On Wednesday, 29. November 2000 14:39, Wolfgang Rohdewald wrote:
>
> I have a multipart/alternative message which does not define a charset in
> the main headers, but the message part does. The charset of the message
> part is not used by KMMessage::asQuotedString. This patch uses the charset
> of the part if and only if the main headers don't define it.
>
> Problem (why I found this): IF codecForLocale() is used, the umlauts
> appear rather strangely in the quoted text. The UTF-8 codec is used.
> An a umlaut (  and the next char become =EF=BF=N^H=BD (quoted-printable)
> or ? (8bit)
>
> Why ? I suspect this is a known problem.
>
>
> --- ../kmail.org/kmmessage.cpp  Wed Nov 29 14:19:32 2000
> +++ kmmessage.cpp       Wed Nov 29 14:21:31 2000
> @@ -619,6 +619,13 @@
>          isInline = (stricmp(msgPart.contentDisposition(), "inline") == 0);
>
>        if (isInline) {
> +        if (cset.isEmpty() || cset.isNull()) {
> +          cset = msgPart.charset();
> +          if (!cset.isEmpty() && !cset.isNull())
> +            codec = KGlobal::charsets()->codecForName(charset());
> +          else
> +            codec = QTextCodec::codecForLocale();
> +        }
>          if (stricmp(msgPart.typeStr(),"text") == 0 ||
>              stricmp(msgPart.typeStr(),"message") == 0) {
>            Kpgp* pgp = Kpgp::getKpgp();

Committed, thanks for the patch.

BTW: If a QString isNull() then it is also isEmpty().

Regards,
Michael Häckel
_______________________________________________
Kmail Developers mailing list
Kmail@master.kde.org
http://master.kde.org/mailman/listinfo/kmail