Python syntax: mutating/reducing a defaultdict?

Question

I wasn't sure what the term is, but basically I have a word_set in the form of a defaultdict [(word, value), ...] that came from a function that parsed some raw data.

I have other functions: reduceVal(word_set, min_val), reduceWord(word_set, *args), etc that removes pairs that: have a value less than min_val, have their (word) in [args], respectively. They all pretty much follow the same structure, e.g.

def reduceVal(word_set, value):
    "Returns word_set with (k, v) pairs where v > value)"
    rtn_set = defaultdict()
    for (k, v) in word_set.items():
        if v > value:
            rtn_set.update({k:v})
    return rtn_set

I was wondering if there was a more concise, or pythonic, way of expressing this, without building a new rtn_set or maybe even defining an entire function

Answer 1

Use a dict-comprehension, defaultdict is not required here:

new_dict = {k:v for k, v in word_set.items() if v > value}

on py2.x you should use word_set.iteritems() as it returns an iterator.

Answer 2

If you learn to use dict comprehensions you don't need to have separate functions for all of these different filters. For instance, reduceVal and reduceWord could be replaced with:

# Python 2.7+
{w: v for w, v in word_set.items() if v > value}
{w: v for w, v in word_set.items() if w in args}

# Python 2.6 and earlier
dict((w, v) for w, v in word_set.iteritems() if v > value)
dict((w, v) for w, v in word_set.iteritems() if w in args)