Writing Your Own Image Plugin#

Pillow uses a plugin model which allows you to add your own decoders and encoders to the library, without any changes to the library itself. Such plugins usually have names like XxxImagePlugin.py, where Xxx is a unique format name (usually an abbreviation).

Warning

Pillow >= 2.1.0 no longer automatically imports any file in the Python path with a name ending in ImagePlugin.py. You will need to import your image plugin manually.

Pillow decodes files in two stages:

  1. It loops over the available image plugins in the loaded order, and calls the plugin’s _accept function with the first 16 bytes of the file. If the _accept function returns true, the plugin’s _open method is called to set up the image metadata and image tiles. The _open method is not for decoding the actual image data.

  2. When the image data is requested, the ImageFile.load method is called, which sets up a decoder for each tile and feeds the data to it.

An image plugin should contain a format handler derived from the PIL.ImageFile.ImageFile base class. This class should provide an _open method, which reads the file header and set at least the internal _size and _mode attributes so that mode and size are populated. To be able to load the file, the method must also create a list of tile descriptors, which contain a decoder name, extents of the tile, and any decoder-specific data. The format handler class must be explicitly registered, via a call to the Image module.

Note

For performance reasons, it is important that the _open method quickly rejects files that do not have the appropriate contents.

Example#

The following plugin supports a simple format, which has a 128-byte header consisting of the words “SPAM” followed by the width, height, and pixel size in bits. The header fields are separated by spaces. The image data follows directly after the header, and can be either bi-level, grayscale, or 24-bit true color.

SpamImagePlugin.py:

from PIL import Image, ImageFile


def _accept(prefix):
    return prefix[:4] == b"SPAM"


class SpamImageFile(ImageFile.ImageFile):

    format = "SPAM"
    format_description = "Spam raster image"

    def _open(self):

        header = self.fp.read(128).split()

        # size in pixels (width, height)
        self._size = int(header[1]), int(header[2])

        # mode setting
        bits = int(header[3])
        if bits == 1:
            self._mode = "1"
        elif bits == 8:
            self._mode = "L"
        elif bits == 24:
            self._mode = "RGB"
        else:
            msg = "unknown number of bits"
            raise SyntaxError(msg)

        # data descriptor
        self.tile = [("raw", (0, 0) + self.size, 128, (self.mode, 0, 1))]


Image.register_open(SpamImageFile.format, SpamImageFile, _accept)

Image.register_extensions(
    SpamImageFile.format,
    [
        ".spam",
        ".spa",  # DOS version
    ],
)

The format handler must always set the internal _size and _mode attributes so that size and mode are populated. If these are not set, the file cannot be opened. To simplify the plugin, the calling code considers exceptions like SyntaxError, KeyError, IndexError, EOFError and struct.error as a failure to identify the file.

Note that the image plugin must be explicitly registered using PIL.Image.register_open(). Although not required, it is also a good idea to register any extensions used by this format.

Once the plugin has been imported, it can be used:

from PIL import Image
import SpamImagePlugin

with Image.open("hopper.spam") as im:
    pass

The tile attribute#

To be able to read the file as well as just identifying it, the tile attribute must also be set. This attribute consists of a list of tile descriptors, where each descriptor specifies how data should be loaded to a given region in the image.

In most cases, only a single descriptor is used, covering the full image. PsdImagePlugin.PsdImageFile uses multiple tiles to combine channels within a single layer, given that the channels are stored separately, one after the other.

The tile descriptor is a 4-tuple with the following contents:

(decoder, region, offset, parameters)

The fields are used as follows:

decoder

Specifies which decoder to use. The raw decoder used here supports uncompressed data, in a variety of pixel formats. For more information on this decoder, see the description below.

A list of C decoders can be seen under codecs section of the function array in _imaging.c. Python decoders are registered within the relevant plugins.

region

A 4-tuple specifying where to store data in the image.

offset

Byte offset from the beginning of the file to image data.

parameters

Parameters to the decoder. The contents of this field depends on the decoder specified by the first field in the tile descriptor tuple. If the decoder doesn’t need any parameters, use None for this field.

Note that the tile attribute contains a list of tile descriptors, not just a single descriptor.

Decoders#

The raw decoder#

The raw decoder is used to read uncompressed data from an image file. It can be used with most uncompressed file formats, such as PPM, BMP, uncompressed TIFF, and many others. To use the raw decoder with the PIL.Image.frombytes() function, use the following syntax:

image = Image.frombytes(
    mode, size, data, "raw",
    raw_mode, stride, orientation
    )

When used in a tile descriptor, the parameter field should look like:

(raw_mode, stride, orientation)

The fields are used as follows:

raw_mode

The pixel layout used in the file, and is used to properly convert data to PIL’s internal layout. For a summary of the available formats, see the table below.

stride

The distance in bytes between two consecutive lines in the image. If 0, the image is assumed to be packed (no padding between lines). If omitted, the stride defaults to 0.

orientation

Whether the first line in the image is the top line on the screen (1), or the bottom line (-1). If omitted, the orientation defaults to 1.

The raw mode field is used to determine how the data should be unpacked to match PIL’s internal pixel layout. PIL supports a large set of raw modes; for a complete list, see the table in the Unpack.c module. The following table describes some commonly used raw modes:

mode

description

1

1-bit bilevel, stored with the leftmost pixel in the most
significant bit. 0 means black, 1 means white.

1;I

1-bit inverted bilevel, stored with the leftmost pixel in the
most significant bit. 0 means white, 1 means black.

1;R

1-bit reversed bilevel, stored with the leftmost pixel in the
least significant bit. 0 means black, 1 means white.

L

8-bit grayscale. 0 means black, 255 means white.

L;I

8-bit inverted grayscale. 0 means white, 255 means black.

P

8-bit palette-mapped image.

RGB

24-bit true colour, stored as (red, green, blue).

BGR

24-bit true colour, stored as (blue, green, red).

RGBX

24-bit true colour, stored as (red, green, blue, pad). The pad
pixels may vary.

RGB;L

24-bit true colour, line interleaved (first all red pixels, then
all green pixels, finally all blue pixels).

Note that for the most common cases, the raw mode is simply the same as the mode.

The Python Imaging Library supports many other decoders, including JPEG, PNG, and PackBits. For details, see the decode.c source file, and the standard plugin implementations provided with the library.

Decoding floating point data#

PIL provides some special mechanisms to allow you to load a wide variety of formats into a mode F (floating point) image memory.

You can use the raw decoder to read images where data is packed in any standard machine data type, using one of the following raw modes:

mode

description

F

32-bit native floating point.

F;8

8-bit unsigned integer.

F;8S

8-bit signed integer.

F;16

16-bit little endian unsigned integer.

F;16S

16-bit little endian signed integer.

F;16B

16-bit big endian unsigned integer.

F;16BS

16-bit big endian signed integer.

F;16N

16-bit native unsigned integer.

F;16NS

16-bit native signed integer.

F;32

32-bit little endian unsigned integer.

F;32S

32-bit little endian signed integer.

F;32B

32-bit big endian unsigned integer.

F;32BS

32-bit big endian signed integer.

F;32N

32-bit native unsigned integer.

F;32NS

32-bit native signed integer.

F;32F

32-bit little endian floating point.

F;32BF

32-bit big endian floating point.

F;32NF

32-bit native floating point.

F;64F

64-bit little endian floating point.

F;64BF

64-bit big endian floating point.

F;64NF

64-bit native floating point.

The bit decoder#

If the raw decoder cannot handle your format, PIL also provides a special “bit” decoder that can be used to read various packed formats into a floating point image memory.

To use the bit decoder with the PIL.Image.frombytes() function, use the following syntax:

image = Image.frombytes(
    mode, size, data, "bit",
    bits, pad, fill, sign, orientation
    )

When used in a tile descriptor, the parameter field should look like:

(bits, pad, fill, sign, orientation)

The fields are used as follows:

bits

Number of bits per pixel (2-32). No default.

pad

Padding between lines, in bits. This is either 0 if there is no padding, or 8 if lines are padded to full bytes. If omitted, the pad value defaults to 8.

fill

Controls how data are added to, and stored from, the decoder bit buffer.

fill=0

Add bytes to the LSB end of the decoder buffer; store pixels from the MSB end.

fill=1

Add bytes to the MSB end of the decoder buffer; store pixels from the MSB end.

fill=2

Add bytes to the LSB end of the decoder buffer; store pixels from the LSB end.

fill=3

Add bytes to the MSB end of the decoder buffer; store pixels from the LSB end.

If omitted, the fill order defaults to 0.

sign

If non-zero, bit fields are sign extended. If zero or omitted, bit fields are unsigned.

orientation

Whether the first line in the image is the top line on the screen (1), or the bottom line (-1). If omitted, the orientation defaults to 1.

Writing Your Own File Codec in C#

There are 3 stages in a file codec’s lifetime:

  1. Setup: Pillow looks for a function in the decoder or encoder registry, falling back to a function named [codecname]_decoder or [codecname]_encoder on the internal core image object. That function is called with the args tuple from the tile.

  2. Transforming: The codec’s decode or encode function is repeatedly called with chunks of image data.

  3. Cleanup: If the codec has registered a cleanup function, it will be called at the end of the transformation process, even if there was an exception raised.

Setup#

The current conventions are that the codec setup function is named PyImaging_[codecname]DecoderNew or PyImaging_[codecname]EncoderNew and defined in decode.c or encode.c. The Python binding for it is named [codecname]_decoder or [codecname]_encoder and is set up from within the _imaging.c file in the codecs section of the function array.

The setup function needs to call PyImaging_DecoderNew or PyImaging_EncoderNew and at the very least, set the decode or encode function pointer. The fields of interest in this object are:

decode/encode

Function pointer to the decode or encode function, which has access to im, state, and the buffer of data to be transformed.

cleanup

Function pointer to the cleanup function, has access to state.

im

The target image, will be set by Pillow.

state

An ImagingCodecStateInstance, will be set by Pillow. The context member is an opaque struct that can be used by the codec to store any format specific state or options.

pulls_fd/pushes_fd

If the decoder has pulls_fd or the encoder has pushes_fd set to 1, state->fd will be a pointer to the Python file like object. The codec may use the functions in codec_fd.c to read or write directly with the file like object rather than have the data pushed through a buffer.

New in version 3.3.0.

Transforming#

The decode or encode function is called with the target (core) image, the codec state structure, and a buffer of data to be transformed.

It is the codec’s responsibility to pull as much data as possible out of the buffer and return the number of bytes consumed. The next call to the codec will include the previous unconsumed tail. The codec function will be called multiple times as the data processed.

Alternatively, if pulls_fd or pushes_fd is set, then the decode or encode function is called once, with an empty buffer. It is the codec’s responsibility to transform the entire tile in that one call. Using this will provide a codec with more freedom, but that freedom may mean increased memory usage if the entire tile is held in memory at once by the codec.

If an error occurs, set state->errcode and return -1.

Return -1 on success, without setting the errcode.

Cleanup#

The cleanup function is called after the codec returns a negative value, or if there is an error. This function should free any allocated memory and release any resources from external libraries.

Writing Your Own File Codec in Python#

Python file decoders and encoders should derive from PIL.ImageFile.PyDecoder and PIL.ImageFile.PyEncoder respectively, and should at least override the decode or encode method. They should be registered using PIL.Image.register_decoder() and PIL.Image.register_encoder(). As in the C implementation of the file codecs, there are three stages in the lifetime of a Python-based file codec:

  1. Setup: Pillow looks for the codec in the decoder or encoder registry, then instantiates the class.

  2. Transforming: The instance’s decode method is repeatedly called with a buffer of data to be interpreted, or the encode method is repeatedly called with the size of data to be output.

    Alternatively, if the decoder’s _pulls_fd property (or the encoder’s _pushes_fd property) is set to True, then decode and encode will only be called once. In the decoder, self.fd can be used to access the file-like object. Using this will provide a codec with more freedom, but that freedom may mean increased memory usage if entire file is held in memory at once by the codec.

    In decode, once the data has been interpreted, set_as_raw can be used to populate the image.

  3. Cleanup: The instance’s cleanup method is called once the transformation is complete. This can be used to clean up any resources used by the codec.

    If you set _pulls_fd or _pushes_fd to True however, then you probably chose to perform any cleanup tasks at the end of decode or encode.

For an example PIL.ImageFile.PyDecoder, see DdsImagePlugin. For a plugin that uses both PIL.ImageFile.PyDecoder and PIL.ImageFile.PyEncoder, see BlpImagePlugin