fix(parsers): walk slides in display order and read SmartArt and chart text

The PPTX walker sorted physical slide part names, but a deck reordered in
PowerPoint keeps its old part names and changes only p:sldIdLst, so it
was indexed out of order. Slides now follow the presentation's id list
resolved through its rels, skipping ids whose part is missing and falling
back to part numbering only when nothing resolves. Graphic frames that
hold SmartArt or a chart were dropped entirely; the diagram data part's
dgm:pt text bodies and a modest chart summary (title, axis titles,
series, categories) are now emitted, with connector text included. Every
relationship target is clamped under ppt/ and read through the per-part
size cap.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

Author: waleedlatif1

apps/sim/lib/file-parsers/ooxml-presentation.test.ts

@@ -28,25 +28,72 @@ interface DeckSlide {
   index: number
   spTree: string
   notesSpTree?: string
+  /** Extra `<Relationship>` elements for the slide's own rels part. */
+  extraRels?: string
 }
 
-async function buildDeck(slides: DeckSlide[]): Promise<Buffer> {
+const RELS_NS = 'xmlns="http://schemas.openxmlformats.org/package/2006/relationships"'
+const REL_TYPE = 'http://schemas.openxmlformats.org/officeDocument/2006/relationships'
+
+/** Lists the given slide part names in `p:sldIdLst` order, resolved through the presentation rels. */
+function presentationParts(zip: JSZip, order: number[]): void {
+  const ids = order.map((n, i) => `<p:sldId id="${256 + i}" r:id="rId${n}"/>`).join('')
+  zip.file(
+    'ppt/presentation.xml',
+    `<?xml version="1.0" encoding="UTF-8" standalone="yes"?><p:presentation ${NS}><p:sldIdLst>${ids}</p:sldIdLst></p:presentation>`
+  )
+  const rels = order
+    .map(
+      (n) => `<Relationship Id="rId${n}" Type="${REL_TYPE}/slide" Target="slides/slide${n}.xml"/>`
+    )
+    .join('')
+  zip.file(
+    'ppt/_rels/presentation.xml.rels',
+    `<?xml version="1.0" encoding="UTF-8" standalone="yes"?><Relationships ${RELS_NS}>${rels}</Relationships>`
+  )
+}
+
+async function buildDeck(slides: DeckSlide[], order?: number[]): Promise<Buffer> {
   const zip = new JSZip()
   zip.file('[Content_Types].xml', '<Types/>')
   zip.file('ppt/media/image1.png', Buffer.from([0x89, 0x50, 0x4e, 0x47]))
+  if (order) presentationParts(zip, order)
   for (const slide of slides) {
     zip.file(`ppt/slides/slide${slide.index}.xml`, slideXml(slide.spTree))
+    const rels: string[] = []
     if (slide.notesSpTree !== undefined) {
+      rels.push(
+        `<Relationship Id="rId2" Type="${REL_TYPE}/notesSlide" Target="../notesSlides/notesSlide${slide.index}.xml"/>`
+      )
+      zip.file(`ppt/notesSlides/notesSlide${slide.index}.xml`, notesXml(slide.notesSpTree))
+    }
+    if (slide.extraRels) rels.push(slide.extraRels)
+    if (rels.length > 0) {
       zip.file(
         `ppt/slides/_rels/slide${slide.index}.xml.rels`,
-        `<?xml version="1.0" encoding="UTF-8" standalone="yes"?><Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships"><Relationship Id="rId2" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/notesSlide" Target="../notesSlides/notesSlide${slide.index}.xml"/></Relationships>`
+        `<?xml version="1.0" encoding="UTF-8" standalone="yes"?><Relationships ${RELS_NS}>${rels.join('')}</Relationships>`
       )
-      zip.file(`ppt/notesSlides/notesSlide${slide.index}.xml`, notesXml(slide.notesSpTree))
     }
   }
   return zip.generateAsync({ type: 'nodebuffer' }) as Promise<Buffer>
 }
 
+const DIAGRAM_FRAME = `<p:graphicFrame><p:nvGraphicFramePr><p:cNvPr id="6" name="Diagram 5"/><p:cNvGraphicFramePr/><p:nvPr/></p:nvGraphicFramePr><a:graphic><a:graphicData uri="http://schemas.openxmlformats.org/drawingml/2006/diagram"><dgm:relIds xmlns:dgm="http://schemas.openxmlformats.org/drawingml/2006/diagram" r:dm="rId3" r:lo="rId4" r:qs="rId5" r:cs="rId6"/></a:graphicData></a:graphic></p:graphicFrame>`
+
+const CHART_FRAME = `<p:graphicFrame><p:nvGraphicFramePr><p:cNvPr id="7" name="Chart 6"/><p:cNvGraphicFramePr/><p:nvPr/></p:nvGraphicFramePr><a:graphic><a:graphicData uri="http://schemas.openxmlformats.org/drawingml/2006/chart"><c:chart xmlns:c="http://schemas.openxmlformats.org/drawingml/2006/chart" r:id="rId3"/></a:graphicData></a:graphic></p:graphicFrame>`
+
+function diagramDataXml(points: string[]): string {
+  const pts = points
+    .map(
+      (text, i) =>
+        `<dgm:pt modelId="{${i}}"><dgm:t><a:bodyPr/><a:p><a:r><a:t>${text}</a:t></a:r></a:p></dgm:t></dgm:pt>`
+    )
+    .join('')
+  return `<?xml version="1.0" encoding="UTF-8" standalone="yes"?><dgm:dataModel xmlns:dgm="http://schemas.openxmlformats.org/drawingml/2006/diagram" xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main"><dgm:ptLst><dgm:pt modelId="{doc}" type="doc"><dgm:t><a:p/></dgm:t></dgm:pt>${pts}</dgm:ptLst></dgm:dataModel>`
+}
+
+const CHART_XML = `<?xml version="1.0" encoding="UTF-8" standalone="yes"?><c:chartSpace xmlns:c="http://schemas.openxmlformats.org/drawingml/2006/chart" xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main"><c:chart><c:title><c:tx><c:rich><a:p><a:r><a:t>Revenue by </a:t></a:r><a:r><a:t>quarter</a:t></a:r></a:p></c:rich></c:tx></c:title><c:plotArea><c:barChart><c:ser><c:idx val="0"/><c:tx><c:strRef><c:f>Sheet1!$B$1</c:f><c:strCache><c:pt idx="0"><c:v>Sales</c:v></c:pt></c:strCache></c:strRef></c:tx><c:cat><c:strRef><c:strCache><c:pt idx="0"><c:v>1st Qtr</c:v></c:pt><c:pt idx="1"><c:v>2nd Qtr</c:v></c:pt></c:strCache></c:strRef></c:cat><c:val><c:numRef><c:numCache><c:pt idx="0"><c:v>10</c:v></c:pt></c:numCache></c:numRef></c:val></c:ser><c:ser><c:idx val="1"/><c:tx><c:strRef><c:strCache><c:pt idx="0"><c:v>Costs</c:v></c:pt></c:strCache></c:strRef></c:tx><c:cat><c:strRef><c:strCache><c:pt idx="0"><c:v>1st Qtr</c:v></c:pt><c:pt idx="1"><c:v>2nd Qtr</c:v></c:pt></c:strCache></c:strRef></c:cat></c:ser></c:barChart><c:catAx><c:axId val="1"/><c:title><c:tx><c:rich><a:p><a:r><a:t>Quarter</a:t></a:r></a:p></c:rich></c:tx></c:title></c:catAx><c:valAx><c:axId val="2"/></c:valAx></c:plotArea></c:chart></c:chartSpace>`
+
 describe('extractPresentationText', () => {
   afterEach(() => {
     vi.restoreAllMocks()
@@ -196,6 +243,83 @@ describe('extractPresentationText', () => {
     })
   })
 
+  it('follows the presentation sldIdLst order rather than part numbering', async () => {
+    const buffer = await buildDeck(
+      [
+        { index: 1, spTree: shape('One') },
+        { index: 2, spTree: shape('Two') },
+        { index: 3, spTree: shape('Three') },
+      ],
+      [3, 1, 2]
+    )
+
+    expect(await extractPresentationText(buffer)).toBe('Three\n\nOne\n\nTwo')
+  })
+
+  it('skips slide ids whose target is missing and falls back when none resolve', async () => {
+    const withMissing = await buildDeck(
+      [
+        { index: 1, spTree: shape('One') },
+        { index: 2, spTree: shape('Two') },
+      ],
+      [2, 9, 1]
+    )
+    expect(await extractPresentationText(withMissing)).toBe('Two\n\nOne')
+
+    const noneResolve = await buildDeck([{ index: 1, spTree: shape('Only') }], [7])
+    expect(await extractPresentationText(noneResolve)).toBe('Only')
+  })
+
+  it('reads SmartArt text from the diagram data part in document order', async () => {
+    const zip = new JSZip()
+    zip.file('ppt/slides/slide1.xml', slideXml(shape('Process', 'title') + DIAGRAM_FRAME))
+    zip.file(
+      'ppt/slides/_rels/slide1.xml.rels',
+      `<Relationships ${RELS_NS}><Relationship Id="rId3" Type="${REL_TYPE}/diagramData" Target="../diagrams/data1.xml"/><Relationship Id="rId4" Type="${REL_TYPE}/diagramLayout" Target="../diagrams/layout1.xml"/></Relationships>`
+    )
+    zip.file('ppt/diagrams/data1.xml', diagramDataXml(['Plan', 'Build', 'Ship']))
+    zip.file('ppt/diagrams/layout1.xml', '<dgm:layoutDef xmlns:dgm="x"/>')
+    const buffer = (await zip.generateAsync({ type: 'nodebuffer' })) as Buffer
+
+    expect(await extractPresentationText(buffer)).toBe('Process\n\nPlan\nBuild\nShip')
+  })
+
+  it('summarizes a chart as title, axis titles, series, and categories', async () => {
+    const zip = new JSZip()
+    zip.file('ppt/slides/slide1.xml', slideXml(CHART_FRAME))
+    zip.file(
+      'ppt/slides/_rels/slide1.xml.rels',
+      `<Relationships ${RELS_NS}><Relationship Id="rId3" Type="${REL_TYPE}/chart" Target="../charts/chart1.xml"/></Relationships>`
+    )
+    zip.file('ppt/charts/chart1.xml', CHART_XML)
+    const buffer = (await zip.generateAsync({ type: 'nodebuffer' })) as Buffer
+
+    expect(await extractPresentationText(buffer)).toBe(
+      '[Chart]\nRevenue by quarter\nQuarter\nSales\nCosts\n1st Qtr\n2nd Qtr\n[/Chart]'
+    )
+  })
+
+  it('ignores a diagram target that escapes ppt/', async () => {
+    const zip = new JSZip()
+    zip.file('ppt/slides/slide1.xml', slideXml(shape('Body') + DIAGRAM_FRAME))
+    zip.file(
+      'ppt/slides/_rels/slide1.xml.rels',
+      `<Relationships ${RELS_NS}><Relationship Id="rId3" Type="${REL_TYPE}/diagramData" Target="../../docProps/data1.xml"/></Relationships>`
+    )
+    zip.file('docProps/data1.xml', diagramDataXml(['Leaked']))
+    const buffer = (await zip.generateAsync({ type: 'nodebuffer' })) as Buffer
+
+    expect(await extractPresentationText(buffer)).toBe('Body')
+  })
+
+  it('includes text carried by a connector shape', async () => {
+    const spTree = `<p:cxnSp><p:nvCxnSpPr><p:cNvPr id="9" name="Connector 8"/><p:cNvCxnSpPr/><p:nvPr/></p:nvCxnSpPr><p:txBody><a:bodyPr/><a:p><a:r><a:t>Yes</a:t></a:r></a:p></p:txBody></p:cxnSp>${shape('After')}`
+
+    expect(await extractPresentationText(await buildDeck([{ index: 1, spTree }]))).toBe(
+      'Yes\nAfter'
+    )
+  })
+
   it('rejects when the signal is already aborted', async () => {
     const controller = new AbortController()
     controller.abort()