A little script I wrote to perform a quick census on my music library, to help me identify which artists/albums contain the most mp3s/m4as. This was intended to help me replace my music library with oggs, but could be used for all sorts of other handy things, too.

countfiles

#!/usr/bin/env python3
import os
from sys import argv

# Walk through folders recursively, list the full path and number of (extension) files found in each.

basefolder = os.path.expanduser(argv[1])
filetype = str(argv[2]).lower()
output = []

# New system:
for folder in os.listdir(basefolder):
    filecounter = 0
    for subf, childf, files in os.walk(os.path.join(basefolder,folder)):
        for filen in files:
            if filen[:2] == "._":
                # Necessary because Rhythmbox (at least) populates some library folders with hidden
                # files of format "._{filename}.mp3" for each mp3 in that folder: this can screw up
                # the count when you simply count how many mp3s are present.
                continue
            if filetype == filen.lower()[len(filen)-len(filetype):]:
                # Above len/len voodoo just compares filetype to only the last letters of filenames.
                # This avoids a file containing the file extension by accident being counted.
                filecounter += 1
    if filecounter:
        #print(folder, "contains {0} mp3s.".format(filecounter))
        output.append([filecounter, folder])

# As first list item is an integer, will be sorted by this. Largest last.
output.sort()

print("Count, Path")
for x in output[:]:
    newx = str(x[0]) + "\t" + x[1]
    output[output.index(x)] = newx
print("\r\n".join(output))