`Funny global thing' in ghc-4
Please, what the folowing `Funny global thing' might mean in ghc-4 ? make source/pol/factor/Pfact0_.o /usr/ghc/4/bin/ghc -c source/pol/factor/Pfact0_.hs -fglasgow-exts -optC-fallow-overlapping-instances -optC-fallow-undecidable-instances -fvia-C -K1500k -syslib misc -cpp -recomp -hi-diffs -iexport -Iexport -odir export -fno-warn-overlapping-patterns -ohi export/Pfact0_.hi -H1k -Onot ghc: module version changed to 1; reason: no old .hi file Funny global thing?: cdX7_btm: Funny global thing?: cdZf_btm: -- Sergey Mechveliani [EMAIL PROTECTED]
Re: `Funny global thing' in ghc-4
Your question was already on the mailing list before: Simon Marlow wrote: This means you've got a pretty complex bit of code that needed a bitmap with more than 32 entries to describe a stack frame. Cool :-) I wrote: can the "Funny global thing?" message be ignored? (It does not crash the compiler.) Simon Marlow wrote: Yep, the warning can be ignored. Best wishes, Jan ___ '---|-- | __, _ _ EMail: [EMAIL PROTECTED] | / | / |/ | WWWeb: http://www.uni-passau.de/~laitenbe/ |/\_/|_/ | |_/ /| Laitenberger --(-|-- \|
RE: GHC-3.03-current bugs
Sven Panne [EMAIL PROTECTED] writes: ... But something really strange is going on here: -- Foo.hs - module Foo where malloc :: IO Int malloc = _casm_ ``%r = 42;'' -- Bar.hs - module Bar where import Foo blah :: IO Int blah = malloc --- panne@liesl:/tmp ghc -O -funfold-casms-in-hi-file -fglasgow-exts -c Foo.hs -o Foo.o ghc: module version changed to 1; reason: no old .hi file panne@liesl:/tmp ghc -O -funfold-casms-in-hi-file -fglasgow-exts -c Bar.hs -o Bar.o -dshow-passes *** Reader: *** Renamer: panic! (the `impossible' happened): IdInfo parse failed malloc Thanks for a fine report, sorry for taking so long in replying. It's a buglet, which somehow managed to get past our entire ISO 9001 Quality Assurance Department. Fixed. --Sigbjorn
RE: H/Direct-0.12 buglets
Sven Panne [EMAIL PROTECTED] writes: To install H/Direct-0.12 on Linux, two things had to be changed: * Move an #include to a more probable place (just guessing...): Thanks, the fix was almost right, the "comPrim.h" include should appear inside the __CYGWIN32__ #ifdef, since it's a replacement header file that allows us to continue using gcc/egcs. Some remarks regarding the examples: * [pure]int rand(void); ??? I really hope that rand is *not* a pure function... :-} thx, fixed. * int strncmp([in,string]char* s1,[in,string]char* s2,[in]int n); Shouldn't this read: int strncmp([in]FastString s1,[in]FastString s2,[in]int n); Yep, that would make it more useful. --Sigbjorn
Re: Haskell 98 progress
Hi Simon, you spake of simple-context restriction: My default position is not to change. Question: who, apart from Ralf, has actually tripped over the lack of contexts of the form (C (a t1 .. tn)) in Haskell 1.4? *raises hand and jumps up and down* Me, mememe! I provided you with an example of this many moons ago (unless I'm going crazy(ier)). I'm moderately agog at your default default: it _only_ restores principal types and fixes some programs, so that's not a big enough reason to do it? OK, that's a tad unfair. But assuming you can show the change is at any rate sound, then at worst it's a conservative, harmless extension, unless I'm missing something huge and vast and horrible (never unlikely). Slan, Alex.
RE: MonadZero (concluded)
At 01:58 -0800 1998/11/09, Simon Peyton-Jones wrote: Following many protests, the right thing to do seems to be to move MonadPlus to the Monad library. Specifically: class Monad m = MonadPlus m where mzero :: m a mplus :: m a - m a - m a It seems me that the MonadPlus is just a monad whose algebras are monoids. So perhaps it should be renamed to reflect that fact. Hans Aberg * Email: Hans Aberg mailto:[EMAIL PROTECTED] * Home Page: http://www.matematik.su.se/~haberg/ * AMS member listing: http://www.ams.org/cml/
A short study of fuzzy oscillator
I have posted a literate Haskell module "Fuzzy_oscillator" in our collection of Haskell modules: http://www.numeric-quest.com/haskell/ You may also download it as a gzipped bundle (containingg 9 plots) according to downloading instructions on that page. Credits go to Gary Meehan and Mike Joy from University of warwick for their module Fuzzy.hs. Summary --- This module presents a short study of one-degree-of-freedom (1DOF) oscillator subjected to resonant sinusoidal excitation force and whose vibrations are controlled by a family of fuzzy controllers: + Velocity controller + Displacement controller + Displacement-velocity controller Haskell simulations demonstrate that each controller performs extremely well -- reducing resonant amplitudes to comparatively small values, even though none of the controllers is thoroughly tuned and our choice of kinematic fuzzy subsets is very simplistic. This suggests that fuzzy logic offers very powerful tools to vibration control engineers; no special skills for solving nonlinear equations are required and purely experimental approach would suffice to design good controllers for such applications. The obvious physical realizations of fuzzy controllers are active devices that employ sensors, processors and actuators. Less appreciated is, however, fact that traditional passive control -- such as dry friction, viscous damping, nonlinear springs or snubbers -- can be also modeled by fuzzy logic. Jan
RE: MonadZero (concluded)
Following many protests, the right thing to do seems to be to move MonadPlus to the Monad library. Specifically: class Monad m = MonadPlus m where mzero :: m a mplus :: m a - m a - m a filterM :: MonadZero m = (a - m Bool) - [a] - m [a] guard :: MonadZero m = Bool - m () mfilter :: MonadZero m = (a - Bool) - m a - m a concatM :: MonadPlus m = [m a] - m a Alex, you'll have to use `mplus` instead of (++); or you can define a new operator (+++) to mean `mplus`; or you can hide the list (++) and redefine it to be `mplus`. I guess that 95% of the mailing list is tired of MonadZero. If anyone has further thoughts, pls send them to me only (and, of course, any other individuals you like). Simon
Re: derive conflicts with multiply-defined and module level import
On Sat, 7 Nov 1998, Fergus Henderson wrote: Well, it depends on what you call *strong*. The only reason that I heard is that it prevents users from making possibly unwanted errors. Another reason is that allowing definitions to be split up without any special syntax indicating this would harm readability. If I see a definition, I can't be sure it's complete without examining the whole module. I am very sympathetic to readability arguments, but I also agree with erik that programmers should have a way out when necessary. Code generation is one of those cases where one can sacrifice readability. In that context, your proposal to allow them to be separated, but only with some special declaration, e.g. a keyword "noncontiguous", prefixed to each clauses, that wouldn't be so bad for readability. Would be something that I agree with. But if we have to add new syntax to make it work then it is getting to be more trouble than its worth. Since the application is code generation, I don't think the extra syntax is a problem. -Alex- ___ S. Alexander Jacobson i2x Media 1-212-697-0184 voice1-212-697-1427 fax
Re: Haskell 98: getting there
At 03:12 -0800 1998/11/09, Simon Peyton-Jones wrote: * Default default. Still undecided (sigh). Should it be (Int, Float) (Integer, Double) (Integer, Rational) Several folk want Integer, but don't say whether they want Float,Double, Rational. I think the default should be (Integer, Rational, Double) :-); none of these types are really mathematically interchangeable. It is prudent to use Double (64-bit IEEE floating numbers) over Float (32-bit IEEE floating numbers) as a default, as 32-bit CPU's and better are likely to compute Float operations by first converting them to Doubles: So Floats are then are then slower with less precision, and the memory difference requirement is normally not of any importance. Hans Aberg * Email: Hans Aberg mailto:[EMAIL PROTECTED] * Home Page: http://www.matematik.su.se/~haberg/ * AMS member listing: http://www.ams.org/cml/
hugs and ghc compatibility and features
I have been reading about the integrated runtime system between GHC and Hugs. Now that GHC 4.0 is out, what is the status of the hugs ghc integration project? 1. Does hugs now support mutually recursive modules? 2. Does the new GHC support TREX? If yes, how does one enable it? 3. Are the literate modes now compatible? (they weren't before) 4. Does Hugs now support MPTC, Existential types, etc? If so, where is this new version of hugs and how long until public release? -Alex- ___ S. Alexander Jacobson i2x Media 1-212-697-0184 voice1-212-697-1427 fax
RE: Haskel Type Question
I'm guessing that the problem is that fos -0.5 [1,1,1,1] is being parsed as ((fos) - (0.5)) [1,1,1,1] so that the 0.5 implies an instance of class Fractional which implies by the type of (-) (-) :: Num a = a - a - a that fos is also an instance of class Fractional But since it isnt we get an error if you do a fos (-0.5) [1,1,1,1] this is parsed as fos (negate 0.5) [1,1,1,1] ... etc This is the reason I prefer the notation negate x = ~x and subtract x y = x - y which means that (-x) is a section and (~x) is a negation As I say I am not completely sure this is the 'exact' reason since I thought that the above example would parse as (fos) - (0.5 [1,1,1,1]) Clearly garbage! it will however certainly be along these lines. Hope this helps Chris -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Matthew Donadio Sent: Monday, November 09, 1998 2:02 PM To: Haskell List Subject: Haskel Type Question I have two functions fos:: Num a - [a] - [a] fos a x = fos' a 0 x fos':: Num a - a - [a] - [a] fos' _ _ [] = [] fos' a y1 (x:xs) = y : fos' a y xs where y = a * y1 + x Why does fos -0.5 [ 1, 1, 1, 1 ] give me [a] - b - [b] - [b] is not an instance of class "Fractional" while fos (-0.5) [ 1, 1, 1, 1 ] evaluates just fine? I'm using Hugs 1.4. Thanks. -- Matt Donadio ([EMAIL PROTECTED]) | 43 Leopard Rd, Suite 102 Sr. Software Engineer | Paoli, PA 19301-1552 Image Signal Processing, Inc. | Phone: +1 610 407 4391 http://www.isptechinc.com | FAX: +1 610 407 4405
