mirror of
https://git.mirrors.martin98.com/https://github.com/infiniflow/ragflow.git
synced 2025-05-25 15:48:07 +08:00
f1c98aad6b
### What problem does this PR solve? _Briefly describe what this PR aims to solve. Include background context that will help reviewers understand the purpose of the PR._ ### Type of change - [x] Documentation Update - [x] Refactoring --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
65 lines
1.6 KiB
Python
65 lines
1.6 KiB
Python
import json
|
|
import os
|
|
import time
|
|
import logging
|
|
import re
|
|
|
|
from api.utils.file_utils import get_project_base_directory
|
|
|
|
|
|
class Dealer:
|
|
def __init__(self, redis=None):
|
|
|
|
self.lookup_num = 100000000
|
|
self.load_tm = time.time() - 1000000
|
|
self.dictionary = None
|
|
path = os.path.join(get_project_base_directory(), "rag/res", "synonym.json")
|
|
try:
|
|
self.dictionary = json.load(open(path, 'r'))
|
|
except Exception as e:
|
|
logging.warning("Missing synonym.json")
|
|
self.dictionary = {}
|
|
|
|
if not redis:
|
|
logging.warning(
|
|
"Real-time synonym is disabled, since no redis connection.")
|
|
if not len(self.dictionary.keys()):
|
|
logging.warning(f"Fail to load synonym")
|
|
|
|
self.redis = redis
|
|
self.load()
|
|
|
|
def load(self):
|
|
if not self.redis:
|
|
return
|
|
|
|
if self.lookup_num < 100:
|
|
return
|
|
tm = time.time()
|
|
if tm - self.load_tm < 3600:
|
|
return
|
|
|
|
self.load_tm = time.time()
|
|
self.lookup_num = 0
|
|
d = self.redis.get("kevin_synonyms")
|
|
if not d:
|
|
return
|
|
try:
|
|
d = json.loads(d)
|
|
self.dictionary = d
|
|
except Exception as e:
|
|
logging.error("Fail to load synonym!" + str(e))
|
|
|
|
def lookup(self, tk):
|
|
self.lookup_num += 1
|
|
self.load()
|
|
res = self.dictionary.get(re.sub(r"[ \t]+", " ", tk.lower()), [])
|
|
if isinstance(res, str):
|
|
res = [res]
|
|
return res
|
|
|
|
|
|
if __name__ == '__main__':
|
|
dl = Dealer()
|
|
print(dl.dictionary)
|