feat: warn (AWT106) when a factory body hides an IAsyncInitializable/IDisposable capability behind its declared return type

[vbreuss]

Taint and disposability for a factory registration are read off the producer method's declared return type, so a concrete type that implements IAsyncInitializable or IDisposable but is returned behind a plainer interface is invisible to the container: its async initialization never runs, or its disposable is never disposed. This adds a best-effort, zero-false-positive lint that inspects the factory method's own return expressions (expression-bodied and block-bodied, never descending into nested lambdas or local functions) and fires only when the statically determined returned type is provably a non-abstract, non-interface named type carrying a capability the declared return type does not expose. Metadata-only factories and unanalyzable returns are left silent; helper-returned or runtime-selected implementations are accepted false negatives.

Severity Warning, since the lint has false negatives by design. One descriptor with a {capability} placeholder (async-initializable / disposable) keeps the AWT numbering tight.

[github-actions]

Test Results

   18 files  ± 0     18 suites  ±0   59s ⏱️ ±0s
  248 tests +13    247 ✅ +13  1 💤 ±0  0 ❌ ±0 
1 232 runs  +39  1 231 ✅ +39  1 💤 ±0  0 ❌ ±0 

Results for commit 144b59e. ± Comparison against base commit 14c9c20.

♻️ This comment has been updated with latest results.

[github-actions]

🚀 Benchmark Results

Details

BenchmarkDotNet v0.15.8, Linux Ubuntu 24.04.4 LTS (Noble Numbat)

AMD EPYC 7763 2.45GHz, 1 CPU, 4 logical and 2 physical cores

.NET SDK 10.0.301

[Host] : .NET 10.0.9 (10.0.9, 10.0.926.27113), X64 RyuJIT x86-64-v3
Job=InProcess  Toolchain=InProcessEmitToolchain  IterationCount=15

LaunchCount=1  WarmupCount=10

Resolve	Size	Mean	Error	StdDev	Ratio	Allocated	Alloc Ratio
baseline*	8	9.945 ns	0.0056 ns	0.0044 ns	0.71	-	NA
Awaiten	8	13.937 ns	0.0158 ns	0.0132 ns	1.00	-	NA
MsDI	8	8.170 ns	0.5996 ns	0.5608 ns	0.59	-	NA
Autofac	8	113.541 ns	1.0272 ns	0.8578 ns	8.15	656 B	NA
Jab	8	2.652 ns	0.0035 ns	0.0027 ns	0.19	-	NA
PureDI	8	5.434 ns	0.0088 ns	0.0078 ns	0.39	-	NA
DryIoc	8	9.079 ns	0.0151 ns	0.0134 ns	0.65	-	NA
SimpleInjector	8	11.234 ns	0.0080 ns	0.0075 ns	0.81	-	NA
baseline*	256	360.425 ns	0.7201 ns	0.6014 ns	0.86	-	NA
Awaiten	256	420.991 ns	0.1860 ns	0.1452 ns	1.00	-	NA
MsDI	256	7.687 ns	0.2130 ns	0.1888 ns	0.02	-	NA
Autofac	256	116.171 ns	2.3662 ns	2.2134 ns	0.28	656 B	NA
Jab	256	42.786 ns	0.0584 ns	0.0488 ns	0.10	-	NA
PureDI	256	7.644 ns	0.0055 ns	0.0043 ns	0.02	-	NA
DryIoc	256	10.355 ns	0.0111 ns	0.0086 ns	0.02	-	NA
SimpleInjector	256	14.867 ns	0.0900 ns	0.0842 ns	0.04	-	NA

Details

BenchmarkDotNet v0.15.8, Linux Ubuntu 24.04.4 LTS (Noble Numbat)

AMD EPYC 7763 2.45GHz, 1 CPU, 4 logical and 2 physical cores

.NET SDK 10.0.301

[Host] : .NET 10.0.9 (10.0.9, 10.0.926.27113), X64 RyuJIT x86-64-v3
Job=InProcess  Toolchain=InProcessEmitToolchain  IterationCount=15

LaunchCount=1  WarmupCount=10

Realistic	Mean	Error	StdDev	Ratio	Allocated	Alloc Ratio
baseline*	242.5 ns	2.52 ns	2.36 ns	1.00	560 B	1.00
Awaiten	243.3 ns	1.63 ns	1.44 ns	1.00	560 B	1.00
MsDI	591.7 ns	6.06 ns	5.67 ns	2.43	1104 B	1.97
Autofac	8,123.8 ns	129.50 ns	121.13 ns	33.40	13696 B	24.46
Jab	176.1 ns	3.26 ns	3.05 ns	0.72	432 B	0.77
DryIoc	392.0 ns	7.77 ns	7.27 ns	1.61	944 B	1.69
SimpleInjector	723.8 ns	5.72 ns	5.35 ns	2.98	1096 B	1.96
PureDI	178.5 ns	1.72 ns	1.61 ns	0.73	632 B	1.13

Details

BenchmarkDotNet v0.15.8, Linux Ubuntu 24.04.4 LTS (Noble Numbat)

AMD EPYC 7763 2.45GHz, 1 CPU, 4 logical and 2 physical cores

.NET SDK 10.0.301

[Host] : .NET 10.0.9 (10.0.9, 10.0.926.27113), X64 RyuJIT x86-64-v3
Job=InProcess  Toolchain=InProcessEmitToolchain  IterationCount=15

LaunchCount=1  WarmupCount=10

Build	Size	Mean	Error	StdDev	Ratio	Allocated	Alloc Ratio
baseline*	8	17.30 ns	0.412 ns	0.366 ns	1.06	136 B	1.00
Awaiten	8	16.28 ns	0.181 ns	0.160 ns	1.00	136 B	1.00
MsDI	8	1,456.55 ns	22.952 ns	21.469 ns	89.46	5688 B	41.82
Autofac	8	28,667.86 ns	118.665 ns	99.091 ns	1,760.66	33098 B	243.37
Jab	8	10.59 ns	0.662 ns	0.619 ns	0.65	96 B	0.71
PureDI	8	14.95 ns	0.422 ns	0.394 ns	0.92	128 B	0.94
DryIoc	8	645.69 ns	7.548 ns	7.060 ns	39.66	1472 B	10.82
SimpleInjector	8	11,692.77 ns	170.573 ns	142.436 ns	718.12	24760 B	182.06
baseline*	256	98.35 ns	2.402 ns	2.247 ns	1.15	2120 B	1.00
Awaiten	256	85.50 ns	2.595 ns	2.167 ns	1.00	2120 B	1.00
MsDI	256	14,597.65 ns	141.723 ns	132.568 ns	170.84	61016 B	28.78
Autofac	256	714,923.82 ns	8,759.137 ns	8,193.302 ns	8,366.90	741509 B	349.77
Jab	256	75.45 ns	2.183 ns	2.042 ns	0.88	2080 B	0.98
PureDI	256	88.57 ns	2.235 ns	2.090 ns	1.04	2112 B	1.00
DryIoc	256	44,365.43 ns	199.622 ns	166.694 ns	519.22	79167 B	37.34
SimpleInjector	256	348,008.02 ns	5,859.271 ns	5,480.766 ns	4,072.81	572992 B	270.28

baseline* rows show the corresponding Awaiten benchmark from the most recent successful main branch build with results, for regression comparison.

[github-actions]

👽 Mutation Results

Awaiten

Details

File	Score	Killed	Survived	Timeout	No Coverage	Ignored	Compile Errors	Runtime Errors	Total Detected	Total Undetected	Total Mutants

The final mutation score is NaN%

Coverage Thresholds: high:80 low:60 break:0

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
96.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud