§Handling file upload
§Uploading files in a form using multipart/form-data
The standard way to upload files in a web application is to use a form with a special multipart/form-data
encoding, which lets you mix standard form data with file attachment data.
Note: The HTTP method used to submit the form must be
POST
(notGET
).
Start by writing an HTML form:
@helper.form(action = routes.ScalaFileUploadController.upload, 'enctype -> "multipart/form-data") {
<input type="file" name="picture">
<p>
<input type="submit">
</p>
}
Add a CSRF token to the form, unless you have the CSRF filter disabled. The CSRF filter checks the multi-part form in the order the fields are listed, so put the CSRF token before the file input field. This improves efficiency and avoids a token-not-found error if the file size exceeds play.filters.csrf.body.bufferSize
.
Now define the upload
action using a multipartFormData
body parser:
def upload = Action(parse.multipartFormData) { request =>
request.body
.file("picture")
.map { picture =>
// only get the last part of the filename
// otherwise someone can send a path like ../../home/foo/bar.txt to write to other files on the system
val filename = Paths.get(picture.filename).getFileName
picture.ref.moveTo(Paths.get(s"/tmp/picture/$filename"), replace = true)
Ok("File uploaded")
}
.getOrElse {
Redirect(routes.ScalaFileUploadController.index).flashing("error" -> "Missing file")
}
}
The ref
attribute give you a reference to a TemporaryFile
. This is the default way the multipartFormData
parser handles file upload.
Note: As always, you can also use the
anyContent
body parser and retrieve it asrequest.body.asMultipartFormData
.
At last, add a POST
router
POST / controllers.ScalaFileUploadController.upload()
§Direct file upload
Another way to send files to the server is to use Ajax to upload the file asynchronously in a form. In this case the request body will not have been encoded as multipart/form-data
, but will just contain the plain file content.
In this case we can just use a body parser to store the request body content in a file. For this example, let’s use the temporaryFile
body parser:
def upload = Action(parse.temporaryFile) { request =>
request.body.moveTo(Paths.get("/tmp/picture/uploaded"), replace = true)
Ok("File uploaded")
}
§Writing your own body parser
If you want to handle the file upload directly without buffering it in a temporary file, you can just write your own BodyParser
. In this case, you will receive chunks of data that you are free to push anywhere you want.
If you want to use multipart/form-data
encoding, you can still use the default multipartFormData
parser by providing a FilePartHandler[A]
and using a different Sink to accumulate data. For example, you can use a FilePartHandler[File]
rather than a TemporaryFile by specifying an Accumulator(fileSink)
:
type FilePartHandler[A] = FileInfo => Accumulator[ByteString, FilePart[A]]
def handleFilePartAsFile: FilePartHandler[File] = {
case FileInfo(partName, filename, contentType) =>
val perms = java.util.EnumSet.of(OWNER_READ, OWNER_WRITE)
val attr = PosixFilePermissions.asFileAttribute(perms)
val path = JFiles.createTempFile("multipartBody", "tempFile", attr)
val file = path.toFile
val fileSink = FileIO.toPath(path)
val accumulator = Accumulator(fileSink)
accumulator.map {
case IOResult(count, status) =>
FilePart(partName, filename, contentType, file)
}(ec)
}
def uploadCustom = Action(parse.multipartFormData(handleFilePartAsFile)) { request =>
val fileOption = request.body.file("name").map {
case FilePart(key, filename, contentType, file) =>
file.toPath
}
Ok(s"File uploaded: $fileOption")
}
§Cleaning up temporary files
Uploading files uses a TemporaryFile
API which relies on storing files in a temporary filesystem, accessible through the ref
attribute. All TemporaryFile
references come from a TemporaryFileCreator
trait, and the implementation can be swapped out as necessary, and there’s now an atomicMoveWithFallback
method that uses StandardCopyOption.ATOMIC_MOVE
if available.
Uploading files is an inherently dangerous operation, because unbounded file upload can cause the filesystem to fill up – as such, the idea behind TemporaryFile
is that it’s only in scope at completion and should be moved out of the temporary file system as soon as possible. Any temporary files that are not moved are deleted.
However, under certain conditions, garbage collection does not occur in a timely fashion. As such, there’s also a play.api.libs.Files.TemporaryFileReaper
that can be enabled to delete temporary files on a scheduled basis using the Akka scheduler, distinct from the garbage collection method.
The reaper is disabled by default, and is enabled through application.conf
:
play.temporaryFile {
reaper {
enabled = true
initialDelay = "5 minutes"
interval = "30 seconds"
olderThan = "30 minutes"
}
}
The above configuration will delete files that are more than 30 minutes old, using the “olderThan” property. It will start the reaper five minutes after the application starts, and will check the filesystem every 30 seconds thereafter. The reaper is not aware of any existing file uploads, so protracted file uploads may run into the reaper if the system is not carefully configured.