Introduce service naming framework

[jbertran]

What does this PR do?

This PR:

• Adds interaction with the DD_TRACE_SPAN_ATTRIBUTE_SCHEMA environment variable
• Provides the framework for 2 versions of the service and operation naming schema:
  • v0 is the current version, i.e. a representation of the existing state of the plugin. One could checkout the commit with the introduction of this version, and have unchanged tests still pass.
  • v1 is the new naming schema, made consistent across all tracers.
• Introduces a testing harness for easy tests of v0 and v1 naming
• Applies this versioned naming for all messaging integrations - rhea, amqp(10|lib), kafkajs, google-cloud-pubsub

Motivation

This is part of the service representation initiative, which provides in an opt-in way a new manner of naming services and operations, and will introduce peer.service soon.

The opt-in nature of the change guarantee that we do not break naming for existing deployments.

[github-actions]

Overall package size

Self size: 4 MB
Deduped: 64.22 MB
No deduping: 64.26 MB

Dependency sizes

name	version	self size	total size
@datadog/native-iast-taint-tracking	1.4.0	14.89 MB	14.9 MB
@datadog/pprof	2.2.0	13.71 MB	14.59 MB
@datadog/native-appsec	3.1.0	13.31 MB	13.32 MB
@datadog/native-metrics	1.6.0	7.88 MB	7.89 MB
protobufjs	7.1.2	2.76 MB	6.55 MB
@datadog/native-iast-rewriter	2.0.1	2.09 MB	2.1 MB
opentracing	0.14.7	194.81 kB	194.81 kB
semver	7.3.8	88.2 kB	118.6 kB
@datadog/sketches-js	2.1.0	109.9 kB	109.9 kB
lodash.sortby	4.7.0	75.76 kB	75.76 kB
lru-cache	7.14.0	74.95 kB	74.95 kB
ipaddr.js	2.0.1	59.52 kB	59.52 kB
ignore	5.2.0	48.87 kB	48.87 kB
import-in-the-middle	1.3.5	34.34 kB	38.81 kB
istanbul-lib-coverage	3.2.0	29.34 kB	29.34 kB
retry	0.10.1	27.44 kB	27.44 kB
lodash.uniq	4.5.0	25.01 kB	25.01 kB
limiter	1.1.5	23.17 kB	23.17 kB
lodash.kebabcase	4.1.1	17.75 kB	17.75 kB
lodash.pick	4.4.0	16.33 kB	16.33 kB
node-abort-controller	3.0.1	14.33 kB	14.33 kB
crypto-randomuuid	1.0.0	11.18 kB	11.18 kB
diagnostics_channel	1.1.0	7.07 kB	7.07 kB
path-to-regexp	0.1.7	6.78 kB	6.78 kB
koalas	1.0.2	6.47 kB	6.47 kB
methods	1.1.2	5.29 kB	5.29 kB
module-details-from-path	1.0.3	4.47 kB	4.47 kB

_{🤖 This report was automatically generated by heaviest-objects-in-the-universe}

[codecov]

Codecov Report

Merging #2941 (4c0e856) into master (a0c2514) will decrease coverage by 0.30%.
The diff coverage is 52.12%.

@@            Coverage Diff             @@
##           master    #2941      +/-   ##
==========================================
- Coverage   87.19%   86.89%   -0.30%     
==========================================
  Files         317      304      -13     
  Lines       11512    10791     -721     
  Branches       33       33              
==========================================
- Hits        10038     9377     -661     
+ Misses       1474     1414      -60     

Impacted Files	Coverage Δ
packages/dd-trace/src/service-naming/schemas/v0.js	13.04% <13.04%> (ø)
packages/dd-trace/src/plugins/tracing.js	57.14% <20.00%> (-6.20%)	⬇️
.../dd-trace/src/service-naming/schemas/definition.js	30.00% <30.00%> (ø)
packages/dd-trace/src/service-naming/schemas/v1.js	35.71% <35.71%> (ø)
packages/dd-trace/src/plugins/client.js	50.00% <50.00%> (-16.67%)	⬇️
packages/dd-trace/src/service-naming/index.js	66.66% <66.66%> (ø)
packages/datadog-plugin-amqp10/src/consumer.js	88.88% <100.00%> (ø)
packages/datadog-plugin-amqp10/src/producer.js	90.90% <100.00%> (ø)
packages/datadog-plugin-amqplib/src/client.js	100.00% <100.00%> (ø)
packages/datadog-plugin-amqplib/src/consumer.js	100.00% <100.00%> (ø)
... and 12 more

... and 18 files with indirect coverage changes

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[jbertran]

Ongoing to-do list:

• Minimize diff in plugin source
• Rename incoming/outgoing to match naming terminology (Inbound/Outbound)
• Verify if plugins all currently fit into the plugin class inheritance tree (and if so, use instance class instead of instance methods to resolve naming)
  • No, even if some of them are trivial to make fit into the model (see HTTP for example)
• Investigate ways to minimize test diff (too much indentation because we wrap the existing tests)
• Pass service as a positional argument in service name resolution
  • Easy, but we don't go further than this (no trickery to freeze service name in the schema and prevent us from passing it altogether)
• Investigate if moving to a literal is possible instead of a function in operation name resolution
  • Not in the current state of at least the couchbase plugin

[pr-commenter]

Benchmarks

Comparing candidate commit 4c0e856 in PR branch jbertran/service-naming-framework with baseline commit a0c2514 in branch master.

Found 0 performance improvements and 28 performance regressions! Performance is the same for 639 metrics, 41 unstable metrics.

scenario:log-with-debug-14

• 🟥 cpu_user_time [+7.634ms; +10.893ms] or [+5.656%; +8.071%]
• 🟥 execution_time [+10.168ms; +11.128ms] or [+6.495%; +7.108%]

scenario:log-with-error-14

• 🟥 cpu_user_time [+8.511ms; +11.487ms] or [+6.298%; +8.499%]
• 🟥 execution_time [+9.899ms; +10.868ms] or [+6.314%; +6.932%]

scenario:log-skip-log-14

• 🟥 cpu_user_time [+8.104ms; +11.315ms] or [+5.979%; +8.348%]
• 🟥 execution_time [+10.726ms; +11.641ms] or [+6.846%; +7.430%]

scenario:log-without-log-14

• 🟥 cpu_user_time [+8.886ms; +12.292ms] or [+6.581%; +9.102%]
• 🟥 execution_time [+9.980ms; +10.976ms] or [+6.343%; +6.976%]

scenario:startup-with-tracer-14

• 🟥 execution_time [+10.279ms; +11.172ms] or [+6.630%; +7.206%]

scenario:log-without-log-16

• 🟥 cpu_user_time [+8.669ms; +11.831ms] or [+6.316%; +8.620%]
• 🟥 execution_time [+10.562ms; +11.058ms] or [+6.573%; +6.881%]

scenario:log-with-debug-16

• 🟥 cpu_user_time [+7.644ms; +10.731ms] or [+5.574%; +7.825%]
• 🟥 execution_time [+10.015ms; +11.270ms] or [+6.243%; +7.025%]

scenario:log-with-error-16

• 🟥 cpu_user_time [+7.783ms; +10.999ms] or [+5.673%; +8.017%]
• 🟥 execution_time [+9.662ms; +10.777ms] or [+6.008%; +6.701%]

scenario:log-skip-log-16

• 🟥 cpu_user_time [+7.160ms; +10.450ms] or [+5.218%; +7.615%]
• 🟥 execution_time [+9.863ms; +11.190ms] or [+6.126%; +6.951%]

scenario:startup-with-tracer-16

• 🟥 cpu_user_time [+9.955ms; +17.076ms] or [+7.406%; +12.703%]
• 🟥 execution_time [+10.071ms; +10.750ms] or [+6.354%; +6.783%]

scenario:log-with-debug-18

• 🟥 cpu_user_time [+8.164ms; +11.634ms] or [+5.866%; +8.359%]
• 🟥 execution_time [+10.148ms; +11.490ms] or [+6.140%; +6.952%]

scenario:log-skip-log-18

• 🟥 cpu_user_time [+7.454ms; +10.923ms] or [+5.343%; +7.830%]
• 🟥 execution_time [+10.605ms; +11.385ms] or [+6.410%; +6.882%]

scenario:log-with-error-18

• 🟥 cpu_user_time [+8.193ms; +11.511ms] or [+5.904%; +8.296%]
• 🟥 execution_time [+10.017ms; +11.027ms] or [+6.053%; +6.663%]

scenario:log-without-log-18

• 🟥 cpu_user_time [+8.202ms; +11.791ms] or [+5.902%; +8.485%]
• 🟥 execution_time [+9.985ms; +11.293ms] or [+6.004%; +6.790%]

scenario:startup-with-tracer-18

• 🟥 execution_time [+11.659ms; +12.246ms] or [+7.167%; +7.528%]

[jbertran]

LGTM, but the benchmarks seem to disagree. Let's make sure that there is no regression but then it's fine to merge 👍

The benchmark workflows were resource-starved, rerunning was enough.