Whitespace as syntax (was Re: Python Rocks!)

Dennis Lee Bieber wlfraed at ix.netcom.com
Wed Feb 9 01:06:20 EST 2000


On 8 Feb 2000 09:19:50 GMT, Thomas Hamelryck <thamelry at vub.ac.be>
declaimed the following in comp.lang.python:

> 
> I am in exactly the same position. It was a serious design error.
> I wonder how the CP4A people are going to explain to complete
> computer illiterates that a python program can contain errors that
> are _not even visible_ when you look at the code. 
>
	The same way the DEC (Compaq <G>) folks can explain that a
Fortran program can have errors which are not visible if the source is
printed on paper.

	Namely, DEC VMS F77 counts a tab as 1 character.

0        1         2         3         4         5         6
123456789012345678901234567890123456789012345678901234567890123456
	This_variable	=	Something	*	Another

	Are there 63 characters on that line?

	Not as I entered it -- to VMS F77 there are only 36. Lines can
extend to "character 72" before the excess is considered in the "comment
/ sequence # zone".  As a result, one can enter valid lines using tabs
that extend beyond column 72 (and even off the edge of the traditional
80 column terminal display). If that source line is fed through some
utility that turns the tabs into equivalent spaces, the valid line now
fails to compile (and if you are lucky you get a compile error -- it is
possible for the "truncation" of stuff past column 72 to result in a
valid, but erroneous, line).

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <



More information about the Python-list mailing list