spider/js_reverse
1
0
Fork 0
mirror of https://github.com/luzhisheng/js_reverse.git synced 2025-04-20 03:59:57 +08:00
Code Issues Packages Projects Releases Wiki Activity

pdf 处理

Browse Source
This commit is contained in:
aiyingfeng 2023-07-24 18:47:16 +08:00
parent c0cdf2f8b5
commit f33a2dac51
1 changed files with 4 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
pdf处理/program/discern.py
View File

@ -2,6 +2,7 @@ import image_compare
import pandas as pd import pandas as pd
import pdfplumber import pdfplumber
import PyPDF2 import PyPDF2
from datetime import datetime
from image_text_ocr import ImageTextOcr from image_text_ocr import ImageTextOcr
import os import os
import cv2 import cv2
@ -18,7 +19,9 @@ class Discern(object):
def export_excel(self, export): def export_excel(self, export):
# 将字典列表转换为DataFrame # 将字典列表转换为DataFrame
pf = pd.DataFrame(list(export)) pf = pd.DataFrame(list(export))
file_path = pd.ExcelWriter('../docs/结果.xlsx') current_time = datetime.now()
formatted_time = current_time.strftime('%Y-%m-%d %H:%M:%S')
file_path = pd.ExcelWriter(f'../docs/{formatted_time}.xlsx')
# 替换空单元格 # 替换空单元格
pf.fillna(' ', inplace=True) pf.fillna(' ', inplace=True)
# 输出 # 输出