]>
Commit | Line | Data |
---|---|---|
d77c3dfd FV |
1 | #!/usr/bin/env python |
2 | # -*- coding: utf-8 -*- | |
3 | ||
c496ca96 PH |
4 | import datetime |
5 | import email.utils | |
f45c185f | 6 | import errno |
d77c3dfd | 7 | import gzip |
03f9daab | 8 | import io |
f4bfd65f | 9 | import json |
d77c3dfd | 10 | import locale |
02dbf93f | 11 | import math |
d77c3dfd | 12 | import os |
4eb7f1d1 | 13 | import pipes |
c496ca96 | 14 | import platform |
d77c3dfd | 15 | import re |
13ebea79 | 16 | import ssl |
c496ca96 | 17 | import socket |
d77c3dfd | 18 | import sys |
01951dda | 19 | import traceback |
d77c3dfd | 20 | import zlib |
d77c3dfd | 21 | |
01ba00ca | 22 | try: |
59ae15a5 | 23 | import urllib.request as compat_urllib_request |
01ba00ca | 24 | except ImportError: # Python 2 |
59ae15a5 | 25 | import urllib2 as compat_urllib_request |
01ba00ca PH |
26 | |
27 | try: | |
59ae15a5 | 28 | import urllib.error as compat_urllib_error |
01ba00ca | 29 | except ImportError: # Python 2 |
59ae15a5 | 30 | import urllib2 as compat_urllib_error |
01ba00ca PH |
31 | |
32 | try: | |
59ae15a5 | 33 | import urllib.parse as compat_urllib_parse |
01ba00ca | 34 | except ImportError: # Python 2 |
59ae15a5 | 35 | import urllib as compat_urllib_parse |
01ba00ca | 36 | |
799c0763 PH |
37 | try: |
38 | from urllib.parse import urlparse as compat_urllib_parse_urlparse | |
39 | except ImportError: # Python 2 | |
40 | from urlparse import urlparse as compat_urllib_parse_urlparse | |
41 | ||
6543f0dc JMF |
42 | try: |
43 | import urllib.parse as compat_urlparse | |
44 | except ImportError: # Python 2 | |
45 | import urlparse as compat_urlparse | |
46 | ||
01ba00ca | 47 | try: |
59ae15a5 | 48 | import http.cookiejar as compat_cookiejar |
01ba00ca | 49 | except ImportError: # Python 2 |
59ae15a5 | 50 | import cookielib as compat_cookiejar |
01ba00ca | 51 | |
3e669f36 | 52 | try: |
59ae15a5 | 53 | import html.entities as compat_html_entities |
9f37a959 | 54 | except ImportError: # Python 2 |
59ae15a5 | 55 | import htmlentitydefs as compat_html_entities |
3e669f36 | 56 | |
a8156c1d | 57 | try: |
59ae15a5 | 58 | import html.parser as compat_html_parser |
9f37a959 | 59 | except ImportError: # Python 2 |
59ae15a5 | 60 | import HTMLParser as compat_html_parser |
a8156c1d | 61 | |
348d0a7a | 62 | try: |
59ae15a5 | 63 | import http.client as compat_http_client |
9f37a959 | 64 | except ImportError: # Python 2 |
59ae15a5 | 65 | import httplib as compat_http_client |
348d0a7a | 66 | |
2eabb802 | 67 | try: |
0e283428 | 68 | from urllib.error import HTTPError as compat_HTTPError |
2eabb802 PH |
69 | except ImportError: # Python 2 |
70 | from urllib2 import HTTPError as compat_HTTPError | |
71 | ||
e0df6211 PH |
72 | try: |
73 | from urllib.request import urlretrieve as compat_urlretrieve | |
74 | except ImportError: # Python 2 | |
75 | from urllib import urlretrieve as compat_urlretrieve | |
76 | ||
77 | ||
5910e210 PH |
78 | try: |
79 | from subprocess import DEVNULL | |
80 | compat_subprocess_get_DEVNULL = lambda: DEVNULL | |
81 | except ImportError: | |
82 | compat_subprocess_get_DEVNULL = lambda: open(os.path.devnull, 'w') | |
83 | ||
9f37a959 | 84 | try: |
59ae15a5 | 85 | from urllib.parse import parse_qs as compat_parse_qs |
9f37a959 | 86 | except ImportError: # Python 2 |
59ae15a5 PH |
87 | # HACK: The following is the correct parse_qs implementation from cpython 3's stdlib. |
88 | # Python 2's version is apparently totally broken | |
89 | def _unquote(string, encoding='utf-8', errors='replace'): | |
90 | if string == '': | |
91 | return string | |
92 | res = string.split('%') | |
93 | if len(res) == 1: | |
94 | return string | |
95 | if encoding is None: | |
96 | encoding = 'utf-8' | |
97 | if errors is None: | |
98 | errors = 'replace' | |
99 | # pct_sequence: contiguous sequence of percent-encoded bytes, decoded | |
100 | pct_sequence = b'' | |
101 | string = res[0] | |
102 | for item in res[1:]: | |
103 | try: | |
104 | if not item: | |
105 | raise ValueError | |
106 | pct_sequence += item[:2].decode('hex') | |
107 | rest = item[2:] | |
108 | if not rest: | |
109 | # This segment was just a single percent-encoded character. | |
110 | # May be part of a sequence of code units, so delay decoding. | |
111 | # (Stored in pct_sequence). | |
112 | continue | |
113 | except ValueError: | |
114 | rest = '%' + item | |
115 | # Encountered non-percent-encoded characters. Flush the current | |
116 | # pct_sequence. | |
117 | string += pct_sequence.decode(encoding, errors) + rest | |
118 | pct_sequence = b'' | |
119 | if pct_sequence: | |
120 | # Flush the final pct_sequence | |
121 | string += pct_sequence.decode(encoding, errors) | |
122 | return string | |
123 | ||
124 | def _parse_qsl(qs, keep_blank_values=False, strict_parsing=False, | |
125 | encoding='utf-8', errors='replace'): | |
126 | qs, _coerce_result = qs, unicode | |
127 | pairs = [s2 for s1 in qs.split('&') for s2 in s1.split(';')] | |
128 | r = [] | |
129 | for name_value in pairs: | |
130 | if not name_value and not strict_parsing: | |
131 | continue | |
132 | nv = name_value.split('=', 1) | |
133 | if len(nv) != 2: | |
134 | if strict_parsing: | |
135 | raise ValueError("bad query field: %r" % (name_value,)) | |
136 | # Handle case of a control-name with no equal sign | |
137 | if keep_blank_values: | |
138 | nv.append('') | |
139 | else: | |
140 | continue | |
141 | if len(nv[1]) or keep_blank_values: | |
142 | name = nv[0].replace('+', ' ') | |
143 | name = _unquote(name, encoding=encoding, errors=errors) | |
144 | name = _coerce_result(name) | |
145 | value = nv[1].replace('+', ' ') | |
146 | value = _unquote(value, encoding=encoding, errors=errors) | |
147 | value = _coerce_result(value) | |
148 | r.append((name, value)) | |
149 | return r | |
150 | ||
151 | def compat_parse_qs(qs, keep_blank_values=False, strict_parsing=False, | |
152 | encoding='utf-8', errors='replace'): | |
153 | parsed_result = {} | |
154 | pairs = _parse_qsl(qs, keep_blank_values, strict_parsing, | |
155 | encoding=encoding, errors=errors) | |
156 | for name, value in pairs: | |
157 | if name in parsed_result: | |
158 | parsed_result[name].append(value) | |
159 | else: | |
160 | parsed_result[name] = [value] | |
161 | return parsed_result | |
348d0a7a | 162 | |
3e669f36 | 163 | try: |
59ae15a5 | 164 | compat_str = unicode # Python 2 |
3e669f36 | 165 | except NameError: |
59ae15a5 | 166 | compat_str = str |
3e669f36 PH |
167 | |
168 | try: | |
59ae15a5 | 169 | compat_chr = unichr # Python 2 |
3e669f36 | 170 | except NameError: |
59ae15a5 | 171 | compat_chr = chr |
3e669f36 | 172 | |
b31756c1 FV |
173 | def compat_ord(c): |
174 | if type(c) is int: return c | |
175 | else: return ord(c) | |
176 | ||
468e2e92 FV |
177 | # This is not clearly defined otherwise |
178 | compiled_regex_type = type(re.compile('')) | |
179 | ||
3e669f36 | 180 | std_headers = { |
ae8f7871 | 181 | 'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20100101 Firefox/10.0 (Chrome)', |
59ae15a5 PH |
182 | 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7', |
183 | 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8', | |
184 | 'Accept-Encoding': 'gzip, deflate', | |
185 | 'Accept-Language': 'en-us,en;q=0.5', | |
3e669f36 | 186 | } |
f427df17 | 187 | |
d77c3dfd | 188 | def preferredencoding(): |
59ae15a5 | 189 | """Get preferred encoding. |
d77c3dfd | 190 | |
59ae15a5 PH |
191 | Returns the best encoding scheme for the system, based on |
192 | locale.getpreferredencoding() and some further tweaks. | |
193 | """ | |
194 | try: | |
195 | pref = locale.getpreferredencoding() | |
196 | u'TEST'.encode(pref) | |
197 | except: | |
198 | pref = 'UTF-8' | |
bae611f2 | 199 | |
59ae15a5 | 200 | return pref |
d77c3dfd | 201 | |
8cd10ac4 | 202 | if sys.version_info < (3,0): |
59ae15a5 PH |
203 | def compat_print(s): |
204 | print(s.encode(preferredencoding(), 'xmlcharrefreplace')) | |
8cd10ac4 | 205 | else: |
59ae15a5 PH |
206 | def compat_print(s): |
207 | assert type(s) == type(u'') | |
208 | print(s) | |
d77c3dfd | 209 | |
f4bfd65f PH |
210 | # In Python 2.x, json.dump expects a bytestream. |
211 | # In Python 3.x, it writes to a character stream | |
212 | if sys.version_info < (3,0): | |
213 | def write_json_file(obj, fn): | |
214 | with open(fn, 'wb') as f: | |
215 | json.dump(obj, f) | |
216 | else: | |
217 | def write_json_file(obj, fn): | |
218 | with open(fn, 'w', encoding='utf-8') as f: | |
219 | json.dump(obj, f) | |
220 | ||
59ae56fa PH |
221 | if sys.version_info >= (2,7): |
222 | def find_xpath_attr(node, xpath, key, val): | |
223 | """ Find the xpath xpath[@key=val] """ | |
5de3ece2 | 224 | assert re.match(r'^[a-zA-Z]+$', key) |
54543467 | 225 | assert re.match(r'^[a-zA-Z0-9@\s]*$', val) |
59ae56fa PH |
226 | expr = xpath + u"[@%s='%s']" % (key, val) |
227 | return node.find(expr) | |
228 | else: | |
229 | def find_xpath_attr(node, xpath, key, val): | |
230 | for f in node.findall(xpath): | |
231 | if f.attrib.get(key) == val: | |
232 | return f | |
233 | return None | |
234 | ||
d7e66d39 JMF |
235 | # On python2.6 the xml.etree.ElementTree.Element methods don't support |
236 | # the namespace parameter | |
237 | def xpath_with_ns(path, ns_map): | |
238 | components = [c.split(':') for c in path.split('/')] | |
239 | replaced = [] | |
240 | for c in components: | |
241 | if len(c) == 1: | |
242 | replaced.append(c[0]) | |
243 | else: | |
244 | ns, tag = c | |
245 | replaced.append('{%s}%s' % (ns_map[ns], tag)) | |
246 | return '/'.join(replaced) | |
247 | ||
d77c3dfd | 248 | def htmlentity_transform(matchobj): |
59ae15a5 PH |
249 | """Transforms an HTML entity to a character. |
250 | ||
251 | This function receives a match object and is intended to be used with | |
252 | the re.sub() function. | |
253 | """ | |
254 | entity = matchobj.group(1) | |
255 | ||
256 | # Known non-numeric HTML entity | |
257 | if entity in compat_html_entities.name2codepoint: | |
258 | return compat_chr(compat_html_entities.name2codepoint[entity]) | |
259 | ||
260 | mobj = re.match(u'(?u)#(x?\\d+)', entity) | |
261 | if mobj is not None: | |
262 | numstr = mobj.group(1) | |
263 | if numstr.startswith(u'x'): | |
264 | base = 16 | |
265 | numstr = u'0%s' % numstr | |
266 | else: | |
267 | base = 10 | |
268 | return compat_chr(int(numstr, base)) | |
269 | ||
270 | # Unknown entity in name, return its literal representation | |
271 | return (u'&%s;' % entity) | |
d77c3dfd | 272 | |
a8156c1d | 273 | 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 |
a921f407 JMF |
274 | class BaseHTMLParser(compat_html_parser.HTMLParser): |
275 | def __init(self): | |
276 | compat_html_parser.HTMLParser.__init__(self) | |
277 | self.html = None | |
278 | ||
279 | def loads(self, html): | |
280 | self.html = html | |
281 | self.feed(html) | |
282 | self.close() | |
283 | ||
284 | class AttrParser(BaseHTMLParser): | |
43e8fafd ND |
285 | """Modified HTMLParser that isolates a tag with the specified attribute""" |
286 | def __init__(self, attribute, value): | |
287 | self.attribute = attribute | |
288 | self.value = value | |
59ae15a5 PH |
289 | self.result = None |
290 | self.started = False | |
291 | self.depth = {} | |
59ae15a5 PH |
292 | self.watch_startpos = False |
293 | self.error_count = 0 | |
a921f407 | 294 | BaseHTMLParser.__init__(self) |
59ae15a5 PH |
295 | |
296 | def error(self, message): | |
297 | if self.error_count > 10 or self.started: | |
298 | raise compat_html_parser.HTMLParseError(message, self.getpos()) | |
299 | self.rawdata = '\n'.join(self.html.split('\n')[self.getpos()[0]:]) # skip one line | |
300 | self.error_count += 1 | |
301 | self.goahead(1) | |
302 | ||
59ae15a5 PH |
303 | def handle_starttag(self, tag, attrs): |
304 | attrs = dict(attrs) | |
305 | if self.started: | |
306 | self.find_startpos(None) | |
43e8fafd | 307 | if self.attribute in attrs and attrs[self.attribute] == self.value: |
59ae15a5 PH |
308 | self.result = [tag] |
309 | self.started = True | |
310 | self.watch_startpos = True | |
311 | if self.started: | |
312 | if not tag in self.depth: self.depth[tag] = 0 | |
313 | self.depth[tag] += 1 | |
314 | ||
315 | def handle_endtag(self, tag): | |
316 | if self.started: | |
317 | if tag in self.depth: self.depth[tag] -= 1 | |
318 | if self.depth[self.result[0]] == 0: | |
319 | self.started = False | |
320 | self.result.append(self.getpos()) | |
321 | ||
322 | def find_startpos(self, x): | |
323 | """Needed to put the start position of the result (self.result[1]) | |
324 | after the opening tag with the requested id""" | |
325 | if self.watch_startpos: | |
326 | self.watch_startpos = False | |
327 | self.result.append(self.getpos()) | |
328 | handle_entityref = handle_charref = handle_data = handle_comment = \ | |
329 | handle_decl = handle_pi = unknown_decl = find_startpos | |
330 | ||
331 | def get_result(self): | |
332 | if self.result is None: | |
333 | return None | |
334 | if len(self.result) != 3: | |
335 | return None | |
336 | lines = self.html.split('\n') | |
337 | lines = lines[self.result[1][0]-1:self.result[2][0]] | |
338 | lines[0] = lines[0][self.result[1][1]:] | |
339 | if len(lines) == 1: | |
340 | lines[-1] = lines[-1][:self.result[2][1]-self.result[1][1]] | |
341 | lines[-1] = lines[-1][:self.result[2][1]] | |
342 | return '\n'.join(lines).strip() | |
3b024e17 PH |
343 | # Hack for https://github.com/rg3/youtube-dl/issues/662 |
344 | if sys.version_info < (2, 7, 3): | |
345 | AttrParser.parse_endtag = (lambda self, i: | |
346 | i + len("</scr'+'ipt>") | |
347 | if self.rawdata[i:].startswith("</scr'+'ipt>") | |
348 | else compat_html_parser.HTMLParser.parse_endtag(self, i)) | |
9e6dd238 FV |
349 | |
350 | def get_element_by_id(id, html): | |
43e8fafd ND |
351 | """Return the content of the tag with the specified ID in the passed HTML document""" |
352 | return get_element_by_attribute("id", id, html) | |
353 | ||
354 | def get_element_by_attribute(attribute, value, html): | |
355 | """Return the content of the tag with the specified attribute in the passed HTML document""" | |
356 | parser = AttrParser(attribute, value) | |
59ae15a5 PH |
357 | try: |
358 | parser.loads(html) | |
359 | except compat_html_parser.HTMLParseError: | |
360 | pass | |
361 | return parser.get_result() | |
9e6dd238 | 362 | |
a921f407 JMF |
363 | class MetaParser(BaseHTMLParser): |
364 | """ | |
365 | Modified HTMLParser that isolates a meta tag with the specified name | |
366 | attribute. | |
367 | """ | |
368 | def __init__(self, name): | |
369 | BaseHTMLParser.__init__(self) | |
370 | self.name = name | |
371 | self.content = None | |
372 | self.result = None | |
373 | ||
374 | def handle_starttag(self, tag, attrs): | |
375 | if tag != 'meta': | |
376 | return | |
377 | attrs = dict(attrs) | |
378 | if attrs.get('name') == self.name: | |
379 | self.result = attrs.get('content') | |
380 | ||
381 | def get_result(self): | |
382 | return self.result | |
383 | ||
384 | def get_meta_content(name, html): | |
385 | """ | |
386 | Return the content attribute from the meta tag with the given name attribute. | |
387 | """ | |
388 | parser = MetaParser(name) | |
389 | try: | |
390 | parser.loads(html) | |
391 | except compat_html_parser.HTMLParseError: | |
392 | pass | |
393 | return parser.get_result() | |
394 | ||
9e6dd238 FV |
395 | |
396 | def clean_html(html): | |
59ae15a5 PH |
397 | """Clean an HTML snippet into a readable string""" |
398 | # Newline vs <br /> | |
399 | html = html.replace('\n', ' ') | |
6b3aef80 FV |
400 | html = re.sub(r'\s*<\s*br\s*/?\s*>\s*', '\n', html) |
401 | html = re.sub(r'<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html) | |
59ae15a5 PH |
402 | # Strip html tags |
403 | html = re.sub('<.*?>', '', html) | |
404 | # Replace html entities | |
405 | html = unescapeHTML(html) | |
7decf895 | 406 | return html.strip() |
9e6dd238 FV |
407 | |
408 | ||
d77c3dfd | 409 | def sanitize_open(filename, open_mode): |
59ae15a5 PH |
410 | """Try to open the given filename, and slightly tweak it if this fails. |
411 | ||
412 | Attempts to open the given filename. If this fails, it tries to change | |
413 | the filename slightly, step by step, until it's either able to open it | |
414 | or it fails and raises a final exception, like the standard open() | |
415 | function. | |
416 | ||
417 | It returns the tuple (stream, definitive_file_name). | |
418 | """ | |
419 | try: | |
420 | if filename == u'-': | |
421 | if sys.platform == 'win32': | |
422 | import msvcrt | |
423 | msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY) | |
898280a0 | 424 | return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename) |
59ae15a5 PH |
425 | stream = open(encodeFilename(filename), open_mode) |
426 | return (stream, filename) | |
427 | except (IOError, OSError) as err: | |
f45c185f PH |
428 | if err.errno in (errno.EACCES,): |
429 | raise | |
59ae15a5 | 430 | |
f45c185f PH |
431 | # In case of error, try to remove win32 forbidden chars |
432 | alt_filename = os.path.join( | |
433 | re.sub(u'[/<>:"\\|\\\\?\\*]', u'#', path_part) | |
434 | for path_part in os.path.split(filename) | |
435 | ) | |
436 | if alt_filename == filename: | |
437 | raise | |
438 | else: | |
439 | # An exception here should be caught in the caller | |
440 | stream = open(encodeFilename(filename), open_mode) | |
441 | return (stream, alt_filename) | |
d77c3dfd FV |
442 | |
443 | ||
444 | def timeconvert(timestr): | |
59ae15a5 PH |
445 | """Convert RFC 2822 defined time string into system timestamp""" |
446 | timestamp = None | |
447 | timetuple = email.utils.parsedate_tz(timestr) | |
448 | if timetuple is not None: | |
449 | timestamp = email.utils.mktime_tz(timetuple) | |
450 | return timestamp | |
1c469a94 | 451 | |
796173d0 | 452 | def sanitize_filename(s, restricted=False, is_id=False): |
59ae15a5 PH |
453 | """Sanitizes a string so it could be used as part of a filename. |
454 | If restricted is set, use a stricter subset of allowed characters. | |
796173d0 | 455 | Set is_id if this is not an arbitrary string, but an ID that should be kept if possible |
59ae15a5 PH |
456 | """ |
457 | def replace_insane(char): | |
458 | if char == '?' or ord(char) < 32 or ord(char) == 127: | |
459 | return '' | |
460 | elif char == '"': | |
461 | return '' if restricted else '\'' | |
462 | elif char == ':': | |
463 | return '_-' if restricted else ' -' | |
464 | elif char in '\\/|*<>': | |
465 | return '_' | |
627dcfff | 466 | if restricted and (char in '!&\'()[]{}$;`^,#' or char.isspace()): |
59ae15a5 PH |
467 | return '_' |
468 | if restricted and ord(char) > 127: | |
469 | return '_' | |
470 | return char | |
471 | ||
472 | result = u''.join(map(replace_insane, s)) | |
796173d0 PH |
473 | if not is_id: |
474 | while '__' in result: | |
475 | result = result.replace('__', '_') | |
476 | result = result.strip('_') | |
477 | # Common case of "Foreign band name - English song title" | |
478 | if restricted and result.startswith('-_'): | |
479 | result = result[2:] | |
480 | if not result: | |
481 | result = '_' | |
59ae15a5 | 482 | return result |
d77c3dfd FV |
483 | |
484 | def orderedSet(iterable): | |
59ae15a5 PH |
485 | """ Remove all duplicates from the input iterable """ |
486 | res = [] | |
487 | for el in iterable: | |
488 | if el not in res: | |
489 | res.append(el) | |
490 | return res | |
d77c3dfd FV |
491 | |
492 | def unescapeHTML(s): | |
59ae15a5 PH |
493 | """ |
494 | @param s a string | |
495 | """ | |
496 | assert type(s) == type(u'') | |
d77c3dfd | 497 | |
59ae15a5 PH |
498 | result = re.sub(u'(?u)&(.+?);', htmlentity_transform, s) |
499 | return result | |
d77c3dfd FV |
500 | |
501 | def encodeFilename(s): | |
59ae15a5 PH |
502 | """ |
503 | @param s The name of the file | |
504 | """ | |
d77c3dfd | 505 | |
59ae15a5 | 506 | assert type(s) == type(u'') |
d77c3dfd | 507 | |
59ae15a5 PH |
508 | # Python 3 has a Unicode API |
509 | if sys.version_info >= (3, 0): | |
510 | return s | |
0f00efed | 511 | |
59ae15a5 PH |
512 | if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5: |
513 | # Pass u'' directly to use Unicode APIs on Windows 2000 and up | |
514 | # (Detecting Windows NT 4 is tricky because 'major >= 4' would | |
515 | # match Windows 9x series as well. Besides, NT 4 is obsolete.) | |
516 | return s | |
517 | else: | |
6df40dcb PH |
518 | encoding = sys.getfilesystemencoding() |
519 | if encoding is None: | |
520 | encoding = 'utf-8' | |
521 | return s.encode(encoding, 'ignore') | |
d77c3dfd | 522 | |
8271226a PH |
523 | def decodeOption(optval): |
524 | if optval is None: | |
525 | return optval | |
526 | if isinstance(optval, bytes): | |
527 | optval = optval.decode(preferredencoding()) | |
528 | ||
529 | assert isinstance(optval, compat_str) | |
530 | return optval | |
1c256f70 | 531 | |
4539dd30 PH |
532 | def formatSeconds(secs): |
533 | if secs > 3600: | |
534 | return '%d:%02d:%02d' % (secs // 3600, (secs % 3600) // 60, secs % 60) | |
535 | elif secs > 60: | |
536 | return '%d:%02d' % (secs // 60, secs % 60) | |
537 | else: | |
538 | return '%d' % secs | |
539 | ||
dca08720 | 540 | def make_HTTPS_handler(opts_no_check_certificate): |
13ebea79 PH |
541 | if sys.version_info < (3, 2): |
542 | import httplib | |
543 | ||
544 | class HTTPSConnectionV3(httplib.HTTPSConnection): | |
545 | def __init__(self, *args, **kwargs): | |
546 | httplib.HTTPSConnection.__init__(self, *args, **kwargs) | |
547 | ||
548 | def connect(self): | |
549 | sock = socket.create_connection((self.host, self.port), self.timeout) | |
550 | if self._tunnel_host: | |
551 | self.sock = sock | |
552 | self._tunnel() | |
553 | try: | |
554 | self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_SSLv3) | |
de79c46c | 555 | except ssl.SSLError: |
13ebea79 PH |
556 | self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_SSLv23) |
557 | ||
558 | class HTTPSHandlerV3(compat_urllib_request.HTTPSHandler): | |
559 | def https_open(self, req): | |
560 | return self.do_open(HTTPSConnectionV3, req) | |
561 | return HTTPSHandlerV3() | |
ea6d901e | 562 | else: |
13ebea79 | 563 | context = ssl.SSLContext(ssl.PROTOCOL_SSLv3) |
ea6d901e PH |
564 | context.set_default_verify_paths() |
565 | ||
566 | context.verify_mode = (ssl.CERT_NONE | |
dca08720 | 567 | if opts_no_check_certificate |
ea6d901e | 568 | else ssl.CERT_REQUIRED) |
acebc9cd | 569 | return compat_urllib_request.HTTPSHandler(context=context) |
ea6d901e | 570 | |
1c256f70 PH |
571 | class ExtractorError(Exception): |
572 | """Error during info extraction.""" | |
2eabb802 | 573 | def __init__(self, msg, tb=None, expected=False, cause=None): |
9a82b238 PH |
574 | """ tb, if given, is the original traceback (so that it can be printed out). |
575 | If expected is set, this is a normal error message and most likely not a bug in youtube-dl. | |
576 | """ | |
577 | ||
578 | if sys.exc_info()[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError): | |
579 | expected = True | |
580 | if not expected: | |
298f833b | 581 | 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.' |
1c256f70 | 582 | super(ExtractorError, self).__init__(msg) |
d5979c5d | 583 | |
1c256f70 | 584 | self.traceback = tb |
8cc83b8d | 585 | self.exc_info = sys.exc_info() # preserve original exception |
2eabb802 | 586 | self.cause = cause |
1c256f70 | 587 | |
01951dda PH |
588 | def format_traceback(self): |
589 | if self.traceback is None: | |
590 | return None | |
591 | return u''.join(traceback.format_tb(self.traceback)) | |
592 | ||
1c256f70 | 593 | |
55b3e45b JMF |
594 | class RegexNotFoundError(ExtractorError): |
595 | """Error when a regex didn't match""" | |
596 | pass | |
597 | ||
598 | ||
d77c3dfd | 599 | class DownloadError(Exception): |
59ae15a5 | 600 | """Download Error exception. |
d77c3dfd | 601 | |
59ae15a5 PH |
602 | This exception may be thrown by FileDownloader objects if they are not |
603 | configured to continue on errors. They will contain the appropriate | |
604 | error message. | |
605 | """ | |
8cc83b8d FV |
606 | def __init__(self, msg, exc_info=None): |
607 | """ exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """ | |
608 | super(DownloadError, self).__init__(msg) | |
609 | self.exc_info = exc_info | |
d77c3dfd FV |
610 | |
611 | ||
612 | class SameFileError(Exception): | |
59ae15a5 | 613 | """Same File exception. |
d77c3dfd | 614 | |
59ae15a5 PH |
615 | This exception will be thrown by FileDownloader objects if they detect |
616 | multiple files would have to be downloaded to the same file on disk. | |
617 | """ | |
618 | pass | |
d77c3dfd FV |
619 | |
620 | ||
621 | class PostProcessingError(Exception): | |
59ae15a5 | 622 | """Post Processing exception. |
d77c3dfd | 623 | |
59ae15a5 PH |
624 | This exception may be raised by PostProcessor's .run() method to |
625 | indicate an error in the postprocessing task. | |
626 | """ | |
7851b379 PH |
627 | def __init__(self, msg): |
628 | self.msg = msg | |
d77c3dfd FV |
629 | |
630 | class MaxDownloadsReached(Exception): | |
59ae15a5 PH |
631 | """ --max-downloads limit has been reached. """ |
632 | pass | |
d77c3dfd FV |
633 | |
634 | ||
635 | class UnavailableVideoError(Exception): | |
59ae15a5 | 636 | """Unavailable Format exception. |
d77c3dfd | 637 | |
59ae15a5 PH |
638 | This exception will be thrown when a video is requested |
639 | in a format that is not available for that video. | |
640 | """ | |
641 | pass | |
d77c3dfd FV |
642 | |
643 | ||
644 | class ContentTooShortError(Exception): | |
59ae15a5 | 645 | """Content Too Short exception. |
d77c3dfd | 646 | |
59ae15a5 PH |
647 | This exception may be raised by FileDownloader objects when a file they |
648 | download is too small for what the server announced first, indicating | |
649 | the connection was probably interrupted. | |
650 | """ | |
651 | # Both in bytes | |
652 | downloaded = None | |
653 | expected = None | |
d77c3dfd | 654 | |
59ae15a5 PH |
655 | def __init__(self, downloaded, expected): |
656 | self.downloaded = downloaded | |
657 | self.expected = expected | |
d77c3dfd | 658 | |
acebc9cd | 659 | class YoutubeDLHandler(compat_urllib_request.HTTPHandler): |
59ae15a5 PH |
660 | """Handler for HTTP requests and responses. |
661 | ||
662 | This class, when installed with an OpenerDirector, automatically adds | |
663 | the standard headers to every HTTP request and handles gzipped and | |
664 | deflated responses from web servers. If compression is to be avoided in | |
665 | a particular request, the original request in the program code only has | |
666 | to include the HTTP header "Youtubedl-No-Compression", which will be | |
667 | removed before making the real request. | |
668 | ||
669 | Part of this code was copied from: | |
670 | ||
671 | http://techknack.net/python-urllib2-handlers/ | |
672 | ||
673 | Andrew Rowls, the author of that code, agreed to release it to the | |
674 | public domain. | |
675 | """ | |
676 | ||
677 | @staticmethod | |
678 | def deflate(data): | |
679 | try: | |
680 | return zlib.decompress(data, -zlib.MAX_WBITS) | |
681 | except zlib.error: | |
682 | return zlib.decompress(data) | |
683 | ||
684 | @staticmethod | |
685 | def addinfourl_wrapper(stream, headers, url, code): | |
686 | if hasattr(compat_urllib_request.addinfourl, 'getcode'): | |
687 | return compat_urllib_request.addinfourl(stream, headers, url, code) | |
688 | ret = compat_urllib_request.addinfourl(stream, headers, url) | |
689 | ret.code = code | |
690 | return ret | |
691 | ||
acebc9cd PH |
692 | def http_request(self, req): |
693 | for h,v in std_headers.items(): | |
59ae15a5 PH |
694 | if h in req.headers: |
695 | del req.headers[h] | |
335959e7 | 696 | req.add_header(h, v) |
59ae15a5 PH |
697 | if 'Youtubedl-no-compression' in req.headers: |
698 | if 'Accept-encoding' in req.headers: | |
699 | del req.headers['Accept-encoding'] | |
700 | del req.headers['Youtubedl-no-compression'] | |
3446dfb7 | 701 | if 'Youtubedl-user-agent' in req.headers: |
335959e7 PH |
702 | if 'User-agent' in req.headers: |
703 | del req.headers['User-agent'] | |
704 | req.headers['User-agent'] = req.headers['Youtubedl-user-agent'] | |
3446dfb7 | 705 | del req.headers['Youtubedl-user-agent'] |
59ae15a5 PH |
706 | return req |
707 | ||
acebc9cd | 708 | def http_response(self, req, resp): |
59ae15a5 PH |
709 | old_resp = resp |
710 | # gzip | |
711 | if resp.headers.get('Content-encoding', '') == 'gzip': | |
aa3e9507 PH |
712 | content = resp.read() |
713 | gz = gzip.GzipFile(fileobj=io.BytesIO(content), mode='rb') | |
714 | try: | |
715 | uncompressed = io.BytesIO(gz.read()) | |
716 | except IOError as original_ioerror: | |
717 | # There may be junk add the end of the file | |
718 | # See http://stackoverflow.com/q/4928560/35070 for details | |
719 | for i in range(1, 1024): | |
720 | try: | |
721 | gz = gzip.GzipFile(fileobj=io.BytesIO(content[:-i]), mode='rb') | |
722 | uncompressed = io.BytesIO(gz.read()) | |
723 | except IOError: | |
724 | continue | |
725 | break | |
726 | else: | |
727 | raise original_ioerror | |
728 | resp = self.addinfourl_wrapper(uncompressed, old_resp.headers, old_resp.url, old_resp.code) | |
59ae15a5 PH |
729 | resp.msg = old_resp.msg |
730 | # deflate | |
731 | if resp.headers.get('Content-encoding', '') == 'deflate': | |
732 | gz = io.BytesIO(self.deflate(resp.read())) | |
733 | resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code) | |
734 | resp.msg = old_resp.msg | |
735 | return resp | |
0f8d03f8 | 736 | |
acebc9cd PH |
737 | https_request = http_request |
738 | https_response = http_response | |
bf50b038 JMF |
739 | |
740 | def unified_strdate(date_str): | |
741 | """Return a string with the date in the format YYYYMMDD""" | |
742 | upload_date = None | |
743 | #Replace commas | |
744 | date_str = date_str.replace(',',' ') | |
745 | # %z (UTC offset) is only supported in python>=3.2 | |
746 | date_str = re.sub(r' (\+|-)[\d]*$', '', date_str) | |
19e1d359 JMF |
747 | format_expressions = [ |
748 | '%d %B %Y', | |
749 | '%B %d %Y', | |
750 | '%b %d %Y', | |
751 | '%Y-%m-%d', | |
752 | '%d/%m/%Y', | |
753 | '%Y/%m/%d %H:%M:%S', | |
754 | '%d.%m.%Y %H:%M', | |
755 | '%Y-%m-%dT%H:%M:%SZ', | |
59040888 PH |
756 | '%Y-%m-%dT%H:%M:%S.%fZ', |
757 | '%Y-%m-%dT%H:%M:%S.%f0Z', | |
2e1fa03b | 758 | '%Y-%m-%dT%H:%M:%S', |
19e1d359 | 759 | ] |
bf50b038 JMF |
760 | for expression in format_expressions: |
761 | try: | |
762 | upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d') | |
763 | except: | |
764 | pass | |
765 | return upload_date | |
766 | ||
cbdbb766 | 767 | def determine_ext(url, default_ext=u'unknown_video'): |
73e79f2a PH |
768 | guess = url.partition(u'?')[0].rpartition(u'.')[2] |
769 | if re.match(r'^[A-Za-z0-9]+$', guess): | |
770 | return guess | |
771 | else: | |
cbdbb766 | 772 | return default_ext |
73e79f2a | 773 | |
d4051a8e JMF |
774 | def subtitles_filename(filename, sub_lang, sub_format): |
775 | return filename.rsplit('.', 1)[0] + u'.' + sub_lang + u'.' + sub_format | |
776 | ||
bd558525 | 777 | def date_from_str(date_str): |
37254abc JMF |
778 | """ |
779 | Return a datetime object from a string in the format YYYYMMDD or | |
780 | (now|today)[+-][0-9](day|week|month|year)(s)?""" | |
781 | today = datetime.date.today() | |
782 | if date_str == 'now'or date_str == 'today': | |
783 | return today | |
784 | match = re.match('(now|today)(?P<sign>[+-])(?P<time>\d+)(?P<unit>day|week|month|year)(s)?', date_str) | |
785 | if match is not None: | |
786 | sign = match.group('sign') | |
787 | time = int(match.group('time')) | |
788 | if sign == '-': | |
789 | time = -time | |
790 | unit = match.group('unit') | |
791 | #A bad aproximation? | |
792 | if unit == 'month': | |
793 | unit = 'day' | |
794 | time *= 30 | |
795 | elif unit == 'year': | |
796 | unit = 'day' | |
797 | time *= 365 | |
798 | unit += 's' | |
799 | delta = datetime.timedelta(**{unit: time}) | |
800 | return today + delta | |
bd558525 JMF |
801 | return datetime.datetime.strptime(date_str, "%Y%m%d").date() |
802 | ||
803 | class DateRange(object): | |
804 | """Represents a time interval between two dates""" | |
805 | def __init__(self, start=None, end=None): | |
806 | """start and end must be strings in the format accepted by date""" | |
807 | if start is not None: | |
808 | self.start = date_from_str(start) | |
809 | else: | |
810 | self.start = datetime.datetime.min.date() | |
811 | if end is not None: | |
812 | self.end = date_from_str(end) | |
813 | else: | |
814 | self.end = datetime.datetime.max.date() | |
37254abc | 815 | if self.start > self.end: |
bd558525 JMF |
816 | raise ValueError('Date range: "%s" , the start date must be before the end date' % self) |
817 | @classmethod | |
818 | def day(cls, day): | |
819 | """Returns a range that only contains the given day""" | |
820 | return cls(day,day) | |
821 | def __contains__(self, date): | |
822 | """Check if the date is in the range""" | |
37254abc JMF |
823 | if not isinstance(date, datetime.date): |
824 | date = date_from_str(date) | |
825 | return self.start <= date <= self.end | |
bd558525 JMF |
826 | def __str__(self): |
827 | return '%s - %s' % ( self.start.isoformat(), self.end.isoformat()) | |
c496ca96 PH |
828 | |
829 | ||
830 | def platform_name(): | |
831 | """ Returns the platform name as a compat_str """ | |
832 | res = platform.platform() | |
833 | if isinstance(res, bytes): | |
834 | res = res.decode(preferredencoding()) | |
835 | ||
836 | assert isinstance(res, compat_str) | |
837 | return res | |
c257baff PH |
838 | |
839 | ||
7459e3a2 PH |
840 | def write_string(s, out=None): |
841 | if out is None: | |
842 | out = sys.stderr | |
843 | assert type(s) == type(u'') | |
844 | ||
845 | if ('b' in getattr(out, 'mode', '') or | |
846 | sys.version_info[0] < 3): # Python 2 lies about mode of sys.stderr | |
847 | s = s.encode(preferredencoding(), 'ignore') | |
848 | out.write(s) | |
849 | out.flush() | |
850 | ||
851 | ||
48ea9cea PH |
852 | def bytes_to_intlist(bs): |
853 | if not bs: | |
854 | return [] | |
855 | if isinstance(bs[0], int): # Python 3 | |
856 | return list(bs) | |
857 | else: | |
858 | return [ord(c) for c in bs] | |
859 | ||
c257baff | 860 | |
cba892fa | 861 | def intlist_to_bytes(xs): |
862 | if not xs: | |
863 | return b'' | |
864 | if isinstance(chr(0), bytes): # Python 2 | |
865 | return ''.join([chr(x) for x in xs]) | |
866 | else: | |
867 | return bytes(xs) | |
c38b1e77 PH |
868 | |
869 | ||
870 | def get_cachedir(params={}): | |
871 | cache_root = os.environ.get('XDG_CACHE_HOME', | |
872 | os.path.expanduser('~/.cache')) | |
873 | return params.get('cachedir', os.path.join(cache_root, 'youtube-dl')) | |
c1c9a79c PH |
874 | |
875 | ||
876 | # Cross-platform file locking | |
877 | if sys.platform == 'win32': | |
878 | import ctypes.wintypes | |
879 | import msvcrt | |
880 | ||
881 | class OVERLAPPED(ctypes.Structure): | |
882 | _fields_ = [ | |
883 | ('Internal', ctypes.wintypes.LPVOID), | |
884 | ('InternalHigh', ctypes.wintypes.LPVOID), | |
885 | ('Offset', ctypes.wintypes.DWORD), | |
886 | ('OffsetHigh', ctypes.wintypes.DWORD), | |
887 | ('hEvent', ctypes.wintypes.HANDLE), | |
888 | ] | |
889 | ||
890 | kernel32 = ctypes.windll.kernel32 | |
891 | LockFileEx = kernel32.LockFileEx | |
892 | LockFileEx.argtypes = [ | |
893 | ctypes.wintypes.HANDLE, # hFile | |
894 | ctypes.wintypes.DWORD, # dwFlags | |
895 | ctypes.wintypes.DWORD, # dwReserved | |
896 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow | |
897 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh | |
898 | ctypes.POINTER(OVERLAPPED) # Overlapped | |
899 | ] | |
900 | LockFileEx.restype = ctypes.wintypes.BOOL | |
901 | UnlockFileEx = kernel32.UnlockFileEx | |
902 | UnlockFileEx.argtypes = [ | |
903 | ctypes.wintypes.HANDLE, # hFile | |
904 | ctypes.wintypes.DWORD, # dwReserved | |
905 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow | |
906 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh | |
907 | ctypes.POINTER(OVERLAPPED) # Overlapped | |
908 | ] | |
909 | UnlockFileEx.restype = ctypes.wintypes.BOOL | |
910 | whole_low = 0xffffffff | |
911 | whole_high = 0x7fffffff | |
912 | ||
913 | def _lock_file(f, exclusive): | |
914 | overlapped = OVERLAPPED() | |
915 | overlapped.Offset = 0 | |
916 | overlapped.OffsetHigh = 0 | |
917 | overlapped.hEvent = 0 | |
918 | f._lock_file_overlapped_p = ctypes.pointer(overlapped) | |
919 | handle = msvcrt.get_osfhandle(f.fileno()) | |
920 | if not LockFileEx(handle, 0x2 if exclusive else 0x0, 0, | |
921 | whole_low, whole_high, f._lock_file_overlapped_p): | |
922 | raise OSError('Locking file failed: %r' % ctypes.FormatError()) | |
923 | ||
924 | def _unlock_file(f): | |
925 | assert f._lock_file_overlapped_p | |
926 | handle = msvcrt.get_osfhandle(f.fileno()) | |
927 | if not UnlockFileEx(handle, 0, | |
928 | whole_low, whole_high, f._lock_file_overlapped_p): | |
929 | raise OSError('Unlocking file failed: %r' % ctypes.FormatError()) | |
930 | ||
931 | else: | |
932 | import fcntl | |
933 | ||
934 | def _lock_file(f, exclusive): | |
935 | fcntl.lockf(f, fcntl.LOCK_EX if exclusive else fcntl.LOCK_SH) | |
936 | ||
937 | def _unlock_file(f): | |
938 | fcntl.lockf(f, fcntl.LOCK_UN) | |
939 | ||
940 | ||
941 | class locked_file(object): | |
942 | def __init__(self, filename, mode, encoding=None): | |
943 | assert mode in ['r', 'a', 'w'] | |
944 | self.f = io.open(filename, mode, encoding=encoding) | |
945 | self.mode = mode | |
946 | ||
947 | def __enter__(self): | |
948 | exclusive = self.mode != 'r' | |
949 | try: | |
950 | _lock_file(self.f, exclusive) | |
951 | except IOError: | |
952 | self.f.close() | |
953 | raise | |
954 | return self | |
955 | ||
956 | def __exit__(self, etype, value, traceback): | |
957 | try: | |
958 | _unlock_file(self.f) | |
959 | finally: | |
960 | self.f.close() | |
961 | ||
962 | def __iter__(self): | |
963 | return iter(self.f) | |
964 | ||
965 | def write(self, *args): | |
966 | return self.f.write(*args) | |
967 | ||
968 | def read(self, *args): | |
969 | return self.f.read(*args) | |
4eb7f1d1 JMF |
970 | |
971 | ||
972 | def shell_quote(args): | |
a6a173c2 JMF |
973 | quoted_args = [] |
974 | encoding = sys.getfilesystemencoding() | |
975 | if encoding is None: | |
976 | encoding = 'utf-8' | |
977 | for a in args: | |
978 | if isinstance(a, bytes): | |
979 | # We may get a filename encoded with 'encodeFilename' | |
980 | a = a.decode(encoding) | |
981 | quoted_args.append(pipes.quote(a)) | |
982 | return u' '.join(quoted_args) | |
9d4660ca PH |
983 | |
984 | ||
f4d96df0 PH |
985 | def takewhile_inclusive(pred, seq): |
986 | """ Like itertools.takewhile, but include the latest evaluated element | |
987 | (the first element so that Not pred(e)) """ | |
988 | for e in seq: | |
989 | yield e | |
990 | if not pred(e): | |
991 | return | |
992 | ||
993 | ||
9d4660ca PH |
994 | def smuggle_url(url, data): |
995 | """ Pass additional data in a URL for internal use. """ | |
996 | ||
997 | sdata = compat_urllib_parse.urlencode( | |
998 | {u'__youtubedl_smuggle': json.dumps(data)}) | |
999 | return url + u'#' + sdata | |
1000 | ||
1001 | ||
1002 | def unsmuggle_url(smug_url): | |
1003 | if not '#__youtubedl_smuggle' in smug_url: | |
1004 | return smug_url, None | |
1005 | url, _, sdata = smug_url.rpartition(u'#') | |
1006 | jsond = compat_parse_qs(sdata)[u'__youtubedl_smuggle'][0] | |
1007 | data = json.loads(jsond) | |
1008 | return url, data | |
02dbf93f PH |
1009 | |
1010 | ||
02dbf93f PH |
1011 | def format_bytes(bytes): |
1012 | if bytes is None: | |
1013 | return u'N/A' | |
1014 | if type(bytes) is str: | |
1015 | bytes = float(bytes) | |
1016 | if bytes == 0.0: | |
1017 | exponent = 0 | |
1018 | else: | |
1019 | exponent = int(math.log(bytes, 1024.0)) | |
1020 | suffix = [u'B', u'KiB', u'MiB', u'GiB', u'TiB', u'PiB', u'EiB', u'ZiB', u'YiB'][exponent] | |
1021 | converted = float(bytes) / float(1024 ** exponent) | |
1022 | return u'%.2f%s' % (converted, suffix) |