openjpa-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Rick Curtis <curti...@gmail.com>
Subject Re: down-level entities?
Date Wed, 11 Aug 2010 18:43:50 GMT
Patrick -

It looks like svnversion doesn't work quite as I expected when I made this
change before.... anyway, as of 984522 this should now be working as I
intended.

Thanks,
Rick

On Tue, Aug 10, 2010 at 8:14 PM, Rick Curtis <curtisr7@gmail.com> wrote:

> Yep, sounds like I didn't narrow the scope of this message enough.
>
> I don't like the idea of having the manually increment a value in
> PCEnhancer anytime the file is changed because I think it'll fall to the
> wayside pretty quickly. I wonder if I could change the revision from the
> openjpa-parent revision to the revision of the PCEnhancer file.? This way
> we'll only log a message if the PCEnhancer is really updated. This still may
> log a few false positive messages, but it will be closer to the intent of my
> original change.
>
> If that sounds reasonable I'll give it a shot in the am to see if that'll
> squash some of those messages.
>
> Thanks,
> Rick
>
>
> On Tue, Aug 10, 2010 at 7:46 PM, Patrick Linskey <plinskey@gmail.com>wrote:
>
>> Hi,
>>
>> I'm getting warnings most of the time that I run my build complaining
>> about down-level entities. Are we maybe logging this too eagerly?
>>
>> Right now, it looks like every time that a new OpenJPA build is created,
>> we record different enhancement values in the entities. Since I'm using
>> 2.1.0-SNAPSHOT, that means that if I don't do a clean build every day, I see
>> messages about down-level entities. I think that this level of logging is
>> useful, but changes to the enhancement contract are very rare indeed. It
>> seems like we should just increment something deliberately when such a
>> contractual change happens, rather than using the svn revision information
>> as a proxy for it.
>>
>> Thoughts?
>>
>> -Patrick
>>
>> --
>> Patrick Linskey
>> 202 669 5907
>>
>>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message