Implement lilypad.configure({ auto_llm: true }) for OpenAI in Lilypad TypeScript SDK #601
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
…ilypad TypeScript SDK
…e unused EventSummaryResponse
… with dom.iterable
…lidation constants
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## update-typescript-sdk #601 +/- ##
==========================================================
- Coverage 99.92% 89.74% -10.18%
==========================================================
Files 190 211 +21
Lines 10115 13372 +3257
Branches 0 226 +226
==========================================================
+ Hits 10107 12001 +1894
- Misses 8 1370 +1362
- Partials 0 1 +1
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:
|
…es, and enhance span type definitions
…lypad TypeScript SDK
…variable handling
…variable handling
… and error handling
…ling and module export checks
koxudaxi
commented
Jul 9, 2025
Comment on lines
1
to
61
| # AutoLLM Import Order Constraint | ||
|
|
||
| ## Important Limitation | ||
|
|
||
| When using the `autoLlm: true` option, there is a critical constraint regarding module import order: | ||
|
|
||
| ### ❌ This will NOT work: | ||
| ```typescript | ||
| import OpenAI from "openai"; // OpenAI imported BEFORE configure() | ||
| import { configure } from "lilypad"; | ||
|
|
||
| await configure({ | ||
| autoLlm: true // Hook cannot intercept already-loaded module | ||
| }); | ||
|
|
||
| const client = new OpenAI(); // Will NOT be instrumented | ||
| ``` | ||
|
|
||
| ### ✅ This WILL work: | ||
| ```typescript | ||
| import { configure } from "lilypad"; | ||
|
|
||
| // Configure FIRST | ||
| await configure({ | ||
| autoLlm: true | ||
| }); | ||
|
|
||
| // Import OpenAI AFTER configure() | ||
| const { default: OpenAI } = await import("openai"); | ||
| const client = new OpenAI(); // Will be instrumented correctly | ||
| ``` | ||
|
|
||
| ## Technical Explanation | ||
|
|
||
| The `autoLlm` feature uses Node.js `Module._load` hooks to intercept and wrap the OpenAI module at load time. This approach has an inherent limitation: | ||
|
|
||
| - **Module._load hooks only work when a module is loaded for the first time** | ||
| - If the module is already in the require cache, the hook will not be triggered | ||
| - This is a fundamental constraint of Node.js module system, not specific to our implementation | ||
|
|
||
| ## Recommended Approaches | ||
|
|
||
| ### 1. Use Dynamic Import (Most Flexible) | ||
| ```typescript | ||
| await configure({ autoLlm: true }); | ||
| const { default: OpenAI } = await import("openai"); | ||
| ``` | ||
|
|
||
| ### 2. Use --require Flag (Most Reliable) | ||
| ```bash | ||
| node --require ./dist/register.js your-app.js | ||
| ``` | ||
| This ensures hooks are installed before any application code runs. | ||
|
|
||
| ### 3. Manual Wrapping (Most Control) | ||
| ```typescript | ||
| import OpenAI from "openai"; | ||
| import { wrapOpenAI } from "lilypad"; | ||
|
|
||
| const client = wrapOpenAI(new OpenAI()); | ||
| ``` |
There was a problem hiding this comment.
@brenkao @willbakst
Unlike Python, there appear to be limitations when applying patches (module hooks). If you know of a solution, please share it.
We should also decide and document which approach to recommend as a library before publishing the package. Related to this, I've discovered that things become even more complicated when parent spans exist, so I'll comment on that in a separate PR.
…environment variable cleanup
feat: add span and session functionality with type definitions
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.