def filename_from_url(url, clean=False):
"""Returns a reasonable filename for saving the given url.
"""
check_u(url)
try:
match = URI_PATTERN.match(url)
if match is None:
# This code path will never be executed.
return unicode_to_filename(url)
filename = match.group(2)
query = match.group(4)
if not filename:
ret = query
elif not query:
ret = filename
else:
root, ext = os.path.splitext(filename)
ret = u"%s-%s%s" % (root, query, ext)
ret = unquote(ret)
if ret is None:
ret = u'unknown'
if clean:
return clean_filename(ret)
else:
return unicode_to_filename(ret)
except (TypeError, KeyError, AttributeError, UnicodeDecodeError):
return unicode_to_filename(u'unknown')
def setup_new(self, url, item, content_type=None, channel_name=None):
check_u(url)
if content_type:
check_u(content_type)
self.orig_url = self.url = url
self.item_list = []
self.child_deleted = False
self.main_item_id = None
self.dlid = generate_dlid()
if content_type is None:
# HACK: Some servers report the wrong content-type for
# torrent files. We try to work around that by assuming
# if the enclosure states that something is a torrent,
# it's a torrent. Thanks to [email protected].
if item.enclosure_type == u'application/x-bittorrent':
content_type = item.enclosure_type
self.content_type = u""
self.delete_files = True
self.channel_name = channel_name
self.manualUpload = False
self._update_retry_time_dc = None
self.status_updates_frozen = False
self.last_update = time.time()
self.reset_status_attributes()
if content_type is None:
self.content_type = u""
else:
self.content_type = content_type
if self.content_type == u'':
self.get_content_type()
else:
self.run_downloader()
def unicode_to_filename(filename, path=None):
"""Takes in a unicode string representation of a filename (NOT a
file path) and creates a valid byte representation of it
attempting to preserve extensions.
.. Note::
This is not guaranteed to give the same results every time it
is run, nor is it guaranteed to reverse the results of
filename_to_unicode.
"""
check_u(filename)
if path:
check_b(path)
else:
path = os.getcwd()
# keep this a little shorter than the max length, so we can
# add a number to the end
max_len = os.statvfs(path)[statvfs.F_NAMEMAX] - 5
for mem in ("/", "\000", "\\", ":", "*", "?", "\"", "'",
"<", ">", "|", "&", "\r", "\n"):
filename = filename.replace(mem, "_")
new_filename = encode_fn(filename)
while len(new_filename) > max_len:
filename = shorten_fn(filename)
new_filename = encode_fn(filename)
return new_filename
def _scrape_youtube_url(url, callback):
check_u(url)
components = urlparse.urlsplit(url)
params = cgi.parse_qs(components[3])
video_id = None
if components[2] == u'/watch' and 'v' in params:
try:
video_id = params['v'][0]
except IndexError:
pass
elif components[2].startswith('/v/'):
m = re.compile(r'/v/([\w-]+)').match(components[2])
if m is not None:
video_id = m.group(1)
if video_id is None:
logging.warning('_scrape_youtube_url: unable to scrape YouTube Video URL')
callback(None)
return
try:
url = u"http://www.youtube.com/get_video_info?video_id=%s&el=embedded&ps=default&eurl=" % video_id
httpclient.grab_url(
url,
lambda x: _youtube_callback_step2(x, video_id, callback),
lambda x: _youtube_errback(x, callback))
except StandardError:
logging.exception("youtube_callback: unable to scrape YouTube Video URL")
callback(None)
def create_downloader(url, contentType, dlid):
check_u(url)
check_u(contentType)
if contentType == u'application/x-bittorrent':
return BTDownloader(url, dlid)
else:
return HTTPDownloader(url, dlid, expectedContentType=contentType)
def _scrape_youtube_url(url, callback):
check_u(url)
components = urlparse.urlsplit(url)
params = cgi.parse_qs(components[3])
video_id = None
if components[2] == u'/watch' and 'v' in params:
try:
video_id = params['v'][0]
except IndexError:
pass
elif components[2].startswith('/v/'):
m = re.compile(r'/v/([\w-]+)').match(components[2])
if m is not None:
video_id = m.group(1)
if video_id is None:
logging.warning('_scrape_youtube_url: unable to scrape YouTube Video URL')
callback(None)
return
try:
url = u"http://www.youtube.com/get_video_info?video_id=%s&el=embedded&ps=default&eurl=" % video_id
httpclient.grab_url(
url,
lambda x: _youtube_callback_step2(x, video_id, callback),
lambda x: _youtube_errback(x, callback))
except StandardError:
logging.exception("youtube_callback: unable to scrape YouTube Video URL")
callback(None)
def __setitem__(self, key, value):
check_u(key)
super(DeviceDatabase, self).__setitem__(key, value)
if self.parent:
self.parent.notify_changed()
else:
self.notify_changed()
def __setitem__(self, key, value):
check_u(key)
super(DeviceDatabase, self).__setitem__(key, value)
if self.parent:
self.parent.notify_changed()
else:
self.notify_changed()
def filename_from_url(url, clean=False):
"""Returns a reasonable filename for saving the given url.
"""
check_u(url)
try:
match = URI_PATTERN.match(url)
if match is None:
# This code path will never be executed.
return unicode_to_filename(url)
filename = match.group(2)
query = match.group(4)
if not filename:
ret = query
elif not query:
ret = filename
else:
root, ext = os.path.splitext(filename)
ret = u"%s-%s%s" % (root, query, ext)
ret = unquote(ret)
if ret is None:
ret = u'unknown'
if clean:
return clean_filename(ret)
else:
return unicode_to_filename(ret)
except (SystemExit, KeyboardInterrupt):
raise
except:
return unicode_to_filename(u'unknown')
def unicode_to_filename(filename, path=None):
"""Takes in a unicode string representation of a filename (NOT a
file path) and creates a valid byte representation of it
attempting to preserve extensions.
.. Note::
This is not guaranteed to give the same results every time it
is run, nor is it guaranteed to reverse the results of
filename_to_unicode.
"""
check_u(filename)
if path:
check_b(path)
else:
path = os.getcwd()
# keep this a little shorter than the max length, so we can
# add a number to the end
max_len = os.statvfs(path)[statvfs.F_NAMEMAX] - 5
for mem in ("/", "\000", "\\", ":", "*", "?", "\"", "'", "<", ">", "|",
"&", "\r", "\n"):
filename = filename.replace(mem, "_")
new_filename = encode_fn(filename)
while len(new_filename) > max_len:
filename = shorten_fn(filename)
new_filename = encode_fn(filename)
return new_filename
def setup_new(self, url, item, content_type=None, channel_name=None):
check_u(url)
if content_type:
check_u(content_type)
self.orig_url = self.url = url
self.item_list = []
self.child_deleted = False
self.main_item_id = None
self.dlid = generate_dlid()
if content_type is None:
# HACK: Some servers report the wrong content-type for
# torrent files. We try to work around that by assuming
# if the enclosure states that something is a torrent,
# it's a torrent. Thanks to [email protected].
if item.enclosure_type == u'application/x-bittorrent':
content_type = item.enclosure_type
self.content_type = u""
self.delete_files = True
self.channel_name = channel_name
self.manualUpload = False
self.status_updates_frozen = False
self.last_update = time.time()
self.reset_status_attributes()
if content_type is None:
self.content_type = u""
else:
self.content_type = content_type
if self.content_type == u'':
self.get_content_type()
else:
self.run_downloader()
def __getitem__(self, key):
check_u(key)
value = super(DeviceDatabase, self).__getitem__(key)
if isinstance(value, dict) and not isinstance(value, DeviceDatabase):
value = DeviceDatabase(value, self.parent or self)
# don't trip the changed signal
super(DeviceDatabase, self).__setitem__(key, value)
return value
def __getitem__(self, key):
check_u(key)
value = super(DeviceDatabase, self).__getitem__(key)
if isinstance(value, dict) and not isinstance(value, DeviceDatabase):
value = DeviceDatabase(value, self.parent or self)
# don't trip the changed signal
super(DeviceDatabase, self).__setitem__(key, value)
return value
def shorten_fn(filename):
check_u(filename)
first, last = os.path.splitext(filename)
if first:
return u"".join([first[:-1], last])
return unicode(last[:-1])
def shorten_fn(filename):
check_u(filename)
first, last = os.path.splitext(filename)
if first:
return u"".join([first[:-1], last])
return unicode(last[:-1])
def try_scraping_url(url, callback):
check_u(url)
scrape = _get_scrape_function_for(url)
if scrape is not None:
scrape(url,
lambda newurl, content_type=u"video/x-flv", title=None: _actual_url_callback(url, callback, newurl, content_type, title))
else:
callback(url)
def try_scraping_url(url, callback):
check_u(url)
scrape = _get_scrape_function_for(url)
if scrape is not None:
scrape(url,
lambda newurl, content_type=u"video/x-flv", title=None: _actual_url_callback(url, callback, newurl, content_type, title))
else:
callback(url)
def filename_to_unicode(filename, path=None):
"""Given a filename in raw bytes, return the unicode representation
Since this is not guaranteed to give the same results every time
it is run, not is it garanteed to reverse the results of
unicode_to_filename
"""
if path:
check_u(path)
check_u(filename)
return filename
def setup_new(self, typ):
"""Construct a TabOrder. typ should be either ``channel`` or
``playlist``.
"""
check_u(typ)
self.type = typ
self.tab_ids = []
self._setup_views()
to_sort = self.id_to_tab.values()
to_sort.sort(key=lambda x: x.get_title().lower())
for tab in to_sort:
self.tab_ids.append(tab.id)
def setup_new(self, typ):
"""Construct a TabOrder. typ should be either ``channel`` or
``playlist``.
"""
check_u(typ)
self.type = typ
self.tab_ids = []
self._setup_views()
to_sort = self.id_to_tab.values()
to_sort.sort(key=lambda x: x.get_title().lower())
for tab in to_sort:
self.tab_ids.append(tab.id)
def start_new_download(url, dlid, contentType, channelName):
"""Creates a new downloader object.
Returns id on success, None on failure.
"""
check_u(url)
check_u(contentType)
if channelName:
check_f(channelName)
dl = create_downloader(url, contentType, dlid)
dl.channelName = channelName
_downloads[dlid] = dl
def filename_to_unicode(filename, path=None):
"""Given a filename in raw bytes, return the unicode
representation.
Since this is not guaranteed to give the same results every time
it is run, not is it garanteed to reverse the results of
unicode_to_filename.
"""
if path:
check_u(path)
check_u(filename)
return filename
def shortenFilename(filename):
check_u(filename)
# Find the first part and the last part
pieces = filename.split(u".")
lastpart = pieces[-1]
if len(pieces) > 1:
firstpart = u".".join(pieces[:-1])
else:
firstpart = u""
# If there's a first part, use that, otherwise shorten what we have
if len(firstpart) > 0:
return u"%s.%s" % (firstpart[:-1], lastpart)
else:
return filename[:-1]
def shortenFilename(filename):
check_u(filename)
# Find the first part and the last part
pieces = filename.split(u".")
lastpart = pieces[-1]
if len(pieces) > 1:
firstpart = u".".join(pieces[:-1])
else:
firstpart = u""
# If there's a first part, use that, otherwise shorten what we have
if len(firstpart) > 0:
return u"%s.%s" % (firstpart[:-1],lastpart)
else:
return filename[:-1]
def unicode_to_filename(filename, path=None):
"""Takes in a unicode string representation of a filename and
creates a valid byte representation of it attempting to preserve
extensions
This is not guaranteed to give the same results every time it is
run, not is it garanteed to reverse the results of
filename_to_unicode
"""
@returns_unicode
def shorten_filename(filename):
check_u(filename)
# Find the first part and the last part
pieces = filename.split(u".")
lastpart = pieces[-1]
if len(pieces) > 1:
firstpart = u".".join(pieces[:-1])
else:
firstpart = u""
# If there's a first part, use that, otherwise shorten what we have
if len(firstpart) > 0:
return u"%s.%s" % (firstpart[:-1], lastpart)
else:
return filename[:-1]
check_u(filename)
if path:
check_u(path)
else:
path = os.getcwd()
# Keep this a little shorter than the max length, so we can run
# nextFilename
MAX_LEN = 200
badchars = ('/', '\000', '\\', ':', '*', '?', "'", '"',
'<', '>', '|', "\n", "\r")
for mem in badchars:
filename = filename.replace(mem, "_")
new_filename = filename
while len(new_filename) > MAX_LEN:
new_filename = shorten_filename(new_filename)
return new_filename
def unicode_to_filename(filename, path=None):
"""Takes in a unicode string representation of a filename and
creates a valid byte representation of it attempting to preserve
extensions
This is not guaranteed to give the same results every time it is
run, not is it garanteed to reverse the results of
filename_to_unicode
"""
@returns_unicode
def shorten_filename(filename):
check_u(filename)
# Find the first part and the last part
pieces = filename.split(u".")
lastpart = pieces[-1]
if len(pieces) > 1:
firstpart = u".".join(pieces[:-1])
else:
firstpart = u""
# If there's a first part, use that, otherwise shorten what we have
if len(firstpart) > 0:
return u"%s.%s" % (firstpart[:-1], lastpart)
else:
return filename[:-1]
check_u(filename)
if path:
check_u(path)
else:
path = os.getcwd()
# Keep this a little shorter than the max length, so we can run
# nextFilename
MAX_LEN = 200
badchars = ('/', '\000', '\\', ':', '*', '?', "'", '"',
'<', '>', '|', "\n", "\r")
for mem in badchars:
filename = filename.replace(mem, "_")
new_filename = filename
while len(new_filename) > MAX_LEN:
new_filename = shorten_filename(new_filename)
return new_filename
def unicode_to_filename(filename, path = None):
check_u(filename)
if path:
check_b(path)
else:
path = os.getcwd()
# Keep this a little shorter than the max length, so we can run
# nextFilename
MAX_LEN = os.statvfs(path)[statvfs.F_NAMEMAX]-5
for mem in ("/", "\000", "\\", ":", "*", "?", "'", "\"", "<", ">", "|", "&", "\r", "\n"):
filename = filename.replace(mem, "_")
new_filename = filename.encode('utf-8','replace')
while len(new_filename) > MAX_LEN:
filename = shortenFilename(filename)
new_filename = filename.encode('utf-8','replace')
return new_filename
def setup_new(self, url, allowedURLs=None):
check_u(url)
# FIXME - clean up the allowedURLs thing here
self.allowedURLs = []
self.url = url
self.updated_url = url
self.title = None
self.userTitle = None
self.client = None
self.lastVisitedURL = None
self.setup_new_icon_cache()
self.favicon = None
self.firstTime = True
if url:
self.historyLocation = 0
self.history = [self.url]
else:
self.historyLocation = None
self.history = []
self.download_guide()
def unicode_to_filename(filename, path=None):
check_u(filename)
if path:
check_b(path)
else:
path = os.getcwd()
# Keep this a little shorter than the max length, so we can run
# nextFilename
MAX_LEN = os.statvfs(path)[statvfs.F_NAMEMAX] - 5
for mem in ("/", "\000", "\\", ":", "*", "?", "'", "\"", "<", ">", "|",
"&", "\r", "\n"):
filename = filename.replace(mem, "_")
new_filename = filename.encode('utf-8', 'replace')
while len(new_filename) > MAX_LEN:
filename = shortenFilename(filename)
new_filename = filename.encode('utf-8', 'replace')
return new_filename
def test_check_u(self):
util.check_u(None)
util.check_u(u'abc')
util.check_u(u'&*@!#)*) !@)( !@# !)@(#')
self.assertRaises(util.MiroUnicodeError, util.check_u, 'abc')
self.assertRaises(util.MiroUnicodeError, util.check_u, '&*@!#)*) !@)( !@# !)@(#')
def test_check_u(self):
util.check_u(None)
util.check_u(u'abc')
util.check_u(u'&*@!#)*) !@)( !@# !)@(#')
self.assertRaises(util.MiroUnicodeError, util.check_u, 'abc')
self.assertRaises(util.MiroUnicodeError, util.check_u,
'&*@!#)*) !@)( !@# !)@(#')
def setup_new(self, url, allowedURLs=None):
check_u(url)
# FIXME - clean up the allowedURLs thing here
self.allowedURLs = []
self.url = url
self.updated_url = url
self.title = None
self.userTitle = None
self.client = None
self.lastVisitedURL = None
self.setup_new_icon_cache()
self.favicon = None
self.firstTime = True
self.store = self.STORE_NOT_STORE
if url:
self.historyLocation = 0
self.history = [self.url]
else:
self.historyLocation = None
self.history = []
self.setup_common()
self.download_guide()
def __init__(self, name, title, url, sort_order=0, filename=None):
check_u(name)
check_u(title)
check_u(url)
self.name = name
self.title = title
self.url = url
self.sort_order = sort_order
if filename is not None:
self.filename = os.path.normcase(filename)
# used for changing icon location on themed searches
else:
self.filename = None
def unmake_url_safe(s):
"""Undoes make_url_safe (assuming it was passed a FilenameType)
"""
# unquote the byte string
check_u(s)
return urllib.unquote(s.encode("ascii"))
def make_url_safe(string, safe='/'):
"""Takes in a byte string or a unicode string and does the right
thing to make a URL
"""
check_u(string)
return urllib.quote(string.encode('utf_8'), safe=safe).decode('ascii')
def unmake_url_safe(string):
"""Undoes make_url_safe.
"""
check_u(string)
return urllib.unquote(string.encode('ascii')).decode('utf_8')
def _actual_url_callback(url, callback, new_url, content_type, title):
if new_url:
check_u(new_url)
callback(new_url, content_type=content_type, title=title)
def _get_scrape_function_for(url):
check_u(url)
for scrape_info in SCRAPER_INFO_MAP:
if scrape_info['pattern'].match(url) is not None:
return scrape_info['func']
return None
def _actual_url_callback(url, callback, new_url, content_type, title):
if new_url:
check_u(new_url)
callback(new_url, content_type=content_type, title=title)
def _get_scrape_function_for(url):
check_u(url)
for scrape_info in SCRAPER_INFO_MAP:
if scrape_info['pattern'].match(url) is not None:
return scrape_info['func']
return None
def unmake_url_safe(string):
"""Undoes make_url_safe (assuming it was passed a filenameType)
"""
# unquote the byte string
check_u(string)
return urllib.unquote(string.encode('ascii'))
def unmake_url_safe(string):
"""Undoes make_url_safe (assuming it was passed a filenameType)
"""
# unquote the byte string
check_u(string)
return urllib.unquote(string.encode('ascii'))
def unmake_url_safe(string):
"""Undoes make_url_safe.
"""
check_u(string)
return urllib.unquote(string.encode('ascii')).decode('utf_8')
def make_url_safe(string, safe='/'):
"""Takes in a byte string or a unicode string and does the right
thing to make a URL
"""
check_u(string)
return urllib.quote(string.encode('utf_8'), safe=safe).decode('ascii')
