On 07/23/2015 10:22 AM, Steven D'Aprano wrote:
(3) The hard case, when you mix f and non-f strings.
f'{spam}' '{eggs}'
Notwithstanding raw strings, the behaviour which makes sense to me is that the implicit string concatenation occurs first, followed by format. So, semantically, if the parser sees the above, it should concat the string:
=> f'{spam}{eggs}'
then transform it to a call to format:
=> format(spam) + format(eggs)
I think this should be... => f'{spam}{{eggs}}' The advantage that has is you could call it's format method manually again to set the eggs name in a different context. It would also work as expected in the case the second stirng is a f-string. '{spam}' f'{eggs}' f'{{spam}}{eggs}' So if any part of an implicitly concatenated string is an f-string, then the whole becomes an f-string, and the parts that were not have their braces escaped. The part that bothers me is it seems like the "f" should be a unary operator rather than a string prefix. As a prefix: s = f'{spam}{{eggs}}' # spam s2 = s.format(eggs=eggs) # eggs As an unary operator: s = ? '{spam}{{eggs}}' # spam s2 = ? s # eggs (? == some to be determined symbol) They are just normal strings in the second case. Cheers, Ron