Improve performance for sample listing index

[ramonski]

Description of the issue/feature this PR addresses

This PR improves the performance of the listing_searchable_text ZCTextIndex of the sample catalog by adding only explicit terms into the index, instead of all the metadata columns defined.

Currently, the following values are part of the index:

• Sample ID
• Batch ID
• Client ID
• Client Name
• Client Order Number
• Client Reference
• Client Sample ID
• Sample Type Title
• Sample Point Title

This improves the performance by 10% when creating new samples.

Furthermore, the stop words are kept in the Lexicon.

Current behavior before PR

All metadata columns are used for the sample listing_searchable_text index

Desired behavior after PR is merged

Only the above mentioned values are used for the listing_searchable_text index

--
I confirm I have tested this PR thoroughly and coded it according to PEP8
and Plone's Python styleguide standards.

[toropok]

@ramonski @xispa
Hi,

That seems you're intentionally removed get_searchable_text_tokens call from Samples indexer. How do you suppose to extend Sample entity custom search fields? Earlier custom ListingSearchableTextProvider class has to be implemented.

thanks!

[ramonski]

Hi @toropok,
apologies, we overlooked this functionality.
Would you mind to hook it into the sample indexer directly?
The get_searchable_text_tokens will be sooner or later removed then.
Thanks