Data Comparison

Introduction

For more data-centric XML resources, the comparison pipeline may have a number of design considerations and priorities different from those for comparing document-centric resources (as described in the previous section). This section outlines comparison features that are more significant in this context, but of course, many features described in the User Guide to Document Comparison page may also apply.  You may also wish to consider the XML Data Compare product.

Numeric Tolerances

For comparison of floating point numbers there may be a requirement to ignore value differences within a specified tolerance, this tolerance can be implemented via output filters based on existing filter resources included in XML Compare, Numeric Tolerances is a worked example of this.

Comparing Large Datasets

When comparing large datasets there are some extra factors to consider, these are covered in the Comparing Large Files guide.

Ignoring Changes

For cases where changes in data are expected but not deemed significant, changes can be 'ignored' in the processing pipeline, a technique for this is explained in the sample: Ignoring Changes.