X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/523a0ad06fdeae419ce4c1474cfa58c9bbc63f63..fe738fbf71a871a5dca457a2743571fd1131b23e:/youtube_dl/utils.py diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index 66ae41e..201ed25 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -1,19 +1,20 @@ #!/usr/bin/env python # -*- coding: utf-8 -*- +import datetime +import email.utils import errno import gzip import io import json import locale import os +import platform import re +import socket import sys import traceback import zlib -import email.utils -import socket -import datetime try: import urllib.request as compat_urllib_request @@ -35,6 +36,11 @@ try: 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 @@ -55,6 +61,17 @@ try: 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 @@ -198,6 +215,20 @@ else: 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 + def htmlentity_transform(matchobj): """Transforms an HTML entity to a character. @@ -224,7 +255,17 @@ 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 -class AttrParser(compat_html_parser.HTMLParser): +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 @@ -232,10 +273,9 @@ class AttrParser(compat_html_parser.HTMLParser): self.result = None self.started = False self.depth = {} - self.html = None 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: @@ -244,11 +284,6 @@ class AttrParser(compat_html_parser.HTMLParser): 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: @@ -309,6 +344,38 @@ def get_element_by_attribute(attribute, value, html): 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: + pass + return parser.get_result() + def clean_html(html): """Clean an HTML snippet into a readable string""" @@ -470,15 +537,20 @@ def make_HTTPS_handler(opts): class ExtractorError(Exception): """Error during info extraction.""" - def __init__(self, msg, tb=None): - """ tb, if given, is the original traceback (so that it can be printed out). """ - - if not sys.exc_info()[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError): - msg = msg + u'; please report this issue on GitHub.' + 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: @@ -599,8 +671,23 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler): 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': @@ -619,7 +706,16 @@ def unified_strdate(date_str): date_str = date_str.replace(',',' ') # %z (UTC offset) is only supported in python>=3.2 date_str = re.sub(r' (\+|-)[\d]*$', '', date_str) - format_expressions = ['%d %B %Y', '%B %d %Y', '%b %d %Y', '%Y-%m-%d', '%d/%m/%Y', '%Y/%m/%d %H:%M:%S'] + format_expressions = [ + '%d %B %Y', + '%B %d %Y', + '%b %d %Y', + '%Y-%m-%d', + '%d/%m/%Y', + '%Y/%m/%d %H:%M:%S', + '%d.%m.%Y %H:%M', + '%Y-%m-%dT%H:%M:%SZ', + ] for expression in format_expressions: try: upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d') @@ -627,6 +723,16 @@ def unified_strdate(date_str): pass return upload_date +def determine_ext(url, default_ext=u'unknown_video'): + guess = url.partition(u'?')[0].rpartition(u'.')[2] + if re.match(r'^[A-Za-z0-9]+$', 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 @@ -678,3 +784,43 @@ class DateRange(object): 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)