archive
/
md-pdf
Archived
1
Fork 0
This repository has been archived on 2023-03-26. You can view files and clone it, but cannot push or open issues or pull requests.
md-pdf/md_pdf/build/pandoc.py

42 lines
1.0 KiB
Python

import pypandoc
from bs4 import BeautifulSoup
import os.path
from md_pdf.consts import PROJECT_DIR
CSL_FILE = os.path.join(PROJECT_DIR, 'assets', 'harverd.csl')
def fix_references_title(content):
soup = BeautifulSoup(content, 'html.parser')
reference_element = soup.find('div', class_='references')
if reference_element is not None:
title = soup.new_tag('h1')
title.string = "References"
reference_element.insert_before(title)
return soup.prettify()
def output_html(html, out_dir):
with open(os.path.join(out_dir, 'output.html'), 'w') as f:
f.write(html)
def build_document(files_content, bibliography):
args = [
'-s',
]
filters = []
if bibliography is not None:
args += [
'--bibliography={}'.format(bibliography),
'--csl={}'.format(CSL_FILE)
]
filters.append('pandoc-citeproc')
html = pypandoc.convert_text(files_content, 'html', format='md',
extra_args=args,
filters=filters
)
return fix_references_title(html)