Closed akshayknarayan closed 3 years ago
Merging #12 (f2aee57) into master (ddc0b9a) will decrease coverage by
0.26%
. The diff coverage is97.87%
.
@@ Coverage Diff @@
## master #12 +/- ##
==========================================
- Coverage 95.32% 95.05% -0.27%
==========================================
Files 7 7
Lines 1027 1315 +288
==========================================
+ Hits 979 1250 +271
- Misses 48 65 +17
Impacted Files | Coverage Δ | |
---|---|---|
src/lib.rs | 74.10% <ø> (-11.05%) |
:arrow_down: |
src/layer.rs | 86.00% <88.23%> (+0.70%) |
:arrow_up: |
src/builder.rs | 90.90% <100.00%> (+0.90%) |
:arrow_up: |
src/subscriber.rs | 75.75% <100.00%> (+1.60%) |
:arrow_up: |
tests/layer.rs | 100.00% <100.00%> (ø) |
|
tests/subscriber.rs | 100.00% <100.00%> (ø) |
|
src/group.rs | 100.00% <0.00%> (ø) |
|
... and 2 more |
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact)
,ø = not affected
,? = missing data
Powered by Codecov. Last update ddc0b9a...f2aee57. Read the comment docs.
Looks like minimum rust version might need to be increased.
I almost wonder if this should just be on by default — what do you think?
Looks like minimum rust version might need to be increased.
Go for it!
re: default, tracing_subscriber
leaves it off by default: https://docs.rs/tracing-subscriber/0.2.17/tracing_subscriber/fmt/struct.Layer.html#method.with_span_events
I can see the argument that for this crate's purposes it should be on, though, since we're all about timing things (I definitely want this :), and I always set no_span_recursion
). Maybe a v0.4.4
with this change, and a v0.5
changing the default?
I can see the argument that for this crate's purposes it should be on, though, since we're all about timing things (I definitely want this :), and I always set
no_span_recursion
). Maybe av0.4.4
with this change, and av0.5
changing the default?
Love it!
How are we doing on this? I think we are just waiting on feedback re: whether this is a terrible mistake?
Released as 0.4.4 :tada:
tracing-subscriber has support for span-close events (https://docs.rs/tracing-subscriber/0.2.17/tracing_subscriber/fmt/struct.Layer.html#method.with_span_events), but these events aren't available to other layers.
For timing purposes, we already track the time from the start of the span, so this lets us track the time from the last event in a span to its close also. We can also get timing information about a span even if there are no events inside it.