Is it possible to check that a list is not empty before grabbing the first element using as few lines as possible in Python?

Question

I keep having to do this operation all over the place in my python code. I'm willing to bet there is an easier (aka one-line) way to do this.

results = getResults()
if len(results) > 0:
    result = results[0]

I actually don't need "results" anywhere else, and I should only run "getResults" once.

Any ideas?

Be aware also that in the recommended Python style guidelines ( [PEP 8](http://www.python.org/dev/peps/pep-0008/)), the method should be named `get_results` rather than `getResults`. — Chris Morgan, Nov 21 '10 at 11:27

score 8 · Accepted Answer · answered Nov 21 '10 at 11:25

8

You haven't specified what result should be if results is empty, but this is one option (assuming Python 2.6 or greater):

result = results[0] if results else None

answered Nov 21 '10 at 11:25

Alec Thomas

khachik · Answer 2 · 2010-11-21T13:18:45.557

2

result = len(results) and results[0] or None. In case if results[0] is not 0, or False, or any empty container [], (), '', {}, set(), ....

edited Nov 21 '10 at 13:18

answered Nov 21 '10 at 10:41

khachik

I think you should make it clearer that this will _only_ work if `results[0]` evaluates to `True`. If `results == range(10)`, say, it will give None. – Katriel Nov 21 '10 at 12:42
@katrielex my comment (the second sentence in the post) doesn't make it clear? – khachik Nov 21 '10 at 13:05
@khachik: no. _Any_ empty container evaluates to False -- `set()`, `tuple()`, `{}`, `frozenset()`, `0.0`, ... -- it will fail for _any_ of these. – Katriel Nov 21 '10 at 13:17
@khachik: closer =) but still unfortunately not true... a) `0j`, `0L`, `0.0` and b) any class can define a `__nonzero__` method which returns False, in which case instances of that class will evaluate to False. The point is that _anything_ which evaluates to False will cause fail. [Although now I'm just nitpicking; these are at best unlikely to come up.] – Katriel Nov 21 '10 at 13:23
@katrielalex I understand that point, thanks :) also there is `__bool__` in 3.x, there is None... Isn't it obvious since `and` is used? – khachik Nov 21 '10 at 13:40
@khachik: yes, I guess so. I think the answer is fine, I'm just nitpicking. =) – Katriel Nov 21 '10 at 13:42

score 1 · Answer 3 · answered Nov 21 '10 at 12:02

1

next(iter(getResults()), None) for Python 2.6

answered Nov 21 '10 at 12:02

Rosh Oxymoron

score 0 · Answer 4 · answered Nov 21 '10 at 10:57

If you don't actually need results anywhere, consider a function that returns just the first result, and use that instead.

def getFirstResult():
    results = getResults()
    if len(results) > 0:
        return results[0]
    else:
        return None

score 0 · Answer 5 · answered Nov 21 '10 at 11:21

0

If you can change getResults() slightly along these lines:

def getResults():
    # ...
    getResults.seq = ... # save results in a func attribute
    return getResults.seq

It would allow you to write:

results = getResults.seq[0] if getResults() else None

answered Nov 21 '10 at 11:21

martineau

getResults is just an example. It would actually be a different function whenever I use it, and I don't want to have to write a wrapper for every time I use it. – Joe Hillenbrand Nov 21 '10 at 11:50
@Joehillen: This isn't a wrapper, it a permanent change to the function itself that is backwards compatible with previous usage but also allows the one-liner example usage to be used when desired. It's also possible to do it without modifying the function -- see the question [pythonic way to rewrite an assignment in an if statement](http://stackoverflow.com/questions/3744382/pythonic-way-to-rewrite-an-assignment-in-an-if-statement). – martineau Nov 21 '10 at 12:05

5 Answers5