stevedlawrence commented on a change in pull request #135: Show different error 
message for OOM
URL: https://github.com/apache/incubator-daffodil/pull/135#discussion_r232822408
 
 

 ##########
 File path: 
daffodil-runtime1/src/main/scala/org/apache/daffodil/processors/parsers/Parser.scala
 ##########
 @@ -271,7 +271,7 @@ class ChoiceParser(ctxt: RuntimeData, val childParsers: 
Vector[Parser])
       case t: Throwable => {
         if (pBefore != null) {
           markLeakCausedByException = true
-          if (!t.isInstanceOf[SchemaDefinitionDiagnosticBase] && 
!t.isInstanceOf[UnsuppressableException]) {
+          if (!t.isInstanceOf[SchemaDefinitionDiagnosticBase] && 
!t.isInstanceOf[UnsuppressableException] && !t.isInstanceOf[OutOfMemoryError]) {
 
 Review comment:
   This pattern of schema checking also appears in SequenceParserBases.scala, 
we probably need a similar check there.
   
   Also, rather than checking the very specific OutOfMemoryError, perhaps we 
instead want to check it's parent java.lang.Error,? The javadoc says 
java.lang.Error "indicates a serious problem that a reasonable application 
should not try to catch". So on any java.lang.Error we still attempt to clean 
up the MarkPool but don't try to warn about an unexpected exception thrown. The 
below message is really trying to tell when Daffodil failed to catch an 
exception that it should have caught elsewhere else.
   
   This also makes me wonder if we shouldn't even be catch/using Throwable's 
anywhere, but instead only catch java.lang.Exceptions? I'm pretty sure we use 
Throwables all over the place, so it's not worth changing for this bug, but 
might be worth creating a new bug to reexamine our usage of Throwables vs 
Exceptions.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

Reply via email to