feat: Better Errors in SQL Lab (#15432)

* snowflake errors

* added big query

* added to setup error messages, first test

* all big query testing added

* added snowflake test

* added syntax error

* added syntax errors to most used databases

docs/src/pages/docs/Miscellaneous/issue_codes.mdx

@@ -271,3 +271,21 @@ One or more parameters specified in the query are malformatted.
 ```
 
 The query contains one or more malformed template parameters. Please check your query and confirm that all template parameters are surround by double braces, for example, "{{ ds }}". Then, try running your query again.
+The object does not exist in this database.
+```
+
+## Issue 1029
+
+```
+The object does not exist in this database.
+```
+
+Either the schema, column, or table do not exist in the database.
+
+## Issue 1030
+
+```
+The query potentially has a syntax error.
+```
+
+The query might have a syntax error. Please check and run again.

superset-frontend/src/components/ErrorMessage/types.ts

@@ -39,6 +39,8 @@ export const ErrorTypeEnum = {
   CONNECTION_DATABASE_PERMISSIONS_ERROR:
     'CONNECTION_DATABASE_PERMISSIONS_ERROR',
   CONNECTION_MISSING_PARAMETERS_ERRORS: 'CONNECTION_MISSING_PARAMETERS_ERRORS',
+  OBJECT_DOES_NOT_EXIST_ERROR: 'OBJECT_DOES_NOT_EXIST_ERROR',
+  SYNTAX_ERROR: 'SYNTAX_ERROR',
 
   // Viz errors
   VIZ_GET_DF_ERROR: 'VIZ_GET_DF_ERROR',

superset-frontend/src/setup/setupErrorMessages.ts

@@ -107,6 +107,14 @@ export default function setupErrorMessages() {
     ErrorTypeEnum.SCHEMA_DOES_NOT_EXIST_ERROR,
     DatabaseErrorMessage,
   );
+  errorMessageComponentRegistry.registerValue(
+    ErrorTypeEnum.OBJECT_DOES_NOT_EXIST_ERROR,
+    DatabaseErrorMessage,
+  );
+  errorMessageComponentRegistry.registerValue(
+    ErrorTypeEnum.SYNTAX_ERROR,
+    DatabaseErrorMessage,
+  );
   errorMessageComponentRegistry.registerValue(
     ErrorTypeEnum.CONNECTION_DATABASE_PERMISSIONS_ERROR,
     DatabaseErrorMessage,

superset/db_engine_specs/athena.py

@@ -14,12 +14,20 @@
 # KIND, either express or implied.  See the License for the
 # specific language governing permissions and limitations
 # under the License.
+import re
 from datetime import datetime
-from typing import Optional
+from typing import Any, Dict, Optional, Pattern, Tuple
+
+from flask_babel import gettext as __
 
 from superset.db_engine_specs.base import BaseEngineSpec
+from superset.errors import SupersetErrorType
 from superset.utils import core as utils
 
+SYNTAX_ERROR_REGEX = re.compile(
+    ": mismatched input '(?P<syntax_error>.*?)'. Expecting: "
+)
+
 
 class AthenaEngineSpec(BaseEngineSpec):
     engine = "awsathena"
@@ -41,6 +49,17 @@ class AthenaEngineSpec(BaseEngineSpec):
                                     date_add('day', 1, CAST({col} AS TIMESTAMP))))",
     }
 
+    custom_errors: Dict[Pattern[str], Tuple[str, SupersetErrorType, Dict[str, Any]]] = {
+        SYNTAX_ERROR_REGEX: (
+            __(
+                "Please check your query for syntax errors at or "
+                'near "%(syntax_error)s". Then, try running your query again.'
+            ),
+            SupersetErrorType.SYNTAX_ERROR,
+            {},
+        ),
+    }
+
     @classmethod
     def convert_dttm(cls, target_type: str, dttm: datetime) -> Optional[str]:
         tt = target_type.upper()

superset/db_engine_specs/bigquery.py

@@ -44,6 +44,24 @@
     + "permission in project (?P<project>.+?)"
 )
 
+TABLE_DOES_NOT_EXIST_REGEX = re.compile(
+    'Table name "(?P<table>.*?)" missing dataset while no default '
+    "dataset is set in the request"
+)
+
+COLUMN_DOES_NOT_EXIST_REGEX = re.compile(
+    r"Unrecognized name: (?P<column>.*?) at \[(?P<location>.+?)\]"
+)
+
+SCHEMA_DOES_NOT_EXIST_REGEX = re.compile(
+    r"bigquery error: 404 Not found: Dataset (?P<dataset>.*?):"
+    r"(?P<schema>.*?) was not found in location"
+)
+
+SYNTAX_ERROR_REGEX = re.compile(
+    'Syntax error: Expected end of input but got identifier "(?P<syntax_error>.+?)"'
+)
+
 ma_plugin = MarshmallowPlugin()
 
 
@@ -127,6 +145,35 @@ class BigQueryEngineSpec(BaseEngineSpec):
             SupersetErrorType.CONNECTION_DATABASE_PERMISSIONS_ERROR,
             {},
         ),
+        TABLE_DOES_NOT_EXIST_REGEX: (
+            __(
+                'The table "%(table)s" does not exist. '
+                "A valid table must be used to run this query.",
+            ),
+            SupersetErrorType.TABLE_DOES_NOT_EXIST_ERROR,
+            {},
+        ),
+        COLUMN_DOES_NOT_EXIST_REGEX: (
+            __('We can\'t seem to resolve column "%(column)s" at line %(location)s.'),
+            SupersetErrorType.COLUMN_DOES_NOT_EXIST_ERROR,
+            {},
+        ),
+        SCHEMA_DOES_NOT_EXIST_REGEX: (
+            __(
+                'The schema "%(schema)s" does not exist. '
+                "A valid schema must be used to run this query."
+            ),
+            SupersetErrorType.SCHEMA_DOES_NOT_EXIST_ERROR,
+            {},
+        ),
+        SYNTAX_ERROR_REGEX: (
+            __(
+                "Please check your query for syntax errors at or near "
+                '"%(syntax_error)s". Then, try running your query again.'
+            ),
+            SupersetErrorType.SYNTAX_ERROR,
+            {},
+        ),
     }
 
     @classmethod

superset/db_engine_specs/gsheets.py

@@ -14,12 +14,17 @@
 # KIND, either express or implied.  See the License for the
 # specific language governing permissions and limitations
 # under the License.
-from typing import Optional
+import re
+from typing import Any, Dict, Optional, Pattern, Tuple
 
+from flask_babel import gettext as __
 from sqlalchemy.engine.url import URL
 
 from superset import security_manager
 from superset.db_engine_specs.sqlite import SqliteEngineSpec
+from superset.errors import SupersetErrorType
+
+SYNTAX_ERROR_REGEX = re.compile('SQLError: near "(?P<server_error>.*?)": syntax error')
 
 
 class GSheetsEngineSpec(SqliteEngineSpec):
@@ -30,6 +35,17 @@ class GSheetsEngineSpec(SqliteEngineSpec):
     allows_joins = False
     allows_subqueries = True
 
+    custom_errors: Dict[Pattern[str], Tuple[str, SupersetErrorType, Dict[str, Any]]] = {
+        SYNTAX_ERROR_REGEX: (
+            __(
+                'Please check your query for syntax errors near "%(server_error)s". '
+                "Then, try running your query again."
+            ),
+            SupersetErrorType.SYNTAX_ERROR,
+            {},
+        ),
+    }
+
     @classmethod
     def modify_url_for_impersonation(
         cls, url: URL, impersonate_user: bool, username: Optional[str]

superset/db_engine_specs/mysql.py

@@ -52,6 +52,11 @@
 )
 CONNECTION_UNKNOWN_DATABASE_REGEX = re.compile("Unknown database '(?P<database>.*?)'")
 
+SYNTAX_ERROR_REGEX = re.compile(
+    "check the manual that corresponds to your MySQL server "
+    "version for the right syntax to use near '(?P<server_error>.*)"
+)
+
 
 class MySQLEngineSpec(BaseEngineSpec, BasicParametersMixin):
     engine = "mysql"
@@ -134,6 +139,14 @@ class MySQLEngineSpec(BaseEngineSpec, BasicParametersMixin):
             SupersetErrorType.CONNECTION_UNKNOWN_DATABASE_ERROR,
             {"invalid": ["database"]},
         ),
+        SYNTAX_ERROR_REGEX: (
+            __(
+                'Please check your query for syntax errors near "%(server_error)s". '
+                "Then, try running your query again."
+            ),
+            SupersetErrorType.SYNTAX_ERROR,
+            {},
+        ),
     }
 
     @classmethod

superset/db_engine_specs/postgres.py

@@ -85,6 +85,8 @@ class FixedOffsetTimezone(_FixedOffset):
     r"does not exist\s+LINE (?P<location>\d+?)"
 )
 
+SYNTAX_ERROR_REGEX = re.compile('syntax error at or near "(?P<syntax_error>.*?)"')
+
 
 class PostgresBaseEngineSpec(BaseEngineSpec):
     """ Abstract class for Postgres 'like' databases """
@@ -151,6 +153,14 @@ class PostgresBaseEngineSpec(BaseEngineSpec):
             SupersetErrorType.COLUMN_DOES_NOT_EXIST_ERROR,
             {},
         ),
+        SYNTAX_ERROR_REGEX: (
+            __(
+                "Please check your query for syntax errors at or "
+                'near "%(syntax_error)s". Then, try running your query again.'
+            ),
+            SupersetErrorType.SYNTAX_ERROR,
+            {},
+        ),
     }
 
     @classmethod

superset/db_engine_specs/snowflake.py

@@ -15,18 +15,31 @@
 # specific language governing permissions and limitations
 # under the License.
 import json
+import re
 from datetime import datetime
-from typing import Optional, TYPE_CHECKING
+from typing import Any, Dict, Optional, Pattern, Tuple, TYPE_CHECKING
 from urllib import parse
 
+from flask_babel import gettext as __
 from sqlalchemy.engine.url import URL
 
 from superset.db_engine_specs.postgres import PostgresBaseEngineSpec
+from superset.errors import SupersetErrorType
 from superset.utils import core as utils
 
 if TYPE_CHECKING:
     from superset.models.core import Database
 
+# Regular expressions to catch custom errors
+OBJECT_DOES_NOT_EXIST_REGEX = re.compile(
+    r"Object (?P<object>.*?) does not exist or not authorized."
+)
+
+SYNTAX_ERROR_REGEX = re.compile(
+    "syntax error line (?P<line>.+?) at position (?P<position>.+?) "
+    "unexpected '(?P<syntax_error>.+?)'."
+)
+
 
 class SnowflakeEngineSpec(PostgresBaseEngineSpec):
     engine = "snowflake"
@@ -54,6 +67,22 @@ class SnowflakeEngineSpec(PostgresBaseEngineSpec):
         "P1Y": "DATE_TRUNC('YEAR', {col})",
     }
 
+    custom_errors: Dict[Pattern[str], Tuple[str, SupersetErrorType, Dict[str, Any]]] = {
+        OBJECT_DOES_NOT_EXIST_REGEX: (
+            __("%(object)s does not exist in this database."),
+            SupersetErrorType.OBJECT_DOES_NOT_EXIST_ERROR,
+            {},
+        ),
+        SYNTAX_ERROR_REGEX: (
+            __(
+                "Please check your query for syntax errors at or "
+                'near "%(syntax_error)s". Then, try running your query again.'
+            ),
+            SupersetErrorType.SYNTAX_ERROR,
+            {},
+        ),
+    }
+
     @classmethod
     def adjust_database_uri(
         cls, uri: URL, selected_schema: Optional[str] = None

superset/errors.py

@@ -49,6 +49,8 @@ class SupersetErrorType(str, Enum):
     CONNECTION_UNKNOWN_DATABASE_ERROR = "CONNECTION_UNKNOWN_DATABASE_ERROR"
     CONNECTION_DATABASE_PERMISSIONS_ERROR = "CONNECTION_DATABASE_PERMISSIONS_ERROR"
     CONNECTION_MISSING_PARAMETERS_ERROR = "CONNECTION_MISSING_PARAMETERS_ERROR"
+    OBJECT_DOES_NOT_EXIST_ERROR = "OBJECT_DOES_NOT_EXIST_ERROR"
+    SYNTAX_ERROR = "SYNTAX_ERROR"
 
     # Viz errors
     VIZ_GET_DF_ERROR = "VIZ_GET_DF_ERROR"
@@ -320,6 +322,17 @@ class SupersetErrorType(str, Enum):
             ),
         },
     ],
+    SupersetErrorType.OBJECT_DOES_NOT_EXIST_ERROR: [
+        {
+            "code": 1029,
+            "message": _(
+                "Issue 1029 - The object does not exist in the given database."
+            ),
+        },
+    ],
+    SupersetErrorType.SYNTAX_ERROR: [
+        {"code": 1030, "message": _("Issue 1029 - The query has a syntax error."),},
+    ],
 }
 
 

tests/db_engine_specs/athena_tests.py

@@ -15,6 +15,7 @@
 # specific language governing permissions and limitations
 # under the License.
 from superset.db_engine_specs.athena import AthenaEngineSpec
+from superset.errors import ErrorLevel, SupersetError, SupersetErrorType
 from tests.db_engine_specs.base_tests import TestDbEngineSpec
 
 
@@ -31,3 +32,26 @@ def test_convert_dttm(self):
             AthenaEngineSpec.convert_dttm("TIMESTAMP", dttm),
             "from_iso8601_timestamp('2019-01-02T03:04:05.678900')",
         )
+
+    def test_extract_errors(self):
+        """
+        Test that custom error messages are extracted correctly.
+        """
+        msg = ": mismatched input 'fromm'. Expecting: "
+        result = AthenaEngineSpec.extract_errors(Exception(msg))
+        assert result == [
+            SupersetError(
+                message='Please check your query for syntax errors at or near "fromm". Then, try running your query again.',
+                error_type=SupersetErrorType.SYNTAX_ERROR,
+                level=ErrorLevel.ERROR,
+                extra={
+                    "engine_name": "Amazon Athena",
+                    "issue_codes": [
+                        {
+                            "code": 1030,
+                            "message": "Issue 1030 - The query has a syntax error.",
+                        }
+                    ],
+                },
+            )
+        ]