def href(self, basedir=None):
'''
Return a URL pointing to this resource. If it is a file on the filesystem
the URL is relative to `basedir`.
`basedir`: If None, the basedir of this resource is used (see :method:`set_basedir`).
If this resource has no basedir, then the current working directory is used as the basedir.
'''
if basedir is None:
if self._basedir:
basedir = self._basedir
else:
basedir = os.getcwdu()
if self.path is None:
return self._href
f = self.fragment.encode('utf-8') if isinstance(
self.fragment, unicode_type) else self.fragment
frag = '#' + quote(f) if self.fragment else ''
if self.path == basedir:
return '' + frag
try:
rpath = relpath(self.path, basedir)
except OSError: # On windows path and basedir could be on different drives
rpath = self.path
if isinstance(rpath, unicode_type):
rpath = rpath.encode('utf-8')
return quote(rpath.replace(os.sep, '/')) + frag
def href(self, basedir=None):
'''
Return a URL pointing to this resource. If it is a file on the filesystem
the URL is relative to `basedir`.
`basedir`: If None, the basedir of this resource is used (see :method:`set_basedir`).
If this resource has no basedir, then the current working directory is used as the basedir.
'''
if basedir is None:
if self._basedir:
basedir = self._basedir
else:
basedir = os.getcwdu()
if self.path is None:
return self._href
f = self.fragment.encode('utf-8') if isinstance(self.fragment, unicode_type) else self.fragment
frag = '#'+quote(f) if self.fragment else ''
if self.path == basedir:
return ''+frag
try:
rpath = relpath(self.path, basedir)
except OSError: # On windows path and basedir could be on different drives
rpath = self.path
if isinstance(rpath, unicode_type):
rpath = rpath.encode('utf-8')
return quote(rpath.replace(os.sep, '/'))+frag
def link_replacer(base, url):
if url.startswith('#'):
frag = urlunquote(url[1:])
if not frag:
return url
changed.add(base)
return resource_template.format(encode_url(base, frag))
purl = urlparse(url)
if purl.netloc or purl.query:
return url
if purl.scheme and purl.scheme != 'file':
return url
if not purl.path or purl.path.startswith('/'):
return url
url, frag = purl.path, purl.fragment
name = container.href_to_name(url, base)
if name:
if container.has_name_and_is_not_empty(name):
frag = urlunquote(frag)
url = resource_template.format(encode_url(name, frag))
else:
if isinstance(name, unicode_type):
name = name.encode('utf-8')
url = 'missing:' + force_unicode(quote(name), 'utf-8')
changed.add(base)
return url
def link_replacer(base, url):
if url.startswith('#'):
frag = urlunquote(url[1:])
if not frag:
return url
changed.add(base)
return resource_template.format(encode_url(base, frag))
purl = urlparse(url)
if purl.netloc or purl.query:
return url
if purl.scheme and purl.scheme != 'file':
return url
if not purl.path or purl.path.startswith('/'):
return url
url, frag = purl.path, purl.fragment
name = self.href_to_name(url, base)
if name:
if self.has_name_and_is_not_empty(name):
frag = urlunquote(frag)
url = resource_template.format(encode_url(name, frag))
else:
if isinstance(name, unicode_type):
name = name.encode('utf-8')
url = 'missing:' + force_unicode(quote(name), 'utf-8')
changed.add(base)
return url
def qquote(val, use_plus=True):
if not isinstance(val, bytes):
val = val.encode('utf-8')
ans = quote_plus(val) if use_plus else quote(val)
if isinstance(ans, bytes):
ans = ans.decode('utf-8')
return ans
def resource_adder(self, link_, base=None):
from polyglot.urllib import quote
link, frag = self.link_to_local_path(link_, base=base)
if link is None:
return link_
try:
if base and not os.path.isabs(link):
link = os.path.join(base, link)
link = os.path.abspath(link)
except:
return link_
if not os.access(link, os.R_OK):
return link_
if os.path.isdir(link):
self.log.warn(link_, 'is a link to a directory. Ignoring.')
return link_
if not self.is_case_sensitive(tempfile.gettempdir()):
link = link.lower()
if link not in self.added_resources:
bhref = os.path.basename(link)
id, href = self.oeb.manifest.generate(
id='added', href=sanitize_file_name(bhref))
guessed = self.guess_type(href)[0]
media_type = guessed or self.BINARY_MIME
if media_type == 'text/plain':
self.log.warn('Ignoring link to text file %r' % link_)
return None
if media_type == self.BINARY_MIME:
# Check for the common case, images
try:
img = what(link)
except EnvironmentError:
pass
else:
if img:
media_type = self.guess_type(
'dummy.' + img)[0] or self.BINARY_MIME
self.oeb.log.debug('Added', link)
self.oeb.container = self.DirContainer(os.path.dirname(link),
self.oeb.log,
ignore_opf=True)
# Load into memory
item = self.oeb.manifest.add(id, href, media_type)
# bhref refers to an already existing file. The read() method of
# DirContainer will call unquote on it before trying to read the
# file, therefore we quote it here.
if isinstance(bhref, unicode_type):
bhref = bhref.encode('utf-8')
item.html_input_href = as_unicode(quote(bhref))
if guessed in self.OEB_STYLES:
item.override_css_fetch = partial(self.css_import_handler,
os.path.dirname(link))
item.data
self.added_resources[link] = href
nlink = self.added_resources[link]
if frag:
nlink = '#'.join((nlink, frag))
return nlink
def canonicalize_url(url):
# mechanize does not handle quoting automatically
if re.search(r'\s+', url) is not None:
purl = list(urlparse(url))
for i in range(2, 6):
purl[i] = quote(purl[i])
url = urlunparse(purl)
return url
def fetch_url(self, url):
data = None
self.log.debug('Fetching', url)
st = time.time()
# Check for a URL pointing to the local filesystem and special case it
# for efficiency and robustness. Bypasses delay checking as it does not
# apply to local fetches. Ensures that unicode paths that are not
# representable in the filesystem_encoding work.
is_local = 0
if url.startswith('file://'):
is_local = 7
elif url.startswith('file:'):
is_local = 5
if is_local > 0:
url = url[is_local:]
if iswindows and url.startswith('/'):
url = url[1:]
with open(url, 'rb') as f:
data = response(f.read())
data.newurl = 'file:'+url # This is what mechanize does for
# local URLs
self.log.debug('Fetched %s in %.1f seconds' % (url, time.time() - st))
return data
delta = time.time() - self.last_fetch_at
if delta < self.delay:
time.sleep(self.delay - delta)
# mechanize does not handle quoting automatically
if re.search(r'\s+', url) is not None:
if isinstance(url, unicode_type):
url = url.encode('utf-8')
purl = list(urlparse(url))
for i in range(2, 6):
purl[i] = quote(purl[i])
url = urlunparse(purl).decode('utf-8')
open_func = getattr(self.browser, 'open_novisit', self.browser.open)
try:
with closing(open_func(url, timeout=self.timeout)) as f:
data = response(f.read()+f.read())
data.newurl = f.geturl()
except URLError as err:
if hasattr(err, 'code') and err.code in responses:
raise FetchError(responses[err.code])
if getattr(err, 'reason', [0])[0] == 104 or \
getattr(getattr(err, 'args', [None])[0], 'errno', None) in (-2,
-3): # Connection reset by peer or Name or service not known
self.log.debug('Temporary error, retrying in 1 second')
time.sleep(1)
with closing(open_func(url, timeout=self.timeout)) as f:
data = response(f.read()+f.read())
data.newurl = f.geturl()
else:
raise err
finally:
self.last_fetch_at = time.time()
self.log.debug('Fetched %s in %f seconds' % (url, time.time() - st))
return data
def resource_adder(self, link_, base=None):
from polyglot.urllib import quote
link, frag = self.link_to_local_path(link_, base=base)
if link is None:
return link_
try:
if base and not os.path.isabs(link):
link = os.path.join(base, link)
link = os.path.abspath(link)
except:
return link_
if not os.access(link, os.R_OK):
return link_
if os.path.isdir(link):
self.log.warn(link_, 'is a link to a directory. Ignoring.')
return link_
if not self.is_case_sensitive(tempfile.gettempdir()):
link = link.lower()
if link not in self.added_resources:
bhref = os.path.basename(link)
id, href = self.oeb.manifest.generate(id='added', href=sanitize_file_name(bhref))
guessed = self.guess_type(href)[0]
media_type = guessed or self.BINARY_MIME
if media_type == 'text/plain':
self.log.warn('Ignoring link to text file %r'%link_)
return None
if media_type == self.BINARY_MIME:
# Check for the common case, images
try:
img = what(link)
except EnvironmentError:
pass
else:
if img:
media_type = self.guess_type('dummy.'+img)[0] or self.BINARY_MIME
self.oeb.log.debug('Added', link)
self.oeb.container = self.DirContainer(os.path.dirname(link),
self.oeb.log, ignore_opf=True)
# Load into memory
item = self.oeb.manifest.add(id, href, media_type)
# bhref refers to an already existing file. The read() method of
# DirContainer will call unquote on it before trying to read the
# file, therefore we quote it here.
if isinstance(bhref, unicode_type):
bhref = bhref.encode('utf-8')
item.html_input_href = unicode_type(quote(bhref))
if guessed in self.OEB_STYLES:
item.override_css_fetch = partial(
self.css_import_handler, os.path.dirname(link))
item.data
self.added_resources[link] = href
nlink = self.added_resources[link]
if frag:
nlink = '#'.join((nlink, frag))
return nlink
def canonicalize_url(url):
# mechanize does not handle quoting automatically
if re.search(r'\s+', url) is not None:
if isinstance(url, unicode_type):
url = url.encode('utf-8')
purl = list(urlparse(url))
for i in range(2, 6):
purl[i] = as_bytes(quote(purl[i]))
url = urlunparse(purl).decode('utf-8')
return url
def book_filename(rd, book_id, mi, fmt, as_encoded_unicode=False):
au = authors_to_string(mi.authors or [_('Unknown')])
title = mi.title or _('Unknown')
ext = (fmt or '').lower()
if ext == 'kepub' and 'Kobo Touch' in rd.inheaders.get('User-Agent', ''):
ext = 'kepub.epub'
fname = '%s - %s_%s.%s' % (title[:30], au[:30], book_id, ext)
if as_encoded_unicode:
# See https://tools.ietf.org/html/rfc6266
fname = sanitize_file_name_unicode(fname).encode('utf-8')
fname = quote(fname).decode('ascii')
else:
fname = ascii_filename(fname).replace('"', '_')
return fname
def icon_map():
global _icon_map
with _icon_map_lock:
if _icon_map is None:
from calibre.gui2 import gprefs
_icon_map = category_icon_map.copy()
custom_icons = gprefs.get('tags_browser_category_icons', {})
for k, v in iteritems(custom_icons):
if os.access(os.path.join(config_dir, 'tb_icons', v), os.R_OK):
_icon_map[k] = '_' + quote(v)
_icon_map['file_type_icons'] = {
k:'mimetypes/%s.png' % v for k, v in iteritems(EXT_MAP)
}
return _icon_map
def icon_map():
global _icon_map
with _icon_map_lock:
if _icon_map is None:
from calibre.gui2 import gprefs
_icon_map = category_icon_map.copy()
custom_icons = gprefs.get('tags_browser_category_icons', {})
for k, v in iteritems(custom_icons):
if os.access(os.path.join(config_dir, 'tb_icons', v), os.R_OK):
_icon_map[k] = '_' + quote(v)
_icon_map['file_type_icons'] = {
k:'mimetypes/%s.png' % v for k, v in iteritems(EXT_MAP)
}
return _icon_map
def book_filename(rd, book_id, mi, fmt, as_encoded_unicode=False):
au = authors_to_string(mi.authors or [_('Unknown')])
title = mi.title or _('Unknown')
ext = (fmt or '').lower()
if ext == 'kepub' and 'Kobo Touch' in rd.inheaders.get('User-Agent', ''):
ext = 'kepub.epub'
fname = '%s - %s_%s.%s' % (title[:30], au[:30], book_id, ext)
if as_encoded_unicode:
# See https://tools.ietf.org/html/rfc6266
fname = sanitize_file_name(fname).encode('utf-8')
fname = unicode_type(quote(fname))
else:
fname = ascii_filename(fname).replace('"', '_')
return fname
def book_filename(rd, book_id, mi, fmt, as_encoded_unicode=False):
au = authors_to_string(mi.authors or [_('Unknown')])
title = mi.title or _('Unknown')
ext = (fmt or '').lower()
fname = f'{title[:30]} - {au[:30]}_{book_id}.{ext}'
if as_encoded_unicode:
# See https://tools.ietf.org/html/rfc6266
fname = sanitize_file_name(fname).encode('utf-8')
fname = str(quote(fname))
else:
fname = ascii_filename(fname).replace('"', '_')
if ext == 'kepub' and 'Kobo Touch' in rd.inheaders.get('User-Agent', ''):
fname = fname.replace('!', '_')
fname += '.epub'
return fname
def a(filename,
data=None,
status=OK,
method='POST',
username='******',
add_duplicates='n',
job_id=1):
r, data = make_request(conn,
'/cdb/add-book/{}/{}/{}'.format(
job_id, add_duplicates,
quote(filename.encode('utf-8'))),
username=username,
password='******',
prefix='',
method=method,
data=data)
ae(status, r.status)
return data
def uniquote(raw):
if isinstance(raw, str):
raw = raw.encode('utf-8')
return str(quote(raw))
def urls_from_identifiers(identifiers): # {{{
identifiers = {k.lower(): v for k, v in iteritems(identifiers)}
ans = []
keys_left = set(identifiers)
def add(name, k, val, url):
ans.append((name, k, val, url))
keys_left.discard(k)
rules = msprefs['id_link_rules']
if rules:
formatter = EvalFormatter()
for k, val in iteritems(identifiers):
val = val.replace('|', ',')
vals = {
'id':
unicode_type(
quote(val if isinstance(val, bytes) else val.encode('utf-8'
)))
}
items = rules.get(k) or ()
for name, template in items:
try:
url = formatter.safe_format(template, vals, '', vals)
except Exception:
import traceback
traceback.format_exc()
continue
add(name, k, val, url)
for plugin in all_metadata_plugins():
try:
for id_type, id_val, url in plugin.get_book_urls(identifiers):
add(plugin.get_book_url_name(id_type, id_val, url), id_type,
id_val, url)
except Exception:
pass
isbn = identifiers.get('isbn', None)
if isbn:
add(isbn, 'isbn', isbn, 'https://www.worldcat.org/isbn/' + isbn)
doi = identifiers.get('doi', None)
if doi:
add('DOI', 'doi', doi, 'https://dx.doi.org/' + doi)
arxiv = identifiers.get('arxiv', None)
if arxiv:
add('arXiv', 'arxiv', arxiv, 'https://arxiv.org/abs/' + arxiv)
oclc = identifiers.get('oclc', None)
if oclc:
add('OCLC', 'oclc', oclc, 'https://www.worldcat.org/oclc/' + oclc)
issn = check_issn(identifiers.get('issn', None))
if issn:
add(issn, 'issn', issn, 'https://www.worldcat.org/issn/' + issn)
q = {'http', 'https', 'file'}
for k, url in iteritems(identifiers):
if url and re.match(r'ur[il]\d*$', k) is not None:
url = url[:8].replace('|', ':') + url[8:].replace('|', ',')
if url.partition(':')[0].lower() in q:
parts = urlparse(url)
name = parts.netloc or parts.path
add(name, k, url, url)
for k in tuple(keys_left):
val = identifiers.get(k)
if val:
url = val[:8].replace('|', ':') + val[8:].replace('|', ',')
if url.partition(':')[0].lower() in q:
parts = urlparse(url)
name = parts.netloc or parts.path
add(name, k, url, url)
return ans
def uniquote(raw):
if isinstance(raw, unicode_type):
raw = raw.encode('utf-8')
return unicode_type(quote(raw))
def uniquote(raw):
if isinstance(raw, unicode_type):
raw = raw.encode('utf-8')
return unicode_type(quote(raw))
def a(filename, data=None, status=OK, method='POST', username='******', add_duplicates='n', job_id=1):
r, data = make_request(conn, '/cdb/add-book/{}/{}/{}'.format(job_id, add_duplicates, quote(filename.encode('utf-8'))),
username=username, password='******', prefix='', method=method, data=data)
ae(status, r.status)
return data
