def parse_full_name(self):
"""
The main parse method for the parser. This method is run upon
assignment to the :py:attr:`full_name` attribute or instantiation.
Basic flow is to hand off to :py:func:`pre_process` to handle
nicknames. It then splits on commas and chooses a code path depending
on the number of commas.
:py:func:`parse_pieces` then splits those parts on spaces and
:py:func:`join_on_conjunctions` joins any pieces next to conjunctions.
"""
self.title_list = []
self.first_list = []
self.middle_list = []
self.last_list = []
self.suffix_list = []
self.nickname_list = []
self.unparsable = True
self.pre_process()
self._full_name = self.collapse_whitespace(self._full_name)
# break up full_name by commas
parts = [x.strip() for x in self._full_name.split(",")]
log.debug("full_name: {0}".format(self._full_name))
log.debug("parts: {0}".format(parts))
if len(parts) == 1:
# no commas, title first middle middle middle last suffix
# part[0]
pieces = self.parse_pieces(parts)
p_len = len(pieces)
for i, piece in enumerate(pieces):
try:
nxt = pieces[i + 1]
except IndexError:
nxt = None
# title must have a next piece, unless it's just a title
if self.is_title(piece) \
and (nxt or p_len == 1) \
and not self.first:
self.title_list.append(piece)
continue
if not self.first:
self.first_list.append(piece)
continue
if self.are_suffixes(pieces[i+1:]) or \
(
# if the next piece is the last piece and a roman
# numeral but this piece is not an initial
self.is_roman_numeral(nxt) and i == p_len - 2
and not self.is_an_initial(piece)
):
self.last_list.append(piece)
self.suffix_list += pieces[i + 1:]
break
if not nxt:
self.last_list.append(piece)
continue
self.middle_list.append(piece)
else:
# if all the end parts are suffixes and there is more than one piece
# in the first part. (Suffixes will never appear after last names
# only, and allows potential first names to be in suffixes, e.g.
# "Johnson, Bart"
if self.are_suffixes(parts[1].split(' ')) \
and len(parts[0].split(' ')) > 1:
# suffix comma:
# title first middle last [suffix], suffix [suffix] [, suffix]
# parts[0], parts[1:...]
self.suffix_list += parts[1:]
pieces = self.parse_pieces(parts[0].split(' '))
log.debug("pieces: {0}".format(u(pieces)))
for i, piece in enumerate(pieces):
try:
nxt = pieces[i + 1]
except IndexError:
nxt = None
if self.is_title(piece) \
and (nxt or len(pieces) == 1) \
and not self.first:
self.title_list.append(piece)
continue
if not self.first:
self.first_list.append(piece)
continue
if self.are_suffixes(pieces[i + 1:]):
self.last_list.append(piece)
self.suffix_list = pieces[i + 1:] + self.suffix_list
break
if not nxt:
self.last_list.append(piece)
continue
self.middle_list.append(piece)
else:
# lastname comma:
# last [suffix], title first middles[,] suffix [,suffix]
# parts[0], parts[1], parts[2:...]
pieces = self.parse_pieces(parts[1].split(' '), 1)
log.debug("pieces: {0}".format(u(pieces)))
# lastname part may have suffixes in it
lastname_pieces = self.parse_pieces(parts[0].split(' '), 1)
for piece in lastname_pieces:
# the first one is always a last name, even if it look like
# a suffix
if self.is_suffix(piece) and len(self.last_list) > 0:
self.suffix_list.append(piece)
else:
self.last_list.append(piece)
for i, piece in enumerate(pieces):
try:
nxt = pieces[i + 1]
except IndexError:
nxt = None
if self.is_title(piece) \
and (nxt or len(pieces) == 1) \
and not self.first:
self.title_list.append(piece)
continue
if not self.first:
self.first_list.append(piece)
continue
if self.is_suffix(piece):
self.suffix_list.append(piece)
continue
self.middle_list.append(piece)
try:
if parts[2]:
self.suffix_list += parts[2:]
except IndexError:
pass
if len(self) < 0:
log.info("Unparsable: \"{}\" ".format(self.original))
else:
self.unparsable = False
self.post_process()
def parse_full_name(self):
"""
The main parse method for the parser. This method is run upon
assignment to the :py:attr:`full_name` attribute or instantiation.
Basic flow is to hand off to :py:func:`pre_process` to handle
nicknames. It then splits on commas and chooses a code path depending
on the number of commas.
:py:func:`parse_pieces` then splits those parts on spaces and
:py:func:`join_on_conjunctions` joins any pieces next to conjunctions.
"""
self.title_list = []
self.first_list = []
self.middle_list = []
self.last_list = []
self.suffix_list = []
self.nickname_list = []
self.unparsable = True
self.pre_process()
self._full_name = self.collapse_whitespace(self._full_name)
# break up full_name by commas
parts = [x.strip() for x in self._full_name.split(",")]
log.debug("full_name: %s", self._full_name)
log.debug("parts: %s", parts)
if len(parts) == 1:
# no commas, title first middle middle middle last suffix
# part[0]
pieces = self.parse_pieces(parts)
p_len = len(pieces)
for i, piece in enumerate(pieces):
try:
nxt = pieces[i + 1]
except IndexError:
nxt = None
# title must have a next piece, unless it's just a title
if self.is_title(piece) \
and (nxt or p_len == 1) \
and not self.first:
self.title_list.append(piece)
continue
if not self.first:
if p_len == 1 and self.nickname:
self.last_list.append(piece)
continue
self.first_list.append(piece)
continue
if self.are_suffixes(pieces[i+1:]) or \
(
# if the next piece is the last piece and a roman
# numeral but this piece is not an initial
self.is_roman_numeral(nxt) and i == p_len - 2
and not self.is_an_initial(piece)
):
self.last_list.append(piece)
self.suffix_list += pieces[i+1:]
break
if not nxt:
self.last_list.append(piece)
continue
self.middle_list.append(piece)
else:
# if all the end parts are suffixes and there is more than one piece
# in the first part. (Suffixes will never appear after last names
# only, and allows potential first names to be in suffixes, e.g.
# "Johnson, Bart"
if self.are_suffixes(parts[1].split(' ')) \
and len(parts[0].split(' ')) > 1:
# suffix comma:
# title first middle last [suffix], suffix [suffix] [, suffix]
# parts[0], parts[1:...]
self.suffix_list += parts[1:]
pieces = self.parse_pieces(parts[0].split(' '))
log.debug("pieces: %s", u(pieces))
for i, piece in enumerate(pieces):
try:
nxt = pieces[i + 1]
except IndexError:
nxt = None
if self.is_title(piece) \
and (nxt or len(pieces) == 1) \
and not self.first:
self.title_list.append(piece)
continue
if not self.first:
self.first_list.append(piece)
continue
if self.are_suffixes(pieces[i+1:]):
self.last_list.append(piece)
self.suffix_list = pieces[i+1:] + self.suffix_list
break
if not nxt:
self.last_list.append(piece)
continue
self.middle_list.append(piece)
else:
# lastname comma:
# last [suffix], title first middles[,] suffix [,suffix]
# parts[0], parts[1], parts[2:...]
pieces = self.parse_pieces(parts[1].split(' '), 1)
log.debug("pieces: %s", u(pieces))
# lastname part may have suffixes in it
lastname_pieces = self.parse_pieces(parts[0].split(' '), 1)
for piece in lastname_pieces:
# the first one is always a last name, even if it looks like
# a suffix
if self.is_suffix(piece) and len(self.last_list) > 0:
self.suffix_list.append(piece)
else:
self.last_list.append(piece)
for i, piece in enumerate(pieces):
try:
nxt = pieces[i + 1]
except IndexError:
nxt = None
if self.is_title(piece) \
and (nxt or len(pieces) == 1) \
and not self.first:
self.title_list.append(piece)
continue
if not self.first:
self.first_list.append(piece)
continue
if self.is_suffix(piece):
self.suffix_list.append(piece)
continue
self.middle_list.append(piece)
try:
if parts[2]:
self.suffix_list += parts[2:]
except IndexError:
pass
if len(self) < 0:
log.info("Unparsable: \"%s\" ", self.original)
else:
self.unparsable = False
self.post_process()