net7.0 & C# style improvements

CoreyKaylor commented 1 year ago

Summary of PR:

Added usage of LibraryImport when targeting net7.0
Marshall environment name string as UTF-8 with the LibraryImport options
Using nint throughout over previous IntPtr
File-scoped namespaces
Fixed test being excluded on Linux for using environment from multi-process
Corrected numerous inconsistencies and analysis warnings where it made sense

Are there other language features this library would benefit from that I missed?

I'm going to run the benchmarks out of curiosity to see the differences between net6.0, net7.0 (master), and net7.0 (net7-csharp-improvements)

CoreyKaylor commented 1 year ago

(master) net6.0 READ


BenchmarkDotNet=v0.12.1, OS=macOS 13.2 (22D49) [Darwin 22.3.0]
Apple M1 Max, 1 CPU, 10 logical and 10 physical cores
.NET Core SDK=7.0.102
  [Host]     : .NET Core 6.0.13 (CoreCLR 6.0.1322.58009, CoreFX 6.0.1322.58009), Arm64 RyuJIT
  DefaultJob : .NET Core 6.0.13 (CoreCLR 6.0.1322.58009, CoreFX 6.0.1322.58009), Arm64 RyuJIT

Method	OpsPerTransaction	ValueSize	KeyOrder	Mean	Error	StdDev	Gen 0	Gen 1	Gen 2	Allocated
Read	1	8	Sequential	340.3 ns	2.02 ns	1.89 ns	0.1450	-	-	304 B
Read	1	64	Sequential	338.7 ns	3.52 ns	3.30 ns	0.1450	-	-	304 B
Read	1	256	Sequential	336.6 ns	3.42 ns	3.03 ns	0.1450	-	-	304 B
Read	100	8	Sequential	8,187.4 ns	11.03 ns	10.31 ns	0.1373	-	-	304 B
Read	100	64	Sequential	8,129.8 ns	15.42 ns	14.42 ns	0.1373	-	-	304 B
Read	100	256	Sequential	8,566.1 ns	22.25 ns	18.58 ns	0.1373	-	-	304 B
Read	1000	8	Sequential	116,099.3 ns	439.43 ns	389.55 ns	0.1221	-	-	304 B
Read	1000	64	Sequential	124,493.6 ns	293.72 ns	274.75 ns	-	-	-	304 B
Read	1000	256	Sequential	132,650.8 ns	308.18 ns	273.20 ns	-	-	-	304 B

CoreyKaylor commented 1 year ago

(master) net6.0 WRITE


BenchmarkDotNet=v0.12.1, OS=macOS 13.2 (22D49) [Darwin 22.3.0]
Apple M1 Max, 1 CPU, 10 logical and 10 physical cores
.NET Core SDK=7.0.102
  [Host]     : .NET Core 6.0.13 (CoreCLR 6.0.1322.58009, CoreFX 6.0.1322.58009), Arm64 RyuJIT
  DefaultJob : .NET Core 6.0.13 (CoreCLR 6.0.1322.58009, CoreFX 6.0.1322.58009), Arm64 RyuJIT

Method	OpsPerTransaction	ValueSize	KeyOrder	Mean	Error	StdDev	Gen 0	Gen 1	Gen 2	Allocated
Write	1	8	Sequential	50.64 μs	0.836 μs	0.741 μs	0.1221	-	-	304 B
Write	1	64	Sequential	50.14 μs	0.226 μs	0.212 μs	0.1221	-	-	304 B
Write	1	256	Sequential	50.33 μs	0.509 μs	0.425 μs	0.1221	-	-	304 B
Write	100	8	Sequential	59.25 μs	0.324 μs	0.271 μs	0.1221	-	-	304 B
Write	100	64	Sequential	60.11 μs	0.334 μs	0.296 μs	0.1221	-	-	304 B
Write	100	256	Sequential	71.69 μs	0.870 μs	0.814 μs	0.1221	-	-	304 B
Write	1000	8	Sequential	195.10 μs	0.589 μs	0.492 μs	-	-	-	304 B
Write	1000	64	Sequential	232.84 μs	1.007 μs	0.893 μs	-	-	-	304 B
Write	1000	256	Sequential	25.15 μs	0.030 μs	0.027 μs	0.1221	-	-	304 B

CoreyKaylor commented 1 year ago

(master) net7.0 READ


BenchmarkDotNet=v0.12.1, OS=macOS 13.2 (22D49) [Darwin 22.3.0]
Apple M1 Max, 1 CPU, 10 logical and 10 physical cores
.NET Core SDK=7.0.102
  [Host]     : .NET Core 7.0.2 (CoreCLR 7.0.222.60605, CoreFX 7.0.222.60605), Arm64 RyuJIT
  DefaultJob : .NET Core 7.0.2 (CoreCLR 7.0.222.60605, CoreFX 7.0.222.60605), Arm64 RyuJIT

Method	OpsPerTransaction	ValueSize	KeyOrder	Mean	Error	StdDev	Gen 0	Gen 1	Gen 2	Allocated
Read	1	8	Sequential	210.6 ns	0.71 ns	0.66 ns	0.0484	-	-	304 B
Read	1	64	Sequential	207.3 ns	3.04 ns	2.85 ns	0.0484	-	-	304 B
Read	1	256	Sequential	200.3 ns	2.29 ns	2.03 ns	0.0484	-	-	304 B
Read	100	8	Sequential	8,171.7 ns	18.80 ns	17.59 ns	0.0458	-	-	304 B
Read	100	64	Sequential	8,036.6 ns	17.22 ns	15.26 ns	0.0458	-	-	304 B
Read	100	256	Sequential	8,593.7 ns	16.99 ns	14.19 ns	0.0458	-	-	304 B
Read	1000	8	Sequential	116,029.8 ns	201.47 ns	188.45 ns	-	-	-	304 B
Read	1000	64	Sequential	124,566.1 ns	320.33 ns	299.64 ns	-	-	-	304 B
Read	1000	256	Sequential	132,605.0 ns	401.22 ns	375.30 ns	-	-	-	304 B

CoreyKaylor commented 1 year ago

(master) net7.0 WRITE


BenchmarkDotNet=v0.12.1, OS=macOS 13.2 (22D49) [Darwin 22.3.0]
Apple M1 Max, 1 CPU, 10 logical and 10 physical cores
.NET Core SDK=7.0.102
  [Host]     : .NET Core 7.0.2 (CoreCLR 7.0.222.60605, CoreFX 7.0.222.60605), Arm64 RyuJIT
  DefaultJob : .NET Core 7.0.2 (CoreCLR 7.0.222.60605, CoreFX 7.0.222.60605), Arm64 RyuJIT

Method	OpsPerTransaction	ValueSize	KeyOrder	Mean	Error	StdDev	Gen 0	Gen 1	Gen 2	Allocated
Write	1	8	Sequential	50.44 μs	0.442 μs	0.414 μs	-	-	-	304 B
Write	1	64	Sequential	50.35 μs	0.261 μs	0.256 μs	-	-	-	304 B
Write	1	256	Sequential	50.02 μs	0.354 μs	0.314 μs	-	-	-	304 B
Write	100	8	Sequential	59.23 μs	0.506 μs	0.422 μs	-	-	-	304 B
Write	100	64	Sequential	60.41 μs	1.147 μs	1.017 μs	-	-	-	304 B
Write	100	256	Sequential	71.16 μs	0.668 μs	0.625 μs	-	-	-	304 B
Write	1000	8	Sequential	193.64 μs	0.866 μs	0.723 μs	-	-	-	304 B
Write	1000	64	Sequential	227.80 μs	0.838 μs	0.743 μs	-	-	-	304 B
Write	1000	256	Sequential	24.33 μs	0.057 μs	0.050 μs	0.0305	-	-	304 B

CoreyKaylor commented 1 year ago

(net7-csharp-improvements) net7.0 READ


BenchmarkDotNet=v0.12.1, OS=macOS 13.2 (22D49) [Darwin 22.3.0]
Apple M1 Max, 1 CPU, 10 logical and 10 physical cores
.NET Core SDK=7.0.102
  [Host]     : .NET Core 7.0.2 (CoreCLR 7.0.222.60605, CoreFX 7.0.222.60605), Arm64 RyuJIT
  DefaultJob : .NET Core 7.0.2 (CoreCLR 7.0.222.60605, CoreFX 7.0.222.60605), Arm64 RyuJIT

Method	OpsPerTransaction	ValueSize	KeyOrder	Mean	Error	StdDev	Gen 0	Gen 1	Gen 2	Allocated
Read	1	8	Sequential	211.1 ns	1.32 ns	1.24 ns	0.0484	-	-	304 B
Read	1	64	Sequential	205.5 ns	3.08 ns	2.88 ns	0.0484	-	-	304 B
Read	1	256	Sequential	206.2 ns	2.88 ns	2.69 ns	0.0484	-	-	304 B
Read	100	8	Sequential	7,944.8 ns	25.29 ns	19.74 ns	0.0458	-	-	304 B
Read	100	64	Sequential	7,955.4 ns	19.95 ns	18.66 ns	0.0458	-	-	304 B
Read	100	256	Sequential	8,559.4 ns	17.13 ns	16.02 ns	0.0458	-	-	304 B
Read	1000	8	Sequential	115,440.9 ns	135.02 ns	119.70 ns	-	-	-	304 B
Read	1000	64	Sequential	123,140.9 ns	588.45 ns	550.44 ns	-	-	-	304 B
Read	1000	256	Sequential	131,738.7 ns	431.69 ns	382.68 ns	-	-	-	304 B

CoreyKaylor commented 1 year ago

(net7-csharp-improvements) net7.0 WRITE


BenchmarkDotNet=v0.12.1, OS=macOS 13.2 (22D49) [Darwin 22.3.0]
Apple M1 Max, 1 CPU, 10 logical and 10 physical cores
.NET Core SDK=7.0.102
  [Host]     : .NET Core 7.0.2 (CoreCLR 7.0.222.60605, CoreFX 7.0.222.60605), Arm64 RyuJIT
  DefaultJob : .NET Core 7.0.2 (CoreCLR 7.0.222.60605, CoreFX 7.0.222.60605), Arm64 RyuJIT

Method	OpsPerTransaction	ValueSize	KeyOrder	Mean	Error	StdDev	Gen 0	Gen 1	Gen 2	Allocated
Write	1	8	Sequential	50.71 μs	0.970 μs	0.908 μs	-	-	-	304 B
Write	1	64	Sequential	50.25 μs	0.585 μs	0.489 μs	-	-	-	304 B
Write	1	256	Sequential	50.40 μs	0.574 μs	0.448 μs	-	-	-	304 B
Write	100	8	Sequential	59.59 μs	0.350 μs	0.292 μs	-	-	-	304 B
Write	100	64	Sequential	60.01 μs	0.367 μs	0.325 μs	-	-	-	304 B
Write	100	256	Sequential	71.29 μs	0.936 μs	0.830 μs	-	-	-	304 B
Write	1000	8	Sequential	194.27 μs	3.038 μs	2.693 μs	-	-	-	304 B
Write	1000	64	Sequential	232.27 μs	1.020 μs	0.852 μs	-	-	-	304 B
Write	1000	256	Sequential	24.14 μs	0.042 μs	0.037 μs	0.0305	-	-	304 B

CoreyKaylor / Lightning.NET

net7.0 & C# style improvements #158

(master) net6.0 READ

(master) net6.0 WRITE

(master) net7.0 READ

(master) net7.0 WRITE

(net7-csharp-improvements) net7.0 READ

(net7-csharp-improvements) net7.0 WRITE