#! /usr/bin/python
# Copyright (C) 2016 and later: Unicode, Inc. and others.
# License & terms of use: http://www.unicode.org/copyright.html
# Copyright (C) 2009-2011, International Business Machines Corporation, Google and Others.
# All rights reserved.
#
# Script to check and fix svn property settings for ICU source files.
# Also check for the correct line endings on files with svn:eol-style = native
#
# THIS SCRIPT DOES NOT WORK ON WINDOWS
# It only works correctly on platforms where the native line ending is a plain \n
#
# usage:
# icu-svnprops-check.py [options]
#
# options:
# -f | --fix Fix any problems that are found
# -h | --help Print a usage line and exit.
#
# The tool operates recursively on the directory from which it is run.
# Only files from the svn repository are checked.
# No changes are made to the repository; only the working copy will be altered.
import sys
import os
import os.path
import re
import getopt
# file_types: The parsed form of the svn auto-props specification.
# A list of file types - .cc, .cpp, .txt, etc.
# each element is a [type, proplist]
# "type" is a regular expression string that will match a file name
# prop list is another list, one element per property.
# Each property item is a two element list, [prop name, prop value]
file_types = list()
def parse_auto_props():
aprops = svn_auto_props.splitlines()
for propline in aprops:
if re.match("\s*(#.*)?$", propline): # Match comment and blank lines
continue
if re.match("\s*\[auto-props\]", propline): # Match the [auto-props] line.
continue
if not re.match("\s*[^\s]+\s*=", propline): # minimal syntax check for <file-type> =
print "Bad line from autoprops definitions: " + propline
continue
file_type, string_proplist = propline.split("=", 1)
#transform the file type expression from autoprops into a normal regular expression.
# e.g. "*.cpp" ==> ".*\.cpp$"
file_type = file_type.strip()
file_type = file_type.replace(".", "\.")
file_type = file_type.replace("*", ".*")
file_type = file_type + "$"
# example string_proplist at this point: " svn:eol-style=native;svn:executable"
# split on ';' into a list of properties. The negative lookahead and lookbehind
# in the split regexp are to prevent matching on ';;', which is an escaped ';'
# within a property value.
string_proplist = re.split("(?<!;);(?!;)", string_proplist)
proplist = list()
for prop in string_proplist:
if prop.find("=") >= 0:
prop_name, prop_val = prop.split("=", 1)
else:
# properties with no explicit value, e.g. svn:executable
prop_name, prop_val = prop, ""
prop_name = prop_name.strip()
prop_val = prop_val.strip()
# unescape any ";;" in a property value, e.g. the mime-type from
# *.java = svn:eol-style=native;svn:mime-type=text/plain;;charset=utf-8
prop_val = prop_val.replace(";;", ";");
proplist.append((prop_name, prop_val))
file_types.append((file_type, proplist))
# print file_types
def runCommand(cmd):
output_file = os.popen(cmd);
output_text = output_file.read();
exit_status = output_file.close();
if exit_status:
print >>sys.stderr, '"', cmd, '" failed. Exiting.'
sys.exit(exit_status)
return output_text
svn_auto_props = runCommand("svn propget svn:auto-props http://source.icu-project.org/repos/icu")
def usage():
print "usage: " + sys.argv[0] + " [-f | --fix] [-h | --help]"
#
# UTF-8 file check. For text files with svn:mime-type=text/anything, check the specified charset
# file_name: name of a text file.
# base_mime_type: svn:mime-type property from the auto-props settings for this file type.
# actual_mime_type: existing svn:mime-type property value for the file.
# return: The correct svn:mime-type property value,
# either the original, if it looks OK, otherwise the value from auto-props
#
def check_utf8(file_name, base_mime_type, actual_mime_type):
f = open(file_name, 'r')
bytes = f.read()
f.close()
file_is_utf8 = True
try:
bytes.decode("UTF-8")
except UnicodeDecodeError:
file_is_utf8 = False
if not file_is_utf8 and actual_mime_type.find("utf-8") >= 0:
print "Error: %s is not valid utf-8, but has a utf-8 mime type." % file_name
return actual_mime_type
if file_is_utf8 and actual_mime_type.find("charset") >=0 and actual_mime_type.find("utf-8") < 0:
print "Warning: %s is valid utf-8, but has a mime-type of %s." % (file_name, actual_mime_type)
if ord(bytes[0]) == 0xef:
if not file_name.endswith(".txt"):
print "Warning: file %s contains a UTF-8 BOM: " % file_name
# If the file already has a charset in its mime-type, don't make any change.
if actual_mime_type.find("charset=") >= 0:
return actual_mime_type;
return base_mime_type
def main(argv):
fix_problems = False;
try:
opts, args = getopt.getopt(argv, "fh", ("fix", "help"))
except getopt.GetoptError:
print "unrecognized option: " + argv[0]
usage()
sys.exit(2)
for opt, arg in opts:
if opt in ("-h", "--help"):
usage()
sys.exit()
if opt in ("-f", "--fix"):
fix_problems = True
if args:
print "unexpected command line argument"
usage()
sys.exit()
parse_auto_props()
output = runCommand("svn ls -R ");
file_list = output.splitlines()
for f in file_list:
if os.path.isdir(f):
# print "Skipping dir " + f
continue
if not os.path.isfile(f):
print "Repository file not in working copy: " + f
continue;
for file_pattern, props in file_types:
if re.match(file_pattern, f):
# print "doing " + f
for propname, propval in props:
actual_propval = runCommand("svn propget --strict " + propname + " " + f)
#print propname + ": " + actual_propval
if propname == "svn:mime-type" and propval.find("text/") == 0:
# check for UTF-8 text files, should have svn:mime-type=text/something; charset=utf8
propval = check_utf8(f, propval, actual_propval)
if not (propval == actual_propval or (propval == "" and actual_propval == "*")):
print "svn propset %s '%s' %s" % (propname, propval, f)
if fix_problems:
os.system("svn propset %s '%s' %s" % (propname, propval, f))
if __name__ == "__main__":
main(sys.argv[1:])