[receiver/windowseventlog] Add suppress_rendering_info parameter and simplify internal logic.#34720
Conversation
pjanotti
left a comment
There was a problem hiding this comment.
Looks to be going in the right direction.
6c24134 to
c22a815
Compare
1372b67 to
1560b54
Compare
edc9933 to
28b81d0
Compare
|
@djaglowski ping me when you think this one is ready (or if you want another pass while still a draft) |
|
Thanks @pjanotti, I'm still just trying to get it to pass CI. There have been a bunch of unrelated tests failing but now it's just https://github.com/open-telemetry/opentelemetry-collector-contrib/actions/runs/10706954934/job/29685874116?pr=34720#step:6:556, which may or may not be related. I don't know if you have any insight on that one but if you do it would be very helpful. |
|
@djaglowski there is something really going on with these tests, when I submitted the PR with the overall changes to the tests I wasn't getting any failure on my machine, but, now it seems almost constant. Nothing conclusive yet but will be looking more deeply into it. |
I noticed on #34720 and #35026 that execution of the test continued beyond a failure of `require.EventuallyWithT`. Based on the description alone, I would expect that using `assert` within `require.EventuallyWithT` should cause execution to stop if the assertion fails, but it appears this may not be the case. However, this change apparently works as intended.
This is a step towards #34720 which should have no user facing impact. It consolidates the event model by removing the notion of a "raw" event. The behavior of `raw` flag should remain the same because we still use different functions to build the output from the unified event.
…simplify internal logic.
1570f19 to
f69210b
Compare
f69210b to
e4a6b3d
Compare
|
@pjanotti I believe this is ready for a review now |
f31a247 to
177a7a5
Compare
177a7a5 to
b53999b
Compare
180311c to
8bb2bf5
Compare
…y#35032) I noticed on open-telemetry#34720 and open-telemetry#35026 that execution of the test continued beyond a failure of `require.EventuallyWithT`. Based on the description alone, I would expect that using `assert` within `require.EventuallyWithT` should cause execution to stop if the assertion fails, but it appears this may not be the case. However, this change apparently works as intended.
This is a step towards open-telemetry#34720 which should have no user facing impact. It consolidates the event model by removing the notion of a "raw" event. The behavior of `raw` flag should remain the same because we still use different functions to build the output from the unified event.
…simplify internal logic. (open-telemetry#34720) **Description:** This PR contains several changes described in open-telemetry#34131. It does not go as far as breaking out a separate parsing component, but I think it is enough to satisfy the known use cases. - Add `suppress_rendering_info` parameter, which acts orthogonally to `raw` flag. - Remove `RemoteServer` field from `EventXML`. Instead, set `attributes["remote_server"]` if remote collection is used. **Link to tracking Issue:** Resolves open-telemetry#34131
…metry#35461) This PR contains some cleanup following open-telemetry#34720. This slightly changes some log messages. It also propagates errors encountered by sending to the next operator (which was done for other operators in open-telemetry#33847).
Description:
This PR contains several changes described in #34131. It does not go as far as breaking out a separate parsing component, but I think it is enough to satisfy the known use cases.
suppress_rendering_infoparameter, which acts orthogonally torawflag.RemoteServerfield fromEventXML. Instead, setattributes["remote_server"]if remote collection is used.Link to tracking Issue:
Resolves #34131