6856630: Restructure jaxp/jaxws repository

Andrew John Hughes gnu_andrew at member.fsf.org
Wed Oct 28 23:03:10 UTC 2009 

2009/10/28 Jonathan Gibbons <Jonathan.Gibbons at sun.com>:
> Andrew John Hughes wrote:
>
> 2009/10/28 Andrew John Hughes <gnu_andrew at member.fsf.org>:
>
>
> 2009/10/28 Kelly O'Hair <Kelly.Ohair at sun.com>:
>
>
> Andrew John Hughes wrote:
>
>
> 2009/10/28 Jonathan Gibbons <Jonathan.Gibbons at sun.com>:
>
>
> Andrew John Hughes wrote:
>
> 2009/10/28 Andrew John Hughes <gnu_andrew at member.fsf.org>:
>
>
> 2009/10/28 Kelly O'Hair <Kelly.Ohair at sun.com>:
>
>
> Joseph D. Darcy wrote:
>
>
> Andrew John Hughes wrote:
>
>
> 2009/10/23 Kelly O'Hair <Kelly.Ohair at sun.com>:
>
>
>
> Jonathan Gibbons wrote:
>
>
>
> Kelly O'Hair wrote:
>
>
>
> Jonathan Gibbons wrote:
>
>
>
> Kelly O'Hair wrote:
>
>
>
> [big snip]
>
>
> DOH!  Sorry...
>
> Yes, these jaxp and jaxws forests can probably go away, we won't
> be using them.
>
> The current plan is that jaxp/jaxws changes (new bundles) will go
> through the TL forest.
>
>
> -kto
>
>
>
>
> I'm guilty of also thinking that Jonathan was referring to the jaxws
> and jaxp repositories per forest, rather than the specific forests.
> On that note, i18n could probably die too because apparently that team
> always use the swing forest for commits.
>
> It would be nice to one day get rid of the jaxp and jaxws trees too.
> I don't actually see why they were created as trees to begin with,
> given they've always been upstream and not a source of many commits.
> The one to actual split up would be jdk, as I can feel Mercurial
> struggling with it a bit already on jdk7.  But I don't know how
> feasible that is, if at all.  Maybe Jigsaw will help there.
>
> One thing that does worry me -- what happens when the jaxws or jaxp
> code needs security updates?
>
>
>
> Yes, the need to support security fixes was considered as part of this
> new
> delivery model.  Ultimately a revised source bundle with the security
> fixes
> will need to be produced.  Until then, the fixes can be represented as
> patches which are applied to the sources before the build.  Kelly can
> speak
> to the implementation details of the patch mechanism.
>
>
> It's crude, but should serve in an emergency. See the patches/README.
> After a bundle is unzipped, all patches are applied, none right now.
>
> I hope that we can always just get updated bundles.
>
> Originally, the jaxp and jaxws (and corba) workspaces were created
> mainly as a place to move files from (trim) the original "j2se"
> workspace,
> and we had no idea where we were going with it all, except that we
> knew these were out of place in the j2se workspace, which became
> the jdk repository.
>
> The jaxp and jaxws repos could just go away someday, but I'll leave
> that for another day. ;^)
>
> Let's just call it evolution. ;^)
>
>
>
> Oh yes, I'm not saying right now -- something for JDK8 perhaps? :)
> Certainly, the trivial change Jonathan mentions could be done when
> creating the jdk8 forests.
> It would be nice to share the common stuff between jaxp and jaxws too,
> and I suspect that may be easier if they are in the same toplevel
> openjdk repository.
>
>
>
> -kto
>
>
>
>
> -Joe
>
>
> --
> Andrew :-)
>
> Free Java Software Engineer
> Red Hat, Inc. (http://www.redhat.com)
>
> Support Free Java!
> Contribute to GNU Classpath and the OpenJDK
> http://www.gnu.org/software/classpath
> http://openjdk.java.net
>
> PGP Key: 94EFD9D8 (http://subkeys.pgp.net)
> Fingerprint: F8EF F1EA 401E 2E60 15FA  7927 142C 2591 94EF D9D8
>
>
>
> I was just testing a webrev for the ALT_DROPS_DIR change on tl and hit
> this:
>
> -jaxp_src-url-bundle:
>     [echo] Downloading from
> https://jaxp.dev.java.net/files/documents/913/144160/jdk7-jaxp-m5.zip
>    [mkdir] Created dir: /mnt/builder/tl/jaxp/drop/bundles
>      [get] Getting:
> https://jaxp.dev.java.net/files/documents/913/144160/jdk7-jaxp-m5.zip
>      [get] To: /mnt/builder/tl/jaxp/drop/bundles/jdk7-jaxp-m5.zip.temp
>      [get] Error getting
> https://jaxp.dev.java.net/files/documents/913/144160/jdk7-jaxp-m5.zip
> to /mnt/builder/tl/jaxp/drop/bundles/jdk7-jaxp-m5.zip.temp
>
> BUILD FAILED
> javax.net.ssl.SSLException: java.lang.RuntimeException: Unexpected
> error: java.security.InvalidAlgorithmParameterException: the
> trustAnchors parameter must be non-empty
>        at sun.security.ssl.Alerts.getSSLException(Alerts.java:208)
>        at sun.security.ssl.SSLSocketImpl.fatal(SSLSocketImpl.java:1627)
>        at sun.security.ssl.SSLSocketImpl.fatal(SSLSocketImpl.java:1590)
>        at
> sun.security.ssl.SSLSocketImpl.handleException(SSLSocketImpl.java:1573)
>        at
> sun.security.ssl.SSLSocketImpl.startHandshake(SSLSocketImpl.java:1166)
>        at
> sun.security.ssl.SSLSocketImpl.startHandshake(SSLSocketImpl.java:1143)
>        at
> sun.net.www.protocol.https.HttpsClient.afterConnect(HttpsClient.java:423)
>        at
>
> sun.net.www.protocol.https.AbstractDelegateHttpsURLConnection.connect(AbstractDelegateHttpsURLConnection.java:185)
>        at
>
> sun.net.www.protocol.https.HttpsURLConnectionImpl.connect(HttpsURLConnectionImpl.java:153)
>        at org.apache.tools.ant.taskdefs.Get.doGet(Get.java:145)
>        at org.apache.tools.ant.taskdefs.Get.execute(Get.java:78)
>        at
> org.apache.tools.ant.UnknownElement.execute(UnknownElement.java:288)
>        at sun.reflect.GeneratedMethodAccessor1.invoke(Unknown Source)
>        at
>
> sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
>        at java.lang.reflect.Method.invoke(Method.java:616)
>        at
>
> org.apache.tools.ant.dispatch.DispatchUtils.execute(DispatchUtils.java:106)
>        at org.apache.tools.ant.Task.perform(Task.java:348)
>        at org.apache.tools.ant.Target.execute(Target.java:357)
>        at org.apache.tools.ant.Target.performTasks(Target.java:385)
>        at
> org.apache.tools.ant.Project.executeSortedTargets(Project.java:1337)
>        at org.apache.tools.ant.Project.executeTarget(Project.java:1306)
>        at
>
> org.apache.tools.ant.helper.DefaultExecutor.executeTargets(DefaultExecutor.java:41)
>        at org.apache.tools.ant.Project.executeTargets(Project.java:1189)
>        at org.apache.tools.ant.Main.runBuild(Main.java:758)
>        at org.apache.tools.ant.Main.startAnt(Main.java:217)
>        at org.apache.tools.ant.launch.Launcher.run(Launcher.java:257)
>        at org.apache.tools.ant.launch.Launcher.main(Launcher.java:104)
>
> Firstly, I obviously have to wonder why it's still downloading, but
> this seems to be caused by the new URL.
>
>
> I'm curious: is there an option to prevent downloading the source bundle?
> It seems to me that if you think you've set up the build to not need to
> download anything, it would be nice if the build failed if the bits were
> not
> available, rather than have it try and find the bits.  Automatic
> downloads
> sound bad, IMO.
>
> -- Jon
>
>
>
> It would be a nice additional option, I don't think it already exists.
>  At present, it happens accidentally in that the JDK crashes
> instead... :(
>
> This is interesting:
>
> /mnt/builder/tl/jaxws/build/xml_generated/build-drop-jaxws_src.xml:117:
> Checksum on file
> /mnt/builder/tl/jaxws/drop/bundles/jdk7-jaxws-2009_09_28.zip is
>              debb949440c5a15ce999cfefbbc56526, not
> f5010ebf636db9f465a61a7a74944543
>
> Did the jaxws bundle change without being renamed?
>
>
> I certainly hope not.
>
> These are the md5 sums I have on all the bundles:
>
> bonsai<11> /usr/bin/sum -x md5  /java/devtools/share/jdk*drops/*.zip
> 7a50bb540a27cdd0001885630088b758
> /java/devtools/share/jdk6-drops/jdk6-jaf-2009_10_27.zip
> 0bb03bbd7b1b6d87cc65772c6adb2d6a
> /java/devtools/share/jdk6-drops/jdk6-jaxp-2009_10_27.zip
> 3ea5728706169498b722b898a1008acf
> /java/devtools/share/jdk6-drops/jdk6-jaxws-2009_10_27.zip
> eb8cb7a4a7f14e211fbe2354878a2472
> /java/devtools/share/jdk7-drops/jdk7-jaf-2009_08_28.zip
> d99f4777bc4c42d7759f7c0fcf87ef5d
> /java/devtools/share/jdk7-drops/jdk7-jaxp-2009_08_28.zip
> 8800970d03bab1fff188dcfafc346f5d
> /java/devtools/share/jdk7-drops/jdk7-jaxp-2009_09_28.zip
> 8b58ce7919cda8e32a9afc5cb4b58bb1
> /java/devtools/share/jdk7-drops/jdk7-jaxp-m5.zip
> debb949440c5a15ce999cfefbbc56526
> /java/devtools/share/jdk7-drops/jdk7-jaxws-2009_08_28.zip
> f5010ebf636db9f465a61a7a74944543
> /java/devtools/share/jdk7-drops/jdk7-jaxws-2009_09_28.zip
>
> -kto
>
>
>
> Ok it appears I have 08_28 masquerading as 09_28.  Thanks.
> --
> Andrew :-)
>
> Free Java Software Engineer
> Red Hat, Inc. (http://www.redhat.com)
>
> Support Free Java!
> Contribute to GNU Classpath and the OpenJDK
> http://www.gnu.org/software/classpath
> http://openjdk.java.net
>
> PGP Key: 94EFD9D8 (http://subkeys.pgp.net)
> Fingerprint: F8EF F1EA 401E 2E60 15FA  7927 142C 2591 94EF D9D8
>
>
>
> Ok here's the webrev for the ALT_DROPS_DIR change against tl:
>
> http://cr.openjdk.java.net/~andrew/drops/webrev.02/
>
> Ok to push?  Do we have a bug ID for this?
>
> Thanks,
>
>
> Andrew,
>
> Some of the comments in the Makefiles need to be updated -- for example,
> references to devtools in lines 89, 100
>
> -- Jon
>

Well spotted; revised version here:

http://cr.openjdk.java.net/~andrew/drops/webrev.03/
-- 
Andrew :-)

Free Java Software Engineer
Red Hat, Inc. (http://www.redhat.com)

Support Free Java!
Contribute to GNU Classpath and the OpenJDK
http://www.gnu.org/software/classpath
http://openjdk.java.net

PGP Key: 94EFD9D8 (http://subkeys.pgp.net)
Fingerprint: F8EF F1EA 401E 2E60 15FA  7927 142C 2591 94EF D9D8

More information about the build-dev mailing list