Search for parts in the Digi-Key catalog by keyword using KeywordSearch. Then make a PartDetails call to retrieve all real time information about the part including pricing. PartDetails works best with Digi-Key part numbers as some manufacturers overlap other manufacturer part numbers.
digikey-api is an Digkey Part Search API client for Python 3.6+. API response data is returned as Python objects that attempt to make it easy to get the data you want. Not all endpoints have been implemented.
pip install digikey-api
mkdir -p path/to/cache/dir
export DIGIKEY_CLIENT_ID="client_id"
export DIGIKEY_CLIENT_SECRET="client_secret"
export DIGIKEY_STORAGE_PATH="path/to/cache/dir"The cache dir is used to store the OAUTH access and refresh token, if you delete it you will need to login again.
Register an app on the Digikey API portal: Digi-Key API V3. You will need the client ID and the client secret to use the API. You will also need a Digi-Key account to authenticate, using the Oauth2 process.
When registering an app the OAuth Callback needs to be set to https://localhost:8139/digikey_callback.
Python will automatically spawn a browser to allow you to authenticate using the Oauth2 process. After obtaining a token the library will cache the access token and use the refresh token to automatically refresh your credentials.
You can test your application using the sandbox API, the data returned from a Sandbox API may not be complete, but the structure of the Sandbox API response will be a representation of what to expect in Production.
For valid responses make sure you ue the client ID and secret for a Production App. Otherwise, it is possible that dummy data is returned and you will pull your hair as to why it doesn't work.
import os
from pathlib import Path
import digikey
from digikey.v3.productinformation import KeywordSearchRequest
from digikey.v3.batchproductdetails import BatchProductDetailsRequest
CACHE_DIR = Path('path/to/cache/dir')
os.environ['DIGIKEY_CLIENT_ID'] = 'client_id'
os.environ['DIGIKEY_CLIENT_SECRET'] = 'client_secret'
os.environ['DIGIKEY_CLIENT_SANDBOX'] = 'False'
os.environ['DIGIKEY_STORAGE_PATH'] = CACHE_DIR
# Query product number
dkpn = '296-6501-1-ND'
part = digikey.product_details(dkpn)
# Search for parts
search_request = KeywordSearchRequest(keywords='CRCW080510K0FKEA', record_count=10)
result = digikey.keyword_search(body=search_request)
# Only if BatchProductDetails endpoint is explicitly enabled
# Search for Batch of Parts/Product
mpn_list = ["0ZCK0050FF2E", "LR1F1K0"] #Length upto 50
batch_request = BatchProductDetailsRequest(products=mpn_list)
part_results = digikey.batch_product_details(body=batch_request)Logging is not forced upon the user but can be enabled according to convention:
import logging
logger = logging.getLogger(__name__)
logger.setLevel(logging.DEBUG)
digikey_logger = logging.getLogger('digikey')
digikey_logger.setLevel(logging.DEBUG)
handler = logging.StreamHandler()
handler.setLevel(logging.DEBUG)
logger.addHandler(handler)
digikey_logger.addHandler(handler)All functions from the PartSearch API have been implemented.
digikey.keyword_search()digikey.product_details()digikey.digi_reel_pricing()digikey.suggested_parts()digikey.manufacturer_product_details()
The one function from the BatchProductDetailsAPI API has been implemented.
digikey.batch_product_details()
All functions from the OrderDetails API have been implemented.
digikey.salesorder_history()digikey.status_salesorder_id()
TODO
The API has a limited amount of requests you can make per time interval Digikey Rate Limits.
It is possible to retrieve the number of max requests and current requests by passing an optional api_limits kwarg to an API function:
api_limit = {}
search_request = KeywordSearchRequest(keywords='CRCW080510K0FKEA', record_count=10)
result = digikey.keyword_search(body=search_request, api_limits=api_limit)The dict will be filled with the information returned from the API:
{
'api_requests_limit': 1000,
'api_requests_remaining': 139
}Sometimes the API does not return any rate limit data, the values will then be set to None.