]>
Commit | Line | Data |
---|---|---|
d77c3dfd FV |
1 | #!/usr/bin/env python |
2 | # -*- coding: utf-8 -*- | |
3 | ||
ecc0c5ee PH |
4 | from __future__ import unicode_literals |
5 | ||
912b38b4 | 6 | import calendar |
676eb3f2 | 7 | import codecs |
62e609ab | 8 | import contextlib |
e3946f98 | 9 | import ctypes |
c496ca96 PH |
10 | import datetime |
11 | import email.utils | |
f45c185f | 12 | import errno |
be4a824d | 13 | import functools |
d77c3dfd | 14 | import gzip |
b7ab0590 | 15 | import itertools |
03f9daab | 16 | import io |
f4bfd65f | 17 | import json |
d77c3dfd | 18 | import locale |
02dbf93f | 19 | import math |
347de493 | 20 | import operator |
d77c3dfd | 21 | import os |
4eb7f1d1 | 22 | import pipes |
c496ca96 | 23 | import platform |
d77c3dfd | 24 | import re |
13ebea79 | 25 | import ssl |
c496ca96 | 26 | import socket |
b53466e1 | 27 | import struct |
1c088fa8 | 28 | import subprocess |
d77c3dfd | 29 | import sys |
181c8655 | 30 | import tempfile |
01951dda | 31 | import traceback |
bcf89ce6 | 32 | import xml.etree.ElementTree |
d77c3dfd | 33 | import zlib |
d77c3dfd | 34 | |
8c25f81b | 35 | from .compat import ( |
8f9312c3 | 36 | compat_basestring, |
8c25f81b | 37 | compat_chr, |
8c25f81b | 38 | compat_html_entities, |
be4a824d | 39 | compat_http_client, |
8c25f81b | 40 | compat_parse_qs, |
be4a824d | 41 | compat_socket_create_connection, |
8c25f81b PH |
42 | compat_str, |
43 | compat_urllib_error, | |
44 | compat_urllib_parse, | |
45 | compat_urllib_parse_urlparse, | |
46 | compat_urllib_request, | |
47 | compat_urlparse, | |
7d4111ed | 48 | shlex_quote, |
8c25f81b | 49 | ) |
4644ac55 S |
50 | |
51 | ||
468e2e92 FV |
52 | # This is not clearly defined otherwise |
53 | compiled_regex_type = type(re.compile('')) | |
54 | ||
3e669f36 | 55 | std_headers = { |
18313934 | 56 | 'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20150101 Firefox/20.0 (Chrome)', |
59ae15a5 PH |
57 | 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7', |
58 | 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8', | |
59 | 'Accept-Encoding': 'gzip, deflate', | |
60 | 'Accept-Language': 'en-us,en;q=0.5', | |
3e669f36 | 61 | } |
f427df17 | 62 | |
5f6a1245 | 63 | |
7105440c YCH |
64 | ENGLISH_MONTH_NAMES = [ |
65 | 'January', 'February', 'March', 'April', 'May', 'June', | |
66 | 'July', 'August', 'September', 'October', 'November', 'December'] | |
67 | ||
68 | ||
d77c3dfd | 69 | def preferredencoding(): |
59ae15a5 | 70 | """Get preferred encoding. |
d77c3dfd | 71 | |
59ae15a5 PH |
72 | Returns the best encoding scheme for the system, based on |
73 | locale.getpreferredencoding() and some further tweaks. | |
74 | """ | |
75 | try: | |
76 | pref = locale.getpreferredencoding() | |
28e614de | 77 | 'TEST'.encode(pref) |
59ae15a5 PH |
78 | except: |
79 | pref = 'UTF-8' | |
bae611f2 | 80 | |
59ae15a5 | 81 | return pref |
d77c3dfd | 82 | |
f4bfd65f | 83 | |
181c8655 | 84 | def write_json_file(obj, fn): |
1394646a | 85 | """ Encode obj as JSON and write it to fn, atomically if possible """ |
181c8655 | 86 | |
92120217 | 87 | fn = encodeFilename(fn) |
61ee5aeb | 88 | if sys.version_info < (3, 0) and sys.platform != 'win32': |
ec5f6016 JMF |
89 | encoding = get_filesystem_encoding() |
90 | # os.path.basename returns a bytes object, but NamedTemporaryFile | |
91 | # will fail if the filename contains non ascii characters unless we | |
92 | # use a unicode object | |
93 | path_basename = lambda f: os.path.basename(fn).decode(encoding) | |
94 | # the same for os.path.dirname | |
95 | path_dirname = lambda f: os.path.dirname(fn).decode(encoding) | |
96 | else: | |
97 | path_basename = os.path.basename | |
98 | path_dirname = os.path.dirname | |
99 | ||
73159f99 S |
100 | args = { |
101 | 'suffix': '.tmp', | |
ec5f6016 JMF |
102 | 'prefix': path_basename(fn) + '.', |
103 | 'dir': path_dirname(fn), | |
73159f99 S |
104 | 'delete': False, |
105 | } | |
106 | ||
181c8655 PH |
107 | # In Python 2.x, json.dump expects a bytestream. |
108 | # In Python 3.x, it writes to a character stream | |
109 | if sys.version_info < (3, 0): | |
73159f99 | 110 | args['mode'] = 'wb' |
181c8655 | 111 | else: |
73159f99 S |
112 | args.update({ |
113 | 'mode': 'w', | |
114 | 'encoding': 'utf-8', | |
115 | }) | |
116 | ||
117 | tf = tempfile.NamedTemporaryFile(**args) | |
181c8655 PH |
118 | |
119 | try: | |
120 | with tf: | |
121 | json.dump(obj, tf) | |
1394646a IK |
122 | if sys.platform == 'win32': |
123 | # Need to remove existing file on Windows, else os.rename raises | |
124 | # WindowsError or FileExistsError. | |
125 | try: | |
126 | os.unlink(fn) | |
127 | except OSError: | |
128 | pass | |
181c8655 PH |
129 | os.rename(tf.name, fn) |
130 | except: | |
131 | try: | |
132 | os.remove(tf.name) | |
133 | except OSError: | |
134 | pass | |
135 | raise | |
136 | ||
137 | ||
138 | if sys.version_info >= (2, 7): | |
59ae56fa PH |
139 | def find_xpath_attr(node, xpath, key, val): |
140 | """ Find the xpath xpath[@key=val] """ | |
cbf915f3 PH |
141 | assert re.match(r'^[a-zA-Z-]+$', key) |
142 | assert re.match(r'^[a-zA-Z0-9@\s:._-]*$', val) | |
ab4ee31e | 143 | expr = xpath + "[@%s='%s']" % (key, val) |
59ae56fa PH |
144 | return node.find(expr) |
145 | else: | |
146 | def find_xpath_attr(node, xpath, key, val): | |
4eefbfdb PH |
147 | # Here comes the crazy part: In 2.6, if the xpath is a unicode, |
148 | # .//node does not match if a node is a direct child of . ! | |
8f9312c3 | 149 | if isinstance(xpath, compat_str): |
4eefbfdb PH |
150 | xpath = xpath.encode('ascii') |
151 | ||
59ae56fa PH |
152 | for f in node.findall(xpath): |
153 | if f.attrib.get(key) == val: | |
154 | return f | |
155 | return None | |
156 | ||
d7e66d39 JMF |
157 | # On python2.6 the xml.etree.ElementTree.Element methods don't support |
158 | # the namespace parameter | |
5f6a1245 JW |
159 | |
160 | ||
d7e66d39 JMF |
161 | def xpath_with_ns(path, ns_map): |
162 | components = [c.split(':') for c in path.split('/')] | |
163 | replaced = [] | |
164 | for c in components: | |
165 | if len(c) == 1: | |
166 | replaced.append(c[0]) | |
167 | else: | |
168 | ns, tag = c | |
169 | replaced.append('{%s}%s' % (ns_map[ns], tag)) | |
170 | return '/'.join(replaced) | |
171 | ||
d77c3dfd | 172 | |
bf0ff932 | 173 | def xpath_text(node, xpath, name=None, fatal=False): |
d74bebd5 PH |
174 | if sys.version_info < (2, 7): # Crazy 2.6 |
175 | xpath = xpath.encode('ascii') | |
176 | ||
bf0ff932 | 177 | n = node.find(xpath) |
42bdd9d0 | 178 | if n is None or n.text is None: |
bf0ff932 PH |
179 | if fatal: |
180 | name = xpath if name is None else name | |
181 | raise ExtractorError('Could not find XML element %s' % name) | |
182 | else: | |
183 | return None | |
184 | return n.text | |
185 | ||
186 | ||
9e6dd238 | 187 | def get_element_by_id(id, html): |
43e8fafd ND |
188 | """Return the content of the tag with the specified ID in the passed HTML document""" |
189 | return get_element_by_attribute("id", id, html) | |
190 | ||
12ea2f30 | 191 | |
43e8fafd ND |
192 | def get_element_by_attribute(attribute, value, html): |
193 | """Return the content of the tag with the specified attribute in the passed HTML document""" | |
9e6dd238 | 194 | |
38285056 PH |
195 | m = re.search(r'''(?xs) |
196 | <([a-zA-Z0-9:._-]+) | |
197 | (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]+|="[^"]+"|='[^']+'))*? | |
198 | \s+%s=['"]?%s['"]? | |
199 | (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]+|="[^"]+"|='[^']+'))*? | |
200 | \s*> | |
201 | (?P<content>.*?) | |
202 | </\1> | |
203 | ''' % (re.escape(attribute), re.escape(value)), html) | |
204 | ||
205 | if not m: | |
206 | return None | |
207 | res = m.group('content') | |
208 | ||
209 | if res.startswith('"') or res.startswith("'"): | |
210 | res = res[1:-1] | |
a921f407 | 211 | |
38285056 | 212 | return unescapeHTML(res) |
a921f407 | 213 | |
9e6dd238 FV |
214 | |
215 | def clean_html(html): | |
59ae15a5 | 216 | """Clean an HTML snippet into a readable string""" |
dd622d7c PH |
217 | |
218 | if html is None: # Convenience for sanitizing descriptions etc. | |
219 | return html | |
220 | ||
59ae15a5 PH |
221 | # Newline vs <br /> |
222 | html = html.replace('\n', ' ') | |
6b3aef80 FV |
223 | html = re.sub(r'\s*<\s*br\s*/?\s*>\s*', '\n', html) |
224 | html = re.sub(r'<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html) | |
59ae15a5 PH |
225 | # Strip html tags |
226 | html = re.sub('<.*?>', '', html) | |
227 | # Replace html entities | |
228 | html = unescapeHTML(html) | |
7decf895 | 229 | return html.strip() |
9e6dd238 FV |
230 | |
231 | ||
d77c3dfd | 232 | def sanitize_open(filename, open_mode): |
59ae15a5 PH |
233 | """Try to open the given filename, and slightly tweak it if this fails. |
234 | ||
235 | Attempts to open the given filename. If this fails, it tries to change | |
236 | the filename slightly, step by step, until it's either able to open it | |
237 | or it fails and raises a final exception, like the standard open() | |
238 | function. | |
239 | ||
240 | It returns the tuple (stream, definitive_file_name). | |
241 | """ | |
242 | try: | |
28e614de | 243 | if filename == '-': |
59ae15a5 PH |
244 | if sys.platform == 'win32': |
245 | import msvcrt | |
246 | msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY) | |
898280a0 | 247 | return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename) |
59ae15a5 PH |
248 | stream = open(encodeFilename(filename), open_mode) |
249 | return (stream, filename) | |
250 | except (IOError, OSError) as err: | |
f45c185f PH |
251 | if err.errno in (errno.EACCES,): |
252 | raise | |
59ae15a5 | 253 | |
f45c185f PH |
254 | # In case of error, try to remove win32 forbidden chars |
255 | alt_filename = os.path.join( | |
b74e86f4 PH |
256 | re.sub('[/<>:"\\|\\\\?\\*]', '#', path_part) |
257 | for path_part in os.path.split(filename) | |
258 | ) | |
f45c185f PH |
259 | if alt_filename == filename: |
260 | raise | |
261 | else: | |
262 | # An exception here should be caught in the caller | |
263 | stream = open(encodeFilename(filename), open_mode) | |
264 | return (stream, alt_filename) | |
d77c3dfd FV |
265 | |
266 | ||
267 | def timeconvert(timestr): | |
59ae15a5 PH |
268 | """Convert RFC 2822 defined time string into system timestamp""" |
269 | timestamp = None | |
270 | timetuple = email.utils.parsedate_tz(timestr) | |
271 | if timetuple is not None: | |
272 | timestamp = email.utils.mktime_tz(timetuple) | |
273 | return timestamp | |
1c469a94 | 274 | |
5f6a1245 | 275 | |
796173d0 | 276 | def sanitize_filename(s, restricted=False, is_id=False): |
59ae15a5 PH |
277 | """Sanitizes a string so it could be used as part of a filename. |
278 | If restricted is set, use a stricter subset of allowed characters. | |
796173d0 | 279 | Set is_id if this is not an arbitrary string, but an ID that should be kept if possible |
59ae15a5 PH |
280 | """ |
281 | def replace_insane(char): | |
282 | if char == '?' or ord(char) < 32 or ord(char) == 127: | |
283 | return '' | |
284 | elif char == '"': | |
285 | return '' if restricted else '\'' | |
286 | elif char == ':': | |
287 | return '_-' if restricted else ' -' | |
288 | elif char in '\\/|*<>': | |
289 | return '_' | |
627dcfff | 290 | if restricted and (char in '!&\'()[]{}$;`^,#' or char.isspace()): |
59ae15a5 PH |
291 | return '_' |
292 | if restricted and ord(char) > 127: | |
293 | return '_' | |
294 | return char | |
295 | ||
2aeb06d6 PH |
296 | # Handle timestamps |
297 | s = re.sub(r'[0-9]+(?::[0-9]+)+', lambda m: m.group(0).replace(':', '_'), s) | |
28e614de | 298 | result = ''.join(map(replace_insane, s)) |
796173d0 PH |
299 | if not is_id: |
300 | while '__' in result: | |
301 | result = result.replace('__', '_') | |
302 | result = result.strip('_') | |
303 | # Common case of "Foreign band name - English song title" | |
304 | if restricted and result.startswith('-_'): | |
305 | result = result[2:] | |
5a42414b PH |
306 | if result.startswith('-'): |
307 | result = '_' + result[len('-'):] | |
a7440261 | 308 | result = result.lstrip('.') |
796173d0 PH |
309 | if not result: |
310 | result = '_' | |
59ae15a5 | 311 | return result |
d77c3dfd | 312 | |
5f6a1245 | 313 | |
a2aaf4db S |
314 | def sanitize_path(s): |
315 | """Sanitizes and normalizes path on Windows""" | |
316 | if sys.platform != 'win32': | |
317 | return s | |
318 | drive, _ = os.path.splitdrive(s) | |
319 | unc, _ = os.path.splitunc(s) | |
320 | unc_or_drive = unc or drive | |
321 | norm_path = os.path.normpath(remove_start(s, unc_or_drive)).split(os.path.sep) | |
322 | if unc_or_drive: | |
323 | norm_path.pop(0) | |
324 | sanitized_path = [ | |
325 | re.sub('[/<>:"\\|\\\\?\\*]', '#', path_part) | |
326 | for path_part in norm_path] | |
327 | if unc_or_drive: | |
328 | sanitized_path.insert(0, unc_or_drive + os.path.sep) | |
329 | return os.path.join(*sanitized_path) | |
330 | ||
331 | ||
d77c3dfd | 332 | def orderedSet(iterable): |
59ae15a5 PH |
333 | """ Remove all duplicates from the input iterable """ |
334 | res = [] | |
335 | for el in iterable: | |
336 | if el not in res: | |
337 | res.append(el) | |
338 | return res | |
d77c3dfd | 339 | |
912b38b4 | 340 | |
4e408e47 PH |
341 | def _htmlentity_transform(entity): |
342 | """Transforms an HTML entity to a character.""" | |
343 | # Known non-numeric HTML entity | |
344 | if entity in compat_html_entities.name2codepoint: | |
345 | return compat_chr(compat_html_entities.name2codepoint[entity]) | |
346 | ||
347 | mobj = re.match(r'#(x?[0-9]+)', entity) | |
348 | if mobj is not None: | |
349 | numstr = mobj.group(1) | |
28e614de | 350 | if numstr.startswith('x'): |
4e408e47 | 351 | base = 16 |
28e614de | 352 | numstr = '0%s' % numstr |
4e408e47 PH |
353 | else: |
354 | base = 10 | |
355 | return compat_chr(int(numstr, base)) | |
356 | ||
357 | # Unknown entity in name, return its literal representation | |
28e614de | 358 | return ('&%s;' % entity) |
4e408e47 PH |
359 | |
360 | ||
d77c3dfd | 361 | def unescapeHTML(s): |
912b38b4 PH |
362 | if s is None: |
363 | return None | |
364 | assert type(s) == compat_str | |
d77c3dfd | 365 | |
4e408e47 PH |
366 | return re.sub( |
367 | r'&([^;]+);', lambda m: _htmlentity_transform(m.group(1)), s) | |
d77c3dfd | 368 | |
8bf48f23 PH |
369 | |
370 | def encodeFilename(s, for_subprocess=False): | |
59ae15a5 PH |
371 | """ |
372 | @param s The name of the file | |
373 | """ | |
d77c3dfd | 374 | |
8bf48f23 | 375 | assert type(s) == compat_str |
d77c3dfd | 376 | |
59ae15a5 PH |
377 | # Python 3 has a Unicode API |
378 | if sys.version_info >= (3, 0): | |
379 | return s | |
0f00efed | 380 | |
59ae15a5 | 381 | if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5: |
28e614de | 382 | # Pass '' directly to use Unicode APIs on Windows 2000 and up |
59ae15a5 PH |
383 | # (Detecting Windows NT 4 is tricky because 'major >= 4' would |
384 | # match Windows 9x series as well. Besides, NT 4 is obsolete.) | |
8bf48f23 PH |
385 | if not for_subprocess: |
386 | return s | |
387 | else: | |
388 | # For subprocess calls, encode with locale encoding | |
389 | # Refer to http://stackoverflow.com/a/9951851/35070 | |
390 | encoding = preferredencoding() | |
59ae15a5 | 391 | else: |
6df40dcb | 392 | encoding = sys.getfilesystemencoding() |
8bf48f23 PH |
393 | if encoding is None: |
394 | encoding = 'utf-8' | |
395 | return s.encode(encoding, 'ignore') | |
396 | ||
f07b74fc PH |
397 | |
398 | def encodeArgument(s): | |
399 | if not isinstance(s, compat_str): | |
400 | # Legacy code that uses byte strings | |
401 | # Uncomment the following line after fixing all post processors | |
7af808a5 | 402 | # assert False, 'Internal error: %r should be of type %r, is %r' % (s, compat_str, type(s)) |
f07b74fc PH |
403 | s = s.decode('ascii') |
404 | return encodeFilename(s, True) | |
405 | ||
406 | ||
8271226a PH |
407 | def decodeOption(optval): |
408 | if optval is None: | |
409 | return optval | |
410 | if isinstance(optval, bytes): | |
411 | optval = optval.decode(preferredencoding()) | |
412 | ||
413 | assert isinstance(optval, compat_str) | |
414 | return optval | |
1c256f70 | 415 | |
5f6a1245 | 416 | |
4539dd30 PH |
417 | def formatSeconds(secs): |
418 | if secs > 3600: | |
419 | return '%d:%02d:%02d' % (secs // 3600, (secs % 3600) // 60, secs % 60) | |
420 | elif secs > 60: | |
421 | return '%d:%02d' % (secs // 60, secs % 60) | |
422 | else: | |
423 | return '%d' % secs | |
424 | ||
a0ddb8a2 | 425 | |
be4a824d PH |
426 | def make_HTTPS_handler(params, **kwargs): |
427 | opts_no_check_certificate = params.get('nocheckcertificate', False) | |
0db261ba | 428 | if hasattr(ssl, 'create_default_context'): # Python >= 3.4 or 2.7.9 |
be5f2c19 | 429 | context = ssl.create_default_context(ssl.Purpose.SERVER_AUTH) |
0db261ba | 430 | if opts_no_check_certificate: |
be5f2c19 | 431 | context.check_hostname = False |
0db261ba | 432 | context.verify_mode = ssl.CERT_NONE |
a2366922 | 433 | try: |
be4a824d | 434 | return YoutubeDLHTTPSHandler(params, context=context, **kwargs) |
a2366922 PH |
435 | except TypeError: |
436 | # Python 2.7.8 | |
437 | # (create_default_context present but HTTPSHandler has no context=) | |
438 | pass | |
439 | ||
440 | if sys.version_info < (3, 2): | |
d7932313 | 441 | return YoutubeDLHTTPSHandler(params, **kwargs) |
aa37e3d4 | 442 | else: # Python < 3.4 |
d7932313 | 443 | context = ssl.SSLContext(ssl.PROTOCOL_TLSv1) |
ea6d901e | 444 | context.verify_mode = (ssl.CERT_NONE |
dca08720 | 445 | if opts_no_check_certificate |
ea6d901e | 446 | else ssl.CERT_REQUIRED) |
303b479e | 447 | context.set_default_verify_paths() |
be4a824d | 448 | return YoutubeDLHTTPSHandler(params, context=context, **kwargs) |
ea6d901e | 449 | |
732ea2f0 | 450 | |
1c256f70 PH |
451 | class ExtractorError(Exception): |
452 | """Error during info extraction.""" | |
5f6a1245 | 453 | |
d11271dd | 454 | def __init__(self, msg, tb=None, expected=False, cause=None, video_id=None): |
9a82b238 PH |
455 | """ tb, if given, is the original traceback (so that it can be printed out). |
456 | If expected is set, this is a normal error message and most likely not a bug in youtube-dl. | |
457 | """ | |
458 | ||
459 | if sys.exc_info()[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError): | |
460 | expected = True | |
d11271dd PH |
461 | if video_id is not None: |
462 | msg = video_id + ': ' + msg | |
410f3e73 | 463 | if cause: |
28e614de | 464 | msg += ' (caused by %r)' % cause |
9a82b238 | 465 | if not expected: |
732ea2f0 PH |
466 | if ytdl_is_updateable(): |
467 | update_cmd = 'type youtube-dl -U to update' | |
468 | else: | |
469 | update_cmd = 'see https://yt-dl.org/update on how to update' | |
470 | msg += '; please report this issue on https://yt-dl.org/bug .' | |
471 | msg += ' Make sure you are using the latest version; %s.' % update_cmd | |
472 | msg += ' Be sure to call youtube-dl with the --verbose flag and include its complete output.' | |
1c256f70 | 473 | super(ExtractorError, self).__init__(msg) |
d5979c5d | 474 | |
1c256f70 | 475 | self.traceback = tb |
8cc83b8d | 476 | self.exc_info = sys.exc_info() # preserve original exception |
2eabb802 | 477 | self.cause = cause |
d11271dd | 478 | self.video_id = video_id |
1c256f70 | 479 | |
01951dda PH |
480 | def format_traceback(self): |
481 | if self.traceback is None: | |
482 | return None | |
28e614de | 483 | return ''.join(traceback.format_tb(self.traceback)) |
01951dda | 484 | |
1c256f70 | 485 | |
416c7fcb PH |
486 | class UnsupportedError(ExtractorError): |
487 | def __init__(self, url): | |
488 | super(UnsupportedError, self).__init__( | |
489 | 'Unsupported URL: %s' % url, expected=True) | |
490 | self.url = url | |
491 | ||
492 | ||
55b3e45b JMF |
493 | class RegexNotFoundError(ExtractorError): |
494 | """Error when a regex didn't match""" | |
495 | pass | |
496 | ||
497 | ||
d77c3dfd | 498 | class DownloadError(Exception): |
59ae15a5 | 499 | """Download Error exception. |
d77c3dfd | 500 | |
59ae15a5 PH |
501 | This exception may be thrown by FileDownloader objects if they are not |
502 | configured to continue on errors. They will contain the appropriate | |
503 | error message. | |
504 | """ | |
5f6a1245 | 505 | |
8cc83b8d FV |
506 | def __init__(self, msg, exc_info=None): |
507 | """ exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """ | |
508 | super(DownloadError, self).__init__(msg) | |
509 | self.exc_info = exc_info | |
d77c3dfd FV |
510 | |
511 | ||
512 | class SameFileError(Exception): | |
59ae15a5 | 513 | """Same File exception. |
d77c3dfd | 514 | |
59ae15a5 PH |
515 | This exception will be thrown by FileDownloader objects if they detect |
516 | multiple files would have to be downloaded to the same file on disk. | |
517 | """ | |
518 | pass | |
d77c3dfd FV |
519 | |
520 | ||
521 | class PostProcessingError(Exception): | |
59ae15a5 | 522 | """Post Processing exception. |
d77c3dfd | 523 | |
59ae15a5 PH |
524 | This exception may be raised by PostProcessor's .run() method to |
525 | indicate an error in the postprocessing task. | |
526 | """ | |
5f6a1245 | 527 | |
7851b379 PH |
528 | def __init__(self, msg): |
529 | self.msg = msg | |
d77c3dfd | 530 | |
5f6a1245 | 531 | |
d77c3dfd | 532 | class MaxDownloadsReached(Exception): |
59ae15a5 PH |
533 | """ --max-downloads limit has been reached. """ |
534 | pass | |
d77c3dfd FV |
535 | |
536 | ||
537 | class UnavailableVideoError(Exception): | |
59ae15a5 | 538 | """Unavailable Format exception. |
d77c3dfd | 539 | |
59ae15a5 PH |
540 | This exception will be thrown when a video is requested |
541 | in a format that is not available for that video. | |
542 | """ | |
543 | pass | |
d77c3dfd FV |
544 | |
545 | ||
546 | class ContentTooShortError(Exception): | |
59ae15a5 | 547 | """Content Too Short exception. |
d77c3dfd | 548 | |
59ae15a5 PH |
549 | This exception may be raised by FileDownloader objects when a file they |
550 | download is too small for what the server announced first, indicating | |
551 | the connection was probably interrupted. | |
552 | """ | |
553 | # Both in bytes | |
554 | downloaded = None | |
555 | expected = None | |
d77c3dfd | 556 | |
59ae15a5 PH |
557 | def __init__(self, downloaded, expected): |
558 | self.downloaded = downloaded | |
559 | self.expected = expected | |
d77c3dfd | 560 | |
5f6a1245 | 561 | |
c5a59d93 | 562 | def _create_http_connection(ydl_handler, http_class, is_https, *args, **kwargs): |
be4a824d PH |
563 | hc = http_class(*args, **kwargs) |
564 | source_address = ydl_handler._params.get('source_address') | |
565 | if source_address is not None: | |
566 | sa = (source_address, 0) | |
567 | if hasattr(hc, 'source_address'): # Python 2.7+ | |
568 | hc.source_address = sa | |
569 | else: # Python 2.6 | |
570 | def _hc_connect(self, *args, **kwargs): | |
571 | sock = compat_socket_create_connection( | |
572 | (self.host, self.port), self.timeout, sa) | |
573 | if is_https: | |
d7932313 PH |
574 | self.sock = ssl.wrap_socket( |
575 | sock, self.key_file, self.cert_file, | |
576 | ssl_version=ssl.PROTOCOL_TLSv1) | |
be4a824d PH |
577 | else: |
578 | self.sock = sock | |
579 | hc.connect = functools.partial(_hc_connect, hc) | |
580 | ||
581 | return hc | |
582 | ||
583 | ||
acebc9cd | 584 | class YoutubeDLHandler(compat_urllib_request.HTTPHandler): |
59ae15a5 PH |
585 | """Handler for HTTP requests and responses. |
586 | ||
587 | This class, when installed with an OpenerDirector, automatically adds | |
588 | the standard headers to every HTTP request and handles gzipped and | |
589 | deflated responses from web servers. If compression is to be avoided in | |
590 | a particular request, the original request in the program code only has | |
591 | to include the HTTP header "Youtubedl-No-Compression", which will be | |
592 | removed before making the real request. | |
593 | ||
594 | Part of this code was copied from: | |
595 | ||
596 | http://techknack.net/python-urllib2-handlers/ | |
597 | ||
598 | Andrew Rowls, the author of that code, agreed to release it to the | |
599 | public domain. | |
600 | """ | |
601 | ||
be4a824d PH |
602 | def __init__(self, params, *args, **kwargs): |
603 | compat_urllib_request.HTTPHandler.__init__(self, *args, **kwargs) | |
604 | self._params = params | |
605 | ||
606 | def http_open(self, req): | |
607 | return self.do_open(functools.partial( | |
c5a59d93 | 608 | _create_http_connection, self, compat_http_client.HTTPConnection, False), |
be4a824d PH |
609 | req) |
610 | ||
59ae15a5 PH |
611 | @staticmethod |
612 | def deflate(data): | |
613 | try: | |
614 | return zlib.decompress(data, -zlib.MAX_WBITS) | |
615 | except zlib.error: | |
616 | return zlib.decompress(data) | |
617 | ||
618 | @staticmethod | |
619 | def addinfourl_wrapper(stream, headers, url, code): | |
620 | if hasattr(compat_urllib_request.addinfourl, 'getcode'): | |
621 | return compat_urllib_request.addinfourl(stream, headers, url, code) | |
622 | ret = compat_urllib_request.addinfourl(stream, headers, url) | |
623 | ret.code = code | |
624 | return ret | |
625 | ||
acebc9cd | 626 | def http_request(self, req): |
33ac271b | 627 | for h, v in std_headers.items(): |
3d5f7a39 JK |
628 | # Capitalize is needed because of Python bug 2275: http://bugs.python.org/issue2275 |
629 | # The dict keys are capitalized because of this bug by urllib | |
630 | if h.capitalize() not in req.headers: | |
33ac271b | 631 | req.add_header(h, v) |
59ae15a5 PH |
632 | if 'Youtubedl-no-compression' in req.headers: |
633 | if 'Accept-encoding' in req.headers: | |
634 | del req.headers['Accept-encoding'] | |
635 | del req.headers['Youtubedl-no-compression'] | |
989b4b2b PH |
636 | |
637 | if sys.version_info < (2, 7) and '#' in req.get_full_url(): | |
638 | # Python 2.6 is brain-dead when it comes to fragments | |
639 | req._Request__original = req._Request__original.partition('#')[0] | |
640 | req._Request__r_type = req._Request__r_type.partition('#')[0] | |
641 | ||
59ae15a5 PH |
642 | return req |
643 | ||
acebc9cd | 644 | def http_response(self, req, resp): |
59ae15a5 PH |
645 | old_resp = resp |
646 | # gzip | |
647 | if resp.headers.get('Content-encoding', '') == 'gzip': | |
aa3e9507 PH |
648 | content = resp.read() |
649 | gz = gzip.GzipFile(fileobj=io.BytesIO(content), mode='rb') | |
650 | try: | |
651 | uncompressed = io.BytesIO(gz.read()) | |
652 | except IOError as original_ioerror: | |
653 | # There may be junk add the end of the file | |
654 | # See http://stackoverflow.com/q/4928560/35070 for details | |
655 | for i in range(1, 1024): | |
656 | try: | |
657 | gz = gzip.GzipFile(fileobj=io.BytesIO(content[:-i]), mode='rb') | |
658 | uncompressed = io.BytesIO(gz.read()) | |
659 | except IOError: | |
660 | continue | |
661 | break | |
662 | else: | |
663 | raise original_ioerror | |
664 | resp = self.addinfourl_wrapper(uncompressed, old_resp.headers, old_resp.url, old_resp.code) | |
59ae15a5 PH |
665 | resp.msg = old_resp.msg |
666 | # deflate | |
667 | if resp.headers.get('Content-encoding', '') == 'deflate': | |
668 | gz = io.BytesIO(self.deflate(resp.read())) | |
669 | resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code) | |
670 | resp.msg = old_resp.msg | |
671 | return resp | |
0f8d03f8 | 672 | |
acebc9cd PH |
673 | https_request = http_request |
674 | https_response = http_response | |
bf50b038 | 675 | |
5de90176 | 676 | |
be4a824d PH |
677 | class YoutubeDLHTTPSHandler(compat_urllib_request.HTTPSHandler): |
678 | def __init__(self, params, https_conn_class=None, *args, **kwargs): | |
679 | compat_urllib_request.HTTPSHandler.__init__(self, *args, **kwargs) | |
680 | self._https_conn_class = https_conn_class or compat_http_client.HTTPSConnection | |
681 | self._params = params | |
682 | ||
683 | def https_open(self, req): | |
4f264c02 JMF |
684 | kwargs = {} |
685 | if hasattr(self, '_context'): # python > 2.6 | |
686 | kwargs['context'] = self._context | |
687 | if hasattr(self, '_check_hostname'): # python 3.x | |
688 | kwargs['check_hostname'] = self._check_hostname | |
be4a824d PH |
689 | return self.do_open(functools.partial( |
690 | _create_http_connection, self, self._https_conn_class, True), | |
4f264c02 | 691 | req, **kwargs) |
be4a824d PH |
692 | |
693 | ||
08b38d54 | 694 | def parse_iso8601(date_str, delimiter='T', timezone=None): |
912b38b4 PH |
695 | """ Return a UNIX timestamp from the given date """ |
696 | ||
697 | if date_str is None: | |
698 | return None | |
699 | ||
08b38d54 PH |
700 | if timezone is None: |
701 | m = re.search( | |
702 | r'(\.[0-9]+)?(?:Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)', | |
703 | date_str) | |
704 | if not m: | |
912b38b4 PH |
705 | timezone = datetime.timedelta() |
706 | else: | |
08b38d54 PH |
707 | date_str = date_str[:-len(m.group(0))] |
708 | if not m.group('sign'): | |
709 | timezone = datetime.timedelta() | |
710 | else: | |
711 | sign = 1 if m.group('sign') == '+' else -1 | |
712 | timezone = datetime.timedelta( | |
713 | hours=sign * int(m.group('hours')), | |
714 | minutes=sign * int(m.group('minutes'))) | |
6ad4013d | 715 | date_format = '%Y-%m-%d{0}%H:%M:%S'.format(delimiter) |
305d0683 | 716 | dt = datetime.datetime.strptime(date_str, date_format) - timezone |
912b38b4 PH |
717 | return calendar.timegm(dt.timetuple()) |
718 | ||
719 | ||
42bdd9d0 | 720 | def unified_strdate(date_str, day_first=True): |
bf50b038 | 721 | """Return a string with the date in the format YYYYMMDD""" |
64e7ad60 PH |
722 | |
723 | if date_str is None: | |
724 | return None | |
bf50b038 | 725 | upload_date = None |
5f6a1245 | 726 | # Replace commas |
026fcc04 | 727 | date_str = date_str.replace(',', ' ') |
bf50b038 | 728 | # %z (UTC offset) is only supported in python>=3.2 |
026fcc04 | 729 | date_str = re.sub(r' ?(\+|-)[0-9]{2}:?[0-9]{2}$', '', date_str) |
42bdd9d0 | 730 | # Remove AM/PM + timezone |
9bb8e0a3 | 731 | date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str) |
42bdd9d0 | 732 | |
19e1d359 JMF |
733 | format_expressions = [ |
734 | '%d %B %Y', | |
0f99566c | 735 | '%d %b %Y', |
19e1d359 JMF |
736 | '%B %d %Y', |
737 | '%b %d %Y', | |
78ff59d0 PP |
738 | '%b %dst %Y %I:%M%p', |
739 | '%b %dnd %Y %I:%M%p', | |
740 | '%b %dth %Y %I:%M%p', | |
a69801e2 | 741 | '%Y %m %d', |
19e1d359 | 742 | '%Y-%m-%d', |
fe556f1b | 743 | '%Y/%m/%d', |
19e1d359 | 744 | '%Y/%m/%d %H:%M:%S', |
5d73273f | 745 | '%Y-%m-%d %H:%M:%S', |
e9be9a6a | 746 | '%Y-%m-%d %H:%M:%S.%f', |
19e1d359 | 747 | '%d.%m.%Y %H:%M', |
b047de6f | 748 | '%d.%m.%Y %H.%M', |
19e1d359 | 749 | '%Y-%m-%dT%H:%M:%SZ', |
59040888 PH |
750 | '%Y-%m-%dT%H:%M:%S.%fZ', |
751 | '%Y-%m-%dT%H:%M:%S.%f0Z', | |
2e1fa03b | 752 | '%Y-%m-%dT%H:%M:%S', |
7ff5d5c2 | 753 | '%Y-%m-%dT%H:%M:%S.%f', |
5de90176 | 754 | '%Y-%m-%dT%H:%M', |
19e1d359 | 755 | ] |
42bdd9d0 PH |
756 | if day_first: |
757 | format_expressions.extend([ | |
776dc399 S |
758 | '%d.%m.%Y', |
759 | '%d/%m/%Y', | |
760 | '%d/%m/%y', | |
42bdd9d0 PH |
761 | '%d/%m/%Y %H:%M:%S', |
762 | ]) | |
763 | else: | |
764 | format_expressions.extend([ | |
776dc399 S |
765 | '%m.%d.%Y', |
766 | '%m/%d/%Y', | |
767 | '%m/%d/%y', | |
42bdd9d0 PH |
768 | '%m/%d/%Y %H:%M:%S', |
769 | ]) | |
bf50b038 JMF |
770 | for expression in format_expressions: |
771 | try: | |
772 | upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d') | |
5de90176 | 773 | except ValueError: |
bf50b038 | 774 | pass |
42393ce2 PH |
775 | if upload_date is None: |
776 | timetuple = email.utils.parsedate_tz(date_str) | |
777 | if timetuple: | |
778 | upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d') | |
bf50b038 JMF |
779 | return upload_date |
780 | ||
5f6a1245 | 781 | |
28e614de | 782 | def determine_ext(url, default_ext='unknown_video'): |
f4776371 S |
783 | if url is None: |
784 | return default_ext | |
28e614de | 785 | guess = url.partition('?')[0].rpartition('.')[2] |
73e79f2a PH |
786 | if re.match(r'^[A-Za-z0-9]+$', guess): |
787 | return guess | |
788 | else: | |
cbdbb766 | 789 | return default_ext |
73e79f2a | 790 | |
5f6a1245 | 791 | |
d4051a8e | 792 | def subtitles_filename(filename, sub_lang, sub_format): |
28e614de | 793 | return filename.rsplit('.', 1)[0] + '.' + sub_lang + '.' + sub_format |
d4051a8e | 794 | |
5f6a1245 | 795 | |
bd558525 | 796 | def date_from_str(date_str): |
37254abc JMF |
797 | """ |
798 | Return a datetime object from a string in the format YYYYMMDD or | |
799 | (now|today)[+-][0-9](day|week|month|year)(s)?""" | |
800 | today = datetime.date.today() | |
f8795e10 | 801 | if date_str in ('now', 'today'): |
37254abc | 802 | return today |
f8795e10 PH |
803 | if date_str == 'yesterday': |
804 | return today - datetime.timedelta(days=1) | |
37254abc JMF |
805 | match = re.match('(now|today)(?P<sign>[+-])(?P<time>\d+)(?P<unit>day|week|month|year)(s)?', date_str) |
806 | if match is not None: | |
807 | sign = match.group('sign') | |
808 | time = int(match.group('time')) | |
809 | if sign == '-': | |
810 | time = -time | |
811 | unit = match.group('unit') | |
5f6a1245 | 812 | # A bad aproximation? |
37254abc JMF |
813 | if unit == 'month': |
814 | unit = 'day' | |
815 | time *= 30 | |
816 | elif unit == 'year': | |
817 | unit = 'day' | |
818 | time *= 365 | |
819 | unit += 's' | |
820 | delta = datetime.timedelta(**{unit: time}) | |
821 | return today + delta | |
bd558525 | 822 | return datetime.datetime.strptime(date_str, "%Y%m%d").date() |
5f6a1245 JW |
823 | |
824 | ||
e63fc1be | 825 | def hyphenate_date(date_str): |
826 | """ | |
827 | Convert a date in 'YYYYMMDD' format to 'YYYY-MM-DD' format""" | |
828 | match = re.match(r'^(\d\d\d\d)(\d\d)(\d\d)$', date_str) | |
829 | if match is not None: | |
830 | return '-'.join(match.groups()) | |
831 | else: | |
832 | return date_str | |
833 | ||
5f6a1245 | 834 | |
bd558525 JMF |
835 | class DateRange(object): |
836 | """Represents a time interval between two dates""" | |
5f6a1245 | 837 | |
bd558525 JMF |
838 | def __init__(self, start=None, end=None): |
839 | """start and end must be strings in the format accepted by date""" | |
840 | if start is not None: | |
841 | self.start = date_from_str(start) | |
842 | else: | |
843 | self.start = datetime.datetime.min.date() | |
844 | if end is not None: | |
845 | self.end = date_from_str(end) | |
846 | else: | |
847 | self.end = datetime.datetime.max.date() | |
37254abc | 848 | if self.start > self.end: |
bd558525 | 849 | raise ValueError('Date range: "%s" , the start date must be before the end date' % self) |
5f6a1245 | 850 | |
bd558525 JMF |
851 | @classmethod |
852 | def day(cls, day): | |
853 | """Returns a range that only contains the given day""" | |
5f6a1245 JW |
854 | return cls(day, day) |
855 | ||
bd558525 JMF |
856 | def __contains__(self, date): |
857 | """Check if the date is in the range""" | |
37254abc JMF |
858 | if not isinstance(date, datetime.date): |
859 | date = date_from_str(date) | |
860 | return self.start <= date <= self.end | |
5f6a1245 | 861 | |
bd558525 | 862 | def __str__(self): |
5f6a1245 | 863 | return '%s - %s' % (self.start.isoformat(), self.end.isoformat()) |
c496ca96 PH |
864 | |
865 | ||
866 | def platform_name(): | |
867 | """ Returns the platform name as a compat_str """ | |
868 | res = platform.platform() | |
869 | if isinstance(res, bytes): | |
870 | res = res.decode(preferredencoding()) | |
871 | ||
872 | assert isinstance(res, compat_str) | |
873 | return res | |
c257baff PH |
874 | |
875 | ||
b58ddb32 PH |
876 | def _windows_write_string(s, out): |
877 | """ Returns True if the string was written using special methods, | |
878 | False if it has yet to be written out.""" | |
879 | # Adapted from http://stackoverflow.com/a/3259271/35070 | |
880 | ||
881 | import ctypes | |
882 | import ctypes.wintypes | |
883 | ||
884 | WIN_OUTPUT_IDS = { | |
885 | 1: -11, | |
886 | 2: -12, | |
887 | } | |
888 | ||
a383a98a PH |
889 | try: |
890 | fileno = out.fileno() | |
891 | except AttributeError: | |
892 | # If the output stream doesn't have a fileno, it's virtual | |
893 | return False | |
aa42e873 PH |
894 | except io.UnsupportedOperation: |
895 | # Some strange Windows pseudo files? | |
896 | return False | |
b58ddb32 PH |
897 | if fileno not in WIN_OUTPUT_IDS: |
898 | return False | |
899 | ||
e2f89ec7 | 900 | GetStdHandle = ctypes.WINFUNCTYPE( |
b58ddb32 | 901 | ctypes.wintypes.HANDLE, ctypes.wintypes.DWORD)( |
6ac4e806 | 902 | (b"GetStdHandle", ctypes.windll.kernel32)) |
b58ddb32 PH |
903 | h = GetStdHandle(WIN_OUTPUT_IDS[fileno]) |
904 | ||
e2f89ec7 | 905 | WriteConsoleW = ctypes.WINFUNCTYPE( |
b58ddb32 PH |
906 | ctypes.wintypes.BOOL, ctypes.wintypes.HANDLE, ctypes.wintypes.LPWSTR, |
907 | ctypes.wintypes.DWORD, ctypes.POINTER(ctypes.wintypes.DWORD), | |
6ac4e806 | 908 | ctypes.wintypes.LPVOID)((b"WriteConsoleW", ctypes.windll.kernel32)) |
b58ddb32 PH |
909 | written = ctypes.wintypes.DWORD(0) |
910 | ||
6ac4e806 | 911 | GetFileType = ctypes.WINFUNCTYPE(ctypes.wintypes.DWORD, ctypes.wintypes.DWORD)((b"GetFileType", ctypes.windll.kernel32)) |
b58ddb32 PH |
912 | FILE_TYPE_CHAR = 0x0002 |
913 | FILE_TYPE_REMOTE = 0x8000 | |
e2f89ec7 | 914 | GetConsoleMode = ctypes.WINFUNCTYPE( |
b58ddb32 PH |
915 | ctypes.wintypes.BOOL, ctypes.wintypes.HANDLE, |
916 | ctypes.POINTER(ctypes.wintypes.DWORD))( | |
6ac4e806 | 917 | (b"GetConsoleMode", ctypes.windll.kernel32)) |
b58ddb32 PH |
918 | INVALID_HANDLE_VALUE = ctypes.wintypes.DWORD(-1).value |
919 | ||
920 | def not_a_console(handle): | |
921 | if handle == INVALID_HANDLE_VALUE or handle is None: | |
922 | return True | |
8fb3ac36 PH |
923 | return ((GetFileType(handle) & ~FILE_TYPE_REMOTE) != FILE_TYPE_CHAR or |
924 | GetConsoleMode(handle, ctypes.byref(ctypes.wintypes.DWORD())) == 0) | |
b58ddb32 PH |
925 | |
926 | if not_a_console(h): | |
927 | return False | |
928 | ||
d1b9c912 PH |
929 | def next_nonbmp_pos(s): |
930 | try: | |
931 | return next(i for i, c in enumerate(s) if ord(c) > 0xffff) | |
932 | except StopIteration: | |
933 | return len(s) | |
934 | ||
935 | while s: | |
936 | count = min(next_nonbmp_pos(s), 1024) | |
937 | ||
b58ddb32 | 938 | ret = WriteConsoleW( |
d1b9c912 | 939 | h, s, count if count else 2, ctypes.byref(written), None) |
b58ddb32 PH |
940 | if ret == 0: |
941 | raise OSError('Failed to write string') | |
d1b9c912 PH |
942 | if not count: # We just wrote a non-BMP character |
943 | assert written.value == 2 | |
944 | s = s[1:] | |
945 | else: | |
946 | assert written.value > 0 | |
947 | s = s[written.value:] | |
b58ddb32 PH |
948 | return True |
949 | ||
950 | ||
734f90bb | 951 | def write_string(s, out=None, encoding=None): |
7459e3a2 PH |
952 | if out is None: |
953 | out = sys.stderr | |
8bf48f23 | 954 | assert type(s) == compat_str |
7459e3a2 | 955 | |
b58ddb32 PH |
956 | if sys.platform == 'win32' and encoding is None and hasattr(out, 'fileno'): |
957 | if _windows_write_string(s, out): | |
958 | return | |
959 | ||
7459e3a2 PH |
960 | if ('b' in getattr(out, 'mode', '') or |
961 | sys.version_info[0] < 3): # Python 2 lies about mode of sys.stderr | |
104aa738 PH |
962 | byt = s.encode(encoding or preferredencoding(), 'ignore') |
963 | out.write(byt) | |
964 | elif hasattr(out, 'buffer'): | |
965 | enc = encoding or getattr(out, 'encoding', None) or preferredencoding() | |
966 | byt = s.encode(enc, 'ignore') | |
967 | out.buffer.write(byt) | |
968 | else: | |
8bf48f23 | 969 | out.write(s) |
7459e3a2 PH |
970 | out.flush() |
971 | ||
972 | ||
48ea9cea PH |
973 | def bytes_to_intlist(bs): |
974 | if not bs: | |
975 | return [] | |
976 | if isinstance(bs[0], int): # Python 3 | |
977 | return list(bs) | |
978 | else: | |
979 | return [ord(c) for c in bs] | |
980 | ||
c257baff | 981 | |
cba892fa | 982 | def intlist_to_bytes(xs): |
983 | if not xs: | |
984 | return b'' | |
eb4157fd | 985 | return struct_pack('%dB' % len(xs), *xs) |
c38b1e77 PH |
986 | |
987 | ||
c1c9a79c PH |
988 | # Cross-platform file locking |
989 | if sys.platform == 'win32': | |
990 | import ctypes.wintypes | |
991 | import msvcrt | |
992 | ||
993 | class OVERLAPPED(ctypes.Structure): | |
994 | _fields_ = [ | |
995 | ('Internal', ctypes.wintypes.LPVOID), | |
996 | ('InternalHigh', ctypes.wintypes.LPVOID), | |
997 | ('Offset', ctypes.wintypes.DWORD), | |
998 | ('OffsetHigh', ctypes.wintypes.DWORD), | |
999 | ('hEvent', ctypes.wintypes.HANDLE), | |
1000 | ] | |
1001 | ||
1002 | kernel32 = ctypes.windll.kernel32 | |
1003 | LockFileEx = kernel32.LockFileEx | |
1004 | LockFileEx.argtypes = [ | |
1005 | ctypes.wintypes.HANDLE, # hFile | |
1006 | ctypes.wintypes.DWORD, # dwFlags | |
1007 | ctypes.wintypes.DWORD, # dwReserved | |
1008 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow | |
1009 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh | |
1010 | ctypes.POINTER(OVERLAPPED) # Overlapped | |
1011 | ] | |
1012 | LockFileEx.restype = ctypes.wintypes.BOOL | |
1013 | UnlockFileEx = kernel32.UnlockFileEx | |
1014 | UnlockFileEx.argtypes = [ | |
1015 | ctypes.wintypes.HANDLE, # hFile | |
1016 | ctypes.wintypes.DWORD, # dwReserved | |
1017 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow | |
1018 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh | |
1019 | ctypes.POINTER(OVERLAPPED) # Overlapped | |
1020 | ] | |
1021 | UnlockFileEx.restype = ctypes.wintypes.BOOL | |
1022 | whole_low = 0xffffffff | |
1023 | whole_high = 0x7fffffff | |
1024 | ||
1025 | def _lock_file(f, exclusive): | |
1026 | overlapped = OVERLAPPED() | |
1027 | overlapped.Offset = 0 | |
1028 | overlapped.OffsetHigh = 0 | |
1029 | overlapped.hEvent = 0 | |
1030 | f._lock_file_overlapped_p = ctypes.pointer(overlapped) | |
1031 | handle = msvcrt.get_osfhandle(f.fileno()) | |
1032 | if not LockFileEx(handle, 0x2 if exclusive else 0x0, 0, | |
1033 | whole_low, whole_high, f._lock_file_overlapped_p): | |
1034 | raise OSError('Locking file failed: %r' % ctypes.FormatError()) | |
1035 | ||
1036 | def _unlock_file(f): | |
1037 | assert f._lock_file_overlapped_p | |
1038 | handle = msvcrt.get_osfhandle(f.fileno()) | |
1039 | if not UnlockFileEx(handle, 0, | |
1040 | whole_low, whole_high, f._lock_file_overlapped_p): | |
1041 | raise OSError('Unlocking file failed: %r' % ctypes.FormatError()) | |
1042 | ||
1043 | else: | |
1044 | import fcntl | |
1045 | ||
1046 | def _lock_file(f, exclusive): | |
2582bebe | 1047 | fcntl.flock(f, fcntl.LOCK_EX if exclusive else fcntl.LOCK_SH) |
c1c9a79c PH |
1048 | |
1049 | def _unlock_file(f): | |
2582bebe | 1050 | fcntl.flock(f, fcntl.LOCK_UN) |
c1c9a79c PH |
1051 | |
1052 | ||
1053 | class locked_file(object): | |
1054 | def __init__(self, filename, mode, encoding=None): | |
1055 | assert mode in ['r', 'a', 'w'] | |
1056 | self.f = io.open(filename, mode, encoding=encoding) | |
1057 | self.mode = mode | |
1058 | ||
1059 | def __enter__(self): | |
1060 | exclusive = self.mode != 'r' | |
1061 | try: | |
1062 | _lock_file(self.f, exclusive) | |
1063 | except IOError: | |
1064 | self.f.close() | |
1065 | raise | |
1066 | return self | |
1067 | ||
1068 | def __exit__(self, etype, value, traceback): | |
1069 | try: | |
1070 | _unlock_file(self.f) | |
1071 | finally: | |
1072 | self.f.close() | |
1073 | ||
1074 | def __iter__(self): | |
1075 | return iter(self.f) | |
1076 | ||
1077 | def write(self, *args): | |
1078 | return self.f.write(*args) | |
1079 | ||
1080 | def read(self, *args): | |
1081 | return self.f.read(*args) | |
4eb7f1d1 JMF |
1082 | |
1083 | ||
4644ac55 S |
1084 | def get_filesystem_encoding(): |
1085 | encoding = sys.getfilesystemencoding() | |
1086 | return encoding if encoding is not None else 'utf-8' | |
1087 | ||
1088 | ||
4eb7f1d1 | 1089 | def shell_quote(args): |
a6a173c2 | 1090 | quoted_args = [] |
4644ac55 | 1091 | encoding = get_filesystem_encoding() |
a6a173c2 JMF |
1092 | for a in args: |
1093 | if isinstance(a, bytes): | |
1094 | # We may get a filename encoded with 'encodeFilename' | |
1095 | a = a.decode(encoding) | |
1096 | quoted_args.append(pipes.quote(a)) | |
28e614de | 1097 | return ' '.join(quoted_args) |
9d4660ca PH |
1098 | |
1099 | ||
f4d96df0 PH |
1100 | def takewhile_inclusive(pred, seq): |
1101 | """ Like itertools.takewhile, but include the latest evaluated element | |
1102 | (the first element so that Not pred(e)) """ | |
1103 | for e in seq: | |
1104 | yield e | |
1105 | if not pred(e): | |
1106 | return | |
1107 | ||
1108 | ||
9d4660ca PH |
1109 | def smuggle_url(url, data): |
1110 | """ Pass additional data in a URL for internal use. """ | |
1111 | ||
1112 | sdata = compat_urllib_parse.urlencode( | |
28e614de PH |
1113 | {'__youtubedl_smuggle': json.dumps(data)}) |
1114 | return url + '#' + sdata | |
9d4660ca PH |
1115 | |
1116 | ||
79f82953 | 1117 | def unsmuggle_url(smug_url, default=None): |
83e865a3 | 1118 | if '#__youtubedl_smuggle' not in smug_url: |
79f82953 | 1119 | return smug_url, default |
28e614de PH |
1120 | url, _, sdata = smug_url.rpartition('#') |
1121 | jsond = compat_parse_qs(sdata)['__youtubedl_smuggle'][0] | |
9d4660ca PH |
1122 | data = json.loads(jsond) |
1123 | return url, data | |
02dbf93f PH |
1124 | |
1125 | ||
02dbf93f PH |
1126 | def format_bytes(bytes): |
1127 | if bytes is None: | |
28e614de | 1128 | return 'N/A' |
02dbf93f PH |
1129 | if type(bytes) is str: |
1130 | bytes = float(bytes) | |
1131 | if bytes == 0.0: | |
1132 | exponent = 0 | |
1133 | else: | |
1134 | exponent = int(math.log(bytes, 1024.0)) | |
28e614de | 1135 | suffix = ['B', 'KiB', 'MiB', 'GiB', 'TiB', 'PiB', 'EiB', 'ZiB', 'YiB'][exponent] |
02dbf93f | 1136 | converted = float(bytes) / float(1024 ** exponent) |
28e614de | 1137 | return '%.2f%s' % (converted, suffix) |
f53c966a | 1138 | |
1c088fa8 | 1139 | |
be64b5b0 PH |
1140 | def parse_filesize(s): |
1141 | if s is None: | |
1142 | return None | |
1143 | ||
1144 | # The lower-case forms are of course incorrect and inofficial, | |
1145 | # but we support those too | |
1146 | _UNIT_TABLE = { | |
1147 | 'B': 1, | |
1148 | 'b': 1, | |
1149 | 'KiB': 1024, | |
1150 | 'KB': 1000, | |
1151 | 'kB': 1024, | |
1152 | 'Kb': 1000, | |
1153 | 'MiB': 1024 ** 2, | |
1154 | 'MB': 1000 ** 2, | |
1155 | 'mB': 1024 ** 2, | |
1156 | 'Mb': 1000 ** 2, | |
1157 | 'GiB': 1024 ** 3, | |
1158 | 'GB': 1000 ** 3, | |
1159 | 'gB': 1024 ** 3, | |
1160 | 'Gb': 1000 ** 3, | |
1161 | 'TiB': 1024 ** 4, | |
1162 | 'TB': 1000 ** 4, | |
1163 | 'tB': 1024 ** 4, | |
1164 | 'Tb': 1000 ** 4, | |
1165 | 'PiB': 1024 ** 5, | |
1166 | 'PB': 1000 ** 5, | |
1167 | 'pB': 1024 ** 5, | |
1168 | 'Pb': 1000 ** 5, | |
1169 | 'EiB': 1024 ** 6, | |
1170 | 'EB': 1000 ** 6, | |
1171 | 'eB': 1024 ** 6, | |
1172 | 'Eb': 1000 ** 6, | |
1173 | 'ZiB': 1024 ** 7, | |
1174 | 'ZB': 1000 ** 7, | |
1175 | 'zB': 1024 ** 7, | |
1176 | 'Zb': 1000 ** 7, | |
1177 | 'YiB': 1024 ** 8, | |
1178 | 'YB': 1000 ** 8, | |
1179 | 'yB': 1024 ** 8, | |
1180 | 'Yb': 1000 ** 8, | |
1181 | } | |
1182 | ||
1183 | units_re = '|'.join(re.escape(u) for u in _UNIT_TABLE) | |
4349c07d PH |
1184 | m = re.match( |
1185 | r'(?P<num>[0-9]+(?:[,.][0-9]*)?)\s*(?P<unit>%s)' % units_re, s) | |
be64b5b0 PH |
1186 | if not m: |
1187 | return None | |
1188 | ||
4349c07d PH |
1189 | num_str = m.group('num').replace(',', '.') |
1190 | mult = _UNIT_TABLE[m.group('unit')] | |
1191 | return int(float(num_str) * mult) | |
be64b5b0 PH |
1192 | |
1193 | ||
caefb1de PH |
1194 | def month_by_name(name): |
1195 | """ Return the number of a month by (locale-independently) English name """ | |
1196 | ||
caefb1de | 1197 | try: |
7105440c YCH |
1198 | return ENGLISH_MONTH_NAMES.index(name) + 1 |
1199 | except ValueError: | |
1200 | return None | |
1201 | ||
1202 | ||
1203 | def month_by_abbreviation(abbrev): | |
1204 | """ Return the number of a month by (locale-independently) English | |
1205 | abbreviations """ | |
1206 | ||
1207 | try: | |
1208 | return [s[:3] for s in ENGLISH_MONTH_NAMES].index(abbrev) + 1 | |
caefb1de PH |
1209 | except ValueError: |
1210 | return None | |
18258362 JMF |
1211 | |
1212 | ||
5aafe895 | 1213 | def fix_xml_ampersands(xml_str): |
18258362 | 1214 | """Replace all the '&' by '&' in XML""" |
5aafe895 PH |
1215 | return re.sub( |
1216 | r'&(?!amp;|lt;|gt;|apos;|quot;|#x[0-9a-fA-F]{,4};|#[0-9]{,4};)', | |
28e614de | 1217 | '&', |
5aafe895 | 1218 | xml_str) |
e3946f98 PH |
1219 | |
1220 | ||
1221 | def setproctitle(title): | |
8bf48f23 | 1222 | assert isinstance(title, compat_str) |
e3946f98 PH |
1223 | try: |
1224 | libc = ctypes.cdll.LoadLibrary("libc.so.6") | |
1225 | except OSError: | |
1226 | return | |
6eefe533 PH |
1227 | title_bytes = title.encode('utf-8') |
1228 | buf = ctypes.create_string_buffer(len(title_bytes)) | |
1229 | buf.value = title_bytes | |
e3946f98 | 1230 | try: |
6eefe533 | 1231 | libc.prctl(15, buf, 0, 0, 0) |
e3946f98 PH |
1232 | except AttributeError: |
1233 | return # Strange libc, just skip this | |
d7dda168 PH |
1234 | |
1235 | ||
1236 | def remove_start(s, start): | |
1237 | if s.startswith(start): | |
1238 | return s[len(start):] | |
1239 | return s | |
29eb5174 PH |
1240 | |
1241 | ||
2b9faf55 PH |
1242 | def remove_end(s, end): |
1243 | if s.endswith(end): | |
1244 | return s[:-len(end)] | |
1245 | return s | |
1246 | ||
1247 | ||
29eb5174 | 1248 | def url_basename(url): |
9b8aaeed | 1249 | path = compat_urlparse.urlparse(url).path |
28e614de | 1250 | return path.strip('/').split('/')[-1] |
aa94a6d3 PH |
1251 | |
1252 | ||
1253 | class HEADRequest(compat_urllib_request.Request): | |
1254 | def get_method(self): | |
1255 | return "HEAD" | |
7217e148 PH |
1256 | |
1257 | ||
9732d77e | 1258 | def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1): |
28746fbd PH |
1259 | if get_attr: |
1260 | if v is not None: | |
1261 | v = getattr(v, get_attr, None) | |
9572013d PH |
1262 | if v == '': |
1263 | v = None | |
9732d77e PH |
1264 | return default if v is None else (int(v) * invscale // scale) |
1265 | ||
9572013d | 1266 | |
40a90862 JMF |
1267 | def str_or_none(v, default=None): |
1268 | return default if v is None else compat_str(v) | |
1269 | ||
9732d77e PH |
1270 | |
1271 | def str_to_int(int_str): | |
48d4681e | 1272 | """ A more relaxed version of int_or_none """ |
9732d77e PH |
1273 | if int_str is None: |
1274 | return None | |
28e614de | 1275 | int_str = re.sub(r'[,\.\+]', '', int_str) |
9732d77e | 1276 | return int(int_str) |
608d11f5 PH |
1277 | |
1278 | ||
9732d77e PH |
1279 | def float_or_none(v, scale=1, invscale=1, default=None): |
1280 | return default if v is None else (float(v) * invscale / scale) | |
43f775e4 PH |
1281 | |
1282 | ||
608d11f5 | 1283 | def parse_duration(s): |
8f9312c3 | 1284 | if not isinstance(s, compat_basestring): |
608d11f5 PH |
1285 | return None |
1286 | ||
ca7b3246 S |
1287 | s = s.strip() |
1288 | ||
608d11f5 | 1289 | m = re.match( |
9d22a7df | 1290 | r'''(?ix)(?:P?T)? |
e8df5cee PH |
1291 | (?: |
1292 | (?P<only_mins>[0-9.]+)\s*(?:mins?|minutes?)\s*| | |
1293 | (?P<only_hours>[0-9.]+)\s*(?:hours?)| | |
1294 | ||
3e675fab | 1295 | \s*(?P<hours_reversed>[0-9]+)\s*(?:[:h]|hours?)\s*(?P<mins_reversed>[0-9]+)\s*(?:[:m]|mins?|minutes?)\s*| |
6a68bb57 | 1296 | (?: |
8f4b58d7 PH |
1297 | (?: |
1298 | (?:(?P<days>[0-9]+)\s*(?:[:d]|days?)\s*)? | |
1299 | (?P<hours>[0-9]+)\s*(?:[:h]|hours?)\s* | |
1300 | )? | |
6a68bb57 PH |
1301 | (?P<mins>[0-9]+)\s*(?:[:m]|mins?|minutes?)\s* |
1302 | )? | |
e8df5cee PH |
1303 | (?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*(?:s|secs?|seconds?)? |
1304 | )$''', s) | |
608d11f5 PH |
1305 | if not m: |
1306 | return None | |
e8df5cee PH |
1307 | res = 0 |
1308 | if m.group('only_mins'): | |
1309 | return float_or_none(m.group('only_mins'), invscale=60) | |
1310 | if m.group('only_hours'): | |
1311 | return float_or_none(m.group('only_hours'), invscale=60 * 60) | |
1312 | if m.group('secs'): | |
1313 | res += int(m.group('secs')) | |
3e675fab PH |
1314 | if m.group('mins_reversed'): |
1315 | res += int(m.group('mins_reversed')) * 60 | |
608d11f5 PH |
1316 | if m.group('mins'): |
1317 | res += int(m.group('mins')) * 60 | |
e8df5cee PH |
1318 | if m.group('hours'): |
1319 | res += int(m.group('hours')) * 60 * 60 | |
3e675fab PH |
1320 | if m.group('hours_reversed'): |
1321 | res += int(m.group('hours_reversed')) * 60 * 60 | |
8f4b58d7 PH |
1322 | if m.group('days'): |
1323 | res += int(m.group('days')) * 24 * 60 * 60 | |
7adcbe75 PH |
1324 | if m.group('ms'): |
1325 | res += float(m.group('ms')) | |
608d11f5 | 1326 | return res |
91d7d0b3 JMF |
1327 | |
1328 | ||
1329 | def prepend_extension(filename, ext): | |
5f6a1245 | 1330 | name, real_ext = os.path.splitext(filename) |
28e614de | 1331 | return '{0}.{1}{2}'.format(name, ext, real_ext) |
d70ad093 PH |
1332 | |
1333 | ||
1334 | def check_executable(exe, args=[]): | |
1335 | """ Checks if the given binary is installed somewhere in PATH, and returns its name. | |
1336 | args can be a list of arguments for a short output (like -version) """ | |
1337 | try: | |
1338 | subprocess.Popen([exe] + args, stdout=subprocess.PIPE, stderr=subprocess.PIPE).communicate() | |
1339 | except OSError: | |
1340 | return False | |
1341 | return exe | |
b7ab0590 PH |
1342 | |
1343 | ||
95807118 | 1344 | def get_exe_version(exe, args=['--version'], |
cae97f65 | 1345 | version_re=None, unrecognized='present'): |
95807118 PH |
1346 | """ Returns the version of the specified executable, |
1347 | or False if the executable is not present """ | |
1348 | try: | |
cae97f65 | 1349 | out, _ = subprocess.Popen( |
95807118 PH |
1350 | [exe] + args, |
1351 | stdout=subprocess.PIPE, stderr=subprocess.STDOUT).communicate() | |
1352 | except OSError: | |
1353 | return False | |
cae97f65 PH |
1354 | if isinstance(out, bytes): # Python 2.x |
1355 | out = out.decode('ascii', 'ignore') | |
1356 | return detect_exe_version(out, version_re, unrecognized) | |
1357 | ||
1358 | ||
1359 | def detect_exe_version(output, version_re=None, unrecognized='present'): | |
1360 | assert isinstance(output, compat_str) | |
1361 | if version_re is None: | |
1362 | version_re = r'version\s+([-0-9._a-zA-Z]+)' | |
1363 | m = re.search(version_re, output) | |
95807118 PH |
1364 | if m: |
1365 | return m.group(1) | |
1366 | else: | |
1367 | return unrecognized | |
1368 | ||
1369 | ||
b7ab0590 | 1370 | class PagedList(object): |
dd26ced1 PH |
1371 | def __len__(self): |
1372 | # This is only useful for tests | |
1373 | return len(self.getslice()) | |
1374 | ||
9c44d242 PH |
1375 | |
1376 | class OnDemandPagedList(PagedList): | |
1377 | def __init__(self, pagefunc, pagesize): | |
1378 | self._pagefunc = pagefunc | |
1379 | self._pagesize = pagesize | |
1380 | ||
b7ab0590 PH |
1381 | def getslice(self, start=0, end=None): |
1382 | res = [] | |
1383 | for pagenum in itertools.count(start // self._pagesize): | |
1384 | firstid = pagenum * self._pagesize | |
1385 | nextfirstid = pagenum * self._pagesize + self._pagesize | |
1386 | if start >= nextfirstid: | |
1387 | continue | |
1388 | ||
1389 | page_results = list(self._pagefunc(pagenum)) | |
1390 | ||
1391 | startv = ( | |
1392 | start % self._pagesize | |
1393 | if firstid <= start < nextfirstid | |
1394 | else 0) | |
1395 | ||
1396 | endv = ( | |
1397 | ((end - 1) % self._pagesize) + 1 | |
1398 | if (end is not None and firstid <= end <= nextfirstid) | |
1399 | else None) | |
1400 | ||
1401 | if startv != 0 or endv is not None: | |
1402 | page_results = page_results[startv:endv] | |
1403 | res.extend(page_results) | |
1404 | ||
1405 | # A little optimization - if current page is not "full", ie. does | |
1406 | # not contain page_size videos then we can assume that this page | |
1407 | # is the last one - there are no more ids on further pages - | |
1408 | # i.e. no need to query again. | |
1409 | if len(page_results) + startv < self._pagesize: | |
1410 | break | |
1411 | ||
1412 | # If we got the whole page, but the next page is not interesting, | |
1413 | # break out early as well | |
1414 | if end == nextfirstid: | |
1415 | break | |
1416 | return res | |
81c2f20b PH |
1417 | |
1418 | ||
9c44d242 PH |
1419 | class InAdvancePagedList(PagedList): |
1420 | def __init__(self, pagefunc, pagecount, pagesize): | |
1421 | self._pagefunc = pagefunc | |
1422 | self._pagecount = pagecount | |
1423 | self._pagesize = pagesize | |
1424 | ||
1425 | def getslice(self, start=0, end=None): | |
1426 | res = [] | |
1427 | start_page = start // self._pagesize | |
1428 | end_page = ( | |
1429 | self._pagecount if end is None else (end // self._pagesize + 1)) | |
1430 | skip_elems = start - start_page * self._pagesize | |
1431 | only_more = None if end is None else end - start | |
1432 | for pagenum in range(start_page, end_page): | |
1433 | page = list(self._pagefunc(pagenum)) | |
1434 | if skip_elems: | |
1435 | page = page[skip_elems:] | |
1436 | skip_elems = None | |
1437 | if only_more is not None: | |
1438 | if len(page) < only_more: | |
1439 | only_more -= len(page) | |
1440 | else: | |
1441 | page = page[:only_more] | |
1442 | res.extend(page) | |
1443 | break | |
1444 | res.extend(page) | |
1445 | return res | |
1446 | ||
1447 | ||
81c2f20b | 1448 | def uppercase_escape(s): |
676eb3f2 | 1449 | unicode_escape = codecs.getdecoder('unicode_escape') |
81c2f20b | 1450 | return re.sub( |
a612753d | 1451 | r'\\U[0-9a-fA-F]{8}', |
676eb3f2 PH |
1452 | lambda m: unicode_escape(m.group(0))[0], |
1453 | s) | |
b53466e1 | 1454 | |
d05cfe06 S |
1455 | |
1456 | def escape_rfc3986(s): | |
1457 | """Escape non-ASCII characters as suggested by RFC 3986""" | |
8f9312c3 | 1458 | if sys.version_info < (3, 0) and isinstance(s, compat_str): |
d05cfe06 | 1459 | s = s.encode('utf-8') |
ecc0c5ee | 1460 | return compat_urllib_parse.quote(s, b"%/;:@&=+$,!~*'()?#[]") |
d05cfe06 S |
1461 | |
1462 | ||
1463 | def escape_url(url): | |
1464 | """Escape URL as suggested by RFC 3986""" | |
1465 | url_parsed = compat_urllib_parse_urlparse(url) | |
1466 | return url_parsed._replace( | |
1467 | path=escape_rfc3986(url_parsed.path), | |
1468 | params=escape_rfc3986(url_parsed.params), | |
1469 | query=escape_rfc3986(url_parsed.query), | |
1470 | fragment=escape_rfc3986(url_parsed.fragment) | |
1471 | ).geturl() | |
1472 | ||
b53466e1 | 1473 | try: |
28e614de | 1474 | struct.pack('!I', 0) |
b53466e1 PH |
1475 | except TypeError: |
1476 | # In Python 2.6 (and some 2.7 versions), struct requires a bytes argument | |
1477 | def struct_pack(spec, *args): | |
1478 | if isinstance(spec, compat_str): | |
1479 | spec = spec.encode('ascii') | |
1480 | return struct.pack(spec, *args) | |
1481 | ||
1482 | def struct_unpack(spec, *args): | |
1483 | if isinstance(spec, compat_str): | |
1484 | spec = spec.encode('ascii') | |
1485 | return struct.unpack(spec, *args) | |
1486 | else: | |
1487 | struct_pack = struct.pack | |
1488 | struct_unpack = struct.unpack | |
62e609ab PH |
1489 | |
1490 | ||
1491 | def read_batch_urls(batch_fd): | |
1492 | def fixup(url): | |
1493 | if not isinstance(url, compat_str): | |
1494 | url = url.decode('utf-8', 'replace') | |
28e614de | 1495 | BOM_UTF8 = '\xef\xbb\xbf' |
62e609ab PH |
1496 | if url.startswith(BOM_UTF8): |
1497 | url = url[len(BOM_UTF8):] | |
1498 | url = url.strip() | |
1499 | if url.startswith(('#', ';', ']')): | |
1500 | return False | |
1501 | return url | |
1502 | ||
1503 | with contextlib.closing(batch_fd) as fd: | |
1504 | return [url for url in map(fixup, fd) if url] | |
b74fa8cd JMF |
1505 | |
1506 | ||
1507 | def urlencode_postdata(*args, **kargs): | |
1508 | return compat_urllib_parse.urlencode(*args, **kargs).encode('ascii') | |
bcf89ce6 PH |
1509 | |
1510 | ||
0990305d PH |
1511 | try: |
1512 | etree_iter = xml.etree.ElementTree.Element.iter | |
1513 | except AttributeError: # Python <=2.6 | |
1514 | etree_iter = lambda n: n.findall('.//*') | |
1515 | ||
1516 | ||
bcf89ce6 PH |
1517 | def parse_xml(s): |
1518 | class TreeBuilder(xml.etree.ElementTree.TreeBuilder): | |
1519 | def doctype(self, name, pubid, system): | |
1520 | pass # Ignore doctypes | |
1521 | ||
1522 | parser = xml.etree.ElementTree.XMLParser(target=TreeBuilder()) | |
1523 | kwargs = {'parser': parser} if sys.version_info >= (2, 7) else {} | |
0990305d PH |
1524 | tree = xml.etree.ElementTree.XML(s.encode('utf-8'), **kwargs) |
1525 | # Fix up XML parser in Python 2.x | |
1526 | if sys.version_info < (3, 0): | |
1527 | for n in etree_iter(tree): | |
1528 | if n.text is not None: | |
1529 | if not isinstance(n.text, compat_str): | |
1530 | n.text = n.text.decode('utf-8') | |
1531 | return tree | |
e68301af PH |
1532 | |
1533 | ||
a1a530b0 PH |
1534 | US_RATINGS = { |
1535 | 'G': 0, | |
1536 | 'PG': 10, | |
1537 | 'PG-13': 13, | |
1538 | 'R': 16, | |
1539 | 'NC': 18, | |
1540 | } | |
fac55558 PH |
1541 | |
1542 | ||
146c80e2 S |
1543 | def parse_age_limit(s): |
1544 | if s is None: | |
d838b1bd | 1545 | return None |
146c80e2 | 1546 | m = re.match(r'^(?P<age>\d{1,2})\+?$', s) |
d838b1bd | 1547 | return int(m.group('age')) if m else US_RATINGS.get(s, None) |
146c80e2 S |
1548 | |
1549 | ||
fac55558 | 1550 | def strip_jsonp(code): |
609a61e3 PH |
1551 | return re.sub( |
1552 | r'(?s)^[a-zA-Z0-9_]+\s*\(\s*(.*)\);?\s*?(?://[^\n]*)*$', r'\1', code) | |
478c2c61 PH |
1553 | |
1554 | ||
e05f6939 PH |
1555 | def js_to_json(code): |
1556 | def fix_kv(m): | |
e7b6d122 PH |
1557 | v = m.group(0) |
1558 | if v in ('true', 'false', 'null'): | |
1559 | return v | |
1560 | if v.startswith('"'): | |
1561 | return v | |
1562 | if v.startswith("'"): | |
1563 | v = v[1:-1] | |
1564 | v = re.sub(r"\\\\|\\'|\"", lambda m: { | |
1565 | '\\\\': '\\\\', | |
1566 | "\\'": "'", | |
1567 | '"': '\\"', | |
1568 | }[m.group(0)], v) | |
1569 | return '"%s"' % v | |
e05f6939 PH |
1570 | |
1571 | res = re.sub(r'''(?x) | |
d305dd73 PH |
1572 | "(?:[^"\\]*(?:\\\\|\\['"nu]))*[^"\\]*"| |
1573 | '(?:[^'\\]*(?:\\\\|\\['"nu]))*[^'\\]*'| | |
8f4b58d7 | 1574 | [a-zA-Z_][.a-zA-Z_0-9]* |
e05f6939 PH |
1575 | ''', fix_kv, code) |
1576 | res = re.sub(r',(\s*\])', lambda m: m.group(1), res) | |
1577 | return res | |
1578 | ||
1579 | ||
478c2c61 PH |
1580 | def qualities(quality_ids): |
1581 | """ Get a numeric quality value out of a list of possible values """ | |
1582 | def q(qid): | |
1583 | try: | |
1584 | return quality_ids.index(qid) | |
1585 | except ValueError: | |
1586 | return -1 | |
1587 | return q | |
1588 | ||
acd69589 PH |
1589 | |
1590 | DEFAULT_OUTTMPL = '%(title)s-%(id)s.%(ext)s' | |
0a871f68 | 1591 | |
a020a0dc PH |
1592 | |
1593 | def limit_length(s, length): | |
1594 | """ Add ellipses to overly long strings """ | |
1595 | if s is None: | |
1596 | return None | |
1597 | ELLIPSES = '...' | |
1598 | if len(s) > length: | |
1599 | return s[:length - len(ELLIPSES)] + ELLIPSES | |
1600 | return s | |
48844745 PH |
1601 | |
1602 | ||
1603 | def version_tuple(v): | |
5f9b8394 | 1604 | return tuple(int(e) for e in re.split(r'[-.]', v)) |
48844745 PH |
1605 | |
1606 | ||
1607 | def is_outdated_version(version, limit, assume_new=True): | |
1608 | if not version: | |
1609 | return not assume_new | |
1610 | try: | |
1611 | return version_tuple(version) < version_tuple(limit) | |
1612 | except ValueError: | |
1613 | return not assume_new | |
732ea2f0 PH |
1614 | |
1615 | ||
1616 | def ytdl_is_updateable(): | |
1617 | """ Returns if youtube-dl can be updated with -U """ | |
1618 | from zipimport import zipimporter | |
1619 | ||
1620 | return isinstance(globals().get('__loader__'), zipimporter) or hasattr(sys, 'frozen') | |
7d4111ed PH |
1621 | |
1622 | ||
1623 | def args_to_str(args): | |
1624 | # Get a short string representation for a subprocess command | |
1625 | return ' '.join(shlex_quote(a) for a in args) | |
2ccd1b10 PH |
1626 | |
1627 | ||
c460bdd5 PH |
1628 | def mimetype2ext(mt): |
1629 | _, _, res = mt.rpartition('/') | |
1630 | ||
1631 | return { | |
1632 | 'x-ms-wmv': 'wmv', | |
1633 | 'x-mp4-fragmented': 'mp4', | |
1634 | }.get(res, res) | |
1635 | ||
1636 | ||
2ccd1b10 PH |
1637 | def urlhandle_detect_ext(url_handle): |
1638 | try: | |
1639 | url_handle.headers | |
1640 | getheader = lambda h: url_handle.headers[h] | |
1641 | except AttributeError: # Python < 3 | |
1642 | getheader = url_handle.info().getheader | |
1643 | ||
b55ee18f PH |
1644 | cd = getheader('Content-Disposition') |
1645 | if cd: | |
1646 | m = re.match(r'attachment;\s*filename="(?P<filename>[^"]+)"', cd) | |
1647 | if m: | |
1648 | e = determine_ext(m.group('filename'), default_ext=None) | |
1649 | if e: | |
1650 | return e | |
1651 | ||
c460bdd5 | 1652 | return mimetype2ext(getheader('Content-Type')) |
05900629 PH |
1653 | |
1654 | ||
1655 | def age_restricted(content_limit, age_limit): | |
1656 | """ Returns True iff the content should be blocked """ | |
1657 | ||
1658 | if age_limit is None: # No limit set | |
1659 | return False | |
1660 | if content_limit is None: | |
1661 | return False # Content available for everyone | |
1662 | return age_limit < content_limit | |
61ca9a80 PH |
1663 | |
1664 | ||
1665 | def is_html(first_bytes): | |
1666 | """ Detect whether a file contains HTML by examining its first bytes. """ | |
1667 | ||
1668 | BOMS = [ | |
1669 | (b'\xef\xbb\xbf', 'utf-8'), | |
1670 | (b'\x00\x00\xfe\xff', 'utf-32-be'), | |
1671 | (b'\xff\xfe\x00\x00', 'utf-32-le'), | |
1672 | (b'\xff\xfe', 'utf-16-le'), | |
1673 | (b'\xfe\xff', 'utf-16-be'), | |
1674 | ] | |
1675 | for bom, enc in BOMS: | |
1676 | if first_bytes.startswith(bom): | |
1677 | s = first_bytes[len(bom):].decode(enc, 'replace') | |
1678 | break | |
1679 | else: | |
1680 | s = first_bytes.decode('utf-8', 'replace') | |
1681 | ||
1682 | return re.match(r'^\s*<', s) | |
a055469f PH |
1683 | |
1684 | ||
1685 | def determine_protocol(info_dict): | |
1686 | protocol = info_dict.get('protocol') | |
1687 | if protocol is not None: | |
1688 | return protocol | |
1689 | ||
1690 | url = info_dict['url'] | |
1691 | if url.startswith('rtmp'): | |
1692 | return 'rtmp' | |
1693 | elif url.startswith('mms'): | |
1694 | return 'mms' | |
1695 | elif url.startswith('rtsp'): | |
1696 | return 'rtsp' | |
1697 | ||
1698 | ext = determine_ext(url) | |
1699 | if ext == 'm3u8': | |
1700 | return 'm3u8' | |
1701 | elif ext == 'f4m': | |
1702 | return 'f4m' | |
1703 | ||
1704 | return compat_urllib_parse_urlparse(url).scheme | |
cfb56d1a PH |
1705 | |
1706 | ||
1707 | def render_table(header_row, data): | |
1708 | """ Render a list of rows, each as a list of values """ | |
1709 | table = [header_row] + data | |
1710 | max_lens = [max(len(compat_str(v)) for v in col) for col in zip(*table)] | |
1711 | format_str = ' '.join('%-' + compat_str(ml + 1) + 's' for ml in max_lens[:-1]) + '%s' | |
1712 | return '\n'.join(format_str % tuple(row) for row in table) | |
347de493 PH |
1713 | |
1714 | ||
1715 | def _match_one(filter_part, dct): | |
1716 | COMPARISON_OPERATORS = { | |
1717 | '<': operator.lt, | |
1718 | '<=': operator.le, | |
1719 | '>': operator.gt, | |
1720 | '>=': operator.ge, | |
1721 | '=': operator.eq, | |
1722 | '!=': operator.ne, | |
1723 | } | |
1724 | operator_rex = re.compile(r'''(?x)\s* | |
1725 | (?P<key>[a-z_]+) | |
1726 | \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s* | |
1727 | (?: | |
1728 | (?P<intval>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)| | |
1729 | (?P<strval>(?![0-9.])[a-z0-9A-Z]*) | |
1730 | ) | |
1731 | \s*$ | |
1732 | ''' % '|'.join(map(re.escape, COMPARISON_OPERATORS.keys()))) | |
1733 | m = operator_rex.search(filter_part) | |
1734 | if m: | |
1735 | op = COMPARISON_OPERATORS[m.group('op')] | |
1736 | if m.group('strval') is not None: | |
1737 | if m.group('op') not in ('=', '!='): | |
1738 | raise ValueError( | |
1739 | 'Operator %s does not support string values!' % m.group('op')) | |
1740 | comparison_value = m.group('strval') | |
1741 | else: | |
1742 | try: | |
1743 | comparison_value = int(m.group('intval')) | |
1744 | except ValueError: | |
1745 | comparison_value = parse_filesize(m.group('intval')) | |
1746 | if comparison_value is None: | |
1747 | comparison_value = parse_filesize(m.group('intval') + 'B') | |
1748 | if comparison_value is None: | |
1749 | raise ValueError( | |
1750 | 'Invalid integer value %r in filter part %r' % ( | |
1751 | m.group('intval'), filter_part)) | |
1752 | actual_value = dct.get(m.group('key')) | |
1753 | if actual_value is None: | |
1754 | return m.group('none_inclusive') | |
1755 | return op(actual_value, comparison_value) | |
1756 | ||
1757 | UNARY_OPERATORS = { | |
1758 | '': lambda v: v is not None, | |
1759 | '!': lambda v: v is None, | |
1760 | } | |
1761 | operator_rex = re.compile(r'''(?x)\s* | |
1762 | (?P<op>%s)\s*(?P<key>[a-z_]+) | |
1763 | \s*$ | |
1764 | ''' % '|'.join(map(re.escape, UNARY_OPERATORS.keys()))) | |
1765 | m = operator_rex.search(filter_part) | |
1766 | if m: | |
1767 | op = UNARY_OPERATORS[m.group('op')] | |
1768 | actual_value = dct.get(m.group('key')) | |
1769 | return op(actual_value) | |
1770 | ||
1771 | raise ValueError('Invalid filter part %r' % filter_part) | |
1772 | ||
1773 | ||
1774 | def match_str(filter_str, dct): | |
1775 | """ Filter a dictionary with a simple string syntax. Returns True (=passes filter) or false """ | |
1776 | ||
1777 | return all( | |
1778 | _match_one(filter_part, dct) for filter_part in filter_str.split('&')) | |
1779 | ||
1780 | ||
1781 | def match_filter_func(filter_str): | |
1782 | def _match_func(info_dict): | |
1783 | if match_str(filter_str, info_dict): | |
1784 | return None | |
1785 | else: | |
1786 | video_title = info_dict.get('title', info_dict.get('id', 'video')) | |
1787 | return '%s does not pass filter %s, skipping ..' % (video_title, filter_str) | |
1788 | return _match_func | |
91410c9b PH |
1789 | |
1790 | ||
1791 | class PerRequestProxyHandler(compat_urllib_request.ProxyHandler): | |
2461f79d PH |
1792 | def __init__(self, proxies=None): |
1793 | # Set default handlers | |
1794 | for type in ('http', 'https'): | |
1795 | setattr(self, '%s_open' % type, | |
1796 | lambda r, proxy='__noproxy__', type=type, meth=self.proxy_open: | |
1797 | meth(r, proxy, type)) | |
1798 | return compat_urllib_request.ProxyHandler.__init__(self, proxies) | |
1799 | ||
91410c9b | 1800 | def proxy_open(self, req, proxy, type): |
2461f79d | 1801 | req_proxy = req.headers.get('Ytdl-request-proxy') |
91410c9b PH |
1802 | if req_proxy is not None: |
1803 | proxy = req_proxy | |
2461f79d PH |
1804 | del req.headers['Ytdl-request-proxy'] |
1805 | ||
1806 | if proxy == '__noproxy__': | |
1807 | return None # No Proxy | |
91410c9b PH |
1808 | return compat_urllib_request.ProxyHandler.proxy_open( |
1809 | self, req, proxy, type) |