Select a number of random results (TRIRIGA)

Mark Grawburg

<p>Hello!</p>
<p> </p>
<p>I do BIRT Report development work with the TRIRIGA platform, currently. I received a spec that called for "captur(ing) a randomly selected 20% or 5 active records (whichever is greater)". Being a report with TRIRIGA, I am not using SQL to create my datasets. The dataset is exported from the client system. </p>
<p> </p>
<p>I am at a loss as to how to grab the random records. Does anyone have any experience with this? How would you go about randomly selecting a calculated amount of records? Thanks for the help and feedback!</p>

Clement Wong

<p>According to this IBM TRIRIGA Application Platform Version 3 Release 4.2 doc (@ <a data-ipb='nomediaparse' href='http://www.ibm.com/support/knowledgecenter/SSHEB3_3.4.2/com.ibm.tap.doc/pdfs/pdf_tap_report.pdf):'>http://www.ibm.com/support/knowledgecenter/SSHEB3_3.4.2/com.ibm.tap.doc/pdfs/pdf_tap_report.pdf):</a></p&gt;
<p> </p>
<p style="margin-left:40px;"><em><strong>Develop a Report in BIRT</strong></em><br>
 </p>
<p style="margin-left:40px;"><em>To develop a BIRT report, you use the BIRT Report Designer to modify the imported.rptdesign file. The BIRT design specifics are beyond the scope of this guide. For detailed information about developing a BIRT report, see the Eclipse –  BIRT<br>
documentation.</em></p>
<p style="margin-left:40px;"><br><em>When you develop a report in BIRT, be aware of the following IBM TRIRIGA/BIRT-specific information.</em></p>
<p style="margin-left:40px;"><br><em><strong>Data Source</strong></em><br>
 </p>
<p style="margin-left:40px;"><em>All data access in a BIRT report is through data sources. A starter report includes a single data source named TRIRIGA. TRIRIGA is a scripted data source and all data sets that obtain their data from an IBM TRIRIGA server need to use this data source. Use this data source as much as possible for all your data access because it goes through the IBM TRIRIGA meta -layer and includes built-in business logic around security and other features.  If the TRIRIGA data source does not provide all the needed data, you can use any of the data access methods that BIRT supports, including JDBC and XML. For direct database access to the TRIRIGA database, you can use a JDBC data source. You must add this data source manually, and it must be named TRIRIGA_DB. By naming it TRIRIGA_DB, at runtime the IBM TRIRIGA Platform dynamically modifies the report so that it interacts with the IBM TRIRIGA database associated to the server within which the report is running.</em></p>
<p> </p>
<p> </p>
<p>and given your requirements, it would seem that going directly against the TRIRIGA database would be your best option.</p>