Re: [sv-ec] Mantis 3028 and other constraints issues

On 12/09/2010 07:44, Bresticker, Shalom wrote:

> 10.9.1, "Array assignment patterns", describes "a recursive
> descent into each subarray of the array".

Indeed, but since there's an explicit type on the pattern this
can descend into packed arrays too. For 3028 we stop as soon
as we reach an element that is not an unpacked array, I think.

I know that we have all agreed that there is no such thing as
a multi-dimensional array in SV. However, there are quite a
few language constructs that care about more than one dimension
(foreach, assignment patterns, "inside") and I think it would
probably be a good idea to have some terminology to refer to
the various different kinds of leaf element that you might
wish to dig out of such arrays of arrays. At present, though,
we don't have such terminology. I felt I had little choice
but to invent some in this case.

> In the new production, the semicolon after
> "uniqueness_constraint" should be red.

Thanks. I'll fix that in time for the meeting. ("Blue for
additions", but that's overruled by "must be red in LRM text",
if I understand you correctly?)

-- 
Jonathan Bromley
-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.