helper/schema: write "attr.#": "0" for empty maps

[phinze]

This fixes some perpetual diffs I saw in Atlas AccTests where an empty
map (map[string]interface{}{}) was being d.Set for "metadata_full".

Because the MapFieldWriter was not distinguishing between empty and nil,
this trigger the "map delete" logic and no count was written to the
state. This caused subsequent plans to improperly report a diff.

Here we redefine the map delete functionality to explicitly trigger only
on nil, so we catch the .# field for empty maps.

[phinze]

@mitchellh I know we've had a lot of churn in these types of situations, but I believe this change is still correct and good. At least, it makes sense to me. 😀

[mitchellh]

Let's see what the other acceptance tests say... I feel like we did this before and it had issues with computed diffs but we'll see. :)

[phinze]

So far I've validated:

• Fixes heroku test failures with "all_config_vars.#": "" => "<computed>"
• Fixes atlas test failures with "full_metadata.#": "" => "<computed>"
• Does not break any google acceptance tests
• Running all _basic$ AWS acctest now, good so far

[phinze]

Okay AWS is looking good too - @mitchellh how you feelin?

[mitchellh]

Alright let's do it

[ghost]

I'm going to lock this issue because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues.

If you have found a problem that seems similar to this, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.