Skip to main content

Python library for integrating Lenddo Services. https://www.lenddo.com

Project description

Python Lenddo API Client

Table of Contents

Introduction

The lenddo_api_client python module provides the class LenddoAPIClient, a generic interface to all the Lenddo HTTP APIs. The client makes authenticated HTTP requests using the HMAC-SHA1 signing method.

Installation

To install our client use pip:

$ pip install lenddo

Version 1.1.0 supports both python 2.7.x and python 3.x. If any bug is encountered kindly report the issue here.

Manual Installation

The python Lenddo API client library has no third-party dependencies. The only installation step is to copy and drop the directory lenddo_api_client/ into a directory in your PYTHONPATH. The client has been tested with python versions 2.7.x and 3.6.x.

General Usage

In order to make API calls, first instantiate a LenddoAPIClient initialized with your API id, your API secret, the base URL of the API resource you intend to use. You may then make HTTP requests using the client's get, post, put, delete and options methods, which are named after the respective HTTP methods. Arguments to these methods take the form (resource-type, resource-id, parameter-dictionary):

  • resource-type describes the resource being requested (for example ApplicationScore, PartnerToken)
  • resource-id identifies the specific instance of the requested resource
  • parameter-dictionary contains additional parameters supported by the call. In the case of a GET request, these are converted to the request query string. In a POST or PUT request, they are JSON-encoded and submitted as the request body.

Optionally, you can use proxies when connecting to external services.

Example: Requesting one of your application's Lenddo Score

In this example we make a GET request to the ApplicationScore API to obtain the score for a application with id example-application. ApplicationScore is hosted at scoreservice.lenddo.com.

Request a Lenddo score with the LenddoAPIClient

from lenddo_api_client import LenddoAPIClient

client = LenddoAPIClient('your-api-client-id', 'your-api-client-secret',
	'https://scoreservice.lenddo.com')
response = client.get('ApplicationScore', 'example-application')

This results in a signed HTTP GET request to https://scoreservice.lenddo.com/ApplicationScore/example-application.

Request a Lenddo score behind a HTTP proxy

Say your HTTP proxy is running at the http://192.168.1.100 address in your internal network:

from lenddo_api_client import LenddoAPIClient

HTTP_PROXIES = {
    'http': 'http://192.168.1.100',
    'https': 'http://192.168.1.100',
}

client = LenddoAPIClient(
            'your-api-client-id', 'your-api-client-secret', 
            'https://scoreservice.lenddo.com', proxies=HTTP_PROXIES
         )
response = client.get('ApplicationScore', 'example-application')         

The HTTP_PROXIES dictionary is based on python's 'proxies' dict.

Exceptions and Error Handling

Exceptions raised by the client are standard exceptions documented in standard library docs. The sdk client provides a wrapper module for these exceptions in lenddo_api_client.errors.

exception description 2.x 3.x
HTTPError Raised when an API call returns a non-success status. link link
URLError Raised on network error. link link

The following is the same request as in the previous example, this time with error handling included:

import json

from lenddo_api_client import LenddoAPIClient
from lenddo_api_client.errors import HTTPError, URLError


client = LenddoAPIClient(
    'your-api-client-id', 
    'your-api-client-secret',
    'https://scoreservice.lenddo.com'
)
try:
    response = client.get('ApplicationScore', 'example-application')
except HTTPError as e:
    print('API call failed with status %d' % e.code)

    # Error responses from the Lenddo APIs still return JSON bodies describing error details,
    # only now we have to decode the JSON ourselves.
    print(json.loads(e.read()))
except URLError as e:
    print('API call failed with reason %s' % e.reason)

For the sake of brevity, we omit error handling in subsequent examples. In production code, calls should be made taking into account at least the possibility of HTTPError.

Submitting Applications to Lenddo: Using the API Client as a White Label Solution

You may submit data directly to Lenddo while keeping your own branding, thus utilizing Lenddo services without having users leave your ecosystem. For each user application, this is accomplished in two phases: first by making POST calls to the PartnerToken resource, which saves OAuth tokens associated to your user's application, and finally by committing the job using the CommitPartnerJob resource. Both resources are hosted at networkservice.lenddo.com.

Applications are identified by a application_id which must be supplied as parameter to both calls. In addition, the partner_script_id parameter specifies how Lenddo informs you of the results. You may only commit one job per application_id/partner_script_id pair.

The PartnerToken API call

During the first phase of the job flow, make calls to PartnerToken to send social network OAuth tokens to Lenddo. Each call to PartnerToken returns a profile_id which you must save in order to send it along with all other profile_ids associated to this application when you finally commit the job. To add multiple tokens for the same application, make multiple PartnerToken calls, each time saving the returned profile_id.

PartnerToken takes the following arguments, all required unless stated otherwise:

  • application_id - a string that identifies the application to associate with an OAuth token It must match the application_id you use in the CommitPartnerJob step.
  • provider - the token provider. Valid values are: Facebook, LinkedIn, Yahoo, WindowsLive, or Google
  • token data - a dictionary of OAuth token data with keys key and extra_data.

    Note: All tokens must be OAuth 2.0.

    • key - the access token proper, a string
    • extra_data - optional dictionary of additional OAuth fields returned by the token provider.

Note

In order for Lenddo to generate the profile_id, it has to use the supplied OAuth token, which may fail (for example, when the user has denied permission or the token has expired). When this happens, PartnerToken will return with HTTP 400 and the body of the response will contain both the error code and response body that Lenddo received when trying to use the token. The provider's HTTP response is under provider_status_code and the provider's response body is under provider_response. The format of the provider_response varies among OAuth providers; see the OAuth provider's documentation for details.

Example

import json

from lenddo_api_client import LenddoAPIClient
from lenddo_api_client.errors import HTTPError


client = LenddoAPIClient(
    'your-api-client-id', 
    'your-api-client-secret',
    'https://networkservice.lenddo.com'
)
try:
    response = client.post('PartnerToken', None, {
        'application_id' : 'example-application',
        'provider' : 'Facebook',
        'token_data' : { 'key' : 'example-access-token' }})
except HTTPError as e:
    print('API call failed with HTTP status %d' % e.code)
    print(json.loads(e.read()))

profile_id = response['profile_id']
# ... and store this profile_id - application_id association

Errors

To inspect error details, catch HTTPError, call the exception object's 'read()' method, and decode the JSON response. In the table below, 'Error Name' refers to the string found in the response body's 'name' field.

Error Name HTTP Status Code Description
BAD_REQUEST 400 Request was malformed, or missing required data.
INVALID_TOKEN 400 Token data was missing required fields or fields had invalid values.
TOKEN_FAILURE 400 Failure upon attempt to use the token. provider_status_code contains the provider's HTTP status code. provider_response contains the provider's error response body.
INTERNAL_ERROR 500 An internal error occurred. If this persists please contact a Lenddo Representative.

The CommitPartnerJob API Call

Once there are no more tokens to associate to a given application, use the CommitPartnerJob call to have Lenddo compute a score for your user. To obtain the results of the job, see [Requesting Results from Lenddo] (#requesting-results-from-lenddo).

CommitPartnerJob takes the following required arguments:

  • partner script id - Please reference the developer section of the partner dashboard. This will define how you're notified of scoring results.
  • application id - an application id that, coupled with the partner script id, identifies this job. For any given application, this is the same application_id used in the PartnerToken call. You can use this value to retrieve score results.
  • profile ids - a list of profile_ids gathered from the results of the PartnerToken call.

In addition to these arguments, CommitPartnerJob accepts the following:

  • verification_data - A dictionary representing a verification form
  • partner_data - A dictionary pf form data.

Example

# In the actual workflow, profile_ids would be the list of the ids obtained from PartnerToken
# responses associated to this application.
profile_ids = ['123FB']

# Assume the same client instance as used in the PartnerToken section.
# This client connects to networkservice.lenddo.com.
client.post('CommitPartnerJob', None, {
    'partner_script_id' : 'your-partner-script-id',
    'application_id' : 'example-application',
    'profile_ids' : profile_ids})

Errors

To inspect error details, catch HTTPError, call the exception object's read() method, and decode the JSON response. In the table below, 'Error Name' refers to the string found in the response body's 'name' field.

Error Name HTTP Status Code Description
BAD_REQUEST 400 Request was malformed, or missing required data.
INTERNAL_ERROR 500 An internal error occurred. If this persists please contact a Lenddo Representative.
PARTNER_CLIENT_ALREADY_PROCESSED 400 The specified application_id has already been used.

Requesting Results from Lenddo

You may obtain score and verification results from Lenddo by making calls to the ApplicationVerification and ApplicationScore APIs hosted at scoreservice.lenddo.com.

Scores

To obtain score results, make a GET request to ApplicationScore:

from lenddo_api_client import LenddoAPIClient
client = LenddoAPIClient('your-api-client-id', 'your-api-client-secret',
	'https://scoreservice.lenddo.com')
response = client.get('ApplicationScore', 'example-application', {'partner_script_id' : 'your-partner-script-id'})
score = response['score']
flags = response['flags']

Verification Results

To obtain verification results, make a GET request to ApplicationVerification:

from lenddo_api_client import LenddoAPIClient
client = LenddoAPIClient('your-api-client-id', 'your-api-client-secret',
	'https://scoreservice.lenddo.com')
response = client.get('ApplicationVerification', 'example-application', {'partner_script_id' : 'your-partner-script-id'})
results = response['verifications']
flags = response['flags']
name_results = results['name']

Application Decision Results

To obtain application decision results, make a GET request to ApplicationVerification:

from lenddo_api_client import LenddoAPIClient
client = LenddoAPIClient('your-api-client-id', 'your-api-client-secret',
	'https://scoreservice.lenddo.com')
response = client.get('ApplicationDecision', 'example-application', {'partner_script_id' : 'your-partner-script-id'})
results = response['decision']
flags = response['flags']

FICO Score Results

To obtain FICO score results, make a GET request to FICOScore:

from lenddo_api_client import LenddoAPIClient
client = LenddoAPIClient('your-api-client-id', 'your-api-client-secret',
	'https://scoreservice.lenddo.com')
response = client.get('FICOScore', 'example-application', {'partner_script_id' : 'your-partner-script-id'})
results = response['score']
pd = response['pd']

CommitPartnerJob partner_data format for FICO scoring

The partner_data format for a FICO score application has has three required fields:

  • application dictionary, required
  • bureauData dictionar
  • applicant dictionary

applicationData.application

The application object has the following fields.

Key Type Required Description
applicationId string Yes
date string in ISO 8601 format Yes
productType string Yes one of "Mortgage", "Personal Loans", "Overdraft", "Cards", "Auto Loans", "Payday Loan", "SME", or "Others"
channel string Yes one of "Partner/Store", "Internet", "Branch", "Small business hub", "Pre-approved", "Top-up", "Mobile App" or "ATM"
product object Yes see description below

applicationData.application.product

The product object has the following fields.

Key Type Required Description
loanAmount integer Yes
paymentMethod string Yes one of "Auto debit" or "Payment through transaction account"
indicatorForPresenceOfGuarantor boolean Yes
loanPurpose string Yes one of "Loan for personal expenses", "Working capital loan", "Holiday", "Loan for making interest payment", "Education", "Renovation", "Refinancing", or "Others"

applicationData.bureauData

The "bureauData" object has two fields, "enquiries" (array of objects) and "tradelines" (array of objects).

applicationData.bureauData.enquiries

Each enquiry object in the enquiries array has the following fields.

Key Type Required Description
amount integer Yes the amount of the enquiry
applicationId string Yes
enquiryDate string in DDMMYYYY format Yes
entityType string Yes
memberCd string Yes
memberRef string Yes
processedDate string in DDMMYYYY format Yes the date when the enquiry was made
productCd string Yes the product applied for
purpose integer Yes purpose of the enquiry
controlNum integer No a system-generated enquiry control number

applicationData.bureauData.tradelines

Each tradeline object in the tradelines array has the following fields.

Key Type Required Description
accountNum string Yes
accountStatus integer Yes
accountType integer Yes
accountTypeCd integer Yes
amtOverdue number No the amount past due as of the date in the Date Reported field
applicationId string Yes
cashLimit number No
closeDate string in DDMMYYYY format Yes the date the account was closed
collateralType integer Yes
collateralValue number No
controlNum integer No a system-generated enquiry control number
creditLimit number No for Credit Card (Account Type 10) and Fleet Card (Account Type 16, this field contains the highest amount of credit used in the history of the account
currBalance number No the entire amount of credit or loan outstanding, including the current and overdue portion, if any, together with interest last applied, as of the date in the Date Reported field
delqHisMMYY string Yes cycle delinquency (up to 48 months)
installmentAmt number No
interestRate number No
lastPayment number No
lastPaymentDate string in DDMMYYYY format Yes
memberCd string Yes
memberRef string Yes
openDate string in DDMMYYYY format Yes date of first disbursement of the account
ownershipInd integer Yes 1 = Individual, 2 = Authorised User (refers to supplementary credit card holder), 3 = Guarantor, 4 = Joint
payHistEndDate string in DDMMYYYY Yes the date of the end of the payment history
payHistStartDate string in DDMMYYYY format Yes date of the beginning of the payment history
processedDate string in DDMMYYYY format Yes
productCd string Yes
repaymentTenure integer Yes term of loan
reportedDate integer No Yes
sanctionAmt number No for all other accounts, this field contains the amount of loan sanctioned
settlementAmt number No settlement amount
suitFiledCurrStatus string Yes suit filed status of the current month
suitFiledStatusMMYY string Yes suit filed status history (up to 48 months))
termFrequency string Yes payment frequency
writeoffPrincipalAmt number No
writeoffSettledStatus string Yes
writeoffTotalAmt number No write-off amount - total

applicant

The applicant object has the following fields.

Key Type Required Description
dob string in ISO 8601 format Yes
mobileTelephoneProvided boolean No
gender string No one of "MALE" or "FEMALE"
maritalStatus string No one of "Married", "Living together", "Single", "Widowed", "Divorced/Separated", or "Other"
residenceRegion string No
emailProvided boolean No
employmentStatus string No one of "Salaried", "Own Business/ Self-Employed", "Retired/Pensioner", "Part-time Employed / Freelance / Contractual Employee / Laborer", "Housewife/Other", or "Student"
yearsInCurrentJob integer No
homeTelephoneProvided boolean No
educationLevel string No one of "Undergraduate", "Graduate", "Post Graduate", or "Diploma and Specialized Courses"
yearsOfService integer No
residentialStatus string No one of "Self Owned", "Company Provided", "With Parents", "Rented", "Paying Guest", "Leased", or "Other"
workTelephoneProvided boolean No
dependents integer No
numOfMonthsInCurrentAddress integer No
monthlyIncome integer No
existingCustomer object No object with fields "timeWithTheBank" (integer) and "noOfAccountsWithTheBank" (integer)

Errors

To inspect error details, catch HTTPError, call the exception object's read() method, and decode the JSON response. In the table below, 'Error Name' refers to the string found in the response body's 'name' field.

Error Name HTTP Status Code Description
BAD_REQUEST 400 Request was malformed, or missing required data.
INTERNAL_ERROR 500 An internal error occurred. If this persists please contact a Lenddo Representative.
NOT_FOUND 404 The requested application_id was not found.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

lenddo-1.2.0.tar.gz (17.1 kB view details)

Uploaded Source

Built Distribution

lenddo-1.2.0-py3-none-any.whl (11.9 kB view details)

Uploaded Python 3

File details

Details for the file lenddo-1.2.0.tar.gz.

File metadata

  • Download URL: lenddo-1.2.0.tar.gz
  • Upload date:
  • Size: 17.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/1.12.1 pkginfo/1.4.2 requests/2.20.0 setuptools/40.5.0 requests-toolbelt/0.8.0 tqdm/4.28.1 CPython/3.6.5

File hashes

Hashes for lenddo-1.2.0.tar.gz
Algorithm Hash digest
SHA256 fc8bc63505abae1c4cf5968ba978652a6bb2926f5d6155e2326f8d10ab88866a
MD5 ce72fefd5aedc49e5cf91455ecd3e110
BLAKE2b-256 72ca80f651bb6c9470c019ce1b9fe8af6edc49dd9a5d28a87140e15ed3444dc4

See more details on using hashes here.

File details

Details for the file lenddo-1.2.0-py3-none-any.whl.

File metadata

  • Download URL: lenddo-1.2.0-py3-none-any.whl
  • Upload date:
  • Size: 11.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/1.12.1 pkginfo/1.4.2 requests/2.20.0 setuptools/40.5.0 requests-toolbelt/0.8.0 tqdm/4.28.1 CPython/3.6.5

File hashes

Hashes for lenddo-1.2.0-py3-none-any.whl
Algorithm Hash digest
SHA256 a39a2b39f60eee8ca65c25042ce8984ae10e4105df177e956d2fabe7fa0fee13
MD5 fc236c66c4cd47037f665e7a2689674b
BLAKE2b-256 1cca990eb22c37965050967d96c519710626610142e58063eeb22ee8b37d37ac

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page