J
J Shrimps, Jr.
Have a linked delimited text file with
several thousand rows. Text
file is created from IBM main frame.
There is one row ("header row")
where the first three characters are ".//"
uniquely identifying all the records
that follow, until the next header row.
Going to query the header row only,
'cause in the 13th position of that row,
and for the next eight positions, contain
the name of the program (file is a dump
of all JCL programs for a person's account
- rows 1 - 50 could be program one,
rows 51 -72 will be the next program,
rows 73 - 90 the next, etc,
but the first line of each program
starts with ".//".)
So, I want to query the linked text file
using mid([Field1],13,8) on rows that
start with ".//" , and return every
consecutive row following, until the next
"header row" - again, starting with
".//" - which means I have all the records
associated with the header row I queried.
Do I have to use some kind of
db.recordset
rs.movenext
type thing for this?
several thousand rows. Text
file is created from IBM main frame.
There is one row ("header row")
where the first three characters are ".//"
uniquely identifying all the records
that follow, until the next header row.
Going to query the header row only,
'cause in the 13th position of that row,
and for the next eight positions, contain
the name of the program (file is a dump
of all JCL programs for a person's account
- rows 1 - 50 could be program one,
rows 51 -72 will be the next program,
rows 73 - 90 the next, etc,
but the first line of each program
starts with ".//".)
So, I want to query the linked text file
using mid([Field1],13,8) on rows that
start with ".//" , and return every
consecutive row following, until the next
"header row" - again, starting with
".//" - which means I have all the records
associated with the header row I queried.
Do I have to use some kind of
db.recordset
rs.movenext
type thing for this?