The Create Data Base file with Journal Fields command allows you to
create an externally described file that contains the standard
journaling fields (such as the user, date, RR number) and the field
names from a file that is being journaled. The new form of the
command uses the DSPJRN outfile to output records to the file. PRTDB
(simple listing of named fields) or PRTDBS (allows select/sort also)
could be used to process the entries.
The old form of the command is CRTDBFJRN2 which creates the file, but
does not do DSPJRN. See the instructions for how to use CRTDBFJRN2.
The normal use of the system DSPJRN command provides a single field
(JOESD) with the image of the data base record. You can process the
outfile, but you must write a program that handles the data within
your record image in a program described manner.
To use CRTDBFJRN, the existing data base file must be a physical file
that is being journaled.
Using the CRTDBFJRN function allows you to analyze the results with a
Query or a program. You may want to consider the CMPJRNIMG2 tool
also. CMPJRNIMG2 will allow you to print specific field comparisons.
A typical CRTDBFJRN command would be:
CRTDBFJRN FROMFILE(FILEA)
CRTDBFJRN would create a temporary source file in QTEMP. The DDS for
the QADSPJR2 model file in QSYS used by DSPJRN would be retrieved and
the JOESD field (entry data) deleted. By default, the *TYPE2
OUTFILFMT is used.
The DDS for FILEA would be retrieved to a separate member. Any
comments are deleted and any file and record level keywords. The 'K'
(Key fields) specifications (and any following records) would be
deleted.
The modified DDS for FILEA would be added to the modified QADSPJR2
retrieved source and the To file created. Note that the To file will
not have any key fields. The default for the To file is TAAJRNDBF in
QTEMP.
DSPJRN is then used to output to a temporary file in QTEMP. CPYF
with FMTOPT(*NOCHK) is then used to copy the DSPJRN output to the
created To file.
The default provides for initialization of the data for the DL
(deleted after image) journal entries if IMAGES(*AFTER) is specified
for journaling as opposed to blank values as output by DSPJRN for the
JOESD field. The decimal fields will contain zeros allowing
processing of the output without concern for decimal data errors. If
IMAGES(*BOTH) are specified, no change occurs to the DL entries.
The file is then ready to be processed. If TAAJRNDBF in QTEMP is
used, the file is deleted first. After the output occurs, you may
copy the file to a permanent library.
Using MINENTDTA for a journal
-----------------------------
The Minimize Entry Data function causes primarily the changed fields
to be captured in the journal entry. Therefore, if CRTDBFJRN is
used, the output records will tend to have many fields with blanks or
zeros.
Depending on what you are trying to do with the outfile may make the
MINENTDTA function incompatible with CRTDBFJRN.
However, in a simple case where you are comparing a before and after
image of the same relative record, the information should be
adequate.
Data mapping errors
-------------------
For most types of entries, no data mapping errors will occur. There
are some conditions where this can occur. An example is when the
file contains date and/or time fields and a delete occurs of a
previously deleted record. In this case, the data fields will be
blank and CRTDBFJRN will fail with a data mapping error because
blanks are invalid in such a field.
If you are having data mapping errors, check the entries to see if
you have valid data in 'entry specific data'. If invalid entries are
all for a specific type (such as DL), it is possible to use the ENTYP
parameter and name the types that are valid to convert.
In some cases it may be possible to use CRTDBFJRN2 and have a program
that deletes the records which are causing an error before the use of
CPYF.
CRTDBFJRN escape messages you can monitor for
----------------------------------------------
CPF9898 Sent for various error conditions (not a physical
file, not being journaled, etc)
Escape messages from based on functions will be re-sent.
Using CRTDBFJRN2
----------------
CRTDBFJRN2 is the old form of the tool. It only creates the file.
After the new file is created, you must use DSPJRN to output the
records from the journal. This is a two step process:
- Use DSPJRN to output to a file in QTEMP
- Use CPYF to copy to the file created by CRTDBFJRN
specifying FMTOPT(*NOCHK)
You must be careful with the specifications as no check occurs that
you are generating the correct data base records. You must do the
following:
** Name the journal where the file to be processed is being
journaled.
** Specify JRNCDE = R. These are the Record image entries that
are captured in the journal. No other JRNCDE values should be
used.
** Specify OUTFILFMT = *TYPE2. This is the format of the
QADSPJR2 model file that is used to retrieve the DDS from.
** Specify the name of a temporary file in QTEMP for the OUTFILE
parameter.
The DSPJRN command to use for the previous example would be:
DSPJRN JRN(yyy) FILE(xxx/FILEA) +
JRNCDE(R) OUTPUT(*OUTFILE) +
OUTFILFMT(*TYPE2) OUTFILE(QTEMP/TMPOUTP)
This will default to output all records for the file.
** Use CPYF to copy to the file created by CRTDBFJRN.
CPYF FROMFILE(QTEMP/TMPOUTP) +
TOFILE(xxx/FILEA) MBROPT(*ADD) +
FMTOPT(*NOCHK)
Depending on your file length, you may see messages from CPYF
regarding truncation.
The new file is now ready to be read. You can perform selection
within a query. You may also want to perform selection on DSPJRN to
minimize the number of records written by using the JOB, PGM, or
USRPRF parameters on DSPJRN.
If the file being journaled has more than a single member, you will
probably want to specify that the records for a single member be
output (specify the member name as part of the FILE parameter on the
DSPJRN command.
Querying the file after data is written to it
---------------------------------------------
If you just want to print the data, use PRTDB.
PRTDB2 FILE(TAAJRNDBF)
If you want to select/sort and print the file, use PRTDBS.
PRTDBS FILE(TAAJRNDBF)
You may be interested in sequencing or selecting on the following:
JODATE - Date of entry (job date format)
JOJOB - Job name that caused the entry
JOUSPF - User name that caused the entry
JOUSER may also be used for most cases
JOPGM - Program name that was used to cause the entry
JOCTRR - Relative record number
xxx - One of your field names (such as selecting a value)
You can use the TAAQRY tool for simple requests. A Query product or
writing a specific program could be used for more complex requests.
CRTDBFJRN Command parameters *CMD
----------------------------
FROMFILE The qualified name of the data base file that is
being journaled. It must be a physical file. The
library value defaults to *LIBL. *CURLIB or a
specific library may also be used.
FROMDATE The date and time of the first journal entry to be
considered. The default is *FIRST to use the first
journal entry in the receiver range.
The special value *CURRENT may be entered to mean
today's date.
A specific date may be entered in job format. If no
date is entered, a date of Jan 1, 1940 is used.
A specific time may be entered in HHMMSS format. If
no time is entered, a time of 000000 is used.
TODATE The date and time of the last journal entry to be
considered. The default is *LAST to use the current
date and the last journal entry in the receiver
chain.
A specific date may be entered in job format. If no
date is entered, the current date is used.
A specific time may be entered in HHMMSS format. If
no time is entered, a time of 235959 is used.
FROMMBR The member of the From file to be considered.
*FIRST is the default. *LAST or a specific member
may be specified.
TOFILE The qualified name of the To file to be output. The
default is TAAJRNDBF in QTEMP. If the default is
used, an existing file of the same name is deleted
first to ensure a correct definition.
A specific file name in QTEMP or a permanent library
may be entered. If so, the file must not exist.
OUTFILFMT The outfile format to be used. The value is passed
thru to the DSPJRN command.
*TYPE2 is the default. *TYPE1, *TYPE3, *TYPE4, or
*TYPE5 may also be specified.
ENTTYP The entry type to be considered. *RCD is the
default for record types.
A list of specific entry types may be entered
including:
BR Before image
DL Deleted record
DR Record deleted for rollback
PT Record added to physical file
PX Records a RRN in a physical file
UB Before image
UP After image
UR After image of a record updated for rollback
INZDLTDFT A *YES/*NO value for whether to initialize the
values for the deleted records (record type DL) if
IMAGES(*AFTER) is specified for the journaled file.
If IMAGES(*BOTH) is specified for the journaled
file, the INZDLTDFT parameter is ignored and the
output record will contain the values of the record
before it was deleted.
*YES is the default to initialize the DL records for
IMAGES(*AFTER) type files so that valid data will
exist. Decimal fields will contain zeros.
*NO provides a compatible interface to the use of
the command prior to July 15, 2010. All data in the
user fields are initialized to blanks as exists in
the JOESD field of the DL journal entry if
IMAGES(*AFTER) exists for the journaled file.
CRTDBFJRN2 Command parameters *CMD
-----------------------------
DBF The qualified name of the data base file that is
being journaled. It must be a physical file. The
library value defaults to *LIBL. *CURLIB may also
be used.
NEWFILE The qualified name of the new file that will be
created by the command. It will contain the
QADSPJR2 fields (without JOESD). Added to this will
be the fields from the file named in the DBF
parameter (without comments, record level keywords,
key fields, etc). The file must not exist. The
library value defaults to *CURLIB.
Once the file has been created, it may be used for
different uses with DSPJRN. See the tool discussion
for how to specify the DSPJRN and CPYF commands when
writing to this file.
SRCFILE The qualified name of the source file that will
contain the combined DDS from QADSPJR2 and the file
named in the DBF parameter. The default is *TEMP
meaning that a source file will be used in QTEMP and
you do not want to retain the generated source.
If a specific source file is named, a blank library
name will be assumed to be *LIBL. *LIBL or *CURLIB
may also be entered.
Restrictions
------------
The file named must be a physical file and must be journaled.
The RTVPFSRC command is used to simulate the DDS used for your actual
file. RTVPFSRC has some restrictions.
See the previous comments relative to data mapping errors.
Prerequisites
-------------
The following TAA Tools must be on your system:
CHKMBR2 Check member 2
CHKOBJ2 Check object 2
CHKOBJ3 Check object 3
EDTVAR Edit variable
EXTLST2 Extract list 2
RTVDBFA Retrieve DBF attributes
RTVOBJJRNA Retrieve object journal attributes
RTVPFSRC Retrieve PF source
SNDCOMPMSG Send completion message
SNDESCINF Send escape information
SNDESCMSG Send escape message
SNDJLGMSG Send job log message
Implementation
--------------
None, the tool is ready to use.
Objects used by the tool
------------------------
Object Type Attribute Src member Src file
------ ---- --------- ---------- ----------
CRTDBFJRN *CMD TAAJRNU QATTCMD
CRTDBFJRN2 *CMD TAAJRNU2 QATTCMD
TAAJRNUC *PGM CLP TAAJRNUC QATTCL
TAAJRNUC2 *PGM CLP TAAJRNUC2 QATTCL
TAAJRNUR *PGM RPG TAAJRNUR QATTRPG
TAAJRNUR11 *PGM RPG TAAJRNUR11 QATTRPG
Structure
---------
CRTDBFJRN Cmd
TAAJRNUC CL Pgm
TAAJRNUR RPG Pgm
TAAJRNUR11 RPG Pgm
CRTDBFJRN2 Cmd
TAAJRNUC2 CL Pgm
TAAJRNUR RPG Pgm
|
