import os
import re
import fitz # pip install pymupdf
import csv
filenames = os.listdir()
prefix = '公开发行证券的公司信息披露内容与格式准则第2号'
pdf = [f for f in filenames if f.startswith(prefix) and f.endswith('.pdf')]# 筛选出格式准则文件
year = [f[-12:-5] for f in pdf]# 提取修订年份
def getText(pdf):
text = ''
doc = fitz.open(pdf)
for page in doc:
text += page.getText()
doc.close()
return(text)# 提取PDF文本，因有一份无法被pdfplumber库提取，故采用pymupdf
def getSubtext(pdf):
text = getText(pdf)
p1 = re.compile('第二章\s*年度报告正文(.*)第三章\s*年度报告摘要',
re.DOTALL)
subtext = p1.search(text).group(0)
return(subtext)
def getTOC(pdf):
subtext = getSubtext(pdf)
p = re.compile('(?<=\\n)(第\w{1,2}节)\s+(.*)(?=\\n)')
listOftuple = p.findall(subtext)
return(listOftuple)
toc = [getTOC(f) for f in pdf]
def getTOC_content(pdf):
subtext = getSubtext(pdf)
p = re.compile('(?<=\\n)(第\w{1,2}节)\s+(.*)(?=\\n)')
index = []
for match in p.finditer(subtext):
s = match.start()
e = match.end()
index.append((s,e))
content = []
for i in range(len(index)-1):
s = index[i][1]
e = index[i+1][0]
content.append(subtext[s:e])
last_start = index[-1][1]
last_end = subtext.rfind('\n第三章')
content.append(subtext[last_start:last_end])
return((index,content))
list_of_content = [getTOC_content(f) for f in pdf]
def export_to_html(pdf,toc,content):
f = open('homework2_output_template.html', encoding='utf-8')
html = f.read()
f.close()
template = template = '%s
%s
'
toc_list = [t[0]+' '+ t[1] for t in toc]
div = [template % (t,c) for (t,c) in zip(toc_list, content)]
div = ''.join(div)
html = html % div
f = open('homework2_%s.html' % pdf, 'w',encoding='utf-8')
f.write(html)
f.close()
return()
content_list = [c[1] for c in list_of_content]
export_to_html(pdf[0], toc[0], content_list[0])
提取不同修订版目录准则代码,获取文本,获取目录函数 getTOC(),获取节内容函数 getTOC_content(),导出为HTML格式