LSCS query using metadata

We want to query for documents using tags (metadata) that we have added via the taxonomy (branch.taxonomies file).

When we tag the document the values are correct ie "Graduates" but when we query using LSCS the value returned is the code ie i2ei52ab

As a result it seems we must use the internal coded value in our request eg

q=(TeamSite/Templating/DCR/Type:ArticleFCT)AND(TeamSite/Metadata/Audience:i2ei52ab,i2ei52ac)

Is that correct or should we be able to use the 'english' values in our query?

If so what is the best practice for mapping the value to the code as we have been advised the branch.taxonomies should not be promoted to the runtime?

Find more posts tagged with

Comments

Rick Poulin

SL_CM wrote:
Is that correct or should we be able to use the 'english' values in our query?

If so what is the best practice for mapping the value to the code as we have been advised the branch.taxonomies should not be promoted to the runtime?

I haven't used the taxonomy manager much, but that seems to match the behaviour that others have reported on these forums. As such, I'd expect you have to map the English value to its ID. OOTB, the PLC attempts to deploy the branch.taxonomies to a database, which your runtime would presumably query. I'd ask whoever advised you not to deploy your taxonomy why that's the case, and what their alternate solution is.

cesarecastro

Two of the biggest advantages that I see from using an ID instead of the English are the following:

1) Multilingual Enterprise Taxonomies: A node could be translated in to multiple languagues and you would retain the ability to query your entire repository for a particular taxonomy node no matter that language that it is in.

2) Taxonmy updates: For example, changing your node from Tax to Taxes, would not required to retag all of your assets but only change the value in your taxonomy mananger.

I am an not arguing on favor of one option over the other. The two that I mentioned are based on the realities and contraints that I am currently dealing with.

Cesar

Rick Poulin

cesarecastro wrote:
Two of the biggest advantages that I see from using an ID instead of the English are the following:

I don't think anybody here needs convincing. It's not about how it should work, it's about dealing with how the product actually works. Smiley Tongue

Bowker

Here is another reason to use the cryptic keys...

on a multi-value item you could have the case of one taxonomy item being "north" and another one being "northeast".

If you have a document tagged "north","northeast" and another document tagged "northeast"

If you did a search for exactly "north" it would not find any (probably). if you did "contains" "north" you would get both. Between the technology and business groups we call this the "substring issue."

Use the index values!

SL_CM

OK that's helpful thanks, I'll stick to the codes and focus on fixing the issue of accessing the taxonomy on the runtime