Craig L Russell wrote:
>
> On May 14, 2009, at 12:21 AM, Bill Shannon wrote:
>
>> Craig L Russell wrote:
>>> Hi Bill,
>>> On May 13, 2009, at 5:39 PM, Bill Shannon wrote:
>>>> I have a bunch of code in JavaMail that looks roughly like this:
>>>>
>>>> try {
>>>> // do something that might fail for unknown reasons
>>>> } catch (Exception ex) {
>>>> // clean up, close connections, etc.
>>>> throw MyException();
>>>> // or...
>>>> return null;
>>>> // or...
>>>> continue;
>>>> }
>>>>
>>>> The "do something" can fail for lots of reasons - NPE, IOException,
>>>> etc. -
>>>> usually several levels deep. No matter what goes wrong I want to
>>>> detect
>>>> the failure and clean things up, especially making sure not to leave
>>>> any
>>>> connections open.
>>>>
>>>> FindBugs complains about this because "do something" doesn't throw
>>>> Exception.
>
>>>>
>>> If do something doesn't throw Exception, you can catch
>>> RuntimeException or Throwable depending on whether you're prepared
>>> for just RuntimeException or Error conditions.
>>
>> Often "do something" throws some checked exception, so I need to handle
>> it *and* the possible RuntimeExceptions.
>
> This seems to be the crux of the FindBugs complaint.
>
> So IOException gets wrapped in RuntimeException which catch
> (RuntimeException) would catch and FindBugs would not complain.
No, "do something" often is declared to throw a checked exception.
In addition to these "normal" failures, it can fail for many unexpected
reasons, throwing a RuntimeException. I need to catch both. I could
have two catch blocks, of course, but that's pretty ugly too.