Repository navigation
fix: isolate website limits and scope explorer counts - #920
Conversation
|
@codex review |
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: d4b9fd85e7
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
|
@codex review |
|
Codex Review: Didn't find any major issues. Can't wait for the next one! Reviewed commit: ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
If Codex has suggestions, it will comment; otherwise it will react with 👍. Codex can also answer questions or update the PR. Try commenting "@codex address that feedback". |
Image-heavy project pages can exhaust a visitor's shared API allowance, leaving both assets and API requests rate-limited. Large repositories also lose folder counts when the explorer asks for a repository-wide map exceeding the memory guard.
Give
/w/its own Redis-backed visitor budget, configurable withWEBSITE_RATE_LIMITand defaulting to 2,000 requests per 15 minutes. Keep the existing website slowdown, trusted-proxy identity, access checks, and API limits. Folder counts now load once per directory after its listing, returning only that folder and its immediate children from the active tree. Responses stay bounded and the aggregation has a two-second deadline. Browsing continues without unavailable badges after a count failure, without repeated count retries or presenting loaded entries as the total file count. Legacy callers without a scope retain the existing guarded response.Validation:
Root counts still scan the active subtree in MongoDB; this change bounds results and database execution time rather than making recursive counting constant-time. No production changes or index migration required.