Search directories for unused header files

gistfile1.txt

import os
import re
from collections import defaultdict

# Directories to inspect all .h, .c, .cpp files inside
TARGET_DIRECTORIES = ["source", "include"]

def getfilepaths(target_directory):
	all_files = []

	for root, dirs, files in os.walk(target_directory):
		for file in files:
			fullpath = os.path.join(root, file)
			all_files.append(fullpath)
			# file_name, file_ext = os.path.splitext(file)
			# print file_ext

	return all_files


def get_included_files(filepath):
	included_files = []

	for line in open(filepath):
		if "#include" in line:
			result = re.search("#include \"(?P<filename>[A-Za-z0-9\._]+)\".*", line)
			if result:
				filename = result.group("filename")
				included_files.append(filename)

	return included_files

if __name__ == "__main__":

	all_filepaths = []

	# Get a list of all file paths we're targetting.
	for directory in TARGET_DIRECTORIES:
		all_filepaths.extend(getfilepaths(directory))


	# A dict of {header: [file paths that include this header]}
	header_includers = defaultdict(set)

	for filepath in all_filepaths:
		included_files = get_included_files(filepath)

		for included_file in included_files:
			header_includers[included_file].add(filepath)


	# Print the include graph nicely.
	for file, callers in header_includers.items():
		print file
		for caller in callers:
			print "\t%s" % caller

		print "\n"

	# Find the headers that were never used
	used_headers = header_includers.keys()
	unused_headers = [filepath for filepath in all_filepaths if ".h" in filepath and os.path.basename(filepath) not in used_headers]


	print "UNUSED HEADER FILES"
	for header in unused_headers:
		print "\t%s" % header