Welcome !

What is TXT2XML ?

TXT2XML is a Rexx script that converts a text file to and from an XML file using a COBOL copybook as reference. Rexx is a scripting language created by IBM. COBOL is a programming language mainly used on big computers also called mainframes.

What is XML2COB ?

XML2COB is a complementary tool of TXT2XML. It generates a COBOL copybook using an XML file as template. You can use this COBOL copybook to convert XML files to and from text files with TXT2XML.

For which operating system ?

TXT2XML and XML2COB were, at the beginning, written for IBM MVS, OS/390 & Z/0S mainframe operating systems. But, thanks to Regina Rexx and Tiny Cobol, TXT2XML and XML2COB run on Windows and Linux platforms. TXT2XML and XML2COB shoud also run, but was never tested, on HP-UX, AIX, UNIX, OS/2, BeOS, AmigaOS, Epoc32, AtheOS, QNX, MacOS X, SkyOS, ...

Is it a batch (shell) script or a GUI tool ?

Both ! Starting with version 1.25, calling TXT2XML and XML2COB without any parameters will start, thanks to Thinlet, a Java GUI (Graphical User Interface). On IBM MVS, an ISPF is started if TXT2XML or XML2COB is invoked without any parameters

Pre-requisites

To run TXT2XML and XML2COB on non-mainframe platforms (i.e. Windows, Linux, ...), you have to install :

See the documentation for more details.

How does it work ?

During text to XML conversion, the COBOL items "meta-data" (name, level, type, sign, default value, length, starting position and occurence) are read from the copybook. A record is read from the text file, item data is extracted from the text record and the XML element with his content is written in the XML file. The name of the XML element is the capitalized COBOL item name with dashes ("-") translated to underscores("_").
During XML to text conversion, the COBOL items "meta-data" are read from the copybook. The XML elements are read one by one and checked against COBOL copybook items. At each XML first child element, a text record is written
During the COBOL copybook generation, the XML elements, attributes and contents are analyzed to compute (level) or extract (name, type, sign, length and occurence) their "meta-data". For example, the name of the COBOL item is the XML element capitalized name with underscores("_") translated to dashes ("-"). This analysis is restarted at each XML first child element. At the end of the XML file, the starting position of each item is computed for reporting and the COBOL copybook is written.

Are there any limitations in the conversion or generation process ?

Yes. See the documentation for more details. For example, REDEFINES clauses, two or more dimension arrays in COBOL copybooks and XML entities are not supported.