]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/utils.py
update changelog.
[youtubedl] / youtube_dl / utils.py
index 44f939053adf5fcf6a2ab1ad4cb9c0eac5cedd7a..bfb8f6bcd971dad03d5236c8e607b59ff81c667a 100644 (file)
@@ -1,15 +1,21 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 
+import datetime
+import email.utils
+import errno
 import gzip
 import io
 import gzip
 import io
+import json
 import locale
 import os
 import locale
 import os
+import pipes
+import platform
 import re
 import re
+import socket
 import sys
 import sys
+import traceback
 import zlib
 import zlib
-import email.utils
-import json
 
 try:
     import urllib.request as compat_urllib_request
 
 try:
     import urllib.request as compat_urllib_request
@@ -31,6 +37,11 @@ try:
 except ImportError: # Python 2
     from urlparse import urlparse as compat_urllib_parse_urlparse
 
 except ImportError: # Python 2
     from urlparse import urlparse as compat_urllib_parse_urlparse
 
+try:
+    import urllib.parse as compat_urlparse
+except ImportError: # Python 2
+    import urlparse as compat_urlparse
+
 try:
     import http.cookiejar as compat_cookiejar
 except ImportError: # Python 2
 try:
     import http.cookiejar as compat_cookiejar
 except ImportError: # Python 2
@@ -51,6 +62,23 @@ try:
 except ImportError: # Python 2
     import httplib as compat_http_client
 
 except ImportError: # Python 2
     import httplib as compat_http_client
 
+try:
+    from urllib.error import HTTPError as compat_HTTPError
+except ImportError:  # Python 2
+    from urllib2 import HTTPError as compat_HTTPError
+
+try:
+    from urllib.request import urlretrieve as compat_urlretrieve
+except ImportError:  # Python 2
+    from urllib import urlretrieve as compat_urlretrieve
+
+
+try:
+    from subprocess import DEVNULL
+    compat_subprocess_get_DEVNULL = lambda: DEVNULL
+except ImportError:
+    compat_subprocess_get_DEVNULL = lambda: open(os.path.devnull, 'w')
+
 try:
     from urllib.parse import parse_qs as compat_parse_qs
 except ImportError: # Python 2
 try:
     from urllib.parse import parse_qs as compat_parse_qs
 except ImportError: # Python 2
@@ -140,13 +168,21 @@ try:
 except NameError:
     compat_chr = chr
 
 except NameError:
     compat_chr = chr
 
+def compat_ord(c):
+    if type(c) is int: return c
+    else: return ord(c)
+
+# This is not clearly defined otherwise
+compiled_regex_type = type(re.compile(''))
+
 std_headers = {
 std_headers = {
-    'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20100101 Firefox/10.0',
+    'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20100101 Firefox/10.0 (Chrome)',
     'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
     'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
     'Accept-Encoding': 'gzip, deflate',
     'Accept-Language': 'en-us,en;q=0.5',
 }
     'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
     'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
     'Accept-Encoding': 'gzip, deflate',
     'Accept-Language': 'en-us,en;q=0.5',
 }
+
 def preferredencoding():
     """Get preferred encoding.
 
 def preferredencoding():
     """Get preferred encoding.
 
@@ -169,6 +205,44 @@ else:
         assert type(s) == type(u'')
         print(s)
 
         assert type(s) == type(u'')
         print(s)
 
+# In Python 2.x, json.dump expects a bytestream.
+# In Python 3.x, it writes to a character stream
+if sys.version_info < (3,0):
+    def write_json_file(obj, fn):
+        with open(fn, 'wb') as f:
+            json.dump(obj, f)
+else:
+    def write_json_file(obj, fn):
+        with open(fn, 'w', encoding='utf-8') as f:
+            json.dump(obj, f)
+
+if sys.version_info >= (2,7):
+    def find_xpath_attr(node, xpath, key, val):
+        """ Find the xpath xpath[@key=val] """
+        assert re.match(r'^[a-zA-Z]+$', key)
+        assert re.match(r'^[a-zA-Z0-9@\s]*$', val)
+        expr = xpath + u"[@%s='%s']" % (key, val)
+        return node.find(expr)
+else:
+    def find_xpath_attr(node, xpath, key, val):
+        for f in node.findall(xpath):
+            if f.attrib.get(key) == val:
+                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.
 
 def htmlentity_transform(matchobj):
     """Transforms an HTML entity to a character.
 
@@ -195,17 +269,27 @@ def htmlentity_transform(matchobj):
     return (u'&%s;' % entity)
 
 compat_html_parser.locatestarttagend = re.compile(r"""<[a-zA-Z][-.a-zA-Z0-9:_]*(?:\s+(?:(?<=['"\s])[^\s/>][^\s/=>]*(?:\s*=+\s*(?:'[^']*'|"[^"]*"|(?!['"])[^>\s]*))?\s*)*)?\s*""", re.VERBOSE) # backport bugfix
     return (u'&%s;' % entity)
 
 compat_html_parser.locatestarttagend = re.compile(r"""<[a-zA-Z][-.a-zA-Z0-9:_]*(?:\s+(?:(?<=['"\s])[^\s/>][^\s/=>]*(?:\s*=+\s*(?:'[^']*'|"[^"]*"|(?!['"])[^>\s]*))?\s*)*)?\s*""", re.VERBOSE) # backport bugfix
-class IDParser(compat_html_parser.HTMLParser):
-    """Modified HTMLParser that isolates a tag with the specified id"""
-    def __init__(self, id):
-        self.id = id
+class BaseHTMLParser(compat_html_parser.HTMLParser):
+    def __init(self):
+        compat_html_parser.HTMLParser.__init__(self)
+        self.html = None
+
+    def loads(self, html):
+        self.html = html
+        self.feed(html)
+        self.close()
+
+class AttrParser(BaseHTMLParser):
+    """Modified HTMLParser that isolates a tag with the specified attribute"""
+    def __init__(self, attribute, value):
+        self.attribute = attribute
+        self.value = value
         self.result = None
         self.started = False
         self.depth = {}
         self.result = None
         self.started = False
         self.depth = {}
-        self.html = None
         self.watch_startpos = False
         self.error_count = 0
         self.watch_startpos = False
         self.error_count = 0
-        compat_html_parser.HTMLParser.__init__(self)
+        BaseHTMLParser.__init__(self)
 
     def error(self, message):
         if self.error_count > 10 or self.started:
 
     def error(self, message):
         if self.error_count > 10 or self.started:
@@ -214,16 +298,11 @@ class IDParser(compat_html_parser.HTMLParser):
         self.error_count += 1
         self.goahead(1)
 
         self.error_count += 1
         self.goahead(1)
 
-    def loads(self, html):
-        self.html = html
-        self.feed(html)
-        self.close()
-
     def handle_starttag(self, tag, attrs):
         attrs = dict(attrs)
         if self.started:
             self.find_startpos(None)
     def handle_starttag(self, tag, attrs):
         attrs = dict(attrs)
         if self.started:
             self.find_startpos(None)
-        if 'id' in attrs and attrs['id'] == self.id:
+        if self.attribute in attrs and attrs[self.attribute] == self.value:
             self.result = [tag]
             self.started = True
             self.watch_startpos = True
             self.result = [tag]
             self.started = True
             self.watch_startpos = True
@@ -259,10 +338,52 @@ class IDParser(compat_html_parser.HTMLParser):
             lines[-1] = lines[-1][:self.result[2][1]-self.result[1][1]]
         lines[-1] = lines[-1][:self.result[2][1]]
         return '\n'.join(lines).strip()
             lines[-1] = lines[-1][:self.result[2][1]-self.result[1][1]]
         lines[-1] = lines[-1][:self.result[2][1]]
         return '\n'.join(lines).strip()
+# Hack for https://github.com/rg3/youtube-dl/issues/662
+if sys.version_info < (2, 7, 3):
+    AttrParser.parse_endtag = (lambda self, i:
+        i + len("</scr'+'ipt>")
+        if self.rawdata[i:].startswith("</scr'+'ipt>")
+        else compat_html_parser.HTMLParser.parse_endtag(self, i))
 
 def get_element_by_id(id, html):
 
 def get_element_by_id(id, html):
-    """Return the content of the tag with the specified id in the passed HTML document"""
-    parser = IDParser(id)
+    """Return the content of the tag with the specified ID in the passed HTML document"""
+    return get_element_by_attribute("id", id, html)
+
+def get_element_by_attribute(attribute, value, html):
+    """Return the content of the tag with the specified attribute in the passed HTML document"""
+    parser = AttrParser(attribute, value)
+    try:
+        parser.loads(html)
+    except compat_html_parser.HTMLParseError:
+        pass
+    return parser.get_result()
+
+class MetaParser(BaseHTMLParser):
+    """
+    Modified HTMLParser that isolates a meta tag with the specified name 
+    attribute.
+    """
+    def __init__(self, name):
+        BaseHTMLParser.__init__(self)
+        self.name = name
+        self.content = None
+        self.result = None
+
+    def handle_starttag(self, tag, attrs):
+        if tag != 'meta':
+            return
+        attrs = dict(attrs)
+        if attrs.get('name') == self.name:
+            self.result = attrs.get('content')
+
+    def get_result(self):
+        return self.result
+
+def get_meta_content(name, html):
+    """
+    Return the content attribute from the meta tag with the given name attribute.
+    """
+    parser = MetaParser(name)
     try:
         parser.loads(html)
     except compat_html_parser.HTMLParseError:
     try:
         parser.loads(html)
     except compat_html_parser.HTMLParseError:
@@ -274,12 +395,13 @@ def clean_html(html):
     """Clean an HTML snippet into a readable string"""
     # Newline vs <br />
     html = html.replace('\n', ' ')
     """Clean an HTML snippet into a readable string"""
     # Newline vs <br />
     html = html.replace('\n', ' ')
-    html = re.sub('\s*<\s*br\s*/?\s*>\s*', '\n', html)
+    html = re.sub(r'\s*<\s*br\s*/?\s*>\s*', '\n', html)
+    html = re.sub(r'<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html)
     # Strip html tags
     html = re.sub('<.*?>', '', html)
     # Replace html entities
     html = unescapeHTML(html)
     # Strip html tags
     html = re.sub('<.*?>', '', html)
     # Replace html entities
     html = unescapeHTML(html)
-    return html
+    return html.strip()
 
 
 def sanitize_open(filename, open_mode):
 
 
 def sanitize_open(filename, open_mode):
@@ -297,16 +419,24 @@ def sanitize_open(filename, open_mode):
             if sys.platform == 'win32':
                 import msvcrt
                 msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY)
             if sys.platform == 'win32':
                 import msvcrt
                 msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY)
-            return (sys.stdout, filename)
+            return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename)
         stream = open(encodeFilename(filename), open_mode)
         return (stream, filename)
     except (IOError, OSError) as err:
         stream = open(encodeFilename(filename), open_mode)
         return (stream, filename)
     except (IOError, OSError) as err:
-        # In case of error, try to remove win32 forbidden chars
-        filename = re.sub(u'[/<>:"\\|\\\\?\\*]', u'#', filename)
+        if err.errno in (errno.EACCES,):
+            raise
 
 
-        # An exception here should be caught in the caller
-        stream = open(encodeFilename(filename), open_mode)
-        return (stream, filename)
+        # In case of error, try to remove win32 forbidden chars
+        alt_filename = os.path.join(
+                        re.sub(u'[/<>:"\\|\\\\?\\*]', u'#', path_part)
+                        for path_part in os.path.split(filename)
+                       )
+        if alt_filename == filename:
+            raise
+        else:
+            # An exception here should be caught in the caller
+            stream = open(encodeFilename(filename), open_mode)
+            return (stream, alt_filename)
 
 
 def timeconvert(timestr):
 
 
 def timeconvert(timestr):
@@ -383,7 +513,64 @@ def encodeFilename(s):
         # match Windows 9x series as well. Besides, NT 4 is obsolete.)
         return s
     else:
         # match Windows 9x series as well. Besides, NT 4 is obsolete.)
         return s
     else:
-        return s.encode(sys.getfilesystemencoding(), 'ignore')
+        encoding = sys.getfilesystemencoding()
+        if encoding is None:
+            encoding = 'utf-8'
+        return s.encode(encoding, 'ignore')
+
+def decodeOption(optval):
+    if optval is None:
+        return optval
+    if isinstance(optval, bytes):
+        optval = optval.decode(preferredencoding())
+
+    assert isinstance(optval, compat_str)
+    return optval
+
+def formatSeconds(secs):
+    if secs > 3600:
+        return '%d:%02d:%02d' % (secs // 3600, (secs % 3600) // 60, secs % 60)
+    elif secs > 60:
+        return '%d:%02d' % (secs // 60, secs % 60)
+    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()
+    else:
+        import ssl
+        context = ssl.SSLContext(ssl.PROTOCOL_SSLv23)
+        context.set_default_verify_paths()
+        
+        context.verify_mode = (ssl.CERT_NONE
+                               if opts.no_check_certificate
+                               else ssl.CERT_REQUIRED)
+        return compat_urllib_request.HTTPSHandler(context=context)
+
+class ExtractorError(Exception):
+    """Error during info extraction."""
+    def __init__(self, msg, tb=None, expected=False, cause=None):
+        """ tb, if given, is the original traceback (so that it can be printed out).
+        If expected is set, this is a normal error message and most likely not a bug in youtube-dl.
+        """
+
+        if sys.exc_info()[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError):
+            expected = True
+        if not expected:
+            msg = msg + u'; please report this issue on https://yt-dl.org/bug . Be sure to call youtube-dl with the --verbose flag and include its complete output. Make sure you are using the latest version; type  youtube-dl -U  to update.'
+        super(ExtractorError, self).__init__(msg)
+
+        self.traceback = tb
+        self.exc_info = sys.exc_info()  # preserve original exception
+        self.cause = cause
+
+    def format_traceback(self):
+        if self.traceback is None:
+            return None
+        return u''.join(traceback.format_tb(self.traceback))
+
 
 class DownloadError(Exception):
     """Download Error exception.
 
 class DownloadError(Exception):
     """Download Error exception.
@@ -392,7 +579,10 @@ class DownloadError(Exception):
     configured to continue on errors. They will contain the appropriate
     error message.
     """
     configured to continue on errors. They will contain the appropriate
     error message.
     """
-    pass
+    def __init__(self, msg, exc_info=None):
+        """ exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """
+        super(DownloadError, self).__init__(msg)
+        self.exc_info = exc_info
 
 
 class SameFileError(Exception):
 
 
 class SameFileError(Exception):
@@ -410,7 +600,8 @@ class PostProcessingError(Exception):
     This exception may be raised by PostProcessor's .run() method to
     indicate an error in the postprocessing task.
     """
     This exception may be raised by PostProcessor's .run() method to
     indicate an error in the postprocessing task.
     """
-    pass
+    def __init__(self, msg):
+        self.msg = msg
 
 class MaxDownloadsReached(Exception):
     """ --max-downloads limit has been reached. """
 
 class MaxDownloadsReached(Exception):
     """ --max-downloads limit has been reached. """
@@ -441,14 +632,6 @@ class ContentTooShortError(Exception):
         self.downloaded = downloaded
         self.expected = expected
 
         self.downloaded = downloaded
         self.expected = expected
 
-
-class Trouble(Exception):
-    """Trouble helper exception
-
-    This is an exception to be handled with
-    FileDownloader.trouble
-    """
-
 class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
     """Handler for HTTP requests and responses.
 
 class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
     """Handler for HTTP requests and responses.
 
@@ -483,22 +666,42 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
         return ret
 
     def http_request(self, req):
         return ret
 
     def http_request(self, req):
-        for h in std_headers:
+        for h,v in std_headers.items():
             if h in req.headers:
                 del req.headers[h]
             if h in req.headers:
                 del req.headers[h]
-            req.add_header(h, std_headers[h])
+            req.add_header(h, v)
         if 'Youtubedl-no-compression' in req.headers:
             if 'Accept-encoding' in req.headers:
                 del req.headers['Accept-encoding']
             del req.headers['Youtubedl-no-compression']
         if 'Youtubedl-no-compression' in req.headers:
             if 'Accept-encoding' in req.headers:
                 del req.headers['Accept-encoding']
             del req.headers['Youtubedl-no-compression']
+        if 'Youtubedl-user-agent' in req.headers:
+            if 'User-agent' in req.headers:
+                del req.headers['User-agent']
+            req.headers['User-agent'] = req.headers['Youtubedl-user-agent']
+            del req.headers['Youtubedl-user-agent']
         return req
 
     def http_response(self, req, resp):
         old_resp = resp
         # gzip
         if resp.headers.get('Content-encoding', '') == 'gzip':
         return req
 
     def http_response(self, req, resp):
         old_resp = resp
         # gzip
         if resp.headers.get('Content-encoding', '') == 'gzip':
-            gz = gzip.GzipFile(fileobj=io.BytesIO(resp.read()), mode='r')
-            resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code)
+            content = resp.read()
+            gz = gzip.GzipFile(fileobj=io.BytesIO(content), mode='rb')
+            try:
+                uncompressed = io.BytesIO(gz.read())
+            except IOError as original_ioerror:
+                # There may be junk add the end of the file
+                # See http://stackoverflow.com/q/4928560/35070 for details
+                for i in range(1, 1024):
+                    try:
+                        gz = gzip.GzipFile(fileobj=io.BytesIO(content[:-i]), mode='rb')
+                        uncompressed = io.BytesIO(gz.read())
+                    except IOError:
+                        continue
+                    break
+                else:
+                    raise original_ioerror
+            resp = self.addinfourl_wrapper(uncompressed, old_resp.headers, old_resp.url, old_resp.code)
             resp.msg = old_resp.msg
         # deflate
         if resp.headers.get('Content-encoding', '') == 'deflate':
             resp.msg = old_resp.msg
         # deflate
         if resp.headers.get('Content-encoding', '') == 'deflate':
@@ -509,3 +712,262 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
 
     https_request = http_request
     https_response = http_response
 
     https_request = http_request
     https_response = http_response
+
+def unified_strdate(date_str):
+    """Return a string with the date in the format YYYYMMDD"""
+    upload_date = None
+    #Replace commas
+    date_str = date_str.replace(',',' ')
+    # %z (UTC offset) is only supported in python>=3.2
+    date_str = re.sub(r' (\+|-)[\d]*$', '', date_str)
+    format_expressions = [
+        '%d %B %Y',
+        '%B %d %Y',
+        '%b %d %Y',
+        '%Y-%m-%d',
+        '%d/%m/%Y',
+        '%Y/%m/%d %H:%M:%S',
+        '%d.%m.%Y %H:%M',
+        '%Y-%m-%dT%H:%M:%SZ',
+        '%Y-%m-%dT%H:%M:%S',
+    ]
+    for expression in format_expressions:
+        try:
+            upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d')
+        except:
+            pass
+    return upload_date
+
+def determine_ext(url, default_ext=u'unknown_video'):
+    guess = url.partition(u'?')[0].rpartition(u'.')[2]
+    if re.match(r'^[A-Za-z0-9]+$', guess):
+        return guess
+    else:
+        return default_ext
+
+def subtitles_filename(filename, sub_lang, sub_format):
+    return filename.rsplit('.', 1)[0] + u'.' + sub_lang + u'.' + sub_format
+
+def date_from_str(date_str):
+    """
+    Return a datetime object from a string in the format YYYYMMDD or
+    (now|today)[+-][0-9](day|week|month|year)(s)?"""
+    today = datetime.date.today()
+    if date_str == 'now'or date_str == 'today':
+        return today
+    match = re.match('(now|today)(?P<sign>[+-])(?P<time>\d+)(?P<unit>day|week|month|year)(s)?', date_str)
+    if match is not None:
+        sign = match.group('sign')
+        time = int(match.group('time'))
+        if sign == '-':
+            time = -time
+        unit = match.group('unit')
+        #A bad aproximation?
+        if unit == 'month':
+            unit = 'day'
+            time *= 30
+        elif unit == 'year':
+            unit = 'day'
+            time *= 365
+        unit += 's'
+        delta = datetime.timedelta(**{unit: time})
+        return today + delta
+    return datetime.datetime.strptime(date_str, "%Y%m%d").date()
+    
+class DateRange(object):
+    """Represents a time interval between two dates"""
+    def __init__(self, start=None, end=None):
+        """start and end must be strings in the format accepted by date"""
+        if start is not None:
+            self.start = date_from_str(start)
+        else:
+            self.start = datetime.datetime.min.date()
+        if end is not None:
+            self.end = date_from_str(end)
+        else:
+            self.end = datetime.datetime.max.date()
+        if self.start > self.end:
+            raise ValueError('Date range: "%s" , the start date must be before the end date' % self)
+    @classmethod
+    def day(cls, day):
+        """Returns a range that only contains the given day"""
+        return cls(day,day)
+    def __contains__(self, date):
+        """Check if the date is in the range"""
+        if not isinstance(date, datetime.date):
+            date = date_from_str(date)
+        return self.start <= date <= self.end
+    def __str__(self):
+        return '%s - %s' % ( self.start.isoformat(), self.end.isoformat())
+
+
+def platform_name():
+    """ Returns the platform name as a compat_str """
+    res = platform.platform()
+    if isinstance(res, bytes):
+        res = res.decode(preferredencoding())
+
+    assert isinstance(res, compat_str)
+    return res
+
+
+def write_string(s, out=None):
+    if out is None:
+        out = sys.stderr
+    assert type(s) == type(u'')
+
+    if ('b' in getattr(out, 'mode', '') or
+            sys.version_info[0] < 3):  # Python 2 lies about mode of sys.stderr
+        s = s.encode(preferredencoding(), 'ignore')
+    out.write(s)
+    out.flush()
+
+
+def bytes_to_intlist(bs):
+    if not bs:
+        return []
+    if isinstance(bs[0], int):  # Python 3
+        return list(bs)
+    else:
+        return [ord(c) for c in bs]
+
+
+def intlist_to_bytes(xs):
+    if not xs:
+        return b''
+    if isinstance(chr(0), bytes):  # Python 2
+        return ''.join([chr(x) for x in xs])
+    else:
+        return bytes(xs)
+
+
+def get_cachedir(params={}):
+    cache_root = os.environ.get('XDG_CACHE_HOME',
+                                os.path.expanduser('~/.cache'))
+    return params.get('cachedir', os.path.join(cache_root, 'youtube-dl'))
+
+
+# Cross-platform file locking
+if sys.platform == 'win32':
+    import ctypes.wintypes
+    import msvcrt
+
+    class OVERLAPPED(ctypes.Structure):
+        _fields_ = [
+            ('Internal', ctypes.wintypes.LPVOID),
+            ('InternalHigh', ctypes.wintypes.LPVOID),
+            ('Offset', ctypes.wintypes.DWORD),
+            ('OffsetHigh', ctypes.wintypes.DWORD),
+            ('hEvent', ctypes.wintypes.HANDLE),
+        ]
+
+    kernel32 = ctypes.windll.kernel32
+    LockFileEx = kernel32.LockFileEx
+    LockFileEx.argtypes = [
+        ctypes.wintypes.HANDLE,     # hFile
+        ctypes.wintypes.DWORD,      # dwFlags
+        ctypes.wintypes.DWORD,      # dwReserved
+        ctypes.wintypes.DWORD,      # nNumberOfBytesToLockLow
+        ctypes.wintypes.DWORD,      # nNumberOfBytesToLockHigh
+        ctypes.POINTER(OVERLAPPED)  # Overlapped
+    ]
+    LockFileEx.restype = ctypes.wintypes.BOOL
+    UnlockFileEx = kernel32.UnlockFileEx
+    UnlockFileEx.argtypes = [
+        ctypes.wintypes.HANDLE,     # hFile
+        ctypes.wintypes.DWORD,      # dwReserved
+        ctypes.wintypes.DWORD,      # nNumberOfBytesToLockLow
+        ctypes.wintypes.DWORD,      # nNumberOfBytesToLockHigh
+        ctypes.POINTER(OVERLAPPED)  # Overlapped
+    ]
+    UnlockFileEx.restype = ctypes.wintypes.BOOL
+    whole_low = 0xffffffff
+    whole_high = 0x7fffffff
+
+    def _lock_file(f, exclusive):
+        overlapped = OVERLAPPED()
+        overlapped.Offset = 0
+        overlapped.OffsetHigh = 0
+        overlapped.hEvent = 0
+        f._lock_file_overlapped_p = ctypes.pointer(overlapped)
+        handle = msvcrt.get_osfhandle(f.fileno())
+        if not LockFileEx(handle, 0x2 if exclusive else 0x0, 0,
+                          whole_low, whole_high, f._lock_file_overlapped_p):
+            raise OSError('Locking file failed: %r' % ctypes.FormatError())
+
+    def _unlock_file(f):
+        assert f._lock_file_overlapped_p
+        handle = msvcrt.get_osfhandle(f.fileno())
+        if not UnlockFileEx(handle, 0,
+                            whole_low, whole_high, f._lock_file_overlapped_p):
+            raise OSError('Unlocking file failed: %r' % ctypes.FormatError())
+
+else:
+    import fcntl
+
+    def _lock_file(f, exclusive):
+        fcntl.lockf(f, fcntl.LOCK_EX if exclusive else fcntl.LOCK_SH)
+
+    def _unlock_file(f):
+        fcntl.lockf(f, fcntl.LOCK_UN)
+
+
+class locked_file(object):
+    def __init__(self, filename, mode, encoding=None):
+        assert mode in ['r', 'a', 'w']
+        self.f = io.open(filename, mode, encoding=encoding)
+        self.mode = mode
+
+    def __enter__(self):
+        exclusive = self.mode != 'r'
+        try:
+            _lock_file(self.f, exclusive)
+        except IOError:
+            self.f.close()
+            raise
+        return self
+
+    def __exit__(self, etype, value, traceback):
+        try:
+            _unlock_file(self.f)
+        finally:
+            self.f.close()
+
+    def __iter__(self):
+        return iter(self.f)
+
+    def write(self, *args):
+        return self.f.write(*args)
+
+    def read(self, *args):
+        return self.f.read(*args)
+
+
+def shell_quote(args):
+    return ' '.join(map(pipes.quote, 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