May 2015 - core-libs-dev

RFR [9] 8080502: Changing accessing module resources
by Miroslav Kos 29 May '15

29 May '15

Hi everybody, this is a review request for 8080502: Changing accessing module resources. This change adapts to Jigsaw JDK runtime - it isn't possible to access module resource using Class::getResource() any more. JBS: https://bugs.openjdk.java.net/browse/JDK-8080502 webrev: http://cr.openjdk.java.net/~mkos/8080502/jaxws.01/index.html testing: JAX-WS unit test; basically it is fix for existing tests (JCK, ...) Thanks Miran

2 3

RFR 9: JDK-8075678 : java.time javadoc error in DateTimeFormatter::parsedLeapSecond
by Roger Riggs 28 May '15

28 May '15

Please review 3 editorial issues in the java.time package: Issues: 8075678 <https://bugs.openjdk.java.net/browse/JDK-8075678>: java.time javadoc error in DateTimeFormatter::parsedLeapSecond 8075676 <https://bugs.openjdk.java.net/browse/JDK-8075676>: java.time package javadoc typos 8068276 <https://bugs.openjdk.java.net/browse/JDK-8068276>: java.time.chrono.HijrahChronology.eraOf() assertions may lead to misunderstanding Webrev: http://cr.openjdk.java.net/~rriggs/webrev-typo-8075678/ Thanks, Roger

3 2

JDK 9 RFR of JDK-8081359: Update bug reporting URL
by joe darcy 28 May '15

28 May '15

Hello, As the venerable URL http://bugreport.sun.com/bugreport/ has been retired and replaced with http://bugreport.java.com/bugreport/ Occurrences of the old URL in the source base should be updated. Please review the patch below which does this: --- a/src/java.base/share/native/libjava/System.c Thu May 28 09:11:14 2015 +0800 +++ b/src/java.base/share/native/libjava/System.c Wed May 27 22:36:07 2015 -0700 @@ -111,7 +111,7 @@ #ifndef VENDOR /* Third party may overwrite this. */ #define VENDOR "Oracle Corporation" #define VENDOR_URL "http://java.oracle.com/" -#define VENDOR_URL_BUG "http://bugreport.sun.com/bugreport/" +#define VENDOR_URL_BUG "http://bugreport.java.com/bugreport/" #endif #define JAVA_MAX_SUPPORTED_VERSION 52 diff -r de8fc052f1b6 src/jdk.rmic/share/classes/sun/tools/javac/resources/javac.properties --- a/src/jdk.rmic/share/classes/sun/tools/javac/resources/javac.properties Thu May 28 09:11:14 2015 +0800 +++ b/src/jdk.rmic/share/classes/sun/tools/javac/resources/javac.properties Wed May 27 22:36:07 2015 -0700 @@ -498,9 +498,9 @@ javac.err.cant.write=\ Can''t write: {0} javac.err.fatal.error=\ - An error has occurred in the compiler; please file a bug report (http://bugreport.sun.com/bugreport/) + An error has occurred in the compiler; please file a bug report (http://bugreport.java.com/bugreport/) javac.err.fatal.exception=\ - An exception has occurred in the compiler; please file a bug report (http://bugreport.sun.com/bugreport/) + An exception has occurred in the compiler; please file a bug report (http://bugreport.java.com/bugreport/) javac.err.no.inner.classes=\ Support for inner classes has been disabled. javac.err.uncaught.exception=\ These were the only remaining occurrences of the old URL in the OpenJDK code base. The occurrence in langtools was previously updated (JDK-8078560). Thanks, -Joe

2 1

RFR (XS: JDK-8080991 Compilation error with recent clang ....
by Kumar Srinivasan 28 May '15

28 May '15

Hi, I have already reviewed it, contributed by Staffan Larsen. This is an fyi, I will push this changeset if no one has any objections. https://bugs.openjdk.java.net/browse/JDK-8080991 http://cr.openjdk.java.net/~ksrini/8080991/webrev/ Thanks Kumar

1 0

RFR: 8081392: getNodeValue should return NULL value for Element nodes
by Aleksej Efimov 28 May '15

28 May '15

Hi, Please, review JDK9 fix [1] for a bug [2] in 'getNodeValue' JAXP function: According to w3c [3] getNodeValue should return 'null' value for Element nodes. This behavior was changed by JDK-8032908 [4] fix - in current implementation the node content is returned. To address this issue the JDK-8032908 fix was reverted and 'getTextContent' function was modified properly to address issue reported by JDK-8032908. JDK-8032908 test was modified to test that 'getNodeValue' returns 'null' for Element nodes. DocumentExtFunc test was modified to use 'getTextContent' instead of 'getNodeValue' for node content retrival. No related test failures on local and JPRT builds were observed. With Best Regards, Aleksej [1] http://cr.openjdk.java.net/~aefimov/8081392/9/00 [2] https://bugs.openjdk.java.net/browse/JDK-8081392 [3] http://www.w3.org/TR/2000/REC-DOM-Level-2-Core-20001113/core.html#ID-195064… [4] https://bugs.openjdk.java.net/browse/JDK-8032908

2 1

RFR JDK-8038310: Re-examine integration of extended Charsets
by Xueming Shen 28 May '15

28 May '15

Hi, Please help review the change of changing the hard-wired extended charsets loading mechanism to ServiceLoader.loadInstalled(), for the module system. With the fix for JDK-8069588 in JDK 9 all charsets needed to startup in supported environments are in the base module. This change is to use ServiceLoader to load the extended charsets to remove the java.base dependency on jdk.charsets moduel. Issues: JDK-8038310: Re-examine integration of extended Charsets JDK-8069588: Need to avoid attempting to load extended charsets (from jdk.charsets module) before module system initialization runs webrev: http://cr.openjdk.java.net/~sherman/8069588_8038310/webrev Thanks -Sherman

3 5

RFR: 8077242: Add default method CharSequence.getChars(int srcBegin, int srcEnd, char[] dst, int dstBegin)
by Ivan Gerasimov 28 May '15

28 May '15

Hi everyone! The String class has getChars(int srcBegin, int srcEnd, char[] dst, int dstBegin) which is used to efficiently extract substrings, avoiding unnecessary copying. The AbstractStringBuilder has the method with the same signature. I propose to add a default implementation of this method to the CharSequence interface. It can used, for example, in AbstractStringBuilder.append(CharSequence s, int start, int end) instead of the for-loop. This, in turn, allows to simplify AbstractStringBuilder.append(CharSequence s). In the webrev, I also included a few optimizations of the kind: stringBuilder.append(string.substring(from, to)) => stringBuilder.append(string, from, to). The later variant becomes more effective after the proposed fix. BUGURL: https://bugs.openjdk.java.net/browse/JDK-8077242 WEBREV: http://cr.openjdk.java.net/~igerasim/8077242/00/webrev/ If people agree that this change is useful, I'll file a CCC request. Sincerely yours, Ivan

5 7

Review Request for 8081347: Add @modules to jdk_core tests
by Mandy Chung 28 May '15

28 May '15

jtreg supports a new @modules to specify the module and the internal API that a test depends on. This is to prepare when the module boundaries are enforced at runtime. This initial patch is contributed by several of us including Alexander Kulyakhtin and Alan. Webrev at: http://cr.openjdk.java.net/~mchung/jdk9/webrevs/8081347/webrev.00/ Thanks Mandy

4 4

RFR 9: 8077350 : JEP 102 Process API Updates Implementation
by Roger Riggs 28 May '15

28 May '15

Thanks to everyone who has reviewed the Process API updates and contributed comments and improvements. Unless there are any showstoppers, I'd like to commit the work completed so far and treat any issues that come up as incremental changes. Please comment on these changes or if I missed giving credit to any reviewers. Webrev: http://cr.openjdk.java.net/~rriggs/webrev-process-8077350/ javadoc: http://cr.openjdk.java.net/~rriggs/ph-apidraft/ Thanks, Roger

1 0

JDK 9 RFR of JDK-8081245: MHIllegalAccess.java failing across platforms
by joe darcy 28 May '15

28 May '15

Hello, The latest HotSpot integration into JDK 9 dev introduced a test failure in java/lang/invoke/8022701/MHIllegalAccess.java.MHIllegalAccess.java Please review the change below to problem list the test until such time as the fix for the test (JDK-8080428) propagates to dev: --- a/test/ProblemList.txt Thu May 28 09:11:14 2015 +0800 +++ b/test/ProblemList.txt Wed May 27 22:43:49 2015 -0700 @@ -123,6 +123,9 @@ # 8029891 java/lang/ClassLoader/deadlock/GetResource.java generic-all +# 8080428 +java/lang/invoke/8022701/MHIllegalAccess.java.MHIllegalAccess.java generic-all + ############################################################################ # jdk_instrument Thanks, -Joe

3 2