Fix build failure with JDK 9 EA b148 #470

Godin · 2016-12-09T18:16:54Z

Failed to execute goal org.codehaus.groovy.maven:gmaven-plugin:1.0:execute (parse-version) on project org.jacoco.build

Caused by: java.lang.reflect.InaccessibleObjectException: Unable to make java.lang.String(char[],int,int,java.lang.Void) accessible: module java.base does not "opens java.lang" to unnamed module @603cabc4
	at java.base/jdk.internal.reflect.Reflection.throwInaccessibleObjectException(Reflection.java:427)
	at java.base/java.lang.reflect.AccessibleObject.checkCanSetAccessible(AccessibleObject.java:201)
	at java.base/java.lang.reflect.Constructor.checkCanSetAccessible(Constructor.java:184)
	at java.base/java.lang.reflect.Constructor.setAccessible(Constructor.java:177)
	at org.codehaus.groovy.reflection.CachedConstructor.<init>(CachedConstructor.java:32)
	at org.codehaus.groovy.reflection.CachedClass.getConstructors(CachedClass.java:233)
	at groovy.lang.MetaClassImpl.<init>(MetaClassImpl.java:115)
	at groovy.lang.MetaClassRegistry$MetaClassCreationHandle.createNormalMetaClass(MetaClassRegistry.java:102)
	at groovy.lang.MetaClassRegistry$MetaClassCreationHandle.create(MetaClassRegistry.java:92)
	at org.codehaus.groovy.runtime.metaclass.MetaClassRegistryImpl.getGlobalMetaClass(MetaClassRegistryImpl.java:252)
	at org.codehaus.groovy.runtime.metaclass.MetaClassRegistryImpl.access$100(MetaClassRegistryImpl.java:45)
	at org.codehaus.groovy.runtime.metaclass.MetaClassRegistryImpl$LocallyKnownClasses.getFromGlobal(MetaClassRegistryImpl.java:112)
	at org.codehaus.groovy.runtime.metaclass.MetaClassRegistryImpl$LocallyKnownClasses.getMetaClass(MetaClassRegistryImpl.java:88)
	at org.codehaus.groovy.runtime.metaclass.MetaClassRegistryImpl$MyThreadLocal.getMetaClass(MetaClassRegistryImpl.java:361)
	at org.codehaus.groovy.runtime.metaclass.MetaClassRegistryImpl.getMetaClass(MetaClassRegistryImpl.java:265)
	at org.codehaus.groovy.runtime.InvokerHelper.invokePojoMethod(InvokerHelper.java:765)
	at org.codehaus.groovy.runtime.InvokerHelper.invokeMethod(InvokerHelper.java:754)
	at org.codehaus.groovy.runtime.ScriptBytecodeAdapter.invokeMethodN(ScriptBytecodeAdapter.java:170)
	at script1481306263465.run(script1481306263465.groovy:6)
	at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
	at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
	at java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
	at java.base/java.lang.reflect.Method.invoke(Method.java:538)
	at org.codehaus.groovy.maven.runtime.support.ScriptExecutorSupport.invokeMethod(ScriptExecutorSupport.java:158)
	... 26 more

According to https://twitter.com/CedricChampeau/status/807285853580103684 Groovy should be updated to 2.4.8

marchof · 2016-12-15T07:03:13Z

.travis.sh

@@ -86,6 +86,8 @@ case "$JDK" in
  ;;
 9-ea | 9-ea-stable)
  # see https://bugs.openjdk.java.net/browse/JDK-8131041 about "java.locale.providers"


This comment refers to the line below the next line of code. Moving down?

@marchof you mean change order of comments? Unfortunately it can't be placed exactly above line or on line to which it refers.

I see. What about concatenating MAVEN_OPTS step-by-step instead of using a long multi-line?

@marchof not sure what you mean - one comment is about java.locale.providers that is not part of MAVEN_OPTS. Maybe possible to construct whole command in variable step-by-step, but I'm afraid to run into problem with escaping of quotes. Following suggestion of @bjkail below I'm going to split this command on multiple.

In fact will be better to get rid of this java.locale.providers completely, but that's another story and I'll try to handle it separately.

@Godin Something like this:

# Groovy version should be updated to get rid of "--add-opens" options (see https://twitter.com/CedricChampeau/status/807285853580103684) MAVEN_OPTS="--add-opens java.base/java.lang=ALL-UNNAMED --add-opens java.base/java.io=ALL-UNNAMED" MAVEN_OPTS=$MAVEN_OPTS mvn -V -B -e verify -Dbytecode.version=1.9 # see https://bugs.openjdk.java.net/browse/JDK-8131041 about "java.locale.providers" MAVEN_OPTS=$MAVEN_OPTS -DargLine=-Djava.locale.providers=JRE,SPI

@marchof I'm not sure that -DargLine can be passed via MAVEN_OPTS - usually last one is used to pass options that control Maven JVM, while -D is argument to Maven invocation

@marchof ouch, and what you wrote isn't gonna work as is, and looks like attempt to construct the whole command in variable.

Anyway command has been divided on two, where comments are above of respective parts - see updated commit. Seems quite acceptable to me given that actually all this should be gone eventually.

marchof

@Godin Didn't we talk about removing at at all? Otherwise we should document the required JVM options in URLStreamHandlerRuntime JavaDoc.

Godin · 2016-12-15T15:45:39Z

@marchof Well, there is separate ticket about removal - #471 , which anyway requires change in documentation - see page about implementation details, and IMO it will be nice to preserve information about it on this page.

Also for instrumentation with companion classes reflection seems mandatory, so maybe we'll update code of agent to allow reflection. And since there is --add-opens option in JDK 9 I wasn't entirely sure about removal. However sure that we'd better fix build failure to ease process of testing of further JDK 9 EA builds.

WDYT?

bjkail · 2016-12-15T15:48:35Z

.travis.sh

@@ -86,6 +86,8 @@ case "$JDK" in
  ;;
 9-ea | 9-ea-stable)
  # see https://bugs.openjdk.java.net/browse/JDK-8131041 about "java.locale.providers"
+  # Groovy version should be updated to get rid of "--add-opens" options (see https://twitter.com/CedricChampeau/status/807285853580103684)
+  MAVEN_OPTS="--add-opens java.base/java.lang=ALL-UNNAMED --add-opens java.base/java.io=ALL-UNNAMED" \
  mvn -V -B -e verify -Dbytecode.version=1.9 \


Very minor, but perhaps indent the mvn line? The trailing \ on the MAVEN_OPTS line can be difficult to spot.

@bjkail indeed, done

marchof · 2016-12-15T17:05:33Z

@Godin I'm ok with this fix to make the latest JDK 9 EA work again.

Godin added the component: build label Dec 9, 2016

Godin self-assigned this Dec 9, 2016

Godin force-pushed the issue-470 branch 3 times, most recently from 587a943 to 70804ba Compare December 14, 2016 23:39

Godin added this to the 0.7.9 milestone Dec 14, 2016

Godin requested a review from marchof December 15, 2016 00:05

marchof reviewed Dec 15, 2016

View reviewed changes

marchof requested changes Dec 15, 2016

View reviewed changes

bjkail reviewed Dec 15, 2016

View reviewed changes

marchof approved these changes Dec 15, 2016

View reviewed changes

Fix build failure with JDK 9 EA b148

f7660de

Godin force-pushed the issue-470 branch from 70804ba to f7660de Compare December 15, 2016 17:12

marchof merged commit 2bc56b2 into master Dec 15, 2016

marchof deleted the issue-470 branch December 15, 2016 20:22

jacoco locked and limited conversation to collaborators Jan 11, 2017

marchof moved this from TODO to DONE in Java 9 May 7, 2017

marchof moved this from DONE to TODO in Java 9 May 7, 2017

Godin moved this from TODO to DONE in Java 9 Jul 8, 2017

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Fix build failure with JDK 9 EA b148 #470

Fix build failure with JDK 9 EA b148 #470

Godin commented Dec 9, 2016 •

edited

marchof Dec 15, 2016

Godin Dec 15, 2016

marchof Dec 15, 2016

Godin Dec 15, 2016

Godin Dec 15, 2016

marchof Dec 15, 2016

Godin Dec 15, 2016

Godin Dec 15, 2016

marchof left a comment

Godin commented Dec 15, 2016

bjkail Dec 15, 2016

Godin Dec 15, 2016

marchof commented Dec 15, 2016

Fix build failure with JDK 9 EA b148 #470

Fix build failure with JDK 9 EA b148 #470

Conversation

Godin commented Dec 9, 2016 • edited

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

marchof left a comment

Choose a reason for hiding this comment

Godin commented Dec 15, 2016

Choose a reason for hiding this comment

Choose a reason for hiding this comment

marchof commented Dec 15, 2016

Godin commented Dec 9, 2016 •

edited