linear whitespace

[Matthias Andree]

David Relson <relson at osagesoftware.com> writes:

> Subject: =?ISO-8859-1?Q?linear?= =?ISO-8859-1?Q?whitespace?=
>
> The current lexer gives "linear whitespace".  Removing spaces would, in
> this case, be wrong.

No, removing spaces would be the right thing to do. Properly decodable
encodings are:

=?ISO-8859-1?Q?linear_?= =?ISO-8859-1?Q?whitespace?=
=?ISO-8859-1?Q?linear?= =?ISO-8859-1?Q?_whitespace?=
=?ISO-8859-1?Q?linear_whitespace?=
=?US-ASCII?Q?linear_whitespace?=

The proper and canonical encoding is none:

linear whitespace

The rule is, encode as little as necessary.

-- 
Matthias Andree