- Glasgow-haskell-users

integrating ghc as editor's code tools
by Paulo Henrique 19 May '07

19 May '07

I´m trying to integrated ghc and Crimson editor (Windows), to get a small and efficient package. I found how to integrate ghc and JCreator, but the package has more than 100MB. I configured Crimson's Tools -> Preferences->User Tools with this values: Menu text: GHC Command: C:\ghc\ghc-6.6.1\bin\ghc.exe (or C:\ghc\ghc-6.6.1\bin\ghci).exe Argument: $(FileName) Initial Dir: $[FileDir] Hot Key: no one and I have choose "Capture Output". After wite and save a test file, I called the ghc tool. the answe I get was: -with ghc: ---------- Capture Output ---------- > "C:\ghc\ghc-6.6.1\bin\ghc.exe" FirstTeste.hs > Terminated with exit code 0. with ghci: ---------- Capture Output ---------- > "C:\ghc\ghc-6.6.1\bin\ghci.exe" FirstTeste.hs > Terminated with exit code 0. How can I solve this problem? If anyone has a better solution (another editor), I would like to know it. thanks in advance, Paulo Henrique

1 0

Re: GNUreadline.framework and GMP.framework
by Christian Maeder 19 May '07

19 May '07

It seems our GMP.framework is only suited for Powerpc. I'll look after it. Thanks for your hints and sorry for the inconveniences. (Meanwhile you could try to get the GMP.framework for intel macs from elsewhere.) Christian P.S. There is also a problem with the GNUreadline.framework in your home directory on Powerpcs only. When you link your own binaries you have to add -optl-F$(HOME)/Library/Frameworks -optl-framework -optlGNUreadline to ghc's command line. Jay McCarthy schrieb: > Your frameworks: > > I've made binary distributions for powerpc and intel macs that rely on a > GNUreadline.framework and a GMP.framework that can be found (for both > architectures) here: > > http://www.informatik.uni-bremen.de/agbkb/forschung/formal_methods/CoFI/het… > > http://www.informatik.uni-bremen.de/agbkb/forschung/formal_methods/CoFI/het… > > http://www.informatik.uni-bremen.de/agbkb/forschung/formal_methods/CoFI/het… > > > Do not work on Intel: > > jay $ wget > http://www.informatik.uni-bremen.de/agbkb/forschung/formal_methods/CoFI/het… > > --09:37:01-- > http://www.informatik.uni-bremen.de/agbkb/forschung/formal_methods/CoFI/het… > > => `GNUreadline-framework.zip' > Resolving www.informatik.uni-bremen.de... 134.102.224.5 > Connecting to www.informatik.uni-bremen.de|134.102.224.5|:80... connected. > HTTP request sent, awaiting response... 200 OK > Length: 2,342,369 (2.2M) [application/zip] > > 100%[====================================>] 2,342,369 345.15K/s > ETA 00:00 > > 09:37:10 (264.54 KB/s) - `GNUreadline-framework.zip' saved > [2342369/2342369] > > ~/Desktop > jay $ wget > http://www.informatik.uni-bremen.de/agbkb/forschung/formal_methods/CoFI/het… > > --09:37:12-- > http://www.informatik.uni-bremen.de/agbkb/forschung/formal_methods/CoFI/het… > > => `GMP.framework.zip' > Resolving www.informatik.uni-bremen.de... 134.102.224.5 > Connecting to www.informatik.uni-bremen.de|134.102.224.5|:80... connected. > HTTP request sent, awaiting response... 200 OK > Length: 564,957 (552K) [application/zip] > > 100%[====================================>] 564,957 228.26K/s > > 09:37:15 (227.77 KB/s) - `GMP.framework.zip' saved [564957/564957] > > ~/Desktop > jay $ unzip GNUreadline-framework.zip -d $HOME/Library/Frameworks > Archive: GNUreadline-framework.zip > creating: /Users/jay/Library/Frameworks/GNUreadline.framework/ > creating: /Users/jay/Library/Frameworks/GNUreadline.framework/Versions/ > creating: /Users/jay/Library/Frameworks/GNUreadline.framework/Versions/A/ > creating: > /Users/jay/Library/Frameworks/GNUreadline.framework/Versions/A/Headers/ > creating: > /Users/jay/Library/Frameworks/GNUreadline.framework/Versions/A/Headers/readline/ > > inflating: > /Users/jay/Library/Frameworks/GNUreadline.framework/Versions/A/Headers/readline/rltypedefs.h > > inflating: > /Users/jay/Library/Frameworks/GNUreadline.framework/Versions/A/Headers/readline/chardefs.h > > inflating: > /Users/jay/Library/Frameworks/GNUreadline.framework/Versions/A/Headers/readline/keymaps.h > > inflating: > /Users/jay/Library/Frameworks/GNUreadline.framework/Versions/A/Headers/readline/history.h > > inflating: > /Users/jay/Library/Frameworks/GNUreadline.framework/Versions/A/Headers/readline/tilde.h > > inflating: > /Users/jay/Library/Frameworks/GNUreadline.framework/Versions/A/Headers/readline/rlstdc.h > > inflating: > /Users/jay/Library/Frameworks/GNUreadline.framework/Versions/A/Headers/readline/rlconf.h > > inflating: > /Users/jay/Library/Frameworks/GNUreadline.framework/Versions/A/Headers/readline/readline.h > > inflating: > /Users/jay/Library/Frameworks/GNUreadline.framework/Versions/A/GNUreadline > linking: /Users/jay/Library/Frameworks/GNUreadline.framework/GNUreadline > -> Versions/A/GNUreadline > linking: /Users/jay/Library/Frameworks/GNUreadline.framework/Headers > -> Versions/A/Headers > creating: /Users/jay/Library/Frameworks/GNUreadline.framework/info/ > inflating: > /Users/jay/Library/Frameworks/GNUreadline.framework/info/readline.info > inflating: > /Users/jay/Library/Frameworks/GNUreadline.framework/info/rluserman.info > inflating: > /Users/jay/Library/Frameworks/GNUreadline.framework/info/history.info > inflating: /Users/jay/Library/Frameworks/GNUreadline.framework/info/dir > inflating: /Users/jay/Library/Frameworks/GNUreadline.framework/README > inflating: /Users/jay/Library/Frameworks/GNUreadline.framework/COPYING > inflating: /Users/jay/Library/Frameworks/GNUreadline.framework/CHANGELOG > inflating: > /Users/jay/Library/Frameworks/GNUreadline.framework/build-framework-R.sh > inflating: > /Users/jay/Library/Frameworks/GNUreadline.framework/readline-5.2.tar.gz > finishing deferred symbolic links: > /Users/jay/Library/Frameworks/GNUreadline.framework/GNUreadline -> > Versions/A/GNUreadline > /Users/jay/Library/Frameworks/GNUreadline.framework/Headers -> > Versions/A/Headers > ~/Desktop > jay $ unzip GMP.framework.zip -d $HOME/Library/Frameworks > Archive: GMP.framework.zip > creating: /Users/jay/Library/Frameworks/GMP.framework/ > inflating: /Users/jay/Library/Frameworks/GMP.framework/build-framework.sh > inflating: /Users/jay/Library/Frameworks/GMP.framework/ChangeLog > inflating: /Users/jay/Library/Frameworks/GMP.framework/COPYING.LIB > linking: /Users/jay/Library/Frameworks/GMP.framework/GMP -> > Versions/A/GMP > linking: /Users/jay/Library/Frameworks/GMP.framework/Headers -> > Versions/A/Headers > creating: /Users/jay/Library/Frameworks/GMP.framework/info/ > inflating: /Users/jay/Library/Frameworks/GMP.framework/info/gmp.info > inflating: /Users/jay/Library/Frameworks/GMP.framework/info/gmp.info-1 > inflating: /Users/jay/Library/Frameworks/GMP.framework/info/gmp.info-2 > inflating: /Users/jay/Library/Frameworks/GMP.framework/README > creating: /Users/jay/Library/Frameworks/GMP.framework/Versions/ > creating: /Users/jay/Library/Frameworks/GMP.framework/Versions/A/ > inflating: /Users/jay/Library/Frameworks/GMP.framework/Versions/A/GMP > creating: /Users/jay/Library/Frameworks/GMP.framework/Versions/A/Headers/ > inflating: > /Users/jay/Library/Frameworks/GMP.framework/Versions/A/Headers/gmp.h > finishing deferred symbolic links: > /Users/jay/Library/Frameworks/GMP.framework/GMP -> Versions/A/GMP > /Users/jay/Library/Frameworks/GMP.framework/Headers -> Versions/A/Headers > ~/Desktop > jay $ ghc > dyld: Library not loaded: GMP.framework/Versions/A/GMP > Referenced from: > /Users/jay/Development/dists/ghc/ghc-6.6.1/lib/i386-apple-darwin/ghc-6.6.1 > Reason: no suitable image found. Did find: > /Users/jay/Library/Frameworks/GMP.framework/Versions/A/GMP: > mach-o, but wrong architecture > Trace/BPT trap > ~/Desktop > jay $ >

1 2

Re: Wanted: warning option for usages of unary minus
by John Meacham 19 May '07

19 May '07

On Thu, May 17, 2007 at 06:40:04PM +0200, Twan van Laarhoven wrote: > Simon Marlow wrote: > >... > > > >Really? I'm beginning to have second thoughts about the proposed change > >to negation for Haskell'. The main reason, and this isn't pointed out > >as well as it should be on the wiki, is that "x-1" will cease to be an > >infix application of (-), it will parse as x applied to the literal > >(-1). And this is different from "x - 1" > > There is one other alternative for parsing: > "-" is a unary minus if and only if it is > a) preceded by whitespace or one of "[({;,", and > b) not followed by whitespace. > > So: > x - 1 == (-) x 1 > x-1 == (-) x 1 > x -1 == x (negate 1) > x -(1) == x (negate 1) > x (-1) == x (negate 1) > x (- 1) == x (\y -> y - 1) also (c) has a digit after it. also note that a big point of this is that we can get rid of 'negate' in the translation, so the literal -1 now desugars to (fromInteger -1) rather than the current (negate (fromInteger 1)) (which requires 2 dictionary lookups for a simple constant! sigh. we also have the option of removing negate from the 'Num' class if so desired. John -- John Meacham - ⑆repetae.net⑆john⑈

1 0

Libraries makefile
by Stefan Holdermans 17 May '07

17 May '07

I noticed that the makefile in ghc/libraries is changed recently--- could it be broken? When building GHC HEAD on Mac OS X (Darwin, Power PC), I got: ... make -C libraries boot rm -f -rf bootstrapping.Cabal cp -R Cabal bootstrapping.Cabal bootstrapping.Cabal $ -name "*.o" -o -name "*.hi" $ \ -exec rm -f -f {} \; /bin/sh: line 1: bootstrapping.Cabal: command not found make[1]: *** [stamp/bootstrapping.Cabal] Error 127 make: *** [stage1] Error 2

1 0

Re: Wanted: warning option for usages of unary minus
by Isaac Dupree 17 May '07

17 May '07

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Twan van Laarhoven wrote: > There is one other alternative for parsing: > "-" is a unary minus if and only if it is > a) preceded by whitespace or one of "[({;,", and > b) not followed by whitespace. > > So: > x - 1 == (-) x 1 > x-1 == (-) x 1 > x -1 == x (negate 1) > x -(1) == x (negate 1) > x (-1) == x (negate 1) > x (- 1) == x (\y -> y - 1) > > Just an idea. Indeed, and in some language syntax designs it would certainly be a good system for prefix operators. Existing parsers may have some difficulty. How about > {-comment-}-1 ? how about > WeirdNumber{value=2,weird=True}-1 ? Although likely to make any actual code work, it seems a bit complicated from the mindset of current Haskell parsing/lexing. "(b) not followed by whitespace." can be replaced by (b) followed by a digit if desired not to allow it for negating arbitrary expressions. Isaac -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFGTJRgHgcxvIWYTTURAqpMAJ9rpCFwzOG/ZSF0qpM/hD/mFKrQ1wCfSRCK 2nKiBzRs/8thmgrdBT+SowA= =lFCl -----END PGP SIGNATURE-----

1 0

All warnings but type signatures
by Joel Reymont 13 May '07

13 May '07

Is there a way to enable all warnings but the following: Hope/Item/DBDesc.hs:6:0: Warning: Definition but no type signature for `item_db' Overall, is there an advantage to type signatures in GHC as opposed to letting the compiler derive them? Thanks, Joel -- http://wagerlabs.com/

3 2

{-# INLINE me_harder #-}
by Duncan Coutts 13 May '07

13 May '07

So I'm trying to program the GHC term rewriting system again (ie the mighty simplifier) and it's not doing what I want it to do without me using unnecessarily large hammers. The crux is that I have a simple function that I want to be inlined early so that my other rules can match on the thing it expands into, however despite the {-# INLINE #-} pragma, it doesn't get inlined until a much later phase when it's too late for the rule to match. So currently I'm forced to use {-# RULES #-} to achieve the same effect, ie expanding the definition in an early phase. This feel like a hack of course. So either the INLINE heuristics could be tweaked to make these cases work or perhaps we should consider adding some other pragma for when we don't want to go by heuristics but are instead deliberately trying to do term rewriting. Anyway, here's the example. It's binary deserialisation. We have a couple important primitives: read :: Int -> (Ptr Word8 -> IO a) -> Get a and the applicative combinators (<$>) (<*>) for Get: fmapGet :: (a -> b) -> Get a -> Get b apGet :: Get (a -> b) -> Get a -> Get b So that we'll be able to match on these, we delay their inlining: {-# INLINE [0] apGet #-} {-# INLINE [0] fmapGet #-} {-# INLINE [0] read #-} The important rules linking these are: {-# RULES "fmap/read" forall f n a. f `fmapGet` read n a = ... "read/read" forall n m f x. read n f `apGet` read m x = ... #-} and finally we have: word8 :: Get Word8 word8 = read 1 peek {-# INLINE word8 #-} So when we write something like: foo :: Get (Word8,Word8,Word8) foo = (,,) <$> Get.word8 <*> Get.word8 <*> Get.word8 we want to expand word8 into it's definition in terms of read and then have the rules fire (which shares the bounds checks). Looking at the simplifier iterations we get to this step: ==================== Simplifier phase 2, iteration 1 out of 4 ==================== PutTest.foo :: Get.Get (GHC.Word.Word8, GHC.Word.Word8, GHC.Word.Word8) [Exported] [] PutTest.foo = Get.apGet @ GHC.Word.Word8 @ (GHC.Word.Word8, GHC.Word.Word8, GHC.Word.Word8) (Get.apGet @ GHC.Word.Word8 @ (GHC.Word.Word8 -> (GHC.Word.Word8, GHC.Word.Word8, GHC.Word.Word8)) (Get.fmapGet @ GHC.Word.Word8 @ (GHC.Word.Word8 -> GHC.Word.Word8 -> (GHC.Word.Word8, GHC.Word.Word8, GHC.Word.Word8)) (Data.Tuple.(,,) @ GHC.Word.Word8 @ GHC.Word.Word8 @ GHC.Word.Word8) Get.word8) Get.word8) Get.word8 Here is where we really must inline word8 or we're going to miss our opportunity once apGet and fmapGet get inlined. Sadly it doesn't inline it here (in phase 2), or in the next phase (in fact it does nothing in phase 1 at all). We get all the way to phase 0 and then go and inline everything (including word8). The behaviour is the same for 6.6 and 6.7.recentish (one month old). If we hit word8 with a bigger hammer: {-# RULES "inline word8" word8 = read 1 peekWord8 #-} then the whole thing works perfectly and we get really nice STG code at the end. I was under the impression that GHC considered a function more 'interesting' if it was mentioned in the LHS of a rule as is the case here. What is the right thing for me to do in this case? Just use the rule to do the inlining? The full source is here: http://haskell.org/~duncan/binary/ Get.hs and GetTest.hs (In the same dir is an example showing the opposite problem, GHC inlining a function when I asked explicitly for it not to. I reported that problem previously. Though I now realise I never filed a bug in trac for that one. I'll file bugs for both.) On the good news side, I'm getting excellent performance results for the serialisation and judging from the STG code for the deserialisation I expect it'll be great too. Duncan

4 7

non-owner-writable files generated in ghc build process?
by Isaac Dupree 12 May '07

12 May '07

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 When I was doing `rm -r` on a build tree it pointed out that driver/split/ghc-split.prl and driver/mangler/ghc-asm.prl were write-protected. Tacking this down, they're generated from .lprl with unlit... then (in mk/suffix.mk) they are `chmod 444`-ed. Is this a problem? Is the purpose to remind people that those are generated files, so don't change them? (hmm... if someone has a restrictive umask like 027, will this circumvent it? would `chmod -w` make more sense?) I know, I can use `rm -rf` in order to unquestioningly follow UNIX permissions model... Isaac -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFGNJmgHgcxvIWYTTURAs5IAKCw5ND5vKLimSkIaclz5onqnNHVyACeKCPq bBtTcqD+/qoG3gnDcAtLHc0= =PUgp -----END PGP SIGNATURE-----

3 3

GHC 6.6.1 Windows installer, test version
by Neil Mitchell 12 May '07

12 May '07

Hi, I've prepared a GHC 6.6.1 Windows installer. Before I offer this up as an official installer, could people please test it? http://www.haskell.org/ghc/dist/6.6.1/ghc-6.6.1-i386-windows-test1.exe The new features of this installer (relative to the 6.6 one) are: * Prettier .hs icons * Automatically adds GHC and Cabal binaries to the %PATH% * 10Mb smaller (better compression) Thanks to Duncan for answering installer related questions. Neil

5 20

ghc-6.6.1 win32 msi available
by Sigbjorn Finne 12 May '07

12 May '07

In case anyone's interested, http://www.galois.com/~sof/msi/ghc-6-6-1.msi contains a Windows installer for 6.6.1; most (all?) libraries/ are in there; no C++ bits (sorry.) enjoy --sigbjorn [And, if it's your preference, a ghc-6-6-1.zip is also available from that same dir.]

3 3