Bug#550823: Updating load-path cache based on modification times probably a bad idea (was: Race condition between Octave 3.2.3 and unlink())

2009-10-21 Thread Judd Storrs
I'm still looking to see if there's a generic way to determine the file-system time resolution. I have seen stated multiple places that FAT/FAT32 is limited to a time resolution of 2 seconds. For example http://docs.python.org/library/os.html#os.stat says: Note The exact meaning and resolution of

Bug#550823: Updating load-path cache based on modification times probably a bad idea (was: Race condition between Octave 3.2.3 and unlink())

2009-10-20 Thread Judd Storrs
If we're already keeping track of when we last read a directory/file, wouldn't it be easiest to make the comparison know about the tolerance? i.e. use something equivalent to if modtime + tolerance > cachetime reparse endif Then when the file/directory is older than the tolerance full caching

Bug#409849: First rough draft of an afni Debian package

2009-10-20 Thread Judd Storrs
> > The place is fine, but how did you determine the subset of headers that > should go into it. If we can automize that, it might scale a bit better > with the future AFNI development. > However, the more serious problem is that having a -dev package also > implies that the libs can be used else

Bug#409849: First rough draft of an afni Debian package

2009-10-19 Thread Judd Storrs
Excellent work! I have two updates. The first adds an afni-dev package to include headers needed to use libmri etc as well as to compile AFNI plugins. I wasn't sure where these should go--this places them inside /usr/include/afni. The second patch allows either openmotif or lesstif to be used. A