RFR: 8344925: translet-name ignored, when package-name is also set via TransformerFactory.setAttribute

[Severin Gehwolf]

On Thu, 28 Nov 2024 01:09:47 GMT, Zheng Feng <duke at openjdk.org> wrote:

> …String)
> 
> This bug seems from the improper handling of package and class name initialization in `XSLTC` since there is a default value of  `_packageName` with `die.verwandlung`. This fix is just adjusting to call `setPackageName` before `setClassName`.
> 
> I've done the tire1 and tire2 tests. The tire1 tests passed and tire2 reports
> 
> 
> ==============================
> Test summary
> ==============================
>    TEST                                              TOTAL  PASS  FAIL ERROR  
>>> jtreg:test/hotspot/jtreg:tier2                      745   744     1     0 <<
>>> jtreg:test/jdk:tier2                               4142  4137     2     3 <<
>    jtreg:test/langtools:tier2                           12    12     0     0  
>    jtreg:test/jaxp:tier2                               514   514     0     0  
>    jtreg:test/docs:tier2                                 1     1     0     0  
> ==============================
> TEST FAILURE
> 
> 
> The failed tests are
> 
> JT Harness : Tests that failed
> runtime/Thread/ThreadCountLimit.java#id0: Stress test that reaches the process limit for thread count, or time limit.
> 
> JT Harness : Tests that failed
> java/net/InetAddress/CheckJNI.java: java -Xcheck:jni failing in net code on Solaris / [Datagram]Socket.getLocalAddress() failure
> java/net/Socket/LinkLocal.java: Connecting to a link-local IPv6 address should not causes a SocketException to be thrown.
> 
> JT Harness : Tests that had errors
> java/net/URL/EarlyOrDelayedParsing.java: URL built-in protocol handlers should parse the URL early to avoid constructing URLs for which openConnection would later throw an exception, when possible.
> java/net/ipv6tests/UdpTest.java: IPv6 support for Windows XP and 2003 server.
> java/nio/channels/DatagramChannel/SendReceiveMaxSize.java: Check that it is possible to send and receive datagrams of maximum size on macOS.

A change similar to the proposal here has been integrated with https://github.com/openjdk/jdk/pull/23446 I'd suggest to close this one. It looks like there was some miscommunication happening. In this case the person proposing the PR isn't OpenJDK author, so couldn't assign the bug to himself. We'll work with the team to catch cases like this in the future.

-------------

PR Comment: https://git.openjdk.org/jdk/pull/22425#issuecomment-2642453124