Cecilia Alm wrote: > 1) Unordered maps/collections like sets and dictionaries also seem to > support iterating over set members or dictionary keys with "for x in > y" syntax. As far as I can tell, the following three ways generally > behave the same way, or is there a difference in behavior between: > > a) for key in dictionary: > b) for key in dictionary.keys(): > c) mykeys = dictionary.keys() > for k in mykeys:
These are equivalent in most common usage but there is a difference which can be significant. Calling dict.keys() creates an actual list with all the keys in it which is not created when you iterate the dictionary directly. In most cases this doesn't matter but for a huge dictionary it might. Also if you are adding or deleting from the dict during the iteration then dict.keys() is safer because the list of keys is created before the add and delete. Using dict.iterkeys() is more equivalent to 'for key in dictionary' because it doesn't create the intermediate list. Of course using c) you will also have the variable mykeys available which is not the case in a) and b) > > 2) The following three ways for manipulating objects in a list with a > function will generally do the same thing, right? (but the last two > have a shorter syntax). > > a) newls = [] > for entry in mylist: > newls.append(function(entry)) > > b) newlist = map(function, mylist) > > c) newlist = [function(entry) for entry in mylist] Yes, pretty much. After a) and c) the entry variable will still be defined. For c) this is considered a bug and IIRC will change with Python 3. IMO modern usage tends toward c). Kent _______________________________________________ Tutor maillist - Tutor@python.org http://mail.python.org/mailman/listinfo/tutor
