Thanks for all the interesting details. Lots of stuff to look into when I get a chance. It's been very interesting to see the evolution of how these "big data" companies saw the opportunities to sell the same services they needed to develop for their own business. Like
"Oh, we need to scale our infrastructure to 100x our current size? How about we scale to 1000x then, and sell the other 900x?" Dude, calm down, you're just a a shopping.... hmm, that's pretty nice.
John connor wrote: ↑Wed Dec 19, 2018 6:23 am
I've seen other people use those non-standard phpBB emoticons. Where do they come from?
Input of emoji is done "natively" on many platforms, by which I mean the input method is simply using
the Unicode consortium-defined values that represent the different emoji, in pretty much the same way we all agree on using the Unicode values which represent "A", "B", "C", etc. i.e. The actual message, and the actual input, knows nothing of "an image"; and the emoji is simply just another character value like every other character in the message.
(As opposed to "for each attempt to reference an emoji, actually send an
<a href=>
or
[img]
link to an image they wanted to be placed among the text." Which is the way these kinds of things started out on the web, way back when. And is probably how it's still done now for all the wild "custom emoji images" systems out there, which fall outside of the Unicode-defined specification.)
The first hurdle is for this Unicode character value to be stored correctly as part of the message. This used to not be possible in phpBB, but now (since 3.2) the Unicode character values can be correctly represented & used when writing and viewing messages in phpBB. With the correct Unicode character value being saved and regurgitated now, technically you could just "do nothing else." And simply let the individual browser/device/platform being used to read the message decide how to deal with these Unicode-defined emoji character value references when they show up in the presented html data, same as they have to do with every other character in the data.
But that would allow for varied and inconsistent display of the emojis, since not all platforms render them the same, and not all platforms render them at all. So it's actually in phpBB's
viewing of the message (and not phpBB proper, but actually the s9e framework code phpBB utilizes) that the Unicode-defined character values are "replaced with a consistent set of images to represent the referenced emoticon."
Which -- finally to the answer of your question -- currently involves the s9e framework replacing those Unicode character references
with emojiOne-based images, which hosts/pulls those images from the CDN you identified. And in the future could be replaced with "let's represent such characters using some other consistent mechanism."
So to be clear, the person inputting those emoticon references had no control and no idea how phpBB might actually end up rendering and representing those emoticons once you, John Conner, actually saw & read their message. Those persons are simply on their iPhone or iPad using their iOS input keyboard, or on Andriod using the input keyboard, etc. Or in my case, simply on Windows 10 using the "Windows key + period" entry keyboard. All of which are performing standards-based Unicode-defined emoji entry.
Now, with all that said, if you want to read more from people who actually know what they're talking about, jump in on the following thread:
https://area51.phpbb.com/phpBB/viewtopi ... 20#p301481