Skip to content

A team of long-lived bots for DeepSeek Harness, with main bots, bot-to-bot messages and group chats.

Handle single-quoted rel ids, sectPr order and hyperlink anchors

Fei24113df596cddf7e
diff --git a/ds-bot/office.js b/ds-bot/office.js
index ab2df87b..70625bdf 100644
--- a/ds-bot/office.js
+++ b/ds-bot/office.js
@@ -7842,7 +7842,7 @@ function addComment(pkg, getDoc, changed, { paragraph, text, anchor, author = "B
if (anchor) {
const span = phraseRuns(doc, p, anchor);
if (!span) throw new Error(`anchor "${anchor}" not found in ${paragraph}`);
- p.insertBefore(rangeStart, span.firstRun);
+ span.firstRun.parentNode.insertBefore(rangeStart, span.firstRun);
span.lastRun.parentNode.insertBefore(rangeEnd, span.lastRun.nextSibling);
span.lastRun.parentNode.insertBefore(ref, rangeEnd.nextSibling);
} else {
@@ -7944,7 +7944,8 @@ function resolveChanges(doc, accept) {
}
for (const c of [...parent.childNodes]) if (!keep.has(c.tagName) && c !== el) parent.removeChild(c);
if (stored) {
- const anchor = [...parent.childNodes].find((c) => keep.has(c.tagName)) ?? null;
+ const kept = [...parent.childNodes].filter((c) => keep.has(c.tagName));
+ const anchor = parent.tagName === "w:sectPr" ? kept.length > 0 ? kept.at(-1).nextSibling : parent.firstChild === el ? null : parent.firstChild : kept[0] ?? el;
for (const s of [...stored.childNodes]) parent.insertBefore(s, anchor);
}
parent.removeChild(el);
@@ -8281,7 +8282,7 @@ function pptxClean(pkg, { dryRun = false } = {}) {
const xml = pkg.get(owner) ?? Buffer.alloc(0);
for (const [id, rel] of [...map]) {
if (rel.mode !== "Internal" || !CONTENT_REL.test(rel.type)) continue;
- if (!xml.includes(`"${id}"`)) map.delete(id);
+ if (!xml.includes(`"${id}"`) && !xml.includes(`'${id}'`)) map.delete(id);
}
}
const listed = new Set(deckOrder(pkg).map((e) => e.part));
diff --git a/ds-bot/src/office/docx.js b/ds-bot/src/office/docx.js
index f999257e..d5e27eb5 100644
--- a/ds-bot/src/office/docx.js
+++ b/ds-bot/src/office/docx.js
@@ -506,7 +506,8 @@ function addComment(pkg, getDoc, changed, { paragraph, text, anchor, author = 'B
if (anchor) {
const span = phraseRuns(doc, p, anchor)
if (!span) throw new Error(`anchor "${anchor}" not found in ${paragraph}`)
- p.insertBefore(rangeStart, span.firstRun)
+ // The phrase can sit inside a w:hyperlink — insert beside the run itself.
+ span.firstRun.parentNode.insertBefore(rangeStart, span.firstRun)
span.lastRun.parentNode.insertBefore(rangeEnd, span.lastRun.nextSibling)
span.lastRun.parentNode.insertBefore(ref, rangeEnd.nextSibling)
} else {
@@ -594,7 +595,13 @@ export function resolveChanges(doc, accept) {
if (parent.tagName === 'w:sectPr') { keep.add('w:headerReference'); keep.add('w:footerReference') }
for (const c of [...parent.childNodes]) if (!keep.has(c.tagName) && c !== el) parent.removeChild(c)
if (stored) {
- const anchor = [...parent.childNodes].find(c => keep.has(c.tagName)) ?? null
+ // pPr keeps its mark rPr/sectPr at the end; a sectPr's header/footer
+ // references come first, so restored children go after them (the
+ // change element is removed anyway).
+ const kept = [...parent.childNodes].filter(c => keep.has(c.tagName))
+ const anchor = parent.tagName === 'w:sectPr'
+ ? (kept.length > 0 ? kept.at(-1).nextSibling : parent.firstChild === el ? null : parent.firstChild)
+ : (kept[0] ?? el)
for (const s of [...stored.childNodes]) parent.insertBefore(s, anchor)
}
parent.removeChild(el)
diff --git a/ds-bot/src/office/pptx.js b/ds-bot/src/office/pptx.js
index 828200cf..cbed3a59 100644
--- a/ds-bot/src/office/pptx.js
+++ b/ds-bot/src/office/pptx.js
@@ -353,7 +353,8 @@ export function pptxClean(pkg, { dryRun = false } = {}) {
const xml = pkg.get(owner) ?? Buffer.alloc(0)
for (const [id, rel] of [...map]) {
if (rel.mode !== 'Internal' || !CONTENT_REL.test(rel.type)) continue
- if (!xml.includes(`"${id}"`)) map.delete(id)
+ // Non-Office generators can write single-quoted attributes.
+ if (!xml.includes(`"${id}"`) && !xml.includes(`'${id}'`)) map.delete(id)
}
}
// Slides not in sldIdLst.
diff --git a/ds-bot/test/office.test.js b/ds-bot/test/office.test.js
index 46157b26..9cd98099 100644
--- a/ds-bot/test/office.test.js
+++ b/ds-bot/test/office.test.js
@@ -10,8 +10,8 @@ import assert from 'node:assert'
import { readZip } from '../src/office/zip.js'
import { Package } from '../src/office/package.js'
import { checkPackage, fixPackage } from '../src/office/check.js'
-import { docxRead, docxEdit } from '../src/office/docx.js'
-import { pptxInventory, pptxLayout } from '../src/office/pptx.js'
+import { docxRead, docxEdit, resolveChanges } from '../src/office/docx.js'
+import { pptxInventory, pptxLayout, pptxClean } from '../src/office/pptx.js'
import { xlsxInventory, xlsxRead } from '../src/office/xlsx.js'
import { decodePng, encodePng, contactSheet } from '../src/office/png.js'
import { parseXml, elementsOf, childrenOf, textOf } from '../src/office/xml.js'
@@ -424,6 +424,49 @@ describe('check rules (review 10, 11)', () => {
})
})
+describe('re-review fixes', () => {
+ test('clean keeps a single-quoted r:embed', () => {
+ const pkg = Package.open(fixture('sample.pptx'))
+ const rel = [...pkg.relsOf('ppt/slides/slide1.xml').values()].find(r => r.type.endsWith('/image'))
+ const xml = String(pkg.get('ppt/slides/slide1.xml')).replace(`"${rel.id}"`, `'${rel.id}'`)
+ pkg.put('ppt/slides/slide1.xml', xml)
+ assert.ok(pptxClean(pkg, { dryRun: true }).removed.every(n => n !== 'ppt/media/image1.png'))
+ })
+ test('sectPr reject puts header/footer references before restored properties', () => {
+ const pkg = Package.open(fixture('sample.docx'))
+ const doc = pkg.xml('word/document.xml')
+ const sectPr = elementsOf(doc.documentElement, 'w:sectPr')[0]
+ const ref = doc.createElement('w:headerReference')
+ ref.setAttribute('w:type', 'default')
+ sectPr.insertBefore(ref, sectPr.firstChild)
+ const change = doc.createElement('w:sectPrChange')
+ const stored = doc.createElement('w:sectPr')
+ const sz = doc.createElement('w:pgSz')
+ sz.setAttribute('w:w', '11906'); sz.setAttribute('w:h', '16838')
+ stored.appendChild(sz); change.appendChild(stored)
+ sectPr.appendChild(change)
+ resolveChanges(doc, false)
+ const names = childrenOf(sectPr).map(c => c.tagName)
+ assert.equal(names[0], 'w:headerReference')
+ assert.ok(names.includes('w:pgSz'))
+ assert.ok(!names.includes('w:sectPrChange'))
+ })
+ test('comment anchor inside a w:hyperlink wraps the phrase', () => {
+ const pkg = Package.open(fixture('sample.docx'))
+ const doc = pkg.xml('word/document.xml')
+ const p = parseXml(`<w:p xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main"><w:hyperlink w:history="1" w:anchor="top"><w:r><w:t xml:space="preserve">click here text</w:t></w:r></w:hyperlink></w:p>`).documentElement
+ childrenOf(doc.documentElement, 'w:body')[0].appendChild(doc.importNode(p, true))
+ pkg.putXml('word/document.xml', doc)
+ const dir = tmp(), out = join(dir, 'hl.docx')
+ writeFileSync(join(dir, 'hl-src.docx'), pkg.save())
+ cli('docx', 'edit', join(dir, 'hl-src.docx'), '-o', out, '--ops',
+ JSON.stringify([{ comment: { paragraph: 'p7', text: 'hi', anchor: 'here' } }]))
+ const read = docxRead(Package.open(out))
+ assert.ok(read.comments.some(c => c.text === 'hi' && c.anchorsTo === 'p7'))
+ assert.deepEqual(checkPackage(Package.open(out)), [])
+ })
+})
+
describe('JSON findings shape (review 14)', () => {
test('findings come back ok:true with exit 1', () => {
const dir = tmp()