Uploaded Files and Upload Handlers

Uploaded files

  • class UploadedFile[源代码]

  • During file uploads, the actual file data is stored in request.FILES. Each entry in this dictionary is anUploadedFile object (or a subclass) — a simple wrapper around an uploadedfile. You'll usually use one of these methods to access the uploaded content:

  • UploadedFile.read()

  • Read the entire uploaded data from the file. Be careful with this method:if the uploaded file is huge it can overwhelm your system if you try toread it into memory. You'll probably want to use chunks() instead; seebelow.

  • UploadedFile.multiplechunks(_chunk_size=None)

  • Returns True if the uploaded file is big enough to require reading inmultiple chunks. By default this will be any file larger than 2.5 megabytes,but that's configurable; see below.

  • UploadedFile.chunks(chunk_size=None)

  • A generator returning chunks of the file. If multiple_chunks() isTrue, you should use this method in a loop instead of read().

In practice, it's often easiest simply to use chunks() all the time.Looping over chunks() instead of using read() ensures that largefiles don't overwhelm your system's memory.

Here are some useful attributes of UploadedFile:

  • UploadedFile.name

  • The name of the uploaded file (e.g. my_file.txt).

  • UploadedFile.size

  • The size, in bytes, of the uploaded file.

  • UploadedFile.content_type

  • The content-type header uploaded with the file (e.g. text/plain_or _application/pdf). Like any data supplied by the user, youshouldn't trust that the uploaded file is actually this type. You'll stillneed to validate that the file contains the content that the content-typeheader claims — "trust but verify."

  • UploadedFile.content_type_extra

  • A dictionary containing extra parameters passed to the content-typeheader. This is typically provided by services, such as Google App Engine,that intercept and handle file uploads on your behalf. As a result yourhandler may not receive the uploaded file content, but instead a URL orother pointer to the file. (see RFC 2388 section 5.3).

  • UploadedFile.charset

  • For text/* content-types, the character set (i.e. utf8)supplied by the browser. Again, "trust but verify" is the best policy here.

注解

Like regular Python files, you can read the file line-by-line simply byiterating over the uploaded file:

  1. for line in uploadedfile:
  2.     do_something_with(line)

Lines are split using universal newlines. The following are recognizedas ending a line: the Unix end-of-line convention '\n', the Windowsconvention '\r\n', and the old Macintosh convention '\r'.

Subclasses of UploadedFile include:

  • class TemporaryUploadedFile[源代码]

  • A file uploaded to a temporary location (i.e. stream-to-disk). This classis used by theTemporaryFileUploadHandler. Inaddition to the methods from UploadedFile, it has one additionalmethod:

  • TemporaryUploadedFile.temporary_file_path()[源代码]

  • Returns the full path to the temporary uploaded file.

  • class InMemoryUploadedFile[源代码]

  • A file uploaded into memory (i.e. stream-to-memory). This class is usedby the MemoryFileUploadHandler.

Built-in upload handlers

Together the MemoryFileUploadHandler andTemporaryFileUploadHandler provide Django's default file uploadbehavior of reading small files into memory and large ones onto disk. Theyare located in django.core.files.uploadhandler.

  • class MemoryFileUploadHandler[源代码]

  • File upload handler to stream uploads into memory (used for small files).

  • class TemporaryFileUploadHandler[源代码]

  • Upload handler that streams data into a temporary file usingTemporaryUploadedFile.

Writing custom upload handlers

  • class FileUploadHandler[源代码]
  • All file upload handlers should be subclasses ofdjango.core.files.uploadhandler.FileUploadHandler. You can define uploadhandlers wherever you wish.

Required methods

Custom file upload handlers must define the following methods:

  • FileUploadHandler.receivedata_chunk(_raw_data, start)[源代码]
  • Receives a "chunk" of data from the file upload.

raw_data is a bytestring containing the uploaded data.

start is the position in the file where this raw_data chunkbegins.

The data you return will get fed into the subsequent upload handlers'receive_data_chunk methods. In this way, one handler can be a"filter" for other handlers.

Return None from receive_data_chunk to short-circuit remainingupload handlers from getting this chunk. This is useful if you'restoring the uploaded data yourself and don't want future handlers tostore a copy of the data.

If you raise a StopUpload or a SkipFile exception, the uploadwill abort or the file will be completely skipped.

  • FileUploadHandler.filecomplete(_file_size)[源代码]
  • Called when a file has finished uploading.

The handler should return an UploadedFile object that will be storedin request.FILES. Handlers may also return None to indicate thatthe UploadedFile object should come from subsequent upload handlers.

Optional methods

Custom upload handlers may also define any of the following optional methods orattributes:

  • FileUploadHandler.chunk_size
  • Size, in bytes, of the "chunks" Django should store into memory and feedinto the handler. That is, this attribute controls the size of chunksfed into FileUploadHandler.receive_data_chunk.

For maximum performance the chunk sizes should be divisible by 4 andshould not exceed 2 GB (231 bytes) in size. When there aremultiple chunk sizes provided by multiple handlers, Django will use thesmallest chunk size defined by any handler.

The default is 64*210 bytes, or 64 KB.

  • FileUploadHandler.newfile(_field_name, file_name, content_type, content_length, charset, content_type_extra)[源代码]
  • Callback signaling that a new file upload is starting. This is calledbefore any data has been fed to any upload handlers.

field_name is a string name of the file <input> field.

file_name is the filename provided by the browser.

content_type is the MIME type provided by the browser — E.g.'image/jpeg'.

content_length is the length of the image given by the browser.Sometimes this won't be provided and will be None.

charset is the character set (i.e. utf8) given by the browser.Like content_length, this sometimes won't be provided.

content_type_extra is extra information about the file from thecontent-type header. See UploadedFile.content_type_extra.

This method may raise a StopFutureHandlers exception to preventfuture handlers from handling this file.

  • FileUploadHandler.upload_complete()[源代码]

  • Callback signaling that the entire upload (all files) has completed.

  • FileUploadHandler.handleraw_input(_input_data, META, content_length, boundary, encoding)[源代码]

  • Allows the handler to completely override the parsing of the rawHTTP input.

input_data is a file-like object that supports read()-ing.

META is the same object as request.META.

content_length is the length of the data in input_data. Don'tread more than content_length bytes from input_data.

boundary is the MIME boundary for this request.

encoding is the encoding of the request.

Return None if you want upload handling to continue, or a tuple of(POST, FILES) if you want to return the new data structures suitablefor the request directly.