fix: updates tests based on revised hacker_news tables (#1591)

This fixes four broken tests that failed due to an unexpected change in the Google Public Dataset: Hacker News. The `comments` table was deleted and only the `full` table remained. This edit updates the name of the table in four tests and updates the names of columns in the table as well as updates the expected results for one of the tests. Fixes #1590 🦕
googleapis · Jun 21, 2023 · d73cf49 · d73cf49
1 parent 41799b4
commit d73cf49
Show file tree

Hide file tree

Showing 2 changed files with 25 additions and 24 deletions.
diff --git a/tests/system/test_client.py b/tests/system/test_client.py
@@ -1706,8 +1706,8 @@ def test_dbapi_fetch_w_bqstorage_client_large_result_set(self):
 
         cursor.execute(
             """
-            SELECT id, `by`, time_ts
-            FROM `bigquery-public-data.hacker_news.comments`
+            SELECT id, `by`, timestamp
+            FROM `bigquery-public-data.hacker_news.full`
             ORDER BY `id` ASC
             LIMIT 100000
         """
@@ -1717,27 +1717,28 @@ def test_dbapi_fetch_w_bqstorage_client_large_result_set(self):
 
         field_name = operator.itemgetter(0)
         fetched_data = [sorted(row.items(), key=field_name) for row in result_rows]
-
         # Since DB API is not thread safe, only a single result stream should be
         # requested by the BQ storage client, meaning that results should arrive
         # in the sorted order.
+
         expected_data = [
             [
-                ("by", "sama"),
-                ("id", 15),
-                ("time_ts", datetime.datetime(2006, 10, 9, 19, 51, 1, tzinfo=UTC)),
+                ("by", "pg"),
+                ("id", 1),
+                ("timestamp", datetime.datetime(2006, 10, 9, 18, 21, 51, tzinfo=UTC)),
             ],
             [
-                ("by", "pg"),
-                ("id", 17),
-                ("time_ts", datetime.datetime(2006, 10, 9, 19, 52, 45, tzinfo=UTC)),
+                ("by", "phyllis"),
+                ("id", 2),
+                ("timestamp", datetime.datetime(2006, 10, 9, 18, 30, 28, tzinfo=UTC)),
             ],
             [
-                ("by", "pg"),
-                ("id", 22),
-                ("time_ts", datetime.datetime(2006, 10, 10, 2, 18, 22, tzinfo=UTC)),
+                ("by", "phyllis"),
+                ("id", 3),
+                ("timestamp", datetime.datetime(2006, 10, 9, 18, 40, 33, tzinfo=UTC)),
             ],
         ]
+
         self.assertEqual(fetched_data, expected_data)
 
     def test_dbapi_dry_run_query(self):
@@ -1769,8 +1770,8 @@ def test_dbapi_connection_does_not_leak_sockets(self):
 
         cursor.execute(
             """
-            SELECT id, `by`, time_ts
-            FROM `bigquery-public-data.hacker_news.comments`
+            SELECT id, `by`, timestamp
+            FROM `bigquery-public-data.hacker_news.full`
             ORDER BY `id` ASC
             LIMIT 100000
         """

diff --git a/tests/system/test_pandas.py b/tests/system/test_pandas.py
@@ -740,21 +740,21 @@ def test_load_table_from_dataframe_w_explicit_schema_source_format_csv_floats(
 
 def test_query_results_to_dataframe(bigquery_client):
     QUERY = """
-    SELECT id, author, time_ts, dead
-    FROM `bigquery-public-data.hacker_news.comments`
+    SELECT id, `by`, timestamp, dead
+    FROM `bigquery-public-data.hacker_news.full`
     LIMIT 10
     """
 
     df = bigquery_client.query(QUERY).result().to_dataframe()
 
     assert isinstance(df, pandas.DataFrame)
     assert len(df) == 10  # verify the number of rows
-    column_names = ["id", "author", "time_ts", "dead"]
+    column_names = ["id", "by", "timestamp", "dead"]
     assert list(df) == column_names  # verify the column names
     exp_datatypes = {
         "id": int,
-        "author": str,
-        "time_ts": pandas.Timestamp,
+        "by": str,
+        "timestamp": pandas.Timestamp,
         "dead": bool,
     }
     for _, row in df.iterrows():
@@ -766,8 +766,8 @@ def test_query_results_to_dataframe(bigquery_client):
 
 def test_query_results_to_dataframe_w_bqstorage(bigquery_client):
     query = """
-    SELECT id, author, time_ts, dead
-    FROM `bigquery-public-data.hacker_news.comments`
+    SELECT id, `by`, timestamp, dead
+    FROM `bigquery-public-data.hacker_news.full`
     LIMIT 10
     """
 
@@ -779,12 +779,12 @@ def test_query_results_to_dataframe_w_bqstorage(bigquery_client):
 
     assert isinstance(df, pandas.DataFrame)
     assert len(df) == 10  # verify the number of rows
-    column_names = ["id", "author", "time_ts", "dead"]
+    column_names = ["id", "by", "timestamp", "dead"]
     assert list(df) == column_names
     exp_datatypes = {
         "id": int,
-        "author": str,
-        "time_ts": pandas.Timestamp,
+        "by": str,
+        "timestamp": pandas.Timestamp,
         "dead": bool,
     }
     for index, row in df.iterrows():