SonarQube Plugins

OutOfMemory error on simple Flex Project due to FlexPMD

Details

  • Type: Bug Bug
  • Status: Open Open
  • Priority: Major Major
  • Resolution: Unresolved
  • Affects Version/s: FLEX-1.2
  • Fix Version/s: None
  • Component/s: Flex
  • Labels:
  • Environment:
    Ubuntu 12.04
    Sonar 3.3
    MAVEN_OPTS=-Xmx2048m -XX:MaxPermSize=256m
  • Number of attachments :
    0

Description

After upgrade to latest versions of Sonar and Flex plugin some code analysis has started to fail.

Full stack trace here:

INFO: Rules number in the ruleSet: 85
Oct 29, 2012 1:03:16 PM de.bokelberg.flex.parser.AS3Scanner$XMLVerifier verify
WARNING: The markup in the document preceding the root element must be well-formed. at org.apache.xerces.util.ErrorHandlerWrapper.createSAXParseException(Unknown Source)
org.apache.xerces.util.ErrorHandlerWrapper.fatalError(Unknown Source)
org.apache.xerces.impl.XMLErrorReporter.reportError(Unknown Source)
[ERROR] Java heap space -> [Help 1]
java.lang.OutOfMemoryError: Java heap space
	at java.util.Arrays.copyOf(Arrays.java:2894)
	at java.lang.AbstractStringBuilder.expandCapacity(AbstractStringBuilder.java:117)
	at java.lang.AbstractStringBuilder.append(AbstractStringBuilder.java:407)
	at java.lang.StringBuffer.append(StringBuffer.java:241)
	at de.bokelberg.flex.parser.AS3Parser.parseMetaData(AS3Parser.java:1577)
	at de.bokelberg.flex.parser.AS3Parser.parseClassContent(AS3Parser.java:157)
	at de.bokelberg.flex.parser.AS3Parser.parseClass(AS3Parser.java:942)
	at de.bokelberg.flex.parser.AS3Parser.parsePackageContent(AS3Parser.java:290)
	at de.bokelberg.flex.parser.AS3Parser.parsePackage(AS3Parser.java:1768)
	at de.bokelberg.flex.parser.AS3Parser.parseCompilationUnit(AS3Parser.java:204)
	at de.bokelberg.flex.parser.AS3Parser.parseLines(AS3Parser.java:1554)
	at de.bokelberg.flex.parser.AS3Parser.buildAst(AS3Parser.java:109)
	at com.adobe.ac.pmd.files.FileSetUtils.tryToBuildAst(FileSetUtils.java:182)
	at com.adobe.ac.pmd.files.FileSetUtils.buildAst(FileSetUtils.java:78)
	at com.adobe.ac.pmd.FlexPmdViolations.computeAsts(FlexPmdViolations.java:126)
	at com.adobe.ac.pmd.FlexPmdViolations.computeAsts(FlexPmdViolations.java:109)
	at com.adobe.ac.pmd.FlexPmdViolations.computeViolations(FlexPmdViolations.java:88)
	at com.adobe.ac.pmd.engines.AbstractFlexPmdEngine.computeViolations(AbstractFlexPmdEngine.java:142)
	at com.adobe.ac.pmd.engines.AbstractFlexPmdEngine.executeReport(AbstractFlexPmdEngine.java:119)
	at org.sonar.plugins.flex.flexpmd.FlexPmdSensor.execute(FlexPmdSensor.java:85)
	at org.sonar.plugins.flex.flexpmd.FlexPmdSensor.analyse(FlexPmdSensor.java:63)
	at org.sonar.batch.phases.SensorsExecutor.execute(SensorsExecutor.java:64)
	at org.sonar.batch.phases.Phases.execute(Phases.java:97)
	at org.sonar.batch.bootstrap.ProjectModule.doStart(ProjectModule.java:128)
	at org.sonar.batch.bootstrap.Module.start(Module.java:83)
	at org.sonar.batch.bootstrap.BatchModule.analyze(BatchModule.java:114)
	at org.sonar.batch.bootstrap.BatchModule.doStart(BatchModule.java:104)
	at org.sonar.batch.bootstrap.Module.start(Module.java:83)
	at org.sonar.batch.bootstrap.BootstrapModule.doStart(BootstrapModule.java:121)
	at org.sonar.batch.bootstrap.Module.start(Module.java:83)
	at org.sonar.batch.bootstrapper.Batch.startBatch(Batch.java:73)
	at org.sonar.batch.bootstrapper.Batch.execute(Batch.java:60)

I've increased my java heap space to 2GB in order to give it a try, but I don't think that's exactly a problem with my machine.

I've tried with different quality profiles (new default and a previous existing one) but it doesn't make any difference.

Also, another quality analysis has stop working on another complex project. In that case the anylisis hangs forever.

Issue Links

is duplicated by

Bug - A problem which impairs or prevents the functions of the product. SONARPLUGINS-2414 Abode FlexPMD : OutOfMemory when parsing some MXML files

  • Major - Major loss of function.
  • Closed - The issue is considered finished, the resolution is correct. Issues which are not closed can be reopened.
relates to

Improvement - An improvement or enhancement to an existing feature or task. SONARPLUGINS-2413 Do not use FlexPMD

  • Major - Major loss of function.
  • Open - The issue is open and ready for the assignee to start work on it.

Activity

Ascending order - Click to sort in descending order
Hide
Permalink
Miguel Martín-Forero added a comment - - edited

I also tried using a Sonar Runner and the problem persists: it gives an OutOfMemory error no matter how much max heap size it is assigned (and it occures very fast).

It only works if it has 0 active rules and fails with only 1 active rule (I've tried activating different rules but this doesn't help).

Info about my JDK installation:
java version "1.7.0_07"
OpenJDK Runtime Environment (IcedTea7 2.3.2) (7u7-2.3.2a-0ubuntu0.12.04.1)
OpenJDK 64-Bit Server VM (build 23.2-b09, mixed mode)

Show
Miguel Martín-Forero added a comment - - edited I also tried using a Sonar Runner and the problem persists: it gives an OutOfMemory error no matter how much max heap size it is assigned (and it occures very fast). It only works if it has 0 active rules and fails with only 1 active rule (I've tried activating different rules but this doesn't help). Info about my JDK installation: java version "1.7.0_07" OpenJDK Runtime Environment (IcedTea7 2.3.2) (7u7-2.3.2a-0ubuntu0.12.04.1) OpenJDK 64-Bit Server VM (build 23.2-b09, mixed mode)
Hide
Permalink
Miguel Martín-Forero added a comment -

I've finally come to a solution (I don't know if it deserve to be called a workaround though) to these problems.

I was having two related issues:

  • An OutOfMemory error whichs was triggered very fast.
  • An analysis that was hanging forever

These problems were provoked by two different files.

1. The OutOfMemory was caused by this code: <fx:Script fb:purpose="styling">. Removing the fb:purpose="styling" part solved the issue.
2. Analysis freeze was cause by a single comment inside a switch case block ... :S like the following one:
if (condition) {
switch (value)

{ /* A harmless comment? */ case option: ... }

}

Removing the comment solved it too.

I've reduced the original code for cleanliness.

Hope this helps to solve these issues in the plugin itself.

Show
Miguel Martín-Forero added a comment - I've finally come to a solution (I don't know if it deserve to be called a workaround though) to these problems. I was having two related issues: An OutOfMemory error whichs was triggered very fast. An analysis that was hanging forever These problems were provoked by two different files. 1. The OutOfMemory was caused by this code: <fx:Script fb:purpose="styling">. Removing the fb:purpose="styling" part solved the issue. 2. Analysis freeze was cause by a single comment inside a switch case block ... :S like the following one: if (condition) { switch (value) { /* A harmless comment? */ case option: ... } } Removing the comment solved it too. I've reduced the original code for cleanliness. Hope this helps to solve these issues in the plugin itself.
Hide
Permalink
Evgeny Mandrikov added a comment - - edited

Hi Miguel,

Thanks for your analysis! Now it's absolutely clear that both problems come from FlexPMD, which we want to fully replace by SSLR. And actually I don't see other ways to solve the issue since it comes from FlexPMD, which is out of our responsibility and we don't want to continue hacking/patching it.

Show
Evgeny Mandrikov added a comment - - edited Hi Miguel, Thanks for your analysis! Now it's absolutely clear that both problems come from FlexPMD, which we want to fully replace by SSLR. And actually I don't see other ways to solve the issue since it comes from FlexPMD, which is out of our responsibility and we don't want to continue hacking/patching it.
Hide
Permalink
Miguel Martín-Forero added a comment -

Hi Evgeny,

thank you very much for your comment. Hoping to see the replacement of the buggy FlexPMD by SSLR soon

In the meantime, I'll try to avoid these problems in order to have a happier developer life. fb:purpose="styling" shouldn't be in my code in the first place. I'm a little more worried about the second issue: this problem demonstrate the buggyness of FlexPMD.

Show
Miguel Martín-Forero added a comment - Hi Evgeny, thank you very much for your comment. Hoping to see the replacement of the buggy FlexPMD by SSLR soon In the meantime, I'll try to avoid these problems in order to have a happier developer life. fb:purpose="styling" shouldn't be in my code in the first place. I'm a little more worried about the second issue: this problem demonstrate the buggyness of FlexPMD.

People

  • Assignee:
    Unassigned
    Reporter:
    Miguel Martín-Forero
Vote (0)
Watch (4)

Dates

  • Created:
    Updated: