def check_filenames(container):
errors = []
all_names = set(container.name_path_map) - container.names_that_must_not_be_changed
for name in all_names:
if urlquote(name) != name:
errors.append(EscapedName(name))
return errors
def name_to_href(self, name, base=None):
'''Convert a name to a href relative to base, which must be a name or
None in which case self.root is used as the base'''
fullpath = self.name_to_abspath(name)
basepath = self.root if base is None else os.path.dirname(self.name_to_abspath(base))
path = relpath(fullpath, basepath).replace(os.sep, '/')
return urlquote(path)
def check_filenames(container):
errors = []
all_names = set(container.name_path_map) - container.names_that_must_not_be_changed
for name in all_names:
if urlquote(name) != name:
errors.append(EscapedName(name))
return errors
def read_image(self, href):
if href not in self.images:
item = self.oeb.manifest.hrefs.get(
href) or self.oeb.manifest.hrefs.get(urlquote(href))
try:
if item is None or not isinstance(item.data, bytes):
self.log.warning('Failed to find image:', href)
return
except FileNotFoundError:
self.log.warning('Failed to find image:', href)
return
try:
fmt, width, height = identify(item.data)
except Exception:
self.log.warning('Replacing corrupted image with blank: %s' %
href)
item.data = I('blank.png',
data=True,
allow_user_override=False)
fmt, width, height = identify(item.data)
image_fname = 'media/' + self.create_filename(href, fmt)
image_rid = self.document_relationships.add_image(image_fname)
self.images[href] = Image(image_rid, image_fname, width, height,
fmt, item)
item.unload_data_from_memory()
return self.images[href]
def name_to_href(self, name, base=None):
'''Convert a name to a href relative to base, which must be a name or
None in which case self.root is used as the base'''
fullpath = self.name_to_abspath(name)
basepath = self.root if base is None else os.path.dirname(self.name_to_abspath(base))
path = relpath(fullpath, basepath).replace(os.sep, '/')
return urlquote(path)
def __init__(self, name):
BaseError.__init__(self, _('Filename contains unsafe characters'), name)
qname = urlquote(name)
self.sname = make_filename_safe(name)
self.HELP = _(
'The filename {0} contains unsafe characters, that must be escaped, like'
' this {1}. This can cause problems with some e-book readers. To be'
' absolutely safe, use only the English alphabet [a-z], the numbers [0-9],'
' underscores and hyphens in your file names. While many other characters'
' are allowed, they may cause problems with some software.').format(name, qname)
self.INDIVIDUAL_FIX = _(
'Rename the file {0} to {1}').format(name, self.sname)
def __init__(self, name):
BaseError.__init__(self, _('Filename contains unsafe characters'), name)
qname = urlquote(name)
self.sname = make_filename_safe(name)
self.HELP = _(
'The filename {0} contains unsafe characters, that must be escaped, like'
' this {1}. This can cause problems with some e-book readers. To be'
' absolutely safe, use only the English alphabet [a-z], the numbers [0-9],'
' underscores and hyphens in your file names. While many other characters'
' are allowed, they may cause problems with some software.').format(name, qname)
self.INDIVIDUAL_FIX = _(
'Rename the file {0} to {1}').format(name, self.sname)
def __init__(self, name):
from calibre.utils.filenames import ascii_filename
BaseError.__init__(self, _('Filename contains unsafe characters'), name)
qname = urlquote(name)
def esc(n):
return ''.join(x if x in URL_SAFE else '_' for x in n)
self.sname = '/'.join(esc(ascii_filename(x)) for x in name.split('/'))
self.HELP = _(
'The filename {0} contains unsafe characters, that must be escaped, like'
' this {1}. This can cause problems with some ebook readers. To be'
' absolutely safe, use only the English alphabet [a-z], the numbers [0-9],'
' underscores and hyphens in your file names. While many other characters'
' are allowed, they may cause problems with some software.').format(name, qname)
self.INDIVIDUAL_FIX = _(
'Rename the file {0} to {1}').format(name, self.sname)
def __init__(self, name):
from calibre.utils.filenames import ascii_filename
BaseError.__init__(self, _('Filename contains unsafe characters'), name)
qname = urlquote(name)
def esc(n):
return ''.join(x if x in URL_SAFE else '_' for x in n)
self.sname = '/'.join(esc(ascii_filename(x)) for x in name.split('/'))
self.HELP = _(
'The filename {0} contains unsafe characters, that must be escaped, like'
' this {1}. This can cause problems with some ebook readers. To be'
' absolutely safe, use only the English alphabet [a-z], the numbers [0-9],'
' underscores and hyphens in your file names. While many other characters'
' are allowed, they may cause problems with some software.').format(name, qname)
self.INDIVIDUAL_FIX = _(
'Rename the file {0} to {1}').format(name, self.sname)
def donode(item, parent, base, subpath):
for child in item:
title = child.title
if not title:
continue
raw = unquote_path(child.href or '')
rsrcname = os.path.basename(raw)
rsrcpath = os.path.join(subpath, rsrcname)
if (not os.path.exists(os.path.join(base, rsrcpath)) and os.path.exists(os.path.join(base, raw))):
rsrcpath = raw
if '%' not in rsrcpath:
rsrcpath = urlquote(rsrcpath)
if not raw:
rsrcpath = ''
c = DIV(A(title, href=rsrcpath))
donode(child, c, base, subpath)
parent.append(c)
def donode(item, parent, base, subpath):
for child in item:
title = child.title
if not title:
continue
raw = unquote_path(child.href or '')
rsrcname = os.path.basename(raw)
rsrcpath = os.path.join(subpath, rsrcname)
if (not os.path.exists(os.path.join(base, rsrcpath)) and os.path.exists(os.path.join(base, raw))):
rsrcpath = raw
if '%' not in rsrcpath:
rsrcpath = urlquote(rsrcpath)
if not raw:
rsrcpath = ''
c = DIV(A(title, href=rsrcpath))
donode(child, c, base, subpath)
parent.append(c)
def serialize_hyperlink(self, parent, link):
item, url, tooltip = link
purl = urlparse(url)
href = purl.path
def make_link(parent, anchor=None, id=None, tooltip=None):
kw = {}
if anchor is not None:
kw['w_anchor'] = anchor
elif id is not None:
kw['r_id'] = id
if tooltip:
kw['w_tooltip'] = tooltip
return self.namespace.makeelement(parent, 'w:hyperlink', **kw)
if not purl.scheme:
href = item.abshref(href)
if href not in self.document_hrefs:
href = urlquote(href)
if href in self.document_hrefs:
key = (href, purl.fragment or self.top_anchor)
if key in self.anchor_map:
bmark = self.anchor_map[key]
else:
bmark = self.anchor_map[(href, self.top_anchor)]
return make_link(parent, anchor=bmark, tooltip=tooltip)
else:
self.log.warn(
'Ignoring internal hyperlink with href (%s) pointing to unknown destination'
% url)
if purl.scheme in {'http', 'https', 'ftp'}:
if url not in self.external_links:
self.external_links[
url] = self.document_relationships.add_relationship(
url,
self.namespace.names['LINKS'],
target_mode='External')
return make_link(parent,
id=self.external_links[url],
tooltip=tooltip)
return parent
def create_oebbook(self, htmlpath, basedir, opts, log, mi):
import uuid
from calibre.ebooks.conversion.plumber import create_oebbook
from calibre.ebooks.oeb.base import (DirContainer, rewrite_links,
urlnormalize, urldefrag,
BINARY_MIME, OEB_STYLES, xpath,
urlquote)
from calibre import guess_type
from calibre.ebooks.oeb.transforms.metadata import \
meta_info_to_oeb_metadata
from calibre.ebooks.html.input import get_filelist
from calibre.ebooks.metadata import string_to_authors
from calibre.utils.localization import canonicalize_lang
import css_parser, logging
css_parser.log.setLevel(logging.WARN)
self.OEB_STYLES = OEB_STYLES
oeb = create_oebbook(log,
None,
opts,
self,
encoding=opts.input_encoding,
populate=False)
self.oeb = oeb
metadata = oeb.metadata
meta_info_to_oeb_metadata(mi, metadata, log)
if not metadata.language:
l = canonicalize_lang(getattr(opts, 'language', None))
if not l:
oeb.logger.warn('Language not specified')
l = get_lang().replace('_', '-')
metadata.add('language', l)
if not metadata.creator:
a = getattr(opts, 'authors', None)
if a:
a = string_to_authors(a)
if not a:
oeb.logger.warn('Creator not specified')
a = [self.oeb.translate(__('Unknown'))]
for aut in a:
metadata.add('creator', aut)
if not metadata.title:
oeb.logger.warn('Title not specified')
metadata.add('title', self.oeb.translate(__('Unknown')))
bookid = unicode_type(uuid.uuid4())
metadata.add('identifier', bookid, id='uuid_id', scheme='uuid')
for ident in metadata.identifier:
if 'id' in ident.attrib:
self.oeb.uid = metadata.identifier[0]
break
filelist = get_filelist(htmlpath, basedir, opts, log)
filelist = [f for f in filelist if not f.is_binary]
htmlfile_map = {}
for f in filelist:
path = f.path
oeb.container = DirContainer(os.path.dirname(path),
log,
ignore_opf=True)
bname = os.path.basename(path)
id, href = oeb.manifest.generate(id='html',
href=sanitize_file_name(bname))
htmlfile_map[path] = href
item = oeb.manifest.add(id, href, 'text/html')
if path == htmlpath and '%' in path:
bname = urlquote(bname)
item.html_input_href = bname
oeb.spine.add(item, True)
self.added_resources = {}
self.log = log
self.log('Normalizing filename cases')
for path, href in htmlfile_map.items():
if not self.is_case_sensitive(path):
path = path.lower()
self.added_resources[path] = href
self.urlnormalize, self.DirContainer = urlnormalize, DirContainer
self.urldefrag = urldefrag
self.guess_type, self.BINARY_MIME = guess_type, BINARY_MIME
self.log('Rewriting HTML links')
for f in filelist:
path = f.path
dpath = os.path.dirname(path)
oeb.container = DirContainer(dpath, log, ignore_opf=True)
href = htmlfile_map[path]
try:
item = oeb.manifest.hrefs[href]
except KeyError:
item = oeb.manifest.hrefs[urlnormalize(href)]
rewrite_links(item.data, partial(self.resource_adder, base=dpath))
for item in oeb.manifest.values():
if item.media_type in self.OEB_STYLES:
dpath = None
for path, href in self.added_resources.items():
if href == item.href:
dpath = os.path.dirname(path)
break
css_parser.replaceUrls(
item.data, partial(self.resource_adder, base=dpath))
toc = self.oeb.toc
self.oeb.auto_generated_toc = True
titles = []
headers = []
for item in self.oeb.spine:
if not item.linear:
continue
html = item.data
title = ''.join(xpath(html, '/h:html/h:head/h:title/text()'))
title = re.sub(r'\s+', ' ', title.strip())
if title:
titles.append(title)
headers.append('(unlabled)')
for tag in ('h1', 'h2', 'h3', 'h4', 'h5', 'strong'):
expr = '/h:html/h:body//h:%s[position()=1]/text()'
header = ''.join(xpath(html, expr % tag))
header = re.sub(r'\s+', ' ', header.strip())
if header:
headers[-1] = header
break
use = titles
if len(titles) > len(set(titles)):
use = headers
for title, item in zip(use, self.oeb.spine):
if not item.linear:
continue
toc.add(title, item.href)
oeb.container = DirContainer(getcwd(), oeb.log, ignore_opf=True)
return oeb
def create_oebbook(self, htmlpath, basedir, opts, log, mi):
import uuid
from calibre.ebooks.conversion.plumber import create_oebbook
from calibre.ebooks.oeb.base import (DirContainer,
rewrite_links, urlnormalize, urldefrag, BINARY_MIME, OEB_STYLES,
xpath, urlquote)
from calibre import guess_type
from calibre.ebooks.oeb.transforms.metadata import \
meta_info_to_oeb_metadata
from calibre.ebooks.html.input import get_filelist
from calibre.ebooks.metadata import string_to_authors
from calibre.utils.localization import canonicalize_lang
import css_parser, logging
css_parser.log.setLevel(logging.WARN)
self.OEB_STYLES = OEB_STYLES
oeb = create_oebbook(log, None, opts, self,
encoding=opts.input_encoding, populate=False)
self.oeb = oeb
metadata = oeb.metadata
meta_info_to_oeb_metadata(mi, metadata, log)
if not metadata.language:
l = canonicalize_lang(getattr(opts, 'language', None))
if not l:
oeb.logger.warn(u'Language not specified')
l = get_lang().replace('_', '-')
metadata.add('language', l)
if not metadata.creator:
a = getattr(opts, 'authors', None)
if a:
a = string_to_authors(a)
if not a:
oeb.logger.warn('Creator not specified')
a = [self.oeb.translate(__('Unknown'))]
for aut in a:
metadata.add('creator', aut)
if not metadata.title:
oeb.logger.warn('Title not specified')
metadata.add('title', self.oeb.translate(__('Unknown')))
bookid = str(uuid.uuid4())
metadata.add('identifier', bookid, id='uuid_id', scheme='uuid')
for ident in metadata.identifier:
if 'id' in ident.attrib:
self.oeb.uid = metadata.identifier[0]
break
filelist = get_filelist(htmlpath, basedir, opts, log)
filelist = [f for f in filelist if not f.is_binary]
htmlfile_map = {}
for f in filelist:
path = f.path
oeb.container = DirContainer(os.path.dirname(path), log,
ignore_opf=True)
bname = os.path.basename(path)
id, href = oeb.manifest.generate(id='html', href=sanitize_file_name(bname))
htmlfile_map[path] = href
item = oeb.manifest.add(id, href, 'text/html')
if path == htmlpath and '%' in path:
bname = urlquote(bname)
item.html_input_href = bname
oeb.spine.add(item, True)
self.added_resources = {}
self.log = log
self.log('Normalizing filename cases')
for path, href in htmlfile_map.items():
if not self.is_case_sensitive(path):
path = path.lower()
self.added_resources[path] = href
self.urlnormalize, self.DirContainer = urlnormalize, DirContainer
self.urldefrag = urldefrag
self.guess_type, self.BINARY_MIME = guess_type, BINARY_MIME
self.log('Rewriting HTML links')
for f in filelist:
path = f.path
dpath = os.path.dirname(path)
oeb.container = DirContainer(dpath, log, ignore_opf=True)
href = htmlfile_map[path]
try:
item = oeb.manifest.hrefs[href]
except KeyError:
item = oeb.manifest.hrefs[urlnormalize(href)]
rewrite_links(item.data, partial(self.resource_adder, base=dpath))
for item in oeb.manifest.values():
if item.media_type in self.OEB_STYLES:
dpath = None
for path, href in self.added_resources.items():
if href == item.href:
dpath = os.path.dirname(path)
break
css_parser.replaceUrls(item.data,
partial(self.resource_adder, base=dpath))
toc = self.oeb.toc
self.oeb.auto_generated_toc = True
titles = []
headers = []
for item in self.oeb.spine:
if not item.linear:
continue
html = item.data
title = ''.join(xpath(html, '/h:html/h:head/h:title/text()'))
title = re.sub(r'\s+', ' ', title.strip())
if title:
titles.append(title)
headers.append('(unlabled)')
for tag in ('h1', 'h2', 'h3', 'h4', 'h5', 'strong'):
expr = '/h:html/h:body//h:%s[position()=1]/text()'
header = ''.join(xpath(html, expr % tag))
header = re.sub(r'\s+', ' ', header.strip())
if header:
headers[-1] = header
break
use = titles
if len(titles) > len(set(titles)):
use = headers
for title, item in zip(use, self.oeb.spine):
if not item.linear:
continue
toc.add(title, item.href)
oeb.container = DirContainer(getcwd(), oeb.log, ignore_opf=True)
return oeb
