fix(library): ignore sentence punctuation after bare internal URLs in check:library-content (#8540)

* fix(library): ignore sentence punctuation after bare internal URLs in check:library-content

* fix(library): only strip trailing punctuation from bare URLs, not explicit link targets

Author: waleedlatif1

scripts/check-library-content.test.ts

@@ -77,6 +77,7 @@ describe('check-library-content', () => {
       [
         '## Overview',
         'See [the guide](/library/kept-guide) and https://www.sim.ai/library/kept-guide#intro.',
+        'Read https://www.sim.ai/library/kept-guide. Then https://www.sim.ai/library/moved-post, too.',
         '![diagram](/library/clean/diagram.png) and [tags](/library/tags) are not posts.',
         'An apex https://sim.ai/library/missing link belongs to check:site-urls.',
         '| a | b |',
@@ -177,6 +178,8 @@ describe('check-library-content', () => {
         '[b](https://www.sim.ai/library/old-guide)',
         '<a href="/blog/moved-post">c</a>',
         '[d](/blog/kept-guide)',
+        'Also see https://www.sim.ai/library/gone.',
+        '[e](/library/kept-guide.) and <a href="https://www.sim.ai/library/kept-guide.">f</a>',
       ].join('\n')
     )
     expect(await findingsFor('library', 'post')).toEqual([
@@ -200,6 +203,23 @@ describe('check-library-content', () => {
         rule: 'internal-link',
         message: '/blog/kept-guide does not exist (no apps/sim/content/blog/kept-guide/index.mdx).',
       },
+      {
+        line: 14,
+        rule: 'internal-link',
+        message: '/library/gone does not exist (no apps/sim/content/library/gone/index.mdx).',
+      },
+      {
+        line: 15,
+        rule: 'internal-link',
+        message:
+          '/library/kept-guide. does not exist (no apps/sim/content/library/kept-guide./index.mdx).',
+      },
+      {
+        line: 15,
+        rule: 'internal-link',
+        message:
+          '/library/kept-guide. does not exist (no apps/sim/content/library/kept-guide./index.mdx).',
+      },
     ])
   })
 

scripts/check-library-content.ts

@@ -76,6 +76,10 @@ export interface PostRef {
  */
 const INTERNAL_LINK =
   /(?:https?:\/\/www\.sim\.ai|(?<=\]\(\s*|href=\{?["'`]))\/(library|blog|customers)\/([^\s)"'`#?/<>\]]+)(\/[^\s)"'`#?<>\]]*)?/g
+/** Sentence punctuation that ends a bare URL in prose (`…see https://www.sim.ai/library/x.`). */
+const TRAILING_PUNCTUATION = /[.,;:!]+$/
+/** Text just before a Markdown link target or `href` value, whose URL ends at its delimiter. */
+const LINK_TARGET_OPENER = /(?:\]\(\s*|href=\{?["'`])$/
 const MARKDOWN_LINK = /\[[^\]\n]*\]\([^)\n]*\)/
 const FAQ_HEADING = /^#{1,6}\s+FAQs?\s*:?\s*$/i
 const CODE_FENCE = /^\s*(```|~~~)/
@@ -330,7 +334,8 @@ export async function checkPost(
     }
     for (const match of text.matchAll(INTERNAL_LINK)) {
       const linkSection = match[1] as Section
-      const target = match[2]
+      const isBareUrl = !LINK_TARGET_OPENER.test(text.slice(0, match.index))
+      const target = isBareUrl ? match[2].replace(TRAILING_PUNCTUATION, '') : match[2]
       const rest = match[3] ?? ''
       // A deeper path is a public asset (`/library/<slug>/cover.jpg`) or a static sub-route.
       if (rest !== '' && rest !== '/') continue