On Tue, 29 Jun 2004, Mike Taylor wrote:
>> From: Robert Sanderson <[log in to unmask]>
> (or whatever). There is no analogous flexibility in scan responses --
> and I am inclined to think that is as it should be, contra Matthew's
> proposal for solving this problem.
I don't think that making <value> the same type as <recordData> is the way
to go either, TBH.
> Perhaps CQL itself should be more specific about what is allowed in
> identifiers? As presently written, the specification allows control
> characters in index names, for example.
Printable characters only, preferably simple ascii or UTF-8, but
applications should be prepared to accept UTF-16 as well?
> I disapprove heartily of this -- you're conflating two essentially
> unrelated issues here: what kind of thing the term is (word, image,
> audio-file) and how it's encoded for transfer. Which is of course
> precisely why MIME defines two separate headers.
> I think we should just have two new relation modifiers:
> which indicate the encoding of the term; and if it turns out in the
> future that there is also a need to convey content-type, then that
> should be handled by a new relation modified when it's needed.
> foo.image =/cql.base64/cql.content-type="image/gif" "R0lGODlh..."
eg to allow:
# 7-Bit - Data is sent as US-ASC data.
# 8-Bit - 8 bit characters are included in short lines.
# BASE64 - Used for binary files. Three bytes are transformed into 4 ASC
characters in lines limited to a length of 76 characters.
# Binary - Long lines are sent using 8 bit characters. These lines may not
be transportable using SMTP.
# Quoted-Printable - Used for ASC text. Line length is limited to 76
# X-Token - Defines private encoding values prefixed with an "X-".
,'/:. Dr Robert Sanderson ([log in to unmask])
,'--/::(@)::. Special Collections and Archives, extension 3142
,'---/::::::::::. University of Liverpool
I L L U M I N A T I L5R Shop: http://www.cardsnotwords.com/