ok. httpwg.org/specs/rfc9112.html forbids parsing HTTP/1.1 headers as Unicode (because it introduces ambiguity about what a LF is). Does that mean I theoretically have to decode URL encoding (%30) ?
by the way, I put in unsafe code so it doesn't crash if the bytes in the path name make up "surrogates". I still don't really understand how surrogates work, so I don't know if this is necessary. For example, It's really hard to type them, if that even makes sense.
does the mega challenge require any two words in the standard alphabetical order are sorted relative to each other and what does standard alphabetical order mean?
@Seggan when I need to regularly give someone a file, sometimes he just brings an iPad, which is one of the devices where I think they don't have a terminal, right after iPhones and Android devices without Termux on them
pretty cool. this may replace another dumb thing i sometimes do with ephemeral web servers, which is transfering files to my phone if I don't have a cable
legal but hackery sounding thing I just did: For some reason, I can only download bundled example circuits for a logical circuit simulator we're using at university if I am at the university. something to do with all the other stuff on that subsite being public. so I just ssh into my account there, curl the zip file and then python3 -m http.server it through a TCP tunnel in the ssh to my computer. I've definitely learned the python3 part on this site.
Mascarpone is an Esoteric programming language with a so-called Meta-Circular interpreter, and it prints "Uncaught infinite loop!" as an error message when I run
@UnrelatedString oh, as a new question? that would solve the problem that, much like with the C specification, we can't really apply an unambiguous specification to existing solutions that used wiggle-room which previously existed
less relevant ambiguity: how do I determine what the possessive form of someone's neopronouns are if xe (example) doesn't tell me? are they xeir neopronouns or xis neopronouns?
since It's only about the storage tech, I think I'm going to go with Box<[Token]> because i'll most likely copy out of that directly later anyway. the upside of iterator is being able to use both a once() iterator and any other, but uh doesn't matter
so if the token is a built-in it transpiles into itself, but if it is a string, it's supposed to become a construction that makes a string. for codegolf.stackexchange.com/questions/11690/…
I'm writing a function which most often returns one of its input (Token) as output, but sometimes creates a longer sequence of Tokens as the result. what should the result type be?
