Re: [gobo-eiffel-develop] Precondition violated in XPath test

[Colin A. <col...@go...>]

It's up to you.
6155 has fewer bugs.
I shall continue to work on the code base at 6155, and correct the
bugs as I find them (I'll check this one at the weekend). So please
don't revert the trunk, as otherwise I shall get in a mess again.
(A warning to others - this is how I think I got into the mess - I was
creating some new classes, but then I decided they were in the wrong
clusters - so as I had already issued svn add command, I issued svn mv
commands to move them to the correct clusters. Which was fine until I
tried to update or commit - then svn kept saying it couldn't issue an
add as it was already added. The only way I could make progress was by
doing an svn rm --force, whilst keeping a copy of the class in my
Emacs buffer, and promptly adding it again - but there were a number
of classes involved, and I missed one.
In future, I am going to make it a rule to only move classes when the
commit will consist of nothing but class moves).

As for the SE 1.2r7 issue, I had tested this sort of thing before (I
have removed it elsewhere).
I didn't specifically test against SE on Monday morning as I had no time.

On 01/11/2007, Eric Bezault <er...@go...> wrote:
> Eric Bezault wrote:
> > Colin Adams wrote:
> >> Note that this is in the area of the class that I had to reconstruct
> >> by hand on Monday morning. I THINK I only re-ran the tests with gec
> >> that morning.
> >
> > OK, I managed to restore version 6152, which is the one that has
> > been accidentally removed at version 6153. Can I use it instead of
> > the one that you reconstructed by hand at version 6155?
>
> At least, the test passes now.
>
> --
> Eric Bezault
> mailto:er...@go...
> http://www.gobosoft.com
>