UPDATE_NGRAM_DML = '''
UPDATE ngram SET freq = freq + 1 WHERE words = ?;
'''
PRUNE_NGRAM_DML = '''
DELETE FROM ngram WHERE freq <= ?;
'''
config = MyConfig()
#maximum combine number
N = config.getMaximumCombineNumber()
#change cwd to the word recognizer directory
words_dir = config.getWordRecognizerDir()
os.chdir(words_dir)
#chdir done
def handleOneDocument(infile, cur, length):
print(infile, length)
infilestatuspath = infile + config.getStatusPostfix()
infilestatus = utils.load_status(infilestatuspath)
if not utils.check_epoch(infilestatus, 'Segment'):
raise utils.EpochError('Please segment first.\n')
if utils.check_epoch(infilestatus, 'Populate'):
return False
sep = config.getWordSep()
#!/usr/bin/python3
import os
import os.path
import sqlite3
from argparse import ArgumentParser
from operator import itemgetter
from math import log
from sys import float_info
import utils
from myconfig import MyConfig
from dirwalk import walkIndex
config = MyConfig()
#change cwd to the word recognizer directory
words_dir = config.getWordRecognizerDir()
os.chdir(words_dir)
#chdir done
############################################################
# Create Bigram Database #
############################################################
CREATE_BIGRAM_DDL = '''
CREATE TABLE bigram (
prefix TEXT NOT NULL,
postfix TEXT NOT NULL,
freq INTEGER NOT NULL
);
'''
