stairlight 0.2.0

Table-level data lineage tool

Stairlight

Stairlight is a table-level data lineage tool, detects table dependencies from 'Transform' SQL files such as plain SELECT, 'CREATE TABLE AS SELECT', 'CREATE VIEW AS SELECT'.

Installation

$ pip install stairlight

Getting Started

There are three steps to use.

# Step 1: Initialize and set data location settings
$ stairlight init
'./stairlight.yaml' has created.
Please edit it to set your data sources.

# Step 2: Map SQL files and tables, and add metadata
$ stairlight check
'./mapping_yyyyMMddhhmmss.yaml' has created.
Please map undefined tables and parameters, and append to your latest file.

# Step 3: Get a table dependency map
$ stairlight

Description

Input

• SQL files
• Configuration files (YAML)
  • stairlight.yaml: SQL file locations and include/exclude files.
  • mapping.yaml: Mapping SQL files and tables.

Output

• Dependency map (JSON)

  Example

  {
      "PROJECT_d.DATASET_e.TABLE_f": {
          "PROJECT_j.DATASET_k.TABLE_l": {
              "type": "fs",
              "file": "tests/sql/main/test_e.sql",
              "uri": "/foo/bar/stairlight/tests/sql/main/test_e.sql",
              "lines": [
                  {
                      "num": 1,
                      "str": "SELECT * FROM PROJECT_j.DATASET_k.TABLE_l WHERE 1 = 1"
                  }
              ]
          },
          "PROJECT_g.DATASET_h.TABLE_i": {
              "type": "gcs",
              "file": "sql/test_b/test_b.sql",
              "uri": "gs://baz/sql/test_b/test_b.sql",
              "lines": [
                  {
                      "num": 23,
                      "str": "    PROJECT_g.DATASET_h.TABLE_i AS b",
                  }
              ],
              "bucket": "stairlight"
          },
          "PROJECT_C.DATASET_C.TABLE_C": {
              "type": "gcs",
              "file": "sql/test_b/test_b.sql",
              "uri": "gs://baz/sql/test_b/test_b.sql",
              "lines": [
                  {
                      "num": 6,
                      "str": "        PROJECT_C.DATASET_C.TABLE_C",
                  }
              ],
              "bucket": "stairlight"
          },
          "PROJECT_d.DATASET_d.TABLE_d": {
              "type": "gcs",
              "file": "sql/test_b/test_b.sql",
              "uri": "gs://baz/sql/test_b/test_b.sql",
              "lines": [
                  {
                      "num": 15,
                      "str": "        PROJECT_d.DATASET_d.TABLE_d",
                  }
              ],
              "bucket": "stairlight"
          }
      },
      "PROJECT_j.DATASET_k.TABLE_l": {
          "PROJECT_d.DATASET_e.TABLE_f": {
              "type": "fs",
              "file": "tests/sql/main/test_d.sql",
              "uri": "/foo/bar/stairlight/tests/sql/main/test_d.sql",
              "lines": [
                  {
                      "num": 1,
                      "str": "SELECT * FROM PROJECT_d.DATASET_e.TABLE_f WHERE 1 = 1"
                  }
              ]
          }
      },
      "PROJECT_d.DATASET_d.TABLE_d": {
          "PROJECT_e.DATASET_e.TABLE_e": {
              "type": "fs",
              "file": "tests/sql/main/test_f.sql",
              "uri": "/foo/bar/stairlight/tests/sql/main/test_f.sql",
              "lines": [
                  {
                      "num": 1,
                      "str": "SELECT * FROM PROJECT_e.DATASET_e.TABLE_e WHERE 1 = 1"
                  }
              ]
          }
      },
  }
  

Configuration

Config files used for unit test in CI can be found here.

stairlight.yaml

'stairlight.yaml' is for setting up Stairlight itself.

It is responsible for specifying the destination of SQL files to be read, and for specifying the prefix of mapping files.

SQL files can be read from the following storage.

• Local file system(with python pathlib module)
• Google Cloud Storage

mapping.yaml

'mapping.yaml' is used to define relationships between input files and tables.

mapping section

This section is used to define relationships between SQL files and tables created as a result of query execution.

The params attribute allows you to reflect settings in jinja template variables embedded in SQL files. If multiple settings are applied to a SQL file using jinja template, the file will be read as if there were the same number of files as the number of settings.

metadata section

This section is mainly used to set labels to tables written in SQL files.

Command and option

$ stairlight --help
usage: stairlight [-h] [-c CONFIG] [--save SAVE | --load LOAD] {init,check,up,down} ...

A table-level data lineage tool, detects table dependencies from 'Transform' SQL files.
Without positional arguments, return a table dependency map as JSON format.

positional arguments:
  {init,check,up,down}
    init                create a new Stairlight configuration file
    check               create a new configuration file about undefined mappings
    up                  return upstairs ( table | SQL file ) list
    down                return downstairs ( table | SQL file ) list

optional arguments:
  -h, --help            show this help message and exit
  -c CONFIG, --config CONFIG
                        set a Stairlight configuration directory
  --save SAVE           save results to a file
  --load LOAD           load results from a file

init

init creates a new Stairlight configuration file.

$ stairlight init --help
usage: stairlight init [-h] [-c CONFIG] [-s SAVE | -l LOAD]

optional arguments:
  -h, --help            show this help message and exit
  -c CONFIG, --config CONFIG
                        set a Stairlight configuration directory.

check

check creates a new configuration file about undefined mappings. The option specification is the same as init.

up

up outputs a list of tables or SQL files located upstream from the specified table.

• Use table(-t, --table) or label(-l, --label) option to specify tables to search.
• Recursive option(-r, --recursive) is set, Stairlight will find tables recursively and output as a list.
• Verbose option(-v, --verbose) is set, Stairlight will add detailed information and output it as a dict.

$ stairlight up --help
usage: stairlight up [-h] [-c CONFIG] [--save SAVE | --load LOAD] (-t TABLE | -l LABEL)
                  [-o {table,file}] [-v] [-r]

optional arguments:
  -h, --help            show this help message and exit
  -c CONFIG, --config CONFIG
                        set a Stairlight configuration directory
  --save SAVE           save results to a file
  --load LOAD           load results from a file
  -t TABLE, --table TABLE
                        table names that Stairlight searches for, can be specified
                        multiple times. e.g. -t PROJECT_a.DATASET_b.TABLE_c -t
                        PROJECT_d.DATASET_e.TABLE_f
  -l LABEL, --label LABEL
                        labels set for the table in mapping configuration, can be
                        specified multiple times. The separator between key and value
                        should be a colon(:). e.g. -l key_1:value_1 -l key_2:value_2
  -o {table,file}, --output {table,file}
                        output type
  -v, --verbose         return verbose results
  -r, --recursive       search recursively

down

down outputs a list of tables or SQL files located downstream from the specified table. The option specification is the same as up.

Use as a library

Stairlight can also be used as a library.

tosh2230/stairlight-app is a sample web application rendering table dependency graph with Stairlight, using Graphviz, Streamlit and Google Cloud Run.