When doing a decompress of a file (say the SAG EMPLOYEES loaded with default records from the install tape) should the counter of any MU or PE ever show a value of x’00’?
To me if this was done the decompressed data could then not be compressed since the decompress still fills the field position with nulls (i.e. blanks for alpha or zeros for numeric).
I am confused by the statement “the decompress still fills the field position with nulls (i.e. blanks for alpha or zeros for numeric).”
Nulls are x’00’ values, while blanks (x’40’) or zeros (x’F0’) are not null. To me, the x’00’ for an MU or PE counter with no subsequent compressed / decompressed field values is correct if there were no elements for either one.
When a field is null-suppressed, what is the value in the “filler” when it’s decompressed? That is all I was referring to, I have always heard the value of x’00’ referred to as ‘low values’ not nulls.
If the counter value of x’00’ for a MU or PE seems valid, why can’t those records be compressed? In versions of Adabas before v8 the counter was never x’00’, it always had a value of at least x’01’ with filler (so the term ‘nulls’ doesn’t confuse anyone) to represent the bytes for the associated fields.