[monorepo] read empty content from its hash, in a shared blob store base (#36831)

The hash of empty content is a property of the content rather than of anything stored: every project's empty file has it, whether or not a blob was ever written under it, and a file may reference it before anything has. Eight blob stores were each paying a lookup -- a database query, an HTTP request, a read of a cache that offline mode may be the only thing standing in for -- to be told what the hash already says, and clsi had already reinvented the answer on its own. overleaf-editor-core gains BlobStoreBase, which answers getString for that hash and delegates the rest to the fetchString its subclasses provide. getObject comes with it: four stores spelled out the same JSON.parse of getString, and a fifth wrapped it in a log line. Adopted by history-v1's BlobStore, BackupBlobStore, HashCheckBlobStore and recover_zip's RecoveryBlobStore, by project-history's BlobStore, by clsi's BlobStore, and by web's HistoryBlobStore, SimpleBlobStore and FrontendBlobStore. history-v1 keeps its own answers in getStream, getBlob and getBlobs, which no other store implements. Writing and copying a blob are untouched, so a project that wrote empty content still has a blob for it and a read answers the same either way. Two paths could not handle a project holding a tracked change or a comment at all. Loading such a file needs the ranges blob its rangesHash points at, and hashing one again needs to write it back, so a store missing getObject or putObject fails outright: - recover_zip died on `blobStore.getObject is not a function`, making every such project unrecoverable. The test covering it asserts a tracked delete is absent from the zip, which the script already asked for and never reached. - GET /projects/:project_id/latest/hashed_content returned 500 from the same cause, then would have hit a missing putObject in snapshot.store. HashCheckBlobStore becomes a class over the base and gains putObject. Co-authored-by: Claude Opus 5 <noreply@anthropic.com> GitOrigin-RevId: c84f62465beff6cf77ffac26aa772984e1b554cb
2026-11-12 13:13:45 +01:00 · 2026-09-17 12:06:30 +00:00
parent 7608d738a6
commit 6caa5fc3a3
18 changed files with 646 additions and 71 deletions
@@ -1,5 +1,6 @@
 import { expect } from 'chai'
 import fetchMock from 'fetch-mock'
+import { File } from 'overleaf-editor-core'
 import { ProjectSnapshot } from '@/infrastructure/project-snapshot'

 describe('ProjectSnapshot', function () {
@@ -714,4 +715,90 @@ describe('ProjectSnapshot', function () {
      )
    })
  })
+
+  describe('blob with the hash of empty content', function () {
+    // Nothing is stored under that hash, on the server or here, so the content is
+    // answered from the hash rather than fetched. A new empty doc references it before
+    // anything has written it, so a fetch would 404 and take the whole snapshot down.
+
+    afterEach(function () {
+      fetchMock.removeRoutes().clearHistory()
+    })
+
+    it('loads an empty doc without fetching the blob', async function () {
+      const emptyChunk = {
+        history: {
+          snapshot: { files: {} },
+          changes: [
+            {
+              operations: [
+                {
+                  pathname: 'fresh.tex',
+                  file: { hash: File.EMPTY_FILE_HASH, stringLength: 0 },
+                },
+              ],
+              timestamp: '2025-01-01T12:00:00.000Z',
+            },
+          ],
+        },
+        startVersion: 0,
+      }
+
+      fetchMock.post(`/project/${projectId}/flush`, 200)
+      fetchMock.getOnce(`/project/${projectId}/latest/history`, {
+        chunk: emptyChunk,
+      })
+      // Served only if the short circuit does not answer first, in which case the
+      // content below is what comes back.
+      fetchMock.get(
+        `/project/${projectId}/blob/${File.EMPTY_FILE_HASH}`,
+        'fetched from the server'
+      )
+
+      await snapshot.refresh()
+
+      expect(snapshot.getDocPaths()).to.have.members(['fresh.tex'])
+      expect(snapshot.getDocContents('fresh.tex')).to.equal('')
+    })
+
+    it('applies edits made on top of an empty doc', async function () {
+      const insertedText = '\\documentclass{article}\n'
+      const emptyChunk = {
+        history: {
+          snapshot: { files: {} },
+          changes: [
+            {
+              operations: [
+                {
+                  pathname: 'fresh.tex',
+                  file: { hash: File.EMPTY_FILE_HASH, stringLength: 0 },
+                },
+              ],
+              timestamp: '2025-01-01T12:00:00.000Z',
+            },
+            {
+              operations: [
+                { pathname: 'fresh.tex', textOperation: [insertedText] },
+              ],
+              timestamp: '2025-01-01T12:01:00.000Z',
+            },
+          ],
+        },
+        startVersion: 0,
+      }
+
+      fetchMock.post(`/project/${projectId}/flush`, 200)
+      fetchMock.getOnce(`/project/${projectId}/latest/history`, {
+        chunk: emptyChunk,
+      })
+      fetchMock.get(
+        `/project/${projectId}/blob/${File.EMPTY_FILE_HASH}`,
+        'fetched from the server'
+      )
+
+      await snapshot.refresh()
+
+      expect(snapshot.getDocContents('fresh.tex')).to.equal(insertedText)
+    })
+  })
 })
@@ -0,0 +1,73 @@
+import { beforeEach, describe, expect, it, vi } from 'vitest'
+import sinon from 'sinon'
+import { Readable } from 'node:stream'
+import { Blob } from 'overleaf-editor-core'
+
+const modulePath = '../../../../app/src/Features/History/HistoryBlobStore.mjs'
+
+const HASH = 'a1b2c3d4e5f60718293a4b5c6d7e8f9012345678'
+
+describe('HistoryBlobStore', function () {
+  beforeEach(async function (ctx) {
+    ctx.HistoryManager = {
+      promises: { requestBlob: sinon.stub() },
+    }
+    vi.doMock(
+      '../../../../app/src/Features/History/HistoryManager.mjs',
+      () => ({
+        default: ctx.HistoryManager,
+      })
+    )
+
+    const HistoryBlobStore = (await import(modulePath)).default
+    ctx.blobStore = new HistoryBlobStore('history-id')
+  })
+
+  describe('getString', function () {
+    it('reads the blob of the given hash', async function (ctx) {
+      ctx.HistoryManager.promises.requestBlob.resolves({
+        stream: Readable.from(['hello ', 'world']),
+      })
+
+      expect(await ctx.blobStore.getString(HASH)).to.equal('hello world')
+      expect(ctx.HistoryManager.promises.requestBlob).to.have.been.calledWith(
+        'history-id',
+        HASH
+      )
+    })
+
+    it('passes on errors from history', async function (ctx) {
+      ctx.HistoryManager.promises.requestBlob.rejects(new Error('boom'))
+
+      await expect(ctx.blobStore.getString(HASH)).to.be.rejectedWith('boom')
+    })
+
+    it('answers the hash of empty content without asking history', async function (ctx) {
+      // Nothing is stored under it, so history-v1 answers it without a lookup of its
+      // own: the round-trip to ask can only be slow or fail.
+      ctx.HistoryManager.promises.requestBlob.resolves({
+        stream: Readable.from(['fetched from history-v1']),
+      })
+
+      expect(await ctx.blobStore.getString(Blob.EMPTY_HASH)).to.equal('')
+    })
+
+    it('answers the hash of empty content when history is down', async function (ctx) {
+      ctx.HistoryManager.promises.requestBlob.rejects(new Error('boom'))
+
+      expect(await ctx.blobStore.getString(Blob.EMPTY_HASH)).to.equal('')
+    })
+  })
+
+  describe('getObject', function () {
+    it('parses the blob of the given hash as JSON', async function (ctx) {
+      ctx.HistoryManager.promises.requestBlob.resolves({
+        stream: Readable.from(['{"comments":[]}']),
+      })
+
+      expect(await ctx.blobStore.getObject(HASH)).to.deep.equal({
+        comments: [],
+      })
+    })
+  })
+})