fix(connectors): address review findings on listing and hashing

- microsoft-excel: `fetchWorksheets` read only the first Graph page and never
  followed `@odata.nextLink`. A workbook with more sheets than fit in one page
  dropped the remainder from the listing without setting `listingCapped`, so
  the sync engine reconciled those documents away as deleted. The walk now
  pages, bounded by MAX_WORKSHEETS, and only follows a nextLink that stays on
  the Graph origin, since the link is server-supplied and carries the token.

- google-slides: the listing `contentHash` covered only the file id and
  modified time, so toggling the speaker-notes option left every stored hash
  matching and no presentation was ever re-hydrated with the new scope. The
  setting is now part of the hash, in the single shared stub builder so the
  list and hydrate paths stay identical.

- mintlify: `pathPrefix` filtered with a bare `startsWith`, so a prefix of
  `/guides` also matched a sibling like `/guides-archive`. It now shares the
  `/`-boundary rule `withinBasePath` already used, extracted as `isUnderPath`.

Author: waleedlatif1

apps/sim/connectors/google-slides/google-slides.ts

@@ -218,15 +218,20 @@ function shouldIncludeSpeakerNotes(sourceConfig: Record<string, unknown>): boole
  * Creates a lightweight stub from a Drive file entry. Content is deferred
  * and only fetched via getDocument for new or changed documents.
  */
-function fileToStub(file: DriveFile): ExternalDocument {
+function fileToStub(file: DriveFile, includeSpeakerNotes: boolean): ExternalDocument {
   return {
     externalId: file.id,
     title: file.name || 'Untitled',
     content: '',
     contentDeferred: true,
     mimeType: 'text/plain',
     sourceUrl: file.webViewLink || `https://docs.google.com/presentation/d/${file.id}/edit`,
-    contentHash: `gslides:${file.id}:${file.modifiedTime ?? ''}`,
+    /**
+     * The speaker-notes setting selects what the rendered content contains, so it
+     * belongs in the hash. Without it, toggling the option leaves every stored
+     * hash matching and no presentation is ever re-hydrated with the new scope.
+     */
+    contentHash: `gslides:${file.id}:${file.modifiedTime ?? ''}:${includeSpeakerNotes ? 'n1' : 'n0'}`,
     metadata: {
       modifiedTime: file.modifiedTime,
       createdTime: file.createdTime,
@@ -314,7 +319,8 @@ export const googleSlidesConnector: ConnectorConfig = {
     const maxDocs = sourceConfig.maxDocs ? Number(sourceConfig.maxDocs) : 0
     const previouslyFetched = (syncContext?.totalDocsFetched as number) ?? 0
 
-    let documents = files.map(fileToStub)
+    const includeSpeakerNotes = shouldIncludeSpeakerNotes(sourceConfig)
+    let documents = files.map((file) => fileToStub(file, includeSpeakerNotes))
     let slicedSome = false
     if (maxDocs > 0) {
       const remaining = maxDocs - previouslyFetched
@@ -376,11 +382,8 @@ export const googleSlidesConnector: ConnectorConfig = {
     if (file.trashed) return null
     if (file.mimeType !== PRESENTATION_MIME_TYPE) return null
 
-    const content = await fetchPresentationContent(
-      accessToken,
-      file.id,
-      shouldIncludeSpeakerNotes(sourceConfig)
-    )
+    const includeSpeakerNotes = shouldIncludeSpeakerNotes(sourceConfig)
+    const content = await fetchPresentationContent(accessToken, file.id, includeSpeakerNotes)
 
     /**
      * An image-only deck carries no extractable text. Surfacing it as a skipped
@@ -389,10 +392,15 @@ export const googleSlidesConnector: ConnectorConfig = {
      * presentation would simply be missing and re-fetched on every sync.
      */
     if (!content.trim()) {
-      return { ...fileToStub(file), content: '', contentDeferred: false, skippedReason: NO_TEXT }
+      return {
+        ...fileToStub(file, includeSpeakerNotes),
+        content: '',
+        contentDeferred: false,
+        skippedReason: NO_TEXT,
+      }
     }
 
-    return { ...fileToStub(file), content, contentDeferred: false }
+    return { ...fileToStub(file, includeSpeakerNotes), content, contentDeferred: false }
   },
 
   validateConfig: async (

apps/sim/connectors/microsoft-excel/microsoft-excel.ts

@@ -39,6 +39,13 @@ const CONTENT_FORMAT_VERSION = 'v2'
  */
 const MAX_WORKSHEETS = 500
 
+/**
+ * Origin every Graph response must stay on. `@odata.nextLink` is server-supplied
+ * and carries the bearer token when followed, so a link pointing anywhere else is
+ * dropped rather than requested.
+ */
+const GRAPH_API_BASE = 'https://graph.microsoft.com/'
+
 /** Maximum rows read from a single worksheet's used range. */
 const MAX_ROWS = 5000
 
@@ -82,6 +89,7 @@ interface Worksheet {
 
 interface WorksheetListResponse {
   value?: Worksheet[]
+  '@odata.nextLink'?: string
 }
 
 interface WorkbookItem {
@@ -243,18 +251,33 @@ async function fetchWorkbookItem(
 
 /** Lists the workbook's worksheets in tab order. */
 async function fetchWorksheets(accessToken: string, basePath: string): Promise<Worksheet[]> {
-  const response = await fetchWithRetry(
-    `${basePath}/workbook/worksheets?$select=id,name,position,visibility&$orderby=position`,
-    {
+  const worksheets: Worksheet[] = []
+  let url: string | undefined =
+    `${basePath}/workbook/worksheets?$select=id,name,position,visibility&$orderby=position`
+
+  /**
+   * Graph paginates collection responses, so a workbook with more sheets than fit
+   * in one page must follow `@odata.nextLink`. Reading only the first page would
+   * drop the remainder from the listing without setting `listingCapped`, and the
+   * sync engine would then reconcile those documents away as deleted. The walk is
+   * bounded by `MAX_WORKSHEETS`, whose truncation the caller does flag.
+   */
+  while (url && worksheets.length <= MAX_WORKSHEETS) {
+    const response = await fetchWithRetry(url, {
       method: 'GET',
       headers: { Authorization: `Bearer ${accessToken}`, Accept: 'application/json' },
-    }
-  )
+    })
+
+    if (!response.ok) await graphError(response, 'Failed to list worksheets')
 
-  if (!response.ok) await graphError(response, 'Failed to list worksheets')
+    const data = (await response.json()) as WorksheetListResponse
+    worksheets.push(...(data.value ?? []))
+
+    const next = data['@odata.nextLink']
+    url = next && next.startsWith(GRAPH_API_BASE) ? next : undefined
+  }
 
-  const data = (await response.json()) as WorksheetListResponse
-  return data.value ?? []
+  return worksheets
 }
 
 /**

apps/sim/connectors/mintlify/mintlify.ts

@@ -323,9 +323,18 @@ async function discoverFromSitemap(
  */
 function withinBasePath(pages: MintlifyPageLink[], site: MintlifySite): MintlifyPageLink[] {
   if (!site.basePath) return pages
-  return pages.filter(
-    (page) => page.path === site.basePath || page.path.startsWith(`${site.basePath}/`)
-  )
+  return pages.filter((page) => isUnderPath(page.path, site.basePath))
+}
+
+/**
+ * Whether `path` is `prefix` itself or sits beneath it.
+ *
+ * A bare `startsWith` would also match a sibling whose name merely begins with
+ * the prefix — `/guides` would capture `/guides-archive` — so the boundary `/`
+ * is required.
+ */
+function isUnderPath(path: string, prefix: string): boolean {
+  return path === prefix || path.startsWith(`${prefix}/`)
 }
 
 /**
@@ -469,7 +478,7 @@ export const mintlifyConnector: ConnectorConfig = {
       }
 
       const filtered = pathPrefix
-        ? discovered.filter((page) => page.path.startsWith(pathPrefix))
+        ? discovered.filter((page) => isUnderPath(page.path, pathPrefix))
         : discovered
 
       if (filtered.length > maxPages && syncContext) {
@@ -591,7 +600,7 @@ export const mintlifyConnector: ConnectorConfig = {
       }
 
       const pathPrefix = resolvePathPrefix(sourceConfig.pathPrefix)
-      if (pathPrefix && !pages.some((page) => page.path.startsWith(pathPrefix))) {
+      if (pathPrefix && !pages.some((page) => isUnderPath(page.path, pathPrefix))) {
         return { valid: false, error: `No pages match the path prefix "${pathPrefix}"` }
       }