[elephant-devel] Re: Made any progress on that ACL issue, Andrew B.?

[Robert L. Read]

Thanks to Andrew Philpot and Andrew Blumberg, I believe the minor fix
noted noted
below makes 0.3.0 completely green on ACL as well as SBCL.

I will commit this change today, and publish 0.3.1 or 0.4.0 today.

Also, Waldo found a serious bug in the documentation; thank your for
that.  I will 
fix that today as well.


On Fri, 2006-01-06 at 01:04 -0600, Andrew Blumberg wrote:

> yes, that sounds like a good idea to me.
> 
>  					- andrew
> 
> On Thu, 5 Jan 2006, Robert L. Read wrote:
> 
> > After commenting out the initform, I am green on all tests including
> > my migration tests.
> >
> > Andrew Blumberg, should I publish this fix into the code base?
> >
> > Mr. Philpot, I will continue to try to help with any additional problems
> > you might have.
> >
> > Thanks to both of you.
> >
> >
> >
> > On Thu, 2006-01-05 at 14:33 -0600, Andrew Blumberg wrote:
> >
> >> hi andrew,
> >>
> >>  	i'm glad to hear the fix works.  thanks for testing it.  just to
> >> be clear, the migration tests are all aimed at moving between different
> >> repositories?  (i spent a fair bit of time getting class update stuff to
> >> be in a reasonable state during the last revision i wrote, and if that's
> >> what is meant by migration i'll be sad).
> >>
> >>  								- andrew
> >>
> >> On Thu, 5 Jan 2006, Andrew Philpot wrote:
> >>
> >>>
> >>> Andrew B. was right all along here.  I rebuilt all from scratch, with
> >>> the initform commented out, and followed the test procedure more
> >>> closely, and I succeeded except for the migration stuff.
> >>>
> >>> 5 out of 107 total tests failed: ELEPHANT-TESTS::MIGRATE1,
> >>>   ELEPHANT-TESTS::MIGRATE2, ELEPHANT-TESTS::MIGRATE3,
> >>>   ELEPHANT-TESTS::MIGRATE4, ELEPHANT-TESTS::MIGRATE5.
> >>>
> >>> Robert, what do you get if you replace the definition of defclass
> >>> PERSISTENT with
> >>>
> >>> (defclass persistent ()
> >>>  ((%oid :accessor oid :initarg :from-oid)
> >>>   ;; This is just an idea for storing connections in the persistent
> >>>   ;; objects; these should be transient as well, if that flag exists!
> >>>   ;; In the case of sleepy cat, this is the controller-db from
> >>>   ;; the store-controller.  In the case of SQL this is
> >>>   ;; the connection spec (since the connection might be broken?)
> >>>   ;; It probably would be better to put a string in here in the case
> >>>   ;; of sleepycat...
> >>>   (dbconnection-spec-pst :type (or list string)
> >>> 			  :accessor :dbcn-spc-pst
> >>> 			  :initarg :dbconnection-spec-pst
> >>> 			  ;; :initform '()
> >>> 			  )
> >>>   )
> >>>  (:documentation
> >>>   "Abstract superclass for all persistent classes (common
> >>> to user-defined classes and collections.)"))
> >>>
> >>> Andrew P.
> >>>
> >>>
> >
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.common-lisp.net/pipermail/elephant-devel/attachments/20060106/b91ddac5/attachment.html>