Re: The MOVE problem

The original question came from a "compiler writer" not from a programmer. The
issue is not whether this SHOULD ever be written - but what should a compiler do
when it finds it. Of course, a compiler could reject the code - but then it
wouldn't conform to any Standard (at least from the '74 Standard on).

--
Bill Klein
wmklein <at> ix.netcom.com
"Judson McClendon" <judmc@xxxxxxxxxxxxx> wrote in message
news:ssZVi.44717$b9.24845@xxxxxxxxxxxxxxxxxxxxxxxxx
Thanks Bill, I thought you would come back with a definitive response. :-)

MOVE A(B) TO B C(B).

I would never consider writing such code as in this example. Code should
be written to be as simple and easy to understand as possible, and this
example violates that principle. Using two statements would make the intent
obvious.
--
Judson McClendon judmc@xxxxxxxxxxxxx (remove zero)
Sun Valley Systems http://sunvaley.com
"For God so loved the world that He gave His only begotten Son, that
whoever believes in Him should not perish but have everlasting life."


"William M. Klein" <wmklein@xxxxxxxxxxxxxxxxx> wrote:
The standard requires subscripts of the sending item o be evaluated once at
the beginning of execution of the statement and of the receiving item
immediately before it gets its data. Pages 478-479 of the '02 Standard has:

***

Item identification for identifier-2 is performed immediately before the data
is moved to the respective data tem. If identifier-2 is a zero-length item,
the MOVE statement leaves identifier-2 unchanged.

If identifier-1 is reference modified, subscripted, or is a
function-identifier, the reference modifier, subscript, or
function-identifier is evaluated only once, immediately before data is moved
to the first of the receiving operands.
...

The result of the statement

MOVE a (b) TO b, c (b)

is equivalent to:

MOVE a (b) TO temp
MOVE temp TO b
MOVE temp to c (b)

where 'temp' is an intermediate result item provided by the implementor

***

Now having said that, the concept of "item identification" was new in the
'02 Standard (as I recall) and although this EXACT same "smple" appears in
the '85 Standard (with the "temp" data item), my memory is that many (most?)
compilers use "temp" as the subscript for the receiving item.

Page VI-103 of the '85 Standard had the rule

"Any length evaluation or subscripting associated with identifier-2 is
evaluated immediately before the data is moved to the respective data item."

and then has exaclty the example given above.
--
Bill Klein
wmklein <at> ix.netcom.com




.

Relevant Pages

  • Re: The MOVE problem
    ... Item identification for identifier-2 is performed immediately before the data is ... MOVE a TO temp ... use "temp" as the subscript for the receiving item. ... "Any length evaluation or subscripting associated with identifier-2 is ...
    (comp.lang.cobol)
  • Re: The MOVE problem
    ... Thanks Bill, I thought you would come back with a definitive response. ... Item identification for identifier-2 is performed immediately before the data is moved to the respective data tem. ... MOVE a TO temp ... subscript for the receiving item. ...
    (comp.lang.cobol)
  • Re: Saving data
    ... if you tell your compiler to conform to a C standard it is required to produce at least one diagnostic. ... char char1; ... fwrite (temp, 1, 1, fp); ...
    (comp.lang.c)
  • Re: [Q] about free(ptr)
    ... That shuts up the compiler, ... double temp; ... it may not be immediately obvious what all the vpp ... So we can get rid of *vpp by ...
    (comp.lang.c)
  • Re: Struggling with libraries
    ... void insert(char* insrt, char* source, int place){ ... Your compiler lets you declare that? ... strcpy(source, temp); ... Ok, you are modifying the source, but not the insrt string. ...
    (comp.lang.c)