def _constructTrie(self):
trie = Trie()
# Written in multiple lines instead of loop to prevent dealing with hashing
trie.insertList(
tuple(
map(lambda x: x.lower(), [
str(member)
for name, member in lh.SQLKeyword.__members__.items()
])), self.terminate_map['Keyword'])
trie.insertList(
tuple(
map(lambda x: x.lower(), [
str(member)
for name, member in lh.SQLFunction.__members__.items()
])), self.terminate_map['Function'])
trie.insertList(
tuple(
map(lambda x: x.lower(), [
str(member)
for name, member in lh.SQLDataType.__members__.items()
])), self.terminate_map['DataType'])
trie.insertList(
tuple(
map(lambda x: x.lower(), [
str(member)
for name, member in lh.SQLOperator.__members__.items()
])), self.terminate_map['Operator'])
return trie
def __init__(self, lower=True):
self.lower = lower
self.trie = Trie()
self.ent2type = {} # word list to type
self.ent2id = {'<UNK>': 0} # word list to id
self.space = ''
class LexiconBuilder(object):
def __init__(self, tokens, value):
self.lexicon = Trie()
self.build(tokens, value) # 每次都会初始化pkl
# def _build(self):
# cnt = 0
# with open(self._path, 'r', encoding='utf-8') as rf:
# for num, line in enumerate(rf, 1):
# line = line.strip()
# if line:
# [token] = line.split()
# self.lexicon.add(token, value)
# cnt += 1
def build(self, tokens, value):
"""
:param tokens: word_list:['电视机', '电视', ...]
:param value: word_onto:'tool','show_name',...
:return:
"""
for token in tokens:
self.lexicon.add(token, value)
def show(self, num):
return self.lexicon.show(num)
def search(self, word):
return self.lexicon.search(word, value_flag=True, verbose=False)
class Gazetteer:
def __init__(self, lower): #lower = False
self.trie = Trie()
self.ent2type = {} ## word list to type
self.ent2id = {"<UNK>": 0} ## word list to id
self.lower = lower
self.space = ""
def enumerateMatchList(self, word_list):
if self.lower:
word_list = [word.lower() for word in word_list]
match_list = self.trie.enumerateMatch(word_list, self.space)
return match_list
def insert(self, word_list, source):
if self.lower:
word_list = [word.lower() for word in word_list]
self.trie.insert(word_list)
string = self.space.join(word_list)
if string not in self.ent2type:
self.ent2type[string] = source
if string not in self.ent2id:
self.ent2id[string] = len(self.ent2id)
def searchId(self, word_list):
if self.lower:
word_list = [word.lower() for word in word_list]
string = self.space.join(word_list)
if string in self.ent2id:
return self.ent2id[string]
return self.ent2id["<UNK>"]
def searchType(self, word_list):
if self.lower:
word_list = [word.lower() for word in word_list]
string = self.space.join(word_list)
if string in self.ent2type:
return self.ent2type[string]
print(
"Error in finding entity type at gazetteer.py, exit program! String:",
string)
exit(0)
def size(self):
return len(self.ent2type)
def clean(self):
self.trie = Trie()
self.ent2type = {}
self.ent2id = {"<UNK>": 0}
self.space = ""
class Lexicon(object):
def __init__(self, lower=True):
self.lower = lower
self.trie = Trie()
self.ent2type = {} # word list to type
self.ent2id = {'<UNK>': 0} # word list to id
self.space = ''
def enumerate_match_list(self, word_list):
if self.lower:
word_list = [word.lower() for word in word_list]
match_list = self.trie.enumerate_match(word_list, self.space)
return match_list
def insert(self, word_list, source):
if self.lower:
word_list = [word.lower() for word in word_list]
self.trie.insert(word_list)
string = self.space.join(word_list)
if string not in self.ent2type:
self.ent2type[string] = source
if string not in self.ent2id:
self.ent2id[string] = len(self.ent2id)
def search_id(self, word_list):
if self.lower:
word_list = [word.lower() for word in word_list]
string = self.space.join(word_list)
if string in self.ent2id:
return self.ent2id[string]
return self.ent2id['<UNK>']
def search_type(self, word_list):
if self.lower:
word_list = [word.lower() for word in word_list]
string = self.space.join(word_list)
if string in self.ent2type:
return self.ent2type[string]
print('Error in finding entity type at lexicon.py, exit programming')
exit(0)
def size(self):
return len(self.ent2type)
def clean(self):
self.trie = Trie()
self.ent2type = {}
self.ent2id = {}
self.space = ''
class Gazetteer:
def __init__(self, lower):
self.trie = Trie()
self.ent2type = {} ## word list to type
self.ent2id = {"<UNK>":0} ## word list to id
self.lower = lower
self.space = ""
def enumerateMatchList(self, word_list):
if self.lower:
word_list = [word.lower() for word in word_list]
match_list = self.trie.enumerateMatch(word_list, self.space)
return match_list
def insert(self, word_list, source):
if self.lower:
word_list = [word.lower() for word in word_list]
self.trie.insert(word_list)
string = self.space.join(word_list)
if string not in self.ent2type:
self.ent2type[string] = source
if string not in self.ent2id:
self.ent2id[string] = len(self.ent2id)
def searchId(self, word_list):
if self.lower:
word_list = [word.lower() for word in word_list]
string = self.space.join(word_list)
if string in self.ent2id:
return self.ent2id[string]
return self.ent2id["<UNK>"]
def searchType(self, word_list):
if self.lower:
word_list = [word.lower() for word in word_list]
string = self.space.join(word_list)
if string in self.ent2type:
return self.ent2type[string]
print ("Error in finding entity type at gazetteer.py, exit program! String:", string)
exit(0)
def size(self):
return len(self.ent2type)
def __init__(self, lower):
self.trie = Trie()
self.ent2type = {} ## word list to type
self.ent2id = {"<UNK>": 0} ## word list to id
self.lower = lower
self.space = ""
def __init__(self, tokens, value):
self.lexicon = Trie()
self.build(tokens, value) # 每次都会初始化pkl
def clean(self):
self.trie = Trie()
self.ent2type = {}
self.ent2id = {"<UNK>": 0}
self.space = ""
import os
from flask_restful import Resource
from flask import request
from flask import jsonify
from utils.utils import json_reader, get_movie_by_index, build_movie_hash_table
from utils.trie import Trie
from utils.search import MovieNameStrategy, GenreStrategy, ShowingTimeStrategy, SearchMovie
parent_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
json_path = f'{parent_dir}/mockData/movies.json'
movies = json_reader(json_path)
trie = Trie()
Trie.build_movie_trie(trie, movies)
movie_hash_table = build_movie_hash_table(movies)
class MovieListResource(Resource):
def get(self):
genre = request.args.get('genre')
name = request.args.get('name')
showing_time = request.args.get('showing_time')
data_obj = {'name': name, 'genre': genre, 'showing_time': showing_time}
if name is None and genre is None and showing_time is None:
result = movies
def __init__(self, lower):
self.trie = Trie()
self.ent2type = {} ## word list to type
self.ent2id = {"<UNK>":0} ## word list to id
self.lower = lower
self.space = ""
def clean(self):
self.trie = Trie()
self.ent2type = {}
self.ent2id = {}
self.space = ''