In particular, this fixes problems on windows where project.load
may be passed a path from Jamfile, and then load-jamfile may use
a path using the exact case as stored in the file system -- different
from the path passed to project.load.
[SVN r59581]
In particular, this fixes problems on windows where project.load
may be passed a path from Jamfile, and then load-jamfile may use
a path using the exact case as stored in the file system -- different
from the path passed to project.load.
[SVN r59581]
Patch from Spencer E. Olson.
Here's the explanation from email:
The version of pathscale that I have access to is 3.2 and there doesn't seem
to be a pathf77 executable. The man pages do not reference it, and there is
also not a -f77 option that specifies that the code is fortran 77. Rather,
the man pages indicate that pathscale only recognizes the difference between
free form and fixed form fortran instead of F77 vs F90 vs F95.
Does anyone have an available installation that differs from this? If not,
I'll submit another small patch for changing this to using pathf90 to compile
f77 code--I've tested it and it appears to work well enough.
[SVN r59508]
Patch from Spencer E. Olson.
Here's the explanation from email:
The version of pathscale that I have access to is 3.2 and there doesn't seem
to be a pathf77 executable. The man pages do not reference it, and there is
also not a -f77 option that specifies that the code is fortran 77. Rather,
the man pages indicate that pathscale only recognizes the difference between
free form and fixed form fortran instead of F77 vs F90 vs F95.
Does anyone have an available installation that differs from this? If not,
I'll submit another small patch for changing this to using pathf90 to compile
f77 code--I've tested it and it appears to work well enough.
[SVN r59508]
This is mostly for completeness but also because I've occasionally noticed
output isn't flushed from a terminated process (when it is killed for exceeding
the time limit). I'm sure this won't break anything and am slightly hopeful
that it might ensure output is flushed before the terminating process is killed.
[SVN r58785]
This is mostly for completeness but also because I've occasionally noticed
output isn't flushed from a terminated process (when it is killed for exceeding
the time limit). I'm sure this won't break anything and am slightly hopeful
that it might ensure output is flushed before the terminating process is killed.
[SVN r58785]
