Metadata api

Syntax#

Parameters#

Parameter name	Description
reportType	Report type. Allowed Values: ga. Where ga corresponds to the Core Reporting API.

Remarks#

The Metadata API returns the list and attributes of columns (i.e. dimensions and metrics) exposed in the Google Analytics reporting APIs (v2,v3 and v4). Attributes returned include UI name, description, segments support, and more.

You can use the Metadata API to:

Automatically discover new columns
Access all dimensions and metrics attributes for Google Analytics reporting APIs

This is the same list as is in the Dimensions & Metrics Explorer.

Note: Real-time and Multi-Channel Funnels dimensions and metrics are not currently available.

Rest Example

Calls to the Metadata API are with HTTP Get:

Using Public API Key

GET https://www.googleapis.com/analytics/v3/metadata/ga/columns?key={YOUR_API_KEY}

Using Access token from either Oauth2 or Service account authentication

GET https://www.googleapis.com/analytics/v3/metadata/ga/columns?access_token={Authentcated_Access_Token}

Java example

uses the Java Client library

/**
 * 1. Execute a Metadata Request
 * An application can request columns data by calling the list method on the Analytics service object.
 * The method requires an reportType parameter that specifies the column data to retrieve.
 * For example, the following code requests columns for the ga report type.
 */

try {
  Columns results = getMetadata(analytics);
  // Success

} catch (GoogleJsonResponseException e) {
  // Catch API specific errors.
  handleApiError(e);
} catch (IOException e) {
  // Catch general parsing network errors.
  e.printStackTrace();
}

/**
 * 2. Print out the Columns data
 * The components of the result can be printed out as follows:
 */

private static Columns getMetadata(Analytics analytics) throws IOException {
  String reportType = "ga";
  return analytics.metadata()
      .columns()
      .list(reportType)
      .execute();
}


private static void printMetadataReport(Columns results) {
  System.out.println("Metadata Response Report");
  printReportInfo(results);
  printAttributes(results.getAttributeNames());
  printColumns(results.getItems());
}


private static void printReportInfo(Columns results) {
  System.out.println("## Metadata Report Info ##");
  System.out.println("Kind: " + results.getKind());
  System.out.println("Etag: " + results.getEtag());
  System.out.println("Total Results: " + results.getTotalResults());
  System.out.println();
}


private static void printAttributes(List<String> attributeNames) {
  System.out.println("## Attribute Names ##");
  for (String attribute : attributeNames) {
    System.out.println(attribute);
  }
}


private static void printColumns(List<Column> columns) {
  System.out.println("## Columns ##");

  for (Column column : columns) {
    System.out.println();
    System.out.println("Column ID: " + column.getId());
    System.out.println("Kind: " + column.getKind());

    Map<String, String> columnAttributes = column.getAttributes();

    for (Map.Entry<String, String> attribute: columnAttributes.entrySet()) {
      System.out.println(attribute.getKey() + ": " + attribute.getValue());
    }
  }
}

Note: first version copied from Metadata.list

PHP Example

Uses the PHP client library

/**
 * 1. Execute a Metadata Request
 * An application can request columns data by calling the list method on the Analytics service object.
 * The method requires an reportType parameter that specifies the column data to retrieve.
 * For example, the following code requests columns for the ga report type.
 */

try {

  $results = $analytics->metadata_columns->listMetadataColumns('ga');
  // Success

} catch (apiServiceException $e) {
  // Handle API service exceptions.
  $error = $e->getMessage();
}


/**
 * 2. Print out the Columns data
 * The components of the result can be printed out as follows:
 */

function printMetadataReport($results) {
  print '<h1>Metadata Report</h1>';
  printReportInfo($results);
  printAttributes($results);
  printColumns($results);
}


function printReportInfo(&$results) {
  $html = '<h2>Report Info</h2>';
  $html .= <<<HTML
<pre>
Kind                  = {$results->getKind()}
Etag                  = {$results->getEtag()}
Total Results         = {$results->getTotalResults()}
</pre>
HTML;
  print $html;
}


function printAttributes(&$results) {
  $html = '<h2>Attribute Names</h2><ul>';
  $attributes = $results->getAttributeNames();
  foreach ($attributes as $attribute) {
    $html .= '<li>'. $attribute . '</li>';
  }
  $html .= '</ul>';
  print $html;
}


function printColumns(&$results) {
  $columns = $results->getItems();
  if (count($columns) > 0) {
    $html = '<h2>Columns</h2>';
    foreach ($columns as $column) {
      $html .= '<h3>' . $column->getId() . '</h3>';
      $column_attributes = $column->getAttributes();
      foreach ($column_attributes as $name=>$value) {
        $html .= <<<HTML
<pre>
{$name}: {$value}
</pre>
HTML;
      }
    }
  } else {
    $html = '<p>No Results Found.</p>';
  }
  print $html;
}

Note: original version copied from metadata.list

Python Example

Uses the Python client library

# 1. Execute a Metadata Request
# An application can request columns data by calling the list method on the Analytics service object.
# The method requires an reportType parameter that specifies the column data to retrieve.
# For example, the following code requests columns for the ga report type.

try:
  results = service.metadata().columns().list(reportType='ga').execute()

except TypeError, error:
  # Handle errors in constructing a query.
  print ('There was an error in constructing your query : %s' % error)

except HttpError, error:
  # Handle API errors.
  print ('Arg, there was an API error : %s : %s' %
         (error.resp.status, error._get_reason()))

# 2. Print out the Columns data
# The components of the result can be printed out as follows:

def print_metadata_report(results):
  print 'Metadata Response Report'
  print_report_info(results)
  print_attributes(results.get('attributeNames'))
  print_columns(results)


def print_report_info(columns):
  print "Metadata Report Info"
  if columns:
    print 'Kind           = %s' % columns.get('kind')
    print 'Etag           = %s' % columns.get('etag')
    print 'Total Results  = %s' % columns.get('totalResults')


def print_attributes(attributes):
  if attributes:
    print 'Attribute Names:'
    for attribute in attributes:
      print attribute

def print_columns(columns_data):
  if columns_data:
    print 'Columns:'

    columns = columns_data.get('items', [])

    for column in columns:
      print
      print '%15s = %35s' % ('Column ID', column.get('id'))
      print '%15s = %35s' % ('Kind', column.get('kind'))

      column_attributes = column.get('attributes', [])

      for name, value in column_attributes.iteritems():
        print '%15s = %35s' % (name, value)

Note: original version copied from metadata.list

C# example

Uses the .Net Client library

PM> Install-Package Google.Apis.Analytics.v3

var metadataService = new AnalyticsMetaDataService(new BaseClientService.Initializer()
            {
                ApiKey = {Public API KEY},
                ApplicationName = "Metadata api",
            });

var result = Service.Metadata.Columns.List("ga").Execute();