def __init__(self, file_stream, url):
DocumentParser.__init__(self, file_stream)
self._url = url
self._namespace_URI = 'http://www.loc.gov/METS/'
self._mods_namespace_URI = 'http://www.loc.gov/mods/v3'
#read the content of the file
self._content_str = self._file_stream.read()
self._logical_structure = None
self._physical_structure = None
self._meta_data = None
self._relation = None
self._file_list = None
#some METS files contain uppercase mets directive
#self._content_str = self._content_str.replace('METS=', 'mets=')
#self._content_str = self._content_str.replace('', '')
#self._content_str = self._content_str.replace('MODS=', 'mods=')
#self._content_str = self._content_str.replace('', '')
try:
self._doc = parseString(self._content_str)
except Exception:
raise ParserError.InvalidDocument("The file is invalid. (is it" \
"corrupted?)")
if self._check_xml() is not True:
raise ParserError.InvalidDocument("The file is invalid. (is it" \
"corrupted?)")
def parse_verb(word_list):
skip(word_list, 'stop')
if peek(word_list) == 'verb':
return match(word_list, 'verb')
else:
raise ParserError("Expected a verb next.")
def _parse_assignment(self):
equals_token = self.expect(TokenType.EQUALS)
self.next()
left_side = self.stack.pop()
if not isinstance(left_side, Identifier) and not isinstance(left_side, ArrayDeclaration):
raise ParserError(
'unexpected left side of assignment, expected Identifier or ArrayDeclaration, but got {}'.format(
repr(left_side)))
self.skip_whitespaces(include_newlines=True)
token = self.token()
if token.token_type == TokenType.WORD:
right_side = self._parse_identifier()
elif token.token_type in [TokenType.QUOTE, TokenType.DOUBLE_QUOTES]:
right_side = self._parse_string_literal()
elif token.token_type == TokenType.NUMBER:
right_side = self._parse_constant()
elif token.token_type == TokenType.L_CURLY:
right_side = self._parse_array()
else:
raise ParsingError('unexpected right side of assignment: {}'.format(repr(token)))
semicolon_token = self.expect(TokenType.SEMICOLON)
self.index += 1
return Assignment(left_side, equals_token, right_side, semicolon_token)
def _parse_array(self):
l_curly_token = self.expect(TokenType.L_CURLY)
token = self.next()
children = []
while token.token_type != TokenType.R_CURLY and self.index < len(self.tokens):
if token.token_type in [TokenType.WHITESPACE, TokenType.TAB, TokenType.NEWLINE]:
self.skip_whitespaces(include_newlines=True)
token = self.token()
if token.token_type == TokenType.NUMBER:
children.append(self._parse_constant())
elif token.token_type in [TokenType.QUOTE, TokenType.DOUBLE_QUOTES]:
children.append(self._parse_string_literal())
elif token.token_type == TokenType.WORD:
children.append(self._parse_identifier())
elif token.token_type == TokenType.L_CURLY:
children.append(self._parse_array())
else:
raise ParserError('encountered unexpected token while parsing array: {}'.format(repr(token)))
self.skip_whitespaces(include_newlines=True)
self.expect([TokenType.COMMA, TokenType.R_CURLY])
if self.token().token_type == TokenType.COMMA:
token = self.next()
continue
else:
break
r_curly_token = self.expect(TokenType.R_CURLY)
self.index += 1
return Array(l_curly_token, children, r_curly_token)
def _get_record(self):
"""Get the record object in the xml file."""
self._file_stream.seek(0)
content_str = self._file_stream.read()
doc = parseString(content_str)
records = doc.getElementsByTagNameNS(self._namespace_URI, 'mods')
# get the id number of the first record
if len(records) == 0:
raise ParserError.InvalidDocument(
"XML/Mods Core document should contains at lease one record!")
if len(records) > 1:
raise ParserError.InvalidDocument(
"XML/Mods Core document should not contains more than "\
"one record!")
return records[0]
def parse_object(word_list):
skip(word_list, 'stop')
next_word = peek(word_list)
if next_word == 'noun':
return match(word_list, 'noun')
elif next_word == 'direction':
return match(word_list, 'direction')
else:
raise ParserError("Expected a noun or direction")
def parse_subject(word_list):
skip(word_list, 'stop')
next_word = peek(word_list)
if next_word == 'noun':
return match(word_list, 'noun')
elif next_word == 'verb':
return ('noun', 'player')
else:
raise ParserError("Expected a verb next.")
def _parse_array_declaration(self):
l_square = self.expect(TokenType.L_SQUARE)
r_square = self.expect_next(TokenType.R_SQUARE)
self.index += 1
identifier = self.stack.pop()
if not isinstance(identifier, Identifier):
raise ParserError('expected identifier for array declaration, but got {}'.format(repr(identifier)))
return ArrayDeclaration(identifier, l_square, r_square)
def parse(version_str):
m = re.match(
r"""
^JDK # 先頭JDK
([1-9][0-9]*) # familyNumber1桁目は0位外の数字で始まりn桁の数字
u # 固定文字 u
([0-9]+)$ # 1桁以上の数字(0可)
""", version_str, re.VERBOSE)
if m is None:
raise ParserError("invalid version string.")
familyNumber = int(m.group(1))
updateNumber = int(m.group(2))
return JavaVersion(familyNumber, updateNumber)
def __init__(self, node, argument):
'''Exception raised when a error is found'''
ParserError.__init__(self, "(Line %i, Col %i) " % \
(node.lineno, node.col_offset) + argument)