Draft
Conversation
Clippy Allow Annotation ReportComparing clippy allow annotations between branches:
Summary by Rule
Annotation Counts by File
Annotation Stats by Crate
About This ReportThis report tracks Clippy allow annotations for specific rules, showing how they've changed in this PR. Decreasing the number of these annotations generally improves code quality. |
BenchmarksComparisonBenchmark execution time: 2026-03-05 15:34:32 Comparing candidate commit 682eca3 in PR branch Found 0 performance improvements and 0 performance regressions! Performance is the same for 57 metrics, 2 unstable metrics. CandidateCandidate benchmark detailsGroup 1
Group 2
Group 3
Group 4
Group 5
Group 6
Group 7
Group 8
Group 9
Group 10
Group 11
Group 12
Group 13
Group 14
Group 15
Group 16
Group 17
Group 18
Group 19
BaselineOmitted due to size. |
Codecov Report❌ Patch coverage is Additional details and impacted files@@ Coverage Diff @@
## main #1661 +/- ##
==========================================
- Coverage 71.24% 71.22% -0.03%
==========================================
Files 427 428 +1
Lines 62805 63275 +470
==========================================
+ Hits 44748 45065 +317
- Misses 18057 18210 +153
🚀 New features to boost your workflow:
|
Contributor
Artifact Size Benchmark Reportaarch64-alpine-linux-musl
aarch64-unknown-linux-gnu
libdatadog-x64-windows
libdatadog-x86-windows
x86_64-alpine-linux-musl
x86_64-unknown-linux-gnu
|
lloeki
force-pushed
the
lloeki/ruby-trace-exporter
branch
from
81de782 to
682eca3
Compare
Contributor
paullegranddc
left a comment
paullegranddc
left a comment
There was a problem hiding this comment.
The code is a bit verbose but LGTM overall
| // Macros (local copies, matching trace_exporter.rs pattern) | ||
| // --------------------------------------------------------------------------- | ||
|
|
||
| macro_rules! gen_error { |
Contributor
There was a problem hiding this comment.
There are better ways to share the defs than copying the macros, but I guess it doesn't matter if we're going to throw away the code
Contributor
There was a problem hiding this comment.
I have a PR here that move the macros definitions so they can be reused
#1699
Comment on lines
+335
to
+336
| out_low: &mut u64, | ||
| out_high: &mut u64, |
Contributor
There was a problem hiding this comment.
The doc should point out that out_low and out_high should be initialized memory so people don't do this
uint64_t high, low;
// should be this so the refs are initialized
// uint64_t high = 0, low = 0;
ddog_tracer_span_get_trace_id(handle, &low, &high)
Member
Author
There was a problem hiding this comment.
I'm wondering if we should make it a self-contained struct instead:
typedef struct trace_uint128 {
uint64_t low;
uint64_t high;
} trace_uint128_t;
trace_uint128_t trace_id = {0};
ddog_tracer_span_get_trace_id(handle, &trace_id);makes it slightly less likely to misuse.
Comment on lines
+240
to
+242
| // --------------------------------------------------------------------------- | ||
| // Span getters (for reading field values back to C / Ruby) | ||
| // --------------------------------------------------------------------------- |
Member
Author
There was a problem hiding this comment.
I added those for some tests: https://github.com/DataDog/dd-trace-rb/pull/5422/changes#diff-41c54aacdd9cf0c9f8adf9d5610bda1b5abadf1a09e629ffb8ff3578b0b178b3R195
While full black box testing works, these help a lot to identify where and how things break.
From experience with libddwaf which lacked getters for the longest time and needed some gnarly gdb inspection, getters definitely come handy a lot at a minimum for logging/debugging/testing, even when unused in production codepaths.
I do think we might need them in the future in a couple of actual production codepath places but I'll have to check. Not an immediate need though.
Member
There was a problem hiding this comment.
An option I've used in profiling is, since libdatadog is quite fast, do "integration-style" testing -- that is, ask datadog to serialize profiles (in this case would be traces/spans) and then parse the result, rather than exposing the internal state of profiles in libdatadog (for which there are no getters, in many cases).
For instance in https://github.com/DataDog/dd-trace-rb/blob/3e3e699d75ed770dadca08438bfbf20ad530a380/spec/datadog/profiling/collectors/stack_spec.rb#L771-L779 it's made pretty by the helpers but that gathered_stack is the result of going the whole way: add a stack trace into libdatadog, serialize it using protobuf into a pprof file, get the bytes, parse them back, etc.
Member
Author
There was a problem hiding this comment.
that is, ask datadog to serialize profiles (in this case would be traces/spans) and then parse the result, rather than exposing the internal state of profiles in libdatadog (for which there are no getters, in many cases).
That is essentially what the other "full black box testing" test I mentioned is doing, only by way of pushing them over HTTP and then deserialising that ;)
In terms of complexity we're far from internal state of profiles though, given that spans (native or not) are mostly a glorified Hash/Struct: whatever you put in behind key :foo you should be able to get back 1:1 just as you put it in.
It's extremely likely that we will need getters though, as we most probably want native spans to be able to non-breaking-change drop-in replace Ruby spans 1:1 and there are currently read operations happening on spans.
| /// | ||
| /// Returns an error if the slice is not valid UTF-8. | ||
| #[inline] | ||
| fn charslice_to_bytesstring(s: CharSlice) -> Result<BytesString, Box<ExporterError>> { |
Contributor
There was a problem hiding this comment.
Not sure BytesString is the right type here because it's kind of inneficient but this is a refactor for us common component rather than for you
Member
Author
There was a problem hiding this comment.
As long as it doesn't structurally compromise a future evolution I'm fine with using something a bit slower.
Make it work, make it right, make it fast!
lloeki
commented
Mar 10, 2026
| use libdd_common_ffi::CharSlice; | ||
| use libdd_data_pipeline::trace_exporter::TraceExporter; | ||
| use libdd_tinybytes::BytesString; | ||
| use libdd_trace_utils::span::v04::SpanBytes; |
Member
Author
There was a problem hiding this comment.
@paullegranddc Do you feel like libdd-data-pipeline-ffi is the proper location for exposing this tracer.rs API?
I'm not sure libdd_trace_utils is it either as it seems to be "lower level" / "wire oriented" (given the version) so to speak and this TracerSpan looks like a higher level API.
I see libdd-profiling*, maybe there should be some corresponding libdd-tracing?
Thoughts?
Contributor
There was a problem hiding this comment.
The naming of this crate isn't so great, but I don't see a use case for only manipulating spans, without also needing the trace exporter API.
We already have so many crates, so let's not introduce a new one for code that doesn't really need decoupling for now
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What does this PR do?
add necessary bits to build and expose the trace API for Ruby's trace exporter
Motivation
Getting the ball rolling on discussion with actual code
Additional Notes
DO NOT MERGE
How to test the change?
See DataDog/dd-trace-rb#5422