add support for corpora default_embedding field (#1696)

* fix mispelling

* re-implement re-embedding

* always load base embedding to fetch counts

* format

* lint

* fix tests

* lint

* fix accept handling

* test log

* more debug

* more

* more

* more

* more

* remove logging

* logging

* jsonify

* remove debugging logs

* lint

* clean up errors a bit

* fix issue found in PR review

* add support for corpora default_embedding

* fix botched merge

* PR review

* PR review

client/src/actions/index.js

@@ -58,7 +58,7 @@ const doInitialDataLoad = () =>
     dispatch({ type: "initial data load start" });
 
     try {
-      const [, schema] = await Promise.all([
+      const [config, schema] = await Promise.all([
         configFetch(dispatch),
         schemaFetch(dispatch),
         userColorsFetchAndLoad(dispatch),
@@ -75,6 +75,15 @@ const doInitialDataLoad = () =>
         obsCrossfilter,
       });
       dispatch({ type: "initial data load complete" });
+
+      const defaultEmbedding = config?.parameters?.["default_embedding"];
+      const layoutSchema = schema?.schema?.layout?.obs ?? [];
+      if (
+        defaultEmbedding &&
+        layoutSchema.some((s) => s.name === defaultEmbedding)
+      ) {
+        dispatch(embActions.layoutChoiceAction(defaultEmbedding));
+      }
     } catch (error) {
       dispatch({ type: "initial data load error", error });
     }

server/common/app_config.py

@@ -242,11 +242,17 @@ class AppConfig(object):
             "about_legal_privacy": dataset_config.app__about_legal_privacy,
         }
 
-        # dataset_props
+        # corpora dataset_props
         # TODO/Note: putting info from the dataset into the /config is not ideal.
         # However, it is definitely not part of /schema, and we do not have a top-level
         # route for data properties.  Consider creating one at some point.
         corpora_props = data_adaptor.get_corpora_props()
+        if corpora_props and "default_embedding" in corpora_props:
+            default_embedding = corpora_props["default_embedding"]
+            if isinstance(default_embedding, str) and default_embedding.startswith("X_"):
+                default_embedding = default_embedding[2:]  # drop X_ prefix
+            if default_embedding in data_adaptor.get_embedding_names():
+                parameters["default_embedding"] = default_embedding
 
         data_adaptor.update_parameters(parameters)
         if annotation:

server/test/test_corpora.py

@@ -1,13 +1,19 @@
 import unittest
 import anndata
 import json
+import tempfile
+import shutil
+from http import HTTPStatus
+import requests
 
 from server.common.corpora import (
     corpora_get_versions_from_anndata,
     corpora_is_version_supported,
     corpora_get_props_from_anndata,
 )
-from server.test import PROJECT_ROOT
+from server.test import PROJECT_ROOT, start_test_server, stop_test_server
+
+VERSION = "v0.2"
 
 
 class CorporaAPITest(unittest.TestCase):
@@ -71,3 +77,70 @@ class CorporaAPITest(unittest.TestCase):
 
     def _get_h5ad(self):
         return anndata.read_h5ad(f"{PROJECT_ROOT}/example-dataset/pbmc3k.h5ad")
+
+
+class CorporaRESTAPITest(unittest.TestCase):
+    """ Confirm endpoints reflect Corpora-specific features """
+
+    @classmethod
+    def setCorporaFields(cls, path):
+        adata = anndata.read_h5ad(path)
+        corpora_props = {
+            "version": {
+                "corpora_schema_version": "1.0.0",
+                "corpora_encoding_version": "0.1.0"
+            },
+            "title": "PBMC3K",
+            "contributors": json.dumps([
+                {"name": "name"}
+            ]),
+            "layer_descriptions": {
+                "X": "raw counts"
+            },
+            "organism": "human",
+            "organism_ontology_term_id": "unknown",
+            "project_name": "test project",
+            "project_description": "test description",
+            "project_links": json.dumps([
+                {"link_name": "test link", "link_type": "SUMMARY", "link_url": "https://a.u.r.l/"}
+            ]),
+            "default_embedding": "X_tsne"
+        }
+        adata.uns.update(corpora_props)
+        adata.write(path)
+
+    @classmethod
+    def setUpClass(cls):
+        cls.tmp_dir = tempfile.TemporaryDirectory()
+        src = f"{PROJECT_ROOT}/example-dataset/pbmc3k.h5ad"
+        dst = f"{cls.tmp_dir.name}/pbmc3k.h5ad"
+        shutil.copyfile(src, dst)
+        cls.setCorporaFields(dst)
+        cls.ps, cls.server = start_test_server([dst])
+
+    @classmethod
+    def tearDownClass(cls):
+        stop_test_server(cls.ps)
+        cls.tmp_dir.cleanup()
+
+    def setUp(self):
+        self.session = requests.Session()
+        self.url_base = f"{self.server}/api/{VERSION}/"
+
+    def test_config(self):
+        endpoint = "config"
+        url = f"{self.url_base}{endpoint}"
+        result = self.session.get(url)
+        self.assertEqual(result.status_code, HTTPStatus.OK)
+        self.assertEqual(result.headers["Content-Type"], "application/json")
+
+        result_data = result.json()
+        self.assertIsInstance(result_data["config"]["corpora_props"], dict)
+        self.assertIsInstance(result_data["config"]["parameters"], dict)
+
+        corpora_props = result_data["config"]["corpora_props"]
+        parameters = result_data["config"]["parameters"]
+
+        self.assertEqual(corpora_props["version"]["corpora_schema_version"], "1.0.0")
+        self.assertEqual(corpora_props["organism"], "human")
+        self.assertEqual(parameters["default_embedding"], "tsne")