From 22f370322412074174cde20ecfd14ec03657ab63 Mon Sep 17 00:00:00 2001
From: lyg <1543117173@qq.com>
Date: 星期一, 07 七月 2025 16:20:25 +0800
Subject: [PATCH] 生成数据库
---
knowledgebase/utils.py | 40 ++++++++++++++++++++++++++++++++++------
1 files changed, 34 insertions(+), 6 deletions(-)
diff --git a/knowledgebase/utils.py b/knowledgebase/utils.py
index ab7d2d1..ae9fe1f 100644
--- a/knowledgebase/utils.py
+++ b/knowledgebase/utils.py
@@ -3,6 +3,7 @@
import os
import json
import re
+import asyncio
def get_bit_mask(start, end):
@@ -15,7 +16,7 @@
return mask
-def generate_md5(input_string):
+def generate_text_md5(input_string):
# 鍒涘缓涓�涓� md5 鍝堝笇瀵硅薄
md5_hash = hashlib.md5()
@@ -28,19 +29,46 @@
return md5_digest
+def generate_bytes_md5(input_bytes):
+ # 鍒涘缓涓�涓� md5 鍝堝笇瀵硅薄
+ md5_hash = hashlib.md5()
+
+ # 鏇存柊鍝堝笇瀵硅薄鍐呭
+ md5_hash.update(input_bytes)
+
+ # 鑾峰彇鍝堝笇鍊肩殑鍗佸叚杩涘埗琛ㄧず
+ md5_digest = md5_hash.hexdigest()
+
+ return md5_digest
+
+
+async def get_md5_async(file: bytes):
+ md5 = await asyncio.to_thread(generate_bytes_md5, file)
+ return md5
+
+
+async def save_to_file_async(data: bytes, file_path: str):
+ await asyncio.to_thread(save_bytes_to_file, data, file_path)
+
+
def file_exists(cache_file: str):
return os.path.exists(cache_file)
-def read_from_file(cache_file: str) -> str:
- with open(cache_file, 'r', encoding='utf-8') as f:
+def read_from_file(file: str) -> str:
+ with open(file, 'r', encoding='utf-8') as f:
text = f.read()
return text
-def save_to_file(text, cache_file):
- with open(cache_file, 'w', encoding='utf-8') as f:
+def save_text_to_file(text: str, file: str):
+ with open(file, 'w', encoding='utf-8') as f:
f.write(text)
+
+
+def save_bytes_to_file(bytes_data: bytes, file: str):
+ with open(file, 'wb') as f:
+ f.write(bytes_data)
def replace_tpl_paras(tpl_text: str, data: dict):
@@ -58,4 +86,4 @@
:param text:
:return:
"""
- return re.sub(r'[\\/:*?"<>|]', '_', text)
+ return re.sub(r'[\t\\/:*?"<>|]', '_', text)
--
Gitblit v1.9.1