Description of XMLPARSE

Download count: 3 this month, 5923 altogether.
Downloads for XMLPARSE:
  VMARC archive: v-28K

XMLPARSE - XML parsing CMS/TSO Pipelines filter

Recent changes:

  • 2005-08-10 Significant enhancements:
    1. Handle ">" correctly within content data. The first ">" after a tag is still taken to be the end of the tag so any ">" within an attribute value must still be represented as ">".
    2. Provide independent options as to whether tag names, content data, attribute names and attribute values should be converted to upper case or left in mixed case.
    3. Support optional element and attribute validation against an in-line DTD.
  • 2003-10-29 Significant enhancements:
    1. Remove unhelpful assumption that input stream is in card image format with sequence numbers. The code now checks whether the first record is 80 bytes with digits in columns 73-80 and assumes sequence numbers are present if so, otherwise it uses the full length of each input record.
    2. Support CDATA sections containing literal occurrences of markup characters and entities. Recognition of such sections relies on square brackets being at code points hex AD and BD, as for code page 1047.
    3. Support new WHITESPACE option which preserves all white space (blanks, tabs and newlines) in character data and attributes, allowing the calling program to decide which fields are to be treated as CDATA and which are to be normalized. The sample functions for accessing attributes and text data have been updated to support this option.
    4. Document known restrictions and limitations more completely.
  • 2002-08-28 First released

The XMLPARSE package contains a REXX filter for CMS/TSO Pipelines which reads an XML document and converts it to a tree structure stored in a REXX stem. The tree structure can then be navigated either sequentially or by looking up specific tag or attribute names. A set of sample REXX tree navigation functions is included.

The filter is provided as REXX source code, but may be compiled to improve performance for large XML documents.

If you have any comments, questions or suggestions relating to this package, please let me know.

Jonathan Scott - IBM Hursley, UK
E-mail: jonathan_scott@vnet.ibm.com