fix: Add status exception on Span.recordException

[jufab]

Which problem is this PR solving?

• Span.recordException not set status

Short description of the changes

• Add status on recordException method

[codecov]

Codecov Report

Merging #1490 into master will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master    #1490   +/-   ##
=======================================
  Coverage   93.82%   93.82%           
=======================================
  Files         154      154           
  Lines        4762     4763    +1     
  Branches      951      951           
=======================================
+ Hits         4468     4469    +1     
  Misses        294      294           

Impacted Files	Coverage Δ
packages/opentelemetry-tracing/src/Span.ts	100.00% <100.00%> (ø)

[obecny]

please fix the lint errors (npm run lint:fix), other than that lgtm

[dyladan]

I just looked back at the spec and it doesn't say anything about setting status when recordException is called.

[jufab]

I just looked back at the spec and it doesn't say anything about setting status when recordException is called.

@dyladan you right and i read this one : there is nothing about status.
But, when i try only with this spec, there's no change in backend collector (like jaeger)

but with a status code (13 for INTERNAL but maybe not the good one...)

So i think status code is important for an exception : that's why i do this PR.

@obecny no prob, i try to fix it.

[vmarchaud]

The span.Status API should be removed from the spec fairly soon (open-telemetry/oteps#134), do we want to merge this since it will be removed anyway ?

[jufab]

Perhaps not a removal but a change?

open-telemetry/opentelemetry-specification#706 (comment)

merging it now, maybe, it's not a good idea...

[dyladan]

An exception does not always mean the request failed. It is entirely possible to have a recorded exception which is caught and handled successfully. With span.status being removed or changed soon, I believe we should close this PR.

[jufab]

Ok, i close it ;)