[ENH] fix bug pivot_longer

[samukweku]

PR Description

Please describe the changes proposed in the pull request:

• fix bug where sub structures are series, not dataframes, preventing melt from occuring.

This PR resolves #827.

PR Checklist

Please ensure that you have done the following:

1. PR in from a fork off your branch. Do not PR from <your_username>:dev, but rather from <your_username>:<feature-branch_name>.

2. If you're not on the contributors list, add yourself to AUTHORS.rst.

3. Add a line to CHANGELOG.md under the latest version header (i.e. the one that is "on deck") describing the contribution.
   • Do use some discretion here; if there are multiple PRs that are related, keep them in a single line.

Automatic checks

There will be automatic checks run on the PR. These include:

• Building a preview of the docs on Netlify
• Automatically linting the code
• Making sure the code is documented
• Making sure that all tests are passed
• Making sure that code coverage doesn't go down.

Relevant Reviewers

Please tag maintainers to review.

• @ericmjl

[samukweku]

the test error is from janitors/finance.py. It seems the url is not working anymore, and one may have to register on the site to get api keys. Not sure what the alternative is;

[ericmjl]

@samukweku I triggered a test re-run. Let's see what happens.

[ericmjl]

@samukweku looks like the tests are still failing.

Good digging before on this - if indeed we need an API key to do currency conversions, then it's going to be challenging for users and for us. We might need to either:

1. Find a different service, or
2. Deprecate the finance APIs.

For now, I don't want this to be a blocker for you. Could you mark the tests as xfail and let's get the bug fix in first? Also, can you add in a comment that references my response, so we have context on why it's marked xfail?

[samukweku]

@ericmjl, there is a notebook for currency conversion that also fails, since it is based on the API?

[ericmjl]

Looks like that one will also have to be taken offline. Can you disable them in the sphinx configuration file? There should be a conf.py. I found the docs; the info in there should give you a headstart!

[codecov-io]

Codecov Report

Merging #828 (a5bef34) into dev (d743e3a) will decrease coverage by 0.30%.
The diff coverage is 98.36%.

@@            Coverage Diff             @@
##              dev     #828      +/-   ##
==========================================
- Coverage   95.62%   95.32%   -0.31%     
==========================================
  Files          19       19              
  Lines        1989     1966      -23     
==========================================
- Hits         1902     1874      -28     
- Misses         87       92       +5     

[samukweku]

@ericmjl I think I'm missing something regarding the conf.py/individual notebook execution stop; if you have an idea i'll gladly take that and resolve this. explicitly disable jupyter notebook execution

[ericmjl]

@samukweku I just pushed a small change to conf.py. Let's see if that fixes the problem.

[ericmjl]

@samukweku I went back and forth on what would be the best way to resolve this issue. The best way is to look for a new exchange rates API, while disabling the function so that we can retain the state of "all tests passing". We can invite users to help us find a new API that we can hit.

[ericmjl]

@anzelpwj and @zjpoh, sorry to bother you two here, but I'm struggling to figure out why the spark session thingy is failing. Would either of you be kind enough to help out here?

[jk3587]

nit: I suggest we change the error to say we "call" the api instead of "hit" (or anything other than "hit").

[ericmjl]

@jk3587 good call on that one (*cough cough*) 😃. I will change that.

[ericmjl]

Oh my, we're finally done here!

[ericmjl]

@samukweku going to merge now.