Janet PEGs don't accept unicode??

nope! not easily, at least. you can match unicode characters, but NOTHING is made easy for you, the way a regex /u flag will. that said, its still possible with use of (to)

look, for example, at a truncated version of a PEG i’m writing to pull info from a french/english dictionary:

(def dict-result-peg
  ~{
    # ... catch some stuff
    :pronunciation (* "/" (to "/") "/")
    # ... and so on
   })
 

i wanna catch all the character in the words IPA prnounciation, which luckily enough are surrounded by /s. since janets PEGs don’t allow for a character class range of unicode prononciation strings, i instead use a (to "/") to grab all the text between delimiters. it’s imprecise, but in this instance perfect, as i know ill never get a / in the prnonciation guide