def latex_extract_images(self, directory):
images = filter(lambda x: x.media_type.startswith('image'),
self.oeb.manifest)
if not images:
return {}
# gather image ID mappings for further references
reference = self.oeb.spine.items[0]
references = {}
with CurrentDir(directory):
# create output directory if needed
image_dir = self.latex_get_image_directory()
if not os.path.exists(image_dir):
os.makedirs(image_dir)
for image in images:
image_name = re.sub(r'[\\/]', r'_', image.id)
image_path = os.path.join(image_dir, image_name)
references[reference.relhref(image.href)] = image_path
with open(image_path, 'wb') as f:
f.write(image.data)
return references
def get_simple_book(fmt='epub'):
cache = get_cache()
ans = os.path.join(cache, 'simple.' + fmt)
src = os.path.join(os.path.dirname(__file__), 'simple.html')
if needs_recompile(ans, src):
with TemporaryDirectory('bpt') as tdir:
with CurrentDir(tdir):
raw = lopen(src, 'rb').read().decode('utf-8')
raw = add_resources(
raw, {
'LMONOI':
P('fonts/liberation/LiberationMono-Italic.ttf'),
'LMONOR':
P('fonts/liberation/LiberationMono-Regular.ttf'),
'IMAGE1': I('marked.png'),
'IMAGE2': I('textures/light_wood.png'),
})
shutil.copy2(I('lt.png'), '.')
x = 'index.html'
with lopen(x, 'wb') as f:
f.write(raw.encode('utf-8'))
build_book(x,
ans,
args=[
'--level1-toc=//h:h2', '--language=en',
'--authors=Kovid Goyal', '--cover=lt.png'
])
return ans
def render_html(path_to_html, width=590, height=750, as_xhtml=True):
from PyQt4.QtWebKit import QWebPage
from PyQt4.Qt import QEventLoop, QPalette, Qt, QUrl, QSize
from calibre.gui2 import is_ok_to_use_qt
if not is_ok_to_use_qt():
return None
path_to_html = os.path.abspath(path_to_html)
with CurrentDir(os.path.dirname(path_to_html)):
page = QWebPage()
pal = page.palette()
pal.setBrush(QPalette.Background, Qt.white)
page.setPalette(pal)
page.setViewportSize(QSize(width, height))
page.mainFrame().setScrollBarPolicy(Qt.Vertical, Qt.ScrollBarAlwaysOff)
page.mainFrame().setScrollBarPolicy(Qt.Horizontal, Qt.ScrollBarAlwaysOff)
loop = QEventLoop()
renderer = HTMLRenderer(page, loop)
page.loadFinished.connect(renderer, type=Qt.QueuedConnection)
if as_xhtml:
page.mainFrame().setContent(open(path_to_html, 'rb').read(),
'application/xhtml+xml', QUrl.fromLocalFile(path_to_html))
else:
page.mainFrame().load(QUrl.fromLocalFile(path_to_html))
loop.exec_()
renderer.loop = renderer.page = None
page.loadFinished.disconnect()
del page
del loop
if isinstance(renderer.exception, ParserError) and as_xhtml:
return render_html(path_to_html, width=width, height=height,
as_xhtml=False)
return renderer
def test_dir_container(self):
def create_book(source):
with ZipFile(P('quick_start/eng.epub',
allow_user_override=False)) as zf:
zf.extractall(source)
with CurrentDir(source):
self.assertTrue(os.path.exists('images/cover.jpg'))
with open('.gitignore', 'wb') as f:
f.write(b'nothing')
os.mkdir('.git')
with open('.git/xxx', 'wb') as f:
f.write(b'xxx')
with TemporaryDirectory('-polish-dir-container') as source:
create_book(source)
c = get_container(source)
c.remove_item('images/cover.jpg')
with c.open('images/test-container.xyz', 'wb') as f:
f.write(b'xyz')
c.commit()
with CurrentDir(source):
self.assertTrue(os.path.exists('.gitignore'))
self.assertTrue(os.path.exists('.git/xxx'))
self.assertTrue(os.path.exists('images/test-container.xyz'))
self.assertFalse(os.path.exists('images/cover.jpg'))
def create_book(mi, path, fmt='epub', opf_name='metadata.opf', html_name='start.xhtml', toc_name='toc.ncx'):
''' Create an empty book in the specified format at the specified location. '''
path = os.path.abspath(path)
lang = 'und'
opf = metadata_to_opf(mi, as_string=False)
for l in opf.xpath('//*[local-name()="language"]'):
if l.text:
lang = l.text
break
lang = lang_as_iso639_1(lang) or lang
opfns = OPF_NAMESPACES['opf']
m = opf.makeelement('{%s}manifest' % opfns)
opf.insert(1, m)
i = m.makeelement('{%s}item' % opfns, href=html_name, id='start')
i.set('media-type', guess_type('a.xhtml'))
m.append(i)
i = m.makeelement('{%s}item' % opfns, href=toc_name, id='ncx')
i.set('media-type', guess_type(toc_name))
m.append(i)
s = opf.makeelement('{%s}spine' % opfns, toc="ncx")
opf.insert(2, s)
i = s.makeelement('{%s}itemref' % opfns, idref='start')
s.append(i)
CONTAINER = '''\
<?xml version="1.0"?>
<container version="1.0" xmlns="urn:oasis:names:tc:opendocument:xmlns:container">
<rootfiles>
<rootfile full-path="{0}" media-type="application/oebps-package+xml"/>
</rootfiles>
</container>
'''.format(prepare_string_for_xml(opf_name, True)).encode('utf-8')
HTML = P('templates/new_book.html', data=True).decode('utf-8').replace(
'_LANGUAGE_', prepare_string_for_xml(lang, True)
).replace(
'_TITLE_', prepare_string_for_xml(mi.title)
).replace(
'_AUTHORS_', prepare_string_for_xml(authors_to_string(mi.authors))
).encode('utf-8')
h = parse(HTML)
pretty_html_tree(None, h)
HTML = serialize(h, 'text/html')
ncx = etree.tostring(create_toc(mi, opf, html_name, lang), encoding='utf-8', xml_declaration=True, pretty_print=True)
pretty_xml_tree(opf)
opf = etree.tostring(opf, encoding='utf-8', xml_declaration=True, pretty_print=True)
if fmt == 'azw3':
with TemporaryDirectory('create-azw3') as tdir, CurrentDir(tdir):
for name, data in ((opf_name, opf), (html_name, HTML), (toc_name, ncx)):
with open(name, 'wb') as f:
f.write(data)
c = Container(os.path.dirname(os.path.abspath(opf_name)), opf_name, DevNull())
opf_to_azw3(opf_name, path, c)
else:
with ZipFile(path, 'w', compression=ZIP_STORED) as zf:
zf.writestr('mimetype', b'application/epub+zip', compression=ZIP_STORED)
zf.writestr('META-INF/', b'', 0755)
zf.writestr('META-INF/container.xml', CONTAINER)
zf.writestr(opf_name, opf)
zf.writestr(html_name, HTML)
zf.writestr(toc_name, ncx)
def get_metadata(stream):
from calibre.ebooks.metadata.archive import is_comic
from calibre.ebooks.metadata.meta import get_metadata
path = getattr(stream, 'name', False)
if not path:
pt = PersistentTemporaryFile('_rar-meta.rar')
pt.write(stream.read())
pt.close()
path = pt.name
path = os.path.abspath(path)
file_names = list(names(path))
if is_comic(file_names):
return get_metadata(stream, 'cbr')
for f in file_names:
stream_type = os.path.splitext(f)[1].lower()
if stream_type:
stream_type = stream_type[1:]
if stream_type in ('lit', 'opf', 'prc', 'mobi', 'fb2', 'epub',
'rb', 'imp', 'pdf', 'lrf', 'azw'):
with TemporaryDirectory() as tdir:
with CurrentDir(tdir):
stream = extract_member(path,
match=None,
name=f,
as_file=True)[1]
return get_metadata(stream, stream_type)
raise ValueError('No ebook found in RAR archive')
def test(): # {{{
from calibre.ptempfile import TemporaryDirectory
from calibre import CurrentDir
from glob import glob
img = image_from_data(I('lt.png', data=True, allow_user_override=False))
with TemporaryDirectory() as tdir, CurrentDir(tdir):
save_image(img, 'test.jpg')
ret = optimize_jpeg('test.jpg')
if ret is not None:
raise SystemExit('optimize_jpeg failed: %s' % ret)
ret = encode_jpeg('test.jpg')
if ret is not None:
raise SystemExit('encode_jpeg failed: %s' % ret)
shutil.copyfile(I('lt.png'), 'test.png')
ret = optimize_png('test.png')
if ret is not None:
raise SystemExit('optimize_png failed: %s' % ret)
if glob('*.bak'):
raise SystemExit('Spurious .bak files left behind')
quantize_image(img)
oil_paint_image(img)
gaussian_sharpen_image(img)
gaussian_blur_image(img)
despeckle_image(img)
remove_borders_from_image(img)
image_to_data(img, fmt='GIF')
def get_metadata(stream):
from calibre.ebooks.metadata import MetaInformation
from calibre.ptempfile import TemporaryDirectory
from calibre.ebooks.mobi.reader.headers import MetadataHeader
from calibre.ebooks.mobi.reader.mobi6 import MobiReader
from calibre.utils.img import save_cover_data_to
from calibre import CurrentDir
stream.seek(0)
try:
raw = stream.read(3)
except Exception:
raw = b''
stream.seek(0)
if raw == b'TPZ':
from calibre.ebooks.metadata.topaz import get_metadata
return get_metadata(stream)
from calibre.utils.logging import Log
log = Log()
try:
mi = MetaInformation(os.path.basename(stream.name), [_('Unknown')])
except:
mi = MetaInformation(_('Unknown'), [_('Unknown')])
mh = MetadataHeader(stream, log)
if mh.title and mh.title != _('Unknown'):
mi.title = mh.title
if mh.exth is not None:
if mh.exth.mi is not None:
mi = mh.exth.mi
else:
size = 1024**3
if hasattr(stream, 'seek') and hasattr(stream, 'tell'):
pos = stream.tell()
stream.seek(0, 2)
size = stream.tell()
stream.seek(pos)
if size < 4 * 1024 * 1024:
with TemporaryDirectory('_mobi_meta_reader') as tdir:
with CurrentDir(tdir):
mr = MobiReader(stream, log)
parse_cache = {}
mr.extract_content(tdir, parse_cache)
if mr.embedded_mi is not None:
mi = mr.embedded_mi
if hasattr(mh.exth, 'cover_offset'):
cover_index = mh.first_image_index + mh.exth.cover_offset
data = mh.section_data(int(cover_index))
else:
try:
data = mh.section_data(mh.first_image_index)
except Exception:
data = b''
if data and what(None,
data) in {'jpg', 'jpeg', 'gif', 'png', 'bmp', 'webp'}:
try:
mi.cover_data = ('jpg', save_cover_data_to(data))
except Exception:
log.exception('Failed to read MOBI cover')
return mi
def convert(self, oeb_book, output_path, input_plugin, opts, log):
from urllib import unquote
from lxml import etree
self.log, self.opts = log, opts
if not os.path.exists(output_path):
os.makedirs(output_path)
from calibre.ebooks.oeb.base import OPF_MIME, NCX_MIME, PAGE_MAP_MIME
with CurrentDir(output_path):
results = oeb_book.to_opf2(page_map=True)
for key in (OPF_MIME, NCX_MIME, PAGE_MAP_MIME):
href, root = results.pop(key, [None, None])
if root is not None:
raw = etree.tostring(root, pretty_print=True,
encoding='utf-8', xml_declaration=True)
if key == OPF_MIME:
# Needed as I can't get lxml to output opf:role and
# not output <opf:metadata> as well
raw = re.sub(r'(<[/]{0,1})opf:', r'\1', raw)
with open(href, 'wb') as f:
f.write(raw)
for item in oeb_book.manifest:
path = os.path.abspath(unquote(item.href))
dir = os.path.dirname(path)
if not os.path.exists(dir):
os.makedirs(dir)
with open(path, 'wb') as f:
f.write(str(item))
item.unload_data_from_memory(memory=path)
def get_metadata(stream):
from calibre.ebooks.metadata.meta import get_metadata
from calibre.ebooks.metadata.archive import is_comic
stream_type = None
zf = ZipFile(stream, 'r')
names = zf.namelist()
if is_comic(names):
# Is probably a comic
return get_metadata(stream, 'cbz')
for f in names:
stream_type = os.path.splitext(f)[1].lower()
if stream_type:
stream_type = stream_type[1:]
if stream_type in ('lit', 'opf', 'prc', 'mobi', 'fb2', 'epub',
'rb', 'imp', 'pdf', 'lrf', 'azw', 'azw1', 'azw3'):
with TemporaryDirectory() as tdir:
with CurrentDir(tdir):
path = zf.extract(f)
mi = get_metadata(open(path,'rb'), stream_type)
if stream_type == 'opf' and mi.application_id is None:
try:
# zip archive opf files without an application_id were assumed not to have a cover
# reparse the opf and if cover exists read its data from zip archive for the metadata
nmi = zip_opf_metadata(path, zf)
nmi.timestamp = None
return nmi
except:
pass
mi.timestamp = None
return mi
raise ValueError('No ebook found in ZIP archive (%s)' % os.path.basename(getattr(stream, 'name', '') or '<stream>'))
def extract_raster_image(wmf_data):
try:
wmf, wmf_err = plugins['wmf']
except KeyError:
raise Unavailable('libwmf not available on this platform')
if wmf_err:
raise Unavailable(wmf_err)
if iswindows:
import sys, os
appdir = sys.app_dir
if isinstance(appdir, unicode):
appdir = appdir.encode(filesystem_encoding)
fdir = os.path.join(appdir, 'wmffonts')
wmf.set_font_dir(fdir)
data = ''
with TemporaryDirectory('wmf2png') as tdir:
with CurrentDir(tdir):
wmf.render(wmf_data)
images = list(sorted(glob.glob('*.png')))
if not images:
raise NoRaster('No raster images in WMF')
data = open(images[0], 'rb').read()
im = Image()
im.load(data)
pw = PixelWand()
pw.color = '#ffffff'
im.rotate(pw, 180)
return im.export('png')
def test_lopen():
from calibre.ptempfile import TemporaryDirectory
from calibre import CurrentDir
n = u'f\xe4llen'
with TemporaryDirectory() as tdir:
with CurrentDir(tdir):
with lopen(n, 'w') as f:
f.write('one')
print 'O_CREAT tested'
with lopen(n, 'w+b') as f:
f.write('two')
with lopen(n, 'r') as f:
if f.read() == 'two':
print 'O_TRUNC tested'
else:
raise Exception('O_TRUNC failed')
with lopen(n, 'ab') as f:
f.write('three')
with lopen(n, 'r+') as f:
if f.read() == 'twothree':
print 'O_APPEND tested'
else:
raise Exception('O_APPEND failed')
with lopen(n, 'r+') as f:
f.seek(3)
f.write('xxxxx')
f.seek(0)
if f.read() == 'twoxxxxx':
print 'O_RANDOM tested'
else:
raise Exception('O_RANDOM failed')
def get_comics_from_collection(self, stream):
from calibre.libunzip import extract as zipextract
tdir = PersistentTemporaryDirectory('_comic_collection')
zipextract(stream, tdir)
comics = []
with CurrentDir(tdir):
if not os.path.exists('comics.txt'):
raise ValueError(
('%s is not a valid comic collection'
' no comics.txt was found in the file') % stream.name)
with open('comics.txt', 'rb') as f:
raw = f.read()
if raw.startswith(codecs.BOM_UTF16_BE):
raw = raw.decode('utf-16-be')[1:]
elif raw.startswith(codecs.BOM_UTF16_LE):
raw = raw.decode('utf-16-le')[1:]
elif raw.startswith(codecs.BOM_UTF8):
raw = raw.decode('utf-8')[1:]
else:
raw = raw.decode('utf-8')
for line in raw.splitlines():
line = line.strip()
if not line:
continue
fname, title = line.partition(':')[0], line.partition(':')[-1]
fname = fname.replace('#', '_')
fname = os.path.join(tdir, *fname.split('/'))
if not title:
title = os.path.basename(fname).rpartition('.')[0]
if os.access(fname, os.R_OK):
comics.append([title, fname])
if not comics:
raise ValueError('%s has no comics' % stream.name)
return comics
def generate_toc(self, oeb_book, ref_url, output_dir):
'''
Generate table of contents
'''
from lxml import etree
from urllib import unquote
from calibre.ebooks.oeb.base import element
from calibre.utils.cleantext import clean_xml_chars
with CurrentDir(output_dir):
def build_node(current_node, parent=None):
if parent is None:
parent = etree.Element('ul')
elif len(current_node.nodes):
parent = element(parent, ('ul'))
for node in current_node.nodes:
point = element(parent, 'li')
href = relpath(abspath(unquote(node.href)),
dirname(ref_url))
if isinstance(href, bytes):
href = href.decode('utf-8')
link = element(point, 'a', href=clean_xml_chars(href))
title = node.title
if isinstance(title, bytes):
title = title.decode('utf-8')
if title:
title = re.sub(r'\s+', ' ', title)
link.text = clean_xml_chars(title)
build_node(node, point)
return parent
wrap = etree.Element('div')
wrap.append(build_node(oeb_book.toc))
return wrap
def test(): # {{{
from calibre.ptempfile import TemporaryDirectory
from calibre import CurrentDir
from glob import glob
img = image_from_data(I('lt.png', data=True, allow_user_override=False))
with TemporaryDirectory() as tdir, CurrentDir(tdir):
save_image(img, 'test.jpg')
ret = optimize_jpeg('test.jpg')
if ret is not None:
raise SystemExit('optimize_jpeg failed: %s' % ret)
ret = encode_jpeg('test.jpg')
if ret is not None:
raise SystemExit('encode_jpeg failed: %s' % ret)
shutil.copyfile(I('lt.png'), 'test.png')
ret = optimize_png('test.png')
if ret is not None:
raise SystemExit('optimize_png failed: %s' % ret)
if glob('*.bak'):
raise SystemExit('Spurious .bak files left behind')
quantize_image(img)
oil_paint_image(img)
gaussian_sharpen_image(img)
gaussian_blur_image(img)
despeckle_image(img)
remove_borders_from_image(img)
image_to_data(img, fmt='GIF')
raw = subprocess.Popen([get_exe_path('JxrDecApp'), '-h'], creationflags=0x08 if iswindows else 0, stdout=subprocess.PIPE).stdout.read()
if b'JPEG XR Decoder Utility' not in raw:
raise SystemExit('Failed to run JxrDecApp')
def generate_toc(self, oeb_book, ref_url, output_dir):
'''
Generate table of contents
'''
from lxml import etree
from urllib.parse import unquote
from calibre.ebooks.oeb.base import element
with CurrentDir(output_dir):
def build_node(current_node, parent=None):
if parent is None:
parent = etree.Element('ul')
elif len(current_node.nodes):
parent = element(parent, ('ul'))
for node in current_node.nodes:
point = element(parent, 'li')
href = relpath(abspath(unquote(node.href)),
dirname(ref_url))
link = element(point, 'a', href=href)
title = node.title
if title:
title = re.sub(r'\s+', ' ', title)
link.text = title
build_node(node, point)
return parent
wrap = etree.Element('div')
wrap.append(build_node(oeb_book.toc))
return wrap
def freeze_locales(self):
zf = self.DEST + '.zip'
from calibre import CurrentDir
from calibre.utils.zipfile import ZipFile, ZIP_DEFLATED
with ZipFile(zf, 'w', ZIP_DEFLATED) as zf:
with CurrentDir(self.DEST):
zf.add_dir('.')
shutil.rmtree(self.DEST)
def create_book(source):
with ZipFile(P('quick_start/eng.epub', allow_user_override=False)) as zf:
zf.extractall(source)
with CurrentDir(source):
self.assertTrue(os.path.exists('images/cover.jpg'))
with open('.gitignore', 'wb') as f:
f.write(b'nothing')
os.mkdir('.git')
with open('.git/xxx', 'wb') as f:
f.write(b'xxx')
def extract_content(self, output_dir):
from calibre.ebooks.pml.pmlconverter import pml_to_html
output_dir = os.path.abspath(output_dir)
if not os.path.exists(output_dir):
os.makedirs(output_dir)
pml = ''
for i in range(1, self.header_record.num_text_pages + 1):
self.log.debug('Extracting text page %i' % i)
pml += self.get_text_page(i)
title = self.mi.title
if not isinstance(title, unicode_type):
title = title.decode('utf-8', 'replace')
html = '<html><head><title>%s</title></head><body>%s</body></html>' % \
(title, pml_to_html(pml))
with CurrentDir(output_dir):
with open('index.html', 'wb') as index:
self.log.debug('Writing text to index.html')
index.write(html.encode('utf-8'))
if not os.path.exists(os.path.join(output_dir, 'images/')):
os.makedirs(os.path.join(output_dir, 'images/'))
images = []
with CurrentDir(os.path.join(output_dir, 'images/')):
for i in range(self.header_record.non_text_offset,
len(self.sections)):
name, img = self.get_image(i)
if name:
name = as_unicode(name)
images.append(name)
with open(name, 'wb') as imgf:
self.log.debug('Writing image %s to images/' % name)
imgf.write(img)
opf_path = self.create_opf(output_dir, images)
return opf_path
def run(epub, opts, log):
with TemporaryDirectory('_epub-fix') as tdir:
with CurrentDir(tdir):
zf = ZipFile(epub)
zf.extractall()
zf.close()
container = Container(tdir, log)
for fixer in epub_fixers():
fix = getattr(opts, fixer.fix_name, False)
fixer.run(container, opts, log, fix=fix)
container.write(epub)
def dump(self, ddir):
with open(os.path.join(ddir, self.name + '.html'), 'wb') as f:
f.write(self.text)
base = os.path.join(ddir, self.name + '-parts')
os.mkdir(base)
with CurrentDir(base):
with open('skeleton.html', 'wb') as f:
f.write(self.skeleton)
for i, text in enumerate(self.sections):
with open('sect-%04d.html' % i, 'wb') as f:
f.write(text)
def do_explode(path, dest):
from calibre.ebooks.mobi.reader.mobi6 import MobiReader
from calibre.ebooks.mobi.reader.mobi8 import Mobi8Reader
with open(path, 'rb') as stream:
mr = MobiReader(stream, default_log, None, None)
with CurrentDir(dest):
mr = Mobi8Reader(mr, default_log, for_tweak=True)
opf = os.path.abspath(mr())
obfuscated_fonts = mr.encrypted_fonts
return opf, obfuscated_fonts
def convert(self, oeb_book, output_path, input_plugin, opts, log):
from polyglot.urllib import unquote
from lxml import etree
self.log, self.opts = log, opts
if not os.path.exists(output_path):
os.makedirs(output_path)
from calibre.ebooks.oeb.base import OPF_MIME, NCX_MIME, PAGE_MAP_MIME, OEB_STYLES
from calibre.ebooks.oeb.normalize_css import condense_sheet
with CurrentDir(output_path):
results = oeb_book.to_opf2(page_map=True)
for key in (OPF_MIME, NCX_MIME, PAGE_MAP_MIME):
href, root = results.pop(key, [None, None])
if root is not None:
if key == OPF_MIME:
try:
self.workaround_nook_cover_bug(root)
except:
self.log.exception(
'Something went wrong while trying to'
' workaround Nook cover bug, ignoring')
try:
self.workaround_pocketbook_cover_bug(root)
except:
self.log.exception(
'Something went wrong while trying to'
' workaround Pocketbook cover bug, ignoring')
self.migrate_lang_code(root)
self.adjust_mime_types(root)
raw = etree.tostring(root,
pretty_print=True,
encoding='utf-8',
xml_declaration=True)
if key == OPF_MIME:
# Needed as I can't get lxml to output opf:role and
# not output <opf:metadata> as well
raw = re.sub(br'(<[/]{0,1})opf:', br'\1', raw)
with lopen(href, 'wb') as f:
f.write(raw)
for item in oeb_book.manifest:
if (not self.opts.expand_css and item.media_type in OEB_STYLES
and hasattr(item.data, 'cssText')
and 'nook' not in self.opts.output_profile.short_name):
condense_sheet(item.data)
path = os.path.abspath(unquote(item.href))
dir = os.path.dirname(path)
if not os.path.exists(dir):
os.makedirs(dir)
with lopen(path, 'wb') as f:
f.write(item.bytes_representation)
item.unload_data_from_memory(memory=path)
def render_cover(opf, opf_path, zf, reader=None):
from calibre.ebooks import render_html_svg_workaround
from calibre.utils.logging import default_log
cpage = opf.first_spine_item()
if not cpage:
return
if reader is not None and reader.encryption_meta.is_encrypted(cpage):
return
with TemporaryDirectory('_epub_meta') as tdir:
with CurrentDir(tdir):
zf.extractall()
opf_path = opf_path.replace('/', os.sep)
cpage = os.path.join(tdir, os.path.dirname(opf_path), cpage)
if not os.path.exists(cpage):
return
if isosx:
# On OS X trying to render a HTML cover which uses embedded
# fonts more than once in the same process causes a crash in Qt
# so be safe and remove the fonts as well as any @font-face
# rules
for f in walk('.'):
if os.path.splitext(f)[1].lower() in ('.ttf', '.otf'):
os.remove(f)
ffpat = re.compile(br'@font-face.*?{.*?}',
re.DOTALL | re.IGNORECASE)
with open(cpage, 'r+b') as f:
raw = f.read()
f.truncate(0)
f.seek(0)
raw = ffpat.sub(b'', raw)
f.write(raw)
from calibre.ebooks.chardet import xml_to_unicode
raw = xml_to_unicode(raw,
strip_encoding_pats=True,
resolve_entities=True)[0]
from lxml import html
for link in html.fromstring(raw).xpath('//link'):
href = link.get('href', '')
if href:
path = os.path.join(os.path.dirname(cpage), href)
if os.path.exists(path):
with open(path, 'r+b') as f:
raw = f.read()
f.truncate(0)
f.seek(0)
raw = ffpat.sub(b'', raw)
f.write(raw)
return render_html_svg_workaround(cpage, default_log)
def test_lopen():
from calibre.ptempfile import TemporaryDirectory
from calibre import CurrentDir
n = 'f\xe4llen'
print('testing lopen()')
if iswindows:
import msvcrt, win32api
def assert_not_inheritable(f):
if win32api.GetHandleInformation(msvcrt.get_osfhandle(
f.fileno())) & 0b1:
raise SystemExit('File handle is inheritable!')
else:
import fcntl
def assert_not_inheritable(f):
if not fcntl.fcntl(f, fcntl.F_GETFD) & fcntl.FD_CLOEXEC:
raise SystemExit('File handle is inheritable!')
def copen(*args):
ans = lopen(*args)
assert_not_inheritable(ans)
return ans
with TemporaryDirectory() as tdir, CurrentDir(tdir):
with copen(n, 'w') as f:
f.write('one')
print('O_CREAT tested')
with copen(n, 'w+b') as f:
f.write(b'two')
with copen(n, 'r') as f:
if f.read() == 'two':
print('O_TRUNC tested')
else:
raise Exception('O_TRUNC failed')
with copen(n, 'ab') as f:
f.write(b'three')
with copen(n, 'r+') as f:
if f.read() == 'twothree':
print('O_APPEND tested')
else:
raise Exception('O_APPEND failed')
with copen(n, 'r+') as f:
f.seek(3)
f.write('xxxxx')
f.seek(0)
if f.read() == 'twoxxxxx':
print('O_RANDOM tested')
else:
raise Exception('O_RANDOM failed')
def do_explode(path, dest):
with open(path, 'rb') as stream:
mr = MobiReader(stream, default_log, None, None)
with CurrentDir(dest):
mr = Mobi8Reader(mr, default_log)
opf = os.path.abspath(mr())
try:
os.remove('debug-raw.html')
except:
pass
return opf
def dump_images(self, output_dir):
'''
This is primarily used for debugging and 3rd party tools to
get the images in the file.
'''
if not os.path.exists(output_dir):
os.makedirs(output_dir)
with CurrentDir(output_dir):
for i in range(0, self.header_record.num_image_pages):
name, img = self.get_image(self.header_record.image_data_offset + i)
with open(name, 'wb') as imgf:
imgf.write(img)
def create_opf(self, output_dir, pages, images):
with CurrentDir(output_dir):
opf = OPFCreator(output_dir, self.mi)
manifest = []
for page in pages + images:
manifest.append((page, None))
opf.create_manifest(manifest)
opf.create_spine(pages)
with open('metadata.opf', 'wb') as opffile:
opf.render(opffile)
return os.path.join(output_dir, 'metadata.opf')
def create_opf(self, output_dir, images):
with CurrentDir(output_dir):
opf = OPFCreator(output_dir, self.mi)
manifest = [('index.html', None)]
for i in images:
manifest.append((os.path.join('images/', i), None))
opf.create_manifest(manifest)
opf.create_spine(['index.html'])
with open('metadata.opf', 'wb') as opffile:
opf.render(opffile)
return os.path.join(output_dir, 'metadata.opf')
def encrypt_fonts(self, uris, tdir, uuid): # {{{
from binascii import unhexlify
key = re.sub(r'[^a-fA-F0-9]', '', uuid)
if len(key) < 16:
raise ValueError('UUID identifier %r is invalid' % uuid)
key = unhexlify((key + key)[:32])
key = tuple(map(ord, key))
paths = []
with CurrentDir(tdir):
paths = [os.path.join(*x.split('/')) for x in uris]
uris = dict(list(zip(uris, paths)))
fonts = []
for uri in list(uris.keys()):
path = uris[uri]
if isinstance(path, str):
path = path.encode(filesystem_encoding)
if not os.path.exists(path):
uris.pop(uri)
continue
self.log.debug('Encrypting font:', uri)
with open(path, 'r+b') as f:
data = f.read(1024)
if len(data) >= 1024:
f.seek(0)
for i in range(1024):
f.write(chr(ord(data[i]) ^ key[i % 16]))
else:
self.log.warn('Font', path, 'is invalid, ignoring')
if not isinstance(uri, str):
uri = uri.decode('utf-8')
fonts.append('''
<enc:EncryptedData>
<enc:EncryptionMethod Algorithm="http://ns.adobe.com/pdf/enc#RC"/>
<enc:CipherData>
<enc:CipherReference URI="%s"/>
</enc:CipherData>
</enc:EncryptedData>
''' % (uri.replace('"', '\\"')))
if fonts:
ans = '''<encryption
xmlns="urn:oasis:names:tc:opendocument:xmlns:container"
xmlns:enc="http://www.w3.org/2001/04/xmlenc#"
xmlns:deenc="http://ns.adobe.com/digitaleditions/enc">
'''
ans += ('\n'.join(fonts)).encode('utf-8')
ans += '\n</encryption>'
return ans