Re: Same prefixes (was Re: It's That Time Again! etc.)

["Richard B Baird" <rbaird@xxxxxxxxxxxxxxxxxxxxx>]

all,

A comment on this line of posts, and then a request for clarification....

The reason for prefixing field names for different files has little to do
with external vs internal field definitions, and everything to do with RPGs
sometimes convenient and othertimes scary behaviour of having only one
instance of each field name - change the value once, and it changes the
value in all iterations of the field name, unlike cobol, and i suspect most
other languages where qualified field names are the norm.

having said all of that, I'm a little confused about this line of posts....

using QUALIFIED data-structures doesn't help when reading or chaining to
files, unless you use program described files with a ds name in the result
field, or you immediately move the record format to the datastructure after
read, then move back before updating.  messy and potentially dangerous!

on the other hand, using some sort of drop-prefix has it's own set of
problems.  once you drop a prefix, then all fields from different files get
changed if one gets changed - whether you intend it to or not - which
negates the need for a move-corresponding, again, unless you use prefix()
and qualified at the ds level instead of the file level, and do the move to
datastructure after read and back before update.

Am I making sense here?  or am I missing something?