Re: Small change in doc snarf output

[Michael Livshin]

Marius Vollmer <address@hidden> writes:

> Michael Livshin <address@hidden> writes:
> 
> > there's no problem whatsoever to change the snarfer behaviour, of
> > course, so if everybody prefers the old way I'll reinstate it.
> 
> We should do it the `correct way'.  There is a difference between [foo
> bar] and [foo [bar]] and only the author of the function can decided
> which one is correct.  So we need a way for the docstring to specify
> how the optional arguments should be marked.

right.

for the time being, I reinstated the [foo [bar]] convention, since it
has the virtue of being compliant with the way the evaluator treats
optional arguments...

-- 
The only complaint I have against WinDoze is that it doesn't 
always fail at install time.       -- Mike Andrews, in s.d.m