Directives 

Formats

Where:

Description

The REPEAT DATA directive is used to control the automatic replication of all PRINT, WRITE, INSERT, UPDATE or REMOVE directives sent to one file (the primary file) to another file (the replicated file).

The purpose of this directive is to allow programs to create mirror copies of reports and/or data files without having to issue identical PRINT or WRITE statements to two separate files. While data replication is active, all output directives to the primary file will be sent to the replicated file.

When repeating the output, the system will internally process the directive twice - once for the primary file and again for the replicated file.

Note:
Auto-increment/decrement options will only be done once.

A unique error handler can be specified to trap any errors that might occur during the replication process. If no error handler is present, normal error processing will occur with the value in LFA indicating the file on which the error occurred.

(The REPEAT DATA directive was added in PxPlus v 7.00.)

Format 1

Replicate Output Sent to One File on Another

REPEAT DATA TO file1 ON file2, ERR=stmtref

This format initiates the replication of all PRINT, WRITE, INSERT, UPDATE or REMOVE directives sent to file1 on file2. Once the directive is executed, all subsequent output directives to file1 will also be sent to file2. Both files must be open at the time the directive is executed.

When doing REPEAT DATA, the system re-processes the WRITE or REMOVE directive. This allows you to replicate the data to files with different IOLists or formats. For example, you might have a Customer master file with a large amount of detail and a potential cross reference key file that could be used to provide additional keys. The updates would effectively allow you to have your alternate keys in the secondary file always in sync with the main master file. It also would allow you to write the data to an external database where the format may be different. The main file might have fixed fields using the CHR or LEN formatting, whereas on the database, you might want everything in unique columns.

Note:
The system will check to see if the file you are repeating data for itself is the target of data replication. If a looping condition exists within the data replication (file1 repeats file2 and file2 repeats file1), an Error #89: File access denied - I/O operation pending will be generated.

Format 2

Terminate Replication

REPEAT DATA TO file1 END, ERR=stmtref

This format terminates the replication of the output going to file1.

Format 3

Define a Control File for Auto-Replication

REPEAT DATA CONTROL FILE ctrlfile$, ERR=stmtref

This format defines the control file that the system will check to see if automatic replication is desired. If the pathname of a file being updated matches an entry in this control file, the system will initiate replication automatically. The pathname check is performed whenever the first output operation is done to a file.

Note:
This directive format can only be used when the Data Mirroring (SQL Mirror) module is enabled within PxPlus.

For additional information on the format of this file, see Data Mirroring.

Format 4

Define an Error Handler for Replication

REPEAT DATA ERROR_HANDLER errhdlr$, ERR=stmtref

This format defines an error handler that will be invoked whenever an error occurs while attempting to replicate an output directive, at which point the program defined by errhdlr$ will be called with the following information:

The error handler is intended to allow the application to log the problem encountered and to take the necessary corrective action. Generally, it should EXIT in order to resume execution of the original program.

If the error handler exits with an error itself (EXIT nnn), the application will have the original error code reported and the standard program exception logic processing will apply.

Example

In the following example, the printed report will be sent both to the viewer and to a PDF file. All PRINT directives will send output to both files.

     begin
     open (1)"*viewer*"
     open (2)"*pdf*;file=\junk.pdf"
     repeat data to 1 on 2
          print (1)'font'("Courier New",1),'DF',
          while 1
               read data COMPANY$,CUSTNO$,NAME$,ADDR$,OWES,end=*break
               if LINESLEFT<1 or COMPANY$<>CURCOMP$ \
                    then gosub NEWPAGE
               print (1)CUSTNO$,@(10),NAME$,@(40),ADDR$,@(70),OWES:"-$###,##0.00"
               LINESLEFT--
          wend
     end
!
NEWPAGE:
     if PAGENO<>0 \
          then print (1)'FF',
     PAGENO++
     print (1)"Page:",PAGENO:"###0",@(30),"Client list for "+COMPANY$
     print (1)
     print (1)"Client",@(10),"Name",@(40),"Address",@(70),"Balance"
     print (1)"------",@(10),dim(20,"-"),@(40),dim(20,"-"),@(70),"------------"
     print (1)
     LINESLEFT=50
     CURCOMP$=COMPANY$
     return
     data "ABC","000047","Cyclops Car Dealer","6675 Cherry",3.12
     data "ABC","000122","Global Undergarments","823 Maple Lodge Court",99.54
     data "ABC","000192","The Hungry Incorporated","5582 2nd Avenue",.55
     data "ABC","000295","The Hungry Magazine","8576 Major Mackenzie",71.74
     data "ABC","000310","New Dimensions Company","8052 Center Avenue",5.93
     data "ABC","000332","Cyclops Computing","7723 Fantasy Island",7.61
     data "DEF","000355","Kitty Kay Undergarments","2124 Major Mackenzie",4.55
     data "DEF","000385","New Age Importers","1743 Allstate Parkway",73.84
     data "DEF","000461","SOTA Golf club","4377 Major Mackenzie",63.66
     data "DEF","000555","Mike was here","8450 Buga-Buga Drive",11.43

File replication can be cascaded for output to multiple files. For example, the logic in NEWPAGE could be changed to produce one PDF file for each company:

NEWPAGE:
     if PAGENO<>0 \
          then print (1)'FF',
     close (3)
     open (3)"*pdf*;file=\"+COMPANY$+".pdf"
     print (3)'font'("Courier New",1),'DF', ! Set font
     repeat data to 2 on 3
          PAGENO++
          print (1)"Page:",PAGENO:"###0",@(30),"Client list for "+COMPANY$
          print (1)
          print (1)"Client",@(10),"Name",@(40),"Address",@(70),"Balance"
          print (1)"------",@(10),dim(20,"-"),@(40),dim(20,"-"),@(70),"------------"
          print (1)
          LINESLEFT=50
          CURCOMP$=COMPANY$
          return