Merge pull request #576 from pyinat/taxonomy

Add Observation.taxon.ancestors based on identification data

CONTRIBUTORS.md

@@ -4,16 +4,17 @@ questions, and other contributions that have helped to improve pyinaturalist:
 
 * [AE Hamrick](https://github.com/AEHamrick)
 * [abubelinha](https://github.com/abubelinha)
+* [Arky](https://github.com/arky)
 * [Ben Armstrong](https://github.com/synrg)
 * [Dr. Tom August](https://github.com/AugustT)
 * [Eduardo Ramírez](https://github.com/eduramirezh)
 * [Felipe S Barros](https://github.com/FelipeSBarros)
+* [jumear](https://github.com/jumear)
 * [Ken-ichi Ueda](https://github.com/kueda)
 * [Matt Muir](https://forum.inaturalist.org/u/muir)
 * [Nicolas Noé](https://github.com/niconoe)
 * [Patrick Leary](https://github.com/pleary)
 * [Peter Desmet](https://github.com/peterdesmet)
-* [pisum](https://forum.inaturalist.org/u/pisum)
 * [Sam R](https://github.com/samtreesandbushes)
 * [SDJ Brown](https://github.com/sdjbrown)
 * [Stijn Van Hoey](https://github.com/stijnvanhoey)

HISTORY.md

@@ -10,6 +10,7 @@
 * Add `root_id` filter to `taxon.make_tree()` to explicitly set the root taxon instead of determining it automatically
 * Fix `taxon.make_tree()` rank filtering to allow skipping any number of rank levels
 * `taxon.make_tree()` now returns copies of original taxon objects instead of modifying them in-place
+* Update `Observation.taxon.ancestors` based on identification data, if available
 
 ### Rate limits, timeouts, and error handling
 * Increase default request timeout from 10 to 20 seconds

pyinaturalist/models/observation.py

@@ -319,6 +319,17 @@ def __init__(self, **kwargs):
         # Set identifications_count if missing
         if kwargs.get('identifications') and not kwargs.get('identifications_count'):
             kwargs['identifications_count'] = len(kwargs['identifications'])
+
+        # Add ancestor info to Observation.taxon based on identification data, if available
+        if kwargs.get('taxon') and (idents := kwargs.get('identifications')):
+            taxa_by_id = {
+                t['id']: t
+                for t in chain.from_iterable(
+                    [i.get('taxon', {}).get('ancestors', []) for i in idents]
+                )
+            }
+            ancestors = [taxa_by_id.get(i) for i in kwargs['taxon'].get('ancestor_ids')]
+            kwargs['taxon']['ancestors'] = list(filter(None, ancestors))
         self.__attrs_init__(**kwargs)  # type: ignore
 
     @property

test/test_models.py

@@ -600,6 +600,24 @@ def test_observation__cumulative_ids__most_agree():
     assert obs.cumulative_ids == (2, 3)
 
 
+def test_observation__add_ancestors():
+    obs = Observation.from_json(j_observation_2)
+    assert [f'{t.rank} {t.name}' for t in obs.taxon.ancestors] == [
+        'kingdom Animalia',
+        'phylum Arthropoda',
+        'subphylum Hexapoda',
+        'class Insecta',
+        'subclass Pterygota',
+        'order Lepidoptera',
+        'superfamily Papilionoidea',
+        'family Nymphalidae',
+        'subfamily Danainae',
+        'tribe Danaini',
+        'subtribe Danaina',
+        'genus Danaus',
+    ]
+
+
 def test_observations():
     obs_list = Observations.from_json_list(
         [j_observation_1, j_observation_1, j_observation_2, j_observation_3_ofvs]