PyPI 官网下载 | ibooks_highlight_exporter-0.1.11.tar.gz

import os import re from collections import OrderedDict from lxml import etree from ibooks_highlight_exporter.helpers import find_first_file_with_extension from ibooks_highlight_exporter.base import HeadingPointer def get_epub_content_path(epub_directory_path: str) -> str: return find_first_file_with_extension(epub_directory_path, "opf") def get_epub_toc_path(epub_directory_path: str) -> str: return find_first_file_with_extension(epub_directory_path, "ncx") def get_parts(epub_directory_path: str) -> OrderedDict: opf = etree.parse(open(get_epub_content_path(epub_directory_path), "r")) parts = OrderedDict() for item in opf.xpath( "/*[local-name() = 'package']/*[local-name() = 'manifest']/*[local-name() = 'item']" ): if item.get("media-type") == "application/xhtml+xml": parts[item.get("id")] = item.get("href") return parts def parse_epub_cfi(epub_cfi: str) -> (str, int): regex = re.compile(r"epubcfi\(/6/\d+\[(.+)\]!/4(.*?)/(\d+)") m = regex.search(epub_cfi) if m is None: raise ValueError("Invalid epubcfi!") else: part_id = m.group(1) try: position = int(m.group(3)) except ValueError: raise ValueError("Invalid epubcfi!") return part_id, position def get_heading_pointers(epub_path): toc_path = get_epub_toc_path(epub_path) toc = etree.parse(open(toc_path, "r")) first_level_nav_points = toc.xpath( "/*[local-name() = 'ncx']/*[local-name() = 'navMap']/*[local-name() = 'navPoint']" ) chapter_links = [] for nav_point in first_level_nav_points: chapter_links += traverse_nav_point(nav_point) return find_chapter_positions(chapter_links, epub_path) def traverse_nav_point(nav_point, level=0): src = nav_point.xpath("./*[local-name() = 'content']")[0].get("src") if "#" in src: html_path, a_id = tuple(src.split("#")) else: html_path = src a_id = None chapter_name = nav_point.xpath( "./*[local-name() = 'navLabel']/*[local-name() = 'text']" )[0].text sub_nav_points = nav_point.xpath("./*[local-name() = 'navPoint']") chapter_links = [(html_path, chapter_name, a_id, level)] for sub_nav_point in sub_nav_points: chapter_links += traverse_nav_point(sub_nav_point, level + 1) return chapter_links def find_chapter_positions(chapter_links, epub_path): heading_pointers = [] for chapter_link in chapter_links: tree = etree.parse(open(os.path.join(epub_path, chapter_link[0]), "r")) body_xpath = "/*[local-name() = 'html'] /*[local-name() = 'body']" body = tree.xpath(body_xpath)[0] if chapter_link[2] is not None: el = body.xpath(f"//*[@id='{chapter_link[2]}']")[0] ancestor_before_body = el for ancestor in el.iterancestors(): if ancestor.tag.endswith("body"): break ancestor_before_body = ancestor position = body.index(ancestor_before_body) * 2 else: position = 0 heading_pointers.append( HeadingPointer( chapter_link[0], int(position), chapter_link[1], chapter_link[3] ) ) return heading_pointers