P*rl in Latin, whither Python?

[Hartmann Schaffer]

In article <3A0DEBA1.C0F2E725 at gte.net>,
Steve Williams  <sandj.williams at gte.net> wrote:
> ...
>Fortran allowed datanames like DOI=1.  I recall a discussion on parsing

no

>DOI=1=1,10.

i suspect you are referring to an incident where a space probe (i
think it was to venus) got lost to a typing error like that.  the
problem with fortran was that white spaces were completely
insignificant, so the parser had to look ahead to determine the
meaning of what it encountered.  the case in question was something
like: 

loops have the format
      do <statement-label> <variable> = <expr> , <expr> [, <expr>]

where statement label was the (numeric) label of the last statement of
the loop body.  a typical loop would start with something like

      do 100 i= 1, 10

again, spaces were insignificant.  however, if the , was missing, this
would degenerate into
      do100i= 110

which is a simple assignment statement, so the fact that there was the
intent to write a loop would be lost to the compiler, as well as the
fact that the variable do100i would never be used

hs