lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Steven A Rowe <sar...@syr.edu>
Subject RE: Lucene-Solr-tests-only-trunk - Build # 436 - Failure
Date Fri, 22 Oct 2010 19:21:22 GMT
I agree that compilation warnings are not useless.  In general, anyway.

But for the case of an email alerting the dev list to a broken compile, the email should point
out what's breaking the compile.  The warnings are useless in this context (i.e., telling
us what broke the build).

On 10/22/2010 at 3:09 PM, Robert Muir wrote:
> On Fri, Oct 22, 2010 at 3:02 PM, Steven A Rowe <sarowe@syr.edu> wrote:
> > It would be nice if compilation warnings were turned off, so that
> > compilation *errors* would be shown instead of warnings here
> > (compilation output is otherwise useless) ...
> 
> Personally I don't think the warnings are useless, they pointed us to
> some problems in lucene (such as missing 'break' from a switch
> statement).
> 
> Its just that someone needs to actually fix the warnings in solr... i
> only fixed the ones i was comfortable with myself.
> 
> I think this stuff is important and we should fix it, if we aren't
> going to fix (most of) the basic warnings from the compiler then we
> should go ahead and cancel ideas like
> https://issues.apache.org/jira/browse/SOLR-143 (integration of PMD) as
> Won't fix.

Mime
View raw message