----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/10095/#review18319 -----------------------------------------------------------
Ship it! Ship It! - Chris Mattmann On March 23, 2013, 2:44 p.m., Ross Laidlaw wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/10095/ > ----------------------------------------------------------- > > (Updated March 23, 2013, 2:44 p.m.) > > > Review request for oodt and Chris Mattmann. > > > Description > ------- > > The readTags method in org.apache.oodt.cas.product.rss.RSSConfigReader > assumes that org.w3c.dom.Element.getAttribute(java.lang.String) will return > null if a tag does not have a 'source' attribute definition in the RSS > configuration file. Instead the getAttribute method returns an empty String. > This means that by default readTags will give every tag a blank source > attribute. > > We could use the org.w3c.dom.Element.getAttributeNode(java.lang.String) > instead. This method takes the attribute name as a String argument and > returns null if the attribute does not exist for the element. With this > change, if any tags in the RSS configuration file have an explicitly defined > empty source attribute (e.g. <tag name="example" source=""/>), readTags will > add the empty source attribute to the attribute list for the tag. But if a > tag does not have an explicit source attribute definition, readTags will no > longer create a blank source attribute for the tag. > > > This addresses bug OODT-582. > https://issues.apache.org/jira/browse/OODT-582 > > > Diffs > ----- > > > /trunk/webapp/fmprod/src/main/java/org/apache/oodt/cas/product/rss/RSSConfigReader.java > 1460085 > > Diff: https://reviews.apache.org/r/10095/diff/ > > > Testing > ------- > > I implemented a new JUnit test class RSSConfigReaderTest to test for this > bug. Please see OODT-583 for details of the test class. > > > Thanks, > > Ross Laidlaw > >
