Rhamphoryncus <[EMAIL PROTECTED]> wrote: ... > > > I believe the prefered method to flatten a list of lists is this: > > > > > > shallow = [] > > > for i in deep: > > > shallow.extend(i) > > > > > > Yes, it's three lines. It's also very easy to read. reduce() and > > > sum() are not. > > > > I'd like to squeeze in the listcomp version, not because it is one line > > shorter, but because I, and maybe others prefer short listcomps such as > > the folowing: > > > > shallow = [] > > [shallow.extend(i) for i in deep] > > I'm sure this has been mentioned before, but listcomps are for when you > want to store the list and use it for further things, not for when you > want a side effect. TOOWTDI. > > And of course, if saving a line was the reason: > > shallow = [] > for i in deep: shallow.extend(i)
Another alternative equivalent to shallow = sum(deep, []) but based on the "proper" usage of list comprehensions is shallow = [ item for sublist in deep for item in sublist ] In terms of performance, however, the simple loop that you (rhamph) posted is generally best -- e.g., with Python 2.5c1 on a MacbookPro: brain:~/downloads alex$ python -mtimeit -s'deep=[range(9)]*9' 's=sum(deep,[])' 100000 loops, best of 3: 11.2 usec per loop brain:~/downloads alex$ python -mtimeit -s'deep=[range(9)]*9' 's=[] > for sublist in deep: s.extend(sublist)' 100000 loops, best of 3: 6.92 usec per loop brain:~/downloads alex$ python -mtimeit -s'deep=[range(9)]*9' 's=[] [s.extend(sublist) for sublist in deep]' 100000 loops, best of 3: 8.48 usec per loop brain:~/downloads alex$ python -mtimeit -s'deep=[range(9)]*9' 's=[item for sublist in deep for item in sublist]' 100000 loops, best of 3: 17.1 usec per loop Alex -- http://mail.python.org/mailman/listinfo/python-list
