Ok, So we have a script that takes emails sent to thunderbird, convertes part of the message to html and saves it to a MySQL. Every file, every part written is set to UTF-8. Finally, on my end of the work, the CRM (written in PHP5.3 expected output Chrome and Firefox), I pull the message, along with other info and display something resembling GMail, but as a "task list" for our employees.
The problem I'm having, if you havn't guessed already, some customer emails are obviously using different encodings. Thus, some (not all, and certainly not majority) of the e-mails don't show all characters correctly.
At first I made use of utf8_encode to get the email messages to look right, and this helps with most email messages coming from the database, however, a few slip by with bad characters.
In the DB these "bad apostrophes" appear as ’, but after utf8_encode they come through as �??. I've tried various encoding things to guess and change as needed, however, this tends to hurt the vast majority of the other emails.
Any suggestions, on one end of the pipe or the other, how I might get these few emails to match everything else, or how i might at least create a possible preg_replace filter at the end or something?
update
it seems even the emails with bad characters are passed to end php as utf-8 according to mb_detect_encoding. This is before any extra encoding. iconv does detect the ones that ahve problems, but this really gives me no way to solve them and just puts a php error box up on the screen instead of a simple FALSE return that it says it's supposed to give, so this too seems to be no solution.
