On Jun 14, 2010, at 5:35 AM, Dr. David Kirkby wrote:
On 06/14/10 01:17 PM, Dr. David Kirkby wrote:
On 06/14/10 12:18 PM, Tim Joseph Dumol wrote:
As for Cython and gcc, the Sage notebook uses pure Python. I do
acknowledge that there's a minuscule chance that a Python update
could
change runtime behaviour.
But what is used to build python? - gcc of course! So we have *at
least*
the following possibilities which could result in a problem.
* gcc update
* python update
* someone patching python (it is already at patch level 8 or so in
Sage)
* operating system update
* port to another platform (Cygwin, OpenSolaris and FreeBSD are all
being worked on.)
* someone's computer may be mis-configured.
I should have also considered anything that is needed to build
python, could change the behavior of python. That includes
* zlib - which will be updated in 4.4.4.alpha1
* termcap
* readline
* sqlite
* gnutls
* libpng
Some I accept are unlikely, but none are impossible.
I have yet to see any Python code that behaves differently on
different platforms except for code using platform dependent features,
numerical noise (and even then only on Sparc), or external recourses
(including timing). It's possible, but there's much more fruitful
things to test.
Is there *any* other motivation for removing this from the testing,
apart from increasing the percentage of doctest coverage?
That I feel is the most important question.
If not, it boils down to sacrificing quality for better looking
statistics.
Which is what I think we are doing.
It's not just about better looking statistics, it's also about more
accurately reflecting the state of the codebase.
To clarify, in this case the reason that code needs to be there is
just to unpickle old notebooks, meaning that there needs to be a class
of a specific name in a specific module (again looked up by name) or
the old pickle won't know how to reconstruct itself. Most (all?) of
the actual code there could probably be deleted without affecting
unpickleability (todo once we have good tests of that).
- Robert
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-devel+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
