#8201: Fortran not found on Linux if gfortran not present
----------------------+-----------------------------------------------------
Reporter: mraum | Owner: Martin Raum
Type: defect | Status: closed
Priority: critical | Milestone: sage-4.3.4
Component: build | Resolution: wontfix
Keywords: fortran | Author:
Upstream: N/A | Reviewer: David Kirkby
Merged: | Work_issues: A symoblic link should solve this
persons very unique and unusual circumstances.
----------------------+-----------------------------------------------------
Changes (by drkirkby):
* status: needs_review => closed
* reviewer: => David Kirkby
* resolution: => wontfix
* work_issues: => A symoblic link should solve this persons very unique
and unusual circumstances.
Comment:
Replying to [comment:3 mraum]:
> SAGE_FORTRAN is ignored, if gfortran doesn't exist.
> Namely, the script will abort. That's why I had to create a patch.
Creating links for gfortran is a hack would most probably create new
problems anywhere else at another time.
What makes you think a symbolic link is a hack? It is very standard
practice on Unix systems to create symbolic links for this very purpose.
There are loads of them created in Sage. (Some people call them soft
links).
Many shell scripts start
{{{
#!/bin/sh
}}}
But most linux distros do not have a program called /bin/sh. Rather
/bin/sh is simply a link to bash, dash or some other shell. On sage.math
{{{
kir...@sage:~$ ls -l /bin/sh
lrwxrwxrwx 1 root root 4 2010-02-02 13:49 /bin/sh -> dash
kir...@sage:~$ ls -l /bin/dash
-rwxr-xr-x 1 root root 100856 2009-03-09 06:18 /bin/dash
}}}
When you type 'sh' on sage.math you are really running 'dash'
> Moreover, I don't have admin rights.
You don't need admin rights to create a symbolic link. Something like:
{{{
$ ln -s /usr/local/bin/gfortran-4.2 $HOME/bin/gfortran
}}}
will do fine. Then make sure $HOME/bin is in your path before
/usr/local/bin. Your will then have 'gfortran'.
> I don't have access to Solaris; And I don't have experience with it. But
if the GNU suit is installed it should work; if not it won't, but scipy
has to blamed for this not fortran itself.
Do you have an account on sage.math or similar? If so, you have an account
on 't2' - or if not, one is easily created. For all practical purposes you
will not find it significantly different from Linux.
> If you can test this on Soloaris I would be happy to integrate any
changes you suggest.
Unfortunately, I simply do not have time to test everyones changes on
Solaris - let alone generate patches if they do not work.
http://www.sagemath.org/doc/developer/inclusion.html
makes it clear that for a package to be included in Sage, it must support
Solaris.
'''Some Sage developers are willing to help you port to OS X, Solaris and
Windows. But this is no guarantee and you or your project are expected to
do the heavy lifting and also support those ports upstream if there is no
Sage developer who is willing to share the burden.'''
Your solution seems like a ''sledgehammer to crack a walnut'' if you know
what I mean by that. A symbolic link has far greater chance of just
working, without risking breaking things for everyone else. In stead you
propose making changes that could impact anyone, without actually testing
them.
'''I'm marking this as wontfix, as I believe the patch is unnecessary. You
may find another reviewer see this differently, though I doubt you will
find any experienced system administrator see it that way'''
Dave
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/8201#comment:4>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/sage-trac?hl=en.
