I've found that the Twemoji images were being blocked by Fanboy's Annoyances (on sites like Emojipedia, or while using the Twemoji extension for Google Chrome); to reduce false positives without also making hundreds of ordinary whitelist rules for very similar URL patterns, I'm thinking the best approach is a regex:
Code: Select all
@@/1?[a-f0-9]{4}-twitter\.png/$image
All of the Twemoji images have filenames containing patterns of the form <codepoint>-twitter.png (often prepended by filesize indicators and appended by PageSpeed caching indicators), and the codepoint is either a 4-digit hex number, or a 5-digit hex number beginning with 1. Maybe this regex could be further refined, but even the deregifier won't unroll this into a huge set of simple filters.
Now it could be possible for someone to use the Twemoji API to generate a Twemoji whitelist containing simple filters for each Twemoji image, and my guess is that these few hundred filters (one per emoji in the Unicode standard, even fewer because Twemoji doesn't support
all of them) would be faster than that one regex (which is much cruder and corresponds to 2^17 simple filters).
There's a buzzin' in my brain I really can't explain; I think about it before they make me go to bed.