A Python library for querying/updating Saleforce.com data via SOAP API
Project description
Introduction
This is a distutils-packaged and updated version of the beatbox module by Simon Fell, which is a Python implementation of a client for the Salesforce.com Partner Web Services API.
This module contains 2 versions of the Salesforce.com client:
- XMLClient
The original beatbox version of the client which returns xmltramp objects.
- PythonClient
Marshalls the returned objects into proper Python data types. e.g. integer fields return integers.
Compatibility
Beatbox supports version 16.0 of the Salesforce Partner Web Services API. However, the following API calls have not been implemented at this time:
convertLead
emptyRecycleBin
invalidateSessions
logout
merge
process
queryAll
undelete
describeSObject
sendEmail
Beatbox has been tested with Python 2.4 and Python 2.6.
Basic Usage Examples
- Instantiate a Python Salesforce.com client:
>>> svc = beatbox.PythonClient() >>> svc.login('username', 'passwordTOKEN')
(Note that interacting with Salesforce.com via the API requires the use of a ‘security token’ which must be appended to the password.)
- Query for contacts with last name ‘Doe’:
>>> res = svc.query("SELECT Id, FirstName, LastName FROM Contact WHERE LastName='Doe'") >>> res[0] {'LastName': 'Doe', 'type': 'Contact', 'Id': '0037000000eRf6vAAC', 'FirstName': 'John'} >>> res[0].Id '0037000000eRf6vAAC'
- Add a new Lead:
>>> contact = {'type': 'Lead', 'LastName': 'Glick', 'FirstName': 'David', 'Company': 'Individual'} >>> res = svc.create(contact)
- Get the ID of the newly created Lead:
>>> res[0]['id'] '00Q7000000RVyiHEAT'
More Examples
The examples folder contains the examples for the original beatbox. For examples on how to use the PythonClient see src/beatbox/tests/test_pythonClient.py.
Some of these other products that have been built on top of beatbox can also provide example of use:
Alternatives
David Lanstein has created a Python Salesforce Toolkit that is based on the suds SOAP library. Based on limited tests it appears to be somewhat slower than beatbox for operations that return a lot of data; however, it may be a better option if you want to be able to automatically generate a service proxy for a new WSDL (such as for the Enterprise web services API).
Ron Hess from Salesforce.com has adapted beatbox for use with Google App Engine. See http://code.google.com/p/force-app-engine/
Running Tests
First, we need to add some custom fields to the Contacts object in your Salesforce instance:
Login to your Salesforce.com instance
Browse to Setup –> Customize –> Contacts –> Fields –> “New” button
- Add a Picklist (multi-select) labeled “Favorite Fruit”, then add
Apple
Orange
Pear
Leave default of 3 lines and field name should default to “Favorite_Fruit”
Add a Number labeled “Favorite Integer”, with 18 places, 0 decimal places
Add a Number labeled “Favorite Float”, with 13 places, 5 decimal places
Create a sfconfig file in your python path with the following format:
USERNAME='your salesforce username' PASSWORD='your salesforce passwordTOKEN'
where TOKEN is your Salesforce API login token.
Add ‘./src’ to your PYTHONPATH
Run the tests:
python src/beatbox/tests/test_beatbox.py python src/beatbox/tests/test_pythonClient.py
Changelog
16.1 (2010-03-11)
Catch and retry on exceptions from the socket library, in addition to ones from httplib. This fixes a regression introduced in version 16.0. [davisagli]
16.0 (2009-11-12)
Don’t strip newlines when marshalling the values of textarea fields. [davisagli]
Make sure to add a field to fieldsToNull if its Python value is None. [rhettg, davisagli]
Fix issue where numbers of type long weren’t converted to a string. [spleeman, davisagli]
Only catch HTTP exceptions when retrying a connection. [spleeman, davisagli]
16.0b1 (2009-09-08)
Log beatbox calls at the debug level. [davisagli]
Fixed a string exception for compatibility with Python 2.6. [davisagli]
Added support for SOSL searches via the search method. Thanks to Alex Tokar of Web Collective. [davisagli]
Added an optional cache for the sObject type descriptions needed for marshalling query results into Python objects. This can avoid an extra describeSObjects API call for each query, but means that the information could become stale if the type metadata is modified in Salesforce.com. The cache is off by default. Turn it on by passing cacheTypeDescriptions=True when instantiating a Python client. The cache may be reset by calling the flushTypeDescriptionsCache method of the Python client. [davisagli]
Support a full SOQL statement as a parameter to the query method of the Python client. The old 3-part method signature (fields, sObjectType, conditionalExpression) should continue to work. [davisagli]
In the Python client, support relationship queries and other queries that may return multiple types of objects. Object type descriptions (required for marshalling field values into the correct Python type) are cached for the duration of the query after the first time they are used. Thanks to Melnychuk Taras of Quintagroup. [davisagli]
In the Python client, queries now return a list-like QueryRecordSet holding a sequence of dict-like QueryRecord objects, instead of a dict containing a list of dicts. This allows for more Pythonic access such as results[0].Id instead of results[‘results’][0][‘Id’]. The old syntax should still work. Thanks to Melnychuk Taras of Quintagroup. [davisagli]
Update to use version 16.0 of the Salesforce.com partner WSDL. [davisagli]
0.12 (2009-05-13)
Use the default serverUrl value if the passed value evaluates to boolean False. [davisagli]
0.11 (2009-05-13)
Access ‘created’ instead of ‘isCreated’ in the upsert result. This closes http://code.google.com/p/salesforce-beatbox/issues/detail?id=4 [davisagli]
0.10 (2009-05-06)
Added optional serverUrl parameter when creating a Client. [davisagli]
pre 0.9.1.1
ancient history
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.