Repository navigation
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
This branch was successfully deployed
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Note
On hold, evaluating a merge with the existing Markdown fidelity checker.
Option: keep only the small
onDropComponenthook from this PR and use it to feed the fidelity checker's unsupported-component list from the actual transform. Today, that list drifts from the transform (for example, it still reportsDocCardsandFeedPageas unsupported) and cannot see some dropped elements (img,em,CopyText).Description
Unhandled JSX components are silently removed when MDX pages are converted to the LLM TXT bundles (
llms-full*.txt). This PR adds a diagnostic that reports which components are dropped, on which pages, and roughly how much text is lost. It does not change the generated Markdown.[TEMP - to be addressed ASAP] The first report already shows a significant gap:
MethodTabsremoves about 91K characters (66 occurrences on 21 ACE/CREC pages), so the UI and API procedures of these guides are missing from the bundles. A separate PR will fix that.Changes
transformMarkdownrecords each unhandled component it removes (name and approximate text length). Only successful transforms are counted. Go/TypeScript variants of a page count once (per-page maximum).npm run generate:llmswritesreports/component-coverage.json(ignored by Git). Invalidate-llms, components missing fromsrc/scripts/component-coverage-baseline.jsonproduce non-blocking warnings and a job summary listing the affected pages. The report is uploaded as an artifact.npm run report:component-coverageregeneratespublic/reports/component-coverage.jsonwithout rewriting the TXT files. Run it weekly from an up-to-datemainand commit the file alone in a dedicated PR. Content PRs never touch it, and the build pipeline is unchanged.generate-llmsnow rejects unknown options, missing option values, and unknown sections, and setsprocess.exitCodeinstead of callingprocess.exit..gitignorenow anchorsreports/to the repository root, so the public report can be versioned.validate-llmsjob also runs on LLM-related config and dependency changes, and its final checks run even if regeneration fails.Limitation:
approxCharsonly measures text inside the removed MDX node. Self-closing components (for example<Common callout="…" />) report 0 even when their rendered content is lost.Validation
src/lib/markdown,src/scripts), typecheck, ESLint, and Prettier pass.report:component-coveragetakes about 15 s locally. A failed run exits with a non-zero code and keeps the previous public report.