"Nikolay Diakov" <[EMAIL PROTECTED]> wrote:

> I see you do the wrapping in a RuntimeException trick. Perhaps you
> can introduce a special exception derived from RuntimeException that
> you would throw in that case. It would basically mean "The
> underlying FS does something we cannot tolerate so we fail fast."

I think in this case I'd like to make a "LockReleaseFailedException"
that subclasses from IOException, since the root cause (inability to
delete) really is an IOException type of issue.  This would be a very
minor break to backwards compatibility: all methods that call
Lock.release in Lucene themselves already throw IOException.

I'll open an issue & post a patch!

Mike

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to