1 from datetime import date
4 from ebooklib import epub
7 from librarian import functions, OutputFile, get_resource, XHTMLNS
8 from librarian.cover import make_cover
9 from librarian.embeds.mathml import MathML
11 from librarian.fonts import strip_font
16 self.element = etree.XML('''<html xmlns="http://www.w3.org/1999/xhtml"><head><link rel="stylesheet" href="style.css" type="text/css"/><meta http-equiv="Content-Type" content="text/html; charset=utf-8"/><title>WolneLektury.pl</title></head><body/></html>''')
20 return self.element.find('.//' + XHTMLNS('title'))
24 return self.element.find('.//' + XHTMLNS('body'))
30 def __init__(self, base_url=None, fundraising=None, cover=None):
31 self._base_url = base_url or 'file:///home/rczajka/for/fnp/librarian/temp~/maly/img/'
32 self.fundraising = fundraising
33 self.footnotes = etree.Element('div', id='footnotes')
34 self.make_cover = cover or make_cover
38 # 'header': self.header,
39 'footnotes': self.footnotes,
41 self.current_cursors = []
47 return self.current_cursors[-1]
49 def enter_fragment(self, fragment):
50 self.current_cursors.append(self.cursors[fragment])
52 def exit_fragment(self):
53 self.current_cursors.pop()
55 def create_fragment(self, name, element):
56 assert name not in self.cursors
57 self.cursors[name] = element
59 def forget_fragment(self, name):
60 del self.cursors[name]
64 if self._base_url is not None:
67 return 'https://wolnelektury.pl/media/book/pictures/{}/'.format(self.document.meta.url.slug)
70 # Base URL should be on Document level, not builder.
71 def build(self, document, **kwargs):
72 """Should return an OutputFile with the output."""
73 raise NotImplementedError()
76 class EpubBuilder(Builder):
77 file_extension = 'epub'
78 isbn_field = 'isbn_epub'
81 def __init__(self, *args, **kwargs):
84 super().__init__(*args, **kwargs)
86 def build(self, document, **kwargs):
87 # replace_characters -- nie, robimy to na poziomie elementów
89 # hyphenator (\00ad w odp. miejscach) -- jeśli już, to też powinno to się dziać na poziomie elementów
90 # spójniki (\u00a0 po)-- jeśli już, to na poziomie elementów
91 # trick na dywizy: ­⁠-
94 # początek z KAŻDEGO PLIKU xml
96 # zliczamy zbiór użytych znaków
99 # mieliśmy taką flagę less-advertising, używaną tylko dla Prestigio; już nie używamy.
101 # @editors = document.editors() (jako str)
102 # @funders = join(meta.funders)
103 # @thanks = meta.thanks
106 self.output = output = epub.EpubBook()
107 self.document = document
113 self.add_title_page()
122 'Początek utworu', # i18n
125 self.output.guide.append({
128 "href": "part1.xhtml"
132 self.build_document(self.document)
137 self.add_annotations()
138 self.add_support_page()
142 e = len(self.output.spine) - 3 - 3
143 nfunds = len(self.fundraising)
147 # COUNTING CHARACTERS?
148 for f in range(nfunds):
149 spine_index = int(4 + (f / nfunds * e) + f)
153 etree.XML('<div id="book-text"><div class="fundraising">' + self.fundraising[f % len(self.fundraising)] + '</div></div>')
155 self.add_html(h.element, file_name='fund%d.xhtml' % f, spine=spine_index)
159 output_file = tempfile.NamedTemporaryFile(
160 prefix='librarian', suffix='.epub',
163 epub.write_epub(output_file.name, output, {'epub3_landmark': False})
164 return OutputFile.from_filename(output_file.name)
166 def build_document(self, document):
167 self.toc_precedences = []
172 document.tree.getroot().epub_build(self)
173 if document.meta.parts:
176 self.start_element('div', {'class': 'title-page'})
177 self.start_element('h1', {'class': 'title'})
178 self.push_text(document.meta.title)
192 for child in document.children:
194 self.add_toc_entry(None, child.meta.title, 0)
195 self.build_document(child)
197 self.shift_toc_base()
200 def add_title_page(self):
202 html.title.text = "Strona tytułowa"
203 bt = etree.SubElement(html.body, 'div', **{'id': 'book-text'})
204 tp = etree.SubElement(bt, 'div', **{'class': 'title-page'})
206 # Tak jak jest teraz – czy może być jednocześnie
208 # i „dzieło nadrzędne”
209 # wcześniej mogło być dzieło nadrzędne,
211 e = self.document.tree.find('//autor_utworu')
213 etree.SubElement(tp, 'h2', **{'class': 'author'}).text = e.raw_printable_text(self)
214 e = self.document.tree.find('//nazwa_utworu')
216 etree.SubElement(tp, 'h1', **{'class': 'title'}).text = e.raw_printable_text(self)
219 for author in self.document.meta.authors:
220 etree.SubElement(tp, 'h2', **{'class': 'author'}).text = author.readable()
221 etree.SubElement(tp, 'h1', **{'class': 'title'}).text = self.document.meta.title
223 # <xsl:apply-templates select="//nazwa_utworu | //podtytul | //dzielo_nadrzedne" mode="poczatek"/>
225 # <xsl:apply-templates select="//dc:creator" mode="poczatek"/>
226 # <xsl:apply-templates select="//dc:title | //podtytul | //dzielo_nadrzedne" mode="poczatek"/>
228 etree.SubElement(tp, 'p', **{"class": "info"}).text = '\u00a0'
230 if self.document.meta.translators:
231 p = etree.SubElement(tp, 'p', **{'class': 'info'})
232 p.text = 'tłum. ' + ', '.join(t.readable() for t in self.document.meta.translators)
234 #<p class="info">[Kopia robocza]</p>
236 p = etree.XML("""<p class="info">
237 <a>Ta lektura</a>, podobnie jak tysiące innych, jest dostępna on-line na stronie
238 <a href="http://www.wolnelektury.pl/">wolnelektury.pl</a>.
240 p[0].attrib['href'] = str(self.document.meta.url)
243 if self.document.meta.thanks:
244 etree.SubElement(tp, 'p', **{'class': 'info'}).text = self.document.meta.thanks
246 tp.append(etree.XML("""
248 Utwór opracowany został w ramach projektu<a href="http://www.wolnelektury.pl/"> Wolne Lektury</a> przez<a href="http://www.nowoczesnapolska.org.pl/"> fundację Nowoczesna Polska</a>.
252 if getattr(self.document.meta, self.isbn_field):
253 etree.SubElement(tp, 'p', **{"class": "info"}).text = getattr(self.document.meta, self.isbn_field)
255 tp.append(etree.XML("""<p class="footer info">
256 <a href="http://www.wolnelektury.pl/"><img src="logo_wolnelektury.png" alt="WolneLektury.pl" /></a>
261 file_name='title.xhtml',
263 toc='Strona tytułowa' # TODO: i18n
267 get_resource('res/wl-logo-small.png'),
268 file_name='logo_wolnelektury.png',
269 media_type='image/png'
272 def set_metadata(self):
273 self.output.set_identifier(
274 str(self.document.meta.url))
275 self.output.set_language(
276 functions.lang_code_3to2(self.document.meta.language)
278 self.output.set_title(self.document.meta.title)
280 for i, author in enumerate(self.document.meta.authors):
281 self.output.add_author(
283 file_as=six.text_type(author),
284 uid='creator{}'.format(i)
286 for translator in self.document.meta.translators:
287 self.output.add_author(
288 translator.readable(),
289 file_as=six.text_type(translator),
291 uid='translator{}'.format(i)
293 for publisher in self.document.meta.publisher:
294 self.output.add_metadata("DC", "publisher", publisher)
296 self.output.add_metadata("DC", "date", self.document.meta.created_at)
302 item = epub.EpubNav()
303 item.add_link(href='style.css', rel='stylesheet', type='text/css')
304 self.output.add_item(item)
305 self.output.spine.append(item)
306 self.output.add_item(epub.EpubNcx())
308 self.output.toc.append(
318 def add_support_page(self):
320 get_resource('epub/support.xhtml'),
322 toc='Wesprzyj Wolne Lektury'
326 get_resource('res/jedenprocent.png'),
327 media_type='image/png'
330 get_resource('epub/style.css'),
331 media_type='text/css'
335 def add_file(self, path=None, content=None,
336 media_type='application/xhtml+xml',
337 file_name=None, uid=None,
338 spine=False, toc=None):
341 # jakieś tam ścieśnianie białych znaków?
344 with open(path, 'rb') as f:
346 if file_name is None:
347 file_name = path.rsplit('/', 1)[-1]
350 uid = file_name.split('.', 1)[0]
352 item = epub.EpubItem(
355 media_type=media_type,
359 self.output.add_item(item)
362 self.output.spine.append(item)
364 self.output.spine.insert(spine, item)
367 self.output.toc.append(
375 def add_html(self, html_tree, **kwargs):
376 html = etree.tostring(
377 html_tree, pretty_print=True, xml_declaration=True,
379 doctype='<!DOCTYPE html>'
382 html = librarian.epub.squeeze_whitespace(html)
391 for fname in ('DejaVuSerif.ttf', 'DejaVuSerif-Bold.ttf',
392 'DejaVuSerif-Italic.ttf', 'DejaVuSerif-BoldItalic.ttf'):
395 get_resource('fonts/' + fname),
399 media_type='font/ttf'
402 def start_chunk(self):
403 if getattr(self, 'current_chunk', None) is not None:
404 if not len(self.current_chunk):
407 self.current_chunk = etree.Element(
411 self.cursors[None] = self.current_chunk
412 self.current_cursors.append(self.current_chunk)
414 self.section_number = 0
417 def close_chunk(self):
418 assert self.cursor is self.current_chunk
419 ###### -- what if we're inside?
426 self.chunk_counter = chunk_no + 1
429 html.body.append(self.current_chunk)
432 ## html container from template.
435 file_name='part%d.xhtml' % chunk_no,
439 self.current_chunk = None
440 self.current_cursors.pop()
442 def start_element(self, tag, attr):
443 self.current_cursors.append(
444 etree.SubElement(self.cursor, tag, **attr)
447 def end_element(self):
448 self.current_cursors.pop()
450 def push_text(self, text):
451 self.chars.update(text)
453 self.cursor[-1].tail = (self.cursor[-1].tail or '') + text
455 self.cursor.text = (self.cursor.text or '') + text
458 def assign_image_number(self):
459 image_number = getattr(self, 'image_number', 0)
460 self.image_number = image_number + 1
463 def assign_footnote_number(self):
464 number = getattr(self, 'footnote_number', 1)
465 self.footnote_number = number + 1
468 def assign_section_number(self):
469 number = getattr(self, 'section_number', 1)
470 self.section_number = number + 1
473 def assign_mathml_number(self):
474 number = getattr(self, 'mathml_number', 0)
475 self.mathml_number = number + 1
479 def add_toc_entry(self, fragment, name, precedence):
481 while self.toc_precedences and self.toc_precedences[-1] >= precedence:
482 self.toc_precedences.pop()
484 self.toc_precedences = []
486 real_level = self.toc_base + len(self.toc_precedences)
488 self.toc_precedences.append(precedence)
492 part_number = getattr(
497 filename = 'part%d.xhtml' % part_number
498 uid = filename.split('.')[0]
500 filename += '#' + fragment
501 uid += '-' + fragment
503 toc = self.output.toc
504 for l in range(1, real_level):
505 if isinstance(toc[-1], epub.Link):
506 toc[-1] = [toc[-1], []]
517 def shift_toc_base(self):
521 def add_last_page(self):
523 m = self.document.meta
525 html.title.text = 'Strona redakcyjna'
526 d = etree.SubElement(html.body, 'div', id='book-text')
528 newp = lambda: etree.SubElement(d, 'p', {'class': 'info'})
532 "Wszystkie zasoby Wolnych Lektur możesz swobodnie wykorzystywać, "
533 "publikować i rozpowszechniać pod warunkiem zachowania warunków "
534 "licencji i zgodnie z "
536 a = etree.SubElement(p, "a", href="https://wolnelektury.pl/info/zasady-wykorzystania/")
537 a.text = "Zasadami wykorzystania Wolnych Lektur"
540 etree.SubElement(p, "br")
544 p[-1].tail = "Ten utwór jest udostępniony na licencji "
545 etree.SubElement(p, 'a', href=m.license).text = m.license_description
547 p[-1].tail = 'Ten utwór jest w domenie publicznej.'
549 etree.SubElement(p, "br")
552 "Wszystkie materiały dodatkowe (przypisy, motywy literackie) są "
555 etree.SubElement(p, 'a', href='https://artlibre.org/licence/lal/pl/').text = 'Licencji Wolnej Sztuki 1.3'
557 etree.SubElement(p, "br")
559 "Fundacja Nowoczesna Polska zastrzega sobie prawa do wydania "
560 "krytycznego zgodnie z art. Art.99(2) Ustawy o prawach autorskich "
561 "i prawach pokrewnych. Wykorzystując zasoby z Wolnych Lektur, "
562 "należy pamiętać o zapisach licencji oraz zasadach, które "
566 etree.SubElement(p, 'a', href='https://wolnelektury.pl/info/zasady-wykorzystania/').text = 'Zasadach wykorzystania Wolnych Lektur'
567 p[-1].tail = '. Zapoznaj się z nimi, zanim udostępnisz dalej nasze książki.'
570 p.text = 'E-book można pobrać ze strony: '
572 p, 'a', href=str(m.url),
574 ', '.join(p.readable() for p in m.authors),
580 newp().text = 'Tekst opracowany na podstawie: ' + m.source_name
584 """ + ", ".join(p for p in m.publisher)
587 newp().text = m.description
591 newp().text = 'Opracowanie redakcyjne i przypisy: %s.' % (
592 ', '.join(e.readable() for e in sorted(self.document.editors())))
595 etree.SubElement(d, 'p', {'class': 'minor-info'}).text = '''Publikację wsparli i wsparły:
596 %s.''' % (', '.join(m.funders))
600 p.text = 'Okładka na podstawie: '
610 if getattr(m, self.isbn_field):
611 newp().text = getattr(m, self.isbn_field)
613 newp().text = '\u00a0'
616 p.attrib['class'] = 'minor-info'
618 Plik wygenerowany dnia '''
619 span = etree.SubElement(p, 'span', id='file_date')
620 span.text = str(date.today())
626 file_name='last.xhtml',
627 toc='Strona redakcyjna',
632 def add_annotations(self):
633 if not len(self.footnotes):
637 html.title.text = 'Przypisy'
638 d = etree.SubElement(
653 d.extend(self.footnotes)
657 file_name='annotations.xhtml',
663 # TODO: allow other covers
665 cover_maker = self.make_cover
667 cover_file = six.BytesIO()
668 cover = cover_maker(self.document.meta, width=600)
669 cover.save(cover_file)
670 cover_name = 'cover.%s' % cover.ext()
672 self.output.set_cover(
673 file_name=cover_name,
674 content=cover_file.getvalue(),
677 ci = ('''<?xml version="1.0" encoding="UTF-8"?>
679 <html xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops" lang="en" xml:lang="en">
681 <title>Okładka</title>
683 body { margin: 0em; padding: 0em; }
684 img { width: 100%%; }
688 <img src="cover.%s" alt="Okładka" />
690 </html>''' % cover.ext()).encode('utf-8')
691 self.add_file(file_name='cover.xhtml', content=ci)
693 self.output.spine.append(('cover', 'no'))
694 self.output.guide.append({
696 'href': 'cover.xhtml',
700 def mathml(self, element):
701 name = "math%d.png" % self.assign_mathml_number()
703 content=MathML(element).to_latex().to_png().data,
704 media_type='image/png',