gitbook-html2print/build.py

#!/usr/bin/python
# -*- coding: utf-8 -*-

import sys, os
import shutil
import markdown
# import mistune
from bs4 import BeautifulSoup
import pypandoc
import json
import re

# import json
# import getopt
# import urllib

BOOKS_SRC = 'book-src'
BUILD_d = "build"

def main():
   if not os.path.isdir(BUILD_d):
      os.mkdir(BUILD_d)

   for book in os.listdir(BOOKS_SRC):
      if os.path.isdir(os.path.join(BOOKS_SRC, book)):
         # print(book)
         parse_book(book)


def parse_book(book):
   book_name = book.replace('.git', '')
   print("- - -")
   print(book_name)
   print("- - -")

   # build destination
   book_build_d = os.path.join(BUILD_d,book_name)
   if os.path.isdir(book_build_d):
      shutil.rmtree(book_build_d, ignore_errors=True)
   os.mkdir(book_build_d)

   # table of content (ordered list of markdown files)
   sum_p = os.path.join(BOOKS_SRC, book, "SUMMARY.md")
   if not os.path.isfile(sum_p):
      print("No summary file, can't generate html")
      return

   sum_f = open(sum_p)
   sum_str = sum_f.read()
   # print(sum_str)
   sum_html = markdown.markdown(sum_str)
   # print(sum_html)
   sum_dom = BeautifulSoup(sum_html, 'html.parser')
   # print(sum_dom)
   toc = parse_summary(sum_dom.ul, {})
   print(toc)

   generate_html(book, toc, book_build_d)

def parse_summary(ul, toc):
   i=0
   for li in ul.find_all('li',recursive=False):
      # print('li')
      for a in li.find_all('a',recursive=False):
         # print(a.get_text(strip=True))
         # print(a['href'])
         href = a['href']
         href = re.sub(r'^/', '', href)
         toc[i] = {
            'label':a.get_text(strip=True),
            'file':href
         }
      i = i+1

   return toc

def generate_html(book, toc, book_build_d):
   railway = []
   for p in toc:
      print(toc[p]['file'])
      # generate html with pandoc

      # files
      md_f = toc[p]['file']
      html_f = md_f.replace('.md', '.html')
      html_f = html_f.replace('README', 'index')
      html_f = html_f.replace('/', '-')
      in_f = os.path.join(BOOKS_SRC, book, md_f)
      print(in_f)
      out_f = os.path.join(book_build_d,html_f)
      print(out_f)

      # pandoc options
      filters = []
      pdoc_args = ['--mathjax',
                   '--smart']

      # pandoc command line
      output = pypandoc.convert_file(in_f,
                               to='html5',
                               format='md',
                               extra_args=pdoc_args,
                               filters=filters,
                               outputfile=out_f)

      # save reference in railway
      railway.append({'label':toc[p]['label'], 'file':html_f})

   # save railway as json file
   with open(book_build_d+'/railway.json', 'w') as fp:
      json.dump(railway, fp, ensure_ascii=False, indent="\t")


if __name__ == "__main__":
   main()