def _parse_string(source: Source, string_regex=BASIC_STRING, separator='"') -> str:
source.expect(separator)
res = []
while True:
if not source.consume_regex(string_regex):
raise DoesNotMatch('Invalid string starting from {text}'
.format(text=source._text[:100]))
res.append(source.last_consumed)
# start of some escape character
if not source.consume('\\'):
break
# do nothing if new line chars encountered
# corresponds to \
if source.consume_regex(NEWLINE_ESCAPE_CHARS_REGEX):
pass
# read unicode characters
elif source.consume_regex(SHORT_UNICODE_REGEX) or source.consume_regex(LONG_UNICODE_REGEX):
res.append(chr(int(source.last_consumed, 16)))
else:
# fail if no escape character follows
source.expect_regex(ESCAPE_CHARS_REGEX)
res.append(ESCAPES_MAPPING[source.last_consumed])
source.expect(separator)
return ''.join(res)
def parse_line_comment(source: Source) -> Comment:
source.expect('#')
source.consume_regex(COMMENT_REGEX)
text = source.last_consumed
source.expect_regex(re.compile(r'(?P<res>\n|\Z)'))
return Comment(text)
def parse_kv_entry(source: Source) -> KVEntry:
key = parse_keyword(source)
source.consume_regex(WHITESPACE_REGEX)
source.expect('=')
source.consume_regex(WHITESPACE_REGEX)
val = parse_value(source)
return KVEntry(key, val)
def _parse_literal_string(source: Source, string_regex=LITERAL_STRING, separator='\'') -> str:
source.expect(separator)
is_matched = source.consume_regex(string_regex)
if not is_matched:
raise DoesNotMatch()
parsed_str = source.last_consumed
source.expect(separator)
return parsed_str
def _parse_inline_table(source: Source) -> InlineTable:
table = InlineTable()
source.expect('{')
if not source.consume('}'):
source.consume_regex(WHITESPACE_REGEX)
kv_entry = parse_kv_entry(source)
if source.last_consumed == '}':
raise InvalidTomlError('Cannot have nested inline tables.')
table[kv_entry.key] = kv_entry.val
while source.consume(','):
source.consume_regex(WHITESPACE_REGEX)
kv_entry = parse_kv_entry(source)
table[kv_entry.key] = kv_entry.val
source.consume_regex(WHITESPACE_REGEX)
source.expect('}')
return table
def _parse_array(source: Source) -> List[ValueType]:
array_type = None # type: type
items = [] # type: List[ValueType]
source.expect('[')
try:
while True:
source.consume_regex(ENCLOSING_WHITESPACE_CHARS)
parsed_val = parse_value(source)
if array_type is not None and \
not isinstance(parsed_val, array_type):
raise MixedTypesArray
if array_type is None:
array_type = type(parsed_val)
items.append(parsed_val)
source.consume_regex(ENCLOSING_WHITESPACE_CHARS)
source.expect(',')
except (ExpectationError, DoesNotMatch):
pass
source.consume_regex(ENCLOSING_WHITESPACE_CHARS)
source.expect(']')
return items
def test_expect_any_text_chunk_false(self):
text = 'World'
s = Source(text)
with self.assertRaises(ExpectationError):
s.expect('Hello')
def test_expect_any_text_chunk_true(self):
text = 'Hello'
s = Source(text)
# not raises anything
s.expect(text)
def parse_table(source: Source) -> ParsedTable:
source.expect('[')
keys = _parse_table_name(source)
source.expect(']')
return ParsedTable(keys)