- Glasgow-haskell-users

RE: Preferred way of submitting patches
by Simon Peyton-Jones 21 Dec '05

21 Dec '05

Einar Simon M's on holiday for Xmas; but he's the man to answer this when he gets back. Simon | -----Original Message----- | From: glasgow-haskell-users-bounces(a)haskell.org [mailto:glasgow-haskell-users- | bounces(a)haskell.org] On Behalf Of Einar Karttunen | Sent: 17 December 2005 16:21 | To: glasgow-haskell-users(a)haskell.org | Subject: Preferred way of submitting patches | | Hello | | What is the preferred way of submitting patches at the moment? | I tried using Trac but it seems that users without an account | on haskell.org cannot attach files... Is using Trac and | linking to the patch hosted externally the way to go? | | - Einar Karttunen | _______________________________________________ | Glasgow-haskell-users mailing list | Glasgow-haskell-users(a)haskell.org | http://www.haskell.org/mailman/listinfo/glasgow-haskell-users

1 0

RE: instance inference
by Simon Peyton-Jones 19 Dec '05

19 Dec '05

Martin Sulzmann is the world expert on this stuff. Maybe he'll know. Simon | -----Original Message----- | From: glasgow-haskell-users-bounces(a)haskell.org [mailto:glasgow-haskell-users- | bounces(a)haskell.org] On Behalf Of Ross Paterson | Sent: 19 December 2005 10:16 | To: Simon Peyton-Jones | Cc: glasgow-haskell-users(a)haskell.org | Subject: Re: instance inference | | As a termination test, how about no restrictions on context and head | except: | | Each assertion in the context must satisfy | * the variables of the assertion are a sub-multiset of those of the | head (though they may be the same), and | * the assertion has fewer type constructors and variables (taken | together and counting repetitions) than the head. | | This ensures that under any ground substitution the context assertion has | fewer type constructors than the head. | | It would accept all instances accepted by the current test, plus | | instance C a | instance C (s a) => C (Sized s a) | instance (C1 a, C2 b) => C a b | instance C1 Int a => C2 Bool [a] | instance C1 Int a => C2 [a] b | instance C a a => C [a] [a] | | But not things like | | instance (C1 a, C2 a, C3 a) => C a | | because that would need to check other instances as well. | | _______________________________________________ | Glasgow-haskell-users mailing list | Glasgow-haskell-users(a)haskell.org | http://www.haskell.org/mailman/listinfo/glasgow-haskell-users

2 1

RE: [STM] Better name for waitUntil?
by Simon Peyton-Jones 19 Dec '05

19 Dec '05

waitUntil sounds like a plausible name. Better than itsAfter or timeAfter. But perhaps sleepUntil would be better, because it's a variant of sleep? Simon | -----Original Message----- | From: glasgow-haskell-users-bounces(a)haskell.org [mailto:glasgow-haskell-users- | bounces(a)haskell.org] On Behalf Of Tomasz Zielonka | Sent: 19 December 2005 09:28 | To: GHC-users | Subject: [STM] Better name for waitUntil? | | Hello! | | I am finishing my waitUntil implementation for STM. It's a new | STM action that will allow to wait for a given time point. | | waitUntil :: UTCTime -> STM () | | (UTCTime is from Ashley Yakeley's Time library.) | | The intended semantics is that | | waitUntil t | | will succeed as soon as time passes t, or more precisely when (now >= t). | | Right now I am to find a better name for waitUntil. Maybe it should be | more declarative, like "itsAfter" or "timeAfter"? I am not a native english | speaker, so I am probably missing many options. Any ideas? | | I also think about adding something like "itsBefore", which would mean | (now < t). | | How about "it'sAfter" ? | Maybe only "after", without "it's" ? | | Thanks in advance for your help. | | PS. I imagine that it would be possible to implement waitUntil more | efficiently in GHC's RTS, but my implementation can be a nice | showcase for STM's expressivity, and it's still quite efficient | (I used some (quite elegant :) hacks to implement subsecond | precision waitUntil efficiently). | | Best regards | Tomasz | | -- | I am searching for a programmer who is good at least in some of | [Haskell, ML, C++, Linux, FreeBSD, math] for work in Warsaw, Poland | _______________________________________________ | Glasgow-haskell-users mailing list | Glasgow-haskell-users(a)haskell.org | http://www.haskell.org/mailman/listinfo/glasgow-haskell-users

2 1

[STM] Better name for waitUntil?
by Tomasz Zielonka 19 Dec '05

19 Dec '05

Hello! I am finishing my waitUntil implementation for STM. It's a new STM action that will allow to wait for a given time point. waitUntil :: UTCTime -> STM () (UTCTime is from Ashley Yakeley's Time library.) The intended semantics is that waitUntil t will succeed as soon as time passes t, or more precisely when (now >= t). Right now I am to find a better name for waitUntil. Maybe it should be more declarative, like "itsAfter" or "timeAfter"? I am not a native english speaker, so I am probably missing many options. Any ideas? I also think about adding something like "itsBefore", which would mean (now < t). How about "it'sAfter" ? Maybe only "after", without "it's" ? Thanks in advance for your help. PS. I imagine that it would be possible to implement waitUntil more efficiently in GHC's RTS, but my implementation can be a nice showcase for STM's expressivity, and it's still quite efficient (I used some (quite elegant :) hacks to implement subsecond precision waitUntil efficiently). Best regards Tomasz -- I am searching for a programmer who is good at least in some of [Haskell, ML, C++, Linux, FreeBSD, math] for work in Warsaw, Poland

1 0

Preferred way of submitting patches
by Einar Karttunen 17 Dec '05

17 Dec '05

Hello What is the preferred way of submitting patches at the moment? I tried using Trac but it seems that users without an account on haskell.org cannot attach files... Is using Trac and linking to the patch hosted externally the way to go? - Einar Karttunen

1 0

RE: Preparing a Registered hc-file-bundle (Was BootstrappingwithHCfiles)
by Simon Marlow 16 Dec '05

16 Dec '05

On 15 December 2005 16:19, Andrew Walrond wrote: > Hi Simon, > > On Wednesday 14 December 2005 09:49, Simon Marlow wrote: >> Actually, looking at ghc/Makefile, it is possible that ghc/lib isn't >> getting built early enough. Try changing: >> >> SUBDIRS = includes rts docs compiler lib utils driver >> >> to >> >> SUBDIRS = includes rts lib docs compiler utils driver >> > > Did that. Now I get to here: > > ------------------------------------------------------------------------ > ===fptools== Recursively making `all' in compat ... > PWD = /home/andrew/test/ghc/bs/target/ghc-6.4.1/ghc/lib > ------------------------------------------------------------------------ > ------------------------------------------------------------------------ > ==fptools== make all - --no-print-directory -r; > in /home/andrew/test/ghc/bs/target/ghc-6.4.1/ghc/lib/compat > ------------------------------------------------------------------------ > make[2]: *** No rule to make target `Compat/Directory.o', needed by > `ghccompat.o'. Stop. > make[1]: *** [all] Error 1 > make: *** [all] Error 1 > make: Leaving directory > `/home/andrew/test/ghc/bs/target/ghc-6.4.1/ghc' > > Any clues how to proceed? I have no idea what has gone wrong here. We ought to be able to make Compat/Directory.o. However, we don't need ghccompat.o, so you could try adding: GhcWithInterpter=NO to ghc/lib/compat/Makefile, right after the NO_INSTALL_LIBRARY setting. If that doesn't work, please run 'make -p -d' in ghc/lib/compat and send me the output. Cheers, Simon

2 1

RE: utf8 strings: memory optimization and case-ignoring comparision
by Simon Marlow 16 Dec '05

16 Dec '05

On 14 December 2005 20:35, Bulat Ziganshin wrote: > i use utf8-packed strings in my program and have to ask 2 questions > about them: > > 1. i need function to do case-ignoring comparision of such strings. > stricmp is not appropriate because it don't know about utf8. can be > the existing Unicode support in Data.Char used for these or can the > appropriate support will be added? you should be able to use toUpper/toLower from Data.Char in GHC 6.4.1. > 2. what is the most memory-efficient representaion for such strings? > now i use John Meacham's library > (http://repetae.net/john/repos/jhc/PackedString.hs) which declares: > > newtype PackedString = PS (UArray Int Word8) > > but this uses two Ints just to hold index bounds: > > data UArray i e = UArray !i !i ByteArray# I don't know why an extra 8/16 bytes per string is that worrying - if you have so many small strings perhaps you should be sharing them via a hash table? > i want to use just memory ptr and put NUL at the end of array (my > strings never contain NUL chars). but what type i must use for this > ptr? ByteArray/ByteArray#, ForeignPtr, StablePtr, Ptr?? and which > function > i must use to quickly allocate memory i need? my packed strings will > be only unpacked and passed to "unsafe" C functions: stricmp, strcpy, > strcat; i plan to not use any other operations ForeignPtr and mallocForeignPtr are the way to go these days. In GHC 6.6 these will be much faster than before. Cheers, Simon

2 1

RE: Preparing a Registered hc-file-bundle (Was Bootstrapping withHCfiles)
by Simon Marlow 15 Dec '05

15 Dec '05

On 14 December 2005 09:46, Simon Marlow wrote: > On 13 December 2005 18:58, Andrew Walrond wrote: > >> Well, I gave it a blast anyway. All went well creating the hc files >> until the last command: >> >>> $ make hc-file-bundle Project=Ghc >> >> at which point it failed with >> >> tar czf ghc-6.4.1-x86_64-unknown-linux-hc.tar.gz `cat hc-files-to-go` >> tar: ghc-6.4.1/ghc/rts/AutoApply_thr.hc: Cannot stat: No such file >> or directory tar: ghc-6.4.1/ghc/rts/AutoApply_thr_p.hc: Cannot stat: >> No such file or directory tar: ghc-6.4.1/ghc/rts/AutoApply_debug.hc: >> Cannot stat: No such file or directory tar: Error exit delayed from >> previous errors >> make: *** [hc-file-bundle] Error 2 > > Yes, these are harmless. They are caused by the fact that > AutoApply.cmm is auto-generated in several different ways > (AutoApply_thr.cmm, AutoApply_debug.cmm etc.), and the .hc files for > these are named AutoApply_thr.thr_hc, AutoAPply_debug.debug_hc, and > so on. The Makefile code for hc-bundle doesn't know about this > special case. > >> Googling suggests this might be harmless, so I tried bootstrapping on >> the target machine using the created tarball, but that failed with >> >> /bin/ld: cannot find -lghccompat >> collect2: ld returned 1 exit status >> make[1]: *** [stage1/ghc-6.4.1] Error 1 >> make: *** [all] Error 1 >> make: Leaving directory >> `/home/andrew/test/ghc/bs/target/ghc-6.4.1/ghc' > > Please send us a full log of the build. libghccompat.a should be in > ghc/lib/compat, and it should be built quite early on. Actually, looking at ghc/Makefile, it is possible that ghc/lib isn't getting built early enough. Try changing: SUBDIRS = includes rts docs compiler lib utils driver to SUBDIRS = includes rts lib docs compiler utils driver Cheers, Simon

2 1

utf8 strings: memory optimization and case-ignoring comparision
by Bulat Ziganshin 14 Dec '05

14 Dec '05

Hello glasgow-haskell-users, i use utf8-packed strings in my program and have to ask 2 questions about them: 1. i need function to do case-ignoring comparision of such strings. stricmp is not appropriate because it don't know about utf8. can be the existing Unicode support in Data.Char used for these or can the appropriate support will be added? 2. what is the most memory-efficient representaion for such strings? now i use John Meacham's library (http://repetae.net/john/repos/jhc/PackedString.hs) which declares: newtype PackedString = PS (UArray Int Word8) but this uses two Ints just to hold index bounds: data UArray i e = UArray !i !i ByteArray# i want to use just memory ptr and put NUL at the end of array (my strings never contain NUL chars). but what type i must use for this ptr? ByteArray/ByteArray#, ForeignPtr, StablePtr, Ptr?? and which function i must use to quickly allocate memory i need? my packed strings will be only unpacked and passed to "unsafe" C functions: stricmp, strcpy, strcat; i plan to not use any other operations -- Best regards, Bulat mailto:bulatz@HotPOP.com

1 0

RE: Preparing a Registered hc-file-bundle (Was Bootstrapping withHC files)
by Simon Marlow 14 Dec '05

14 Dec '05

On 13 December 2005 18:58, Andrew Walrond wrote: > Well, I gave it a blast anyway. All went well creating the hc files > until the last command: > >> $ make hc-file-bundle Project=Ghc > > at which point it failed with > > tar czf ghc-6.4.1-x86_64-unknown-linux-hc.tar.gz `cat hc-files-to-go` > tar: ghc-6.4.1/ghc/rts/AutoApply_thr.hc: Cannot stat: No such file or > directory > tar: ghc-6.4.1/ghc/rts/AutoApply_thr_p.hc: Cannot stat: No such file > or directory > tar: ghc-6.4.1/ghc/rts/AutoApply_debug.hc: Cannot stat: No such file > or directory > tar: Error exit delayed from previous errors > make: *** [hc-file-bundle] Error 2 Yes, these are harmless. They are caused by the fact that AutoApply.cmm is auto-generated in several different ways (AutoApply_thr.cmm, AutoApply_debug.cmm etc.), and the .hc files for these are named AutoApply_thr.thr_hc, AutoAPply_debug.debug_hc, and so on. The Makefile code for hc-bundle doesn't know about this special case. > Googling suggests this might be harmless, so I tried bootstrapping on > the target machine using the created tarball, but that failed with > > /bin/ld: cannot find -lghccompat > collect2: ld returned 1 exit status > make[1]: *** [stage1/ghc-6.4.1] Error 1 > make: *** [all] Error 1 > make: Leaving directory > `/home/andrew/test/ghc/bs/target/ghc-6.4.1/ghc' Please send us a full log of the build. libghccompat.a should be in ghc/lib/compat, and it should be built quite early on. Cheers, Simon

1 0