jim barchuk wrote:
> The assembler/compiler/interpreter should just sit back and not frigging
> *interfere* with what I do because unless I tell it it has no idea what
> I intend. It should comment *only* after I have already dropped a brick,
> only if it knows it to be a brick, and my foot is directly under it, not
> trying to decide whether I might move my foot away before it hits it or
> might move my foot under it later even though it's not in the way right
> now. Maybe it's a Yankee 'don't tread on me' thing. The software should
> just stay outta my face and simply do what I tell it to do. :)

What Tim was suggesting could be achieved with a minimally intrusive type
checking.  I've actually contemplated adding something like that to my
preprocessor.  Right now, the bit number symbols are just integers, and
can therefore be used anywhere an integer is allowed.

It would be nice if RP0, for example, didn't just have the value 5, but
was also known to represent a bit within STATUS.  Using it in a bit
manipulation instruction to any register except STATUS and INDF would
result in an assembly time error.  In the relatively unusual case where it
is valid to use the RP0 symbol as other than a bit number within STATUS
(like setting a shadow register, for example), a special syntax would be
required to selectively disable the type checking.  Something like the
Pascal ORD function (returns the raw integer value of any ordinal value)
would be simple and effective.  Having to write ORD(RP0) on rare occasions
would be well worth the error checking it provided in the vast majority of
cases.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listserv@mitvma.mit.edu with SET PICList DIGEST in the body