Skip to content

add on chunk events to genai spans #200

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Oct 16, 2025
Merged

add on chunk events to genai spans #200

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
279e9ba
add on chunk events to genai spans
dinmukhamedm Oct 15, 2025
fa4a934
small update in doc comment
dinmukhamedm Oct 16, 2025
7945365
wrap add event in try except
dinmukhamedm Oct 16, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 8 additions & 0 deletions src/lmnr/opentelemetry_lib/opentelemetry/instrumentation/google_genai/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -380,6 +380,10 @@ def _build_from_streaming_response(
aggregated_usage_metadata = defaultdict(int)
model_version = None
for chunk in response:
try:
span.add_event("llm.content.completion.chunk")
except Exception:
pass
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The try/except block around span.add_event swallows all exceptions. Consider logging the exception (even at a debug level) so that unexpected failures in adding events aren’t silently ignored.

Suggested change
pass
logger.debug("Exception in span.add_event: ", exc_info=True)

# Important: do all processing in a separate sync function, that is
# wrapped in @dont_throw. If we did it here, the @dont_throw on top of
# this function would not be able to catch the errors, as they are
Expand Down Expand Up @@ -430,6 +434,10 @@ async def _abuild_from_streaming_response(
aggregated_usage_metadata = defaultdict(int)
model_version = None
async for chunk in response:
try:
span.add_event("llm.content.completion.chunk")
except Exception:
pass
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Similarly, the async version wraps span.add_event in a try/except without logging. It would be beneficial to log errors for visibility, even if instrumentation should continue.

Suggested change
pass
logger.exception('Failed to add event to span')

# Important: do all processing in a separate sync function, that is
# wrapped in @dont_throw. If we did it here, the @dont_throw on top of
# this function would not be able to catch the errors, as they are
Expand Down
6 changes: 3 additions & 3 deletions src/lmnr/sdk/laminar.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -671,12 +671,12 @@ def set_span_attributes(
instrumentation.
Example:
```python
with L.start_as_current_span(
with Laminar.start_as_current_span(
name="my_span_name", input=input["messages"], span_type="LLM"
):
response = await my_custom_call_to_openai(input)
L.set_span_output(response["choices"][0]["message"]["content"])
L.set_span_attributes({
Laminar.set_span_output(response["choices"][0]["message"]["content"])
Laminar.set_span_attributes({
Attributes.PROVIDER: 'openai',
Attributes.REQUEST_MODEL: input["model"],
Attributes.RESPONSE_MODEL: response["model"],
Expand Down
4 changes: 4 additions & 0 deletions tests/test_google_genai.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -847,8 +847,10 @@ def test_google_genai_streaming(span_exporter: InMemorySpanExporter):
],
)
final_response = ""
chunk_count = 0
for chunk in stream:
final_response += chunk.text or ""
chunk_count += 1

spans = span_exporter.get_finished_spans()
assert len(spans) == 1
Expand Down Expand Up @@ -885,6 +887,8 @@ def test_google_genai_streaming(span_exporter: InMemorySpanExporter):
assert span.attributes["gen_ai.usage.input_tokens"] == 7
assert span.attributes["gen_ai.usage.output_tokens"] == 166
assert span.attributes["llm.usage.total_tokens"] == 175 # 173 + 2 (thinking tokens)
assert len(span.events) == chunk_count
assert all(event.name == "llm.content.completion.chunk" for event in span.events)


@pytest.mark.vcr
Expand Down