More Efficient Way of Reading Large List of dicts

Question

For my project I use the mutagen library to read ID3 tags from 5000+ mp3 files. After reading them I construct the following objects using them.

    class Track:
    def __init__(self, artist, title, album=None):
        self.artist = artist
        self.title = title
        self.album = None

    def __str__(self):
        return "Track: %s : %s" % (self.artist,self.title, )

    def set_album(self,album):
        self.album = album

class Album:
    def __init__(self, artist, title, year='', genre='', tracks=None):
        self.artist = artist
        self.year = year
        self.genre = genre
        self.title = title 
        self.tracks = []

    def __str__(self):
        return "Album: %s : %s [%d]" % (self.artist,self.title,len(self.tracks))

    def add_track(self,track):
        self.tracks.append(track)

The problem is some files are missing some required tags(title missing,artist missing, or both), causing KeyValueError

 #'TALB' (album title), 'TIT2' (track title), 'TPE1' (artist), 'TDRC' (year), and 'TCON' (genre)
    for root, dirs, files in os.walk(dir):
        for filename in files:
            if filename.lower().endswith(e):
                fullname = os.path.join(root, filename)
                try:
                    audio = mutagen.File(fullname)
                    track = Track(audio['TPE1'],audio['TIT2'])
                    album = Album(audio['TPE1'], audio['TALB'], audio['TDRC'], audio['TCON'])
                excpet Exception as e:
                                print "Error on %s. %s " % (filename,type(e).__name__)  

This loads all the files that have all the tags, which is not good enough. I solved this problem by using ifs, it works fine and is fast enough. However I wonder if there is a better way of handling this.

Answer 1

If your default value can be an empty string instead of None you could use a defaultdict.

>>> 
>>> from collections import defaultdict
>>> d = defaultdict(str)
>>> d['a'] = 'data'
>>> d['b'] = 1
>>> d
defaultdict(<type 'str'>, {'a': 'data', 'b': 1})
>>> a = d['a']
>>> b = d['b']
>>> c = d['c']
>>> a, b, c
('data', 1, '')
>>> d
defaultdict(<type 'str'>, {'a': 'data', 'c': '', 'b': 1})
>>> 

Answer 2

Yes, use dict.get():

process(data.get('a'), data.get('b'), data.get('c'), data.get('d'))

This is similar to doing dict[key], except instead of raising a KeyError, it will return None, or it can return the second parameter passed (default is None).

It's similar to:

try:
    blah = data['a']
except ValueError:
    blah = None

Or:

if 'a' in data:
    blah = data['a']
else:
    blah = None

Answer 3

Use data.get("a") - this will return the value of data["a"] if a is present as a key in data, otherwise it will return None.

Answer 4

Your statement

a = data['a'] if 'a' in key else None

is equivalent to this more concise (and more efficient) statement:

a = data.get('a')

---

You might also consider just passing data in to process:

process(data)

---

Finally, there are also "slick" ways of doing this, which probably aren't warranted here, but fyi something along these lines might serve you:

process(**data)