Velocity Reviews > delete items from list by indices

# delete items from list by indices

blumenkraft
Guest
Posts: n/a

 09-23-2009
Hi,

I have some list:
x = [8, 9, 1, 7]
and list of indices I want to delete from x:
indices_to_delete = [0, 3], so after deletion x must be equal to [9,
1].

What is the fastest way to do this? Is there any builtin?

Thanks.

Chris Rebert
Guest
Posts: n/a

 09-23-2009
On Wed, Sep 23, 2009 at 1:25 AM, blumenkraft <(E-Mail Removed)> wrote:
> Hi,
>
> I have some list:
> x = [8, 9, 1, 7]
> and list of indices I want to delete from x:
> indices_to_delete = [0, 3], so after deletion x must be equal to [9,
> 1].
>
> What is the fastest way to do this? Is there any builtin?

#untested & unbenchmarked since it's 1am
offset = 0
for index in indices_to_delete:
del x[index-offset]
offset += 1

Cheers,
Chris
--
http://blog.rebertia.com

Peter Otten
Guest
Posts: n/a

 09-23-2009
blumenkraft wrote:

> I have some list:
> x = [8, 9, 1, 7]
> and list of indices I want to delete from x:
> indices_to_delete = [0, 3], so after deletion x must be equal to [9,
> 1].
>
> What is the fastest way to do this? Is there any builtin?

Why's that obsession with speed?

>>> items = ["a", "b", "c", "d"]
>>> delenda = [0, 3]
>>> for i in sorted(delenda, reverse=True):

.... del items[i]
....
>>> items

['b', 'c']

>>> items = ["a", "b", "c", "d"]
>>> delenda = set([0, 3])
>>> items = [item for index, item in enumerate(items) if index not in

delenda]
>>> items

['b', 'c']

If you really need to modify the list in place change

items = [item for ...]

to

items[:] = [item for ...]

Try these and come back to complain if any of the above slows down your
script significantly...

Peter

Ishwor
Guest
Posts: n/a

 09-23-2009
Hi

2009/9/23 blumenkraft <(E-Mail Removed)>:
> Hi,
>
> I have some list:
> x = [8, 9, 1, 7]
> and list of indices I want to delete from x:
> indices_to_delete = [0, 3], so after deletion x must be equal to [9,
> 1].
>
> What is the fastest way to do this? Is there any builtin?

Try this-
>>> x = [8, 9, 1, 7]
>>> [x.pop(i) for i in sorted(indices_to_delete,reverse=True)]

[7, 8]
>>> x

[9, 1]

Built-in used here is `sorted' and method on list used here is `pop'.
With regards to efficiency you may want to use the methods of list
which is more intuitive afaik and useful as its more reflective of
effect on the type list. It's a trivial choice here but later it might
help.
--
Regards,
Ishwor Gurung

blumenkraft
Guest
Posts: n/a

 09-23-2009
On 23 ÓÅÎ, 12:48, Peter Otten <(E-Mail Removed)> wrote:
> blumenkraft wrote:
> > I have some list:
> > x = [8, 9, 1, 7]
> > and list of indices I want to delete from x:
> > indices_to_delete = [0, 3], so after deletion x must be equal to [9,
> > 1].

>
> > What is the fastest way to do this? Is there any builtin?

>
> Why's that obsession with speed?
>
> >>> items = ["a", "b", "c", "d"]
> >>> delenda = [0, 3]
> >>> for i in sorted(delenda, reverse=True):

>
> ... š š del items[i]
> ...>>> items
>
> ['b', 'c']
>
> >>> items = ["a", "b", "c", "d"]
> >>> delenda = set([0, 3])
> >>> items = [item for index, item in enumerate(items) if index not in

> delenda]
> >>> items

>
> ['b', 'c']
>
> If you really need to modify the list in place change
>
> items = [item for ...]
>
> to
>
> items[:] = [item for ...]
>
> Try these and come back to complain if any of the above slows down your
> script significantly...
>
> Peter

Thanks, it helped (I didn't know about keyword argument "reverse" in
sorted function)