[GHC] #5000: Eliminate absent arguments in non-strict positions
#5000: Eliminate absent arguments in non-strict positions
-+--
Reporter: batterseapower| Owner:
Type: feature request | Status: new
Priority: normal| Component: Compiler
Version: 7.0.2 |Keywords:
Testcase:| Blockedby:
Os: Unknown/Multiple |Blocking:
Architecture: Unknown/Multiple | Failure: Runtime performance bug
-+--
One of my programs has this is the output of GHC -O2:
{{{
Rec {
h28_s14N
:: forall t_aMK t_aML.
Main.Nat - t_aMK - t_aML - Main.Nat - Main.Nat
[LclId,
Arity=4,
Unf=Unf{Src=vanilla, TopLvl=True, Arity=4, Value=True,
ConLike=True, Cheap=True, Expandable=True,
Guidance=IF_ARGS [0 0 0 0] 6 2}]
h28_s14N =
\ (@ t_aMK)
(@ t_aML)
(az94z94z941_u439_amh :: Main.Nat)
(awS1_u462_ami :: t_aMK)
(awS1_u466_amj :: t_aML)
(x_u590_amk :: Main.Nat) -
Main.S
(h29_s14O
@ t_aMK
@ t_aML
az94z94z941_u439_amh
awS1_u462_ami
awS1_u466_amj
x_u590_amk)
h29_s14O [Occ=LoopBreaker]
:: forall t_aMK t_aML.
Main.Nat - t_aMK - t_aML - Main.Nat - Main.Nat
[LclId,
Arity=4,
Unf=Unf{Src=vanilla, TopLvl=True, Arity=4, Value=True,
ConLike=True, Cheap=True, Expandable=True,
Guidance=IF_ARGS [0 0 0 5] 9 0}]
h29_s14O =
\ (@ t_aMK)
(@ t_aML)
(az94z94z941_u439_amm :: Main.Nat)
(awS1_u462_amn :: t_aMK)
(awS1_u466_amo :: t_aML)
(x_u590_amp :: Main.Nat) -
case x_u590_amp of _ {
Main.Z - h21'_al8 @ Main.Nat az94z94z941_u439_amm;
Main.S x_u595_amq -
h28_s14N
@ t_aMK
@ t_aML
az94z94z941_u439_amm
awS1_u462_amn
awS1_u466_amo
x_u595_amq
}
end Rec }
}}}
This is ridiculous because the two awS1 arguments are totally dead. This
greatly increases allocations made by the program.
I speculate that this occurs because the recursive call to h29 is guarded
by a use of the S constructor, making it non-strict.
--
Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/5000
GHC http://www.haskell.org/ghc/
The Glasgow Haskell Compiler
___
Glasgow-haskell-bugs mailing list
Glasgow-haskell-bugs@haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #5000: Eliminate absent arguments in non-strict positions
#5000: Eliminate absent arguments in non-strict positions
-+--
Reporter: batterseapower| Owner:
Type: feature request | Status: new
Priority: normal| Component: Compiler
Version: 7.0.2 |Keywords:
Testcase:| Blockedby:
Os: Unknown/Multiple |Blocking:
Architecture: Unknown/Multiple | Failure: Runtime performance bug
-+--
Description changed by batterseapower:
Old description:
One of my programs has this is the output of GHC -O2:
{{{
Rec {
h28_s14N
:: forall t_aMK t_aML.
Main.Nat - t_aMK - t_aML - Main.Nat - Main.Nat
[LclId,
Arity=4,
Unf=Unf{Src=vanilla, TopLvl=True, Arity=4, Value=True,
ConLike=True, Cheap=True, Expandable=True,
Guidance=IF_ARGS [0 0 0 0] 6 2}]
h28_s14N =
\ (@ t_aMK)
(@ t_aML)
(az94z94z941_u439_amh :: Main.Nat)
(awS1_u462_ami :: t_aMK)
(awS1_u466_amj :: t_aML)
(x_u590_amk :: Main.Nat) -
Main.S
(h29_s14O
@ t_aMK
@ t_aML
az94z94z941_u439_amh
awS1_u462_ami
awS1_u466_amj
x_u590_amk)
h29_s14O [Occ=LoopBreaker]
:: forall t_aMK t_aML.
Main.Nat - t_aMK - t_aML - Main.Nat - Main.Nat
[LclId,
Arity=4,
Unf=Unf{Src=vanilla, TopLvl=True, Arity=4, Value=True,
ConLike=True, Cheap=True, Expandable=True,
Guidance=IF_ARGS [0 0 0 5] 9 0}]
h29_s14O =
\ (@ t_aMK)
(@ t_aML)
(az94z94z941_u439_amm :: Main.Nat)
(awS1_u462_amn :: t_aMK)
(awS1_u466_amo :: t_aML)
(x_u590_amp :: Main.Nat) -
case x_u590_amp of _ {
Main.Z - h21'_al8 @ Main.Nat az94z94z941_u439_amm;
Main.S x_u595_amq -
h28_s14N
@ t_aMK
@ t_aML
az94z94z941_u439_amm
awS1_u462_amn
awS1_u466_amo
x_u595_amq
}
end Rec }
}}}
This is ridiculous because the two awS1 arguments are totally dead. This
greatly increases allocations made by the program.
I speculate that this occurs because the recursive call to h29 is guarded
by a use of the S constructor, making it non-strict.
New description:
One of my programs has this is the output of the post-worker/wrapper
simplification:
{{{
Rec {
h28_s14N
:: forall t_aMK t_aML.
Main.Nat - t_aMK - t_aML - Main.Nat - Main.Nat
[LclId,
Arity=4,
Str=DmdType ,
Unf=Unf{Src=vanilla, TopLvl=True, Arity=4, Value=True,
ConLike=True, Cheap=True, Expandable=True,
Guidance=IF_ARGS [0 0 0 0] 6 2}]
h28_s14N =
\ (@ t_aMK)
(@ t_aML)
(az94z94z941_u439_amh :: Main.Nat)
(awS1_u462_ami :: t_aMK)
(awS1_u466_amj :: t_aML)
(x_u590_amk :: Main.Nat) -
Main.S
(h29_s14O
@ t_aMK
@ t_aML
az94z94z941_u439_amh
awS1_u462_ami
awS1_u466_amj
x_u590_amk)
h29_s14O [Occ=LoopBreaker]
:: forall t_aMK t_aML.
Main.Nat - t_aMK - t_aML - Main.Nat - Main.Nat
[LclId,
Arity=4,
Str=DmdType LLLS,
Unf=Unf{Src=vanilla, TopLvl=True, Arity=4, Value=True,
ConLike=True, Cheap=True, Expandable=True,
Guidance=IF_ARGS [0 0 0 3] 7 0}]
h29_s14O =
\ (@ t_aMK)
(@ t_aML)
(az94z94z941_u439_amm :: Main.Nat)
(awS1_u462_amn :: t_aMK)
(awS1_u466_amo :: t_aML)
(x_u590_amp :: Main.Nat) -
case x_u590_amp of _ {
Main.Z - az94z94z941_u439_amm;
Main.S x_u595_amq [Dmd=Just L] -
h28_s14N
@ t_aMK
@ t_aML
az94z94z941_u439_amm
awS1_u462_amn
awS1_u466_amo
x_u595_amq
}
end Rec }
}}}
This is ridiculous because the two awS1 arguments are totally dead. This
greatly increases allocations made by the program.
These absent arguments should be removed by the strictness analyser,
However I speculate that this occurs because the recursive call to h29 is
guarded by a use of the S constructor, making it non-strict.
--
--
Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/5000#comment:1
GHC http://www.haskell.org/ghc/
The Glasgow Haskell Compiler
___
Glasgow-haskell-bugs mailing list
Glasgow-haskell-bugs@haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #5000: Eliminate absent arguments in non-strict positions
#5000: Eliminate absent arguments in non-strict positions -+-- Reporter: batterseapower | Owner: Type: feature request | Status: closed Priority: normal | Component: Compiler Version: 7.0.2| Resolution: worksforme Keywords: | Testcase: Blockedby: | Os: Unknown/Multiple Blocking: | Architecture: Unknown/Multiple Failure: Runtime performance bug | -+-- Changes (by batterseapower): * status: new = closed * resolution: = worksforme Comment: Urk, never mind: this was fixed sometime between 7.0.1.20101215 and HEAD -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/5000#comment:2 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler ___ Glasgow-haskell-bugs mailing list Glasgow-haskell-bugs@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #4999: build fails on powerpc: error: 'ObjectCode' has no member named 'misalignment'
#4999: build fails on powerpc: error: 'ObjectCode' has no member named
'misalignment'
+---
Reporter: nomeata | Owner:
Type: bug | Status: new
Priority: normal | Component: Runtime System
Version: 7.0.2|Keywords:
Testcase: | Blockedby:
Os: Linux|Blocking:
Architecture: powerpc | Failure: Building GHC failed
+---
Comment(by erikd):
I have a patch that works (very thoroughly tested) against 7.0.2 which
I'll add as an attachment.
I've tried to apply the same patch against GHC HEAD and I've run into this
problem:
{{{
rts/Linker.c:1073:7:
error: 'alloc_blocks_lim' undeclared here (not in a function)
rts/Linker.c:1073:7: warning: missing initializer
rts/Linker.c:1073:7:
warning: (near initialization for 'rtsSyms[377].addr')
}}}
which seems completely unrelated to what I did and has me completely
baffled.
--
Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/4999#comment:4
GHC http://www.haskell.org/ghc/
The Glasgow Haskell Compiler
___
Glasgow-haskell-bugs mailing list
Glasgow-haskell-bugs@haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #1408: groupWhen – a groupBy that compares consecutive values
#1408: groupWhen – a groupBy that compares consecutive values --+- Reporter: Joachim Breitner mail@…| Owner: Type: feature request | Status: closed Priority: normal | Milestone: Component: libraries/base |Version: 6.6.1 Resolution: wontfix | Keywords: Testcase: | Blockedby: Difficulty: Easy (less than 1 hour) | Os: Unknown/Multiple Blocking: | Architecture: Unknown/Multiple Failure: None/Unknown | --+- Changes (by Lemming): * cc: ghc@… (added) * failure: = None/Unknown Comment: You may like to use my groupBy in: http://hackage.haskell.org/packages/archive/utility-ht/0.0.5.1/doc/html /Data-List-HT.html#v:groupBy -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/1408#comment:8 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler ___ Glasgow-haskell-bugs mailing list Glasgow-haskell-bugs@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #1408: groupWhen – a groupBy that compares consecutive values
#1408: groupWhen – a groupBy that compares consecutive values --+- Reporter: Joachim Breitner mail@…| Owner: Type: feature request | Status: closed Priority: normal | Milestone: Component: libraries/base |Version: 7.0.2 Resolution: wontfix | Keywords: Testcase: | Blockedby: Difficulty: Easy (less than 1 hour) | Os: Unknown/Multiple Blocking: | Architecture: Unknown/Multiple Failure: None/Unknown | --+- Changes (by Lemming): * version: 6.6.1 = 7.0.2 Comment: It seems to worry many people: http://www.haskell.org/pipermail/haskell-cafe/2011-March/089832.html http://www.haskell.org/pipermail/libraries/2007-August/008039.html http://www.haskell.org/pipermail/haskell-cafe/2007-April/024971.html http://www.haskell.org/haskellwiki/List_function_suggestions#Generalize_groupBy_and_friends -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/1408#comment:9 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler ___ Glasgow-haskell-bugs mailing list Glasgow-haskell-bugs@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
[GHC] #5001: makeCorePair: arity missing
#5001: makeCorePair: arity missing
-+--
Reporter: maeder| Owner:
Type: bug | Status: new
Priority: normal| Component: Compiler
Version: 7.0.2 |Keywords:
Testcase:| Blockedby:
Os: Unknown/Multiple |Blocking:
Architecture: Unknown/Multiple | Failure: Incorrect warning at
compile-time
-+--
when compiling our hets sources with optimization I get unexpected
messages like:
{{{
makeCorePair: arity missing $cmap_sentence{v au5bz} [lid]
makeCorePair: arity missing $cmap_symbol{v au5bC} [lid]
}}}
I think it is emitted by the desugarer i.e. for the source file
http://trac.informatik.uni-
bremen.de:8080/hets/browser/trunk/Comorphisms/OWL2CommonLogic.hs
I've noticed no consequences (bugs), though. Maybe it should only be
emitted when a certain verbosity level is set.
--
Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/5001
GHC http://www.haskell.org/ghc/
The Glasgow Haskell Compiler
___
Glasgow-haskell-bugs mailing list
Glasgow-haskell-bugs@haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
[GHC] #5002: 7.0.2 ignores a context which 7.0.1 picks up
#5002: 7.0.2 ignores a context which 7.0.1 picks up
+---
Reporter: patrick_premont | Owner:
Type: bug | Status: new
Priority: normal | Component: Compiler
Version: 7.0.2|Keywords:
Testcase: | Blockedby:
Os: Windows |Blocking:
Architecture: x86 | Failure: GHC rejects valid program
+---
GHC 7.0.2 rejects programs which 7.0.1 accepts.
A passed context is not used, and the compiler (expectedly) fails to
deduce an instance. Patching the code so that it works again is not
difficult. Some type annotations can do the trick. So this is not a
critical issue but it is a bit surprising.
I have simplified my code as much as possible so that it still shows
the error in 7.0.2 (okIn701). I have included two further
simplifications which produce no error : okInBoth and okInBoth'. I see
why okInBoth is more simple (it side steps a type function), but I do
not see why okInBoth' would avoid the problem.
The code also compiles if we remove the instance
declaration for class B. In that case the type of a in okIn701 can be
infered, and the context for that type is provided.
I have seen the following comment by dimitris in ticket #4981, which
seems related. I know why GHC is not picking the given up: it has to
do with the fact that we have not saturated all possible equalities
before we look for instances, but luckily this is something Simon and
I are planning to fix pretty soon. Ticket #4981 seems to be an issue
with 7.0.1. Here we see an apparent regression with 7.0.2, so I
thought I would bring it up in case it is an unexpected change in
behavior.
The diagnosis of ticket #3018 may be applicable to the code here: we
may be asking too much of the compiler. As an additional
simplification attempt, I have added function fromTicket3018, but it
compiles fine with 7.0.1 and 7.0.2.
Also if the instance for 'B' is restricted to '[a]' (we can then
remove the UndecidableInstances extension), and the type 'a' is
replaced by '[a]' in 'okIn701', then it compiles fine. Are instances
that match everything applied more eagerly ? If so then this
compilatin problem should be quite rare.
{{{
{-# LANGUAGE TypeFamilies, FlexibleInstances, UndecidableInstances,
FlexibleContexts #-}
class A a
class B a where b :: a - ()
instance A a = B [a] where b = undefined
newtype Y a = Y (a - ())
okIn701 :: B [a] = Y [a]
okIn701 = wrap $ const () . b
okInBoth' :: B a = Y a
okInBoth' = wrap $ b
okInBoth :: B a = Y a
okInBoth = Y $ const () . b
class Wrapper a where
type Wrapped a
wrap :: Wrapped a - a
instance Wrapper (Y a) where
type Wrapped (Y a) = a - ()
wrap = Y
fromTicket3018 :: Eq [a] = a - ()
fromTicket3018 x = let {g :: Int - Int; g = [x]==[x] `seq` id} in ()
main = undefined
}}}
--
Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/5002
GHC http://www.haskell.org/ghc/
The Glasgow Haskell Compiler
___
Glasgow-haskell-bugs mailing list
Glasgow-haskell-bugs@haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #4970: time002 and time004 (ghci) test failures on OS X 64 bit
#4970: time002 and time004 (ghci) test failures on OS X 64 bit --+- Reporter: gwright | Owner: igloo Type: bug | Status: new Priority: high | Milestone: 7.2.1 Component: GHCi |Version: 7.0.1 Resolution: | Keywords: Testcase: | Blockedby: Difficulty: | Os: MacOS X Blocking: | Architecture: x86_64 (amd64) Failure: Incorrect result at runtime | --+- Comment(by gwright): Replying to [comment:9 igloo]: It looks like the right thing to do is to define a Haskell type for the C `suseconds_t`, and use that. At least, http://pubs.opengroup.org/onlinepubs/009695399/basedefs/sys/time.h.html and my Linux `gettimeofday(2)` agree on that. That should work on OS X as well, since `__darwin_suseconds_t` is a typedef for `suseconds_t` (defined in `/usr/include/sys/types.h`); `__darwin_suseconds_t` is the type of the `tv_usec` field in `struct timeval`. -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/4970#comment:11 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler ___ Glasgow-haskell-bugs mailing list Glasgow-haskell-bugs@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
[GHC] #5003: test 4038 (ghci) fails on OS X 64 bit.
#5003: test 4038 (ghci) fails on OS X 64 bit.
---+
Reporter: gwright | Owner:
Type: bug | Status: new
Priority: normal | Component: GHCi
Version: 7.0.2 |Keywords:
Testcase: | Blockedby:
Os: MacOS X |Blocking:
Architecture: x86_64 (amd64) | Failure: Runtime crash
---+
Test 4038 (ghci) sometimes fails on OS X 64 bit. When the failure
happens, the test crashes with a segfault or bus error:
{{{
gwright-macbook inplace/bin/ghc-stage2 --interactive 4038.hs
GHCi, version 7.1.20110302: http://www.haskell.org/ghc/ :? for help
Loading package ghc-prim ... linking ... done.
Loading package integer-gmp ... linking ... done.
Loading package base ... linking ... done.
Loading package ffi-1.0 ... linking ... done.
[1 of 1] Compiling Main ( 4038.hs, interpreted )
Ok, modules loaded: Main.
*Main main
Bus error
gwright-macbook
}}}
Under gdb, I get
{{{
gwright-macbook gdb
GNU gdb 6.3.50-20050815 (Apple version gdb-1510) (Wed Sep 22 02:45:02 UTC
2010)
Copyright 2004 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you
are
welcome to change it and/or distribute copies of it under certain
conditions.
Type show copying to see the conditions.
There is absolutely no warranty for GDB. Type show warranty for
details.
This GDB was configured as x86_64-apple-darwin.
(gdb) source gdbinit
(gdb) source gdb_ghc
Reading symbols for shared libraries . done
(gdb) run
Starting program: /Users/gwright/tmp/ghc-7-branch/ghc/inplace/lib/ghc-
stage2 +RTS -V0 -i0 -RTS -B/Users/gwright/tmp/ghc-7-branch/ghc/inplace/lib
-pgmc /usr/bin/gcc-4.2 -pgma /usr/bin/gcc-4.2 -pgml /usr/bin/gcc-4.2 -pgmP
/usr/bin/gcc-4.2 -E -undef -traditional --interactive 4038.hs
Reading symbols for shared libraries . done
GHCi, version 7.0.1.20101221: http://www.haskell.org/ghc/ :? for help
Loading package ghc-prim ... linking ... done.
Loading package integer-gmp ... linking ... done.
Loading package base ... linking ... done.
Loading package ffi-1.0 ... linking ... done.
[1 of 1] Compiling Main ( 4038.hs, interpreted )
Ok, modules loaded: Main.
*Main main
Program received signal EXC_BAD_ACCESS, Could not access memory.
Reason: KERN_PROTECTION_FAILURE at address: 0x000104300fd8
[Switching to process 249]
0x000101f3d8a9 in stg_makeStablePtrzh ()
(gdb)
}}}
The `gdb_ghc` file used to start `gdb` contained:
{{{
file /Users/gwright/tmp/ghc-7-branch/ghc/inplace/lib/ghc-stage2
set args +RTS -V0 -i0 -RTS
-B/Users/gwright/tmp/ghc-7-branch/ghc/inplace/lib -pgmc /usr/bin/gcc-4.2
-pgma /usr/bin/gcc-4.2 -pgml /usr/bin/gcc-4.2 -pgmP /usr/bin/gcc-4.2 -E
-undef -traditional --interactive 4038.hs
}}}
I had copied the failing `4038.hs` file to the top level directory.
The address of the failure, `0x104300fd8`, is in the block of memory
mapped by the linker for `HSghc-prim-0.2.0.0.o`.
--
Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/5003
GHC http://www.haskell.org/ghc/
The Glasgow Haskell Compiler
___
Glasgow-haskell-bugs mailing list
Glasgow-haskell-bugs@haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
[GHC] #5004: yesod-0.7.1 fails to build on ghc 7.0.2
#5004: yesod-0.7.1 fails to build on ghc 7.0.2
---+
Reporter: pbrisbin| Owner:
Type: bug | Status: new
Priority: normal | Component: Compiler
Version: 7.0.2 |Keywords:
Testcase: | Blockedby:
Os: Linux |Blocking:
Architecture: x86_64 (amd64) | Failure: Compile-time crash
---+
Arch linux's ghc package was upgraded to 7.0.2 this weekend. I
unregistered all cabal packages, removed ~/.cabal/packages/*/* and ~/.ghc
in an effort to start fresh.
I successfully installed quite a few packages via cabal install.
You can find my current `ghc-pkg list`
[http://pbrisbin.com/static/fileshare/ghc_list.txt here].
At `cabal install yesod`, I get the following:
{{{
...
Loading package MonadCatchIO-mtl-0.3.0.2 ... linking ... done.
Loading package Cabal-1.10.1.0 ... linking ... done.
Loading package ghc-binary-0.5.0.2 ... linking ... done.
Loading package bin-package-db-0.0.0.0 ... linking ... done.
Loading package hpc-0.5.0.6 ... linking ... done.
Loading package ghc-7.0.2 ... ghc: This ELF file contains no symtab
ghc: panic! (the 'impossible' happened)
(GHC version 7.0.2 for x86_64-unknown-linux):
loadArchive /usr/lib/ghc-7.0.2/ghc-7.0.2/libHSghc-7.0.2.a:
failed
Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug
cabal: Error: some packages failed to install:
yesod-0.7.1 failed during the building phase. The exception was:
ExitFailure 1
}}}
I'm reporting here because I was specifically asked to by ghc (I almost
always assume these things are my fault), please let me know what
additional info you would like in this report.
Thanks.
--
Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/5004
GHC http://www.haskell.org/ghc/
The Glasgow Haskell Compiler
___
Glasgow-haskell-bugs mailing list
Glasgow-haskell-bugs@haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #5004: yesod-0.7.1 causes panic on Arch's ghc 7.0.2 (was: yesod-0.7.1 fails to build on ghc 7.0.2)
#5004: yesod-0.7.1 causes panic on Arch's ghc 7.0.2 ---+ Reporter: pbrisbin| Owner: Type: bug | Status: new Priority: normal | Component: Compiler Version: 7.0.2 |Keywords: Testcase: | Blockedby: Os: Linux |Blocking: Architecture: x86_64 (amd64) | Failure: Compile-time crash ---+ -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/5004#comment:1 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler ___ Glasgow-haskell-bugs mailing list Glasgow-haskell-bugs@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #4247: getCPUTime on x86_64 Mac OS X 10.6
#4247: getCPUTime on x86_64 Mac OS X 10.6 ---+ Reporter: quark |Owner: igloo Type: bug | Status: new Priority: normal |Milestone: 7.2.1 Component: libraries/base | Version: 6.10.4 Keywords: | Testcase: Blockedby: | Difficulty: Os: MacOS X | Blocking: Architecture: x86_64 (amd64) | Failure: Incorrect result at runtime ---+ Comment(by gwright): This is essentially the same bug as #4970, so it looks like more than one file needs to be fixed. -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/4247#comment:8 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler ___ Glasgow-haskell-bugs mailing list Glasgow-haskell-bugs@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #1408: groupWhen – a groupBy that compares consecutive values
#1408: groupWhen – a groupBy that compares consecutive values --+- Reporter: Joachim Breitner mail@…| Owner: Type: feature request | Status: closed Priority: normal | Milestone: Component: libraries/base |Version: 7.0.2 Resolution: wontfix | Keywords: Testcase: | Blockedby: Difficulty: Easy (less than 1 hour) | Os: Unknown/Multiple Blocking: | Architecture: Unknown/Multiple Failure: None/Unknown | --+- Changes (by Remi): * cc: rturk@… (added) -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/1408#comment:10 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler ___ Glasgow-haskell-bugs mailing list Glasgow-haskell-bugs@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #4318: Crash while building HEAD on OS X
#4318: Crash while building HEAD on OS X ---+ Reporter: gwright |Owner: Type: bug | Status: new Priority: normal |Milestone: 7.2.1 Component: Compiler| Version: 6.13 Keywords: | Testcase: Blockedby: | Difficulty: Os: MacOS X | Blocking: Architecture: x86_64 (amd64) | Failure: Building GHC failed ---+ Comment(by gwright): This bug should be closed. The patch that fixed most of the problem (incorrect lookup of external symbols) was applied although I didn't attach the patch to this ticket. The most obvious remaining linker bug was repaired by #4867. -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/4318#comment:21 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler ___ Glasgow-haskell-bugs mailing list Glasgow-haskell-bugs@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #4995: Compiling pandoc with llvm backend fails with panic
#4995: Compiling pandoc with llvm backend fails with panic
+---
Reporter: jgm |Owner: dterei
Type: bug | Status: new
Priority: normal |Milestone: 7.2.1
Component: Compiler (LLVM) | Version: 7.0.2
Keywords: llvm pandoc | Testcase:
Blockedby: | Difficulty:
Os: MacOS X | Blocking:
Architecture: x86 | Failure: None/Unknown
+---
Comment(by dterei):
jgm would you be able to do this for me please:
{{{
ghc -fllvm -O2 -c -keep-llvm-files -keep-s-files
src/Text/Pandoc/Readers/LaTeX.hs
}}}
Then send me/attach to this bug the LaTex.s and LaTeX.ll files that should
appear. I have only limited access to an OSX machine, enough to build ghc
and run the ghc test suite but installing cabal packages gets a little
difficult. So I think this will be the quickest and easiest way for me to
be able to fix the problem.
Thanks
--
Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/4995#comment:5
GHC http://www.haskell.org/ghc/
The Glasgow Haskell Compiler
___
Glasgow-haskell-bugs mailing list
Glasgow-haskell-bugs@haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #4995: Compiling pandoc with llvm backend fails with panic
#4995: Compiling pandoc with llvm backend fails with panic
+---
Reporter: jgm |Owner: dterei
Type: bug | Status: new
Priority: normal |Milestone: 7.2.1
Component: Compiler (LLVM) | Version: 7.0.2
Keywords: llvm pandoc | Testcase:
Blockedby: | Difficulty:
Os: MacOS X | Blocking:
Architecture: x86 | Failure: None/Unknown
+---
Comment(by jgm):
Unfortunately there's a 256K limit on attachments, and these files are far
larger.
{{{
-rw-r--r-- 1 jgm jgm 31678880 Mar 7 19:00
src/Text/Pandoc/Readers/LaTeX.ll
-rw-r--r-- 1 jgm jgm 9603608 Mar 7 19:00
src/Text/Pandoc/Readers/LaTeX.s
}}}
Should I email them to you? If so, send me an email directly so I know
where to send them.
jgm at berkeley dot edu
--
Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/4995#comment:6
GHC http://www.haskell.org/ghc/
The Glasgow Haskell Compiler
___
Glasgow-haskell-bugs mailing list
Glasgow-haskell-bugs@haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #5003: test 4038 (ghci) fails on OS X 64 bit.
#5003: test 4038 (ghci) fails on OS X 64 bit. ---+ Reporter: gwright | Owner: gwright Type: bug | Status: new Priority: normal | Component: GHCi Version: 7.0.2 |Keywords: Testcase: | Blockedby: Os: MacOS X |Blocking: Architecture: x86_64 (amd64) | Failure: Runtime crash ---+ Changes (by PHO): * cc: pho@… (added) -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/5003#comment:2 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler ___ Glasgow-haskell-bugs mailing list Glasgow-haskell-bugs@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: [GHC] #4867: ghci displays negative floats incorrectly (was: Incorrect result from trig functions)
#4867: ghci displays negative floats incorrectly (was: Incorrect result from trig functions) --+- Reporter: gwright | Owner: gwright Type: bug | Status: closed Priority: high | Milestone: 7.0.2 Component: GHCi |Version: 7.0.1 Resolution: fixed| Keywords: Testcase: | Blockedby: Difficulty: | Os: MacOS X Blocking: | Architecture: x86_64 (amd64) Failure: Incorrect result at runtime | --+- Changes (by PHO): * cc: pho@… (added) -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/4867#comment:50 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler ___ Glasgow-haskell-bugs mailing list Glasgow-haskell-bugs@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
Re: semaphore functions in the wrong place during cross-build
On Sat, Mar 5, 2011 at 4:39 PM, Ian Lynagh ig...@earth.li wrote: On Fri, Mar 04, 2011 at 10:46:13AM -0800, Tyler Pirtle wrote: I'm not sure where the right place is to configure this. While linking hsunix + some others, the compiler assumes that -lrt is the appropriate library to find semaphore-related functions, but for the particular toolchain i'm building against, those functions are actually in -lpthread. Is there an easy way i can configure this? What version are you using? This sounds like it might be: http://hackage.haskell.org/trac/ghc/ticket/4523 Ah! It was a bug in 7.0.1 indeed. 7.0.2 fixed it. Thanks, Ian! Fixed in 7.0.2 and HEAD. Thanks Ian ___ Glasgow-haskell-users mailing list Glasgow-haskell-users@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-users
Re: Configuring specific packges in libraries/ during ghc build
I ultimately found what I was looking for, but i'm wondering if there's a better source of documentation for it. There are variables strewn around the build system that look like: libraries/time_dist-install_EXTRA_HC_OPTS += -fno-warn-unused-do-bind Or, sometimes more generally, $1_$2_CONFIGURE_OPTS += --enable-shared By looking at the last bit of these variables I'm able to configure my package correctly and so it's al very good (and i can define these in my mk/build.mk). But is there a listing of ALL of the suffixes that are available here (CONFIGURE_OPTS, CC_OPTS, HC_OPTS, etc) ? On Fri, Mar 4, 2011 at 1:04 PM, Tyler Pirtle tee...@gmail.com wrote: My toolchain is a little silly. ;) It doesn't have ncurses in the obvious place, so by the time my build gets around to doing libraries/terminfo, the terminfo configure bails because it can't find the curses headers. I can work around this, by building the library manually, but is there an easy way for me to get a hold of specific cabal options when calling specific packages during build? inplace/bin/ghc-cabal configure --with-ghc=/home/teeler/Downloads/ghc-7.0.2/inplace/bin/dummy-ghc --with-ghc-pkg=/home/teeler/Downloads/ghc-7.0.2/inplace/bin/ghc-pkg --with-gcc=/some/weird/bin/gcc --configure-option=--with-cc=/some/weird/bin/gcc --enable-library-for-ghci --configure-option=CFLAGS= -fno-stack-protector --configure-option=LDFLAGS= --configure-option=CPPFLAGS= -- dist-install libraries/terminfo I need to add some extra options to the configure command here. ___ Glasgow-haskell-users mailing list Glasgow-haskell-users@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-users
bin-package-db
Hello, The release note of GHC 7.0.2 says: 1.5.6.3. bin-package-db This is an internal package, and should not be used. What is the intention of this change? The author of cabal-delete hesitates to register cabal-delete into HackageDB due to this words. Shouldn't a package-maintenance command even use this package? If so, what package should be used? --Kazu ___ Glasgow-haskell-users mailing list Glasgow-haskell-users@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-users
Re: [Haskell] ANNOUNCE: silently-0.0.1 (prevent IO actions from writing to stdout)
On 7 March 2011 02:28, Trystan Spangler trysta...@comcast.net wrote: Announcing Silently, a package with two simple functions to run an IO action while preventing it from writing to stdout (or the given handle) Useful package! I tiny remark about the code: -- | Run an IO action while ignoring all output to stdout. silently :: IO a - IO a silently = hSilently stdout -- | Run an IO action while ignoring all output to the given handle. hSilently :: Handle - IO a - IO a hSilently handle action = do oldHandle - hDuplicate handle (tmpFile, tmpHandle) - openTempFile . silently hDuplicateTo tmpHandle handle finally action (hDuplicateTo oldHandle handle hClose tmpHandle removeFile tmpFile) Think about what happens if an asynchronous exception is thrown to the thread executing hSilently just after opening the temp file. It will abort the computation which causes the temp file to be left on your system and keep the tmpHandle open. You need to use bracket to ensure the temp file we always be closed and deleted. I send you a pull-request. Regards, Bas ___ Haskell mailing list Haskell@haskell.org http://www.haskell.org/mailman/listinfo/haskell
Re: [arch-haskell] Spec ready to go!
Hi Rémy, Is there anything depending on haskell-packedstring in the haskell repo? no, there isn't. The package appears to be obsolete. Take care, Peter ___ arch-haskell mailing list arch-hask...@haskell.org http://www.haskell.org/mailman/listinfo/arch-haskell
[Haskell] ANNOUNCE: silently-0.0.2 (thread safe and can capture output)
Silently has doubled in size with two new functions to capture the output of an IO action: capture :: IO a - IO (String, a) hCapture :: Handle - IO a - IO (String, a) All 4 functions now clean up any temp file even if there's an asynchronous exception but it still uses temp files behind-the-scenes; I'll get to that when I can in a week or so. http://hackage.haskell.org/package/silently https://github.com/trystan/silently A big thanks to everyone who had comments and suggestions! Trystan Spangler ___ Haskell mailing list Haskell@haskell.org http://www.haskell.org/mailman/listinfo/haskell
[Haskell] Linker flags for foreign export.
I am having no luck generating a foreign export and calling it from C. The problem seems to be that I must spoon feed the linker. My working example is the files under this directory on Github: https://github.com/solidsnack/bash/tree/106e4de8358a726984d54859fb0b4aeaf264156f/hs/exports Running `make' to produce the executable that imports the binding yields a number of unresolved linkages: :; make import ghc --make -fPIC -dynamic -c exports.hs [1 of 1] Compiling LanguageBashExports ( exports.hs, exports.o ) ghc -shared exports.o exports_stub.o -o exports.so \ -L/usr/lib/ghc-6.12.1/ -lHSrts \ -L/usr/lib/ghc-6.12.1/ghc-prim-0.2.0.0/ -lHSghc-prim-0.2.0.0 \ -L/usr/lib/ghc-6.12.1/integer-gmp-0.2.0.0/ -lHSinteger-gmp-0.2.0.0 \ -L/usr/lib/ghc-6.12.1/base-4.2.0.0/ -lHSbase-4.2.0.0 \ -lffi gcc -g -Wall -O2 -fPIC -Wall -o import \ -I/usr/lib/ghc-6.12.1/include/ \ import.c exports.so exports.so: undefined reference to `timer_settime' exports.so: undefined reference to `__stginit_base_DataziWord_dyn' exports.so: undefined reference to `clock_gettime' exports.so: undefined reference to `__stginit_base_Prelude_dyn' exports.so: undefined reference to `__stginit_base_ForeignziCziTypes_dyn' exports.so: undefined reference to `timer_delete' exports.so: undefined reference to `timer_create' exports.so: undefined reference to `hs_free_stable_ptr' collect2: ld returned 1 exit status make: *** [import] Error 1 If I add linker directives for libHSrts and librt to the compilation of import.c, the only unresolved symbols remaining are the STG init functions for the Prelude, Data.Word and Foreign.C.Types: :; make import gcc -g -Wall -O2 -fPIC -Wall -o import \ -I/usr/lib/ghc-6.12.1/include/ \ import.c exports.so \ -L/usr/lib/ghc-6.12.1/ -lHSrts \ -lrt exports.so: undefined reference to `__stginit_base_DataziWord_dyn' exports.so: undefined reference to `__stginit_base_Prelude_dyn' exports.so: undefined reference to `__stginit_base_ForeignziCziTypes_dyn' collect2: ld returned 1 exit status make: *** [import] Error 1 This hide and go seek with symbols really makes me feel like I'm just doing it wrong, though. I linked exports.so statically so why shouldn't it include `hs_free_stable_ptr' already (as well as the stginit functions)? Not sure why I need to mention librt at this step and not at the previous one. The truth is, I don't write C very often and in fact I am only doing this to create a bridge to Ruby. How should I be approaching this? Is there an SO-chasing option? -- Jason Dusek Linux User #510144 | http://counter.li.org/ ___ Haskell mailing list Haskell@haskell.org http://www.haskell.org/mailman/listinfo/haskell
Re: [Haskell-cafe] Haskell IDE
I use kate too. I tried and liked leksah, but the fact that everything is a project with a cabal file felt to heavy for me when I just want to hack on a single .hs file. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] How to daemonize a threaded Haskell program?
Sterling, Gregory, Brandon and David thanks for your suggestions. On 6 March 2011 05:38, David Anderson d...@natulte.net wrote: I humbly recommend doing such daemonizations from outside your program. Programs that daemonize on startup make it very difficult to monitor them by direct means, instead forcing you to rely on PID files and other mechanisms which may not always be available or fresh. Agreed, I already noticed that debugging is a bit harder since I have no stdout and stderr anymore. For reference, Upstart, the new PID 1 on Ubuntu and friends, has a horrible hack[1] built in specifically to keep track of processes that helpfully daemonize themselves, so that it can offer additional services like restarting crashed services or notifying the owner of a problem. As has been pointed out elsewhere on thread, there are plenty of standalone programs that implement daemonization if you end up not using service management software to do it for you. I plan to run the daemon on a Ubuntu server. Do you know if Upstart is able to daemonize a process? Regards, Bas ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] How to daemonize a threaded Haskell program?
On Mon, Mar 7, 2011 at 12:39 AM, Bas van Dijk v.dijk@gmail.com wrote: Sterling, Gregory, Brandon and David thanks for your suggestions. On 6 March 2011 05:38, David Anderson d...@natulte.net wrote: I humbly recommend doing such daemonizations from outside your program. Programs that daemonize on startup make it very difficult to monitor them by direct means, instead forcing you to rely on PID files and other mechanisms which may not always be available or fresh. Agreed, I already noticed that debugging is a bit harder since I have no stdout and stderr anymore. For reference, Upstart, the new PID 1 on Ubuntu and friends, has a horrible hack[1] built in specifically to keep track of processes that helpfully daemonize themselves, so that it can offer additional services like restarting crashed services or notifying the owner of a problem. As has been pointed out elsewhere on thread, there are plenty of standalone programs that implement daemonization if you end up not using service management software to do it for you. I plan to run the daemon on a Ubuntu server. Do you know if Upstart is able to daemonize a process? It's perfectly simple. In fact, it's the default assumption of Upstart that your program doesn't daemonize, and that Upstart should help it. You have to specify if your program violates that assumption. Here's a sample file. It specifies a service that; should run whenever the main runlevels are entered (i.e. basic system init like mounting filesystems is done, but no other guarantee); stops when the system leaves those runlevels; runs a binary that does not fork or daemonize; respawns the service if the process should die for any reason; can be manually stopped and restarted by the admin if he feels like it. $ cat /etc/init/demo.conf description Demo service that doesn't daemonize by itself start on runlevel [2345] stop on runlevel [!2345] respawn exec /path/to/binary $ sudo start demo demo start/running, process 27469 $ sudo stop demo demo stop/waiting $ sudo status demo demo stop/waiting $ Hope this helps. The upstart configuration format is a little sparsely documented, when I wrote scripts for it, I used the existing files in /etc/init as a rough guide. - Dave Regards, Bas ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] ANNOUNCE: cinvoke 0.1 released
Hi Remi, On 6 March 2011 13:38, Remi Turk rt...@science.uva.nl wrote: I am happy to finally announce cinvoke 0.1, a binding to the C library cinvoke[1], allowing functions to be loaded and called whose names and types are not known before run-time. As the author of the libffi package (http://hackage.haskell.org/package/libffi-0.1) which does a similar thing, could you say when it would be appropriate to use one or the other package? Cheers, Max ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] Overlaping Parsec rules
Hi,
to solve this ICFP task http://www.cs.cornell.edu/icfp/task.htm I'm currnetly
working on the parser. With the hint from Thu (reading Phillip Wadlers monadic
parser paper) and consulting
http://en.wikibooks.org/wiki/Write_Yourself_a_Scheme_in_48_Hours/Parsing I
produce a first working version of the parser.
After this great moment I completed the token definition and near all parsing
rules. For the complete code have a look at
http://code.google.com/p/hgmltracer/source/browse/#svn%2Ftrunk%2FhGmlTracer
data GmlToken =
-- structures
FunctionToken TokenSequence |
ArrayToken TokenSequence|
-- control operators
ApplyToken |
IfToken |
-- number operators
AddiToken |
AddfToken |
ACosToken |
ASinToken |
ClampfToken |
CosToken|
DiviToken |
DivfToken |
EqiToken|
EqfToken|
FloorToken |
FracToken |
LessiToken |
LessfToken |
ModiToken |
MuliToken |
MulfToken |
NegiToken |
NegfToken |
ToRealToken |
SinToken|
SqrtToken |
SubiToken |
SubfToken |
-- points
GetXToken |
GetYToken |
GetZToken |
PointToken |
-- arrays
GetToken|
LengthToken |
-- environment
IdentifierToken String |
BinderToken String |
-- types
BoolToken Bool |
IntToken Int|
RealToken Double|
StringToken String deriving Show
And
parseGml :: String - [GmlToken]
parseGml input = case parse parseList gml input of
Left err - error (Parse error: ++ (show err))
Right gml - gml
parseList = sepBy parseGml' spaces
parseGml' =
-- structures
--parseFunction
--| parseArray
-- control operators
parseControlOperators
-- number operators
| parseNumberOperators
-- points
| parsePointOperators
-- arrays
| parseArrayOperators
-- types
| parseBool
| parseInteger
| parseString
-- environment
| parseIdentifier
| parseBinder
parseArray = parseSequence '[' ']'
parseFunction = parseSequence '{' '}'
parseSequence start end = do char start
array - parseList
char end
return $ ArrayToken array
parseControlOperators = parseApply | parseIf
parseApply = do string apply
return $ ApplyToken
parseIf = do string if
return $ IfToken
parseNumberOperators = do string addi
return $ AddiToken
| do string addf
return $ AddfToken
| do string acos
return $ ACosToken
| do string asind
return $ ASinToken
| do string clampf
return $ ClampfToken
| do string cos
return $ CosToken
| do string divi
return $ DiviToken
| do string divf
return $ DivfToken
| do string eqi
return $ EqiToken
| do string eqf
return $ EqfToken
| do string floor
return $ FloorToken
| do string frac
return $ FracToken
| do string lessi
return $ LessiToken
| do string lessf
return $ LessfToken
| do string modi
return $ ModiToken
| do string muli
return $ MuliToken
| do string mulf
return $ MulfToken
| do string negi
return $ NegiToken
[Haskell-cafe] Explicitly Typed Exceptions in Haskell 98 (Was: Idiomatic error handling in Haskell)
On Wed, 2 Mar 2011, Henning Thielemann wrote: On Wed, 2 Mar 2011, Rouan van Dalen wrote: I would like to know what is the preferred Haskell mechanism for handling exceptions in the IO monad? I am not concerned with mechanisms such as Maybe / Either, but would like to know about exception mechanisms inside the IO monad. The 2 I know of are: o) throwDyn o) ioError and catch I do need the exceptions to be extendable. So which is the preferred way to handle exceptions in Haskell for new libs? I recently had an idea of how to design extensible explicit type safe exceptions in Haskell 98, i.e. with single parameter type classes and non-overlapping instances. It seems to work quite well, the only drawback is that you have to define n^2 instances for n exceptions. This is much inspired by: http://users.dsic.upv.es/~jiborra/papers/explicitexceptions.pdf Consider two exceptions: ReadException and WriteException. In order to be able to freely combine these exceptions, we use type classes, since type constraints of two function calls are automatically merged. import Control.Monad.Exception.Synchronous (ExceptionalT, ) class ThrowsRead e where throwRead :: e class ThrowsWrite e where throwWrite :: e readFile :: ThrowsRead e = FilePath - ExceptionalT e IO String writeFile :: ThrowsWrite e = FilePath - String - ExceptionalT e IO () For example for copyFile src dst = writeFile dst = readFile src the compiler automatically infers copyFile :: (ThrowsWrite e, ThrowsRead e) = FilePath - FilePath - ExceptionalT e IO () Instead of ExceptionalT you can also use EitherT or ErrorT. It's also simple to add parameters to throwRead and throwWrite, such that you can pass more precise information along with the exception. I just want to keep it simple for now. With those definitions you can already write a nice library and defer the decision of the particular exception types to the library user. The user might define something like data ApplicationException = ReadException | WriteException instance ThrowsRead ApplicationException where throwRead = ReadException instance ThrowsWrite ApplicationException where throwWrite = WriteException Using ApplicationException however it is cumbersome to handle only ReadException and propagate WriteException. The user might write something like case e of ReadException - handleReadException WriteException - throwT throwWrite in order to handle a ReadException and regenerate a 'ThrowWrite e = e' type variable, instead of the concrete ApplicationException type. He may choose to switch on multi-parameter type classes and overlapping instances, define an exception type like 'data EE l' and then use the technique from control-monad-exception for exception handling with the ExceptionalT monads. Now I like to propose a technique for handling a particular set of exceptions in Haskell 98: data ReadException e = ReadException | NoReadException e instance ThrowsRead (ReadException e) where throwRead = ReadException instance ThrowsWrite e = ThrowsWrite (ReadException e) where throwWrite = NoReadException throwWrite data WriteException e = WriteException | NoWriteException e instance ThrowsRead e = ThrowsRead (WriteException e) where throwRead = NoWriteException throwRead instance ThrowsWrite (WriteException e) where throwWrite = WriteException Defining exception types as a sum of this particular exception and another exception lets us compose concrete types that can carry a certain set of exceptions on the fly. This is very similar to switching from particular monads to monad transformers. Thanks to the type class approach the order of composition needs not to be fixed by the throwing function but is determined by the order of catching. We even do not have to fix the nested exception type fully when catching an exception. It is enough to fix the part that is interesting for 'catch': import Control.Monad.Exception.Synchronous (Exceptional(Success,Exception)) catchRead :: ReadException e - Exceptional e String catchRead ReadException = Success catched a read exception catchRead (NoReadException e) = Exception e throwReadWrite :: (ThrowsRead e, ThrowsWrite e) = e throwReadWrite = asTypeOf throwRead throwWrite exampleCatchRead :: (ThrowsWrite e) = Exceptional e String exampleCatchRead = catchRead throwReadWrite Note how in exampleCatchRead the constraint ThrowsRead is removed from the constraint list of throwReadWrite. As I said, the nasty thing is, that the library has to define n^2 instances for n exceptions. Even worse, if an application imports package A and package B with their sets of exceptions, you have to make the exception types of A instances of the exception class of B and vice versa, and these are orphan instances. However I am still uncertain, how sophisticated an exception system really must be. In principle it must be
Re: [Haskell-cafe] Overlaping Parsec rules
You should parse keywords using:
keyword s = try (string s) notFollowedBy (letter | digit)
C.
Am 07.03.2011 11:34, schrieb Hauschild, Klaus (EXT):
Hi,
to solve this ICFP task _http://www.cs.cornell.edu/icfp/task.htm_ I'm
currnetly working on the parser. With the hint from Thu (reading Phillip
Wadlers monadic parser paper) and consulting
_http://en.wikibooks.org/wiki/Write_Yourself_a_Scheme_in_48_Hours/Parsing_
I produce a first working version of the parser.
After this great moment I completed the token definition and near all
parsing rules. For the complete code have a look at
_http://code.google.com/p/hgmltracer/source/browse/#svn%2Ftrunk%2FhGmlTracer_
data GmlToken =
-- structures
FunctionToken TokenSequence |
ArrayToken TokenSequence|
-- control operators
ApplyToken |
IfToken |
-- number operators
AddiToken |
AddfToken |
ACosToken |
ASinToken |
ClampfToken |
CosToken|
DiviToken |
DivfToken |
EqiToken|
EqfToken|
FloorToken |
FracToken |
LessiToken |
LessfToken |
ModiToken |
MuliToken |
MulfToken |
NegiToken |
NegfToken |
ToRealToken |
SinToken|
SqrtToken |
SubiToken |
SubfToken |
-- points
GetXToken |
GetYToken |
GetZToken |
PointToken |
-- arrays
GetToken|
LengthToken |
-- environment
IdentifierToken String |
BinderToken String |
-- types
BoolToken Bool |
IntToken Int|
RealToken Double|
StringToken String deriving Show
And
parseGml :: String - [GmlToken]
parseGml input = case parse parseList gml input of
Left err - error (Parse error: ++ (show err))
Right gml - gml
parseList = sepBy parseGml' spaces
parseGml' =
-- structures
--parseFunction
--| parseArray
-- control operators
parseControlOperators
-- number operators
| parseNumberOperators
-- points
| parsePointOperators
-- arrays
| parseArrayOperators
-- types
| parseBool
| parseInteger
| parseString
-- environment
| parseIdentifier
| parseBinder
parseArray = parseSequence '[' ']'
parseFunction = parseSequence '{' '}'
parseSequence start end = do char start
array - parseList
char end
return $ ArrayToken array
parseControlOperators = parseApply | parseIf
parseApply = do string apply
return $ ApplyToken
parseIf = do string if
return $ IfToken
parseNumberOperators = do string addi
return $ AddiToken
| do string addf
return $ AddfToken
| do string acos
return $ ACosToken
| do string asind
return $ ASinToken
| do string clampf
return $ ClampfToken
| do string cos
Re: [Haskell-cafe] [Haskell] ANNOUNCE: silently-0.0.1 (prevent IO actions from writing to stdout)
Hello! I wouldn't expect that a function that ignores the output directs it to a temporary file. I think that should be clearly stated on the documentation, as this may be undesirable. I don't know if there is a better way of solving this problem, though. (Note, however, that on Unix systems you could probably direct the output to /dev/null.) Cheers, -- Felipe. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] groupBy huh?
Jacek Generowicz schrieb: Hi Cafe, It seems that I don't understand what groupBy does. I expect it to group together elements as long as adjacent ones satisfy the predicate, so I would expect ALL four of the following to give one group of 3 and a group of 1. Prelude :m + Data.List Prelude Data.List groupBy () abcb [abcb] Prelude Data.List groupBy () abca [abc,a] Prelude Data.List groupBy () [1,2,3,2] [[1,2,3,2]] Prelude Data.List groupBy () [1,2,3,1] [[1,2,3],[1]] What am I missing? http://hackage.haskell.org/packages/archive/utility-ht/0.0.5.1/doc/html/Data-List-HT.html#v:groupBy ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Overlaping Parsec rules
Thanks Christian,
I adapted the keyword parser and now n - negi does not occur.
But there are still other problems to solve. If I activate the parseFunction
the parser will answer to fact.gml unexpected end of input, expecting space or
}.
Any ideas?
-Ursprüngliche Nachricht-
Von: Christian Maeder [mailto:christian.mae...@dfki.de]
Gesendet: Montag, 7. März 2011 12:23
An: Hauschild, Klaus (EXT)
Cc: haskell-cafe@haskell.org
Betreff: Re: Overlaping Parsec rules
You should parse keywords using:
keyword s = try (string s) notFollowedBy (letter | digit)
C.
Am 07.03.2011 11:34, schrieb Hauschild, Klaus (EXT):
Hi,
to solve this ICFP task _http://www.cs.cornell.edu/icfp/task.htm_ I'm
currnetly working on the parser. With the hint from Thu (reading Phillip
Wadlers monadic parser paper) and consulting
_http://en.wikibooks.org/wiki/Write_Yourself_a_Scheme_in_48_Hours/Parsing_
I produce a first working version of the parser.
After this great moment I completed the token definition and near all
parsing rules. For the complete code have a look at
_http://code.google.com/p/hgmltracer/source/browse/#svn%2Ftrunk%2FhGmlTracer_
data GmlToken =
-- structures
FunctionToken TokenSequence |
ArrayToken TokenSequence|
-- control operators
ApplyToken |
IfToken |
-- number operators
AddiToken |
AddfToken |
ACosToken |
ASinToken |
ClampfToken |
CosToken|
DiviToken |
DivfToken |
EqiToken|
EqfToken|
FloorToken |
FracToken |
LessiToken |
LessfToken |
ModiToken |
MuliToken |
MulfToken |
NegiToken |
NegfToken |
ToRealToken |
SinToken|
SqrtToken |
SubiToken |
SubfToken |
-- points
GetXToken |
GetYToken |
GetZToken |
PointToken |
-- arrays
GetToken|
LengthToken |
-- environment
IdentifierToken String |
BinderToken String |
-- types
BoolToken Bool |
IntToken Int|
RealToken Double|
StringToken String deriving Show
And
parseGml :: String - [GmlToken]
parseGml input = case parse parseList gml input of
Left err - error (Parse error: ++ (show err))
Right gml - gml
parseList = sepBy parseGml' spaces
parseGml' =
-- structures
--parseFunction
--| parseArray
-- control operators
parseControlOperators
-- number operators
| parseNumberOperators
-- points
| parsePointOperators
-- arrays
| parseArrayOperators
-- types
| parseBool
| parseInteger
| parseString
-- environment
| parseIdentifier
| parseBinder
parseArray = parseSequence '[' ']'
parseFunction = parseSequence '{' '}'
parseSequence start end = do char start
array - parseList
char end
return $ ArrayToken array
parseControlOperators = parseApply | parseIf
parseApply = do string apply
return $ ApplyToken
parseIf = do string if
return $ IfToken
parseNumberOperators = do string addi
[Haskell-cafe] ghc in debian sid
I tried to move from debian testing to sid and got the message that a whole load of libghc**'s would be removed. Are more recent ones lined up? ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] ghc in debian sid
There are other instabilities (sid is after all unstable) so staying with testing for now. On Mar 7, 5:59 pm, rustom rustompm...@gmail.com wrote: I tried to move from debian testing to sid and got the message that a whole load of libghc**'s would be removed. Are more recent ones lined up? ___ Haskell-Cafe mailing list Haskell-C...@haskell.orghttp://www.haskell.org/mailman/listinfo/haskell-cafe ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Overlaping Parsec rules
Am 07.03.2011 13:48, schrieb Hauschild, Klaus (EXT):
Thanks Christian,
I adapted the keyword parser and now n - negi does not occur.
But there are still other problems to solve. If I activate the parseFunction
the parser will answer to fact.gml unexpected end of input, expecting space
or }.
Your redefinition of spaces (= skipMany1 space) is problematic and:
parseList = sepBy parseGml' spaces
Allow for the original (optional) spaces after parseGml':
parseGml'' = liftM2 const parseGml' spaces
parseList = many parseGml''
C.
P.S. why do you call? many (noneOf )
manyTill anyChar newline
or just: many (noneOf \n)
(a trailing newline will be skipped by spaces)
Any ideas?
-Ursprüngliche Nachricht-
Von: Christian Maeder [mailto:christian.mae...@dfki.de]
Gesendet: Montag, 7. März 2011 12:23
An: Hauschild, Klaus (EXT)
Cc: haskell-cafe@haskell.org
Betreff: Re: Overlaping Parsec rules
You should parse keywords using:
keyword s = try (string s) notFollowedBy (letter | digit)
C.
Am 07.03.2011 11:34, schrieb Hauschild, Klaus (EXT):
Hi,
to solve this ICFP task _http://www.cs.cornell.edu/icfp/task.htm_ I'm
currnetly working on the parser. With the hint from Thu (reading Phillip
Wadlers monadic parser paper) and consulting
_http://en.wikibooks.org/wiki/Write_Yourself_a_Scheme_in_48_Hours/Parsing_
I produce a first working version of the parser.
After this great moment I completed the token definition and near all
parsing rules. For the complete code have a look at
_http://code.google.com/p/hgmltracer/source/browse/#svn%2Ftrunk%2FhGmlTracer_
data GmlToken =
-- structures
FunctionToken TokenSequence |
ArrayToken TokenSequence|
-- control operators
ApplyToken |
IfToken |
-- number operators
AddiToken |
AddfToken |
ACosToken |
ASinToken |
ClampfToken |
CosToken|
DiviToken |
DivfToken |
EqiToken|
EqfToken|
FloorToken |
FracToken |
LessiToken |
LessfToken |
ModiToken |
MuliToken |
MulfToken |
NegiToken |
NegfToken |
ToRealToken |
SinToken|
SqrtToken |
SubiToken |
SubfToken |
-- points
GetXToken |
GetYToken |
GetZToken |
PointToken |
-- arrays
GetToken|
LengthToken |
-- environment
IdentifierToken String |
BinderToken String |
-- types
BoolToken Bool |
IntToken Int|
RealToken Double|
StringToken String deriving Show
And
parseGml :: String - [GmlToken]
parseGml input = case parse parseList gml input of
Left err - error (Parse error: ++ (show err))
Right gml - gml
parseList = sepBy parseGml' spaces
parseGml' =
-- structures
--parseFunction
--| parseArray
-- control operators
parseControlOperators
-- number operators
| parseNumberOperators
-- points
| parsePointOperators
-- arrays
| parseArrayOperators
-- types
| parseBool
| parseInteger
| parseString
-- environment
| parseIdentifier
| parseBinder
parseArray = parseSequence '[' ']'
parseFunction = parseSequence '{' '}'
parseSequence start end = do char start
Re: [Haskell-cafe] Overlaping Parsec rules
Hi Klaus Unless you have very specific white-space considerations, you really want to be using Parsec's Token and LanguageDef modules. There are examples in the legacy Parsec 2.0 distribution available from Daan Leijen's old website: http://legacy.cs.uu.nl/daan/parsec.html Best wishes Stephen ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Overlaping Parsec rules
On Monday 07 March 2011 13:48:31, Hauschild, Klaus (EXT) wrote: But there are still other problems to solve. One thing that I saw which wasn't mentioned before (or I missed it) is that you need to try parsing binders before parsing identifiers. Currently, all binders are happily accepted as identifiers. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Overlaping Parsec rules
Am 07.03.2011 14:20, schrieb Stephen Tetley: Hi Klaus Unless you have very specific white-space considerations, you really want to be using Parsec's Token and LanguageDef modules. Well, I would want to use Parsec's Token and LanguageDef modules just to get the white spaces (and comments to be ignored) right. Simply skip spaces and comments after every lexical token (by your own wrapper function like Text.ParserCombinators.Parsec.Token.lexeme). C. There are examples in the legacy Parsec 2.0 distribution available from Daan Leijen's old website: http://legacy.cs.uu.nl/daan/parsec.html Best wishes Stephen ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Haskell IDE
2011/3/3 JP Moresmau jpmores...@gmail.com: Hello, I'm one of the maintainers of EclipseFP. It is a real alternative: it works, it is maintained, supported and enhanced. I use it for my own projects, and of course I use it to work on the version of the scion library that ships with it, so we eat our own dogfood :-). A new minor version is going to come out in the next couple of weeks. Why don't you give it a try? We appreciate any feedback! from http://eclipsefp.sourceforge.net/eclipsefp2.html I understand that building scion is automatically done, but can be optionally done it it's too long. but from http://eclipsefp.sourceforge.net/build.html I see that I do have to build it. But : git clone http://github.com/JPMoresmau/scion.git Initialized empty Git repository in /data/code/scion/.git/ remote: Counting objects: 3563, done. remote: Compressing objects: 100% (1310/1310), done. remote: Total 3563 (delta 2137), reused 3342 (delta 1970) Receiving objects: 100% (3563/3563), 580.09 KiB | 395 KiB/s, done. Resolving deltas: 100% (2137/2137), done. david@pcdavid:~/code$ cd scion/ david@pcdavid:~/code/scion$ sudo cabal install [sudo] password for david: Resolving dependencies... cabal: dependencies conflict: ghc-6.12.3 requires Cabal ==1.8.0.6 however Cabal-1.8.0.6 was excluded because scion-0.1.0.8 requires Cabal ==1.6.* I'm not sure what to do now. Trying to install eclipsefp without scion built in case it works, but I'm not used to eclipse, I could use some screenshots to take me step by step through the process. For example, My project menu has open grayed, and file/import wants me to choose an import source (I chose Existing project into workspace) but then do I copy projects into workspace ( I did not ) I have eclipse already setup with android development (I have just a toy project, I planned to learn about android programming), when I imported the projects they appeared at the same place. Is there a way not to mix android and haskell projects ? Do I create a new workspace, or assign working sets ? Anyway the projects are in the workspace, and there's 123 warnings; but I can't really tell if the build failed or not. Selecting all the project, and right-click - refresh didn't seem to change anything. But it seems it worked, because step 7 does launch a new eclipse window, and I can see haskell in the preferences. Except that I had an error message regarding scion, and stupid me, clicked ok before I could remember what was written. I tried rebuild scion, but nothing seemed to happen. I closed the first Eclipse window (the one that has all the eclipsefp projects in the package explorer), but then the second one closed too. I lauched eclipse again, run configurations, etc and this time it looks like scion is building (well something happens in the Console tab) While this is building, I'm wondering if there's a way to launch eclise directly in the right mode. Not having to go to run configurations, etc. Ah, the build failed. Complains that HUnit is missing. So I cabal install it, close my Eclipse window, run configurations again on the first Eclipse window, and the build resumes. And build success. Now to open a haskell source file and play with Eclipse. /feedback David. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Haskell IDE
2011/3/7 David Virebayre dav.vire+hask...@gmail.com And build success. Now to open a haskell source file and play with Eclipse. ... and I keep having those Problem occurred popups: 'Occurrences has encountered a problem An internal error has occurred. In the detail: An internal error has occurred. java.lang.NullPointerException In show error log tab, 3 messages; all have 'net.sf.eclipsefp.haskell.ui' in the Plug-in column : TWICE: eclipse.buildId=M20100211-1343 java.version=1.6.0_20 java.vendor=Sun Microsystems Inc. BootLoader constants: OS=linux, ARCH=x86, WS=gtk, NL=fr_FR Framework arguments: -product org.eclipse.sdk.ide Command-line arguments: -product org.eclipse.sdk.ide -data /data/code/eclipse/../runtime-New_configuration -dev file:/data/code/eclipse/.metadata/.plugins/org.eclipse.pde.core/New_configuration/dev.properties -os linux -ws gtk -arch x86 Info Mon Mar 07 15:15:13 CET 2011 cabal executable: /usr/local/bin/cabal, cabal-install 0.8.2, Cabal library version 1.8.0.2 AND ONCE: eclipse.buildId=M20100211-1343 java.version=1.6.0_20 java.vendor=Sun Microsystems Inc. BootLoader constants: OS=linux, ARCH=x86, WS=gtk, NL=fr_FR Framework arguments: -product org.eclipse.sdk.ide Command-line arguments: -product org.eclipse.sdk.ide -data /data/code/eclipse/../runtime-New_configuration -dev file:/data/code/eclipse/.metadata/.plugins/org.eclipse.pde.core/New_configuration/dev.properties -os linux -ws gtk -arch x86 Info Mon Mar 07 15:03:33 CET 2011 cabal executable: /home/david/bin/cabalinit, cabal-install , Cabal library version Cheers, David. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Haskell IDE
I use leksah, and I want to say thanks to the people that maintain it. I want also to encourage them to continuing its valuable work. from the last release of Leksah, I particulary appreciate its: Multiplatform support Ease of installation in all platforms including windows capability to works with many cabal packages in a single workspace its detection of dependencies and rebuilding of the affected packages Integration of GHCI and the GHCI debugger integrated building of documentation referencing and navigation trough the source of the installed packages and of course all the other traditional IDE capabilities 2011/3/7 David Virebayre dav.vire+hask...@gmail.com 2011/3/3 JP Moresmau jpmores...@gmail.com: Hello, I'm one of the maintainers of EclipseFP. It is a real alternative: it works, it is maintained, supported and enhanced. I use it for my own projects, and of course I use it to work on the version of the scion library that ships with it, so we eat our own dogfood :-). A new minor version is going to come out in the next couple of weeks. Why don't you give it a try? We appreciate any feedback! from http://eclipsefp.sourceforge.net/eclipsefp2.html I understand that building scion is automatically done, but can be optionally done it it's too long. but from http://eclipsefp.sourceforge.net/build.html I see that I do have to build it. But : git clone http://github.com/JPMoresmau/scion.git Initialized empty Git repository in /data/code/scion/.git/ remote: Counting objects: 3563, done. remote: Compressing objects: 100% (1310/1310), done. remote: Total 3563 (delta 2137), reused 3342 (delta 1970) Receiving objects: 100% (3563/3563), 580.09 KiB | 395 KiB/s, done. Resolving deltas: 100% (2137/2137), done. david@pcdavid:~/code$ cd scion/ david@pcdavid:~/code/scion$ sudo cabal install [sudo] password for david: Resolving dependencies... cabal: dependencies conflict: ghc-6.12.3 requires Cabal ==1.8.0.6 however Cabal-1.8.0.6 was excluded because scion-0.1.0.8 requires Cabal ==1.6.* I'm not sure what to do now. Trying to install eclipsefp without scion built in case it works, but I'm not used to eclipse, I could use some screenshots to take me step by step through the process. For example, My project menu has open grayed, and file/import wants me to choose an import source (I chose Existing project into workspace) but then do I copy projects into workspace ( I did not ) I have eclipse already setup with android development (I have just a toy project, I planned to learn about android programming), when I imported the projects they appeared at the same place. Is there a way not to mix android and haskell projects ? Do I create a new workspace, or assign working sets ? Anyway the projects are in the workspace, and there's 123 warnings; but I can't really tell if the build failed or not. Selecting all the project, and right-click - refresh didn't seem to change anything. But it seems it worked, because step 7 does launch a new eclipse window, and I can see haskell in the preferences. Except that I had an error message regarding scion, and stupid me, clicked ok before I could remember what was written. I tried rebuild scion, but nothing seemed to happen. I closed the first Eclipse window (the one that has all the eclipsefp projects in the package explorer), but then the second one closed too. I lauched eclipse again, run configurations, etc and this time it looks like scion is building (well something happens in the Console tab) While this is building, I'm wondering if there's a way to launch eclise directly in the right mode. Not having to go to run configurations, etc. Ah, the build failed. Complains that HUnit is missing. So I cabal install it, close my Eclipse window, run configurations again on the first Eclipse window, and the build resumes. And build success. Now to open a haskell source file and play with Eclipse. /feedback David. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Haskell IDE
The eclipseFP Help forum or mailing list is probably a better medium for EclipseFP questions than haskell-cafe, I'd say... EclipseFP uses the scion library, but comes bundled with it, so it should be able to build it (including downloading dependencies from hackage) itself without you building scion manually (if you still want to do it, there is a flag you can set to specify the version of Cabal you're using). I think you're confusing the instructions to install EclipseFP and use it to develop in Haskell with the instructions on the build page that are for people that wish to hack EclipseFP itself. Basically you need to install EclipseFP using the Eclipse install new software features and take it from there. You can of course have different workspaces, one for your haskell projects, one for your android projects, but it doesn't matter if you have projects of different types in the same workspace. Good luck, and if you're still confused feel free to contact me directly. JP On Mon, Mar 7, 2011 at 3:19 PM, David Virebayre dav.vire+hask...@gmail.com wrote: 2011/3/3 JP Moresmau jpmores...@gmail.com: Hello, I'm one of the maintainers of EclipseFP. It is a real alternative: it works, it is maintained, supported and enhanced. I use it for my own projects, and of course I use it to work on the version of the scion library that ships with it, so we eat our own dogfood :-). A new minor version is going to come out in the next couple of weeks. Why don't you give it a try? We appreciate any feedback! from http://eclipsefp.sourceforge.net/eclipsefp2.html I understand that building scion is automatically done, but can be optionally done it it's too long. but from http://eclipsefp.sourceforge.net/build.html I see that I do have to build it. But : git clone http://github.com/JPMoresmau/scion.git Initialized empty Git repository in /data/code/scion/.git/ remote: Counting objects: 3563, done. remote: Compressing objects: 100% (1310/1310), done. remote: Total 3563 (delta 2137), reused 3342 (delta 1970) Receiving objects: 100% (3563/3563), 580.09 KiB | 395 KiB/s, done. Resolving deltas: 100% (2137/2137), done. david@pcdavid:~/code$ cd scion/ david@pcdavid:~/code/scion$ sudo cabal install [sudo] password for david: Resolving dependencies... cabal: dependencies conflict: ghc-6.12.3 requires Cabal ==1.8.0.6 however Cabal-1.8.0.6 was excluded because scion-0.1.0.8 requires Cabal ==1.6.* I'm not sure what to do now. Trying to install eclipsefp without scion built in case it works, but I'm not used to eclipse, I could use some screenshots to take me step by step through the process. For example, My project menu has open grayed, and file/import wants me to choose an import source (I chose Existing project into workspace) but then do I copy projects into workspace ( I did not ) I have eclipse already setup with android development (I have just a toy project, I planned to learn about android programming), when I imported the projects they appeared at the same place. Is there a way not to mix android and haskell projects ? Do I create a new workspace, or assign working sets ? Anyway the projects are in the workspace, and there's 123 warnings; but I can't really tell if the build failed or not. Selecting all the project, and right-click - refresh didn't seem to change anything. But it seems it worked, because step 7 does launch a new eclipse window, and I can see haskell in the preferences. Except that I had an error message regarding scion, and stupid me, clicked ok before I could remember what was written. I tried rebuild scion, but nothing seemed to happen. I closed the first Eclipse window (the one that has all the eclipsefp projects in the package explorer), but then the second one closed too. I lauched eclipse again, run configurations, etc and this time it looks like scion is building (well something happens in the Console tab) While this is building, I'm wondering if there's a way to launch eclise directly in the right mode. Not having to go to run configurations, etc. Ah, the build failed. Complains that HUnit is missing. So I cabal install it, close my Eclipse window, run configurations again on the first Eclipse window, and the build resumes. And build success. Now to open a haskell source file and play with Eclipse. /feedback David. -- JP Moresmau http://jpmoresmau.blogspot.com/ ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Haskell IDE
On Monday 07 March 2011 16:08:01, Alberto G. Corona wrote: I use leksah, and I want to say thanks to the people that maintain it. I want also to encourage them to continuing its valuable work. from the last release of Leksah, I particulary appreciate its: Multiplatform support Ease of installation in all platforms including windows capability to works with many cabal packages in a single workspace its detection of dependencies and rebuilding of the affected packages Integration of GHCI and the GHCI debugger integrated building of documentation referencing and navigation trough the source of the installed packages and of course all the other traditional IDE capabilities That sounds nice, so I thought I'd try out leksah again. Unfortunately, the dependencies rule out GHC-7 (base 4.3, Cabal 1.9, containers 0.4 were the ones that sprang to the eyes immediately). Bummer Maybe someone could try relaxing the bounds and build it with GHC-7, and - if it works - upload a new version? (I could try if I get a go-ahead from Hamish or Jürgen) ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] [Haskell] ANNOUNCE: silently-0.0.1 (prevent IO actions from writing to stdout)
Thanks for the feedback! I'll definitely update the documentation. I agree that temp files are a weird way to go about it but they're the least surprising and dangerous way I could think of handling it. Unix has /dev/null and Windows has a NUL device that works the same way so I could use those. * I could see what OS it is and try those options and use a temp file only if it's an unexpected os. Unfortunately System.Info.os reports windows (and some other os's) as mingw32, I think it's because of how GHC was compiled. Who knows what the real os is or what each os will be reported as in the future or how they appear to other compilers and runtimes. * I could write to /dev/null or NUL if they exist and use a temp file if neither does, but if there's a legitimate /dev/null or NUL file then having some unrelated program write to them would be very very bad and a major pain for whoever has to figure that one out. I'd rather not use temp files but both of these seem problematic. Does anyone have a better idea or think that either of those are acceptable? Trystan Spangler - Original Message - From: Felipe Almeida Lessa felipe.le...@gmail.com To: Trystan Spangler trysta...@comcast.net Cc: haskell-cafe@haskell.org Sent: Monday, March 7, 2011 3:41:07 AM Subject: Re: [Haskell] ANNOUNCE: silently-0.0.1 (prevent IO actions from writing to stdout) Hello! I wouldn't expect that a function that ignores the output directs it to a temporary file. I think that should be clearly stated on the documentation, as this may be undesirable. I don't know if there is a better way of solving this problem, though. (Note, however, that on Unix systems you could probably direct the output to /dev/null.) Cheers, -- Felipe. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] silently-0.0.1 (prevent IO actions from writing to stdout)
On Mon, 7 Mar 2011, trysta...@comcast.net wrote: * I could see what OS it is and try those options and use a temp file only if it's an unexpected os. Unfortunately System.Info.os reports windows (and some other os's) as mingw32, I think it's because of how GHC was compiled. Who knows what the real os is or what each os will be reported as in the future or how they appear to other compilers and runtimes. You could add a check to Cabal and then choose a platform-dependent module using If os(windows) Hs-Source-Dirs: windows/src Else Hs-Source-Dirs: unix/src ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] silently-0.0.1 (prevent IO actions from writing to stdout)
I was not aware of that! Very cool. Trystan Spangler - Original Message - From: Henning Thielemann lemm...@henning-thielemann.de To: trystan s trysta...@comcast.net Cc: Felipe Almeida Lessa felipe.le...@gmail.com, haskell-cafe@haskell.org Sent: Monday, March 7, 2011 9:56:16 AM Subject: Re: silently-0.0.1 (prevent IO actions from writing to stdout) On Mon, 7 Mar 2011, trysta...@comcast.net wrote: * I could see what OS it is and try those options and use a temp file only if it's an unexpected os. Unfortunately System.Info.os reports windows (and some other os's) as mingw32, I think it's because of how GHC was compiled. Who knows what the real os is or what each os will be reported as in the future or how they appear to other compilers and runtimes. You could add a check to Cabal and then choose a platform-dependent module using If os(windows) Hs-Source-Dirs: windows/src Else Hs-Source-Dirs: unix/src ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] ANNOUNCE: cinvoke 0.1 released
On Mon, Mar 07, 2011 at 09:41:27AM +, Max Bolingbroke wrote: Hi Remi, On 6 March 2011 13:38, Remi Turk rt...@science.uva.nl wrote: I am happy to finally announce cinvoke 0.1, a binding to the C library cinvoke[1], allowing functions to be loaded and called whose names and types are not known before run-time. As the author of the libffi package (http://hackage.haskell.org/package/libffi-0.1) which does a similar thing, could you say when it would be appropriate to use one or the other package? Cheers, Max Of course: - libffi doesn't do library/function loading; you'll need to use System.Posix.DynamicLinker or System.Win32.DLL for that. cinvoke will not only load your libraries and functions, but even collect the garbage afterwards. - Things seem to have changed, but back when I first looked at cinvoke, getting libffi to run under windows didn't seem too realistic. - If you need to pass C structs (by value), you'll have to use libffi: cinvoke doesn't support them at all. - The current version of libffi is not exception safe (I do have some code lying around here though...) - cinvoke is actually haddockized (although hackage still hasn't generated the docs, apparently). Groeten, Remi ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] ANNOUNCE: cinvoke 0.1 released
On Mon, Mar 7, 2011 at 7:32 PM, Remi Turk rt...@science.uva.nl wrote: On Mon, Mar 07, 2011 at 09:41:27AM +, Max Bolingbroke wrote: Hi Remi, On 6 March 2011 13:38, Remi Turk rt...@science.uva.nl wrote: I am happy to finally announce cinvoke 0.1, a binding to the C library cinvoke[1], allowing functions to be loaded and called whose names and types are not known before run-time. As the author of the libffi package (http://hackage.haskell.org/package/libffi-0.1) which does a similar thing, could you say when it would be appropriate to use one or the other package? Cheers, Max Of course: - libffi doesn't do library/function loading; you'll need to use System.Posix.DynamicLinker or System.Win32.DLL for that. cinvoke will not only load your libraries and functions, but even collect the garbage afterwards. - Things seem to have changed, but back when I first looked at cinvoke, getting libffi to run under windows didn't seem too realistic. - If you need to pass C structs (by value), you'll have to use libffi: cinvoke doesn't support them at all. - The current version of libffi is not exception safe (I do have some code lying around here though...) - cinvoke is actually haddockized (although hackage still hasn't generated the docs, apparently). It's reporting a build failure. Groeten, Remi ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe -- Work is punishment for failing to procrastinate effectively. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] ANNOUNCE: cinvoke 0.1 released
On Monday 07 March 2011 21:42:16, Gábor Lehel wrote: It's reporting a build failure. Missing C library. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] ANNOUNCE: cinvoke 0.1 released
On Mon, Mar 07, 2011 at 10:00:47PM +0100, Daniel Fischer wrote: On Monday 07 March 2011 21:42:16, Gábor Lehel wrote: It's reporting a build failure. Missing C library. cinvoke (the C library) is obviously not installed on the testing machine. Does that really mean no library with uncommon C dependencies gets documentation on hackage? Remi ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] ANNOUNCE: cinvoke 0.1 released
On Monday 07 March 2011 22:14:38, Remi Turk wrote: On Mon, Mar 07, 2011 at 10:00:47PM +0100, Daniel Fischer wrote: On Monday 07 March 2011 21:42:16, Gábor Lehel wrote: It's reporting a build failure. Missing C library. cinvoke (the C library) is obviously not installed on the testing machine. Does that really mean no library with uncommon C dependencies gets documentation on hackage? Remi Basically, yes. As far as I know, documentation is only built for libraries that build on hackage. Maybe it would be a good idea to have the opportunity to upload haddock bundles to hackage too for such libraries. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] ANNOUNCE: cinvoke 0.1 released
On Mon, Mar 07, 2011 at 10:31:25PM +0100, Daniel Fischer wrote: On Monday 07 March 2011 22:14:38, Remi Turk wrote: cinvoke (the C library) is obviously not installed on the testing machine. Does that really mean no library with uncommon C dependencies gets documentation on hackage? Remi Basically, yes. As far as I know, documentation is only built for libraries that build on hackage. Maybe it would be a good idea to have the opportunity to upload haddock bundles to hackage too for such libraries. That sucks :( Uploading haddock bundles could solve the problem, though I don't currently understand why being able to successfully configure a package is a prerequisite to generating the docs. Anyway, I just put the docs online somewhere else with a link from the homepage: http://haskell.org/haskellwiki/Library/cinvoke ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Learn You a Haskell for Great Good - a few doubts
On Sat, Mar 5, 2011 at 5:06 AM, wren ng thornton w...@freegeek.org wrote: On 3/4/11 4:33 PM, Alexander Solla wrote: On Thu, Mar 3, 2011 at 10:14 PM, wren ng thorntonw...@freegeek.org wrote: On 3/3/11 2:58 AM, Antti-Juhani Kaijanaho wrote: On Thu, Mar 03, 2011 at 12:29:44PM +0530, Karthick Gururaj wrote: Thanks - is this the same unit that accompanies IO in IO () ? In any case, my question is answered since it is not a tuple. It can be viewed as the trivial 0-tuple. Except that this is problematic since Haskell doesn't have 1-tuples (which would be distinct from plain values in that they have an extra bottom). I don't get this line of thought. I understand what you're saying, but why even bother trying to distinguish between bottoms when they can't be compared by equality, or even computed? If we have, data OneTuple a = One a Then _|_ /= One _|_ That is vacuously true. I will demonstrate the source of the contradiction later. But you also have _|_ == One _|_, by evaluation: Just undefined Just *** Exception: Prelude.undefined This can be detected by seq: the left-hand side doesn't terminate, whereas the right-hand side does. And moreover, this can mess up other things (e.g., monads) by introducing too much laziness. Space leaks are quite a serious matter and they have nothing to do with trying to compare uncomputable values. Do you want a seemingly insignificant refactoring to cause your program to suddenly hang forever? Or to take up gobs more space than it used to? 'seq' is not a function, since it breaks referential transparency and possibly extensionality in function composition. By construction, seq a b = b, and yet seq undefined b /= b. Admittedly, the Haskell report and the GHC implementation, diverge on this issue. Haskell98 specifically defines seq in terms of a comparison with bottom, whereas GHC merely reduces the first argument to WHNF. In any case, the reduction is a side effect, with which can lead to inconsistent semantics if 'seq' is included in the language. It is nice to know that we can work in a consistent language if we avoid certain constructs, such as 'seq', 'unsafePerformIO', and probably others. In addition to making the core language conceptually simpler, it means that we can be sure we aren't inadvertently destroying the correctness guarantees introduced by the Howard-Curry correspondence theorem. Nope, it contains one. Just ask any proof theorist, or anyone who uses witnesses to capture information in the type system. I have studied enough proof theory, model theory, and lattice theory to know that there is room for both positions. Just because you /can/ lift a lattice into one with bottom, it doesn't mean you /should/, if it means losing conceptual clarity. In particular, I don't see why you want to generate an algebra of special cases that add no expressiveness, and include them in the language, when you can use a quotient construction to remove them from the language. As a practical matter, 'seq' is necessary. But it should be treated as a special case (like unsafePerformIO), because it IS one. Also, there is no need for a set to contain an element for it to be named or quantified over. The empty set can be a witness just as well as a singleton set. EmptyDataDecls works just as well whether you interpret (undefined :: Blah) to mean a Blah that is not defined or a Blah that is the value 'undefined'. Indeed, the latter is paradoxical. 'undefined = undefined' type checks, but it is not well-founded -- which is exactly why it is not defined! 'undefined' is not a value. It is the name for a thing which cannot be evaluated. We happen to know that there are a lot of things which cannot be evaluated, but the quotient construction treats them all the same. (There is the practical issue of GHC handling Prelude.undefined differently than the other bottoms. But I won't complain about that ;0) If you choose to interpret all bottoms as being the same non-existent, unquantifiable (in the language of Haskell) proto-value, you get the isomorphism between types a and (a), as types. Nope, because we have notBottom :: OneTuple a - Bool notBottom x = x `seq` True whereas isBottom :: a - Bool isBottom x = x `seq` True seq is not a function, and is removed from consideration by the quotient construction I have mentioned, specifically because of how it behaves on bottom and the fact that the Haskell98 report defines it in terms of comparing values to bottom, an operation which is known to be impossible. notBottom and isBottom are also not functions, for the same reason. Indeed, those are the semantics in use by the language. A value written (a) is interpreted as a. A type written (a) is interpreted as a. That's a syntactic matter. Those are parentheses of grouping, not parentheses of tuple construction. For example, you can say: (,) a b or (,,) a b c
Re: [Haskell-cafe] Learn You a Haskell for Great Good - a few doubts
On Mon, Mar 7, 2011 at 3:38 PM, Alexander Solla alex.so...@gmail.comwrote: This can be detected by seq: the left-hand side doesn't terminate, whereas the right-hand side does. And moreover, this can mess up other things (e.g., monads) by introducing too much laziness. Space leaks are quite a serious matter and they have nothing to do with trying to compare uncomputable values. Do you want a seemingly insignificant refactoring to cause your program to suddenly hang forever? Or to take up gobs more space than it used to? 'seq' is not a function, since it breaks referential transparency and possibly extensionality in function composition. By construction, seq a b = b, and yet seq undefined b /= b. Admittedly, the Haskell report and the GHC implementation, diverge on this issue. Haskell98 specifically defines seq in terms of a comparison with bottom, whereas GHC merely reduces the first argument to WHNF. In any case, the reduction is a side effect, with which can lead to inconsistent semantics if 'seq' is included in the language. It is nice to know that we can work in a consistent language if we avoid certain constructs, such as 'seq', 'unsafePerformIO', and probably others. In addition to making the core language conceptually simpler, it means that we can be sure we aren't inadvertently destroying the correctness guarantees introduced by the Howard-Curry correspondence theorem. As a matter of fact, if you read GHC.Prim, you will see that seq is a bottom! seq :: a - b - b seq = let x = x in x The magic semantics of evaluating the first argument are done by the compiler/runtime, and are apparently not expressible in Haskell. This is true of inline lazy unsafeCoerce and dozens of others, all of which are expressed as specialized types with the same equation: let x = x in x ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Learn You a Haskell for Great Good - a few doubts
On Tuesday 08 March 2011 00:58:36, Alexander Solla wrote: As a matter of fact, if you read GHC.Prim, you will see that seq is a bottom! No, you don't. GHC.Prim is a dummy module whose only purpose is to let haddock generate documentation. Every function there has the code let x = x in x, e.g. plusWord# :: Word# - Word# - Word# plusWord# = let x = x in x minusWord# :: Word# - Word# - Word# minusWord# = let x = x in x undefined is not yet available, otherwise probably everything in GHC.Prim would be pseudo-defined as undefined for haddock. seq :: a - b - b seq = let x = x in x The magic semantics of evaluating the first argument are done by the compiler/runtime, and are apparently not expressible in Haskell. Right. But neither is addition of Word# etc., for the primitives, you have to do something special. This is true of inline lazy unsafeCoerce and dozens of others, all of which are expressed as specialized types with the same equation: let x = x in x ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] AusHack 2011
It is time once again to start planning for AusHac (the Hackathon Down Under)! The current proposed dates are 8-10 July. This time we're trying to organise a venue in the Sydney CBD, but if all else fails we'll be at UNSW again. So, is anyone vehemently opposed to having AusHac at this time or place? -- Ivan Lazar Miljenovic ivan.miljeno...@gmail.com IvanMiljenovic.wordpress.com ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Learn You a Haskell for Great Good - a few doubts
On 3/7/11 6:38 PM, Alexander Solla wrote: 'seq' is not a function, since it breaks referential transparency and possibly extensionality in function composition. By construction, seq a b = b, and yet seq undefined b /= b. Admittedly, the Haskell report and the GHC implementation, diverge on this issue. Haskell98 specifically defines seq in terms of a comparison with bottom, whereas GHC merely reduces the first argument to WHNF. In any case, the reduction is a side effect, with which can lead to inconsistent semantics if 'seq' is included in the language. It is nice to know that we can work in a consistent language if we avoid certain constructs, such as 'seq', 'unsafePerformIO', and probably others. In addition to making the core language conceptually simpler, it means that we can be sure we aren't inadvertently destroying the correctness guarantees introduced by the Howard-Curry correspondence theorem. You are free to reason in whichever language you so desire. But that does not mean the semantics of the language you desire are the same as the semantics of Haskell. Fact of the matter is that Haskell has 'seq' and bottom, even if you choose to call them non-functions or non-values. It is not the case that for every pair, ab, we have that: ab == (fst ab, snd ab) Why? Well consider ab = undefined: _|_ /= (_|_,_|_) (undefined, undefined) (*** Exception: Prelude.undefined That is as close to Haskell-equality as you can get for a proto-value that does not have an Eq instance. As a consequence of referential transparency, evaluation induces an equivalence relation. This implies that (_|_, _|_) = _|_ = (_|_, _|_). I value referential transparency, and so reject constructs which violate it. Please demonstrate a proof that _|_ /= (_|_, _|_), so that I can exclude the unsound constructs you will undoubtedly have to use from my interpretation of the language. I am more interested in finding the consistent fragment of what you call Haskell than defending what I call Haskell. The trivial proof is: seq _|_ () == _|_ /= seq (_|_,_|_) () == () But you refuse to believe that 'seq' exists, so here's another proof: case _|_ of (_,_) - () == _|_ /= case (_|_,_|_) of (_,_) - () == () Do you refuse to believe that case analysis exists too? -- Live well, ~wren ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Learn You a Haskell for Great Good - a few doubts
On Tuesday 08 March 2011 00:38:53, Alexander Solla wrote: On Sat, Mar 5, 2011 at 5:06 AM, wren ng thornton w...@freegeek.org wrote: If we have, data OneTuple a = One a Then _|_ /= One _|_ That is vacuously true. I will demonstrate the source of the contradiction later. But you also have _|_ == One _|_, by evaluation: Just undefined Just *** Exception: Prelude.undefined But that shows that _|_ and Just _|_ aren't the same (in Haskell), doesn't it? case x of Just _ - Just something Nothing - Nothing produces Just something for (Just _|_), but not for _|_. This can be detected by seq: the left-hand side doesn't terminate, whereas the right-hand side does. And moreover, this can mess up other things (e.g., monads) by introducing too much laziness. Space leaks are quite a serious matter and they have nothing to do with trying to compare uncomputable values. Do you want a seemingly insignificant refactoring to cause your program to suddenly hang forever? Or to take up gobs more space than it used to? 'seq' is not a function, since it breaks referential transparency Does it, if one assumes that 'seq a b' is *not* the same as 'b' ? and possibly extensionality in function composition. By construction, seq a b = b, and yet seq undefined b /= b. Admittedly, the Haskell report and the GHC implementation, diverge on this issue. Haskell98 specifically defines seq in terms of a comparison with bottom, whereas GHC merely reduces the first argument to WHNF. But reducing to WHNF is precisely what is needed to detect bottom. If a value is a constructor application or a lambda, it's not bottom. In any case, the reduction is a side effect, with which can lead to inconsistent semantics if 'seq' is included in the language. But seq is in the language, as specified by the report. You can argue that it shouldn't and campaign for its removal, but it's in now and speaking about Haskell, one can only sometimes ignore it. It is not the case that for every pair, ab, we have that: ab == (fst ab, snd ab) Why? Well consider ab = undefined: _|_ /= (_|_,_|_) (undefined, undefined) (*** Exception: Prelude.undefined That is as close to Haskell-equality as you can get for a proto-value that does not have an Eq instance. As a consequence of referential transparency, evaluation induces an equivalence relation. This implies that (_|_, _|_) = _|_ = (_|_, _|_). But case x of (_, _) - Okay distinguishes _|_ and (_|_, _|_). In Haskell98 and Haskell2010, they are not the same. I value referential transparency, and so reject constructs which violate it. Please demonstrate a proof that _|_ /= (_|_, _|_), so that I can exclude the unsound constructs you will undoubtedly have to use from Pattern matching my interpretation of the language. I am more interested in finding the consistent fragment of what you call Haskell than defending what I call Haskell. That can be a source of much confusion. Usually, 'Haskell' is understood as the language defined in the report. There's some room for interpretation, but not too much. If you call soemthing too different 'Haskell', people won't understand you. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Learn You a Haskell for Great Good - a few doubts
On 3/7/11 6:58 PM, Alexander Solla wrote: The magic semantics of evaluating the first argument are done by the compiler/runtime, and are apparently not expressible in Haskell. Of course this is true. The only ways of forcing evaluation in Haskell are (a) to perform pattern matches on a value, (b) use 'seq'--- either directly or in its disguised forms: strict data constructors and -XBangPatterns. In order to use pattern matching you need to know what the constructors of the type are in order to force a choice between different constructors; therefore you can't use case analysis to define a function with the type and semantics of 'seq'. But 'seq' is still defined in the Haskell report and therefore a part of Haskell. Many have lamented the problems introduced by a parametric polymorphic 'seq'; if it were just type-class polymorphic then it wouldn't be such a problem. But then a type-class polymorphic 'seq' could lead to maintenance issues similar to those faced by Java's checked exceptions, which is why it was rejected from Haskell. -- Live well, ~wren ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] ANNOUNCE: cinvoke 0.1 released
On Mon, Mar 7, 2011 at 6:32 PM, Remi Turk rt...@science.uva.nl wrote: - If you need to pass C structs (by value), you'll have to use libffi: cinvoke doesn't support them at all. What about CInvStructure[1]? I was just glancing at the documentation when I saw this. Cheers! =) [1] http://www.nongnu.org/cinvoke/doc/cinvoke_8h.html#b39daa4325b8b87aa246cd8acf8a7e65 -- Felipe. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Haskell IDE
Please try the version here https://github.com/jutaro Hopefully we will be in a position to release a new version very soon. Hamish On 8/03/2011, at 5:39, Daniel Fischer daniel.is.fisc...@googlemail.com wrote: On Monday 07 March 2011 16:08:01, Alberto G. Corona wrote: I use leksah, and I want to say thanks to the people that maintain it. I want also to encourage them to continuing its valuable work. from the last release of Leksah, I particulary appreciate its: Multiplatform support Ease of installation in all platforms including windows capability to works with many cabal packages in a single workspace its detection of dependencies and rebuilding of the affected packages Integration of GHCI and the GHCI debugger integrated building of documentation referencing and navigation trough the source of the installed packages and of course all the other traditional IDE capabilities That sounds nice, so I thought I'd try out leksah again. Unfortunately, the dependencies rule out GHC-7 (base 4.3, Cabal 1.9, containers 0.4 were the ones that sprang to the eyes immediately). Bummer Maybe someone could try relaxing the bounds and build it with GHC-7, and - if it works - upload a new version? (I could try if I get a go-ahead from Hamish or Jürgen) ___ Leksah mailing list lek...@projects.haskell.org http://projects.haskell.org/cgi-bin/mailman/listinfo/leksah ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] have anyone tried GHC 7.0.2
Hi, After I installed GHC 7.0.2 and cabal-install 0.10.0 using Cabal library 1.10.1, I found that some package failed to install. For example like Happy, I got the following error: cabal install happy Resolving dependencies... command line: cannot satisfy -package Cabal-1.10.1.0: Cabal-1.10.1.0-64e8f05722ea5bbdf07df2e5a3491105 is unusable due to missing or recursive dependencies: directory-1.1.0.0-85d1d0f9d96dffdacf64f3cc6fba6f2f process-1.0.1.5-4cab1bf0666275ac101dd48c7565b64c (use -v for more information) cabal: Error: some packages failed to install: happy-1.18.6 failed during the configure step. The exception was: ExitFailure 1 Besides happy package, some other packages like ghc-mod, hmatrix also failed to install. Does anyone else have the same issue after installed ghc 7.0.2? By the way, does anyone hear anything about Haskell-platform 2011.2 ? Thanks! -- Qi Qi ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] Questions on list administration
Hello, I recently joined Haskell group and wanted to share a few points. First, I came to Haskell cafe from google groups (groups.google.com - search for Haskell). Two groups are listed: fa.haskell and haskell-cafe. I became a member of Haskell Cafe via the google groups interface (the operation was successful according to google). When I posted a message, I got an automated email (within a few seconds, so I think there was no human moderation): You are not allowed to post to this mailing list, and your message has been automatically rejected. If you think that your messages are being rejected in error, contact the mailing list owner at haskell-cafe-ow...@haskell.org. I wrote to the email address mentioned and didn't get any response. Funny thing is, google reports that my post was successful, and my post was visible on google groups even after I signed out from my google account. I next wrote to the group owner, again from the google groups interface (there is an option to send email to the owner). No response here as well. I then went to Haskell.org - Mailing lists and saw there are atleast three lists of immediate relevance, including one for beginners. The website also describes the posting guidelines and other information. Very useful. I would have stuck to asking questions on the beginner's group if I had known earlier.. Can I ask a couple of changes be done to help others like me? a. In google groups - https://groups.google.com/group/haskell-cafe/about - add a link to the haskell.org - Mailing lists for more info. b. Let users join from google groups interface, or if this is not possible, mention that in the about page c. Change the contents of the reject email (reproduced above) to include the following: If you want to subscribe to the Haskell Cafe mailing list, please visit http://www.haskell.org/mailman/listinfo/haskell-cafe; Aside: the Haskell Cafe mailing list page (at http://www.haskell.org/mailman/listinfo/haskell-cafe) has a dead link at the bottom of the page. The Overview of all haskell.org mailing lists points to http://www.haskell.org/cgi-bin/mailman/listinfo (404 error) instead of http://haskell.org/mailman/listinfo. Thanks! Warm regards, Karthick ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] ANN: theoremquest-0.0.0
On Feb 28, 7:59 pm, Tom Hawkins tomahawk...@gmail.com wrote: I have been wanting to gain a better understanding of interactive theorem proving for some time. And I've often wondered: Can theorem proving be made into a user-friendly game that could attract mass appeal? And if so, could a population of gamers collectively solve interesting and relevant mathematical problems? To try to answer these questions -- and to gain some experience myself with theorem proving -- I started a project called TheoremQuest [1]. TheoremQuest intends to be a multi-player game system, where the game server receives requests by clients, such as theorem queries and inference rule applications. The TheoremQuest server validates deductions, compares them with existing theorems in the database, then returns results. TheoremQuest's deductive system borrows that of John Harrison's HOL Light [2]. There are 2 Hackage packages: theoremquest [3] is a library that declares types for terms, inference rules, and client-server transactions, used by both server and clients; and theoremquest-client [4] is an example client (tq). All the code, including the server-side, is hosted at github [5]. Currently the client-server interface is working, but little else. The library has defined most of the core inference rules, but has none of the basic types or axioms. And the tq client is nothing at all like a game; at this point it is just a command line tool to test the server. Currently tq can ping the server, create a new user, apply inference rules, and print out theorems. Here's how tq can prove x |- x: $ tq newuser tomahawkins tomahawk...@gmail.com Ack $ export THEOREMQUEST_USER=tomahawkins $ tq infer 'ASSUME (Var (Variable x Bool))' Id 1 $ tq theorem 1 assumptions: Var (Variable x Bool) conclusion: Var (Variable x Bool) I'd love to have help with this project, if anyone's interested. I am curious -- how easy is it to use theoremquest for playing with equational theories? In particular equational logic -- see https://www.ideals.illinois.edu/bitstream/handle/2142/11411/A%20Rewriting%20Decision%20Procedure%20for%20Dijkstra-Scholten%27s%20Syllogistic%20Logic%20with%20Complements.pdf?sequence=2 ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
