A newer version of this documentation is available.

View Latest

Sample Code

The Python SDK supports creation and handling of extended attributes.

Subdocument Operations and Extended Attributes

A high-level summary of extended attributes can be found in Extended Attributes. Extended attributes are handled by means of extensions to the Subdocument API.

Python Extended Attributes Example

The following code demonstrates how extended attributes can be used. It assumes that Couchbase Server is established on localhost; that the Full Administrator username and password are Administrator and password respectively; and that the travel-sample bucket is installed. For information on installing the travel-sample bucket, see Sample Buckets.

import logging
import sys
import couchbase.subdocument as SD

from couchbase.cluster import Cluster, PasswordAuthenticator

logging.basicConfig(stream=sys.stderr, level=logging.DEBUG)

# open cluster and authenticate as Cluster Admin
cluster = Cluster('couchbase://10.112.170.101:8091')
cluster.authenticate(PasswordAuthenticator('Administrator', 'password'))

# open travel-sample bucket
bucket = cluster.open_bucket('travel-sample')

# Add key-value pairs to hotel_10138, representing traveller-Ids and associated discount percentages
bucket.mutate_in('hotel_10138',
                SD.upsert('discounts.jsmith123', '20', xattr=True, create_parents=True),
                SD.upsert('discounts.pjones356', '30', xattr=True, create_parents=True),
                # The following lines, "insert" and "remove", simply demonstrate insertion and
                # removal of the same path and value
                SD.insert('discounts.jbrown789', '25', xattr=True, create_parents=True),
                SD.remove('discounts.jbrown789', xattr=True)
                )

# Add key - value pairs to hotel_10142, again representing traveller - Ids and associated discount percentages
bucket.mutate_in('hotel_10142',
                SD.upsert('discounts.jsmith123', '15', xattr=True, create_parents=True),
                SD.upsert('discounts.pjones356', '10', xattr=True, create_parents=True)
                )

# Create a user and assign roles. This user will search for their available discounts.
manager = cluster.cluster_manager()
                manager.user_upsert('jsmith123', 'jsmith123pwd', [
                # Roles required for the reading of data from the bucket
                ('data_reader', 'travel-sample'),
                ('query_select', 'travel-sample'),

                # Roles required for the writing of data into the bucket
                ('data_writer', 'travel-sample'),
                ('query_insert', 'travel-sample'),
                ('query_delete', 'travel-sample'),

                # Role required for the creation of indexes on the bucket
                ('query_manage_index', 'travel-sample')
                ], 'John Smith')

# reconnect using new user
cluster = Cluster('couchbase://10.112.170.101')
cluster.authenticate(PasswordAuthenticator('jsmith123', 'jsmith123pwd'))

# must have at least one open bucket to submit cluster query
bucket = cluster.open_bucket('travel-sample')

# Perform a N1QL Query to return document IDs from the bucket. These IDs will be
# used to reference each document in turn, and check for extended attributes
# corresponding to discounts.
result = cluster.n1ql_query('SELECT id, meta(`travel-sample`).id AS docID FROM `travel-sample`')
results = ''

for row in result:
    # get row document ID
    docID = row['docID']

    # Determine whether a hotel-discount has been applied to this user.
    r = bucket.lookup_in(docID, SD.exists('discounts.jsmith123', xattr=True))
    if r.exists('discounts.jsmith123'):

        # If so, get the discount-percentage.
        r = bucket.lookup_in(docID, SD.get('discounts.jsmith123', xattr=True))
        discount = r['discounts.jsmith123']

        # If the percentage - value is greater than 15, include the document in the
        # results to be returned.
        results = '%s\n%s - %s' % (results, discount, docID)

print("Results returned are: %s" % results)