feat: add Anthropic instrumentation support for TS #618
Conversation
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
koxudaxi
changed the title
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## LILYPAD-209 #618 +/- ##
===============================================
- Coverage 93.74% 88.49% -5.25%
===============================================
Files 229 231 +2
Lines 16332 17757 +1425
Branches 713 763 +50
===============================================
+ Hits 15310 15714 +404
- Misses 1018 2039 +1021
Partials 4 4
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
There was a problem hiding this comment.
There's a lot of code duplication (seems like around 2-4 copies of span creation, message recording, stream chunk processing, etc). I think this will really be difficult to keep track of and maintain as time goes on. I suggest some significant cleanup of the implementation
| AnthropicMessageChunk, | ||
| } from '../types/anthropic'; | ||
|
|
||
| // Import GenAI semantic conventions |
There was a problem hiding this comment.
nit : unnecessary comment
| suppressInternalInstrumentation?: boolean; | ||
| } | ||
|
|
||
| // Symbol to mark instrumented modules |
There was a problem hiding this comment.
nit: probably unnecessary?
| } | ||
|
|
||
| private _applyPatch(moduleExports: any): any { | ||
| // Check if already instrumented by Lilypad |
There was a problem hiding this comment.
More vibe coded comments :P
|
|
||
| // Start span using the OTel way with active context | ||
| const span = tracer.startSpan( | ||
| `anthropic.messages ${params?.model || 'unknown'}`, |
There was a problem hiding this comment.
nit: wonder if it make sense to make the unknown string like unknown-model so it's a little clearer what is unknown?
Comment on lines
+219
to
+233
| { | ||
| kind: SpanKind.CLIENT, | ||
| attributes: { | ||
| [SEMATTRS_GEN_AI_SYSTEM]: 'anthropic', | ||
| [SEMATTRS_GEN_AI_REQUEST_MODEL]: params?.model, | ||
| [SEMATTRS_GEN_AI_REQUEST_TEMPERATURE]: params?.temperature, | ||
| [SEMATTRS_GEN_AI_REQUEST_MAX_TOKENS]: params?.max_tokens, | ||
| [SEMATTRS_GEN_AI_REQUEST_TOP_P]: params?.top_p, | ||
| 'gen_ai.request.top_k': params?.top_k, | ||
| [SEMATTRS_GEN_AI_OPERATION_NAME]: 'chat', | ||
| 'lilypad.type': 'llm', | ||
| 'server.address': 'api.anthropic.com', | ||
| }, | ||
| }, | ||
| activeContext, |
There was a problem hiding this comment.
Looks like this span creation logic is duplicated 3 times in this file, can we refactor to DRY?
Comment on lines
+663
to
+687
| if (chunk.type === 'message_start' && chunk.message) { | ||
| // Initial message with usage info | ||
| if (chunk.message.usage) { | ||
| usage.input_tokens = chunk.message.usage.input_tokens; | ||
| usage.output_tokens = chunk.message.usage.output_tokens; | ||
| } | ||
| } else if (chunk.type === 'content_block_delta' && chunk.delta) { | ||
| // Text content | ||
| if (chunk.delta.type === 'text_delta' && chunk.delta.text) { | ||
| content += chunk.delta.text; | ||
| } | ||
| } else if (chunk.type === 'message_delta' && chunk.delta) { | ||
| // Message delta with stop reason | ||
| if (chunk.delta.stop_reason) { | ||
| stopReason = chunk.delta.stop_reason; | ||
| } | ||
| } else if (chunk.type === 'message_stop' && chunk.usage) { | ||
| // Final usage info | ||
| usage.output_tokens = chunk.usage.output_tokens; | ||
| } | ||
|
|
||
| // Add chunk event | ||
| span.addEvent('gen_ai.chunk', { | ||
| size: chunk.delta?.text?.length || 0, | ||
| }); |
There was a problem hiding this comment.
first copy of chunk processing
Comment on lines
+745
to
+764
| if (chunk.type === 'message_start' && chunk.message) { | ||
| // Initial message with usage info | ||
| if (chunk.message.usage) { | ||
| usage.input_tokens = chunk.message.usage.input_tokens; | ||
| usage.output_tokens = chunk.message.usage.output_tokens; | ||
| } | ||
| } else if (chunk.type === 'content_block_delta' && chunk.delta) { | ||
| // Text content | ||
| if (chunk.delta.type === 'text_delta' && chunk.delta.text) { | ||
| content += chunk.delta.text; | ||
| } | ||
| } else if (chunk.type === 'message_delta' && chunk.delta) { | ||
| // Message delta with stop reason | ||
| if (chunk.delta.stop_reason) { | ||
| stopReason = chunk.delta.stop_reason; | ||
| } | ||
| } else if (chunk.type === 'message_stop' && chunk.usage) { | ||
| // Final usage info | ||
| usage.output_tokens = chunk.usage.output_tokens; | ||
| } |
There was a problem hiding this comment.
second copy of chunk processing
Comment on lines
+600
to
+619
| // Record system message if present | ||
| if (params?.system) { | ||
| span.addEvent('gen_ai.system.message', { | ||
| 'gen_ai.system': 'anthropic', | ||
| content: params.system, | ||
| }); | ||
| } | ||
|
|
||
| // Record messages | ||
| if (params?.messages) { | ||
| params.messages.forEach((message: any) => { | ||
| const content = | ||
| typeof message.content === 'string' ? message.content : JSON.stringify(message.content); | ||
|
|
||
| span.addEvent(`gen_ai.${message.role}.message`, { | ||
| 'gen_ai.system': 'anthropic', | ||
| content: content, | ||
| }); | ||
| }); | ||
| } |
There was a problem hiding this comment.
second copy of message recording logic
Comment on lines
+951
to
+972
| // Record system message if present | ||
| if (params?.system) { | ||
| span.addEvent('gen_ai.system.message', { | ||
| 'gen_ai.system': 'anthropic', | ||
| content: params.system, | ||
| }); | ||
| } | ||
|
|
||
| // Record messages | ||
| if (params?.messages) { | ||
| params.messages.forEach((message: any) => { | ||
| const content = | ||
| typeof message.content === 'string' | ||
| ? message.content | ||
| : JSON.stringify(message.content); | ||
|
|
||
| span.addEvent(`gen_ai.${message.role}.message`, { | ||
| 'gen_ai.system': 'anthropic', | ||
| content: content, | ||
| }); | ||
| }); | ||
| } |
There was a problem hiding this comment.
third copy of message recording logic
Comment on lines
+252
to
+273
| // Record system message if present | ||
| if (params?.system) { | ||
| span.addEvent('gen_ai.system.message', { | ||
| 'gen_ai.system': 'anthropic', | ||
| content: params.system, | ||
| }); | ||
| } | ||
|
|
||
| // Record messages | ||
| if (params?.messages) { | ||
| params.messages.forEach((message: any) => { | ||
| const content = | ||
| typeof message.content === 'string' | ||
| ? message.content | ||
| : JSON.stringify(message.content); | ||
|
|
||
| span.addEvent(`gen_ai.${message.role}.message`, { | ||
| 'gen_ai.system': 'anthropic', | ||
| content: content, | ||
| }); | ||
| }); | ||
| } |
There was a problem hiding this comment.
first copy of message recording logic
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
No description provided.