RE: Same field names -- MIDRANGE-L

Thank you all for your responses.

I was under the impression that if you didn't PREFIX the fields the compile 
would blow.  Some testing shows otherwise.  I could see some concerns with that.

If you don't understand the fields because they are now prefixed, how would you 
understand them better if they had different names?

I've seen files, (in BPCS, the customer master), where many of the fields in 
the file started with CM and then some new, (BPCS, we don't add fields to their 
files) ones didn't.  PREFIX would be consistent.

Some good workarounds, like querying the SQL tables by LIKE '%field%'.  And, 
granted the tools would do a much better job of catching stuff.   But there is 
always the person would bust a leg trying to defeat the tool.  Of course my 
file cross reference tool is just a PDM scan, pretty much works.  I've found it 
much faster than the collection jobs of the tools.





DRF@HeiligMeyers.com on 05/04/99 12:41:48 PM
Please respond to MIDRANGE-L@midrange.com@Internet
To:     MIDRANGE-L@midrange.com@Internet
cc:      

Subject:        RE: Same field names

Hmmm...I've learned something today.

In my opinion, however, using the same field name in multiple files causes
more problems then it solves, if any.  Even though the "PREFIX" keyword is
easy to use, it will cause difficulty analyzing programs if not applied
consistently.  

For example, one programmer may use the prefix "CM" for the customer master
file and another may use "CS".  If, however, the field names were unique to
each file, the need for using "PREFIX" disappears and every program is using
the same field names.

It is true that enforced programming standards could solve this problem, but
why go to the trouble of creating and enforcing such a standard when making
the field names unique makes it unnecessary?

Secondly, the use of "PREFIX" adds a step to program analysis.  Not only
must the analyst know the field names, but also the various prefixes used by
the program for each file.

Lastly, with ten character field names available, the last six to eight
could be the same name throughout the system with the first 2 to 4 used as a
file prefix.  Such a standard would, in my opinion, be easier to enforce and
you can still use query to extract the "field where used" information by
using the substring function.

Donald R. Fisher, III
Project Manager
Heilig-Meyers Furniture Company
(804) 784-7500 ext. 2124
Don.Fisher@HeiligMeyers.com

<clip> Query the 
file QADBIFLD on your system and you can see every file containing this
field.  <clip>
So what do you think of using the same names?  How about requiring database
field names be 
limited to 7 or 8 characters so the PREFIX keyword is cleaner?
+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---


+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---