Re: [MacPerl] bug in MIME::QuotedPrint ?

[Axel Rose <rose@sj.com>]

Thank you Ronald and Paul for responding.

You are right from your point of view.
Nearly every 8bit character has a different representation on
different systems with different character sets.

A german umlaut "ä" happens to be 0xe4 in ISO-8859-P1 charset
(also Latin-1, Windows) but is 0x8a in the Mac charset and
certainly something weird in EBCDIC.

But the whole point of MIME encoding is to transport such characters
between different systems.

As far as I understand, to stick with my example, an "ä" has to be
encoded into "=E4" (same as the HTML entity "ä").
As in HTML it is the receivers task to decode into his local
charset.

If the MacPerl MIME::QuotedPrint doesn't transform a "ä" into
a "=E4" I consider this a major malfunction.

Another test case - I send the =8A to myself, reading with Mac
Eudora, I see a "..." character, called a dieresis, option-.

MacPerl's ord() function is not portable with 8bit characters.
I have no idea whether this is known nor do I have a workaround yet.


Regards


Axel


>#!perl -w
>
>use MIME::QuotedPrint;
>
>$string = "ä";
>$estring = encode_qp( $string );
>print "encode_qp( $string ) = $estring\n";
>
>__END__
>
>Result is
>encode_qp( ä ) = =8A
>
>Running under Linux I get the proper result
>encode_qp( ä ) = =E4
>
>
>Looking into the source code I find
>...
>$res =~ s/([^ \t\n!-<>-~])/sprintf("=%02X", ord($1))/eg;  # rule #2,#3
>$res =~ s/([ \t]+)$/
>	join('', map { sprintf("=%02X", ord($_)) }
>		split('', $1)
>      )/egm;                        # rule #3 (encode whitespace at eol)
...

----------------------------------------------------------------------
Axel Rose, Springer & Jacoby Digital GmbH & Co. KG, mailto:rose@sj.com
  pub PGP key 1024/A21CB825 E0E4 BC69 E001 96E9  2EFD 86CA 9CA1 AAC5
  "... denn alles, was entsteht, ist wert, daß es zugrunde geht ..."

# ===== Want to unsubscribe from this list?
# ===== Send mail with body "unsubscribe" to macperl-request@macperl.org