[WEB + CLSI] Download as docx file feature (#32851)

* using CLSI logic for fetching the project contents and skip the .zip export * Use unique conversion directory for project-to-docx export to avoid corrupting the shared compile directory when a compile runs concurrently * Remove X-Accel-Buffering header — not needed as CLSI does not run behind nginx * moving log before sending the data * Return CLSI stream directly instead of buffering to disk on web Previously convertProjectToDocx wrote the CLSI response to a temp file on disk, then the controller read it back to stream to the client. Now the stream is returned directly and piped to the response, avoiding unnecessary disk I/O on the web server. * Use href redirect for docx export instead of fetching blob into memory * making functions and files more generic so they can be used in future for other documents exports as well * adding export-docx split test * adding unit tests * adding cypress E2E test * format:fix * renaming the route to download from convert * adding new icon for export docx button * format:fix * remove unused showExportDocumentErrorToast export and adding guard against invalid Content-Length header from CLSI * format:fix * refactor(clsi): move promisify(parse) into RequestParser * refactor: generic conversion endpoint with type as route param * refactor: use type→extension map for validated conversion types * refactor(clsi): remove --standalone flag and fix rejection test * fixing the href in cypress test * renaming function * adding type to Metrics.inc * fix: rename exportProjectDocument, add WithLock wrapper and metrics type label * format:fix * fix: hide docx export from anonymous users and add WithLock wrapper * format fix * remove redundant Content-Length validation from DocumentConversionManager * format:fix * removing trailing icon GitOrigin-RevId: e9764fefac2c4b625d23be9e942ea4a8b283c70d
2026-04-24 08:06:10 +00:00
parent b6ec7945f4
commit be5a7b56c8
20 changed files with 772 additions and 41 deletions
@@ -1171,6 +1171,21 @@ function _finaliseRequest(projectId, options, project, docs, files) {
  }
 }

+async function buildDocumentConversionRequest(projectId) {
+  const project = await ProjectGetter.promises.getProject(projectId, {
+    compiler: 1,
+    imageName: 1,
+    'overleaf.history.id': 1,
+    rootDoc_id: 1,
+    rootFolder: 1,
+  })
+  if (project == null) {
+    throw new Errors.NotFoundError(`project does not exist: ${projectId}`)
+  }
+  const projectStateHash = ClsiStateManager.computeHash(project, {})
+  return _buildRequestFromMongo(projectId, {}, project, projectStateHash)
+}
+
 async function wordCount(projectId, userId, file, limits, clsiserverid) {
  const { compileBackendClass, compileGroup } = limits
  const req = await _buildRequest(projectId, userId, limits)
@@ -1297,5 +1312,6 @@ export default {
    getOutputFileStream,
    wordCount,
    syncTeX,
+    buildDocumentConversionRequest,
  },
 }
@@ -5,13 +5,55 @@ import DocumentUpdaterHandler from '../DocumentUpdater/DocumentUpdaterHandler.mj
 import { prepareZipAttachment } from '../../infrastructure/Response.mjs'
 import SessionManager from '../Authentication/SessionManager.mjs'
 import ProjectAuditLogHandler from '../Project/ProjectAuditLogHandler.mjs'
+import DocumentConversionManager from '../Uploads/DocumentConversionManager.mjs'
+import { expressify } from '@overleaf/promise-utils'
+import { pipeline } from 'node:stream/promises'
+
+const SUPPORTED_CONVERSION_TYPES = new Map([['docx', 'docx']])

 // Keep in sync with the logic for PDF files in CompileController
 function getSafeProjectName(project) {
  return project.name.replace(/[^\p{L}\p{Nd}]/gu, '_')
 }

+async function exportProjectConversion(req, res) {
+  const type = req.params.type
+  const extension = SUPPORTED_CONVERSION_TYPES.get(type)
+  if (!extension) {
+    return res.sendStatus(400)
+  }
+  const userId = SessionManager.getLoggedInUserId(req.session)
+  const projectId = req.params.Project_id
+  Metrics.inc('document-exports', 1, { type })
+
+  const project = await ProjectGetter.promises.getProject(projectId, {
+    name: true,
+  })
+
+  const { stream, contentLength } =
+    await DocumentConversionManager.promises.convertProjectToDocument(
+      projectId,
+      userId,
+      type
+    )
+
+  const safeFileName = getSafeProjectName(project)
+  res.setHeader('Content-Length', contentLength)
+  res.attachment(`${safeFileName}.${extension}`)
+  res.setHeader('X-Content-Type-Options', 'nosniff')
+  res.setHeader('X-Accel-Buffering', 'no')
+  ProjectAuditLogHandler.addEntryInBackground(
+    projectId,
+    `project-exported-${type}`,
+    userId,
+    req.ip
+  )
+  await pipeline(stream, res)
+}
+
 export default {
+  exportProjectConversion: expressify(exportProjectConversion),
+
  downloadProject(req, res, next) {
    const userId = SessionManager.getLoggedInUserId(req.session)
    const projectId = req.params.Project_id
@@ -480,6 +480,7 @@ const _ProjectController = {
      'wf-fake-non-english-suggestions',
      'editor-tabs',
      'overleaf-code',
+      'export-docx',
    ].filter(Boolean)

    const getUserValues = async userId =>
@@ -1,5 +1,6 @@
 import Settings from '@overleaf/settings'
 import CompileManager from '../Compile/CompileManager.mjs'
+import ClsiManager from '../Compile/ClsiManager.mjs'
 import fs from 'node:fs'
 import fsPromises from 'node:fs/promises'
 import logger from '@overleaf/logger'
@@ -38,14 +39,7 @@ async function convertDocxToLaTeXZipArchive(path, userId) {
      signal: abortController.signal,
    })

-    const contentLengthHeader = response.headers.get('Content-Length')
-    if (contentLengthHeader == null) {
-      logger.warn(
-        'CLSI did not provide Content-Length header for converted document'
-      )
-      throw new OError('CLSI response missing Content-Length header')
-    }
-    const contentLength = parseInt(contentLengthHeader, 10)
+    const contentLength = parseInt(response.headers.get('Content-Length'), 10)
    if (contentLength > Settings.maxUploadSize) {
      abortController.abort()
      stream.destroy()
@@ -77,8 +71,35 @@ async function convertDocxToLaTeXZipArchive(path, userId) {
  return outputPath
 }

+async function convertProjectToDocument(projectId, userId, type) {
+  const limits = await CompileManager.promises._getUserCompileLimits(userId)
+  const clsiRequest =
+    await ClsiManager.promises.buildDocumentConversionRequest(projectId)
+
+  const clsiUrl = new URL(Settings.apis.clsi.url)
+  clsiUrl.pathname = `/project/${projectId}/user/${userId}/download/project-to-document`
+  clsiUrl.searchParams.set('type', type)
+  clsiUrl.searchParams.set('compileBackendClass', limits.compileBackendClass)
+  clsiUrl.searchParams.set('compileGroup', limits.compileGroup)
+
+  logger.debug(
+    { clsiUrl: clsiUrl.toString(), projectId, userId, type },
+    'sending project to CLSI for document conversion'
+  )
+
+  const { stream, response } = await fetchStreamWithResponse(clsiUrl, {
+    method: 'POST',
+    json: clsiRequest,
+  })
+
+  const contentLength = parseInt(response.headers.get('Content-Length'), 10)
+
+  return { stream, contentLength }
+}
+
 export default {
  promises: {
    convertDocxToLaTeXZipArchive,
+    convertProjectToDocument,
  },
 }
@@ -193,6 +193,10 @@ const rateLimiters = {
    points: 10,
    duration: 60,
  }),
+  documentExport: new RateLimiter('document-export', {
+    points: 5,
+    duration: 60,
+  }),
 }

 async function initialize(webRouter, privateApiRouter, publicApiRouter) {
@@ -764,6 +768,18 @@ async function initialize(webRouter, privateApiRouter, publicApiRouter) {
    ExportsController.exportDownload
  )

+  if (Settings.enablePandocConversions) {
+    webRouter.get(
+      '/project/:Project_id/download/conversion/:type',
+      AuthenticationController.requireLogin(),
+      RateLimiterMiddleware.rateLimit(rateLimiters.documentExport, {
+        params: ['Project_id'],
+      }),
+      AuthorizationMiddleware.ensureUserCanReadProject,
+      ProjectDownloadsController.exportProjectConversion
+    )
+  }
+
  webRouter.get(
    '/Project/:Project_id/download/zip',
    RateLimiterMiddleware.rateLimit(rateLimiters.zipDownload, {
@@ -618,7 +618,9 @@
  "expires": "",
  "expires_in_days": "",
  "expires_on": "",
+  "export_as_docx": "",
  "export_csv": "",
+  "export_document_error": "",
  "export_project_to_github": "",
  "failed": "",
  "failed_to_consent_to_workbench_terms": "",
@@ -7,6 +7,7 @@ import importOverleafModules from '../../../../macros/import-overleaf-module.mac
 import { OLToastContainer } from '@/shared/components/ol/ol-toast-container'
 import clipboardToastGenerators from '@/features/source-editor/components/clipboard-toasts'
 import importDocxFeedbackToastGenerators from '@/features/project-list/components/new-project-button/import-docx-feedback-toast'
+import exportDocumentToastGenerators from '@/features/ide-react/components/toolbar/export-document-toasts'

 const moduleGeneratorsImport = importOverleafModules('toastGenerators') as {
  import: { default: GlobalToastGeneratorEntry[] }
@@ -29,6 +30,7 @@ const GENERATOR_LIST: GlobalToastGeneratorEntry[] = [
  ...moduleGenerators.flat(),
  ...clipboardToastGenerators,
  ...importDocxFeedbackToastGenerators,
+  ...exportDocumentToastGenerators,
 ]
 const GENERATOR_MAP: Map<string, GlobalToastGenerator> = new Map(
  GENERATOR_LIST.map(({ key, generator }) => [key, generator])
@@ -7,6 +7,8 @@ import { useProjectContext } from '@/shared/context/project-context'
 import { useCallback } from 'react'
 import { useTranslation } from 'react-i18next'
 import { useEditorAnalytics } from '@/shared/hooks/use-editor-analytics'
+import getMeta from '@/utils/meta'
+import { useFeatureFlag } from '@/shared/context/split-test-context'

 export const DownloadProjectZip = () => {
  const { t } = useTranslation()
@@ -100,3 +102,43 @@ export const DownloadProjectPDF = () => {
    return button
  }
 }
+
+export const ExportProjectDocx = () => {
+  const { t } = useTranslation()
+  const { projectId } = useProjectContext()
+  const exportDocxEnabled = useFeatureFlag('export-docx')
+  const enablePandocConversions =
+    getMeta('ol-ExposedSettings')?.enablePandocConversions
+  const anonymous = getMeta('ol-anonymous')
+
+  const showExportDocx =
+    exportDocxEnabled && enablePandocConversions && !anonymous
+
+  useCommandProvider(
+    () =>
+      showExportDocx
+        ? [
+            {
+              id: 'export-as-docx',
+              href: `/project/${projectId}/download/conversion/docx`,
+              label: t('export_as_docx'),
+            },
+          ]
+        : [],
+    [t, showExportDocx, projectId]
+  )
+
+  if (!showExportDocx) {
+    return null
+  }
+
+  return (
+    <OLDropdownMenuItem
+      href={`/project/${projectId}/download/conversion/docx`}
+      target="_blank"
+      rel="noreferrer"
+    >
+      {t('export_as_docx')}
+    </OLDropdownMenuItem>
+  )
+}
@@ -0,0 +1,22 @@
+import { GlobalToastGeneratorEntry } from '@/features/ide-react/components/global-toasts'
+import { useTranslation } from 'react-i18next'
+
+const ExportDocumentErrorToast = () => {
+  const { t } = useTranslation()
+  return <span>{t('export_document_error')}</span>
+}
+
+const generators: GlobalToastGeneratorEntry[] = [
+  {
+    key: 'export-document:error',
+    generator: () => ({
+      content: <ExportDocumentErrorToast />,
+      type: 'error',
+      autoHide: true,
+      delay: 5000,
+      isDismissible: true,
+    }),
+  },
+]
+
+export default generators
@@ -87,7 +87,7 @@ export const ToolbarMenuBar = () => {
      { id: 'submit', children: ['submit-project', 'manage-template'] },
      {
        id: 'file-download',
-        children: ['download-as-source-zip', 'download-pdf'],
+        children: ['download-as-source-zip', 'download-pdf', 'export-as-docx'],
      },
      {
        id: 'settings',
@@ -10,7 +10,11 @@ import { useTranslation } from 'react-i18next'
 import importOverleafModules from '../../../../../macros/import-overleaf-module.macro'
 import { useEditorContext } from '@/shared/context/editor-context'
 import { useIdeReactContext } from '@/features/ide-react/context/ide-react-context'
-import { DownloadProjectPDF, DownloadProjectZip } from './download-project'
+import {
+  DownloadProjectPDF,
+  DownloadProjectZip,
+  ExportProjectDocx,
+} from './download-project'
 import { useCallback, useState } from 'react'
 import OLDropdownMenuItem from '@/shared/components/ol/ol-dropdown-menu-item'
 import EditableLabel from './editable-label'
@@ -76,6 +80,7 @@ export const ToolbarProjectTitle = () => {
        )}
        <DownloadProjectPDF />
        <DownloadProjectZip />
+        <ExportProjectDocx />
        <DropdownDivider />
        <DuplicateProject />
        <OLDropdownMenuItem
@@ -822,7 +822,9 @@
  "expires_on": "Expires: __date__",
  "expiry": "Expiry Date",
  "explore_all_plans": "Explore all plans",
+  "export_as_docx": "Export as Word document (.docx)",
  "export_csv": "Export CSV",
+  "export_document_error": "Export failed. Please try again.",
  "export_project_to_github": "Export Project to GitHub",
  "failed": "Failed",
  "failed_to_consent_to_workbench_terms": "Failed to consent to workbench terms. Please try again later.",
@@ -28,7 +28,11 @@ describe('ProjectDownloadsController', function () {
    )

    vi.doMock('../../../../app/src/Features/Project/ProjectGetter.mjs', () => ({
-      default: (ctx.ProjectGetter = {}),
+      default: (ctx.ProjectGetter = {
+        promises: {
+          getProject: sinon.stub(),
+        },
+      }),
    }))

    vi.doMock(
@@ -47,6 +51,32 @@ describe('ProjectDownloadsController', function () {
      })
    )

+    vi.doMock(
+      '../../../../app/src/Features/Authentication/SessionManager.mjs',
+      () => ({
+        default: (ctx.SessionManager = {
+          getLoggedInUserId: sinon
+            .stub()
+            .callsFake(session => session?.user?._id),
+        }),
+      })
+    )
+
+    vi.doMock(
+      '../../../../app/src/Features/Uploads/DocumentConversionManager.mjs',
+      () => ({
+        default: (ctx.DocumentConversionManager = {
+          promises: {
+            convertProjectToDocument: sinon.stub(),
+          },
+        }),
+      })
+    )
+
+    vi.doMock('node:stream/promises', () => ({
+      pipeline: (ctx.pipeline = sinon.stub().resolves()),
+    }))
+
    ctx.ProjectDownloadsController = (await import(modulePath)).default
  })

@@ -208,4 +238,107 @@ describe('ProjectDownloadsController', function () {
      }
    })
  })
+
+  describe('exportProjectConversion', function () {
+    describe('when an unsupported type is requested', function () {
+      beforeEach(async function (ctx) {
+        ctx.req.params = { Project_id: 'test-project-id', type: 'unsupported' }
+        ctx.req.session = { user: { _id: 'test-user-id' } }
+
+        await ctx.ProjectDownloadsController.exportProjectConversion(
+          ctx.req,
+          ctx.res,
+          ctx.next
+        )
+      })
+
+      it('should return 400', function (ctx) {
+        expect(ctx.res.statusCode).to.equal(400)
+      })
+
+      it('should not call the conversion manager', function (ctx) {
+        sinon.assert.notCalled(
+          ctx.DocumentConversionManager.promises.convertProjectToDocument
+        )
+      })
+    })
+
+    describe('with a supported type', function () {
+      beforeEach(async function (ctx) {
+        ctx.projectId = 'test-project-id'
+        ctx.userId = 'test-user-id'
+        ctx.projectName = 'My Test Project'
+        ctx.exportStream = { pipe: sinon.stub() }
+        ctx.contentLength = 9876
+
+        ctx.req.params = { Project_id: ctx.projectId, type: 'docx' }
+        ctx.req.session = { user: { _id: ctx.userId } }
+        ctx.req.ip = '192.168.1.1'
+
+        ctx.res.attachment = sinon.stub().returns(ctx.res)
+
+        ctx.SessionManager.getLoggedInUserId.returns(ctx.userId)
+        ctx.ProjectGetter.promises.getProject.resolves({
+          name: ctx.projectName,
+        })
+        ctx.DocumentConversionManager.promises.convertProjectToDocument.resolves(
+          {
+            stream: ctx.exportStream,
+            contentLength: ctx.contentLength,
+          }
+        )
+
+        await ctx.ProjectDownloadsController.exportProjectConversion(
+          ctx.req,
+          ctx.res,
+          ctx.next
+        )
+      })
+
+      it('should call convertProjectToDocument with the docx type', function (ctx) {
+        sinon.assert.calledWith(
+          ctx.DocumentConversionManager.promises.convertProjectToDocument,
+          ctx.projectId,
+          ctx.userId,
+          'docx'
+        )
+      })
+
+      it('should set the Content-Length header', function (ctx) {
+        expect(ctx.res.headers['Content-Length']).to.equal(ctx.contentLength)
+      })
+
+      it('should set the attachment filename with safe project name', function (ctx) {
+        sinon.assert.calledWith(ctx.res.attachment, 'My_Test_Project.docx')
+      })
+
+      it('should set the X-Content-Type-Options header', function (ctx) {
+        expect(ctx.res.headers['X-Content-Type-Options']).to.equal('nosniff')
+      })
+
+      it('should set the X-Accel-Buffering header', function (ctx) {
+        expect(ctx.res.headers['X-Accel-Buffering']).to.equal('no')
+      })
+
+      it('should add an audit log entry', function (ctx) {
+        sinon.assert.calledWith(
+          ctx.ProjectAuditLogHandler.addEntryInBackground,
+          ctx.projectId,
+          'project-exported-docx',
+          ctx.userId,
+          ctx.req.ip
+        )
+      })
+
+      it('should record the action via Metrics', function (ctx) {
+        ctx.Metrics.inc
+          .calledWith('document-exports', 1, { type: 'docx' })
+          .should.equal(true)
+      })
+
+      it('should stream the document to the response', function (ctx) {
+        sinon.assert.calledWith(ctx.pipeline, ctx.exportStream, ctx.res)
+      })
+    })
+  })
 })
@@ -73,6 +73,18 @@ describe('DocumentConversionManager', function () {
      })
    )

+    ctx.ClsiManager = {
+      promises: {
+        buildDocumentConversionRequest: sinon
+          .stub()
+          .resolves({ some: 'clsi-request' }),
+      },
+    }
+
+    vi.doMock('../../../../app/src/Features/Compile/ClsiManager.mjs', () => ({
+      default: ctx.ClsiManager,
+    }))
+
    ctx.DocumentConversionManager = (await import(MODULE_PATH)).default
  })

@@ -215,43 +227,64 @@ describe('DocumentConversionManager', function () {
        )
      })
    })
+  })

-    describe('when the Content-Length header is missing', function () {
+  describe('convertProjectToDocument', function () {
+    beforeEach(function (ctx) {
+      ctx.projectId = 'test-project-id'
+      ctx.userId = 'test-user-id'
+      ctx.type = 'docx'
+      ctx.mockStream = { destroy: sinon.stub() }
+      ctx.response = {
+        headers: { get: sinon.stub().returns(null) },
+      }
+      ctx.response.headers.get.withArgs('Content-Length').returns('50')
+      ctx.fetchUtils.fetchStreamWithResponse.resolves({
+        stream: ctx.mockStream,
+        response: ctx.response,
+      })
+    })
+
+    describe('successfully', function () {
      beforeEach(async function (ctx) {
-        ctx.path = '/path/to/input.docx'
-        ctx.userId = 'test-user-id'
-        ctx.response = {
-          headers: {
-            get: sinon.stub().returns(null),
-          },
-        }
-
-        ctx.fetchUtils.fetchStreamWithResponse.resolves({
-          stream: 'mocked-fetch-stream',
-          response: ctx.response,
-        })
-
-        await expect(
-          ctx.DocumentConversionManager.promises.convertDocxToLaTeXZipArchive(
-            ctx.path,
-            ctx.userId
+        ctx.result =
+          await ctx.DocumentConversionManager.promises.convertProjectToDocument(
+            ctx.projectId,
+            ctx.userId,
+            ctx.type
          )
-        ).to.be.rejectedWith('document conversion failed')
      })

-      it('should not write the archive to disk', function (ctx) {
-        sinon.assert.notCalled(ctx.fs.createWriteStream)
-        sinon.assert.notCalled(ctx.nodeStream.pipeline)
-      })
-
-      it('should attempt to clean up the output path', function (ctx) {
+      it('should build the CLSI document conversion request', function (ctx) {
        sinon.assert.calledWith(
-          ctx.fsPromises.unlink,
-          sinon.match(
-            /\/path\/to\/dump\/folder\/[0-9a-f]{8}-[0-9a-f]{4}-4[0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}_document-conversion\.zip/
-          )
+          ctx.ClsiManager.promises.buildDocumentConversionRequest,
+          ctx.projectId
        )
      })
+
+      it('should call CLSI with the correct URL', function (ctx) {
+        const expectedUrl = new URL(ctx.Settings.apis.clsi.url)
+        expectedUrl.pathname = `/project/${ctx.projectId}/user/${ctx.userId}/download/project-to-document`
+        expectedUrl.searchParams.set('type', ctx.type)
+        expectedUrl.searchParams.set(
+          'compileBackendClass',
+          'test-backend-class'
+        )
+        expectedUrl.searchParams.set('compileGroup', 'test-compile-group')
+
+        sinon.assert.calledWith(
+          ctx.fetchUtils.fetchStreamWithResponse,
+          sinon.match(url => url.toString() === expectedUrl.toString()),
+          { method: 'POST', json: { some: 'clsi-request' } }
+        )
+      })
+
+      it('should return the stream and content length', function (ctx) {
+        expect(ctx.result).to.deep.equal({
+          stream: ctx.mockStream,
+          contentLength: 50,
+        })
+      })
    })
  })
 })