Hello, I have an issue with gpg signing emails with sup. The header of sent emails look like this: Content-Type: multipart/signed; protocol=""application/pgp-signature""; boundary="=-1204074034-949437-9467-7738-1-="; micalg="pgp-sha1" Note the double quotation marks around application/pgp-signature. (I checked it in the "Sent", and of course on my other e-mail account.) Apparently they throw off at least mutt and claws-mail, so that they don''t recognize the protocol for the attachment. However, if I manually edit the header to have only one set of quotation marks, the signature is recognized. I took a peek on the sources (crypto.rb), and I don''t know how the double quotation marks could be created. Anybody else having this problem? Thanks, Jan --- . gpg key: http://freeshell.de/~teatime/teatime.sig -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 197 bytes Desc: not available Url : http://rubyforge.org/pipermail/sup-talk/attachments/20080226/134de07a/attachment.bin
Excerpts from Jan Spakula''s message of Wed Feb 27 04:40:54 +0100 2008:> Hello, > > I have an issue with gpg signing emails with sup. The header of sent > emails look like this: > Content-Type: multipart/signed; protocol=""application/pgp-signature""; boundary="=-1204074034-949437-9467-7738-1-="; micalg="pgp-sha1" > Note the double quotation marks around > application/pgp-signature. (I checked it in the "Sent", and of course on > my other e-mail account.) Apparently they throw off at least mutt and > claws-mail, so that they don''t recognize the protocol for the > attachment. However, if I manually edit the header to have only one set > of quotation marks, the signature is recognized. > > I took a peek on the sources (crypto.rb), and I don''t know how the > double quotation marks could be created. Anybody else having this > problem?That''s a ruby-mail bug with a working patch [1], however ruby-mail is no longer actively maintained :( [1]: "[2661] set_boundary doubles quotes in params which already contained a quote" http://rubyforge.org/tracker/index.php?func=detail&aid=2661&group_id=446&atid=1756 -- Nicolas Pouillard aka Ertai -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 194 bytes Desc: not available Url : http://rubyforge.org/pipermail/sup-talk/attachments/20080227/94488b5f/attachment.bin
Reformatted excerpts from nicolas.pouillard''s message of 2008-02-27:> That''s a ruby-mail bug with a working patch [1], however > ruby-mail is no longer actively maintained :(I did hear from Matt Armstrong recently that he was starting to maintain it again. At least, he released a 1.0.0 last month that has Ruby 1.9 modifications. But nothing since.> [1]: "[2661] set_boundary doubles quotes in params which already contained a > quote" > http://rubyforge.org/tracker/index.php?func=detail&aid=2661&group_id=446&atid=1756Does the following workaround help? (Completely untested.) diff --git a/lib/sup/crypto.rb b/lib/sup/crypto.rb index 9e16132..ee06958 100644 --- a/lib/sup/crypto.rb +++ b/lib/sup/crypto.rb @@ -40,7 +40,7 @@ class CryptoManager raise Error, (output || "gpg command failed: #{cmd}") unless $?.success? envelope = RMail::Message.new - envelope.header["Content-Type"] = ''multipart/signed; protocol="application/pgp-signature"; micalg=pgp-sha1'' + envelope.header["Content-Type"] = ''multipart/signed; protocol=application/pgp-signature; micalg=pgp-sha1'' envelope.add_part payload signature = RMail::Message.make_attachment output, "application/pgp-signature", nil, "signature.asc" -- William <wmorgan-sup at masanjin.net>
Excerpts from William Morgan''s message of Wed Feb 27 11:50:37 -0600 2008:> Reformatted excerpts from nicolas.pouillard''s message of 2008-02-27: > > [1]: "[2661] set_boundary doubles quotes in params which already contained a > > quote" > > http://rubyforge.org/tracker/index.php?func=detail&aid=2661&group_id=446&atid=1756 > > Does the following workaround help? (Completely untested.) > ...Yes, both fixes work. (Didn''t try them simultaneusly though.) Thanks a lot, Jan
Reformatted excerpts from Jan Spakula''s message of 2008-02-27:> Yes, both fixes work. (Didn''t try them simultaneusly though.)Ok, I''m going to merge this into both next and master. -- William <wmorgan-sup at masanjin.net>