Oppolzer - Informatik / Blog


Blog-Hauptseite      Neuester Artikel      Älterer Artikel      Neuerer Artikel      Älterer gleiche Kategorie      Neuerer gleiche Kategorie

PL1-L - Dokumentation zu PL/1 V2.3

Subject:

Re: PL/I V2R3 documentation....

From:

Bernd Oppolzer <bernd.oppolzer@T-ONLINE.DE>

Reply-To:

PL1 (language) discussions <PL1-L@LISTSERV.DARTMOUTH.EDU>

Date:

2012.03.07 15:25:52


Our site rules say:

only record I/O in locate mode, reports have to be done using machine code
control characters (no ANSI control characters). This may have to do with print
output post processing mechanisms.

Only exception, AFAIK, is writing of variable length records, where move mode is
allowed.

Very old, from the seventies. No discussion about that. Accepted by everybody.

Kind regards

Bernd



Am Mittwoch, 7. März 2012 14:16 schrieben Sie:
> This discussion has now taken an all but predictable turn that is
> nevertheless disheartening.
>
> The generic strictures and rules against the use of stream I/O found
> in some commercial shops were and are ridiculous.
>
> It is of course true that the profligate use of data-directed stream
> I/O is 'inefficient', and thus best relegated to testing statements.
>
> That said, the quondam COBOL programmer's notion that he can do better
> than edited-directed I/O for report writing is flatly wrong.  IBM
> implementations of edited-directed stream I/O have for long  been
> asychronous; and, used with any intelligence (and appropriate JCL),
> they are better, very much better, than the RYO pseudo-record I/O
> schemes that I have seen used to avoid them.
>
> Lest I be misunderstood, let me add that I am not of course arguing
> against the use of record I/O, particular locate-mode record I/O, in
> its proper sphere, in an MFU or the like.
>
> J.G.

Blog-Hauptseite      Neuester Artikel      Älterer Artikel      Neuerer Artikel      Älterer gleiche Kategorie      Neuerer gleiche Kategorie