Mailing-List: contact derby-dev-help@db.apache.org; run by ezmlm
Precedence: bulk
Reply-To: <derby-dev@db.apache.org>
Received-SPF: neutral (asf.osuosl.org: local policy)
Message-ID: <44EA40FC.3080804@apache.org>
Date: Mon, 21 Aug 2006 16:25:48 -0700
From: Daniel John Debrunner <djd@apache.org>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US;
 rv:1.7.3) Gecko/20040910
MIME-Version: 1.0
To: derby-dev@db.apache.org
Subject: Re: LineNumber table in class files and sane=false strangeness
References: <44EA3B59.2020907@apache.org>
In-Reply-To: <44EA3B59.2020907@apache.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Daniel John Debrunner wrote:


> If I remove PermissionsCatalogRowFactory.class and recompile just using
> ant from the top level then it reverts to 2116 bytes in the first codeline.
> 
> Doing a ant clobber followed by an ant -verbose all in the first
> codeline reveals this as the only entry for that class.
> 
>     [javac]
> org\apache\derby\impl\sql\catalog\PermissionsCatalogRowFactory.java
> omitted as
> org/apache/derby/impl/sql/catalog/PermissionsCatalogRowFactory.class is
> up to date.
> 
> So I assume something else is compiling it indirectly with a different
> -g flag, that doesn't match my sane=false setting, any ideas?

Doing this provides a consistent build (thanks Andrew):

ant clobber
ant insanse
ant all

Looking at the previous verbose output for a full build shows all the
compiles had just -g, not -g:none. Maybe just setting the ant property
'sane=false' is no longer enough, one one needs to run 'ant insane'? It
does seem my codeline was in a half insane, half sane mode.

Dan.