Closed luboq closed 5 years ago
@waldens Your "example post" gives a 404.
I was able to reproduce the issue with standard tumblr-utils, however with the changes proposed by #114 (tested using a git clone of this) I was able to successfully download 319 posts. I'm still not sure why it's less than the expected 700+.
I compared the results to an equivalent download done by TumblThree, and I found that the performance of tumblr-utils was far better -- not only did TumblThree find ~100 less likes, but it also just downloaded a dump of images and videos instead of a proper index. As far as I know, TumblThree scrapes the /liked/by
page directly.
snip