Merge pull request #10 from axiom-data-science/testcov

Adds unit tests for ERDDAP Catalog

intake_erddap/erddap_cat.py

@@ -1,8 +1,13 @@
 """Catalog implementation for intake-erddap."""
+from typing import Optional, Type
+
+import pandas as pd
+
 from erddapy import ERDDAP
 from intake.catalog.base import Catalog
 from intake.catalog.local import LocalCatalogEntry
 
+from .erddap import ERDDAPSource
 from .version import __version__
 
 
@@ -18,34 +23,51 @@ class ERDDAPCatalog(Catalog):
     name = "erddap_cat"
     version = __version__
 
-    def __init__(self, server, kwargs_search=None, **kwargs):
+    def __init__(
+        self,
+        server: str,
+        kwargs_search: Optional[dict] = None,
+        erddap_client: Optional[Type[ERDDAP]] = None,
+        **kwargs
+    ):
+        """ERDDAPCatalog initialization
+
+        Arguments
+        ---------
+
+            server
+        """
+        self._erddap_client = erddap_client or ERDDAP
         self.server = server
         self.kwargs_search = kwargs_search
         super(ERDDAPCatalog, self).__init__(**kwargs)
 
-    def _load(self):
-
-        from .erddap import ERDDAPSource  # , ERDDAPSourceAutoPartition
-
-        e = ERDDAP(self.server)
-        e.protocol = "tabledap"
-        e.dataset_id = "allDatasets"
-
+    def _load_df(self):
+        e = self.get_client()
         if self.kwargs_search is not None:
             search_url = e.get_search_url(
                 response="csv",
                 **self.kwargs_search,
                 # variableName=variable,
                 items_per_page=100000,
             )
-            import pandas as pd
-
             df = pd.read_csv(search_url)
-            dataidkey = "Dataset ID"
+            df.rename(columns={"Dataset ID": "datasetID"}, inplace=True)
+            return df
 
-        else:
-            df = e.to_pandas()
-            dataidkey = "datasetID"
+        return e.to_pandas()
+
+    def get_client(self) -> ERDDAP:
+        """Return an initialized ERDDAP Client."""
+        e = self._erddap_client(self.server)
+        e.protocol = "tabledap"
+        e.dataset_id = "allDatasets"
+        return e
+
+    def _load(self):
+        dataidkey = "datasetID"
+        e = self.get_client()
+        df = self._load_df()
 
         self._entries = {}
 

tests/test_erddap_cat.py

@@ -1,7 +1,60 @@
 #!/usr/bin/env pytest
 """Unit tests."""
+from unittest import mock
+
+import intake
+import pandas as pd
+import pytest
+
+from intake_erddap.erddap_cat import ERDDAPCatalog
 
 
 def test_nothing():
     """This test exists to ensure that at least one test works."""
     pass
+
+
+@mock.patch("erddapy.ERDDAP.to_pandas")
+def test_erddap_catalog(mock_to_pandas):
+    results = pd.DataFrame()
+    results["datasetID"] = ["abc123"]
+    mock_to_pandas.return_value = results
+    server = "http://erddap.invalid/erddap"
+    cat = ERDDAPCatalog(server=server)
+    assert list(cat) == ["abc123"]
+
+
+@mock.patch("pandas.read_csv")
+def test_erddap_catalog_searching(mock_read_csv):
+    results = pd.DataFrame()
+    results["datasetID"] = ["abc123"]
+    mock_read_csv.return_value = results
+    kw = {
+        "min_lon": -180,
+        "max_lon": -156,
+        "min_lat": 50,
+        "max_lat": 66,
+        "min_time": "2021-4-1",
+        "max_time": "2021-4-2",
+    }
+    server = "http://erddap.invalid/erddap"
+    cat = ERDDAPCatalog(server=server, kwargs_search=kw)
+    assert list(cat) == ["abc123"]
+
+
+@pytest.mark.integration
+def test_ioos_erddap_catalog_and_source():
+    kw = {
+        "min_lon": -180,
+        "max_lon": -156,
+        "min_lat": 50,
+        "max_lat": 66,
+        "min_time": "2021-4-1",
+        "max_time": "2021-4-2",
+    }
+    server = "http://erddap.sensors.ioos.us/erddap"
+    cat_sensors = intake.open_erddap_cat(server, kwargs_search=kw)
+    df = cat_sensors[list(cat_sensors)[0]].read()
+    assert df is not None
+    assert isinstance(df, pd.DataFrame)
+    assert len(df) > 0