Hi, * Andrey R. Urazov [05/11/02 13:36:50 CEST] wrote: > On Sat, May 11, 2002 at 11:59:15AM +0200, Rocco Rutte wrote:
[...] > > Headers have to encoded word by word if there's something in them, > > which has to encoded. > What do you mean, Rocco? Any special encoding? Yes. > Please explain me how > non-ascii characters should be written in the headers? An example is: | Subject: Re: =?iso-8859-1?Q?Datenm=FCll?= umwandeln '=FC' is the encoded character (�). > > If there's something going wrong - as in your case - have a look at > > 'iconv-hook'. > At the moment I do not see any application of the `iconv-hook', If mutt doesn't recognize the character set of the above example, you could map it via an alias to a know one. > it seems > to me that the problem is not in that mutt cannot accomplish conversion > on account of incorrect name of some encoding in my installation (all is > correct here) but because it does not know how to treat symbols in the > header. Could you give an example? > > > I wonder if the standards defining format of e-mail messages allow > > > for usage of non-ASCII characters in mail headers. > > No, and that's why they have to be encoded. > What do you mean by `being encoded'. Well, all headers may only contain ASCII printable characters. With 'encoding' I mean conversion like in the above example. > > The character set for header encoding does not need to be the same as > > for the body. > Rocco, it's absolutely unclear for me how can I include non-ASCII > characters in message headers the right way If you have a mailer like mutt, just type them. Mutt will do the rest of the work. ;-) You'll have to look at the plain text of your message to see those conversion since they're interpreted before they're displayed. > and say other clients how to > interpret them. All you can do is to correctly convert headers; interpretation is left to the other side. RFC2231 describes it. And if the other side's mailer doesn't support RFC2231, you will have to avoid using non-ASCII characters. Cheers, Rocco.
