[squeak-dev] Re: Completed audit comparing trunk as of today (7/3/09) to 3.10.2 release as represented by update stream

Andreas Raab andreas.raab at gmx.de
Fri Jul 3 22:29:00 UTC 2009


[Sorry folks I'm being brief here because I'm posting while waiting on 
my flight and the number of messages in my inbox is long...]

Thanks Ken. This audit is really helpful. I think the BabySRE extension 
is non-critical and we either ignore it (no senders) or move it into a 
more appropriate category.

The DoIt is interesting since it raises the question whether we need a 
general "reset to factory defaults" button that reinitializes pretty 
much anything that we expect to be current in any image we care about. I 
can't imagine that someone hasn't done something along these lines 
already - any takers?

Lastly, Version number. My proposal is to use the date of the newest 
trunk package. This is a policy that we use at work (i.e., choose the 
newest date amongst the packages that come out of the core repositories) 
which allows to have other packages alongside them. I'm not wed to that 
proposal though; alternatives are very welcome.

Cheers,
   - Andreas

Ken Causey wrote:
> I have completed my audit through the update stream between 7159 and
> 7179 checking the current status of trunk.  The only issues I find:
> 
> 1. 7161 - method categorized as BabySRE extension method and so packaged
> in seperate package not loaded.  Move into proper category and mark as
> deprecated?  As an additional data point, this method exists in
> 3.10-7159 but not in a *BabySRE extension category.
> 
> 2. 7161 - Do-it to refresh objects flap not 'done'.
> 
> 3. Version number.  Policy still to be discussed.
> 
> Otherwise it looks OK to me...
> 
> Oh, ran the tests, one new failure
> 
> TimeTest>>#testGeneralInquiries
> 
> for some reason it seems that the now is not reliably less than or equal
> to now.  Looks like a pain to track down to me.  Any takers?
> 
> Ken
> 
> 
> ------------------------------------------------------------------------
> 
> 




More information about the Squeak-dev mailing list