March 2007 - Hugs-Bugs

[Hugs] #49: LDFLAGS and hugs packages on OS X 10.4
by Hugs 06 Apr '07

06 Apr '07

#49: LDFLAGS and hugs packages on OS X 10.4 ----------------------------------+----------------------------------------- Reporter: guest | Owner: nobody Type: defect | Status: new Priority: minor | Milestone: Component: build system | Version: 200609 Keywords: LDFLAGS X11 packages | ----------------------------------+----------------------------------------- The value of the environnment variable LDFLAGS is not used to compile hug's packages. Consequently, make fails to build X11 package (on OS X 10.4.8) ; here is the make output : Preprocessing library X11-1.2... ../../libraries/tools/hsc2hs -D__HUGS__ -Iinclude -I/usr/X11R6/include --lflag=-lX11 -o Graphics/X11/Types.hs Graphics/X11/Types.hsc /usr/bin/ld: can't locate file for: -lX11 collect2: ld returned 1 exit status linking Graphics/X11/Types_hsc_make.o failed command was: /usr/bin/gcc -lX11 Graphics/X11/Types_hsc_make.o -o Graphics/X11/Types_hsc_make ../../packages/Cabal/examples/hapax.hs: got error code while preprocessing: Graphics.X11.Types Skipping X11 package My LDFLAGS variable whose value is "-L/usr/X11R6/lib -L/Users/julien/local/lib" has not been used. Simply add -L/usr/X11R6/lib to the gcc command and it compile fine. Another work around is to create a symbolic link : sudo ln -s /usr/X11R6/lib/libX11.a /usr/lib/libX11.a -- Ticket URL: <http://hackage.haskell.org/trac/hugs/ticket/49> Hugs <http://www.haskell.org/hugs/> Hugs 98, an interpreter for Haskell

1 3

[Hugs] #36: Statistics modes does not give wall time elapsed
by Hugs 06 Apr '07

06 Apr '07

#36: Statistics modes does not give wall time elapsed -------------------------+-------------------------------------------------- Reporter: neil | Owner: nobody Type: enhancement | Status: new Priority: minor | Milestone: Component: hugs | Version: 200605 Keywords: | -------------------------+-------------------------------------------------- When statistics are turned on, it gives reduction count, memory useage count, and garbage collection time. However, it does not give the amount of time that was actually consumed, which would be nice to have. -- Ticket URL: <http://hackage.haskell.org/trac/hugs/ticket/36> Hugs <http://www.haskell.org/hugs/> Hugs 98, an interpreter for Haskell

1 1

[Hugs] #61: [Patch] Readline command completion for everything
by Hugs 01 Apr '07

01 Apr '07

#61: [Patch] Readline command completion for everything -------------------------+-------------------------------------------------- Reporter: gsauthof | Owner: nobody Type: enhancement | Status: new Priority: major | Milestone: Component: hugs | Version: 200609 Keywords: | -------------------------+-------------------------------------------------- Hi, I wrote a patch that improves the readline command completion feature of hugs. I.e: - added completion of function names and constructors - added completion of command names (:<cmd>) - added completion of module names (e.g. after :browse , etc.) - only do filename completion where it makes sense (e.g. after :load , etc.) - expand use of tilde (~) in filenames (previously it was completed but not recognized by hugs) - read and write the history from/to ~/.hugs_history - make hugs recognize :help Patch is attached. Best regards Georg Sauthoff -- Ticket URL: <http://hackage.haskell.org/trac/hugs/ticket/61> Hugs <http://www.haskell.org/hugs/> Hugs 98, an interpreter for Haskell

1 2

System.Directory.getTemporaryDirectory broken on Windows
by Neil Mitchell 31 Mar '07

31 Mar '07

Hi, getTemporaryDirectory is defined as: getTemporaryDirectory :: IO FilePath getTemporaryDirectory = do #if defined(mingw32_HOST_OS) allocaBytes long_path_size $ \pPath -> do r <- c_GetTempPath (fromIntegral long_path_size) pPath peekCString pPath #else catch (getEnv "TMPDIR") (\ex -> return "/tmp") #endif However, taking a look at the generated files that ship with WinHugs Sep 2006, the code is the #else case - a catch and TMPDIR. This means that function doesn't work on Windows properly Any idea how this happened? I will try recompiling the libraries after Easter. Should mingw32_HOST_OS be defined with our build system on Windows? Thanks Neil

1 0

Re: [Hugs] #2: User's Guide entry for WinHugs
by Hugs 29 Mar '07

29 Mar '07

1 0

[Hugs] #51: The WinHugs uninstaller.exe doesn't work!
by Hugs 29 Mar '07

29 Mar '07

#51: The WinHugs uninstaller.exe doesn't work! -----------------------------+---------------------------------------------- Reporter: guest | Owner: neil Type: defect | Status: new Priority: major | Milestone: Component: winhugs | Version: current Keywords: setup uninstall | -----------------------------+---------------------------------------------- Hi guys! I was trying to uninstall WinHugs in order to install a newer version, and I found that the uninstaller.exe program doesn't work. I opened the uninstaller.exe with a hex editor, and I found that it's not a valid executable file. It's some weird combination of HTML and executable code. These are the first 15 lines of the uninstaller.exe file: {{{ <html> <head> <META HTTP-EQUIV="Content-Type" content="text/html; charset=Windows-1252"> </head> <body> <pre> <table width=100% bgcolor=#CFCFE5><tr> <td> <font face=arial size=+3> Build Log </font></table><table width=* cellspacing=0 cellpadding=0><tr><td width=0 bgcolor=#EDEDF5> </td><td width=0 bgcolor=#FFFFFF> </td><td width=*><pre> <h3>------- Build started: Project: Uninstaller, Configuration: Release|Win32 ------- </h3> </pre></table><table width=100% bgcolor=#DFDFE5><tr><td><font face=arial size=+2> Command Lines </font></table><table width=* cellspacing=0 cellpadding=0><tr><td width=0 bgcolor=#EDEDF5> </td><td width=0 bgcolor=#FFFFFF> </td><td width=*><pre>Uninstaller - up-to-date.</pre></table><table width=100% height=20 bgcolor=#CFCFE5><tr><td><font face=arial size=+2> </font></table></body></html> }}} I hope this is not difficult to fix. Greetings, -- L e a n d r o < c a f e s o l o [at] g m a i l [dot] c o m > -- Ticket URL: <http://hackage.haskell.org/trac/hugs/ticket/51> Hugs <http://www.haskell.org/hugs/> Hugs 98, an interpreter for Haskell

1 4

[Hugs] #31: Cnfigure script of Hugs98-May2006 source build doesn't check for useable sgml-tools installation
by Hugs 26 Mar '07

26 Mar '07

#31: Cnfigure script of Hugs98-May2006 source build doesn't check for useable sgml-tools installation --------------------+------------------------------------------------------- Reporter: guest | Owner: nobody Type: defect | Status: new Priority: minor | Milestone: Component: hugs | Version: 200605 Keywords: | --------------------+------------------------------------------------------- I've been unable completely to build and install Hugs98 May2006 on Slackware Linux 10.1, thus: $ tar xvfz hugs98-plus-May2006.tar.gz $ cd hugs98-plus-May2006 $ ./configure --prefix=/usr/local/pkg/hugs98-plus.May2006 $ make because of the following build errors: ... mkdir users_guide /bin/cp hugs-ug.css users_guide /bin/cp winhugs.png users_guide test -z 'docbook2html' || \ docbook2html --dsl hugs-ug.dsl users_guide.sgml Using catalogs: /etc/sgml/catalog Using stylesheet: /local/d0p6/john/hugs98-plus-May2006/docs/users_guide /hugs-ug.dsl Working on: /local/d0p6/john/hugs98-plus- May2006/docs/users_guide/users_guide.sgml jade:/local/d0p6/john/hugs98-plus- May2006/docs/users_guide/users_guide.sgml:14:0:E: URL not supported by this version jade:/local/d0p6/john/hugs98-plus- May2006/docs/users_guide/users_guide.sgml:14:0:E: DTD did not contain element declaration for document type name jade:/local/d0p6/john/hugs98-plus- May2006/docs/users_guide/users_guide.sgml:15:9:E: there is no attribute "ID" jade:/local/d0p6/john/hugs98-plus- May2006/docs/users_guide/users_guide.sgml:15:22:E: element "BOOK" undefined jade:/local/d0p6/john/hugs98-plus- May2006/docs/users_guide/users_guide.sgml:16:9:E: element "BOOKINFO" undefined ... docbook2html comes from package sgml-tools, for which we have version 1.0.9. Shouldn't the hugs98 configure script check that the available versions of docbook2html and the commands that uses in turn are suitable (a sufficiently late version) to complete the Hugs98 build? For the minute, I'll just disable the build of the documentation, but this is disappointing. John A. Murdie (john(a)cs.york.ac.uk) Department of Computer Science University of York -- Ticket URL: <http://hackage.haskell.org/trac/hugs/ticket/31> Hugs <http://www.haskell.org/hugs/> Hugs 98, an interpreter for Haskell

3 6

0/0 => "nan", not "NaN". Also, empty hsc2hs dir and "invalid predicate `-wholename'"
by Ian Lynagh 23 Mar '07

23 Mar '07

Hi all, hugs thinks that show (0/0 :: Double) is "nan", but it will only read "NaN". I'd suggest changing the behaviour of show rather than that of read, so as to match GHC: Prelude> show (0/0 :: Double) "NaN" Prelude> read "nan" :: Double *** Exception: Prelude.read: no parse Prelude> read "NaN" :: Double NaN The result of the show seeme to come from sprintf in doubleToString (src/machdep.c); this C program also prints "nan": #include <stdio.h> int main(void) { char b[100]; sprintf(b, "%e", (double) 0/0); printf("%s\n", b); return 0; } While building CVS/darcs hugs I got darcs get --partial http://darcs.haskell.org/hsc2hs Directory '/home/ian/code/hugs/hugs98/hsc2hs' already exists, creating repository as '/home/ian/code/hugs/hugs98/hsc2hs_0' I think due to an empty CVS directory; I had to jiggle them around for it to build later on. I also got /usr/bin/find: invalid predicate `-wholename' thrice. Replacing -wholename with -path in libraries/tools/make_bootlib let the build go through, but I'm not sure if this is either (a) correct or (b) a superset of the portability of -wholename. Thanks Ian

3 3

copyFile on Windows dies with binary files
by Neil Mitchell 22 Mar '07

22 Mar '07

Hi, copyFile on Windows with a binary file doesn't work terribly well - it truncates at the first binary zero. The code goes: copyFile :: FilePath -> FilePath -> IO () copyFile fromFPath toFPath = do readFile fromFPath >>= writeFile toFPath try (copyPermissions fromFPath toFPath) return () That should be openFileBinary etc. Thanks Neil

1 0

[Hugs] #32: Compile of hugs98-plus-May2006.tar.gz fails on SPARC Solaris 8
by Hugs 21 Mar '07

21 Mar '07

#32: Compile of hugs98-plus-May2006.tar.gz fails on SPARC Solaris 8 --------------------+------------------------------------------------------- Reporter: guest | Owner: nobody Type: defect | Status: new Priority: major | Milestone: Component: hugs | Version: 200605 Keywords: | --------------------+------------------------------------------------------- My machine: {{{ straumli> uname -a SunOS straumli 5.8 Generic_117350-28 sun4u sparc SUNW,Sun-Blade-100 straumli> }}} After configuring (WTF is up with that "ultra-evil" comment anyway?), I compile, and get: {{{ straumli> make cd src; make all make[1]: Entering directory `/users/stremler/Work/Haskell/hugs98-plus- May2006/src' gcc -c -DNDEBUG=1 -g -O2 hugs.c gcc -c -DNDEBUG=1 -g -O2 edit.c gcc -c -DNDEBUG=1 -g -O2 observe.c gcc -c -DNDEBUG=1 -g -O2 builtin.c builtin.c: In function `mkThunk': builtin.c:2151: error: `pc' undeclared (first use in this function) builtin.c:2151: error: (Each undeclared identifier is reported only once builtin.c:2151: error: for each function it appears in.) make[1]: *** [builtin.o] Error 1 make[1]: Leaving directory `/users/stremler/Work/Haskell/hugs98-plus- May2006/src' make: *** [all] Error 2 straumli> }}} I eventually got it to compile: {{{ straumli> diff hugs98-plus-May2006/src/builtin.c hugs98-plus-May2006- modified/src/builtin.c 2151c2151,2152 < unsigned long *adj_code = (unsigned long *)pc; --- > /*unsigned long *adj_code = (unsigned long *)pc;*/ > unsigned long *adj_code = (unsigned long *)thunk->code; straumli> }}} Dunno if that's the _right_ fix. But it seems to work enough for HelloWorld.hs... hugs98-May06-bugreport (at) stremler (dot) net -- Ticket URL: <http://hackage.haskell.org/trac/hugs/ticket/32> Hugs <http://www.haskell.org/hugs/> Hugs 98, an interpreter for Haskell

1 2