Merge pull request #18148 from overleaf/ar-add-output-zip-endpoint-to-clsi

[clsi] Add endpoints to get zip of output files GitOrigin-RevId: a1a935e8170ab5a8d40baa6d96f8e42fe22c2e8c
2026-11-14 22:23:42 +01:00 · 2024-05-02 08:03:44 +00:00
parent d02f175afa
commit 92f62f91c1
7 changed files with 556 additions and 18 deletions
@@ -0,0 +1,90 @@
+let OutputFileArchiveManager
+const archiver = require('archiver')
+const OutputCacheManager = require('./OutputCacheManager')
+const OutputFileFinder = require('./OutputFileFinder')
+const Settings = require('@overleaf/settings')
+const { open } = require('node:fs/promises')
+const path = require('node:path')
+const { NotFoundError } = require('./Errors')
+
+function getContentDir(projectId, userId) {
+  let subDir
+  if (userId != null) {
+    subDir = `${projectId}-${userId}`
+  } else {
+    subDir = projectId
+  }
+  return `${Settings.path.outputDir}/${subDir}/`
+}
+
+module.exports = OutputFileArchiveManager = {
+  async archiveFilesForBuild(projectId, userId, build, files = []) {
+    const contentDir = getContentDir(projectId, userId)
+
+    const validFiles = await (files.length > 0
+      ? this._getRequestedOutputFiles(projectId, userId, build, files)
+      : this._getAllOutputFiles(projectId, userId, build))
+
+    const archive = archiver('zip')
+
+    const missingFiles = files.filter(
+      file => !validFiles.includes(path.basename(file))
+    )
+
+    for (const file of validFiles) {
+      try {
+        const fileHandle = await open(
+          `${contentDir}${OutputCacheManager.path(build, file)}`
+        )
+        const fileStream = fileHandle.createReadStream()
+        archive.append(fileStream, { name: file })
+      } catch (error) {
+        missingFiles.push(file)
+      }
+    }
+
+    if (missingFiles.length > 0) {
+      archive.append(missingFiles.join('\n'), {
+        name: 'missing_files.txt',
+      })
+    }
+
+    await archive.finalize()
+
+    return archive
+  },
+
+  async _getAllOutputFiles(projectId, userId, build) {
+    const contentDir = getContentDir(projectId, userId)
+
+    try {
+      const { outputFiles } = await OutputFileFinder.promises.findOutputFiles(
+        [],
+        `${contentDir}${OutputCacheManager.path(build, '.')}`
+      )
+
+      return outputFiles.map(({ path }) => path)
+    } catch (error) {
+      if (
+        error.code === 'ENOENT' ||
+        error.code === 'ENOTDIR' ||
+        error.code === 'EACCES'
+      ) {
+        throw new NotFoundError('Output files not found')
+      }
+      throw error
+    }
+  },
+
+  async _getRequestedOutputFiles(projectId, userId, build, files) {
+    const outputFiles = new Set(
+      await OutputFileArchiveManager._getAllOutputFiles(
+        projectId,
+        userId,
+        build
+      )
+    )
+
+    return files.filter(file => outputFiles.has(file))
+  },
+}