January 2008 - Glasgow-haskell-users

Read instance of StdGen returns no parse
by Denis Bueno 02 Feb '08

02 Feb '08

Hi all, According to http://haskell.org/ghc/docs/latest/html/libraries/random/System-Random.html… the Read StdGen instance should never fail. However, in GHC 6.8.2, it appears to: > ghci GHCi, version 6.8.2: http://www.haskell.org/ghc/ :? for help Loading package base ... linking ... done. Prelude> :m System.Random Prelude System.Random> read "exsqueeze me" :: StdGen Loading package old-locale-1.0.0.0 ... linking ... done. Loading package old-time-1.0.0.0 ... linking ... done. Loading package random-1.0.0.0 ... linking ... done. *** Exception: Prelude.read: no parse Prelude System.Random> Leaving GHCi. It first fails for me on strings of length seven. Is this a bug, or have I done something wrong? -- Denis

2 2

Using "GHC as a library" with own functions makes runStmt return RunBreak
by Mads Lindstrøm 01 Feb '08

01 Feb '08

Hi I am trying to use "GHC as a library". From runStmt, I want to load and run functions from a home-grown module. Lets call my home-grown module HomeGrown.hs. I have tried to model my application after http://haskell.org/sitewiki/images/1/17/Interactive-6.8.hs / http://haskell.org/sitewiki/images/f/f9/MyPrelude.hs from http://haskell.org/haskellwiki/GHC/As_a_library . This library loads MyPrelude.hs and runs myPutStrLn & myGetLine + more. However, when I try to run code from HomeGrown.hs like: runStmt session "HomeGrown.foo" GHC.SingleStep runStmt keeps returning RunBreak (of type RunResult). From what I can read of the WWW, RunBreak is related to some debugger and indicate that GHC encountered some breakpoint. However, I did not set any break points. Comparing Interactive-6.8 and MyPrelude.hs to my own code, the largest difference seemed to be that Interactive-6.8 do not only use MyPrelude.hs by calling `addTarget` and `setContext`, it also imports MyPrelude.hs as a regular Haskell import (import MyPrelude). Thus I tried to remove this import to see what would happen. Interactive-6.8 compiled and started just fine. However, whenever it needed to use any functions in MyPrelude, runStmt also returned RunBreak. So is it only possible to load home-grown/non-standard modules when also doing a regular import? Greetings, Mads Lindstrøm

2 3

GHC on Solaris Nevada
by Felix Martini 28 Jan '08

28 Jan '08

Hello all, I am trying to get a working GHC on Solaris Nevada (Solaris 11). I've tried the binary versions for Solaris 10 made by Christian Maeder. I couldn't get GHC 6.8.2 to install properly and GHC 6.6 does not work (it crashes the moment it needs to link to a library). After that i tried to make an unregistered bootstrap of GHC 6.6 using Linux as the host OS. The bootstrapping failed on Solaris when making genapply or cabal-setup. The linker reports symbol referencing errors. Has someone already successfully build GHC on Solaris 11? Regards, Felix

2 6

Re: [Haskell-cafe] Internships at GHC HQ
by Tim Chevalier 25 Jan '08

25 Jan '08

On 1/25/08, Dan Licata <drl(a)cs.cmu.edu> wrote: > A further plug: > > I did an internship with Simon PJ last summer (implementing view > patterns in GHC, among other things), and this is a great opportunity if > you're interested in PL research. There is a lot of interesting work > going on at MSR Cambridge, the atmosphere is very friendly, and > Cambridge is a lovely place to spend a summer. > > If anyone wants an "intern's-eye" view of the experience, feel free to > e-mail me! > I second most parts of this (including the "feel free to email me part"). I hear that the summer is busiest with respect to interns, but those who can arrange it with their graduate programs ought to consider applying for an internship during the school year. I don't know what things are like now, but I found that there were plenty of other interns around to keep me company even during the fall, and Cambridge is a lovely place to spend an autumn as well. Cheers, Tim -- Tim Chevalier * http://cs.pdx.edu/~tjc * Often in error, never in doubt "The more you talk, the more I get / a sense of something that hasn't happened yet / The more you talk, the more I want to know / the way I'll remember you when I go." -- Ani DiFranco

1 0

GHCi is being too "helpful"
by Dinko Tenev 25 Jan '08

25 Jan '08

Hi everyone, When evaluating an IO monad in GHCi from the 6.6.1 distribution, the REPL tries to be a bit too smart about printing out the returned value. For example, () values are not printed: Prelude System.Directory> setCurrentDirectory "work" Prelude System.Directory> ...but String, or, as far as I can see, any other values, do get printed: Prelude System.Directory> getCurrentDirectory "/home/shinobi" Prelude System.Directory> set In my particular case, the value printed is a 250,000-element list -- waiting for the value to print out is not an option. I don't want to see the whole list at once, but do need to examine parts of it, subject it to intermediate transformations, etc. Is there any way to suppress this output and revert to the old behavior, where the value was simply bound to "it", without being printed? Thanks, D. Tenev

3 2

Network.CGI changes
by Frederik Eaton 25 Jan '08

25 Jan '08

Hello all, About two years ago, I wrote a web page for one of my projects, using Network.CGI. I chose that over WASH because it had a simple interface and I thought it would be more stable as a result. Now, Bjorn Bringert has replaced the interface with a completely different one. There is a cgi-compat, but it doesn't build with GHC 6.8.2. Does anyone know a quick way to get my web page functioning again? Is there an early version of Network.CGI (with the 'wrapper' interface) that still compiles under GHC 6.8.2? I'm a bit reluctant to convert my code to the new interface, because it will probably change completely again in a year or so, and I'd rather not have to keep revisiting my project and rewriting the web page... Anyway, I am attaching the source file. Thanks in advance, Frederik P.S. Sometimes programmers make a rule of changing the names of things when they change the interface; this makes it possible to keep the old interface alongside the new one, so code which depends on the old interface don't have to change. Is there a reason why the Haskell community does not adopt this practice? It seems that people are always saying that there is a lack of manpower in this community, perhaps that is an excellent reason to make *fewer* incompatible changes, rather than more? -- http://ofb.net/~frederik/

5 7

Fwd: Arrow without `>>>'
by Valery V. Vorotyntsev 24 Jan '08

24 Jan '08

The problem has been resolved. Kudos to Ross Paterson. ---------- Forwarded message ---------- From: Valery V. Vorotyntsev <valery.vv(a)gmail.com> Date: Jan 24, 2008 3:34 PM Subject: Re: Arrow without `>>>' To: libraries(a)haskell.org On 1/24/08, Ross Paterson <ross(a)soi.city.ac.uk> wrote: > > The right question: how to build `arrows' in such circumstances? > > To build with the darcs version of the base library (built with GHC), > you'll need the darcs version of the arrows library, which I think is > right now: > > darcs get http://darcs.haskell.org/packages/arrows Yes, after pulling the latest change from repo installation went smoothly. Thank you very much, Ross!

1 0

ghc -O
by Matthew Naylor 24 Jan '08

24 Jan '08

Hello GHC gurus, specifying "-O" or "-O2" to GHC enables various optimisations to the frontend of the compiler, but I wonder does it turn on any optimistions to the backend/code-generator that are independent of the frontend? I can imagine that knowledge such as strictness which is computed by frontend would enable optimisations in the backend, but I'm more interested in whether the backend would do anything independent (e.g. peephole optimistion, C compiler options) with "-O" but not without. Thanks, Matt.

2 1

Link failures using GHC 6.8.2 on OS X 10.1.11 (PPC)
by Robin Houston 23 Jan '08

23 Jan '08

This morning I downloaded and installed ghc-6.8.2-powerpc-apple-darwin.tar.bz2. I'm now getting link errors when I use ghc. For example, the simple program import qualified Data.IntSet as IntSet main = print $ IntSet.empty gives the errors: /usr/bin/ld: Undefined symbols: ___stginit_containerszm0zi1zi0zi1_DataziIntSet_ _containerszm0zi1zi0zi1_DataziIntSet_empty_closure _containerszm0zi1zi0zi1_DataziIntSet_zdf3_closure These symbols are defined in /usr/local/lib/ghc-6.8.2/lib/containers-0.1.0.1/libHScontainers-0.1.0.1.a, so apparently the linker is not finding this library. The same program does work in ghci, and in ghc-6.6 (which I was using previously). I've searched the list archives, and I haven't found any mention of this problem. Any help would be much appreciated! Robin

3 3

Re: Integrating editline with ghc
by Christian Maeder 22 Jan '08

22 Jan '08

(my previous mail went to the bugs list by accident) Christian Maeder wrote: > I would like to know from package maintainers if there packages can be > easily ported from libreadline to libedit. > > The best indication for this would be if the package is also happy with > a restricted interface of readline (i.e. readline-compat) or requires > the full GNU readline. > > At least testing this compatibility makes sense using a readline package > with a temporarily reduced interface (without the need to change the > packages depending on readline.) In short, consider a split up of the readline package into readline-basics and readline-exts (both with GPL licence). Unfortunately it is not possible to have the package readline to be the union of readline-basics and readline-exts. So readline-basic (or readline-compat) would be a duplication of a reduced readline package. For porting packages from libreadline to libedit it would be sufficient to replace readline-basics with editline (and change the licence) in *.cabal files. Packages depending on more then readline-basics cannot be ported. I'm only worried if any package maintainer would bother to change the build-depends from readline to readline-basic (if not forced to change it anyway, i.e. to readline-basic and readline-ext). Alternatively, if the interface of readline is reduced to that of readline-basic, package maintainers could hope that their package is still okay and if not (after compilation errors) add readline-ext to their build-depends. Is this really such a crazy idea? Christian

1 0