proxy: fix metrics capture for v1/responses

[mostlygeek]

properly parse anthropic compatible usage data from streaming responses.

closes: #577

[coderabbitai]

Walkthrough

The PR modifies metrics monitoring to handle nested usage structures in SSE streaming responses. When parsing streaming data, if usage is not found at the top level, the code now checks for usage under response.usage. Additionally, header parsing in filterAcceptEncoding is improved using safer string operations (strings.SplitSeq and strings.Cut). A test case validates the nested usage parsing behavior.

Changes

Cohort / File(s)	Summary
Streaming Response Metrics Parsing proxy/metrics_monitor.go	Enhanced processStreamingResponse to check for usage under response.usage when not found at top level. Improved filterAcceptEncoding to use strings.SplitSeq and strings.Cut for safer header encoding parsing.
Metrics Test Coverage proxy/metrics_monitor_test.go	Added test case to TestMetricsMonitor_StreamingResponse validating parsing of v1 SSE responses with usage tokens nested under response.usage.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~12 minutes

Possibly related PRs

• Add metrics logging and an Activity page to show requests #195: Introduces the metrics streaming parsing logic that is being extended to support nested usage structures in this PR.
• Fix token metrics parsing #199: Modifies streaming-response metrics parsing to add support for usage under response.usage and enhance streaming payload searching.
• proxy: extract metrics for v1/messages #419: Modifies metrics_monitor.go's streaming metrics parsing to handle nested usage and timing structures in similar ways.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 33.33% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly and concisely summarizes the main change: fixing metrics capture for the v1/responses endpoint by properly parsing usage data.
Description check	✅ Passed	The description is related to the changeset, explaining that it properly parses Anthropic-compatible usage data from streaming responses and references the relevant issue #577.
Linked Issues check	✅ Passed	The PR addresses the primary coding requirement from issue #577: parsing nested Responses usage payloads in streaming metrics by checking response.usage when top-level usage is absent.
Out of Scope Changes check	✅ Passed	The changes are limited to metrics parsing for streaming responses and directly address the issue scope; no unrelated modifications are present in the changeset.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

• 📝 Generate docstrings (stacked PR)
• 📝 Generate docstrings (commit on current branch)

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch mg/responses-streaming-metrics

📝 Coding Plan

• Generate coding plan for human review comments

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

Tip

CodeRabbit can enforce grammar and style rules using `languagetool`.

Configure the reviews.tools.languagetool setting to enable/disable rules and categories. Refer to the LanguageTool Community to learn more.

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Inline comments:
In `@proxy/metrics_monitor.go`:
- Around line 513-515: The code extracts the encoding token with encoding, _, _
:= strings.Cut(strings.TrimSpace(part), ";") but doesn't trim encoding
afterwards, so values like "gzip " fail the supported[strings.ToLower(encoding)]
check; fix by trimming encoding (e.g., encoding = strings.TrimSpace(encoding))
before lowercasing and checking the supported map in the same block where part,
encoding and filtered are handled (referencing the variables encoding, part,
supported and the strings.Cut call).

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 20900357-3e65-4ecf-9afb-eb30a080abe5

📥 Commits

Reviewing files that changed from the base of the PR and between 29a38fd and 3f7be88.

📒 Files selected for processing (2)

• proxy/metrics_monitor.go
• proxy/metrics_monitor_test.go

[coderabbitai]

⚠️ Potential issue | 🟡 Minor

Trim the encoding token after strings.Cut to handle valid OWS before ;q=.

At Line 513, headers like gzip ;q=1.0 leave encoding as "gzip " and fail the supported check.

💡 Proposed fix

-		encoding, _, _ := strings.Cut(strings.TrimSpace(part), ";")
+		encoding, _, _ := strings.Cut(strings.TrimSpace(part), ";")
+		encoding = strings.TrimSpace(encoding)
 		if supported[strings.ToLower(encoding)] {
 			filtered = append(filtered, strings.TrimSpace(part))
 		}

📝 Committable suggestion

‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.

Suggested change

	encoding, _, _ := strings.Cut(strings.TrimSpace(part), ";")
	if supported[strings.ToLower(encoding)] {
	filtered = append(filtered, strings.TrimSpace(part))
	encoding, _, _ := strings.Cut(strings.TrimSpace(part), ";")
	encoding = strings.TrimSpace(encoding)
	if supported[strings.ToLower(encoding)] {
	filtered = append(filtered, strings.TrimSpace(part))
	}

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@proxy/metrics_monitor.go` around lines 513 - 515, The code extracts the
encoding token with encoding, _, _ := strings.Cut(strings.TrimSpace(part), ";")
but doesn't trim encoding afterwards, so values like "gzip " fail the
supported[strings.ToLower(encoding)] check; fix by trimming encoding (e.g.,
encoding = strings.TrimSpace(encoding)) before lowercasing and checking the
supported map in the same block where part, encoding and filtered are handled
(referencing the variables encoding, part, supported and the strings.Cut call).