Passing traceids/spanids to library functions and correlating calls

I have a python library which implements Open census logging.

How to pass traceid from main application to the library functions so that all calls from main application can be correlated with library functions? Also, is it possible to see those calls in Application map. Is there some sample code covering this scenario? Do we also need to take care about passing span-ids?

Here is test code:

Wheel Package contains util.py. This is part of a library This contains some functions like

def test():
  logger.info("Test Log")
  with tracer.span(name="span1"):
     test2()

Now this library function is called from some other application main.py

from util import test

logger = logging.getlogger()
log_exporter = AzureExporter(connection_string=app_insights_cs)
tracer = Tracer(exporter=log_exporter, sampler=AlwaysOnSampler()
with tracer.span(name="test"):
   test()
``

I would like correlate calls from main to test() and then test2() function.

[anandchugh]

Thanks @lzchen I was able to resolve this issue by doing like this:

In main.py, first we can created tracer and pass it to util.py In util.py, create tracer using span_context of passed tracer

parent_tracer = <tracer-from-main.py>
tracer = Tracer(
                span_context=parent_tracer.span_context,
                exporter=log_exporter,
                sampler=sampler,
            )

[lzchen]

Is it not possible to just set traceid rather than complete tracer in library?

Yes you can use telemetry_processor to modify traceid before export. https://docs.microsoft.com/en-us/azure/azure-monitor/app/api-filtering-sampling#opencensus-python-telemetry-processors

The field you are looking for is envelope.tags['ai.operation.id'].