[Bug c++/70529] Unhelpful diagnostic for hex float literals, inconsistent parsing

["jakub at gcc dot gnu.org" <gcc-bugzilla at gcc dot gnu dot org>]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70529

--- Comment #11 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
pp-number:
  digit
  . digit
  pp-number digit
  pp-number identifier-nondigit
  pp-number e sign
  pp-number E sign
  pp-number .
identifier-nondigit:
  nondigit
  universal-character-name
  other implementation-defined characters
nondigit: one of
  a b c d e f g h i j k l m
  n o p q r s t u v w x y z
  A B C D E F G H I J K L M
  N O P Q R S T U V W X Y Z _
digit: one of
  0 1 2 3 4 5 6 7 8 9
sign: one of
  + -
Thus
0x123P2
is valid pp-number, because it starts with a digit, is followed by nondigit,
then 3 times digit, then nondigit and then another digit, all matching the
above grammar.  But 0x123P-2 is not a pp-number, because in the grammar
sign is only allowed after e or E letters.
0.123P2E-3E+3e-2E+2 is also a pp-number.  After lexing this one is of course
rejected, because it is not a valid floating point constant.