headers
Alexander Hansen | 2 May 20:33
Picon
Gravatar

SBCL-1.0.25/Intel on the tracker

I just tested out sbcl-1.0.25 for Intel

http://sourceforge.net/tracker/?func=detail&aid=2780107&group_id=17203&atid=414256

on 10.5.6.  It looks to have built all right--though I've not done any
exhaustive testing.  I have a few issues to resolve before committing
the package:

1)  The PowerPC version is out of sync, so I've cc'ed that maintainer. 
Since maxima is now intimately tied to sbcl, having separate versions
for the two architectures means that I would have to do that for maxima
(which could be done, but I'd rather not).
2)  We've had build differences (as in failures) with 10.4 vs. 10.5
before, and since I don't have a 10.4/Intel test box, I'd like to have a
report from somebody with access to that platform.
3)  I'd like to roll the current sbcl, maxima, and wxmaxima packages
over to the stable tree, if there are no objections.

--

-- 
Alexander Hansen
Fink User Liaison

------------------------------------------------------------------------------
Register Now & Save for Velocity, the Web Performance & Operations 
Conference from O'Reilly Media. Velocity features a full day of 
expert-led, hands-on workshops and two days of sessions from industry 
leaders in dedicated Performance & Operations tracks. Use code vel09scf 
and Save an extra 15% before 5/3. http://p.sf.net/sfu/velocityconf
Permalink | Reply | 2 comments |
headers
Jack Howarth | 4 May 06:50
Picon

paraview(-mpi)-3.4.0-1000 packaging

Martin,
   I've updated the revised paraview(-mpi) package on...

https://sourceforge.net/tracker/?func=detail&aid=1747995&group_id=17203&atid=414256

to the latest 3.4.0 release using the Fedora 11 patches for building
against qt 4.5.1. The paraview-3.4.0-1000 packaging passes 'fink -m --build-as-nobody'
for the build and the resulting binaries work fine under x86_64 fink. I still need
to do a test build of the matching paraview-mpi packaging to add the missing additional
shlibs entries. Once that is fixed can we get this into unstable as the current paraview
packaging is based on an ancient release of paraview?
              Jack

------------------------------------------------------------------------------
Register Now & Save for Velocity, the Web Performance & Operations 
Conference from O'Reilly Media. Velocity features a full day of 
expert-led, hands-on workshops and two days of sessions from industry 
leaders in dedicated Performance & Operations tracks. Use code vel09scf 
and Save an extra 15% before 5/3. http://p.sf.net/sfu/velocityconf
Permalink | Reply | 2 comments |
headers
Martin Costabel | 4 May 08:59
Picon

Re: paraview(-mpi)-3.4.0-1000 packaging

Jack Howarth wrote:
> Martin,
>    I've updated the revised paraview(-mpi) package on...
> 
> https://sourceforge.net/tracker/?func=detail&aid=1747995&group_id=17203&atid=414256
> 
> to the latest 3.4.0 release using the Fedora 11 patches for building
> against qt 4.5.1. The paraview-3.4.0-1000 packaging passes 'fink -m --build-as-nobody'
> for the build and the resulting binaries work fine under x86_64 fink. I still need
> to do a test build of the matching paraview-mpi packaging to add the missing additional
> shlibs entries. Once that is fixed can we get this into unstable as the current paraview
> packaging is based on an ancient release of paraview?
>               Jack

I have currently no time to look into this, but I have no objection if 
you want to take over the maintainership of paraview.

--

-- 
Martin

------------------------------------------------------------------------------
Register Now & Save for Velocity, the Web Performance & Operations 
Conference from O'Reilly Media. Velocity features a full day of 
expert-led, hands-on workshops and two days of sessions from industry 
leaders in dedicated Performance & Operations tracks. Use code vel09scf 
and Save an extra 15% before 5/3. http://p.sf.net/sfu/velocityconf
Permalink | Reply | 1 comment |
headers
Jack Howarth | 4 May 14:52
Picon

Re: paraview(-mpi)-3.4.0-1000 packaging

On Mon, May 04, 2009 at 08:59:35AM +0200, Martin Costabel wrote:
> Jack Howarth wrote:
>> Martin,
>>    I've updated the revised paraview(-mpi) package on...
>>
>> https://sourceforge.net/tracker/?func=detail&aid=1747995&group_id=17203&atid=414256
>>
>> to the latest 3.4.0 release using the Fedora 11 patches for building
>> against qt 4.5.1. The paraview-3.4.0-1000 packaging passes 'fink -m --build-as-nobody'
>> for the build and the resulting binaries work fine under x86_64 fink. I still need
>> to do a test build of the matching paraview-mpi packaging to add the missing additional
>> shlibs entries. Once that is fixed can we get this into unstable as the current paraview
>> packaging is based on an ancient release of paraview?
>>               Jack
>
> I have currently no time to look into this, but I have no objection if  
> you want to take over the maintainership of paraview.
>
> -- 
> Martin

Martin,
   I'd be happy to take over paraview. I still need to do some test
builds of the paraview-mpi packaging and then I'll update it in
unstable.
               Jack

------------------------------------------------------------------------------
Register Now & Save for Velocity, the Web Performance & Operations 
Conference from O'Reilly Media. Velocity features a full day of
(Continue reading)

Permalink | Reply | 0 comments |
headers
Jack Howarth | 5 May 01:16
Picon

fink 0.29.2-41 breaks gcc44

 The latest fink 0.29.2-41 release breaks the ability to
build gcc44. While it does eliminate the bogus validation
error on gcc44.info, the actual packaging of gcc44 on intel
darwin now fails with...

/bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libgfortran.3*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
/bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libffi.4*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
/bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libstdc++.6*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
/bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libgcc_s.1*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
/bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libgcj.10*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
/bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libgcj-tools.10*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
/bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libgij.10*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
/bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libssp.0*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
/bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libobjc-gnu.2*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
/bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libgomp.1*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
/bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/gcj-4.4.0-10 /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
/bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libgfortran.3*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
mv: rename /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libgfortran.3*.dylib to
/sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/libgfortran.3*.dylib: No
such file or directory
### execution of /bin/mv failed, exit code 1
Removing runtime build-lock...
Removing build-lock package...
/sw/bin/dpkg-lockwait -r fink-buildlock-gcc44-4.4.0-1000

Apparently the %lib substitution is no longer being honored for....

    (%m != x86_64) %p/lib/gcc4.4/%lib/libgfortran.3.dylib 4.0.0 %n (>= 4.3-20070711) 64

Can we get this fixed soon?
(Continue reading)

Permalink | Reply | 1 comment |
headers
Daniel Macks | 5 May 02:09

Re: fink 0.29.2-41 breaks gcc44

Not sure what the val error had been, but last night I corrected a
validation problem in gcc44.info in unstable that was an actual real
problem. Check cvs log for details.

dan

On Mon, May 04, 2009 at 07:16:53PM -0400, Jack Howarth wrote:
>  The latest fink 0.29.2-41 release breaks the ability to
> build gcc44. While it does eliminate the bogus validation
> error on gcc44.info, the actual packaging of gcc44 on intel
> darwin now fails with...
> 
> /bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libgfortran.3*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
> /bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libffi.4*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
> /bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libstdc++.6*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
> /bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libgcc_s.1*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
> /bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libgcj.10*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
> /bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libgcj-tools.10*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
> /bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libgij.10*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
> /bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libssp.0*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
> /bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libobjc-gnu.2*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
> /bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libgomp.1*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
> /bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/gcj-4.4.0-10 /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
> /bin/mv /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libgfortran.3*.dylib /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/
> mv: rename /sw/src/fink.build/root-gcc44-4.4.0-1000/sw/lib/gcc4.4/lib/libgfortran.3*.dylib
to /sw/src/fink.build/root-gcc44-shlibs-4.4.0-1000/sw/lib/gcc4.4/lib/libgfortran.3*.dylib:
No such file or directory
> ### execution of /bin/mv failed, exit code 1
> Removing runtime build-lock...
> Removing build-lock package...
(Continue reading)

Permalink | Reply | 0 comments |
headers
Jack Howarth | 5 May 03:26
Picon

Re: fink 0.29.2-41 breaks gcc44

Daniel,
    You might check with David Morrison on these changes you are making.
He helped craft the current gcc44.info (before you messed with it) so
that it would allow for a 32-bit multilib under x86_64 fink. David said
that the validation errors in gcc44.info were bogus and that he would
look into fixing fink itself (I assume the errors you are referring to
were duplicate packages and splitoffs being detected).

Error: Duplicate declaration of package "gcc44". (gcc44.info)
Error: Duplicate declaration of package "gcc44-shlibs". (gcc44.info)

               Jack
ps I'll have to double check that you haven't broken the x86_64 fink 
build of gcc44 (sigh).

------------------------------------------------------------------------------
The NEW KODAK i700 Series Scanners deliver under ANY circumstances! Your
production scanning environment may not be a perfect world - but thanks to
Kodak, there's a perfect scanner to get the job done! With the NEW KODAK i700
Series Scanner you'll get full speed at 300 dpi even with all image 
processing features enabled. http://p.sf.net/sfu/kodak-com
Permalink | Reply | 501 comments |
headers
Charles Lepple | 5 May 04:15
Picon

Re: abcde on mac intel

On Wed, Apr 29, 2009 at 5:20 AM, Mario Frasca <mariotomo <at> inventati.org> wrote:
> it complains because some outputs are redirected appending a $REDIR to
> the line, where
> REDIR=">&2"
> (at line 2693)

Is there a URL for the Debian version of the script, or is it buried
in a tarball and patch set?

Also, what does the #! line (first line) look like?

--

-- 
- Charles Lepple

------------------------------------------------------------------------------
The NEW KODAK i700 Series Scanners deliver under ANY circumstances! Your
production scanning environment may not be a perfect world - but thanks to
Kodak, there's a perfect scanner to get the job done! With the NEW KODAK i700
Series Scanner you'll get full speed at 300 dpi even with all image 
processing features enabled. http://p.sf.net/sfu/kodak-com
Permalink | Reply | 4 comments |
headers
Daniel Macks | 5 May 03:31

Re: fink 0.29.2-41 breaks gcc44

Yup, he and I are working on it.

dan

On Mon, May 04, 2009 at 09:26:49PM -0400, Jack Howarth wrote:
> Daniel,
>     You might check with David Morrison on these changes you are making.
> He helped craft the current gcc44.info (before you messed with it) so
> that it would allow for a 32-bit multilib under x86_64 fink. David said
> that the validation errors in gcc44.info were bogus and that he would
> look into fixing fink itself (I assume the errors you are referring to
> were duplicate packages and splitoffs being detected).
> 
> Error: Duplicate declaration of package "gcc44". (gcc44.info)
> Error: Duplicate declaration of package "gcc44-shlibs". (gcc44.info)
> 
>                Jack
> ps I'll have to double check that you haven't broken the x86_64 fink 
> build of gcc44 (sigh).
> 
> ------------------------------------------------------------------------------
> The NEW KODAK i700 Series Scanners deliver under ANY circumstances! Your
> production scanning environment may not be a perfect world - but thanks to
> Kodak, there's a perfect scanner to get the job done! With the NEW KODAK i700
> Series Scanner you'll get full speed at 300 dpi even with all image 
> processing features enabled. http://p.sf.net/sfu/kodak-com
> _______________________________________________
> Fink-devel mailing list
> Fink-devel <at> lists.sourceforge.net
> http://news.gmane.org/gmane.os.apple.fink.devel
(Continue reading)

Permalink | Reply | 501 comments |
headers
Martin Costabel | 5 May 09:25
Picon

Re: abcde on mac intel

Charles Lepple wrote:
> On Wed, Apr 29, 2009 at 5:20 AM, Mario Frasca <mariotomo <at> inventati.org> wrote:
>> it complains because some outputs are redirected appending a $REDIR to
>> the line, where
>> REDIR=">&2"
>> (at line 2693)

The difference between the two versions is that they are two different 
versions: Fink has abcde-2.2.6, whereas debian has 2.3.99. The script 
has been much reworked in the meantime, so it is not very curious that 
the output is rather different. OTOH, those $REDIRECT constructs are 
still present in 2.3.99, and I am not sure why this then works for Mario.

>  Is there a URL for the Debian version of the script, or is it buried
>  in a tarball and patch set?

<http://http.us.debian.org/debian/pool/main/a/abcde/abcde_2.3.99.6.orig.tar.gz>
The script is on the surface of the source tarball.

> Also, what does the #! line (first line) look like?

#!/bin/sh for the Fink version, #!/bin/bash for 2.3.99, but I would be 
surprised if this made a difference. A little test shows that the $REDIR 
construct does not work. It appears that Apple's bash/sh does not allow 
to store redirect commands in a variable. According to man bash, under 
"SIMPLE COMMAND EXPANSION", redirect commands are "saved for later 
processing" before any variables are expanded.

   ls . asdf 2>file >&2
works as expected (i.e. writes both stdout and stderr into "file"), but
(Continue reading)

Permalink | Reply | 2 comments |

Gmane