X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/dd3e5f7c77134d034eb8592479288e1bb9fe28fd..5ea41f9b810aa661189d3ab8a3cd15b61a3d900f:/librarian/epub.py?ds=sidebyside diff --git a/librarian/epub.py b/librarian/epub.py index c8a8668..b17ca0e 100644 --- a/librarian/epub.py +++ b/librarian/epub.py @@ -22,6 +22,7 @@ from librarian.cover import DefaultEbookCover from librarian import functions, get_resource functions.reg_person_name() +functions.reg_lang_code_3to2() def inner_xml(node): @@ -77,6 +78,7 @@ def replace_characters(node): def replace_chars(text): if text is None: return None + #text = re.sub(r"(?<=\s\w)\s+", u"\u00a0", text) #fix for hanging single letter conjunctions – for future use. return text.replace(u"\ufeff", u"")\ .replace("---", u"\u2014")\ .replace("--", u"\u2013")\ @@ -247,7 +249,7 @@ class TOC(object): nav_label = nav_map.makeelement(NCXNS('navLabel')) text = nav_map.makeelement(NCXNS('text')) - text.text = child.name + text.text = re.sub(r'\n', ' ', child.name) nav_label.append(text) nav_point.append(nav_label) @@ -290,18 +292,41 @@ def chop(main_text): main_xml_part = part_xml[0] # master last_node_part = False + + # the below loop are workaround for a problem with epubs in drama ebooks without acts + is_scene = False + is_act = False + for one_part in main_text: + name = one_part.tag + if name == 'naglowek_scena': + is_scene = True + elif name == 'naglowek_akt': + is_act = True + for one_part in main_text: name = one_part.tag - if name == 'naglowek_czesc': - yield part_xml - last_node_part = True - main_xml_part[:] = [deepcopy(one_part)] - elif not last_node_part and name in ("naglowek_rozdzial", "naglowek_akt", "srodtytul"): - yield part_xml - main_xml_part[:] = [deepcopy(one_part)] + if is_act is False and is_scene is True: + if name == 'naglowek_czesc': + yield part_xml + last_node_part = True + main_xml_part[:] = [deepcopy(one_part)] + elif not last_node_part and name in ("naglowek_scena"): + yield part_xml + main_xml_part[:] = [deepcopy(one_part)] + else: + main_xml_part.append(deepcopy(one_part)) + last_node_part = False else: - main_xml_part.append(deepcopy(one_part)) - last_node_part = False + if name == 'naglowek_czesc': + yield part_xml + last_node_part = True + main_xml_part[:] = [deepcopy(one_part)] + elif not last_node_part and name in ("naglowek_rozdzial", "naglowek_akt", "srodtytul"): + yield part_xml + main_xml_part[:] = [deepcopy(one_part)] + else: + main_xml_part.append(deepcopy(one_part)) + last_node_part = False yield part_xml @@ -310,7 +335,9 @@ def transform_chunk(chunk_xml, chunk_no, annotations, empty=False, _empty_html_s toc = TOC() for element in chunk_xml[0]: - if element.tag in ("naglowek_czesc", "naglowek_rozdzial", "naglowek_akt", "srodtytul"): + if element.tag in ("naglowek_czesc"): + toc.add(node_name(element), "part%d.html#book-text" % chunk_no) + elif element.tag in ("naglowek_rozdzial", "naglowek_akt", "srodtytul"): toc.add(node_name(element), "part%d.html" % chunk_no) elif element.tag in ('naglowek_podrozdzial', 'naglowek_scena'): subnumber = toc.add(node_name(element), "part%d.html" % chunk_no, level=1, is_part=False) @@ -472,7 +499,7 @@ def transform(wldoc, verbose=False, '')) manifest.append(etree.fromstring( '' % (cover_name, bound_cover.mime_type()))) - spine.insert(0, etree.fromstring('')) + spine.insert(0, etree.fromstring('')) opf.getroot()[0].append(etree.fromstring('')) guide.append(etree.fromstring('')) @@ -556,8 +583,7 @@ def transform(wldoc, verbose=False, rmtree(tmpdir) if cwd is not None: os.chdir(cwd) - - zip.writestr('OPS/content.opf', etree.tostring(opf, pretty_print=True)) + zip.writestr('OPS/content.opf', etree.tostring(opf, pretty_print=True, xml_declaration = True, encoding='UTF-8')) title = document.book_info.title attributes = "dtb:uid", "dtb:depth", "dtb:totalPageCount", "dtb:maxPageNumber" for st in attributes: @@ -574,7 +600,7 @@ def transform(wldoc, verbose=False, toc.add(u"Spis treści", "toc.html", index=1) zip.writestr('OPS/toc.html', toc.html().encode('utf-8')) toc.write_to_xml(nav_map) - zip.writestr('OPS/toc.ncx', etree.tostring(toc_file, pretty_print=True)) + zip.writestr('OPS/toc.ncx', etree.tostring(toc_file, pretty_print=True, xml_declaration = True, encoding='UTF-8')) zip.close() return OutputFile.from_filename(output_file.name)