feat(prometheus): add getMetricsRequestHandler-method to Prometheus…

[tapico-weyert]

… exporter

This commit introduces a new method on the PrometheusExporter-class that
allows users of server frameworks with an existing service instance to
call the getMetricsRequestHandler-method as part of a route to return
the collected metrics in Prometheus format.

Fixes #1872

Which problem is this PR solving?

• Allows using an existing server instance to expose the Prometheus endpoint / metrics

Short description of the changes

• Added a new getMetricsRequestHandler-method to the PrometheusExporter-class which calls the internal _exportMetrics-method

[linux-foundation-easycla]

The committers are authorized under a signed CLA.

• ✅ tapico-weyert (d0fefc9)

[codecov]

Codecov Report

Merging #1881 (d0fefc9) into main (d54c5c8) will increase coverage by 0.23%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##             main    #1881      +/-   ##
==========================================
+ Coverage   92.35%   92.58%   +0.23%     
==========================================
  Files         157      174      +17     
  Lines        5104     6032     +928     
  Branches     1085     1278     +193     
==========================================
+ Hits         4714     5585     +871     
- Misses        390      447      +57     

Impacted Files	Coverage Δ
...etry-exporter-prometheus/src/PrometheusExporter.ts	91.04% <100.00%> (-1.27%)	⬇️
...kages/opentelemetry-node/src/NodeTracerProvider.ts	96.55% <0.00%> (-3.45%)	⬇️
...ges/opentelemetry-instrumentation-http/src/http.ts	94.73% <0.00%> (-0.81%)	⬇️
.../opentelemetry-exporter-collector/src/transform.ts	87.85% <0.00%> (ø)
packages/opentelemetry-web/src/types.ts	100.00% <0.00%> (ø)
...s/opentelemetry-instrumentation-fetch/src/fetch.ts	96.00% <0.00%> (ø)
...lemetry-exporter-collector/src/transformMetrics.ts	95.00% <0.00%> (ø)
...kages/opentelemetry-exporter-collector/src/util.ts	100.00% <0.00%> (ø)
...es/opentelemetry-context-zone-peer-dep/src/util.ts	100.00% <0.00%> (ø)
... and 11 more

[MSNev]

Note: I'm not a Prometheus user and have no knowledge of it's inner workings, but I think the name and description of this function is misleading based on what the issue is talking about and looking at the implementation.

Based on m limited understanding it looks like you are really trying to "reuse" the internal pipeline, rather than "return" (i.e. get) the MetricsRequestHandler.

As such I would think (I'll let the Prometheus experts and approvers / maintainers correct me and tell me I'm wrong) the name of the method would better be named as "handleRequest" or "publishMetrics" or something like that based on the way the internal _exportMetrics() function is used from the _requestHandler. There may also be some considerations around the batcher usage if the request is a different URL's...

[tapico-weyert]

Closing this pull request in favour of #1879