February 2005 - Libraries

RE: Data.Graph patch
by Simon Marlow 02 Feb '05

02 Feb '05

On 31 January 2005 22:20, John Meacham wrote: > Here is a patch for Data.Graph which fixes an oversight in > graphFromEdges. graphFromEdges returns a function which turns vertexes > into nodes, but provides no way to go the other direction and figure > out which vertex goes with a node. I simply return the already created > function of type key -> Maybe Vertex in the tuple. > > I provided it as a new function > graphFromEdges' and left the old functionality under the old name. if > we don't care about backwards compatability, I'd recommend just > replacing the one that is already there since it is a more complete > interface and an easy transition for anyone currently using the > library. Does anyone care if we simply change the interface to graphFromEdges? Cheers, Simon

1 0

RE: Question about packages and GHCi
by Simon Marlow 02 Feb '05

02 Feb '05

On 02 February 2005 11:17, Malcolm Wallace wrote: > "Simon Marlow" <simonmar(a)microsoft.com> writes: > >>> (1) When I try to add a package to ghc, I receive the >>> error message that a GHCi lib file cannot be found. >>> >>> > ghc-pkg -u -i fgl.pkg >>> Reading package info from "fgl.pkg"done. >>> Expanding embedded variables... done. >>> warning: can't find GHCi lib `HSfgl.o' >>> Saving old package config file... done. >>> Writing new package config file... done. >>> >>> The missing file `HSfgl.o' is located on the >>> current directory as well as in the directory >>> where all other .o and .a files for GHC are >>> kept (/usr/local/lib/ghc-6.2/). >> >> ghc-pkg is looking in the directories specified in the library_dirs >> field in fgl.pkg. HSfgl.o should be in one of those directories. > > This has puzzled me for a long time as well. I /always/ get this > warning message, even when the HSfoo.o file is in the correct place, > as listed in library_dirs. Strange. I just tried a simple example, and didn't get the warning, so at least it's not completely broken. Can someone who is getting this warning send me: the package config being fed into ghc-pkg, and an ls -l of each directory in library_dirs. Cheers, Simon

1 0

RE: Question about packages and GHCi
by Simon Marlow 02 Feb '05

02 Feb '05

On 02 February 2005 01:52, Martin Erwig wrote: > I apologize if this is the wrong mailing list to turn to, > but I would be grateful for any help with two problems > I have when trying to add a package to GHCi. (I wasn't able > to find sufficient information in the documentation for packages > http://www.haskell.org/ghc/docs/latest/html/users_guide/packages.html.) > > > (1) When I try to add a package to ghc, I receive the > error message that a GHCi lib fil cannot be found. > > > ghc-pkg -u -i fgl.pkg > Reading package info from "fgl.pkg"done. > Expanding embedded variables... done. > warning: can't find GHCi lib `HSfgl.o' > Saving old package config file... done. > Writing new package config file... done. > > The missing file `HSfgl.o' is located on the > current directory as well as in the directory > where all other .o and .a files for GHC are > kept (/usr/local/lib/ghc-6.2/). ghc-pkg is looking in the directories specified in the library_dirs field in fgl.pkg. HSfgl.o should be in one of those directories. > (2) Maybe related to the previous problem is the > following. When I start "ghci -package fgl", I receive > the following error message: > > Loading package base ... linking ... done. > Loading package haskell98 ... linking ... done. > Loading package haskell-src ... linking ... done. > Loading package fgl ... > > GHCi runtime linker: fatal error: I found a duplicate definition > for symbol > __module_registered > whilst processing object file > /usr/local/lib/ghc-6.2/HSfgl.o > This could be caused by: > * Loading two different object files which export the same > symbol > * Specifying the same object file twice on the GHCi command line > * An incorrect `package.conf' entry, causing some object to be > loaded twice. > GHCi cannot safely continue in this situation. Exiting now. > Sorry. I vaguely remembered seeing this before... googling for 'duplicate __module_registered' turned up this: http://www.haskell.org/pipermail/glasgow-haskell-users/2003-December/006 042.html which indicates that perhaps you should add -x to ld when building HSfgl.o? What version of GHC is this, BTW? Cheers, Simon

3 2

RE: Time Resolution
by Simon Marlow 02 Feb '05

02 Feb '05

On 02 February 2005 07:25, Ashley Yakeley wrote: > In article <ashley-EF0986.17424301022005(a)sea.gmane.org>, > Ashley Yakeley <ashley(a)semantic.org> wrote: > >> It looks like we may need three types here: >> >> * one for TAI >> >> * one for POSIX time, which is a broken encoding of UTC >> >> * one for correct encoding of UTC > > This is the sort of thing I mean: > > newtype TAITime = TAITime Integer deriving (...) > > newtype POSIXTime = POSIXTime Integer deriving (...) > > newtype DiffTime = DiffTime Integer deriving (...) > > type JulianDay = Integer > > data UTCTime = UTCTime JulianDay DiffTime > > getCurrentTime :: IO POSIXTime > > posixToUTCTime :: POSIXTime -> UTCTime Why does the user of the library care about the difference between UTC, TAI and POSIX time? The only way I can think that this really needs to be exposed to the programmer is so that they can tell the library whether the system clock is running POSIX or TAI time. For the library API, a single notion of time is sufficient, no? Cheers, Simon

2 1

Time Resolution
by Ashley Yakeley 02 Feb '05

02 Feb '05

What underlying numerical type should be used to store TAI and UTC types, and at what resolution? newtype Clocktime = Clocktime ??? deriving (Eq,Ord, ...) Here are some suggestions. Not all of them are good ones: * Integer, where 1 = 1 microsecond, nanosecond, picosecond But which? Current ClockTime is ps. POSIX "struct timeval" is micro-s, "struct timespec" is ns. TAI64NA uses attoseconds (10^-18 s). <http://cr.yp.to/libtai/tai64.html> + fast(?) + power of ten matches common practice - whatever resolution might not be enough * Integer, where 1 = 1 Planck time (c. 5*10^-44 s) + amusing + unlikely to ever need more accuracy - number of Planck times in an SI second not precisely known - not a power of ten - about 2*10^31 in a picosecond, or about 100 bits * Fixed-size integer type, where 1 = 1 microsecond or etc. 2^64 mus = 580,000 years 2^64 ns = 580 years 2^128 ps = 10^19 years + fast(?) - probably need Int128 * Rational, where 1 = 1 second + 1 = 1 second easy to use + all the resolution you need + no error when dividing, guarantee (a / b) * b == a - might be slower? * Fixed-point type, where 1 = 1 second This could be Integer at some power of ten. + 1 = 1 second easy to use - we'd have to create it - whatever resolution might not be enough * Floating-point type, where 1 = 1 second + 1 = 1 second easy to use + fair amount of resolution + can use sqrt and trig functions easily + fancy NaN values, signed zeros and infinities - resolution variable, bad use of floating point * Type parameter, where 1 = 1 second + flexible - extra complication - can't use directly with Integer except for 1-second resolution -- Ashley Yakeley, Seattle WA

7 22

Question about packages and GHCi
by Martin Erwig 01 Feb '05

01 Feb '05

I apologize if this is the wrong mailing list to turn to, but I would be grateful for any help with two problems I have when trying to add a package to GHCi. (I wasn't able to find sufficient information in the documentation for packages http://www.haskell.org/ghc/docs/latest/html/users_guide/packages.html.) (1) When I try to add a package to ghc, I receive the error message that a GHCi lib fil cannot be found. > ghc-pkg -u -i fgl.pkg Reading package info from "fgl.pkg"done. Expanding embedded variables... done. warning: can't find GHCi lib `HSfgl.o' Saving old package config file... done. Writing new package config file... done. The missing file `HSfgl.o' is located on the current directory as well as in the directory where all other .o and .a files for GHC are kept (/usr/local/lib/ghc-6.2/). (2) Maybe related to the previous problem is the following. When I start "ghci -package fgl", I receive the following error message: Loading package base ... linking ... done. Loading package haskell98 ... linking ... done. Loading package haskell-src ... linking ... done. Loading package fgl ... GHCi runtime linker: fatal error: I found a duplicate definition for symbol __module_registered whilst processing object file /usr/local/lib/ghc-6.2/HSfgl.o This could be caused by: * Loading two different object files which export the same symbol * Specifying the same object file twice on the GHCi command line * An incorrect `package.conf' entry, causing some object to be loaded twice. GHCi cannot safely continue in this situation. Exiting now. Sorry. Any help is greatly appreciated. Thanks, Martin

2 1

RE: Time Resolution
by Simon Marlow 01 Feb '05

01 Feb '05

On 01 February 2005 06:23, Ashley Yakeley wrote: > In article <41FE95DF.50000(a)cql.com>, Seth Kurtzberg <seth(a)cql.com> > wrote: > >> I'm not, I hope, being pedantic here and talking about an irrelevant >> issue. But if you look in any first year engineering textbook (for >> an engineering discipline other than CS), you will find that the >> display of a value with a precision greater than the known precision >> of the inputs is a cardinal sin. It's the wrong answer. The >> roundoff error behavior is clearly going to be different. > > Well, how do you feel about using Rationals everywhere? That way > there's never any question of some irrelevant specified unused > accuracy. > > newtype ClockTime = ClockTime Rational deriving (Eq, etc.) > > I quite like the idea of using Rational for ClockTime, but I worry > that it might be slow. But it does allow you to do things such as > dividing times into n pieces and adding them all up again without > error. Not that keen, I have to say. If we had an Integral type, you can always convert to Rational for the purposes of dividing by 3 and recombining (or whatever) and then use truncate to convert back to Integral. Cheers, Simon

2 1