Reconcile Set handling discrepancies affecting diffs

[t0yv0]

What happened?

There're some indications that Set-typed elements compute different diffs from the expected ones by TF.

This is a tracking issue as we try to correlate examples.

• rds.ParameterGroup diff not clear pulumi-aws#2442
  • Regress AWS-2442 #1255
• DynamoDB GSI minor autoscaling update in AWS is interpreted by Pulumi as GSIs requiring major changes pulumi-aws#2856 possibly
• ECS Container definition permanent diff pulumi-aws#1738 possibly

Example

See above

Output of pulumi about

N/A

Additional context

N/A

Contributing

Vote on this issue by adding a 👍 reaction.
To contribute a fix for this issue, leave a comment (and link to your pull request, if you've opened one already).

[t0yv0]

Quick update on these issues. It appears that sets are working semantically as expected, matching TF, but set diffs present confusingly, which is tracked in an old-ish ticket #186

• AWS-2442 is unrelated to set ordering it appears
• AWS-2856 appears to work as expected but presents confusingly due to 186
• AWS-1728 needs a repro
• On an experimental branch that can cross-test bridged vs TF it seems that bridged providers tolerate set reordering well and match TF in canonicalizing the order of set elements
• One new finding is that IgnoreChanges unable to ignore set element changes #1756 ignoreChanges does not work as you would expect in the situation tracked by Present clearer diff fo single element changes to sets  #186

[t0yv0]

I'll close this. I think the remaining tickets above track what needs to happen here: proper diff display and proper ignoreCHanges tracking. If we find any more issues that bottom out at this as part of AWS backlog combing I'll reopen or link here.

[pulumi-bot]

Cannot close issue:

• does not have required labels: resolution/

Please fix these problems and try again.