search

fhdsl / metricminer

R package that digs up data that matters from APIs, making it dashboard-ready
https://hutchdatascience.org/metricminer/
MIT License
2 stars 0 forks source link

Creating dummy data and better classifying the data types returned by metricminer #52

Closed cansavvy closed 9 months ago

cansavvy commented 9 months ago

Purpose/implementation Section

What changes are being implemented in this Pull Request?

This PR's goal is to have example data. This is for these reasons:

  1. Better for unit testing to have standard data to use
  2. Better to have that data not be something we care about
  3. Better to have this data be available to users to they know what data structures to expect and we can have them use it in tutorials
  4. This also allows me to better classify our data structures so the write functions can handle input data more gracefully

What was your approach?

I'm making a example-data-setup.R script that can be called when we wish to remake these example data.

I'll aslo be making a family of functions that allow folks to pull these example data in and play with them.

I created a new gmail and associated accounts for metricminersoftware@gmail.com -- this means the data are real in a sense but this account doesn't have privileges or access to anything besides empty accounts. This will be a bit safer than using our real accounts. (Although we still will keep these credentials safe of course).

What GitHub issue does your pull request address?

Somewhat related to #20 but doesn't entirely cover the needs of #20

cansavvy commented 9 months ago

There some bug fixes in here but crux of this is this set up file: inst/extdata/example_data/example-data-setup.R

It basically makes a bunch of example datasets users (and us) can use to test things and go through tutorials.

cansavvy commented 9 months ago

@howardbaek There's a lot here. This PR got bigger than I intended to be, but it is all of a similar theme. Maybe we can discuss over our meeting today?

cansavvy commented 9 months ago

This passes checks. Going to merge so we can start to move on. We can always reiterate and fix later. This isn't published on CRAN yet