On 3 Jun 2007, at 19:46, Richard Holland wrote:
<snip>
I won't go as far as Arek and call the lack of this option in 0.5 a
'bug in
the algorithm'... I'd prefer to call the ability to include additional
relations beyond those covered by the default algorithm a 'new feature'
instead. :) After all, the algorithm must have a defined and
predictable
stopping point else it would end up merging the entire schema into a
single table!
</snip>
I agree with you entirely :-) This was never a bug in the algorithm,
because the algorithm
has been around for 5 years now, years before even the idea of MBuilder
was conceived and it worked correctly so far.
This was a bug in the implementation of the algorithm. The 0.5 version
simply cut it's path too early (The focus point for the dm
transformation was the first 1:m table rather than the first link
table as it was customary for our earlier implementations - the
starting point was implemented incorrectly. The stopping point was
correct and it always was there anyway). The good news is that this is
now fixed for 0.6 so people should not be 'short changed' on their
tables anymore :)
Mummi - as a general rule, you should very rarely need to add tables to
the path manually. This is a sign
of an unusual situation and breaking 1:n/n:m cutoff. Most of the times
you should be simply removing tables from the path as the algorithm
is 'greedy' in this respect and tend to include more tables that you
normally require.
anyway, happy building
a.
------------------------------------------------------------------------
-------
Arek Kasprzyk
EMBL-European Bioinformatics Institute.
Wellcome Trust Genome Campus, Hinxton,
Cambridge CB10 1SD, UK.
Tel: +44-(0)1223-494606
Fax: +44-(0)1223-494468
------------------------------------------------------------------------
-------
