google-cloud-bigquery 1.20.0 → 1.23.0

data/lib/google/cloud/bigquery/load_job.rb

@@ -1303,12 +1303,21 @@ module Google
           # Sets the labels to use for the load job.
           #
           # @param [Hash] val A hash of user-provided labels associated with
-          #   the job. You can use these to organize and group your jobs. Label
-          #   keys and values can be no longer than 63 characters, can only
-          #   contain lowercase letters, numeric characters, underscores and
-          #   dashes. International characters are allowed. Label values are
-          #   optional. Label keys must start with a letter and each label in
-          #   the list must have a different key.
+          #   the job. You can use these to organize and group your jobs.
+          #
+          #   The labels applied to a resource must meet the following requirements:
+          #
+          #   * Each resource can have multiple labels, up to a maximum of 64.
+          #   * Each label must be a key-value pair.
+          #   * Keys have a minimum length of 1 character and a maximum length of
+          #     63 characters, and cannot be empty. Values can be empty, and have
+          #     a maximum length of 63 characters.
+          #   * Keys and values can contain only lowercase letters, numeric characters,
+          #     underscores, and dashes. All characters must use UTF-8 encoding, and
+          #     international characters are allowed.
+          #   * The key portion of a label must be unique. However, you can use the
+          #     same key with multiple resources.
+          #   * Keys must start with a lowercase letter or international character.
           #
           # @!group Attributes
           #

data/lib/google/cloud/bigquery/model.rb

@@ -341,14 +341,19 @@ module Google
         # the update to comply with ETag-based optimistic concurrency control.
         #
         # @param [Hash<String, String>] new_labels A hash containing key/value
-        #   pairs.
-        #
-        #   * Label keys and values can be no longer than 63 characters.
-        #   * Label keys and values can contain only lowercase letters, numbers,
-        #     underscores, hyphens, and international characters.
-        #   * Label keys and values cannot exceed 128 bytes in size.
-        #   * Label keys must begin with a letter.
-        #   * Label keys must be unique within a model.
+        #   pairs. The labels applied to a resource must meet the following requirements:
+        #
+        #   * Each resource can have multiple labels, up to a maximum of 64.
+        #   * Each label must be a key-value pair.
+        #   * Keys have a minimum length of 1 character and a maximum length of
+        #     63 characters, and cannot be empty. Values can be empty, and have
+        #     a maximum length of 63 characters.
+        #   * Keys and values can contain only lowercase letters, numeric characters,
+        #     underscores, and dashes. All characters must use UTF-8 encoding, and
+        #     international characters are allowed.
+        #   * The key portion of a label must be unique. However, you can use the
+        #     same key with multiple resources.
+        #   * Keys must start with a lowercase letter or international character.
         #
         # @example
         #   require "google/cloud/bigquery"
@@ -482,6 +487,146 @@ module Google
           Array @gapi_json[:trainingRuns]
         end
 
+        ##
+        # Exports the model to Google Cloud Storage asynchronously, immediately
+        # returning an {ExtractJob} that can be used to track the progress of the
+        # export job. The caller may poll the service by repeatedly calling
+        # {Job#reload!} and {Job#done?} to detect when the job is done, or
+        # simply block until the job is done by calling #{Job#wait_until_done!}.
+        # See also {#extract}.
+        #
+        # The geographic location for the job ("US", "EU", etc.) can be set via
+        # {ExtractJob::Updater#location=} in a block passed to this method. If
+        # the model is a full resource representation (see {#resource_full?}),
+        # the location of the job will automatically be set to the location of
+        # the model.
+        #
+        # @see https://cloud.google.com/bigquery-ml/docs/exporting-models
+        #   Exporting models
+        #
+        # @param [String] extract_url The Google Storage URI to which BigQuery
+        #   should extract the model. This value should be end in an object name
+        #   prefix, since multiple objects will be exported.
+        # @param [String] format The exported file format. The default value is
+        #   `ml_tf_saved_model`.
+        #
+        #   The following values are supported:
+        #
+        #   * `ml_tf_saved_model` - TensorFlow SavedModel
+        #   * `ml_xgboost_booster` - XGBoost Booster
+        # @param [String] job_id A user-defined ID for the extract job. The ID
+        #   must contain only letters (a-z, A-Z), numbers (0-9), underscores
+        #   (_), or dashes (-). The maximum length is 1,024 characters. If
+        #   `job_id` is provided, then `prefix` will not be used.
+        #
+        #   See [Generating a job
+        #   ID](https://cloud.google.com/bigquery/docs/managing-jobs#generate-jobid).
+        # @param [String] prefix A string, usually human-readable, that will be
+        #   prepended to a generated value to produce a unique job ID. For
+        #   example, the prefix `daily_import_job_` can be given to generate a
+        #   job ID such as `daily_import_job_12vEDtMQ0mbp1Mo5Z7mzAFQJZazh`. The
+        #   prefix must contain only letters (a-z, A-Z), numbers (0-9),
+        #   underscores (_), or dashes (-). The maximum length of the entire ID
+        #   is 1,024 characters. If `job_id` is provided, then `prefix` will not
+        #   be used.
+        # @param [Hash] labels A hash of user-provided labels associated with
+        #   the job. You can use these to organize and group your jobs.
+        #
+        #   The labels applied to a resource must meet the following requirements:
+        #
+        #   * Each resource can have multiple labels, up to a maximum of 64.
+        #   * Each label must be a key-value pair.
+        #   * Keys have a minimum length of 1 character and a maximum length of
+        #     63 characters, and cannot be empty. Values can be empty, and have
+        #     a maximum length of 63 characters.
+        #   * Keys and values can contain only lowercase letters, numeric characters,
+        #     underscores, and dashes. All characters must use UTF-8 encoding, and
+        #     international characters are allowed.
+        #   * The key portion of a label must be unique. However, you can use the
+        #     same key with multiple resources.
+        #   * Keys must start with a lowercase letter or international character.
+        #
+        # @yield [job] a job configuration object
+        # @yieldparam [Google::Cloud::Bigquery::ExtractJob::Updater] job a job
+        #   configuration object for setting additional options.
+        #
+        # @return [Google::Cloud::Bigquery::ExtractJob]
+        #
+        # @example
+        #   require "google/cloud/bigquery"
+        #
+        #   bigquery = Google::Cloud::Bigquery.new
+        #   dataset = bigquery.dataset "my_dataset"
+        #   model = dataset.model "my_model"
+        #
+        #   extract_job = model.extract_job "gs://my-bucket/#{model.model_id}"
+        #
+        #   extract_job.wait_until_done!
+        #   extract_job.done? #=> true
+        #
+        # @!group Data
+        #
+        def extract_job extract_url, format: nil, job_id: nil, prefix: nil, labels: nil
+          ensure_service!
+          options = { format: format, job_id: job_id, prefix: prefix, labels: labels }
+          updater = ExtractJob::Updater.from_options service, model_ref, extract_url, options
+          updater.location = location if location # may be model reference
+
+          yield updater if block_given?
+
+          job_gapi = updater.to_gapi
+          gapi = service.extract_table job_gapi
+          Job.from_gapi gapi, service
+        end
+
+        ##
+        # Exports the model to Google Cloud Storage using a synchronous method
+        # that blocks for a response. Timeouts and transient errors are generally
+        # handled as needed to complete the job. See also {#extract_job}.
+        #
+        # The geographic location for the job ("US", "EU", etc.) can be set via
+        # {ExtractJob::Updater#location=} in a block passed to this method. If
+        # the model is a full resource representation (see {#resource_full?}),
+        # the location of the job will automatically be set to the location of
+        # the model.
+        #
+        # @see https://cloud.google.com/bigquery-ml/docs/exporting-models
+        #   Exporting models
+        #
+        # @param [String] extract_url The Google Storage URI to which BigQuery
+        #   should extract the model. This value should be end in an object name
+        #   prefix, since multiple objects will be exported.
+        # @param [String] format The exported file format. The default value is
+        #   `ml_tf_saved_model`.
+        #
+        #   The following values are supported:
+        #
+        #   * `ml_tf_saved_model` - TensorFlow SavedModel
+        #   * `ml_xgboost_booster` - XGBoost Booster
+        # @yield [job] a job configuration object
+        # @yieldparam [Google::Cloud::Bigquery::ExtractJob::Updater] job a job
+        #   configuration object for setting additional options.
+        #
+        # @return [Boolean] Returns `true` if the extract operation succeeded.
+        #
+        # @example
+        #   require "google/cloud/bigquery"
+        #
+        #   bigquery = Google::Cloud::Bigquery.new
+        #   dataset = bigquery.dataset "my_dataset"
+        #   model = dataset.model "my_model"
+        #
+        #   model.extract "gs://my-bucket/#{model.model_id}"
+        #
+        # @!group Data
+        #
+        def extract extract_url, format: nil, &block
+          job = extract_job extract_url, format: format, &block
+          job.wait_until_done!
+          ensure_job_succeeded! job
+          true
+        end
+
         ##
         # Permanently deletes the model.
         #
@@ -734,6 +879,17 @@ module Google
         def ensure_full_data!
           reload! unless resource_full?
         end
+
+        def ensure_job_succeeded! job
+          return unless job.failed?
+          begin
+            # raise to activate ruby exception cause handling
+            raise job.gapi_error
+          rescue StandardError => e
+            # wrap Google::Apis::Error with Google::Cloud::Error
+            raise Google::Cloud::Error.from_error(e)
+          end
+        end
       end
     end
   end

data/lib/google/cloud/bigquery/project.rb

@@ -153,13 +153,21 @@ module Google
         #   is 1,024 characters. If `job_id` is provided, then `prefix` will not
         #   be used.
         # @param [Hash] labels A hash of user-provided labels associated with
-        #   the job. You can use these to organize and group your jobs. Label
-        #   keys and values can be no longer than 63 characters, can only
-        #   contain lowercase letters, numeric characters, underscores and
-        #   dashes. International characters are allowed. Label values are
-        #   optional. Label keys must start with a letter and each label in the
-        #   list must have a different key. See [Requirements for
-        #   labels](https://cloud.google.com/bigquery/docs/creating-managing-labels#requirements).
+        #   the job. You can use these to organize and group your jobs.
+        #
+        #   The labels applied to a resource must meet the following requirements:
+        #
+        #   * Each resource can have multiple labels, up to a maximum of 64.
+        #   * Each label must be a key-value pair.
+        #   * Keys have a minimum length of 1 character and a maximum length of
+        #     63 characters, and cannot be empty. Values can be empty, and have
+        #     a maximum length of 63 characters.
+        #   * Keys and values can contain only lowercase letters, numeric characters,
+        #     underscores, and dashes. All characters must use UTF-8 encoding, and
+        #     international characters are allowed.
+        #   * The key portion of a label must be unique. However, you can use the
+        #     same key with multiple resources.
+        #   * Keys must start with a lowercase letter or international character.
         # @yield [job] a job configuration object
         # @yieldparam [Google::Cloud::Bigquery::CopyJob::Updater] job a job
         #   configuration object for setting additional options.
@@ -411,20 +419,36 @@ module Google
         #   See [Generating a job
         #   ID](https://cloud.google.com/bigquery/docs/managing-jobs#generate-jobid).
         # @param [Hash] labels A hash of user-provided labels associated with
-        #   the job. You can use these to organize and group your jobs. Label
-        #   keys and values can be no longer than 63 characters, can only
-        #   contain lowercase letters, numeric characters, underscores and
-        #   dashes. International characters are allowed. Label values are
-        #   optional. Label keys must start with a letter and each label in the
-        #   list must have a different key. See [Requirements for
-        #   labels](https://cloud.google.com/bigquery/docs/creating-managing-labels#requirements).
+        #   the job. You can use these to organize and group your jobs.
+        #
+        #   The labels applied to a resource must meet the following requirements:
+        #
+        #   * Each resource can have multiple labels, up to a maximum of 64.
+        #   * Each label must be a key-value pair.
+        #   * Keys have a minimum length of 1 character and a maximum length of
+        #     63 characters, and cannot be empty. Values can be empty, and have
+        #     a maximum length of 63 characters.
+        #   * Keys and values can contain only lowercase letters, numeric characters,
+        #     underscores, and dashes. All characters must use UTF-8 encoding, and
+        #     international characters are allowed.
+        #   * The key portion of a label must be unique. However, you can use the
+        #     same key with multiple resources.
+        #   * Keys must start with a lowercase letter or international character.
         # @param [Array<String>, String] udfs User-defined function resources
-        #   used in the query. May be either a code resource to load from a
-        #   Google Cloud Storage URI (`gs://bucket/path`), or an inline resource
+        #   used in a legacy SQL query. May be either a code resource to load from
+        #   a Google Cloud Storage URI (`gs://bucket/path`), or an inline resource
         #   that contains code for a user-defined function (UDF). Providing an
         #   inline code resource is equivalent to providing a URI for a file
-        #   containing the same code. See [User-Defined
-        #   Functions](https://cloud.google.com/bigquery/docs/reference/standard-sql/user-defined-functions).
+        #   containing the same code.
+        #
+        #   This parameter is used for defining User Defined Function (UDF)
+        #   resources only when using legacy SQL. Users of standard SQL should
+        #   leverage either DDL (e.g. `CREATE [TEMPORARY] FUNCTION ...`) or the
+        #   Routines API to define UDF resources.
+        #
+        #   For additional information on migrating, see: [Migrating to
+        #   standard SQL - Differences in user-defined JavaScript
+        #   functions](https://cloud.google.com/bigquery/docs/reference/standard-sql/migrating-from-legacy-sql#differences_in_user-defined_javascript_functions)
         # @param [Integer] maximum_billing_tier Deprecated: Change the billing
         #   tier to allow high-compute queries.
         # @yield [job] a job configuration object
@@ -709,9 +733,12 @@ module Google
         #   sql = "SELECT name FROM `my_project.my_dataset.my_table`"
         #   data = bigquery.query sql
         #
+        #   # Iterate over the first page of results
         #   data.each do |row|
         #     puts row[:name]
         #   end
+        #   # Retrieve the next page of results
+        #   data = data.next if data.next?
         #
         # @example Query using legacy SQL:
         #   require "google/cloud/bigquery"
@@ -721,9 +748,12 @@ module Google
         #   sql = "SELECT name FROM [my_project:my_dataset.my_table]"
         #   data = bigquery.query sql, legacy_sql: true
         #
+        #   # Iterate over the first page of results
         #   data.each do |row|
         #     puts row[:name]
         #   end
+        #   # Retrieve the next page of results
+        #   data = data.next if data.next?
         #
         # @example Retrieve all rows: (See {Data#all})
         #   require "google/cloud/bigquery"
@@ -746,9 +776,12 @@ module Google
         #                         "WHERE id = ?",
         #                         params: [1]
         #
+        #   # Iterate over the first page of results
         #   data.each do |row|
         #     puts row[:name]
         #   end
+        #   # Retrieve the next page of results
+        #   data = data.next if data.next?
         #
         # @example Query using named query parameters:
         #   require "google/cloud/bigquery"
@@ -760,9 +793,12 @@ module Google
         #                         "WHERE id = @id",
         #                         params: { id: 1 }
         #
+        #   # Iterate over the first page of results
         #   data.each do |row|
         #     puts row[:name]
         #   end
+        #   # Retrieve the next page of results
+        #   data = data.next if data.next?
         #
         # @example Query using named query parameters with types:
         #   require "google/cloud/bigquery"
@@ -775,9 +811,12 @@ module Google
         #                         params: { ids: [] },
         #                         types: { ids: [:INT64] }
         #
+        #   # Iterate over the first page of results
         #   data.each do |row|
         #     puts row[:name]
         #   end
+        #   # Retrieve the next page of results
+        #   data = data.next if data.next?
         #
         # @example Execute a DDL statement:
         #   require "google/cloud/bigquery"
@@ -816,9 +855,12 @@ module Google
         #     query.table = dataset.table "my_table", skip_lookup: true
         #   end
         #
+        #   # Iterate over the first page of results
         #   data.each do |row|
         #     puts row[:name]
         #   end
+        #   # Retrieve the next page of results
+        #   data = data.next if data.next?
         #
         def query query, params: nil, types: nil, external: nil, max: nil, cache: true, dataset: nil, project: nil,
                   standard_sql: nil, legacy_sql: nil, &block
@@ -880,9 +922,12 @@ module Google
         #   data = bigquery.query "SELECT * FROM my_ext_table",
         #                         external: { my_ext_table: csv_table }
         #
+        #   # Iterate over the first page of results
         #   data.each do |row|
         #     puts row[:name]
         #   end
+        #   # Retrieve the next page of results
+        #   data = data.next if data.next?
         #
         def external url, format: nil
           ext = External.from_urls url, format
@@ -1084,18 +1129,22 @@ module Google
         #   part of the larger set of results to view. Optional.
         # @param [Integer] max Maximum number of jobs to return. Optional.
         # @param [String] filter A filter for job state. Optional.
-        # @param [Time] min_created_at Min value for {Job#created_at}. When
-        #   provided, only jobs created after or at this time are returned.
-        #   Optional.
-        # @param [Time] max_created_at Max value for {Job#created_at}. When
-        #   provided, only jobs created before or at this time are returned.
-        #   Optional.
         #
         #   Acceptable values are:
         #
         #   * `done` - Finished jobs
         #   * `pending` - Pending jobs
         #   * `running` - Running jobs
+        # @param [Time] min_created_at Min value for {Job#created_at}. When
+        #   provided, only jobs created after or at this time are returned.
+        #   Optional.
+        # @param [Time] max_created_at Max value for {Job#created_at}. When
+        #   provided, only jobs created before or at this time are returned.
+        #   Optional.
+        # @param [Google::Cloud::Bigquery::Job, String] parent_job A job
+        #   object or a job ID. If set, retrieve only child jobs of the
+        #   specified parent. Optional. See {Job#job_id}, {Job#num_child_jobs},
+        #   and {Job#parent_job_id}.
         #
         # @return [Array<Google::Cloud::Bigquery::Job>] (See
         #   {Google::Cloud::Bigquery::Job::List})
@@ -1144,13 +1193,63 @@ module Google
         #     # process job
         #   end
         #
-        def jobs all: nil, token: nil, max: nil, filter: nil,
-                 min_created_at: nil, max_created_at: nil
+        # @example Retrieve child jobs by setting `parent_job`:
+        #   require "google/cloud/bigquery"
+        #
+        #   bigquery = Google::Cloud::Bigquery.new
+        #
+        #   multi_statement_sql = <<~SQL
+        #     -- Declare a variable to hold names as an array.
+        #     DECLARE top_names ARRAY<STRING>;
+        #     -- Build an array of the top 100 names from the year 2017.
+        #     SET top_names = (
+        #     SELECT ARRAY_AGG(name ORDER BY number DESC LIMIT 100)
+        #     FROM `bigquery-public-data.usa_names.usa_1910_current`
+        #     WHERE year = 2017
+        #     );
+        #     -- Which names appear as words in Shakespeare's plays?
+        #     SELECT
+        #     name AS shakespeare_name
+        #     FROM UNNEST(top_names) AS name
+        #     WHERE name IN (
+        #     SELECT word
+        #     FROM `bigquery-public-data.samples.shakespeare`
+        #     );
+        #   SQL
+        #
+        #   job = bigquery.query_job multi_statement_sql
+        #
+        #   job.wait_until_done!
+        #
+        #   child_jobs = bigquery.jobs parent_job: job
+        #
+        #   child_jobs.each do |child_job|
+        #     script_statistics = child_job.script_statistics
+        #     puts script_statistics.evaluation_kind
+        #     script_statistics.stack_frames.each do |stack_frame|
+        #       puts stack_frame.text
+        #     end
+        #   end
+        #
+        def jobs all: nil,
+                 token: nil,
+                 max: nil,
+                 filter: nil,
+                 min_created_at: nil,
+                 max_created_at: nil,
+                 parent_job: nil
           ensure_service!
-          options = { all: all, token: token, max: max, filter: filter, min_created_at: min_created_at,
-                      max_created_at: max_created_at }
-          gapi = service.list_jobs options
-          Job::List.from_gapi gapi, service, options
+          parent_job = parent_job.job_id if parent_job.is_a? Job
+          options = {
+            parent_job_id: parent_job,
+            all: all,
+            token: token,
+            max: max, filter: filter,
+            min_created_at: min_created_at,
+            max_created_at: max_created_at
+          }
+          gapi = service.list_jobs(**options)
+          Job::List.from_gapi gapi, service, **options
         end
 
         ##
@@ -1222,9 +1321,12 @@ module Google
         #                         "WHERE time_of_date = @time",
         #                         params: { time: fourpm }
         #
+        #   # Iterate over the first page of results
         #   data.each do |row|
         #     puts row[:name]
         #   end
+        #   # Retrieve the next page of results
+        #   data = data.next if data.next?
         #
         # @example Create Time with fractional seconds:
         #   require "google/cloud/bigquery"
@@ -1237,9 +1339,12 @@ module Google
         #                         "WHERE time_of_date >= @time",
         #                         params: { time: precise_time }
         #
+        #   # Iterate over the first page of results
         #   data.each do |row|
         #     puts row[:name]
         #   end
+        #   # Retrieve the next page of results
+        #   data = data.next if data.next?
         #
         def time hour, minute, second
           Bigquery::Time.new "#{hour}:#{minute}:#{second}"
@@ -1356,46 +1461,58 @@ module Google
         end
 
         ##
-        # Extracts the data from the provided table to a Google Cloud Storage
-        # file using an asynchronous method. In this method, an {ExtractJob} is
-        # immediately returned. The caller may poll the service by repeatedly
-        # calling {Job#reload!} and {Job#done?} to detect when the job is done,
-        # or simply block until the job is done by calling
+        # Extracts the data from a table or exports a model to Google Cloud Storage
+        # asynchronously, immediately returning an {ExtractJob} that can be used to
+        # track the progress of the export job.  The caller may poll the service by
+        # repeatedly calling {Job#reload!} and {Job#done?} to detect when the job
+        # is done, or simply block until the job is done by calling
         # #{Job#wait_until_done!}. See {#extract} for the synchronous version.
-        # Use this method instead of {Table#extract_job} to extract data from
-        # source tables in other projects.
+        #
+        # Use this method instead of {Table#extract_job} or {Model#extract_job} to
+        # extract data from source tables or models in other projects.
         #
         # The geographic location for the job ("US", "EU", etc.) can be set via
         # {ExtractJob::Updater#location=} in a block passed to this method.
         #
-        # @see https://cloud.google.com/bigquery/exporting-data-from-bigquery
-        #   Exporting Data From BigQuery
+        # @see https://cloud.google.com/bigquery/docs/exporting-data
+        #   Exporting table data
+        # @see https://cloud.google.com/bigquery-ml/docs/exporting-models
+        #   Exporting models
         #
-        # @param [String, Table] table The source table from which to extract
-        #   data. This can be a table object; or a string ID as specified by the
-        #   [Standard SQL Query
+        # @param [Table, Model, String] source The source table or model for
+        #   the extract operation. This can be a table or model object; or a
+        #   table ID string as specified by the [Standard SQL Query
         #   Reference](https://cloud.google.com/bigquery/docs/reference/standard-sql/query-syntax#from-clause)
         #   (`project-name.dataset_id.table_id`) or the [Legacy SQL Query
         #   Reference](https://cloud.google.com/bigquery/query-reference#from)
         #   (`project-name:dataset_id.table_id`).
         # @param [Google::Cloud::Storage::File, String, Array<String>]
         #   extract_url The Google Storage file or file URI pattern(s) to which
-        #   BigQuery should extract the table data.
-        # @param [String] format The exported file format. The default value is
-        #   `csv`.
+        #   BigQuery should extract. For a model export this value should be a
+        #   string ending in an object name prefix, since multiple objects will
+        #   be exported.
+        # @param [String] format The exported file format. The default value for
+        #   tables is `csv`. Tables with nested or repeated fields cannot be
+        #   exported as CSV. The default value for models is `ml_tf_saved_model`.
         #
-        #   The following values are supported:
+        #   Supported values for tables:
         #
         #   * `csv` - CSV
         #   * `json` - [Newline-delimited JSON](http://jsonlines.org/)
         #   * `avro` - [Avro](http://avro.apache.org/)
+        #
+        #   Supported values for models:
+        #
+        #   * `ml_tf_saved_model` - TensorFlow SavedModel
+        #   * `ml_xgboost_booster` - XGBoost Booster
         # @param [String] compression The compression type to use for exported
         #   files. Possible values include `GZIP` and `NONE`. The default value
-        #   is `NONE`.
+        #   is `NONE`. Not applicable when extracting models.
         # @param [String] delimiter Delimiter to use between fields in the
-        #   exported data. Default is <code>,</code>.
-        # @param [Boolean] header Whether to print out a header row in the
-        #   results. Default is `true`.
+        #   exported table data. Default is `,`. Not applicable when extracting
+        #   models.
+        # @param [Boolean] header Whether to print out a header row in table
+        #   exports. Default is `true`. Not applicable when extracting models.
         # @param [String] job_id A user-defined ID for the extract job. The ID
         #   must contain only letters (a-z, A-Z), numbers (0-9), underscores
         #   (_), or dashes (-). The maximum length is 1,024 characters. If
@@ -1412,40 +1529,60 @@ module Google
         #   is 1,024 characters. If `job_id` is provided, then `prefix` will not
         #   be used.
         # @param [Hash] labels A hash of user-provided labels associated with
-        #   the job. You can use these to organize and group your jobs. Label
-        #   keys and values can be no longer than 63 characters, can only
-        #   contain lowercase letters, numeric characters, underscores and
-        #   dashes. International characters are allowed. Label values are
-        #   optional. Label keys must start with a letter and each label in the
-        #   list must have a different key. See [Requirements for
-        #   labels](https://cloud.google.com/bigquery/docs/creating-managing-labels#requirements).
+        #   the job. You can use these to organize and group your jobs.
+        #
+        #   The labels applied to a resource must meet the following requirements:
+        #
+        #   * Each resource can have multiple labels, up to a maximum of 64.
+        #   * Each label must be a key-value pair.
+        #   * Keys have a minimum length of 1 character and a maximum length of
+        #     63 characters, and cannot be empty. Values can be empty, and have
+        #     a maximum length of 63 characters.
+        #   * Keys and values can contain only lowercase letters, numeric characters,
+        #     underscores, and dashes. All characters must use UTF-8 encoding, and
+        #     international characters are allowed.
+        #   * The key portion of a label must be unique. However, you can use the
+        #     same key with multiple resources.
+        #   * Keys must start with a lowercase letter or international character.
         # @yield [job] a job configuration object
         # @yieldparam [Google::Cloud::Bigquery::ExtractJob::Updater] job a job
         #   configuration object for setting additional options.
         #
         # @return [Google::Cloud::Bigquery::ExtractJob]
         #
-        # @example
+        # @example Export table data
         #   require "google/cloud/bigquery"
         #
         #   bigquery = Google::Cloud::Bigquery.new
         #
         #   table_id = "bigquery-public-data.samples.shakespeare"
-        #   extract_job = bigquery.extract_job table_id,
-        #                                      "gs://my-bucket/shakespeare.csv"
+        #   extract_job = bigquery.extract_job table_id, "gs://my-bucket/shakespeare.csv"
         #   extract_job.wait_until_done!
         #   extract_job.done? #=> true
         #
+        # @example Export a model
+        #   require "google/cloud/bigquery"
+        #
+        #   bigquery = Google::Cloud::Bigquery.new
+        #   dataset = bigquery.dataset "my_dataset"
+        #   model = dataset.model "my_model"
+        #
+        #   extract_job = bigquery.extract model, "gs://my-bucket/#{model.model_id}"
+        #
         # @!group Data
         #
-        def extract_job table, extract_url, format: nil, compression: nil, delimiter: nil, header: nil, job_id: nil,
+        def extract_job source, extract_url, format: nil, compression: nil, delimiter: nil, header: nil, job_id: nil,
                         prefix: nil, labels: nil
           ensure_service!
           options = { format: format, compression: compression, delimiter: delimiter, header: header, job_id: job_id,
                       prefix: prefix, labels: labels }
+          source_ref = if source.respond_to? :model_ref
+                         source.model_ref
+                       else
+                         Service.get_table_ref source, default_ref: project_ref
+                       end
 
-          table_ref = Service.get_table_ref table, default_ref: project_ref
-          updater = ExtractJob::Updater.from_options service, table_ref, extract_url, options
+          updater = ExtractJob::Updater.from_options service, source_ref, extract_url, options
 
           yield updater if block_given?
 
@@ -1455,51 +1592,63 @@ module Google
         end
 
         ##
-        # Extracts the data from the provided table to a Google Cloud Storage
-        # file using a synchronous method that blocks for a response. Timeouts
+        # Extracts the data from a table or exports a model to Google Cloud Storage
+        # using a synchronous method that blocks for a response. Timeouts
         # and transient errors are generally handled as needed to complete the
-        # job. See {#extract_job} for the asynchronous version. Use this method
-        # instead of {Table#extract} to extract data from source tables in other
-        # projects.
+        # job. See {#extract_job} for the asynchronous version.
+        #
+        # Use this method instead of {Table#extract} or {Model#extract} to
+        # extract data from source tables or models in other projects.
         #
         # The geographic location for the job ("US", "EU", etc.) can be set via
         # {ExtractJob::Updater#location=} in a block passed to this method.
         #
-        # @see https://cloud.google.com/bigquery/exporting-data-from-bigquery
-        #   Exporting Data From BigQuery
+        # @see https://cloud.google.com/bigquery/docs/exporting-data
+        #   Exporting table data
+        # @see https://cloud.google.com/bigquery-ml/docs/exporting-models
+        #   Exporting models
         #
-        # @param [String, Table] table The source table from which to extract
-        #   data. This can be a table object; or a string ID as specified by the
-        #   [Standard SQL Query
+        # @param [Table, Model, String] source The source table or model for
+        #   the extract operation. This can be a table or model object; or a
+        #   table ID string as specified by the [Standard SQL Query
         #   Reference](https://cloud.google.com/bigquery/docs/reference/standard-sql/query-syntax#from-clause)
         #   (`project-name.dataset_id.table_id`) or the [Legacy SQL Query
         #   Reference](https://cloud.google.com/bigquery/query-reference#from)
         #   (`project-name:dataset_id.table_id`).
         # @param [Google::Cloud::Storage::File, String, Array<String>]
         #   extract_url The Google Storage file or file URI pattern(s) to which
-        #   BigQuery should extract the table data.
-        # @param [String] format The exported file format. The default value is
-        #   `csv`.
+        #   BigQuery should extract. For a model export this value should be a
+        #   string ending in an object name prefix, since multiple objects will
+        #   be exported.
+        # @param [String] format The exported file format. The default value for
+        #   tables is `csv`. Tables with nested or repeated fields cannot be
+        #   exported as CSV. The default value for models is `ml_tf_saved_model`.
         #
-        #   The following values are supported:
+        #   Supported values for tables:
         #
         #   * `csv` - CSV
         #   * `json` - [Newline-delimited JSON](http://jsonlines.org/)
         #   * `avro` - [Avro](http://avro.apache.org/)
+        #
+        #   Supported values for models:
+        #
+        #   * `ml_tf_saved_model` - TensorFlow SavedModel
+        #   * `ml_xgboost_booster` - XGBoost Booster
         # @param [String] compression The compression type to use for exported
         #   files. Possible values include `GZIP` and `NONE`. The default value
-        #   is `NONE`.
+        #   is `NONE`. Not applicable when extracting models.
         # @param [String] delimiter Delimiter to use between fields in the
-        #   exported data. Default is <code>,</code>.
-        # @param [Boolean] header Whether to print out a header row in the
-        #   results. Default is `true`.
+        #   exported table data. Default is `,`. Not applicable when extracting
+        #   models.
+        # @param [Boolean] header Whether to print out a header row in table
+        #   exports. Default is `true`. Not applicable when extracting models.
         # @yield [job] a job configuration object
         # @yieldparam [Google::Cloud::Bigquery::ExtractJob::Updater] job a job
         #   configuration object for setting additional options.
         #
         # @return [Boolean] Returns `true` if the extract operation succeeded.
         #
-        # @example
+        # @example Export table data
         #   require "google/cloud/bigquery"
         #
         #   bigquery = Google::Cloud::Bigquery.new
@@ -1507,10 +1656,19 @@ module Google
         #   bigquery.extract "bigquery-public-data.samples.shakespeare",
         #                    "gs://my-bucket/shakespeare.csv"
         #
+        # @example Export a model
+        #   require "google/cloud/bigquery"
+        #
+        #   bigquery = Google::Cloud::Bigquery.new
+        #   dataset = bigquery.dataset "my_dataset"
+        #   model = dataset.model "my_model"
+        #
+        #   bigquery.extract model, "gs://my-bucket/#{model.model_id}"
+        #
         # @!group Data
         #
-        def extract table, extract_url, format: nil, compression: nil, delimiter: nil, header: nil, &block
-          job = extract_job table, extract_url,
+        def extract source, extract_url, format: nil, compression: nil, delimiter: nil, header: nil, &block
+          job = extract_job source, extract_url,
                             format:      format,
                             compression: compression,
                             delimiter:   delimiter,