From 10a46fd33c86146eef0bb388b8002330fc829981 Mon Sep 17 00:00:00 2001
From: Eric Radman <eradman@starfishstorage.com>
Date: Wed, 15 May 2024 23:28:42 -0400
Subject: [PATCH] Revert "show pg and athena column comments and table
 descriptions as antd tooltip if they are defined (#6582)" (#6971)

This reverts commit c12d45077a992e9d5d21bc891d4961c9a2ff73f6.

This commit did not sort tables properly by schema, then name
---
 .../app/components/queries/SchemaBrowser.jsx  | 74 ++++++-------------
 redash/models/__init__.py                     | 11 +--
 redash/query_runner/athena.py                 | 39 ++--------
 redash/query_runner/pg.py                     | 18 +----
 tests/models/test_data_sources.py             | 20 ++---
 tests/query_runner/test_athena.py             | 23 +-----
 6 files changed, 43 insertions(+), 142 deletions(-)

diff --git a/client/app/components/queries/SchemaBrowser.jsx b/client/app/components/queries/SchemaBrowser.jsx
index 730f4acf88..ed8d2402ed 100644
--- a/client/app/components/queries/SchemaBrowser.jsx
+++ b/client/app/components/queries/SchemaBrowser.jsx
@@ -16,7 +16,6 @@ import LoadingState from "../items-list/components/LoadingState";
 const SchemaItemColumnType = PropTypes.shape({
   name: PropTypes.string.isRequired,
   type: PropTypes.string,
-  comment: PropTypes.string,
 });
 
 export const SchemaItemType = PropTypes.shape({
@@ -48,30 +47,13 @@ function SchemaItem({ item, expanded, onToggle, onSelect, ...props }) {
   return (
     <div {...props}>
       <div className="schema-list-item">
-        {item.description ? (
-          <Tooltip
-            title={item.description}
-            mouseEnterDelay={0}
-            mouseLeaveDelay={0}
-            placement="right"
-            arrowPointAtCenter>
-            <PlainButton className="table-name" onClick={onToggle}>
-              <i className="fa fa-table m-r-5" aria-hidden="true" />
-              <strong>
-                <span title={item.name}>{tableDisplayName}</span>
-                {!isNil(item.size) && <span> ({item.size})</span>}
-              </strong>
-            </PlainButton>
-          </Tooltip>
-        ) : (
-          <PlainButton className="table-name" onClick={onToggle}>
-            <i className="fa fa-table m-r-5" aria-hidden="true" />
-            <strong>
-              <span title={item.name}>{tableDisplayName}</span>
-              {!isNil(item.size) && <span> ({item.size})</span>}
-            </strong>
-          </PlainButton>
-        )}
+        <PlainButton className="table-name" onClick={onToggle}>
+          <i className="fa fa-table m-r-5" aria-hidden="true" />
+          <strong>
+            <span title={item.name}>{tableDisplayName}</span>
+            {!isNil(item.size) && <span> ({item.size})</span>}
+          </strong>
+        </PlainButton>
         <Tooltip
           title="Insert table name into query text"
           mouseEnterDelay={0}
@@ -91,34 +73,22 @@ function SchemaItem({ item, expanded, onToggle, onSelect, ...props }) {
             map(item.columns, column => {
               const columnName = get(column, "name");
               const columnType = get(column, "type");
-              const columnComment = get(column, "comment");
-              if (columnComment) {
-                return (
-                  <Tooltip title={columnComment} mouseEnterDelay={0} mouseLeaveDelay={0} placement="rightTop">
-                    <PlainButton
-                      key={columnName}
-                      className="table-open-item"
-                      onClick={e => handleSelect(e, columnName)}>
-                      <div>
-                        {columnName} {columnType && <span className="column-type">{columnType}</span>}
-                      </div>
-
-                      <div className="copy-to-editor">
-                        <i className="fa fa-angle-double-right" aria-hidden="true" />
-                      </div>
-                    </PlainButton>
-                  </Tooltip>
-                );
-              }
               return (
-                <PlainButton key={columnName} className="table-open-item" onClick={e => handleSelect(e, columnName)}>
-                  <div>
-                    {columnName} {columnType && <span className="column-type">{columnType}</span>}
-                  </div>
-                  <div className="copy-to-editor">
-                    <i className="fa fa-angle-double-right" aria-hidden="true" />
-                  </div>
-                </PlainButton>
+                <Tooltip
+                  title="Insert column name into query text"
+                  mouseEnterDelay={0}
+                  mouseLeaveDelay={0}
+                  placement="rightTop">
+                  <PlainButton key={columnName} className="table-open-item" onClick={e => handleSelect(e, columnName)}>
+                    <div>
+                      {columnName} {columnType && <span className="column-type">{columnType}</span>}
+                    </div>
+
+                    <div className="copy-to-editor">
+                      <i className="fa fa-angle-double-right" aria-hidden="true" />
+                    </div>
+                  </PlainButton>
+                </Tooltip>
               );
             })
           )}
diff --git a/redash/models/__init__.py b/redash/models/__init__.py
index 7ee0d99741..91ddd91ee3 100644
--- a/redash/models/__init__.py
+++ b/redash/models/__init__.py
@@ -227,16 +227,7 @@ def get_schema(self, refresh=False):
 
     def _sort_schema(self, schema):
         return [
-            {
-                "name": i["name"],
-                "description": i.get("description"),
-                "columns": sorted(
-                    i["columns"],
-                    key=lambda col: (
-                        ("partition" in col["type"], col.get("idx", 0), col["name"]) if isinstance(col, dict) else col
-                    ),
-                ),
-            }
+            {"name": i["name"], "columns": sorted(i["columns"], key=lambda x: x["name"] if isinstance(x, dict) else x)}
             for i in sorted(schema, key=lambda x: x["name"])
         ]
 
diff --git a/redash/query_runner/athena.py b/redash/query_runner/athena.py
index 0a5f648417..acde734f02 100644
--- a/redash/query_runner/athena.py
+++ b/redash/query_runner/athena.py
@@ -21,9 +21,7 @@
 
 try:
     import boto3
-    import pandas as pd
     import pyathena
-    from pyathena.pandas_cursor import PandasCursor
 
     enabled = True
 except ImportError:
@@ -190,35 +188,10 @@ def __get_schema_from_glue(self):
                         logger.warning("Glue table doesn't have StorageDescriptor: %s", table_name)
                         continue
                     if table_name not in schema:
-                        columns = []
-                        for cols in table["StorageDescriptor"]["Columns"]:
-                            c = {
-                                "name": cols["Name"],
-                            }
-                            if "Type" in cols:
-                                c["type"] = cols["Type"]
-                            if "Comment" in cols:
-                                c["comment"] = cols["Comment"]
-                            columns.append(c)
-
-                        schema[table_name] = {
-                            "name": table_name,
-                            "columns": columns,
-                            "description": table.get("Description"),
-                        }
-                        for idx, partition in enumerate(table.get("PartitionKeys", [])):
-                            schema[table_name]["columns"].append(
-                                {
-                                    "name": partition["Name"],
-                                    "type": "partition",
-                                    "idx": idx,
-                                }
-                            )
-                            if "Type" in partition:
-                                _type = partition["Type"]
-                                c["type"] = f"partition ({_type})"
-                            if "Comment" in partition:
-                                c["comment"] = partition["Comment"]
+                        column = [columns["Name"] for columns in table["StorageDescriptor"]["Columns"]]
+                        schema[table_name] = {"name": table_name, "columns": column}
+                        for partition in table.get("PartitionKeys", []):
+                            schema[table_name]["columns"].append(partition["Name"])
         return list(schema.values())
 
     def get_schema(self, get_stats=False):
@@ -252,7 +225,6 @@ def run_query(self, query, user):
             kms_key=self.configuration.get("kms_key", None),
             work_group=self.configuration.get("work_group", "primary"),
             formatter=SimpleFormatter(),
-            cursor_class=PandasCursor,
             **self._get_iam_credentials(user=user),
         ).cursor()
 
@@ -260,8 +232,7 @@ def run_query(self, query, user):
             cursor.execute(query)
             column_tuples = [(i[0], _TYPE_MAPPINGS.get(i[1], None)) for i in cursor.description]
             columns = self.fetch_columns(column_tuples)
-            df = cursor.as_pandas().replace({pd.NA: None})
-            rows = df.to_dict(orient="records")
+            rows = [dict(zip(([c["name"] for c in columns]), r)) for i, r in enumerate(cursor.fetchall())]
             qbytes = None
             athena_query_id = None
             try:
diff --git a/redash/query_runner/pg.py b/redash/query_runner/pg.py
index 46bf3cc0b8..dc74aff2cc 100644
--- a/redash/query_runner/pg.py
+++ b/redash/query_runner/pg.py
@@ -108,8 +108,6 @@ def build_schema(query_result, schema):
         column = row["column_name"]
         if row.get("data_type") is not None:
             column = {"name": row["column_name"], "type": row["data_type"]}
-            if "column_comment" in row:
-                column["comment"] = row["column_comment"]
 
         schema[table_name]["columns"].append(column)
 
@@ -224,9 +222,7 @@ def _get_tables(self, schema):
         SELECT s.nspname as table_schema,
                c.relname as table_name,
                a.attname as column_name,
-               null as data_type,
-               null as column_comment,
-               null as idx
+               null as data_type
         FROM pg_class c
         JOIN pg_namespace s
         ON c.relnamespace = s.oid
@@ -242,16 +238,8 @@ def _get_tables(self, schema):
         SELECT table_schema,
                table_name,
                column_name,
-               data_type,
-               pgd.description,
-               isc.ordinal_position
-        FROM information_schema.columns as isc
-        LEFT JOIN pg_catalog.pg_statio_all_tables as st
-        ON isc.table_schema = st.schemaname
-        AND isc.table_name = st.relname
-        LEFT JOIN pg_catalog.pg_description pgd
-        ON pgd.objoid=st.relid
-        AND pgd.objsubid=isc.ordinal_position
+               data_type
+        FROM information_schema.columns
         WHERE table_schema NOT IN ('pg_catalog', 'information_schema')
         """
 
diff --git a/tests/models/test_data_sources.py b/tests/models/test_data_sources.py
index 82bd4b5a7a..7db4d463a3 100644
--- a/tests/models/test_data_sources.py
+++ b/tests/models/test_data_sources.py
@@ -8,7 +8,7 @@
 
 class DataSourceTest(BaseTestCase):
     def test_get_schema(self):
-        return_value = [{"name": "table", "columns": [], "description": None}]
+        return_value = [{"name": "table", "columns": []}]
 
         with mock.patch("redash.query_runner.pg.PostgreSQL.get_schema") as patched_get_schema:
             patched_get_schema.return_value = return_value
@@ -18,7 +18,7 @@ def test_get_schema(self):
             self.assertEqual(return_value, schema)
 
     def test_get_schema_uses_cache(self):
-        return_value = [{"name": "table", "columns": [], "description": None}]
+        return_value = [{"name": "table", "columns": []}]
         with mock.patch("redash.query_runner.pg.PostgreSQL.get_schema") as patched_get_schema:
             patched_get_schema.return_value = return_value
 
@@ -29,12 +29,12 @@ def test_get_schema_uses_cache(self):
             self.assertEqual(patched_get_schema.call_count, 1)
 
     def test_get_schema_skips_cache_with_refresh_true(self):
-        return_value = [{"name": "table", "columns": [], "description": None}]
+        return_value = [{"name": "table", "columns": []}]
         with mock.patch("redash.query_runner.pg.PostgreSQL.get_schema") as patched_get_schema:
             patched_get_schema.return_value = return_value
 
             self.factory.data_source.get_schema()
-            new_return_value = [{"name": "new_table", "columns": [], "description": None}]
+            new_return_value = [{"name": "new_table", "columns": []}]
             patched_get_schema.return_value = new_return_value
             schema = self.factory.data_source.get_schema(refresh=True)
 
@@ -43,11 +43,10 @@ def test_get_schema_skips_cache_with_refresh_true(self):
 
     def test_schema_sorter(self):
         input_data = [
-            {"name": "zoo", "columns": ["is_zebra", "is_snake", "is_cow"], "description": None},
+            {"name": "zoo", "columns": ["is_zebra", "is_snake", "is_cow"]},
             {
                 "name": "all_terain_vehicle",
                 "columns": ["has_wheels", "has_engine", "has_all_wheel_drive"],
-                "description": None,
             },
         ]
 
@@ -55,9 +54,8 @@ def test_schema_sorter(self):
             {
                 "name": "all_terain_vehicle",
                 "columns": ["has_all_wheel_drive", "has_engine", "has_wheels"],
-                "description": None,
             },
-            {"name": "zoo", "columns": ["is_cow", "is_snake", "is_zebra"], "description": None},
+            {"name": "zoo", "columns": ["is_cow", "is_snake", "is_zebra"]},
         ]
 
         real_output = self.factory.data_source._sort_schema(input_data)
@@ -66,11 +64,10 @@ def test_schema_sorter(self):
 
     def test_model_uses_schema_sorter(self):
         orig_schema = [
-            {"name": "zoo", "columns": ["is_zebra", "is_snake", "is_cow"], "description": None},
+            {"name": "zoo", "columns": ["is_zebra", "is_snake", "is_cow"]},
             {
                 "name": "all_terain_vehicle",
                 "columns": ["has_wheels", "has_engine", "has_all_wheel_drive"],
-                "description": None,
             },
         ]
 
@@ -78,9 +75,8 @@ def test_model_uses_schema_sorter(self):
             {
                 "name": "all_terain_vehicle",
                 "columns": ["has_all_wheel_drive", "has_engine", "has_wheels"],
-                "description": None,
             },
-            {"name": "zoo", "columns": ["is_cow", "is_snake", "is_zebra"], "description": None},
+            {"name": "zoo", "columns": ["is_cow", "is_snake", "is_zebra"]},
         ]
 
         with mock.patch("redash.query_runner.pg.PostgreSQL.get_schema") as patched_get_schema:
diff --git a/tests/query_runner/test_athena.py b/tests/query_runner/test_athena.py
index 2c386ead11..2ac4ee42fe 100644
--- a/tests/query_runner/test_athena.py
+++ b/tests/query_runner/test_athena.py
@@ -75,9 +75,7 @@ def test_external_table(self):
             {"DatabaseName": "test1"},
         )
         with self.stubber:
-            assert query_runner.get_schema() == [
-                {"columns": [{"name": "row_id", "type": "int"}], "name": "test1.jdbc_table", "description": None}
-            ]
+            assert query_runner.get_schema() == [{"columns": ["row_id"], "name": "test1.jdbc_table"}]
 
     def test_partitioned_table(self):
         """
@@ -126,16 +124,7 @@ def test_partitioned_table(self):
             {"DatabaseName": "test1"},
         )
         with self.stubber:
-            assert query_runner.get_schema() == [
-                {
-                    "columns": [
-                        {"name": "sk", "type": "partition (int)"},
-                        {"name": "category", "type": "partition", "idx": 0},
-                    ],
-                    "name": "test1.partitioned_table",
-                    "description": None,
-                }
-            ]
+            assert query_runner.get_schema() == [{"columns": ["sk", "category"], "name": "test1.partitioned_table"}]
 
     def test_view(self):
         query_runner = Athena({"glue": True, "region": "mars-east-1"})
@@ -167,9 +156,7 @@ def test_view(self):
             {"DatabaseName": "test1"},
         )
         with self.stubber:
-            assert query_runner.get_schema() == [
-                {"columns": [{"name": "sk", "type": "int"}], "name": "test1.view", "description": None}
-            ]
+            assert query_runner.get_schema() == [{"columns": ["sk"], "name": "test1.view"}]
 
     def test_dodgy_table_does_not_break_schema_listing(self):
         """
@@ -209,9 +196,7 @@ def test_dodgy_table_does_not_break_schema_listing(self):
             {"DatabaseName": "test1"},
         )
         with self.stubber:
-            assert query_runner.get_schema() == [
-                {"columns": [{"name": "region", "type": "string"}], "name": "test1.csv", "description": None}
-            ]
+            assert query_runner.get_schema() == [{"columns": ["region"], "name": "test1.csv"}]
 
     def test_no_storage_descriptor_table(self):
         """