Example: Ruby "hello world" application

This code sample is a "hello world" application that runs on Ruby. The sample illustrates how to complete the following tasks:

• Connect to a Cloud Bigtable instance.
• Create a new table.
• Write data to the table.
• Read the data back.
• Delete the table.

Running the sample

This code sample uses the Ruby Client for Cloud Bigtable package of the Google Cloud Client Library for Ruby to communicate with Cloud Bigtable.

To run this sample program, follow the instructions for the sample on GitHub.

Using the Cloud Client Library with Cloud Bigtable

The sample application connects to Cloud Bigtable and demonstrates some simple operations.

Requiring the client library

The sample requires google/cloud/bigtable, which provides the Bigtable module.

bigtable/hello_world.rb

View on GitHub

require "google/cloud/bigtable"

Connecting to Cloud Bigtable

Establish the variables you will use in your application, replacing "YOUR_PROJECT_ID" with the ID of a valid Google Cloud project. Then create a new Bigtable object that you will use to connect to Cloud Bigtable.

bigtable/hello_world.rb

View on GitHub

project_id = "YOUR_PROJECT_ID"
table_id = "Hello-Bigtable"
instance_id = "my-instance"
column_family = "cf"
column_qualifier = "greeting"

bigtable = Google::Cloud::Bigtable.new project_id: project_id

Creating a table

Check to see if your table already exists. If it doesn't, call the create_table() method to create a Table object. The table has a single column family that retains one version of each value.

bigtable/hello_world.rb

View on GitHub

if bigtable.table(instance_id, table_id).exists?
  puts "#{table_id} is already exists."
  exit 0
else
  table = bigtable.create_table instance_id, table_id do |column_families|
    column_families.add(
      column_family,
      Google::Cloud::Bigtable::GcRule.max_versions(1)
    )
  end

  puts "Table #{table_id} created."
end

Writing rows to a table

Next, use a string array of greetings to create some new rows for the table. For each greeting, create an entry using the table's new_mutation_entry() method. Next, use the entry's set_cell() method to assign the column family, column qualifier, the greeting, and a timestamp to the entry. Finally, write that entry to the table using the table's mutate_row() method.

Note: In this sample application, set_Cell() is only called once per mutation, but you can call it up to 10,000 times per single mutation.

bigtable/hello_world.rb

View on GitHub

puts "Write some greetings to the table #{table_id}"
greetings = ["Hello World!", "Hello Bigtable!", "Hello Ruby!"]

# Insert rows one by one
# Note: To perform multiple mutation on multiple rows use `mutate_rows`.
greetings.each_with_index do |value, i|
  puts " Writing,  Row key: greeting#{i}, Value: #{value}"

  entry = table.new_mutation_entry "greeting#{i}"
  entry.set_cell(
    column_family,
    column_qualifier,
    value,
    timestamp: (Time.now.to_f * 1_000_000).round(-3)
  )

  table.mutate_row entry
end

Creating a filter

Before you read the data that you wrote, create a filter to limit the data that Cloud Bigtable returns. This filter tells Cloud Bigtable to return only the most recent version of each value, even if the table contains older versions that haven't been garbage-collected.

bigtable/hello_world.rb

View on GitHub

# Only retrieve the most recent version of the cell.
filter = Google::Cloud::Bigtable::RowFilter.cells_per_column 1

Reading a row by its row key

Create a row object, then call the read_row() method, passing in the filter, to get one version of each value in that row.

bigtable/hello_world.rb

View on GitHub

puts "Reading a single row by row key"
row = table.read_row "greeting0", filter: filter
puts "Row key: #{row.key}, Value: #{row.cells[column_family].first.value}"

puts "Reading the entire table"
table.read_rows.each do |row|
  puts "Row key: #{row.key}, Value: #{row.cells[column_family].first.value}"
end

puts "Deleting the table #{table_id}"
table.delete

Scanning all table rows

Call the read_rows() method, passing in the filter, to get all of the rows in the table. Because you passed in the filter, Cloud Bigtable returns only one version of each value.

bigtable/hello_world.rb

View on GitHub

puts "Reading the entire table"
table.read_rows.each do |row|
  puts "Row key: #{row.key}, Value: #{row.cells[column_family].first.value}"
end

Deleting a table

Delete the table with the table's delete() method.

bigtable/hello_world.rb

View on GitHub

puts "Deleting the table #{table_id}"
table.delete

Putting it all together

Here is the full code sample without comments.

bigtable/hello_world.rb

View on GitHub

# frozen_string_literal: true

require "google/cloud/bigtable"

project_id = "YOUR_PROJECT_ID"
table_id = "Hello-Bigtable"
instance_id = "my-instance"
column_family = "cf"
column_qualifier = "greeting"

bigtable = Google::Cloud::Bigtable.new project_id: project_id

if bigtable.table(instance_id, table_id).exists?
  puts "#{table_id} is already exists."
  exit 0
else
  table = bigtable.create_table instance_id, table_id do |column_families|
    column_families.add(
      column_family,
      Google::Cloud::Bigtable::GcRule.max_versions(1)
    )
  end

  puts "Table #{table_id} created."
end

puts "Write some greetings to the table #{table_id}"
greetings = ["Hello World!", "Hello Bigtable!", "Hello Ruby!"]

# Insert rows one by one
# Note: To perform multiple mutation on multiple rows use `mutate_rows`.
greetings.each_with_index do |value, i|
  puts " Writing,  Row key: greeting#{i}, Value: #{value}"

  entry = table.new_mutation_entry "greeting#{i}"
  entry.set_cell(
    column_family,
    column_qualifier,
    value,
    timestamp: (Time.now.to_f * 1_000_000).round(-3)
  )

  table.mutate_row entry
end

# Only retrieve the most recent version of the cell.
filter = Google::Cloud::Bigtable::RowFilter.cells_per_column 1

puts "Reading a single row by row key"
row = table.read_row "greeting0", filter: filter
puts "Row key: #{row.key}, Value: #{row.cells[column_family].first.value}"

puts "Reading the entire table"
table.read_rows.each do |row|
  puts "Row key: #{row.key}, Value: #{row.cells[column_family].first.value}"
end

puts "Deleting the table #{table_id}"
table.delete