Data2Xml - Multiple Level 01

[Brian Reynolds]

Hi Bruce,

I have a copybook with multiple level 01 elements, and am trying to generate an XML doc from this using Cobol2Xml.newCobol2Xml. The setSplitCopybook function throws a "not yet implemented" exception, so I was wondering what sort of effort do you feel would be involved in doing this? I'd be happy to give it a shot myself if you can provide me with some high-level pointers.

E.g.

000100 01  DATA-HDR.
000300     03  MSG-HDR-FIELDS.
000400         05  VRICWS-MSG-NAME PIC  X(30).
000600         05  VRICWS-MSG-VER  PIC  X(6).
000900 01  DATA-OUTPUT.                                                 
001100     03  SVC-RETURN PIC  X(01).                                  
001300         88  SVC-PASSED           VALUE 'P'.                       
001400         88  SVC-FAILED           VALUE 'F'.                       
001600     03  RETURN-CODE-DATA.                                         
001900         05  RC-DATA OCCURS 20 TIMES.
002200             07  RETURN-CODE PIC  X(10).                                  
002400             07  RETURN-DESC PIC  X(79).                                  

Msgname                       0001  P1110080021This is a description of the error

I had a quick look at this in a debugger, and the cause is obviously the pos field in FieldDetail starts at 0 when we come around to processing the DATA-OUTPUT data.

Thanks.

[Bruce Martin]

Sorry for the delay responding, I started to yesterday but got distracted before
I finished. I am getting snowed under with requests at the moment.

With regards to supporting multiple records, there are 2 changes required:

1) Splitting the copybook up into seperate records
2) Working out which record you have

With regards 1) you should be able to manage this for Split=01 and Split=Highest-Repeating,
As you want split=01, I will provide more information about this (tomorrow ???).
Split=Redefine will be more difficult and I suggest leaving it for the moment

With regards 2) it looks like the Position in the file determines the record type
(i.e. the first record is the Header-Record. This is probably best done by me

Bruce

[Bruce Martin]

Currentl

                                                                  +-------------> JRecord LayoutDetail class 
                                                                  !
Cobol-Copybook ---> (cb2xml pgm) ----> xml-document  +
                                                                  !
                                                                  +-->(jaxb) --> Copybook/Item class's

Basically the Copybook/Item are used to create the Xml structure and
the LayoutDetail is used to retrieve values via JRecord.

If you you Split=01, you will get:

LayoutDetail +--- RecordDetail name=Record1
                          !
                          +--- RecordDetail name=Record2
                          !
                          +--- RecordDetail name=Record3

Copybook --- Item +----- Item nameToUse=Record1
                  ! 
                  +----- Item nameToUse=Record2
                  ! 
                  +----- Item nameToUse=Record2

in class Cobol2GroupXml, method cobol2xml , the following code needs to change

    if (items.size() == 1) {
         Item item = items.get(0);
            while ((l = r.read()) != null) {
                //System.out.println(l.getFullLine());
                writeItem(writer, l, item, new IntStack());
            }
        } else {
            while ((l = r.read()) != null) {
                writer.writeStartElement("Line");
                writeItems(writer, l, items, new IntStack());
                writer.writeEndElement();
            }
       }

[Bruce Martin]

I have done most of it now; just testing + Change to support arrays in Xml2Data program to go

[Bruce Martin]

I have released a new version of JRecord:

• https://sourceforge.net/projects/jrecord/files/jrecord/Version_0.80.8g

where the cobol <==> xml programs support multi-rcord files. You must use
the program interface (to set Record-Selection-Criteria):

    Cobol2GroupXml.newCobol2Xml(copybookFileName)
                       .setFileOrganization(Constants.IO_BIN_TEXT)
                       .setDialect(ICopybookDialects.FMT_INTEL)
                      .setSplitCopybook(CopybookLoader.SPLIT_01_LEVEL)
                           .setRecordSelection("PO-Record", newFieldSelection("Record-Type","H1"))
                           .setRecordSelection("Product-Record", newFieldSelection("Record-Type","D1"))
                           .setRecordSelection("Location-Record", newFieldSelection("Record-Type","S1"))
                   .cobol2xml(new FileInputStream(dataFileName), os);

---

At the moment, each different record appears a level 2 in the hierarchy. But does have parent Record field. This is not currently used in JRecord (in the RecordEditor it is) this would allow
files like

PO-Header-Record
Product-Record
Store-Record
Stroe-Record

   i.e

   Po        
      +---  Product
                                 +--- Store 
                                 +--- Store
       +--- Product

to be represented properly in Xml

[Brian Reynolds]

Bruce,
Awesome, thanks for this.