feat(common): add priority option on middleware decorator

[Romakita]

Closes: #2516

[github-actions]

Benchmarks

• Machine: linux x64 | 4 vCPUs | 15.6GB Mem
• Node: v16.20.2
• Run: Wed Nov 15 2023 08:19:39 GMT+0000 (Coordinated Universal Time)
• Method: autocannon -c 100 -d 10 -p 10 localhost:3000 (two rounds; one to warm-up, one to measure)

	Version	Router	Requests/s	Latency	Throughput/Mb
fastify	3.29.4	✓	55545.6	17.51	9.90
nest-fastify	8.4.3	✓	46611.2	20.95	8.31
koa	2.13.4	✗	40288.0	24.33	7.18
fastify-injector	3.29.4	✓	28044.4	35.12	6.39
express	4.18.1	✓	12246.2	80.95	2.18
tsed-koa	7.44.1	✓	11615.2	85.39	10.14
fastify-big-json	3.29.4	✓	11104.8	89.27	127.76
nest	8.4.3	✓	10585.1	93.54	2.54
express-injector	4.18.1	✗	10199.4	97.18	2.33
tsed-express	7.44.1	✓	9527.1	103.96	1.74
express-morgan	4.18.1	✗	7332.0	135.09	1.31

Explanation

The benchmark shows a performance difference between the frameworks. We note that Ts.ED is often last. In fact, Ts.ED uses features useful to a production application which reduce its performance.

For example, Ts.ED initializes a sandbox (async_hook) for each request in order to work in an isolated context if necessary.
It also initializes the elements necessary for monitoring requests in a log manager.

All this at a necessary cost that reflects the reality of a production application ;)

[EinfachHans]

Looks understandable for me. A priority to define the middleware order 👍🏼

Question: What happens to the priority when i override a middleware using OverrideProvider?

Only thing that i would add, would be some docs? About the priority option in general and about the default priorities set by this project (-10 for PlatformAcceptMimesMiddleware and 10 for PlatformMulterMiddleware).

[github-actions]

Benchmarks

• Machine: linux x64 | 4 vCPUs | 15.6GB Mem
• Node: v16.20.2
• Run: Thu Nov 16 2023 07:55:59 GMT+0000 (Coordinated Universal Time)
• Method: autocannon -c 100 -d 10 -p 10 localhost:3000 (two rounds; one to warm-up, one to measure)

	Version	Router	Requests/s	Latency	Throughput/Mb
fastify	3.29.4	✓	54544.0	17.85	9.73
nest-fastify	8.4.3	✓	49376.0	19.75	8.81
koa	2.13.4	✗	38316.8	25.60	6.83
fastify-injector	3.29.4	✓	27680.7	35.57	6.31
express	4.18.1	✓	12334.2	80.36	2.20
tsed-koa	7.44.1	✓	11821.5	83.83	10.33
fastify-big-json	3.29.4	✓	11344.0	87.40	130.53
nest	8.4.3	✓	10607.6	93.56	2.55
express-injector	4.18.1	✗	10283.3	96.46	2.34
tsed-express	7.44.1	✓	9478.7	104.58	1.74
express-morgan	4.18.1	✗	7344.0	135.05	1.31

Explanation

The benchmark shows a performance difference between the frameworks. We note that Ts.ED is often last. In fact, Ts.ED uses features useful to a production application which reduce its performance.

For example, Ts.ED initializes a sandbox (async_hook) for each request in order to work in an isolated context if necessary.
It also initializes the elements necessary for monitoring requests in a log manager.

All this at a necessary cost that reflects the reality of a production application ;)

[github-actions]

Benchmarks

• Machine: linux x64 | 4 vCPUs | 15.6GB Mem
• Node: v16.20.2
• Run: Thu Nov 16 2023 08:39:25 GMT+0000 (Coordinated Universal Time)
• Method: autocannon -c 100 -d 10 -p 10 localhost:3000 (two rounds; one to warm-up, one to measure)

	Version	Router	Requests/s	Latency	Throughput/Mb
fastify	3.29.4	✓	56182.4	17.30	10.02
nest-fastify	8.4.3	✓	48809.6	19.98	8.71
koa	2.13.4	✗	39299.2	24.95	7.01
fastify-injector	3.29.4	✓	27993.6	35.19	6.38
express	4.18.1	✓	12452.0	79.64	2.22
tsed-koa	7.45.0	✓	11957.5	82.88	10.44
fastify-big-json	3.29.4	✓	11654.2	85.04	134.09
nest	8.4.3	✓	10983.6	90.15	2.64
express-injector	4.18.1	✗	10624.0	93.33	2.42
tsed-express	7.45.0	✓	9544.2	103.92	1.75
express-morgan	4.18.1	✗	7198.4	137.42	1.28

Explanation

The benchmark shows a performance difference between the frameworks. We note that Ts.ED is often last. In fact, Ts.ED uses features useful to a production application which reduce its performance.

For example, Ts.ED initializes a sandbox (async_hook) for each request in order to work in an isolated context if necessary.
It also initializes the elements necessary for monitoring requests in a log manager.

All this at a necessary cost that reflects the reality of a production application ;)

[github-actions]

Benchmarks

• Machine: linux x64 | 4 vCPUs | 15.6GB Mem
• Node: v16.20.2
• Run: Thu Nov 16 2023 08:57:22 GMT+0000 (Coordinated Universal Time)
• Method: autocannon -c 100 -d 10 -p 10 localhost:3000 (two rounds; one to warm-up, one to measure)

	Version	Router	Requests/s	Latency	Throughput/Mb
fastify	3.29.4	✓	54965.8	17.70	9.80
nest-fastify	8.4.3	✓	49017.6	19.90	8.74
koa	2.13.4	✗	40108.8	24.42	7.15
fastify-injector	3.29.4	✓	28636.4	34.38	6.53
express	4.18.1	✓	12002.4	82.44	2.14
tsed-koa	7.45.0	✓	11950.9	82.82	10.44
fastify-big-json	3.29.4	✓	11563.3	85.75	133.01
nest	8.4.3	✓	11093.6	89.28	2.67
express-injector	4.18.1	✗	10742.9	92.31	2.45
tsed-express	7.45.0	✓	9526.7	104.11	1.74
express-morgan	4.18.1	✗	7269.5	136.37	1.30

Explanation

The benchmark shows a performance difference between the frameworks. We note that Ts.ED is often last. In fact, Ts.ED uses features useful to a production application which reduce its performance.

For example, Ts.ED initializes a sandbox (async_hook) for each request in order to work in an isolated context if necessary.
It also initializes the elements necessary for monitoring requests in a log manager.

All this at a necessary cost that reflects the reality of a production application ;)

[github-actions]

Benchmarks

• Machine: linux x64 | 4 vCPUs | 15.6GB Mem
• Node: v16.20.2
• Run: Thu Nov 16 2023 09:07:40 GMT+0000 (Coordinated Universal Time)
• Method: autocannon -c 100 -d 10 -p 10 localhost:3000 (two rounds; one to warm-up, one to measure)

	Version	Router	Requests/s	Latency	Throughput/Mb
fastify	3.29.4	✓	56297.6	17.26	10.04
nest-fastify	8.4.3	✓	48656.0	20.06	8.68
koa	2.13.4	✗	40403.2	24.25	7.21
fastify-injector	3.29.4	✓	28934.5	34.03	6.59
express	4.18.1	✓	12368.0	80.16	2.21
tsed-koa	7.45.0	✓	11999.2	82.54	10.48
fastify-big-json	3.29.4	✓	11696.8	84.69	134.58
nest	8.4.3	✓	10811.3	91.75	2.60
express-injector	4.18.1	✗	10580.8	93.74	2.41
tsed-express	7.45.0	✓	10165.0	97.60	1.86
express-morgan	4.18.1	✗	7296.4	135.87	1.30

Explanation

The benchmark shows a performance difference between the frameworks. We note that Ts.ED is often last. In fact, Ts.ED uses features useful to a production application which reduce its performance.

For example, Ts.ED initializes a sandbox (async_hook) for each request in order to work in an isolated context if necessary.
It also initializes the elements necessary for monitoring requests in a log manager.

All this at a necessary cost that reflects the reality of a production application ;)

[Romakita]

🎉 This PR is included in version 7.46.0 🎉

The release is available on:

• v7.46.0
• GitHub release

Your semantic-release bot 📦🚀