fix: skip some tests when run against the emulator

[skuruppu]

Thank you for opening a Pull Request! Before submitting your PR, there are a few things you can do to make sure it goes smoothly:

• Make sure to open an issue as a bug/issue before writing your code! That way we can discuss the change, evaluate designs, and agree on the general idea
• Ensure the tests and linter pass
• Code coverage does not decrease (if any source code was changed)
• Appropriate docs were updated (if necessary)

Fixes #932 🦕

[codecov]

Codecov Report

Merging #933 into master will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master     #933   +/-   ##
=======================================
  Coverage   98.20%   98.20%           
=======================================
  Files          21       21           
  Lines       19959    19959           
  Branches     1069     1069           
=======================================
  Hits        19601    19601           
  Misses        354      354           
  Partials        4        4           

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 5f2d52d...b07cc61. Read the comment docs.

[hengfengli]

Hi @AVaksman, @bcoe, do you have any idea why ci/windows fails? I can see it also fails in another PR for the same reason: #931

[hengfengli]

@skuruppu I have skipped all needed tests and fixed related errors. I am not sure:

1. why "ci / windows" fails. It happens in other PRs as well.
2. Who should I assign this as a reviewer? I guess you're the PR creator and I can't assign it to you.

[stephenplusplus]

Would doing this here work:?

if (emulatorEnabled) {
  this.skip();
}

[hengfengli]

This doesn't work. I tried this at the beginning.

[hengfengli]

I got the following error:

     Error: `this.skip` forbidden

[skuruppu]

I did some research. The documentation says to just do a return instead of a skip. I updated the code.

[hengfengli]

Good catch. Thanks for making this more elegant.

[hengfengli]

@skuruppu I think we need to clean up the leftover instances in the GCP project. It fails to add a new instance.

[skuruppu]

I agree, it's weird though because the tests actually try to clean up the instances before it creates a new one. I'm not sure why that's not working. I suppose it's because we try to create an instance before deleting the old ones?

[hengfengli]

I guess the instance ID is generated by date:

const INSTANCE_ID = process.env.SPANNERTEST_INSTANCE || `test-instance-${date}`;

The old one from yesterday is not deleted.

[skuruppu]

I created #955, let's see if it runs.

[hengfengli]

LGTM.