Merge remote-tracking branch 'upstream/master'

[youtube-dl] / youtube_dl / utils.py

diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 4dcf18991487d7a180a5d4b903bad4b38b3165db..ab1049cc0d99b17adcfc83f951d38e89a89a4bc9 100644 (file)
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -1,15 +1,19 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 

+import errno

 import gzip
 import io
 import gzip
 import io

+import json

 import locale
 import os
 import re
 import sys
 import locale
 import os
 import re
 import sys

+import traceback

 import zlib
 import email.utils
 import zlib
 import email.utils

-import json
+import socket
+import datetime

 
 try:
     import urllib.request as compat_urllib_request
 
 try:
     import urllib.request as compat_urllib_request


@@ -31,6 +35,11 @@ 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 +60,12 @@ try:
 except ImportError: # Python 2
     import httplib as compat_http_client
 
 except ImportError: # Python 2
     import httplib as compat_http_client
 

+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


@@ -140,6 +155,13 @@ 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 = {
     'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20100101 Firefox/10.0',
     'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
 std_headers = {
     'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20100101 Firefox/10.0',
     'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',


@@ -147,6 +169,7 @@ std_headers = {
     'Accept-Encoding': 'gzip, deflate',
     'Accept-Language': 'en-us,en;q=0.5',
 }
     'Accept-Encoding': 'gzip, deflate',
     'Accept-Language': 'en-us,en;q=0.5',
 }

+

 def preferredencoding():
     """Get preferred encoding.
 
 def preferredencoding():
     """Get preferred encoding.
 


@@ -169,6 +192,31 @@ 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]+$', key)
+        assert re.match(r'^[a-zA-Z@\s]*$', 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
+

 def htmlentity_transform(matchobj):
     """Transforms an HTML entity to a character.
 
 def htmlentity_transform(matchobj):
     """Transforms an HTML entity to a character.
 


@@ -195,10 +243,11 @@ 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 AttrParser(compat_html_parser.HTMLParser):
+    """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 = {}


@@ -223,7 +272,7 @@ class IDParser(compat_html_parser.HTMLParser):
         attrs = dict(attrs)
         if self.started:
             self.find_startpos(None)
         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 +308,20 @@ 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:
     try:
         parser.loads(html)
     except compat_html_parser.HTMLParseError:


@@ -274,12 +333,13 @@ 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 +357,24 @@ 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 +451,63 @@ 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 YoutubeDLHandlerHTTPS()
+    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 YoutubeDLHandlerHTTPS(context=context)
+
+class ExtractorError(Exception):
+    """Error during info extraction."""
+    def __init__(self, msg, tb=None, expected=False):
+        """ 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
+
+    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 +516,10 @@ 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 +537,8 @@ 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. """


@@ -442,14 +570,7 @@ class ContentTooShortError(Exception):
         self.expected = expected
 
 
         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):
+class YoutubeDLHandler_Template:  # Old-style class, like HTTPHandler

     """Handler for HTTP requests and responses.
 
     This class, when installed with an OpenerDirector, automatically adds
     """Handler for HTTP requests and responses.
 
     This class, when installed with an OpenerDirector, automatically adds


@@ -482,18 +603,23 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
         ret.code = code
         return ret
 
         ret.code = code
         return ret
 

-    def http_request(self, req):
-        for h in std_headers:
+    def _http_request(self, req):
+        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
 
         return req
 

-    def http_response(self, req, resp):
+    def _http_response(self, req, resp):

         old_resp = resp
         # gzip
         if resp.headers.get('Content-encoding', '') == 'gzip':
         old_resp = resp
         # gzip
         if resp.headers.get('Content-encoding', '') == 'gzip':


@@ -506,3 +632,91 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
             resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code)
             resp.msg = old_resp.msg
         return resp
             resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code)
             resp.msg = old_resp.msg
         return resp

+
+
+class YoutubeDLHandler(YoutubeDLHandler_Template, compat_urllib_request.HTTPHandler):
+    http_request = YoutubeDLHandler_Template._http_request
+    http_response = YoutubeDLHandler_Template._http_response
+
+
+class YoutubeDLHandlerHTTPS(YoutubeDLHandler_Template, compat_urllib_request.HTTPSHandler):
+    https_request = YoutubeDLHandler_Template._http_request
+    https_response = YoutubeDLHandler_Template._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]*$', '', date_str)
+    format_expressions = ['%d %B %Y', '%B %d %Y', '%b %d %Y', '%Y-%m-%d', '%d/%m/%Y', '%Y/%m/%d %H:%M:%S', '%d.%m.%Y %H:%M']
+    for expression in format_expressions:
+        try:
+            upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d')
+        except:
+            pass
+    return upload_date
+
+def determine_ext(url, default_ext=u'unknown_video'):
+    guess = url.partition(u'?')[0].rpartition(u'.')[2]
+    if re.match(r'^[A-Za-z0-9]+$', 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())