2010-07-31 06:52:47 +04:00
|
|
|
#
|
|
|
|
# The Python Imaging Library.
|
|
|
|
# $Id$
|
|
|
|
#
|
|
|
|
# PPM support for PIL
|
|
|
|
#
|
|
|
|
# History:
|
|
|
|
# 96-03-24 fl Created
|
|
|
|
# 98-03-06 fl Write RGBA images (as RGB, that is)
|
|
|
|
#
|
|
|
|
# Copyright (c) Secret Labs AB 1997-98.
|
|
|
|
# Copyright (c) Fredrik Lundh 1996.
|
|
|
|
#
|
|
|
|
# See the README file for information on usage and redistribution.
|
|
|
|
#
|
|
|
|
|
|
|
|
|
2017-01-17 16:22:18 +03:00
|
|
|
from . import Image, ImageFile
|
2010-07-31 06:52:47 +04:00
|
|
|
|
2015-08-25 15:27:18 +03:00
|
|
|
__version__ = "0.2"
|
|
|
|
|
2010-07-31 06:52:47 +04:00
|
|
|
#
|
|
|
|
# --------------------------------------------------------------------
|
|
|
|
|
2017-11-03 00:29:06 +03:00
|
|
|
b_whitespace = b'\x20\x09\x0a\x0b\x0c\x0d'
|
py3k: The big push
There are two main issues fixed with this commit:
* bytes vs. str: All file, image, and palette data are now handled as
bytes. A new _binary module consolidates the hacks needed to do this
across Python versions. tostring/fromstring methods have been renamed to
tobytes/frombytes, but the Python 2.6/2.7 versions alias them to the old
names for compatibility. Users should move to tobytes/frombytes.
One other potentially-breaking change is that text data in image files
(such as tags, comments) are now explicitly handled with a specific
character encoding in mind. This works well with the Unicode str in
Python 3, but may trip up old code expecting a straight byte-for-byte
translation to a Python string. This also required a change to Gohlke's
tags tests (in Tests/test_file_png.py) to expect Unicode strings from
the code.
* True div vs. floor div: Many division operations used the "/" operator
to do floor division, which is now the "//" operator in Python 3. These
were fixed.
As of this commit, on the first pass, I have one failing test (improper
handling of a slice object in a C module, test_imagepath.py) in Python 3,
and three that that I haven't tried running yet (test_imagegl,
test_imagegrab, and test_imageqt). I also haven't tested anything on
Windows. All but the three skipped tests run flawlessly against Pythons
2.6 and 2.7.
2012-10-21 01:01:53 +04:00
|
|
|
|
2010-07-31 06:52:47 +04:00
|
|
|
MODES = {
|
|
|
|
# standard
|
py3k: The big push
There are two main issues fixed with this commit:
* bytes vs. str: All file, image, and palette data are now handled as
bytes. A new _binary module consolidates the hacks needed to do this
across Python versions. tostring/fromstring methods have been renamed to
tobytes/frombytes, but the Python 2.6/2.7 versions alias them to the old
names for compatibility. Users should move to tobytes/frombytes.
One other potentially-breaking change is that text data in image files
(such as tags, comments) are now explicitly handled with a specific
character encoding in mind. This works well with the Unicode str in
Python 3, but may trip up old code expecting a straight byte-for-byte
translation to a Python string. This also required a change to Gohlke's
tags tests (in Tests/test_file_png.py) to expect Unicode strings from
the code.
* True div vs. floor div: Many division operations used the "/" operator
to do floor division, which is now the "//" operator in Python 3. These
were fixed.
As of this commit, on the first pass, I have one failing test (improper
handling of a slice object in a C module, test_imagepath.py) in Python 3,
and three that that I haven't tried running yet (test_imagegl,
test_imagegrab, and test_imageqt). I also haven't tested anything on
Windows. All but the three skipped tests run flawlessly against Pythons
2.6 and 2.7.
2012-10-21 01:01:53 +04:00
|
|
|
b"P4": "1",
|
|
|
|
b"P5": "L",
|
|
|
|
b"P6": "RGB",
|
2010-07-31 06:52:47 +04:00
|
|
|
# extensions
|
py3k: The big push
There are two main issues fixed with this commit:
* bytes vs. str: All file, image, and palette data are now handled as
bytes. A new _binary module consolidates the hacks needed to do this
across Python versions. tostring/fromstring methods have been renamed to
tobytes/frombytes, but the Python 2.6/2.7 versions alias them to the old
names for compatibility. Users should move to tobytes/frombytes.
One other potentially-breaking change is that text data in image files
(such as tags, comments) are now explicitly handled with a specific
character encoding in mind. This works well with the Unicode str in
Python 3, but may trip up old code expecting a straight byte-for-byte
translation to a Python string. This also required a change to Gohlke's
tags tests (in Tests/test_file_png.py) to expect Unicode strings from
the code.
* True div vs. floor div: Many division operations used the "/" operator
to do floor division, which is now the "//" operator in Python 3. These
were fixed.
As of this commit, on the first pass, I have one failing test (improper
handling of a slice object in a C module, test_imagepath.py) in Python 3,
and three that that I haven't tried running yet (test_imagegl,
test_imagegrab, and test_imageqt). I also haven't tested anything on
Windows. All but the three skipped tests run flawlessly against Pythons
2.6 and 2.7.
2012-10-21 01:01:53 +04:00
|
|
|
b"P0CMYK": "CMYK",
|
2010-07-31 06:52:47 +04:00
|
|
|
# PIL extensions (for test purposes only)
|
py3k: The big push
There are two main issues fixed with this commit:
* bytes vs. str: All file, image, and palette data are now handled as
bytes. A new _binary module consolidates the hacks needed to do this
across Python versions. tostring/fromstring methods have been renamed to
tobytes/frombytes, but the Python 2.6/2.7 versions alias them to the old
names for compatibility. Users should move to tobytes/frombytes.
One other potentially-breaking change is that text data in image files
(such as tags, comments) are now explicitly handled with a specific
character encoding in mind. This works well with the Unicode str in
Python 3, but may trip up old code expecting a straight byte-for-byte
translation to a Python string. This also required a change to Gohlke's
tags tests (in Tests/test_file_png.py) to expect Unicode strings from
the code.
* True div vs. floor div: Many division operations used the "/" operator
to do floor division, which is now the "//" operator in Python 3. These
were fixed.
As of this commit, on the first pass, I have one failing test (improper
handling of a slice object in a C module, test_imagepath.py) in Python 3,
and three that that I haven't tried running yet (test_imagegl,
test_imagegrab, and test_imageqt). I also haven't tested anything on
Windows. All but the three skipped tests run flawlessly against Pythons
2.6 and 2.7.
2012-10-21 01:01:53 +04:00
|
|
|
b"PyP": "P",
|
|
|
|
b"PyRGBA": "RGBA",
|
|
|
|
b"PyCMYK": "CMYK"
|
2010-07-31 06:52:47 +04:00
|
|
|
}
|
|
|
|
|
2014-08-26 17:47:10 +04:00
|
|
|
|
2010-07-31 06:52:47 +04:00
|
|
|
def _accept(prefix):
|
py3k: The big push
There are two main issues fixed with this commit:
* bytes vs. str: All file, image, and palette data are now handled as
bytes. A new _binary module consolidates the hacks needed to do this
across Python versions. tostring/fromstring methods have been renamed to
tobytes/frombytes, but the Python 2.6/2.7 versions alias them to the old
names for compatibility. Users should move to tobytes/frombytes.
One other potentially-breaking change is that text data in image files
(such as tags, comments) are now explicitly handled with a specific
character encoding in mind. This works well with the Unicode str in
Python 3, but may trip up old code expecting a straight byte-for-byte
translation to a Python string. This also required a change to Gohlke's
tags tests (in Tests/test_file_png.py) to expect Unicode strings from
the code.
* True div vs. floor div: Many division operations used the "/" operator
to do floor division, which is now the "//" operator in Python 3. These
were fixed.
As of this commit, on the first pass, I have one failing test (improper
handling of a slice object in a C module, test_imagepath.py) in Python 3,
and three that that I haven't tried running yet (test_imagegl,
test_imagegrab, and test_imageqt). I also haven't tested anything on
Windows. All but the three skipped tests run flawlessly against Pythons
2.6 and 2.7.
2012-10-21 01:01:53 +04:00
|
|
|
return prefix[0:1] == b"P" and prefix[1] in b"0456y"
|
2010-07-31 06:52:47 +04:00
|
|
|
|
2014-08-26 17:47:10 +04:00
|
|
|
|
2010-07-31 06:52:47 +04:00
|
|
|
##
|
|
|
|
# Image plugin for PBM, PGM, and PPM images.
|
|
|
|
|
|
|
|
class PpmImageFile(ImageFile.ImageFile):
|
|
|
|
|
|
|
|
format = "PPM"
|
|
|
|
format_description = "Pbmplus image"
|
|
|
|
|
2014-08-26 17:47:10 +04:00
|
|
|
def _token(self, s=b""):
|
|
|
|
while True: # read until next whitespace
|
2010-07-31 06:52:47 +04:00
|
|
|
c = self.fp.read(1)
|
py3k: The big push
There are two main issues fixed with this commit:
* bytes vs. str: All file, image, and palette data are now handled as
bytes. A new _binary module consolidates the hacks needed to do this
across Python versions. tostring/fromstring methods have been renamed to
tobytes/frombytes, but the Python 2.6/2.7 versions alias them to the old
names for compatibility. Users should move to tobytes/frombytes.
One other potentially-breaking change is that text data in image files
(such as tags, comments) are now explicitly handled with a specific
character encoding in mind. This works well with the Unicode str in
Python 3, but may trip up old code expecting a straight byte-for-byte
translation to a Python string. This also required a change to Gohlke's
tags tests (in Tests/test_file_png.py) to expect Unicode strings from
the code.
* True div vs. floor div: Many division operations used the "/" operator
to do floor division, which is now the "//" operator in Python 3. These
were fixed.
As of this commit, on the first pass, I have one failing test (improper
handling of a slice object in a C module, test_imagepath.py) in Python 3,
and three that that I haven't tried running yet (test_imagegl,
test_imagegrab, and test_imageqt). I also haven't tested anything on
Windows. All but the three skipped tests run flawlessly against Pythons
2.6 and 2.7.
2012-10-21 01:01:53 +04:00
|
|
|
if not c or c in b_whitespace:
|
2010-07-31 06:52:47 +04:00
|
|
|
break
|
2014-04-09 09:43:57 +04:00
|
|
|
if c > b'\x79':
|
|
|
|
raise ValueError("Expected ASCII value, found binary")
|
2010-07-31 06:52:47 +04:00
|
|
|
s = s + c
|
2014-04-09 09:43:57 +04:00
|
|
|
if (len(s) > 9):
|
|
|
|
raise ValueError("Expected int, got > 9 digits")
|
2010-07-31 06:52:47 +04:00
|
|
|
return s
|
|
|
|
|
|
|
|
def _open(self):
|
|
|
|
|
|
|
|
# check magic
|
|
|
|
s = self.fp.read(1)
|
py3k: The big push
There are two main issues fixed with this commit:
* bytes vs. str: All file, image, and palette data are now handled as
bytes. A new _binary module consolidates the hacks needed to do this
across Python versions. tostring/fromstring methods have been renamed to
tobytes/frombytes, but the Python 2.6/2.7 versions alias them to the old
names for compatibility. Users should move to tobytes/frombytes.
One other potentially-breaking change is that text data in image files
(such as tags, comments) are now explicitly handled with a specific
character encoding in mind. This works well with the Unicode str in
Python 3, but may trip up old code expecting a straight byte-for-byte
translation to a Python string. This also required a change to Gohlke's
tags tests (in Tests/test_file_png.py) to expect Unicode strings from
the code.
* True div vs. floor div: Many division operations used the "/" operator
to do floor division, which is now the "//" operator in Python 3. These
were fixed.
As of this commit, on the first pass, I have one failing test (improper
handling of a slice object in a C module, test_imagepath.py) in Python 3,
and three that that I haven't tried running yet (test_imagegl,
test_imagegrab, and test_imageqt). I also haven't tested anything on
Windows. All but the three skipped tests run flawlessly against Pythons
2.6 and 2.7.
2012-10-21 01:01:53 +04:00
|
|
|
if s != b"P":
|
2012-10-11 07:52:53 +04:00
|
|
|
raise SyntaxError("not a PPM file")
|
2010-07-31 06:52:47 +04:00
|
|
|
mode = MODES[self._token(s)]
|
|
|
|
|
|
|
|
if mode == "1":
|
|
|
|
self.mode = "1"
|
|
|
|
rawmode = "1;I"
|
|
|
|
else:
|
|
|
|
self.mode = rawmode = mode
|
|
|
|
|
|
|
|
for ix in range(3):
|
2012-10-17 07:39:56 +04:00
|
|
|
while True:
|
|
|
|
while True:
|
2010-07-31 06:52:47 +04:00
|
|
|
s = self.fp.read(1)
|
py3k: The big push
There are two main issues fixed with this commit:
* bytes vs. str: All file, image, and palette data are now handled as
bytes. A new _binary module consolidates the hacks needed to do this
across Python versions. tostring/fromstring methods have been renamed to
tobytes/frombytes, but the Python 2.6/2.7 versions alias them to the old
names for compatibility. Users should move to tobytes/frombytes.
One other potentially-breaking change is that text data in image files
(such as tags, comments) are now explicitly handled with a specific
character encoding in mind. This works well with the Unicode str in
Python 3, but may trip up old code expecting a straight byte-for-byte
translation to a Python string. This also required a change to Gohlke's
tags tests (in Tests/test_file_png.py) to expect Unicode strings from
the code.
* True div vs. floor div: Many division operations used the "/" operator
to do floor division, which is now the "//" operator in Python 3. These
were fixed.
As of this commit, on the first pass, I have one failing test (improper
handling of a slice object in a C module, test_imagepath.py) in Python 3,
and three that that I haven't tried running yet (test_imagegl,
test_imagegrab, and test_imageqt). I also haven't tested anything on
Windows. All but the three skipped tests run flawlessly against Pythons
2.6 and 2.7.
2012-10-21 01:01:53 +04:00
|
|
|
if s not in b_whitespace:
|
2010-07-31 06:52:47 +04:00
|
|
|
break
|
2015-08-28 19:05:08 +03:00
|
|
|
if s == b"":
|
2018-06-24 15:32:25 +03:00
|
|
|
raise ValueError(
|
|
|
|
"File does not extend beyond magic number")
|
py3k: The big push
There are two main issues fixed with this commit:
* bytes vs. str: All file, image, and palette data are now handled as
bytes. A new _binary module consolidates the hacks needed to do this
across Python versions. tostring/fromstring methods have been renamed to
tobytes/frombytes, but the Python 2.6/2.7 versions alias them to the old
names for compatibility. Users should move to tobytes/frombytes.
One other potentially-breaking change is that text data in image files
(such as tags, comments) are now explicitly handled with a specific
character encoding in mind. This works well with the Unicode str in
Python 3, but may trip up old code expecting a straight byte-for-byte
translation to a Python string. This also required a change to Gohlke's
tags tests (in Tests/test_file_png.py) to expect Unicode strings from
the code.
* True div vs. floor div: Many division operations used the "/" operator
to do floor division, which is now the "//" operator in Python 3. These
were fixed.
As of this commit, on the first pass, I have one failing test (improper
handling of a slice object in a C module, test_imagepath.py) in Python 3,
and three that that I haven't tried running yet (test_imagegl,
test_imagegrab, and test_imageqt). I also haven't tested anything on
Windows. All but the three skipped tests run flawlessly against Pythons
2.6 and 2.7.
2012-10-21 01:01:53 +04:00
|
|
|
if s != b"#":
|
2010-07-31 06:52:47 +04:00
|
|
|
break
|
|
|
|
s = self.fp.readline()
|
|
|
|
s = int(self._token(s))
|
|
|
|
if ix == 0:
|
|
|
|
xsize = s
|
|
|
|
elif ix == 1:
|
|
|
|
ysize = s
|
|
|
|
if mode == "1":
|
|
|
|
break
|
2014-04-08 09:12:33 +04:00
|
|
|
elif ix == 2:
|
|
|
|
# maxgrey
|
|
|
|
if s > 255:
|
|
|
|
if not mode == 'L':
|
2014-08-26 17:47:10 +04:00
|
|
|
raise ValueError("Too many colors for band: %s" % s)
|
2014-04-08 09:24:48 +04:00
|
|
|
if s < 2**16:
|
2014-04-08 09:12:33 +04:00
|
|
|
self.mode = 'I'
|
|
|
|
rawmode = 'I;16B'
|
|
|
|
else:
|
2014-08-26 17:47:10 +04:00
|
|
|
self.mode = 'I'
|
2014-04-08 09:12:33 +04:00
|
|
|
rawmode = 'I;32B'
|
2014-08-26 17:47:10 +04:00
|
|
|
|
2018-09-30 05:58:02 +03:00
|
|
|
self._size = xsize, ysize
|
2010-07-31 06:52:47 +04:00
|
|
|
self.tile = [("raw",
|
|
|
|
(0, 0, xsize, ysize),
|
|
|
|
self.fp.tell(),
|
|
|
|
(rawmode, 0, 1))]
|
|
|
|
|
2014-08-26 17:47:10 +04:00
|
|
|
|
2010-07-31 06:52:47 +04:00
|
|
|
#
|
|
|
|
# --------------------------------------------------------------------
|
|
|
|
|
|
|
|
def _save(im, fp, filename):
|
|
|
|
if im.mode == "1":
|
py3k: The big push
There are two main issues fixed with this commit:
* bytes vs. str: All file, image, and palette data are now handled as
bytes. A new _binary module consolidates the hacks needed to do this
across Python versions. tostring/fromstring methods have been renamed to
tobytes/frombytes, but the Python 2.6/2.7 versions alias them to the old
names for compatibility. Users should move to tobytes/frombytes.
One other potentially-breaking change is that text data in image files
(such as tags, comments) are now explicitly handled with a specific
character encoding in mind. This works well with the Unicode str in
Python 3, but may trip up old code expecting a straight byte-for-byte
translation to a Python string. This also required a change to Gohlke's
tags tests (in Tests/test_file_png.py) to expect Unicode strings from
the code.
* True div vs. floor div: Many division operations used the "/" operator
to do floor division, which is now the "//" operator in Python 3. These
were fixed.
As of this commit, on the first pass, I have one failing test (improper
handling of a slice object in a C module, test_imagepath.py) in Python 3,
and three that that I haven't tried running yet (test_imagegl,
test_imagegrab, and test_imageqt). I also haven't tested anything on
Windows. All but the three skipped tests run flawlessly against Pythons
2.6 and 2.7.
2012-10-21 01:01:53 +04:00
|
|
|
rawmode, head = "1;I", b"P4"
|
2010-07-31 06:52:47 +04:00
|
|
|
elif im.mode == "L":
|
py3k: The big push
There are two main issues fixed with this commit:
* bytes vs. str: All file, image, and palette data are now handled as
bytes. A new _binary module consolidates the hacks needed to do this
across Python versions. tostring/fromstring methods have been renamed to
tobytes/frombytes, but the Python 2.6/2.7 versions alias them to the old
names for compatibility. Users should move to tobytes/frombytes.
One other potentially-breaking change is that text data in image files
(such as tags, comments) are now explicitly handled with a specific
character encoding in mind. This works well with the Unicode str in
Python 3, but may trip up old code expecting a straight byte-for-byte
translation to a Python string. This also required a change to Gohlke's
tags tests (in Tests/test_file_png.py) to expect Unicode strings from
the code.
* True div vs. floor div: Many division operations used the "/" operator
to do floor division, which is now the "//" operator in Python 3. These
were fixed.
As of this commit, on the first pass, I have one failing test (improper
handling of a slice object in a C module, test_imagepath.py) in Python 3,
and three that that I haven't tried running yet (test_imagegl,
test_imagegrab, and test_imageqt). I also haven't tested anything on
Windows. All but the three skipped tests run flawlessly against Pythons
2.6 and 2.7.
2012-10-21 01:01:53 +04:00
|
|
|
rawmode, head = "L", b"P5"
|
2014-04-08 09:23:04 +04:00
|
|
|
elif im.mode == "I":
|
|
|
|
if im.getextrema()[1] < 2**16:
|
|
|
|
rawmode, head = "I;16B", b"P5"
|
|
|
|
else:
|
|
|
|
rawmode, head = "I;32B", b"P5"
|
2010-07-31 06:52:47 +04:00
|
|
|
elif im.mode == "RGB":
|
py3k: The big push
There are two main issues fixed with this commit:
* bytes vs. str: All file, image, and palette data are now handled as
bytes. A new _binary module consolidates the hacks needed to do this
across Python versions. tostring/fromstring methods have been renamed to
tobytes/frombytes, but the Python 2.6/2.7 versions alias them to the old
names for compatibility. Users should move to tobytes/frombytes.
One other potentially-breaking change is that text data in image files
(such as tags, comments) are now explicitly handled with a specific
character encoding in mind. This works well with the Unicode str in
Python 3, but may trip up old code expecting a straight byte-for-byte
translation to a Python string. This also required a change to Gohlke's
tags tests (in Tests/test_file_png.py) to expect Unicode strings from
the code.
* True div vs. floor div: Many division operations used the "/" operator
to do floor division, which is now the "//" operator in Python 3. These
were fixed.
As of this commit, on the first pass, I have one failing test (improper
handling of a slice object in a C module, test_imagepath.py) in Python 3,
and three that that I haven't tried running yet (test_imagegl,
test_imagegrab, and test_imageqt). I also haven't tested anything on
Windows. All but the three skipped tests run flawlessly against Pythons
2.6 and 2.7.
2012-10-21 01:01:53 +04:00
|
|
|
rawmode, head = "RGB", b"P6"
|
2010-07-31 06:52:47 +04:00
|
|
|
elif im.mode == "RGBA":
|
py3k: The big push
There are two main issues fixed with this commit:
* bytes vs. str: All file, image, and palette data are now handled as
bytes. A new _binary module consolidates the hacks needed to do this
across Python versions. tostring/fromstring methods have been renamed to
tobytes/frombytes, but the Python 2.6/2.7 versions alias them to the old
names for compatibility. Users should move to tobytes/frombytes.
One other potentially-breaking change is that text data in image files
(such as tags, comments) are now explicitly handled with a specific
character encoding in mind. This works well with the Unicode str in
Python 3, but may trip up old code expecting a straight byte-for-byte
translation to a Python string. This also required a change to Gohlke's
tags tests (in Tests/test_file_png.py) to expect Unicode strings from
the code.
* True div vs. floor div: Many division operations used the "/" operator
to do floor division, which is now the "//" operator in Python 3. These
were fixed.
As of this commit, on the first pass, I have one failing test (improper
handling of a slice object in a C module, test_imagepath.py) in Python 3,
and three that that I haven't tried running yet (test_imagegl,
test_imagegrab, and test_imageqt). I also haven't tested anything on
Windows. All but the three skipped tests run flawlessly against Pythons
2.6 and 2.7.
2012-10-21 01:01:53 +04:00
|
|
|
rawmode, head = "RGB", b"P6"
|
2010-07-31 06:52:47 +04:00
|
|
|
else:
|
2012-10-11 07:52:53 +04:00
|
|
|
raise IOError("cannot write mode %s as PPM" % im.mode)
|
py3k: The big push
There are two main issues fixed with this commit:
* bytes vs. str: All file, image, and palette data are now handled as
bytes. A new _binary module consolidates the hacks needed to do this
across Python versions. tostring/fromstring methods have been renamed to
tobytes/frombytes, but the Python 2.6/2.7 versions alias them to the old
names for compatibility. Users should move to tobytes/frombytes.
One other potentially-breaking change is that text data in image files
(such as tags, comments) are now explicitly handled with a specific
character encoding in mind. This works well with the Unicode str in
Python 3, but may trip up old code expecting a straight byte-for-byte
translation to a Python string. This also required a change to Gohlke's
tags tests (in Tests/test_file_png.py) to expect Unicode strings from
the code.
* True div vs. floor div: Many division operations used the "/" operator
to do floor division, which is now the "//" operator in Python 3. These
were fixed.
As of this commit, on the first pass, I have one failing test (improper
handling of a slice object in a C module, test_imagepath.py) in Python 3,
and three that that I haven't tried running yet (test_imagegl,
test_imagegrab, and test_imageqt). I also haven't tested anything on
Windows. All but the three skipped tests run flawlessly against Pythons
2.6 and 2.7.
2012-10-21 01:01:53 +04:00
|
|
|
fp.write(head + ("\n%d %d\n" % im.size).encode('ascii'))
|
2014-04-08 09:23:04 +04:00
|
|
|
if head == b"P6":
|
py3k: The big push
There are two main issues fixed with this commit:
* bytes vs. str: All file, image, and palette data are now handled as
bytes. A new _binary module consolidates the hacks needed to do this
across Python versions. tostring/fromstring methods have been renamed to
tobytes/frombytes, but the Python 2.6/2.7 versions alias them to the old
names for compatibility. Users should move to tobytes/frombytes.
One other potentially-breaking change is that text data in image files
(such as tags, comments) are now explicitly handled with a specific
character encoding in mind. This works well with the Unicode str in
Python 3, but may trip up old code expecting a straight byte-for-byte
translation to a Python string. This also required a change to Gohlke's
tags tests (in Tests/test_file_png.py) to expect Unicode strings from
the code.
* True div vs. floor div: Many division operations used the "/" operator
to do floor division, which is now the "//" operator in Python 3. These
were fixed.
As of this commit, on the first pass, I have one failing test (improper
handling of a slice object in a C module, test_imagepath.py) in Python 3,
and three that that I haven't tried running yet (test_imagegl,
test_imagegrab, and test_imageqt). I also haven't tested anything on
Windows. All but the three skipped tests run flawlessly against Pythons
2.6 and 2.7.
2012-10-21 01:01:53 +04:00
|
|
|
fp.write(b"255\n")
|
2014-04-08 09:23:04 +04:00
|
|
|
if head == b"P5":
|
2014-04-09 09:44:24 +04:00
|
|
|
if rawmode == "L":
|
2014-04-08 09:23:04 +04:00
|
|
|
fp.write(b"255\n")
|
|
|
|
elif rawmode == "I;16B":
|
|
|
|
fp.write(b"65535\n")
|
|
|
|
elif rawmode == "I;32B":
|
|
|
|
fp.write(b"2147483648\n")
|
2014-08-26 17:47:10 +04:00
|
|
|
ImageFile._save(im, fp, [("raw", (0, 0)+im.size, 0, (rawmode, 0, 1))])
|
2010-07-31 06:52:47 +04:00
|
|
|
|
|
|
|
# ALTERNATIVE: save via builtin debug function
|
|
|
|
# im._dump(filename)
|
|
|
|
|
|
|
|
#
|
|
|
|
# --------------------------------------------------------------------
|
|
|
|
|
2018-03-03 12:54:00 +03:00
|
|
|
|
2015-07-04 16:29:58 +03:00
|
|
|
Image.register_open(PpmImageFile.format, PpmImageFile, _accept)
|
|
|
|
Image.register_save(PpmImageFile.format, _save)
|
2010-07-31 06:52:47 +04:00
|
|
|
|
2016-04-25 07:59:02 +03:00
|
|
|
Image.register_extensions(PpmImageFile.format, [".pbm", ".pgm", ".ppm"])
|