On Tue, Sep 11, 2012 at 12:16 AM, Martin Grigorov <mgrigo...@apache.org> wrote:
> On Tue, Sep 11, 2012 at 8:44 AM,  <ivaynb...@apache.org> wrote:
>> fix compilation errors for environments with nullability annotation 
>> validation
>>
>>
>> Project: http://git-wip-us.apache.org/repos/asf/wicket/repo
>> Commit: http://git-wip-us.apache.org/repos/asf/wicket/commit/d86f5e78
>> Tree: http://git-wip-us.apache.org/repos/asf/wicket/tree/d86f5e78
>> Diff: http://git-wip-us.apache.org/repos/asf/wicket/diff/d86f5e78
>>
>> Branch: refs/heads/master
>> Commit: d86f5e781bbd0fe14d0ced13b58fc5e9fff57e37
>> Parents: aae3a99
>> Author: Igor Vaynberg <igor.vaynb...@gmail.com>
>> Authored: Mon Sep 10 20:05:28 2012 -0700
>> Committer: Igor Vaynberg <igor.vaynb...@gmail.com>
>> Committed: Mon Sep 10 20:05:28 2012 -0700
>>
>> ----------------------------------------------------------------------
>>  wicket-experimental/wicket-atmosphere/pom.xml      |    6 ++++++
>>  .../org/apache/wicket/atmosphere/EventFilter.java  |   10 +++++++++-
>>  .../wicket/atmosphere/NoFilterPredicate.java       |   11 ++++++++++-
>>  3 files changed, 25 insertions(+), 2 deletions(-)
>> ----------------------------------------------------------------------
>>
>> +               <dependency>
>> +                       <groupId>com.google.code.findbugs</groupId>
>> +                       <artifactId>jsr305</artifactId>
>> +                       <version>1.3.9</version>
>
> Any reason why not to use latest version of this plugin - 2.0.1 ?

i used the same version that guava depends on. in fact it shouldve
come through as transitive, but didnt for some reason. wasnt sure if
2.x is backwards compatible.

>>         @Override
>> -       public boolean apply(EventSubscription input)
>> +       public boolean apply(@Nullable EventSubscription input)
>>         {
>>                 return input.getFilter().apply(event);
>
> Doesn't that mean that we need to check whether 'input' is null before
> calling #getFilter() ?

this is stil on my todo list to look into to. the null processing
shouldve forced that check or errored out at compile time, thats the
point. something is funky.

-igor

Reply via email to