def img_extract(): #The function to Extract the images from the Document
print(Style.BRIGHT + Fore.CYAN + '\nEnter Your Option:\n')
print(Style.BRIGHT + Fore.YELLOW +
'1 - Extracting the Images in Specific Pages.')
print(Style.BRIGHT + Fore.YELLOW +
'2 - Extracting Every Images of a Document.')
print(Style.BRIGHT + Fore.RED + '\nYour Option >>>', end=' ')
n = int(input())
if (n != 1 and n != 2):
raise Exception(
'Please enter only the above given numerical values...')
input_path, filename = get_path()
pdf = pikepdf.Pdf.open(os.path.join(input_path, filename))
if (n == 1):
print(
Style.BRIGHT + Fore.BLUE +
'\nEnter the Page Numbers of the Images to be Extracted (Each number should be separated by Comma(,)): ',
end='')
page_nums = list(map(int, input().split(',')))
for page in page_nums:
l = list(pdf.pages[page - 1].images.keys())
if len(l) == 0:
print(Style.DIM + Fore.LIGHTRED_EX +
'There is No Image present in the Given Page:{}'.format(
page))
else:
for image in l:
raw_image = pdf.pages[page - 1].images[image]
pdfimage = pikepdf.PdfImage(raw_image)
pdfimage.extract_to(fileprefix=os.path.join(
os.getcwd(),
str(page) + '_' + image[1:] + '_' + filename))
print(Style.BRIGHT + Fore.RED + '\n----------FINISHED------------')
pdf.close()
else:
for page in range(len(pdf.pages)):
l = list(pdf.pages[page].images.keys())
if len(l) == 0:
print(Style.DIM + Fore.LIGHTRED_EX +
'There is No Image present in the Page:{}'.format(page))
else:
for image in l:
raw_image = pdf.pages[page].images[image]
pdfimage = pikepdf.PdfImage(raw_image)
pdfimage.extract_to(fileprefix=os.path.join(
os.getcwd(),
str(page) + '_' + image[1:] + '_' + filename))
print(Style.BRIGHT + Fore.RED +
'\n----------FINISHED------------')
pdf.close()
def test_jbig2_lossy(lossy, resources, outpdf):
args = [
resources / 'ccitt.pdf',
outpdf,
'--image-dpi',
'200',
'--optimize',
3,
'--jpg-quality',
'50',
'--png-quality',
'20',
'--plugin',
'tests/plugins/tesseract_noop.py',
]
if lossy:
args.append('--jbig2-lossy')
check_ocrmypdf(*args)
pdf = pikepdf.open(outpdf)
pim = pikepdf.PdfImage(next(iter(pdf.pages[0].images.values())))
assert pim.filters[0] == '/JBIG2Decode'
if lossy:
assert '/JBIG2Globals' in pim.decode_parms[0]
else:
assert len(pim.decode_parms) == 0
def test_flate_to_jbig2(resources, outdir):
# This test requires an image that pngquant is capable of converting to
# to 1bpp - so use an existing 1bpp image, convert up, confirm it can
# convert down
with Image.open(fspath(resources / 'typewriter.png')) as im:
assert im.mode in ('1', 'P')
im = im.convert('L')
im.save(fspath(outdir / 'type8.png'))
check_ocrmypdf(
outdir / 'type8.png',
outdir / 'out.pdf',
'--image-dpi',
'100',
'--png-quality',
'50',
'--optimize',
'3',
'--plugin',
'tests/plugins/tesseract_noop.py',
)
pdf = pikepdf.open(outdir / 'out.pdf')
pim = pikepdf.PdfImage(next(iter(pdf.pages[0].images.values())))
assert pim.filters[0] == '/JBIG2Decode'
def __init__(
self,
*,
name='',
pdfimage: Optional[Object] = None,
inline: Optional[Object] = None,
shorthand=None,
):
self._name = str(name)
self._shorthand = shorthand
if inline is not None:
self._origin = 'inline'
pim = inline.iimage
elif pdfimage is not None:
self._origin = 'xobject'
pim = pikepdf.PdfImage(pdfimage)
else:
raise ValueError("Either pdfimage or inline must be set")
self._width = pim.width
self._height = pim.height
# If /ImageMask is true, then this image is a stencil mask
# (Images that draw with this stencil mask will have a reference to
# it in their /Mask, but we don't actually need that information)
if pim.image_mask:
self._type = 'stencil'
else:
self._type = 'image'
self._bpc = int(pim.bits_per_component)
try:
self._enc = FRIENDLY_ENCODING.get(pim.filters[0], 'image')
except IndexError:
self._enc = '?'
try:
self._color = FRIENDLY_COLORSPACE.get(pim.colorspace, '?')
except NotImplementedError:
self._color = '?'
if self._enc == Encoding.jpeg2000:
self._color = Colorspace.jpeg2000
if self._color == Colorspace.icc:
# Check the ICC profile to determine actual colorspace
pim_icc = pim.icc
if pim_icc.profile.xcolor_space == 'GRAY':
self._comp = 1
elif pim_icc.profile.xcolor_space == 'CMYK':
self._comp = 4
else:
self._comp = 3
else:
self._comp = FRIENDLY_COMP.get(self._color, '?')
# Bit of a hack... infer grayscale if component count is uncertain
# but encoding only supports monochrome.
if self._comp == '?' and self._enc in (Encoding.ccitt,
Encoding.jbig2):
self._comp = FRIENDLY_COMP[Colorspace.gray]
def check_pim(imobj, idx):
pim = pikepdf.PdfImage(imobj)
assert pim.mode == 'DeviceN'
assert pim.is_device_n
assert not pim.is_separation
assert pim.indexed == idx
assert repr(pim)
with pytest.raises(pikepdf.models.image.HifiPrintImageNotTranscodableError):
pim.extract_to(stream=BytesIO())
def test_palette_nonrgb(base, hival, palette, expect_type):
pdf = pikepdf.new()
imobj = Stream(
pdf,
b'\x00\x01\x02\x03' * 4,
BitsPerComponent=8,
ColorSpace=Array([Name.Indexed, base, hival, palette]),
Width=16,
Height=1,
Type=Name.XObject,
Subtype=Name.Image,
)
pim = pikepdf.PdfImage(imobj)
assert pim.palette == (expect_type, palette)
def test_palette_nonrgb(base, hival, bits, palette, expect_type, expect_mode):
pdf = pikepdf.new()
imobj = Stream(
pdf,
b'\x00\x01\x02\x03' * 16,
BitsPerComponent=bits,
ColorSpace=Array([Name.Indexed, base, hival, palette]),
Width=16,
Height=4,
Type=Name.XObject,
Subtype=Name.Image,
)
pim = pikepdf.PdfImage(imobj)
assert pim.palette == (expect_type, palette)
pim.extract_to(stream=BytesIO())
# To view images:
# pim.extract_to(fileprefix=f'palette_nonrgb_{expect_type}_{bits}')
assert pim.mode == expect_mode
def __init__(self, *, name='', pdfimage=None, inline=None,
shorthand=None):
self._name = str(name)
self._shorthand = shorthand
if inline is not None:
self._origin = 'inline'
pim = inline.iimage
elif pdfimage is not None:
self._origin = 'xobject'
pim = pikepdf.PdfImage(pdfimage)
self._width = pim.width
self._height = pim.height
# If /ImageMask is true, then this image is a stencil mask
# (Images that draw with this stencil mask will have a reference to
# it in their /Mask, but we don't actually need that information)
if pim.image_mask:
self._type = 'stencil'
else:
self._type = 'image'
self._bpc = int(pim.bits_per_component)
try:
self._enc = FRIENDLY_ENCODING.get(pim.filters[0], 'image')
except IndexError:
self._enc = '?'
try:
self._color = FRIENDLY_COLORSPACE.get(pim.colorspace, '?')
except NotImplementedError:
self._color = '?'
if self._enc == Encoding.jpeg2000:
self._color = Colorspace.jpeg2000
self._comp = FRIENDLY_COMP.get(self._color, '?')
# Bit of a hack... infer grayscale if component count is uncertain
# but encoding must be monochrome. This happens if a monochrome image
# has an ICC profile attached. Better solution would be to examine
# the ICC profile.
if self._comp == '?' and self._enc in (Encoding.ccitt, 'jbig2'):
self._comp = FRIENDLY_COMP[Colorspace.gray]
def extract_image_filter(pike, root, log, image, xref):
if image.Subtype != Name.Image:
return None
if image.Length < 100:
log.debug("Skipping small image, xref %s", xref)
return None
pim = pikepdf.PdfImage(image)
if len(pim.filter_decodeparms) > 1:
log.debug("Skipping multiply filtered, xref %s", xref)
return None
filtdp = pim.filter_decodeparms[0]
if pim.bits_per_component > 8:
return None # Don't mess with wide gamut images
if filtdp[0] == Name.JPXDecode:
return None # Don't do JPEG2000
return pim, filtdp
def test_dict_or_array_dict():
pdf = pikepdf.new()
imobj = Stream(
pdf,
b'dummy',
BitsPerComponent=1,
ColorSpace=Name.DeviceGray,
DecodeParms=Array([Dictionary(
BlackIs1=False,
Columns=16,
K=-1,
)]),
Filter=Array([Name.CCITTFaxDecode]),
Height=16,
Width=16,
Type=Name.XObject,
Subtype=Name.Image,
)
pim = pikepdf.PdfImage(imobj)
assert pim.decode_parms[
0].K == -1 # Check that array of dict is unpacked properly
def decodeImage(page, prefix):
xObject = page.images
count = 0
for name, img in xObject.items():
count += 1
if img['/Filter'] in ['/DCTDecode', '/JPXDecode', '/FlatDecode']:
try:
pdfimage = pikepdf.PdfImage(img)
input = imdecode(
np.asarray(bytearray(img.get_raw_stream_buffer())),
IMREAD_COLOR)
suffixes = {
'/JPXDecode': 'jp2',
'/DCTDecode': 'jpg',
'/FlatDecode': 'png'
}
suffixes[img['/Filter']]
imwrite(
prefix + '_%03d.%s' % (count, suffixes[img['/Filter']]),
input)
except Exception as ex:
log(ERROR,
'Cannot extract %s:%s due to %s' % (prefix, name, str(ex)))
def extract_image(*, pike, root, log, image, xref, jbig2s,
pngs, jpegs, options):
if image.Subtype != '/Image':
return False
if image.Length < 100:
log.debug("Skipping small image, xref {}".format(xref))
return False
pim = pikepdf.PdfImage(image)
if len(pim.filter_decodeparms) > 1:
log.debug("Skipping multiply filtered, xref {}".format(xref))
return False
filtdp = pim.filter_decodeparms[0]
if pim.bits_per_component > 8:
return False # Don't mess with wide gamut images
if filtdp[0] == '/JPXDecode':
return False # Don't do JPEG2000
if pim.bits_per_component == 1 \
and filtdp != '/JBIG2Decode' \
and jbig2enc.available():
try:
imgname = Path(root / '{:08d}'.format(xref))
with imgname.open('wb') as f:
ext = pim.extract_to(stream=f)
imgname.rename(imgname.with_suffix(ext))
except pikepdf.UnsupportedImageTypeError:
return False
jbig2s.append((xref, ext))
elif filtdp[0] == '/DCTDecode' \
and options.optimize >= 2:
# This is a simple heuristic derived from some training data, that has
# about a 70% chance of guessing whether the JPEG is high quality,
# and possibly recompressible, or not. The number itself doesn't mean
# anything.
# bytes_per_pixel = int(raw_jpeg.Length) / (w * h)
# jpeg_quality_estimate = 117.0 * (bytes_per_pixel ** 0.213)
# if jpeg_quality_estimate < 65:
# return False
# We could get the ICC profile here, but there's no need to look at it
# for quality transcoding
# if icc:
# stream = BytesIO(raw_jpeg.read_raw_bytes())
# iccbytes = icc.read_bytes()
# with Image.open(stream) as im:
# im.save(jpg_name(root, xref), icc_profile=iccbytes)
try:
imgname = Path(root / '{:08d}'.format(xref))
with imgname.open('wb') as f:
ext = pim.extract_to(stream=f)
imgname.rename(imgname.with_suffix(ext))
except pikepdf.UnsupportedImageTypeError:
return False
jpegs.append(xref)
elif pim.indexed \
and pim.colorspace in pim.SIMPLE_COLORSPACES \
and options.optimize >= 3:
# Try to improve on indexed images - these are far from low hanging
# fruit in most cases
pim.as_pil_image().save(png_name(root, xref))
pngs.append(xref)
elif not pim.indexed and pim.colorspace in pim.SIMPLE_COLORSPACES:
# An optimization opportunity here, not currently taken, is directly
# generating a PNG from compressed data
pim.as_pil_image().save(png_name(root, xref))
pngs.append(xref)
else:
return False
return True
def test_display_image(pal):
im0 = pal.pages[0].Resources.XObject['/Im0']
pim = pikepdf.PdfImage(im0)
result = pim._repr_png_()
assert result[1:4] == b'PNG'
