URL Parsing Bug

Hello o/ I hope this is the right place to notify you of this, I noticed in another message that crabmail was incorrectly parsing the following string: https://www.paritybit.ca—which It seems that it’s including the “—which” and reading the em-dash as part of the URL. The expected behaviour is that the “—which” is not part of the URL since “.ca—which” is not a valid top-level domain and therefore a broken link is created. Crabmail is parsing it and creating a link with punycode (it creates the following link: https://www.paritybit.xn--cawhich-316c/), but I would expect from my use of other software (such as Mastodon, my mailing client, etc.) that the link ends at the end of the TLD, “.ca”. Thanks! — Jake Bauer (~jbauer) https://jbauer.ca @jbauer@bsd.network

> Hello o/ > > I hope this is the right place to notify you of this, I noticed in > another message that crabmail was incorrectly parsing the following > string: > > https://www.paritybit.ca—which > > It seems that it’s including the “—which” and reading the > em-dash as part of the URL. Thanks Jake! This is indeed the right place. This appears to be an issue with one of my dependencies (although I'm not 100% sure if it could parse this unambiguously.) -- I brought this to the attention of the maintainer: https://github.com/robinst/linkify/issues/28 -- Alex https://alex.flounder.online gemini://alex.flounder.online