Skip to content


Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
Python wrapper for all FamilySearch APIs
branch: master
Failed to load latest commit information.
familysearch Bump version to 0.3pre
.gitignore Ignore *.html files (documentation generated by rst2html)
CHANGELOG.rst Bump version to 0.3pre
LICENSE Initial commit Add test suite to, which requires setuptools instead of dist…
README.rst Add project links and more installation instructions to README
ROADMAP.rst Update roadmap in preparation for release of 0.2



python-fs-stack provides a Python package that simplifies access to the FamilySearch REST-style API.

Home Page:
Source Code:


  • Python 2.4 or later
  • simplejson, if using Python older than 2.6
  • wsgi_intercept 0.5.0 or later (only required to run test suite)


Using pip:

pip install python-fs-stack

or using easy_install (from setuptools or distribute):

easy_install python-fs-stack

or (after downloading manually):

python install

Example Usage

First, import the FamilySearch class:

from familysearch import FamilySearch

Authenticating with FamilySearch

python-fs-stack supports several ways of initiating a session with FamilySearch, including Basic Authentication, OAuth, and resuming a previous session.

Log in immediately with Basic Authentication:

fs = FamilySearch('ClientApp/1.0', 'developer_key', 'username', 'password')

Log in in a separate step with Basic Authentication:

fs = FamilySearch('ClientApp/1.0', 'developer_key')
fs.login('username', 'password')

Log in in two steps with Basic Authentication:

fs = FamilySearch('ClientApp/1.0', 'developer_key')
fs.authenticate('username', 'password')

Log in with OAuth:

import webbrowser
fs = FamilySearch('ClientApp/1.0', 'developer_key')
# [Enter username and password into browser window that opens]
verifier = [verifier from resulting web page]

Resume a previous session:

fs = FamilySearch('ClientApp/1.0', 'developer_key', session='session_id')

Use the production system instead of the reference system:

fs = FamilySearch('ClientApp/1.0', 'developer_key', base='')

Maintaining and Ending a Session

Keep the current session active:


Log out:


Accessing Family Tree Information

Print current user's family tree details:

print fs.person()

To specify a person ID to retrieve, pass the ID as an argument:

print fs.person('ABCD-123')

To print multiple family tree entries, pass a list of IDs as an argument. To pass additional parameters to the API, simply pass them as named arguments:

print fs.person(['ABCD-123', 'EFGH-456'], events='all', children='all')

Print the latest version of a list of persons (this request is more lightweight than a full person request, so it supports more IDs at once):

for person in fs.version(['ABCD-123', 'EFGH-456']):
    print person['id'], person['version']

Print the contents of a persona:

print fs.persona('ABCD-123')

Print current user's pedigree:

print fs.pedigree()

Format the pedigree output more nicely:

import pprint

Searching for Persons in the Family Tree

Search for a male named John Smith:

results ='John', familyName='Smith', gender='Male', maxResults=10)

Retrieve the second page of the previous search:

more_results =[0]['contextId'], maxResults=10, startIndex=10)

Search for an exact match for John Smith (use an options dict to specify options with periods in their names):

results ={'givenName.exact': 'John', 'familyName.exact': 'Smith'}, gender='Male', maxResults=10)

Searching for Possible Duplicates

Search for possible duplicates of a person:

matches = fs.match('ABCD-123')

Compute match score between two persons:

match = fs.match('ABCD-123', id='EFGH-456')

Search for possible duplicates matching specified parameters:

matches = fs.match(givenName='John', familyName='Smith', gender='Male', birthDate='1900', birthPlace='USA', deathDate='1950', deathPlace='USA')

Standardizing Places, Names, and Dates

Look up a place by name:

place ='paris')

Look up a place by ID:

place =

Look up a list of places by ID:

places =[5061509, 5061446])

Look up a place by name, showing only the most likely result, returning results in another locale:

place ='germany', filter=True, locale='de')

Standardize a name:

name ='John Smith')

Standardize a list of names:

names =['John Smith', 'Jane Doe'])

Standardize a date:

date ='1-1-11')

Standardize a list of dates:

dates =['1-1-11', 'december 31 1999'])
Something went wrong with that request. Please try again.