[Python] get_census_version_directory returns sorted elements (#728)

* get_census_version_directory returns sorted elements * typing
chanzuckerberg · Aug 25, 2023 · 62e15e1 · 62e15e1
1 parent fd03fb6
commit 62e15e1
Show file tree

Hide file tree

Showing 2 changed files with 34 additions and 13 deletions.
diff --git a/api/python/cellxgene_census/src/cellxgene_census/_release_directory.py b/api/python/cellxgene_census/src/cellxgene_census/_release_directory.py
@@ -6,6 +6,7 @@
 
 Methods to retrieve information about versions of the publicly hosted Census object.
 """
+from collections import OrderedDict
 from typing import Dict, Literal, Optional, Union, cast
 
 import requests
@@ -159,6 +160,7 @@ def get_census_version_directory() -> Dict[CensusVersionName, CensusVersionDescr
     """
     response = requests.get(CELL_CENSUS_RELEASE_DIRECTORY_URL)
     response.raise_for_status()
+
     directory: CensusDirectory = cast(CensusDirectory, response.json())
 
     # Resolve all aliases for easier use
@@ -180,7 +182,17 @@ def get_census_version_directory() -> Dict[CensusVersionName, CensusVersionDescr
             cast(CensusVersionDescription, directory[census_version])["alias"] = alias
 
     # Cast is safe, as we have removed all aliases
-    return cast(Dict[CensusVersionName, CensusVersionDescription], directory)
+    unordered_directory = cast(Dict[CensusVersionName, CensusVersionDescription], directory)
+
+    # Sort by aliases and release date, descending
+
+    aliases = [(k, v) for k, v in unordered_directory.items() if v.get("alias") is not None]
+    releases = [(k, v) for k, v in unordered_directory.items() if v.get("alias") is None]
+    ordered_directory = OrderedDict()
+    for k, v in aliases + sorted(releases, key=lambda k: k[0], reverse=True):
+        ordered_directory[k] = v
+
+    return ordered_directory
 
 
 def get_census_mirror_directory() -> Dict[CensusMirrorName, CensusMirror]:

diff --git a/api/python/cellxgene_census/tests/test_directory.py b/api/python/cellxgene_census/tests/test_directory.py
@@ -8,32 +8,32 @@
 from cellxgene_census._release_directory import CELL_CENSUS_MIRRORS_DIRECTORY_URL, CELL_CENSUS_RELEASE_DIRECTORY_URL
 
 DIRECTORY_JSON = {
-    "stable": "2022-10-01",
-    "latest": "2022-11-01",
-    "2022-11-01": {
-        "release_date": "2022-11-30",
-        "release_build": "2022-11-01",
+    "2022-10-01": {
+        "release_date": "2022-10-30",
+        "release_build": "2022-10-01",
         "soma": {
-            "uri": "s3://cellxgene-data-public/cell-census/2022-11-01/soma/",
+            "uri": "s3://cellxgene-data-public/cell-census/2022-10-01/soma/",
             "s3_region": "us-west-2",
         },
         "h5ads": {
-            "uri": "s3://cellxgene-data-public/cell-census/2022-11-01/h5ads/",
+            "uri": "s3://cellxgene-data-public/cell-census/2022-10-01/h5ads/",
             "s3_region": "us-west-2",
         },
     },
-    "2022-10-01": {
-        "release_date": "2022-10-30",
-        "release_build": "2022-10-01",
+    "2022-11-01": {
+        "release_date": "2022-11-30",
+        "release_build": "2022-11-01",
         "soma": {
-            "uri": "s3://cellxgene-data-public/cell-census/2022-10-01/soma/",
+            "uri": "s3://cellxgene-data-public/cell-census/2022-11-01/soma/",
             "s3_region": "us-west-2",
         },
         "h5ads": {
-            "uri": "s3://cellxgene-data-public/cell-census/2022-10-01/h5ads/",
+            "uri": "s3://cellxgene-data-public/cell-census/2022-11-01/h5ads/",
             "s3_region": "us-west-2",
         },
     },
+    "stable": "2022-10-01",
+    "latest": "2022-11-01",
     # An explicitly dangling tag, to confirm we handle correct
     # Underscore indicates expected failure to test below
     "_dangling": "no-such-tag",
@@ -74,6 +74,15 @@ def test_get_census_version_directory(directory_mock: Any) -> None:
     for tag in directory:
         assert directory[tag] == cellxgene_census.get_census_version_description(tag)
 
+    # Verify that the directory is sorted according to this criteria:
+    # 1. Aliases first
+    # 2. Non aliases after, in reverse order
+    dir_list = list(directory)
+    assert dir_list[0] == "stable"
+    assert dir_list[1] == "latest"
+    assert dir_list[2] == "2022-11-01"
+    assert dir_list[3] == "2022-10-01"
+
 
 def test_get_census_version_description_errors() -> None:
     with pytest.raises(ValueError):