update-license-header.py

update-license-header.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-
#
# Copyright 2005-2017 The Mumble Developers. All rights reserved.
# Use of this source code is governed by a BSD-style license
# that can be found in the LICENSE file at the root of the
# Mumble source tree or at <https://www.mumble.info/LICENSE>.

from __future__ import (unicode_literals, print_function, division)

import os

def licenseHeader():
	s = ''
	with open('LICENSE.header', 'rb') as f:
		s = f.read()
	return s

def fileHasLicenseHeader(fn):
	f = open(fn, 'rb')
	buf = f.read(4096)

	clue1 = b'Copyright' in buf
	clue2 = b'The Mumble Developers. All rights reserved.' in buf
	clue3 = b'Mumble source tree or at <https://www.mumble.info/LICENSE>.' in buf

	return clue1 and clue2 and clue3

def updateLicenseHeader(fn):
	hdr = licenseHeader()
	splitHdr = hdr.split(b'\n')

	# Read the original file
	with open(fn, 'rb') as f:
		buf = f.read()
	print(type(buf))

	# Figure out which comment style
	# the file uses. Also, try to figure
	# out if it uses Unix-style line endings,
	# or Windows-style line endings.
	lastLine = splitHdr[-2]
	lastLineNoComments = lastLine.replace(b'//', b'', 1)

	lastLineIndex = buf.find(lastLineNoComments)
	if lastLineIndex < 0:
		raise Exception('bad file')

	commentToken = b''
	idx = lastLineIndex
	while idx >= 0:
		idx -= 1
		if buf[idx] == '\n':
			break
		commentToken = buf[idx] + commentToken

	newline = '\n'
	if buf[idx-1] == '\r':
		newline = '\r\n'

	# Construct the new header
	newHeader = b''
	for line in splitHdr:
		if len(line) == 0:
			continue
		newHeader += line.replace(b'//', commentToken, 1) + newline

	# Now, find the beginning of the license
	# header in the file.
	needle = commentToken + b' Copyright'

	idx = lastLineIndex
	while idx >= 0:
		oi = buf[idx:idx+len(needle)]
		if oi == needle:
			break
		idx -= 1
	if idx == -1:
		raise Exception('could not find beginning of license header in {0}'.format(fn))
	headerBeginIndex = idx

	# OK, we have the beginning.
	# Now, find the end. (Inlcuding
	# newline characters.)
	idx = lastLineIndex
	while True:
		if buf[idx] == '\r' and buf[idx+1] == '\n':
			if newline != '\r\n':
				raise Exception('found CRLF when searching for end-of-header but the file does not use CRLF')
			break
		if buf[idx] == '\n':
			if newline != '\n':
				raise Exception('found LF when searching for end-of-header but the file does not use LF')
			break
		idx += 1

	byteAfterHeaderIdx = idx + len(newline)

	# OK, let's re-write the file.

	with open(fn, 'wb') as nf:
		if headerBeginIndex > 0:
			nf.write(buf[0:headerBeginIndex])
		nf.write(newHeader)
		nf.write(buf[byteAfterHeaderIdx:])

	return

def main():
	for root, dirs, files in os.walk('.'):
		for fn in files:
			absFn = os.path.join(root, fn)
			# Skip git
			if os.path.join('.git', '') in absFn:
				continue
			# Skip LICENSE.header
			if absFn == os.path.join(root, 'LICENSE.header'):
				continue
			# Skip self
			if absFn == os.path.join(root, 'scripts', 'update-license-header.py'):
				continue
			hasHeader = fileHasLicenseHeader(absFn)
			if hasHeader:
				updateLicenseHeader(absFn)
				print('Updated license header in file at {0}'.format(absFn))
				continue

if __name__ == '__main__':
	main()