def extract_images(self, processed_records, output_dir):
self.log.debug('Extracting images...')
output_dir = os.path.abspath(os.path.join(output_dir, 'images'))
if not os.path.exists(output_dir):
os.makedirs(output_dir)
image_index = 0
self.image_names = []
start = getattr(self.book_header, 'first_image_index', -1)
if start > self.num_sections or start < 0:
# BAEN PRC files have bad headers
start = 0
for i in range(start, self.num_sections):
if i in processed_records:
continue
processed_records.append(i)
data = self.sections[i][0]
image_index += 1
if data[:4] in {b'FLIS', b'FCIS', b'SRCS', b'\xe9\x8e\r\n',
b'RESC', b'BOUN', b'FDST', b'DATP', b'AUDI', b'VIDE'}:
# This record is a known non image type, not need to try to
# load the image
continue
path = os.path.join(output_dir, '%05d.jpg' % image_index)
try:
if what(None, data) not in {'jpg', 'jpeg', 'gif', 'png', 'bmp'}:
continue
save_cover_data_to(data, path, minify_to=(10000, 10000))
except Exception:
continue
self.image_names.append(os.path.basename(path))
def extract_images(self, processed_records, output_dir):
self.log.debug('Extracting images...')
output_dir = os.path.abspath(os.path.join(output_dir, 'images'))
if not os.path.exists(output_dir):
os.makedirs(output_dir)
image_index = 0
self.image_names = []
start = getattr(self.book_header, 'first_image_index', -1)
if start > self.num_sections or start < 0:
# BAEN PRC files have bad headers
start = 0
for i in range(start, self.num_sections):
if i in processed_records:
continue
processed_records.append(i)
data = self.sections[i][0]
image_index += 1
if data[:4] in {b'FLIS', b'FCIS', b'SRCS', b'\xe9\x8e\r\n',
b'RESC', b'BOUN', b'FDST', b'DATP', b'AUDI', b'VIDE'}:
# This record is a known non image type, not need to try to
# load the image
continue
path = os.path.join(output_dir, '%05d.jpg' % image_index)
try:
if what(None, data) not in {'jpg', 'jpeg', 'gif', 'png', 'bmp'}:
continue
save_cover_data_to(data, path, minify_to=(10000, 10000))
except Exception:
continue
self.image_names.append(os.path.basename(path))
def main(args=sys.argv):
parser = option_parser()
opts, args = parser.parse_args(args)
buf = BytesIO()
log = create_log(buf)
abort = Event()
patch_plugins()
authors = []
if opts.authors:
authors = string_to_authors(opts.authors)
identifiers = {}
if opts.isbn:
identifiers['isbn'] = opts.isbn
allowed_plugins = frozenset(opts.allowed_plugin)
results = identify(log,
abort,
title=opts.title,
authors=authors,
identifiers=identifiers,
timeout=int(opts.timeout),
allowed_plugins=allowed_plugins or None)
if not results:
print(log, file=sys.stderr)
prints('No results found', file=sys.stderr)
raise SystemExit(1)
result = results[0]
cf = None
if opts.cover and results:
cover = download_cover(log,
title=opts.title,
authors=authors,
identifiers=result.identifiers,
timeout=int(opts.timeout))
if cover is None and not opts.opf:
prints('No cover found', file=sys.stderr)
else:
save_cover_data_to(cover[-1], opts.cover)
result.cover = cf = opts.cover
log = buf.getvalue()
result = (metadata_to_opf(result)
if opts.opf else unicode(result).encode('utf-8'))
if opts.verbose:
print(log, file=sys.stderr)
print(result)
if not opts.opf and opts.cover:
prints('Cover :', cf)
return 0
def main(args=sys.argv):
parser = option_parser()
opts, args = parser.parse_args(args)
buf = BytesIO()
log = create_log(buf)
abort = Event()
patch_plugins()
authors = []
if opts.authors:
authors = string_to_authors(opts.authors)
identifiers = {}
for idspec in opts.identifier:
k, v = idspec.partition(':')[::2]
if not k or not v:
raise SystemExit('Not a valid identifier: {}'.format(idspec))
identifiers[k] = v
if opts.isbn:
identifiers['isbn'] = opts.isbn
allowed_plugins = frozenset(opts.allowed_plugin)
results = identify(log, abort, title=opts.title, authors=authors,
identifiers=identifiers, timeout=int(opts.timeout),
allowed_plugins=allowed_plugins or None)
if not results:
print(log, file=sys.stderr)
prints('No results found', file=sys.stderr)
raise SystemExit(1)
result = results[0]
cf = None
if opts.cover and results:
cover = download_cover(log, title=opts.title, authors=authors,
identifiers=result.identifiers, timeout=int(opts.timeout))
if cover is None and not opts.opf:
prints('No cover found', file=sys.stderr)
else:
save_cover_data_to(cover[-1], opts.cover)
result.cover = cf = opts.cover
log = buf.getvalue()
result = (metadata_to_opf(result) if opts.opf else
unicode_type(result).encode('utf-8'))
if opts.verbose:
print(log, file=sys.stderr)
print(result)
if not opts.opf and opts.cover:
prints('Cover :', cf)
return 0
def main(args=sys.argv):
parser = option_parser()
opts, args = parser.parse_args(args)
buf = BytesIO()
log = create_log(buf)
abort = Event()
patch_plugins()
authors = []
if opts.authors:
authors = string_to_authors(opts.authors)
identifiers = {}
for idspec in opts.identifier:
k, v = idspec.partition(':')[::2]
if not k or not v:
raise SystemExit('Not a valid identifier: {}'.format(idspec))
identifiers[k] = v
if opts.isbn:
identifiers['isbn'] = opts.isbn
allowed_plugins = frozenset(opts.allowed_plugin)
results = identify(log, abort, title=opts.title, authors=authors,
identifiers=identifiers, timeout=int(opts.timeout),
allowed_plugins=allowed_plugins or None)
if not results:
prints(buf.getvalue(), file=sys.stderr)
prints('No results found', file=sys.stderr)
raise SystemExit(1)
result = results[0]
cf = None
if opts.cover and results:
cover = download_cover(log, title=opts.title, authors=authors,
identifiers=result.identifiers, timeout=int(opts.timeout))
if cover is None:
if not opts.opf:
prints('No cover found', file=sys.stderr)
else:
save_cover_data_to(cover[-1], opts.cover)
result.cover = cf = opts.cover
if opts.verbose:
prints(buf.getvalue(), file=sys.stderr)
if opts.opf:
getattr(sys.stdout, 'buffer', sys.stdout).write(metadata_to_opf(result))
print()
else:
prints(str(result))
if not opts.opf and opts.cover:
prints('Cover :', cf)
return 0
def extract_images(self, processed_records, output_dir):
self.log.debug('Extracting images...')
output_dir = os.path.abspath(os.path.join(output_dir, 'images'))
if not os.path.exists(output_dir):
os.makedirs(output_dir)
image_index = 0
self.image_names = []
image_name_map = {}
start = getattr(self.book_header, 'first_image_index', -1)
if start > self.num_sections or start < 0:
# BAEN PRC files have bad headers
start = 0
for i in range(start, self.num_sections):
if i in processed_records:
continue
processed_records.append(i)
data = self.sections[i][0]
image_index += 1
if data[:4] in {
b'FLIS', b'FCIS', b'SRCS', b'\xe9\x8e\r\n', b'RESC',
b'BOUN', b'FDST', b'DATP', b'AUDI', b'VIDE'
}:
# This record is a known non image type, no need to try to
# load the image
continue
try:
imgfmt = what(None, data)
except Exception:
continue
if imgfmt not in {'jpg', 'jpeg', 'gif', 'png', 'bmp'}:
continue
if imgfmt == 'jpeg':
imgfmt = 'jpg'
if imgfmt == 'gif':
try:
data = gif_data_to_png_data(data)
imgfmt = 'png'
except AnimatedGIF:
pass
path = os.path.join(output_dir, '%05d.%s' % (image_index, imgfmt))
image_name_map[image_index] = os.path.basename(path)
if imgfmt == 'png':
with open(path, 'wb') as f:
f.write(data)
else:
try:
save_cover_data_to(data, path, minify_to=(10000, 10000))
except Exception:
continue
self.image_names.append(os.path.basename(path))
return image_name_map
def main(args=sys.argv):
parser = option_parser()
opts, args = parser.parse_args(args)
buf = BytesIO()
log = create_log(buf)
abort = Event()
authors = []
if opts.authors:
authors = string_to_authors(opts.authors)
identifiers = {}
if opts.isbn:
identifiers['isbn'] = opts.isbn
results = identify(log, abort, title=opts.title, authors=authors,
identifiers=identifiers, timeout=int(opts.timeout))
if not results:
print (log, file=sys.stderr)
prints('No results found', file=sys.stderr)
raise SystemExit(1)
result = results[0]
cf = None
if opts.cover and results:
cover = download_cover(log, title=opts.title, authors=authors,
identifiers=result.identifiers, timeout=int(opts.timeout))
if cover is None and not opts.opf:
prints('No cover found', file=sys.stderr)
else:
save_cover_data_to(cover[-1], opts.cover)
result.cover = cf = opts.cover
log = buf.getvalue()
result = (metadata_to_opf(result) if opts.opf else
unicode(result).encode('utf-8'))
if opts.verbose:
print (log, file=sys.stderr)
print (result)
if not opts.opf and opts.cover:
prints('Cover :', cf)
return 0
def fb2mlize_images(self):
'''
This function uses the self.image_hrefs dictionary mapping. It is populated by the dump_text function.
'''
from calibre.ebooks.oeb.base import OEB_RASTER_IMAGES
images = []
for item in self.oeb_book.manifest:
# Don't write the image if it's not referenced in the document's text.
if item.href not in self.image_hrefs:
continue
if item.media_type in OEB_RASTER_IMAGES:
try:
if item.media_type != 'image/jpeg':
imdata = save_cover_data_to(item.data, compression_quality=70)
raw_data = b64encode(imdata)
else:
raw_data = b64encode(item.data)
# Don't put the encoded image on a single line.
data = ''
col = 1
for char in raw_data:
if col == 72:
data += '\n'
col = 1
col += 1
data += char
images.append('<binary id="%s" content-type="image/jpeg">%s\n</binary>' % (self.image_hrefs[item.href], data))
except Exception as e:
self.log.error('Error: Could not include file %s because '
'%s.' % (item.href, e))
return ''.join(images)
def fb2mlize_images(self):
'''
This function uses the self.image_hrefs dictionary mapping. It is populated by the dump_text function.
'''
from calibre.ebooks.oeb.base import OEB_RASTER_IMAGES
images = []
for item in self.oeb_book.manifest:
# Don't write the image if it's not referenced in the document's text.
if item.href not in self.image_hrefs:
continue
if item.media_type in OEB_RASTER_IMAGES:
try:
if item.media_type != 'image/jpeg':
imdata = save_cover_data_to(item.data,
compression_quality=70)
raw_data = b64encode(imdata)
else:
raw_data = b64encode(item.data)
# Don't put the encoded image on a single line.
data = ''
col = 1
for char in raw_data:
if col == 72:
data += '\n'
col = 1
col += 1
data += char
images.append(
'<binary id="%s" content-type="image/jpeg">%s\n</binary>'
% (self.image_hrefs[item.href], data))
except Exception as e:
self.log.error('Error: Could not include file %s because '
'%s.' % (item.href, e))
return ''.join(images)
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 fb2mlize_images(self):
'''
This function uses the self.image_hrefs dictionary mapping. It is populated by the dump_text function.
'''
from calibre.ebooks.oeb.base import OEB_RASTER_IMAGES
images = []
for item in self.oeb_book.manifest:
# Don't write the image if it's not referenced in the document's text.
if item.href not in self.image_hrefs:
continue
if item.media_type in OEB_RASTER_IMAGES:
try:
if item.media_type not in ('image/jpeg', 'image/png'):
imdata = save_cover_data_to(item.data,
compression_quality=70)
raw_data = as_base64_unicode(imdata)
content_type = 'image/jpeg'
else:
raw_data = as_base64_unicode(item.data)
content_type = item.media_type
# Don't put the encoded image on a single line.
step = 72
data = '\n'.join(raw_data[i:i + step]
for i in range(0, len(raw_data), step))
images.append(
'<binary id="%s" content-type="%s">%s</binary>' %
(self.image_hrefs[item.href], content_type, data))
except Exception as e:
self.log.error('Error: Could not include file %s because '
'%s.' % (item.href, e))
return '\n'.join(images)
def get_cover(self, id, thumbnail=False, thumb_width=60, thumb_height=80):
try:
cherrypy.response.headers['Content-Type'] = 'image/jpeg'
cherrypy.response.timeout = 3600
cover = self.db.cover(id, index_is_id=True)
if cover is None:
cover = self.default_cover
updated = self.build_time
else:
updated = self.db.cover_last_modified(id, index_is_id=True)
cherrypy.response.headers['Last-Modified'] = self.last_modified(updated)
if thumbnail:
quality = tweaks['content_server_thumbnail_compression_quality']
if quality < 50:
quality = 50
elif quality > 99:
quality = 99
return scale_image(cover, thumb_width, thumb_height, compression_quality=quality)[-1]
return save_cover_data_to(cover, None, minify_to=(self.max_cover_width, self.max_cover_height))
except Exception as err:
import traceback
cherrypy.log.error('Failed to generate cover:')
cherrypy.log.error(traceback.print_exc())
raise cherrypy.HTTPError(404, 'Failed to generate cover: %r'%err)
def get_cover(self, id, thumbnail=False, thumb_width=60, thumb_height=80):
try:
cherrypy.response.headers['Content-Type'] = 'image/jpeg'
cherrypy.response.timeout = 3600
cover = self.db.cover(id, index_is_id=True)
if cover is None:
cover = self.default_cover
updated = self.build_time
else:
updated = self.db.cover_last_modified(id, index_is_id=True)
cherrypy.response.headers['Last-Modified'] = self.last_modified(
updated)
if thumbnail:
quality = tweaks[
'content_server_thumbnail_compression_quality']
if quality < 50:
quality = 50
elif quality > 99:
quality = 99
return scale_image(cover,
thumb_width,
thumb_height,
compression_quality=quality)[-1]
return save_cover_data_to(cover,
None,
minify_to=(self.max_cover_width,
self.max_cover_height))
except Exception as err:
import traceback
cherrypy.log.error('Failed to generate cover:')
cherrypy.log.error(traceback.print_exc())
raise cherrypy.HTTPError(404, 'Failed to generate cover: %r' % err)
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:
raw = ''
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:
data = ''
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 image_to_hexstring(self, data):
# Images must be hex-encoded in 128 character lines
data = save_cover_data_to(data)
width, height = identify(data)[1:]
lines = []
v = memoryview(data)
for i in range(0, len(data), 64):
lines.append(hexlify(v[i:i + 64]))
hex_string = b'\n'.join(lines).decode('ascii')
return hex_string, width, height
def process_result(log, result):
plugin, data = result
try:
if getattr(plugin, 'auto_trim_covers', False):
img = image_from_data(data)
nimg = remove_borders_from_image(img)
if nimg is not img:
data = image_to_data(nimg)
fmt, width, height = identify(data)
if width < 0 or height < 0:
raise ValueError('Could not read cover image dimensions')
if width < 50 or height < 50:
raise ValueError('Image too small')
data = save_cover_data_to(data)
except Exception:
log.exception('Invalid cover from', plugin.name)
return None
return (plugin, width, height, fmt, data)
def process_result(log, result):
plugin, data = result
try:
if getattr(plugin, 'auto_trim_covers', False):
img = image_from_data(data)
nimg = remove_borders_from_image(img)
if nimg is not img:
data = image_to_data(nimg)
fmt, width, height = identify(data)
if width < 0 or height < 0:
raise ValueError('Could not read cover image dimensions')
if width < 50 or height < 50:
raise ValueError('Image too small')
data = save_cover_data_to(data)
except Exception:
log.exception('Invalid cover from', plugin.name)
return None
return (plugin, width, height, fmt, data)
def rescale_image(data, maxsizeb=IMAGE_MAX_SIZE, dimen=None):
'''
Convert image setting all transparent pixels to white and changing format
to JPEG. Ensure the resultant image has a byte size less than
maxsizeb.
If dimen is not None, generate a thumbnail of
width=dimen, height=dimen or width, height = dimen (depending on the type
of dimen)
Returns the image as a bytestring
'''
if dimen is not None:
if hasattr(dimen, '__len__'):
width, height = dimen
else:
width = height = dimen
data = scale_image(data,
width=width,
height=height,
compression_quality=90)[-1]
else:
# Replace transparent pixels with white pixels and convert to JPEG
data = save_cover_data_to(data)
if len(data) <= maxsizeb:
return data
orig_data = data # save it in case compression fails
quality = 90
while len(data) > maxsizeb and quality >= 5:
data = image_to_data(image_from_data(orig_data),
compression_quality=quality)
quality -= 5
if len(data) <= maxsizeb:
return data
orig_data = data
scale = 0.9
while len(data) > maxsizeb and scale >= 0.05:
img = image_from_data(data)
w, h = img.width(), img.height()
img = resize_image(img, int(scale * w), int(scale * h))
data = image_to_data(img, compression_quality=quality)
scale -= 0.05
return data
def image_to_hexstring(self, data):
data = save_cover_data_to(data)
width, height = identify(data)[1:]
raw_hex = ''
for char in data:
raw_hex += hex(ord(char)).replace('0x', '').rjust(2, '0')
# Images must be broken up so that they are no longer than 129 chars
# per line
hex_string = ''
col = 1
for char in raw_hex:
if col == 129:
hex_string += '\n'
col = 1
col += 1
hex_string += char
return (hex_string, width, height)
def image_to_hexstring(self, data):
data = save_cover_data_to(data)
width, height = identify(data)[1:]
raw_hex = ''
for char in data:
raw_hex += hex(ord(char)).replace('0x', '').rjust(2, '0')
# Images must be broken up so that they are no longer than 129 chars
# per line
hex_string = ''
col = 1
for char in raw_hex:
if col == 129:
hex_string += '\n'
col = 1
col += 1
hex_string += char
return (hex_string, width, height)
def rescale_image(data, maxsizeb=IMAGE_MAX_SIZE, dimen=None):
'''
Convert image setting all transparent pixels to white and changing format
to JPEG. Ensure the resultant image has a byte size less than
maxsizeb.
If dimen is not None, generate a thumbnail of
width=dimen, height=dimen or width, height = dimen (depending on the type
of dimen)
Returns the image as a bytestring
'''
if dimen is not None:
if hasattr(dimen, '__len__'):
width, height = dimen
else:
width = height = dimen
data = scale_image(data, width=width, height=height, compression_quality=90)[-1]
else:
# Replace transparent pixels with white pixels and convert to JPEG
data = save_cover_data_to(data)
if len(data) <= maxsizeb:
return data
orig_data = data # save it in case compression fails
quality = 90
while len(data) > maxsizeb and quality >= 5:
data = image_to_data(image_from_data(orig_data), compression_quality=quality)
quality -= 5
if len(data) <= maxsizeb:
return data
orig_data = data
scale = 0.9
while len(data) > maxsizeb and scale >= 0.05:
img = image_from_data(data)
w, h = img.width(), img.height()
img = resize_image(img, int(scale*w), int(scale*h))
data = image_to_data(img, compression_quality=quality)
scale -= 0.05
return data
def convert(self, oeb_book, output_path, input_plugin, opts, log):
from lxml import etree
from calibre.ebooks.oeb.base import OEB_IMAGES, SVG_MIME
from calibre.ebooks.metadata.opf2 import OPF, metadata_to_opf
from calibre.utils.zipfile import ZipFile
from calibre.utils.filenames import ascii_filename
# HTML
if opts.htmlz_css_type == 'inline':
from calibre.ebooks.htmlz.oeb2html import OEB2HTMLInlineCSSizer
OEB2HTMLizer = OEB2HTMLInlineCSSizer
elif opts.htmlz_css_type == 'tag':
from calibre.ebooks.htmlz.oeb2html import OEB2HTMLNoCSSizer
OEB2HTMLizer = OEB2HTMLNoCSSizer
else:
from calibre.ebooks.htmlz.oeb2html import OEB2HTMLClassCSSizer as OEB2HTMLizer
with TemporaryDirectory(u'_htmlz_output') as tdir:
htmlizer = OEB2HTMLizer(log)
html = htmlizer.oeb2html(oeb_book, opts)
fname = u'index'
if opts.htmlz_title_filename:
from calibre.utils.filenames import shorten_components_to
fname = shorten_components_to(100, (ascii_filename(
unicode_type(oeb_book.metadata.title[0])), ))[0]
with open(os.path.join(tdir, fname + u'.html'), 'wb') as tf:
if isinstance(html, unicode_type):
html = html.encode('utf-8')
tf.write(html)
# CSS
if opts.htmlz_css_type == 'class' and opts.htmlz_class_style == 'external':
with open(os.path.join(tdir, u'style.css'), 'wb') as tf:
tf.write(htmlizer.get_css(oeb_book))
# Images
images = htmlizer.images
if images:
if not os.path.exists(os.path.join(tdir, u'images')):
os.makedirs(os.path.join(tdir, u'images'))
for item in oeb_book.manifest:
if item.media_type in OEB_IMAGES and item.href in images:
if item.media_type == SVG_MIME:
data = unicode_type(
etree.tostring(item.data,
encoding=unicode_type))
else:
data = item.data
fname = os.path.join(tdir, u'images',
images[item.href])
with open(fname, 'wb') as img:
img.write(data)
# Cover
cover_path = None
try:
cover_data = None
if oeb_book.metadata.cover:
term = oeb_book.metadata.cover[0].term
cover_data = oeb_book.guide[term].item.data
if cover_data:
from calibre.utils.img import save_cover_data_to
cover_path = os.path.join(tdir, u'cover.jpg')
with lopen(cover_path, 'w') as cf:
cf.write('')
save_cover_data_to(cover_data, cover_path)
except:
import traceback
traceback.print_exc()
# Metadata
with open(os.path.join(tdir, u'metadata.opf'), 'wb') as mdataf:
opf = OPF(
io.BytesIO(
etree.tostring(oeb_book.metadata.to_opf1(),
encoding='UTF-8')))
mi = opf.to_book_metadata()
if cover_path:
mi.cover = u'cover.jpg'
mdataf.write(metadata_to_opf(mi))
htmlz = ZipFile(output_path, 'w')
htmlz.add_dir(tdir)
def serialize_cover_data(new_cdata, cpath):
from calibre.utils.img import save_cover_data_to
return save_cover_data_to(new_cdata,
data_fmt=os.path.splitext(cpath)[1][1:])
def _encode_into_jpeg(data):
data = save_cover_data_to(data)
return as_base64_unicode(data)
def _encode_into_jpeg(data):
data = save_cover_data_to(data)
return b64encode(data)
def _write_new_cover(new_cdata, cpath):
from calibre.utils.img import save_cover_data_to
new_cover = PersistentTemporaryFile(suffix=os.path.splitext(cpath)[1])
new_cover.close()
save_cover_data_to(new_cdata, new_cover.name)
return new_cover
def extract_content(self, output_dir):
# Each text record is independent (unless the continuation
# value is set in the previous record). Put each converted
# text recored into a separate file. We will reference the
# home.html file as the first file and let the HTML input
# plugin assemble the order based on hyperlinks.
with CurrentDir(output_dir):
for uid, num in self.uid_text_secion_number.items():
self.log.debug('Writing record with uid: %s as %s.html' % (uid, uid))
with open('%s.html' % uid, 'wb') as htmlf:
html = u'<html><body>'
section_header, section_data = self.sections[num]
if section_header.type == DATATYPE_PHTML:
html += self.process_phtml(section_data.data, section_data.header.paragraph_offsets)
elif section_header.type == DATATYPE_PHTML_COMPRESSED:
d = self.decompress_phtml(section_data.data)
html += self.process_phtml(d, section_data.header.paragraph_offsets).decode(self.get_text_uid_encoding(section_header.uid), 'replace')
html += '</body></html>'
htmlf.write(html.encode('utf-8'))
# Images.
# Cache the image sizes in case they are used by a composite image.
images = set()
if not os.path.exists(os.path.join(output_dir, 'images/')):
os.makedirs(os.path.join(output_dir, 'images/'))
with CurrentDir(os.path.join(output_dir, 'images/')):
# Single images.
for uid, num in self.uid_image_section_number.items():
section_header, section_data = self.sections[num]
if section_data:
idata = None
if section_header.type == DATATYPE_TBMP:
idata = section_data
elif section_header.type == DATATYPE_TBMP_COMPRESSED:
if self.header_record.compression == 1:
idata = decompress_doc(section_data)
elif self.header_record.compression == 2:
idata = zlib.decompress(section_data)
try:
save_cover_data_to(idata, '%s.jpg' % uid, compression_quality=70)
images.add(uid)
self.log.debug('Wrote image with uid %s to images/%s.jpg' % (uid, uid))
except Exception as e:
self.log.error('Failed to write image with uid %s: %s' % (uid, e))
else:
self.log.error('Failed to write image with uid %s: No data.' % uid)
# Composite images.
# We're going to use the already compressed .jpg images here.
for uid, num in self.uid_composite_image_section_number.items():
try:
section_header, section_data = self.sections[num]
# Get the final width and height.
width = 0
height = 0
for row in section_data.layout:
row_width = 0
col_height = 0
for col in row:
if col not in images:
raise Exception('Image with uid: %s missing.' % col)
w, h = identify(lopen('%s.jpg' % col, 'rb'))[1:]
row_width += w
if col_height < h:
col_height = h
if width < row_width:
width = row_width
height += col_height
# Create a new image the total size of all image
# parts. Put the parts into the new image.
with Canvas(width, height) as canvas:
y_off = 0
for row in section_data.layout:
x_off = 0
largest_height = 0
for col in row:
im = image_from_data(lopen('%s.jpg' % col, 'rb').read())
canvas.compose(im, x_off, y_off)
w, h = im.width(), im.height()
x_off += w
if largest_height < h:
largest_height = h
y_off += largest_height
with lopen('%s.jpg' % uid) as out:
out.write(canvas.export(compression_quality=70))
self.log.debug('Wrote composite image with uid %s to images/%s.jpg' % (uid, uid))
except Exception as e:
self.log.error('Failed to write composite image with uid %s: %s' % (uid, e))
# Run the HTML through the html processing plugin.
from calibre.customize.ui import plugin_for_input_format
html_input = plugin_for_input_format('html')
for opt in html_input.options:
setattr(self.options, opt.option.name, opt.recommended_value)
self.options.input_encoding = 'utf-8'
odi = self.options.debug_pipeline
self.options.debug_pipeline = None
# Determine the home.html record uid. This should be set in the
# reserved values in the metadata recored. home.html is the first
# text record (should have hyper link references to other records)
# in the document.
try:
home_html = self.header_record.home_html
if not home_html:
home_html = self.uid_text_secion_number.items()[0][0]
except:
raise Exception('Could not determine home.html')
# Generate oeb from html conversion.
oeb = html_input.convert(open('%s.html' % home_html, 'rb'), self.options, 'html', self.log, {})
self.options.debug_pipeline = odi
return oeb
def _encode_into_jpeg(data):
data = save_cover_data_to(data)
return b64encode(data)
def convert(self, oeb_book, output_path, input_plugin, opts, log):
from lxml import etree
from calibre.ebooks.oeb.base import OEB_IMAGES, SVG_MIME
from calibre.ebooks.metadata.opf2 import OPF, metadata_to_opf
from calibre.utils.zipfile import ZipFile
from calibre.utils.filenames import ascii_filename
# HTML
if opts.htmlz_css_type == 'inline':
from calibre.ebooks.htmlz.oeb2html import OEB2HTMLInlineCSSizer
OEB2HTMLizer = OEB2HTMLInlineCSSizer
elif opts.htmlz_css_type == 'tag':
from calibre.ebooks.htmlz.oeb2html import OEB2HTMLNoCSSizer
OEB2HTMLizer = OEB2HTMLNoCSSizer
else:
from calibre.ebooks.htmlz.oeb2html import OEB2HTMLClassCSSizer as OEB2HTMLizer
with TemporaryDirectory(u'_htmlz_output') as tdir:
htmlizer = OEB2HTMLizer(log)
html = htmlizer.oeb2html(oeb_book, opts)
fname = u'index'
if opts.htmlz_title_filename:
from calibre.utils.filenames import shorten_components_to
fname = shorten_components_to(100, (ascii_filename(unicode_type(oeb_book.metadata.title[0])),))[0]
with open(os.path.join(tdir, fname+u'.html'), 'wb') as tf:
if isinstance(html, unicode_type):
html = html.encode('utf-8')
tf.write(html)
# CSS
if opts.htmlz_css_type == 'class' and opts.htmlz_class_style == 'external':
with open(os.path.join(tdir, u'style.css'), 'wb') as tf:
tf.write(htmlizer.get_css(oeb_book))
# Images
images = htmlizer.images
if images:
if not os.path.exists(os.path.join(tdir, u'images')):
os.makedirs(os.path.join(tdir, u'images'))
for item in oeb_book.manifest:
if item.media_type in OEB_IMAGES and item.href in images:
if item.media_type == SVG_MIME:
data = etree.tostring(item.data, encoding='unicode')
else:
data = item.data
fname = os.path.join(tdir, u'images', images[item.href])
with open(fname, 'wb') as img:
img.write(data)
# Cover
cover_path = None
try:
cover_data = None
if oeb_book.metadata.cover:
term = oeb_book.metadata.cover[0].term
cover_data = oeb_book.guide[term].item.data
if cover_data:
from calibre.utils.img import save_cover_data_to
cover_path = os.path.join(tdir, u'cover.jpg')
with lopen(cover_path, 'w') as cf:
cf.write('')
save_cover_data_to(cover_data, cover_path)
except:
import traceback
traceback.print_exc()
# Metadata
with open(os.path.join(tdir, u'metadata.opf'), 'wb') as mdataf:
opf = OPF(io.BytesIO(etree.tostring(oeb_book.metadata.to_opf1(), encoding='UTF-8')))
mi = opf.to_book_metadata()
if cover_path:
mi.cover = u'cover.jpg'
mdataf.write(metadata_to_opf(mi))
htmlz = ZipFile(output_path, 'w')
htmlz.add_dir(tdir)
def serialize_cover_data(new_cdata, cpath):
from calibre.utils.img import save_cover_data_to
return save_cover_data_to(new_cdata, data_fmt=os.path.splitext(cpath)[1][1:])
def _write_new_cover(new_cdata, cpath):
from calibre.utils.img import save_cover_data_to
new_cover = PersistentTemporaryFile(suffix=os.path.splitext(cpath)[1])
new_cover.close()
save_cover_data_to(new_cdata, new_cover.name)
return new_cover
def _encode_into_jpeg(data):
data = save_cover_data_to(data)
return as_base64_unicode(data)
def extract_content(self, output_dir):
# Each text record is independent (unless the continuation
# value is set in the previous record). Put each converted
# text recored into a separate file. We will reference the
# home.html file as the first file and let the HTML input
# plugin assemble the order based on hyperlinks.
with CurrentDir(output_dir):
for uid, num in self.uid_text_secion_number.items():
self.log.debug('Writing record with uid: %s as %s.html' %
(uid, uid))
with open('%s.html' % uid, 'wb') as htmlf:
html = u'<html><body>'
section_header, section_data = self.sections[num]
if section_header.type == DATATYPE_PHTML:
html += self.process_phtml(
section_data.data,
section_data.header.paragraph_offsets)
elif section_header.type == DATATYPE_PHTML_COMPRESSED:
d = self.decompress_phtml(section_data.data)
html += self.process_phtml(
d, section_data.header.paragraph_offsets).decode(
self.get_text_uid_encoding(section_header.uid),
'replace')
html += '</body></html>'
htmlf.write(html.encode('utf-8'))
# Images.
# Cache the image sizes in case they are used by a composite image.
images = set()
if not os.path.exists(os.path.join(output_dir, 'images/')):
os.makedirs(os.path.join(output_dir, 'images/'))
with CurrentDir(os.path.join(output_dir, 'images/')):
# Single images.
for uid, num in self.uid_image_section_number.items():
section_header, section_data = self.sections[num]
if section_data:
idata = None
if section_header.type == DATATYPE_TBMP:
idata = section_data
elif section_header.type == DATATYPE_TBMP_COMPRESSED:
if self.header_record.compression == 1:
idata = decompress_doc(section_data)
elif self.header_record.compression == 2:
idata = zlib.decompress(section_data)
try:
save_cover_data_to(idata,
'%s.jpg' % uid,
compression_quality=70)
images.add(uid)
self.log.debug(
'Wrote image with uid %s to images/%s.jpg' %
(uid, uid))
except Exception as e:
self.log.error(
'Failed to write image with uid %s: %s' % (uid, e))
else:
self.log.error(
'Failed to write image with uid %s: No data.' % uid)
# Composite images.
# We're going to use the already compressed .jpg images here.
for uid, num in self.uid_composite_image_section_number.items():
try:
section_header, section_data = self.sections[num]
# Get the final width and height.
width = 0
height = 0
for row in section_data.layout:
row_width = 0
col_height = 0
for col in row:
if col not in images:
raise Exception('Image with uid: %s missing.' %
col)
w, h = identify(lopen('%s.jpg' % col, 'rb'))[1:]
row_width += w
if col_height < h:
col_height = h
if width < row_width:
width = row_width
height += col_height
# Create a new image the total size of all image
# parts. Put the parts into the new image.
with Canvas(width, height) as canvas:
y_off = 0
for row in section_data.layout:
x_off = 0
largest_height = 0
for col in row:
im = image_from_data(
lopen('%s.jpg' % col, 'rb').read())
canvas.compose(im, x_off, y_off)
w, h = im.width(), im.height()
x_off += w
if largest_height < h:
largest_height = h
y_off += largest_height
with lopen('%s.jpg' % uid) as out:
out.write(canvas.export(compression_quality=70))
self.log.debug(
'Wrote composite image with uid %s to images/%s.jpg' %
(uid, uid))
except Exception as e:
self.log.error(
'Failed to write composite image with uid %s: %s' %
(uid, e))
# Run the HTML through the html processing plugin.
from calibre.customize.ui import plugin_for_input_format
html_input = plugin_for_input_format('html')
for opt in html_input.options:
setattr(self.options, opt.option.name, opt.recommended_value)
self.options.input_encoding = 'utf-8'
odi = self.options.debug_pipeline
self.options.debug_pipeline = None
# Determine the home.html record uid. This should be set in the
# reserved values in the metadata recored. home.html is the first
# text record (should have hyper link references to other records)
# in the document.
try:
home_html = self.header_record.home_html
if not home_html:
home_html = self.uid_text_secion_number.items()[0][0]
except:
raise Exception('Could not determine home.html')
# Generate oeb from html conversion.
oeb = html_input.convert(open('%s.html' % home_html, 'rb'),
self.options, 'html', self.log, {})
self.options.debug_pipeline = odi
return oeb