Handmade CJK coverage util

cjk.txt

CJK Radical Supplement      ,2E80-2EFF
Kangxi Radicals             ,2F00-2FDF
CJK symbols and punctuation ,3000-303F
Hiragana                    ,3040-309F
Katakana                    ,30A0-30FF
CJK strokes                 ,31C0-31EF
Katakana Common             ,31F0-31FF
                            ,3200-33FF
CJK compatibility           ,3300-33FF
CJK Unified Ideographs      ,4E00-9FFF
CJK compatibility           ,F900-FAFF
                            ,FE30-FE4F
Katakana halfwidth          ,FF00-FFEF
Kana Supplement             ,1B000-1B0FF

Example.sh

$ ls -1 font-repo/
HelveticaNeue.ttf
LucidaGrande.ttc
meiryo.ttc
ヒラギノ丸ゴ ProN W4.ttc
ヒラギノ角ゴシック W8.ttc
$ python handmade-cjk.py
num desired: 23216
num unsupported: 6640
extras (supported by not desired): 6563

handmade-cjk.py

# -*- coding: utf-8 -*-
from __future__ import unicode_literals
import os
from itertools import chain
from fontTools.ttLib import TTFont

def hex_range_to_dec(hex_range):
    lower, upper = map(lambda o: int(o, 16), hex_range.split('-'))
    # hex ranges are inclusive, and python's range end is exclusive, so +1
    return range(lower, upper+1)

def desired_chars(range_lines):
    needed = set()
    for line in range_lines:
        needed |= set(hex_range_to_dec(line))
    return needed

def supported_by(font_list):
    provided = set()
    for font_name in font_list:
        ttf = TTFont(font_name, fontNumber=0)
        provided |= set(chain.from_iterable([charCode for charCode in table.cmap.keys()] for table in ttf["cmap"].tables))
    return provided

if __name__ == '__main__':
    with open('handmade-cjk.txt', 'r') as f:
        desired_ranges = [line[:-1].rsplit(',', 1)[1] for line in f.readlines()]
        desired = desired_chars(desired_ranges)

    repo_path = './font-repo/'
    files = [os.path.join(repo_path, fn) for fn in os.listdir(repo_path)]
    supported = supported_by(files)

    unsupported = desired - supported
    extra = supported - desired

    print "num desired: %s" % len(desired)
    print "num unsupported: %s" % len(unsupported)
    print "extras (supported by not desired): %s" % len(extra)

Strictly in "works for me" territory, but:
Requires fontTools, assumes your font repo dir is ./font-repo, assumes hex ranges are inclusive and in a specific file format.

It looks like "num desired" is measured in bytes, and "num supported" is measured in characters