X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;ds=sidebyside;f=youtube_dl%2Futils.py;h=b9bff5fde87d91a5956e978c98880f05034ac6ab;hb=20c3893f0ea7c341ed015a15cec951ab4f7c8748;hp=25b67db0665af756ee32dac797f18a647004ff1f;hpb=5910e210f48826c6d078ef3744b25d209535c3ae;p=youtube-dl.git
diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 25b67db06..b9bff5fde 100644
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -1,15 +1,19 @@
#!/usr/bin/env python
# -*- coding: utf-8 -*-
+import errno
import gzip
import io
+import json
import locale
import os
import re
import sys
+import traceback
import zlib
import email.utils
-import json
+import socket
+import datetime
try:
import urllib.request as compat_urllib_request
@@ -146,6 +150,13 @@ try:
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',
@@ -153,6 +164,7 @@ std_headers = {
'Accept-Encoding': 'gzip, deflate',
'Accept-Language': 'en-us,en;q=0.5',
}
+
def preferredencoding():
"""Get preferred encoding.
@@ -175,6 +187,17 @@ else:
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)
+
def htmlentity_transform(matchobj):
"""Transforms an HTML entity to a character.
@@ -201,10 +224,11 @@ 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
-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 = {}
@@ -229,7 +253,7 @@ class IDParser(compat_html_parser.HTMLParser):
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
@@ -265,10 +289,20 @@ 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()
+# 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("")
+ if self.rawdata[i:].startswith("")
+ else compat_html_parser.HTMLParser.parse_endtag(self, i))
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:
@@ -280,12 +314,13 @@ def clean_html(html):
"""Clean an HTML snippet into a readable string"""
# Newline vs
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)
- return html
+ return html.strip()
def sanitize_open(filename, open_mode):
@@ -303,16 +338,24 @@ def sanitize_open(filename, open_mode):
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:
- # 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):
@@ -389,7 +432,63 @@ def encodeFilename(s):
# 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):
+ """ 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.'
+ 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.
@@ -398,7 +497,10 @@ class DownloadError(Exception):
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):
@@ -416,7 +518,8 @@ class PostProcessingError(Exception):
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. """
@@ -447,14 +550,6 @@ class ContentTooShortError(Exception):
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.
@@ -489,14 +584,19 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
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]
- 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-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):
@@ -515,3 +615,77 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
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]*$', '', date_str)
+ format_expressions = ['%d %B %Y', '%B %d %Y', '%b %d %Y', '%Y-%m-%d', '%d/%m/%Y', '%Y/%m/%d %H:%M:%S', '%d.%m.%Y %H:%M']
+ for expression in format_expressions:
+ try:
+ upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d')
+ except:
+ pass
+ return upload_date
+
+def determine_ext(url):
+ guess = url.partition(u'?')[0].rpartition(u'.')[2]
+ if re.match(r'^[A-Za-z0-9]+$', guess):
+ return guess
+ else:
+ return u'unknown_video'
+
+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[+-])(?P