add test for get_hash_digest_column in TestUtils

cellxgene_schema_cli/cellxgene_schema/write_labels.py

@@ -2,14 +2,13 @@
 import traceback
 from typing import Dict, List, Optional
 
-import numpy as np
 import pandas as pd
 
 from cellxgene_schema import ontology
 from cellxgene_schema.env import SCHEMA_REFERENCE_BASE_URL, SCHEMA_REFERENCE_FILE_NAME
 from cellxgene_schema.validate import ONTOLOGY_CHECKER, Validator
 
-from .utils import enforce_canonical_format, getattr_anndata, get_hash_digest_column
+from .utils import enforce_canonical_format, get_hash_digest_column, getattr_anndata
 
 logger = logging.getLogger(__name__)
 
@@ -328,17 +327,6 @@ def _remove_categories_with_zero_values(self):
     def _build_schema_reference_url(self, schema_version: str):
         return f"{SCHEMA_REFERENCE_BASE_URL}/{schema_version}/{SCHEMA_REFERENCE_FILE_NAME}"
 
-    def _get_observation_joinid_column(self):
-        """
-        Set column with unique join ID for each row in obs dataframe.
-        """
-        return (
-            self.adata.obs.index.to_series()
-            .map(xxh3_64_intdigest)
-            .astype(np.uint64)
-            .apply(lambda v: b85encode(v.to_bytes(8, "big")).decode("ascii"))
-        )
-
     def write_labels(self, add_labels_file: str):
         """
         From a valid (per cellxgene's schema) h5ad, this function writes a new h5ad file with ontology/gene labels added

cellxgene_schema_cli/tests/test_utils.py

@@ -1,8 +1,10 @@
 import numpy as np
+from pandas import Series
 import pytest
 from anndata import AnnData
 from cellxgene_schema.utils import (
     enforce_canonical_format,
+    get_hash_digest_column,
     map_ontology_term,
     remove_deprecated_features,
     replace_ontology_term,
@@ -107,3 +109,12 @@ def test_adata_with_canonical_raw_X(self, adata_with_raw):
     def test_adata_with_canonical_X(self, adata_without_raw):
         enforce_canonical_format(adata)
         assert adata_without_raw.X.has_canonical_format is True
+
+
+class TestGetHashDigestColumn:
+    def test_get_hash_digest_column(self, adata_with_raw):
+        hash_digest_column = get_hash_digest_column(adata_with_raw.obs)
+        assert isinstance(hash_digest_column, Series)
+        for val in hash_digest_column:
+            assert isinstance(val, str)
+            assert len(val) == 10   # 5 ASCII chars / 4 bytes, each ID is 8 bytes