现需将pdf 转换至Excel ,
目前实现方式:将PDF的TABLE部分与 非 TABLE部分分别导出至Excel两个sheet中
1)、识别PDF中的表格块
2)、将PDF转换为Word格式
3)、提取Word中非表格的文本数据
4)、对文本与表格重复的行进行去重
5)、合并导出至Excel不同sheet页中
# coding=UTF8
import datetime
from docx import Document
from pdf2docx import Converter
import pandas as pd
import numpy as np
import pdfplumber
import os
import fitz# TODO 输出PDF表格数据至Excel
def extractTables(filepath):with pdfplumber.open(filepath) as pdf:tables = []for i in range(0, len(pdf.pages)):page = pdf.pages[i]tables.append(page.extract_tables())df = pd.DataFrame()df_seperation = pd.DataFrame([np.nan, np.nan]) # 创建空白的,用于充当分隔行for i in range(0, len(tables)):tabular = tables[i] # 选取第i页的表格if len(tabular) > 0: # 如果该页存在表格的话for j in range(0, len(tabular)): # j 表示第几个表格df_temp = pd.DataFrame(tabular[j])df = pd.concat([df, df_seperation, df_temp]) # 更新总表格return df# TODO pdf 转Word
def extractWord(pdffilepath, wordfilepath):cv = Converter(pdffilepath)cv.convert(wordfilepath)cv.close()# TODO 获取非表格内容
def getDocLines(wordfilepath):doc = Document(wordfilepath)paragraphs = doc.paragraphslines = []for paragraph in paragraphs:line = paragraph.text.strip()if not line:continuelines.append(line)# aspose用的体验板,带有页眉# lines = lines[100:]# print(lines)df = pd.DataFrame(lines)return df# TODO 删除与table重复的行数据
def txt(tabledf, txtdf):lines = []for line in tabledf[0]:lines.append(str(line))# 获取txt与Excel重复的数据repeat_txt = []for line in lines:for txt in txtdf[0]:if line.find(txt) != -1:repeat_txt.append(txt)txtdf = txtdf.drop(txtdf[txtdf[0] == txt].index)return txtdfif __name__ == '__main__':filepath = 'D:\develop_python\Python_Demo\PDF_TO_EXCEL\YM2021\\'outpath = 'D:\develop_python\Python_Demo\PDF_TO_EXCEL\YM2021\\'pdffile = '南通-2.21'pdffilepath = filepath + str(pdffile) + '.pdf'excelfilepath = filepath + str(pdffile) + '.xlsx'wordfilepath = filepath + str(pdffile) + '.docx'# 获取PDF表格数据try:starttime = datetime.datetime.now()print('执行开始', starttime)if not os.path.exists(outpath):os.makedirs(outpath)print()print('正在读取表格数据........')tabledf = extractTables(pdffilepath)print('表格数据读取完成........')print()print('正在转换Word......')extractWord(pdffilepath, wordfilepath)print('Word转换完成......')print()print('正在解析非表格文本数据......')txtdf = getDocLines(wordfilepath)# 删除与table重复的行数据df = txt(tabledf, txtdf)print('非表格数据解析完成......')print()print('正在输出Excel文件......')writer = pd.ExcelWriter(excelfilepath)tabledf.to_excel(writer, sheet_name='EXCEL', index=False)# txtdf.to_excel(writer, sheet_name='txt', index=False)df.to_excel(writer, sheet_name='txt', index=False)writer.save()print('Excel文件输出成功......')print()endtime = datetime.datetime.now()print('执行结束', endtime)print('耗时', endtime - starttime)except Exception as e:print(Exception, e.args)