feat(ffe): expose observeFullEvaluationData config-level FFI getter by vjfridge · Pull Request #2373 · DataDog/libdatadog

vjfridge · 2026-08-19T13:58:30Z

Summary

Adds the top-level UFC observeFullEvaluationData boolean end-to-end through libdatadog and exposes a config-level FFI getter so SDKs can read it without re-parsing the raw JSON string. This is the libdatadog half of the "Protecting PII in flagevaluations track" initiative; the server-side UFC field ships via ddoghq/dd-source#22826 and the RC schema via ddoghq/dd-go#3627.

When true: SDK emits unhashed targeting_key + full evaluation context (today's behavior).
When false (default): SDK emits a SHA-256 hex hash of the targeting key and omits the context.

Default is false (privacy-preserving). A fresh deploy is a no-op for customers until an SDK ships the read-side change and someone toggles the field on for an env.

Changes

UniversalFlagConfigWire (libdd-ffe): new #[serde(default)] bool observe_full_evaluation_data field, a sibling of environment at the UFC top level. #[serde(default)] keeps older configurations (without the key) parsing to false.
CompiledFlagsConfig: carries the field through the wire -> compiled transform.
Configuration::observe_full_evaluation_data(): Rust getter, mirroring the existing created_at() / environment() accessors.
ddog_ffe_configuration_get_observe_full_evaluation_data (libdd-ffe-ffi): new #[no_mangle] C accessor returning bool. The datadog/ffe.h header regenerates via cbindgen.
examples/ffi/ffe.c: prints the flag after ddog_ffe_configuration_new to demonstrate usage.

Test plan

cargo check -p libdd-ffe -p libdd-ffe-ffi
cargo +nightly-2026-07-26 fmt --all -- --check
cargo +nightly-2026-07-26 clippy -p libdd-ffe -p libdd-ffe-ffi --all-targets -- -D warnings
cargo nextest run -p libdd-ffe -p libdd-ffe-ffi (incl. 4 new unit tests: field defaults to false when absent, parses true when present, exposed via Configuration true/false)
cargo test -p libdd-ffe --doc
cargo ffi-test — ffe C example compiles, links against the regenerated header (new symbol present), and runs green against the ffe-system-test-data fixture.

Refs: FFL-3038

vjfridge · 2026-08-19T13:58:39Z

Note to self: Follow-ups to propagate to SDK repos

This PR adds the ddog_ffe_configuration_get_observe_full_evaluation_data symbol to libdatadog, but it only takes effect once SDKs wrap it so they dont need to re-parsing the raw JSON. Required follow-ups, in merge order:

libdatadog release Nothing downstream can consume the getter until a release goes out for this PR.
datadog/libdatadog-rb bump the vendored libdatadog to the release from step 1, then add observe_full_evaluation_data to the Core::FeatureFlags::Configuration C extension wrapping ddog_ffe_configuration_get_observe_full_evaluation_data
dd-trace-rb consume the new libdatadog-rb release and read observe_full_evaluation_data from the C extension instead of parsing the config string.

Other language SDKs to check

The same getter should be exposed by the other libdatadog FFI consumers once they pick up the release:

Python (dd-trace-py / libdatadog Python bindings) — wrap the new symbol if it ships FFE bindings.
Any other SDK whose C extension links libdd-ffe-ffi and currently parses the UFC JSON for this field.

datadog-datadog-prod-us1-2 · 2026-08-19T14:10:50Z

🎉 All green!

🧪 All tests passed
❄️ No new flaky tests detected

🎯 Code Coverage (details)
• Patch Coverage: 99.08%
• Overall Coverage: 76.62% (+0.20%)

_{This comment will be updated automatically if new data arrives.

🔗 Commit SHA: f1baf07 | Docs | View more details | Give us feedback!}

pr-commenter · 2026-08-19T14:12:49Z

Benchmarks

Comparison

Benchmark execution time: 2026-08-20 15:11:49

Comparing candidate commit f1baf07 in PR branch vickie/ffl-3038-add-observe_full_evaluation_data-accessor with baseline commit 24022a9 in branch main.

Found 0 performance improvements and 3 performance regressions! Performance is the same for 11 metrics, 0 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 ----------------------------------'

scenario:flagevaluation_evp/payloads/scale/2500flags_500users_20fields

🟥 execution_time [+846.814µs; +866.901µs] or [+4.026%; +4.121%]

scenario:flagevaluation_evp/payloads/stress/10flags_1000users_250fields

🟥 execution_time [+6.991ms; +7.117ms] or [+7.299%; +7.431%]
🟥 throughput [-722.479op/s; -709.797op/s] or [-6.920%; -6.799%]

Candidate

Candidate benchmark details

Group 1

cpu_model	git_commit_sha	git_commit_date	git_branch
Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz	`f1baf07`	1787237900	vickie/ffl-3038-add-observe_full_evaluation_data-accessor

scenario	metric	min	mean ± sd	median ± mad	p75	p95	p99	max	peak_to_median_ratio	skewness	kurtosis	cv	sem	runs	sample_size
flagevaluation_evp/coalescer/scale/2500flags_500users_20fields	execution_time	5.399ms	5.415ms ± 0.007ms	5.415ms ± 0.004ms	5.419ms	5.426ms	5.430ms	5.432ms	0.31%	0.001	-0.398	0.12%	0.000ms	1	200
flagevaluation_evp/coalescer/scale/2500flags_500users_20fields	throughput	460270.402op/s	461691.308op/s ± 558.215op/s	461681.988op/s ± 380.399op/s	462063.100op/s	462652.001op/s	462846.579op/s	463023.542op/s	0.29%	0.004	-0.399	0.12%	39.472op/s	1	200
flagevaluation_evp/coalescer/stress/10flags_1000users_250fields	execution_time	7.938ms	7.951ms ± 0.008ms	7.950ms ± 0.003ms	7.954ms	7.960ms	7.992ms	8.008ms	0.73%	3.625	19.694	0.11%	0.001ms	1	200
flagevaluation_evp/coalescer/stress/10flags_1000users_250fields	throughput	124880.323op/s	125771.118op/s ± 132.557op/s	125792.778op/s ± 53.677op/s	125840.196op/s	125909.280op/s	125967.047op/s	125983.009op/s	0.15%	-3.598	19.476	0.11%	9.373op/s	1	200
flagevaluation_evp/coalescer/typical/100flags_50users_10fields	execution_time	180.008µs	185.268µs ± 38.810µs	182.406µs ± 0.252µs	182.746µs	183.505µs	184.004µs	731.321µs	300.93%	14.031	194.913	20.90%	2.744µs	1	200
flagevaluation_evp/coalescer/typical/100flags_50users_10fields	throughput	136738.860op/s	545824.351op/s ± 29125.388op/s	548226.955op/s ± 757.517op/s	548838.721op/s	549492.275op/s	553253.855op/s	555530.248op/s	1.33%	-13.959	193.559	5.32%	2059.476op/s	1	200

scenario	metric	95% CI mean	Shapiro-Wilk pvalue	Ljung-Box pvalue (lag=1)	Dip test pvalue
flagevaluation_evp/coalescer/scale/2500flags_500users_20fields	execution_time	[5.414ms; 5.416ms] or [-0.017%; +0.017%]	None	None	None
flagevaluation_evp/coalescer/scale/2500flags_500users_20fields	throughput	[461613.944op/s; 461768.671op/s] or [-0.017%; +0.017%]	None	None	None
flagevaluation_evp/coalescer/stress/10flags_1000users_250fields	execution_time	[7.950ms; 7.952ms] or [-0.015%; +0.015%]	None	None	None
flagevaluation_evp/coalescer/stress/10flags_1000users_250fields	throughput	[125752.747op/s; 125789.489op/s] or [-0.015%; +0.015%]	None	None	None
flagevaluation_evp/coalescer/typical/100flags_50users_10fields	execution_time	[179.889µs; 190.646µs] or [-2.903%; +2.903%]	None	None	None
flagevaluation_evp/coalescer/typical/100flags_50users_10fields	throughput	[541787.853op/s; 549860.850op/s] or [-0.740%; +0.740%]	None	None	None

Group 2

cpu_model	git_commit_sha	git_commit_date	git_branch
Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz	`f1baf07`	1787237900	vickie/ffl-3038-add-observe_full_evaluation_data-accessor

scenario	metric	min	mean ± sd	median ± mad	p75	p95	p99	max	peak_to_median_ratio	skewness	kurtosis	cv	sem	runs	sample_size
single_flag_killswitch/rules-based	execution_time	156.723ns	159.792ns ± 2.413ns	159.289ns ± 1.318ns	160.848ns	164.147ns	166.964ns	169.920ns	6.67%	1.227	1.697	1.51%	0.171ns	1	200

scenario	metric	95% CI mean	Shapiro-Wilk pvalue	Ljung-Box pvalue (lag=1)	Dip test pvalue
single_flag_killswitch/rules-based	execution_time	[159.457ns; 160.126ns] or [-0.209%; +0.209%]	None	None	None

Group 3

cpu_model	git_commit_sha	git_commit_date	git_branch
Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz	`f1baf07`	1787237900	vickie/ffl-3038-add-observe_full_evaluation_data-accessor

scenario	metric	min	mean ± sd	median ± mad	p75	p95	p99	max	peak_to_median_ratio	skewness	kurtosis	cv	sem	runs	sample_size
flagevaluation_evp/payloads/scale/2500flags_500users_20fields	execution_time	21.791ms	21.892ms ± 0.040ms	21.892ms ± 0.024ms	21.914ms	21.966ms	21.981ms	22.003ms	0.51%	0.183	0.053	0.18%	0.003ms	1	200
flagevaluation_evp/payloads/scale/2500flags_500users_20fields	throughput	113621.456op/s	114196.084op/s ± 208.615op/s	114197.228op/s ± 124.331op/s	114323.732op/s	114564.847op/s	114663.358op/s	114726.658op/s	0.46%	-0.172	0.051	0.18%	14.751op/s	1	200
flagevaluation_evp/payloads/stress/10flags_1000users_250fields	execution_time	102.187ms	102.839ms ± 0.346ms	102.844ms ± 0.233ms	103.062ms	103.349ms	103.491ms	104.780ms	1.88%	0.797	3.624	0.34%	0.024ms	1	200
flagevaluation_evp/payloads/stress/10flags_1000users_250fields	throughput	9543.828op/s	9724.067op/s ± 32.599op/s	9723.460op/s ± 21.940op/s	9746.972op/s	9773.054op/s	9783.643op/s	9785.970op/s	0.64%	-0.748	3.341	0.33%	2.305op/s	1	200
flagevaluation_evp/payloads/typical/100flags_50users_10fields	execution_time	596.601µs	599.242µs ± 1.526µs	598.780µs ± 1.246µs	600.891µs	601.550µs	601.901µs	602.036µs	0.54%	0.221	-1.428	0.25%	0.108µs	1	200
flagevaluation_evp/payloads/typical/100flags_50users_10fields	throughput	166103.071op/s	166878.567op/s ± 424.834op/s	167006.185op/s ± 347.878op/s	167238.423op/s	167451.833op/s	167554.974op/s	167616.267op/s	0.37%	-0.217	-1.428	0.25%	30.040op/s	1	200

scenario	metric	95% CI mean	Shapiro-Wilk pvalue	Ljung-Box pvalue (lag=1)	Dip test pvalue
flagevaluation_evp/payloads/scale/2500flags_500users_20fields	execution_time	[21.887ms; 21.898ms] or [-0.025%; +0.025%]	None	None	None
flagevaluation_evp/payloads/scale/2500flags_500users_20fields	throughput	[114167.172op/s; 114224.996op/s] or [-0.025%; +0.025%]	None	None	None
flagevaluation_evp/payloads/stress/10flags_1000users_250fields	execution_time	[102.791ms; 102.887ms] or [-0.047%; +0.047%]	None	None	None
flagevaluation_evp/payloads/stress/10flags_1000users_250fields	throughput	[9719.549op/s; 9728.585op/s] or [-0.046%; +0.046%]	None	None	None
flagevaluation_evp/payloads/typical/100flags_50users_10fields	execution_time	[599.030µs; 599.454µs] or [-0.035%; +0.035%]	None	None	None
flagevaluation_evp/payloads/typical/100flags_50users_10fields	throughput	[166819.689op/s; 166937.444op/s] or [-0.035%; +0.035%]	None	None	None

Group 4

cpu_model	git_commit_sha	git_commit_date	git_branch
Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz	`f1baf07`	1787237900	vickie/ffl-3038-add-observe_full_evaluation_data-accessor

scenario	metric	min	mean ± sd	median ± mad	p75	p95	p99	max	peak_to_median_ratio	skewness	kurtosis	cv	sem	runs	sample_size
sdk_test_data/rules-based	execution_time	142.671µs	145.905µs ± 5.834µs	144.610µs ± 0.676µs	145.645µs	150.917µs	168.367µs	204.827µs	41.64%	7.253	61.899	3.99%	0.412µs	1	200

scenario	metric	95% CI mean	Shapiro-Wilk pvalue	Ljung-Box pvalue (lag=1)	Dip test pvalue
sdk_test_data/rules-based	execution_time	[145.096µs; 146.713µs] or [-0.554%; +0.554%]	None	None	None

Baseline

Baseline benchmark details

Group 1

cpu_model	git_commit_sha	git_commit_date	git_branch
Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz	`24022a9`	1787233149	main

scenario	metric	min	mean ± sd	median ± mad	p75	p95	p99	max	peak_to_median_ratio	skewness	kurtosis	cv	sem	runs	sample_size
flagevaluation_evp/coalescer/scale/2500flags_500users_20fields	execution_time	5.394ms	5.414ms ± 0.008ms	5.414ms ± 0.004ms	5.418ms	5.425ms	5.431ms	5.454ms	0.74%	0.596	3.409	0.14%	0.001ms	1	200
flagevaluation_evp/coalescer/scale/2500flags_500users_20fields	throughput	458337.573op/s	461764.085op/s ± 649.141op/s	461749.716op/s ± 359.687op/s	462119.904op/s	462808.942op/s	463224.213op/s	463446.685op/s	0.37%	-0.574	3.319	0.14%	45.901op/s	1	200
flagevaluation_evp/coalescer/stress/10flags_1000users_250fields	execution_time	7.958ms	7.973ms ± 0.007ms	7.973ms ± 0.004ms	7.976ms	7.982ms	8.005ms	8.010ms	0.47%	1.936	7.490	0.09%	0.000ms	1	200
flagevaluation_evp/coalescer/stress/10flags_1000users_250fields	throughput	124842.735op/s	125417.734op/s ± 109.556op/s	125430.847op/s ± 58.044op/s	125485.028op/s	125550.964op/s	125601.165op/s	125656.687op/s	0.18%	-1.921	7.410	0.09%	7.747op/s	1	200
flagevaluation_evp/coalescer/typical/100flags_50users_10fields	execution_time	181.249µs	183.830µs ± 0.548µs	183.702µs ± 0.210µs	184.002µs	184.937µs	185.309µs	186.316µs	1.42%	0.518	5.159	0.30%	0.039µs	1	200
flagevaluation_evp/coalescer/typical/100flags_50users_10fields	throughput	536723.998op/s	543984.343op/s ± 1619.989op/s	544358.857op/s ± 622.529op/s	544878.338op/s	545342.718op/s	548185.545op/s	551726.878op/s	1.35%	-0.456	5.228	0.30%	114.551op/s	1	200

scenario	metric	95% CI mean	Shapiro-Wilk pvalue	Ljung-Box pvalue (lag=1)	Dip test pvalue
flagevaluation_evp/coalescer/scale/2500flags_500users_20fields	execution_time	[5.413ms; 5.415ms] or [-0.019%; +0.019%]	None	None	None
flagevaluation_evp/coalescer/scale/2500flags_500users_20fields	throughput	[461674.121op/s; 461854.050op/s] or [-0.019%; +0.019%]	None	None	None
flagevaluation_evp/coalescer/stress/10flags_1000users_250fields	execution_time	[7.972ms; 7.974ms] or [-0.012%; +0.012%]	None	None	None
flagevaluation_evp/coalescer/stress/10flags_1000users_250fields	throughput	[125402.551op/s; 125432.917op/s] or [-0.012%; +0.012%]	None	None	None
flagevaluation_evp/coalescer/typical/100flags_50users_10fields	execution_time	[183.754µs; 183.906µs] or [-0.041%; +0.041%]	None	None	None
flagevaluation_evp/coalescer/typical/100flags_50users_10fields	throughput	[543759.828op/s; 544208.858op/s] or [-0.041%; +0.041%]	None	None	None

Group 2

cpu_model	git_commit_sha	git_commit_date	git_branch
Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz	`24022a9`	1787233149	main

scenario	metric	min	mean ± sd	median ± mad	p75	p95	p99	max	peak_to_median_ratio	skewness	kurtosis	cv	sem	runs	sample_size
single_flag_killswitch/rules-based	execution_time	155.856ns	159.741ns ± 2.613ns	159.260ns ± 1.656ns	161.317ns	164.691ns	167.195ns	168.960ns	6.09%	0.890	0.322	1.63%	0.185ns	1	200

scenario	metric	95% CI mean	Shapiro-Wilk pvalue	Ljung-Box pvalue (lag=1)	Dip test pvalue
single_flag_killswitch/rules-based	execution_time	[159.379ns; 160.103ns] or [-0.227%; +0.227%]	None	None	None

Group 3

cpu_model	git_commit_sha	git_commit_date	git_branch
Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz	`24022a9`	1787233149	main

scenario	metric	min	mean ± sd	median ± mad	p75	p95	p99	max	peak_to_median_ratio	skewness	kurtosis	cv	sem	runs	sample_size
flagevaluation_evp/payloads/scale/2500flags_500users_20fields	execution_time	20.914ms	21.035ms ± 0.060ms	21.025ms ± 0.036ms	21.066ms	21.145ms	21.247ms	21.278ms	1.20%	1.192	2.284	0.29%	0.004ms	1	200
flagevaluation_evp/payloads/scale/2500flags_500users_20fields	throughput	117491.967op/s	118848.336op/s ± 340.261op/s	118905.482op/s ± 205.072op/s	119081.403op/s	119270.048op/s	119394.142op/s	119539.227op/s	0.53%	-1.168	2.195	0.29%	24.060op/s	1	200
flagevaluation_evp/payloads/stress/10flags_1000users_250fields	execution_time	95.334ms	95.784ms ± 0.297ms	95.854ms ± 0.153ms	95.974ms	96.059ms	96.110ms	98.208ms	2.46%	2.454	20.303	0.31%	0.021ms	1	200
flagevaluation_evp/payloads/stress/10flags_1000users_250fields	throughput	10182.420op/s	10440.205op/s ± 32.100op/s	10432.499op/s ± 16.628op/s	10473.142op/s	10481.122op/s	10484.165op/s	10489.463op/s	0.55%	-2.307	18.824	0.31%	2.270op/s	1	200
flagevaluation_evp/payloads/typical/100flags_50users_10fields	execution_time	577.619µs	580.269µs ± 1.252µs	580.372µs ± 0.942µs	581.146µs	582.405µs	583.108µs	583.547µs	0.55%	0.092	-0.624	0.22%	0.089µs	1	200
flagevaluation_evp/payloads/typical/100flags_50users_10fields	throughput	171365.691op/s	172334.603op/s ± 371.778op/s	172303.193op/s ± 279.636op/s	172626.423op/s	172913.165op/s	173075.221op/s	173124.356op/s	0.48%	-0.083	-0.630	0.22%	26.289op/s	1	200

scenario	metric	95% CI mean	Shapiro-Wilk pvalue	Ljung-Box pvalue (lag=1)	Dip test pvalue
flagevaluation_evp/payloads/scale/2500flags_500users_20fields	execution_time	[21.027ms; 21.044ms] or [-0.040%; +0.040%]	None	None	None
flagevaluation_evp/payloads/scale/2500flags_500users_20fields	throughput	[118801.179op/s; 118895.493op/s] or [-0.040%; +0.040%]	None	None	None
flagevaluation_evp/payloads/stress/10flags_1000users_250fields	execution_time	[95.743ms; 95.826ms] or [-0.043%; +0.043%]	None	None	None
flagevaluation_evp/payloads/stress/10flags_1000users_250fields	throughput	[10435.756op/s; 10444.653op/s] or [-0.043%; +0.043%]	None	None	None
flagevaluation_evp/payloads/typical/100flags_50users_10fields	execution_time	[580.096µs; 580.443µs] or [-0.030%; +0.030%]	None	None	None
flagevaluation_evp/payloads/typical/100flags_50users_10fields	throughput	[172283.079op/s; 172386.128op/s] or [-0.030%; +0.030%]	None	None	None

Group 4

cpu_model	git_commit_sha	git_commit_date	git_branch
Intel(R) Xeon(R) Platinum 8259CL CPU @ 2.50GHz	`24022a9`	1787233149	main

scenario	metric	min	mean ± sd	median ± mad	p75	p95	p99	max	peak_to_median_ratio	skewness	kurtosis	cv	sem	runs	sample_size
sdk_test_data/rules-based	execution_time	140.028µs	143.702µs ± 5.713µs	142.448µs ± 0.637µs	143.286µs	148.373µs	167.353µs	203.346µs	42.75%	7.393	65.500	3.97%	0.404µs	1	200

scenario	metric	95% CI mean	Shapiro-Wilk pvalue	Ljung-Box pvalue (lag=1)	Dip test pvalue
sdk_test_data/rules-based	execution_time	[142.910µs; 144.494µs] or [-0.551%; +0.551%]	None	None	None

dd-octo-sts · 2026-08-19T14:33:40Z

Artifact Size Benchmark Report

aarch64-alpine-linux-musl

Artifact	Baseline	Commit	Change
/aarch64-alpine-linux-musl/lib/libdatadog_profiling.so	8.20 MB	8.26 MB	+.76% (+64.00 KB) 🔍
/aarch64-alpine-linux-musl/lib/libdatadog_profiling.a	89.47 MB	89.51 MB	+.04% (+41.45 KB) 🔍

aarch64-unknown-linux-gnu

Artifact	Baseline	Commit	Change
/aarch64-unknown-linux-gnu/lib/libdatadog_profiling.a	100.65 MB	100.69 MB	+.04% (+41.53 KB) 🔍
/aarch64-unknown-linux-gnu/lib/libdatadog_profiling.so	11.06 MB	11.06 MB	+.01% (+1.82 KB) 🔍

libdatadog-x64-windows

Artifact	Baseline	Commit	Change
/libdatadog-x64-windows/debug/dynamic/datadog_profiling_ffi.dll	26.62 MB	26.64 MB	+.05% (+16.00 KB) 🔍
/libdatadog-x64-windows/debug/dynamic/datadog_profiling_ffi.lib	94.56 KB	94.96 KB	+.42% (+414 B) 🔍
/libdatadog-x64-windows/debug/dynamic/datadog_profiling_ffi.pdb	180.69 MB	180.74 MB	+.03% (+56.00 KB) 🔍
/libdatadog-x64-windows/debug/static/datadog_profiling_ffi.lib	772.58 MB	772.82 MB	+.03% (+247.01 KB) 🔍
/libdatadog-x64-windows/release/dynamic/datadog_profiling_ffi.dll	8.72 MB	8.73 MB	+.06% (+5.50 KB) 🔍
/libdatadog-x64-windows/release/dynamic/datadog_profiling_ffi.lib	94.56 KB	94.96 KB	+.42% (+414 B) 🔍
/libdatadog-x64-windows/release/dynamic/datadog_profiling_ffi.pdb	25.64 MB	25.65 MB	+.03% (+8.00 KB) 🔍
/libdatadog-x64-windows/release/static/datadog_profiling_ffi.lib	51.06 MB	51.09 MB	+.04% (+25.26 KB) 🔍

libdatadog-x86-windows

Artifact	Baseline	Commit	Change
/libdatadog-x86-windows/debug/dynamic/datadog_profiling_ffi.dll	23.20 MB	23.22 MB	+.06% (+14.50 KB) 🔍
/libdatadog-x86-windows/debug/dynamic/datadog_profiling_ffi.lib	96.04 KB	96.45 KB	+.42% (+420 B) 🔍
/libdatadog-x86-windows/debug/dynamic/datadog_profiling_ffi.pdb	185.60 MB	185.66 MB	+.02% (+56.00 KB) 🔍
/libdatadog-x86-windows/debug/static/datadog_profiling_ffi.lib	759.08 MB	759.15 MB	+0% (+68.95 KB) 👌
/libdatadog-x86-windows/release/dynamic/datadog_profiling_ffi.dll	6.74 MB	6.74 MB	+.06% (+4.50 KB) 🔍
/libdatadog-x86-windows/release/dynamic/datadog_profiling_ffi.lib	96.04 KB	96.45 KB	+.42% (+420 B) 🔍
/libdatadog-x86-windows/release/dynamic/datadog_profiling_ffi.pdb	27.58 MB	27.59 MB	+.05% (+16.00 KB) 🔍
/libdatadog-x86-windows/release/static/datadog_profiling_ffi.lib	48.60 MB	48.62 MB	+.04% (+24.72 KB) 🔍

x86_64-alpine-linux-musl

Artifact	Baseline	Commit	Change
/x86_64-alpine-linux-musl/lib/libdatadog_profiling.a	79.74 MB	79.77 MB	+.04% (+35.69 KB) 🔍
/x86_64-alpine-linux-musl/lib/libdatadog_profiling.so	9.16 MB	9.16 MB	+.04% (+4.00 KB) 🔍

x86_64-unknown-linux-gnu

Artifact	Baseline	Commit	Change
/x86_64-unknown-linux-gnu/lib/libdatadog_profiling.a	95.38 MB	95.41 MB	+.03% (+35.98 KB) 🔍
/x86_64-unknown-linux-gnu/lib/libdatadog_profiling.so	11.16 MB	11.17 MB	+.04% (+5.48 KB) 🔍

Model the top-level UFC `observeFullEvaluationData` boolean through the wire -> compiled -> Configuration layers and add `ddog_ffe_configuration_get_observe_full_evaluation_data` so SDKs can read whether to emit full flag-evaluation data without re-parsing the raw JSON string. Defaults to `false` (privacy-preserving) when the field is absent from the server payload, so older configurations are a no-op. This is the libdatadog half of the "Protecting PII in flagevaluations track" initiative; the server-side UFC field already ships via ddoghq/dd-source#22826 and the RC schema via ddoghq/dd-go#3627. Refs: FFL-3038 Generated with Claude Code

Address review findings on the observeFullEvaluationData accessor. A `null` or mistyped value for this optional field previously failed the parse of the whole configuration, which discarded every flag and forced SDKs onto their defaults. Parse it leniently instead: any value that is not a JSON boolean now falls back to `false`, the privacy-preserving default. A malformed value can no longer take down unrelated flags. `TryParse<bool>` was not usable here because `#[serde(default)]` needs `Default`, which `TryParse` does not implement. Routing through `serde_json::Value` also consumes nested shapes (`{}`, `[]`) cleanly, which `Option::<bool>::deserialize` does not guarantee after a type error. The new FFI symbol had no assertion anywhere. The shared test fixture omits the key, so `cargo ffi-test` only ever exercised the `false` path. Add Rust tests that drive the real `ddog_ffe_configuration_new` -> getter -> `ddog_ffe_configuration_drop` sequence for the true, false, absent, and malformed cases, and make the C example assert the expected default rather than only printing it. Also re-wrap the three new doc comments to satisfy `cargo fmt --check`. Refs: FFL-3038 Generated with Claude Code Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 5dd17bc9b7

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".

pavlokhrebto · 2026-08-19T18:27:24Z

lgtm, please address Codex review comments

Generated with Claude Code

Aaalibaba42

lgtm o/

…ker, libdd-data-pipeline, li... (#2482) # Release proposal for libdd-capabilities-impl, libdd-crashtracker, libdd-data-pipeline, libdd-ddsketch, libdd-ffe, libdd-http-client, libdd-ipc, libdd-library-config, libdd-live-debugger, libdd-otel-thread-ctx, libdd-remote-config, libdd-shared-runtime, libdd-telemetry, libdd-trace-utils, libdd-tracer-flare and their dependencies This PR contains version bumps based on public API changes and commits since last release. ### ⚠️ Crates left out of this proposal affected by its major bumps These publishable workspace crates are not part of this release but their dependency requirement was rewritten on this branch while their published version still requires the old major. If they are a dependency on your deployment not including them in the release could result in duplicate packages or symbol incompatibility. - `libdd-common` `5.2.0` → `6.0.0` affects: `libdd-profiling`, `libdd-sampling` - `libdd-trace-utils` `11.0.0` → `12.0.0` affects: `libdd-sampling` ## libdd-capabilities **Next version:** `3.0.1` **Semver bump:** `patch` **Tag:** `libdd-capabilities-v3.0.1` ### Commits - refactor: migrate HTTP & networking deps to workspace level (phase 4bis) (#2350) - feat: do not entirely disable connection pooling for periodic connections (#2440) ## libdd-common **Next version:** `6.0.0` **Semver bump:** `major` **Tag:** `libdd-common-v6.0.0` ### Commits - refactor: migrate HTTP & networking deps to workspace level (phase 4bis) (#2350) - feat: do not entirely disable connection pooling for periodic connections (#2440) - feat(data-pipeline)!: add agentless stats export (#2309) - feat(data-pipeline): add runtime-independent agentless sending (#2389) - feat(common)!: add HTTPS_PROXY support for hyper_backend (#2421) ## libdd-ipc-macros **Next version:** `1.0.1` **Semver bump:** `patch` **Tag:** `libdd-ipc-macros-v1.0.1` ### Commits - feat(sidecar)!: support appsec helper-rust integration with sidecar (#2310) ## libdd-otel-thread-ctx **Next version:** `1.1.0` **Semver bump:** `minor` **Tag:** `libdd-otel-thread-ctx-v1.1.0` ### Commits - feat(otel-thread-ctx): add update-and-attach operation (#2443) ## libdd-tinybytes **Next version:** `1.1.3` **Semver bump:** `patch` **Tag:** `libdd-tinybytes-v1.1.3` ### Commits - refactor: migrate HTTP & networking deps to workspace level (phase 4bis) (#2350) ## libdd-capabilities-impl **Next version:** `5.0.0` **Semver bump:** `major` **Tag:** `libdd-capabilities-impl-v5.0.0` ### ⚠️ major bump forced due to: - `libdd-common`: ^5.1.1 → ^6.0.0 ### Commits - refactor: migrate HTTP & networking deps to workspace level (phase 4bis) (#2350) - feat: do not entirely disable connection pooling for periodic connections (#2440) - feat(data-pipeline): add runtime-independent agentless sending (#2389) ## libdd-http-client **Next version:** `2.0.0` **Semver bump:** `major` **Tag:** `libdd-http-client-v2.0.0` ### ⚠️ major bump forced due to: - `libdd-common`: ^5.1.1 → ^6.0.0 ### Commits - refactor: migrate HTTP & networking deps to workspace level (phase 4bis) (#2350) - feat(data-pipeline): add runtime-independent agentless sending (#2389) - feat(common)!: add HTTPS_PROXY support for hyper_backend (#2421) ## libdd-remote-config **Next version:** `5.0.0` **Semver bump:** `major` **Tag:** `libdd-remote-config-v5.0.0` ### ⚠️ major bump forced due to: - `libdd-common`: ^5.2.0 → ^6.0.0 ### Commits - fix(remote-config): refresh fetcher identity (#2469) - refactor: migrate HTTP & networking deps to workspace level (phase 4bis) (#2350) - fix(remote-config): reuse injected sleep capability (#2429) ## libdd-shared-runtime **Next version:** `4.0.0` **Semver bump:** `major` **Tag:** `libdd-shared-runtime-v4.0.0` ### ⚠️ major bump forced due to: - `libdd-common`: ^5.2.0 → ^6.0.0 ### Commits - fix(shared-runtime): allow disabling worker fork restart (#2464) ## libdd-trace-utils **Next version:** `12.0.0` **Semver bump:** `major` **Tag:** `libdd-trace-utils-v12.0.0` ### ⚠️ major bump forced due to: - `libdd-common`: ^5.2.0 → ^6.0.0 ### Commits - feat(trace-utils): add v1-native JSON log encoder brick (#2371) - feat(trace-utils): add v1-native agentless JSON encoder brick (#2370) - refactor: migrate HTTP & networking deps to workspace level (phase 4bis) (#2350) - fix(trace-stats): read OTel HTTP names for the status and method dimensions (#2323) - feat(data-pipeline): emit native trace export telemetry (#2338) - feat(data-pipeline)!: add agentless stats export (#2309) - fix(trace-utils): use vec map dedup when serializing (#2422) - feat(data-pipeline): add runtime-independent agentless sending (#2389) - fix(compression): align zstd behavior across targets (#2400) - feat(trace-utils)!: add from owned to SpanText (#2403) ## libdd-ffe **Next version:** `2.0.0` **Semver bump:** `major` **Tag:** `libdd-ffe-v2.0.0` ### ⚠️ major bump forced due to: - `libdd-common`: ^5.1.1 → ^6.0.0 - `libdd-remote-config`: ^3.0.0 → ^4.1.0 - `libdd-trace-protobuf`: ^4.0.1 → ^5.0.0 ### Commits - refactor: migrate HTTP & networking deps to workspace level (phase 4bis) (#2350) - feat(ffe): support arbitrary semver core parts (#2413) - feat(ffe)!: send the split serial id on exposure events [EX-3425] (#2402) - feat(ffe): expose observeFullEvaluationData config-level FFI getter (#2373) - fix(ffe): report rejected flags as parse errors (#2339) - test: skip/shorten slow miri jobs (#2331) ## libdd-dogstatsd-client **Next version:** `6.0.0` **Semver bump:** `major` **Tag:** `libdd-dogstatsd-client-v6.0.0` ### ⚠️ major bump forced due to: - `libdd-common`: ^5.2.0 → ^6.0.0 ### Commits - refactor: migrate HTTP & networking deps to workspace level (phase 4bis) (#2350) ## libdd-telemetry **Next version:** `8.0.0` **Semver bump:** `major` **Tag:** `libdd-telemetry-v8.0.0` ### ⚠️ major bump forced due to: - `libdd-common`: ^5.2.0 → ^6.0.0 ### Commits - refactor: migrate HTTP & networking deps to workspace level (phase 4bis) (#2350) - refactor(telemetry): avoid doing two separate http requests in stop telemetry (#2435) ## libdd-trace-obfuscation **Next version:** `8.0.0` **Semver bump:** `major` **Tag:** `libdd-trace-obfuscation-v8.0.0` ### ⚠️ major bump forced due to: - `libdd-common`: ^5.2.0 → ^6.0.0 - `libdd-trace-utils`: ^11.0.0 → ^12.0.0 ### Commits - fix(trace-obfuscation): scan all span meta for credit-card obfuscation (#2472) - refactor: migrate HTTP & networking deps to workspace level (phase 4bis) (#2350) - feat(data-pipeline)!: Obfuscate v04 spans in agentless context (#2418) ## libdd-tracer-flare **Next version:** `3.0.0` **Semver bump:** `major` **Tag:** `libdd-tracer-flare-v3.0.0` ### ⚠️ major bump forced due to: - `libdd-common`: ^5.2.0 → ^6.0.0 - `libdd-trace-utils`: ^11.0.0 → ^12.0.0 ### Commits - refactor: migrate HTTP & networking deps to workspace level (phase 4bis) (#2350) ## libdd-crashtracker **Next version:** `3.0.0` **Semver bump:** `major` **Tag:** `libdd-crashtracker-v3.0.0` ### ⚠️ major bump forced due to: - `libdd-common`: ^5.2.0 → ^6.0.0 ### Commits - refactor: migrate HTTP & networking deps to workspace level (phase 4bis) (#2350) - fix(crashtracking): filter out frames above faulting frame (#2428) - feat(sidecar)!: support appsec helper-rust integration with sidecar (#2310) - chore(crashtracking): use RAII remote ptrace API (#2416) - chore(crashtracking): bump libdd-libunwind-sys to v1.0.3 (#2414) ## libdd-data-pipeline-core **Next version:** `1.0.0` **Semver bump:** `major` **Tag:** `libdd-data-pipeline-core-v1.0.0` **Warning:** this is an initial release. Please verify that the version and commits included are correct. ## libdd-trace-stats **Next version:** `9.0.0` **Semver bump:** `major` **Tag:** `libdd-trace-stats-v9.0.0` ### ⚠️ major bump forced due to: - `libdd-common`: ^5.2.0 → ^6.0.0 - `libdd-trace-obfuscation`: ^7.0.0 → ^8.0.0 - `libdd-trace-utils`: ^11.0.0 → ^12.0.0 ### Commits - refactor: migrate HTTP & networking deps to workspace level (phase 4bis) (#2350) - fix(trace-stats): read OTel HTTP names for the status and method dimensions (#2323) - feat(data-pipeline)!: add agentless stats export (#2309) - feat(trace-utils)!: add from owned to SpanText (#2403) ## libdd-data-pipeline **Next version:** `10.0.0` **Semver bump:** `major` **Tag:** `libdd-data-pipeline-v10.0.0` ### ⚠️ major bump forced due to: - `libdd-common`: ^5.2.0 → ^6.0.0 - `libdd-trace-obfuscation`: ^7.0.0 → ^8.0.0 - `libdd-trace-stats`: ^8.0.0 → ^9.0.0 - `libdd-trace-utils`: ^11.0.0 → ^12.0.0 ### Commits - feat(trace-utils): add v1-native JSON log encoder brick (#2371) - chore: prepare crate for publishing (#2466) - refactor: migrate HTTP & networking deps to workspace level (phase 4bis) (#2350) - fix(trace-stats): read OTel HTTP names for the status and method dimensions (#2323) - fix(data-pipeline): pass obfuscation config to OTLP stats (#2444) - feat(data-pipeline): emit native trace export telemetry (#2338) - feat(data-pipeline): add fork-safe OTLP gRPC trace transport (#2273) - feat(data-pipeline)!: add agentless stats export (#2309) - feat(data-pipeline)!: Obfuscate v04 spans in agentless context (#2418) - feat(data-pipeline): add runtime-independent agentless sending (#2389) - feat(trace-utils)!: add from owned to SpanText (#2403) ## libdd-ipc **Next version:** `2.0.0` **Semver bump:** `major` **Tag:** `libdd-ipc-v2.0.0` ### ⚠️ major bump forced due to: - `libdd-common`: ^5.2.0 → ^6.0.0 - `libdd-trace-stats`: ^8.0.0 → ^9.0.0 ### Commits - fix(ipc): drop the signal feature from libdd-ipc (#2431) ## libdd-live-debugger **Next version:** `1.0.0` **Semver bump:** `major` **Tag:** `libdd-live-debugger-v1.0.0` **Warning:** this is an initial release. Please verify that the version and commits included are correct. [EX-3425]: https://datadoghq.atlassian.net/browse/EX-3425?atlOrigin=eyJpIjoiNWRkNTljNzYxNjVmNDY3MDlhMDU5Y2ZhYzA5YTRkZjUiLCJwIjoiZ2l0aHViLWNvbS1KU1cifQ --------- Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: hoolioh <107922352+hoolioh@users.noreply.github.com>

vjfridge requested review from a team as code owners August 19, 2026 13:58

vjfridge requested review from leoromanovsky and pavlokhrebto and removed request for a team August 19, 2026 13:58

vjfridge marked this pull request as draft August 19, 2026 13:59

vjfridge force-pushed the vickie/ffl-3038-add-observe_full_evaluation_data-accessor branch 2 times, most recently from 1b7d17e to 6a33037 Compare August 19, 2026 15:58

vjfridge commented Aug 19, 2026

View reviewed changes

Comment thread libdd-ffe-ffi/src/configuration.rs Outdated

vjfridge commented Aug 19, 2026

View reviewed changes

Comment thread libdd-ffe/src/rules_based/ufc/compiled_flag_config.rs Outdated

vjfridge commented Aug 19, 2026

View reviewed changes

Comment thread libdd-ffe/src/rules_based/ufc/models.rs Outdated

vjfridge force-pushed the vickie/ffl-3038-add-observe_full_evaluation_data-accessor branch from b021c65 to ea343ad Compare August 19, 2026 16:07

vjfridge marked this pull request as ready for review August 19, 2026 18:04

chatgpt-codex-connector Bot reviewed Aug 19, 2026

View reviewed changes

Comment thread libdd-ffe-ffi/src/configuration.rs

Comment thread examples/ffi/ffe.c Outdated

pavlokhrebto approved these changes Aug 19, 2026

View reviewed changes

fix(ffe): allow opt-in configs in C example

f1baf07

Generated with Claude Code

Aaalibaba42 approved these changes Aug 20, 2026

View reviewed changes

gh-worker-dd-devflow-36fce6 Bot added mergequeue-status: queued mergequeue-status: in_progress and removed mergequeue-status: queued labels Aug 20, 2026

gh-worker-dd-mergequeue-cf854d Bot merged commit 378be45 into main Aug 20, 2026
78 checks passed

gh-worker-dd-devflow-36fce6 Bot added mergequeue-status: done and removed mergequeue-status: in_progress labels Aug 20, 2026

gh-worker-dd-mergequeue-cf854d Bot deleted the vickie/ffl-3038-add-observe_full_evaluation_data-accessor branch August 20, 2026 18:47

dd-octo-sts Bot mentioned this pull request Sep 8, 2026

chore(release): proposal for libdd-capabilities-impl, libdd-crashtracker, libdd-data-pipeline, li... #2482

Merged

Conversation

vjfridge commented Aug 19, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Changes

Test plan

Uh oh!

vjfridge commented Aug 19, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Note to self: Follow-ups to propagate to SDK repos

Other language SDKs to check

Uh oh!

datadog-datadog-prod-us1-2 Bot commented Aug 19, 2026 • edited by datadog-prod-us1-3 Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pr-commenter Bot commented Aug 19, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Benchmarks

Comparison

Explanation

More details about the CI and significant changes

scenario:flagevaluation_evp/payloads/scale/2500flags_500users_20fields

scenario:flagevaluation_evp/payloads/stress/10flags_1000users_250fields

Candidate

Group 1

Group 2

Group 3

Group 4

Baseline

Group 1

Group 2

Group 3

Group 4

Uh oh!

dd-octo-sts Bot commented Aug 19, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Artifact Size Benchmark Report

Uh oh!

Uh oh!

Uh oh!

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

Uh oh!

pavlokhrebto commented Aug 19, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Aaalibaba42 left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

vjfridge commented Aug 19, 2026 •

edited

Loading

vjfridge commented Aug 19, 2026 •

edited

Loading

datadog-datadog-prod-us1-2 Bot commented Aug 19, 2026 •

edited by datadog-prod-us1-3 Bot

Loading

pr-commenter Bot commented Aug 19, 2026 •

edited

Loading

dd-octo-sts Bot commented Aug 19, 2026 •

edited

Loading

pavlokhrebto commented Aug 19, 2026 •

edited

Loading