Wikidata reconciliation with OpenRefine—get only those subjects with English as the language flag

Question

Leptospermum scoparium (Q1520028) has 8 common name (P1843) entries.

I have data in OpenRefine reconciled to this object:

How can I get only those P1843 entries that are in English? The reconciliation documentation says to use Len to get, for example, the English label of a given object, but I can't work out the SPARQL syntax that would deliver what I want.

I have tried P1843/Len and P1843|Len but these operators don't work the way I thought:

I realise I can limit the number of values returned per row, but the values are returned in the order in which they appear, meaning in this instance the Welsh common name appears first.

If I can't do what I need through a SPARQL-like syntax on the reconcile screen, is there some way to obtain the JSON response from the reconciliation and parse it? I've tried jsonize(cell.recon.match) but that just gives:

{"id":"Q1520028","name":"Leptospermum scoparium","types":["Q16521"],"score":100}

Answer 1

Currently there is no way to do this simply via the reconciliation API. What you could do instead is use the "Add columns by fetching URLs" feature to retrieve the JSON representation of the item from Wikidata (for instance with the expression "https://www.wikidata.org/wiki/Special:EntityData/"+cell.recon.match.id+".json").

Then manipulate that object to get what you need (probably doable in GREL, but personally I would go for Python).