Duplicate files still being downloaded despite MD5 checker active

[FuzzyWuggzy]

I recently started using the MD5 checker during batch downloads so that I wouldn't have to use external tools, but I've discovered that with some sources it doesn't skip the identical post and downloads it anyway, there are no filename conflicts as I use the %website% token.

A quick example is if I search md5:6024ece49f9e4c8067692ee956c55a1f, it returns the file on three sites, konachan.com, tbib.org (non standard site) and yande.re, it downloads the post on the first site, fails the MD5 check and downloads the post on the second, then finally acknowledges that the third is identical and skips it, this seems to be an issue with sources like tbib.org

Here's a log of the download:

Grabber_log.txt

Thank you

[Bionus]

I see two main issues that seem to be causing your problem here:

• You're using the "HTML" parsing mode for tbib.org, which does not return the original MD5 in it, unlike the XML API
• The file hosted on tbib.org for the image with md5 6024ece49f9e4c8067692ee956c55a1f is hosted at https://tbib.org/images/9149/2936e08e1b3e6c8d985cdacfc7ae38a1d1d85065.png, and the Gelbooru HTML parser, by default, assumes the filename to be the MD5 in HTML mode

Any reason you overrode the default tbib.org settings to use HTML over XML?