2metre <news55@[EMAIL PROTECTED]
> wrote:
> frank@[EMAIL PROTECTED]
wrote:
>> Code snippets never conform to a Pascal standard.
> Thats an interesting viewpoint. ISO 7185 includes many code snippets.
> They are normally termed 'examples'. They are used to illustate a point.
> As far as I'm concerned, if it's okay for the standard, it's okay for
me.
: NOTES are included in this International Standard only for purposes
: of clarification, and aid in the use of the standard. NOTES are
: informative only and are not a part of the International Standard.
:
: Examples in this International Standard are equivalent to NOTES.
Since neither of us wants to be unprecise about the standard, we
surely have to attest that examples are not a part of the standard.
>> all too often users send incomplete code omitting some relevant
>> information.
> I think my snippet included all the neccessary information to illustate
> my point for discussion.
For this purpose the first sentence would have sufficed. Seeing the
example it was not clear if that was the only issue at hand.
(Especially given that it contained invalid identifiers, starting
with a number, which were not the result of cutting for brevity I
suppose.)
>> I've gotten tired of this. If one expects free help,
> Free help? Its a discussion point. This is the place to discuss ISO
> Pascal conformance isn't it?
This is no contradiction. You started the discussion because you
wanted to get some information. I knew the answer already and gained
nothing by writing it (same for others who might have answered
instead). It can't hurt be rather more polite and helpful when one
wants to be helped. (And yes, providing a self-contained example is
helpful to responders.)
>> one could at least bother to make a self-contained test case. BTW,
>> given a complete program I could just it run through GPC in ISO 7185
>> mode, and it will (most likely) tell me if it conforms.
> That's presuming GPC has correctly interpreted the standard. You do. I
> don't. (But nor do I suggest that it doesn't).
I know that it doesn't (yet) in all regards. But I can think about
the output myself in case of doubt. It's still better to have the
output as a start. (E.g., I might have forgotten this rule, or your
code might have respected this rule and unintentionally violated
another rule I wasn't aware of. A compiler might have notified me.)
> For the sake of completeness, I think I should point out that your
> answer appears incomplete.
> The standard states:
> "The control-variable shall be an entire-variable whose identifier is
> declared in the variable-declaration part of the block
> closest-containing the for-statement."
> Let me stress: not *any* entire-variable but one that matches the
> following qualification: "whose identifier is declared in the
> variable-declaration part of the block closest-containing the
> for-statement."
Yes, you're right. (BTW, I didn't say *any*, but "a", just for the
sake of completeness, of course.)
> > See the current thread "Current test version
>> of Chess 0.5" in c.l.p.misc where the same issue came up and other
>> people have provided comments as well.
> If someone discusses ISO 7185 in alt.s*x.bondage should I refrain from
> discussing it here as well?
I didn't mean to stop the discussion with this comment, just trying
to give additional helpful information (because those who posted
there might not repeat everything here or might not even read here).
You seem to have a talent for taking offence at being helpful ...
Frank
--
Frank Heckenbach, frank@[EMAIL PROTECTED]
http://fjf.gnu.de/
GnuPG and PGP keys: http://fjf.gnu.de/plan
(7977168E)
Pascal code, BP CRT bugfix: http://fjf.gnu.de/programs.html
Free GNU Pascal Compiler: http://www.gnu-pascal.de/
|
