def pdf_printer(self, pdf_input_path):
args = ['pdf_printer',
'-dNOPAUSE',
'-sDEVICE=mswinpr2',
pdf_input_path]
encoding = locale.getpreferredencoding()
args = [a.encode(encoding) for a in args]
ghostscript.Ghostscript(*args)
ghostscript.cleanup()
def pdf_to_jpg(pdf_input_path, jpeg_name):
args = [
"pdf2jpeg", # actual value doesn't matter
"-dNOPAUSE",
"-sDEVICE=jpeg",
"-r300",
f'-sOutputFile={jpeg_name}-%03d.jpg',
pdf_input_path
]
encoding = locale.getpreferredencoding()
args = [a.encode(encoding) for a in args]
with ghostscript.Ghostscript(*args) as g:
ghostscript.cleanup()
def pdf2jpeg(pdf_input_path, jpeg_output_path):
ghostscript.cleanup()
args = [
"pef2jpeg", # actual value doesn't matter
"-dNOPAUSE",
"-sDEVICE=jpeg",
"-r144",
"-sOutputFile=" + jpeg_output_path,
pdf_input_path
]
encoding = locale.getpreferredencoding()
args = [a.encode(encoding) for a in args]
ghostscript.Ghostscript(*args)
def convert_to_image(self):
"""
Convert the PDF to tiff image.
:return: self (allows chaining of methods, since the methods do not return any additional info).
"""
if os.path.exists(self.pdf_file_spec):
start_conversion = perf_counter()
args = [
"pdf2tiff",
"-dNOPAUSE",
"-dSAFER",
"-dBATCH",
f"-dNumRenderingThreads={self.threads}",
f"-q",
f"-sDEVICE={self.IMAGE_FORMAT}",
f"-r{self.dpi}",
f"-sOutputFile={os.path.abspath(f'{self.output_file}-%00d.{self.extension}')}",
f"{self.pdf_file_spec}",
]
encoding = locale.getpreferredencoding()
args = [a.encode(encoding) for a in args]
# Convert the PDF to the TIFF (Need to clean up instance after execution,
# to allow conversion of additional documents)
try:
gs_apis = ghostscript.Ghostscript(*args)
gs_apis.exit()
ghostscript.cleanup()
except Exception as exc:
print(f"\tERROR ({self.name}): Exception: {exc}")
# Measure time to convert the PDF to image files.
self.conversion_duration = perf_counter() - start_conversion
print(
f"{self.name}: Conversion took: {self.conversion_duration:0.4f} seconds."
)
# Specified PDF was not found.
else:
print(f"{self.name}: Unable to find '{self.pdf_file_spec}'")
return self
def pdf2png(self, pdf_input_path, png_output_path):
args = [
"pdf2png", # actual value doesn't matter
"-dNOPAUSE",
"-sDEVICE=pngmono",
"-r300",
"-sOutputFile=" + png_output_path,
pdf_input_path
]
encoding = locale.getpreferredencoding()
args = [a.encode(encoding) for a in args]
try:
ghostscript.Ghostscript(*args)
ghostscript.cleanup()
except:
print("Erro", ghostscript.GhostscriptError)
def pdf2jpeg(self, pdf_input_path, jpeg_output_path):
args = [
"gs", # actual value doesn't matter
"--permit-file-read=" + os.path.dirname(pdf_input_path),
"--permit-file-write=" + jpeg_output_path,
"-dNOPAUSE",
#"-dBATCH",
"-sDEVICE=jpeg",
"-dTextAlphaBits=4",
"-r300",
#"-sOutputFile=" + jpeg_output_path,
"-o a%03d.jpg",
pdf_input_path
]
encoding = locale.getpreferredencoding()
args = [a.encode(encoding) for a in args]
with ghostscript.Ghostscript(*args) as g:
ghostscript.cleanup()
def process_input_from_request(request):
file_name = mime_type = ""
try:
image = request.FILES['image']
print("IMAGE ", image)
# import pdb;pdb.set_trace()
except MultiValueDictKeyError:
try:
file_url = request.POST['image']
if not file_url:
raise MultiValueDictKeyError
file_name = file_url.split('/')[-1]
image = download_image(file_url) if is_uri(file_url) else False
if not image:
raise Exception
except MultiValueDictKeyError:
return HttpResponse({
'status': 'FAIL',
'status_code': 204,
'message': "No Content",
'file_name': None,
'file_type': None,
'url': str(request.path)
}), [], "", "", ""
except Exception as err:
return HttpResponse({
'status': 'FAIL',
'status_code': 204,
'message': "Invalid URI",
'file_name': None,
'file_type': None,
'url': str(request.path)
}), [], "", "", ""
file_name = image.name if not file_name else file_name
mime_type = magic.from_buffer(image.read(1024), mime=True)
file_size = image.getbuffer().nbytes if isinstance(
image, io.BytesIO) else image.size
try:
assert hasattr(image, "read")
except AssertionError:
return HttpResponse({
'status': 'FAIL',
'status_code': 204,
'message': "File Not Readable",
'file_name': file_name,
'file_type': mime_type.split('/')[-1],
'url': str(request.path)
}), [], "", "", ""
if mime_type in ['application/pdf']:
image.seek(0) if isinstance(image, io.BytesIO) else image.open()
pages = []
max_confidence_index = ''
int_, temp_local_filename = tempfile.mkstemp()
f = os.fdopen(int_, 'wb')
f.write(image.read()) # write the tmp file
f.close()
temp_local_dir = tempfile.mkdtemp()
gs_args = [
"pdf2png",
"-dSAFER -dBATCH -dNOPAUSE",
"-r300",
"-sDEVICE=pnggray",
"-dTextAlphaBits=4 -sPAPERSIZE=a4",
"-o",
temp_local_dir + "page-%02d.png",
temp_local_filename,
]
encoding = locale.getpreferredencoding()
gs_args = [gs_arg.encode(encoding) for gs_arg in gs_args]
with ghostscript.Ghostscript(*gs_args) as g:
ghostscript.cleanup()
files = sorted(glob.glob(temp_local_dir + "*.png"))
print(files)
return None, files, file_name, mime_type, file_size
elif mime_type == 'image/tiff':
file = []
_, temp_local_filename = tempfile.mkstemp()
image.seek(0) if isinstance(image, io.BytesIO) else image.open()
pdf_tiff = wi(file=image, resolution=180)
for i, page in enumerate(pdf_tiff.sequence):
with wi(page) as page_image:
page_image.alpha_channel = False
img_buffer = np.asarray(bytearray(
page_image.make_blob(format='jpeg')),
dtype='uint8')
bytesio = io.BytesIO(img_buffer)
image = Image.open(bytesio)
image.save(temp_local_filename + '_' + str(i) + '.jpeg')
file.append(temp_local_filename + '_' + str(i) + '.jpeg')
# import pdb;pdb.set_trace()
return None, file, file_name, mime_type, file_size
# return None, [temp_local_filename], file_name, mime_type, file_size
elif mime_type == 'text/html':
return HttpResponse(
status='FAIL',
status_code=204,
message="Not a file object",
file_name=None,
file_type=None,
url=str(
request.path)).to_dict(), [], file_name, mime_type, file_size
else:
temp_local_filename = tempfile.mkstemp()
fix_image_orientation_using_exif(image, temp_local_filename)
return None, [temp_local_filename[1]], file_name, mime_type, file_size
