Charset Nightmare...

[rocketchef]

Does anyone have advice in relation to character sets used by form submissions?

I have an email application that is generating wierd characters in certain email clients.

Typical usage is for someone to cut'n'paste something from a Word Document into one of my forms. In certain email clients, (such as Novell GroupWise Internet Agent) speechmarks generate wierd characters.

" prints fine
” does not
(you may need to view source to see difference)

I guess this is because Word generates unicode, but the decode function doesn't make a difference. Any advice on detecting character sets etc? I've checked the manual but I can't see anything obvious...

Cheers,
Ben

[Anon]

Its more to do with the smart quotes in word than the character set. If you look at the two things as hex values you will find that they are different. You might be able to deal with this by converting the entire submitted form into hexcode and then dropping or replacing any that sit outside of the printable range. I think its codes 28 to 117 in decimal. But I might be wrong. Hope that helps

Mark.