[Tutor] question about a list of lists

[Kent Johnson]

shawn bright wrote:
> Thanks Kent,
>    i am going with option A, the helper set, because i also need to 
> count the occurances and this seems to be the easiest solution.

If you need the number of unique items that is just the length of the 
final list. If you need the number of occurrences of each initial item 
then a dict mapping initial item to count (or maybe a list of lists) 
would be the way to go.

Kent

> 
> thanks for your help.
> 
> shawn
> 
> On 1/28/07, *Kent Johnson* <kent37 at tds.net <mailto:kent37 at tds.net>> wrote:
> 
>     shawn bright wrote:
>      > lo there all.
>      >
>      > i have a list of lists that i want to build, only if an item is
>     not in
>      > the list already.
>      >
>      > kinda like this
>      > new_list = []
>      > for item in lists: # item will look something like [var1, var2,
>     var3]
>      >     if item[0] in new_list ( only the first element of each list
>     ) like
>      > new_list[0][0]
>      >
>      > basicly, i want to know if item[0]   is one of the items[0] in my
>     new_list
>      >
>      > whats a good pythonic way to do this? i mean, i have a couple of
>     way to
>      > do this, but they are ugly.
> 
>     One way to do this is to keep a helper set that contains the first
>     elements of each list. Something like
>     new_list = []
>     firsts = set()
>     for item in lists:
>        if item[0] not in firsts:
>          new_list.append(item)
>          firsts.add(item[0]
> 
>     If you don't care about the order of the result, and if two lists have
>     duplicate first items you are happy to use the first, then you could use
>     a dict mapping first item to list:
> 
>     new_list = dict((item[0], item) for item in lists).values()
> 
>     Kent
> 
>