Parsing through the Federal Register

fedregparse.py

import os
from lxml import etree
import csv
import sys

titles = {}
for title in range(1,50):
    titles[title] = 0

for froot, dirs, files in os.walk('.', topdown=True):
    for name in files:
        if name.lower().endswith('.xml'):
            fname = os.path.join(froot, name)  # This is the filename for the XML file
            # TODO: Check for Zero bytes... right now, I just deleted the file....             
            root = etree.parse(fname)
            for title in range(1,50):
                rules = root.xpath("//REGTEXT[@TITLE='" + str(title) + "']/ancestor::RULE")
                titles[title] = titles[title] + len(rules)

w = csv.writer(sys.stderr)
w.writerows(titles.items())

Data comes from the zip file here: https://www.govinfo.gov/bulkdata/FR