HDB Extract to CSV
The HDB(EXTRACT) operand requests CICS PA to generate CSV extracts from HDB data.
CICSPA HDB(EXTRACT(hdbname),
[OUTPUT(ddname),]
[DDNAME(ddname),]
[STATnnnn(ddname),]
[HSTGnnnn(ddname),]
[DELIMIT('field-delimiter'),]
[LABELS|NOLABELS,]
[FLOAT,]
[SUFACTOR(hdbname(nnnnn.nnn)),]
[INTERVAL(hh:mm:ss),]
[FIELDS(field1[(options)],...),]
[SELECT(PERFORMANCE(INCLUDE|EXCLUDE(field1(values1),...),...)),]
[SELECT2(PERFORMANCE(INCLUDE|EXCLUDE(field1(values1),...),...))])
- EXTRACT
- Specifies the name of the HDB from which to extract data. The HDB must be defined in the Repository (DDname CPAHDBRG).
- OUTPUT
- Specifies the DDname for the Recap report output. If not specified, the CICS PA dialog assigns a DDname in the format HXTS0001 to uniquely identify the output.
- DDNAME (performance HDBs only)
- Specifies the DDname for the performance extract data set. Dialog default: HDBX0001
- STATnnnn, HSTGnnnn (statistics HDBs only)
- Specifies the DDname for the extract data set for each statistics report that you want to extract, where nnnn is the statistics ID. STATnnnn identifies a CICS® Transaction Server statistics report; HSTGnnnn identifies a CICS Transaction Gateway statistics report. For example, HSTG000A(TGCMCSV) instructs CICS PA to extract CICS Transaction Gateway Connection Manager statistics report data to the extract data set identified by the DDname TGCMCSV. Dialog default: DDname matches keyword; for example, STAT010A(STAT010A).
- DELIMIT
- Specifies the field delimiter, enclosed in quotes, to be used to separate each data field in the extract data set. The default is a semicolon DELIMIT(';').
- LABELS|NOLABELS
- LABELS indicates that the first record to be written to
the extract data set is to be a field labels record. This is the default.
NOLABELS indicates that CICS PA is not to write a field labels record to the extract data set.
- FLOAT (performance HDBs only)
- Write numeric fields in the extract in S390 FLOAT format.
Specify FLOAT format if you plan to import the extract into a DB2® table. When the DB2 Load Utility is used, it will interpret all numerical fields reliably and consistently in FLOAT format.
If FLOAT is not specified, the numeric fields are written in a mixture of integer, real and exponential using character digits. This is the default and is suitable when importing the extract data into a PC spreadsheet tool.
- SUFACTOR
- Specifies a CPU SU conversion factor to apply
to the HDB container. The SUFACTOR operand includes two keywords to
identify the HDB name and its associated conversion factor. The value
must be a decimal number or integer in the range 1 - 999999999 (nine
9s).
When reporting or extracting HDB records the CPU SU conversion factor will only be applied when field CPUSU is not present in the HDB record. In other words, if field CPUSU is in the HDB record, it will be reported as is, otherwise, it will be calculated using the CPU SU conversion factor.
- INTERVAL
- This operand applies to Summary HDBs. It specifies a time interval
when the extract summarizes transaction activity over time. The interval
is in the range 1 second to 24 hours in the format hh:mm:ss for
hours, minutes, and seconds.
Data in a Summary HDB is already summarized by the interval that was used to load the data. This is the value specified in the HDB or, if Time Interval was not specified in the HDB, the value defined in the Template.
You can further summarize the data by specifying a multiple of the interval that was used to load the data. Specify a value in the range 00:00:01 (1 second) to 24:00:00 (24 hours). For example, specify 00:15:00 if you want to summarize transaction activity over 15 minute intervals. If you are reviewing many days worth of data then you might specify 24:00:00 (24 hours) so that you can view the daily trend.
Notes:- If you specify a reporting interval that is equal to or less than the interval that was used to load the data, the report or extract uses the data as-is, without further summarization.
- If you do not specify a reporting interval it defaults to the interval that was used to load the data, unless that value is less than 1 minute, in which case the reporting interval is set to 1 minute.
A time interval of less than one hour must fit evenly into the hour. CICS PA will round it down to the nearest interval that aligns to the hour. For example, 1.35 is reduced to 00:01:30 minutes which will produce 40 interval report lines for each hour of data.
A time interval of more than one hour must fit evenly into the day. CICS PA will round it down to the nearest interval that aligns to the day. For example, 10.30.23 is reduced to 08:00:00 hours which will produce 3 interval report lines for each day of data.
Minutes take precedence for an abbreviated entry. For example:- 1
- becomes 00:01:00
- 1.1
- becomes 00:01:00 (rounded down from 00:01:01)
- 1.1.1
- becomes 01:00:00 (rounded down from 01:01:01)
- FIELDS
- Specifies which fields are extracted, the order in which they appear in the extract, and their summarization presentation. Only fields that are specified in the HDB Template can be specified. Fields not contained in the HDB are written as Missing.
- SELECT, SELECT2
- Specifies
what data to include or exclude from the extract based on data field
values. See Using SELECT statements for an explanation
and examples.
You can use both SELECT and SELECT2 to perform record filtering. The CICS PA dialog generates SELECT2 statements in the JCL when you use a report form that has active selection criteria. You can add a SELECT statement by editing the generated JCL. If both SELECT and SELECT2 are specified, the record must pass selection by both specifications for it to be included in the extract.
The following restrictions apply to selection criteria.
- The SELECT and SELECT2 options can specify only fields that are contained in the HDB. If the selection criteria specify a field that is not present, the extract returns an error that identifies the field.
- For a performance summary report, as distinct from a performance list report, CICS PA ignores selection criteria for summarized fields because it would be meaningless to apply selection criteria to them. However, you can apply selection criteria for key fields, that is character and time stamp fields.