bug#40219: 26.3; f90-end-of-block doesn't work with a prefix argument (and possibly it's semantics is wrong)

[Lars Ingebrigtsen]

Vladimir Nikishkin <lockywolf@gmail.com> writes:

> If you try C-3 C-M-n, the command will fail, saying "missing block end".
> Frankly, the semantic of the command is unclear. That is, the command is
> expected to move the point "to the end of the current block" -- that is
> "outside of the block's scope". And with the numeric prefix argument it will 
> "do
> the same thing several times", that is "it will exit n nested scopes",
> rather than "go forward n balanced blocks". Additionally, it's Emacs
> Lisp's counterpart bound to C-M-n does _not_ exit the scope of the list,
> but rather moves between blocks (list elements) within an enclosing
> scope.
>
> This is not to be confused with C-c C-e (f90-next-block), because
> f90-next-block moves between _not_ balanced blocks. That is, it will
> enter each new scope, such as the (do) loop on line 16.

The semantics are unclear, but I think it looks like this works as
designed.

subroutine parse_symbol( )
  do
  end do
end subroutine parse_symbol

If point is at the end of the second line, then `C-u 2 C-M-n' will end
up after

end subroutine parse_symbol

because it's gone "out" of two blocks.

I don't know Fortran well, but the way the commands work here make
sense, I think, so I'm closing this bug report.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no