High Level Python and QuickBase Interfacing

The quickbase-client Python package focuses on “code that writes code” to allow users to interact with the QuickBase API from a high level to provide easier development interfacing with QuickBase.

Comment

Inspiration

  • When originally working with the QuickBase API through python, we found the most difficult part being always referring to fields by some numerical ID in the code. I found myself writing many many variables just to alias these IDs, to make the code more readable.
  • Similarly, it was sometimes difficult to build queries just using a string in the QuickBase query language
  • QuickBase recently rolled out the new RESTful JSON API, we wanted to specifically target these new endpoints
  • More readable code is easier to maintain, and code that is easier to maintain makes our systems more responsive and robust

What it does

  • This open source Python package serves as a utility for developers to interface with QuickBase using a higher-level Python wrapper.
  • It includes a utility which goes through the QuickBase API (specifically the get tables for app, and the get fields for table endpoints), to generate a Python class that can then be used to interact with records in QuickBase tables as if they are python objects. It generates classes that can be used like below:
    client = GitHubIssue.client(user_tok=os.environ['QB_USER_TOKEN'])
    new_issue = GitHubIssue(
        title='Something broke',   # you get friendly-kwargs for fields without worrying about ID's
        description='Please fix!',
        date_opened=date.today()   # things like Python date objects will be serialized
    )
    response = client.add_record(new_issue)
    print(response.json())  # all methods (except for query) return the requests Response object
  • It also includes utilities for helping build query strings to run QuickBase queries from Python code.
    from quickbase_client.query import on_or_before_, eq_, and_

    schema = GitHubIssue.schema
    q = and_(
        eq_(schema.date_opened, schema.date_created),
        on_or_before_(schema.date_closed, date(2020, 11, 16))
    )
    print(q.where)  # ({'9'.EX.'_FID_1'}AND{'10'.OBF.'11-16-2020'})
    recs = client.query(q)  # recs will be GitHubIssue objects unless passing raw=True
    print([str(r) for r in recs])  # ['<GitHubIssue title="Made And Closed Today" id="10000">']

How I built it

  • Using python and the QuickBase API, with a test QuickBase app here

What's next for High Level Python and QuickBase Interfacing

  • We decided to make this project open source to build the QuickBase developer community and collectively enhance the package
  • We will continue to enhance for our own use, and add more features
    • Some features include handling composite fields like Addresses, other QueryBuilding classes, and more.

Built With

Share this project:

Updates