test(openfeature): add null targeting key test coverage

[leoromanovsky]

What does this PR do?

Adds test coverage for OpenFeature spec OF.2 (optional targeting key) to prove libdatadog handles nil targeting key correctly via the Ruby FFI layer.

Motivation:

The OpenFeature spec (Requirement 3.1.1, OF.2) defines targeting key as optional. The correct behavior (matching libdatadog's reference implementation) is:

	Static (no rules, no shards)	Split (shards)	Targeted (rules)
"" empty string	Returns value	Hashes "", returns value	Evaluates rules with id="", returns value
null/missing	Returns value	TARGETING_KEY_MISSING error, returns default	Evaluates rules (no id fallback), returns value or falls through

Ruby's evaluator delegates to libdatadog which already handles this correctly — the C extension passes NULL when targeting_key is absent from the context hash. This PR adds explicit test coverage proving the behavior.

Change log entry

No customer-visible change. Test-only — adds coverage for optional targeting key behavior that already works correctly.

Additional Notes:

• Tests placed in spec/datadog/open_feature/ so @DataDog/feature-flagging-and-experimentation-sdk is auto-requested via CODEOWNERS
• Tests call Configuration#get_assignment directly to exercise the Rust FFI layer without mocks
• Used partial shard range (0-4999 of 10000) instead of full range to avoid libdatadog's insignificant shard optimization
• No logic changes to Ruby code

Question for reviewers: The existing evaluation_engine_spec.rb mocks NativeEvaluator, so there's no integration spec testing the full OpenFeature provider → native evaluator → libdatadog path with real flag configs. Should we add one?

How to test the change?

bundle exec rspec spec/datadog/open_feature/null_targeting_key_spec.rb

3 scenarios:

1. Static flag + nil targeting key → returns value with STATIC reason
2. Sharded flag + nil targeting key → returns TARGETING_KEY_MISSING error
3. Rule-match flag + nil targeting key → returns value with TARGETING_MATCH reason

Next Steps

• Add test-case-null-targeting-key.json to ffe-system-test-data shared fixtures (separate fixture centralization project)
• Wire the fixture loop spec to cover null targeting key via JSON fixtures
• System test coverage for null targeting key across all tracers

[github-actions]

Thank you for updating Change log entry section 👏

^{Visited at: 2026-03-27 21:29:50 UTC}

[datadog-datadog-prod-us1]

⚠️ Tests

✨ Fix all issues with BitsAI or with Cursor

⚠️ Other Violations

❄️ 3 New flaky tests detected

    OpenFeature OF.2: Optional Targeting Key with nil targeting key evaluates rule-match flag when rule matches on non-id attribute from rspec     (Fix with Cursor)

    OpenFeature OF.2: Optional Targeting Key with nil targeting key evaluates static flag successfully from rspec     (Fix with Cursor)

    OpenFeature OF.2: Optional Targeting Key with nil targeting key returns TARGETING_KEY_MISSING for sharded flag from rspec     (Fix with Cursor)

ℹ️ Info

No other issues found (see more)

🧪 All tests passed

🎯 Code Coverage (details)
• Patch Coverage: 100.00%
• Overall Coverage: 95.43% (+0.08%)

_{This comment will be updated automatically if new data arrives.
🔗 Commit SHA: b4fc266 | Docs | Datadog PR Page | Was this helpful? React with 👍/👎 or give us feedback!}

[pr-commenter]

Benchmarks

Benchmark execution time: 2026-04-13 17:24:38

Comparing candidate commit b4fc266 in PR branch leo.romanovsky/fix-optional-targeting-key with baseline commit 5f68ddc in branch master.

Found 0 performance improvements and 0 performance regressions! Performance is the same for 45 metrics, 1 unstable metrics.

Explanation

This is an A/B test comparing a candidate commit's performance against that of a baseline commit. Performance changes are noted in the tables below as:

• 🟩 = significantly better candidate vs. baseline
• 🟥 = significantly worse candidate vs. baseline

We compute a confidence interval (CI) over the relative difference of means between metrics from the candidate and baseline commits, considering the baseline as the reference.

If the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD), the change is considered significant.

Feel free to reach out to #apm-benchmarking-platform on Slack if you have any questions.

More details about the CI and significant changes

You can imagine this CI as a range of values that is likely to contain the true difference of means between the candidate and baseline commits.

CIs of the difference of means are often centered around 0%, because often changes are not that big:

---------------------------------(------|---^--------)-------------------------------->
                              -0.6%    0%  0.3%     +1.2%
                                 |          |        |
         lower bound of the CI --'          |        |
sample mean (center of the CI) -------------'        |
         upper bound of the CI ----------------------'

As described above, a change is considered significant if the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD).

For instance, for an execution time metric, this confidence interval indicates a significantly worse performance:

----------------------------------------|---------|---(---------^---------)---------->
                                       0%        1%  1.3%      2.2%      3.1%
                                                  |   |         |         |
       significant impact threshold --------------'   |         |         |
                      lower bound of CI --------------'         |         |
       sample mean (center of the CI) --------------------------'         |
                      upper bound of CI ----------------------------------'

[p-datadog]

One question below.

[p-datadog]

question (non-blocking): These tests call the native C extension directly (Configuration.new → get_assignment), but spec:open_feature doesn't have a .enhance([:compile]) dependency in the Rakefile — unlike spec:core_with_libdatadog_api (Rakefile:260) which does.

Right now the extension is available because the batch seed happens to place a compile-triggering task (like core_with_libdatadog_api) in the same CI job, and it runs first. A different seed could put open_feature in a batch without one, and the skip guard here would silently skip all three tests.

Would it make sense to add something like this to the Rakefile?

Rake::Task['spec:open_feature'].enhance(["compile:libdatadog_api.#{RUBY_VERSION[/\d+.\d+/]}_#{RUBY_PLATFORM}"])

[p-datadog]

@codex review

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: b4fc266283

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Cover explicit nil targeting_key in OF.2 tests

These examples are labeled as “nil targeting key” coverage, but they pass contexts without targeting_key ({} / {'email' => ...}), which only exercises the missing key path. This can give false confidence for OF.2 null handling: if the Ruby→FFI context conversion ever diverges between missing and explicit nil, this spec would still pass while the null case regresses. Add at least one call with {'targeting_key' => nil, ...} to assert the advertised behavior.

Useful? React with 👍 / 👎.