数据探索可视化平台：一个的现代数据探索和可视化平台

<!-- Licensed to the Apache Software Foundation (ASF) under one or more contributor license agreements. See the NOTICE file distributed with this work for additional information regarding copyright ownership. The ASF licenses this file to you under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License. --> # Database engine specifications Superset uses [SQLAlchemy](https://www.sqlalchemy.org/) as an abstraction layer for running queries and fetching metadata from tables (like column names and types). Unfortunately, while SQLAlchemy offers enough functionality to allow connecting Superset to dozens of databases, there are still implementation details that differ across them. Because of this, Superset has an additional abstraction on top of SQLAlchemy, called a "database engine specification" or, simply, "DB engine spec". DB engine specs were created initially because there's no SQL standard for computing aggregations at different time grains. For example, to compute a daily metric in Trino or Postgres we could run a query like this: ```sql SELECT date_trunc('day', CAST(time_column) AS TIMESTAMP) AS day, COUNT(*) AS metric FROM some_table GROUP BY 1 ``` For MySQL, instead of using the `date_trunc` function, we would need to write: ```sql SELECT DATE(time_column) AS day, COUNT(*) AS metric FROM some_table GROUP BY 1 ``` Over time, more and more functionality was added to DB engine specs, including validating SQL, estimating the cost of queries before they are run, and understanding the semantics of error messages. These are all described in detail in this document, and in the table below you can see a summary of what features are supported by each database. Note that DB engine specs are completely optional. Superset can connect to any database supported by SQLAlchemy (or 3rd party dialects) even if there's no DB engine spec associated with it. But DB engine specs greatly improve the experience of working with a database in Superset. ## Features The table below (generated via `python superset/db_engine_specs/lib.py`) summarizes the information about the status of all DB engine specs in Superset (note that this excludes 3rd party DB engine specs): | Feature | Amazon Athena | Amazon DynamoDB | Amazon Redshift | Apache Drill | Apache Druid | Apache Hive | Apache Impala | Apache Kylin | Apache Pinot | Apache Solr | Apache Spark SQL | Ascend | Aurora MySQL (Data API) | Aurora PostgreSQL (Data API) | Azure Synapse | ClickHouse | ClickHouse Connect (Superset) | CockroachDB | CrateDB | Databricks | Databricks Interactive Cluster | Databricks SQL Endpoint | Dremio | DuckDB | ElasticSearch (OpenDistro SQL) | ElasticSearch (SQL API) | Exasol | Firebird | Firebolt | Google BigQuery | Google Sheets | IBM Db2 | IBM Netezza Performance Server | KustoKQL | KustoSQL | Microsoft SQL Server | MySQL | Ocient | Oracle | PostgreSQL | Presto | RisingWave | Rockset | SAP HANA | SQLite | Shillelagh | Snowflake | StarRocks | Teradata | Trino | Vertica | base | | ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| ---| | Module | superset.db_engine_specs.athena | superset.db_engine_specs.dynamodb | superset.db_engine_specs.redshift | superset.db_engine_specs.drill | superset.db_engine_specs.druid | superset.db_engine_specs.hive | superset.db_engine_specs.impala | superset.db_engine_specs.kylin | superset.db_engine_specs.pinot | superset.db_engine_specs.solr | superset.db_engine_specs.spark | superset.db_engine_specs.ascend | superset.db_engine_specs.aurora | superset.db_engine_specs.aurora | superset.db_engine_specs.mssql | superset.db_engine_specs.clickhouse | superset.db_engine_specs.clickhouse | superset.db_engine_specs.cockroachdb | superset.db_engine_specs.crate | superset.db_engine_specs.databricks | superset.db_engine_specs.databricks | superset.db_engine_specs.databricks | superset.db_engine_specs.dremio | superset.db_engine_specs.duckdb | superset.db_engine_specs.elasticsearch | superset.db_engine_specs.elasticsearch | superset.db_engine_specs.exasol | superset.db_engine_specs.firebird | superset.db_engine_specs.firebolt | superset.db_engine_specs.bigquery | superset.db_engine_specs.gsheets | superset.db_engine_specs.db2 | superset.db_engine_specs.netezza | superset.db_engine_specs.kusto | superset.db_engine_specs.kusto | superset.db_engine_specs.mssql | superset.db_engine_specs.mysql | superset.db_engine_specs.ocient | superset.db_engine_specs.oracle | superset.db_engine_specs.postgres | superset.db_engine_specs.presto | superset.db_engine_specs.risingwave | superset.db_engine_specs.rockset | superset.db_engine_specs.hana | superset.db_engine_specs.sqlite | superset.db_engine_specs.shillelagh | superset.db_engine_specs.snowflake | superset.db_engine_specs.starrocks | superset.db_engine_specs.teradata | superset.db_engine_specs.trino | superset.db_engine_specs.vertica | superset.db_engine_specs.presto | | Method used to limit the rows in the subquery | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | WRAP_SQL | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FETCH_MANY | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | WRAP_SQL | FORCE_LIMIT | WRAP_SQL | WRAP_SQL | WRAP_SQL | FORCE_LIMIT | FORCE_LIMIT | WRAP_SQL | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | WRAP_SQL | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | WRAP_SQL | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | | Supports JOINs | True | True | True | True | False | True | True | True | False | False | True | True | True | True | True | True | True | True | True | True | True | True | True | True | False | False | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | | Supports subqueries | True | True | True | True | True | True | True | True | False | False | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | | Allows aliases in the SELECT statement | True | True | True | True | True | True | True | True | False | True | True | True | True | True | True | True | True | True | True | True | True | True | False | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | | Allows referencing aliases in the ORDER BY statement | True | True | True | True | True | True | True | True | False | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | True | | Supports secondary t