Here's a solution using a background colour expression that relates directly to the example as provided, in case it helps anyone.

My query for data set is as follows, rather than...

The form that preserves ampersands in the approach originally used would be... (notice the required additional nested SELECT). This is also explained in the replies and responses to the blog...

Thanks for the question, though I agree that STUFF is simpler.

I also wanted to add that if you may have odd characters in your text, you will fall foul of...

From the information given, I'm missing why the update fails. We're told that it's a primary key, but that doesn't stop it being updated.

create table test (id int NOT NULL...

I might be getting the wrong end of the stick, but I interpret it as an approach using the RIM design as a working database design - expressing it in...

Though wary of re-igniting the discussion (informative and lively though it has been), I wonder how this relates to the HL7 RIM model (look up RIMBAA) - as I (only...

As I understand it, that's exactly what Anchor Modeling (see earlier post) is trying to achieve, since new attributes can be added without detriment to the existing schema and with...

Not EAV, but I'd be interested in people's views on Anchor Modelling (www.anchormodeling.com) which basically normalises out every column except the primary key, but provides decent generated code for accessing...

Hmm, I answered too quickly (being on holiday now), thinking that without specifying the column list, the INSERT would expect values for both columns and hence would fail the inserts...

Thanks for the explanation.

I'd assumed that the inner SELECT without a FROM would fail syntax.

I am now enlightened, so the QOD has done its job for...

Thanks for the approach, which makes a lot of sense.

Has anyone done any testing to see whether it would be worth taking an adaptive approach to this? What I mean...