def __init__(self, path_to_ebook, tdir, log=None):
log = log or default_log
book_fmt, opfpath, input_fmt = extract_book(path_to_ebook, tdir, log=log)
ContainerBase.__init__(self, tdir, opfpath, log)
excluded_names = {
name for name, mt in self.mime_map.iteritems() if
name == self.opf_name or mt == guess_type('a.ncx') or name.startswith('META-INF/')
}
self.book_render_data = data = {
'version': self.RENDER_VERSION,
'toc':get_toc(self).as_dict,
'spine':[name for name, is_linear in self.spine_names],
'link_uid': uuid4(),
'is_comic': input_fmt.lower() in {'cbc', 'cbz', 'cbr', 'cb7'},
'manifest': list(set(self.name_path_map) - excluded_names),
}
# Mark the spine as dirty since we have to ensure it is normalized
for name in data['spine']:
self.parsed(name), self.dirty(name)
self.virtualize_resources()
self.commit()
for name in excluded_names:
os.remove(self.name_path_map[name])
with lopen(os.path.join(self.root, 'calibre-book-manifest.json'), 'wb') as f:
f.write(json.dumps(self.book_render_data, ensure_ascii=False).encode('utf-8'))
def __init__(self, parent=None):
QNetworkAccessManager.__init__(self, parent)
self.mathjax_prefix = str(uuid4())
self.mathjax_base = '%s://%s/%s/' % (FAKE_PROTOCOL, FAKE_HOST, self.mathjax_prefix)
self.root = self.orig_root = os.path.dirname(P('viewer/blank.html', allow_user_override=False))
self.mime_map, self.single_pages, self.codec_map = {}, set(), {}
self.mathjax_dir = P('mathjax', allow_user_override=False)
def __init__(self, prefs):
self.log = default_log
self.current_frag = None
self.com_id = str(uuid4())
QWebEnginePage.__init__(self)
secure_webengine(self.settings(), for_viewer=True)
self.titleChanged.connect(self.title_changed)
self.loadFinished.connect(self.show_frag)
s = QWebEngineScript()
s.setName('toc.js')
s.setInjectionPoint(QWebEngineScript.InjectionPoint.DocumentCreation)
s.setRunsOnSubFrames(True)
s.setWorldId(QWebEngineScript.ScriptWorldId.ApplicationWorld)
js = P('toc.js', allow_user_override=False,
data=True).decode('utf-8').replace('COM_ID', self.com_id, 1)
if 'preview_background' in prefs.defaults and 'preview_foreground' in prefs.defaults:
from calibre.gui2.tweak_book.preview import get_editor_settings
settings = get_editor_settings(prefs)
else:
if is_dark_theme():
settings = {
'is_dark_theme': True,
'bg': dark_color.name(),
'fg': dark_text_color.name(),
'link': dark_link_color.name(),
}
else:
settings = {}
js = js.replace('SETTINGS', json.dumps(settings), 1)
dark_mode_css = P('dark_mode.css',
data=True,
allow_user_override=False).decode('utf-8')
js = js.replace('CSS', json.dumps(dark_mode_css), 1)
s.setSourceCode(js)
self.scripts().insert(s)
def write_to_unicode(self, objects):
try:
name = self.metrics.postscript_name
except KeyError:
name = uuid4()
cmap = CMap(name, self.metrics.glyph_map, compress=self.compress)
self.font_dict['ToUnicode'] = objects.add(cmap)
def __init__(self, name, glyph_map, compress=False):
Stream.__init__(self, compress)
current_map = OrderedDict()
maps = []
for glyph_id in sorted(glyph_map):
if len(current_map) > 99:
maps.append(current_map)
current_map = OrderedDict()
val = []
for c in glyph_map[glyph_id]:
c = ord(c)
val.append(to_hex_string(c))
glyph_id = '<%s>'%to_hex_string(glyph_id)
current_map[glyph_id] = '<%s>'%''.join(val)
if current_map:
maps.append(current_map)
mapping = []
for m in maps:
meat = '\n'.join('%s %s'%(k, v) for k, v in iteritems(m))
mapping.append('%d beginbfchar\n%s\nendbfchar'%(len(m), meat))
try:
name = name.encode('ascii').decode('ascii')
except Exception:
name = uuid4()
self.write(self.skeleton.format(name=name, mapping='\n'.join(mapping)))
def write_to_unicode(self, objects):
try:
name = self.metrics.postscript_name
except KeyError:
name = uuid4()
cmap = CMap(name, self.metrics.glyph_map, compress=self.compress)
self.font_dict['ToUnicode'] = objects.add(cmap)
def __init__(self, parent=None):
QNetworkAccessManager.__init__(self, parent)
self.mathjax_prefix = str(uuid4())
self.mathjax_base = '%s://%s/%s/' % (FAKE_PROTOCOL, FAKE_HOST, self.mathjax_prefix)
self.root = self.orig_root = os.path.dirname(P('viewer/blank.html', allow_user_override=False))
self.mime_map, self.single_pages, self.codec_map = {}, set(), {}
self.mathjax_dir = P('mathjax', allow_user_override=False)
def __init__(self, name, glyph_map, compress=False):
Stream.__init__(self, compress)
current_map = OrderedDict()
maps = []
for glyph_id in sorted(glyph_map):
if len(current_map) > 99:
maps.append(current_map)
current_map = OrderedDict()
val = []
for c in glyph_map[glyph_id]:
c = ord(c)
val.append(to_hex_string(c))
glyph_id = '<%s>'%to_hex_string(glyph_id)
current_map[glyph_id] = '<%s>'%''.join(val)
if current_map:
maps.append(current_map)
mapping = []
for m in maps:
meat = '\n'.join('%s %s'%(k, v) for k, v in iteritems(m))
mapping.append('%d beginbfchar\n%s\nendbfchar'%(len(m), meat))
try:
name = name.encode('ascii').decode('ascii')
except Exception:
name = uuid4()
self.write(self.skeleton.format(name=name, mapping='\n'.join(mapping)))
def __init__(self, path_to_ebook, tdir, log=None, book_hash=None):
log = log or default_log
book_fmt, opfpath, input_fmt = extract_book(path_to_ebook, tdir, log=log)
ContainerBase.__init__(self, tdir, opfpath, log)
excluded_names = {
name for name, mt in self.mime_map.iteritems() if
name == self.opf_name or mt == guess_type('a.ncx') or name.startswith('META-INF/') or
name == 'mimetype'
}
self.book_render_data = data = {
'version': RENDER_VERSION,
'toc':get_toc(self).as_dict,
'spine':[name for name, is_linear in self.spine_names],
'link_uid': uuid4(),
'book_hash': book_hash,
'is_comic': input_fmt.lower() in {'cbc', 'cbz', 'cbr', 'cb7'},
}
# Mark the spine as dirty since we have to ensure it is normalized
for name in data['spine']:
self.parsed(name), self.dirty(name)
self.virtualized_names = set()
self.virtualize_resources()
def manifest_data(name):
return {'size':os.path.getsize(self.name_path_map[name]), 'is_virtualized': name in self.virtualized_names, 'mimetype':self.mime_map.get(name)}
data['files'] = {name:manifest_data(name) for name in set(self.name_path_map) - excluded_names}
self.commit()
for name in excluded_names:
os.remove(self.name_path_map[name])
with lopen(os.path.join(self.root, 'calibre-book-manifest.json'), 'wb') as f:
f.write(json.dumps(self.book_render_data, ensure_ascii=False).encode('utf-8'))
def __init__(self, path_to_ebook, tdir, log=None, book_hash=None):
log = log or default_log
book_fmt, opfpath, input_fmt = extract_book(path_to_ebook, tdir, log=log)
ContainerBase.__init__(self, tdir, opfpath, log)
excluded_names = {
name for name, mt in self.mime_map.iteritems() if
name == self.opf_name or mt == guess_type('a.ncx') or name.startswith('META-INF/') or
name == 'mimetype'
}
raster_cover_name, titlepage_name = self.create_cover_page(input_fmt.lower())
self.book_render_data = data = {
'version': RENDER_VERSION,
'toc':get_toc(self).as_dict,
'spine':[name for name, is_linear in self.spine_names],
'link_uid': uuid4(),
'book_hash': book_hash,
'is_comic': input_fmt.lower() in {'cbc', 'cbz', 'cbr', 'cb7'},
'raster_cover_name': raster_cover_name,
'title_page_name': titlepage_name,
'has_maths': False,
'total_length': 0,
'spine_length': 0,
}
# Mark the spine as dirty since we have to ensure it is normalized
for name in data['spine']:
self.parsed(name), self.dirty(name)
self.transform_css()
self.virtualized_names = set()
self.virtualize_resources()
def manifest_data(name):
mt = (self.mime_map.get(name) or 'application/octet-stream').lower()
ans = {
'size':os.path.getsize(self.name_path_map[name]),
'is_virtualized': name in self.virtualized_names,
'mimetype':mt,
'is_html': mt in OEB_DOCS,
}
if ans['is_html']:
root = self.parsed(name)
ans['length'] = l = get_length(root)
self.book_render_data['total_length'] += l
if name in data['spine']:
self.book_render_data['spine_length'] += l
ans['has_maths'] = hm = check_for_maths(root)
if hm:
self.book_render_data['has_maths'] = True
return ans
data['files'] = {name:manifest_data(name) for name in set(self.name_path_map) - excluded_names}
self.commit()
for name in excluded_names:
os.remove(self.name_path_map[name])
with lopen(os.path.join(self.root, 'calibre-book-manifest.json'), 'wb') as f:
f.write(json.dumps(self.book_render_data, ensure_ascii=False).encode('utf-8'))
def add_all_links(container, margin_files):
uuid = uuid4()
name_anchor_map = {}
for name, is_linear in container.spine_names:
root = container.parsed(name)
name_anchor_map[name] = frozenset(root.xpath('//*/@id'))
for margin_file in margin_files:
name = margin_file.name
anchors = name_anchor_map.get(name, set())
add_anchors_markup(container.parsed(name), uuid, anchors)
container.dirty(name)
return uuid
def prepare_convert(temp_path, key, st):
tdir = tempfile.mkdtemp(dir=temp_path)
now = time.time()
return {
'path': os.path.basename(tdir),
'id': uuid4(),
'status': 'working',
'mtime': now,
'atime': now,
'key': key,
'file_mtime': st.st_mtime,
'file_size': st.st_size,
'cache_size': 0,
}
def make_book(options, comic_file, log):
global _log, _options
_options = options
print(options)
_log = log
log.prints(INFO, "Extracting images.")
path = get_work_folder(comic_file)
image_path = os.path.join(path, "OEBPS", "Images")
log.prints(INFO, "Processing images.")
img_directory_processing(image_path)
chapter_names = sanitize_tree(image_path)
options['uuid'] = str(uuid4())
_log.prints(INFO, "Creating EPUB file...")
return build_epub(path, options, chapter_names)
def __init__(self):
self.log = default_log
self.current_frag = None
self.com_id = unicode_type(uuid4())
QWebEnginePage.__init__(self)
secure_webengine(self.settings(), for_viewer=True)
self.titleChanged.connect(self.title_changed)
self.loadFinished.connect(self.show_frag)
s = QWebEngineScript()
s.setName('toc.js')
s.setInjectionPoint(QWebEngineScript.InjectionPoint.DocumentReady)
s.setRunsOnSubFrames(True)
s.setWorldId(QWebEngineScript.ScriptWorldId.ApplicationWorld)
s.setSourceCode(P('toc.js', allow_user_override=False, data=True).decode('utf-8').replace('COM_ID', self.com_id))
self.scripts().insert(s)
def prepare_convert(temp_path, key, st, book_path):
tdir = tempfile.mkdtemp(dir=temp_path, prefix=f'c{next(td_counter)}-')
now = time.time()
return {
'path': os.path.basename(tdir),
'id': uuid4(),
'status': 'working',
'mtime': now,
'atime': now,
'key': key,
'file_mtime': st.st_mtime,
'file_size': st.st_size,
'cache_size': 0,
'book_path': book_path,
}
def __init__(self, metrics, num, objects, compress):
self.metrics, self.compress = metrics, compress
self.is_otf = self.metrics.is_otf
self.subset_tag = unicode_type(
re.sub('.', lambda m: chr(int(m.group())+ord('A')), oct(num).replace('o', '')
)).rjust(6, 'A')
self.font_stream = FontStream(metrics.is_otf, compress=compress)
try:
psname = metrics.postscript_name
except Exception:
psname = uuid4()
self.font_descriptor = Dictionary({
'Type': Name('FontDescriptor'),
'FontName': Name('%s+%s'%(self.subset_tag, psname)),
'Flags': 0b100, # Symbolic font
'FontBBox': Array(metrics.pdf_bbox),
'ItalicAngle': metrics.post.italic_angle,
'Ascent': metrics.pdf_ascent,
'Descent': metrics.pdf_descent,
'CapHeight': metrics.pdf_capheight,
'AvgWidth': metrics.pdf_avg_width,
'StemV': metrics.pdf_stemv,
})
self.descendant_font = Dictionary({
'Type':Name('Font'),
'Subtype':Name('CIDFontType' + ('0' if metrics.is_otf else '2')),
'BaseFont': self.font_descriptor['FontName'],
'FontDescriptor':objects.add(self.font_descriptor),
'CIDSystemInfo':Dictionary({
'Registry':String('Adobe'),
'Ordering':String('Identity'),
'Supplement':0,
}),
})
if not self.is_otf:
self.descendant_font['CIDToGIDMap'] = Name('Identity')
self.font_dict = Dictionary({
'Type':Name('Font'),
'Subtype':Name('Type0'),
'Encoding':Name('Identity-H'),
'BaseFont':self.descendant_font['BaseFont'],
'DescendantFonts':Array([objects.add(self.descendant_font)]),
})
self.used_glyphs = set()
def __init__(self, metrics, num, objects, compress):
self.metrics, self.compress = metrics, compress
self.is_otf = self.metrics.is_otf
self.subset_tag = str(
re.sub('.', lambda m: codepoint_to_chr(int(m.group())+ord('A')), oct(num).replace('o', '')
)).rjust(6, 'A')
self.font_stream = FontStream(metrics.is_otf, compress=compress)
try:
psname = metrics.postscript_name
except Exception:
psname = uuid4()
self.font_descriptor = Dictionary({
'Type': Name('FontDescriptor'),
'FontName': Name('%s+%s'%(self.subset_tag, psname)),
'Flags': 0b100, # Symbolic font
'FontBBox': Array(metrics.pdf_bbox),
'ItalicAngle': metrics.post.italic_angle,
'Ascent': metrics.pdf_ascent,
'Descent': metrics.pdf_descent,
'CapHeight': metrics.pdf_capheight,
'AvgWidth': metrics.pdf_avg_width,
'StemV': metrics.pdf_stemv,
})
self.descendant_font = Dictionary({
'Type':Name('Font'),
'Subtype':Name('CIDFontType' + ('0' if metrics.is_otf else '2')),
'BaseFont': self.font_descriptor['FontName'],
'FontDescriptor':objects.add(self.font_descriptor),
'CIDSystemInfo':Dictionary({
'Registry':String('Adobe'),
'Ordering':String('Identity'),
'Supplement':0,
}),
})
if not self.is_otf:
self.descendant_font['CIDToGIDMap'] = Name('Identity')
self.font_dict = Dictionary({
'Type':Name('Font'),
'Subtype':Name('Type0'),
'Encoding':Name('Identity-H'),
'BaseFont':self.descendant_font['BaseFont'],
'DescendantFonts':Array([objects.add(self.descendant_font)]),
})
self.used_glyphs = set()
def __init__(self, path_to_ebook, tdir, log=None, book_hash=None):
log = log or default_log
book_fmt, opfpath, input_fmt = extract_book(path_to_ebook, tdir, log=log)
ContainerBase.__init__(self, tdir, opfpath, log)
# We do not add zero byte sized files as the IndexedDB API in the
# browser has no good way to distinguish between zero byte files and
# load failures.
excluded_names = {
name for name, mt in iteritems(self.mime_map) if
name == self.opf_name or mt == guess_type('a.ncx') or name.startswith('META-INF/') or
name == 'mimetype' or not self.has_name_and_is_not_empty(name)}
raster_cover_name, titlepage_name = self.create_cover_page(input_fmt.lower())
toc = get_toc(self).to_dict(count())
spine = [name for name, is_linear in self.spine_names]
spineq = frozenset(spine)
landmarks = [l for l in get_landmarks(self) if l['dest'] in spineq]
self.book_render_data = data = {
'version': RENDER_VERSION,
'toc':toc,
'spine':spine,
'link_uid': uuid4(),
'book_hash': book_hash,
'is_comic': input_fmt.lower() in {'cbc', 'cbz', 'cbr', 'cb7'},
'raster_cover_name': raster_cover_name,
'title_page_name': titlepage_name,
'has_maths': False,
'total_length': 0,
'spine_length': 0,
'toc_anchor_map': toc_anchor_map(toc),
'landmarks': landmarks,
'link_to_map': {},
}
# Mark the spine as dirty since we have to ensure it is normalized
for name in data['spine']:
self.parsed(name), self.dirty(name)
self.transform_css()
self.virtualized_names = set()
self.virtualize_resources()
def manifest_data(name):
mt = (self.mime_map.get(name) or 'application/octet-stream').lower()
ans = {
'size':os.path.getsize(self.name_path_map[name]),
'is_virtualized': name in self.virtualized_names,
'mimetype':mt,
'is_html': mt in OEB_DOCS,
}
if ans['is_html']:
root = self.parsed(name)
ans['length'] = l = get_length(root)
self.book_render_data['total_length'] += l
if name in data['spine']:
self.book_render_data['spine_length'] += l
ans['has_maths'] = hm = check_for_maths(root)
if hm:
self.book_render_data['has_maths'] = True
ans['anchor_map'] = anchor_map(root)
return ans
data['files'] = {name:manifest_data(name) for name in set(self.name_path_map) - excluded_names}
self.commit()
for name in excluded_names:
os.remove(self.name_path_map[name])
data = json.dumps(self.book_render_data, ensure_ascii=False)
if not isinstance(data, bytes):
data = data.encode('utf-8')
with lopen(os.path.join(self.root, 'calibre-book-manifest.json'), 'wb') as f:
f.write(data)
def __init__(self, path_to_ebook, tdir, log=None, book_hash=None):
log = log or default_log
book_fmt, opfpath, input_fmt = extract_book(path_to_ebook,
tdir,
log=log)
ContainerBase.__init__(self, tdir, opfpath, log)
excluded_names = {
name
for name, mt in self.mime_map.iteritems()
if name == self.opf_name or mt == guess_type('a.ncx')
or name.startswith('META-INF/') or name == 'mimetype'
}
raster_cover_name, titlepage_name = self.create_cover_page(
input_fmt.lower())
toc = get_toc(self).to_dict(count())
spine = [name for name, is_linear in self.spine_names]
spineq = frozenset(spine)
landmarks = [l for l in get_landmarks(self) if l['dest'] in spineq]
self.book_render_data = data = {
'version': RENDER_VERSION,
'toc': toc,
'spine': spine,
'link_uid': uuid4(),
'book_hash': book_hash,
'is_comic': input_fmt.lower() in {'cbc', 'cbz', 'cbr', 'cb7'},
'raster_cover_name': raster_cover_name,
'title_page_name': titlepage_name,
'has_maths': False,
'total_length': 0,
'spine_length': 0,
'toc_anchor_map': toc_anchor_map(toc),
'landmarks': landmarks,
'link_to_map': {},
}
# Mark the spine as dirty since we have to ensure it is normalized
for name in data['spine']:
self.parsed(name), self.dirty(name)
self.transform_css()
self.virtualized_names = set()
self.virtualize_resources()
def manifest_data(name):
mt = (self.mime_map.get(name)
or 'application/octet-stream').lower()
ans = {
'size': os.path.getsize(self.name_path_map[name]),
'is_virtualized': name in self.virtualized_names,
'mimetype': mt,
'is_html': mt in OEB_DOCS,
}
if ans['is_html']:
root = self.parsed(name)
ans['length'] = l = get_length(root)
self.book_render_data['total_length'] += l
if name in data['spine']:
self.book_render_data['spine_length'] += l
ans['has_maths'] = hm = check_for_maths(root)
if hm:
self.book_render_data['has_maths'] = True
ans['anchor_map'] = anchor_map(root)
return ans
data['files'] = {
name: manifest_data(name)
for name in set(self.name_path_map) - excluded_names
}
self.commit()
for name in excluded_names:
os.remove(self.name_path_map[name])
with lopen(os.path.join(self.root, 'calibre-book-manifest.json'),
'wb') as f:
f.write(
json.dumps(self.book_render_data,
ensure_ascii=False).encode('utf-8'))
def __init__(self,
path_to_ebook,
tdir,
log=None,
book_hash=None,
save_bookmark_data=False,
book_metadata=None):
log = log or default_log
book_fmt, opfpath, input_fmt = extract_book(path_to_ebook,
tdir,
log=log)
ContainerBase.__init__(self, tdir, opfpath, log)
self.book_metadata = book_metadata
if save_bookmark_data:
bm_file = 'META-INF/calibre_bookmarks.txt'
self.bookmark_data = None
if self.exists(bm_file):
with self.open(bm_file, 'rb') as f:
self.bookmark_data = f.read()
# We do not add zero byte sized files as the IndexedDB API in the
# browser has no good way to distinguish between zero byte files and
# load failures.
excluded_names = {
name
for name, mt in iteritems(self.mime_map) if name == self.opf_name
or mt == guess_type('a.ncx') or name.startswith('META-INF/')
or name == 'mimetype' or not self.has_name_and_is_not_empty(name)
}
raster_cover_name, titlepage_name = self.create_cover_page(
input_fmt.lower())
toc = get_toc(self).to_dict(count())
if not toc or not toc.get('children'):
toc = from_xpaths(self,
['//h:h1', '//h:h2', '//h:h3']).to_dict(count())
spine = [name for name, is_linear in self.spine_names]
spineq = frozenset(spine)
landmarks = [l for l in get_landmarks(self) if l['dest'] in spineq]
self.book_render_data = data = {
'version': RENDER_VERSION,
'toc': toc,
'book_format': book_fmt,
'spine': spine,
'link_uid': uuid4(),
'book_hash': book_hash,
'is_comic': input_fmt.lower() in {'cbc', 'cbz', 'cbr', 'cb7'},
'raster_cover_name': raster_cover_name,
'title_page_name': titlepage_name,
'has_maths': False,
'total_length': 0,
'spine_length': 0,
'toc_anchor_map': toc_anchor_map(toc),
'landmarks': landmarks,
'link_to_map': {},
}
# Mark the spine as dirty since we have to ensure it is normalized
for name in data['spine']:
self.parsed(name), self.dirty(name)
self.transform_css()
self.virtualized_names = set()
self.virtualize_resources()
def manifest_data(name):
mt = (self.mime_map.get(name)
or 'application/octet-stream').lower()
ans = {
'size': os.path.getsize(self.name_path_map[name]),
'is_virtualized': name in self.virtualized_names,
'mimetype': mt,
'is_html': mt in OEB_DOCS,
}
if ans['is_html']:
root = self.parsed(name)
ans['length'] = l = get_length(root)
self.book_render_data['total_length'] += l
if name in data['spine']:
self.book_render_data['spine_length'] += l
ans['has_maths'] = hm = check_for_maths(root)
if hm:
self.book_render_data['has_maths'] = True
ans['anchor_map'] = anchor_map(root)
return ans
data['files'] = {
name: manifest_data(name)
for name in set(self.name_path_map) - excluded_names
}
self.commit()
for name in excluded_names:
os.remove(self.name_path_map[name])
data = json.dumps(self.book_render_data, ensure_ascii=False)
if not isinstance(data, bytes):
data = data.encode('utf-8')
with lopen(os.path.join(self.root, 'calibre-book-manifest.json'),
'wb') as f:
f.write(data)
def process_exploded_book(book_fmt,
opfpath,
input_fmt,
tdir,
render_manager,
log=None,
book_hash=None,
save_bookmark_data=False,
book_metadata=None,
virtualize_resources=True):
log = log or default_log
container = SimpleContainer(tdir, opfpath, log)
input_plugin = plugin_for_input_format(input_fmt)
is_comic = bool(getattr(input_plugin, 'is_image_collection', False))
def needs_work(mt):
return mt in OEB_STYLES or mt in OEB_DOCS or mt == 'image/svg+xml'
def work_priority(name):
# ensure workers with large files or stylesheets
# have the less names
size = os.path.getsize(container.name_path_map[name]),
is_html = container.mime_map.get(name) in OEB_DOCS
return (0 if is_html else 1), size
if not is_comic:
render_manager.launch_workers(
tuple(n for n, mt in iteritems(container.mime_map)
if needs_work(mt)), container)
bookmark_data = None
if save_bookmark_data:
bm_file = 'META-INF/calibre_bookmarks.txt'
if container.exists(bm_file):
with container.open(bm_file, 'rb') as f:
bookmark_data = f.read()
# We do not add zero byte sized files as the IndexedDB API in the
# browser has no good way to distinguish between zero byte files and
# load failures.
excluded_names = {
name
for name, mt in iteritems(container.mime_map)
if name == container.opf_name or mt == guess_type('a.ncx')
or name.startswith('META-INF/') or name == 'mimetype'
or not container.has_name_and_is_not_empty(name)
}
raster_cover_name, titlepage_name = create_cover_page(
container, input_fmt.lower(), is_comic, book_metadata)
toc = get_toc(container, verify_destinations=False).to_dict(count())
if not toc or not toc.get('children'):
toc = from_xpaths(container,
['//h:h1', '//h:h2', '//h:h3']).to_dict(count())
spine = [name for name, is_linear in container.spine_names]
spineq = frozenset(spine)
landmarks = [l for l in get_landmarks(container) if l['dest'] in spineq]
book_render_data = {
'version': RENDER_VERSION,
'toc': toc,
'book_format': book_fmt,
'spine': spine,
'link_uid': uuid4(),
'book_hash': book_hash,
'is_comic': is_comic,
'raster_cover_name': raster_cover_name,
'title_page_name': titlepage_name,
'has_maths': False,
'total_length': 0,
'spine_length': 0,
'toc_anchor_map': toc_anchor_map(toc),
'landmarks': landmarks,
'link_to_map': {},
}
names = sorted(
(n for n, mt in iteritems(container.mime_map) if needs_work(mt)),
key=work_priority)
results = render_manager(
names,
(tdir, opfpath, virtualize_resources, book_render_data['link_uid'],
container.data_for_clone()), container)
ltm = book_render_data['link_to_map']
html_data = {}
virtualized_names = set()
def merge_ltm(dest, src):
for k, v in iteritems(src):
if k in dest:
dest[k] |= v
else:
dest[k] = v
for link_to_map, hdata, vnames in results:
html_data.update(hdata)
virtualized_names |= vnames
for k, v in iteritems(link_to_map):
if k in ltm:
merge_ltm(ltm[k], v)
else:
ltm[k] = v
def manifest_data(name):
mt = (container.mime_map.get(name)
or 'application/octet-stream').lower()
ans = {
'size': os.path.getsize(container.name_path_map[name]),
'is_virtualized': name in virtualized_names,
'mimetype': mt,
'is_html': mt in OEB_DOCS,
}
if ans['is_html']:
data = html_data[name]
ans['length'] = l = data['length']
book_render_data['total_length'] += l
if name in book_render_data['spine']:
book_render_data['spine_length'] += l
ans['has_maths'] = hm = data['has_maths']
if hm:
book_render_data['has_maths'] = True
ans['anchor_map'] = data['anchor_map']
return ans
book_render_data['files'] = {
name: manifest_data(name)
for name in set(container.name_path_map) - excluded_names
}
container.commit()
for name in excluded_names:
os.remove(container.name_path_map[name])
ltm = book_render_data['link_to_map']
for name, amap in iteritems(ltm):
for k, v in tuple(iteritems(amap)):
amap[k] = tuple(v) # needed for JSON serialization
data = as_bytes(json.dumps(book_render_data, ensure_ascii=False))
with lopen(os.path.join(container.root, 'calibre-book-manifest.json'),
'wb') as f:
f.write(data)
return container, bookmark_data
def do_add(dbctx, paths, one_book_per_directory, recurse, add_duplicates,
otitle, oauthors, oisbn, otags, oseries, oseries_index, ocover,
oidentifiers, olanguages, compiled_rules, oautomerge):
request_id = uuid4()
with add_ctx():
files, dirs = [], []
for path in paths:
path = os.path.abspath(path)
if os.path.isdir(path):
dirs.append(path)
else:
if os.path.exists(path):
files.append(path)
else:
prints(path, 'not found')
file_duplicates, added_ids, merged_ids = [], set(), set()
for book in files:
fmt = os.path.splitext(book)[1]
fmt = fmt[1:] if fmt else None
if not fmt:
continue
aids, mids, dups, book_title = dbctx.run(
'add', 'book', dbctx.path(book), os.path.basename(book), fmt,
add_duplicates, otitle, oauthors, oisbn, otags, oseries,
oseries_index,
serialize_cover(ocover) if ocover else None, oidentifiers,
olanguages, oautomerge, request_id)
added_ids |= set(aids)
merged_ids |= set(mids)
if dups:
file_duplicates.append((book_title, book))
dir_dups = []
scanner = cdb_recursive_find if recurse else cdb_find_in_dir
for dpath in dirs:
for formats in scanner(dpath, one_book_per_directory,
compiled_rules):
cover_data = None
for fmt in formats:
if fmt.lower().endswith('.opf'):
with lopen(fmt, 'rb') as f:
mi = get_metadata(f, stream_type='opf')
if mi.cover_data and mi.cover_data[1]:
cover_data = mi.cover_data[1]
elif mi.cover:
try:
with lopen(mi.cover, 'rb') as f:
cover_data = f.read()
except OSError:
pass
book_title, ids, mids, dups = dbctx.run(
'add', 'format_group', tuple(map(dbctx.path, formats)),
add_duplicates, oautomerge, request_id, cover_data)
if book_title is not None:
added_ids |= set(ids)
merged_ids |= set(mids)
if dups:
dir_dups.append((book_title, formats))
sys.stdout = sys.__stdout__
if dir_dups or file_duplicates:
prints(_('The following books were not added as '
'they already exist in the database '
'(see --duplicates option or --automerge option):'),
file=sys.stderr)
for title, formats in dir_dups:
prints(' ', title, file=sys.stderr)
for path in formats:
prints(' ', path)
if file_duplicates:
for title, path in file_duplicates:
prints(' ', title, file=sys.stderr)
prints(' ', path)
if added_ids:
prints(_('Added book ids: %s') % (', '.join(map(str, added_ids))))
if merged_ids:
prints(
_('Merged book ids: %s') % (', '.join(map(str, merged_ids))))