Avalara Tax Python SDK.
This GitHub repository is the Python SDK for Avalara's world-class tax service, AvaTax. It uses the AvaTax REST v2 API, which is a fully REST implementation and provides a single client for all AvaTax functionality. For more information about AvaTax REST v2, please visit Avalara's Developer Network or view the online Swagger documentation.
Install simply with pip.
pip install Avalara
OR
$ git clone https://github.com/avadev/AvaTax-REST-V2-Python-SDK.git
AvaTax-REST-V2-Python-SDK
directory.$ cd AvaTax-REST-V2-Python-SDK
AvaTax-REST-V2-Python-SDK $ python3 -m venv ENV
AvaTax-REST-V2-Python-SDK $ source ENV/bin/activate
(ENV) AvaTax-REST-V2-Python-SDK $ pip install -e .
(ENV) AvaTax-REST-V2-Python-SDK $ pip install -e .[testing]
Import the AvataxClient from the avalara module
First thing to do is to import the AvataxClient constructor module to your name space, or your python script.
from avalara import AvataxClient
Now we are ready to construct a client object
Create a new AvaTaxClient object:
client = AvataxClient('my test app',
'ver 0.0',
'my test machine',
'sandbox')
The client constructor takes four string parameters, in squence they are app_name(required)
, app_version(required)
, machine_name(optional)
, and environment(required)
.
The app_name, app_version, machine_name will be used to construct the Client Header associated with each call made by this client. It will be returned within the response object to help you keep track of the API calls.
The environment
variable can be either "sandbox"
or production
, they correspond to the two different environments for AvaTax service.
If you are a regular or free trial customer please use "production"
. If you don't have an account, you can sign up for a free trail account on our developer site, this will be a production account as well.
If you wish to obtain a Sandbox account, please contact your Customer Account Manager
Ping the service
Now we have a client object, we can ping the AvaTax REST V2 server to ensure connectivity.
response = client.ping()
# to view respnse text
print(response.text())
# to view json version of the response
print(response.json())
# to view the status code
print(response.status_code())
# to view the raw response
print(response.raw())
Note that the response from all REST calls made using this SDK will be Request object, which contains status code, response text, raw josn, and more information on the respnse.
Add credentials to your client object
Unlike ping
, most methods in our REST V2 API requires you to be authenticated in order to associate those information provided by you with your account.
To find out if a method requires authentication, visit our API Reference page.
To add credential on the current client object:
client = client.add_credentials('USERNAME/ACCOUNT_ID', 'PASSWORD/LICENSE_KEY')
The add_credential
method will hash your username/password, or account_id/license_key pair and attach to every call made by your client object, meaning you only have to add credential once to each client you prepare to use.
To verify that you have added a valid credential, simply call the ping
method again, this time in the response text you should see "authenticated": true
.
To create a transaction using your client object
Now our client object is authenticated, we can call the create_transaction method which calls the CreateTransaction API
transaction_response = client.create_transaction(tax_document)
print(transaction_response.text())
tax_document = {
'addresses': {'SingleLocation': {'city': 'Irvine',
'country': 'US',
'line1': '123 Main Street',
'postalCode': '92615',
'region': 'CA'}},
'commit': False,
'companyCode': 'DEFAULT',
'currencyCode': 'USD',
'customerCode': 'ABC',
'date': '2017-04-12',
'description': 'Yarn',
'lines': [{'amount': 100,
'description': 'Yarn',
'itemCode': 'Y0001',
'number': '1',
'quantity': 1,
'taxCode': 'PS081282'}],
'purchaseOrderNo': '2017-04-12-001',
'type': 'SalesInvoice'}
The create_transaction method takes in a model, in python it's a dictionary type object. Which you will fill out to include all of your transaction information. In this case, we are using the TransactionModel For information on other models use by AvaTax APIs, visit our information page here
Like our SDKs in other languages, the Python SDK includes all methods offered by the AvaTax REST V2 API. To find a mehtod corresponding to a specific API endpoint, simply visit this code page To learn more about integrating our REST API into your system, visit our developer guide that contains information on using the powerful features offered by our API.
We realize that having to format the TransactionModel can be complicated and time consuming, thus we created a tool called Transaction Builder to help you put together a transaction model, and create it!
First import the transaction builder constructor into your name space:
from avalara.transaction_builder import TransactionBuilder
Then, let's create a transaction builder object:
tb = TransactionBuilder(client, "DEFAULT", "SalesInvoice", "ABC")
The builder takes four required parameters, in sequence they are
Now you are free to add transaction details to this object, by using methods like with_address
, with_line
, with_parameter
.
For a fulll list of transaction builder methods available and the parameters they take in, visit the code page
In the end, you may call the create
method on your builder object, which will call the CreateTransaction API with the transaction model you have build so far, and return back the response.
If you wish to run the integration and unit testings, you must store a pair of credentials in the current enviroment.
Add the following to the activate
file in your environment, and restart bash.
OR simply export
them directly:
export SANDBOX_USERNAME='your_sandbox_username'
export SANDBOX_PASSWORD='your_sandbox_password'
# OR
SANDBOX_ACCOUNTID='your_sandbox_account_id'
SANDBOX_LICENSEKEY='your_sandbox_license_key'
Note: Only Sandbox credentials should be used for testing, as the test case will commit/adjust/void dummy transactions on the account to verify functionalities.
Logging is implemented using standard Python logging framework.
AvataxClient
class are decorated with ava_log
decorator.(This is achieved using another decorator at class level, decorate_all_methods
)ava_log
decorator collects relevant request data, response data useful for instrumentation and logs error data in case of exception.AvataxClient
constructor is modified with optional parameter, is_log_req_resp_allowed
(defaulted to False), to control if log entry should contain request and response objects.AvataxClient
to use already configured logger instance. e.g.from logging import config
# configure logging using file config or dictConfig
# or by setting basicConfig (this is default in case no logger is set)
config.fileConfig("logger_config.conf")
logger = logging.getLogger()
# set logger property
client.logger = logger
User feedbacks are highly valued here at Avalara, we want to ensure the best experience for every customer using our services. If you have any concern with this SDK or AvaTax in general, please post your queston/suggestion on our Developer Relation Forum as we will reply to you in a timely manner. If you wish to contribute to this SDK, please fork the repository and submit your pull request from the forked version. We are happy to review your contribution, and merge them if all checks has passed!