Hi List,
I'm in the process of polishing the API for the forthcoming PyTables 2.2. So,
I'll appreciate your feedback in a couple of improvements that implies some
changes in the API:
1) Many people got confused on how the `Row` iterator works. I think it all
boils down to the fact that `Row.__str__` gives a representation of the *data*
pointed to the row, and not the row object itself. So, what people would
think if I change `Row.__str__` behaviour so that:
"""
for row in table:
print row
"""
will yield:
"""
Row accessor pointing to row #0
Row accessor pointing to row #1
Row accessor pointing to row #2
...
"""
instead of current:
"""
(row0_field0_value, row0_field1_value,...)
(row1_field0_value, row1_field1_value,...)
(row2_field0_value, row2_field1_value,...)
...
"""
2) In the process of implementing HDF5 links (only available when compiled
with HDF5 1.8 series), I need to create another `UnImplemented`-like class
(for PyTables compiled against HDF5 1.6 series). Initially, I have thought in
something like `UnImplementedLink`, but then, there is an asymmetry between
current `UnImplemented`, which is meant for unsupported datasets, and new
`UnImplementedLink`, which is meant for unsupported links. So, my plan is to
rename `UnImplemented` to `UnImplementedLeaf` so that, together with
`UnImplementedLink`, they form the two kinds of unsupported nodes. I'm even
thinking that two short names like 'UILeaf` and `UILink` would be handier.
As you can see, both proposals are important: 1) represents a semantic change
in `Row.__str__` special method, while 2) is a change in the name of a not
very widely used (although perhaps I'm wrong here) class like `UnImplemented`.
At the moment, I'm inclined to *do* these changes, but I'd be glad to hear
your opinion first ;-)
--
Francesc Alted
------------------------------------------------------------------------------
Return on Information:
Google Enterprise Search pays you back
Get the facts.
http://p.sf.net/sfu/google-dev2dev
_______________________________________________
Pytables-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/pytables-users