The IndexingOutputConfig component is used to generate XHTML documents from Oracle ATG Web Commerce repository items (see the Indexing Repository Data chapter for detailed information). If your project includes repository data, you can use this page to identify properties you are indexing that may not be adding value to your index. Indexing unnecessary properties increases the size of your index and slows response times. This page can provide guidance on ways you might improve structured data processing and therefore Oracle ATG Web Commerce Search performance.
Note, however, that the data represented on this page does not indicate errors, but consists of informational feedback. Your application may have unusual needs, for which these recommendations do not apply. The following image shows an example of information that may be included in this tab:
The analysis page performs the following checks. See the Determining Which Properties to Index section for guidance on applying this information to your IndexingOutputConfig settings.
Per Item Analysis (meta/text)—Number of meta-properties indexed per content item.
Properties with a Single Value (meta)—Meta-properties that have the same value for all content items with that property.
Properties with Long String Values (meta)—Identifies meta-properties that contain strings more than 50 characters long on average.
Text Fields with only Numbers (text)—Identifies text properties that contain only numbers. Such fields can be useful to index in the case of part numbers and other likely search terms, but are likely to be less useful in the case of dates or Boolean values.
Text Fields with Small Contents (text)—Identifies text properties that contain fewer than 10 characters on average.
Text Fields with Multiple Instances (text)—Identifies text fields that occur more than once per content item; depending on the contents, it may be possible and useful to combine such fields into one field.
