feat:Update LemurBaseParams schema documentation for transcript_ids and input_text #94
Conversation
WalkthroughThe documentation for the Changes
Poem
✨ Finishing Touches🧪 Generate unit tests
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. 🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments. CodeRabbit Commands (Invoked using PR comments)
Other keywords and placeholders
CodeRabbit Configuration File (
|
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=60&v=4){kind=small}
coderabbitai
bot
changed the title
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
⛔ Files ignored due to path filters (1)
src/libs/AssemblyAI/Generated/AssemblyAI.Models.LemurBaseParams.g.csis excluded by!**/generated/**
📒 Files selected for processing (1)
src/libs/AssemblyAI/openapi.yaml(2 hunks)
🔇 Additional comments (1)
src/libs/AssemblyAI/openapi.yaml (1)
5588-5592: Verify that all remaining references to the “100 000 tokens” default were removedThe default context-limit figure was dropped here, but earlier examples / docs might still cite the
100 000value.
Please grep the spec & markdown docs to ensure no stale numbers persist, otherwise users will see conflicting guidance.
| transcript_ids: | ||
| x-label: Transcript IDs | ||
| description: | | ||
| A list of completed transcripts with text. Up to a maximum of 100 files or 100 hours, whichever is lower. | ||
| A list of completed transcripts with text. Up to a maximum of 100 hours. | ||
| Use either transcript_ids or input_text as input into LeMUR. | ||
| type: array |
There was a problem hiding this comment.
💡 Verification agent
🧩 Analysis chain
Description change leaves the array unconstrained – consider adding a formal maxItems or clarifying docs
The previous wording (“100 files or 100 hours”) communicated two independent limits.
After the edit the hours-based limit remains, but the array length is now unspecified.
Because OpenAPI cannot express a duration constraint, clients will infer the only enforced limit is hours and might push thousands of transcript_ids in one call.
If the backend still restricts the list size (e.g. 100 transcripts), add an explicit schema constraint:
items:
x-label: Transcript ID
type: string
format: uuid
+ maxItems: 100 # hard cap retained from the original specAt minimum, extend the sentence to state whether the number of transcripts is unlimited.
Without this, SDKs and generated docs lose an important contract.
🏁 Script executed:
#!/bin/bash
# Locate the transcript_ids definition
grep -R "transcript_ids:" -n src/libs/AssemblyAI/openapi.yaml
# Show the surrounding context of that block
sed -n '5560,5600p' src/libs/AssemblyAI/openapi.yaml
# Check for any existing maxItems constraints in the file
grep -R "maxItems" -n src/libs/AssemblyAI/openapi.yamlLength of output: 1958
Add explicit maxItems to the transcript_ids array
We confirmed that the transcript_ids schema in src/libs/AssemblyAI/openapi.yaml (around lines 5577–5582) lacks a maxItems constraint, even though the description still mentions “100 hours” and the original spec enforced a 100-item cap. Without this, generated clients may assume unlimited IDs.
Please update the schema as follows:
properties:
transcript_ids:
x-label: Transcript IDs
description: |
A list of completed transcripts with text. Up to a maximum of 100 hours.
Use either transcript_ids or input_text as input into LeMUR.
type: array
- items:
+ items:
x-label: Transcript ID
type: string
format: uuid
+ maxItems: 100 # enforce the original 100-transcript limitIf the service truly allows unlimited transcript IDs regardless of count, please clarify in the description instead.
Committable suggestion skipped: line range outside the PR's diff.
🤖 Prompt for AI Agents
In src/libs/AssemblyAI/openapi.yaml around lines 5577 to 5582, the
transcript_ids array schema is missing a maxItems constraint, which should limit
the number of items to 100 as per the original specification. Add a maxItems:
100 property to the transcript_ids array definition to explicitly enforce this
limit and align with the description. If the service actually supports unlimited
transcript IDs, update the description to clarify this instead.
Summary by CodeRabbit