Google Cloud Platform

Blobstore Python API Overview

Python |Java |PHP |Go

The Blobstore API allows your application to serve data objects, called blobs, that are much larger than the size allowed for objects in the Datastore service. Blobs are useful for serving large files, such as video or image files, and for allowing users to upload large data files. Blobs are created by uploading a file through an HTTP request. Typically, your applications will do this by presenting a form with a file upload field to the user. When the form is submitted, the Blobstore creates a blob from the file's contents and returns an opaque reference to the blob, called a blob key, which you can later use to serve the blob. The application can serve the complete blob value in response to a user request, or it can read the value directly using a streaming file-like interface.

  1. Introducing the Blobstore
  2. Using the Blobstore
  3. Uploading a blob
  4. Serving a blob
  5. Complete sample application
  6. Using the Images service with the Blobstore
  7. Using the Blobstore API with Google Cloud Storage
  8. Using BlobReader
  9. Making asynchronous requests
  10. Quotas and limits

Introducing the Blobstore

Google App Engine includes the Blobstore service, which allows applications to serve data objects limited only by the amount of data that can be uploaded or downloaded over a single HTTP connection. These objects are called Blobstore values, or blobs. Blobstore values are served as responses from request handlers and are created as uploads via web forms. Applications do not create blob data directly; instead, blobs are created indirectly, by a submitted web form or other HTTP POST request. Blobstore values can be served to the user, or accessed by the application in a file-like stream, using the Blobstore API.

To prompt a user to upload a Blobstore value, your application presents a web form with a file upload field. The application generates the form's action URL by calling the Blobstore API. The user's browser uploads the file directly to the Blobstore via the generated URL. Blobstore then stores the blob, rewrites the request to contain the blob key, and passes it to a path in your application. A request handler at that path in your application can perform additional form processing.

To serve a blob, your application sets a header on the outgoing response, and App Engine replaces the response with the blob value.

Blobs can't be modified after they're created, though they can be deleted. Each blob has a corresponding blob info record, stored in the datastore, that provides details about the blob, such as its creation time and content type. You can use the blob key to fetch blob info records and query their properties.

An application can read a Blobstore value a portion at a time using an API call. The size of the portion can be up to the maximum size of an API return value. This size is a little less than 32 megabytes, represented in Python by the constant google.appengine.ext.blobstore.MAX_BLOB_FETCH_SIZE. An application cannot create or modify Blobstore values except through files uploaded by the user.

Using the Blobstore

Applications can use the Blobstore to accept large files as uploads from users and to serve those files. Files are called blobs once they're uploaded. Applications don't access blobs directly. Instead, applications work with blobs through blob info entities (represented by the BlobInfo class) in the datastore.

The user creates a blob by submitting an HTML form that includes one or more file input fields. Your application calls create_upload_url() to get the destination (action) of this form, passing the function a URL path of a handler in your application. When the user submits the form, the user's browser uploads the specified files directly to the Blobstore. The Blobstore rewrites the user's request and stores the uploaded file data, replacing the uploaded file data with one or more corresponding blob keys, then passes the rewritten request to the handler at the URL path you provided to create_upload_url(). This handler can do additional processing based on the blob key.

The application can read portions of a Blobstore value using a file-like streaming interface. See The BlobReader Class.

Uploading a blob

To create and upload a blob, follow this procedure:

1. Create an upload URL

Call blobstore.create_upload_url() to create an upload URL for the form that the user will fill out, passing the application path to load when the POST of the form is completed.

There is an asynchronous version, create_upload_url_async(). It allows your application code to continue running while Blobstore generates the upload URL.

2. Create an upload form

The form must include a file upload field, and the form's enctype must be set to multipart/form-data. When the user submits the form, the POST is handled by the Blobstore API, which creates the blob. The API also creates an info record for the blob and stores the record in the datastore, and passes the rewritten request to your application on the given path as a blob key.

3. Implement upload handler

In this handler, you can store the blob key with the rest of your application's data model. The blob key itself remains accessible from the blob info entity in the datastore. Note that after the user submits the form and your handler is called, the blob has already been saved and the blob info added to the datastore. If your application doesn't want to keep the blob, you should delete the blob immediately to prevent it from becoming orphaned:

The webapp framework provides the blobstore_handlers.BlobstoreUploadHandler upload handler class to help you parse the form data. For more information, see the reference for BlobstoreUploadHandler.

When the Blobstore rewrites the user's request, the MIME parts of the uploaded files have their bodies emptied, and the blob key is added as a MIME part header. All other form fields and parts are preserved and passed to the upload handler. If you don't specify a content type, the Blobstore will try to infer it from the file extension. If no content type can be determined, the newly created blob is assigned content type application/octet-stream.

Serving a blob

To serve blobs, you must include a blob download handler as a path in your application. The application serves a blob by setting a header on the outgoing response. The following sample uses the webapp framework. When using webapp, the handler should pass the blob key for the desired blob to self.send_blob(). In this example, the blob key is passed to the download handler as part of the URL. In practice, the download handler can get the blob key by any means you choose, such as through another method or user action.

The webapp framework provides the download handler class blobstore_handlers.BlobstoreDownloadHandler to help you parse the form data. For more information, see the reference for BlobstoreDownloadHandler.

Blobs can be served from any application URL. To serve a blob in your application, you put a special header in the response containing the blob key. App Engine replaces the body of the response with the content of the blob.

Blob byte ranges

The Blobstore supports serving part of a large value instead of the full value in response to a request. To serve a partial value, include the X-AppEngine-BlobRange header in the outgoing response. Its value is a standard HTTP byte range. The byte numbering is zero-based. A blank X-AppEngine-BlobRange instructs the API to ignore the range header and serve the full blob. Example ranges include:

  • 0-499 serves the first 500 bytes of the value (bytes 0 through 499, inclusive).
  • 500-999 serves 500 bytes starting with the 501st byte.
  • 500- serves all bytes starting with the 501st byte to the end of the value.
  • -500 serves the last 500 bytes of the value.

If the byte range is valid for the Blobstore value, the Blobstore sends a 206 Partial Content status code and the requested byte range to the client. If the range is not valid for the value, the Blobstore sends 416 Requested Range Not Satisfiable.

The Blobstore does not support multiple byte ranges in a single request (for example, 100-199,200-299), whether or not they overlap.

The webapp.blobstore_handlers.BlobstoreDownloadHandler class includes features for setting this header using provided byte indices, and for deriving the byte range automatically from a range header provided by the user.

Complete sample application

In the following sample application, the application's main URL loads the form that asks the user for a file to upload, and the upload handler immediately calls the download handler to serve the data. This is to simplify the sample application. In practice, you would probably not use the main URL to request upload data, nor would you immediately serve a blob you had just uploaded.

Using the Images service with the Blobstore

The Images service can use a Blobstore value as the source of a transformation. The source image can be as large as the maximum size for a Blobstore value. The Images service still returns the transformed image to the application, so the transformed image must be smaller than 32 megabytes. This is useful for making thumbnail images of large photographs uploaded by users.

For information on using the Images service with Blobstore values, see the Images Service documentation.

Using the Blobstore API with Google Cloud Storage

You can use the Blobstore API to store blobs in Cloud Storage instead of storing them in Blobstore. You need to set up a bucket as described in the Google Cloud Storage documentation and specify the bucket and filename in the blobstore.blobstore.create_upload_url gs_bucket_name parameter. In your upload handler, you need to process the returned FileInfo metadata and explicitly store the Google Cloud Storage filename needed to retrieve the blob later.

You can also serve Cloud Storage objects using the Blobstore API. The following code snippets shows how to do this:

"""A sample app that operates on GCS files with blobstore API."""

from __future__ import with_statement

import cloudstorage as gcs
import main
import webapp2

from google.appengine.ext import blobstore
from google.appengine.ext.webapp import blobstore_handlers

def CreateFile(filename):
  """Create a GCS file with GCS client lib.

    filename: GCS filename.

    The corresponding string blobkey for this GCS file.
  # Create a GCS file with GCS client.
  with, 'w') as f:

  # Blobstore API requires extra /gs to distinguish against blobstore files.
  blobstore_filename = '/gs' + filename
  # This blob_key works with blobstore APIs that do not expect a
  # corresponding BlobInfo in datastore.
  return blobstore.create_gs_key(blobstore_filename)

class GCSHandler(webapp2.RequestHandler):

  def get(self):
    self.response.headers['Content-Type'] = 'text/plain'
    gcs_filename = main.BUCKET + '/blobstore_demo'
    blob_key = CreateFile(gcs_filename)

    # Fetch data.
    self.response.write('Fetched data %s\n' %
                        blobstore.fetch_data(blob_key, 0, 2))

    # Delete files.

class GCSServingHandler(blobstore_handlers.BlobstoreDownloadHandler):

  def get(self):
    blob_key = CreateFile(main.BUCKET + '/blobstore_serving_demo')

app = webapp2.WSGIApplication([('/blobstore/ops', GCSHandler),
                               ('/blobstore/serve', GCSServingHandler)],

Using BlobReader

An application can read data from Blobstore values using an interface similar to a Python file object. This interface can start reading a value at any byte position, and uses multiple service calls and buffering, so an application can access the full size of the value despite the limit on the size of a single service call response.

The BlobReader class can take one of three values as an argument to its constructor:

The object implements the familiar file methods for reading the value. The application cannot modify the Blobstore value; file methods for writing are not implemented.

from google.appengine.ext import blobstore

# blob_key = ...

# Instantiate a BlobReader for a given Blobstore value.
blob_reader = blobstore.BlobReader(blob_key)

# Instantiate a BlobReader for a given Blobstore value, setting the
# buffer size to 1 MB.
blob_reader = blobstore.BlobReader(blob_key, buffer_size=1048576)

# Instantiate a BlobReader for a given Blobstore value, setting the
# initial read position.
blob_reader = blobstore.BlobReader(blob_key, position=4194304)

# Read the entire value into memory. This may take a while depending
# on the size of the value and the size of the read buffer, and is not
# recommended for large values.
value =

# Set the read position, then read 100 bytes.
data =

# Read the value, one line (up to and including a '\n' character) at a time.
for line in blob_reader:
    # ...

Making asynchronous requests

An application can call some Blobstore functions that work in the background. Blobstore carries out the request while the application does other things. To make the request, the application calls an asynchronous function. The function immediately returns an RPC object; this object represents the request. When the application needs the result of the request, it calls the RPC object's get_result() method.

If the service has not completed the request when the application calls get_result(), the method waits until the request is complete (or has reached the deadline, or an error occurs). The method returns the result object, or raises an exception if an error occurred while carrying out the request. For example, this code snippet

upload_url = blobstore.create_upload_url('/upload')
self.response.out.write("""<form action="%s" method="POST"
                           enctype="multipart/form-data">""" % upload_url)


upload_url_rpc = blobstore.create_upload_url_async('/upload')
upload_url = upload_url_rpc.get_result()
self.response.out.write("""<form action="%s" method="POST"
                           enctype="multipart/form-data">""" % upload_url)

In this example, the application carries out the slow_operation() code at the same time that Blobstore generates the upload URL.

Quotas and limits

Space used for Blobstore values contributes to the Stored Data (billable) quota. Blob info entities in the datastore count towards datastore-related limits.

For more information on system-wide safety quotas, see Quotas.

In addition to system-wide safety quotas, a limit applies specifically to the use of the Blobstore: the maximum size of Blobstore data that can be read by the application with one API call is 32 megabytes.