Docs Menu

GridFS

On this page

In this guide, you can learn how to store and retrieve large files in MongoDB using GridFS. GridFS is a specification implemented by the driver that describes how to split files into chunks when storing them and reassemble them when retrieving them. The driver implementation of GridFS is an abstraction that manages the operations and organization of the file storage.

You should use GridFS if the size of your files exceed the BSON document size limit of 16MB. For more detailed information on whether GridFS is suitable for your use case, see the GridFS server manual page.

See the following sections that describe GridFS operations and how to perform them:

GridFS organizes files in a bucket, a group of MongoDB collections that contain the chunks of files and information describing them. The bucket contains the following collections, named using the convention defined in the GridFS specification:

  • The chunks collection stores the binary file chunks.
  • The files collection stores the file metadata.

When you create a new GridFS bucket, the driver creates the preceding collections, prefixed with the default bucket name fs, unless you specify a different name. The driver also creates an index on each collection to ensure efficient retrieval of the files and related metadata. The driver only creates the GridFS bucket on the first write operation if it does not already exist. The driver only creates indexes if they do not exist and when the bucket is empty. For more information on GridFS indexes, see the server manual page on GridFS Indexes.

When storing files with GridFS, the driver splits the files into smaller chunks, each represented by a separate document in the chunks collection. It also creates a document in the files collection that contains a file id, file name, and other file metadata. You can upload the file from memory or from a stream. See the following diagram to see how GridFS splits the files when uploaded to a bucket.

A diagram that shows how GridFS uploads a file to a bucket

When retrieving files, GridFS fetches the metadata from the files collection in the specified bucket and uses the information to reconstruct the file from documents in the chunks collection. You can read the file into memory or output it to a stream.

To store or retrieve files from GridFS, create a bucket or get a reference to an existing one on a MongoDB database. Call the GridFSBuckets.create() helper method with a MongoDatabase instance as the parameter to instantiate a GridFSBucket. You can use the GridFSBucket instance to call read and write operations on the files in your bucket.

MongoDatabase database = mongoClient.getDatabase("mydb");
GridFSBucket gridFSBucket = GridFSBuckets.create(database);

To create or reference a bucket with a custom name other than the default name fs, pass your bucket name as the second parameter to the create() method as shown below:

GridFSBucket gridFSBucket = GridFSBuckets.create(database, "myCustomBucket");
Note

When you call create(), MongoDB does not create the bucket if it does not exist. Instead, MongoDB creates the bucket as necessary such as when you upload your first file.

For more information on the classes and methods mentioned in this section, see the following API Documentation:

To store a file in a GridFS bucket, you can either upload it from an instance of InputStream or write its data to a GridFSUploadStream.

For either upload process, you can specify configuration information such as file chunk size and other field/value pairs to store as metadata. Set this information on an instance of GridFSUploadOptions as shown in the following code snippet:

GridFSUploadOptions options = new GridFSUploadOptions()
.chunkSizeBytes(1048576) // 1MB chunk size
.metadata(new Document("myField", "myValue"));

See the GridFSUploadOptions API Documentation for more information.

This section shows you how to upload a file to a GridFS bucket using an input stream. The following code example shows how you can use a FileInputStream to read data from a file in your filesystem and upload it to GridFS by performing the following operations:

  • Read from the filesystem using a FileInputStream.
  • Set the chunk size using GridFSUploadOptions.
  • Set a custom metadata field called type to the value "zip archive".
  • Upload a file called project.zip, specifying the GridFS file name as "myProject.zip".
String filePath = "/path/to/project.zip";
try (InputStream streamToUploadFrom = new FileInputStream(filePath) ) {
GridFSUploadOptions options = new GridFSUploadOptions()
.chunkSizeBytes(1048576)
.metadata(new Document("type", "zip archive"));
ObjectId fileId = gridFSBucket.uploadFromStream("myProject.zip", streamToUploadFrom, options);
System.out.println("The file id of the uploaded file is: " + fileId.toHexString());
}

This code example prints the file id of the uploaded file after it is successfully saved in GridFS.

For more information, see the API Documentation on uploadFromStream().

This section shows you how to upload a file to a GridFS bucket by writing to an output stream. The following code example shows how you can write to a GridFSUploadStream to send data to GridFS by performing the following operations:

  • Read a file named "project.zip" from the filesystem into a byte array.
  • Set the chunk size using GridFSUploadOptions.
  • Set a custom metadata field called type to the value "zip archive".
  • Write the bytes to a GridFSUploadStream, assigning the file name "myProject.zip". The stream reads data into a buffer until it reaches the limit specified in the chunkSize setting, and inserts it as a new chunk in the chunks collection.
Path filePath = Paths.get("/path/to/project.zip");
byte[] data = Files.readAllBytes(filePath);
GridFSUploadOptions options = new GridFSUploadOptions()
.chunkSizeBytes(1048576)
.metadata(new Document("type", "zip archive"));
try (GridFSUploadStream uploadStream = gridFSBucket.openUploadStream("myProject.zip", options)) {
uploadStream.write(data);
uploadStream.flush();
System.out.println("The file id of the uploaded file is: " + uploadStream.getObjectId().toHexString());
} catch (Exception e) {
System.err.println("The file upload failed: " + e);
}

This code example prints the file id of the uploaded file after it is successfully saved in GridFS.

Note

If your file upload is not successful, the operation throws an exception and any uploaded chunks become orphaned chunks. An orphaned chunk is a document in a GridFS chunks collection that does not reference any file id in the GridFS files collection. File chunks can become orphaned chunks when an upload or delete operation is interrupted. To remove orphaned chunks, you must identify them using read operations and remove them using write operations.

For more information, see the API Documentation on GridFSUploadStream.

In this section, you can learn how to retrieve file metadata stored in the files collection of the GridFS bucket. The metadata contains information about the file it refers to, including:

  • The id of the file
  • The name of the file
  • The length/size of the file
  • The upload date and time
  • A metadata document in which you can store any other information

To retrieve files from a GridFS bucket, call the find() method on the GridFSBucket instance. The method returns a GridFSFindIterable from which you can access the results.

The following code example shows you how to retrieve and print file metadata from all your files in a GridFS bucket. Among the different ways that you can traverse the retrieved results from the GridFSFindIterable, the example uses a Consumer functional interface to print the following results:

gridFSBucket.find().forEach(new Consumer<GridFSFile>() {
@Override
public void accept(final GridFSFile gridFSFile) {
System.out.println(gridFSFile);
}
});

The next code example shows you how to retrieve and print the file names for all files that match the fields specified in the query filter. The example also calls sort() and limit() on the returned GridFSFindIterable to specify the order and maximum number of results:

Bson query = Filters.eq("metadata.type", "zip archive");
Bson sort = Sorts.ascending("filename");
gridFSBucket.find(query)
.sort(sort)
.limit(5)
.forEach(new Consumer<GridFSFile>() {
@Override
public void accept(final GridFSFile gridFSFile) {
System.out.println(gridFSFile);
}
});

Since metadata is an embedded document, the query filter specifies the type field within the document using dot notation. See the server manual guide on how to Query on Embedded/Nested Documents for more information.

For more information on the classes and methods mentioned in this section, see the following resources:

You can download a file from GridFS directly to a stream or you can save it to memory from a stream. You can specify the file to retrieve using either the file id or file name.

When your bucket contains multiple files that share the same file name, GridFS chooses the latest uploaded version of the file by default. To differentiate between each file that shares the same name, GridFS assigns files that share the same filename a revision number, ordered by upload time.

The original file revision number is "0" and the next most recent file revision number is "1". You can also specify negative values which correspond to the recency of the revision. The revision value "-1" references the most recent revision and "-2" references the next most recent revision.

The following code snippet shows how you can specify the second revision of a file in an instance of GridFSDownloadOptions:

GridFSDownloadOptions downloadOptions = new GridFSDownloadOptions().revision(1);

For more information on the enumeration of revisions, see the API documentation for GridFSDownloadOptions.

You can download a file in a GridFS bucket to an output stream. The following code example shows you how you can call the downloadToStream() method to download the first revision of the file named "myProject.zip" to an OutputStream.

GridFSDownloadOptions downloadOptions = new GridFSDownloadOptions().revision(0);
try (FileOutputStream streamToDownloadTo = new FileOutputStream("/tmp/myProject.zip")) {
gridFSBucket.downloadToStream("myProject.zip", streamToDownloadTo, downloadOptions);
streamToDownloadTo.flush();
}

For more information on this method, see the downloadToStream() API Documentation.

You can download a file in a GridFS bucket to memory by using an input stream. You can call the openDownloadStream() method on the GridFS bucket to open a GridFSDownloadStream, an input stream from which you can read the file.

The following code example shows you how to download a file referenced by the fileId variable into memory and print its contents as a string:

ObjectId fileId = new ObjectId("60345d38ebfcf47030e81cc9");
try (GridFSDownloadStream downloadStream = gridFSBucket.openDownloadStream(fileId)) {
int fileLength = (int) downloadStream.getGridFSFile().getLength();
byte[] bytesToWriteTo = new byte[fileLength];
downloadStream.read(bytesToWriteTo);
System.out.println(new String(bytesToWriteTo, StandardCharsets.UTF_8));
}

For more information on this method, see the openDownloadStream(). API Documentation.

You can update the name of a GridFS file in your bucket by calling the rename() method. You must specify the file to rename by its file id rather than its file name.

Note

The rename() method only supports updating the name of one file at a time. To rename multiple files, retrieve a list of files matching the file name from the bucket, extract the file id values from the files you want to rename, and pass each file id in separate calls to the rename() method.

The following code example shows you how to update the name of the file referenced by the fileId variable to "mongodbTutorial.zip":

ObjectId fileId = new ObjectId("60345d38ebfcf47030e81cc9");
gridFSBucket.rename(fileId, "mongodbTutorial.zip");

For more information on this method, see the rename() API Documentation.

You can remove a file from your GridFS bucket by calling the delete() method. You must specify the file by its file id rather than its file name.

Note

The delete() method only supports deleting one file at a time. To delete multiple files, retrieve the files from the bucket, extract the file id values from the files you want to delete, and pass each file id in separate calls to the delete() method.

The following code example shows you how to delete the file referenced by the fileId variable:

ObjectId fileId = new ObjectId("60345d38ebfcf47030e81cc9");
gridFSBucket.delete(fileId);

For more information on this method, see the delete() API Documentation.

The following code example shows you how to delete the default GridFS bucket on the database named "mydb". If you need to reference a custom named bucket, see the section of this guide on how to create a custom bucket.

MongoDatabase database = mongoClient.getDatabase("mydb");
GridFSBucket gridFSBucket = GridFSBuckets.create(database);
gridFSBucket.drop();

For more information on this method, see the drop() API Documentation.

Give Feedback
© 2021 MongoDB, Inc.

About

  • Careers
  • Legal Notices
  • Privacy Notices
  • Security Information
  • Trust Center
© 2021 MongoDB, Inc.