Auto-generated columns in the BIRT (and IXIASOFT CCMS) data sets

When you select the Group by values operation in the first column of the Data Set Definition dialog, IXIASOFT CCMS automatically generates several additional columns that will be available as data set columns when you are working in the BIRT environment.
  • Documents Count – gives you the number of documents returned by the grouping operation in the first column. Or by the combined criteria of columns one and two, if you’re grouping by two sets of criteria: documents per writer per month, for example.
  • Running Count – totals the values in the Count column in subsequent cells.
  • Values Count – gives the number of values, as opposed to the number of documents, returned by the first column.

You can see them in the CMS data source TSV file.

Documents Count and Running Count

You can see these columns if you open the CCMS data source that’s been produced using a grouping operation in the first column. The CCMS data sources are stored as TSV files, which you can easily open in a spreadsheet application.

The image below shows the TSV file produced if you select Creation Date for your first column and then apply the Group by values operation. It's also filtered to show the results by Year-Month.
  • The first column (column A) contains all the months when documents were created. These range from September 2006 (2006-09) through September 2007 (2007-09).
  • Column D (Documents Count) shows the number of documents that correspond to the criteria in column A. In the first row, we can see that 114 documents were created in 2006-09.
  • Column E (Running Count) progressively sums the Document Count entries. The last entry at the bottom of this column is the total number of documents that meet the criteria – 286.

Count column TSV

Values Count

For many grouping operations the Values Count will be the same as the Documents Count. This is the case for the example like the one shown above—there’s never more than one creation date for a document.

But queries on stamp-based information may give you interesting results. The example below shows the values returned by a query on the Authored Stamp.

The Authored Stamp returns information which includes not only the date that a document was put into the review cycle, but the number of such events. If the document is put into the review cycle ten times before it's approved, then there will be a total of ten date values.

If we look at the first data row, we can see that in May of 2009 (2009-05), 6 documents were put into the review cycle – that's the number in the Documents Count column.

We can also see that the query found 25 occurrences of the date stamp in the documents – that's the number in the Values Count column.

Taken together these values indicate that each of these documents are going through the review process an average of four times apiece. This is one sort of analysis that you could perform on these figures; there might be others.

If you wanted to get a clearer picture of the process, you could set up your data set definition to first group by Year-Month, and then group by User. This would give you the number of times each writer needs to send a topic through review before it is approved.

Values count