Hi John!
On Sun, Aug 23, 2009 at 12:37:11PM +0100, John Cremona wrote:
2009/8/23 Nicolas M. Thiery <address@hidden>:
So you can focus on looking at the code, doc, and tests in the files
mentioned in:
http://trac.sagemath.org/sage_trac/wiki/CategoriesCategoriesReview
possibly simply by browsing:
http://combinat.sagemath.org/hgwebdir.cgi/code/file/tip/sage/categories)
and make sure they makes sense.
I don't really know how this type of reviewing differs from normal
Sage reviewing of patches attached to trac tickets.
Just that you don't need to worry about the technical side (checking
that the patch applies, pass test, ...). Also setting a positive
review will be on the wiki page, file by file, rather than on the trac
ticket. Finally, there is essentially no datastructure/algorithm issue
in the current code (that will change!)
But anyway, I looked at one file, finite_fields.py, and the method
_call_() in there looks wrong -- it raises an error whatever the
input, while the dicstring suggests that it is supposed to try using
__call__() and only change the error message if that does not work.
It's the converse: __call__ tries to do generic stuff (like C(P)
returns P if P is readilly a parent in the category C), and if that
does not work calls _call_ (similar to what happens for __mul__ /
_mul). I actually haven't changed that part; that's how it is in the
original category framework.
See the doc of _call_ / __call__ in Category (sage.category.category.py).
Suggestions to improve the doctests to clarify this are welcome!
If I have completely misunderstood what is going on, then I will
carry on ignoring the category activity until it is finished,
Please don't!
at which point I'm sure I will use it all the time!
Which is why I want as much "early" feedback from you as possible :-)
Cheers,
Nicolas
--
Nicolas M. Thiéry "Isil" <address@hidden>
http://Nicolas.Thiery.name/
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send an email to address@hidden
To unsubscribe from this group, send an email to address@hidden
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---