X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;ds=sidebyside;f=youtube_dl%2Futils.py;h=2d12e2df93f6f1e885600e49f0eb6f587494e19f;hb=d6c7a367e88096bb17e323954002c084477fe908;hp=82a1daeb9075a056aa908e0a91bbd83b8897673a;hpb=81d7f1928cf548160a101f23c5571cce11403f8d;p=youtube-dl diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index 82a1daeb9..2d12e2df9 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -1,6 +1,7 @@ #!/usr/bin/env python # -*- coding: utf-8 -*- +import ctypes import datetime import email.utils import errno @@ -8,10 +9,14 @@ import gzip import io import json import locale +import math import os +import pipes import platform import re +import ssl import socket +import subprocess import sys import traceback import zlib @@ -229,6 +234,19 @@ else: return f return None +# On python2.6 the xml.etree.ElementTree.Element methods don't support +# the namespace parameter +def xpath_with_ns(path, ns_map): + components = [c.split(':') for c in path.split('/')] + replaced = [] + for c in components: + if len(c) == 1: + replaced.append(c[0]) + else: + ns, tag = c + replaced.append('{%s}%s' % (ns_map[ns], tag)) + return '/'.join(replaced) + def htmlentity_transform(matchobj): """Transforms an HTML entity to a character. @@ -521,18 +539,38 @@ def formatSeconds(secs): else: return '%d' % secs -def make_HTTPS_handler(opts): - if sys.version_info < (3,2): - # Python's 2.x handler is very simplistic - return compat_urllib_request.HTTPSHandler() +def make_HTTPS_handler(opts_no_check_certificate): + if sys.version_info < (3, 2): + import httplib + + class HTTPSConnectionV3(httplib.HTTPSConnection): + def __init__(self, *args, **kwargs): + httplib.HTTPSConnection.__init__(self, *args, **kwargs) + + def connect(self): + sock = socket.create_connection((self.host, self.port), self.timeout) + if getattr(self, '_tunnel_host', False): + self.sock = sock + self._tunnel() + try: + self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_SSLv3) + except ssl.SSLError: + self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_SSLv23) + + class HTTPSHandlerV3(compat_urllib_request.HTTPSHandler): + def https_open(self, req): + return self.do_open(HTTPSConnectionV3, req) + return HTTPSHandlerV3() else: - import ssl - context = ssl.SSLContext(ssl.PROTOCOL_SSLv23) - context.set_default_verify_paths() - + context = ssl.SSLContext(ssl.PROTOCOL_SSLv3) context.verify_mode = (ssl.CERT_NONE - if opts.no_check_certificate + if opts_no_check_certificate else ssl.CERT_REQUIRED) + context.set_default_verify_paths() + try: + context.load_default_certs() + except AttributeError: + pass # Python < 3.4 return compat_urllib_request.HTTPSHandler(context=context) class ExtractorError(Exception): @@ -558,6 +596,11 @@ class ExtractorError(Exception): return u''.join(traceback.format_tb(self.traceback)) +class RegexNotFoundError(ExtractorError): + """Error when a regex didn't match""" + pass + + class DownloadError(Exception): """Download Error exception. @@ -715,6 +758,8 @@ def unified_strdate(date_str): '%Y/%m/%d %H:%M:%S', '%d.%m.%Y %H:%M', '%Y-%m-%dT%H:%M:%SZ', + '%Y-%m-%dT%H:%M:%S.%fZ', + '%Y-%m-%dT%H:%M:%S.%f0Z', '%Y-%m-%dT%H:%M:%S', ] for expression in format_expressions: @@ -927,3 +972,122 @@ class locked_file(object): def read(self, *args): return self.f.read(*args) + + +def shell_quote(args): + quoted_args = [] + encoding = sys.getfilesystemencoding() + if encoding is None: + encoding = 'utf-8' + for a in args: + if isinstance(a, bytes): + # We may get a filename encoded with 'encodeFilename' + a = a.decode(encoding) + quoted_args.append(pipes.quote(a)) + return u' '.join(quoted_args) + + +def takewhile_inclusive(pred, seq): + """ Like itertools.takewhile, but include the latest evaluated element + (the first element so that Not pred(e)) """ + for e in seq: + yield e + if not pred(e): + return + + +def smuggle_url(url, data): + """ Pass additional data in a URL for internal use. """ + + sdata = compat_urllib_parse.urlencode( + {u'__youtubedl_smuggle': json.dumps(data)}) + return url + u'#' + sdata + + +def unsmuggle_url(smug_url): + if not '#__youtubedl_smuggle' in smug_url: + return smug_url, None + url, _, sdata = smug_url.rpartition(u'#') + jsond = compat_parse_qs(sdata)[u'__youtubedl_smuggle'][0] + data = json.loads(jsond) + return url, data + + +def format_bytes(bytes): + if bytes is None: + return u'N/A' + if type(bytes) is str: + bytes = float(bytes) + if bytes == 0.0: + exponent = 0 + else: + exponent = int(math.log(bytes, 1024.0)) + suffix = [u'B', u'KiB', u'MiB', u'GiB', u'TiB', u'PiB', u'EiB', u'ZiB', u'YiB'][exponent] + converted = float(bytes) / float(1024 ** exponent) + return u'%.2f%s' % (converted, suffix) + + +def str_to_int(int_str): + int_str = re.sub(r'[,\.]', u'', int_str) + return int(int_str) + + +def get_term_width(): + columns = os.environ.get('COLUMNS', None) + if columns: + return int(columns) + + try: + sp = subprocess.Popen( + ['stty', 'size'], + stdout=subprocess.PIPE, stderr=subprocess.PIPE) + out, err = sp.communicate() + return int(out.split()[1]) + except: + pass + return None + + +def month_by_name(name): + """ Return the number of a month by (locale-independently) English name """ + + ENGLISH_NAMES = [ + u'January', u'February', u'March', u'April', u'May', u'June', + u'July', u'August', u'September', u'October', u'November', u'December'] + try: + return ENGLISH_NAMES.index(name) + 1 + except ValueError: + return None + + +def fix_xml_all_ampersand(xml_str): + """Replace all the '&' by '&' in XML""" + return xml_str.replace(u'&', u'&') + + +def setproctitle(title): + assert isinstance(title, type(u'')) + try: + libc = ctypes.cdll.LoadLibrary("libc.so.6") + except OSError: + return + title = title + buf = ctypes.create_string_buffer(len(title) + 1) + buf.value = title.encode('utf-8') + try: + libc.prctl(15, ctypes.byref(buf), 0, 0, 0) + except AttributeError: + return # Strange libc, just skip this + + +def remove_start(s, start): + if s.startswith(start): + return s[len(start):] + return s + + +def url_basename(url): + m = re.match(r'(?:https?:|)//[^/]+/(?:[^?#]+/)?([^/?#]+)/?(?:[?#]|$)', url) + if not m: + return u'' + return m.group(1)