>That was the root cause of a long-standing exmh bug where GPG >signatures on replies would be corrupted - the trailing blank got >included in the GPG checksum generation calculation, but at the >remote end the canonical text (minus the trailing blank) got used >in the verification. I ended up hacking around it in exmh by >basically doing the moral equivalent of sed 's/[ ]*$//'
Fair enough; I stand corrected. Still not going to fix mhl, though :-) I haven't been paying much attention to encrypted email support, but I might have to in the near term; can you point me to a reference on this? Specifically the part about how canonical text is defined? Or is that just a GPG thing? --Ken _______________________________________________ Nmh-workers mailing list [email protected] https://lists.nongnu.org/mailman/listinfo/nmh-workers
