BPCS Column Heading
"We have found the program to be of great value.  Both users and staff have commented on the usability of queries now that the field descriptions make sense."
Bill Clark - OMG Americas.

The BPCS Column Heading Fix 

Corrects the text and column headings on BPCS V6 

BPCS Column Heading Fix Questions & Answers


Why are the column headings as they are?
How does this product work?
What does is fix?
How long does it take?

Why are the column headings as they are?

The problem with the column headings arose due to a limitation in the AS/SET case tool.  When a file is defined using AS/SET, each of its fields must be based on a reference field.  The reference field provides a generic definition for a similar set of fields e.g. order number, name, date.  When using a reference field, you are allowed to override its name and text, but not its column heading.  As a result, most of the column headings are incorrect or insufficiently descriptive.  Older versions of BPCS did not have this problem because the fields were defined using standard DDS which allows reference fields to be fully overridden.

How does this product work?

This product works by directly updating the file's field descriptions rather than generating a new AS/SET repository or DDS specs.  The advantage with this is that you don't need to recompile anything.  The only attributes which are changed are the fields' text, column headings, and edit code / edit word.  The information for the change comes from a 'field reference work file'.   This is a separate physical file which contains the new information for all the fields in either the version 6.x.xx or 8.0.00 databases.  This file may be modified as required and the Fix re-run for one or all files.

[For those technically minded, the field descriptions are stored in an object called a 'format space'.  This is a hidden object of type *FMT.  It is not stored in a library, but 'linked' to a file.   Each file has a format space, sometimes several files may point to the same format space - this is called format sharing (discussed further in the downloaded document).   If you want to see a format space you can do a 'dump object' on a file, i.e. DMPOBJ filename *FILE , then look for the *FMT object in the spool file.]

What does it fix?

In general the fields’ text have been left alone - because they were mostly correct - and the focus was made on fixing the column headings.   Around 80% of the column headings were changed due them being: (a) misleading or limited in their description; (b) inappropriately sized for the field’s display length; or (c) completely mismatched.  Where possible, new column headings have been made to fit within each field’s display length without over-abbreviating.

Edit codes:  Numeric fields that had no edit code were given edit code of L, and numeric fields with decimal positions having edit code Z were changed to edit code L.  Date fields were given an edit word which will make them display as e.g. 1998/12/25 and time fields an edit word to display as e.g. 12:34:56.

How long does it take?

This depends on your machine's size.  If you are running BPCS version 6 then you will probably require a fairly large box (e.g. models S20, 640, 730 ...).  In this case, to fix a database (all physical and logical files) typically takes between 3 to 5 minutes as a dedicated batch job.  You can also fix one file at a time as required.  To fix a single file (a physical plus its logicals) interactively takes less than a second.