Conversation
|
@sl0thentr0py If we're going to do this, I think the Wdyt? |
|
I spent an hour trying the separation and I think the ideal modeling would be:
But that'll require changing the SDK's modeling standard. So I'll leave it for the SDK team to decide and just add comments about the |
|
@st0012 I kinda agree with you that this inheritance causes coupling and confusion. But the other side is that even in the relay protocols we have a single Just another point to consider, I still haven't made up my mind / thought through completely on what I'd like to have here tbh. |
|
Oh cool. So on the server side it's actually modeled better, but the SDK side's spec and implementation are the "dirty" ones? |
|
Even between SDKs there's a fair amount of inconsistency.
So yes I think basically your comment is the way to go. I'll put it on our backlog but dunno when we'll actually get to it. :) |
|
@sl0thentr0py Since that looks like the right direction, I'd be happy to do it this week. Otherwise, it'd become another debt if we exposed them with all the test helpers. |
|
ok so we do a breaking release then? |
|
Personally I don't think so as users shouldn't have any actual usage to touch From my understanding, the behavioral changes would be
So I don't think it's more like a private API change and doesn't need a breaking release. Wdyt? |
|
sounds good! I'm slightly worried because the specs will also change but we'll just have to review carefully. But you're right that no end-user API is changing. 👍 |
Codecov Report
@@ Coverage Diff @@
## master #1773 +/- ##
==========================================
+ Coverage 98.38% 98.40% +0.01%
==========================================
Files 146 148 +2
Lines 8744 8825 +81
==========================================
+ Hits 8603 8684 +81
Misses 141 141
Continue to review full report at Codecov.
|
56b26da to
07e5082
Compare
|
This pull request has gone three weeks without activity. In another week, I will close it. But! If you comment or otherwise update it, I will reset the clock, and if you label it "A weed is but an unloved flower." ― Ella Wheeler Wilcox 🥀 |
|
@st0012 do you know what we wanna do with this? Should it go out with the next release or do we wait till later? |
|
Since some users start asking for the I like the idea that SDK provides some helpers for testing and encourage users to test the SDK on CI. In the long-term it should reduce the chance for SDK bug to break on production and will also help issue reports. |
sl0thentr0py
left a comment
There was a problem hiding this comment.
@st0012 I merged master and updated some specs, do you want to add a changelog before merging?
|
@sl0thentr0py can we wait for the close api first? I want to refresh the implementation here with it. |
It seems as if changes introduced by getsentry/sentry-ruby#2206 were interfering with Sentry test teardown. Fortunately, getsentry/sentry-ruby#1773 adds some test helpers which manage Sentry test setup/teardown, which resolves this issue.
It seems as if changes introduced by getsentry/sentry-ruby#2206 were interfering with Sentry test teardown. Fortunately, getsentry/sentry-ruby#1773 adds some test helpers which manage Sentry test setup/teardown, which resolves this issue.
It seems as if changes introduced by getsentry/sentry-ruby#2206 were interfering with Sentry test teardown. Fortunately, getsentry/sentry-ruby#1773 adds some test helpers which manage Sentry test setup/teardown, which resolves this issue.
As explained in #1680, the SDK should encourage testing the error capturing logic by providing useful tools to users. So as a first step, we decided to add
Sentry::TestHelperfor the following functionalities:setup_sentry_test- Alters the SDK configuration to for testing. Based on how user sets up the test environment, this may be called once before all tests, or before every test.sentry_events- An easy way to access all the event objects that have been captured.last_sentry_event- Basicallysentry_events.last.teardown_sentry_test- Clears all captured events.Example