Hi, On Thu, May 14, 2020 at 11:51:08AM +0200, Oswald Buddenhagen wrote: > On Wed, May 13, 2020 at 08:03:02PM -0700, Kevin J. McCarthy wrote: > > On Sun, Apr 19, 2020 at 01:03:06AM +0200, Oswald Buddenhagen wrote: > > > on a tangent, mutt's thread linking features do not work if the > > > message-ids lack the <angle brackets>. i presume these might be > > > invalid, but they are rather common nonetheless. someone feels like > > > having a look? > > > > This was on my todo list, but ticket 241 prompted me to work on it a bit > > sooner. I've just committed a change to master that loosens this up a > > bit. > > > thanks, but ... > > from the commit message: > > However, in order to reduce the risk of extracting garbage, insist > > that message-id's outside of angle brackets at least have an '@'. > > well, guess what, the message-ids from aliexpress are just bare hex numbers > (md5s?). > > i think a reasoanble heuristic would be to just accept anything as long as > it's a single string with no whitespace.
Although things should still work the same with such heuristics, I don't think that it is a good plan to silently accept garbage in general and here. The main reason is that this honors behaviour of disrespecting standards and motivates people to extend that. Overall side effects potentially downgrade systems. So with heuristics I propose to display a warning about the illegal format (RFC is clear) to the user when the message is displayed. This notifies about possible problems and transparently shows sender's responsibility. If one of the users then complains at aliexpress I'm fine with it ;-) I'm aware about that warnings could be annoying, but because of the previous statement IMO users must live with it in that world - I wouldn't add an option to silence it. Gero
