def transform_chunk(chunk_xml, chunk_no, annotations, empty=False, _empty_html_static=[]):
""" transforms one chunk, returns a HTML string, a TOC object and a set of used characters """
toc = TOC()
for element in chunk_xml[0]:
if element.tag == "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)
element.set('sub', str(subnumber))
if empty:
if not _empty_html_static:
_empty_html_static.append(open(get_resource('epub/emptyChunk.html')).read())
chars = set()
output_html = _empty_html_static[0]
else:
find_annotations(annotations, chunk_xml, chunk_no)
replace_by_verse(chunk_xml)
html_tree = xslt(chunk_xml, get_resource('epub/xsltScheme.xsl'))
chars = used_chars(html_tree.getroot())
output_html = etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" ' +
'"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
)
return output_html, toc, chars
class PrestigioCover(Cover):
width = 580
height = 783
background_img = get_resource('res/cover-prestigio.png')
author_top = 446
author_margin_left = 118
author_margin_right = 62
author_lineskip = 60
author_color = '#fff'
author_shadow = '#000'
author_font_ttf = get_resource('fonts/JunicodeWL-Italic.ttf')
author_font_size = 50
title_top = 0
title_margin_left = 118
title_margin_right = 62
title_lineskip = 60
title_color = '#fff'
title_shadow = '#000'
title_font_ttf = get_resource('fonts/JunicodeWL-Italic.ttf')
title_font_size = 50
def pretty_title(self):
return u"„%s”" % self.title
def transform_chunk(chunk_xml, chunk_no, annotations, empty=False, _empty_html_static=[]):
""" transforms one chunk, returns a HTML string, a TOC object and a set of used characters """
toc = TOC()
for element in chunk_xml[0]:
if element.tag == "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)
element.set('sub', str(subnumber))
if empty:
if not _empty_html_static:
_empty_html_static.append(open(get_resource('epub/emptyChunk.html')).read())
chars = set()
output_html = _empty_html_static[0]
else:
find_annotations(annotations, chunk_xml, chunk_no)
replace_by_verse(chunk_xml)
html_tree = xslt(chunk_xml, get_resource('epub/xsltScheme.xsl'))
chars = used_chars(html_tree.getroot())
output_html = etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" ' +
'"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
)
return output_html, toc, chars
class GandalfCover(Cover):
width = 600
height = 730
background_img = get_resource('res/cover-gandalf.png')
author_font_ttf = get_resource('fonts/JunicodeWL-Regular.ttf')
author_font_size = 30
title_font_ttf = get_resource('fonts/JunicodeWL-Regular.ttf')
title_font_size = 40
logo_bottom = 25
logo_width = 250
format = 'PNG'
def image(self):
img = super(LogoWLCover, self).image()
metr = Metric(self, self.scale)
gradient = Image.new("RGBA", (metr.width - metr.bar_width, metr.gradient_height), self.gradient_color)
gradient_mask = Image.new("L", (metr.width - metr.bar_width, metr.gradient_height))
draw = ImageDraw.Draw(gradient_mask)
for line in range(0, metr.gradient_height):
draw.line(
(0, line, metr.width - metr.bar_width, line),
fill=int(255 * self.gradient_opacity * line / metr.gradient_height),
)
img.paste(gradient, (metr.bar_width, metr.height - metr.gradient_height), mask=gradient_mask)
cursor = metr.width - metr.gradient_logo_margin_right
logo_top = metr.height - metr.gradient_height / 2 - metr.gradient_logo_height / 2
for logo_path in self.gradient_logos[::-1]:
logo = Image.open(get_resource(logo_path))
logo = logo.resize(
(logo.size[0] * metr.gradient_logo_height / logo.size[1], metr.gradient_logo_height), Image.ANTIALIAS
)
cursor -= logo.size[0]
img.paste(logo, (cursor, logo_top), mask=logo)
cursor -= metr.gradient_logo_spacing
return img
def image(self):
img = super(LogoWLCover, self).image()
metr = Metric(self, self.scale)
gradient = Image.new(
'RGBA', (metr.width - metr.bar_width, metr.gradient_height),
self.gradient_color)
gradient_mask = Image.new(
'L', (metr.width - metr.bar_width, metr.gradient_height))
draw = ImageDraw.Draw(gradient_mask)
for line in range(0, metr.gradient_height):
draw.line((0, line, metr.width - metr.bar_width, line),
fill=int(255 * self.gradient_opacity * line /
metr.gradient_height))
img.paste(gradient,
(metr.bar_width, metr.height - metr.gradient_height),
mask=gradient_mask)
cursor = metr.width - metr.gradient_logo_margin_right
logo_top = metr.height - metr.gradient_height / 2 - metr.gradient_logo_height / 2 - metr.bleed / 2
for logo_path in self.gradient_logos[::-1]:
logo = Image.open(get_resource(logo_path))
logo = logo.resize(
(logo.size[0] * metr.gradient_logo_height / logo.size[1],
metr.gradient_logo_height), Image.ANTIALIAS)
cursor -= logo.size[0]
img.paste(logo, (cursor, logo_top), mask=logo)
cursor -= metr.gradient_logo_spacing
return img
def image(self):
metr = Metric(self, self.scale)
img = Image.new("RGB", (metr.width, metr.height), self.background_color)
if self.background_img:
background = Image.open(self.background_img)
img.paste(background, None, background)
del background
# WL logo
if metr.logo_width:
logo = Image.open(get_resource("res/wl-logo.png"))
logo = logo.resize((metr.logo_width, logo.size[1] * metr.logo_width / logo.size[0]))
img.paste(logo, ((metr.width - metr.logo_width) / 2, img.size[1] - logo.size[1] - metr.logo_bottom))
top = metr.author_top
tbox = TextBox(metr.width - metr.author_margin_left - metr.author_margin_right, metr.height - top)
author_font = ImageFont.truetype(self.author_font_ttf, metr.author_font_size)
tbox.text(self.pretty_author(), self.author_color, author_font, metr.author_lineskip, self.author_shadow)
text_img = tbox.image()
img.paste(text_img, (metr.author_margin_left, top), text_img)
top += text_img.size[1] + metr.title_top
tbox = TextBox(metr.width - metr.title_margin_left - metr.title_margin_right, metr.height - top)
title_font = ImageFont.truetype(self.title_font_ttf, metr.title_font_size)
tbox.text(self.pretty_title(), self.title_color, title_font, metr.title_lineskip, self.title_shadow)
text_img = tbox.image()
img.paste(text_img, (metr.title_margin_left, top), text_img)
return img
def to_png(self):
tmpl = open(get_resource('res/embeds/latex/template.tex'), 'rb').read().decode('utf-8')
tempdir = mkdtemp('-librarian-embed-latex')
fpath = os.path.join(tempdir, 'doc.tex')
with open(fpath, 'wb') as f:
f.write((tmpl % {'code': self.data}).encode('utf-8'))
call(['xelatex', '-interaction=batchmode', '-output-directory', tempdir, fpath], stdout=PIPE, stderr=PIPE)
call(['convert', '-density', '150', os.path.join(tempdir, 'doc.pdf'), '-trim',
os.path.join(tempdir, 'doc.png')])
pngdata = open(os.path.join(tempdir, 'doc.png'), 'rb').read()
shutil.rmtree(tempdir)
return create_embed('image/png', data=pngdata)
def to_png(self):
tmpl = open(get_resource('res/embeds/latex/template.tex')).read().decode('utf-8')
tempdir = mkdtemp('-librarian-embed-latex')
fpath = os.path.join(tempdir, 'doc.tex')
with open(fpath, 'w') as f:
f.write((tmpl % {'code': self.data}).encode('utf-8'))
call(['xelatex', '-interaction=batchmode', '-output-directory', tempdir, fpath], stdout=PIPE, stderr=PIPE)
call(['convert', '-density', '150', os.path.join(tempdir, 'doc.pdf'), '-trim',
os.path.join(tempdir, 'doc.png')])
pngdata = open(os.path.join(tempdir, 'doc.png')).read()
shutil.rmtree(tempdir)
return create_embed('image/png', data=pngdata)
def get_short_lng_code(text):
result = ''
text = ''.join(text)
with open(get_resource('res/ISO-639-2_8859-1.txt'), 'rb') as f:
for line in f:
list = line.strip().split('|')
if list[0] == text:
result = list[2]
if result == '':
return text
else:
return result
def get_short_lng_code(text):
result = ''
text = ''.join(text)
with open(get_resource('res/ISO-639-2_8859-1.txt'), 'rb') as f:
for line in f:
list = line.strip().split('|')
if list[0] == text:
result = list[2]
if result == '':
return text
else:
return result
def lang_code_3to2(context, text):
"""Convert 3-letter language code to 2-letter code"""
result = ''
text = ''.join(text)
with open(get_resource('res/ISO-639-2_8859-1.txt'), 'rb') as f:
for line in f.read().decode('latin1').split('\n'):
list = line.strip().split('|')
if list[0] == text:
result = list[2]
if result == '':
return text
else:
return result
def transform_chunk(chunk_xml, chunk_no, annotations, empty=False, _empty_html_static=[]):
""" transforms one chunk, returns a HTML string, a TOC object and a set of used characters """
toc = TOC()
for element in chunk_xml[0]:
if element.tag in ("naglowek_czesc", "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)
element.set('sub', str(subnumber))
if empty:
if not _empty_html_static:
_empty_html_static.append(open(get_resource('epub/emptyChunk.html')).read())
chars = set()
output_html = _empty_html_static[0]
else:
find_annotations(annotations, chunk_xml, chunk_no)
replace_by_verse(chunk_xml)
html_tree = xslt(chunk_xml, get_resource('epub/xsltScheme.xsl'))
chars = used_chars(html_tree.getroot())
output_html = etree.tostring(html_tree, method="html", pretty_print=True)
return output_html, toc, chars
def lang_code_3to2(context, text):
"""Convert 3-letter language code to 2-letter code"""
result = ''
text = ''.join(text)
with open(get_resource('res/ISO-639-2_8859-1.txt'), 'rb') as f:
for line in f:
list = line.strip().split('|')
if list[0] == text:
result=list[2]
if result == '':
return text
else:
return result
class BookotekaCover(Cover):
width = 2140
height = 2733
background_img = get_resource('res/cover-bookoteka.png')
author_top = 480
author_margin_left = 307
author_margin_right = 233
author_lineskip = 156
author_color = '#d9d919'
author_font_ttf = get_resource('fonts/JunicodeWL-Regular.ttf')
author_font_size = 130
title_top = 400
title_margin_left = 307
title_margin_right = 233
title_lineskip = 168
title_color = '#d9d919'
title_font_ttf = get_resource('fonts/JunicodeWL-Regular.ttf')
title_font_size = 140
format = 'PNG'
def image(self):
img = Image.new('RGB', (self.width, self.height), self.background_color)
if self.background_img:
background = Image.open(self.background_img)
img.paste(background, None, background)
del background
# WL logo
if self.logo_width:
logo = Image.open(get_resource('res/wl-logo.png'))
logo = logo.resize((self.logo_width, logo.size[1] * self.logo_width / logo.size[0]))
img.paste(logo, ((self.width - self.logo_width) / 2, img.size[1] - logo.size[1] - self.logo_bottom))
top = self.author_top
tbox = TextBox(
self.width - self.author_margin_left - self.author_margin_right,
self.height - top,
)
author_font = self.author_font or ImageFont.truetype(
get_resource('fonts/DejaVuSerif.ttf'), 30)
tbox.text(self.pretty_author(), self.author_color, author_font,
self.author_lineskip, self.author_shadow)
text_img = tbox.image()
img.paste(text_img, (self.author_margin_left, top), text_img)
top += text_img.size[1] + self.title_top
tbox = TextBox(
self.width - self.title_margin_left - self.title_margin_right,
self.height - top,
)
title_font = self.author_font or ImageFont.truetype(
get_resource('fonts/DejaVuSerif.ttf'), 40)
tbox.text(self.pretty_title(), self.title_color, title_font,
self.title_lineskip, self.title_shadow)
text_img = tbox.image()
img.paste(text_img, (self.title_margin_left, top), text_img)
return img
def add_source_note(wlpic, img):
from PIL import ImageDraw, ImageFont
from librarian import get_resource
annotated = Image.new(img.mode, (img.size[0], img.size[1] + 40), (255, 255, 255))
annotated.paste(img, (0, 0))
annotation = Image.new('RGB', (img.size[0] * 3, 120), (255, 255, 255))
ImageDraw.Draw(annotation).text(
(30, 15),
wlpic.picture_info.source_name,
(0, 0, 0),
font=ImageFont.truetype(get_resource("fonts/DejaVuSerif.ttf"), 75)
)
annotated.paste(annotation.resize((img.size[0], 40), Image.ANTIALIAS), (0, img.size[1]))
return annotated
def transform(wldoc, verbose=False,
sample=None, cover=None, flags=None):
""" produces a MOBI file
wldoc: a WLDocument
sample=n: generate sample e-book (with at least n paragraphs)
cover: a cover.Cover factory overriding default
flags: less-advertising,
"""
document = deepcopy(wldoc)
del wldoc
book_info = document.book_info
# provide a cover by default
if not cover:
cover = DefaultEbookCover
cover_file = NamedTemporaryFile(suffix='.png', delete=False)
bound_cover = cover(book_info)
bound_cover.save(cover_file)
if bound_cover.uses_dc_cover:
if document.book_info.cover_by:
document.edoc.getroot().set('data-cover-by', document.book_info.cover_by)
if document.book_info.cover_source:
document.edoc.getroot().set('data-cover-source', document.book_info.cover_source)
if not flags:
flags = []
flags = list(flags) + ['with-full-fonts']
epub = document.as_epub(verbose=verbose, sample=sample, html_toc=True,
flags=flags, style=get_resource('epub/style.css'))
if verbose:
kwargs = {}
else:
devnull = open("/dev/null", 'w')
kwargs = {"stdout": devnull, "stderr": devnull}
output_file = NamedTemporaryFile(prefix='librarian', suffix='.mobi', delete=False)
output_file.close()
subprocess.check_call(['ebook-convert', epub.get_filename(), output_file.name,
'--no-inline-toc', '--mobi-file-type=both', '--subset-embedded-fonts', '--mobi-ignore-margins', '--cover=%s' % cover_file.name], **kwargs)
os.unlink(cover_file.name)
return OutputFile.from_filename(output_file.name)
def image(self):
metr = Metric(self, self.scale)
img = Image.new('RGB', (metr.width, metr.height),
self.background_color)
if self.background_img:
background = Image.open(self.background_img)
img.paste(background, None, background)
del background
# WL logo
if metr.logo_width:
logo = Image.open(get_resource('res/wl-logo.png'))
logo = logo.resize(
(metr.logo_width,
int(round(logo.size[1] * metr.logo_width / logo.size[0]))))
img.paste(logo, ((metr.width - metr.logo_width) // 2,
img.size[1] - logo.size[1] - metr.logo_bottom))
top = metr.author_top
tbox = TextBox(
metr.width - metr.author_margin_left - metr.author_margin_right,
metr.height - top,
)
author_font = ImageFont.truetype(self.author_font_ttf,
metr.author_font_size)
tbox.text(self.pretty_author(), self.author_color, author_font,
metr.author_lineskip, self.author_shadow)
text_img = tbox.image()
img.paste(text_img, (metr.author_margin_left, top), text_img)
top += text_img.size[1] + metr.title_top
tbox = TextBox(
metr.width - metr.title_margin_left - metr.title_margin_right,
metr.height - top,
)
title_font = ImageFont.truetype(self.title_font_ttf,
metr.title_font_size)
tbox.text(self.pretty_title(), self.title_color, title_font,
metr.title_lineskip, self.title_shadow)
text_img = tbox.image()
img.paste(text_img, (metr.title_margin_left, top), text_img)
return img
def set_hyph_language(source_tree):
def get_short_lng_code(text):
result = ''
text = ''.join(text)
with open(get_resource('res/ISO-639-2_8859-1.txt'), 'rb') as f:
for line in f:
list = line.strip().split('|')
if list[0] == text:
result = list[2]
if result == '':
return text
else:
return result
bibl_lng = etree.XPath('//dc:language//text()',
namespaces={'dc': str(DCNS)})(source_tree)
short_lng = get_short_lng_code(bibl_lng[0])
try:
return Hyphenator(get_resource('res/hyph-dictionaries/hyph_' +
short_lng + '.dic'))
except:
pass
def set_hyph_language(source_tree):
def get_short_lng_code(text):
result = ''
text = ''.join(text)
with open(get_resource('res/ISO-639-2_8859-1.txt'), 'rb') as f:
for line in f:
list = line.strip().split('|')
if list[0] == text:
result = list[2]
if result == '':
return text
else:
return result
bibl_lng = etree.XPath('//dc:language//text()',
namespaces={'dc': str(DCNS)})(source_tree)
short_lng = get_short_lng_code(bibl_lng[0])
try:
return Hyphenator(get_resource('res/hyph-dictionaries/hyph_' +
short_lng + '.dic'))
except:
pass
def transform(wldoc, verbose=False,
style=None, html_toc=False,
sample=None, cover=None, flags=None):
""" produces a EPUB file
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
"""
def transform_file(wldoc, chunk_counter=1, first=True, sample=None):
""" processes one input file and proceeds to its children """
replace_characters(wldoc.edoc.getroot())
hyphenator = set_hyph_language(wldoc.edoc.getroot())
hyphenate_and_fix_conjunctions(wldoc.edoc.getroot(), hyphenator)
# every input file will have a TOC entry,
# pointing to starting chunk
toc = TOC(wldoc.book_info.title, "part%d.html" % chunk_counter)
chars = set()
if first:
# write book title page
html_tree = xslt(wldoc.edoc, get_resource('epub/xsltTitle.xsl'))
chars = used_chars(html_tree.getroot())
zip.writestr(
'OPS/title.html',
etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"' +
' "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
)
)
# add a title page TOC entry
toc.add(u"Strona tytułowa", "title.html")
elif wldoc.book_info.parts:
# write title page for every parent
if sample is not None and sample <= 0:
chars = set()
html_string = open(get_resource('epub/emptyChunk.html')).read()
else:
html_tree = xslt(wldoc.edoc, get_resource('epub/xsltChunkTitle.xsl'))
chars = used_chars(html_tree.getroot())
html_string = etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"' +
' "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
)
zip.writestr('OPS/part%d.html' % chunk_counter, html_string)
add_to_manifest(manifest, chunk_counter)
add_to_spine(spine, chunk_counter)
chunk_counter += 1
if len(wldoc.edoc.getroot()) > 1:
# rdf before style master
main_text = wldoc.edoc.getroot()[1]
else:
# rdf in style master
main_text = wldoc.edoc.getroot()[0]
if main_text.tag == RDFNS('RDF'):
main_text = None
if main_text is not None:
for chunk_xml in chop(main_text):
empty = False
if sample is not None:
if sample <= 0:
empty = True
else:
sample -= len(chunk_xml.xpath('//strofa|//akap|//akap_cd|//akap_dialog'))
chunk_html, chunk_toc, chunk_chars = transform_chunk(chunk_xml, chunk_counter, annotations, empty)
toc.extend(chunk_toc)
chars = chars.union(chunk_chars)
zip.writestr('OPS/part%d.html' % chunk_counter, chunk_html)
add_to_manifest(manifest, chunk_counter)
add_to_spine(spine, chunk_counter)
chunk_counter += 1
for child in wldoc.parts():
child_toc, chunk_counter, chunk_chars, sample = transform_file(
child, chunk_counter, first=False, sample=sample)
toc.append(child_toc)
chars = chars.union(chunk_chars)
return toc, chunk_counter, chars, sample
document = deepcopy(wldoc)
del wldoc
if flags:
for flag in flags:
document.edoc.getroot().set(flag, 'yes')
# add editors info
editors = document.editors()
if editors:
document.edoc.getroot().set('editors', u', '.join(sorted(
editor.readable() for editor in editors)))
if document.book_info.funders:
document.edoc.getroot().set('funders', u', '.join(
document.book_info.funders))
if document.book_info.thanks:
document.edoc.getroot().set('thanks', document.book_info.thanks)
opf = xslt(document.book_info.to_etree(), get_resource('epub/xsltContent.xsl'))
manifest = opf.find('.//' + OPFNS('manifest'))
guide = opf.find('.//' + OPFNS('guide'))
spine = opf.find('.//' + OPFNS('spine'))
output_file = NamedTemporaryFile(prefix='librarian', suffix='.epub', delete=False)
zip = zipfile.ZipFile(output_file, 'w', zipfile.ZIP_DEFLATED)
# write static elements
mime = zipfile.ZipInfo()
mime.filename = 'mimetype'
mime.compress_type = zipfile.ZIP_STORED
mime.extra = ''
zip.writestr(mime, 'application/epub+zip')
zip.writestr(
'META-INF/container.xml',
'<?xml version="1.0" ?>'
'<container version="1.0" '
'xmlns="urn:oasis:names:tc:opendocument:xmlns:container">'
'<rootfiles><rootfile full-path="OPS/content.opf" '
'media-type="application/oebps-package+xml" />'
'</rootfiles></container>'
)
zip.write(get_resource('res/wl-logo-small.png'),
os.path.join('OPS', 'logo_wolnelektury.png'))
zip.write(get_resource('res/jedenprocent.png'),
os.path.join('OPS', 'jedenprocent.png'))
if not style:
style = get_resource('epub/style.css')
zip.write(style, os.path.join('OPS', 'style.css'))
if cover:
if cover is True:
cover = DefaultEbookCover
cover_file = StringIO()
bound_cover = cover(document.book_info)
bound_cover.save(cover_file)
cover_name = 'cover.%s' % bound_cover.ext()
zip.writestr(os.path.join('OPS', cover_name), cover_file.getvalue())
del cover_file
cover_tree = etree.parse(get_resource('epub/cover.html'))
cover_tree.find('//' + XHTMLNS('img')).set('src', cover_name)
zip.writestr('OPS/cover.html', etree.tostring(
cover_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" ' +
'"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
))
if bound_cover.uses_dc_cover:
if document.book_info.cover_by:
document.edoc.getroot().set('data-cover-by', document.book_info.cover_by)
if document.book_info.cover_source:
document.edoc.getroot().set('data-cover-source', document.book_info.cover_source)
manifest.append(etree.fromstring(
'<item id="cover" href="cover.html" media-type="application/xhtml+xml" />'))
manifest.append(etree.fromstring(
'<item id="cover-image" href="%s" media-type="%s" />' % (cover_name, bound_cover.mime_type())))
spine.insert(0, etree.fromstring('<itemref idref="cover"/>'))
opf.getroot()[0].append(etree.fromstring('<meta name="cover" content="cover-image"/>'))
guide.append(etree.fromstring('<reference href="cover.html" type="cover" title="Okładka"/>'))
annotations = etree.Element('annotations')
toc_file = etree.fromstring(
'<?xml version="1.0" encoding="utf-8"?><!DOCTYPE ncx PUBLIC '
'"-//NISO//DTD ncx 2005-1//EN" '
'"http://www.daisy.org/z3986/2005/ncx-2005-1.dtd">'
'<ncx xmlns="http://www.daisy.org/z3986/2005/ncx/" xml:lang="pl" '
'version="2005-1"><head></head><docTitle></docTitle><navMap>'
'</navMap></ncx>'
)
nav_map = toc_file[-1]
if html_toc:
manifest.append(etree.fromstring(
'<item id="html_toc" href="toc.html" media-type="application/xhtml+xml" />'))
spine.append(etree.fromstring(
'<itemref idref="html_toc" />'))
guide.append(etree.fromstring('<reference href="toc.html" type="toc" title="Spis treści"/>'))
toc, chunk_counter, chars, sample = transform_file(document, sample=sample)
if len(toc.children) < 2:
toc.add(u"Początek utworu", "part1.html")
# Last modifications in container files and EPUB creation
if len(annotations) > 0:
toc.add("Przypisy", "annotations.html")
manifest.append(etree.fromstring(
'<item id="annotations" href="annotations.html" media-type="application/xhtml+xml" />'))
spine.append(etree.fromstring(
'<itemref idref="annotations" />'))
replace_by_verse(annotations)
html_tree = xslt(annotations, get_resource('epub/xsltAnnotations.xsl'))
chars = chars.union(used_chars(html_tree.getroot()))
zip.writestr('OPS/annotations.html', etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" ' +
'"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
))
toc.add("Wesprzyj Wolne Lektury", "support.html")
manifest.append(etree.fromstring(
'<item id="support" href="support.html" media-type="application/xhtml+xml" />'))
spine.append(etree.fromstring(
'<itemref idref="support" />'))
html_string = open(get_resource('epub/support.html')).read()
chars.update(used_chars(etree.fromstring(html_string)))
zip.writestr('OPS/support.html', html_string)
toc.add("Strona redakcyjna", "last.html")
manifest.append(etree.fromstring(
'<item id="last" href="last.html" media-type="application/xhtml+xml" />'))
spine.append(etree.fromstring(
'<itemref idref="last" />'))
html_tree = xslt(document.edoc, get_resource('epub/xsltLast.xsl'))
chars.update(used_chars(html_tree.getroot()))
zip.writestr('OPS/last.html', etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" ' +
'"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
))
if not flags or not 'without-fonts' in flags:
# strip fonts
tmpdir = mkdtemp('-librarian-epub')
try:
cwd = os.getcwd()
except OSError:
cwd = None
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)
else:
subprocess.check_call(optimizer_call, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
zip.write(os.path.join(tmpdir, fname), os.path.join('OPS', fname))
manifest.append(etree.fromstring(
'<item id="%s" href="%s" media-type="application/x-font-truetype" />' % (fname, fname)))
rmtree(tmpdir)
if cwd is not None:
os.chdir(cwd)
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:
meta = toc_file.makeelement(NCXNS('meta'))
meta.set('name', st)
meta.set('content', '0')
toc_file[0].append(meta)
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(('<text>', title, '</text>')))
# write TOC
if html_toc:
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,
xml_declaration=True, encoding="utf-8"))
zip.close()
return OutputFile.from_filename(output_file.name)
def transform(wldoc,
verbose=False,
save_tex=None,
morefloats=None,
cover=None,
flags=None,
customizations=None,
ilustr_path='',
latex_dir=False):
""" produces a PDF file with XeLaTeX
wldoc: a WLDocument
verbose: prints all output from LaTeX
save_tex: path to save the intermediary LaTeX file to
morefloats (old/new/none): force specific morefloats
cover: a cover.Cover factory or True for default
flags: less-advertising,
customizations: user requested customizations regarding various formatting parameters (passed to wl LaTeX class)
"""
# Parse XSLT
try:
book_info = wldoc.book_info
document = load_including_children(wldoc)
root = document.edoc.getroot()
if cover:
if cover is True:
cover = make_cover
bound_cover = cover(book_info, width=1200)
root.set('data-cover-width', str(bound_cover.width))
root.set('data-cover-height', str(bound_cover.height))
if bound_cover.uses_dc_cover:
if book_info.cover_by:
root.set('data-cover-by', book_info.cover_by)
if book_info.cover_source:
root.set('data-cover-source', book_info.cover_source)
if flags:
for flag in flags:
root.set('flag-' + flag, 'yes')
# check for LaTeX packages
if morefloats:
root.set('morefloats', morefloats.lower())
elif package_available('morefloats', 'maxfloats=19'):
root.set('morefloats', 'new')
# add customizations
if customizations is not None:
root.set('customizations', u','.join(customizations))
# add editors info
editors = document.editors()
if editors:
root.set(
'editors',
u', '.join(sorted(editor.readable() for editor in editors)))
if document.book_info.funders:
root.set('funders', u', '.join(document.book_info.funders))
if document.book_info.thanks:
root.set('thanks', document.book_info.thanks)
# hack the tree
move_motifs_inside(document.edoc)
hack_motifs(document.edoc)
parse_creator(document.edoc)
substitute_hyphens(document.edoc)
fix_hanging(document.edoc)
fix_tables(document.edoc)
mark_subauthors(document.edoc)
# wl -> TeXML
style_filename = get_stylesheet("wl2tex")
style = etree.parse(style_filename)
functions.reg_mathml_latex()
# TeXML -> LaTeX
temp = mkdtemp('-wl2pdf')
for ilustr in document.edoc.findall("//ilustr"):
shutil.copy(os.path.join(ilustr_path, ilustr.get("src")), temp)
for sponsor in book_info.sponsors:
ins = etree.Element("data-sponsor", name=sponsor)
logo = sponsor_logo(sponsor)
if logo:
fname = 'sponsor-%s' % os.path.basename(logo)
shutil.copy(logo, os.path.join(temp, fname))
ins.set('src', fname)
root.insert(0, ins)
if book_info.sponsor_note:
root.set("sponsor-note", book_info.sponsor_note)
texml = document.transform(style)
if cover:
with open(os.path.join(temp, 'cover.png'), 'w') as f:
bound_cover.save(f, quality=80)
del document # no longer needed large object :)
tex_path = os.path.join(temp, 'doc.tex')
fout = open(tex_path, 'wb')
process(six.BytesIO(texml), fout, 'utf-8')
fout.close()
del texml
if save_tex:
shutil.copy(tex_path, save_tex)
# LaTeX -> PDF
shutil.copy(get_resource('pdf/wl.cls'), temp)
shutil.copy(get_resource('res/wl-logo.png'), temp)
if latex_dir:
return temp
try:
cwd = os.getcwd()
except OSError:
cwd = None
os.chdir(temp)
# some things work better when compiled twice
# (table of contents, [line numbers - disabled])
for run in range(2):
if verbose:
p = call(['xelatex', tex_path])
else:
p = call(['xelatex', '-interaction=batchmode', tex_path],
stdout=PIPE,
stderr=PIPE)
if p:
raise ParseError("Error parsing .tex file")
if cwd is not None:
os.chdir(cwd)
output_file = NamedTemporaryFile(prefix='librarian',
suffix='.pdf',
delete=False)
pdf_path = os.path.join(temp, 'doc.pdf')
shutil.move(pdf_path, output_file.name)
shutil.rmtree(temp)
return OutputFile.from_filename(output_file.name)
except (XMLSyntaxError, XSLTApplyError) as e:
raise ParseError(e)
def html(self):
with open(get_resource('epub/toc.html')) as f:
t = unicode(f.read(), 'utf-8')
return t % self.html_part()
def html(self):
with open(get_resource('epub/toc.html'), 'rb') as f:
t = f.read().decode('utf-8')
return t % self.html_part()
def get_stylesheet(name):
return get_resource(STYLESHEETS[name])
def transform(wldoc, verbose=False, save_tex=None, morefloats=None,
cover=None, flags=None, customizations=None):
""" produces a PDF file with XeLaTeX
wldoc: a WLDocument
verbose: prints all output from LaTeX
save_tex: path to save the intermediary LaTeX file to
morefloats (old/new/none): force specific morefloats
cover: a cover.Cover factory or True for default
flags: less-advertising,
customizations: user requested customizations regarding various formatting parameters (passed to wl LaTeX class)
"""
# Parse XSLT
try:
book_info = wldoc.book_info
document = load_including_children(wldoc)
root = document.edoc.getroot()
if cover:
if cover is True:
cover = DefaultEbookCover
bound_cover = cover(book_info, width=1200)
root.set('data-cover-width', str(bound_cover.width))
root.set('data-cover-height', str(bound_cover.height))
if bound_cover.uses_dc_cover:
if book_info.cover_by:
root.set('data-cover-by', book_info.cover_by)
if book_info.cover_source:
root.set('data-cover-source',
book_info.cover_source)
if flags:
for flag in flags:
root.set('flag-' + flag, 'yes')
# check for LaTeX packages
if morefloats:
root.set('morefloats', morefloats.lower())
elif package_available('morefloats', 'maxfloats=19'):
root.set('morefloats', 'new')
# add customizations
if customizations is not None:
root.set('customizations', u','.join(customizations))
# add editors info
editors = document.editors()
if editors:
root.set('editors', u', '.join(sorted(
editor.readable() for editor in editors)))
if document.book_info.funders:
root.set('funders', u', '.join(document.book_info.funders))
if document.book_info.thanks:
root.set('thanks', document.book_info.thanks)
# hack the tree
move_motifs_inside(document.edoc)
hack_motifs(document.edoc)
parse_creator(document.edoc)
substitute_hyphens(document.edoc)
fix_hanging(document.edoc)
fix_tables(document.edoc)
# wl -> TeXML
style_filename = get_stylesheet("wl2tex")
style = etree.parse(style_filename)
functions.reg_mathml_latex()
# TeXML -> LaTeX
temp = mkdtemp('-wl2pdf')
for sponsor in book_info.sponsors:
ins = etree.Element("data-sponsor", name=sponsor)
logo = sponsor_logo(sponsor)
if logo:
fname = 'sponsor-%s' % os.path.basename(logo)
shutil.copy(logo, os.path.join(temp, fname))
ins.set('src', fname)
root.insert(0, ins)
if book_info.sponsor_note:
root.set("sponsor-note", book_info.sponsor_note)
texml = document.transform(style)
if cover:
with open(os.path.join(temp, 'cover.png'), 'w') as f:
bound_cover.save(f, quality=80)
del document # no longer needed large object :)
tex_path = os.path.join(temp, 'doc.tex')
fout = open(tex_path, 'w')
process(StringIO(texml), fout, 'utf-8')
fout.close()
del texml
if save_tex:
shutil.copy(tex_path, save_tex)
# LaTeX -> PDF
shutil.copy(get_resource('pdf/wl.cls'), temp)
shutil.copy(get_resource('res/wl-logo.png'), temp)
try:
cwd = os.getcwd()
except OSError:
cwd = None
os.chdir(temp)
if verbose:
p = call(['xelatex', tex_path])
else:
p = call(['xelatex', '-interaction=batchmode', tex_path], stdout=PIPE, stderr=PIPE)
if p:
raise ParseError("Error parsing .tex file")
if cwd is not None:
os.chdir(cwd)
output_file = NamedTemporaryFile(prefix='librarian', suffix='.pdf', delete=False)
pdf_path = os.path.join(temp, 'doc.pdf')
shutil.move(pdf_path, output_file.name)
shutil.rmtree(temp)
return OutputFile.from_filename(output_file.name)
except (XMLSyntaxError, XSLTApplyError), e:
raise ParseError(e)
def to_latex(self):
xslt = etree.parse(get_resource('res/embeds/mathml/mathml2latex.xslt'))
output = self.tree.xslt(xslt)
return create_embed('application/x-latex', data=unicode(output))
def transform(wldoc, verbose=False, style=None, html_toc=False,
sample=None, cover=None, flags=None, hyphenate=False, ilustr_path='', output_type='epub'):
""" produces a EPUB file
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
"""
def transform_file(wldoc, chunk_counter=1, first=True, sample=None):
""" processes one input file and proceeds to its children """
replace_characters(wldoc.edoc.getroot())
hyphenator = set_hyph_language(wldoc.edoc.getroot()) if hyphenate else None
hyphenate_and_fix_conjunctions(wldoc.edoc.getroot(), hyphenator)
# every input file will have a TOC entry,
# pointing to starting chunk
toc = TOC(wldoc.book_info.title, "part%d.html" % chunk_counter)
chars = set()
if first:
# write book title page
html_tree = xslt(wldoc.edoc, get_resource('epub/xsltTitle.xsl'), outputtype=output_type)
chars = used_chars(html_tree.getroot())
zip.writestr(
'OPS/title.html',
etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"' +
' "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
)
)
# add a title page TOC entry
toc.add(u"Strona tytułowa", "title.html")
elif wldoc.book_info.parts:
# write title page for every parent
if sample is not None and sample <= 0:
chars = set()
html_string = open(get_resource('epub/emptyChunk.html')).read()
else:
html_tree = xslt(wldoc.edoc, get_resource('epub/xsltChunkTitle.xsl'))
chars = used_chars(html_tree.getroot())
html_string = etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"' +
' "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
)
zip.writestr('OPS/part%d.html' % chunk_counter, html_string)
add_to_manifest(manifest, chunk_counter)
add_to_spine(spine, chunk_counter)
chunk_counter += 1
if len(wldoc.edoc.getroot()) > 1:
# rdf before style master
main_text = wldoc.edoc.getroot()[1]
else:
# rdf in style master
main_text = wldoc.edoc.getroot()[0]
if main_text.tag == RDFNS('RDF'):
main_text = None
if main_text is not None:
for chunk_xml in chop(main_text):
empty = False
if sample is not None:
if sample <= 0:
empty = True
else:
sample -= len(chunk_xml.xpath('//strofa|//akap|//akap_cd|//akap_dialog'))
chunk_html, chunk_toc, chunk_chars = transform_chunk(chunk_xml, chunk_counter, annotations, empty)
toc.extend(chunk_toc)
chars = chars.union(chunk_chars)
zip.writestr('OPS/part%d.html' % chunk_counter, chunk_html)
add_to_manifest(manifest, chunk_counter)
add_to_spine(spine, chunk_counter)
chunk_counter += 1
for child in wldoc.parts():
child_toc, chunk_counter, chunk_chars, sample = transform_file(
child, chunk_counter, first=False, sample=sample)
toc.append(child_toc)
chars = chars.union(chunk_chars)
return toc, chunk_counter, chars, sample
document = deepcopy(wldoc)
del wldoc
if flags:
for flag in flags:
document.edoc.getroot().set(flag, 'yes')
document.clean_ed_note()
document.clean_ed_note('abstrakt')
# add editors info
editors = document.editors()
if editors:
document.edoc.getroot().set('editors', u', '.join(sorted(
editor.readable() for editor in editors)))
if document.book_info.funders:
document.edoc.getroot().set('funders', u', '.join(
document.book_info.funders))
if document.book_info.thanks:
document.edoc.getroot().set('thanks', document.book_info.thanks)
opf = xslt(document.book_info.to_etree(), get_resource('epub/xsltContent.xsl'))
manifest = opf.find('.//' + OPFNS('manifest'))
guide = opf.find('.//' + OPFNS('guide'))
spine = opf.find('.//' + OPFNS('spine'))
output_file = NamedTemporaryFile(prefix='librarian', suffix='.epub', delete=False)
zip = zipfile.ZipFile(output_file, 'w', zipfile.ZIP_DEFLATED)
functions.reg_mathml_epub(zip)
if os.path.isdir(ilustr_path):
for i, filename in enumerate(os.listdir(ilustr_path)):
file_path = os.path.join(ilustr_path, filename)
zip.write(file_path, os.path.join('OPS', filename))
image_id = 'image%s' % i
manifest.append(etree.fromstring(
'<item id="%s" href="%s" media-type="%s" />' % (image_id, filename, guess_type(file_path)[0])))
# write static elements
mime = zipfile.ZipInfo()
mime.filename = 'mimetype'
mime.compress_type = zipfile.ZIP_STORED
mime.extra = ''
zip.writestr(mime, 'application/epub+zip')
zip.writestr(
'META-INF/container.xml',
'<?xml version="1.0" ?>'
'<container version="1.0" '
'xmlns="urn:oasis:names:tc:opendocument:xmlns:container">'
'<rootfiles><rootfile full-path="OPS/content.opf" '
'media-type="application/oebps-package+xml" />'
'</rootfiles></container>'
)
zip.write(get_resource('res/wl-logo-small.png'),
os.path.join('OPS', 'logo_wolnelektury.png'))
zip.write(get_resource('res/jedenprocent.png'),
os.path.join('OPS', 'jedenprocent.png'))
if not style:
style = get_resource('epub/style.css')
zip.write(style, os.path.join('OPS', 'style.css'))
if cover:
if cover is True:
cover = DefaultEbookCover
cover_file = StringIO()
bound_cover = cover(document.book_info)
bound_cover.save(cover_file)
cover_name = 'cover.%s' % bound_cover.ext()
zip.writestr(os.path.join('OPS', cover_name), cover_file.getvalue())
del cover_file
cover_tree = etree.parse(get_resource('epub/cover.html'))
cover_tree.find('//' + XHTMLNS('img')).set('src', cover_name)
zip.writestr('OPS/cover.html', etree.tostring(
cover_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" ' +
'"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
))
if bound_cover.uses_dc_cover:
if document.book_info.cover_by:
document.edoc.getroot().set('data-cover-by', document.book_info.cover_by)
if document.book_info.cover_source:
document.edoc.getroot().set('data-cover-source', document.book_info.cover_source)
manifest.append(etree.fromstring(
'<item id="cover" href="cover.html" media-type="application/xhtml+xml" />'))
manifest.append(etree.fromstring(
'<item id="cover-image" href="%s" media-type="%s" />' % (cover_name, bound_cover.mime_type())))
spine.insert(0, etree.fromstring('<itemref idref="cover"/>'))
opf.getroot()[0].append(etree.fromstring('<meta name="cover" content="cover-image"/>'))
guide.append(etree.fromstring('<reference href="cover.html" type="cover" title="Okładka"/>'))
annotations = etree.Element('annotations')
toc_file = etree.fromstring(
'<?xml version="1.0" encoding="utf-8"?><!DOCTYPE ncx PUBLIC '
'"-//NISO//DTD ncx 2005-1//EN" '
'"http://www.daisy.org/z3986/2005/ncx-2005-1.dtd">'
'<ncx xmlns="http://www.daisy.org/z3986/2005/ncx/" xml:lang="pl" '
'version="2005-1"><head></head><docTitle></docTitle><navMap>'
'</navMap></ncx>'
)
nav_map = toc_file[-1]
if html_toc:
manifest.append(etree.fromstring(
'<item id="html_toc" href="toc.html" media-type="application/xhtml+xml" />'))
spine.append(etree.fromstring(
'<itemref idref="html_toc" />'))
guide.append(etree.fromstring('<reference href="toc.html" type="toc" title="Spis treści"/>'))
toc, chunk_counter, chars, sample = transform_file(document, sample=sample)
if len(toc.children) < 2:
toc.add(u"Początek utworu", "part1.html")
# Last modifications in container files and EPUB creation
if len(annotations) > 0:
toc.add("Przypisy", "annotations.html")
manifest.append(etree.fromstring(
'<item id="annotations" href="annotations.html" media-type="application/xhtml+xml" />'))
spine.append(etree.fromstring(
'<itemref idref="annotations" />'))
replace_by_verse(annotations)
html_tree = xslt(annotations, get_resource('epub/xsltAnnotations.xsl'))
chars = chars.union(used_chars(html_tree.getroot()))
zip.writestr('OPS/annotations.html', etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" ' +
'"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
))
toc.add("Wesprzyj Wolne Lektury", "support.html")
manifest.append(etree.fromstring(
'<item id="support" href="support.html" media-type="application/xhtml+xml" />'))
spine.append(etree.fromstring(
'<itemref idref="support" />'))
html_string = open(get_resource('epub/support.html')).read()
chars.update(used_chars(etree.fromstring(html_string)))
zip.writestr('OPS/support.html', html_string)
toc.add("Strona redakcyjna", "last.html")
manifest.append(etree.fromstring(
'<item id="last" href="last.html" media-type="application/xhtml+xml" />'))
spine.append(etree.fromstring(
'<itemref idref="last" />'))
html_tree = xslt(document.edoc, get_resource('epub/xsltLast.xsl'), outputtype=output_type)
chars.update(used_chars(html_tree.getroot()))
zip.writestr('OPS/last.html', etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" ' +
'"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
))
if not flags or 'without-fonts' not in flags:
# strip fonts
tmpdir = mkdtemp('-librarian-epub')
try:
cwd = os.getcwd()
except OSError:
cwd = None
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)
else:
dev_null = open(os.devnull, 'w')
subprocess.check_call(optimizer_call, stdout=dev_null, stderr=dev_null)
zip.write(os.path.join(tmpdir, fname), os.path.join('OPS', fname))
manifest.append(etree.fromstring(
'<item id="%s" href="%s" media-type="application/x-font-truetype" />' % (fname, fname)))
rmtree(tmpdir)
if cwd is not None:
os.chdir(cwd)
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:
meta = toc_file.makeelement(NCXNS('meta'))
meta.set('name', st)
meta.set('content', '0')
toc_file[0].append(meta)
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(('<text>', title, '</text>')))
# write TOC
if html_toc:
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,
xml_declaration=True, encoding="utf-8"))
zip.close()
return OutputFile.from_filename(output_file.name)
def transform_file(wldoc, chunk_counter=1, first=True, sample=None):
""" processes one input file and proceeds to its children """
replace_characters(wldoc.edoc.getroot())
hyphenator = set_hyph_language(wldoc.edoc.getroot()) if hyphenate else None
hyphenate_and_fix_conjunctions(wldoc.edoc.getroot(), hyphenator)
# every input file will have a TOC entry,
# pointing to starting chunk
toc = TOC(wldoc.book_info.title, "part%d.html" % chunk_counter)
chars = set()
if first:
# write book title page
html_tree = xslt(wldoc.edoc, get_resource('epub/xsltTitle.xsl'), outputtype=output_type)
chars = used_chars(html_tree.getroot())
zip.writestr(
'OPS/title.html',
etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"' +
' "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
)
)
# add a title page TOC entry
toc.add(u"Strona tytułowa", "title.html")
elif wldoc.book_info.parts:
# write title page for every parent
if sample is not None and sample <= 0:
chars = set()
html_string = open(get_resource('epub/emptyChunk.html')).read()
else:
html_tree = xslt(wldoc.edoc, get_resource('epub/xsltChunkTitle.xsl'))
chars = used_chars(html_tree.getroot())
html_string = etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"' +
' "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
)
zip.writestr('OPS/part%d.html' % chunk_counter, html_string)
add_to_manifest(manifest, chunk_counter)
add_to_spine(spine, chunk_counter)
chunk_counter += 1
if len(wldoc.edoc.getroot()) > 1:
# rdf before style master
main_text = wldoc.edoc.getroot()[1]
else:
# rdf in style master
main_text = wldoc.edoc.getroot()[0]
if main_text.tag == RDFNS('RDF'):
main_text = None
if main_text is not None:
for chunk_xml in chop(main_text):
empty = False
if sample is not None:
if sample <= 0:
empty = True
else:
sample -= len(chunk_xml.xpath('//strofa|//akap|//akap_cd|//akap_dialog'))
chunk_html, chunk_toc, chunk_chars = transform_chunk(chunk_xml, chunk_counter, annotations, empty)
toc.extend(chunk_toc)
chars = chars.union(chunk_chars)
zip.writestr('OPS/part%d.html' % chunk_counter, chunk_html)
add_to_manifest(manifest, chunk_counter)
add_to_spine(spine, chunk_counter)
chunk_counter += 1
for child in wldoc.parts():
child_toc, chunk_counter, chunk_chars, sample = transform_file(
child, chunk_counter, first=False, sample=sample)
toc.append(child_toc)
chars = chars.union(chunk_chars)
return toc, chunk_counter, chars, sample
def image(self):
metr = Metric(self, self.scale)
img = Image.new('RGB', (metr.width, metr.height), self.background_color)
draw = ImageDraw.Draw(img)
if self.epoch in self.epoch_colors:
epoch_color = self.epoch_colors[self.epoch]
else:
epoch_color = '#000'
draw.rectangle((0, 0, metr.bar_width, metr.height), fill=epoch_color)
if self.background_img:
src = Image.open(self.background_img)
trg_size = (metr.width - metr.bar_width, metr.height)
if src.size[0] * trg_size[1] < src.size[1] * trg_size[0]:
resized = (
trg_size[0],
src.size[1] * trg_size[0] / src.size[0]
)
cut = (resized[1] - trg_size[1]) / 2
src = src.resize(resized, Image.ANTIALIAS)
src = src.crop((0, cut, src.size[0], src.size[1] - cut))
else:
resized = (
src.size[0] * trg_size[1] / src.size[1],
trg_size[1],
)
cut = (resized[0] - trg_size[0]) / 2
src = src.resize(resized, Image.ANTIALIAS)
src = src.crop((cut, 0, src.size[0] - cut, src.size[1]))
img.paste(src, (metr.bar_width, 0))
del src
box = TextBox(metr.title_box_width, metr.height, padding_y=metr.box_padding_y)
author_font = ImageFont.truetype(
self.author_font_ttf, metr.author_font_size)
box.text(self.pretty_author(),
font=author_font,
line_height=metr.author_lineskip,
color=self.author_color,
shadow_color=self.author_shadow,
)
box.skip(metr.box_above_line)
box.draw.line((metr.box_line_left, box.height, metr.box_line_right, box.height),
fill=self.author_color, width=metr.box_line_width)
box.skip(metr.box_below_line)
title_font = ImageFont.truetype(
self.title_font_ttf, metr.title_font_size)
box.text(self.pretty_title(),
line_height=metr.title_lineskip,
font=title_font,
color=epoch_color,
shadow_color=self.title_shadow,
)
if self.with_logo:
logo = Image.open(get_resource('res/wl-logo-mono.png'))
logo = logo.resize((metr.logo_width, logo.size[1] * metr.logo_width / logo.size[0]), Image.ANTIALIAS)
alpha = logo.split()[3]
alpha = ImageEnhance.Brightness(alpha).enhance(.75)
logo.putalpha(alpha)
box.skip(metr.logo_top + logo.size[1])
box_img = box.image()
if self.kind == 'Liryka':
# top
box_top = metr.box_top_margin
elif self.kind == 'Epika':
# bottom
box_top = metr.height - metr.box_bottom_margin - box_img.size[1]
else:
# center
box_top = (metr.height - box_img.size[1]) / 2
box_left = metr.bar_width + (metr.width - metr.bar_width -
box_img.size[0]) / 2
draw.rectangle((box_left, box_top,
box_left + box_img.size[0], box_top + box_img.size[1]),
fill='#fff')
img.paste(box_img, (box_left, box_top), box_img)
if self.with_logo:
img.paste(logo,
(box_left + (box_img.size[0] - logo.size[0]) / 2,
box_top + box_img.size[1] - metr.box_padding_y - logo.size[1]), mask=logo)
return img
def html(self):
with open(get_resource('epub/toc.html')) as f:
t = unicode(f.read(), 'utf-8')
return t % self.html_part()
class Cover(object):
"""Abstract base class for cover images generator."""
width = 600
height = 800
background_color = '#fff'
background_img = None
author_top = 100
author_margin_left = 20
author_margin_right = 20
author_lineskip = 40
author_color = '#000'
author_shadow = None
author_font_ttf = get_resource('fonts/DejaVuSerif.ttf')
author_font_size = 30
title_top = 100
title_margin_left = 20
title_margin_right = 20
title_lineskip = 54
title_color = '#000'
title_shadow = None
title_font_ttf = get_resource('fonts/DejaVuSerif.ttf')
title_font_size = 40
logo_bottom = None
logo_width = None
uses_dc_cover = False
format = 'JPEG'
scale = 1
scale_after = 1
exts = {
'JPEG': 'jpg',
'PNG': 'png',
}
mime_types = {
'JPEG': 'image/jpeg',
'PNG': 'image/png',
}
def __init__(self, book_info, format=None, width=None, height=None):
self.author = ", ".join(auth.readable() for auth in book_info.authors)
self.title = book_info.title
if format is not None:
self.format = format
if width and height:
self.height = height * self.width / width
scale = max(
float(width or 0) / self.width,
float(height or 0) / self.height)
if scale >= 1:
self.scale = scale
elif scale:
self.scale_after = scale
def pretty_author(self):
"""Allows for decorating author's name."""
return self.author
def pretty_title(self):
"""Allows for decorating title."""
return self.title
def image(self):
metr = Metric(self, self.scale)
img = Image.new('RGB', (metr.width, metr.height),
self.background_color)
if self.background_img:
background = Image.open(self.background_img)
img.paste(background, None, background)
del background
# WL logo
if metr.logo_width:
logo = Image.open(get_resource('res/wl-logo.png'))
logo = logo.resize((metr.logo_width,
logo.size[1] * metr.logo_width / logo.size[0]))
img.paste(logo, ((metr.width - metr.logo_width) / 2,
img.size[1] - logo.size[1] - metr.logo_bottom))
top = metr.author_top
tbox = TextBox(
metr.width - metr.author_margin_left - metr.author_margin_right,
metr.height - top,
)
author_font = ImageFont.truetype(self.author_font_ttf,
metr.author_font_size)
tbox.text(self.pretty_author(), self.author_color, author_font,
metr.author_lineskip, self.author_shadow)
text_img = tbox.image()
img.paste(text_img, (metr.author_margin_left, top), text_img)
top += text_img.size[1] + metr.title_top
tbox = TextBox(
metr.width - metr.title_margin_left - metr.title_margin_right,
metr.height - top,
)
title_font = ImageFont.truetype(self.title_font_ttf,
metr.title_font_size)
tbox.text(self.pretty_title(), self.title_color, title_font,
metr.title_lineskip, self.title_shadow)
text_img = tbox.image()
img.paste(text_img, (metr.title_margin_left, top), text_img)
return img
def final_image(self):
img = self.image()
if self.scale_after != 1:
img = img.resize((int(round(img.size[0] * self.scale_after)),
int(round(img.size[1] * self.scale_after))),
Image.ANTIALIAS)
return img
def mime_type(self):
return self.mime_types[self.format]
def ext(self):
return self.exts[self.format]
def save(self, *args, **kwargs):
default_kwargs = {
'format': self.format,
'quality': 95,
}
default_kwargs.update(kwargs)
return self.final_image().save(*args, **default_kwargs)
def output_file(self, *args, **kwargs):
imgstr = StringIO()
self.save(imgstr, *args, **kwargs)
return OutputFile.from_string(imgstr.getvalue())
def transform_file(wldoc, chunk_counter=1, first=True, sample=None):
""" processes one input file and proceeds to its children """
replace_characters(wldoc.edoc.getroot())
hyphenator = set_hyph_language(wldoc.edoc.getroot())
hyphenate_and_fix_conjunctions(wldoc.edoc.getroot(), hyphenator)
# every input file will have a TOC entry,
# pointing to starting chunk
toc = TOC(wldoc.book_info.title, "part%d.html" % chunk_counter)
chars = set()
if first:
# write book title page
html_tree = xslt(wldoc.edoc, get_resource('epub/xsltTitle.xsl'))
chars = used_chars(html_tree.getroot())
zip.writestr(
'OPS/title.html',
etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"' +
' "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
)
)
# add a title page TOC entry
toc.add(u"Strona tytułowa", "title.html")
elif wldoc.book_info.parts:
# write title page for every parent
if sample is not None and sample <= 0:
chars = set()
html_string = open(get_resource('epub/emptyChunk.html')).read()
else:
html_tree = xslt(wldoc.edoc, get_resource('epub/xsltChunkTitle.xsl'))
chars = used_chars(html_tree.getroot())
html_string = etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"' +
' "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
)
zip.writestr('OPS/part%d.html' % chunk_counter, html_string)
add_to_manifest(manifest, chunk_counter)
add_to_spine(spine, chunk_counter)
chunk_counter += 1
if len(wldoc.edoc.getroot()) > 1:
# rdf before style master
main_text = wldoc.edoc.getroot()[1]
else:
# rdf in style master
main_text = wldoc.edoc.getroot()[0]
if main_text.tag == RDFNS('RDF'):
main_text = None
if main_text is not None:
for chunk_xml in chop(main_text):
empty = False
if sample is not None:
if sample <= 0:
empty = True
else:
sample -= len(chunk_xml.xpath('//strofa|//akap|//akap_cd|//akap_dialog'))
chunk_html, chunk_toc, chunk_chars = transform_chunk(chunk_xml, chunk_counter, annotations, empty)
toc.extend(chunk_toc)
chars = chars.union(chunk_chars)
zip.writestr('OPS/part%d.html' % chunk_counter, chunk_html)
add_to_manifest(manifest, chunk_counter)
add_to_spine(spine, chunk_counter)
chunk_counter += 1
for child in wldoc.parts():
child_toc, chunk_counter, chunk_chars, sample = transform_file(
child, chunk_counter, first=False, sample=sample)
toc.append(child_toc)
chars = chars.union(chunk_chars)
return toc, chunk_counter, chars, sample
def get_stylesheet(name):
return get_resource(STYLESHEETS[name])
def sponsor_logo(name):
return {
'Narodowe Centrum Kultury': get_resource('res/sponsors/nck.png')
}.get(name.strip())
class WLCover(Cover):
"""Wolne Lektury cover without logos."""
width = 600
height = 833
uses_dc_cover = True
author_font_ttf = get_resource('fonts/JunicodeWL-Regular.ttf')
author_font_size = 20
author_lineskip = 30
title_font_ttf = get_resource('fonts/DejaVuSerif-Bold.ttf')
title_font_size = 30
title_lineskip = 40
title_box_width = 350
box_top_margin = 100
box_bottom_margin = 100
box_padding_y = 20
box_above_line = 10
box_below_line = 15
box_line_left = 75
box_line_right = 275
box_line_width = 2
logo_top = 15
logo_width = 140
bar_width = 35
bar_color = '#000'
box_position = 'middle'
background_color = '#444'
author_color = '#444'
background_img = get_resource('res/cover.png')
format = 'JPEG'
epoch_colors = {
u'Starożytność': '#9e3610',
u'Średniowiecze': '#564c09',
u'Renesans': '#8ca629',
u'Barok': '#a6820a',
u'Oświecenie': '#f2802e',
u'Romantyzm': '#db4b16',
u'Pozytywizm': '#961060',
u'Modernizm': '#7784e0',
u'Dwudziestolecie międzywojenne': '#3044cf',
u'Współczesność': '#06393d',
}
kind_box_position = {
u'Liryka': 'top',
u'Epika': 'bottom',
}
def __init__(self,
book_info,
format=None,
width=None,
height=None,
bleed=0):
super(WLCover, self).__init__(book_info,
format=format,
width=width,
height=height)
# Set box position.
self.box_position = book_info.cover_box_position or \
self.kind_box_position.get(book_info.kind, self.box_position)
# Set bar color.
if book_info.cover_bar_color == 'none':
self.bar_width = 0
else:
self.bar_color = book_info.cover_bar_color or \
self.epoch_colors.get(book_info.epoch, self.bar_color)
# Set title color.
self.title_color = self.epoch_colors.get(book_info.epoch,
self.title_color)
self.bleed = bleed
self.box_top_margin += bleed
self.box_bottom_margin += bleed
self.bar_width += bleed
if book_info.cover_url:
url = book_info.cover_url
bg_src = None
if bg_src is None:
bg_src = URLOpener().open(url)
self.background_img = StringIO(bg_src.read())
bg_src.close()
def pretty_author(self):
return self.author.upper()
def add_box(self, img):
if self.box_position == 'none':
return img
metr = Metric(self, self.scale)
# Write author name.
box = TextBox(metr.title_box_width,
metr.height,
padding_y=metr.box_padding_y)
author_font = ImageFont.truetype(self.author_font_ttf,
metr.author_font_size)
box.text(self.pretty_author(),
font=author_font,
line_height=metr.author_lineskip,
color=self.author_color,
shadow_color=self.author_shadow)
box.skip(metr.box_above_line)
box.draw.line(
(metr.box_line_left, box.height, metr.box_line_right, box.height),
fill=self.author_color,
width=metr.box_line_width)
box.skip(metr.box_below_line)
# Write title.
title_font = ImageFont.truetype(self.title_font_ttf,
metr.title_font_size)
box.text(self.pretty_title(),
line_height=metr.title_lineskip,
font=title_font,
color=self.title_color,
shadow_color=self.title_shadow)
box_img = box.image()
# Find box position.
if self.box_position == 'top':
box_top = metr.box_top_margin
elif self.box_position == 'bottom':
box_top = metr.height - metr.box_bottom_margin - box_img.size[1]
else: # Middle.
box_top = (metr.height - box_img.size[1]) / 2
box_left = metr.bar_width + (metr.width - metr.bar_width -
box_img.size[0]) / 2
# Draw the white box.
ImageDraw.Draw(img).rectangle(
(box_left, box_top, box_left + box_img.size[0],
box_top + box_img.size[1]),
fill='#fff')
# Paste the contents into the white box.
img.paste(box_img, (box_left, box_top), box_img)
return img
def add_cut_lines(self, img):
line_ratio = 0.5
if self.bleed == 0:
return img
metr = Metric(self, self.scale)
draw = ImageDraw.Draw(img)
for corner_x, corner_y in ((0, 0), (metr.width, 0), (0, metr.height),
(metr.width, metr.height)):
dir_x = 1 if corner_x == 0 else -1
dir_y = 1 if corner_y == 0 else -1
for offset in (-1, 0, 1):
draw.line((corner_x, corner_y + dir_y * metr.bleed + offset,
corner_x + dir_x * metr.bleed * line_ratio,
corner_y + dir_y * metr.bleed + offset),
fill='black' if offset == 0 else 'white',
width=1)
draw.line((corner_x + dir_x * metr.bleed + offset, corner_y,
corner_x + dir_x * metr.bleed + offset,
corner_y + dir_y * metr.bleed * line_ratio),
fill='black' if offset == 0 else 'white',
width=1)
return img
def image(self):
metr = Metric(self, self.scale)
img = Image.new('RGB', (metr.width, metr.height),
self.background_color)
draw = ImageDraw.Draw(img)
draw.rectangle((0, 0, metr.bar_width, metr.height),
fill=self.bar_color)
if self.background_img:
src = Image.open(self.background_img)
trg_size = (metr.width - metr.bar_width, metr.height)
if src.size[0] * trg_size[1] < src.size[1] * trg_size[0]:
resized = (trg_size[0],
src.size[1] * trg_size[0] / src.size[0])
cut = (resized[1] - trg_size[1]) / 2
src = src.resize(resized, Image.ANTIALIAS)
src = src.crop((0, cut, src.size[0], src.size[1] - cut))
else:
resized = (
src.size[0] * trg_size[1] / src.size[1],
trg_size[1],
)
cut = (resized[0] - trg_size[0]) / 2
src = src.resize(resized, Image.ANTIALIAS)
src = src.crop((cut, 0, src.size[0] - cut, src.size[1]))
img.paste(src, (metr.bar_width, 0))
del src
img = self.add_box(img)
img = self.add_cut_lines(img)
return img
