1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/addon-sdk/source/python-lib/cuddlefish/property_parser.py Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,111 @@
1.4 +# This Source Code Form is subject to the terms of the Mozilla Public
1.5 +# License, v. 2.0. If a copy of the MPL was not distributed with this
1.6 +# file, You can obtain one at http://mozilla.org/MPL/2.0/.
1.7 +
1.8 +import re
1.9 +import codecs
1.10 +
1.11 +class MalformedLocaleFileError(Exception):
1.12 + pass
1.13 +
1.14 +def parse_file(path):
1.15 + return parse(read_file(path), path)
1.16 +
1.17 +def read_file(path):
1.18 + try:
1.19 + return codecs.open( path, "r", "utf-8" ).readlines()
1.20 + except UnicodeDecodeError, e:
1.21 + raise MalformedLocaleFileError(
1.22 + 'Following locale file is not a valid ' +
1.23 + 'UTF-8 file: %s\n%s"' % (path, str(e)))
1.24 +
1.25 +COMMENT = re.compile(r'\s*#')
1.26 +EMPTY = re.compile(r'^\s+$')
1.27 +KEYVALUE = re.compile(r"\s*([^=:]+)(=|:)\s*(.*)")
1.28 +
1.29 +def parse(lines, path=None):
1.30 + lines = iter(lines)
1.31 + lineNo = 1
1.32 + pairs = dict()
1.33 + for line in lines:
1.34 + if COMMENT.match(line) or EMPTY.match(line) or len(line) == 0:
1.35 + continue
1.36 + m = KEYVALUE.match(line)
1.37 + if not m:
1.38 + raise MalformedLocaleFileError(
1.39 + 'Following locale file is not a valid .properties file: %s\n'
1.40 + 'Line %d is incorrect:\n%s' % (path, lineNo, line))
1.41 +
1.42 + # All spaces are strip. Spaces at the beginning are stripped
1.43 + # by the regular expression. We have to strip spaces at the end.
1.44 + key = m.group(1).rstrip()
1.45 + val = m.group(3).rstrip()
1.46 + val = val.encode('raw-unicode-escape').decode('raw-unicode-escape')
1.47 +
1.48 + # `key` can be empty when key is only made of spaces
1.49 + if not key:
1.50 + raise MalformedLocaleFileError(
1.51 + 'Following locale file is not a valid .properties file: %s\n'
1.52 + 'Key is invalid on line %d is incorrect:\n%s' %
1.53 + (path, lineNo, line))
1.54 +
1.55 + # Multiline value: keep reading lines, while lines end with backslash
1.56 + # and strip spaces at the beginning of lines except the last line
1.57 + # that doesn't end up with backslash, we strip all spaces for this one.
1.58 + if val.endswith("\\"):
1.59 + val = val[:-1]
1.60 + try:
1.61 + # remove spaces before/after and especially the \n at EOL
1.62 + line = lines.next().strip()
1.63 + while line.endswith("\\"):
1.64 + val += line[:-1].lstrip()
1.65 + line = lines.next()
1.66 + lineNo += 1
1.67 + val += line.strip()
1.68 + except StopIteration:
1.69 + raise MalformedLocaleFileError(
1.70 + 'Following locale file is not a valid .properties file: %s\n'
1.71 + 'Unexpected EOF in multiline sequence at line %d:\n%s' %
1.72 + (path, lineNo, line))
1.73 + # Save this new pair
1.74 + pairs[key] = val
1.75 + lineNo += 1
1.76 +
1.77 + normalize_plural(path, pairs)
1.78 + return pairs
1.79 +
1.80 +# Plural forms in properties files are defined like this:
1.81 +# key = other form
1.82 +# key[one] = one form
1.83 +# key[...] = ...
1.84 +# Parse them and merge each key into one object containing all forms:
1.85 +# key: {
1.86 +# other: "other form",
1.87 +# one: "one form",
1.88 +# ...: ...
1.89 +# }
1.90 +PLURAL_FORM = re.compile(r'^(.*)\[(zero|one|two|few|many|other)\]$')
1.91 +def normalize_plural(path, pairs):
1.92 + for key in list(pairs.keys()):
1.93 + m = PLURAL_FORM.match(key)
1.94 + if not m:
1.95 + continue
1.96 + main_key = m.group(1)
1.97 + plural_form = m.group(2)
1.98 + # Allows not specifying a generic key (i.e a key without [form])
1.99 + if not main_key in pairs:
1.100 + pairs[main_key] = {}
1.101 + # Ensure that we always have the [other] form
1.102 + if not main_key + "[other]" in pairs:
1.103 + raise MalformedLocaleFileError(
1.104 + 'Following locale file is not a valid UTF-8 file: %s\n'
1.105 + 'This plural form doesn\'t have a matching `%s[other]` form:\n'
1.106 + '%s\n'
1.107 + 'You have to defined following key:\n%s'
1.108 + % (path, main_key, key, main_key))
1.109 + # convert generic form into an object if it is still a string
1.110 + if isinstance(pairs[main_key], unicode):
1.111 + pairs[main_key] = {"other": pairs[main_key]}
1.112 + # then, add this new plural form
1.113 + pairs[main_key][plural_form] = pairs[key]
1.114 + del pairs[key]
