Joseph Kesselman wrote: > > Maybe we ought to rename "gump" to "grump"? <grin/> (My own image of it has > a lot more to do with the Gump in the Wizard of Oz than with Forrest > Gump... attach a bunch of pieces, sprinkle the powder of life over them, > and see if they'll fly.)
Chuckle! If you take a look at any one single data point, then it would certainly seem that way. See below, however for another perspective. > This report confuses me somewhat. D13 passes smoketest on my machine, > running under Sun JDK 1.3.1_01; I wouldn't have released it otherwise. Just > re-ran it and it's still happy. Admittedly, I'm running directly from the > build's jarfiles rather than unpackaging the zip/tar; we could try that, or > try running on another machine, as a sanity check. First, take a look at: http://jakarta.apache.org/builds/gump/2001-11-12/xml-xalan2-smoketest.html You will see that it passes. Even if you consider all of the projects involved (and in this case, there only is Ant, Xerces, and Xalan to contend with), there is only a small number of changes which occurred in the 24 hour period. There were three source files changed in Xerces - perhaps they were involved? One project making a change breaking another project is not all that uncommon. For a current example, take a look at http://nagoya.apache.org/bugzilla/show_bug.cgi?id=2650 http://jakarta.apache.org/builds/gump/2001-11-13/xml-security.html If you are confident that the problem is not in Xalan, I can help do a little divide and conquer by rolling back selected changed until the problem goes away and thereby identify which change was the root cause. - Sam Ruby
