Re: [Nmh-workers] base64 ... just looking for advice

[David Levine]

Wolfgang wrote:

> Conversion attempt 1:
> 
>       $ mhfixmsg +inbox 174 -decodetext 8bit -verbose -outfile ...
>       mhfixmsg: 174 part 2, decode text/plain; charset="iso-8859-1"
>       mhfixmsg: 174 part 1, decode text/html; charset="iso-8859-1"
> 
> The resulting file is identified by "file" as "ISO-8859 text, with
> very long lines"; it basically looks OK, except that I actually would
> prefer to have it converted to UTF8 instead.

Right, the mhfixmsg default is -notextcharset.   Its man page shows the 
defaults.  To change that, in effect, you can add a -textcharset switch to your 
profile, e.g.,

mhfixmsg: -textcharset utf8

> Conversion attempt 2:
> 
>       $ mhfixmsg +inbox 174 -decodetext 8bit -textcharset utf8 -verbose -outf
> ile ...
> 
> The resulting file is identified by "file" as "Non-ISO extended-ASCII
> text, with very long lines".  Opening the file in "vim" shows that all
> the 8 bit characters like the German umlaut characters are wrong;
> the quoted-printable encoded "f=FCr" for example was not converted
> into "für" as expected, but instead into "für".
> 
> What is needed to properly decode everything into UTF8?

Here's my guess at what's happening:  mhfixmsg only converts the character sets 
of text/plain parts.  Did you look at that or the text/html part in vim?

We could remove that text/plain restriction and let mhfixmsg convert any text 
part.  I don't know if that's a good idea or not.

I assume that your message is a multipart/alternative.  If not, the output from 
mhlist -verbose might be helpful.  You can also do things like :

    $ mhfixmsg +inbox 174 -textcharset utf8 -verbose -outfile - | mhlist 
-verbose -file -

David