You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@superset.apache.org by be...@apache.org on 2023/07/26 00:12:46 UTC
[superset] branch db-diagnostics updated: Working on docs
This is an automated email from the ASF dual-hosted git repository.
beto pushed a commit to branch db-diagnostics
in repository https://gitbox.apache.org/repos/asf/superset.git
The following commit(s) were added to refs/heads/db-diagnostics by this push:
new ae298e0df6 Working on docs
ae298e0df6 is described below
commit ae298e0df6f922be6a4bb198c994c785fbcfa609
Author: Beto Dealmeida <ro...@dealmeida.net>
AuthorDate: Tue Jul 25 17:12:36 2023 -0700
Working on docs
---
superset/db_engine_specs/README.md | 10 ++++++++--
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/superset/db_engine_specs/README.md b/superset/db_engine_specs/README.md
index e6aba4b6f7..e05ac9cab1 100644
--- a/superset/db_engine_specs/README.md
+++ b/superset/db_engine_specs/README.md
@@ -3,7 +3,7 @@ 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 would run a query like this:
+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
@@ -34,7 +34,9 @@ Note that DB engine specs are completely optional. Superset can connect to any d
Features
--------
-| | Details | Base | Amazon Athena | Amazon DynamoDB | Amazon Redshift | Apache Drill | Apache Druid | Apache Hive | Apache Impala | Apache Kylin | Apache Pinot | Apache Solr | Apac [...]
+The table below summarizes the information about the status of all DB engine specs Superset is aware of (note that this includes 3rd party DB engine specs):
+
+| Feature | Details | Base | Amazon Athena | Amazon DynamoDB | Amazon Redshift | Apache Drill | Apache Druid | Apache Hive | Apache Impala | Apache Kylin | Apache Pinot | Apache Solr | Apac [...]
|----------------------------|----------------------------------------------------------------------------------------|-------------------------------|---------------------------------|-----------------------------------|-----------------------------------|--------------------------------|--------------------------------|-------------------------------|---------------------------------|--------------------------------|--------------------------------|-------------------------------|----- [...]
| Module | | superset.db_engine_specs.base | 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 | supe [...]
| Limit method | In general, FORCE_LIMIT > WRAP_SQL > FETCH_MANY | FETCH_MANY | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORCE_LIMIT | FORC [...]
@@ -92,5 +94,9 @@ Features
| Query cost estimation | | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | [...]
| SQL validation | Implemented outside the spec. | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | [...]
+Database information
+--------------------
+A DB engine spec has attributes that describe the underlying database engine, so that Superset can know how to build and run queries. For example, some databases don't support subqueries, which are needed for some of the queries produced by Superset for certain charts. When a database doesn't support subqueries the query is run in two-steps, using the results from the first query to build the second query.
+These attributes and their default values (set in the base class, `BaseEngineSpec`) are described below: