mechanize.Browser() 在 Windows 下下载损坏的 JPG，在 Linux 下工作正常吗？
我有以下代码：
self.bg_br = mechanize.Browser()
self.bg_br.retrieve(self.bg_imageurl, "image2.jpg")
self.bg_file2 = open("image.jpg", mode="w")
self.bg_image = self.bg_br.open(self.bg_imageurl).read()
self.bg_file2.write(self.bg_image)
self.bg_file2.close()
问题是 image.jpg （通过 .read() 方法下载的 image.jpg 已损坏。稍后由 Qt 显示（报告“JPG 损坏”错误） ），当我尝试使用 Windows 应用程序打开该文件时，图像打开正常，但
通过 .retrieve 方法，一切正常，但是，图像我。正在下载每次访问链接时都会重新生成 - 所以这不好，
两种方法在 Linux 下都可以正常工作，但是在 Windows 下，就像我说的，第一种方法给出了损坏的图像。
这
I have the following code:
self.bg_br = mechanize.Browser()
self.bg_br.retrieve(self.bg_imageurl, "image2.jpg")
self.bg_file2 = open("image.jpg", mode="w")
self.bg_image = self.bg_br.open(self.bg_imageurl).read()
self.bg_file2.write(self.bg_image)
self.bg_file2.close()
The problem is that the image.jpg (the one that is downloaded via .read() method is corrupted. This is displayed later by Qt (which reports "JPG corrupted" error), and when I try to open the file with a windows application, the image is garbled. It opens fine, but it's garbled.
Via the .retrieve method, everything works fine, HOWEVER, the image I am downloading is generated anew every time you visit the link - so that's no good.
What's even more puzzling is that both methods work just fine under linux, however under windows, like I said, the first method gives a corrupt image.
Help?
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（1）
您需要使用
'wb'模式打开文件来执行bg_file2操作。在 Unix 下，Python 不区分文本文件和二进制文件，但在 Windows 上它会区分。为了支持平台独立性，只需使用'wb'模式以二进制格式写入图像。我不知道 mechanize 的内部原理，但我的猜测是retrieve方法也执行此操作：您只是看不到它，因为它是由库抽象的。有关更多详细信息，请参阅Python 文档的此部分 。
You need to open the file using the
'wb'mode for yourbg_file2operation. Under Unix, Python doesn't differentiate between text and binary files, but on Windows it does. To support platform independence, just use the'wb'mode to write your image in a binary format. I don't know the internals of mechanize, but my guess is that theretrievemethod does this as well: you just don't see it as it is abstracted by the library.For more details, see this section of the Python docs.