March 2009 - Haskell-Cafe

MissingH bracketCD (aka bracketCWD) bug -- this is the infamous lazy io, right?
by Thomas Hartman 23 Mar '09

23 Mar '09

I got bitten by a bug (well, I call it bug) in bracketCD from HSH/MissingH demonstrated by the following code bracketCD is very useful for sysadminny one-offs, I use it all the time, but..... I suspect that unless people are very careful, this behavior will affect other users of bracketCD, in potentially very subtle and tricky ways. 1) -- is there a more elegant way to deal with lazy io than the hack below? (putStrLn the last character) 2) -- should MissingH function bracketCD be fixed, and if so how? import System.FilePath.Find (find) import System.Directory (getDirectoryContents, getCurrentDirectory, setCurrentDirectory) import System.Path (bracketCWD) -- from MissingH -- the fixed function, a more restrictive type than bracketCWD which lets you do any io, not just showables myBracketCWD :: Show a => FilePath -> IO a -> IO a myBracketCWD fp action = do oldcwd <- getCurrentDirectory setCurrentDirectory fp a <- action putStrLn $ show . last . show $ a -- force evaluation setCurrentDirectory oldcwd return a -- rather than listing "/", lists dir this module is in tWrong = bracketCWD "/" $ return . take 2 =<< listCurrentWithFind -- this lists filesystem root tRight = myBracketCWD "/" $ return . take 2 =<< listCurrentWithFind listCurrentWithFind = System.FilePath.Find.find (return True) (return True) "." -- for some reason, bracketCWD does the right thing with get tGetContents = myBracketCWD "/" $ return . take 2 =<< getDirectoryContents "." Actual code for bracketCWD in MissingH: -- this is from MissingH, seems there's a bug when attempt to wrap a find command bracketCWD :: FilePath -> IO a -> IO a bracketCWD fp action = do oldcwd <- getCurrentDirectory setCurrentDirectory fp finally action (setCurrentDiurectory oldcwd)

2 1

How to send signal from Xmonad to my app?
by Magicloud Magiclouds 23 Mar '09

23 Mar '09

Hi, I have a little program, which shows a message box when gets 'USR1' signal. This works fine when I `kill -USR1 $PID`. But my code in xmonad.hs does not work. No error throws out. And as I inspected, the "pid" is right. Only nothing happened in my app. So I wonder what is wrong? 28 , ((mod4Mask, xK_x), sendSignal2PIDinFile sigUSR1 ".sidebar.pid") 31 sendSignal2PIDinFile signal pidfile = io $ do 32 pid <- readFile pidfile 33 signalProcess signal $ read pid -- 竹密岂妨流水过山高哪阻野云飞

1 0

ANN: hranker: Basic utility for ranking a list of items (e.g. for the logo poll)
by Robin Green 22 Mar '09

22 Mar '09

hranker is a fun little command-line utility I have written to help a user rank a list of items (of any type implementing Show, Eq and Ord). I hope the code is sufficiently clear that it could also serve as an educational piece of code, especially for people wanting to learn how to use the HCL library. It is now available on hackage: http://hackage.haskell.org/cgi-bin/hackage-scripts/package/hranker Just typing hranker or hranker --help will yield a usage message. Here is an example usage of the program: In the unlikely event that anyone else wants to use it to help them prepare their votes in the Haskell logo poll (which is what I used it for just now), here is what to do: 1. cabal install hranker (or download it and runhaskell Setup.lhs etc.) 2. hranker 113 (because 113 is the number of items in the poll) 3. IMPORTANT: Read the warnings the program prints on startup! 4. To control the program, enter the first character of the command you wish to perform, and then press Enter. The command you have chosen may ask you for further information. Commands generally operate on the next unranked item, apart from annotate (which can operate on ranked or unranked items), remove (which always operates on ranked items), and quit. I recommend you use the annotate command a lot - it is really helpful for this kind of situation where you have a lot of numbered items to rank. If you use the remove command, don't remove *all* ranked items, because if you do you it may terminate. 5. If you rank *everything* somehow (remember, ties are allowed), the program will print an inverse mapping (items to ranks, instead of ranks to items) before terminating. If you choose to quit before that point, you won't get that inverse mapping, unless you explicitly invoke the mapping command. There is obvious room for improvement: apart from fixing the remove bug mentioned above, there is currently no simple way to remove or annotate just *one* item at a rank when multiple items are tied at that rank. Have fun! There are some hilarious entries in the logo poll, if you haven't yet voted. -- Robin

1 0

Transforming XML using HXT, I'm confused.
by P Orrifolius 22 Mar '09

22 Mar '09

Hello, I'm trying to transform a fragment of XML using HXT, specifically I'm trying to strip out any disallowed content while retaining as much as possible. For example I might be given "<bar><evil>blah</evil></bar>". I want to get rid of the 'evil' element but retain it's content resulting in "<bar>blah</bar>", rather than just "<bar></bar>". My basic theory was to do a bottom up traversal replacing any invalid elements with their children. Unfortunately I've got myself thoroughly confused. The code below shows what I've tried so far (except the many random permutations of collapseInvalids I've tried!). I think one problem is that validElements is choosing based on the parent element but processing will just be given the child. Also I think some inputs will require multiple passes over the children during the traversal, eg given "...<baz>s6<bar>s7</bar></baz>..." I'd like "...<baz>s6s7</baz>..." but when processing the bar it is valid, it's only once you're processing the baz you see that you need to pull the s7 up. I'd really appreciate any advice, and I'm not wedded to the current code so feel free to throw it all out! :) Thanks P. ps: Is it possible to render an XmlTree as String without invoking IO? I've just been doing runLA (hread >>> collapseInvalids) frag1 and looking at the nodes. module Test where import Text.XML.HXT.Arrow --collapseInvalids = processBottomUp (processChildren ((getChildren `whenNot` validElements) `orElse` this)) -- <foo> can contain <foo>, <bar> or <baz>, not text validInFoo :: (ArrowXml a) => a XmlTree XmlTree validInFoo = foldr1 orElse ( map hasName ["foo", "bar", "baz"] ) -- <bar> can contain <foo>, <baz> or text, not <foo> validInBar :: (ArrowXml a) => a XmlTree XmlTree validInBar = foldr1 orElse ( isText : map hasName ["foo", "baz"] ) -- <baz> can only contain text validInBaz :: (ArrowXml a) => a XmlTree XmlTree validInBaz = foldr1 orElse ( isText : []) validElements :: (ArrowXml a) => a XmlTree XmlTree validElements = choiceA ( hasName "foo" :-> validInFoo : hasName "bar" :-> validInBar : hasName "baz" :-> validInBaz : isText :-> this : []) frag1 = concat ( "s1" : "<foo>" : "<bar>" : "s2" : "<evil>" : "s3" : "<baz>" : "s4" : "</baz>" : "s5" : "</evil>" : "s6" : "</bar>" : "</foo>" : "s7" : []) result1 = concat ( "s1" : "<foo>" : "<bar>" : "s2s3" : "<baz>" : "s4" : "</baz>" : "s5s6" : "</bar>" : "</foo>" : "s7" : []) frag2 = concat ( "s1" : "<foo>" : "s2" : "<evil>" : "s3" : "</evil>" : "<bar>" : "s4" : "<evil>" : "s5" : "<baz>" : "s6" : "<bar>" : "s7" : "</bar>" : "</baz>" : "s8" : "<bar>" : "s9" : "</bar>" : "</evil>" : "s10" : "</bar>" : "</foo>" : "s11" : []) result2 = concat ( "s1" : "<foo>" : "<bar>" : "s4" : "<evil>" : "s5" : "<baz>" : "s6" : "<bar>" : "s7" : "</bar>" : "</baz>" : "s8" : "<bar>" : "s9" : "</bar>" : "</evil>" : "s10" : "</bar>" : "</foo>" : "s11" : [])

1 0

Re: Re: [Haskell-cafe] [Haskell-beginners] appropriateness of haskell for GUIs
by Sanny Sanoff 22 Mar '09

22 Mar '09

Brandon S. Allbery KF8NH wrote: > > > Glade works fine if all you want is to place widgets; but it won't > generate template code in Haskell (it will for C and C++) to connect > the widgets together, which is what I understand from "you have to > type everything yourself". > > I use Template Haskell to extract information from .glade files and produce some haskell glue on the fly. So you need to declare stuff only once (in glade file). Regards, San. -- View this message in context: http://www.nabble.com/Re%3A--Haskell-beginners--appropriateness-of-haskell-… Sent from the Haskell - Haskell-Cafe mailing list archive at Nabble.com.

1 0

Re: [Haskell-cafe] Haskell syntax highlighting support for Trac
by Patai Gergely 22 Mar '09

22 Mar '09

I'm not sure what your problem is. I downloaded Trac not long ago, and Haskell highlighting worked out of the box for me after installing Pygments. Here's the proof (this is the wiki of a course I'm teaching): http://kfarkas.iit.bme.hu/trac/bfp/wiki/EloadasAbsztrakciok1 The code to activate it is the following: {{{ #!hs main = print "Hello world!" }}} Gergely -- http://www.fastmail.fm - One of many happy users: http://www.fastmail.fm/docs/quotes.html

2 1

Re: [Haskell-cafe] Ease of Haskell development on OS X?
by Ross Mellgren 22 Mar '09

22 Mar '09

(back to the list) Answers inline: On Mar 21, 2009, at 2:42 PM, Colin Adams wrote: > Yes, that was the problem, and swapping the PATH order does the trick. > Thanks. no prob. > I must still have the macports stuff installed. Can you tell me how to > get rid of it? if you really want to get rid of it I believe you just have to rm -rf / opt/local and remove any initialization hooks for /opt/local out of / etc/profile and ~/.bash_profile > However I still can't install my program - I get missing dependencies > for gtk glib cairo and svgcairo. I'm not really that savvy with the package registration magics -- on my system after doing sudo make install for gtk2hs, I get the packages properly registered. You can check the registered packages with ghc- pkg list. It could be that the package is registered local for your user, but the cabal install line you're using for the application includes --global (or your cabal configuration) so it won't look in the user config. I think Duncan answered some questions about this same problem recently, though I don't have a mailing list reference. > I tried some of the gtk2hs demos - they work, but, for instance, the > actionmenu demo does not put the menus up at the top of the screen, so > it doesn't look like the framework integration has worked. Even with the integration everything is still GTK-ish. There's a separate framework that comes with the .dmg called ige-mac-integration that allows applications to meld with the mac environment better, but no gtk2hs bindings for this exist at the moment. Bindings would have to be created, and then you'd have to modify the application to make use of them (presumably with CPP or similar switch to control whether you want mac support, or just plain GTK) -Ross > 2009/3/21 Ross Mellgren <rmm-haskell(a)z.odi.ac>: >> I think there must be a version inconsistency with your GLib >> framework -- >> either not the most recent copy of the GTK+ DMG, or your path is >> flipped >> around and you're using a ports version. >> >> Run glib-mkenums --version to see what version you have... I have >> >> rmm@Hugo:~/tmpgtk/GLib.framework/Resources/dev/bin$ ./glib-mkenums >> --version >> glib-mkenums version glib-2.18.1 >> glib-mkenums comes with ABSOLUTELY NO WARRANTY. >> You may redistribute copies of glib-mkenums under the terms of >> the GNU General Public License which can be found in the >> GLib source package. Sources, examples and contact >> information are available at http://www.gtk.org >> >> Particularly, the error you're getting is because your version of >> glib-mkenums doesn't expand @ENUMPREFIX@ -- if I run glib-mkenums >> with no >> arguments, my version reports @ENUMPREFIX@ as a valid subtitution: >> >> rmm@Hugo:~/tmpgtk/GLib.framework/Resources/dev/bin$ ./glib-mkenums >> Usage: glib-mkenums [options] [files...] >> --fhead <text> output file header >> --fprod <text> per input file production >> --ftail <text> output file trailer >> --eprod <text> per enum text (produced prior to value >> itarations) >> --vhead <text> value header, produced before iterating >> over >> enum values >> --vprod <text> value text, produced for each enum value >> --vtail <text> value tail, produced after iterating >> over enum >> values >> --comments <text> comment structure >> --template file template file >> -h, --help show this help message >> -v, --version print version informations >> Production text substitutions: >> @EnumName@ PrefixTheXEnum >> @enum_name@ prefix_the_xenum >> @ENUMNAME@ PREFIX_THE_XENUM >> @ENUMSHORT@ THE_XENUM >> @ENUMPREFIX@ PREFIX >> @VALUENAME@ PREFIX_THE_XVALUE >> @valuenick@ the-xvalue >> @type@ either enum or flags >> @Type@ either Enum or Flags >> @TYPE@ either ENUM or FLAGS >> @filename@ name of current input file >> >> >> Does yours? >> >> You might try moving /Library/Frameworks/GLib.framework/Resources/ >> dev/bin to >> the front of your path before make'ing pango -- >> >> cd pango-1.24.0 >> export PATH=/Library/Frameworks/GLib.framework/Resources/dev/bin: >> $PATH >> make >> >> Hope this helps, >> >> -Ross >> >> On Mar 21, 2009, at 1:52 PM, Colin Adams wrote: >> >>> Attached. >>> >>> >>> >>> 2009/3/21 Ross Mellgren <rmm-haskell(a)z.odi.ac>: >>>> >>>> (taking this off list, to avoid noise) >>>> >>>> Could you attach pango-1.24.0/pango/pango-enum-types.h? Something >>>> hokey >>>> is >>>> going on -- this file is shipped with pango-1.24.0 but may be >>>> overwritten >>>> if >>>> your glib-mkenums does it differently. In any case, since it's >>>> (possibly) >>>> machine generated I would have to see what's going on at those >>>> lines to >>>> make >>>> progress. >>>> >>>> Also, if you could cd pango-1.24.0/pango, and run: >>>> >>>> gcc -E -DHAVE_CONFIG_H -I. -I.. -DG_LOG_DOMAIN=\"Pango\" >>>> -DPANGO_ENABLE_BACKEND -DPANGO_ENABLE_ENGINE >>>> -DSYSCONFDIR=\"/usr/local/etc\" >>>> -DLIBDIR=\"/usr/local/lib\" -I.. -DG_DISABLE_CAST_CHECKS >>>> -I/Library/Frameworks/GLib.framework/Headers >>>> -I/Library/Frameworks/Cairo.framework/Headers -I/usr/X11/ >>>> include >>>> -DFC_WEIGHT_EXTRABLACK=215 -Wall -MT fonts.lo -MD -MP -MF >>>> .deps/fonts.Tpo >>>> -c -o fonts.i fonts.c >>>> >>>> and attach fonts.i, that might shed more light. However, since >>>> the error >>>> you're getting is a preprocessor one, gcc -E may fail as well. >>>> >>>> -Ross >>>> >>>> On Mar 21, 2009, at 1:37 PM, Colin Adams wrote: >>>> >>>>> In file included from ../pango/pango.h:31, >>>>> from pango-impl-utils.h:28, >>>>> from fonts.c:30: >>>>> ../pango/pango-enum-types.h:12:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:14:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:17:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:19:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:22:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:25:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:27:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:29:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:31:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:33:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:36:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:38:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:41:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:43:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:45:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:48:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:51:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:54:9: error: macro names must be >>>>> identifiers >>>>> In file included from ../pango/pango.h:31, >>>>> from pango-impl-utils.h:28, >>>>> from glyphstring.c:26: >>>>> ../pango/pango-enum-types.h:12:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:14:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:17:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:19:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:22:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:25:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:27:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:29:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:31:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:33:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:36:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:38:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:41:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:43:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:45:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:48:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:51:9: error: macro names must be >>>>> identifiers >>>>> ../pango/pango-enum-types.h:54:9: error: macro names must be >>>>> identifiers >>>>> >>>>> >>>>> 2009/3/21 Ross Mellgren <rmm-haskell(a)z.odi.ac>: >>>>>> >>>>>> I didn't get any errors like that (nor do I remember ever >>>>>> having them), >>>>>> though if you want to paste them here maybe I can help with them. >>>>>> >>>>>> -Ross >>>>>> >>>>>> On Mar 21, 2009, at 1:27 PM, Colin Adams wrote: >>>>>> >>>>>>> If I try this, pango fails to compile with lots of error >>>>>>> messages >>>>>>> about error: macro names must be identifiers. >>>>>>> >>>>>>> I think I've seen this before - some well-known Mac OSX >>>>>>> problem? (I'm >>>>>>> a linux man myself - so I'm not used to the mac) >>>>>>> >>>>>>> 2009/3/21 Ross Mellgren <rmm-haskell(a)z.odi.ac>: >>>>>>>> >>>>>>>> I tried making this work, but librsvg requires pango, and >>>>>>>> pango is a >>>>>>>> huge >>>>>>>> pain in the ass -- I managed to get the whole thing to >>>>>>>> compile, but >>>>>>>> now >>>>>>>> it >>>>>>>> can't find any fonts, apparently due to some dynaloading >>>>>>>> issues. >>>>>>>> >>>>>>>> I think if you need any of the extended modules (e.g. >>>>>>>> svgcairo, gl) >>>>>>>> that >>>>>>>> the >>>>>>>> DMG doesn't directly support then you're best served by >>>>>>>> shedding a >>>>>>>> couple >>>>>>>> tears and going with macports. >>>>>>>> >>>>>>>> The steps I used to get the broken version were (FYI -- doing >>>>>>>> this >>>>>>>> causes >>>>>>>> all fonts to load as the no-character square in all gtk2hs >>>>>>>> apps): >>>>>>>> >>>>>>>> export >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> PKG_CONFIG_PATH=/usr/X11/lib/pkgconfig:/usr/local/lib/ >>>>>>>> pkgconfig:/Library/Frameworks/Cairo.framework/Resources/dev/ >>>>>>>> lib/pkgconfig:/Library/Frameworks/GLib.framework/Resources/ >>>>>>>> dev/lib/pkgconfig:/Library/Frameworks/Gtk.framework/Resources/ >>>>>>>> dev/lib/pkgconfig >>>>>>>> >>>>>>>> export >>>>>>>> PATH=$PATH:/Library/Frameworks/GLib.framework/Resources/dev/bin >>>>>>>> >>>>>>>> cd pango-1.24.0 >>>>>>>> CFLAGS="-DFC_WEIGHT_EXTRABLACK=215" ./configure --prefix=/usr/ >>>>>>>> local >>>>>>>> make -j4 >>>>>>>> sudo make install >>>>>>>> >>>>>>>> cd librsvg-2.26.0 >>>>>>>> ./configure --prefix=/usr/local >>>>>>>> make -j4 >>>>>>>> sudo make install >>>>>>>> >>>>>>>> cd gtk2hs-0.10.0 >>>>>>>> ./configure --disable-gio >>>>>>>> make -j4 >>>>>>>> sudo make install >>>>>>>> >>>>>>>> Running apps linked with this version of gtk2hs gets: >>>>>>>> >>>>>>>> (svgviewer:43314): Pango-CRITICAL **: No modules found: >>>>>>>> No builtin or dynamically loaded modules were found. >>>>>>>> PangoFc will not work correctly. >>>>>>>> This probably means there was an error in the creation of: >>>>>>>> '/usr/local/etc/pango/pango.modules' >>>>>>>> You should create this file by running: >>>>>>>> pango-querymodules > '/usr/local/etc/pango/pango.modules' >>>>>>>> >>>>>>>> (svgviewer:43314): Pango-WARNING **: failed to choose a font, >>>>>>>> expect >>>>>>>> ugly >>>>>>>> output. engine-type='PangoRenderATSUI', script='latin' >>>>>>>> >>>>>>>> So if anyone knows what Pango is trying to do, maybe they >>>>>>>> could help >>>>>>>> resolve >>>>>>>> this. >>>>>>>> >>>>>>>> I'm probably going to ditch the use of GTK in my own project >>>>>>>> and use >>>>>>>> FFI >>>>>>>> bindings to Carbon/Win32, since I don't really need widgets >>>>>>>> for my >>>>>>>> own >>>>>>>> stuff >>>>>>>> and the whole family of GTK related packages is a total pain >>>>>>>> in the >>>>>>>> ass >>>>>>>> (as >>>>>>>> are most things in the GNU family of software, IMO) >>>>>>>> >>>>>>>> -Ross >>>>>>>> >>>>>>>> On Mar 21, 2009, at 10:13 AM, Colin Paul Adams wrote: >>>>>>>> >>>>>>>>>>>>>> "Ross" == Ross Mellgren <rmm-haskell(a)z.odi.ac> writes: >>>>>>>>> >>>>>>>>> Ross> While there is not a .dmg for Gtk2Hs, you can use >>>>>>>>> a .dmg >>>>>>>>> Ross> installed GHC with a .dmg installed Gtk, and then build >>>>>>>>> Ross> gtk2hs straight on top of that, without having to >>>>>>>>> deal with >>>>>>>>> Ross> the dual-GHC macports mess.. >>>>>>>>> >>>>>>>>> Ross> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> http://www.haskell.org/haskellwiki/Gtk2hs#Using_the_GTK.2B_OS_X_Framework >>>>>>>>> >>>>>>>>> I just tried this. >>>>>>>>> >>>>>>>>> The configure enables cairo, but does not enable svgcairo. >>>>>>>>> Is there >>>>>>>>> anyway round this, or do I have to revert to the macports >>>>>>>>> gtk? (my >>>>>>>>> application uses svgcairo) >>>>>>>>> -- >>>>>>>>> Colin Adams >>>>>>>>> Preston Lancashire >>>>>>>> >>>>>>>> _______________________________________________ >>>>>>>> Haskell-Cafe mailing list >>>>>>>> Haskell-Cafe(a)haskell.org >>>>>>>> http://www.haskell.org/mailman/listinfo/haskell-cafe >>>>>>>> >>>>>> >>>>>> >>>> >>>> >>> <pango-enum-types.h> >> >>

5 19

about beta NF in lambda calculus
by Algebras Math 22 Mar '09

22 Mar '09

hi, What is the different between 'in beta normal form' and 'has beta normal form' ? Does the former means the current form of the term is already in normal form but the latter means that it is not a normal form yet and can be reduced to be normal form? Like \x.x is in NF and (\x.x) (\x.x) has NF? If above is true, I am confused why we have to distinguish the terms which have NF and be in NF? isn't the terms have NF will eventually become in NF? or there are some way to avoid them becoming in NF? Thanks Alg

5 5

ANN: regex-tdfa 1.1.1
by ChrisK 22 Mar '09

22 Mar '09

This is both a bug fix release and a feature release. The bug fix is a bit embarrassing, the indices were correct but the captured text was wrong in version 1.1.0. Oops. As of version 1.1.1 the following GNU extensions are recognized, all anchors: \` at beginning of entire text \' at end of entire text \< at beginning of word \> at end of word \b at either beginning or end of word \B at neither beginning nor end of word There is a "newSyntax" field of CompOptions that can enable/disable this syntax. These were requested by users of the Yi project. The "home page" is on the wiki and UPDATED a bit at http://haskell.org/haskellwiki/Regular_expressions#regex-tdfa The HADDOCK fails on hackage for now, but is up at http://darcs.haskell.org/packages/regex-unstable/regex-tdfa/doc/html/regex-… The released code is on hackage at http://hackage.haskell.org/cgi-bin/hackage-scripts/package/regex-tdfa The darcs repository is at http://darcs.haskell.org/packages/regex-unstable/regex-tdfa/ Cheers, Chris

1 0

Completely confused by cabal
by Colin Paul Adams 22 Mar '09

22 Mar '09

[colin@susannah ghc]$ ghc-pkg list /usr/local/lib/ghc-6.11.20090319/./package.conf: Cabal-1.7.0, array-0.2.0.1, base-3.0.3.0, base-4.0.0.0, bytestring-0.9.1.4, containers-0.2.0.1, directory-1.0.0.2, (dph-base-0.4.0), (dph-par-0.4.0), (dph-prim-interface-0.4.0), (dph-prim-par-0.4.0), (dph-prim-seq-0.4.0), (dph-seq-0.4.0), editline-0.2.1.0, ffi-1.0, filepath-1.1.0.1, gconf-0.10.0, (ghc-6.11.20090319), ghc-prim-0.1.0.0, gio-0.10.0, glib-0.10.0, haskell98-1.0.1.0, hpc-0.5.0.2, integer-0.1.0.0, old-locale-1.0.0.1, old-time-1.0.0.1, packedstring-0.1.0.1, pretty-1.0.1.0, process-1.0.1.1, random-1.0.0.1, rts-1.0, syb-0.1.0.0, template-haskell-2.3.0.0, unix-2.3.1.0 /home/colin/.ghc/x86_64-linux-6.11.20090319/package.conf: EdisonAPI-1.2.1, EdisonCore-1.2.1.3, HUnit-1.2.0.3, QuickCheck-1.2.0.0, cairo-0.10.0, game-tree-0.1.0.0, gconf-0.10.0, gio-0.10.0, glade-0.10.0, glib-0.10.0, gnomevfs-0.10.0, gstreamer-0.10.0, gtk-0.10.0, gtkglext-0.10.0, mozembed-0.10.0, mtl-1.1.0.2, parallel-1.1.0.0, parsec-2.1.0.1, parsec-3.0.0, soegtk-0.10.0, sourceview-0.10.0, svgcairo-0.10.0, time-1.1.2.3 [colin@susannah ghc]$ su -c 'cabal install parsec --global' Password: Warning: The package list for 'hackage.haskell.org' does not exist. Run 'cabal update' to download it. cabal: There is no package named parsec [colin@susannah ghc]$ su -c 'cabal install parsec-2.0.0 --global' Password: Warning: The package list for 'hackage.haskell.org' does not exist. Run 'cabal update' to download it. cabal: There is no package named parsec [colin@susannah ghc]$ cabal update Downloading the latest package list from hackage.haskell.org [colin@susannah ghc]$ su -c 'cabal install parsec-2.0.0 --global' Password: Warning: The package list for 'hackage.haskell.org' does not exist. Run 'cabal update' to download it. cabal: There is no package named parsec [colin@susannah ghc]$ cabal update --global unrecognized option `--global' Why does cabal install nearly everything in the user repository rather than the global repository? -- Colin Adams Preston Lancashire

4 11