make_tree() updates

HISTORY.md

@@ -9,6 +9,7 @@
 * Add `User.annotated_observations_count` field
 * 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
 
 ### Rate limits, timeouts, and error handling
 * Increase default request timeout from 10 to 20 seconds

pyinaturalist/models/taxon.py

@@ -1,4 +1,5 @@
 import re
+from copy import deepcopy
 from itertools import chain, groupby
 from logging import getLogger
 from typing import Any, Callable, Dict, Iterable, List, Optional
@@ -413,7 +414,8 @@ def make_tree(
     sort_key: Optional[TaxonSortKey] = None,
     root_id: Optional[int] = None,
 ) -> Taxon:
-    """Organize a list of taxa into a taxonomic tree. Expects exactly one root taxon.
+    """Organize a list of taxa into a taxonomic tree, defined by ``children`` and ``ancestors``
+    attributes. Expects exactly one root taxon.
 
     Args:
         taxa: Taxon objects to organize
@@ -427,6 +429,7 @@ def make_tree(
     """
     include_ranks = [r.lower() for r in include_ranks or []]
     sort_key = sort_key if sort_key is not None else _sort_rank_name
+    taxa = [deepcopy(t) for t in taxa]
     root = _find_root(taxa, include_ranks, root_id)
 
     # Group taxa by parent ID, including any ungrafted children added directly to root
@@ -440,6 +443,8 @@ def add_descendants(taxon, ancestors=None) -> Taxon:
         taxon.ancestors = ancestors or []
         for child in get_included_children(taxon):
             child = add_descendants(child, taxon.ancestors + [taxon])
+            child.ancestor_ids = [a.id for a in child.ancestors]
+            child.parent_id = taxon.id
             taxon.children.append(child)
 
         taxon.children = sorted(taxon.children, key=sort_key)
@@ -504,6 +509,10 @@ def _find_and_graft_root(taxa: Iterable[Taxon], include_ranks: Optional[List[str
     # If there are multiple branches, we need to insert a 'Life' root above them
     root = TaxonCount(id=ROOT_TAXON_ID, name='Life', rank='stateofmatter', is_active=True)  # type: ignore
     root.children = root_taxa
+    for t in root.children:
+        t.ancestors = [root]
+        t.ancestor_ids = [root.id]
+        t.parent_id = root.id
     return root
 
 

test/test_models.py

@@ -1116,25 +1116,33 @@ def test_make_tree():
     assert node.name == 'Bombus'
     children = [t.name for t in node.children]
     assert children == ['Bombus', 'Psithyrus', 'Pyrobombus', 'Subterraneobombus', 'Thoracobombus']
+    assert (
+        node.child_ids == [t.id for t in node.children] == [538903, 538893, 538900, 415027, 538902]
+    )
 
     # Ensure ancestors are updated
     assert root.ancestors == []
-    assert [t.id for t in node.ancestors] == [
-        48460,
-        1,
-        47120,
-        372739,
-        47158,
-        184884,
-        47201,
-        124417,
-        326777,
-        47222,
-        630955,
-        47221,
-        199939,
-        538883,
-    ]
+    assert node.parent_id == 538883
+    assert (
+        node.ancestor_ids
+        == [t.id for t in node.ancestors]
+        == [
+            48460,
+            1,
+            47120,
+            372739,
+            47158,
+            184884,
+            47201,
+            124417,
+            326777,
+            47222,
+            630955,
+            47221,
+            199939,
+            538883,
+        ]
+    )
 
 
 def test_make_tree__filtered():
@@ -1169,6 +1177,18 @@ def test_make_tree__filtered():
     ]
 
 
+def test_make_tree__preserves_originals():
+    """Children/ancestors of original taxon objects should be preserved"""
+    taxa = Taxon.from_json_list(j_life_list_2)
+    animalia = taxa[1]
+    animalia.ancestors = [Taxon(name='Life')]
+    animalia.children = [Taxon(name='Arthropoda')]
+    make_tree(taxa, include_ranks=['kingdom', 'class', 'family', 'genus', 'species'])
+
+    assert animalia.ancestors[0].name == 'Life'
+    assert animalia.children[0].name == 'Arthropoda'
+
+
 def test_make_tree__flattened():
     flat_list = make_tree(Taxon.from_json_list(j_life_list_1)).flatten()
     assert [t.id for t in flat_list] == [48460, 1, 2, 3, 573, 574, 889, 890, 980, 981]
@@ -1237,6 +1257,7 @@ def test_make_tree__ungrafted():
     assert root.name == 'Life'
     assert root.children[0].name == 'Animalia'
     assert root.children[1].name == 'Monocots'
+    assert root.children[0].children[0].name == 'Arthropoda'
 
 
 def test_make_tree__explicit_root():