You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This is mostly for the test suite as currently the experience is not great (especially when a problem chain involves dummy-central).
Rough idea:
have a fixture generate a trace id (per test) if enabled
do magic so capsys & co output their stuff to tracing (?)
make env pass that to the Environment, update Environment to add the trace id as a header to xmlrpc (ServerProxy(header) support was added in 3.8)
mangle logging to support tracing output (otel apparently has some integration with logging)
similarly with make_repo and the github session
update the mergebot/forwardbot to forward the trace id when it calls github (can just get it from the request headers)
update DC to take up that trace id and add it to its tracing context
find a way to collate all the tracing stuff back into pytest's stream capture things, so it's all displayed nice and tight in case of failure (see CaptureManager, possibly terminal_reporter, not entirely sure how to make pytest retrieve the tracing data, most likely all three processes need some config to sync on)
The text was updated successfully, but these errors were encountered:
This is mostly for the test suite as currently the experience is not great (especially when a problem chain involves dummy-central).
Rough idea:
env
pass that to theEnvironment
, updateEnvironment
to add the trace id as a header to xmlrpc (ServerProxy(header)
support was added in 3.8)make_repo
and the github sessionThe text was updated successfully, but these errors were encountered: