dsh-cait 1.1.6 → 1.1.7

package/lib/xlsx.js

@@ -33,24 +33,33 @@ function sharedStringTable(xml) {
   ))
 }
 
-function firstSheetPath(entries) {
+function workbookSheets(entries) {
   const workbook = zipText(entries, 'xl/workbook.xml')
   const rels = zipText(entries, 'xl/_rels/workbook.xml.rels')
-  if (workbook && rels) {
-    const sheet = workbook.match(/<sheet\b([^>]*)\/?>/i)
-    const id = sheet ? attributes(sheet[1])['r:id'] : ''
-    if (id) {
-      const relation = [...rels.matchAll(/<Relationship\b([^>]*)\/?>/gi)]
-        .map((match) => attributes(match[1]))
-        .find((item) => item.Id === id)
-      if (relation?.Target) {
-        const target = relation.Target.replace(/^\/+/, '')
-        return target.startsWith('xl/') ? target : `xl/${target}`
-      }
-    }
+  const relationMap = new Map(
+    [...rels.matchAll(/<Relationship\b([^>]*)\/?>/gi)]
+      .map((match) => attributes(match[1]))
+      .filter((item) => item.Id && item.Target)
+      .map((item) => [item.Id, item.Target]),
+  )
+  const sheets = []
+  for (const match of workbook.matchAll(/<sheet\b([^>]*)\/?>/gi)) {
+    const attrs = attributes(match[1])
+    const id = attrs['r:id']
+    const target = relationMap.get(id)
+    if (!target) continue
+    const path = target.replace(/^\/+/, '')
+    sheets.push({
+      name: attrs.name || `Sheet${sheets.length + 1}`,
+      path: path.startsWith('xl/') ? path : `xl/${path}`,
+    })
   }
-  const fallback = entries.find((item) => /^xl\/worksheets\/sheet\d+\.xml$/i.test(item.name))
-  return fallback?.name ?? 'xl/worksheets/sheet1.xml'
+  if (sheets.length) return sheets
+  return entries
+    .map((entry) => entry.name)
+    .filter((name) => /^xl\/worksheets\/sheet\d+\.xml$/i.test(name))
+    .sort((left, right) => Number(left.match(/(\d+)\.xml$/)[1]) - Number(right.match(/(\d+)\.xml$/)[1]))
+    .map((path, index) => ({ name: `Sheet${index + 1}`, path }))
 }
 
 function columnIndex(reference) {
@@ -85,7 +94,12 @@ function parseRows(xml, shared) {
     for (const cellMatch of rowMatch[2].matchAll(/<c\b([^>]*?)(?:\/>|>([\s\S]*?)<\/c>)/gi)) {
       const attrs = attributes(cellMatch[1])
       if (!attrs.r) continue
-      cells.set(columnIndex(attrs.r), cellValue(attrs, cellMatch[2] ?? '', shared))
+      const body = cellMatch[2] ?? ''
+      cells.set(columnIndex(attrs.r), {
+        cell: attrs.r,
+        value: cellValue(attrs, body, shared),
+        formula: /<f\b/i.test(body),
+      })
     }
     if (cells.size) rows.set(rowNumber, cells)
   }
@@ -100,39 +114,53 @@ export function parseXlsx(buffer) {
     throw fail('XLSX_PARSE', `XLSX 文件不是有效的 ZIP/Excel 包: ${error.message}`)
   }
   const shared = sharedStringTable(zipText(entries, 'xl/sharedStrings.xml'))
-  const sheetPath = firstSheetPath(entries)
-  const sheetXml = zipText(entries, sheetPath)
-  if (!sheetXml) throw fail('XLSX_PARSE', 'XLSX 中找不到工作表。')
-  const rows = parseRows(sheetXml, shared)
-  if (!rows.length) throw fail('XLSX_NO_UNITS', 'XLSX 中没有可用的行。')
-
-  const maxColumns = rows.reduce((max, [, cells]) => Math.max(max, ...cells.keys(), -1), -1) + 1
-  const first = rows[0][1]
-  const header = (index) => String(first.get(index) ?? '').trim().toLowerCase()
-  let sourceIndex = -1
-  let targetIndex = -1
-  for (let index = 0; index < maxColumns; index += 1) {
-    if (SOURCE_HEADERS.has(header(index))) sourceIndex = index
-    if (TARGET_HEADERS.has(header(index))) targetIndex = index
-  }
-  const hasHeader = sourceIndex >= 0 && targetIndex >= 0
-  if (!hasHeader) {
-    sourceIndex = 0
-    targetIndex = maxColumns > 1 ? 1 : -1
-  }
+  const sheets = workbookSheets(entries)
+  if (!sheets.length) throw fail('XLSX_PARSE', 'XLSX 中找不到工作表。')
 
   const segments = []
-  for (const [rowNumber, cells] of rows) {
-    if (hasHeader && rowNumber === rows[0][0]) continue
-    const source = String(cells.get(sourceIndex) ?? '').trim()
-    if (!source) continue
-    const target = targetIndex >= 0 ? String(cells.get(targetIndex) ?? '').trim() : ''
-    segments.push({
-      source,
-      target,
-      paragraph: segments.length,
-      meta: { format: 'xlsx', xlsxRow: rowNumber },
-    })
+  for (const [sheetIndex, sheet] of sheets.entries()) {
+    const sheetXml = zipText(entries, sheet.path)
+    if (!sheetXml) continue
+    const rows = parseRows(sheetXml, shared)
+    if (!rows.length) continue
+
+    const maxColumns = rows.reduce((max, [, cells]) => Math.max(max, ...cells.keys(), -1), -1) + 1
+    const first = rows[0][1]
+    const header = (index) => String(first.get(index)?.value ?? '').trim().toLowerCase()
+    let sourceIndex = -1
+    let targetIndex = -1
+    for (let index = 0; index < maxColumns; index += 1) {
+      if (SOURCE_HEADERS.has(header(index))) sourceIndex = index
+      if (TARGET_HEADERS.has(header(index))) targetIndex = index
+    }
+    const hasHeader = sourceIndex >= 0 && targetIndex >= 0
+    if (!hasHeader) {
+      sourceIndex = 0
+      targetIndex = maxColumns > 1 ? 1 : -1
+    }
+
+    for (const [rowNumber, cells] of rows) {
+      if (hasHeader && rowNumber === rows[0][0]) continue
+      const sourceCell = cells.get(sourceIndex)
+      if (!sourceCell || sourceCell.formula) continue
+      const source = String(sourceCell.value ?? '').trim()
+      if (!source) continue
+      const targetCell = targetIndex >= 0 ? cells.get(targetIndex) : null
+      const target = targetCell ? String(targetCell.value ?? '').trim() : ''
+      segments.push({
+        source,
+        target,
+        paragraph: segments.length,
+        meta: {
+          format: 'xlsx',
+          xlsxRow: rowNumber,
+          sheet: sheetIndex + 1,
+          sheetName: sheet.name,
+          cell: sourceCell.cell,
+          ...(targetCell?.cell ? { targetCell: targetCell.cell } : {}),
+        },
+      })
+    }
   }
   if (!segments.length) throw fail('XLSX_NO_UNITS', 'XLSX 中没有可用的源文行。')
   return { format: 'xlsx', segments, warnings: [] }
@@ -153,23 +181,12 @@ function columnName(index) {
   return name
 }
 
-export function buildXlsx(doc) {
-  const rows = [[
-    inlineCell('A1', 'Source'),
-    inlineCell('B1', 'Target'),
-  ]]
-  doc.segments.forEach((segment, index) => {
-    const row = index + 2
-    rows.push([
-      inlineCell(`A${row}`, segment.source || ''),
-      inlineCell(`B${row}`, segment.target || ''),
-    ])
-  })
+function sheetXml(rows) {
   const sheetData = rows
     .map((cells, index) => `    <row r="${index + 1}">${cells.join('')}</row>`)
     .join('\n')
-  const lastRow = rows.length + 1
-  const sheet = `<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+  const lastRow = Math.max(rows.length, 1)
+  return `<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
 <worksheet xmlns="http://schemas.openxmlformats.org/spreadsheetml/2006/main">
   <dimension ref="A1:B${lastRow}"/>
   <sheetViews><sheetView workbookViewId="0"/></sheetViews>
@@ -178,12 +195,105 @@ export function buildXlsx(doc) {
 ${sheetData}
   </sheetData>
 </worksheet>`
+}
+
+function safeSheetName(name, used) {
+  let base = String(name ?? '').replace(/[\[\]:*?\/\\]/g, ' ').trim().slice(0, 31) || 'Sheet'
+  let candidate = base
+  let index = 2
+  while (used.has(candidate.toLowerCase())) {
+    const suffix = ` ${index}`
+    candidate = `${base.slice(0, 31 - suffix.length)}${suffix}`
+    index += 1
+  }
+  used.add(candidate.toLowerCase())
+  return candidate
+}
+
+function replaceCellText(xml, reference, value) {
+  const escaped = escapeXml(value)
+  const pattern = new RegExp(
+    `<c\\b([^>]*\\br="${reference}"[^>]*?)(?:/>|>([\\s\\S]*?)<\\/c>)`,
+    'i',
+  )
+  let found = false
+  const output = String(xml).replace(pattern, (all, rawAttrs, body = '') => {
+    if (/<f\b/i.test(body)) return all
+    found = true
+    const attrs = rawAttrs.replace(/\s+t="[^"]*"/gi, '').replace(/\s*\/\s*$/g, '')
+    return `<c${attrs} t="inlineStr"><is><t xml:space="preserve">${escaped}</t></is></c>`
+  })
+  return { xml: output, found }
+}
+
+/**
+ * Write target text back into the original workbook. Rebuilding a workbook
+ * drops styles, column widths and merged ranges; patching the source zip keeps
+ * everything except the translated cell content intact.
+ */
+export function patchXlsx(doc, template) {
+  let entries
+  try {
+    entries = readZipEntries(template)
+  } catch (error) {
+    throw fail('XLSX_SOURCE_INVALID', `XLSX 源文件不是有效的 ZIP/Excel 包: ${error.message}`)
+  }
+  const sheetPaths = new Map(workbookSheets(entries).map((sheet) => [sheet.name, sheet.path]))
+  const bySheet = new Map()
+  for (const segment of doc.segments) {
+    const sheetName = segment.meta?.sheetName
+    const cell = segment.meta?.targetCell
+    if (!sheetName || !cell) continue
+    if (!bySheet.has(sheetName)) bySheet.set(sheetName, new Map())
+    bySheet.get(sheetName).set(cell, String(segment.target || segment.source || ''))
+  }
+
+  let patched = 0
+  let missing = 0
+  const next = entries.map((entry) => {
+    const sheetName = [...sheetPaths.entries()].find(([, path]) => path === entry.name)?.[0]
+    const replacements = sheetName ? bySheet.get(sheetName) : null
+    if (!replacements?.size) return entry
+    let xml = entry.data.toString('utf8')
+    for (const [cell, text] of replacements) {
+      const result = replaceCellText(xml, cell, text)
+      xml = result.xml
+      if (result.found) patched += 1
+      else missing += 1
+    }
+    return { ...entry, data: Buffer.from(xml, 'utf8') }
+  })
+  return { buffer: writeZip(next), patched, missing }
+}
+
+export function buildXlsx(doc) {
+  const groups = new Map()
+  for (const segment of doc.segments) {
+    const key = segment.meta?.sheetName || `Sheet ${Number(segment.meta?.sheet) || 1}`
+    if (!groups.has(key)) groups.set(key, [])
+    groups.get(key).push(segment)
+  }
+  if (!groups.size) groups.set('Sheet 1', [])
+
+  const used = new Set()
+  const sheets = [...groups.entries()].map(([name, segments], index) => ({
+    name: safeSheetName(name, used),
+    path: `xl/worksheets/sheet${index + 1}.xml`,
+    rows: [
+      [inlineCell('A1', 'Source'), inlineCell('B1', 'Target')],
+      ...segments.map((segment, rowIndex) => [
+        inlineCell(`A${rowIndex + 2}`, segment.source || ''),
+        inlineCell(`B${rowIndex + 2}`, segment.target || ''),
+      ]),
+    ],
+  }))
+
   const contentTypes = `<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
 <Types xmlns="http://schemas.openxmlformats.org/package/2006/content-types">
   <Default Extension="rels" ContentType="application/vnd.openxmlformats-package.relationships+xml"/>
   <Default Extension="xml" ContentType="application/xml"/>
   <Override PartName="/xl/workbook.xml" ContentType="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet.main+xml"/>
-  <Override PartName="/xl/worksheets/sheet1.xml" ContentType="application/vnd.openxmlformats-officedocument.spreadsheetml.worksheet+xml"/>
+${sheets.map((sheet) => `  <Override PartName="/${sheet.path}" ContentType="application/vnd.openxmlformats-officedocument.spreadsheetml.worksheet+xml"/>`).join('\n')}
   <Override PartName="/xl/styles.xml" ContentType="application/vnd.openxmlformats-officedocument.spreadsheetml.styles+xml"/>
 </Types>`
   const rootRels = `<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
@@ -192,12 +302,12 @@ ${sheetData}
 </Relationships>`
   const workbook = `<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
 <workbook xmlns="http://schemas.openxmlformats.org/spreadsheetml/2006/main" xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships">
-  <sheets><sheet name="CAIT" sheetId="1" r:id="rId1"/></sheets>
+  <sheets>${sheets.map((sheet, index) => `<sheet name="${escapeXml(sheet.name)}" sheetId="${index + 1}" r:id="rId${index + 1}"/>`).join('')}</sheets>
 </workbook>`
   const workbookRels = `<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
 <Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships">
-  <Relationship Id="rId1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/worksheet" Target="worksheets/sheet1.xml"/>
-  <Relationship Id="rId2" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/styles" Target="styles.xml"/>
+${sheets.map((sheet, index) => `  <Relationship Id="rId${index + 1}" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/worksheet" Target="worksheets/sheet${index + 1}.xml"/>`).join('\n')}
+  <Relationship Id="rId${sheets.length + 1}" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/styles" Target="styles.xml"/>
 </Relationships>`
   const styles = `<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
 <styleSheet xmlns="http://schemas.openxmlformats.org/spreadsheetml/2006/main">
@@ -212,7 +322,7 @@ ${sheetData}
     { name: '_rels/.rels', data: Buffer.from(rootRels, 'utf8') },
     { name: 'xl/workbook.xml', data: Buffer.from(workbook, 'utf8') },
     { name: 'xl/_rels/workbook.xml.rels', data: Buffer.from(workbookRels, 'utf8') },
-    { name: 'xl/worksheets/sheet1.xml', data: Buffer.from(sheet, 'utf8') },
+    ...sheets.map((sheet) => ({ name: sheet.path, data: Buffer.from(sheetXml(sheet.rows), 'utf8') })),
     { name: 'xl/styles.xml', data: Buffer.from(styles, 'utf8') },
   ])
 }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "dsh-cait",
-  "version": "1.1.6",
+  "version": "1.1.7",
   "license": "Apache-2.0",
   "author": {
     "name": "Jovi Lecca",