[y2ksw]

Originally Posted by sticky

I understand, but why does it replace incorrectly sometimes?

It works about 90% of the time correctly but sometimes it will replace an image with the wrong image.

And can it be edited to use the same image name as the source?

If, for example, the source does not allow hot linking, they send usually a dummy image of 1x1 pixel in order to save bandwidth. If you have CURL enabled, more images are correctly downloaded, since CURL simulates a browser, while raw sockets (fopen) don't.

If you have known sites where you get always wrong images, you may include them into the Ignore list. This way the images are kept on the original server, although you may loose them over time.