def add_and_delete_word():
'''
作为api，实时往字典中增添关键词等
'''
if request.method == 'GET':
word = request.args.get('word', '').strip()
stop_word = request.args.get('stopword', '').strip()
if word == '' and stop_word == '':
print('word is None, add fail.')
raise exceptions.NotAcceptable()
if stop_word:
segment_.add_stop_word(stop_word)
print(f'add stopword: {stop_word} successfully')
return dumps({'status': 'ok'})
# 设置默认权重为2
weight = request.args.get('weight', 2)
weight = float(weight)
segment_.add_word(word, weight)
print(f'word is: {word}, weight is {weight}. add success.')
return dumps({'status': 'ok'})
elif request.method == 'DELETE':
word = request.args.get('word', '').strip()
stop_word = request.args.get('stopword', '').strip()
if word == '' and stop_word == '':
print('word is None, delete fail.')
raise exceptions.NotAcceptable()
if stop_word:
segment_.delete_stop_word(word)
print(f'delete stopword: {stop_word} successfully')
return dumps({'status': 'ok'})
segment_.delete_word(word)
print(f'delete {word} success')
return dumps({'status': 'ok'})
def select_renderer(self, renderers):
"""
Determine which renderer to use for rendering the response body.
Returns a two-tuple of (renderer, content type).
"""
accept_header = request.headers.get('Accept', '*/*')
for client_media_types in parse_accept_header(accept_header):
for renderer in renderers:
server_media_type = MediaType(renderer.media_type)
for client_media_type in client_media_types:
if client_media_type.satisfies(server_media_type):
if server_media_type.precedence > client_media_type.precedence:
return (renderer, server_media_type)
else:
return (renderer, client_media_type)
raise exceptions.NotAcceptable()
def search():
'''
提供api给调用，对传入的文本进行分词检索， 并返回结果
'''
# 获取需要分词的句子
text = request.args.get('text', '')
text = text.strip()
print('question is:', text)
# 传入参数为空抛出异常
if not text: raise exceptions.NotAcceptable()
# 获取分词
jieba_seg, jieba_keywords = get_jieba_seg(text)
fool_seg, fool_keywords = get_fool_seg(text)
# 将分词结果中的词替换为数据库中存在的词
jieba_seg, jieba_index = api.replace(jieba_seg)
adding_keywords = [jieba_seg[index] for index in jieba_index]
# 将关键词中的词也替换为数据库中存在的词， 并将分词中存在的关键词加入到关键词中
jieba_keywords, jieba_index = api.replace(jieba_keywords)
adding_keywords += jieba_keywords
jieba_keywords = adding_keywords
# 去重 保持有序
unique = set()
order_seg = []
for seg in jieba_seg:
if seg not in unique:
order_seg.append(seg)
unique.add(seg)
jieba_seg = order_seg
unique.clear()
order_seg = []
for seg in jieba_keywords:
if seg not in unique:
order_seg.append(seg)
unique.add(seg)
jieba_keywords = order_seg
jieba_keywords = api.remove(jieba_keywords)
# 将分词结果中的词替换为数据库中存在的词
fool_seg, fool_index = api.replace(fool_seg)
adding_keywords = [fool_seg[index] for index in fool_index]
# 将关键词中的词也替换为数据库中存在的词， 并将分词中存在的关键词加入到关键词中
fool_keywords, fool_index = api.replace(fool_keywords)
adding_keywords += fool_keywords
fool_keywords = adding_keywords
# 去重 保持有序
unique = set()
order_seg = []
for seg in fool_seg:
if seg not in unique:
order_seg.append(seg)
unique.add(seg)
fool_seg = order_seg
unique.clear()
order_seg = []
for seg in fool_keywords:
if seg not in unique:
order_seg.append(seg)
unique.add(seg)
fool_keywords = order_seg
# 使用es根据关键词进行检索
jieba_result = api.query(jieba_keywords)
fool_result = api.query(fool_keywords)
# 构造结果作为api返回
jieba_result = api.format(jieba_result)
fool_result = api.format(fool_result)
# 构造结果将结果存入数据库中
jieba_content = [
f'{index}.{data["phoneNum"]},{data["allName"]}\n'
for index, data in enumerate(jieba_result, 1)
]
fool_content = [
f'{index}.{data["phoneNum"]},{data["allName"]}\n'
for index, data in enumerate(fool_result, 1)
]
jieba_seg = ','.join(jieba_seg)
jieba_keywords = ','.join(jieba_keywords)
fool_seg = ','.join(fool_seg)
fool_keywords = ','.join(fool_keywords)
jieba_content = ''.join(jieba_content)
fool_content = ''.join(fool_content)
# 需要存入数据库中的结果
save = [
text, jieba_seg, fool_seg, jieba_keywords, fool_keywords, fool_content,
jieba_content
]
Thread(target=api._save_query, args=(*save, )).start()
# api._save_query(*save)
return dumps({
'jiebaSeg': jieba_seg,
'jiebaKeywords': jieba_keywords,
'foolSeg': fool_seg,
'foolKeywords': fool_keywords,
'jiebaSearchResult': jieba_result,
'foolSearchResult': fool_result,
})