that's because you aren't using a primary heuristic to scan the text for`http`
it's really really REALLY simple if you just think about it
the regexp is not so much for finding but for validating
the brackets at the end of a url before the next space can easily be ignored as a typical human fucking
idk what to say guys
seriously, the rest of teh internet figured tihs out 10 years ago
just as a test to see if #coracle does this correctly
ok, now really pushing it:
nostr:nprofile1qyvhwumn8ghj76rzwghxxmmjv93kcefwwdhkx6tpdshsz9thwden5te0wfjkccte9ejxzmt4wvhxjme0qy88wumn8ghj7mn0wvhxcmmv9uq3qamnwvaz7tm99ehx7uewd3hkctcppemhxue69uhkummn9ekx7mp0qqsf03c2gsmx5ef4c9zmxvlew04gdh7u94afnknp33qvv3c94kvwxgs9w8d3z little tip with the URL recognition, if the http has ( before it you probably can ignore the last )
it is VERY rare situation that someone places a URL, inside brackets, and remembers to close them without adding a space between
so rare, that you can basically assume that if a URL has )) at the end, the second one is probably invalid
Your proof that it is rare that users forget spaces when using brackets is the same as my proof that there are more uses that use brackets then URLs that have them
In other words its all pretty much subjective
no, it's really not, because there is not whitespace inside URLs, URLs are almost always bounded by whitespace
thus, when you find whitespace, it marks boundaries around the matched URL and you know that a second ... fucking... close brace... is actually, not part of the URL! idk how to say this couldn't be more obvious
it's a small piece of logic that you have been spoiled to never learn how to actually write a lexical analyser and this reminds me that the majority of the devs in the world couldn't write a lexical analyser if their life depended on it
i learned how to do it, working from a text that was part of first year computer science when i was 15 years old
you guys both just utterly lost my respect tonight
nostr:nprofile1qyxhwumn8ghj7cnjvghxjme0qyt8wumn8ghj7ct4w35zumn0wd68yvfwvdhk6tcpr3mhxue69uhkyatrddjhgtnrdaexzcmvv5h8xmmrd9skctcpypmhxue69uhkummnw3ez6un9d3shjtnyd9nkjarpd3kk7c3wwfhj7qgswaehxw309ajjumn0wvhxcmmv9uqzqwlsccluhy6xxsr6l9a9uhhxf75g85g8a709tprjcn4e42h053va63qc8k nostr:nprofile1qy2hwumn8ghj7un9d3shjtnyv9kh2uewd9hj7qgawaehxw309a5x7arjd9nksarwdamjumn0wd68yvfwvdhk6tcpzemhxue69uhk6mr9dd6juun9v9k8jtnvdakz7qguwaehxw309a382cmtv46zucm0wfskxmr99eek7cmfv9kz7qg7waehxw309a5xyu3wvdhhyctrd3jjuum0vd5kzmp0d9hxymmcqqszv6q4uryjzr06xfxxew34wwc5hmjfmfpqn229d72gfegsdn2q3fg4nzgum
> you guys both just utterly lost my respect tonight
that's why I'm bothering to respond 😁
also, you are saying this to someone who wrote a fully effective json parser from scratch for the entire nostr protocol spec, that is fully compliant, and always works, except for a few wrongly escaped json strings from old events mostly
