From vogel@titze.de Wed Jul 5 07:00:00 2000 From: vogel@titze.de (Achim Vogel) Date: Wed, 05 Jul 2000 07:00:00 -0000 Subject: Mauve documentation Message-ID: <3963401A.2180F878@titze.de> Hello, we start up using mauve to test kaffe. We plan to write new tests if nesseccary. So my question: Is there any documentation for the test suite ? ( How to write mauve compatible tests ...) Best regards Achim Vogel From dalakakis@titze.de Wed Jul 5 07:09:00 2000 From: dalakakis@titze.de (Stavros Dalakakis) Date: Wed, 05 Jul 2000 07:09:00 -0000 Subject: Mauve-Docu? Message-ID: <39634142.7A690E3D@titze.de> Hi, I am looking for a mauve documentation? Does it exists one? Thanks in advance...! Grettings Stavros DALAKAKIS From tromey@cygnus.com Wed Jul 5 09:49:00 2000 From: tromey@cygnus.com (Tom Tromey) Date: Wed, 05 Jul 2000 09:49:00 -0000 Subject: Mauve documentation References: <3963401A.2180F878@titze.de> Message-ID: <200007051649.JAA20267@ferrule.cygnus.com> Achim> Is there any documentation for the test suite ? Achim> ( How to write mauve compatible tests ...) There is some documentation in the README file that explains how to do it. That might not be quite enough; I don't know. Feel free to ask questions. I will try to answer them as quickly as I can. Also, if you plan to write a lot of tests, I can arrange to get you an account to do checkins. There are some rules for doing this, but they are not very onerous. Tom From tromey@cygnus.com Wed Jul 5 09:51:00 2000 From: tromey@cygnus.com (Tom Tromey) Date: Wed, 05 Jul 2000 09:51:00 -0000 Subject: Mauve-Docu? References: <39634142.7A690E3D@titze.de> Message-ID: <200007051651.JAA20270@ferrule.cygnus.com> >>>>> "Stavros" == Stavros Dalakakis writes: Stavros> I am looking for a mauve documentation? Does it exists one? There is some documentation in the README file that explains how to write tests and how to run the test suite. That might not be quite enough; I don't know. Feel free to ask questions. I will try to answer them as quickly as I can. Tom From peter@erble.freeserve.co.uk Mon Jul 10 06:29:00 2000 From: peter@erble.freeserve.co.uk (Peter Naulls) Date: Mon, 10 Jul 2000 06:29:00 -0000 Subject: PipedStream/receive.java Message-ID: Is it just me, or does the file in the subject not compile: public class receive extends PipedInputStream implements Runnable, Testlet { static Thread main; static PipeTest in; static PipedOutputStream out; PipeTest (PipedOutputStream x) throws IOException { super(x); } . . . The constructor's name is different from the classname, so compilers barf. Peter -- ------------------------------------------------------------------------ Peter Naulls - peter@erble.freeserve.co.uk Homepage - http://free.prohosting.com/~chocky/ Java for RISC OS and ARM - http://free.prohosting.com/~chocky/java/ Debian Linux on RiscPCs - http://www.erble.freeserve.co.uk/ ------------------------------------------------------------------------ From tromey@cygnus.com Mon Jul 10 09:22:00 2000 From: tromey@cygnus.com (Tom Tromey) Date: Mon, 10 Jul 2000 09:22:00 -0000 Subject: PipedStream/receive.java References: Message-ID: <200007101622.JAA00326@ferrule.cygnus.com> >>>>> "Peter" == Peter Naulls writes: Peter> Is it just me, or does the file in the subject not compile: Thanks. I checked in a fix. Tom From a-xubinw@Exchange.Microsoft.com Tue Jul 11 14:25:00 2000 From: a-xubinw@Exchange.Microsoft.com (Xubin Wu (Volt Computer)) Date: Tue, 11 Jul 2000 14:25:00 -0000 Subject: Float.toString test Message-ID: Title: Float.toString test c:\mauve-snapshot-2000-06-17>echo gnu.testlet.java.lang.Float.FloatTest|java gnu.testlet.SimpleTestHarness gnu.testlet.java.lang.Float.FloatTest FAIL: gnu.testlet.java.lang.Float.FloatTest: Error: test_toString failed - 10 (number 1) 1 of 68 tests failed when I run FloatTest, one test case fails which is ???????????????????????????? ???????????????????????????? str = Float.toString( 0.001f ); ???????????????????????????? ???????????????????????????? harness.check(!( !Float.toString( 0.001f ).equals ("0.001" )), ???????????????????????????? ???????????????????????????? ???????????????????????????? "Error: test_toString failed - 10" ); it returns 0.0010, not 0.001. When I passed Float.toString( 0.01f ), it returns 0.01. Does anybody think it is a bug, or something else. Thanks. Sharon From peter@erble.freeserve.co.uk Tue Jul 11 15:46:00 2000 From: peter@erble.freeserve.co.uk (Peter Naulls) Date: Tue, 11 Jul 2000 15:46:00 -0000 Subject: Float.toString test References: Message-ID: In message < BE97ACF1A2EB464886E1DAF6C51BAC6E31E689@DF-SPOT.platinum.corp.microsoft.com > "Xubin Wu (Volt Computer)" wrote: > c:\mauve-snapshot-2000-06-17>echo > gnu.testlet.java.lang.Float.FloatTest|java gnu.testlet.SimpleTestHarness > gnu.testlet.java.lang.Float.FloatTest > FAIL: gnu.testlet.java.lang.Float.FloatTest: Error: test_toString failed > - 10 (number 1) > 1 of 68 tests failed > > when I run FloatTest, one test case fails which is > str = Float.toString( 0.001f ); > harness.check(!( !Float.toString( 0.001f ).equals > ("0.001" )), > "Error: test_toString failed - 10" ); > it returns 0.0010, not 0.001. > When I passed Float.toString( 0.01f ), it returns 0.01. > Does anybody think it is a bug, or something else. It looks like a bug. The 1.1 toString method in java.lang.Float suggest that values in that range should be in a ddd.ddd style: I'm not sure about Java 1.2, but I recall in 1.0.2 this was more lax - and that the behaviour of Sun's 1.0.2 JVM didn't quite match the API. Peter -- ------------------------------------------------------------------------ Peter Naulls - peter@erble.freeserve.co.uk Homepage - http://free.prohosting.com/~chocky/ Java for RISC OS and ARM - http://free.prohosting.com/~chocky/java/ Debian Linux on RiscPCs - http://www.erble.freeserve.co.uk/ ------------------------------------------------------------------------ From a-xubinw@Exchange.Microsoft.com Tue Jul 11 17:01:00 2000 From: a-xubinw@Exchange.Microsoft.com (Xubin Wu (Volt Computer)) Date: Tue, 11 Jul 2000 17:01:00 -0000 Subject: Float.toString test Message-ID: Title: RE: Float.toString test I rewrite the testcase for only test Float.toString() and Double.toString()method. And I found only if the input value is 0.001f or 0.001 (double), I get the error. But when the input is 0.01f or 0.01, the result is correct. Are the 0.001f or 0.001 magic numbers which only I can use to get the error ? see attach. Here is result for java1.2.2 on NT . ---------------------------------------------------------------???? C:\MAUVE_~1\java.lang.float>java ToString Float.toString(0.001f)= 0.0010 Error: test_toString failed - 3 Double.toString(0.001)= 0.0010 Error: test_toString failed - 4 Tests 2/4???????? fails -----Original Message----- From: Peter Naulls [ mailto:peter@erble.freeserve.co.uk ] Sent: Tuesday, July 11, 2000 3:46 PM Cc: mauve-discuss@sourceware.cygnus.com Subject: Re: Float.toString test In message < BE97ACF1A2EB464886E1DAF6C51BAC6E31E689@DF-SPOT.platinum.corp.microsoft.com > ???????????????????????????????????? "Xubin Wu (Volt Computer)" wrote: > c:\mauve-snapshot-2000-06-17>echo > gnu.testlet.java.lang.Float.FloatTest|java gnu.testlet.SimpleTestHarness > gnu.testlet.java.lang.Float.FloatTest > FAIL: gnu.testlet.java.lang.Float.FloatTest: Error: test_toString failed > - 10 (number 1) > 1 of 68 tests failed > > when I run FloatTest, one test case fails which is > ???????????????????? ???????????????????????????? str = Float.toString( 0.001f ); > ???????????????????? ???????????????????????????? harness.check(!( !Float.toString( 0.001f ).equals > ("0.001" )), > ???????????????????? ???????????????????????????? ???????????????????????????? "Error: test_toString failed - 10" ); > it returns 0.0010, not 0.001. > When I passed Float.toString( 0.01f ), it returns 0.01. > Does anybody think it is a bug, or something else. It looks like a bug.???? The 1.1 toString method in java.lang.Float suggest that values in that range should be in a ddd.ddd style: I'm not sure about Java 1.2, but I recall in 1.0.2 this was more lax - and that the behaviour of Sun's 1.0.2 JVM didn't quite match the API. Peter -- ------------------------------------------------------------------------ ????Peter Naulls - peter@erble.freeserve.co.uk ????Homepage - http://free.prohosting.com/~chocky/ ????Java for RISC OS and ARM - http://free.prohosting.com/~chocky/java/ ????Debian Linux on RiscPCs - http://www.erble.freeserve.co.uk/ ------------------------------------------------------------------------ ToString.java -------------- next part -------------- A non-text attachment was scrubbed... Name: ToString.java Type: text/x-c Size: 1216 bytes Desc: not available URL: From lotus_june@yahoo.com Wed Jul 12 18:33:00 2000 From: lotus_june@yahoo.com (sharon wu) Date: Wed, 12 Jul 2000 18:33:00 -0000 Subject: Class.getInterfaces() on [[I Message-ID: <20000713013251.3722.qmail@web5403.mail.yahoo.com> One of Mauve Test cases. \java\lang\Class\ClassTest.java: public void test_getInterfaces() { Class clsss = Class.forName("[[I"); harness.check ( clsss.getInterfaces().length, 1 ); Class clsss = Class.forName("[D"); harness.check ( clsss.getInterfaces().length, 1 ); } These codes want to check getInterfaces() for [[I and [D, and retrun interfaces should be one. I wonder what interface of array of array interger or array of Double should be return? Does anyone has any ideas? Thanks. Sharon __________________________________________________ Do You Yahoo!? Get Yahoo! Mail ???? Free email you can access from anywhere! http://mail.yahoo.com/ From peter@erble.freeserve.co.uk Thu Jul 13 04:57:00 2000 From: peter@erble.freeserve.co.uk (Peter Naulls) Date: Thu, 13 Jul 2000 04:57:00 -0000 Subject: PipedStream/receive.java References: Message-ID: > Peter> Is it just me, or does the file in the subject not compile: > > > Thanks. > I checked in a fix. It compiles, but doesn't run :-) The newInstance method which is called from SimpleTestHarness requires a constructor with no argumnets, which this class doesn't have. Peter -- ------------------------------------------------------------------------ Peter Naulls - peter@erble.freeserve.co.uk Homepage - http://free.prohosting.com/~chocky/ Java for RISC OS and ARM - http://free.prohosting.com/~chocky/java/ Debian Linux on RiscPCs - http://www.erble.freeserve.co.uk/ ------------------------------------------------------------------------ From peter@erble.freeserve.co.uk Thu Jul 13 05:05:00 2000 From: peter@erble.freeserve.co.uk (Peter Naulls) Date: Thu, 13 Jul 2000 05:05:00 -0000 Subject: Class.getInterfaces() on [[I References: <20000713013251.3722.qmail@web5403.mail.yahoo.com> Message-ID: In message < 20000713013251.3722.qmail@web5403.mail.yahoo.com > sharon wu wrote: > > Class clsss = Class.forName("[D"); > harness.check ( clsss.getInterfaces().length, 1 ); > } > These codes want to check getInterfaces() for [[I and > [D, and retrun interfaces should be one. I wonder what > interface of array of array interger or array of > Double should be return? > Does anyone has any ideas? Also 1. Arrays, regardless of dimension only implement 1 interface, java.lang.Cloneable. Peter -- ------------------------------------------------------------------------ Peter Naulls - peter@erble.freeserve.co.uk Homepage - http://free.prohosting.com/~chocky/ Java for RISC OS and ARM - http://free.prohosting.com/~chocky/java/ Debian Linux on RiscPCs - http://www.erble.freeserve.co.uk/ ------------------------------------------------------------------------ From tromey@cygnus.com Thu Jul 13 08:50:00 2000 From: tromey@cygnus.com (Tom Tromey) Date: Thu, 13 Jul 2000 08:50:00 -0000 Subject: Class.getInterfaces() on [[I References: <20000713013251.3722.qmail@web5403.mail.yahoo.com> Message-ID: <200007131550.IAA16282@ferrule.cygnus.com> Peter> Also 1. Arrays, regardless of dimension only implement 1 interface, Peter> java.lang.Cloneable. That's true for jdk 1.0 (which is what this test tests). My understanding is that for 1.1 arrays also implement Serializable. Tom From lotus_june@yahoo.com Thu Jul 13 18:32:00 2000 From: lotus_june@yahoo.com (sharon wu) Date: Thu, 13 Jul 2000 18:32:00 -0000 Subject: ClassTest on class.getClassLoader() Message-ID: <20000714013217.19315.qmail@web5401.mail.yahoo.com> One of Mauve Test cases. \java\lang\Class\ClassTest.java: public void test_getClassloader() { try { Class obj1 = Class.forName("java.lang.String"); ClassLoader ldr = obj1.getClassLoader(); if ( ldr != null ) harness.fail("Error: test_getClassLoader failed - 1"); Class obj2 = Class.forName("ClassTest"); ClassLoader ldr1 = obj2.getClassLoader(); if ( ldr1 != null ) harness.fail("Error: test_getClassLoader failed on ClassTest- 2"); } catch ( Exception e ){ harness.fail("Error: test_getClassLoader failed - 2"); } } I test on jdk1.1.4 I get the result Class name = java.lang.String Classloader = null Class name = ClassTest Classloader = null But when I test on jdk1.2.2 I get the different result Class name = java.lang.String Classloader = null Class name = ClassTest_test_getClassloader Classloader = sun.misc.Launcher$AppClassLoader@e6775003 Error: test_getClassLoader failed - 2 IT seems to me the customer class and system class should both return null. Does somebody know why i get the different result ? which one is correct? Thanks __________________________________________________ Do You Yahoo!? Get Yahoo! Mail ???? Free email you can access from anywhere! http://mail.yahoo.com/ From tromey@cygnus.com Fri Jul 14 00:10:00 2000 From: tromey@cygnus.com (Tom Tromey) Date: Fri, 14 Jul 2000 00:10:00 -0000 Subject: ClassTest on class.getClassLoader() References: <20000714013217.19315.qmail@web5401.mail.yahoo.com> Message-ID: <200007140710.AAA17308@ferrule.cygnus.com> Sharon> But when I test on jdk1.2.2 Sharon> I get the different result Sharon> Class name = java.lang.String Sharon> Classloader = null Sharon> Class name = ClassTest_test_getClassloader Sharon> Classloader = Sharon> sun.misc.Launcher$AppClassLoader@e6775003 Sharon> Error: test_getClassLoader failed - 2 Sharon> IT seems to me the customer class and system class Sharon> should both return null. Does somebody know why i get Sharon> the different result ? which one is correct? "Correct", unfortunately, changes over time. This looks like a change in the JCL. The online 1.2 docs say this: Some implementations may use null to represent the bootstrap class loader. This method will return null in such implementations if this class was loaded by the bootstrap class loader. To me this clearly implies that a non-null result is also acceptable. This is why we have tags in Mauve -- if your implementation doesn't conform to "JLS1.0" (which is how this test is tagged), then this tag should be undefined so these tests won't be run. It would be nice if Mauve's configure script detected the correct tags for a given jdk installation. Someone other than me would have to write the patch, though. Tom From a-xubinw@Exchange.Microsoft.com Fri Jul 14 09:41:00 2000 From: a-xubinw@Exchange.Microsoft.com (Xubin Wu (Volt Computer)) Date: Fri, 14 Jul 2000 09:41:00 -0000 Subject: Class.getInterfaces() on [[I Message-ID: Title: RE: Class.getInterfaces() on [[I For java1.2.2 it retrun two interfaces, Serializable and Cloneable. For JDK1.1.4, only return the empty list of array. -----Original Message----- From: Tom Tromey [ mailto:tromey@cygnus.com ] Sent: Thursday, July 13, 2000 8:50 AM To: Peter Naulls Cc: mauve-discuss@sourceware.cygnus.com Subject: Re: Class.getInterfaces() on [[I Peter> Also 1.???? Arrays, regardless of dimension only implement 1 interface, Peter> java.lang.Cloneable. That's true for jdk 1.0 (which is what this test tests). My understanding is that for 1.1 arrays also implement Serializable. Tom From lotus_june@yahoo.com Fri Jul 14 13:20:00 2000 From: lotus_june@yahoo.com (sharon wu) Date: Fri, 14 Jul 2000 13:20:00 -0000 Subject: ClassTest on class.getClassLoader() Message-ID: <20000714195723.18934.qmail@web5403.mail.yahoo.com> Another thing I notice is that in test_getClassloader //Class obj2 = Class.forName("ClassTest"); Class obj2 = Class.forName("gnu.testlet.java.lang.Class.ClassTest"); I wonder if you need to put all package name here. For me, I need to put all package name here even I set the classpath. And in JLS, it claims that for Class.forName(), you need put fully qualified name. Thanks. --- Tom Tromey wrote: > Sharon> But when I test on jdk1.2.2 > Sharon> I get the different result > Sharon> Class name = java.lang.String > Sharon> Classloader = null > Sharon> Class name = ClassTest_test_getClassloader > Sharon> Classloader = > Sharon> sun.misc.Launcher$AppClassLoader@e6775003 > Sharon> Error: test_getClassLoader failed - 2 > > Sharon> IT seems to me the customer class and system > class > Sharon> should both return null. Does somebody know > why i get > Sharon> the different result ? which one is correct? > > "Correct", unfortunately, changes over time. > > This looks like a change in the JCL. The online 1.2 > docs say this: > > Some implementations may use null to represent > the bootstrap class > loader. This method will return null in such > implementations if this > class was loaded by the bootstrap class loader. > > To me this clearly implies that a non-null result is > also acceptable. > > This is why we have tags in Mauve -- if your > implementation doesn't > conform to "JLS1.0" (which is how this test is > tagged), then this tag > should be undefined so these tests won't be run. > > It would be nice if Mauve's configure script > detected the correct tags > for a given jdk installation. Someone other than me > would have to > write the patch, though. > > Tom __________________________________________________ Do You Yahoo!? Get Yahoo! Mail ???? Free email you can access from anywhere! http://mail.yahoo.com/ From tromey@cygnus.com Fri Jul 14 15:42:00 2000 From: tromey@cygnus.com (Tom Tromey) Date: Fri, 14 Jul 2000 15:42:00 -0000 Subject: ClassTest on class.getClassLoader() References: <20000714195723.18934.qmail@web5403.mail.yahoo.com> Message-ID: <200007142242.PAA18221@ferrule.cygnus.com> sharon> Another thing I notice is that in test_getClassloader sharon> //Class obj2 = Class.forName("ClassTest"); sharon> Class obj2 = sharon> Class.forName("gnu.testlet.java.lang.Class.ClassTest"); sharon> I wonder if you need to put all package name here. For sharon> me, I need to put all package name here even I set the sharon> classpath. And in JLS, it claims that for sharon> Class.forName(), you need put fully qualified name. Sounds right to me. I'm committing the change. Tom From lotus_june@yahoo.com Fri Jul 14 17:57:00 2000 From: lotus_june@yahoo.com (sharon wu) Date: Fri, 14 Jul 2000 17:57:00 -0000 Subject: ClassTest on class.getClassLoader() Message-ID: <20000715005723.25677.qmail@web5405.mail.yahoo.com> java\text\BreakIterator\Worditor.java BreakIterator bi = BreakIterator.getWordInstance (loc); Does somebody know where I get the documentation talking about what is wordbreak, what is sentence break, etc"? so I can configure the test out. Thanks. __________________________________________________ Do You Yahoo!? Get Yahoo! Mail ???? Free email you can access from anywhere! http://mail.yahoo.com/ From lotus_june@yahoo.com Fri Jul 14 18:03:00 2000 From: lotus_june@yahoo.com (sharon wu) Date: Fri, 14 Jul 2000 18:03:00 -0000 Subject: java.text.BreakIterator.getWordInstance() Message-ID: <20000715010320.13514.qmail@web5404.mail.yahoo.com> java\text\BreakIterator\Worditor.java BreakIterator bi = BreakIterator.getWordInstance (loc); Does somebody know where I get the documentation talking about what is wordbreak, what is sentence break, etc"? so I can configure the test out. Thanks. __________________________________________________ Do You Yahoo!? Get Yahoo! Mail ???? Free email you can access from anywhere! http://mail.yahoo.com/ From tromey@cygnus.com Wed Jul 19 10:02:00 2000 From: tromey@cygnus.com (Tom Tromey) Date: Wed, 19 Jul 2000 10:02:00 -0000 Subject: ClassTest on class.getClassLoader() References: <20000715005723.25677.qmail@web5405.mail.yahoo.com> Message-ID: <200007191702.KAA22037@ferrule.cygnus.com> >>>>> "Sharon" == sharon wu writes: Sharon> Does somebody know where I get the documentation Sharon> talking about what is wordbreak, what is sentence Sharon> break, etc"? so I can configure the test out. I think these things are locale-specific and not very well defined. In my implementation I wrote the default implementations based on definitions in the Unicode book. This reference was augmented in some places by hints in the Java Class Libraries book. Tom From NGurevich@NETsilicon.com Wed Jul 26 08:03:00 2000 From: NGurevich@NETsilicon.com (Gurevich, Naum) Date: Wed, 26 Jul 2000 08:03:00 -0000 Subject: Source Contribution Message-ID: <3FEE3089984DD211ABCC00A0C9D346D0016A2B56@postoffice.netsilicon.com> My name is Naum Gurevich. I am a Principal Software engineer with Net Silicon. We are using Mauve test. In process we discover some missing tests or what seems to be a bug. I will send my source code and comments in case that somebody want to use them. <> -------------- next part -------------- A non-text attachment was scrubbed... Name: ArrayTest.java Type: text/x-java Size: 1744 bytes Desc: not available URL: From tromey@cygnus.com Thu Jul 27 18:59:00 2000 From: tromey@cygnus.com (Tom Tromey) Date: Thu, 27 Jul 2000 18:59:00 -0000 Subject: Source Contribution References: <3FEE3089984DD211ABCC00A0C9D346D0016A2B56@postoffice.netsilicon.com> Message-ID: <200007280159.SAA09754@ferrule.cygnus.com> >>>>> "Naum" == Gurevich, Naum writes: Naum> My name is Naum Gurevich. I am a Principal Software engineer Naum> with Net Silicon. We are using Mauve test. In process we Naum> discover some missing tests or what seems to be a bug. I will Naum> send my source code and comments in case that somebody want to Naum> use them. Thanks. This test looks like it tests general runtime stuff and not any particular class. Right now we don't have a framework for doing that in Mauve. I think we ought to add one. Mostly that just means picking a new name for the test. Any suggestions? Tom From peter@erble.freeserve.co.uk Fri Jul 28 02:06:00 2000 From: peter@erble.freeserve.co.uk (Peter Naulls) Date: Fri, 28 Jul 2000 02:06:00 -0000 Subject: Source Contribution References: <200007280159.SAA09754@ferrule.cygnus.com> Message-ID: In message < 200007280159.SAA09754@ferrule.cygnus.com > Tom Tromey wrote: > Thanks. > This test looks like it tests general runtime stuff and not any > particular class. Right now we don't have a framework for doing that > in Mauve. I think we ought to add one. Mostly that just means > picking a new name for the test. Any suggestions? It might be nice to add them under a Virtual Machine section - I'm sure there are plenty of other tests like this. -- ------------------------------------------------------------------------ Peter Naulls - peter@erble.freeserve.co.uk Homepage - http://free.prohosting.com/~chocky/ Java for RISC OS and ARM - http://free.prohosting.com/~chocky/java/ Debian Linux on RiscPCs - http://www.erble.freeserve.co.uk/ ------------------------------------------------------------------------ From NGurevich@NETsilicon.com Fri Jul 28 05:38:00 2000 From: NGurevich@NETsilicon.com (Gurevich, Naum) Date: Fri, 28 Jul 2000 05:38:00 -0000 Subject: Source Contribution Message-ID: <3FEE3089984DD211ABCC00A0C9D346D0016A2B6A@postoffice.netsilicon.com> Anything is fine with me. One more problem with Mauve test Section 10.7 of Java Language Specification explicitly specifies that array implements 2 interfaces, not 1 as assumed in java.lang.Class.ClassTest. The following is a corrected code. I made an additional changes to test_getClassloader: Provide fully qualified name for "gnu.testlet.java.lang.Class.ClassTest". Because this class must be loaded variable ldr1 must be not null. Naum -----Original Message----- From: Tom Tromey [ mailto:tromey@cygnus.com ] Sent: Thursday, July 27, 2000 10:00 PM To: Gurevich, Naum Cc: 'mauve-discuss@sourceware.cygnus.com' Subject: Re: Source Contribution >>>>> "Naum" == Gurevich, Naum writes: Naum> My name is Naum Gurevich. I am a Principal Software engineer Naum> with Net Silicon. We are using Mauve test. In process we Naum> discover some missing tests or what seems to be a bug. I will Naum> send my source code and comments in case that somebody want to Naum> use them. Thanks. This test looks like it tests general runtime stuff and not any particular class. Right now we don't have a framework for doing that in Mauve. I think we ought to add one. Mostly that just means picking a new name for the test. Any suggestions? Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: ClassTest.java Type: text/x-java Size: 10489 bytes Desc: not available URL: From tromey@cygnus.com Fri Jul 28 09:17:00 2000 From: tromey@cygnus.com (Tom Tromey) Date: Fri, 28 Jul 2000 09:17:00 -0000 Subject: Source Contribution References: <3FEE3089984DD211ABCC00A0C9D346D0016A2B6A@postoffice.netsilicon.com> Message-ID: <200007281616.JAA06862@ferrule.cygnus.com> >>>>> "Naum" == Gurevich, Naum writes: Naum> Anything is fine with me. One more problem with Mauve test Naum> Section 10.7 of Java Language Specification explicitly specifies Naum> that array implements 2 interfaces, not 1 as assumed in Naum> java.lang.Class.ClassTest. ClassTest is tagged "JLS1.0". In 1.0 array only implemented a single interface. 1.1 or 1.2 implementations probably shouldn't run this test. That is easy to set up using a tag file. Tom From tromey@cygnus.com Sat Aug 5 16:23:00 2000 From: tromey@cygnus.com (Tom Tromey) Date: Sat, 05 Aug 2000 16:23:00 -0000 Subject: PipedStream/receive.java References: Message-ID: <200008052322.QAA03401@ferrule.cygnus.com> Peter> It compiles, but doesn't run :-) The newInstance method which Peter> is called from SimpleTestHarness requires a constructor with no Peter> argumnets, which this class doesn't have. Thanks. I finally got around to doing this. I haven't had any time to look at Mauve in quite a while... Tom