neocrimesecurity / python-ldap_paged_search

Python library to easily perform large (>1000 records) LDAP queries

python-ldap_paged_search 0.4.1

Summary

ldap_paged_search is a python library to easily perform LDAP queries with more than 1000 results, or to break down queries into smaller result sets to reduce server loads.

Many LDAP servers, such as active directory, will not return more than 1000 results unless paged requests are used. The existing python ldap library does support pageing, but requires some not very intuitive coding to perform it. This library is simply a wrapper to make paged searches easy.

Its interface is also slightly easier to perform queries than the LDAP library it wraps.

Requirements

• Tested on python 2.8
• Default python library includes ldap library

Installation

Via pip or easy_install

$ sudo pip install ldap_paged_search   # If you prefer PIP

$ sudo easy_install ldap_paged_search  # If you prefer easy_install

Manual installation

$ git clone https://github.com/neoCrimeLabs/python-ldap_paged_search.git
$ cd python-ldap_paged_search
$ sudo python setup.py install

Usage

Initial setup

from ldap_paged_search import LdapPagedSearch

# Required values
url             = 'ldap://your.ldap.server'
username        = 'username'      # for anything but active directory
username        = 'domain\\user'  # for active directory
password        = 'yourPassword'

baseDN          = 'dc=company,dc=com'
searchFilter    = '(&(objectCategory=user))'

# Optional values
maxPages        = 0     # 0 = everything
maxPages        = 10    # Return first 10 pages only

attributes      = ['*']                         # Return all fields
attributes      = ['FieldName', 'AnotherField'] # Return specific fields only


pageSize        = 1000  # How many records per page
                        # Usual max is 1000; check your LDAP server docs

Defining a callback method

# Using a callback method to process results uses less memory on large queries
# Not using a callback search() will return all results as a single list

def myCallback(dn,record):
    print dn, record

Query using 'with'

# maxPages, pageSize, attributes, and callback are all OPTIONAL

with LdapPagedSearch(url, username, password, maxPages=2, pageSize=2 ) as l:
    results = l.search(baseDN, searchFilter, attributes = attributes, callback = myCallback)

Alternative query method

# maxPages, pageSize, attributes, and callback are all OPTIONAL

l = LdapPagedSearch(url, username, password, maxPages=2, pageSize=2 )
results = l.search(baseDN, searchFilter, attributes = attributes, callback = myCallback)

Results format

# If you don't set a callback, your results will be returned as follows

[
    ('DistinctName1',
        {  'FieldName':    ['value1', 'value2'],
           'AnotherField': ['value'], }),
    ('DistinctName2',
        {  'FieldName':    ['value1', 'value2'],
           'AnotherField': ['value'], }),
    ...
]

Conditions of Use

I wrote this library for my own use, but realized others may find it useful as there were many forum topics describing this problem.

Unfortunately I cannot guarentee any active support, but will do my best as time permits. That said, I'll happily accept push requests with suitable changes that address the general audience of this library.

Put simply, use this at your own risk. If it works, great! If not, I may not be able to help you. If you fix anything, however, please push it back and I'll likely accept it. :-)

Also, if you use this library in your package, tool, or comercial software, let me know, and I'll list it here!

python-ldap_paged_search's Issues

Missing self reference

Code should be self._doSearch instead of _doSearch @ this line.