Add database functionality and update image

processing

.gitignore

@@ -129,6 +129,8 @@ ENV/
 env.bak/
 venv.bak/
 
+database
+
 # Spyder project settings
 .spyderproject
 .spyproject

main.py

@@ -1,9 +1,35 @@
+#!/usr/bin/env python3
+
 import io
-import oss2
 import requests
+import oss2
+import plyvel
 
 from PIL import Image, ImageFile
 
+
+# 创建或打开一个数据库
+db = plyvel.DB('database', create_if_missing=True)
+
+'''
+# 写入一个键值对
+db.put(b'key', b'value')
+# 获取一个键的值
+value = db.get(b'key')
+# 删除一个键值对
+db.delete(b'key')
+# 批量写入
+with db.write_batch() as wb:
+    for i in range(10000):
+        wb.put(b'key' + str(i).encode(), b'value' + str(i).encode())
+# 迭代数据库中的所有键值对
+for key, value in db:
+    print(key, value)
+# 关闭数据库
+db.close()
+'''
+
+
 # 读取 .env
 from dotenv import dotenv_values
 config = dotenv_values(".env")
@@ -19,39 +45,44 @@ def download_image(url:str) -> Image.Image:
             oss_auth = oss2.Auth(config['OSS_ACCESS_KEY_ID'], config['OSS_ACCESS_KEY_SECRET'])
             return Image.open(io.BytesIO(oss2.Bucket(oss_auth, f'http://{config["OSS_HOST"]}', config['OSS_BUCKET_NAME']).get_object(url).read()))
         except Exception:
-            print('图片下载失败:', url)
+            print('图片从OSS下载失败:', url)
             return None
     else:
         try:
             response = requests.get(url)
             return Image.open(io.BytesIO(response.content))
         except Exception:
-            print('图片下载失败:', url)
+            print('图片从URL下载失败:', url)
             return None
 
-
 import pymysql
 import pymysql.cursors
 import cnocr
 
+# 打开 mysql
 ocr = cnocr.CnOcr(rec_model_name='ch_PP-OCRv3')
 conn = pymysql.connect(host=config['MYSQL_HOST'], user=config['MYSQL_USER'], password=config['MYSQL_PASSWORD'], database=config['MYSQL_NAME'], cursorclass=pymysql.cursors.DictCursor)
 cursor = conn.cursor()
 cursor.execute("SELECT id, content FROM web_images LIMIT 5")
 
-# 获取查询结果
+# 获取查询结果(跳过下载失败的)
-rows = cursor.fetchall()
+for item in cursor.fetchall():
-for row in rows:
+    image = download_image(item['content'])
-    image = download_image(row['content'])
     if image is None:
-        print('图片下载失败，跳过')
         continue
+    # 将只包含那些非空非纯数字且长度大于1的'text'值
     out = ocr.ocr(image)
-    # 这段代码将只包含那些非空、不是纯数字且长度大于1的'text'值
+    texts = [x['text'] for x in out if x['text'] and not x['text'].isdigit() and len(x['text']) > 1]
-    texts = [item['text'] for item in out if item['text'] and not item['text'].isdigit() and len(item['text']) > 1]
+    print(item['id'], texts)
-    print(row['id'], texts)
 
+    # 将结果存入 leveldb
+    # db.put(str(row['id']).encode(), ','.join(texts).encode())
 
 # 关闭游标和连接
 cursor.close()
 conn.close()
+
+# 关闭数据库
+db.close()
+
+print('Done')