def check_file_content(original, converted):
original_pdf = PdfFileReader(open(original, mode="rb"), strict=False)
original_page_num = original_pdf.numPages
with open(converted, mode="rb") as converted_data:
converted_pdf = PdfFileReader(converted_data, strict=False)
page = PageObject(converted_data)
if (page.getContents() is None or original_page_num != converted_pdf.numPages):
return False
return True
def build_folio(pages, padding=None, starting_page_number=None):
assert(len(pages) <= 12)
collation_patterns = {4:[3,0,1,2], 8:[7,0,1,6,5,2,3,4], 12:[11,0,1,10,9,2,3,8,7,4,5,6]}
if not padding:
# If no padding page was provided, create a blank page to pad with.
padding = ('blank', PageObject.createBlankPage(None, 8.5*72, 11*72), 'landscape')
# pad out the folio till it is 4, 8, or 12 pages long.
while len(pages) not in collation_patterns and len(pages) <= 12:
pages.append(padding)
print(len(pages))
assert(len(pages) <= 12)
collated = [pages[n] for n in collation_patterns[len(pages)]]
joined = []
for n in range(int(len(collated)/2)):
page1 = collated[n*2+0]
# if the width is greater than the height, the document is in landscape and will need to be rotated.
rot1 = page1['/MediaBox'][2] > page1['/MediaBox'][3]
page2 = collated[n*2+1]
rot2 = page2['/MediaBox'][2] > page2['/MediaBox'][3]
if starting_page_number:
lpage = starting_page_number + collation_patterns[len(pages)][n*2+0]
rpage = starting_page_number + collation_patterns[len(pages)][n*2+1]
joined.append(create_page(page1, page2, rot1, rot2, lpage, rpage))
else:
joined.append(create_page(page1, page2, rot1, rot2))
return joined
def merge_to_page(page1, page2):
outs = PageObject.createBlankPage(None, 17*72, 11*72)
outs.mergeScaledTranslatedPage(p1, 1, 0, 0)
outs.mergeScaledTranslatedPage(p2, 1, 8.5*72, 0)
outs.scaleTo(11*72,8.5*72)
outs.rotateCounterClockwise(90)
return outs
def pypdf3():
"""Much slower than PyPDF3 method."""
# 5b. Get our files ready
document_reader = PdfFileReader(document)
output_file = PdfFileWriter()
# Number of pages in input document
page_count = document_reader.getNumPages()
# Watermark objects
watermark_reader = PdfFileReader(watermark)
wtrmrk_page = watermark_reader.getPage(0)
wtrmrk_width = (wtrmrk_page.mediaBox.getWidth() / 2) + 0
wtrmrk_height = (wtrmrk_page.mediaBox.getHeight() / 2) + 80
wtrmrk_rotate = -int(Info(watermark_reader).rotate) if Info(watermark_reader).rotate is not None else 0
# 5c. Go through all the input file pages to add a watermark to them
for page_number in range(page_count):
# Merge the watermark with the page
if not self.underneath:
input_page = document_reader.getPage(page_number)
if wtrmrk_rotate != 0:
input_page.mergeRotatedTranslatedPage(wtrmrk_page, wtrmrk_rotate, wtrmrk_width, wtrmrk_height)
else:
wtrmrk_width = 0
wtrmrk_height = 0
input_page.mergeTranslatedPage(wtrmrk_page, wtrmrk_width, wtrmrk_height)
else:
size = Info(document_reader).dimensions
input_page = PageObject().createBlankPage(document_reader, size['w'], size['h'])
if wtrmrk_rotate != 0:
input_page.mergeRotatedTranslatedPage(wtrmrk_page, wtrmrk_rotate, wtrmrk_width, wtrmrk_height)
else:
wtrmrk_width = 0
wtrmrk_height = 0
input_page.mergeTranslatedPage(wtrmrk_page, wtrmrk_width, wtrmrk_height)
input_page.mergePage(document_reader.getPage(page_number))
# Add page from input file to output document
output_file.addPage(input_page)
# 5d. finally, write "output" to PDF
with open(output_filename, "wb") as outputStream:
output_file.write(outputStream)
return output_filename
def pypdf3(self):
reader = PdfFileReader(self.file_name)
writer = PdfFileWriter()
# Number of pages in input document
page_count = reader.getNumPages()
for page_number in range(page_count):
wtrmrk = reader.getPage(page_number)
page = PageObject.createBlankPage(width=self.target_w,
height=self.target_h)
page.mergeScaledTranslatedPage(wtrmrk, self.scale, self.margin_x,
self.margin_y)
writer.addPage(page)
with open(self.output, "wb") as outputStream:
writer.write(outputStream)
return self.output
def create_page(p1, p2, rot1=False, rot2=False, lpage_num=None, rpage_num=None):
outs = PageObject.createBlankPage(None, 17*72, 11*72)
if rot1:
outs.mergeRotatedScaledTranslatedPage(p1, 90, 1, 8.5*72, 0)
else:
outs.mergeScaledTranslatedPage(p1, 1, 0, 0)
if rot2:
outs.mergeRotatedScaledTranslatedPage(p2, 90, 1, 8.5*72*2, 0)
else:
outs.mergeScaledTranslatedPage(p2, 1, 8.5*72, 0)
outs.scaleTo(11*72,8.5*72)
outs.rotateCounterClockwise(90)
temp = NumberPDF(lpage_num, rpage_num)
temp.add_page(orientation='L')
temp.output('/tmp/pntemp.pdf')
num_page = PdfFileReader('/tmp/pntemp.pdf').getPage(0)
outs.mergePage(num_page)
return outs
def upscale(file_name,
scale=1.5,
margin_x=0,
margin_y=0,
suffix='scaled',
tempdir=None):
"""Upscale a PDF to a large size."""
# Set output file name
if tempdir:
output = NamedTemporaryFile(suffix='.pdf', dir=tempdir,
delete=False).name
elif suffix:
output = os.path.join(os.path.dirname(file_name),
add_suffix(file_name, suffix))
else:
output = NamedTemporaryFile(suffix='.pdf').name
reader = PdfFileReader(file_name)
writer = PdfFileWriter()
dims = dimensions(file_name)
target_w = dims['w'] * scale
target_h = dims['h'] * scale
# Number of pages in input document
page_count = reader.getNumPages()
for page_number in range(page_count):
wtrmrk = reader.getPage(page_number)
page = PageObject.createBlankPage(width=target_w, height=target_h)
page.mergeScaledTranslatedPage(wtrmrk, scale, margin_x, margin_y)
writer.addPage(page)
with open(output, "wb") as outputStream:
writer.write(outputStream)
return output