def convert(self, filename):
'''Convert PDF file from sample path to output path.'''
source_pdf_file = os.path.join(sample_path, f'{filename}.pdf')
docx_file = os.path.join(output_path, f'{filename}.docx')
cv = Converter(source_pdf_file)
cv.convert(docx_file)
cv.close()
def init_test(self, filename):
''' Initialize parsed layout and benchmark layout.'''
# parsed layout: first page only
pdf_file = os.path.join(self.sample_dir, f'{filename}.pdf')
docx_file = os.path.join(self.output_dir, f'{filename}.docx')
cv = Converter(pdf_file)
cv.convert(docx_file, pages=[0])
self.test = cv[0] # type: Page
cv.close()
# restore sample layout
cv = Converter(pdf_file)
layout_file = os.path.join(self.layout_dir, f'{filename}.json')
cv.deserialize(layout_file)
self.sample = cv[0] # type: Page
return self
def init_test(self, filename):
''' Initialize parsed layout and benchmark layout.'''
# restore sample layout
layout_file = os.path.join(self.layout_dir, f'{filename}.json')
with open(layout_file, 'r') as f:
raw_dict = json.load(f)
self.sample = Layout().restore(raw_dict)
# parsed layout: first page only
pdf_file = os.path.join(self.sample_dir, f'{filename}.pdf')
docx_file = os.path.join(self.output_dir, f'{filename}.docx')
cv = Converter(pdf_file)
layouts = cv.make_docx(docx_file, pages=[0])
self.test = layouts[0] # type: Layout
cv.close()
return self
def post(self, request, format=None):
data = request.data['file']
print(data.__dict__)
try:
unique_filename = str(uuid.uuid4())
docs_file_name_path = 'docx/' + unique_filename + '.docx'
file_path = 'pdf/' + unique_filename + '.pdf'
path = default_storage.save(file_path, ContentFile(data.read()))
tmp_file = os.path.join(settings.MEDIA_ROOT, path)
cv = Converter(tmp_file)
cv.convert(docs_file_name_path, start=0, end=None)
cv.close()
# response = FileResponse(open(docs_file_name_path, 'rb'),filename=data._name+'.docx')
# os.remove(docs_file_name_path)
os.remove(tmp_file)
return Response(
{
'success': True,
'file':
'https://api.pdfmake.com/media/' + docs_file_name_path
},
status=200)
except:
return Response(status=400)
def extract(self, filename):
# convert pdf to docx
cv = Converter(filename)
wordfile = tempfile.NamedTemporaryFile()
cv.convert(wordfile.name, start=0, end=None)
cv.close()
output = docx2txt.process(wordfile.name)
return output
def pdf_to_word(fileName):
pdf_file = fileName
# 正则获取不含文件类型后缀的部分，用于组成word文档绝对路径
name = re.findall(r'(.*?)\.', pdf_file)[0]
docx_file = f'{name}.docx'
cv = Converter(pdf_file)
cv.convert(docx_file, start=0, end=None)
cv.close()
def test_extracting_table(self):
'''test extracting contents from table.'''
filename = 'demo-table'
pdf_file = os.path.join(self.sample_dir, f'{filename}.pdf')
tables = Converter(pdf_file).extract_tables(end=1)
print(tables)
# compare the last table
sample = [['Input', None, None, None, None, None],
['Description A', 'mm', '30.34', '35.30', '19.30', '80.21'],
['Description B', '1.00', '5.95', '6.16', '16.48', '48.81'],
['Description C', '1.00', '0.98', '0.94', '1.03', '0.32'],
['Description D', 'kg', '0.84', '0.53', '0.52', '0.33'],
['Description E', '1.00', '0.15', None, None, None],
['Description F', '1.00', '0.86', '0.37', '0.78', '0.01']]
assert tables[-1] == sample
def pdf_to_word(entrada_pdf, q, adr, extension_out):
try:
salida_doc = entrada_pdf.split(".")[0] + '.docx'
cv = Converter(entrada_pdf)
cv.convert(salida_doc, start=0, end=None)
cv.close()
os.system("clear")
print(adr)
q.put(salida_doc)
except:
q.put("Error")
def convert_file(file_path):
save_location = r"C:\Users\mdsak\Desktop\pdf-to-docx-converter\Document.docx"
cv = Converter(file_path)
cv.convert(save_location, start=0, end=None)
cv.close()
success_text = tk.Label(frame,
text="Pdf has been successfully converted",
font=(16),
bg="#263D42",
fg="#fff")
success_text.pack()
def _callback_convert(self):
'''Starts the convert of the file or files.'''
# input check
if not self.pdf_paths and not self.docx_folder:
messagebox.showwarning(
title='Neither files or folder selected',
message='Select PDF file or files for convert '
'and Select a folder for the converted files!')
return
if not self.pdf_paths:
messagebox.showwarning(
title='Not files for convert selected',
message='Select PDF file or PDF files for convert!')
return
if not self.docx_folder:
messagebox.showwarning(
title='Not files folder selected',
message='Select a folder for the converted files!')
return
# collect docx files to convert to
docx_paths = []
for pdf_path in self.pdf_paths:
base_name = os.path.basename(pdf_path)
name, ext = os.path.splitext(base_name)
docx_path = os.path.join(self.docx_folder, f'{name}.docx')
docx_paths.append(docx_path)
if any([os.path.exists(path) for path in docx_paths]) and \
not messagebox.askokcancel(title='Existed target file',
message='Docx files with same target name are found under selected folder. '
'Do you want to continue and replace them?'):
return
# now, do the converting work
num_succ, num_fail = 0, 0
for pdf_path, docx_path in zip(self.pdf_paths, docx_paths):
cv = Converter(pdf_path)
try:
cv.convert(docx_path)
except Exception as e:
print(e)
num_fail += 1
else:
num_succ += 1
finally:
cv.close()
messagebox.showinfo(
title='Convert Done!',
message=f'Successful ({num_succ}), Failed ({num_fail}).')
def post(self, request, format=None):
data = request.data['file']
print(data.__dict__)
if data.content_type != 'application/pdf':
return Response(status=400)
unique_filename = str(uuid.uuid4())
docs_file_name_path = 'docx/' + unique_filename + '.docx'
file_path = 'pdf/' + unique_filename + '.pdf'
path = default_storage.save(file_path, ContentFile(data.read()))
tmp_file = os.path.join(settings.MEDIA_ROOT, path)
cv = Converter(tmp_file)
cv.convert(docs_file_name_path, start=0, end=None)
cv.close()
response = FileResponse(open(docs_file_name_path, 'rb'),
filename=data._name + '.docx')
os.remove(docs_file_name_path)
os.remove(tmp_file)
return response
def main():
global trust_ratio
# 读取参数信息，调整全局单句阈值
if len(sys.argv) < 2:
print("参数错误")
exit(-1)
if len(sys.argv) == 3:
trust_ratio = float(sys.argv[2])
# 读取需要进行比对的原文件
filepath = sys.argv[1]
filename, _type = os.path.splitext(filepath)
# 对 pdf 文件先行进行格式转换
if _type == '.pdf':
convert_path = filename + '.docx'
cv = Converter(filepath)
cv.convert(convert_path, multi_processing=True)
cv.close()
filepath = convert_path
doc1 = readDocx(filepath, True)
# 读取本地数据库内的论文，并逐文件进行比对
print("加载数据库...")
path = '.\\database\\word'
filename_list = os.listdir(path)
print(f'数据库共 {len(filename_list)} 篇论文')
progress = ProgressBar(len(filename_list), fmt=ProgressBar.FULL)
t1 = datetime.datetime.now()
print('开始比对...')
for filename in filename_list:
if filename.endswith('.docx'):
doc2 = readDocx(os.path.join(path, filename))
for i in range(len(doc1)):
for j in range(len(doc2)):
compareParagraph(doc1, doc2, i, j, filename)
progress.current += 1
progress()
t2 = datetime.datetime.now()
progress.done()
print('\n比对完成，总用时: ', t2 - t1)
def local_test(filename, make_test_case=False):
pdf_file = os.path.join(output, f'{filename}.pdf')
docx_file = os.path.join(output, f'{filename}.docx')
cv = Converter(pdf_file, docx_file)
# process page by page
for page in cv[0:1]:
# print(page.rotation, page.rotationMatrix)
# print(page.transformationMatrix)
# print(page.rect, page.MediaBox, page.CropBox)
# print(page.xref)
# print(page.getContents())
# print(cv.doc_pdf.xrefObject(page.xref))
# page.cleanContents()
# c = page.readContents().decode(encoding="ISO-8859-1")
# with open('c.txt', 'w') as f:
# f.write(c)
# print(cv.doc_pdf.xrefObject(94))
# with open('x.svg', 'w') as f:
# f.write(page.getSVGimage(text_as_path=False))
# parse layout
cv.make_page(page)
# # extract tables
# tables = cv.extract_tables(page)
# for table in tables:
# print(table)
cv.close() # close pdf
# check results
check_result(pdf_file, docx_file, 'comparison.pdf', make_test_case)
def local_test(sub_path, filename, compare=False, make_test_case=False):
pdf_file = os.path.join(output, sub_path, f'{filename}.pdf')
docx_file = os.path.join(output, sub_path, f'{filename}.docx')
page_index = 0
cv = Converter(pdf_file)
page = cv.fitz_doc[page_index]
# print(page.rotation, page.rotationMatrix)
# print(page.transformationMatrix)
# print(page.rect, page.MediaBox, page.CropBox)
# print(page.xref)
# print(page.getContents())
# print(cv.doc_pdf.xrefObject(page.xref))
# page.cleanContents()
# c = page.readContents().decode(encoding="ISO-8859-1")
# with open('c.txt', 'w') as f:
# f.write(c)
# print(cv.doc_pdf.xrefObject(6))
# print(cv.doc_pdf._getXrefString(7))
# with open('x.svg', 'w') as f:
# f.write(page.getSVGimage(text_as_path=False))
# parse layout
cv.debug_page(page_index, docx_file)
# # extract tables
# tables = cv.extract_tables([page_index])
# for table in tables:
# print(table)
cv.close() # close pdf
# check results
if compare:
check_result(pdf_file, docx_file, 'comparison.pdf', make_test_case)
def Start_PDF_Word(self):
if os.path.isdir(self.filename):
QMessageBox.question(win, '温馨提示！', '程序开始执行时，因为计算量大可能会导致卡顿，这是正常现象，请不要乱点，耐心稍等一会儿！！！',
QMessageBox.Yes | QMessageBox.No, (QMessageBox.Yes))
config_parser = ConfigParser()
config_parser.read('config.cfg', encoding='utf-8')
config = config_parser['default']
for file in os.listdir(self.filename):
extension_name = os.path.splitext(file)[1]
if extension_name != '.pdf':
continue
file_name = os.path.splitext(file)[0]
pdf_file = self.filename + '/' + file
word_file = self.filename + '/' + file_name + '.docx'
cv = Converter(pdf_file)
cv.convert(word_file)
cv.close()
self.textedit_one.moveCursor(QTextCursor.End)
self.textedit_one.insertPlainText(
f'\nPDF文件已成功转换图片文件，请前往桌面查看！！！\n\n生成路径为:{word_file}\n')
else:
QMessageBox.question(win, '温馨提示！', '请导入一个内含PDF文件的文件夹！！！',
QMessageBox.Yes | QMessageBox.No, (QMessageBox.Yes))
from pdf2docx import Converter pdf_file = 'imzalancaklar.pdf' docx_file = 'cv.docx' # convert pdf to docx cv = Converter(pdf_file) cv.convert(docx_file, start=0, end=None) cv.close()
import sys
# Take PDF's path as input
pdf = input("Enter the path to your file: ")
assert os.path.exists(pdf), "File not found at, " + str(pdf)
f = open(pdf, 'r+')
#Ask for custom name for the word doc
doc_name_choice = input(
"Do you want to give a custom name to your file ?(Y/N)")
if (doc_name_choice == 'Y' or doc_name_choice == 'y'):
# User input
doc_name = input("Enter the custom name : ") + ".docx"
else:
# Use the same name as pdf
# Get the file name from the path provided by the user
pdf_name = os.path.basename(pdf)
# Get the name without the extension .pdf
doc_name = os.path.splitext(pdf_name)[0] + ".docx"
# Convert PDF to Word
cv = Converter(pdf)
#Path to the directory
path = os.path.dirname(pdf)
cv.convert(os.path.join(path, "", doc_name), start=0, end=None)
print("Word doc created!")
cv.close()
print('Comparing with sample pdf...')
if compare_layput(pdf_file, docx_pdf_file, output_file, threshold=0.7):
print('Fully matched.')
else:
print(f'Please convert {docx_file} to {docx_pdf_file} in advance.')
if __name__ == '__main__':
script_path = os.path.abspath(__file__) # current script path
output = os.path.dirname(script_path)
filename = 'test'
pdf_file = os.path.join(output, f'{filename}.pdf')
docx_file = os.path.join(output, f'{filename}.docx')
cv = Converter(pdf_file, docx_file)
# process page by page
for page in cv[0:1]:
# print(page.rotation, page.rotationMatrix)
# print(page.transformationMatrix)
# print(page.rect, page.MediaBox, page.CropBox)
# print(page.xref)
# print(page.getContents())
# print(cv.doc_pdf.xrefObject(page.xref))
# page.cleanContents()
# c = page.readContents().decode(encoding="ISO-8859-1")
# with open('c.txt', 'w') as f:
# f.write(c)
def upload():
file = request.files['inputFile']
file.save(os.path.join(app.config['UPLOAD_FOLDER'], (file.filename)))
#Checks if the file format is acceptabe
VALID_FORMATS = {"pdf"} #, "PNG", "docx", "jpg"}
valid = 0
for valid_format in VALID_FORMATS:
if (file.filename)[-len(valid_format):] == valid_format:
newFile = FileContents(name=file.filename[:-len(valid_format)-1], data=file.read())
valid = 1
original_format = valid_format
break
if valid == 0:
return "Error: Wrong Format. Please upload a PDF file."
#TODO: Implement convert logic
newFile.data_pdf = newFile.data
#newFile.data_png = newFile.data
#newFile.data_docx = newFile.data
#Need to upload newFile.data to the folder TMP
#------------convert starts here-----------
if original_format == "pdf":
#-----PDF TO JPG-----
outputDir = "tmp/"
input_path = "tmp/" + newFile.name + ".pdf"
pages = convert_from_path(input_path, 500)
img = pages[0]
# Create a buffer to hold the bytes
buf = BytesIO()
# Save the image as jpeg to the buffer
img.save(buf, 'jpeg')
# Rewind the buffer's file pointer
buf.seek(0)
# Read the bytes from the buffer
image_bytes = buf.read()
# Close the buffer
buf.close()
#Upload file to DB
newFile.data_png = image_bytes
#-----PDF TO DOCX-----
pdf_file = 'tmp/' + newFile.name + ".pdf"
docx_file = 'tmp/' + newFile.name + ".docx"
# convert pdf to docx
cv = Converter(pdf_file)
cv.convert(docx_file, start=0, end=None)
cv.close()
newFile.data_docx = open("tmp/" + newFile.name + ".docx", 'rb').read()
newFile.data_pdf = open("tmp/" + newFile.name + ".pdf", 'rb').read()
#files = glob.glob('tmp')
#newFile.data_pdf = files[0]
#for f in files:
#os.remove(f)
db.session.add(newFile)
db.session.commit()
return redirect('/download/' + str(newFile.id))
from docx2pdf import convert
from pdf2docx import Converter
import sys
import os
from PIL import Image
from fpdf import FPDF
if sys.argv[1] == 'd-p':
convert(sys.argv[2])
print("convert success")
if sys.argv[1] == 'p-d':
cv = Converter(sys.argv[2])
open(sys.argv[3], "a")
cv.convert(sys.argv[3], start=0, end=None)
cv.close()
print("convert success")
if sys.argv[1] == 'i-i':
filename = sys.argv[2]
img = Image.open(filename)
img.save(sys.argv[3])
print("success")
if sys.argv[1] == 'i-b':
filename = sys.argv[2]
img = Image.open(filename).save(sys.argv[3])
print("success")
if sys.argv[1] == 't-p':
filename = sys.argv[2]
def pdf_to_word(pdf_file_path, word_file_path):
cv = Converter(pdf_file_path)
cv.convert(word_file_path)
cv.close()