Code efficiency analysis

[aaronsteers]

As discussed, after we got everything working, I wanted to perform this quick analysis of streams.py and tap.py for any "extra" or unhelpful code snippets which could possibly be moved into the SDK itself. Excepting the final change proposed here in #12, I didn't see anything at all, really, which I would remove.

Here's a quick overview of line counts:

• tap.py (total 47 lines including whitespace):
  • Docstring and imports: 21 lines
  • CLI boilerplate: 3 lines (including code comment)
  • Tap class definition: 15 lines
  • Constants declarations: 5 lines
• streams.py (total 254 lines including whitespace):
  • Docstring and imports: 25 lines
  • Authentication: 20 lines (1 Authenticator class definition and 1 authenticator Stream class property)
  • Custom Source API Logic: 53 lines total over 3 areas:
    • get_next_page_token() for looping and pagination logic: 29 lines
    • get_url_params() for defining the interface with the REST API: 19 lines
    • parse_response() to translate the REST response into 'records': 5 lines
  • Stream schema definition, which specifies all field names and types: 139 lines for approximately 81 fields (including some commented-out WIP)

[aaronsteers]

@jtimeus-slalom - Let me know if you see other opportunities to improve, but overall I'm very happy with these numbers! All up, we are just around 300 lines of code including the stream field definitions, or less than 175 lines of code if we exclude those JSON schema declarations for column types.

[aaronsteers]

No action here so I'll close this item to reduce clutter in the issues list. :)