def testRealAnnotationDefinition(self):
real_definition = """
DefineNetworkTrafficAnnotation("foobar_fetcher", R"(
semantics {
sender: "Foobar Component"
description: "Fetches Foobars for the user."
trigger: "The user requests a new Foobar."
data: "The type of Foobar the user requested."
destination: GOOGLE_OWNED_SERVICE
}
policy {
cookies_allowed: NO
setting: "Privacy and Security > Enable Foobars"
chrome_policy {
FoobarsEnabled {
FoobarsEnabled: false
}
}
})");"""
tokenizer = Tokenizer(real_definition,
'components/foobar/foobar_request_handler.cc',
42)
self.assertEqual('DefineNetworkTrafficAnnotation',
tokenizer.advance('symbol'))
self.assertEqual('(', tokenizer.advance('left_paren'))
self.assertEqual('foobar_fetcher', tokenizer.advance('string_literal'))
self.assertEqual(',', tokenizer.advance('comma'))
self.assertTrue(bool(tokenizer.advance('string_literal')))
self.assertEqual(')', tokenizer.advance('right_paren'))
def _parse_string(self, tokenizer: Tokenizer) -> str:
"""Parse a string value.
It could be a string literal by itself, or multiple string literals
concatenated together. Add a newline to the string for each
concatenation."""
text = tokenizer.advance('string_literal')
while True:
# Perform concatenations.
if tokenizer.maybe_advance('plus') is None:
break
text += '\n'
text += tokenizer.advance('string_literal')
return text
def testAdvanceHappyPath(self):
tokenizer = Tokenizer('"hello", R"(world)", function_name())));',
'foo.txt', 33)
self.assertEqual('hello', tokenizer.advance('string_literal'))
self.assertEqual(',', tokenizer.advance('comma'))
self.assertEqual('world', tokenizer.advance('string_literal'))
self.assertEqual(',', tokenizer.advance('comma'))
self.assertEqual('function_name', tokenizer.advance('symbol'))
self.assertEqual('(', tokenizer.advance('left_paren'))
self.assertEqual(')', tokenizer.advance('right_paren'))
self.assertEqual(')', tokenizer.advance('right_paren'))
def testMaybeAdvance(self):
tokenizer = Tokenizer('"hello", world', 'foo.txt', 33)
self.assertEqual(None, tokenizer.maybe_advance('symbol'))
self.assertEqual('hello', tokenizer.maybe_advance('string_literal'))
self.assertEqual(',', tokenizer.maybe_advance('comma'))
self.assertEqual(None, tokenizer.maybe_advance('left_paren'))
self.assertEqual('world', tokenizer.maybe_advance('symbol'))
self.assertEqual(None, tokenizer.maybe_advance('right_paren'))
def testConcatenatedStrings(self):
tokenizer = Tokenizer('"hello " + "world" + "!"', 'foo.java', 22)
self.assertEqual('hello ', tokenizer.advance('string_literal'))
self.assertEqual('+', tokenizer.advance('plus'))
self.assertEqual('world', tokenizer.advance('string_literal'))
self.assertEqual('+', tokenizer.advance('plus'))
self.assertEqual('!', tokenizer.advance('string_literal'))
def _parse_body(self, body):
"""Tokenizes and parses the arguments given to the definition function."""
# Don't bother parsing CreateMutableNetworkTrafficAnnotationTag(), we don't
# care about its arguments anyways.
if self.type_name == 'Mutable':
return
tokenizer = Tokenizer(body, self.file_path, self.line_number)
# unique_id
self.unique_id = tokenizer.advance('string_literal')
tokenizer.advance('comma')
# extra_id (Partial/BranchedCompleting)
if self.type_name == 'Partial' or self.type_name == 'BranchedCompleting':
self.extra_id = tokenizer.advance('string_literal')
tokenizer.advance('comma')
# partial_annotation (Completing/BranchedCompleting)
if self.type_name == 'Completing' or self.type_name == 'BranchedCompleting':
# Skip the |partial_annotation| argument. It can be a variable_name, or a
# FunctionName(), so skip the parentheses if they're there.
tokenizer.advance('symbol')
if tokenizer.maybe_advance('left_paren'):
tokenizer.advance('right_paren')
tokenizer.advance('comma')
# proto text
self.text = tokenizer.advance('string_literal')
# The function call should end here without any more arguments.
assert tokenizer.advance('right_paren')
def testAdvanceErrorPaths(self):
tokenizer = Tokenizer(' hello , ', 'foo.txt', 33)
tokenizer.advance('symbol')
with self.assertRaisesRegexp(SourceCodeParsingError,
'Expected symbol.+at foo.txt:33'):
# There are no more tokens.
tokenizer.advance('symbol')
tokenizer = Tokenizer('"hello"', 'foo.txt', 33)
with self.assertRaisesRegexp(SourceCodeParsingError,
'Expected comma.+at foo.txt:33'):
# The type doesn't match.
tokenizer.advance('comma')
tokenizer = Tokenizer('{', 'foo.txt', 33)
with self.assertRaisesRegexp(SourceCodeParsingError,
'Expected string_literal.+at foo.txt:33'):
# Not a valid token at all.
tokenizer.advance('string_literal')
def testAdvanceMultiline(self):
tokenizer = Tokenizer('\n\tR"(the quick\nbrown\nfox)"', 'foo.txt', 33)
self.assertEqual('the quick\nbrown\nfox',
tokenizer.advance('string_literal'))
