def merge(self, pdf_one, pdf_two, filename='my.pdf', output_dir='D:/pdf/'):
'''
function:#pdfone为扫描的正面；#pdftwo为扫描的背面；#本函数实现将两个扫描文件按原有的顺序合并起来
:param pdf_one:
:param pdf_two:
:param filename:
:param output_dir:
:return:
'''
input_one = open(pdf_one, 'rb')
input_two = open(pdf_two, 'rb')
pdf_input_one = PdfFileReader(input_one)
pdf_input_two = PdfFileReader(input_two)
numOne = pdf_input_one.getNumPages()
numTwo = pdf_input_two.getNumPages()
print(numOne, numTwo)
pdf_output = PdfFileWriter()
index_one = 0
index_two = numTwo - 1
while True:
if index_one == numOne: break
print(index_one, index_two)
page1 = pdf_input_one.getPage(index_one)
pdf_output.addPage(page1)
page2 = pdf_input_two.getPage(index_two)
pdf_output.addPage(page2)
index_one += 1
index_two -= 1
pdf_name = output_dir + filename
output_stream = open(pdf_name, 'wb')
pdf_output.write(output_stream)
output_stream.close()
input_one.close()
input_two.close()
print('Done!')
def PdfPrettyPrint(inputname, outputname):
inputfile = open(inputname, 'rb')
wrt = PdfFileWriter()
ipt = PdfFileReader(inputfile)
#print ipt.getDocumentInfo()
pdfnums = ipt.getNumPages()
#print pdfnums
i = 0
while i < pdfnums:
page = ipt.getPage(i)
wrt.addPage(page)
if i + 2 < pdfnums:
page = ipt.getPage(i + 2)
wrt.addPage(page)
else:
wrt.addBlankPage()
if i + 1 < pdfnums:
page = ipt.getPage(i + 1)
page.rotateClockwise(180)
wrt.addPage(page)
else:
wrt.addBlankPage()
if i + 3 < pdfnums:
page = ipt.getPage(i + 3)
page.rotateClockwise(180)
wrt.addPage(page)
else:
wrt.addBlankPage()
i = i + 4
fl = open(outputname, "wb")
wrt.write(fl)
inputfile.close()
fl.close()
return True
def generate_a_pdf(filename, num_pages, dir=None):
"""function to generate a random PDF file of N pages with single image per page
taken from https://stackoverflow.com/questions/2925484/place-image-over-pdf
Args:
filename (str): path to save the pdf file
num_pages (int): number of pages to make the pdf file
KWArgs:
dir (str): the path to the directory to save the pdf file
Returns:
str. path to the new pdf file
"""
pdf = PdfFileWriter()
for num in range(1, num_pages+1):
imgTemp = BytesIO()
jpeg_path = make_a_jpeg('{}.jpeg'.format(str(num)), pick_a_color(num))
imgDoc = canvas.Canvas(imgTemp, pagesize=A4)
imgDoc.drawImage(jpeg_path, 25, 45)
imgDoc.save()
pdf.addPage(PdfFileReader(BytesIO(imgTemp.getvalue())).getPage(0))
remove(jpeg_path)
if dir:
path = join(dir, filename)
else:
path = join(getcwd(), filename)
pdf.write(open(path, 'wb'))
return path
def generate_images(path, save_dir_name, is_train):
if not os.path.exists('png_files/'):
os.mkdir('png_files/')
train_images = 'train_images/'
test_images = 'test_images/'
if is_train:
save_directory_path = 'png_files/'+ train_images + save_dir_name + '_annotated_images'
else:
save_directory_path = 'png_files/' + test_images + save_dir_name + '_annotated_images'
if not os.path.exists(save_directory_path):
os.makedirs(save_directory_path)
filename = path
print("Converting " + filename + " from pdf to PNG...")
reader = PdfFileReader(open(filename, mode="rb"))
try:
page_number = reader.getNumPages()
except:
page_number = reader.getNumPages() #PyPDF2 bug
with tempfile.TemporaryDirectory() as path:
images_from_path = convert_from_path(filename, dpi=72, output_folder=path, last_page=page_number, first_page=0)
i = 0
for page in images_from_path:
base_filename = os.path.splitext(os.path.basename(filename))[0] + '_' + str(i + 1) + '.png'
page.save(os.path.join(save_directory_path, base_filename), 'PNG')
i += 1
print('PDF file successfully converted.')
def test_cat(self):
"""Make sure files are properly concatenated."""
run_stapler(['cat', ONEPAGE_PDF, FIVEPAGE_PDF, self.outputfile])
self.assertTrue(os.path.isfile(self.outputfile))
with open(self.outputfile, 'rb') as outputfile:
pdf = PdfFileReader(outputfile)
self.assertEqual(pdf.getNumPages(), 6)
def translate(self):
'''读取pdf内容，并翻译，写入txt文件'''
f = open(self.fullPath, 'rb')
pdf = PdfFileReader(f)
index = 0
for i in range(0, pdf.getNumPages()):
extractedText = pdf.getPage(i).extractText()
content = extractedText.split('\n')
content = self.removeBlankFromList(content)
# 拼接之后的文本，如果单词间歇超过一个空格的，认为是需要换行处理的
content_list = self.enter_symbol(content)
for line in content_list:
line = line.strip()
if line:
ret = translate_func(line)
trans = ret if ret else '翻译失败'
self.write(line + '\n')
self.write(trans)
index += 1
print(index, end=' ', flush=True)
f.close()
Logger().write(self.fileName + '翻译完成，新文档：' + self.new_fullPath)
def metadata(path: Path) -> Metadata:
"""
Reads a given PDF file and produces a Metadata object.
:param path: path to a PDF file
:return: the metadata extracted from the PDF file
"""
with path.open('rb') as f:
reader = PdfFileReader(f)
info = reader.getDocumentInfo()
page_count = reader.getNumPages()
typer.echo(f'PDF metadata: {info}', err=True)
# Decide which possible title to use:
# - the title annotated in the PDF metadata
# - the title read by pdftitle (largest text on the first page)
# - the file name without extension
pdftitle_title = pdftitle.get_title_from_file(str(path))
typer.echo(f'Title according to pdftitle: {pdftitle_title}', err=True)
title_candidates = [t for t in [info.title, pdftitle_title, path.stem] if t is not None]
# The current heuristic is just to use the longest of the three candidates
title = max(title_candidates, key=len)
return Metadata(
title=title,
author=info.author,
page_count=page_count
)
def finalize_print_preparation(self):
"""Take the resulting multi page PDF and split into rotated single pages
Taken from `pythonlibrary.org
<https://www.blog.pythonlibrary.org/2018/04/11/splitting-and-merging-pdfs
-with-python/>`_ in combination with `johndcook.com
<https://www.johndcook.com/blog/2015/05/01/rotating-pdf-pages-with-python/>`_
"""
pdf: PdfFileReader = PdfFileReader(self._full_output_path_)
for page_number in range(pdf.getNumPages()):
pdf_writer: PdfFileWriter = PdfFileWriter()
page: PageObject = pdf.getPage(page_number)
page.rotateCounterClockwise(90)
pdf_writer.addPage(page)
output_filename: str = (f"{self._output_base_filename}_page_"
f"{str(page_number + 1).zfill(2)}.pdf")
with open(
os.path.join(self._output_directory_name, output_filename),
"wb") as pdf_out:
pdf_writer.write(pdf_out)
path_to_pdf = os.path.join(os.getcwd(), self._full_output_path_)
print(f"Create {pdf.getNumPages()} single paged PDFs.\n\n"
f"You can find them concatenated at file://"
f"{path_to_pdf}")
def _create_pdf_from_rtf_files(self):
pdfs = []
self.progress.emit(0)
for count, file in enumerate(self.files):
changed_file = change_filetype(file, "pdf", self.engine)
pdfs.append(changed_file)
self.progress.emit(count + 1)
merger = PdfFileMerger()
pages = []
chapters = []
for file in pdfs:
read_pdf = PdfFileReader(file)
txt = read_pdf.getPage(0)
page_content = txt.extractText()
try:
chapter = helper_functions.get_chapter_from_pdf_txt(
page_content)
chapters.append(chapter)
except:
chapter = os.path.basename(file)
chapter = chapter.split(".")[0]
chapter = chapter.replace("_", " ")
chapters.append(chapter)
pages.append(read_pdf.getNumPages())
merger.append(fileobj=file)
self.pages = pages
self.chapters = chapters
if not self.create_toc:
merger.write(self.master_file_name)
else:
merger.write("tmp.pdf")
merger.close()
self.trash += pdfs
def test_zip(self):
"""Test zip."""
run_stapler(['zip', ONEPAGE_PDF, FIVEPAGE_PDF, self.outputfile])
self.assertTrue(os.path.isfile(self.outputfile))
with open(self.outputfile, 'rb') as outputfile:
pdf = PdfFileReader(outputfile)
self.assertEqual(pdf.getNumPages(), 6)
def test_sel_range(self):
"""Test select of more pages from a PDF file."""
run_stapler(['cat', 'A=' + FIVEPAGE_PDF, 'A2-4', self.outputfile])
self.assertTrue(os.path.isfile(self.outputfile))
with open(self.outputfile, 'rb') as outputfile:
pdf = PdfFileReader(outputfile)
self.assertEqual(pdf.getNumPages(), 3)
def test_del_range(self):
"""Test del command for inverse select multiple pages."""
run_stapler(['del', 'A=' + FIVEPAGE_PDF, 'A2-4', self.outputfile])
self.assertTrue(os.path.isfile(self.outputfile))
with open(self.outputfile, 'rb') as outputfile:
pdf = PdfFileReader(outputfile)
self.assertEqual(pdf.getNumPages(), 2)
def extract_from_file(file: IO[bytes], filename: str, mime_type: str,
file_id: int) -> Tuple[Optional[str], Optional[int]]:
""" Returns the text and the page count """
parsed_text = None
page_count = None
if mime_type == "application/pdf" or mime_type.startswith(
"application/pdf;"):
try:
command = ["pdftotext", filename, "-"]
completed = subprocess.run(command,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
check=True)
parsed_text = completed.stdout.decode("utf-8", "ignore")
if completed.stderr:
logger.info("pdftotext: {}".format(completed.stderr))
except CalledProcessError as e:
logger.exception("File {}: Failed to run pdftotext: {}".format(
file_id, e))
try:
page_count = PdfFileReader(file,
strict=False,
overwriteWarnings=False).getNumPages()
except (PdfReadError, KeyError):
message = "File {}: Pdf does not allow to read the number of pages".format(
file_id)
logger.warning(message)
elif mime_type == "text/text":
parsed_text = file.read()
else:
logger.warning(
f"File {file_id} has an unknown mime type: '{mime_type}'")
return parsed_text, page_count
def extract_pdf_pypdf2(pdf_path):
with open(pdf_path, 'rb') as f:
pdf = PdfFileReader(f)
if pdf.isEncrypted:
pdf.decrypt('')
page_obj = pdf.getPage(2)
return page_obj.extractText()
def RemovePdfOwnerPassword(inputname, outputname):
'''
'''
inputfile = open(inputname, 'rb')
wrt = PdfFileWriter()
ipt = PdfFileReader(inputfile)
try:
ipt.decrypt("")
except KeyError as e:
if e.message == '/Encrypt':
print("%s is not an encrypted pdf" % inputname)
return -1
else:
raise e
print(ipt.getDocumentInfo())
size = ipt.getNumPages()
i = 0
while i < size:
page = ipt.getPage(i)
#print(page.extractText())
wrt.addPage(page)
i = i + 1
fl = open(outputname, "wb")
wrt.write(fl)
inputfile.close()
fl.close()
return 0
def run(self):
if self.beforeHandler(self._id, self.attachUrl):
return
filename = self.tempDir + str(random.random())
filename1 = self.tempDir + str(random.random()) + '.pdf'
try:
urllib.request.urlretrieve(self.attachUrl, filename)
input_stream = open(filename, 'rb')
pdf_input = PdfFileReader(input_stream)
pdf_output = PdfFileWriter()
page = 0
pages = pdf_input.getNumPages() - 1
# remove last page
while page < pages:
pdf_output.addPage(pdf_input.getPage(page))
page += 1
output_stream = open(filename1, 'wb')
pdf_output.write(output_stream)
output_stream.close()
input_stream.close()
if self.success is not None:
self.success(self._id, filename1)
except Exception as e:
if self.error is not None:
self.error(e, self.attachUrl)
finally:
if os.path.exists(filename):
os.remove(filename)
if os.path.exists(filename1):
os.remove(filename1)
def readPDFfile(infile):
pdf = PdfFileReader(infile, "rb"))
content = ""
num = pdf.getNumPages()
for i in range(0, num):
extractedText = pdf.getPage(i).extractText()
content += extractedText + "\n"
return content
def getDataUsingPyPdf2(filename):
pdf = PdfFileReader(open(filename, "rb"))
content = ""
num = pdf.getNumPages()
for i in range(0, num):
extractedText = pdf.getPage(i).extractText()
content += extractedText + "\n"
return content
def getPdffileBookmark(filename, bookmark_file_savepath):
pdf = PdfFileReader(open(filename, "rb"))
pagecount = pdf.getNumPages()
print('pagecount:', pagecount)
pageLabels = {
} #真实页码的索引 indirectRef “{'/Type': '/Fit', '/Page': IndirectObject(7871, 0), '/Title': '封面'}”
for i in range(pagecount):
page = pdf.getPage(i)
pageLabels[page.indirectRef.idnum] = i + 1
# print(page.indirectRef.idnum,i+1)
bookmark_file = codecs.open(bookmark_file_savepath, 'w', encoding='utf-8')
title = []
pagedir = []
bookmark_jibie = []
outlines = pdf.getOutlines()
print(outlines)
index = 0
jibie = 0
for outline in outlines:
index += 1
jibie = 0
print(len(outline), outline)
if type(outline) == PyPDF2.generic.Destination:
# print('dict--------')
# print(list(outline.keys()))
# for x,j in enumerate(list(outline.keys())):
# print(str(outline[j]))
# print(outline['/Title'])
# print(outline['/Type'])
# print(outline.page.idnum)
bookmark_file.write(outline['/Title'] + '\t' +
str(pageLabels[outline.page.idnum]) + '\r\n')
if type(outline) == list:
# print('list')
jibie = 1
for i, outline in enumerate(outline):
if type(outline) == PyPDF2.generic.Destination:
bookmark_file.write('\t' * jibie + outline['/Title'] +
'\t' +
str(pageLabels[outline.page.idnum]) +
'\r\n')
elif type(outline) == list:
jibie = 2
for i, o in enumerate(outline):
if type(outline) == PyPDF2.generic.Destination:
bookmark_file.write(
'\t' * jibie + outline['/Title'] + '\t' +
str(pageLabels[outline.page.idnum]) + '\r\n')
# print('\n')
# if index>=3:
# break
bookmark_file.close()
def convertPDFAlternative(self, path):
from PyPDF2.pdf import PdfFileReader
if not os.path.exists(path):
return False
pdf = PdfFileReader(open(path, "rb"))
for i in range(0, pdf.getNumPages()):
print(i)
extractedText = pdf.getPage(i).extractText()
self.pages.append(extractedText)
return True
def merge_pdf(file_list, output_path):
'''合并 PDF'''
outpdf = PdfFileWriter()
for f in file_list:
f_pdf = PdfFileReader(open(f, 'rb'))
for page in f_pdf.pages:
outpdf.addPage(page)
ous = open(output_path, 'wb')
outpdf.write(ous)
ous.close()
def test_split(self):
"""Make sure a file is properly split into pages."""
run_stapler(['split', FIVEPAGE_PDF])
filelist = os.listdir(self.tmpdir)
self.assertEqual(len(filelist), 5)
for f in os.listdir(self.tmpdir):
with open(os.path.join(self.tmpdir, f), 'rb') as pdf_file:
pdf = PdfFileReader(pdf_file)
self.assertEqual(pdf.getNumPages(), 1)
def get(self, request, *args, **kwargs):
fontname_g = "HeiseiMin-W3"
pdfmetrics.registerFont(UnicodeCIDFont(fontname_g))
reader = PdfFileReader('media/pdf/riyuu-format4.pdf')
writer = PdfFileWriter()
buffer = io.BytesIO()
cc = canvas.Canvas(buffer)
cc.setFont(fontname_g, 11)
initial = 295
before_rect_x = 748
after_rect_x = 776.5
line_height = 11.9
input_list = [{
'label': '便器からの立ち座り',
'before_flag': True,
'after_flag': False
}, {
'label': 'トイレまでの移動',
'before_flag': False,
'after_flag': True
}, {
'label': 'トイレ出入口の出入（扉の開閉含む）',
'before_flag': True,
'after_flag': False
}]
welfare_equipment_material = PdfMaterial.objects.get(
key="welfare_equipment")
cc = self.motion_purpose_draw(cc, before_rect_x, after_rect_x,
welfare_equipment_material.materials,
input_list, initial, line_height)
cc.showPage()
cc.save()
buffer.seek(0)
new_pdf = PdfFileReader(buffer)
existing_page = reader.getPage(0)
existing_page.mergePage(new_pdf.getPage(0))
writer.addPage(existing_page)
new = io.BytesIO()
writer.write(new)
new.seek(0)
print('finish')
return FileResponse(new, as_attachment=True, filename='hello.pdf')
def getDataUsingPyPdf2(filename):
pdf = PdfFileReader(open(filename, "rb"))
content = ""
for i in range(0, pdf.getNumPages()):
#print(str(i))
extractedText = pdf.getPage(i).extractText()
content += extractedText + "\n"
content = " ".join(content.replace("\xa0", " ").strip().split())
return content.encode("ascii", "ignore")
def _extract_pdf_forms(self, fname):
"""Extracts interactive form fields data from a PDF file.
Parameters:
fname (str): Path to PDF file.
Returns:
dict: Form fields data extracted.
"""
f = PdfFileReader(fname)
return f.getFields()
def get(self, request, *args, **kwargs):
fontname_g = "HeiseiKakuGo-W5"
pdfmetrics.registerFont(UnicodeCIDFont(fontname_g))
buffer = io.BytesIO()
cc = canvas.Canvas(buffer)
reader = PdfFileReader('media/pdf/sample.pdf')
existing_page = reader.getPage(0)
cc.setFont(fontname_g, 24)
cc.drawString(0, 820, "テスト")
cc.showPage()
cc.save()
buffer.seek(0)
new_pdf = PdfFileReader(buffer)
existing_page.mergePage(new_pdf.getPage(0))
writer = PdfFileWriter()
writer.addPage(existing_page)
new = io.BytesIO()
writer.write(new)
new.seek(0)
return FileResponse(new, as_attachment=True, filename='hello.pdf')
def test_render_to_pdf_bytes(self):
doc = self._get_docx('complex_fields')
data = doc.render(context={
'complex':
'BEEES. AAAAH. BEEEEES',
'complex2':
'ಠ_ಠ unifying matrix conventions is the way of the future, Max.'
},
format='pdf')
# does pypdf like our output?
PdfFileReader(BytesIO(data))
def pdfSplit(pdf_main, pdf_part):
try:
pdf_read_obj = PdfFileReader(pdf_main)
pdf_write_obj = PdfFileWriter()
page_num = pdf_read_obj.getNumPages()
page_last_obj = pdf_read_obj.getPage(page_num - 1)
page_last_obj.rotateClockwise(90)
pdf_write_obj.addPage(page_last_obj)
pdf_write_obj.write(open(pdf_part, 'wb'))
return page_num - 1
except Exception as e:
return False
def process_files(files):
title_map = {}
for file in files:
# parse filename
title, _, _ = util.parse_file(os.path.basename(file))
# if not already seen, add this file to the map
if title not in title_map:
title_map[title] = file
continue
# we need to merge pdfs into 1
merger = PdfFileMerger()
with open(title_map[title], "rb") as f:
merger.append(PdfFileReader(f))
with open(file, "rb") as f:
merger.append(PdfFileReader(f))
merger.write(title_map[title])
os.remove(file)
return title_map
def set_bglj(self, filename):
if filename.endswith('.pdf'):
self.user_data['bgms'] = '1'
# 获取PDF 页码
try:
pdf_read_obj = PdfFileReader(filename)
self.user_data['bgym'] = pdf_read_obj.getNumPages()
except Exception as e:
self.user_data['bgym'] = 0
print("获取报告页码出错，错误信息：%s" % e)
else:
self.user_data['bgms'] = '0'
self.user_data['bglj'] = self.cur_dir
