynamic adaption mechanisms to the hierarchical, heterogeneous and shared architecture. - Mustafa Aswad (Completing MPhil) is investigating developing GpH programs for multiple architectures, based on a GpH gene-sequence alignment program. More details including a raft of publications is available from http://www.cee.hw.ac.uk/~dsg/gph/ http://www.cee.hw.ac.uk/~dsg/gdh/ http://haskell.cs.yale.edu/communities/ Phil -------------------------------------------------- Phil Trinder Department of Computing and Electrical Engineering Heriot Watt University Riccarton Edinburgh, EH14 4AS E-mail: trinder(a)cee.hw.ac.uk Teleph: +44 (0)131 451 3435 Depart: +44 (0)131 451 3328 Fasmly: +44 (0)131 451 3327 Intrnt: http://www.cee.hw.ac.uk/~trinder

1 0

Re: More suitable data structure needed
by Ashley Yakeley 21 Aug '02

21 Aug '02

At 2002-08-21 01:14, Dr Mark H Phillips wrote: >But what is the best way to represent this in Haskell? (Clearly >I can't do exactly this, because Haskell requires all list elements >to be of the same type.) Create a new type that can allow either a pair or a list of pairs, and use a list of that. -- Ashley Yakeley, Seattle WA

1 0

zipWith, zipWith3, zipWith4.... looks gawky, IMHO
by Coeus＠gmx.de 20 Aug '02

20 Aug '02

Hi all. I'm new to this mailing list. (and still a relative newbie in Haskell - learning GraphicsLib) Because the Wish List did not work (maybe it is my browsers fault), I now write it to this list. I found the zipWithN functions in the standard libs, but imho it would be much more comfortable to use operators like in this example: zipWith6 :: (a->b->c->d->e->f -> g) -> ([a]->[b]->[c]->[d]->[e]->[f] -> [g]) zipWith6 f as bs cs ds es fs = f :< as >< bs >< cs >< ds >< es >< fs infixl 123whatever (:<), (><) -- lower priority than (++) (:<) :: (a->b) -> [a] -> [b] (:<) = map (><) :: [(a->b)] -> [a] -> [b] (><) = zipWith id or better: (><) :: [(a->b)] -> [a] -> [b] (><) (f:fs) (x:xs) = (f x) : (fs >< xs) (><) _ _ = [] The fibs example would now look like this: take 10 fibs where fibs = 1 : 1 : ( (+) :< fibs >< tail fibs ) instead of take 10 fibs where fibs = 1 : 1 : zipWith (+) (fibs) (tail fibs) What does you suggest/what's your oppinion to this? (maybe other operators?) - Marc

2 2