You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
91 lines
3.4 KiB
91 lines
3.4 KiB
#!/usr/bin/env python3
|
|
# Copyright 2020 The Pigweed Authors
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License"); you may not
|
|
# use this file except in compliance with the License. You may obtain a copy of
|
|
# the License at
|
|
#
|
|
# https://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
|
|
# WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
|
|
# License for the specific language governing permissions and limitations under
|
|
# the License.
|
|
"""Tests the tokenized string decode module."""
|
|
|
|
import unittest
|
|
|
|
import tokenized_string_decoding_test_data as tokenized_string
|
|
import varint_test_data
|
|
from pw_tokenizer import decode
|
|
|
|
|
|
def error(msg, value=None):
|
|
"""Formats msg as the message for an argument that failed to parse."""
|
|
if value is None:
|
|
return '<[{}]>'.format(msg)
|
|
return '<[{} ({})]>'.format(msg, value)
|
|
|
|
|
|
class TestDecodeTokenized(unittest.TestCase):
|
|
"""Tests decoding tokenized strings with various arguments."""
|
|
def test_decode_generated_data(self):
|
|
self.assertGreater(len(tokenized_string.TEST_DATA), 100)
|
|
|
|
for fmt, decoded, encoded in tokenized_string.TEST_DATA:
|
|
self.assertEqual(decode.decode(fmt, encoded, True), decoded)
|
|
|
|
def test_unicode_decode_errors(self):
|
|
"""Tests unicode errors, which do not occur in the C++ decoding code."""
|
|
self.assertEqual(decode.decode('Why, %c', b'\x01', True),
|
|
'Why, ' + error('%c ERROR', -1))
|
|
|
|
self.assertEqual(
|
|
decode.decode('%sXY%+ldxy%u', b'\x83N\x80!\x01\x02', True),
|
|
'{}XY{}xy{}'.format(error('%s ERROR', "'N\\x80!'"),
|
|
error('%+ld SKIPPED', -1),
|
|
error('%u SKIPPED', 1)))
|
|
|
|
self.assertEqual(
|
|
decode.decode('%s%lld%9u', b'\x82$\x80\x80', True),
|
|
'{0}{1}{2}'.format(error("%s ERROR ('$\\x80')"),
|
|
error('%lld SKIPPED'), error('%9u SKIPPED')))
|
|
|
|
self.assertEqual(decode.decode('%c', b'\xff\xff\xff\xff\x0f', True),
|
|
error('%c ERROR', -2147483648))
|
|
|
|
def test_ignore_errors(self):
|
|
self.assertEqual(decode.decode('Why, %c', b'\x01'), 'Why, %c')
|
|
|
|
self.assertEqual(decode.decode('%s %d', b'\x01!'), '! %d')
|
|
|
|
def test_pointer(self):
|
|
"""Tests pointer args, which are not natively supported in Python."""
|
|
self.assertEqual(decode.decode('Hello: %p', b'\x00', True),
|
|
'Hello: 0x00000000')
|
|
self.assertEqual(decode.decode('%p%d%d', b'\x02\x80', True),
|
|
'0x00000001<[%d ERROR]><[%d SKIPPED]>')
|
|
|
|
|
|
class TestIntegerDecoding(unittest.TestCase):
|
|
"""Test decoding variable-length integers."""
|
|
def test_decode_generated_data(self):
|
|
test_data = varint_test_data.TEST_DATA
|
|
self.assertGreater(len(test_data), 100)
|
|
|
|
for signed_spec, signed, unsigned_spec, unsigned, encoded in test_data:
|
|
self.assertEqual(
|
|
int(signed),
|
|
decode.FormatSpec.from_string(signed_spec).decode(
|
|
bytearray(encoded)).value)
|
|
|
|
self.assertEqual(
|
|
int(unsigned),
|
|
decode.FormatSpec.from_string(unsigned_spec).decode(
|
|
bytearray(encoded)).value)
|
|
|
|
|
|
if __name__ == '__main__':
|
|
unittest.main()
|