Imgur recently announced that new Terms of Service will go into effect on May 15, 2023, and they will delete old content that is not associated with any user account, as well as any content containing nudity.
Consequently, old forum threads and chat logs may soon stop making sense if they contain links to Imgur.
I already have saved copies of most of my chat logs, so I used grep to extract
from them a list of Imgur URLs, one per line. I use the script download.py
from this repository, passing the file with the list of URLs as standard input,
to download the images from Imgur.
For each image URL of the form *.imgur.com/IMAGE_ID or
*imgur.com/IMAGE_ID.EXTENSION, the script saves a file called IMAGE_ID.ext,
where ext is autodetected. For animated GIFs, the webm format is preferred,
with a fallback to mp4.
For an album located at *.imgur.com/a/ALBUM_ID, the script saves a file called
ALBUM_ID.zip. For a gallery located at *.imgur.com/a/GALLERY_ID, the script
attempts to download and save GALLERY_ID.zip, but if this fails because the
gallery only has a single image and no corresponding album, the script
autodetects the url of that single image, downloads it, and then renames it so
that it starts with GALLERY_ID.
All files are downloaded to the current working directory. If the script detects that there is already a file with the correct name (not including its extension), that URL is skipped.