README.md

Embulk::Input::Bigquery

This is Embulk input plugin from Bigquery.

Installation

install it yourself as:

$ embulk gem install embulk-input-bigquery

Usage

in:
  type: bigquery
  project: 'project-name'
  keyfile: '/home/hogehoge/bigquery-keyfile.json'
  sql: 'SELECT price,category_id FROM [ecsite.products] GROUP BY category_id'
  columns:
    - {name: price, type: long}
    - {name: category_id, type: string}
  max: 2000
out:
  type: stdout

If, table name is changeable, then

in:
  type: bigquery
  project: 'project-name'
  keyfile: '/home/hogehoge/bigquery-keyfile.json'
  sql_erb: 'SELECT price,category_id FROM [ecsite.products_<%= params["date"].strftime("%Y%m")  %>] GROUP BY category_id'
  erb_params:
    date: "require 'date'; (Date.today - 1)"
  columns:
    - {name: price, type: long}
    - {name: category_id, type: long}
    - {name: month, type: timestamp, format: '%Y-%m', eval: 'require "time"; Time.parse(params["date"]).to_i'}

Determine columns from query results if colmuns defination is empty

in:
  type: bigquery
  project: 'project-name'
  keyfile: '/home/hogehoge/bigquery-keyfile.json'
  sql: 'SELECT price,category_id FROM [ecsite.products] GROUP BY category_id'
out:
  type: stdout

Optional Configuration

This plugin uses the gem google-cloud(Google Cloud Client Library for Ruby) and queries data using the synchronous method. Therefore some optional configuration items comply with the Google Cloud Client Library.

• max :
  • default value : null and null value is interpreted as no maximum row count in the Google Cloud Client Library.
• cache :
  • default value : null and null value is interpreted as true in the Google Cloud Client Library.
• standard_sql:
  • default value : null and null value is interpreted as true in the Google Cloud Client Library.
• legacy_sql:
  • default value : null and null value is interpreted as false in the Google Cloud Client Library.