X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/ac6f4dfb3c4e7ac8b22ab56390e61634a8152df6..88cabfe:/librarian/epub.py diff --git a/librarian/epub.py b/librarian/epub.py index bfd7570..4855de8 100644 --- a/librarian/epub.py +++ b/librarian/epub.py @@ -17,11 +17,12 @@ from tempfile import mkdtemp, NamedTemporaryFile from shutil import rmtree from librarian import RDFNS, WLNS, NCXNS, OPFNS, XHTMLNS, OutputFile -from librarian.cover import WLCover +from librarian.cover import DefaultEbookCover from librarian import functions, get_resource functions.reg_person_name() +functions.reg_lang_code_3to2() def inner_xml(node): @@ -290,18 +291,47 @@ def chop(main_text): main_xml_part = part_xml[0] # master last_node_part = False + + # the below loops are workaround for a problem with epubs in drama ebooks without acts + is_scene = False + for one_part in main_text: + name = one_part.tag + if name in ('naglowek_scena'): + is_scene = True + break + if is_scene is True: + is_scene_with_acts = False + for one_part in main_text: + if one_part.tag == 'naglowek_akt': + is_scene_with_acts = True + break + else: + is_scene_with_acts = False + 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_scene_with_acts 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 +340,9 @@ 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) @@ -336,7 +368,7 @@ def transform(wldoc, verbose=False, sample=n: generate sample e-book (with at least n paragraphs) cover: a cover.Cover factory or True for default - flags: less-advertising, without-fonts, working-copy + flags: less-advertising, without-fonts, working-copy, with-full-fonts """ def transform_file(wldoc, chunk_counter=1, first=True, sample=None): @@ -448,7 +480,7 @@ def transform(wldoc, verbose=False, if cover: if cover is True: - cover = WLCover + cover = DefaultEbookCover cover_file = StringIO() bound_cover = cover(document.book_info) @@ -540,21 +572,23 @@ def transform(wldoc, verbose=False, os.chdir(os.path.join(os.path.dirname(os.path.realpath(__file__)), 'font-optimizer')) for fname in 'DejaVuSerif.ttf', 'DejaVuSerif-Bold.ttf', 'DejaVuSerif-Italic.ttf', 'DejaVuSerif-BoldItalic.ttf': - optimizer_call = ['perl', 'subset.pl', '--chars', ''.join(chars).encode('utf-8'), - get_resource('fonts/' + fname), os.path.join(tmpdir, fname)] - if verbose: - print "Running font-optimizer" - subprocess.check_call(optimizer_call) + if not flags or not 'with-full-fonts' in flags: + optimizer_call = ['perl', 'subset.pl', '--chars', ''.join(chars).encode('utf-8'), + get_resource('fonts/' + fname), os.path.join(tmpdir, fname)] + if verbose: + print "Running font-optimizer" + subprocess.check_call(optimizer_call) + else: + subprocess.check_call(optimizer_call, stdout=subprocess.PIPE, stderr=subprocess.PIPE) + zip.write(os.path.join(tmpdir, fname), os.path.join('OPS', fname)) else: - subprocess.check_call(optimizer_call, stdout=subprocess.PIPE, stderr=subprocess.PIPE) - zip.write(os.path.join(tmpdir, fname), os.path.join('OPS', fname)) + zip.write(get_resource('fonts/' + fname), os.path.join('OPS', fname)) manifest.append(etree.fromstring( - '' % (fname, fname))) + '' % (fname, fname))) 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: @@ -562,7 +596,7 @@ def transform(wldoc, verbose=False, meta.set('name', st) meta.set('content', '0') toc_file[0].append(meta) - toc_file[0][0].set('content', ''.join((title, 'WolneLektury.pl'))) + toc_file[0][0].set('content', str(document.book_info.url)) toc_file[0][1].set('content', str(toc.depth())) set_inner_xml(toc_file[1], ''.join(('', title, ''))) @@ -571,7 +605,7 @@ 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)