Uploading Files

There are two ways to upload a file to Canvas: either by sending the file data in a POST request, or by sending Canvas a publicly accessible HTTP or HTTPS URL to the file.

Uploading via POST

There are three steps to uploading a file directly via POST:

Notify Canvas that you are uploading a file with a POST to the file creation endpoint. This POST will include the file name and file size, along with information about what context the file is being created in.
Upload the file using the information returned in the first POST request.
On successful upload, the API will respond with a redirect. This redirect needs to be followed to complete the upload, or the file may not appear.

Step 1: Telling Canvas about the file upload and getting a token

The first step is to POST to the relevant API endpoint, depending one where you want to create the file. For example, to add a file to a course, you'd POST to /api/v1/courses/:course_id/files. Or to upload a file as part of a student homework submission, as the student you'd POST to /api/v1/courses/:course_id/assignments/:assignment_id/submissions/self/files.

Arguments:

name: The filename of the file. Any UTF-8 name is allowed. Path components such as `/` and `\` will be treated as part of the filename, not a path to a sub-folder.
size: The size of the file, in bytes. This field is recommended, as it will let you find out if there's a quota issue before uploading the raw file.
content_type: The content type of the file. If not given, it will be guessed based on the file extension.
parent_folder_id: The id of the folder to store the file in. If this and parent_folder_path are sent an error will be returned. If neither is given, a default folder will be used.
parent_folder_path: The path of the folder to store the file in. The path separator is the forward slash `/`, never a back slash. The folder will be created if it does not already exist. This parameter only applies to file uploads in a context that has folders, such as a user, a course, or a group. If this and parent_folder_id are sent an error will be returned. If neither is given, a default folder will be used.
folder: [deprecated] Use parent_folder_path instead.
on_duplicate: How to handle duplicate filenames. If `overwrite`, then this file upload will overwrite any other file in the folder with the same name. If `rename`, then this file will be renamed if another file in the folder exists with the given name. If no parameter is given, the default is `overwrite`. This doesn't apply to file uploads in a context that doesn't have folders.

Example Request:

curl 'https://<canvas>/api/v1/users/self/files' \
     -F 'name=profile_pic.jpg' \
     -F 'size=302185' \
     -F 'content_type=image/jpeg' \
     -F 'parent_folder_path=my_files/section1' \
     -H "Authorization: Bearer <token>"

Example Response:

{
  "upload_url": "https://some-bucket.s3.amazonaws.com/",
  "upload_params": {
    "key": "/users/1234/files/profile_pic.jpg",
    "acl": "private",
    "Filename": "profile_pic.jpg",
    "AWSAccessKeyId": "some_id",
    "Policy": "some_opaque_string",
    "Signature": "another_opaque_string",
    "Content-Type": "image/jpeg"
  }
}

At this point, the file object has been created in Canvas in a "pending" state, with no content. It will not appear in any listings in the UI until the next two steps are completed.

Step 2: Upload the file data to the URL given in the previous response

Using the data in the JSON response from Step 1, the application can now upload the actual file data, by POSTing a specially formulated request to the URL given in the upload_url field of the response.

Depending on how Canvas is configured, this upload URL might be another URL in the same domain, or a Amazon S3 bucket, or some other URL. In order to work with all Canvas installations, applications should be very careful to follow this documentation and not make any undocumented assumptions about the upload workflow.

This second request must be POSTed as a multipart/form-data request to accomodate the file data. The parameters POSTed with this request come directly from the upload_params part of the JSON response in Step 1.

The only addition is the file parameter which must be posted as the last parameter following all the others.

Example Request:

curl '<upload_url>' \ 
     -F 'key=/users/1234/files/profile_pic.jpg' \ 
     -F 'acl=private' \ 
     -F 'Filename=profile_pic.jpg' \ 
     -F 'AWSAccessKeyId=some_id' \ 
     -F 'Policy=some_opaque_string' \ 
     -F 'Signature=another_opaque_string' \ 
     -F 'Content-Type=image/jpeg' \ 
     -F '<any other fields returned in upload_params>' \
     -F 'file=@my_local_file.jpg'

The access token is not sent with this request.

Example Response:

HTTP/1.1 301 Moved Permanently
Location: https://<canvas>/api/v1/s3_success/1234?uuid=ABCDE

IMPORTANT: The request is signed, and will be denied if any parameters from the upload_params response are added, removed or modified. The parameters in upload_params may vary over time, and between Canvas installs. It's important for the application to copy over all of the parameters, and not rely on the names or values of the params for any functionality.

This example assumes there is a file called my_local_file.jpg in the current directory.

Step 3: Confirm the upload's success

If Step 2 is successful, the response will be a 3XX redirect with a Location header set as normal. The application needs to perform a POST to this location in order to complete the upload, otherwise the new file may not be marked as available. This request is back against Canvas again, and needs to be authenticated using the normal API access token authentication.

Example Request:

curl -X POST '<Location>' \
     -H 'Content-Length: 0' \
     -H "Authorization: Bearer <token>"

Example Response:

{
  'id': 1234,
  'url': '...url to download the file...',
  'content-type': 'image/jpeg',
  'display_name': 'profile_pic.jpg',
  'size': 302185
}

Uploading via URL

If the file is accessible through a public HTTP or HTTPS URL, Canvas can download it directly. However, Canvas downloads the file in the background, so unlike the POST workflow above, the file won't be immediately available when the API call completes. Applications can poll the API to check on the status of the upload.

Step 1: Posting the file URL to Canvas

The first step is the same as with the "Uploading via POST" flow above, with the addition of one new parameter:

url: The full URL to the file to be uploaded. This URL must be accessible by Canvas.

Example Request:

curl 'https://<canvas>/api/v1/users/self/files' \
     -F 'url=http://example.com/my_pic.jpg' \
     -F 'name=profile_pic.jpg' \
     -F 'size=302185' \
     -F 'content_type=image/jpeg' \
     -F 'parent_folder_path=my_files/section1' \
     -H "Authorization: Bearer <token>"

Example Response:

{
  'id': 1234,
  'upload_status': 'pending',
  'status_url' '...url to check status...'
}

Canvas will return a status_url parameter, which is a Canvas API endpoint that the application can periodically poll to check on the status of the upload. Note that the id given in the response will be treated like a deleted file, until the upload completes successfully.

Step 2: Check the status_url to see when the upload is complete

If the application needs to know the outcome of the upload, it can check the status_url.

Example Request:

curl 'https://<canvas>/api/v1/files/1234/5678/status' \
     -H "Authorization: Bearer <token>"

When still pending:

{
  'upload_status': 'pending'
}

When complete:

{
  'upload_status': 'ready',
  // This is the normal attachment JSON response object
  'attachment': {
    'id': 1234,
    'url': '...url to download the file...',
    'content-type': 'image/jpeg',
    'display_name': 'profile_pic.jpg',
    'size': 302185
  }
}

On error:

{
  'upload_status': 'errored',
  'message': 'Invalid response code, expected 200 got 404'
}

Canvas LMS - REST API and Extensions Documentation

Uploading Files

Uploading via POST

Step 1: Telling Canvas about the file upload and getting a token

Step 2: Upload the file data to the URL given in the previous response

Step 3: Confirm the upload's success

Uploading via URL

Step 1: Posting the file URL to Canvas

Step 2: Check the status_url to see when the upload is complete