Josh Kupershmidt <schmiddy@gmail.com> writes:
> The only mild concern I have is if this could possibly lead to
> ambiguous parsing in some situations, though I've played with some
> examples and I haven't seen any yet. It would be nice to have this
> behavior documented somewhere though.
The fine manual currently says (at the head of section 4.1):
A token can be a key word, an identifier, a quoted identifier, a literal
(or constant), or a special character symbol. Tokens are normally
separated by whitespace (space, tab, newline), but need not be if there
is no ambiguity (which is generally only the case if a special character
is adjacent to some other token type).
The parenthetical remark at the end fails to point out the special case
of number-followed-by-identifier-that-doesn't-look-like-an-exponent.
But I'm not sure that it's reasonable to try to shoehorn in a mention
of the case. Might be a good idea to change "generally" to "usually",
though, since "generally" might be read as implying that that's the
exact and only rule.
regards, tom lane