UTF-8 in e-mail subject - quoted printable?

Kukulkan · Post by **Kukulkan** » Tue Feb 26, 2008 3:30 pm

Hi,

I use UTF-8 encoded templates to design messages for sending using gnozals PureSMTP. The used subject is encoded in this template, too. I set

Content-Type: text/html; charset=utf-8;

and the complete body is ok. Using Outlook and Thunderbird, the UTF-8 encoded subject is correct, too. But some e-mail programs (tobit david for example) fail to display the UTF-8 encoded subject of the e-mail.

How do I need to convert the subject for all kind of e-mail clients? Is there a quoted printable encoding algorithm available for PureBasic?

Kukulkan

pdwyer · Post by **pdwyer** » Wed Feb 27, 2008 2:09 am

My understanding is that the subject of an email must be in 7bit ascii to be standard. For subjects that contain other codings (like UTF8 etc) they have a type marker and then the bytes encoded in base64.

like this

Subject: =?utf-8?B?UkEU6IEVNUyAtIFRFU1Q5IOCuOOCueOBmOOFg+OBquOBj+OBpg==?=
=?utf-8?B?

Kukulkan · Post by **Kukulkan** » Wed Feb 27, 2008 9:19 am

Hi pdwyer,

Thank you. Your information completed the puzzle.

I found the following way to manage the problem using PB:

Code: Select all

Subject.s = "UTF-8 encoded string like VersandbestÃ¤tigung"
SubjectBase64.s = Space(Len(Subject.s) * 2)
Base64Encoder(@Subject.s, Len(Subject.s), @SubjectBase64.s, Len(SubjectBase64.s))
Subject.s = "=?UTF-8?B?" + SubjectBase64.s + "?="

Seems to work without any problems...

Kukulkan

srod · Post by **srod** » Wed Feb 27, 2008 10:06 am

Useful. Thanks.

pdwyer · Post by **pdwyer** » Wed Feb 27, 2008 10:16 am

Cheers, I was wondering if that would work...

Now I don't need to spend the time! Thanks

Seymour Clufley · Post by **Seymour Clufley** » Tue Dec 30, 2008 9:59 pm

Is there a quoted printable encoding algorithm available for PureBasic?

Now I need such a function!

I found this VB function but it involves using something called "CDO.Message object".

Can anyone help out?

Kukulkan · Post by **Kukulkan** » Wed Dec 31, 2008 11:33 am

Hi,

Quoted Printable seem to be less complicated than UTF-8. The result may be such a string:

Original: Hätten Hüte ein ß im Namen, wären sie möglicherweise keine Hüte mehr, sondern Hüße.

QP: H=E4tten H=FCte ein =DF im Namen, w=E4ren sie m=F6glicherweise keine H=FCte= mehr, sondern H=FC=DFe.

Every byte that is not between 33–60 and 62–126 will be replaced by = followed by the uppercase hex value.

Here is more information:
http://en.wikipedia.org/wiki/Quoted-printable

Will be only a work of some minutes.

Kukulkan