Update 2magiDialogCut.py

2magiDialogCut.py

@@ -3,27 +3,34 @@ import requests
 import json
 from PIL import Image
 import os
-
-
-def getImgCoordinatesByMagi(image_path:"包含后缀的文件路径") -> "全部对话坐标list,失败返回none":
-    # headers = {
-    #     'authority': 'rogerxavier-fastapi-t5-magi.hf.space',
-    #     'method': 'GET',
-    #     'scheme': 'https',
-    #     'Accept': '*/*',
-    #     'Accept-Encoding': 'gzip, deflate, br, zstd',
-    #     'Accept-Language': 'zh-CN,zh;q=0.9',
-    #     'Cookie': 'spaces-jwt=eyJhbGciOiJFZERTQSJ9.eyJyZWFkIjp0cnVlLCJwZXJtaXNzaW9ucyI6eyJyZXBvLmNvbnRlbnQucmVhZCI6dHJ1ZX0sIm9uQmVoYWxmT2YiOnsia2luZCI6InVzZXIiLCJfaWQiOiI2NDJhNTNiNTE2ZDRkODI5M2M5YjdiNzgiLCJ1c2VyIjoicm9nZXJ4YXZpZXIifSwiaWF0IjoxNzE2NjExMTE3LCJzdWIiOiIvc3BhY2VzL3JvZ2VyeGF2aWVyL2Zhc3RhcGlfdDVfbWFnaSIsImV4cCI6MTcxNjY5NzUxNywiaXNzIjoiaHR0cHM6Ly9odWdnaW5nZmFjZS5jbyJ9.W00jo8kiRgwCpq5aaGhaPE2RP2jLOyvfimjyIfHVhP1gs7NHkBkRzVTFqYv3TRtZoHNPTiFiI5Ehu12KP06sDQ',
-    #     'Dnt': '1',
-    #     'Priority': 'u=1, i',
-    #     'Sec-Ch-Ua': '"Chromium";v="124", "Google Chrome";v="124", "Not-A.Brand";v="99"',
-    #     'Sec-Ch-Ua-Mobile': '?0',
-    #     'Sec-Ch-Ua-Platform': '"Windows"',
-    #     'Sec-Fetch-Dest': 'empty',
-    #     'Sec-Fetch-Mode': 'cors',
-    #     'Sec-Fetch-Site': 'same-origin',
-    #     'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36'
-    # }
+from faker import Faker
+fake = Faker()
+fake_user_agent = fake.user_agent()
+fake_headers = {
+    'scheme': 'https',
+    'Accept': '*/*',
+    'Accept-Encoding': 'gzip, deflate, br, zstd',
+    'Accept-Language': 'zh-CN,zh;q=0.9',
+    'Dnt': '1',
+    'Priority': 'u=1, i',
+    'Sec-Ch-Ua': '"Chromium";v="124", "Google Chrome";v="124", "Not-A.Brand";v="99"',
+    'Sec-Ch-Ua-Mobile': '?0',
+    'Sec-Ch-Ua-Platform': '"Windows"',
+    'Sec-Fetch-Dest': 'empty',
+    'Sec-Fetch-Mode': 'cors',
+    'Sec-Fetch-Site': 'same-origin',
+    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36'
+}
+fake_headers['User-Agent'] = fake_user_agent
+
+
+def getImgCoordinatesByMagi(image_path:"包含后缀的文件路径",fake_headers:dict=None) -> "全部对话坐标list,失败返回none":
+    # 创建一个Session对象
+    session = requests.Session()
+    if fake_headers is not None:
+        session.headers = fake_headers
+        
+    
     with open(image_path, 'rb') as file:
         image_bytes = file.read()
     files = {
@@ -31,8 +38,8 @@ def getImgCoordinatesByMagi(image_path:"包含后缀的文件路径") -> "全部
     }
     magi_fastapi_base = 'https://rogerxavier-fastapi-t5-magi.hf.space/getCoordinates'
     try:
-        # resp = requests.post(magi_fastapi_base, files=files, headers=headers) ##是仓库public后好像就api可用了
-        resp = requests.post(magi_fastapi_base, files=files) ##不使用header看看会不会还限制
+
+        resp = session.post(magi_fastapi_base, files=files) ##用faker header试试
         print(resp.json())
         return resp.json()["panels"]
     except Exception as e:
@@ -59,23 +66,10 @@ def save_img(new_save_path:"新文件的保存路径(包含后缀)",old_img_path
     print(new_save_path)
     # 原始照片文件名
     original_image = old_img_path
-    #打开原始照片
-    # img = Image.open(original_image)
-    # text_bboxes_for_all_images = getImgCoordinatesByMagi(original_image)
-    # if text_bboxes_for_all_images is not None:
-    #     for index, box in enumerate(text_bboxes_for_all_images):
-    #         cropped_img = img.crop(tuple(box))
-    #         # 保存裁剪后的图片，并按照数字大小依次排序命名(包括本身的也要3d,比如 0.jpg - >000.000_cropped.jpg)
-    #
-    #         cropped_img.save(f"{os.path.splitext(new_save_path)[0]}_{index:03d}_cropped.jpg")
-    # else:
-    #     print("图片识别有问题,准备删除")
-    #     os.remove(original_image)
-
 
     ##防止文件打开无法删除
 
-    text_bboxes_for_all_images = getImgCoordinatesByMagi(original_image)
+    text_bboxes_for_all_images = getImgCoordinatesByMagi(image_path =original_image,fake_headers = fake_headers)
     if text_bboxes_for_all_images is not None:
         with Image.open(original_image) as img:
             for index, box in enumerate(text_bboxes_for_all_images):