On Nov 5, 10:51 pm, Ralf-Philipp Weinmann <[EMAIL PROTECTED]
darmstadt.de> wrote:
> Hello list,
Hello Ralf,
>
> I've finished building and testing SAGE 2.8.11 (on ppc)
>
> MACOSX_DEPLOYMENT_TARGET=10.3
>
> This causes a build produced on OSX 10.5 to _almost_ work on 10.4. The
> culprit that forces me to use the "almost" in the above sentence is
> maxima, which has been linked against the system libiconv of 10.5 and
> wants *at least this version* libiconv of which only an older version
> is present on 10.4). I've temporarily worked around this issue by
> installing libiconv through MacPorts and setting DYLD_LIBRARY_PATH
> accordingly (to /opt/local/lib in my case. DYLD_LIBRARY_PATH is the
> equivalent to LD_LIBRARY_PATH on other Unices). A "make check" then
> causes the test cases (bar one, see below) to successfully finish.
>
> The 10.4 SDK (usually) installed in /Developer/SDKs/MacOSX10.4u.sdk on
> 10.5 systems with XCode 3 contains the appropriate (older) library,
> but I haven't had enough time to figure out whether we can have maxima
> link against the old library or whether we need to ship libiconv with
> SAGE to get a build that works flawlessly on both platforms. I tend to
> prefer the latter option at the moment.
>
> * All my tests so far have shown that devel/sage-main/sage/combinat/
> sfa.py fails with a SIGSEGV (both on ppc and Intel, both on 10.5 and
> 10.4). Is this a known problem? I haven't been able to find any ticket
> about this in the TRAC.
We knew about it, but it didn't have its own ticket. It is #1108 now.
>
> Cheers,
> Ralf
Cheers,
Michael
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~----------~----~----~----~------~----~------~--~---
