in reply to Re^2: Decoding an email body, into utf8
in thread Decoding an email body, into utf8

And what output do you expect? It would help to see this in the form of a test eg. How to ask better questions using Test::More and sample data

Note that you will need a utf-8 capable terminal and have the correct locale set in order to view utf-8 data (which the output appears like it might be).

(updated: added link to test example)

  • Comment on Re^3: Decoding an email body, into utf8

Replies are listed 'Best First'.
Re^4: Decoding an email body, into utf8
by ultranerds (Hermit) on Jul 22, 2016 at 13:50 UTC
    Ah, that was it. Bit surprised, as that section shouldn't really have any utf8 in it (just plain text)

    Anyway, all I need to do now is look out for what the email encoding is, and if utf8 convert it.

    Thanks!

    Andy
[reply]

      Words like résumé and piñon are plain text and many punctuation marks are too (dashes, ellipsis, real quotes and apostrophes, etc). UTF-8 is the best (default) encoding for plain text.

[reply]
Re^4: Decoding an email body, into utf8
by ultranerds (Hermit) on Jul 27, 2016 at 05:34 UTC

    Mmmm actually, that doesn't work in some cases:

        my $name = "From: =?UTF-8?B?QW5keSBOZXdieSDDrcOpw7M=?= <andy\@cham
+bresdhotesfrance.com>";

    use MIME::QuotedPrint;

    if ($name  =~ /utf-8|utf8/i) {
        $name= decode_qp($name);
        $name =~ s/([\200-\377]+)/from_utf8({ -string => $1, -charset 
+=> 'ISO-8859-1'})/eg;
    }

    print $IN->header;
    print "NOW: $name";

    [download]

    Prints out:

    NOW: From: =?UTF-8?B?QW5keSBOZXdieSDDrcOpw7M=?=
    ...instead of what I was expecting:

    Andy Newby íóé

    This is a valid email header passed through from Thunderbird. Any ideas why it won't decode?

    Cheers

    Andy
[reply]
[d/l]

      Interesting. Just found a post on StackOverflow, where someone suggested using decode_base64 to decode it. And that seems to work:

          my $name = "=?UTF-8?B?QW5keSBOZXdieSDDrcOpw7M=?=";

    use MIME::Base64;
    $name =~ s|\Q=?UTF-8?B?||i;
    $name = decode_base64($name);
    $name =~ s/([\200-\377]+)/from_utf8({ -string => $1, -charset => '
+ISO-8859-1'})/eg;

      [download]

      I wonder why the other module doesn't work?

      Cheers

      Andy
[reply]
[d/l]
        RFC 2047, section 4 says it. The "?B?" part tells you to use base 64, while "?q?" means "quoted-printable".

        Since you already know CPAN: search term is rfc 2047

[reply]

In Section Seekers of Perl Wisdom