tag:old.nabble.com,2006:forum-1524Nabble - Gnu - Classpath2009-12-14T03:03:38ZGNU Classpath, Essential Libraries for Java, is a GNU project to create free core class libraries for use with virtual machines and compilers for the java programming language. Gnu - Classpath home is <a href="http://www.gnu.org/software/classpath/classpath.html" target="_top" rel="nofollow">here</a>.tag:old.nabble.com,2006:post-26776045New mpc on builder (Was: [cp-testresults] FAIL: gcc build on Mon Dec 14 10:06:45 UTC 2009)2009-12-14T03:03:38Z2009-12-14T03:03:38ZMark WielaardOn Mon, 2009-12-14 at 10:06 +0000, <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26776045&i=0" target="_top" rel="nofollow">cpdev@...</a> wrote:
<div class='shrinkable-quote'><br>> checking for the correct version of mpc.h... no
<br>> configure: error: Building GCC requires GMP 4.2+, MPFR 2.3.2+ and MPC 0.8.0+.
<br>> Try the --with-gmp, --with-mpfr and/or --with-mpc options to specify
<br>> their locations. Source code for these libraries can be found at
<br>> their respective hosting sites as well as at
<br>> ftp://gcc.gnu.org/pub/gcc/infrastructure/. See also
<br>> <a href="http://gcc.gnu.org/install/prerequisites.html" target="_top" rel="nofollow">http://gcc.gnu.org/install/prerequisites.html</a> for additional info. If
<br>> you obtained GMP, MPFR and/or MPC from a vendor distribution package,
<br>> make sure that you have installed both the libraries and the header
<br>> files. They may be located in separate packages.
</div><br>Which wasn't available for Debian etch (4.0 - yes, I know, builder
<br>should really be upgraded...). Installed now by hand in /usr/local.
<br>Sorry for the inconvenience.
<br><br>Cheers,
<br><br>Mark
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Classpath---General-f1525.html" embed="fixTarget[1525]" target="_top" >Gnu - Classpath - General</a></p>tag:old.nabble.com,2006:post-26757461[PATCH] StrictMath: Correct the description of atan()2009-12-10T18:11:22Z2009-12-10T18:11:22ZMaciej W. RozyckiHi,
<br><br>On Fri, 11 Dec 2009, Maciej W. Rozycki wrote:
<br><br>> Noticed in and taken from GCC sources, but present in the current head of
<br>> the classpath tree (0.99-pre) too. This should be obvious -- the
<br>> description of the atan() method incorrectly refers to "arcsin" rather
<br>> than "arctan". Fixed thus.
<br><br> There's more to fix; please use the following patch instead.
<br><br>2009-12-11 Maciej W. Rozycki <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26757461&i=0" target="_top" rel="nofollow">macro@...</a>>
<br><br> * java/lang/StrictMath.java (atan): Correct description.
<br><br> Please apply,
<br><br> Maciej
<br><br>classpath-0.98-20091211-atan.patch
<br>Index: java/lang/StrictMath.java
<br>===================================================================
<br>--- java/lang/StrictMath.java (revision 155148)
<br>+++ java/lang/StrictMath.java (working copy)
<br>@@ -478,12 +478,12 @@ public final strictfp class StrictMath
<br> }
<br>
<br> /**
<br>- * The trigonometric function <em>arcsin</em>. The range of angles returned
<br>+ * The trigonometric function <em>arctan</em>. The range of angles returned
<br> * is -pi/2 to pi/2 radians (-90 to 90 degrees). If the argument is NaN, the
<br> * result is NaN; and the arctangent of 0 retains its sign.
<br> *
<br> * @param x the tan to turn back into an angle
<br>- * @return arcsin(x)
<br>+ * @return arctan(x)
<br> * @see #atan2(double, double)
<br> */
<br> public static double atan(double x)
<br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Classpath---Patches-f1527.html" embed="fixTarget[1527]" target="_top" >Gnu - Classpath - Patches</a></p>tag:old.nabble.com,2006:post-26757460[PATCH] StrictMath: Correct the description of atan()2009-12-10T18:03:28Z2009-12-10T18:03:28ZMaciej W. RozyckiHello,
<br><br> Noticed in and taken from GCC sources, but present in the current head of
<br>the classpath tree (0.99-pre) too. This should be obvious -- the
<br>description of the atan() method incorrectly refers to "arcsin" rather
<br>than "arctan". Fixed thus.
<br><br>2009-12-11 Maciej W. Rozycki <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26757460&i=0" target="_top" rel="nofollow">macro@...</a>>
<br><br> * java/lang/StrictMath.java (atan): Correct description.
<br><br> Please apply,
<br><br> Maciej
<br><br>classpath-0.98-20091211-atan.patch
<br>Index: java/lang/StrictMath.java
<br>===================================================================
<br>--- java/lang/StrictMath.java (revision 155148)
<br>+++ java/lang/StrictMath.java (working copy)
<br>@@ -478,7 +478,7 @@ public final strictfp class StrictMath
<br> }
<br>
<br> /**
<br>- * The trigonometric function <em>arcsin</em>. The range of angles returned
<br>+ * The trigonometric function <em>arctan</em>. The range of angles returned
<br> * is -pi/2 to pi/2 radians (-90 to 90 degrees). If the argument is NaN, the
<br> * result is NaN; and the arctangent of 0 retains its sign.
<br> *
<br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Classpath---Patches-f1527.html" embed="fixTarget[1527]" target="_top" >Gnu - Classpath - Patches</a></p>tag:old.nabble.com,2006:post-26706313Re: Re: classpathx-javamail Digest, Vol 44, Issue 32009-12-08T23:45:03Z2009-12-08T23:45:03ZChris Burdess-2Dhaval Yoganandi wrote:
<br>> Thanks for the update. May I know when the next stable version would be available?
<br><br><br>Sorry but I don't have a schedule for that.
<br>--
<br>Chris Burdess
<br><br><br><br><br><p>From forum: <a href="http://old.nabble.com/classpathx-javamail-f16003.html" embed="fixTarget[16003]" target="_top" >classpathx-javamail</a></p>tag:old.nabble.com,2006:post-26705447Re: classpathx-javamail Digest, Vol 44, Issue 32009-12-08T21:38:02Z2009-12-08T21:38:02ZDhaval Yoganandi<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26705447&i=0" target="_top" rel="nofollow">classpathx-javamail-request@...</a> wrote:
<div class='shrinkable-quote'><br>> Send classpathx-javamail mailing list submissions to
<br>> <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26705447&i=1" target="_top" rel="nofollow">classpathx-javamail@...</a>
<br>>
<br>> To subscribe or unsubscribe via the World Wide Web, visit
<br>> <a href="http://lists.gnu.org/mailman/listinfo/classpathx-javamail" target="_top" rel="nofollow">http://lists.gnu.org/mailman/listinfo/classpathx-javamail</a><br>> or, via email, send a message with subject or body 'help' to
<br>> <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26705447&i=2" target="_top" rel="nofollow">classpathx-javamail-request@...</a>
<br>>
<br>> You can reach the person managing the list at
<br>> <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26705447&i=3" target="_top" rel="nofollow">classpathx-javamail-owner@...</a>
<br>>
<br>> When replying, please edit your Subject line so it is more specific
<br>> than "Re: Contents of classpathx-javamail digest..."
<br>>
<br>>
<br>> Today's Topics:
<br>>
<br>> 1. Re: Exception while sending mail through GNU JavaMail API.
<br>> (Chris Burdess)
<br>>
<br>>
<br>> ----------------------------------------------------------------------
<br>>
<br>> Message: 1
<br>> Date: Tue, 8 Dec 2009 11:29:30 +0000
<br>> From: Chris Burdess <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26705447&i=4" target="_top" rel="nofollow">dog@...</a>>
<br>> Subject: Re: [Classpathx-javamail] Exception while sending mail
<br>> through GNU JavaMail API.
<br>> To: <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26705447&i=5" target="_top" rel="nofollow">classpathx-javamail@...</a>
<br>> Message-ID: <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26705447&i=6" target="_top" rel="nofollow">722BA005-61FE-4BF0-ACB5-17CFEA34E8E2@...</a>>
<br>> Content-Type: text/plain; charset=us-ascii
<br>>
<br>> Dhaval Yoganandi wrote:
<br>>
<br>>> | InputStream source = *new* FileInputStream(file);|
<br>>>
<br>>> | MimeMessage mimeMessage = *new* MimeMessage(mailSession,source);|
<br>>>
<br>>> | mimeMessage.setHeader("Content-Type", "multipart/mixed");|
<br>>>
<br>>> | Transport./send/(mimeMessage);|
<br>>>
<br>>>
<br>>> and I am getting following exception while sending the mail. The code is working fine with sun JVM and the javamail (not GNU javamail) library.
<br>>>
<br>>
<br>> This bug is now fixed in CVS HEAD.
<br>>
</div>Hello Chris,
<br><br>Thanks for the update. May I know when the next stable version would be
<br>available?
<br><br>--
<br>Thanks and Regards,
<br><br>Dhaval Yoganandi,
<br>Jr. Software Engineer,
<br>Elitecore Technologies Ltd.
<br>Cyberoam.
<br><p>From forum: <a href="http://old.nabble.com/classpathx-javamail-f16003.html" embed="fixTarget[16003]" target="_top" >classpathx-javamail</a></p>tag:old.nabble.com,2006:post-26692493Re: Exception while sending mail through GNU JavaMail API.2009-12-08T03:29:30Z2009-12-08T03:29:30ZChris Burdess-2Dhaval Yoganandi wrote:
<div class='shrinkable-quote'><br>> | InputStream source = *new* FileInputStream(file);|
<br>>
<br>> | MimeMessage mimeMessage = *new* MimeMessage(mailSession,source);|
<br>>
<br>> | mimeMessage.setHeader("Content-Type", "multipart/mixed");|
<br>>
<br>> | Transport./send/(mimeMessage);|
<br>>
<br>>
<br>> and I am getting following exception while sending the mail. The code is working fine with sun JVM and the javamail (not GNU javamail) library.
</div><br>This bug is now fixed in CVS HEAD.
<br>--
<br>Chris Burdess
<br><br><br><br><p>From forum: <a href="http://old.nabble.com/classpathx-javamail-f16003.html" embed="fixTarget[16003]" target="_top" >classpathx-javamail</a></p>tag:old.nabble.com,2006:post-26674341Exception while sending mail through GNU JavaMail API.2009-12-07T00:54:09Z2009-12-07T00:54:09ZDhaval YoganandiHello gnu javamail team,
<br><br>I've one problem while sending mail using gnu javamail library. Here is
<br>what I'm using.
<br><br>I'm using JamVm (a third party java virtual machine) with GNU Classpath
<br>0.98. And I'm also using GNU javamail 1.1.2 library compiled with
<br>dependent GNU libraries. Now, I've written a code to send the stored EML
<br>files on the disk to the recipient. See the following code.
<br><br> |Authenticator authenticator=||*null*;|
<br><br>| MailcapCommandMap mc =
<br>(MailcapCommandMap)CommandMap./getDefaultCommandMap/();|
<br><br>| CommandMap./setDefaultCommandMap/(mc); |
<br><br>| |
<br><br>| Properties mailProperties = *new* Properties(); |
<br><br>| mailProperties.put(MailConstants./MAILSMTPHOST/,
<br>mailServerHost);|
<br><br>| mailProperties.put(MailConstants./MAILSMTPPORT/,
<br>mailServerPort);|
<br><br>| mailProperties.put(MailConstants./MAILDEBUG/,
<br>MailConstants./VALUETRUE/);|
<br><br>|
<br>mailProperties.put(MailConstants./MAILTRANSPORTPROTOCOL/,
<br>MailConstants./TRANSPORTPROTOCOLSMTP/); |
<br><br>| mailProperties.put(MailConstants./MAILSMTPAUTH/,
<br>"false");|
<br><br>| mailProperties.put(MailConstants./MAILDEBUG/,
<br>MailConstants./VALUETRUE/);|
<br><br>| Session mailSession =
<br>Session./getInstance/(mailProperties, authenticator);|
<br><br>| InputStream source = *new* FileInputStream(file);|
<br><br>| MimeMessage mimeMessage = *new*
<br>MimeMessage(mailSession,source);|
<br><br>| mimeMessage.setHeader("Content-Type",
<br>"multipart/mixed");|
<br><br>| Address[] fromAddrsss = *new* Address[1];|
<br><br>| Address[] toAddress = *new* Address[1];|
<br><br>| fromAddrsss[0] = *new* InternetAddress(mailFrom);|
<br><br>| toAddress[0] = *new*
<br>InternetAddress(mailTo); |
<br><br>| mimeMessage.addFrom(fromAddrsss);|
<br><br>| mimeMessage.setRecipients(Message.RecipientType./TO/,
<br>toAddress);|
<br><br>| mimeMessage.setSubject(strSubject); |
<br><br>| Transport./send/(mimeMessage);|
<br><br><br>and I am getting following exception while sending the mail. The code is
<br>working fine with sun JVM and the javamail (not GNU javamail) library.
<br><br>Exception is
<br><br>Exception while sending Mail java.lang.ClassCastException
<br><<a href="http://java.sun.com/javase/6/docs/api/java/lang/ClassCastException.html" target="_top" rel="nofollow">http://java.sun.com/javase/6/docs/api/java/lang/ClassCastException.html</a>>:
<br>java/io/ByteArrayInputStreamjava.lang.ClassCastException:
<br>java/io/ByteArrayInputStream
<br><br>at javax.mail.internet.MimeMessage.updateHeaders(MimeMessage.java:1866)
<br><br>at javax.mail.internet.MimeMessage.saveChanges(MimeMessage.java:1843)
<br><br>at javax.mail.Transport.send(Transport.java:74)
<br><br>at my.utilities.SendUtility.sendMail(SendUtility.java:199)
<br><br>at my.utilities.SendUtility.handleReleaseRequest(SendUtility.java:70)
<br><br>at my.servlets.MyServlet.doPost(MyServlet..java:129)
<br><br>at javax.servlet.http.HttpServlet.service(HttpServlet.java:638)
<br><br>at javax.servlet.http.HttpServlet.service(HttpServlet.java:720)
<br><br>at
<br>org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:199)
<br><br><br>at
<br>org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:145)
<br><br><br>at
<br>org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.java:210)
<br><br><br>at
<br>org.apache.catalina.core.StandardPipeline$StandardPipelineValveContext.invokeNext(StandardPipeline.java:596)
<br><br><br>at
<br>org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.java:433)
<br><br>at org.apache.catalina.core.ContainerBase.invoke(ContainerBase.java:955)
<br><br>at
<br>org.apache.catalina.core.StandardContextValve.invoke(StandardContextValve.java:139)
<br><br><br>at
<br>org.apache.catalina.core.StandardPipeline$StandardPipelineValveContext.invokeNext(StandardPipeline.java:596)
<br><br><br>at
<br>org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.java:433)
<br><br>at org.apache.catalina.core.ContainerBase.invoke(ContainerBase.java:955)
<br><br>at
<br>org.apache.catalina.core.StandardContext.invoke(StandardContext.java:2460)
<br><br>at
<br>org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:133)
<br><br><br>at
<br>org.apache.catalina.core.StandardPipeline$StandardPipelineValveContext.invokeNext(StandardPipeline.java:596)
<br><br><br>at
<br>org.apache.catalina.valves.ErrorDispatcherValve.invoke(ErrorDispatcherValve.java:119)
<br><br><br>at
<br>org.apache.catalina.core.StandardPipeline$StandardPipelineValveContext.invokeNext(StandardPipeline.java:594)
<br><br><br>at
<br>org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:117)
<br><br><br>at
<br>org.apache.catalina.core.StandardPipeline$StandardPipelineValveContext.invokeNext(StandardPipeline.java:594)
<br><br><br>at
<br>org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.java:433)
<br><br>at org.apache.catalina.core.ContainerBase.invoke(ContainerBase.java:955)
<br><br>at
<br>org.apache.catalina.core.StandardEngineValve.invoke(StandardEngineValve.java:127)
<br><br><br>at
<br>org.apache.catalina.core.StandardPipeline$StandardPipelineValveContext.invokeNext(StandardPipeline.java:596)
<br><br><br>at
<br>org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.java:433)
<br><br>at org.apache.catalina.core.ContainerBase.invoke(ContainerBase.java:955)
<br><br>at org.apache.ajp.tomcat4.Ajp13Processor.process(Ajp13Processor.java:410)
<br><br>at org.apache.ajp.tomcat4.Ajp13Processor.run(Ajp13Processor.java:529)
<br><br>at java.lang.Thread.run(Thread.java:745)
<br><br><br>--
<br>Thanks and Regards,
<br><br>Dhaval Yoganandi,
<br>Jr. Software Engineer,
<br>Elitecore Technologies Ltd.
<br>Cyberoam.
<br><br><br><p>From forum: <a href="http://old.nabble.com/classpathx-javamail-f16003.html" embed="fixTarget[16003]" target="_top" >classpathx-javamail</a></p>tag:old.nabble.com,2006:post-26612954Re: netif.h present but unusable on Mac2009-12-02T09:31:46Z2009-12-02T09:31:46Zgnu_andrew2009/11/22 Richard O'Keefe <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26612954&i=0" target="_top" rel="nofollow">ok@...</a>>:
<div class='shrinkable-quote'><br>> The host is an Intel Core 2 Duo processor running Mac OS X version 10.5.7
<br>>
<br>> I downloaded the JikesRVM version 3.1.0 and having checked that
<br>> ia32-osx was supported set out to build it. The manual says that
<br>> it needs GNU Classpath, and that I should download and install
<br>> that by doing ant -f build/components/classpath.xml, which I did.
<br>>
<br>> [exec] checking ifaddrs.h usability... yes
<br>> [exec] checking ifaddrs.h presence... yes
<br>> [exec] checking for ifaddrs.h... yes
<br>> [exec] checking netinet/in_systm.h usability... yes
<br>> [exec] checking netinet/in_systm.h presence... yes
<br>> [exec] checking for netinet/in_systm.h... yes
<br>> [exec] checking netinet/ip.h usability... yes
<br>> [exec] checking netinet/ip.h presence... yes
<br>> [exec] checking for netinet/ip.h... yes
<br>> [exec] checking net/if.h usability... no
<br>> [exec] checking net/if.h presence... yes
<br>> [exec] configure: WARNING: net/if.h: present but cannot be compiled
<br>> [exec] configure: WARNING: net/if.h: check for missing prerequisite
<br>> headers?
<br>> [exec] configure: WARNING: net/if.h: see the Autoconf documentation
<br>> [exec] configure: WARNING: net/if.h: section "Present But Cannot Be
<br>> Compiled"
<br>> [exec] configure: WARNING: net/if.h: proceeding with the preprocessor's
<br>> result
<br>> [exec] configure: WARNING: net/if.h: in the future, the compiler will
<br>> take precedence
<br>> [exec] configure: WARNING: ## -------------------------------- ##
<br>> [exec] configure: WARNING: ## Report this to <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26612954&i=1" target="_top" rel="nofollow">classpath@...</a> ##
<br>> [exec] configure: WARNING: ## -------------------------------- ##
<br>> [exec] checking for net/if.h... yes
<br>>
<br>> There was a further problem later down where the Ant script said
<br>> [exec] configure: error: The Java compiler
<br>> /home/cshome/o/ok/Desktop/jikesrvm-3.1.0/components/ecj/3.2/ecj-3.2/ecj
<br>> failed (see config.log, check the CLASSPATH?)
<br>> [exec] checking if
<br>> /home/cshome/o/ok/Desktop/jikesrvm-3.1.0/components/ecj/3.2/ecj-3.2/ecj
<br>> works...
<br>>
<br>>
<br>> Since ...components/ecj doesn't exist, it's not surprising that it
<br>> doesn't "work". This leads me to suspect a problem with the Ant
<br>> script.
<br>>
<br>>
<br>>
</div><br><br>As your mail concerns JikesRVM, I'm including the main JikesRVM list
<br>in this discussion (see <a href="http://jikesrvm.org/Mailing+Lists" target="_top" rel="nofollow">http://jikesrvm.org/Mailing+Lists</a>)
<br><br>My first suggestion would be that you try running the main build.xml
<br>in the root of the RVM tree by invoking 'ant' from that directory.
<br>You will need to set host.name to the desired value from build/hosts,
<br>ia32-osx in your case, so use 'ant -Dhost.name=ia32-osx'. Invoking
<br>the Classpath sub-build separately, as you do above, will mean that
<br>any prerequisites for it are not being built.
<br>--
<br>Andrew :-)
<br><br>Free Java Software Engineer
<br>Red Hat, Inc. (<a href="http://www.redhat.com" target="_top" rel="nofollow">http://www.redhat.com</a>)
<br><br>Support Free Java!
<br>Contribute to GNU Classpath and the OpenJDK
<br><a href="http://www.gnu.org/software/classpath" target="_top" rel="nofollow">http://www.gnu.org/software/classpath</a><br><a href="http://openjdk.java.net" target="_top" rel="nofollow">http://openjdk.java.net</a><br><br>PGP Key: 94EFD9D8 (<a href="http://subkeys.pgp.net" target="_top" rel="nofollow">http://subkeys.pgp.net</a>)
<br>Fingerprint: F8EF F1EA 401E 2E60 15FA 7927 142C 2591 94EF D9D8
<br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Classpath---General-f1525.html" embed="fixTarget[1525]" target="_top" >Gnu - Classpath - General</a></p>tag:old.nabble.com,2006:post-26605608Re: Re: Bug in 1.1.2: Folder.search() alters Folder.mode (works with 1.1.1)2009-12-02T00:58:48Z2009-12-02T00:58:48ZChris Burdess-2Hi Boris,
<br><br>Is that patch working for you?
<br>--
<br>Chris Burdess
<br><br><br><br><p>From forum: <a href="http://old.nabble.com/classpathx-javamail-f16003.html" embed="fixTarget[16003]" target="_top" >classpathx-javamail</a></p>tag:old.nabble.com,2006:post-26699893FOSDEM devroom request accepted2009-12-01T05:46:06Z2009-12-01T05:46:06ZDalibor Topic-5alibor Topic wrote:
<br>> > Hi,
<br>> >
<br>> > I expect to hear back from the FOSDEM devroom team whether the
<br>> > request was accepted (or denied) sometime next week, according to
<br>> > the published schedule. [3]
<br><br>Here's what the FOSDEM devroom team e-mailed me today:
<br><br>"Your request for a devroom at FOSDEM 2010 has been accepted, and we will
<br>put the following at your disposal:
<br>- room AW1.125 with 76 seats,
<br>- a video projector with VGA cable,
<br>- wireless internet,
<br>for 2 days:
<br>- on Saturday 6th February from 13:00 to 19:00
<br> and
<br>- on Sunday 7th February from 09:00 to 17:00
<br>[...]"
<br><br>Yay!
<br><br>OK, with that out of the way, it's same procedure as every year:
<br><br>* there is a wiki page with the basic information:
<br><a href="http://wiki.debian.org/Java/DevJam/2010/Fosdem" target="_top" rel="nofollow">http://wiki.debian.org/Java/DevJam/2010/Fosdem</a><br><br>* there is a wiki page where talk proposals go:
<br><a href="http://wiki.debian.org/Java/DevJam/2010/FosdemProposedTalks" target="_top" rel="nofollow">http://wiki.debian.org/Java/DevJam/2010/FosdemProposedTalks</a><br><br>Talk slots are either for 15 or 30 minutes. Like in the past years,
<br>we'll group related talks into blocks, in order to allow for some
<br>breaks between blocks of sessions. You can submit more then one
<br>proposal if you wish. If we run out of slots (happened a couple of
<br>times in previous years), not all submissions will get picked - so
<br>please try to make your submission sound like a great pick.
<br><br>The call for submissions runs for four weeks until Sunday, January
<br>3rd 2010. You'll know if your talk was picked by January 6th, 2010.
<br><br>cheers,
<br>dalibor topic
<br>--
<br>*******************************************************************
<br>Dalibor Topic Tel: (+49 40) 23 646 738
<br>Java F/OSS Ambassador AIM: robiladonaim
<br>Sun Microsystems GmbH Mobile: (+49 177) 2664 192
<br>Nagelsweg 55 <a href="http://openjdk.java.net" target="_top" rel="nofollow">http://openjdk.java.net</a><br>D-20097 Hamburg mailto:<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26699893&i=0" target="_top" rel="nofollow">Dalibor.Topic@...</a>
<br>Sitz der Gesellschaft: Sonnenallee 1, D-85551 Kirchheim-Heimstetten
<br>Amtsgericht München: HRB 161028
<br>Geschäftsführer: Thomas Schröder, Wolfgang Engels, Wolf Frenkel
<br>Vorsitzender des Aufsichtsrates: Martin Häring
<br><br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Classpath---General-f1525.html" embed="fixTarget[1525]" target="_top" >Gnu - Classpath - General</a></p>tag:old.nabble.com,2006:post-26544652Re: Re: UnsupportedEncodingException for unicode-1-1-utf-72009-11-27T08:26:16Z2009-11-27T08:26:16ZChris Burdess-2Boris Folgmann wrote:
<div class='shrinkable-quote'><br>> Hi Chris,
<br>> Chris Burdess schrieb/wrote:
<br>>> UTF7 is supported, however I have never heard of anything being
<br>>> identified as "unicode-1-1-utf-7". Do you have a copy of the original
<br>>> message to hand?
<br>>
<br>> Cite from <a href="http://www.faqs.org/rfcs/rfc1642.html" target="_top" rel="nofollow">http://www.faqs.org/rfcs/rfc1642.html</a><br>>
<br>> -------------------------------------------------------------------------
<br>> Use of Character Set UTF-7 Within MIME
<br>>
<br>> Character set UTF-7 is safe for mail transmission and therefore may
<br>> be used with any content transfer encoding in MIME (except where line
<br>> length and line break restrictions are violated). Specifically, the 7
<br>> bit encoding for bodies and the Q encoding for headers are both
<br>> acceptable. The MIME character set identifier is UNICODE-1-1-UTF-7.
<br>>
<br>> Example. Here is a text portion of a MIME message containing the
<br>> Unicode sequence "Hi Mom <WHITE SMILING FACE>!" (hexadecimal 0048,
<br>> 0069, 0020, 004D, 006F, 004D, 0020, 263A, 0021).
<br>>
<br>> Content-Type: text/plain; charset=UNICODE-1-1-UTF-7
<br>>
<br>> Hi Mom +Jjo-!
<br>> [...]
<br>> -------------------------------------------------------------------------
</div><br>Ah, my bad. UTF7 is only supported at the IMAP protocol level, not at the MIME encoding level. If you're interested in developing this functionality, the class gnu.inet.imap.UTF7imap may be helpful.
<br>--
<br>Chris Burdess
<br><br><br><br><p>From forum: <a href="http://old.nabble.com/classpathx-javamail-f16003.html" embed="fixTarget[16003]" target="_top" >classpathx-javamail</a></p>tag:old.nabble.com,2006:post-26544512Re: UnsupportedEncodingException for unicode-1-1-utf-72009-11-27T08:16:00Z2009-11-27T08:16:00ZBoris FolgmannHi Chris,
<br><br>Chris Burdess schrieb/wrote:
<br>> UTF7 is supported, however I have never heard of anything being
<br>> identified as "unicode-1-1-utf-7". Do you have a copy of the original
<br>> message to hand?
<br><br>Cite from <a href="http://www.faqs.org/rfcs/rfc1642.html" target="_top" rel="nofollow">http://www.faqs.org/rfcs/rfc1642.html</a><br><br>-------------------------------------------------------------------------
<br>Use of Character Set UTF-7 Within MIME
<br><br> Character set UTF-7 is safe for mail transmission and therefore may
<br> be used with any content transfer encoding in MIME (except where line
<br> length and line break restrictions are violated). Specifically, the 7
<br> bit encoding for bodies and the Q encoding for headers are both
<br> acceptable. The MIME character set identifier is UNICODE-1-1-UTF-7.
<br><br> Example. Here is a text portion of a MIME message containing the
<br> Unicode sequence "Hi Mom <WHITE SMILING FACE>!" (hexadecimal 0048,
<br> 0069, 0020, 004D, 006F, 004D, 0020, 263A, 0021).
<br><br> Content-Type: text/plain; charset=UNICODE-1-1-UTF-7
<br><br> Hi Mom +Jjo-!
<br>[...]
<br>-------------------------------------------------------------------------
<br><br>cu,
<br> boris
<br><br><br><br><p>From forum: <a href="http://old.nabble.com/classpathx-javamail-f16003.html" embed="fixTarget[16003]" target="_top" >classpathx-javamail</a></p>tag:old.nabble.com,2006:post-26543148Re: UnsupportedEncodingException for unicode-1-1-utf-72009-11-27T06:28:22Z2009-11-27T06:28:22ZChris Burdess-2Boris Folgmann wrote:
<br>> I get this exception when parsing an email:
<br>>
<br>> java.io.UnsupportedEncodingException: unicode-1-1-utf-7 at sun.io.Converters.getConverterClass(Converters.java:218) at sun.io.Converters.newConverter(Converters.java:251) at sun.io.ByteToCharConverter.getConverter(ByteToCharConverter.java:68) at sun.nio.cs.StreamDecoder$ConverterSD.<init>(StreamDecoder.java:224) at sun.nio.cs.StreamDecoder$ConverterSD.<init>(StreamDecoder.java:210) at sun.nio.cs.StreamDecoder.forInputStreamReader(StreamDecoder.java:77) at java.io.InputStreamReader.<init>(InputStreamReader.java:83) at gnu.mail.handler.Text.getContent(Text.java:106) at javax.activationDataSourceDataContentHandler.getContent(DataHandler.java:803) at javax.activation.DataHandler.getContent(DataHandler.java:550) at javax.mail.internet.MimeBodyPart.getContent(MimeBodyPart.java:691) at gnu.mail.providers.imap.IMAPBodyPart.getContent(IMAPBodyPart.java:282)
<br>>
<br>> The encoding might be unsupported by JavaSE 5.0, but as it is a common encoding of emails, as you can read in RFC1642 (UTF-7 - A Mail-Safe Transformation Format of Unicode), I don't see why javamail shouldn't support it.
<br>>
<br>> Is it possible to implement a dirty work-around like forcing the BodyPart's encoding to US-ASCII before calling getContent()? I couldn't find a suitable method.
<br><br>UTF7 is supported, however I have never heard of anything being identified as "unicode-1-1-utf-7". Do you have a copy of the original message to hand?
<br>--
<br>Chris Burdess
<br><br><br><br><p>From forum: <a href="http://old.nabble.com/classpathx-javamail-f16003.html" embed="fixTarget[16003]" target="_top" >classpathx-javamail</a></p>tag:old.nabble.com,2006:post-26541993UnsupportedEncodingException for unicode-1-1-utf-72009-11-27T04:38:22Z2009-11-27T04:38:22ZBoris FolgmannHi,
<br><br>I get this exception when parsing an email:
<br><br>java.io.UnsupportedEncodingException: unicode-1-1-utf-7 at
<br>sun.io.Converters.getConverterClass(Converters.java:218) at
<br>sun.io.Converters.newConverter(Converters.java:251) at
<br>sun.io.ByteToCharConverter.getConverter(ByteToCharConverter.java:68) at
<br>sun.nio.cs.StreamDecoder$ConverterSD.<init>(StreamDecoder.java:224) at
<br>sun.nio.cs.StreamDecoder$ConverterSD.<init>(StreamDecoder.java:210) at
<br>sun.nio.cs.StreamDecoder.forInputStreamReader(StreamDecoder.java:77) at
<br>java.io.InputStreamReader.<init>(InputStreamReader.java:83) at
<br>gnu.mail.handler.Text.getContent(Text.java:106) at
<br>javax.activationDataSourceDataContentHandler.getContent(DataHandler.java:803)
<br>at javax.activation.DataHandler.getContent(DataHandler.java:550) at
<br>javax.mail.internet.MimeBodyPart.getContent(MimeBodyPart.java:691) at
<br>gnu.mail.providers.imap.IMAPBodyPart.getContent(IMAPBodyPart.java:282)
<br><br>The encoding might be unsupported by JavaSE 5.0, but as it is a common
<br>encoding of emails, as you can read in RFC1642 (UTF-7 - A Mail-Safe
<br>Transformation Format of Unicode), I don't see why javamail shouldn't
<br>support it.
<br><br>Is it possible to implement a dirty work-around like forcing the BodyPart's
<br>encoding to US-ASCII before calling getContent()? I couldn't find a
<br>suitable method.
<br><br>cu,
<br> boris
<br><br><br><br><p>From forum: <a href="http://old.nabble.com/classpathx-javamail-f16003.html" embed="fixTarget[16003]" target="_top" >classpathx-javamail</a></p>tag:old.nabble.com,2006:post-26526095Re: Possible bug in gnu.inet.util.BASE642009-11-26T00:40:53Z2009-11-26T00:40:53ZChris Burdess-2Jorge Barreiro wrote:
<div class='shrinkable-quote'><br>> I've found what seems to be a bug in the gnu.inet.util.BASE64 class: The
<br>> encoding function never appends the '=' padding characters. I sent this mail
<br>> first because there's some code in the class dealing with this, but it's
<br>> commented, so I don't know if this is intentional. In addition, that library
<br>> wasn't updated in years. Is it deprecated? Is there any alternative?
<br>>
<br>> This class is used in the gnu javax.mail implementation, and is preventing me
<br>> from authenticating with a server.
<br>>
<br>> If you confirm me that the padding characters should be there and this is a
<br>> bug, I can make and send you a patch.
</div><br>I'm pretty sure this was an old bug that was fixed a while ago. Check that you're using the most up-to-date release.
<br><br>If it's still not working, please feel free to send me a patch against CVS HEAD.
<br>--
<br>Chris Burdess
<br><br><br><br>_______________________________________________
<br>Classpath-inetlib mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26526095&i=0" target="_top" rel="nofollow">Classpath-inetlib@...</a>
<br><a href="http://lists.gnu.org/mailman/listinfo/classpath-inetlib" target="_top" rel="nofollow">http://lists.gnu.org/mailman/listinfo/classpath-inetlib</a><br><p>From forum: <a href="http://old.nabble.com/Gnu---Classpath---Inetlib-f1526.html" embed="fixTarget[1526]" target="_top" >Gnu - Classpath - Inetlib</a></p>tag:old.nabble.com,2006:post-26516746Re: netif.h present but unusable on Mac2009-11-25T09:19:38Z2009-11-25T09:19:38ZIan Rogers (nabble)Hi Richard,
<br><br>Jikes RVM is using an older version of Classpath than what is current.
<br>MRP [1] (derived from Jikes RVM) uses the latest GNU Classpath. You
<br>may want to try this as an alternative, although currently you'll need
<br>to download MRP using git.
<br><br>Regards,
<br>Ian
<br><br>[1] <a href="http://mrp.codehaus.org/" target="_top" rel="nofollow">http://mrp.codehaus.org/</a><br><br>2009/11/22 Richard O'Keefe <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26516746&i=0" target="_top" rel="nofollow">ok@...</a>>:
<div class='shrinkable-quote'><br>> The host is an Intel Core 2 Duo processor running Mac OS X version 10.5.7
<br>>
<br>> I downloaded the JikesRVM version 3.1.0 and having checked that
<br>> ia32-osx was supported set out to build it. The manual says that
<br>> it needs GNU Classpath, and that I should download and install
<br>> that by doing ant -f build/components/classpath.xml, which I did.
<br>>
<br>> [exec] checking ifaddrs.h usability... yes
<br>> [exec] checking ifaddrs.h presence... yes
<br>> [exec] checking for ifaddrs.h... yes
<br>> [exec] checking netinet/in_systm.h usability... yes
<br>> [exec] checking netinet/in_systm.h presence... yes
<br>> [exec] checking for netinet/in_systm.h... yes
<br>> [exec] checking netinet/ip.h usability... yes
<br>> [exec] checking netinet/ip.h presence... yes
<br>> [exec] checking for netinet/ip.h... yes
<br>> [exec] checking net/if.h usability... no
<br>> [exec] checking net/if.h presence... yes
<br>> [exec] configure: WARNING: net/if.h: present but cannot be compiled
<br>> [exec] configure: WARNING: net/if.h: check for missing prerequisite
<br>> headers?
<br>> [exec] configure: WARNING: net/if.h: see the Autoconf documentation
<br>> [exec] configure: WARNING: net/if.h: section "Present But Cannot Be
<br>> Compiled"
<br>> [exec] configure: WARNING: net/if.h: proceeding with the preprocessor's
<br>> result
<br>> [exec] configure: WARNING: net/if.h: in the future, the compiler will
<br>> take precedence
<br>> [exec] configure: WARNING: ## -------------------------------- ##
<br>> [exec] configure: WARNING: ## Report this to <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26516746&i=1" target="_top" rel="nofollow">classpath@...</a> ##
<br>> [exec] configure: WARNING: ## -------------------------------- ##
<br>> [exec] checking for net/if.h... yes
<br>>
<br>> There was a further problem later down where the Ant script said
<br>> [exec] configure: error: The Java compiler
<br>> /home/cshome/o/ok/Desktop/jikesrvm-3.1.0/components/ecj/3.2/ecj-3.2/ecj
<br>> failed (see config.log, check the CLASSPATH?)
<br>> [exec] checking if
<br>> /home/cshome/o/ok/Desktop/jikesrvm-3.1.0/components/ecj/3.2/ecj-3.2/ecj
<br>> works...
<br>>
<br>>
<br>> Since ...components/ecj doesn't exist, it's not surprising that it
<br>> doesn't "work". This leads me to suspect a problem with the Ant
<br>> script.
<br>>
<br>>
<br>>
</div><br><p>From forum: <a href="http://old.nabble.com/Gnu---Classpath---General-f1525.html" embed="fixTarget[1525]" target="_top" >Gnu - Classpath - General</a></p>tag:old.nabble.com,2006:post-26510007Possible bug in gnu.inet.util.BASE642009-11-25T01:50:50Z2009-11-25T01:50:50ZJorge BarreiroHello,
<br><br>I've found what seems to be a bug in the gnu.inet.util.BASE64 class: The
<br>encoding function never appends the '=' padding characters. I sent this mail
<br>first because there's some code in the class dealing with this, but it's
<br>commented, so I don't know if this is intentional. In addition, that library
<br>wasn't updated in years. Is it deprecated? Is there any alternative?
<br><br>This class is used in the gnu javax.mail implementation, and is preventing me
<br>from authenticating with a server.
<br><br>If you confirm me that the padding characters should be there and this is a
<br>bug, I can make and send you a patch.
<br><br><br>_______________________________________________
<br>Classpath-inetlib mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26510007&i=0" target="_top" rel="nofollow">Classpath-inetlib@...</a>
<br><a href="http://lists.gnu.org/mailman/listinfo/classpath-inetlib" target="_top" rel="nofollow">http://lists.gnu.org/mailman/listinfo/classpath-inetlib</a><br><p>From forum: <a href="http://old.nabble.com/Gnu---Classpath---Inetlib-f1526.html" embed="fixTarget[1526]" target="_top" >Gnu - Classpath - Inetlib</a></p>tag:old.nabble.com,2006:post-26509802netif.h present but unusable on Mac2009-11-22T13:35:27Z2009-11-22T13:35:27ZRichard O'KeefeThe host is an Intel Core 2 Duo processor running Mac OS X version
<br>10.5.7
<br><br>I downloaded the JikesRVM version 3.1.0 and having checked that
<br>ia32-osx was supported set out to build it. The manual says that
<br>it needs GNU Classpath, and that I should download and install
<br>that by doing ant -f build/components/classpath.xml, which I did.
<br><br> [exec] checking ifaddrs.h usability... yes
<br> [exec] checking ifaddrs.h presence... yes
<br> [exec] checking for ifaddrs.h... yes
<br> [exec] checking netinet/in_systm.h usability... yes
<br> [exec] checking netinet/in_systm.h presence... yes
<br> [exec] checking for netinet/in_systm.h... yes
<br> [exec] checking netinet/ip.h usability... yes
<br> [exec] checking netinet/ip.h presence... yes
<br> [exec] checking for netinet/ip.h... yes
<br> [exec] checking net/if.h usability... no
<br> [exec] checking net/if.h presence... yes
<br> [exec] configure: WARNING: net/if.h: present but cannot be
<br>compiled
<br> [exec] configure: WARNING: net/if.h: check for missing
<br>prerequisite headers?
<br> [exec] configure: WARNING: net/if.h: see the Autoconf
<br>documentation
<br> [exec] configure: WARNING: net/if.h: section "Present But
<br>Cannot Be Compiled"
<br> [exec] configure: WARNING: net/if.h: proceeding with the
<br>preprocessor's result
<br> [exec] configure: WARNING: net/if.h: in the future, the compiler
<br>will take precedence
<br> [exec] configure: WARNING: ##
<br>-------------------------------- ##
<br> [exec] configure: WARNING: ## Report this to
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26509802&i=0" target="_top" rel="nofollow">classpath@...</a> ##
<br> [exec] configure: WARNING: ##
<br>-------------------------------- ##
<br> [exec] checking for net/if.h... yes
<br><br>There was a further problem later down where the Ant script said
<br> [exec] configure: error: The Java compiler /home/cshome/o/ok/
<br>Desktop/jikesrvm-3.1.0/components/ecj/3.2/ecj-3.2/ecj failed (see
<br>config.log, check the CLASSPATH?)
<br> [exec] checking if /home/cshome/o/ok/Desktop/jikesrvm-3.1.0/
<br>components/ecj/3.2/ecj-3.2/ecj works...
<br><br><br>Since ...components/ecj doesn't exist, it's not surprising that it
<br>doesn't "work". This leads me to suspect a problem with the Ant
<br>script.
<br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Classpath---General-f1525.html" embed="fixTarget[1525]" target="_top" >Gnu - Classpath - General</a></p>tag:old.nabble.com,2006:post-26370160Re: Re: Bug in 1.1.2: Folder.search() alters Folder.mode (works with 1.1.1)2009-11-16T03:08:02Z2009-11-16T03:08:02ZChris Burdess-2Boris Folgmann wrote:
<br>> Chris Burdess schrieb/wrote:
<br>>> Boris Folgmann wrote:
<br>>>> I hope you find the problem.
<br>>>
<br>>> Please could you try with the attached patches to the inetlib and mail
<br>>> packages (from CVS HEAD) and let me know if this works for you?
<br>>
<br>> Sorry, no attachments. Is it because I'm reading through news.gmane.org?
<br><br>Hmm, maybe the listserv is stripping them out. Try this:
<br><br>----inetlib.patch----
<br>Index: source/gnu/inet/imap/IMAPConnection.java
<br>===================================================================
<br>RCS file: /sources/classpath/inetlib/source/gnu/inet/imap/IMAPConnection.java,v
<br>retrieving revision 1.33
<br>diff -r1.33 IMAPConnection.java
<br>841a842
<br>> ms.select = true;
<br>Index: source/gnu/inet/imap/MailboxStatus.java
<br>===================================================================
<br>RCS file: /sources/classpath/inetlib/source/gnu/inet/imap/MailboxStatus.java,v
<br>retrieving revision 1.5
<br>diff -r1.5 MailboxStatus.java
<br>64a65,66
<br>> public boolean select;
<br>>
<br><br>----mail.patch----
<br>Index: build.xml
<br>===================================================================
<br>RCS file: /sources/classpathx/mail/build.xml,v
<br>retrieving revision 1.17
<br>diff -r1.17 build.xml
<br>128a129
<br>> </echo>
<br>Index: source/gnu/mail/providers/imap/IMAPFolder.java
<br>===================================================================
<br>RCS file: /sources/classpathx/mail/source/gnu/mail/providers/imap/IMAPFolder.java,v
<br>retrieving revision 1.23
<br>diff -r1.23 IMAPFolder.java
<br>152c152,155
<br>< mode = status.readWrite ? Folder.READ_WRITE : Folder.READ_ONLY;
<br>---
<br>> if (status.select)
<br>> {
<br>> mode = status.readWrite ? Folder.READ_WRITE : Folder.READ_ONLY;
<br>> }
<br><br>--
<br>Chris Burdess
<br><br><br><br><p>From forum: <a href="http://old.nabble.com/classpathx-javamail-f16003.html" embed="fixTarget[16003]" target="_top" >classpathx-javamail</a></p>tag:old.nabble.com,2006:post-26370084Re: Bug in 1.1.2: Folder.search() alters Folder.mode (works with 1.1.1)2009-11-16T03:03:12Z2009-11-16T03:03:12ZBoris FolgmannChris Burdess schrieb/wrote:
<br>> Boris Folgmann wrote:
<br>>> I hope you find the problem.
<br>>
<br>> Please could you try with the attached patches to the inetlib and mail
<br>> packages (from CVS HEAD) and let me know if this works for you?
<br><br>Sorry, no attachments. Is it because I'm reading through news.gmane.org?
<br><br>cu,
<br> boris
<br><br><br><br><p>From forum: <a href="http://old.nabble.com/classpathx-javamail-f16003.html" embed="fixTarget[16003]" target="_top" >classpathx-javamail</a></p>tag:old.nabble.com,2006:post-26369908Re: Re: Bug in 1.1.2: Folder.search() alters Folder.mode (works with 1.1.1)2009-11-16T02:46:59Z2009-11-16T02:46:59ZChris Burdess-2Boris Folgmann wrote:
<br>> I hope you find the problem.
<br><br>Please could you try with the attached patches to the inetlib and mail packages (from CVS HEAD) and let me know if this works for you?
<br>--
<br>Chris Burdess
<br><p>From forum: <a href="http://old.nabble.com/classpathx-javamail-f16003.html" embed="fixTarget[16003]" target="_top" >classpathx-javamail</a></p>tag:old.nabble.com,2006:post-26338953Err...2009-11-13T08:04:28Z2009-11-13T08:04:28ZAndrew Haleyrecursion: n.
<br><br> See recursion.
<br><br>:-)
<br><br>Andrew.
<br><br><br>2009-11-13 Andrew Haley <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26338953&i=0" target="_top" rel="nofollow">aph@...</a>>
<br><br> * java/util/TreeMap.java (anonymous class.size()): Debogosify.
<br> anonymous class.clear(): Likewise.
<br><br>Index: classpath/java/util/TreeMap.java
<br>===================================================================
<br>--- classpath/java/util/TreeMap.javax (revision 153739)
<br>+++ classpath/java/util/TreeMap.java (working copy)
<br>@@ -2745,7 +2745,7 @@
<br> {
<br> public int size()
<br> {
<br>- return size();
<br>+ return DescendingMap.this.size();
<br> }
<br>
<br> public Iterator<DV> iterator()
<br>@@ -2786,7 +2786,7 @@
<br>
<br> public void clear()
<br> {
<br>- clear();
<br>+ DescendingMap.this.clear();
<br> }
<br> };
<br> return values;
<br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Classpath---Patches-f1527.html" embed="fixTarget[1527]" target="_top" >Gnu - Classpath - Patches</a></p>tag:old.nabble.com,2006:post-26337365Re: Bug in 1.1.2: Folder.search() alters Folder.mode (works with 1.1.1)2009-11-13T06:24:08Z2009-11-13T06:24:08ZBoris FolgmannHi Chris,
<br><br>Chris Burdess schrieb/wrote:
<br>> The mode is only changed from a call to update(MailboxStatus,
<br>> boolean). This can happen during various method calls, but not search
<br>> (). Are you sure this is your complete code with respect to IMAPFolder?
<br><br>Yes, I tried some even simpler constructions:
<br><br> public void modeBug() throws MessagingException, IOException
<br> {
<br> try {
<br> store.connect();
<br> Folder folder = store.getFolder("empty IMAP folder");
<br> folder.open(Folder.READ_WRITE);
<br> System.out.println("10 mode: " + folder.getMode());
<br> folder.expunge();
<br> System.out.println("20 mode: " + folder.getMode());
<br> folder.search(remoteDsnTerm); // returns 0 msgs, as folder is empty
<br> System.out.println("30 mode: " + folder.getMode());
<br> folder.close(true);
<br> } finally {
<br> store.close();
<br> }
<br> }
<br><br>The output is:
<br>10 mode: 2
<br>20 mode: 2
<br>30 mode: 1
<br><br>I further investigated and found out that the mode change does not happen
<br>with every search term. Here's mine triggering the bug:
<br><br>remoteDsnTerm = new AndTerm(new SearchTerm[] {
<br> new OrTerm(new SearchTerm[] {
<br> new FromStringTerm("Mail Delivery Subsystem <MAILER-DAEMON@"),
<br> new FromStringTerm("Mail Delivery System <MAILER-DAEMON@"),
<br> new FromStringTerm("Mail Delivery Service <postmaster@")
<br> }),
<br> new RecipientTerm(Message.RecipientType.TO, new
<br>InternetAddress(toAddress)),
<br> new HeaderTerm("content-type", "multipart/report;
<br>report-type=delivery-status;")
<br> });
<br><br>1. The bug happens not if the complete OrTerm is removed.
<br>2. The bug happens not if expunge() is not called before search().
<br><br>I hope you find the problem.
<br><br>cu,
<br> boris
<br><br><br><br><p>From forum: <a href="http://old.nabble.com/classpathx-javamail-f16003.html" embed="fixTarget[16003]" target="_top" >classpathx-javamail</a></p>tag:old.nabble.com,2006:post-26454573Profiling application on JamVm2009-11-13T03:39:13Z2009-11-13T03:39:13ZSidh<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
<title></title>
</head>
<body bgcolor="#ffffff" text="#000000">
<small><font face="Arial">Hello,<br>
<br>
I have deployed web application with Jam VM and gnu classpath,<br>
After deploying web application in Jetty , It grabs 20+ MB Residential
memory<br>
Need to why these much memory required by application / Virtual Machine.<br>
<br>
For need for profiling , Using JProfiler have to add two arguments
while invoking container <br>
<br>
Adding two arguments in invoking Jetty <br>
-Xrunjprofiler:port=8849 -Xbootclasspath/a:/sig/jprofiler6/bin/agent.jar<br>
<br>
JamVm does not accept -Xrunjprofiler:port=8849 as an argument while it
works fine with JVM distribution<br>
exported required library(ibjprofiler.so) path by jprofiler <br>
<br>
I m not getting why JamVm does not accept </font></small><small><font face="Arial"> -Xrunjprofiler:port=8849 argument<br>
Is there any profiling tool which is preferable and working on jamvm<br>
<br>
class path configured with no arguments (all components taken)<br>
<br>
</font></small><small><font face="Arial">regards,<br>
Siddharth<br>
</font></small>
</body>
</html>
<p>From forum: <a href="http://old.nabble.com/Gnu---Classpath---General-f1525.html" embed="fixTarget[1525]" target="_top" >Gnu - Classpath - General</a></p>tag:old.nabble.com,2006:post-26332398Re: Bug in 1.1.2: Folder.search() alters Folder.mode (works with 1.1.1)2009-11-12T23:23:22Z2009-11-12T23:23:22ZChris Burdess-2Boris Folgmann wrote:
<div class='shrinkable-quote'><br>> I've got a code that worked perfectly with mail 1.1.1 but has
<br>> problems with 1.1.2.
<br>>
<br>> ----------------------------------------------------------------
<br>> store.connect();
<br>> Folder folder = store.getFolder("a cyrus imap folder");
<br>> folder.open(Folder.READ_WRITE);
<br>> log.info("Total messages: " + folder.getMessageCount());
<br>>
<br>> Message[] msgs = folder.search(aTerm);
<br>> processMessages(msgs);
<br>> folder.expunge();
<br>> // folder.getMode() still returns Folder.READ_WRITE here
<br>>
<br>> msgs = folder.search(aDifferentTerm);
<br>> // folder.getMode() now returns Folder.READ_ONLY!!
<br>> processMessages(msgs);
<br>> folder.expunge();
<br>>
<br>> ----------------------------------------------------------------
<br>>
<br>> The second folder.expunge() throws
<br>>
<br>> javax.mail.MessagingException: Folder was opened read-only
<br>> at gnu.mail.providers.imap.IMAPFolder.expunge(IMAPFolder.java:
<br>> 530)
<br>>
<br>> Using some more log.info() I found out that the Folder.mode is
<br>> altered by the second search. I couldn't find anything in the
<br>> JavaDocs that this is a regular behaviour. As it worked with the
<br>> previous release, you may know what the problem could be.
</div><br><br>The mode is only changed from a call to update(MailboxStatus,
<br>boolean). This can happen during various method calls, but not search
<br>(). Are you sure this is your complete code with respect to IMAPFolder?
<br>--
<br>Chris Burdess
<br><br><br><br><br><p>From forum: <a href="http://old.nabble.com/classpathx-javamail-f16003.html" embed="fixTarget[16003]" target="_top" >classpathx-javamail</a></p>tag:old.nabble.com,2006:post-26323193Re: Questions about jvmti.h header2009-11-12T09:49:32Z2009-11-12T09:49:32ZKeith SeitzOn 11/03/2009 04:27 AM, Michael Starzinger wrote:
<br><br>> * Are there any plans to install the jvmti.h file (just like it is done
<br>> with jni.h) or are there circumstances preventing such an installation?
<br><br>That is probably just an oversight.
<br><br>> * Why does the mentioned header include jvmti_md.h? All
<br>> machine-dependent definitions should have already been covered by
<br>> jni_md.h. Should the missing jvmti_md.h file be provided by the VM?
<br><br>A VM may need to specialize some behavior in jvmti.h. Gcj did this, and
<br>it provides the missing jvmti_md.h. If jni_md.h supplies everything
<br>necessary, a "blank"/nop jvmti_md.h will suffice.
<br><br>> I also stumbled upon a mailing list post[1] dealing with this topic. But
<br>> it seems to me that only a small part[2] actually made it into the GNU
<br>> Classpath source, the rest remained in the GCJ source.
<br><br>I think this is self-explanatory now: jvmti_md.h is VM-dependent and
<br>should be supplied by the VM vendor. That's why that file never made it
<br>into classpath. It wouldn't help any other VM.
<br><br>Keith
<br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Classpath---General-f1525.html" embed="fixTarget[1525]" target="_top" >Gnu - Classpath - General</a></p>tag:old.nabble.com,2006:post-26322515Re: Classpath 0.98 & jks2009-11-12T09:10:27Z2009-11-12T09:10:27ZArigeadAndrew John Hughes wrote:
<div class='shrinkable-quote'><br>> 2009/11/12 Arigead <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26322515&i=0" target="_top" rel="nofollow">captain.deadly@...</a>>:
<br>>> Hello All,
<br>>> sorry new here and haven't had time to lurk for a while to see the
<br>>> lie of the land.
<br>>>
<br>>> I'm trying to run Jxta on Jamvm 1.5.3 with Classpath 0.98. I'm getting
<br>>> an error because of jks security not being found. I'll post the error
<br>>> I'm getting at the end of this email.
<br>>>
<br>>> A bit of searching on the Internet gives me a few posts by somebody with
<br>>> a similar problem [1] & [2]. I can't find any responses and solutions to
<br>>> this problem.
<br>>>
<br>>> I'm not sure how to get around this issue. I'm sure that there are many
<br>>> implementations of this security algorithm and I wonder can a lump of
<br>>> code be added included into the JRE or can source code be added to
<br>>> classpath. I mean there could I, if I had a clue ;-) add the necessary
<br>>> code into classpath source? I'd prefer if I could pull in something like
<br>>> BouncyCastle, although I'm not sure it would solve the problem as I'm
<br>>> not sure it implements this functionality.
<br>>>
<br>>> I'm not great with Java I'm afraid so I'm not sure that a BounceyCastle
<br>>> could be used to implement functionality that is being requested of the JRE.
<br>>>
<br>>> Sorry for a newbie post and thanks a million for any advice that anybody
<br>>> can offer.
<br>>>
<br>>>
<br>>> [1]
<br>>> <a href="http://old.nabble.com/Problem-in-using-SSL-%28secure-socket%29-with-classpath-0.98-and-Jamvm-1.5.3-td23997596.html" target="_top" rel="nofollow">http://old.nabble.com/Problem-in-using-SSL-%28secure-socket%29-with-classpath-0.98-and-Jamvm-1.5.3-td23997596.html</a><br>>> [2] <a href="http://www.mail-archive.com/classpath@gnu.org/msg15120.html" target="_top" rel="nofollow">http://www.mail-archive.com/classpath@.../msg15120.html</a><br>>>
<br>>>
<br>>>
<br>>>
<br>>> Caused by: java.security.KeyStoreException: jks
<br>>> at java.security.KeyStore.getInstance(KeyStore.java:203)
<br>>> at java.security.KeyStore.getInstance(KeyStore.java:130)
<br>>> at
<br>>> net.jxta.impl.membership.pse.CMKeyStoreManager.<init>(CMKeyStoreManager.java:148)
<br>>> at
<br>>> net.jxta.impl.membership.pse.PSEKeyStoreManagerFactory$PSEKeyStoreManagerFactoryDefault.getInstance(PSEKeyStoreManagerFactory.java:143)
<br>>> ...24 more
<br>>> Caused by: java.security.NoSuchAlgorithmException: Algorithm [jks] of
<br>>> type [KeyStore] from provider
<br>>> [gnu.javax.security.auth.callback.GnuCallbacks: name=GNU-CALLBACKS
<br>>> version=2.1] is not found
<br>>> at gnu.java.security.Engine.getInstance(Engine.java:191)
<br>>> at gnu.java.security.Engine.getInstance(Engine.java:105)
<br>>> at java.security.KeyStore.getInstance(KeyStore.java:188)
<br>>> ...27 more
<br>>>
<br>>>
<br>>>
<br>>>
<br>>>
<br>>>
<br>>
<br>> jks is not a crypto algorithm but a keystore format, and a proprietary
<br>> Sun one until the release of OpenJDK.
<br>> I guess in theory support could now be developed using the Sun source
<br>> code as a reference, but there's not much work on Classpath in general
<br>> at present.
<br>>
<br>> <a href="http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34731" target="_top" rel="nofollow">http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34731</a></div><br>Andrew thanks a million for the Education. That totally answers my
<br>question perfectly. Don't get me a quick easy and dirty solution but at
<br>least now I know where I stand.
<br><br>cheers
<br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Classpath---General-f1525.html" embed="fixTarget[1525]" target="_top" >Gnu - Classpath - General</a></p>tag:old.nabble.com,2006:post-26321291Re: Classpath 0.98 & jks2009-11-12T08:02:31Z2009-11-12T08:02:31Zgnu_andrew2009/11/12 Arigead <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26321291&i=0" target="_top" rel="nofollow">captain.deadly@...</a>>:
<div class='shrinkable-quote'><br>> Hello All,
<br>> sorry new here and haven't had time to lurk for a while to see the
<br>> lie of the land.
<br>>
<br>> I'm trying to run Jxta on Jamvm 1.5.3 with Classpath 0.98. I'm getting
<br>> an error because of jks security not being found. I'll post the error
<br>> I'm getting at the end of this email.
<br>>
<br>> A bit of searching on the Internet gives me a few posts by somebody with
<br>> a similar problem [1] & [2]. I can't find any responses and solutions to
<br>> this problem.
<br>>
<br>> I'm not sure how to get around this issue. I'm sure that there are many
<br>> implementations of this security algorithm and I wonder can a lump of
<br>> code be added included into the JRE or can source code be added to
<br>> classpath. I mean there could I, if I had a clue ;-) add the necessary
<br>> code into classpath source? I'd prefer if I could pull in something like
<br>> BouncyCastle, although I'm not sure it would solve the problem as I'm
<br>> not sure it implements this functionality.
<br>>
<br>> I'm not great with Java I'm afraid so I'm not sure that a BounceyCastle
<br>> could be used to implement functionality that is being requested of the JRE.
<br>>
<br>> Sorry for a newbie post and thanks a million for any advice that anybody
<br>> can offer.
<br>>
<br>>
<br>> [1]
<br>> <a href="http://old.nabble.com/Problem-in-using-SSL-%28secure-socket%29-with-classpath-0.98-and-Jamvm-1.5.3-td23997596.html" target="_top" rel="nofollow">http://old.nabble.com/Problem-in-using-SSL-%28secure-socket%29-with-classpath-0.98-and-Jamvm-1.5.3-td23997596.html</a><br>> [2] <a href="http://www.mail-archive.com/classpath@gnu.org/msg15120.html" target="_top" rel="nofollow">http://www.mail-archive.com/classpath@.../msg15120.html</a><br>>
<br>>
<br>>
<br>>
<br>> Caused by: java.security.KeyStoreException: jks
<br>> at java.security.KeyStore.getInstance(KeyStore.java:203)
<br>> at java.security.KeyStore.getInstance(KeyStore.java:130)
<br>> at
<br>> net.jxta.impl.membership.pse.CMKeyStoreManager.<init>(CMKeyStoreManager.java:148)
<br>> at
<br>> net.jxta.impl.membership.pse.PSEKeyStoreManagerFactory$PSEKeyStoreManagerFactoryDefault.getInstance(PSEKeyStoreManagerFactory.java:143)
<br>> ...24 more
<br>> Caused by: java.security.NoSuchAlgorithmException: Algorithm [jks] of
<br>> type [KeyStore] from provider
<br>> [gnu.javax.security.auth.callback.GnuCallbacks: name=GNU-CALLBACKS
<br>> version=2.1] is not found
<br>> at gnu.java.security.Engine.getInstance(Engine.java:191)
<br>> at gnu.java.security.Engine.getInstance(Engine.java:105)
<br>> at java.security.KeyStore.getInstance(KeyStore.java:188)
<br>> ...27 more
<br>>
<br>>
<br>>
<br>>
<br>>
<br>>
</div><br>jks is not a crypto algorithm but a keystore format, and a proprietary
<br>Sun one until the release of OpenJDK.
<br>I guess in theory support could now be developed using the Sun source
<br>code as a reference, but there's not much work on Classpath in general
<br>at present.
<br><br><a href="http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34731" target="_top" rel="nofollow">http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34731</a><br>--
<br>Andrew :-)
<br><br>Free Java Software Engineer
<br>Red Hat, Inc. (<a href="http://www.redhat.com" target="_top" rel="nofollow">http://www.redhat.com</a>)
<br><br>Support Free Java!
<br>Contribute to GNU Classpath and the OpenJDK
<br><a href="http://www.gnu.org/software/classpath" target="_top" rel="nofollow">http://www.gnu.org/software/classpath</a><br><a href="http://openjdk.java.net" target="_top" rel="nofollow">http://openjdk.java.net</a><br><br>PGP Key: 94EFD9D8 (<a href="http://subkeys.pgp.net" target="_top" rel="nofollow">http://subkeys.pgp.net</a>)
<br>Fingerprint: F8EF F1EA 401E 2E60 15FA 7927 142C 2591 94EF D9D8
<br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Classpath---General-f1525.html" embed="fixTarget[1525]" target="_top" >Gnu - Classpath - General</a></p>tag:old.nabble.com,2006:post-26320883Classpath 0.98 & jks2009-11-12T07:40:13Z2009-11-12T07:40:13ZArigeadHello All,
<br> sorry new here and haven't had time to lurk for a while to see the
<br>lie of the land.
<br><br>I'm trying to run Jxta on Jamvm 1.5.3 with Classpath 0.98. I'm getting
<br>an error because of jks security not being found. I'll post the error
<br>I'm getting at the end of this email.
<br><br>A bit of searching on the Internet gives me a few posts by somebody with
<br>a similar problem [1] & [2]. I can't find any responses and solutions to
<br>this problem.
<br><br>I'm not sure how to get around this issue. I'm sure that there are many
<br>implementations of this security algorithm and I wonder can a lump of
<br>code be added included into the JRE or can source code be added to
<br>classpath. I mean there could I, if I had a clue ;-) add the necessary
<br>code into classpath source? I'd prefer if I could pull in something like
<br>BouncyCastle, although I'm not sure it would solve the problem as I'm
<br>not sure it implements this functionality.
<br><br>I'm not great with Java I'm afraid so I'm not sure that a BounceyCastle
<br>could be used to implement functionality that is being requested of the JRE.
<br><br>Sorry for a newbie post and thanks a million for any advice that anybody
<br>can offer.
<br><br><br>[1]
<br><a href="http://old.nabble.com/Problem-in-using-SSL-%28secure-socket%29-with-classpath-0.98-and-Jamvm-1.5.3-td23997596.html" target="_top" rel="nofollow">http://old.nabble.com/Problem-in-using-SSL-%28secure-socket%29-with-classpath-0.98-and-Jamvm-1.5.3-td23997596.html</a><br>[2] <a href="http://www.mail-archive.com/classpath@gnu.org/msg15120.html" target="_top" rel="nofollow">http://www.mail-archive.com/classpath@.../msg15120.html</a><br><br><br><br><br>Caused by: java.security.KeyStoreException: jks
<br> at java.security.KeyStore.getInstance(KeyStore.java:203)
<br> at java.security.KeyStore.getInstance(KeyStore.java:130)
<br> at
<br>net.jxta.impl.membership.pse.CMKeyStoreManager.<init>(CMKeyStoreManager.java:148)
<br> at
<br>net.jxta.impl.membership.pse.PSEKeyStoreManagerFactory$PSEKeyStoreManagerFactoryDefault.getInstance(PSEKeyStoreManagerFactory.java:143)
<br> ...24 more
<br>Caused by: java.security.NoSuchAlgorithmException: Algorithm [jks] of
<br>type [KeyStore] from provider
<br>[gnu.javax.security.auth.callback.GnuCallbacks: name=GNU-CALLBACKS
<br>version=2.1] is not found
<br> at gnu.java.security.Engine.getInstance(Engine.java:191)
<br> at gnu.java.security.Engine.getInstance(Engine.java:105)
<br> at java.security.KeyStore.getInstance(KeyStore.java:188)
<br> ...27 more
<br><br><br><br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Classpath---General-f1525.html" embed="fixTarget[1525]" target="_top" >Gnu - Classpath - General</a></p>tag:old.nabble.com,2006:post-26309608Bug in 1.1.2: Folder.search() alters Folder.mode (works with 1.1.1)2009-11-11T14:09:50Z2009-11-11T14:09:50ZBoris FolgmannHi,
<br><br>I've got a code that worked perfectly with mail 1.1.1 but has problems with
<br>1.1.2.
<br><br>----------------------------------------------------------------
<br>store.connect();
<br>Folder folder = store.getFolder("a cyrus imap folder");
<br>folder.open(Folder.READ_WRITE);
<br>log.info("Total messages: " + folder.getMessageCount());
<br><br>Message[] msgs = folder.search(aTerm);
<br>processMessages(msgs);
<br>folder.expunge();
<br>// folder.getMode() still returns Folder.READ_WRITE here
<br><br>msgs = folder.search(aDifferentTerm);
<br>// folder.getMode() now returns Folder.READ_ONLY!!
<br>processMessages(msgs);
<br>folder.expunge();
<br><br>----------------------------------------------------------------
<br><br>The second folder.expunge() throws
<br><br>javax.mail.MessagingException: Folder was opened read-only
<br> at gnu.mail.providers.imap.IMAPFolder.expunge(IMAPFolder.java:530)
<br><br>Using some more log.info() I found out that the Folder.mode is altered by
<br>the second search. I couldn't find anything in the JavaDocs that this is a
<br>regular behaviour. As it worked with the previous release, you may know
<br>what the problem could be.
<br><br>I also update jaf from 1.0 to 1.1.1
<br><br>cu,
<br> boris
<br><br><br><br><p>From forum: <a href="http://old.nabble.com/classpathx-javamail-f16003.html" embed="fixTarget[16003]" target="_top" >classpathx-javamail</a></p>tag:old.nabble.com,2006:post-26178892Questions about jvmti.h header2009-11-03T04:27:24Z2009-11-03T04:27:24ZMichael StarzingerHello!
<br><br>I am currently experimenting with JVMTI support in Cacao. My goal is
<br>to have the implementation compile equally against both GNU Classpath
<br>and OpenJDK headers. That's why I stumbled upon some questions about
<br>the JVMTI header included in the GNU Classpath source.
<br><br>* Are there any plans to install the jvmti.h file (just like it is
<br>done with jni.h) or are there circumstances preventing such an
<br>installation?
<br><br>* Why does the mentioned header include jvmti_md.h? All machine-
<br>dependent definitions should have already been covered by jni_md.h.
<br>Should the missing jvmti_md.h file be provided by the VM?
<br><br>I also stumbled upon a mailing list post[1] dealing with this topic.
<br>But it seems to me that only a small part[2] actually made it into the
<br>GNU Classpath source, the rest remained in the GCJ source.
<br><br>References:
<br>[1] <a href="http://gcc.gnu.org/ml/java-patches/2006-q3/msg00368.html" target="_top" rel="nofollow">http://gcc.gnu.org/ml/java-patches/2006-q3/msg00368.html</a><br>[2] <a href="http://developer.classpath.org/pipermail/classpath-patches/2006-August/004013.html" target="_top" rel="nofollow">http://developer.classpath.org/pipermail/classpath-patches/2006-August/004013.html</a><br><br>Thanks for your help,
<br>-michi
<br><br><br><br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Classpath---General-f1525.html" embed="fixTarget[1525]" target="_top" >Gnu - Classpath - General</a></p>tag:old.nabble.com,2006:post-26024960Re: Aliasing violations in gnu_java_nio_VMChannel.c2009-10-23T04:54:46Z2009-10-23T04:54:46ZAndrew HaleyAndrew John Hughes wrote:
<div class='shrinkable-quote'><br>> 2009/10/23 Andrew Haley <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26024960&i=0" target="_top" rel="nofollow">aph@...</a>>:
<br>>> Andrew John Hughes wrote:
<br>>>> 2009/10/22 Andrew Haley <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26024960&i=1" target="_top" rel="nofollow">aph@...</a>>:
<br>>>>> This:
<br>>>>>
<br>>>>>
<br>>>>> #ifdef HAVE_GETPEERNAME
<br>>>>> #ifdef HAVE_INET6
<br>>>>> struct sockaddr_in6 *addr6;
<br>>>>> struct sockaddr_in6 sock_storage;
<br>>>>> socklen_t socklen = sizeof (struct sockaddr_in6);
<br>>>>> #else
<br>>>>> struct sockaddr_in sock_storage;
<br>>>>> socklen_t socklen = sizeof (struct sockaddr_in);
<br>>>>> #endif /* HAVE_INET6 */
<br>>>>>
<br>>>>> struct sockaddr *sockaddr = (struct sockaddr *) &sock_storage;
<br>>>>>
<br>>>>>
<br>>>>> is clearly an aliasing violation: sock_storage is of type struct sockaddr_in6,
<br>>>>> and *sockaddr is of type struct sockaddr.
<br>>>>>
<br>>>>> It would be easy enough to fix the code with a union, but I can't test it
<br>>>>> because gcj doesn't use this code. I could simply compile with
<br>>>>> -fno-strict-aliasing.
<br>>>>>
<br>>>>> Thoughts?
<br>>>>>
<br>>>>> Andrew.
<br>>>>>
<br>>>>>
<br>>>> I posted a patch sometime ago to do exactly that (use a union), but
<br>>>> received no response:
<br>>>>
<br>>>> <a href="http://developer.classpath.org/pipermail/classpath-patches/2009-June/006341.html" target="_top" rel="nofollow">http://developer.classpath.org/pipermail/classpath-patches/2009-June/006341.html</a><br>>>>
<br>>>> It does need more testing than I've had chance to give it so far, I
<br>>>> just haven't had time to do any more work on it.
<br>>> I seriously wonder if it's worth the effort. I just did this:
<br>>>
<br>>> Index: configure.ac
<br>>> ===================================================================
<br>>> RCS file: /sources/classpath/classpath/configure.ac,v
<br>>> retrieving revision 1.244
<br>>> diff -c -u -r1.244 configure.ac
<br>>> --- configure.ac 6 Feb 2009 02:21:23 -0000 1.244
<br>>> +++ configure.ac 23 Oct 2009 09:10:33 -0000
<br>>> @@ -514,7 +514,7 @@
<br>>> dnl CFLAGS that are used for all native code. We want to compile
<br>>> dnl everything with unwinder data so that backtrace() will always
<br>>> dnl work.
<br>>> - EXTRA_CFLAGS='-fexceptions -fasynchronous-unwind-tables'
<br>>> + EXTRA_CFLAGS='-fexceptions -fasynchronous-unwind-tables -fno-strict-aliasing'
<br>>> AC_SUBST(EXTRA_CFLAGS)
<br>>>
<br>>> dnl Strict warning flags which not every module uses.
<br>>>
<br>>> which is a much safer change. It will somewhat pessimize other native code,
<br>>> that's true, but will that make any real difference to anyone these days?
<br>>
<br>> Why not just amend the CFLAGS in the local Makefile.am for cpnet?
</div><br>Sure. It won't make much difference, I suspect.
<br><br>Andrew.
<br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Classpath---General-f1525.html" embed="fixTarget[1525]" target="_top" >Gnu - Classpath - General</a></p>tag:old.nabble.com,2006:post-26024794Re: Aliasing violations in gnu_java_nio_VMChannel.c2009-10-23T04:42:01Z2009-10-23T04:42:01Zgnu_andrew2009/10/23 Andrew Haley <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26024794&i=0" target="_top" rel="nofollow">aph@...</a>>:
<div class='shrinkable-quote'><br>> Andrew John Hughes wrote:
<br>>> 2009/10/22 Andrew Haley <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26024794&i=1" target="_top" rel="nofollow">aph@...</a>>:
<br>>>> This:
<br>>>>
<br>>>>
<br>>>> #ifdef HAVE_GETPEERNAME
<br>>>> #ifdef HAVE_INET6
<br>>>> struct sockaddr_in6 *addr6;
<br>>>> struct sockaddr_in6 sock_storage;
<br>>>> socklen_t socklen = sizeof (struct sockaddr_in6);
<br>>>> #else
<br>>>> struct sockaddr_in sock_storage;
<br>>>> socklen_t socklen = sizeof (struct sockaddr_in);
<br>>>> #endif /* HAVE_INET6 */
<br>>>>
<br>>>> struct sockaddr *sockaddr = (struct sockaddr *) &sock_storage;
<br>>>>
<br>>>>
<br>>>> is clearly an aliasing violation: sock_storage is of type struct sockaddr_in6,
<br>>>> and *sockaddr is of type struct sockaddr.
<br>>>>
<br>>>> It would be easy enough to fix the code with a union, but I can't test it
<br>>>> because gcj doesn't use this code. I could simply compile with
<br>>>> -fno-strict-aliasing.
<br>>>>
<br>>>> Thoughts?
<br>>>>
<br>>>> Andrew.
<br>>>>
<br>>>>
<br>>>
<br>>> I posted a patch sometime ago to do exactly that (use a union), but
<br>>> received no response:
<br>>>
<br>>> <a href="http://developer.classpath.org/pipermail/classpath-patches/2009-June/006341.html" target="_top" rel="nofollow">http://developer.classpath.org/pipermail/classpath-patches/2009-June/006341.html</a><br>>>
<br>>> It does need more testing than I've had chance to give it so far, I
<br>>> just haven't had time to do any more work on it.
<br>>
<br>> I seriously wonder if it's worth the effort. I just did this:
<br>>
<br>> Index: configure.ac
<br>> ===================================================================
<br>> RCS file: /sources/classpath/classpath/configure.ac,v
<br>> retrieving revision 1.244
<br>> diff -c -u -r1.244 configure.ac
<br>> --- configure.ac 6 Feb 2009 02:21:23 -0000 1.244
<br>> +++ configure.ac 23 Oct 2009 09:10:33 -0000
<br>> @@ -514,7 +514,7 @@
<br>> dnl CFLAGS that are used for all native code. We want to compile
<br>> dnl everything with unwinder data so that backtrace() will always
<br>> dnl work.
<br>> - EXTRA_CFLAGS='-fexceptions -fasynchronous-unwind-tables'
<br>> + EXTRA_CFLAGS='-fexceptions -fasynchronous-unwind-tables -fno-strict-aliasing'
<br>> AC_SUBST(EXTRA_CFLAGS)
<br>>
<br>> dnl Strict warning flags which not every module uses.
<br>>
<br>> which is a much safer change. It will somewhat pessimize other native code,
<br>> that's true, but will that make any real difference to anyone these days?
<br>>
<br>> Andrew.
<br>>
</div><br>Why not just amend the CFLAGS in the local Makefile.am for cpnet?
<br>--
<br>Andrew :-)
<br><br>Free Java Software Engineer
<br>Red Hat, Inc. (<a href="http://www.redhat.com" target="_top" rel="nofollow">http://www.redhat.com</a>)
<br><br>Support Free Java!
<br>Contribute to GNU Classpath and the OpenJDK
<br><a href="http://www.gnu.org/software/classpath" target="_top" rel="nofollow">http://www.gnu.org/software/classpath</a><br><a href="http://openjdk.java.net" target="_top" rel="nofollow">http://openjdk.java.net</a><br><br>PGP Key: 94EFD9D8 (<a href="http://subkeys.pgp.net" target="_top" rel="nofollow">http://subkeys.pgp.net</a>)
<br>Fingerprint: F8EF F1EA 401E 2E60 15FA 7927 142C 2591 94EF D9D8
<br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Classpath---General-f1525.html" embed="fixTarget[1525]" target="_top" >Gnu - Classpath - General</a></p>tag:old.nabble.com,2006:post-26023108Re: Aliasing violations in gnu_java_nio_VMChannel.c2009-10-23T02:12:26Z2009-10-23T02:12:26ZAndrew HaleyAndrew John Hughes wrote:
<div class='shrinkable-quote'><br>> 2009/10/22 Andrew Haley <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26023108&i=0" target="_top" rel="nofollow">aph@...</a>>:
<br>>> This:
<br>>>
<br>>>
<br>>> #ifdef HAVE_GETPEERNAME
<br>>> #ifdef HAVE_INET6
<br>>> struct sockaddr_in6 *addr6;
<br>>> struct sockaddr_in6 sock_storage;
<br>>> socklen_t socklen = sizeof (struct sockaddr_in6);
<br>>> #else
<br>>> struct sockaddr_in sock_storage;
<br>>> socklen_t socklen = sizeof (struct sockaddr_in);
<br>>> #endif /* HAVE_INET6 */
<br>>>
<br>>> struct sockaddr *sockaddr = (struct sockaddr *) &sock_storage;
<br>>>
<br>>>
<br>>> is clearly an aliasing violation: sock_storage is of type struct sockaddr_in6,
<br>>> and *sockaddr is of type struct sockaddr.
<br>>>
<br>>> It would be easy enough to fix the code with a union, but I can't test it
<br>>> because gcj doesn't use this code. I could simply compile with
<br>>> -fno-strict-aliasing.
<br>>>
<br>>> Thoughts?
<br>>>
<br>>> Andrew.
<br>>>
<br>>>
<br>>
<br>> I posted a patch sometime ago to do exactly that (use a union), but
<br>> received no response:
<br>>
<br>> <a href="http://developer.classpath.org/pipermail/classpath-patches/2009-June/006341.html" target="_top" rel="nofollow">http://developer.classpath.org/pipermail/classpath-patches/2009-June/006341.html</a><br>>
<br>> It does need more testing than I've had chance to give it so far, I
<br>> just haven't had time to do any more work on it.
</div><br>I seriously wonder if it's worth the effort. I just did this:
<br><br>Index: configure.ac
<br>===================================================================
<br>RCS file: /sources/classpath/classpath/configure.ac,v
<br>retrieving revision 1.244
<br>diff -c -u -r1.244 configure.ac
<br>--- configure.ac 6 Feb 2009 02:21:23 -0000 1.244
<br>+++ configure.ac 23 Oct 2009 09:10:33 -0000
<br>@@ -514,7 +514,7 @@
<br> dnl CFLAGS that are used for all native code. We want to compile
<br> dnl everything with unwinder data so that backtrace() will always
<br> dnl work.
<br>- EXTRA_CFLAGS='-fexceptions -fasynchronous-unwind-tables'
<br>+ EXTRA_CFLAGS='-fexceptions -fasynchronous-unwind-tables -fno-strict-aliasing'
<br> AC_SUBST(EXTRA_CFLAGS)
<br><br> dnl Strict warning flags which not every module uses.
<br><br>which is a much safer change. It will somewhat pessimize other native code,
<br>that's true, but will that make any real difference to anyone these days?
<br><br>Andrew.
<br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Classpath---General-f1525.html" embed="fixTarget[1525]" target="_top" >Gnu - Classpath - General</a></p>tag:old.nabble.com,2006:post-26013285Re: Aliasing violations in gnu_java_nio_VMChannel.c2009-10-22T09:42:04Z2009-10-22T09:42:04Zgnu_andrew2009/10/22 Andrew Haley <<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26013285&i=0" target="_top" rel="nofollow">aph@...</a>>:
<div class='shrinkable-quote'><br>> This:
<br>>
<br>>
<br>> #ifdef HAVE_GETPEERNAME
<br>> #ifdef HAVE_INET6
<br>> struct sockaddr_in6 *addr6;
<br>> struct sockaddr_in6 sock_storage;
<br>> socklen_t socklen = sizeof (struct sockaddr_in6);
<br>> #else
<br>> struct sockaddr_in sock_storage;
<br>> socklen_t socklen = sizeof (struct sockaddr_in);
<br>> #endif /* HAVE_INET6 */
<br>>
<br>> struct sockaddr *sockaddr = (struct sockaddr *) &sock_storage;
<br>>
<br>>
<br>> is clearly an aliasing violation: sock_storage is of type struct sockaddr_in6,
<br>> and *sockaddr is of type struct sockaddr.
<br>>
<br>> It would be easy enough to fix the code with a union, but I can't test it
<br>> because gcj doesn't use this code. I could simply compile with
<br>> -fno-strict-aliasing.
<br>>
<br>> Thoughts?
<br>>
<br>> Andrew.
<br>>
<br>>
</div><br>I posted a patch sometime ago to do exactly that (use a union), but
<br>received no response:
<br><br><a href="http://developer.classpath.org/pipermail/classpath-patches/2009-June/006341.html" target="_top" rel="nofollow">http://developer.classpath.org/pipermail/classpath-patches/2009-June/006341.html</a><br><br>It does need more testing than I've had chance to give it so far, I
<br>just haven't had time to do any more work on it.
<br>--
<br>Andrew :-)
<br><br>Free Java Software Engineer
<br>Red Hat, Inc. (<a href="http://www.redhat.com" target="_top" rel="nofollow">http://www.redhat.com</a>)
<br><br>Support Free Java!
<br>Contribute to GNU Classpath and the OpenJDK
<br><a href="http://www.gnu.org/software/classpath" target="_top" rel="nofollow">http://www.gnu.org/software/classpath</a><br><a href="http://openjdk.java.net" target="_top" rel="nofollow">http://openjdk.java.net</a><br><br>PGP Key: 94EFD9D8 (<a href="http://subkeys.pgp.net" target="_top" rel="nofollow">http://subkeys.pgp.net</a>)
<br>Fingerprint: F8EF F1EA 401E 2E60 15FA 7927 142C 2591 94EF D9D8
<br><br><p>From forum: <a href="http://old.nabble.com/Gnu---Classpath---General-f1525.html" embed="fixTarget[1525]" target="_top" >Gnu - Classpath - General</a></p>