feat: add filter option to de-duplicate blocks in car files

packages/car/package.json

@@ -151,6 +151,7 @@
     "progress-events": "^1.0.0"
   },
   "devDependencies": {
+    "@helia/mfs": "^3.0.6",
     "@helia/unixfs": "^3.0.6",
     "@ipld/dag-pb": "^4.1.0",
     "aegir": "^43.0.1",

packages/car/src/index.ts

@@ -67,6 +67,7 @@ import type { DAGWalker } from '@helia/interface'
 import type { GetBlockProgressEvents, PutManyBlocksProgressEvents } from '@helia/interface/blocks'
 import type { CarReader } from '@ipld/car'
 import type { AbortOptions } from '@libp2p/interfaces'
+import type { Filter } from '@libp2p/utils/filters'
 import type { Blockstore } from 'interface-blockstore'
 import type { CID } from 'multiformats/cid'
 import type { ProgressOptions } from 'progress-events'
@@ -76,6 +77,10 @@ export interface CarComponents {
   dagWalkers: Record<number, DAGWalker>
 }
 
+interface ExportCarOptions {
+  blockFilter: Filter
+}
+
 /**
  * The Car interface provides operations for importing and exporting Car files
  * from Helia's underlying blockstore.
@@ -129,7 +134,7 @@ export interface Car {
    * await eventPromise
    * ```
    */
-  export(root: CID | CID[], writer: Pick<CarWriter, 'put' | 'close'>, options?: AbortOptions & ProgressOptions<GetBlockProgressEvents>): Promise<void>
+  export(root: CID | CID[], writer: Pick<CarWriter, 'put' | 'close'>, options?: ExportCarOptions & AbortOptions & ProgressOptions<GetBlockProgressEvents>): Promise<void>
 
   /**
    * Returns an AsyncGenerator that yields CAR file bytes.
@@ -170,7 +175,7 @@ class DefaultCar implements Car {
     ))
   }
 
-  async export (root: CID | CID[], writer: Pick<CarWriter, 'put' | 'close'>, options?: AbortOptions & ProgressOptions<GetBlockProgressEvents>): Promise<void> {
+  async export (root: CID | CID[], writer: Pick<CarWriter, 'put' | 'close'>, options?: ExportCarOptions & AbortOptions & ProgressOptions<GetBlockProgressEvents>): Promise<void> {
     const deferred = defer<Error | undefined>()
     const roots = Array.isArray(root) ? root : [root]
 
@@ -189,6 +194,14 @@ class DefaultCar implements Car {
     for (const root of roots) {
       void queue.add(async () => {
         await this.#walkDag(root, queue, async (cid, bytes) => {
+          // check if duplicate blocks should be skipped
+          if (options?.blockFilter != null) {
+            // skip blocks that have already been written
+            if (options.blockFilter.has(cid.toString())) {
+              return
+            }
+            options.blockFilter.add(cid.toString())
+          }
           await writer.put({ cid, bytes })
         }, options)
       })
@@ -203,7 +216,7 @@ class DefaultCar implements Car {
     }
   }
 
-  async * stream (root: CID | CID[], options?: AbortOptions & ProgressOptions<GetBlockProgressEvents>): AsyncGenerator<Uint8Array, void, undefined> {
+  async * stream (root: CID | CID[], options?: ExportCarOptions & AbortOptions & ProgressOptions<GetBlockProgressEvents>): AsyncGenerator<Uint8Array, void, undefined> {
     const { writer, out } = CarWriter.create(root)
 
     // has to be done async so we write to `writer` and read from `out` at the

packages/car/test/index.spec.ts

@@ -1,9 +1,12 @@
 /* eslint-env mocha */
 
+import { mfs } from '@helia/mfs'
 import { type UnixFS, unixfs } from '@helia/unixfs'
 import { CarReader } from '@ipld/car'
+import { createScalableCuckooFilter } from '@libp2p/utils/filters'
 import { expect } from 'aegir/chai'
 import { MemoryBlockstore } from 'blockstore-core'
+import { MemoryDatastore } from 'datastore-core'
 import { fixedSize } from 'ipfs-unixfs-importer/chunker'
 import toBuffer from 'it-to-buffer'
 import { car, type Car } from '../src/index.js'
@@ -115,4 +118,55 @@ describe('import/export car file', () => {
     expect(await toBuffer(u.cat(cid2))).to.equalBytes(fileData2)
     expect(await toBuffer(u.cat(cid3))).to.equalBytes(fileData3)
   })
+
+  it('exports a car file without duplicates', async () => {
+    const otherBlockstore = new MemoryBlockstore()
+    const otherUnixFS = unixfs({ blockstore: otherBlockstore })
+    const otherDatastore = new MemoryDatastore()
+    const otherMFS = mfs({ blockstore: otherBlockstore, datastore: otherDatastore })
+    const otherCar = car({ blockstore: otherBlockstore, dagWalkers })
+
+    await otherMFS.mkdir('/testDups')
+    await otherMFS.mkdir('/testDups/sub')
+
+    const sourceCid = await otherUnixFS.addBytes(smallFile)
+    await otherMFS.cp(sourceCid, '/testDups/a.smallfile')
+    await otherMFS.cp(sourceCid, '/testDups/sub/b.smallfile')
+
+    const rootObject = await otherMFS.stat('/testDups/')
+    const rootCid = rootObject.cid
+
+    const writer = memoryCarWriter(rootCid)
+    const blockFilter = createScalableCuckooFilter(5)
+    await otherCar.export(rootCid, writer, {
+      blockFilter
+    })
+
+    const carBytes = await writer.bytes()
+    expect(carBytes.length).to.equal(349)
+  })
+
+  it('exports a car file with duplicates', async () => {
+    const otherBlockstore = new MemoryBlockstore()
+    const otherUnixFS = unixfs({ blockstore: otherBlockstore })
+    const otherDatastore = new MemoryDatastore()
+    const otherMFS = mfs({ blockstore: otherBlockstore, datastore: otherDatastore })
+    const otherCar = car({ blockstore: otherBlockstore, dagWalkers })
+
+    await otherMFS.mkdir('/testDups')
+    await otherMFS.mkdir('/testDups/sub')
+
+    const sourceCid = await otherUnixFS.addBytes(smallFile)
+    await otherMFS.cp(sourceCid, '/testDups/a.smallfile')
+    await otherMFS.cp(sourceCid, '/testDups/sub/b.smallfile')
+
+    const rootObject = await otherMFS.stat('/testDups/')
+    const rootCid = rootObject.cid
+
+    const writer = memoryCarWriter(rootCid)
+    await otherCar.export(rootCid, writer)
+
+    const carBytes = await writer.bytes()
+    expect(carBytes.length).to.equal(399)
+  })
 })