API taxon search doesn't respect tautonyms

Bug Reports
1

Platform: Web

App version number, if a mobile app issue:

Browser, if a website issue:

URLs (aka web addresses) of any relevant observations or pages:
For Pica pica (Eurasian Magpie):

https://api.inaturalist.org/v1/taxa?q=Pica+pica&is_active=true&per_page=1

Returns:
“default_photo”: {
“id”: 2964195,
“license_code”: “cc-by-nc”,

But that picture is a toucan:
“https://inaturalist-open-data.s3.amazonaws.com/photos/2964195/medium.jpg”

https://inaturalist-open-data.s3.amazonaws.com/photos/2964195/medium.jpg

Screenshots of what you are seeing:

Description of problem:

API request for default photo of Eurasian Magpie returns url for the image of a Toucan.

2

I don’t think this is a bug, you searched for a taxon got 759 results and the first one happens to Order Piciformes (from “Tucanos, Pica-Paus e afins”), and this is the correct photo for that taxon. I think for your purposes you want to find an exact match for “Pica pica” (maybe that is a filter you can include that in the url or something?), but anyway I would suggest using the taxon ID (891696) since that is less likely to be misinterpreted.

3

For example: https://api.inaturalist.org/v1/taxa?id=891696

4

Platform: Web

App version number, if a mobile app issue:

Browser, if a website issue: Any

URLs (aka web addresses) of any relevant observations or pages:

https://api.inaturalist.org/v1/taxa?q=Riparia+riparia&is_active=true&per_page=1

Screenshots of what you are seeing:

Description of problem:

Step 1: Visit “https://api.inaturalist.org/v1/taxa?q=Riparia+riparia&is_active=true&per_page=1”

Step 2: Use returned json to extract the default photo id: 78038371

Step 3: Use default photo id to generate this url: “https://inaturalist-open-data.s3.amazonaws.com/photos/78038371/medium.jpg”

Result: The returned image is the seed head of a grass species:

https://inaturalist-open-data.s3.amazonaws.com/photos/78038371/medium.jpg

5

I merged your reports and changed the title to reflect the shared cause of both of them. When you search for a name that has the same genus and species epithet (a tautonym), the API doesn’t process it as a true genus and species combination. This is a known issue that is already documented on github: https://github.com/inaturalist/iNaturalistAPI/issues/605

As isopod-fan said, the pica search is hitting on the Portuguese name of https://www.inaturalist.org/taxa/17550-Piciformes, and the riparia search is hitting on an unaccepted scientific name of https://www.inaturalist.org/taxa/63337-Phalaris-arundinacea. Search by taxon id is the current workaround.

6

I think it is the logic in the search query that is to blame here, when you search for “Pica+pica” the search returns results that include Pica and/or pica (anywhere in the name). Since the case probably isn’t taken into account it is really just looking for results that have Pica/pica and/or Pica/pica in them, so it sees “Tucanos, Pica-Paus e afins”, and Pica/pica is present, as is Pica/pica, so that is as good as it gets, along with the other 759 results that have Pica/pica in them. Then I assume it sorts by popularity or number of observations, so Piciformes is returned as the top result.

If you want this to work then I assume you will need to have the whole query “Pica pica” (including the space) taken into account rather than just the individual words.

7

Note that even if they searched for “Pica+pica”, it would still just be a prefix search, not an exact string search. The iNaturalist API doesn’t support exact string searches: https://github.com/inaturalist/iNaturalistAPI/issues/336.

8

Thanks for the clarifications!