v2.6.0

• Update to Unicode 15.1.

• Add "script" property – also supported in the list and print commands:

  % uni identify -f '%(script l:auto) %(cpoint) %(name)' 'a Ω'
  Script CPoint Name
  Latin  U+0061 LATIN SMALL LETTER A
  Common U+0020 SPACE
  Greek  U+03A9 GREEK CAPITAL LETTER OMEGA
  
  % uni list scripts
  Scripts:
  Name                    Assigned
  Adlam                         83
  Ahom                          54
  Anatolian Hieroglyphs        582
  …
  
  % uni print 'script:linear a'
  Showing script Linear A
       CPoint  Dec    UTF8        HTML       Name (Cat)
  '𐘀'  U+10600 67072  f0 90 98 80 𐘀  LINEAR A SIGN AB001 (Other_Letter)
  '𐘁'  U+10601 67073  f0 90 98 81 𐘁  LINEAR A SIGN AB002 (Other_Letter)
  '𐘂'  U+10602 67074  f0 90 98 82 𐘂  LINEAR A SIGN AB003 (Other_Letter)
  …
  

• Add "unicode" property, which tells you in which Unicode version a codepoint was introduced:

  % uni identify -f '%(unicode l:auto) %(cpoint l:auto) %(name)' a𐘂🫁
  Unicode CPoint  Name
  1.1     U+0061  LATIN SMALL LETTER A
  7.0     U+10602 LINEAR A SIGN AB003
  13.0    U+1FAC1 LUNGS
  

• Show unprintable control characters as the open box (␣, U+2423) instead of the replacement character (�, U+FFFD). It already did that for C1 control characters, and U+FFFD looked more like a bug than intentional. The -raw/-r flag still overrides this.

• Always print Private Use characters as-is for %(char) instead of using U+FFFD replacement character. It's usually safe to print this, and having to use -raw is confusing.

• ls command is now an alias for `list.