def __init__(self, dictfilename, unkstamp=u"UNK", nodump=False):
"""
Constructor.
@param dictfilename is the dictionary file name (HTK-ASCII format)
@param unkstamp is the string used to represent a missing pronunciation
"""
# Symbol to represent missing entries in the dictionary
# (also called unknown entries)
self._unkstamp = unkstamp
self._filename = dictfilename
# The pronunciation dictionary
self._dict = {}
if dictfilename is not None:
data = None
if nodump is False:
# Try first to get the dict from a dump file (at least 2 times faster)
data = rutils.load_from_dump( dictfilename )
# Load from ascii if: 1st load, or, dump load error, or dump older than ascii
if data is None:
self.load_from_ascii( dictfilename )
if nodump is False:
rutils.save_as_dump( self._dict, dictfilename )
logging.info('Get dictionary from ASCII file.')
else:
self._dict = data
logging.info('Get dictionary from dumped file.')
def __init__(self, filename=None, nodump=False, casesensitive=False):
"""
Create a WordsList instance.
@param filename (str) is the word list file name, i.e. a file with 1 column.
@param nodump (Boolean) allows to disable the creation of a dump file.
@param casesensitive (Boolean) the list of word is case-sensitive or not
"""
self._stw = {}
# a list is enough but a dictionary is used because:
# with a dictionary it is faster to read tokens from a file and is also
# faster to find a token in it!
# Set the list of word to be case-sensitive or not.
self.casesensitive = casesensitive
if filename is not None:
data = None
#if nodump is False:
# Try first to get the dict from a dump file
# (at least 2 times faster than the ascii one)
data = rutils.load_from_dump( filename )
# Load from ascii if: 1st load, or, dump load error, or dump older than ascii
if data is None:
self.load_from_ascii( filename )
if nodump is False:
rutils.save_as_dump( self._stw, filename )
logging.info('Got word list from ASCII file.')
else:
self._stw = data
logging.info('Got word list from dumped file.')
def __init__(self, dictfilename=None, nodump=False):
"""
Constructor.
@param dictfilename is the dictionary file name (2 columns)
@param nodump (Boolean) disable the creation of a dump file
"""
self._dict = {}
if dictfilename is not None:
data = None
if nodump is False:
# Try first to get the dict from a dump file (at least 2 times faster)
data = rutils.load_from_dump( dictfilename )
# Load from ascii if: 1st load, or, dump load error, or dump older than ascii
if data is None:
self.load_from_ascii( dictfilename )
if nodump is False:
rutils.save_as_dump( self._dict, dictfilename )
else:
self._dict = data
def __init__(self, filename=None, nodump=False):
"""
Create a new WordsList instance.
@param filename is the word list file name (1 column)
@param nodump (Boolean) disable the creation of a dump file
"""
self._stw = {}
# a list is enough but:
# a dictionary is used because it is faster (to read tokens from
# a file / to find a token in this list).
if filename is not None:
data = None
if nodump is False:
# Try first to get the dict from a dump file (at least 2 times faster)
data = rutils.load_from_dump( filename )
# Load from ascii if: 1st load, or, dump load error, or dump older than ascii
if data is None:
self.load_from_ascii( filename )
if nodump is False:
rutils.save_as_dump( self._stw, filename )
logging.info('Get word list from ASCII file.')
else:
self._stw = data
logging.info('Get word list from dumped file.')
def __init__(self, dictfilename=None, nodump=False):
"""
Constructor.
@param dictfilename is the dictionary file name (2 columns)
@param nodump (Boolean) disable the creation of a dump file
"""
# Symbol to represent missing entries in the dictionary
# (also called unknown entries)
self._filename = dictfilename
# The replacements dictionary
self._dict = {}
if dictfilename is not None:
data = None
if nodump is False:
# Try first to get the dict from a dump file (at least 2 times faster)
data = rutils.load_from_dump( dictfilename )
# Load from ascii if: 1st load, or, dump load error, or dump older than ascii
if data is None:
self.load_from_ascii( dictfilename )
if nodump is False:
rutils.save_as_dump( self._dict, dictfilename )
logging.info('Get dictionary from ASCII file.')
else:
self._dict = data
logging.info('Get dictionary from dumped file.')
def __init__(self, filename=None, nodump=True):
"""
Constructor.
"""
self._sum = 0
self._dict = {}
if filename is not None:
data = None
if nodump is False:
# Try first to get the dict from a dump file (at least 2 times faster)
data = rutils.load_from_dump( filename )
# Load from ascii if: 1st load, or, dump load error, or dump older than ascii
if data is None:
self.load_from_ascii( filename )
if nodump is False:
rutils.save_as_dump( self._dict, filename )
else:
self._dict = data
