I was initially going to ask only about smartcard support but I guessed it was a broader issue.
From what I've read either in #358, #353 or in lib/keystore.go#L19, I understand that the PGP implementation is based on a custom keyring. Please tell me if I'm mistaken.
I think integrating with the existing gpg keyring if it exists and gpg is installed would be a better idea as we would benefit from all the goodies it has to offer, like support for smartcards for example.
Maybe a duplicate of #356 but it's not detailed so it's not very easy to know for sure.
I'm thinking of implementing this as an input filter, by adding support for filters to output MIME. It would then only require a simple
gpg --decryptcommand to get this working. However, that would effectively removeaerc's first-class PGP support, by making it an external feature. As such, I'm not sure how to proceed. Any thoughts?
IMO relying on existing and established tools such as
gpgwould be the best way to provide support for these features without requiring too much development effort (I guess) and without having to reinvent the wheel.Also, users with existing keyrings would have a much nicer experience since it would allow them to use what they already have and not have a separate tool to manage when importing keys and trusting them.
I'm not interested in using gpg. Developer effort is not the concern, quality is the concern. There are 100 bad email clients which would be swayed by the developer effort argument, but aerc is not one of them.
We discussed gpg in detail in #aerc-dev a few days ago. Under no circumstances is gpg support going to be added to aerc; we intend to continue developing our internal keyring.
Thanks for your answer. Is smartcard support planned then? As it was my original issue I could reopen a ticket for this specific use case.
No, I don't intend to support smartcards any time soon. That'll come after PGP is generally implemented and stable.