RFC: Pure Python Decoder objects

PIL/Image.py

@@ -208,6 +208,8 @@ def isImageType(t):
 SAVE = {}
 SAVE_ALL = {}
 EXTENSION = {}
+DECODERS = {}
+ENCODERS = {}
 
 # --------------------------------------------------------------------
 # Modes supported by this version
@@ -413,6 +415,11 @@ def _getdecoder(mode, decoder_name, args, extra=()):
     elif not isinstance(args, tuple):
         args = (args,)
 
+    try:
+        decoder = DECODERS[decoder_name]
+        return decoder(mode, *args + extra)
+    except KeyError:
+        pass
     try:
         # get decoder
         decoder = getattr(core, decoder_name + "_decoder")
@@ -430,6 +437,11 @@ def _getencoder(mode, encoder_name, args, extra=()):
     elif not isinstance(args, tuple):
         args = (args,)
 
+    try:
+        encoder = ENCODERS[encoder_name]
+        return encoder(mode, *args + extra)
+    except KeyError:
+        pass
     try:
         # get encoder
         encoder = getattr(core, encoder_name + "_encoder")
@@ -2526,6 +2538,33 @@ def registered_extensions():
         init()
     return EXTENSION
 
+def register_decoder(name, decoder):
+    """
+    Registers an image decoder.  This function should not be
+    used in application code.
+
+    :param name: The name of the decoder
+    :param decoder: A callable(mode, args) that returns an
+                    ImageFile.PyDecoder object
+
+    .. versionadded:: 4.1.0                
+    """
+    DECODERS[name] = decoder
+
+
+def register_encoder(name, encoder):
+    """
+    Registers an image encoder.  This function should not be
+    used in application code.
+
+    :param name: The name of the encoder
+    :param encoder: A callable(mode, args) that returns an
+                    ImageFile.PyEncoder object
+
+    .. versionadded:: 4.1.0
+    """
+    ENCODERS[name] = encoder
+
 
 # --------------------------------------------------------------------
 # Simple display support.  User code may override this.

PIL/ImageFile.py

@@ -193,10 +193,7 @@ def load(self):
                 decoder = Image._getdecoder(self.mode, decoder_name,
                                       args, self.decoderconfig)
                 seek(offset)
-                try:
-                    decoder.setimage(self.im, extents)
-                except ValueError:
-                    continue
+                decoder.setimage(self.im, extents)
                 if decoder.pulls_fd:
                     decoder.setfd(self.fp)
                     status, err_code = decoder.decode(b"")
@@ -520,3 +517,128 @@ def _safe_read(fp, size):
         data.append(block)
         size -= len(block)
     return b"".join(data)
+
+
+class PyCodecState(object):
+    def __init__(self):
+        self.xsize = 0
+        self.ysize = 0
+        self.xoff = 0
+        self.yoff = 0
+
+    def extents(self):
+        return (self.xoff, self.yoff,
+                self.xoff+self.xsize, self.yoff+self.ysize)
+
+class PyDecoder(object):
+    """
+    Python implementation of a format decoder. Override this class and
+    add the decoding logic in the `decode` method.
+
+    See :ref:`Writing Your Own File Decoder in Python<file-decoders-py>`
+    """
+
+    _pulls_fd = False
+
+    def __init__(self, mode, *args):
+        self.im = None
+        self.state = PyCodecState()
+        self.fd = None
+        self.mode = mode
+        self.init(args)
+
+    def init(self, args):
+        """
+        Override to perform decoder specific initialization
+
+        :param args: Array of args items from the tile entry
+        :returns: None
+        """
+        self.args = args
+
+    @property
+    def pulls_fd(self):
+        return self._pulls_fd
+
+    def decode(self, buffer):
+        """
+        Override to perform the decoding process.
+
+        :param buffer: A bytes object with the data to be decoded.  If `handles_eof`
+             is set, then `buffer` will be empty and `self.fd` will be set.
+        :returns: A tuple of (bytes consumed, errcode). If finished with decoding
+             return <0 for the bytes consumed. Err codes are from `ERRORS`
+        """
+        raise NotImplementedError()
+
+    def cleanup(self):
+        """
+        Override to perform decoder specific cleanup
+
+        :returns: None
+        """
+        pass
+
+    def setfd(self, fd):
+        """
+        Called from ImageFile to set the python file-like object
+
+        :param fd: A python file-like object
+        :returns: None
+        """
+        self.fd = fd
+
+    def setimage(self, im, extents=None):
+        """
+        Called from ImageFile to set the core output image for the decoder
+
+        :param im: A core image object
+        :param extents: a 4 tuple of (x0, y0, x1, y1) defining the rectangle
+            for this tile
+        :returns: None
+        """
+
+        # following c code
+        self.im = im
+
+        if extents:
+            (x0, y0, x1, y1) = extents
+        else:
+            (x0, y0, x1, y1) = (0, 0, 0, 0)
+
+
+        if x0 == 0 and x1 == 0:
+            self.state.xsize, self.state.ysize = self.im.size
+        else:
+            self.state.xoff = x0
+            self.state.yoff = y0
+            self.state.xsize = x1 - x0
+            self.state.ysize = y1 - y0
+
+        if self.state.xsize <= 0 or self.state.ysize <= 0:
+            raise ValueError("Size cannot be negative")
+
+        if (self.state.xsize + self.state.xoff > self.im.size[0] or
+            self.state.ysize + self.state.yoff > self.im.size[1]):
+            raise ValueError("Tile cannot extend outside image")
+
+    def set_as_raw(self, data, rawmode=None):
+        """
+        Convenience method to set the internal image from a stream of raw data
+
+        :param data: Bytes to be set
+        :param rawmode: The rawmode to be used for the decoder. If not specified,
+             it will default to the mode of the image
+        :returns: None
+        """
+
+        if not rawmode:
+            rawmode = self.mode
+        d = Image._getdecoder(self.mode, 'raw', (rawmode))
+        d.setimage(self.im, self.state.extents())
+        s = d.decode(data)
+
+        if s[0] >= 0:
+            raise ValueError("not enough image data")
+        if s[1] != 0:
+            raise ValueError("cannot decode image data")

PIL/MspImagePlugin.py

@@ -1,6 +1,5 @@
 #
 # The Python Imaging Library.
-# $Id$
 #
 # MSP file handling
 #
@@ -9,16 +8,24 @@
 # History:
 #       95-09-05 fl     Created
 #       97-01-03 fl     Read/write MSP images
+#       17-02-21 es     Fixed RLE interpretation
 #
 # Copyright (c) Secret Labs AB 1997.
 # Copyright (c) Fredrik Lundh 1995-97.
+# Copyright (c) Eric Soroos 2017.
 #
 # See the README file for information on usage and redistribution.
 #
-
+# More info on this format: https://archive.org/details/gg243631
+# Page 313:
+# Figure 205. Windows Paint Version 1: "DanM" Format
+# Figure 206. Windows Paint Version 2: "LinS" Format. Used in Windows V2.03
+#
+# See also: http://www.fileformat.info/format/mspaint/egff.htm
 
 from . import Image, ImageFile
-from ._binary import i16le as i16, o16le as o16
+from ._binary import i16le as i16, o16le as o16, i8
+import struct, io
 
 __version__ = "0.1"
 
@@ -60,7 +67,90 @@ def _open(self):
         if s[:4] == b"DanM":
             self.tile = [("raw", (0, 0)+self.size, 32, ("1", 0, 1))]
         else:
-            self.tile = [("msp", (0, 0)+self.size, 32+2*self.size[1], None)]
+            self.tile = [("MSP", (0, 0)+self.size, 32, None)]
+
+
+class MspDecoder(ImageFile.PyDecoder):
+    # The algo for the MSP decoder is from
+    # http://www.fileformat.info/format/mspaint/egff.htm
+    # cc-by-attribution -- That page references is taken from the
+    # Encyclopedia of Graphics File Formats and is licensed by
+    # O'Reilly under the Creative Common/Attribution license
+    #
+    # For RLE encoded files, the 32byte header is followed by a scan
+    # line map, encoded as one 16bit word of encoded byte length per
+    # line.
+    #
+    # NOTE: the encoded length of the line can be 0. This was not
+    # handled in the previous version of this encoder, and there's no
+    # mention of how to handle it in the documentation. From the few
+    # examples I've seen, I've assumed that it is a fill of the
+    # background color, in this case, white.
+    #
+    #
+    # Pseudocode of the decoder:
+    # Read a BYTE value as the RunType
+    #  If the RunType value is zero
+    #   Read next byte as the RunCount
+    #   Read the next byte as the RunValue
+    #   Write the RunValue byte RunCount times
+    #  If the RunType value is non-zero
+    #   Use this value as the RunCount
+    #   Read and write the next RunCount bytes literally
+    # 
+    #  e.g.:
+    #  0x00 03 ff 05 00 01 02 03 04
+    #  would yield the bytes:
+    #  0xff ff ff 00 01 02 03 04
+    #
+    # which are then interpreted as a bit packed mode '1' image
+
+
+    _pulls_fd = True
+
+    def decode(self, buffer):
+
+        img = io.BytesIO()
+        blank_line = bytearray((0xff,)*((self.state.xsize+7)//8))
+        try:
+            self.fd.seek(32)
+            rowmap = struct.unpack_from("<%dH" % (self.state.ysize),
+                                        self.fd.read(self.state.ysize*2))
+        except struct.error:
+            raise IOError("Truncated MSP file in row map")
+
+        for x, rowlen in enumerate(rowmap):
+            try:
+                if rowlen == 0:
+                    img.write(blank_line)
+                    continue
+                row = self.fd.read(rowlen)
+                if len(row) != rowlen:
+                    raise IOError("Truncated MSP file, expected %d bytes on row %s",
+                                  (rowlen, x))
+                idx = 0
+                while idx < rowlen:
+                    runtype = i8(row[idx])
+                    idx += 1
+                    if runtype == 0:
+                        (runcount, runval) = struct.unpack("Bc", row[idx:idx+2])
+                        img.write(runval * runcount)
+                        idx += 2
+                    else:
+                        runcount = runtype
+                        img.write(row[idx:idx+runcount])
+                        idx += runcount
+
+            except struct.error:
+                raise IOError("Corrupted MSP file in row %d" %x)
+
+        self.set_as_raw(img.getvalue(), ("1", 0, 1))
+
+        return 0,0
+
+
+Image.register_decoder('MSP', MspDecoder)
+
 
 #
 # write MSP files (uncompressed only)
@@ -92,6 +182,7 @@ def _save(im, fp, filename):
     # image body
     ImageFile._save(im, fp, [("raw", (0, 0)+im.size, 32, ("1", 0, 1))])
 
+
 #
 # registry