重新生成依赖列表

main.py

@@ -7,10 +7,12 @@ import pymysql
 import cnocr
 import json
 import numpy as np
+import warnings
 from PIL import Image, ImageFile
 from dotenv import dotenv_values
 from elasticsearch import Elasticsearch
 
+warnings.filterwarnings("ignore")
 ImageFile.LOAD_TRUNCATED_IMAGES = True
 config = dotenv_values(".env")
 oss2.defaults.connection_pool_size = 100
@@ -45,15 +47,17 @@ def save_text(conn, id:int, text:str):
 
 def process_images(conn, ocr, es):
     with conn.cursor(pymysql.cursors.SSCursor) as cursor:
-        cursor.execute("SELECT id, content FROM web_images WHERE text!='' LIMIT 10")
+        cursor.execute("SELECT id, content FROM web_images WHERE text='' LIMIT 10000")
         for id, content in cursor.fetchall():
             image = download_image(content)
             if image is None:
                 continue
             item = [x for x in ocr.ocr(image) if x['text'] and not x['text'].isdigit() and len(x['text']) > 1]
-            save_text(conn, id, json.dumps(item, ensure_ascii=False, cls=MyEncoder))
             text = ' '.join([x['text'] for x in item])
+            print(id, text)
+            save_text(conn, id, json.dumps(item, ensure_ascii=False, cls=MyEncoder))
             es.index(index='web_images', id=id, body={'content': text})
+        conn.commit()
 
 def main():
     es = Elasticsearch(config['ELASTICSEARCH_HOST'], basic_auth=(config['ELASTICSEARCH_USERNAME'], config['ELASTICSEARCH_PASSWORD']), verify_certs=False)