williamr/convert_to_eula.py
author Simon Howkins <simonh@symbian.org>
Wed, 14 Apr 2010 12:58:22 +0100
changeset 219 d57b367400c0
parent 137 b9165bfa078f
permissions -rw-r--r--
Updated release notes generation: Added copyright message Proper command line parsing Usage message if inputs are not right Added explanatory preamble to output "NEW" FCLs are marked as such Merge changesets are included in output

#!/usr/bin/python
# Copyright (c) 2009 Symbian Foundation.
# All rights reserved.
# This component and the accompanying materials are made available
# under the terms of the License "Eclipse Public License v1.0"
# which accompanies this distribution, and is available
# at the URL "http://www.eclipse.org/legal/epl-v10.html".
#
# Initial Contributors:
# Symbian Foundation - Initial contribution
# 
# Description:
# Map the SFL license to the EULA license, keeping a copy of the original file
# in a parallel tree for creation of a "repair" kit to reinstate the SFL

import os
import os.path
import re
import codecs

oldtext0 = re.compile('terms of the License "Symbian Foundation License v1.0"(to Symbian Foundation)?')
oldtext1 = re.compile('the URL "http:..www.symbianfoundation.org/legal/sfl-v10.html"')

newtext = [
  'terms of the License "Symbian Foundation License v1.0" to Symbian Foundation members and "Symbian Foundation End User License Agreement v1.0" to non-members',
  'the URL "http://www.symbianfoundation.org/legal/licencesv10.html"'
]

errorfiles = []
multinoticefiles = []
shadowroot = 'shadow_epoc32'

def file_type(file) :
	f = open(file, 'r')
	data = f.read(256)
	f.close()
	if len(data) < 2:
		return None # too short to be worth bothering about anyway
	if data[0] == chr(255) and data[1] == chr(254) :
		return 'utf_16_le'
	if data.find(chr(0)) >= 0 : 
		return None	# zero byte implies binary file
	return 'text'
	
def map_eula(dir, name, encoded) :
	global oldtext0
	global newtext1
	global newtext
	file = os.path.join(dir, name)
	if encoded == 'text':
		f = open(file, 'r')
	else:
		f = codecs.open(file, 'r', encoding=encoded)
	lines = f.readlines()
	# print ">> %s encoded as %s" % (file, f.encoding)
	f.close()
	
	updated = 0
	newlines = []
	while len(lines) > 0:
		line = lines.pop(0)
		pos1 = oldtext0.search(line)
		if pos1 != None:
			# be careful - oldtext is a prefix of newtext
			if pos1.group(1) != None:
				# line already converted - nothing to do
				newlines.append(line)
				continue
			midlines = []
			midlinecount = 1
			while len(lines) > 0:
				nextline = lines.pop(0)
				if not re.match('^\s$', nextline):
					# non-blank line
					if midlinecount == 0:
						break
					midlinecount -= 1
				midlines.append(nextline)
			urlline = nextline
			pos2 = oldtext1.search(urlline)
			if pos2 != None:
				# found it - assume that there's only one instance
				newline = oldtext0.sub(newtext[0], line)
				newurl  = oldtext1.sub(newtext[1], urlline)
				newlines.append(newline)
				newlines.extend(midlines)
				newlines.append(newurl)
				updated += 1
				continue
			else:
			  if updated != 0:
			  	lineno = 1 + len(newlines)
			  	print "Problem in " + file + " at " + lineno + ": incorrectly formatted >"
			  	print line
			  	print midlines
			  	print urlline
			  	global errorfiles
			  	errorfiles.append(file)
			  break
		newlines.append(line)
	
	if updated == 0:
		print " = no change to " + file
		return
	
	if updated > 1:
	  global multinoticefiles
	  multinoticefiles.append(file)
	  print '! found %d SFL notices in %s' % (updated, file)
	
	global shadowroot
	shadowdir = os.path.join(shadowroot, dir)
	if not os.path.exists(shadowdir) :
		os.makedirs(shadowdir)
	newfile = os.path.join(shadowroot,file)
	os.rename(file, newfile)
	if encoded == 'text':
		f = open(file, 'w')
	else:
		f = codecs.open(file, 'w', encoding=encoded)
	f.writelines(newlines)
	f.close()
	print "* updated %s (encoding %s)" % (file, f.encoding)

# process tree

for root, dirs, files in os.walk('epoc32', topdown=True):
	if re.compile('epoc32$').match(root) >= 0:
		if 'build' in dirs:
			dirs.remove('build') # don't recurse into the epoc32/build subtree
	for name in files:
		encoding = file_type(os.path.join(root, name))
		if encoding:
			map_eula(root, name, encoding)
	
print '%d problem files' % len(errorfiles)
print errorfiles

print '%d files with multiple notices' % len(multinoticefiles)
print multinoticefiles