chore: share README gate Markdown scanning

scripts/markdown.ts

@@ -5,6 +5,14 @@ import { gfmFromMarkdown } from 'mdast-util-gfm'
 import { gfm } from 'micromark-extension-gfm'
 import type { Nodes } from 'mdast'
 
+/** One authored Markdown line outside fenced code. */
+export interface MarkdownProseLine {
+  /** 1-based source line number. */
+  index: number
+  /** Source text without normalization. */
+  raw: string
+}
+
 /** Parse GitHub-flavored Markdown with the repository's standard extensions. */
 export function parseMarkdown(source: string): Nodes {
   return fromMarkdown(source, { extensions: [gfm()], mdastExtensions: [gfmFromMarkdown()] })
@@ -21,3 +29,27 @@ export function visitMarkdown(node: Nodes, visitor: (node: Nodes) => boolean | v
     for (const child of node.children) visitMarkdown(child, visitor)
   }
 }
+
+/**
+ * Return source lines outside backtick or tilde fences.
+ * @param source - Markdown source whose prose should be retained verbatim.
+ * @returns unfenced lines with their original 1-based locations.
+ */
+export function markdownProseLines(source: string): MarkdownProseLine[] {
+  let fence: { marker: '`' | '~'; length: number } | undefined
+  const kept: MarkdownProseLine[] = []
+  source.split('\n').forEach((raw, i) => {
+    const token = /^ {0,3}(`{3,}|~{3,})/.exec(raw)?.[1]
+    if (token !== undefined) {
+      const marker = token[0] as '`' | '~'
+      if (fence === undefined) {
+        fence = { marker, length: token.length }
+      } else if (marker === fence.marker && token.length >= fence.length) {
+        fence = undefined
+      }
+      return
+    }
+    if (fence === undefined) kept.push({ index: i + 1, raw })
+  })
+  return kept
+}

scripts/verify-package-readme-limitations.ts

@@ -34,6 +34,7 @@
 
 import { existsSync, globSync, readFileSync } from 'node:fs'
 import { resolve } from 'node:path'
+import { markdownProseLines } from './markdown.ts'
 
 const root = resolve(import.meta.dirname, '..')
 
@@ -60,33 +61,8 @@ function isLimitationsLike(headingText: string): boolean {
   )
 }
 
-interface Line {
-  index: number
-  raw: string
-}
-
 const ATX_HEADING = /^ {0,3}#{1,6}[ \t]+/
 
-/** Split a README into prose lines (fenced code dropped), keeping 1-based line numbers. */
-function proseLines(text: string): Line[] {
-  let fence: { marker: '`' | '~'; length: number } | undefined
-  const kept: Line[] = []
-  text.split('\n').forEach((raw, i) => {
-    const token = /^ {0,3}(`{3,}|~{3,})/.exec(raw)?.[1]
-    if (token !== undefined) {
-      const marker = token[0] as '`' | '~'
-      if (fence === undefined) {
-        fence = { marker, length: token.length }
-      } else if (marker === fence.marker && token.length >= fence.length) {
-        fence = undefined
-      }
-      return
-    }
-    if (fence === undefined) kept.push({ index: i + 1, raw })
-  })
-  return kept
-}
-
 const packageJsons = globSync('packages/*/*/package.json', { cwd: root }).sort()
 const scannedPackages = new Set(packageJsons.map(path => path.slice(0, -'/package.json'.length)))
 const failures: string[] = []
@@ -106,7 +82,7 @@ for (const pkg of scannedPackages) {
     failures.push(`${readme}: package manifest has no sibling README with the \`${CANONICAL}\` section`)
     continue
   }
-  const lines = proseLines(readFileSync(resolve(root, readme), 'utf8'))
+  const lines = markdownProseLines(readFileSync(resolve(root, readme), 'utf8'))
   const headings = lines.filter(line => ATX_HEADING.test(line.raw))
   const limitations = headings.filter(line => isLimitationsLike(line.raw.replace(ATX_HEADING, '')))
 

scripts/verify-package-readme-model-experience.ts

@@ -11,6 +11,7 @@
 
 import { existsSync, globSync, readFileSync } from 'node:fs'
 import { relative, resolve } from 'node:path'
+import { markdownProseLines, type MarkdownProseLine } from './markdown.ts'
 
 const root = resolve(import.meta.dirname, '..')
 const HEADING = '## Model Experience'
@@ -69,10 +70,7 @@ interface Failure {
   message: string
 }
 
-interface Line {
-  index: number
-  raw: string
-}
+type Line = MarkdownProseLine
 
 interface ContextSurface {
   heading: Line
@@ -82,26 +80,6 @@ interface ContextSurface {
   verbatimBlocks: number
 }
 
-/** Split Markdown into prose lines, excluding fenced code that may quote the contract. */
-function proseLines(text: string): Line[] {
-  let fence: { marker: '`' | '~'; length: number } | undefined
-  const kept: Line[] = []
-  text.split('\n').forEach((raw, i) => {
-    const token = /^ {0,3}(`{3,}|~{3,})/.exec(raw)?.[1]
-    if (token !== undefined) {
-      const marker = token[0] as '`' | '~'
-      if (fence === undefined) {
-        fence = { marker, length: token.length }
-      } else if (marker === fence.marker && token.length >= fence.length) {
-        fence = undefined
-      }
-      return
-    }
-    if (fence === undefined) kept.push({ index: i + 1, raw })
-  })
-  return kept
-}
-
 /** Validate H4-plus-markdown literals nested after one context surface's fields. */
 function validateNestedVerbatim(raw: readonly string[]): { blocks: number; error?: string } {
   let cursor = 0
@@ -192,7 +170,7 @@ for (const packageJson of packageJsons) {
 
   const text = readFileSync(abs, 'utf8')
   const rawLines = text.split('\n')
-  const lines = proseLines(text)
+  const lines = markdownProseLines(text)
   const h2Headings = lines.filter(line => H2_HEADING.test(line.raw))
   const modelHeadings = h2Headings.filter(line => line.raw === HEADING)
   if (modelHeadings.length !== 1) {