Test partial corrs API with mix of existing and non-existing control traits

Test that the partial correlations endpoint handles a mix of existing and
non-existing control traits gracefully and issues a warning to the user.

Summary of changes:
* gn3/computations/partial_correlations.py: Issue a warning for all
  non-existing control traits
* gn3/db/partial_correlations.py: update queries - use `INNER JOIN` for tables
  instead of comma-separated list of tables
* tests/integration/conftest.py: Add `db_conn` fixture to provide a database
  connection to the tests. This will probably be changed in the future to
  connect to a temporary database for tests.
* tests/integration/test_partial_correlations.py: Add test to check for
  correct behaviour with a mix of existing and non-existing control traits

4 files changed, 65 insertions, 10 deletions

diff --git a/gn3/computations/partial_correlations.py b/gn3/computations/partial_correlations.py
index 1cc969c..3633a59 100644
--- a/gn3/computations/partial_correlations.py
+++ b/gn3/computations/partial_correlations.py
@@ -628,6 +628,12 @@ def partial_correlations_entry(# pylint: disable=[R0913, R0914, R0911]
         return {
             "status": "not-found",
             "message": "None of the requested control traits were found."}
+    for trait in cntrl_traits:
+        if trait["haveinfo"] == False:
+            warnings.warn(
+                (f"Control traits {trait['trait_fullname']} was not found "
+                 "- continuing without it."),
+                category=UserWarning)
 
     group = primary_trait["db"]["group"]
     primary_trait_data = all_traits_data[primary_trait["trait_name"]]
diff --git a/gn3/db/partial_correlations.py b/gn3/db/partial_correlations.py
index 0931f09..0075cad 100644
--- a/gn3/db/partial_correlations.py
+++ b/gn3/db/partial_correlations.py
@@ -347,10 +347,11 @@ def probeset_traits_info(
         "probe_set_note_by_rw", "flag")
     query = (
         "SELECT ProbeSet.Name AS trait_name, {columns} "
-        "FROM ProbeSet, ProbeSetFreeze, ProbeSetXRef "
-        "WHERE ProbeSetXRef.ProbeSetFreezeId = ProbeSetFreeze.Id "
-        "AND ProbeSetXRef.ProbeSetId = ProbeSet.Id "
-        "AND ProbeSetFreeze.Name IN ({dataset_names}) "
+        "FROM ProbeSet INNER JOIN ProbeSetXRef "
+        "ON ProbeSetXRef.ProbeSetId = ProbeSet.Id "
+        "INNER JOIN ProbeSetFreeze "
+        "ON ProbeSetFreeze.Id = ProbeSetXRef.ProbeSetFreezeId "
+        "WHERE ProbeSetFreeze.Name IN ({dataset_names}) "
         "AND ProbeSet.Name IN ({trait_names})").format(
             columns=", ".join(["ProbeSet.{}".format(x) for x in keys]),
             dataset_names=", ".join(["%s"] * len(dataset_names)),
@@ -376,11 +377,10 @@ def geno_traits_info(
         "SELECT "
         "Geno.Name AS trait_name, {columns} "
         "FROM "
-        "Geno, GenoFreeze, GenoXRef "
-        "WHERE "
-        "GenoXRef.GenoFreezeId = GenoFreeze.Id AND GenoXRef.GenoId = Geno.Id AND "
-        "GenoFreeze.Name IN ({dataset_names}) AND "
-        "Geno.Name IN ({trait_names})").format(
+        "Geno INNER JOIN GenoXRef ON GenoXRef.GenoId = Geno.Id "
+        "INNER JOIN GenoFreeze ON GenoFreeze.Id = GenoXRef.GenoFreezeId "
+        "WHERE GenoFreeze.Name IN ({dataset_names}) "
+        "AND Geno.Name IN ({trait_names})").format(
             columns=", ".join(["Geno.{}".format(x) for x in keys]),
             dataset_names=", ".join(["%s"] * len(dataset_names)),
             trait_names=", ".join(["%s"] * len(traits)))
diff --git a/tests/integration/conftest.py b/tests/integration/conftest.py
index be58d80..e1d1c37 100644
--- a/tests/integration/conftest.py
+++ b/tests/integration/conftest.py
@@ -1,12 +1,22 @@
+"""Module that holds fixtures for integration tests"""
 import pytest
 
 from gn3.app import create_app
+from gn3.db_utils import database_connector
 
 @pytest.fixture(scope="session")
 def client():
+    """Create a test client fixture for tests"""
     # Do some setup
     app = create_app()
     app.config.update({"TESTING": True})
     app.testing = True
     yield app.test_client()
     # Do some teardown/cleanup
+
+
+@pytest.fixture
+def db_conn():
+    """Create a db connection fixture for tests"""
+    ## Update this to use temp db once that is in place
+    return database_connector()[0]
diff --git a/tests/integration/test_partial_correlations.py b/tests/integration/test_partial_correlations.py
index ff6d771..7f9ff30 100644
--- a/tests/integration/test_partial_correlations.py
+++ b/tests/integration/test_partial_correlations.py
@@ -1,7 +1,7 @@
 """Test partial correlations"""
 import pytest
 
-from tests.integration.conftest import client
+from gn3.computations.partial_correlations import partial_correlations_entry
 
 @pytest.mark.integration_test
 @pytest.mark.parametrize(
@@ -176,3 +176,42 @@ def test_partial_correlation_api_with_non_existent_control_traits(client, post_d
     assert (
         response.status_code == 404 and response.is_json and
         response.json.get("status") != "error")
+
+@pytest.mark.integration_test
+@pytest.mark.slow
+@pytest.mark.parametrize(
+    "primary,controls,method,target", (
+        (# Probeset
+            "UCLA_BXDBXH_CARTILAGE_V2::ILM103710672", (
+                "UCLA_BXDBXH_CARTILAGE_V2::nonExisting01",
+                "UCLA_BXDBXH_CARTILAGE_V2::nonExisting02",
+                "UCLA_BXDBXH_CARTILAGE_V2::ILM380019"),
+            "Genetic Correlation, Pearson's r", "BXDPublish"),
+        (# Publish
+            "BXDPublish::17937", (
+                "BXDPublish::17940",
+                "BXDPublish::nonExisting03"),
+            "Genetic Correlation, Spearman's rho", "BXDPublish"),
+        (# Geno
+            "AKXDGeno::D4Mit16", (
+                "AKXDGeno::D1Mit170",
+                "AKXDGeno::nonExisting04",
+                "AKXDGeno::D1Mit135",
+                "AKXDGeno::nonExisting05",
+                "AKXDGeno::nonExisting06"),
+            "SGO Literature Correlation", "BXDPublish")
+    )
+    # Temp -- the data in the database for these is ephemeral, making it
+    #         difficult to test for these without a temp database with the temp
+    #         traits data set to something we are in control of
+     )
+def test_part_corr_api_with_mix_of_existing_and_non_existing_control_traits(
+        db_conn, primary, controls, method, target):
+    """
+    Check that calling the function with a mix of existing and missing control
+    traits raises an warning.
+    """
+    criteria = 10
+    with pytest.warns(UserWarning):
+        partial_correlations_entry(
+            db_conn, primary, controls, method, criteria, target)