Get all docx elements from docx files in a directory tree. Use with this script in this gist (https://gist.github.com/chulman444/b9e0ef2b61a241f47d25746a9a26e9f0)

docx_elements_from_docx_files.py

import sys, os, json, time
from examine_docx_elements import run as getDocxElements

def main():
	print("Started")
	rootdir = sys.argv[1]
	before = time.time()
	output = run(rootdir)
	after = time.time()
	print(output)
	print("Took {} seconds".format(after - before))

def run(rootdir):
	all_docx_elements = []
	ft = FilepathTracker()

	for root, subdirs, files in os.walk(rootdir):
		for f in files:
			filepath = root + "/" + f
			f_splits = f.rsplit(".")

			if len(f_splits) > 1 and f_splits[1] == "docx":
				docx_elements = getDocxElements(filepath)
				all_docx_elements += docx_elements
				ft.track(filepath, docx_elements)

	ft.print()
	unique_docx_elements = set(all_docx_elements)
	return unique_docx_elements

class FilepathTracker():
	def __init__(self):
		self.tracked = {}

	def track(self, filepath, docx_elements):
		for el in docx_elements:
			self.tracked.setdefault(el, []).append(filepath)

	def print(self):
		content = json.dumps(self.tracked, indent=4)
		print(content)
		self.printOccurrenceSummary()

	def printOccurrenceSummary(self):
		oc_summary = {}
		for k in self.tracked.keys():
			oc_summary[k] = len(self.tracked[k])

		print(json.dumps(oc_summary, indent=4))

if __name__=="__main__":
	main()