Question

I find it strange that there is no built in function to just take the entire contents of an excel-produced csv file and just place them on a nested list so that it makes a 2D matrix.

Is there any reason why this is impractical/not standard?

I am putting my own CSV code together to do it, but I am already bumping into annoying hurdles wanting to do simple stuff like write columns instead of rows and whatnot.

Is there any reason I should consider a completely different approach?

the rationale is that using grids in, say, PyQt is so simple and it seems like the same kind of designing could translate seamlessly to reading and writing simple spreadsheets ie:

name , myname
age , myage
xvals, yvals
1,1
2,4
3,9

turns into

[[name,myname],[age,myage],[xvals,yvals],[1,1],[2,4],[3,9]]

Answer 1

Have a look at Numpy and its loadtxt function. It does exactly what you want :

http://docs.scipy.org/doc/numpy/reference/generated/numpy.loadtxt.html

and for writing, you can use the savetxt function :

http://docs.scipy.org/doc/numpy/reference/generated/numpy.savetxt.html

Also there is functions to read directly from Excel files :

http://pypi.python.org/pypi/xlrd/

http://packages.python.org/openpyxl/