"groupby" is brilliant!

[Alex Martelli]

Frank Millman <frank at chagford.com> wrote:

> Benji York wrote:
> > Frank Millman wrote:
> > > reader = csv.reader(open('trans.csv', 'rb'))
> > > rows = []
> > > for row in reader:
> > >     rows.append(row)
> >
> > Why do you create a list of rows instead of just iterating over the
> > reader directly?
> > --
> > Benji York
> 
> A - didn't think of it - good idea
> 
> B - can't always do it -
> B1 - if the file is not sorted, I have to sort the rows first
> B2 - if I need to update the file, I can modify the rows in place, and
> then call
>         csv.writer(open('trans.csv','wb')).writerows(rows)
> 
> BTW, I know that B2 is simplistic - to be safe I should rename, then
> write, then unlink. I will do that for production code.
> 
> BTW2, an alternative to B2 is
>     reader = csv.reader(open('trans.csv', 'rb'))
>     newtrans = open('newtrans.csv','wb')
>     writer = csv.writer(newtrans)
>     for row in reader:
>         [process and modify row]
>         writer.writerow(row)
>     newtrans.close()
>     [unlink and rename]
> 
> Could be useful if the file is large. Food for thought.

BTW, if and when you do need a list for some other purpose,

rows = list(reader)

may be slightly better than the for/append loop; and if you need a
sorted list, perhaps

rows = sorted(reader)

similarly.


Alex