On Tue, 13 Apr 2004, dc wrote:
> there are services that map emoji across carriers, so i doubt the
> gateway changes the mapping.
It does change the mapping.
> I think this is just because you dont have the PC version of the kanji
> fonts on your system (probaly cos its linux in your case) so the default
> mapping for those characters is up in high-digit unicode land where the
> default (on japanese systems) is a 〓〓 eq symbol... so they all look
> the same.
I'm not just viewing the messages on my PC; I'm also viewing them on
my phone. When a message with emoji comes from a docomo phone, passes
through my mail server, and comes back to my docomo phone, I can't
see the emoji any more; instead I see "〓". If we look at a hex dump
of the copy that lands in my computer mailbox, we can see that every
emoji is now the same character. For example, here's a message saying
"hiカート123" where the 1, 2 and 3 are the first three emoji in the
emoji menu on my phone (heart, beating heart, and broken heart):
000003d0 34 39 66 31 63 35 0a 0a 68 69 1b 24 42 25 2b 21 |49f1c5..hi.$B%+!|
000003e0 3c 25 48 22 2e 22 2e 22 2e 1b 28 42 0a |<%H"."."..(B.|
The message starts after the two newlines, "0a 0a". Then we have "hi"
(68 69), an ISO-2022-JP shift-into-Japanese escape sequence (1b 24 42),
"カート" (25 2b, 21 3c, 25 48) and the three emoji, 22 2e, 22 2e, 22 2e,
all three the same. The message ends with an ISO-2022-JP "switch back to
ASCII" escape sequence (1b 28 42) and a newline (0a).
My MTA is not doing any conversion on the message body; it treats it as
binary.
cjs
--
Curt Sampson <cjs_at_cynic.net> +81 90 7737 2974 http://www.NetBSD.org
Don't you know, in this new Dark Age, we're all light. --XTC
Received on Wed Apr 14 03:15:13 2004