dbt-sas 1.3.0a17

dbt (data build tool) adapter for SAS

SAS DBT adapter

The dbt-sas package allows dbt to connect to SAS.

Installation

pip install dbt-sas

Configurations

Basic profile.yml for connecting to SAS:

sas-test:
  outputs:
    dev:
      type: sas
      host: SAS host
      port: 8591
      database: sas (keep 'sas', don't change)
      schema: default libray name
      user: SAS username
      password: SAS password
      autoexec: optional local path of autoexec.sas file
      lib_base_path: optional path (on the SAS server) for new libraries
      threads: 1
      fail_on_warnings: False
      lib_name_strict_mode: False
  target: dev

Key	Required	Description
type	Yes	The specific adapter to use (sas)
host	Yes	SAS server hostname
port	Yes	SAS server port
database	Yes	sas, don't change
schema	Yes	Default SAS libname
username	Yes	The username to use to connect to the server
password	Yes	The password to use for authenticating to the server
handler	Yes	SAS handler ws (default) or saspy
autoexec	No	Local path of autoexec.sas file
lib_base_path	No	Base path on the SAS server for new libraries
fail_on_warnings	No	Raise and error if it encounters a warning
lib_name_strict_mode	No	Enable strict libname/identifier check

Features

Key	Supported	Description
Tests	Yes	Run dbt tests on SAS
Load seed files	Yes	Load seeds from CSV files to SAS
View Materialization	Yes	Materialize dbt models as views in SAS
Table Materialization	Yes	Materialize dbt models as tables in SAS
Ephemeral Materialization	Partial	Materialize dbt ephemeral tables as temporary views in SAS
Incremental Materialization	No
Snapshots	No
Grants	No	SQL grants are not supported by SAS

Limitations

• Schemas (lib names) are limited to 8 characters.
• Table names, column names, and aliases are limited to 32 characters.

Usage

• Create dbt project, choose sas database and set up connection

$ dbt init <project_name>

Query Logging

Enable additional logging by setting the DBT_SAS_LOG environment variable to the log filename, e.g.:

export DBT_SAS_LOG='file.log'

Select the log families, e.g.:

export DBT_SAS_FAMILY='sas,sql'

Log families are:

• sas
• sql
• original_sql

Adapter Testing

• Install dev requirements

$ pip install -r dev_requirements.txt

• Run pytest

$ python -m pytest tests/

Licence

Apache License, Version 2.0

Links

• SASpy
• SQL Procedure