Hi vedaal--
On Tue 2021-02-16 14:29:19 -0500, vedaal(_at_)nym(_dot_)hush(_dot_)com wrote:
On 2/16/2021 at 12:08 PM, "Daniel Kahn Gillmor" wrote:
Proposals for additional cases to test would also be useful.
=====
PDF renders an empty line as whitespace, even if there is a line
return.
This functionally means that the empty line after the PGP header and
the Cyphertext
will be interpreted as being omitted, even if copied from the PDF.
I don't know what PDF you're talking about. Are you referring to a
specific example PDF? to a (common?) form of translation between ASCII
text streams and PDFs? to something else?
If it is possible, I would suggest that both cases be accepted,
i.e. with a LR/empty line after the header and the cyphertext,
and if no such empty line exists.
The goal of the whitespace canonicalization as i understand it is
*exactly* this -- that regardless of how the unprintable characters show
up (or don't), a valid signature will be seen as valid.
However, the thing that is validated is the canonicalized version, not
any of the potential whitespace variants.
--dkg
signature.asc
Description: PGP signature
_______________________________________________
openpgp mailing list
openpgp(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/openpgp