You are viewing a plain text version of this content. The canonical link for it is here.
Posted to derby-dev@db.apache.org by Andrew McIntyre <mc...@gmail.com> on 2006/01/28 00:59:57 UTC
Client SQLException change causing build failure
Hi David,
I think that your changes that went in with revision 371561 are
causing ClientPooledConnection40.java to fail to compile:
compile_jdbc4_target:
[echo] [javac]
C:\derby-trunk\java\client\org\apache\derby\client\ClientPooledConnection40.java:32:
unreported exception java.sql.SQLException; must be caught or declared
to be thrown
[echo] [javac] super(ds,logWriter,user,password);
[echo] [javac] ^
<snip other two errors>
[echo] [javac] 3 errors
[echo]
[echo] BUILD FAILED
Because ClientPooledConnection40 extends ClientPooledConnection, but
the latter now throws SQLExceptions instead of SqlExceptions. Take a
look if you have a moment.
Thanks,
andrew
Re: Client SQLException change causing build failure
Posted by "David W. Van Couvering" <Da...@Sun.COM>.
I just checked the test pages, and tests are running and (generally)
passing with the JDK 1.5 builds done by DBTG.
I did a test build on JDK 1.4 and all is OK. I did a test build on JDK
1.6 and indeed the build fails, but my failures are different, and
appear to be a poor configuration on my part (I have never tried this
before), as I am getting syntax errors when the compiler hits generic types.
Rick -- you also mentioned that JDK 1.6 builds were breaking, and you
were working on a fix. Is this the same as what Andrew's seeing? I
don't want to step on your toes if you're already fixing this.
David
Andrew McIntyre wrote:
> Hi David,
>
> I think that your changes that went in with revision 371561 are
> causing ClientPooledConnection40.java to fail to compile:
>
> compile_jdbc4_target:
> [echo] [javac]
> C:\derby-trunk\java\client\org\apache\derby\client\ClientPooledConnection40.java:32:
> unreported exception java.sql.SQLException; must be caught or declared
> to be thrown
> [echo] [javac] super(ds,logWriter,user,password);
> [echo] [javac] ^
>
> <snip other two errors>
>
> [echo] [javac] 3 errors
> [echo]
> [echo] BUILD FAILED
>
> Because ClientPooledConnection40 extends ClientPooledConnection, but
> the latter now throws SQLExceptions instead of SqlExceptions. Take a
> look if you have a moment.
>
> Thanks,
> andrew
Re: Client SQLException change causing build failure
Posted by Rick Hillegas <Ri...@Sun.COM>.
Sure. I will do that.
Cheers,
-Rick
David W. Van Couvering wrote:
> Great, Rick. Are you going to post this fix as a separate patch?
>
> David
>
> Rick Hillegas wrote:
>
>> What Andrew says. These failures turn up if you build Derby in a 1.4
>> environment but supplement your ant.properties with a pointer to a
>> 1.6 installation. In that case, the build will use the 1.6 compiler
>> to build the optional jdbc 4 support and compilation will choke on
>> some discrepancies introduced by the SQLException rototill.
>>
>> I fixed these build failures in my first patch for bug 862, but I
>> backed them out because they weren't germane to that bug's javadoc
>> focus. The fixes are still available in that first patch in the
>> following files:
>>
>> M java\client\org\apache\derby\client\ClientPooledConnection40.java
>> M java\client\org\apache\derby\client\net\NetConnection40.java
>> M java\client\org\apache\derby\client\am\CallableStatement40.java
>> M java\client\org\apache\derby\client\am\PreparedStatement40.java
>>
>> Regards,
>> -Rick
>>
>> Andrew McIntyre wrote:
>>
>>>
>>> On Jan 28, 2006, at 11:05 AM, David W. Van Couvering wrote:
>>>
>>>> Hi, Andrew. Are you doing your build with JDK 1.6? I did my
>>>> build with 1.4. I'm not trying to pass off blame, just trying to
>>>> understand the circumstances better.
>>>
>>>
>>>
>>>
>>> Hi David,
>>>
>>> I guess this is the 'mixed' build? It's everything but the JDBC 4.0
>>> classes compiled with JDK 1.4, and the JDBC 4.0 classes compiled
>>> with JDK 1.6. It's what you get if you follow the instructions
>>> posted previously to build the JDBC 4.0 classes as a part of the
>>> regular build by setting the ant property ${jdk16} to the location
>>> of a JDK 1.6 install.
>>>
>>> If you don't set the jdk16 ant property, everything works fine. I
>>> just thought I'd mention the problem because I hadn't seen it
>>> raised anywhere else.
>>>
>>> andrew
>>
>>
>>
>>
Re: Client SQLException change causing build failure
Posted by "David W. Van Couvering" <Da...@Sun.COM>.
Great, Rick. Are you going to post this fix as a separate patch?
David
Rick Hillegas wrote:
> What Andrew says. These failures turn up if you build Derby in a 1.4
> environment but supplement your ant.properties with a pointer to a 1.6
> installation. In that case, the build will use the 1.6 compiler to build
> the optional jdbc 4 support and compilation will choke on some
> discrepancies introduced by the SQLException rototill.
>
> I fixed these build failures in my first patch for bug 862, but I backed
> them out because they weren't germane to that bug's javadoc focus. The
> fixes are still available in that first patch in the following files:
>
> M java\client\org\apache\derby\client\ClientPooledConnection40.java
> M java\client\org\apache\derby\client\net\NetConnection40.java
> M java\client\org\apache\derby\client\am\CallableStatement40.java
> M java\client\org\apache\derby\client\am\PreparedStatement40.java
>
> Regards,
> -Rick
>
> Andrew McIntyre wrote:
>
>>
>> On Jan 28, 2006, at 11:05 AM, David W. Van Couvering wrote:
>>
>>> Hi, Andrew. Are you doing your build with JDK 1.6? I did my build
>>> with 1.4. I'm not trying to pass off blame, just trying to
>>> understand the circumstances better.
>>
>>
>>
>> Hi David,
>>
>> I guess this is the 'mixed' build? It's everything but the JDBC 4.0
>> classes compiled with JDK 1.4, and the JDBC 4.0 classes compiled with
>> JDK 1.6. It's what you get if you follow the instructions posted
>> previously to build the JDBC 4.0 classes as a part of the regular
>> build by setting the ant property ${jdk16} to the location of a JDK
>> 1.6 install.
>>
>> If you don't set the jdk16 ant property, everything works fine. I
>> just thought I'd mention the problem because I hadn't seen it raised
>> anywhere else.
>>
>> andrew
>
>
>
Re: Client SQLException change causing build failure
Posted by Rick Hillegas <Ri...@Sun.COM>.
What Andrew says. These failures turn up if you build Derby in a 1.4
environment but supplement your ant.properties with a pointer to a 1.6
installation. In that case, the build will use the 1.6 compiler to build
the optional jdbc 4 support and compilation will choke on some
discrepancies introduced by the SQLException rototill.
I fixed these build failures in my first patch for bug 862, but I backed
them out because they weren't germane to that bug's javadoc focus. The
fixes are still available in that first patch in the following files:
M java\client\org\apache\derby\client\ClientPooledConnection40.java
M java\client\org\apache\derby\client\net\NetConnection40.java
M java\client\org\apache\derby\client\am\CallableStatement40.java
M java\client\org\apache\derby\client\am\PreparedStatement40.java
Regards,
-Rick
Andrew McIntyre wrote:
>
> On Jan 28, 2006, at 11:05 AM, David W. Van Couvering wrote:
>
>> Hi, Andrew. Are you doing your build with JDK 1.6? I did my build
>> with 1.4. I'm not trying to pass off blame, just trying to
>> understand the circumstances better.
>
>
> Hi David,
>
> I guess this is the 'mixed' build? It's everything but the JDBC 4.0
> classes compiled with JDK 1.4, and the JDBC 4.0 classes compiled with
> JDK 1.6. It's what you get if you follow the instructions posted
> previously to build the JDBC 4.0 classes as a part of the regular
> build by setting the ant property ${jdk16} to the location of a JDK
> 1.6 install.
>
> If you don't set the jdk16 ant property, everything works fine. I
> just thought I'd mention the problem because I hadn't seen it raised
> anywhere else.
>
> andrew
Re: Client SQLException change causing build failure
Posted by Andrew McIntyre <mc...@gmail.com>.
On Jan 28, 2006, at 11:05 AM, David W. Van Couvering wrote:
> Hi, Andrew. Are you doing your build with JDK 1.6? I did my build
> with 1.4. I'm not trying to pass off blame, just trying to
> understand the circumstances better.
Hi David,
I guess this is the 'mixed' build? It's everything but the JDBC 4.0
classes compiled with JDK 1.4, and the JDBC 4.0 classes compiled with
JDK 1.6. It's what you get if you follow the instructions posted
previously to build the JDBC 4.0 classes as a part of the regular
build by setting the ant property ${jdk16} to the location of a JDK
1.6 install.
If you don't set the jdk16 ant property, everything works fine. I
just thought I'd mention the problem because I hadn't seen it raised
anywhere else.
andrew
Re: Client SQLException change causing build failure
Posted by "David W. Van Couvering" <Da...@Sun.COM>.
Hi, Andrew. Are you doing your build with JDK 1.6? I did my build with
1.4. I'm not trying to pass off blame, just trying to understand the
circumstances better.
Thanks,
David
Andrew McIntyre wrote:
> Hi David,
>
> I think that your changes that went in with revision 371561 are
> causing ClientPooledConnection40.java to fail to compile:
>
> compile_jdbc4_target:
> [echo] [javac]
> C:\derby-trunk\java\client\org\apache\derby\client\ClientPooledConnection40.java:32:
> unreported exception java.sql.SQLException; must be caught or declared
> to be thrown
> [echo] [javac] super(ds,logWriter,user,password);
> [echo] [javac] ^
>
> <snip other two errors>
>
> [echo] [javac] 3 errors
> [echo]
> [echo] BUILD FAILED
>
> Because ClientPooledConnection40 extends ClientPooledConnection, but
> the latter now throws SQLExceptions instead of SqlExceptions. Take a
> look if you have a moment.
>
> Thanks,
> andrew