Omar wrote:
> I'm looking for a programming language or module that sorta looks and
> feels like MS Excel (I love and think in tables), yet has the power and
> open-endedness of python or javascript. I'm still pretty new to
> python.
>
> any ideas? i've been having some fun with VBA in excel, but I want
> something I can save as en exe and call my own creation, y'know?
>
More than the implementation, I would be curious about the API you (or
anyone else) might envision. I have spent a lot of time making a "Table"
class over about the last year and a half, but I'm not sure what might
be an intuitive interface for most people. First, I think it should work
like a "sorted" dictionary of lists, but, at the same time, a list of
sorted dictionaries. I also want *shorthand* for selection.
For example, does the output below look like an intuitive interface? Or,
more likely, how many people get squeamish when they see this interface?
Do these squeamish people have any better ideas? This is a taste of how
my Table class currently behaves:
py> print t # dependent on its property t.format
Last First Age
Barker Bob 204
Burnet Carol 64
Danson Ted 54
Cooper Alice 78
py> t.headings()
("Last", "First", "Age")
py> t.get_row(1) # approximately equal clarity with 1d slice
['Burnet', 'Carol', 64]
py> t[1] # implicit selection of "first" dimension
['Burnet', 'Carol', 64]
py> t.get_column('Last') # probably clearer than taking a 1d slice
['Barker', 'Burnet', 'Danson', 'Cooper']
py> # the following is probably the trickiest, should it return a Table
py> # should it be illegal?
py> # should t['Last'] be the way to take the "slice" and get the col?
py> t[None, 'Last'] # 1d slice returns list (2nd dim. explicit)
['Barker', 'Burnet', 'Danson', 'Cooper']
py> t2 = t[1:3, ('First', 'Age')] # 2d slice returns a new Table
py> t2
<__main__.Table instance at 0x404f676c>
py> t2.format
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
AttributeError: 't2' object has no attribute 'format'
py> t2.format = "%10s %4d"
py> print t2
First Age
Carol 64
Ted 54
py> t3 = t[1:3,'First':'Age'] # shorthand to take a swath of columns
py> t3.format = "%10s %4d"
py> print t3
First Age
Carol 64
Ted 54
py> t3 = t[1:3, 0:2] # if we know what column numbers we want instead
py> t3.format = "%10s%10s"
py> print t3
Last First
Burnet Carol
Danson Ted
These latter selections might turn some heads, especially for those who
think a a little too deeply about dictionaries. But, unlike a
dictionary, a Table is ordered. If you disagree with the assumption that
tables are ordered, make a table in excel, save it, close it, open it
again and see that the columns are still in the same order. Do this 1000
times to convince yourself. The table will be in the same order every
time! Amazing!
The idea is that tables are not dictionaries--please resist drawing
paralells. If we strangle ourselves with operating under the dictionary
paradigm, we will never get a Table. Trust me, I've thought about it
more than most. (If you want a Table to act like a dictionary, I can
make a RandomTable class just for you.)
Notice also that slicing can be done with a tuple or a list, so type
checking is done in the implementation--but lets forget about our
notions of impementation and how type checking is "bad". I want to get
feedback on the API here without regard to how it might be implemented.
In a Table, the first dimension is row and the second is column (i.e.
the data is organized by rows). My Table class has a lot of additional
functionality. But, I'm curious about how people see such a beast
working, though. Please don't criticize unless you have a better idea
about the API of a Table. I want to hear genuine and concrete ideas and
not abstruse pontification about programming or design! Statements of "I
wouldn't do this thing here" should be immediately followed by
"--rather, I would do this other thing for which I've created a concrete
example below."
If you think this interface is genius, well my ego wants to hear about
that as well, but its not terribly necessary.
James
--
James Stroud
UCLA-DOE Institute for Genomics and Proteomics
Box 951570
Los Angeles, CA 90095
http://www.jamesstroud.com/
--
http://mail.python.org/mailman/listinfo/python-list
