|
|
|
In addition to creating CSV members, CA Earl can write index records to the CAIJSIDX data set as each CSV member is created. This index, or directory, is used to keep track of the CSV files. Each record in the CSV directory has a one byte record type field in the first position of the record. Valid record types are:
Version
Data set name record
Member name record
Each record in the CSV directory has one or more fields in addition to the record type field. These fields have a fixed format and length.
The version record is created when the CAIJSIDX directory data set is initially created, and identifies the level of the host software that created the directory file.
The data set name record contains the name of the data set containing the CSV members that are indicated on subsequent member records.
The member records contain the CSV member name and the date and time the CA Earl run was created, (NOT the date of your data). The constant 'PT' in positions 3 and 4 are a requirement of the download program.
The following example shows what a CA Earl CAIJSIDX directory data set may contain. It also shows an index file tracking an output PDS defined as a GDG. Had the PDS NOT been defined as a GDG, the second CA Earl run would not have created members CSVM03CC and CSVM04CC, but would have rewritten CSVM01CC and CSVM02CC, causing a loss of data if the data had not been downloaded in the interim.
position
1 2 3 4 5 6 7
1...5....0....5....0....5....0....5....0....5....0....5....0....5....0
V JARSR12 SP0
D CONST03.CSV.ERXT0262.PDS.G0001V00
M PT CSVM01CC 2000/09/20 14:14
M PT CSVM02CC 2000/09/20 14:14
D CONST03.CSV.ERXT0262.PDS.G0002V00
M PT CSVM01CC 2000/09/20 14:16
M PT CSVM02CC 2000/09/20 14:16
D CONST03.CSV.ERXT0265.PDS.G0003V00
M PT CSVM01CC 2000/09/20 14:59
M PT CSVM02CC 2000/09/20 14:59
D CONST03.CSV.ERXT0267.PDS.G0003V00
M PT CSVM01CC 2000/09/20 14:59
M PT CSVM02CC 2000/09/20 14:59
D CONST03.CSV.ERXT0262.PDS.G0003V00
M PT CSVM01CC 2000/09/20 15:00
M PT CSVM02CC 2000/09/20 15:00
Only one version record exists, followed by "groups" of data set and member name records that correspond to the various batch executions that have occurred. The CAIJSIDX DD is required during any execution of CA Earl that creates output CSV members, but may be specified as DD DUMMY for users who do not want to maintain this file. Notice that the index file can point to many different data sets that contain CSV files. As many index files and PDS CSV repositories as desired can be maintained by the user. You can have one directory file for each CSV PDS file, one directory file for all CSV PDS files, or a combination of the two.
| Copyright © 2012 CA. All rights reserved. |
|