After Wirt:

        DRA has two programs that are already written that can help anyone
in this situation, DISK2DB and DB2DISK. DB2DISK copies the contents of
either a manual master or detail data set to a flat file. DISK2DB takes a
flat file, in the exact format of the targeted manual master or detail data
set, and copies the data via DBPUT. Neither program does any normalization
of data.

        Please email me if you would like a copy of either or both programs.

                              Eric Sand
                              [log in to unmask]


> Jay asks:
>
> > (1) Is it possible to upload data from a flat file to an Image dataset?
>
> Yes.
>
>
>
> >  (2) If the answer is yes to question above is there an utility using
> which
> I
> >  can update the dataset with data from flat file?
>
> The general procedure is that you need to parse the flat file into its
> constituent components and write them into the dataset using a DBPUT
> intrinsic. Because these are inherently string-oriented operations,
> writing
> your own small routine of approx. 20-30 lines in BASIC is probably the
> easiest thing to do.
>
> You can download both a copy of the BASIC language for the HP3000 and its
> documentation for free from:
>
>      http://www.aics-research.com/basic/index.html
>
>
>
>
> >  (3) If the answer is yes to question above could u Pls. give me an
> exapmle.
>
> Examples exist in the manuals.
>
>
>
>
> >  (4) Is the master dataset updated automatically?
>
> It is the nature of an automatic master to be updated automatically, hence
> the name. If you're designing the database from scratch, I've long ago
> come
> to the conclusion that you do not want to use non-automatic master
> datasets
> in your databases.
>
> Indeed, if I had my way, I would take manual masters out of IMAGE's
> repretoire.
>
>
>
> >  When I am updating say the
> >  detail data set, what are issues/points to remember that I have to take
> >  care.? e.g. My Master dataset may need some entries inorder to update
> the
> >  detail dataset.
>
> You cannot control the ordering of entries into master datasets. Their
> order
> is determined by a hashing key. It's this secondary ordering that allows
> you
> find specific values in the master datasets at high speed rather than
> having
> to read every entry, one at a time.
>
> Wirt Atmar
>
> * To join/leave the list, search archives, change list settings, *
> * etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *