[PYTHON MATRIX-SIG] Numeric Python 1.0 Beta 2 is available

[Geoffrey Furnish]

Konrad Hinsen writes:
 > > Again, I regard this as a central shortcoming of Python.  The system
 > > used in TeX/LaTeX seems a lot better to me, where directory search
 > > paths are specified effectively as being "rooted at" some place, and
 > > all subdirectories of that are searched.  If python worked like that,
 > 
 > Funny to see that suggestion -- I was about to suggest the same in my
 > last message, but deleted the sentence after I had half written it.
 > I had suddenly realized that I would definitely not want Python
 > to go through the subdirectories in my personal module collection,
 > which includes subdirectories such as "old code to be kept for a while",
 > "untested code by others" etc.

I'm not really sure if this objection is very strong, I think it is
probably not.  Suppose you have a python installed with
--prefix=/over/there.  Somewhere else in your filespace, you are
developing your foobar package.  You archive your old stuff under
/here/foobar/oldstuff, not under /over/there/foobar/oldstuff.  I mean
really, who develops in an install tree?  Certainly not me.  But the
advantage of the TeX kpathsea business is that you can install a
package under /over/there/foobar and have no trouble with the files
beeing seen via the search mechanism of the previously compiled
executable.  

I maintian that this would be a huge win for python extension
development. 

-- 
Geoffrey Furnish		email: furnish@llnl.gov
LLNL X/ICF			phone: 510-424-4227	fax: 510-423-0925

_______________
MATRIX-SIG  - SIG on Matrix Math for Python

send messages to: matrix-sig@python.org
administrivia to: matrix-sig-request@python.org
_______________