October 2003 - Libraries

Subprocesses (was: [ANNOUNCE] shell-haskell 1.0.0)
by Simon Marlow 08 Jul '04

08 Jul '04

Libraries people, There are currently two ways to create and communicate with external processes: - System.Cmd.system (very basic, doesn't give you access to the process's output but you can use shell primitives to redirect its input/output). - Posix.runProcess (can specify Handles for the process's input/output, but it doesn't create pipes so you can't actually communicate with the subprocess using these Handles.). - POpen.popen (can specify an input String, and gives you lazy output Strings. No interactive communication is possible.) Ok, *three* ways. System.Cmd.system, Posix.runProcess, POpen.popen, and shell-haskell. Er. *Four* ways. David Sankel's shell-haskell creates pipes for stdin/stdout/stderr, and lets you communicate interactively with the sub-process via Handles. It could do with a way to specify the environment and working directory of the sub-process (ala POpen and runProcess), but apart from that it seems to be the most flexible of the four. It could be simplified slightly by using one Handle rather than two for stdin/stdout. We should think about what functionality we want to provide in the core libraries. My preference would be to go for a full version with three pipes (like shell-haskell) and perhaps one or two more specialised versions covering common cases (I don't know what these should be, though). What do people think? Cheers, Simon -----Original Message----- From: haskell-admin(a)haskell.org [mailto:haskell-admin@haskell.org] On Behalf Of David Sankel Sent: 06 May 2003 03:06 To: haskell(a)haskell.org Subject: [ANNOUNCE] shell-haskell 1.0.0 =================== shell-haskell 1.0.0 =================== shell-haskell provides utilities to communicate with other process's via Haskell code. This library can be used for something as simple as getting the output of another program or as complex as interacting with an interpreter. see http://www.electronconsulting.com/shell-haskell for details. David J. Sankel _______________________________________________ Haskell mailing list Haskell(a)haskell.org http://www.haskell.org/mailman/listinfo/haskell

5 10

--something accepted as argument of option in GetOpt.hs
by Johan Mårtensson 30 Oct '03

30 Oct '03

Hi, I have noticed that GetOpt.hs accepts option arguments starting with a dash. This might of use in some applications but it may also be a nuisance, because if you omit the argument of an ReqArg option this may not be reported as an error because GetOpt just takes the next flag (staring with a '-') as the argument to the option in question. This could be solved by introducing a new constructor in the 'ArgDescr a' datatype, one that not only requires an argument, but an argument not starting with '-'. I think something like this little patch would do the job. 85a86 > | ReqArgNoDash (String -> a) String -- ^ option requires argument 117a119 > fmtShort (ReqArgNoDash _ ad) so = "-" ++ [so] ++ " " ++ ad 122a125 > fmtLong (ReqArgNoDash _ ad) lo = "--" ++ lo ++ "=" ++ ad 179a183,186 > long [ReqArgNoDash _ d] [] [] = (errReq d optStr,[]) > long [ReqArgNoDash _ d] [] r@(('-':_):_) = (errReq d optStr,r) > long [ReqArgNoDash f _] [] (r:rest) = (Opt (f r),rest) > long [ReqArgNoDash f _] ('=':xs) rest = (Opt (f xs),rest) 196a204,207 > short (ReqArgNoDash _ d:_) [] [] = (errReq d optStr,[]) > short (ReqArgNoDash _ d:_) [] r@(('-':_):_) = (errReq d optStr,r) > short (ReqArgNoDash f _:_) [] (r:rest) = (Opt (f r),rest) > short (ReqArgNoDash f _:_) xs rest = (Opt (f xs),rest) regards, -- ------------------------------------------------------------ Johan Martensson, PhD Office: +46 31 772 81 70 R&D Mobile: +46 703749681 Safelogic AB Fax: +46 31 772 80 91 Stena Center 1C johan.martensson(a)safelogic.se SE-412 92 Gothenburg, SWEDEN PGP key ID A8857A60 www.safelogic.se ------------------------------------------------------------

1 0

RE: Special Invitation :-) HC&A Report (November 2003) (fwd)
by Simon Peyton-Jones 29 Oct '03

29 Oct '03

Good. I hope Isaac will do a write up of his own. Re the "improved package mechanism", perhaps add a URL? http://www.haskell.org/~simonmar/packages.html Simon | -----Original Message----- | From: libraries-bounces(a)haskell.org [mailto:libraries-bounces@haskell.org] On Behalf Of Malcolm | Wallace | Sent: 28 October 2003 11:12 | To: libraries(a)haskell.org | Subject: Re: Special Invitation :-) HC&A Report (November 2003) (fwd) | | "C.Reinke" <C.Reinke(a)kent.ac.uk> writes: | | > That includes individual libraries, developments in the hierarchical | > libraries (Simon M., the usual contact for this, is away;), as well | > as the discussions (Haskell WS & after) about library | > infrastructure, and your conclusions and plans. In short, anything | > interesting that happened on this list but may not have reached the | > general Haskeller yet. | | Since Simon is away, I thought I would attempt to write a general | introduction to the meta-activity in the libraries area (see below). | However, rather than usurp his title by stealth, I am posting my | summary to the libraries list so that anyone can edit/improve it. | Please feel free to write a better summary and submit it to Claus! | | (And individuals who have been proposing infrastructural things | might want to add a few more paragraphs of detail on their own | contributions.) | | Regards, | Malcolm | | | ---- | HC&A Report - Hierarchical libraries | | Apart from actually writing libraries that do stuff, recent | meta-activity in the libraries community has concentrated on plans for | making libraries easier to install, to distribute, to maintain, and | therefore also easier to write and contribute to Haskellers at large. | | Isaac Jones <ijones(a)syntaxpolice.org> has been leading the effort in | bringing together a proposal for a common distribution/installation | mechanism. The aim is for a general framework that can be re-used | by anyone wanting to contribute a library, so that they can be sure | it will work uniformly across all the compilers/interpreters without | too much effort. It covers such areas as automated configure/build, | library registration, dependencies on other library packages, and | automatic re-installation when a compiler version is updated. | | Meanwhile, the two Simons at GHC-HQ have been proposing an improved | 'package' mechanism for the compilers/interpreters themselves. | This proposal aims to create a more flexible 'backend' for library | usage, such that libraries can be re-located within the hierarchy, | and different versions of a library can co-exist together. | | In other recent news, the common 'base' package of libraries has | recently been subdivided to separate off non-core functionality, making | the core more stable, whilst enabling independent evolution of the | other libraries. The split-off packages include parsec, QuickCheck, | and the monad transformers. Other packages resident in the haskell.org | CVS tree include OpenGL, GLUT, HaXml, Japi (Java API), ObjectIO, | Win32, X11, HGL (Haskell Graphics Library), haskell-src, network, | readline, and unix. | ---- | _______________________________________________ | Libraries mailing list | Libraries(a)haskell.org | http://www.haskell.org/mailman/listinfo/libraries

2 1

RE: sleeping
by Simon Marlow 29 Oct '03

29 Oct '03

> Nope, that's exactly the part of Control.Concurrent which is > guarded with > #ifdef __GLASGOW_HASKELL__, so it won't work with Hugs and nhc98. > > Nevertheless, I'm sure every platform has a good way of > sleeping, so my > proposal is to move (u)sleep out of the unix package into the base > libraries, but I'm not sure where exactly in the hierarchy it > should be placed. Somewhere under System probably, but it doesn't really fit into any of the existing modules, and adding a whole new module seems like overkill. Perhaps just adding it to System.IO is ok. Remember to make the GHC implementation use threadDelay, because System.Posix.(u?)sleep isn't thread-safe. Cheers, Simon

1 0

Special Invitation :-) HC&A Report (November 2003) (fwd)
by C.Reinke 29 Oct '03

29 Oct '03

Dear Haskell libraries specialists, the Haskell community hopes to hear from you about all the interesting stuff you've been brewing over the last six months, not to mention the even more interesting stuff in the pipeline!-) That includes individual libraries, developments in the hierarchical libraries (Simon M., the usual contact for this, is away;), as well as the discussions (Haskell WS & after) about library infrastructure, and your conclusions and plans. In short, anything interesting that happened on this list but may not have reached the general Haskeller yet. Check the topics and contributors list to see what still needs filling in: http://www.haskell.org/communities/topics.html ----------------------------------------------------------------- Haskell Communities and Activities Report (November 2003 edition) http://www.haskell.org/communities/ Could everybody send in their individual reports ** by the end of THIS week ** ? Please?-) ----------------------------------------------------------------- Thanks, Claus (current editor)

6 5

sleeping
by Sven Panne 28 Oct '03

28 Oct '03

Just a short question: If I'm not mistaken, there is no way in the "base" package to sleep for a given amount of time in portable a way. Of course there is System.Posix.Unistd.sleep and System.Posix.Unistd.usleep in the "unix" package, but shouldn't we have a similar feature in the "base" libraries? Or have I overlooked something obvious? Cheers, S.

4 5

NOBLE
by JAMES UGOMA 24 Oct '03

24 Oct '03

HELLO, I WISH TO BEGIN BY WAY OF INTRODUCTION, I AM JAMES UGOMA, THE FIRST SON OF THE LATE EGOH UGOMA, ONE OF THE FORMOST BLACK FARMERS IN ZIMBABWE RECENTLY MURDERED IN THE LAND DISPUTE IN MY COUNTRY. BEFORE THE DEATH OF MY FATHER, HE HAD TAKEN ME TO JOHANNESBURG IN SOUTH AFRICA, TO DEPOSIT THE SUM OF US$ 14.5MILLION (FOURTEEN POINT FIVE UNITED STATES DOLLARS) IN ONE OF THE PRIVATE SECURITY COMPANIES, AS FAMILY VALUABLES, WHEN HE REALISED THE LOOMING DANGER IN ZIMBABWE. THIS AMOUNT WAS MEANT FOR THE PURCHASE OF NEW MACHINERIES AND CHEMICALS FOR THE FARMS AND ALSO FOE THE ESTABLISHMENT OF NEW FARMS IN SWAZILAND. THIS LAND DISPUTE STARTED WHEN PRESIDENT ROBERT MUGABE INTRODUCED A NEW LAND REFORM WHICH HAS PARTICULARLY TARGETED AT THE RICH, WHITE FARMERS AND A FEW BLACK FARMERS WITH MY FATHER INCLUDED. THIS RESULTED IN THE GRUESOME KILLING OF RICH FARMERS (MAINLY WHITES AND FEW BLACKS) AND THE UNLAWFULL POSSESION OF THEIR PROPERTIES BY MUGABE'S WAR VETERANS,UNDER THE DISGUISE OF FIGHTING FOR THE INTEREST OF THE COUNTRY. IT IS AGAINST THIS BACKGROUND I FLED THE COUNTRY WITH MY FAMILY, TO THE EAST OF GERMANY WHERE WE ARE CURRENTLY SEEKING POLITICAL ASYLUM (REFUGEE). I HAVE DECIDED TO SEEK FOREIGN ASSISTANCE, AS THE GERMAN LAWS PROHITBITS ASYLUM SEEKERS FROM OPARATING BANK ACCOUNTS OR INVOLVE IN FINANCIAL TRANSACTIONS OF ANY KIND.HENCE THIS LETTER TO YOU. ALL I NEED IS A TRUST WORTHY PERSON IN WHOSE CUSTODY I SHALL ENTRUST THE ABOVE SAID AMOUNT AND HELP ME WITH INVESTMENT OPPORTUNTIES. THERE ARE TWO OPTIONS, FIRSTLY, YOU CAN CHOOSE TO HAVE A PERCENTAGE OF THE TOTAL AMOUNT FOR YOUR ASSISTANCE OR GO INTO PARTNERSHIP WITH ME FOR PROPER INVESTMENT OF THE MONEY IN YOUR COUNTRY. WE SHALL DISCUSS IN DETAILS WHEN YOU REPLY. I AWAIT YOUR REPLY, WHILE I IMPLORE YOU TO MAINTAIN THE ABSOLUTE SECRECY REQUIRED TO ENSURE A SAFE AND HITCH FREE TRANSACTION.I SHALL FURNISH YOU WITH MORE DETAILS ON REQUEST.I WISH YOU THE BEST IN LIFE (GENERALLY) AS I AWAIT FOR YOUR POSITIVE RESPONSE.YOU CAN REACH ME ON MY E-MAIL ADDRESS:jamesugoma@netscape.net YOURS ,JAMES UGOMA .

1 0

RE: Calling an external command from a Haskell program
by Simon Marlow 23 Oct '03

23 Oct '03

> > >> I checked libevent, and it doesn't support wait() events. > It would be > >> great if it did... > > > > You can catch sigCHLD in the main event-loop and then invoke wait(). > > [This could even be possible in the current RTS by applying another > > incarnation of Wolfgang's pipe-hack] > > Why another incarnation? The "first incarnation" should do the job, > too. What's wrong with just safe-foreign-calling wait and relying on > the threaded RTS to do the job? Sure, we could do that - and we could do it for the I/O library too and remove all the select() stuff entirely. But that would impose a performance penalty on applications with large numbers of I/O-bound threads, so we like to keep the lightweight mechanism too. The same argument applies to process waiting, although you could argue that it is less convincing there, because there are probably not that many applications which need to wait on thousands of processes. Cheers, Simon

1 0

control your mood swings
by bczulc＠gte.net 22 Oct '03

22 Oct '03

1 0

RE: Calling an external command from a Haskell program
by Simon Marlow 22 Oct '03

22 Oct '03

> In local.libraries, you wrote: > > The sticking point is that GHC needs a > > thread-safe wait() implementation, which means some hacking in the > > scheduler (there isn't even a good way to do this portably, because > > select() doesn't let you wait for process termination). > > *mumbl* libevent *mumbl* I checked libevent, and it doesn't support wait() events. It would be great if it did... Cheers, Simon

3 2