97 lines
3.5 KiB
Python
97 lines
3.5 KiB
Python
import re
|
||
from PyPDF2 import PdfReader
|
||
def extract_common_header(pdf_path):
|
||
|
||
def get_headers(pdf_document, start_page, pages_to_read):
|
||
headers = []
|
||
for i in range(start_page, min(start_page + pages_to_read, len(pdf_document.pages))):
|
||
page = pdf_document.pages[i]
|
||
text = page.extract_text() or ""
|
||
if text:
|
||
# 只取每页的前三行，去除前后的空白字符
|
||
first_lines = [line.strip() for line in text.strip().split('\n')[:3]]
|
||
headers.append(first_lines)
|
||
return headers
|
||
|
||
def find_common_headers(headers):
|
||
if not headers:
|
||
return []
|
||
|
||
# 使用 zip 对齐所有页的对应行
|
||
common_headers = []
|
||
for lines in zip(*headers):
|
||
# 检查所有行是否完全相同
|
||
if all(line == lines[0] for line in lines[1:]):
|
||
common_headers.append(lines[0])
|
||
return common_headers
|
||
|
||
pdf_document = PdfReader(pdf_path)
|
||
total_pages = len(pdf_document.pages)
|
||
|
||
# 定义两个提取策略
|
||
strategies = []
|
||
if total_pages >= 3:
|
||
# 策略1：中间的3页
|
||
middle_page = total_pages // 2
|
||
start_page = max(0, middle_page - 1)
|
||
strategies.append((start_page, 3))
|
||
elif total_pages == 2:
|
||
# 策略1：2页
|
||
strategies.append((0, 2))
|
||
else:
|
||
# 策略1：1页
|
||
strategies.append((0, 1))
|
||
|
||
# 策略2：前三页
|
||
if total_pages >= 3:
|
||
strategies.append((0, 3))
|
||
elif total_pages == 2:
|
||
strategies.append((0, 2))
|
||
elif total_pages == 1:
|
||
strategies.append((0, 1))
|
||
|
||
common_headers = []
|
||
|
||
for idx, (start, count) in enumerate(strategies):
|
||
headers = get_headers(pdf_document, start, count)
|
||
if len(headers) < 2:
|
||
continue # 需要至少2页来比较
|
||
|
||
current_common = find_common_headers(headers)
|
||
if current_common:
|
||
common_headers = current_common
|
||
break # 找到共同部分后退出
|
||
# 如果没有找到，继续下一个策略
|
||
|
||
return '\n'.join(common_headers)
|
||
|
||
def clean_page_content(text, common_header):
|
||
# 首先删除抬头公共部分
|
||
if common_header: # 确保有公共抬头才进行替换
|
||
for header_line in common_header.split('\n'):
|
||
if header_line.strip(): # 只处理非空行
|
||
# 替换首次出现的完整行
|
||
text = re.sub(r'^' + re.escape(header_line.strip()) + r'\n?', '', text, count=1)
|
||
|
||
# 删除页码 eg:89/129 这个代码分三步走可以把89/129完全删除
|
||
text = re.sub(r'^\s*\d+\s*(?=\D)', '', text) # 删除开头的页码，仅当紧跟非数字字符时
|
||
text = re.sub(r'\s+\d+\s*$', '', text) # 删除结尾的页码
|
||
text = re.sub(r'\s*\/\s*\d+\s*', '', text) # 删除形如 /129 的页码
|
||
text = re.sub(r'\s*[—-]\s*\d+\s*[—-]\s*', '', text) # 删除形如 '—2—' 或 '-2-' 的页码
|
||
return text
|
||
|
||
def is_scanned_pdf(file_path):
|
||
with open(file_path, 'rb') as file:
|
||
reader = PdfReader(file)
|
||
for page in reader.pages:
|
||
if page.extract_text().strip(): # 如果有文本
|
||
return False # 不是扫描型
|
||
return True # 全部页面都没有文本
|
||
|
||
if __name__ == '__main__':
|
||
file_path = r"C:\Users\Administrator\Documents\WeChat Files\wxid_d11awe5rp1y722\FileStorage\File\2024-12\2020-安徽-安徽省生态环境厅电梯采购.pdf"
|
||
res=is_scanned_pdf(file_path)
|
||
if res:
|
||
print("扫描型")
|
||
else:
|
||
print("普通型") |