Postgres Pro
Downloads
Home   >   mailing lists

Thread: Buildfarm error: Broken compatability with JDK 1.4

Buildfarm error: Broken compatability with JDK 1.4

From
"Mike Fowler"
Date:
Hi,

The JDBC buildfarm has been broken since Friday as I've discovered this
morning (see
http://www.pgbuildfarm.org/cgi-bin/show_status.pl?member=piapiac). All
branches (8.4,9.0,9.1 & HEAD) are failing with the same reason:

compile:
    [javac] /home/ubuntu/buildarea/HEAD/pgjdbc/build.xml:125: warning:
'includeantruntime' was not set, defaulting to
build.sysclasspath=last; set to false for repeatable builds
    [javac] Compiling 152 source files to
/home/ubuntu/buildarea/HEAD/pgjdbc/build
    [javac]
/home/ubuntu/buildarea/HEAD/pgjdbc/org/postgresql/ds/common/PGObjectFactory.java:119:
cannot resolve symbol
    [javac] symbol  : method parseBoolean (java.lang.String)
    [javac] location: class java.lang.Boolean
    [javac]
ds.setBinaryTransfer(Boolean.parseBoolean(binaryTransfer));
    [javac]                                         ^
    [javac]
/home/ubuntu/buildarea/HEAD/pgjdbc/org/postgresql/jdbc2/AbstractJdbc2ResultSet.java:3129:
cannot resolve symbol
    [javac] symbol  : method valueOf (long)
    [javac] location: class java.lang.Long
    [javac]             throw new PSQLException(GT.tr("Bad value for type
{0} : {1}", new Object[]{targetType, Long.valueOf(val)}),
    [javac]
                                        ^
    [javac] Note: Some input files use or override a deprecated API.
    [javac] Note: Recompile with -deprecation for details.
    [javac] 2 errors

Both of these method calls - Boolean.parseBoolean() and Long.valueOf() -
are introduced in JDK 1.5. Are we dropping support for 1.4? Given that the
workaround to maintain 1.4 compatability is so simple (see attached patch)
I would recommend maintaing support of 1.4.

Regards,

--
Mike Fowler
Registered Linux user: 379787

The Call for Papers for is open for FLOSS UK, contact postgresql2012 AT
flossuk DOT org for details.

Re: Buildfarm error: Broken compatability with JDK 1.4

From
Dave Cramer
Date:
On Mon, Sep 26, 2011 at 9:17 AM, Mike Fowler <mike@mlfowler.com> wrote:
> Hi,
>
> The JDBC buildfarm has been broken since Friday as I've discovered this
> morning (see
> http://www.pgbuildfarm.org/cgi-bin/show_status.pl?member=piapiac). All
> branches (8.4,9.0,9.1 & HEAD) are failing with the same reason:
>
> compile:
>    [javac] /home/ubuntu/buildarea/HEAD/pgjdbc/build.xml:125: warning:
> 'includeantruntime' was not set, defaulting to
> build.sysclasspath=last; set to false for repeatable builds
>    [javac] Compiling 152 source files to
> /home/ubuntu/buildarea/HEAD/pgjdbc/build
>    [javac]
> /home/ubuntu/buildarea/HEAD/pgjdbc/org/postgresql/ds/common/PGObjectFactory.java:119:
> cannot resolve symbol
>    [javac] symbol  : method parseBoolean (java.lang.String)
>    [javac] location: class java.lang.Boolean
>    [javac]
> ds.setBinaryTransfer(Boolean.parseBoolean(binaryTransfer));
>    [javac]                                         ^
>    [javac]
> /home/ubuntu/buildarea/HEAD/pgjdbc/org/postgresql/jdbc2/AbstractJdbc2ResultSet.java:3129:
> cannot resolve symbol
>    [javac] symbol  : method valueOf (long)
>    [javac] location: class java.lang.Long
>    [javac]             throw new PSQLException(GT.tr("Bad value for type
> {0} : {1}", new Object[]{targetType, Long.valueOf(val)}),
>    [javac]
>                                        ^
>    [javac] Note: Some input files use or override a deprecated API.
>    [javac] Note: Recompile with -deprecation for details.
>    [javac] 2 errors
>
> Both of these method calls - Boolean.parseBoolean() and Long.valueOf() -
> are introduced in JDK 1.5. Are we dropping support for 1.4? Given that the
> workaround to maintain 1.4 compatability is so simple (see attached patch)
> I would recommend maintaing support of 1.4.


I would agree that we want to keep support for 1.4

Miiko, I'm guessing this is mostly from your patches.

Dave

Re: Buildfarm error: Broken compatability with JDK 1.4

From
Dave Cramer
Date:
Did I miss something ? I did not see a patch from Mike.

Dave Cramer

dave.cramer(at)credativ(dot)ca
http://www.credativ.ca




On Mon, Sep 26, 2011 at 9:46 AM, Mikko Tiihonen
<mikko.tiihonen@nitorcreations.com> wrote:
> On 09/26/2011 04:41 PM, Dave Cramer wrote:
>>
>> On Mon, Sep 26, 2011 at 9:17 AM, Mike Fowler<mike@mlfowler.com>  wrote:
>>>
>>> Hi,
>>>
>>> The JDBC buildfarm has been broken since Friday as I've discovered this
>>> morning (see
>>> http://www.pgbuildfarm.org/cgi-bin/show_status.pl?member=piapiac). All
>>> branches (8.4,9.0,9.1&  HEAD) are failing with the same reason:
>>>
>>> compile:
>>>    [javac] /home/ubuntu/buildarea/HEAD/pgjdbc/build.xml:125: warning:
>>> 'includeantruntime' was not set, defaulting to
>>> build.sysclasspath=last; set to false for repeatable builds
>>>    [javac] Compiling 152 source files to
>>> /home/ubuntu/buildarea/HEAD/pgjdbc/build
>>>    [javac]
>>>
>>> /home/ubuntu/buildarea/HEAD/pgjdbc/org/postgresql/ds/common/PGObjectFactory.java:119:
>>> cannot resolve symbol
>>>    [javac] symbol  : method parseBoolean (java.lang.String)
>>>    [javac] location: class java.lang.Boolean
>>>    [javac]
>>> ds.setBinaryTransfer(Boolean.parseBoolean(binaryTransfer));
>>>    [javac]                                         ^
>>>    [javac]
>>>
>>> /home/ubuntu/buildarea/HEAD/pgjdbc/org/postgresql/jdbc2/AbstractJdbc2ResultSet.java:3129:
>>> cannot resolve symbol
>>>    [javac] symbol  : method valueOf (long)
>>>    [javac] location: class java.lang.Long
>>>    [javac]             throw new PSQLException(GT.tr("Bad value for type
>>> {0} : {1}", new Object[]{targetType, Long.valueOf(val)}),
>>>    [javac]
>>>                                        ^
>>>    [javac] Note: Some input files use or override a deprecated API.
>>>    [javac] Note: Recompile with -deprecation for details.
>>>    [javac] 2 errors
>>>
>>> Both of these method calls - Boolean.parseBoolean() and Long.valueOf() -
>>> are introduced in JDK 1.5. Are we dropping support for 1.4? Given that
>>> the
>>> workaround to maintain 1.4 compatability is so simple (see attached
>>> patch)
>>> I would recommend maintaing support of 1.4.
>>
>>
>> I would agree that we want to keep support for 1.4
>>
>> Mikko, I'm guessing this is mostly from your patches.
>
> Yes, my fault. I have tried to stay away from new APIs - the breakage was
> unintentional.
>
> I'll be more careful in the future, but I do not have any older than 1.6 JVM
> available
> so it is very useful to have the build farm to catch these errors.
>
> If the patch from Mike does not get committed I'll create one tomorrow.
>
> -Mikko
>

Re: Buildfarm error: Broken compatability with JDK 1.4

From
"Mike Fowler"
Date:
> Did I miss something ? I did not see a patch from Mike.
>
> Dave Cramer
>
> dave.cramer(at)credativ(dot)ca
> http://www.credativ.ca

It might help if I attached the patch, apologies!


--
Mike Fowler
Registered Linux user: 379787

Re: Buildfarm error: Broken compatability with JDK 1.4

From
"Mike Fowler"
Date:
>
>> Did I miss something ? I did not see a patch from Mike.
>>
>> Dave Cramer
>>
>> dave.cramer(at)credativ(dot)ca
>> http://www.credativ.ca
>
> It might help if I attached the patch, apologies!
>


How embarrassing. I'm using an uncopertive web client. The attach button
is right next to the send button and I missed. The patch really is
attached this time :-/

--
Mike Fowler
Registered Linux user: 379787

The Call for Papers for is open for FLOSS UK, contact postgresql2012 AT
flossuk DOT org for details.
Attachment

Re: Buildfarm error: Broken compatability with JDK 1.4

From
Mikko Tiihonen
Date:
On 09/26/2011 04:41 PM, Dave Cramer wrote:
> On Mon, Sep 26, 2011 at 9:17 AM, Mike Fowler<mike@mlfowler.com>  wrote:
>> Hi,
>>
>> The JDBC buildfarm has been broken since Friday as I've discovered this
>> morning (see
>> http://www.pgbuildfarm.org/cgi-bin/show_status.pl?member=piapiac). All
>> branches (8.4,9.0,9.1&  HEAD) are failing with the same reason:
>>
>> compile:
>>     [javac] /home/ubuntu/buildarea/HEAD/pgjdbc/build.xml:125: warning:
>> 'includeantruntime' was not set, defaulting to
>> build.sysclasspath=last; set to false for repeatable builds
>>     [javac] Compiling 152 source files to
>> /home/ubuntu/buildarea/HEAD/pgjdbc/build
>>     [javac]
>> /home/ubuntu/buildarea/HEAD/pgjdbc/org/postgresql/ds/common/PGObjectFactory.java:119:
>> cannot resolve symbol
>>     [javac] symbol  : method parseBoolean (java.lang.String)
>>     [javac] location: class java.lang.Boolean
>>     [javac]
>> ds.setBinaryTransfer(Boolean.parseBoolean(binaryTransfer));
>>     [javac]                                         ^
>>     [javac]
>> /home/ubuntu/buildarea/HEAD/pgjdbc/org/postgresql/jdbc2/AbstractJdbc2ResultSet.java:3129:
>> cannot resolve symbol
>>     [javac] symbol  : method valueOf (long)
>>     [javac] location: class java.lang.Long
>>     [javac]             throw new PSQLException(GT.tr("Bad value for type
>> {0} : {1}", new Object[]{targetType, Long.valueOf(val)}),
>>     [javac]
>>                                         ^
>>     [javac] Note: Some input files use or override a deprecated API.
>>     [javac] Note: Recompile with -deprecation for details.
>>     [javac] 2 errors
>>
>> Both of these method calls - Boolean.parseBoolean() and Long.valueOf() -
>> are introduced in JDK 1.5. Are we dropping support for 1.4? Given that the
>> workaround to maintain 1.4 compatability is so simple (see attached patch)
>> I would recommend maintaing support of 1.4.
>
>
> I would agree that we want to keep support for 1.4
>
> Mikko, I'm guessing this is mostly from your patches.

Yes, my fault. I have tried to stay away from new APIs - the breakage was unintentional.

I'll be more careful in the future, but I do not have any older than 1.6 JVM available
so it is very useful to have the build farm to catch these errors.

If the patch from Mike does not get committed I'll create one tomorrow.

-Mikko