AF-3801 fix reporter runtime error in Dart 2.1.0

[robbecker-wf]

Overview

Under Dart 2.1.0, when running tests that use reporter.log. The quiet variable would be null somehow and cause a runtime error in the conditional check in the _log method.
I imagine this is caused by something like checked mode when running tests. I could not reproduce this runtime error when running a similar thing on the command line (prod mode).

An example failing build due to this error: https://ci.webfilings.com/build/1585120

Changes

• Don't allow these to variables to be null

Testing

• CI passes
• See passing build using this branch: https://github.com/Workiva/over_react_format/pull/25

[aviary3-wk]

Security Insights

No security relevant content was detected by automated scans.

Action Items

• Review PR for security impact; comment "security review required" if needed or unsure
• Verify aviary.yaml coverage of security relevant code

---

Questions or Comments? Reach out on Slack: #support-infosec.

[greglittlefield-wf]

Could it be that Reporter isn't getting passed both arguments during intiialization?

dart_dev/lib/src/reporter.dart

Line 22 in 7eee6da

Reporter reporter = new Reporter();

Based on the existing code,

dart_dev/lib/src/reporter.dart

Lines 30 to 35 in a3b36d0

	bool color = true;
	bool quiet = false;
	Reporter({bool this.color, bool this.quiet});
	String colorBlue(String message) => _color(_blue, message);

, I would expect new Reporter() to have both fields be null.

I feel like it should just be:

bool color;
bool quiet;

Reporter({bool this.color = true, bool this.quiet = false}); 

Could it be that it's only showing up in testing because this is being run in checked mode?

[greglittlefield-wf]

#nit these defaults aren't necessary if we have null-awares below

[robbecker-wf]

I left them in so that when calling the constructor, the IDE would tell you what the defaults are.

[greglittlefield-wf]

#nit these defaults aren't necessary

[robbecker-wf]

Totally. I'll remove them.

[greglittlefield-wf]

I don't think we should have to worry about this case; these should never be null unless someone explicitly sets them to null, in which case the program should probably fail.

[greglittlefield-wf]

+1

[evanweible-wf]

QA +1

• CI passes

@Workiva/release-management-p