- Haskell-Cafe - Haskell.org

RE: simulating dynamic dispatch
by Simon Peyton-Jones 04 Apr '03

04 Apr '03

(Taking this off to haskell-cafe) It's not unreasonable for it to be ambiguous. The argument 'f' says *forall b*. The cast will cast to a *specified* type. But you aren't specifying the type. If you passed a less polymorphic 'f' it might be ok. Simon | -----Original Message----- | From: Hal Daume III [mailto:hdaume@ISI.EDU] | Sent: 21 March 2003 22:29 | To: Simon Peyton-Jones | Cc: oleg(a)pobox.com; haskell(a)haskell.org | | Excellent paper! I have started using some of the techniques already! | | But, back to the real question :). I now no longer think such a thing is | possible. Essentially what I want (given 'cast :: Typeable a, Typeable b | => a -> Maybe b' [1]), is to write something like: | | test :: forall a r . Typeable a => | (forall b . (Typeable b, Foo b) => b -> r) -> | a -> Maybe r | test f a = liftM f (cast a) | | but then 'b' is ambiguous. I'm not sure there is any way around | this. Can someone help? | | - Hal | | [1] is there any reason why parens are requires around class | contexts? i've always found this odd... | | -- | Hal Daume III | hdaume(a)isi.edu | "Arrest this man, he talks in maths." | www.isi.edu/~hdaume | | On Fri, 21 Mar 2003, Simon Peyton-Jones wrote: | | > You might also find the 'cast' function in Section 3 of "Scrap your | > boilerplate" useful. | > http://research.microsoft.com/~simonpj/papers/hmap/ | > I'm not certain, but it has the right smell. | > Simon | > | > | -----Original Message----- | > | From: oleg(a)pobox.com [mailto:oleg@pobox.com] | > | Sent: 21 March 2003 04:19 | > | To: hdaume(a)ISI.EDU; haskell(a)haskell.org | > | Subject: Re: simulating dynamic dispatch | > | | > | | > | > i'm hoping to be able to simulate a sort of dynamic dispatch based | > on | > | > class instances. | > | | > | It seems you want to dispatch based not on a type but on the | > | constraint of a type. | > | | > | You code almost worked. Here's the a bit updated and working version. | > | | > | class Foo a where { foo :: a -> Bool } | > | class Bar a where { bar :: a -> Bool } | > | | > | data FB = forall a . Foo a => MkFoo a | forall a . Bar a => MkBar a | > | | > | instance Foo FB where | > | foo (MkFoo x) = foo x | > | | > | instance Bar FB where | > | bar (MkBar x) = bar x | > | | > | -- some instances for the test | > | instance Foo Int where | > | foo x = x == 0 | > | | > | instance Bar Char where | > | bar x = x == 'a' | > | | > | | > | test x = case x of | > | (MkFoo a) -> Just $ foo a | > | (MkBar a) -> Just $ bar a | > | -- _ -> Nothing | > | | > | | > | -- *Main> test $ MkFoo (0::Int) | > | -- Just True | > | -- *Main> test $ MkFoo (10::Int) | > | -- Just False | > | -- *Main> test $ MkBar 'a' | > | -- Just True | > | -- *Main> test $ MkBar 'b' | > | -- Just False | > | | > | _______________________________________________ | > | Haskell mailing list | > | Haskell(a)haskell.org | > | http://www.haskell.org/mailman/listinfo/haskell | >

1 0

GISTalk - New GIS Forum!
by Tim Johnson 03 Apr '03

03 Apr '03

WaterTalk, StormTalk, and SewerTalk http://www.haestad.com/frm_gtk_030321 You're invited to join GISTalk, a new online forum for the Geographic Information System (GIS) community. GISTalk follows the success of our WaterTalk, StormTalk, and SewerTalk forums, providing GIS professionals a place to ask questions and raise issues regarding geographic and geospatial information systems and receive feedback from GIS experts across the world. GISTalk is your one-stop source for GIS-related information. With over 10,000 engineers active on the forums, you're sure to find answers and opinions for virtually any project you are working on. Professionals interested in joining any or all of the discussion lists can subscribe by visiting the link below... http://www.haestad.com/frm_gtk_030321 This message is intended for civil engineers and water resource professionals. If it has reached haskell-cafe(a)haskell.org in error, reply to this message with a subject line of "stop". LSID: 12274-2076416

1 0

Tokenizing Strings
by jamesd＠mena.org.au 02 Apr '03

02 Apr '03

I have a string that needs to be split/tokenized based on a delimiter. This can easily be accomplished using 'break' if the delimiter is only 1 character (i.e. break isSpace "this is a string"), but I can't see any way of using this for a delimiter with multiple characters. in this case, I have a string containing multiples fields seperated by *two* blank lines (\n\n). I can't just break on the newline character, as single newline characters can be found inside each field. any idea how I can do this without too much hassle? -James (jamesd(a)mena.org.au)

4 4

how to use POSIX function utime?
by David Roundy 28 Mar '03

28 Mar '03

I'm trying to figure out how to use the POSIX function utime. I'm pretty much stumped at the moment, largely because I can't figure out how to do anything useful with ClockTime variables. I can't see any way to make use of GHC.Posix.utime, as it gives a binding to the utime function, but the argument is of type Ptr (), and I can't see how to make this to a pointer to the contents of a pair of ClockTimes. I had thought that I could just write my own utime wrapper in C, but I'm still stumped on getting anything out of a ClockTime! The problem is that I would like my code to be portable, so I can't assume a ClockTime is a UInt32 (or that a time_t is a 32 bit int). Any ideas how I can get around this? The only idea I have at the moment is to define a MyClockTime which is always a UInt64 and then do in my C wrapper I could cast back and forth from time_t to an int64 type. Ugh. -- David Roundy http://www.abridgegame.org

2 2

Re: how to track down infinite loop?
by David Roundy 26 Mar '03

26 Mar '03

On Wed, Mar 26, 2003 at 07:40:30AM -0800, Andy Moran wrote: > On Wednesday 26 March 2003 07:23 am, David Roundy wrote: > > I would try using Debug.Trace.trace. It takes a string and an expression, > and returns the expressions while outputting the string to stderr (I > think). You can choose your own mnemonics to prevent the truncation > problem. Thanks for the hint! I like trace, strongly reminiscent of the good old technique of sprinkling printfs around. :) It turned out the problem was running sequence on an infinite list of type [Maybe a]. In retrospect obvious, but sort of hard to notice (especially since the infinite list is created in a different module from the sequence call. Fortunately, in this case it is safe to just truncate the list to some large size (since the function in question is only used in the quickChecking code, and not in the actual code). -- David Roundy http://civet.berkeley.edu/droundy/

1 0

how to track down infinite loop?
by David Roundy 26 Mar '03

26 Mar '03

Hello. I've got an infinite loop (I think) which I recently introduced to my code (the visible symptom is stack space overflow), and am wondering if there is an easier way to figure out where it might be. The problem being that I've made a rather large number of changes (all of which are mutually dependent). In the past, I've used ghc's heap profiling to track this sort of thing down, as I can pretty easily trace from that what functions were called in the infinite loop. The problem I'm now having is that the string describing the call stack is truncated too much, so I can't see who the real culprit is. I know the best thing would be to go over all my changes with a fine-toothed comb, but it would be nice to have an idea where to look. Actually, it's not all that much new code, but I usually find it hard to see a bug that I've written myself... Any ideas for tricks to see where a program is looping indefinitely? I'm sure I can track down this bug pretty easily, but is seems like this is something one really ought to be able to do... -- David Roundy http://civet.berkeley.edu/droundy/

2 1

RE: Asynchronous exceptions and "resume"
by Simon Marlow 26 Mar '03

26 Mar '03

> Hi all, is there a way, or is it planned to, or has anyone published > articles on... resuming from asynchronous exceptions? > > I mean: it would be useful there was a > > suspend :: ThreadID -> IO () > > where the result is the remaining computation of the other > thread, wich > one could forkIO again, or simply run in the current thread. This is fairly easy to implement in GHC. Simple call Schedule.c:raiseAsync() with NULL for the exception, and make it return the final closure after it has stripped the thread's stack. Entering that closure should restart the thread's computation exactly where it left off. If the thread was blocked on something (eg. takeMVar), then it should restart the blocking operation. The types aren't quite right: the closure you get back will behave like an application of unsafePerformIO, so you will need to wrap it using eg. Control.Exception.evaluate to get an IO action. I'm sure there are plenty of cool uses for this... Cheers, Simon

1 0

Asynchronous exceptions and "resume"
by Nick Name 26 Mar '03

26 Mar '03

Hi all, is there a way, or is it planned to, or has anyone published articles on... resuming from asynchronous exceptions? I mean: it would be useful there was a suspend :: ThreadID -> IO () where the result is the remaining computation of the other thread, wich one could forkIO again, or simply run in the current thread. If this is not possible, it would still be useful that the IO action returned by "suspend" would only "resume" the other thread, as an alternative to a brutal "kill" wich destroys any work done by the other thread. The first form seems interesting when one wants to fork off a long computation, but wants the ability to wait for it again if necessary; pointers to articles are always appreciated. Vincenzo

1 1

(no subject)
by Mrs.Sese-seko 26 Mar '03

26 Mar '03

Dear friend, I am Mrs. Sese-Seko widow of late president Mobutu Sese-Seko of zaire? now known as democratic republic of congo (drc). I am moved to write you this letter, this was in confidence considering my present circumstance and situation. I escaped along with my husband and two of our sons Theophilus and Basher out of Democratic Republic of Congo (DRC) to Abidjan, cote d'ivoire where my family and I settled, while we later moved to settled in morroco where my husband later died of cancer disease. However due to this situation we decided to changed most of my husband's billions of dollars deposited in swiss bank and other countries into other forms of money coded for safe purpose because the new head of state of (dr) Mr, laurent kabila has made arrangement with the swiss government and other european countries to freeze all my late husband's treasures deposited in some european countries. hence my children and I decided laying low in Africa to study the situation till when things gets better. Now that President kabila is dead and the son taking over (Joseph kabila). one of my late husband's chateaux in southern france was confiscated by the french government, and as such I had to change my identity so that my investment will not be traced and confiscated. I have deposited the sum Thirty Million United State Dollars(US$30,000,000,00.) with a security company , for safekeeping. The funds are security coded to prevent them from knowing the content. What I want you to do is to indicate your interest that you will assist us by receiving the money on our behalf. Acknowledge this message, so that I can introduce you to my son (Theophilus) who has the out modalities for the claim of the said funds. I want you to assist in investing this money, but I will not want my identity revealed. I will also want to buy properties and stock in multi-national companies and to engage in other safe and non-speculative investments. May I at this point emphasise the high level of confidentiality, which this business demands, and hope you will not betray the trust and confidence, which I repose in you. in conclusion, if you want to assist us , my son shall put you in the true picture of the business, tell you where the funds are currently being maintained and also discuss other modalities including remuneration for your services. For this reason kindly furnish us your contact information, that is your personal telephone and fax number for confidential purpose. Best regards, Mrs,sese-seko

9 9

Re: African money
by Marcus Lindblom 26 Mar '03

26 Mar '03

From: John Peterson <peterson-john(a)cs.yale.edu> > > If the spam storm gets too heavy on haskell-cafe we can switch to > moderator approval as on the main Haskell list. We have the spam > assassin available at haskell.org but I've been reluctant to turn it > loose on the lists. Maybe the time has come ... Many lists use moderator approval on non-list members. Maybe spam assassin could be turned loose on just those non-member mails, and have those passing through be moderator reviewed? /Marcus

2 1