]>
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 |
02dbf93f | 20 | import xml.etree.ElementTree |
d77c3dfd | 21 | import zlib |
d77c3dfd | 22 | |
01ba00ca | 23 | try: |
59ae15a5 | 24 | import urllib.request as compat_urllib_request |
01ba00ca | 25 | except ImportError: # Python 2 |
59ae15a5 | 26 | import urllib2 as compat_urllib_request |
01ba00ca PH |
27 | |
28 | try: | |
59ae15a5 | 29 | import urllib.error as compat_urllib_error |
01ba00ca | 30 | except ImportError: # Python 2 |
59ae15a5 | 31 | import urllib2 as compat_urllib_error |
01ba00ca PH |
32 | |
33 | try: | |
59ae15a5 | 34 | import urllib.parse as compat_urllib_parse |
01ba00ca | 35 | except ImportError: # Python 2 |
59ae15a5 | 36 | import urllib as compat_urllib_parse |
01ba00ca | 37 | |
799c0763 PH |
38 | try: |
39 | from urllib.parse import urlparse as compat_urllib_parse_urlparse | |
40 | except ImportError: # Python 2 | |
41 | from urlparse import urlparse as compat_urllib_parse_urlparse | |
42 | ||
6543f0dc JMF |
43 | try: |
44 | import urllib.parse as compat_urlparse | |
45 | except ImportError: # Python 2 | |
46 | import urlparse as compat_urlparse | |
47 | ||
01ba00ca | 48 | try: |
59ae15a5 | 49 | import http.cookiejar as compat_cookiejar |
01ba00ca | 50 | except ImportError: # Python 2 |
59ae15a5 | 51 | import cookielib as compat_cookiejar |
01ba00ca | 52 | |
3e669f36 | 53 | try: |
59ae15a5 | 54 | import html.entities as compat_html_entities |
9f37a959 | 55 | except ImportError: # Python 2 |
59ae15a5 | 56 | import htmlentitydefs as compat_html_entities |
3e669f36 | 57 | |
a8156c1d | 58 | try: |
59ae15a5 | 59 | import html.parser as compat_html_parser |
9f37a959 | 60 | except ImportError: # Python 2 |
59ae15a5 | 61 | import HTMLParser as compat_html_parser |
a8156c1d | 62 | |
348d0a7a | 63 | try: |
59ae15a5 | 64 | import http.client as compat_http_client |
9f37a959 | 65 | except ImportError: # Python 2 |
59ae15a5 | 66 | import httplib as compat_http_client |
348d0a7a | 67 | |
2eabb802 | 68 | try: |
0e283428 | 69 | from urllib.error import HTTPError as compat_HTTPError |
2eabb802 PH |
70 | except ImportError: # Python 2 |
71 | from urllib2 import HTTPError as compat_HTTPError | |
72 | ||
e0df6211 PH |
73 | try: |
74 | from urllib.request import urlretrieve as compat_urlretrieve | |
75 | except ImportError: # Python 2 | |
76 | from urllib import urlretrieve as compat_urlretrieve | |
77 | ||
78 | ||
5910e210 PH |
79 | try: |
80 | from subprocess import DEVNULL | |
81 | compat_subprocess_get_DEVNULL = lambda: DEVNULL | |
82 | except ImportError: | |
83 | compat_subprocess_get_DEVNULL = lambda: open(os.path.devnull, 'w') | |
84 | ||
9f37a959 | 85 | try: |
59ae15a5 | 86 | from urllib.parse import parse_qs as compat_parse_qs |
9f37a959 | 87 | except ImportError: # Python 2 |
59ae15a5 PH |
88 | # HACK: The following is the correct parse_qs implementation from cpython 3's stdlib. |
89 | # Python 2's version is apparently totally broken | |
90 | def _unquote(string, encoding='utf-8', errors='replace'): | |
91 | if string == '': | |
92 | return string | |
93 | res = string.split('%') | |
94 | if len(res) == 1: | |
95 | return string | |
96 | if encoding is None: | |
97 | encoding = 'utf-8' | |
98 | if errors is None: | |
99 | errors = 'replace' | |
100 | # pct_sequence: contiguous sequence of percent-encoded bytes, decoded | |
101 | pct_sequence = b'' | |
102 | string = res[0] | |
103 | for item in res[1:]: | |
104 | try: | |
105 | if not item: | |
106 | raise ValueError | |
107 | pct_sequence += item[:2].decode('hex') | |
108 | rest = item[2:] | |
109 | if not rest: | |
110 | # This segment was just a single percent-encoded character. | |
111 | # May be part of a sequence of code units, so delay decoding. | |
112 | # (Stored in pct_sequence). | |
113 | continue | |
114 | except ValueError: | |
115 | rest = '%' + item | |
116 | # Encountered non-percent-encoded characters. Flush the current | |
117 | # pct_sequence. | |
118 | string += pct_sequence.decode(encoding, errors) + rest | |
119 | pct_sequence = b'' | |
120 | if pct_sequence: | |
121 | # Flush the final pct_sequence | |
122 | string += pct_sequence.decode(encoding, errors) | |
123 | return string | |
124 | ||
125 | def _parse_qsl(qs, keep_blank_values=False, strict_parsing=False, | |
126 | encoding='utf-8', errors='replace'): | |
127 | qs, _coerce_result = qs, unicode | |
128 | pairs = [s2 for s1 in qs.split('&') for s2 in s1.split(';')] | |
129 | r = [] | |
130 | for name_value in pairs: | |
131 | if not name_value and not strict_parsing: | |
132 | continue | |
133 | nv = name_value.split('=', 1) | |
134 | if len(nv) != 2: | |
135 | if strict_parsing: | |
136 | raise ValueError("bad query field: %r" % (name_value,)) | |
137 | # Handle case of a control-name with no equal sign | |
138 | if keep_blank_values: | |
139 | nv.append('') | |
140 | else: | |
141 | continue | |
142 | if len(nv[1]) or keep_blank_values: | |
143 | name = nv[0].replace('+', ' ') | |
144 | name = _unquote(name, encoding=encoding, errors=errors) | |
145 | name = _coerce_result(name) | |
146 | value = nv[1].replace('+', ' ') | |
147 | value = _unquote(value, encoding=encoding, errors=errors) | |
148 | value = _coerce_result(value) | |
149 | r.append((name, value)) | |
150 | return r | |
151 | ||
152 | def compat_parse_qs(qs, keep_blank_values=False, strict_parsing=False, | |
153 | encoding='utf-8', errors='replace'): | |
154 | parsed_result = {} | |
155 | pairs = _parse_qsl(qs, keep_blank_values, strict_parsing, | |
156 | encoding=encoding, errors=errors) | |
157 | for name, value in pairs: | |
158 | if name in parsed_result: | |
159 | parsed_result[name].append(value) | |
160 | else: | |
161 | parsed_result[name] = [value] | |
162 | return parsed_result | |
348d0a7a | 163 | |
3e669f36 | 164 | try: |
59ae15a5 | 165 | compat_str = unicode # Python 2 |
3e669f36 | 166 | except NameError: |
59ae15a5 | 167 | compat_str = str |
3e669f36 PH |
168 | |
169 | try: | |
59ae15a5 | 170 | compat_chr = unichr # Python 2 |
3e669f36 | 171 | except NameError: |
59ae15a5 | 172 | compat_chr = chr |
3e669f36 | 173 | |
b31756c1 FV |
174 | def compat_ord(c): |
175 | if type(c) is int: return c | |
176 | else: return ord(c) | |
177 | ||
468e2e92 FV |
178 | # This is not clearly defined otherwise |
179 | compiled_regex_type = type(re.compile('')) | |
180 | ||
3e669f36 | 181 | std_headers = { |
ae8f7871 | 182 | 'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20100101 Firefox/10.0 (Chrome)', |
59ae15a5 PH |
183 | 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7', |
184 | 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8', | |
185 | 'Accept-Encoding': 'gzip, deflate', | |
186 | 'Accept-Language': 'en-us,en;q=0.5', | |
3e669f36 | 187 | } |
f427df17 | 188 | |
d77c3dfd | 189 | def preferredencoding(): |
59ae15a5 | 190 | """Get preferred encoding. |
d77c3dfd | 191 | |
59ae15a5 PH |
192 | Returns the best encoding scheme for the system, based on |
193 | locale.getpreferredencoding() and some further tweaks. | |
194 | """ | |
195 | try: | |
196 | pref = locale.getpreferredencoding() | |
197 | u'TEST'.encode(pref) | |
198 | except: | |
199 | pref = 'UTF-8' | |
bae611f2 | 200 | |
59ae15a5 | 201 | return pref |
d77c3dfd | 202 | |
8cd10ac4 | 203 | if sys.version_info < (3,0): |
59ae15a5 PH |
204 | def compat_print(s): |
205 | print(s.encode(preferredencoding(), 'xmlcharrefreplace')) | |
8cd10ac4 | 206 | else: |
59ae15a5 PH |
207 | def compat_print(s): |
208 | assert type(s) == type(u'') | |
209 | print(s) | |
d77c3dfd | 210 | |
f4bfd65f PH |
211 | # In Python 2.x, json.dump expects a bytestream. |
212 | # In Python 3.x, it writes to a character stream | |
213 | if sys.version_info < (3,0): | |
214 | def write_json_file(obj, fn): | |
215 | with open(fn, 'wb') as f: | |
216 | json.dump(obj, f) | |
217 | else: | |
218 | def write_json_file(obj, fn): | |
219 | with open(fn, 'w', encoding='utf-8') as f: | |
220 | json.dump(obj, f) | |
221 | ||
59ae56fa PH |
222 | if sys.version_info >= (2,7): |
223 | def find_xpath_attr(node, xpath, key, val): | |
224 | """ Find the xpath xpath[@key=val] """ | |
5de3ece2 | 225 | assert re.match(r'^[a-zA-Z]+$', key) |
54543467 | 226 | assert re.match(r'^[a-zA-Z0-9@\s]*$', val) |
59ae56fa PH |
227 | expr = xpath + u"[@%s='%s']" % (key, val) |
228 | return node.find(expr) | |
229 | else: | |
230 | def find_xpath_attr(node, xpath, key, val): | |
231 | for f in node.findall(xpath): | |
232 | if f.attrib.get(key) == val: | |
233 | return f | |
234 | return None | |
235 | ||
d7e66d39 JMF |
236 | # On python2.6 the xml.etree.ElementTree.Element methods don't support |
237 | # the namespace parameter | |
238 | def xpath_with_ns(path, ns_map): | |
239 | components = [c.split(':') for c in path.split('/')] | |
240 | replaced = [] | |
241 | for c in components: | |
242 | if len(c) == 1: | |
243 | replaced.append(c[0]) | |
244 | else: | |
245 | ns, tag = c | |
246 | replaced.append('{%s}%s' % (ns_map[ns], tag)) | |
247 | return '/'.join(replaced) | |
248 | ||
d77c3dfd | 249 | def htmlentity_transform(matchobj): |
59ae15a5 PH |
250 | """Transforms an HTML entity to a character. |
251 | ||
252 | This function receives a match object and is intended to be used with | |
253 | the re.sub() function. | |
254 | """ | |
255 | entity = matchobj.group(1) | |
256 | ||
257 | # Known non-numeric HTML entity | |
258 | if entity in compat_html_entities.name2codepoint: | |
259 | return compat_chr(compat_html_entities.name2codepoint[entity]) | |
260 | ||
261 | mobj = re.match(u'(?u)#(x?\\d+)', entity) | |
262 | if mobj is not None: | |
263 | numstr = mobj.group(1) | |
264 | if numstr.startswith(u'x'): | |
265 | base = 16 | |
266 | numstr = u'0%s' % numstr | |
267 | else: | |
268 | base = 10 | |
269 | return compat_chr(int(numstr, base)) | |
270 | ||
271 | # Unknown entity in name, return its literal representation | |
272 | return (u'&%s;' % entity) | |
d77c3dfd | 273 | |
a8156c1d | 274 | 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 |
275 | class BaseHTMLParser(compat_html_parser.HTMLParser): |
276 | def __init(self): | |
277 | compat_html_parser.HTMLParser.__init__(self) | |
278 | self.html = None | |
279 | ||
280 | def loads(self, html): | |
281 | self.html = html | |
282 | self.feed(html) | |
283 | self.close() | |
284 | ||
285 | class AttrParser(BaseHTMLParser): | |
43e8fafd ND |
286 | """Modified HTMLParser that isolates a tag with the specified attribute""" |
287 | def __init__(self, attribute, value): | |
288 | self.attribute = attribute | |
289 | self.value = value | |
59ae15a5 PH |
290 | self.result = None |
291 | self.started = False | |
292 | self.depth = {} | |
59ae15a5 PH |
293 | self.watch_startpos = False |
294 | self.error_count = 0 | |
a921f407 | 295 | BaseHTMLParser.__init__(self) |
59ae15a5 PH |
296 | |
297 | def error(self, message): | |
298 | if self.error_count > 10 or self.started: | |
299 | raise compat_html_parser.HTMLParseError(message, self.getpos()) | |
300 | self.rawdata = '\n'.join(self.html.split('\n')[self.getpos()[0]:]) # skip one line | |
301 | self.error_count += 1 | |
302 | self.goahead(1) | |
303 | ||
59ae15a5 PH |
304 | def handle_starttag(self, tag, attrs): |
305 | attrs = dict(attrs) | |
306 | if self.started: | |
307 | self.find_startpos(None) | |
43e8fafd | 308 | if self.attribute in attrs and attrs[self.attribute] == self.value: |
59ae15a5 PH |
309 | self.result = [tag] |
310 | self.started = True | |
311 | self.watch_startpos = True | |
312 | if self.started: | |
313 | if not tag in self.depth: self.depth[tag] = 0 | |
314 | self.depth[tag] += 1 | |
315 | ||
316 | def handle_endtag(self, tag): | |
317 | if self.started: | |
318 | if tag in self.depth: self.depth[tag] -= 1 | |
319 | if self.depth[self.result[0]] == 0: | |
320 | self.started = False | |
321 | self.result.append(self.getpos()) | |
322 | ||
323 | def find_startpos(self, x): | |
324 | """Needed to put the start position of the result (self.result[1]) | |
325 | after the opening tag with the requested id""" | |
326 | if self.watch_startpos: | |
327 | self.watch_startpos = False | |
328 | self.result.append(self.getpos()) | |
329 | handle_entityref = handle_charref = handle_data = handle_comment = \ | |
330 | handle_decl = handle_pi = unknown_decl = find_startpos | |
331 | ||
332 | def get_result(self): | |
333 | if self.result is None: | |
334 | return None | |
335 | if len(self.result) != 3: | |
336 | return None | |
337 | lines = self.html.split('\n') | |
338 | lines = lines[self.result[1][0]-1:self.result[2][0]] | |
339 | lines[0] = lines[0][self.result[1][1]:] | |
340 | if len(lines) == 1: | |
341 | lines[-1] = lines[-1][:self.result[2][1]-self.result[1][1]] | |
342 | lines[-1] = lines[-1][:self.result[2][1]] | |
343 | return '\n'.join(lines).strip() | |
3b024e17 PH |
344 | # Hack for https://github.com/rg3/youtube-dl/issues/662 |
345 | if sys.version_info < (2, 7, 3): | |
346 | AttrParser.parse_endtag = (lambda self, i: | |
347 | i + len("</scr'+'ipt>") | |
348 | if self.rawdata[i:].startswith("</scr'+'ipt>") | |
349 | else compat_html_parser.HTMLParser.parse_endtag(self, i)) | |
9e6dd238 FV |
350 | |
351 | def get_element_by_id(id, html): | |
43e8fafd ND |
352 | """Return the content of the tag with the specified ID in the passed HTML document""" |
353 | return get_element_by_attribute("id", id, html) | |
354 | ||
355 | def get_element_by_attribute(attribute, value, html): | |
356 | """Return the content of the tag with the specified attribute in the passed HTML document""" | |
357 | parser = AttrParser(attribute, value) | |
59ae15a5 PH |
358 | try: |
359 | parser.loads(html) | |
360 | except compat_html_parser.HTMLParseError: | |
361 | pass | |
362 | return parser.get_result() | |
9e6dd238 | 363 | |
a921f407 JMF |
364 | class MetaParser(BaseHTMLParser): |
365 | """ | |
366 | Modified HTMLParser that isolates a meta tag with the specified name | |
367 | attribute. | |
368 | """ | |
369 | def __init__(self, name): | |
370 | BaseHTMLParser.__init__(self) | |
371 | self.name = name | |
372 | self.content = None | |
373 | self.result = None | |
374 | ||
375 | def handle_starttag(self, tag, attrs): | |
376 | if tag != 'meta': | |
377 | return | |
378 | attrs = dict(attrs) | |
379 | if attrs.get('name') == self.name: | |
380 | self.result = attrs.get('content') | |
381 | ||
382 | def get_result(self): | |
383 | return self.result | |
384 | ||
385 | def get_meta_content(name, html): | |
386 | """ | |
387 | Return the content attribute from the meta tag with the given name attribute. | |
388 | """ | |
389 | parser = MetaParser(name) | |
390 | try: | |
391 | parser.loads(html) | |
392 | except compat_html_parser.HTMLParseError: | |
393 | pass | |
394 | return parser.get_result() | |
395 | ||
9e6dd238 FV |
396 | |
397 | def clean_html(html): | |
59ae15a5 PH |
398 | """Clean an HTML snippet into a readable string""" |
399 | # Newline vs <br /> | |
400 | html = html.replace('\n', ' ') | |
6b3aef80 FV |
401 | html = re.sub(r'\s*<\s*br\s*/?\s*>\s*', '\n', html) |
402 | html = re.sub(r'<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html) | |
59ae15a5 PH |
403 | # Strip html tags |
404 | html = re.sub('<.*?>', '', html) | |
405 | # Replace html entities | |
406 | html = unescapeHTML(html) | |
7decf895 | 407 | return html.strip() |
9e6dd238 FV |
408 | |
409 | ||
d77c3dfd | 410 | def sanitize_open(filename, open_mode): |
59ae15a5 PH |
411 | """Try to open the given filename, and slightly tweak it if this fails. |
412 | ||
413 | Attempts to open the given filename. If this fails, it tries to change | |
414 | the filename slightly, step by step, until it's either able to open it | |
415 | or it fails and raises a final exception, like the standard open() | |
416 | function. | |
417 | ||
418 | It returns the tuple (stream, definitive_file_name). | |
419 | """ | |
420 | try: | |
421 | if filename == u'-': | |
422 | if sys.platform == 'win32': | |
423 | import msvcrt | |
424 | msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY) | |
898280a0 | 425 | return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename) |
59ae15a5 PH |
426 | stream = open(encodeFilename(filename), open_mode) |
427 | return (stream, filename) | |
428 | except (IOError, OSError) as err: | |
f45c185f PH |
429 | if err.errno in (errno.EACCES,): |
430 | raise | |
59ae15a5 | 431 | |
f45c185f PH |
432 | # In case of error, try to remove win32 forbidden chars |
433 | alt_filename = os.path.join( | |
434 | re.sub(u'[/<>:"\\|\\\\?\\*]', u'#', path_part) | |
435 | for path_part in os.path.split(filename) | |
436 | ) | |
437 | if alt_filename == filename: | |
438 | raise | |
439 | else: | |
440 | # An exception here should be caught in the caller | |
441 | stream = open(encodeFilename(filename), open_mode) | |
442 | return (stream, alt_filename) | |
d77c3dfd FV |
443 | |
444 | ||
445 | def timeconvert(timestr): | |
59ae15a5 PH |
446 | """Convert RFC 2822 defined time string into system timestamp""" |
447 | timestamp = None | |
448 | timetuple = email.utils.parsedate_tz(timestr) | |
449 | if timetuple is not None: | |
450 | timestamp = email.utils.mktime_tz(timetuple) | |
451 | return timestamp | |
1c469a94 | 452 | |
796173d0 | 453 | def sanitize_filename(s, restricted=False, is_id=False): |
59ae15a5 PH |
454 | """Sanitizes a string so it could be used as part of a filename. |
455 | If restricted is set, use a stricter subset of allowed characters. | |
796173d0 | 456 | Set is_id if this is not an arbitrary string, but an ID that should be kept if possible |
59ae15a5 PH |
457 | """ |
458 | def replace_insane(char): | |
459 | if char == '?' or ord(char) < 32 or ord(char) == 127: | |
460 | return '' | |
461 | elif char == '"': | |
462 | return '' if restricted else '\'' | |
463 | elif char == ':': | |
464 | return '_-' if restricted else ' -' | |
465 | elif char in '\\/|*<>': | |
466 | return '_' | |
627dcfff | 467 | if restricted and (char in '!&\'()[]{}$;`^,#' or char.isspace()): |
59ae15a5 PH |
468 | return '_' |
469 | if restricted and ord(char) > 127: | |
470 | return '_' | |
471 | return char | |
472 | ||
473 | result = u''.join(map(replace_insane, s)) | |
796173d0 PH |
474 | if not is_id: |
475 | while '__' in result: | |
476 | result = result.replace('__', '_') | |
477 | result = result.strip('_') | |
478 | # Common case of "Foreign band name - English song title" | |
479 | if restricted and result.startswith('-_'): | |
480 | result = result[2:] | |
481 | if not result: | |
482 | result = '_' | |
59ae15a5 | 483 | return result |
d77c3dfd FV |
484 | |
485 | def orderedSet(iterable): | |
59ae15a5 PH |
486 | """ Remove all duplicates from the input iterable """ |
487 | res = [] | |
488 | for el in iterable: | |
489 | if el not in res: | |
490 | res.append(el) | |
491 | return res | |
d77c3dfd FV |
492 | |
493 | def unescapeHTML(s): | |
59ae15a5 PH |
494 | """ |
495 | @param s a string | |
496 | """ | |
497 | assert type(s) == type(u'') | |
d77c3dfd | 498 | |
59ae15a5 PH |
499 | result = re.sub(u'(?u)&(.+?);', htmlentity_transform, s) |
500 | return result | |
d77c3dfd FV |
501 | |
502 | def encodeFilename(s): | |
59ae15a5 PH |
503 | """ |
504 | @param s The name of the file | |
505 | """ | |
d77c3dfd | 506 | |
59ae15a5 | 507 | assert type(s) == type(u'') |
d77c3dfd | 508 | |
59ae15a5 PH |
509 | # Python 3 has a Unicode API |
510 | if sys.version_info >= (3, 0): | |
511 | return s | |
0f00efed | 512 | |
59ae15a5 PH |
513 | if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5: |
514 | # Pass u'' directly to use Unicode APIs on Windows 2000 and up | |
515 | # (Detecting Windows NT 4 is tricky because 'major >= 4' would | |
516 | # match Windows 9x series as well. Besides, NT 4 is obsolete.) | |
517 | return s | |
518 | else: | |
6df40dcb PH |
519 | encoding = sys.getfilesystemencoding() |
520 | if encoding is None: | |
521 | encoding = 'utf-8' | |
522 | return s.encode(encoding, 'ignore') | |
d77c3dfd | 523 | |
8271226a PH |
524 | def decodeOption(optval): |
525 | if optval is None: | |
526 | return optval | |
527 | if isinstance(optval, bytes): | |
528 | optval = optval.decode(preferredencoding()) | |
529 | ||
530 | assert isinstance(optval, compat_str) | |
531 | return optval | |
1c256f70 | 532 | |
4539dd30 PH |
533 | def formatSeconds(secs): |
534 | if secs > 3600: | |
535 | return '%d:%02d:%02d' % (secs // 3600, (secs % 3600) // 60, secs % 60) | |
536 | elif secs > 60: | |
537 | return '%d:%02d' % (secs // 60, secs % 60) | |
538 | else: | |
539 | return '%d' % secs | |
540 | ||
dca08720 | 541 | def make_HTTPS_handler(opts_no_check_certificate): |
13ebea79 PH |
542 | if sys.version_info < (3, 2): |
543 | import httplib | |
544 | ||
545 | class HTTPSConnectionV3(httplib.HTTPSConnection): | |
546 | def __init__(self, *args, **kwargs): | |
547 | httplib.HTTPSConnection.__init__(self, *args, **kwargs) | |
548 | ||
549 | def connect(self): | |
550 | sock = socket.create_connection((self.host, self.port), self.timeout) | |
551 | if self._tunnel_host: | |
552 | self.sock = sock | |
553 | self._tunnel() | |
554 | try: | |
555 | self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_SSLv3) | |
556 | except ssl.SSLError as e: | |
557 | self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_SSLv23) | |
558 | ||
559 | class HTTPSHandlerV3(compat_urllib_request.HTTPSHandler): | |
560 | def https_open(self, req): | |
561 | return self.do_open(HTTPSConnectionV3, req) | |
562 | return HTTPSHandlerV3() | |
ea6d901e | 563 | else: |
13ebea79 | 564 | context = ssl.SSLContext(ssl.PROTOCOL_SSLv3) |
ea6d901e PH |
565 | context.set_default_verify_paths() |
566 | ||
567 | context.verify_mode = (ssl.CERT_NONE | |
dca08720 | 568 | if opts_no_check_certificate |
ea6d901e | 569 | else ssl.CERT_REQUIRED) |
acebc9cd | 570 | return compat_urllib_request.HTTPSHandler(context=context) |
ea6d901e | 571 | |
1c256f70 PH |
572 | class ExtractorError(Exception): |
573 | """Error during info extraction.""" | |
2eabb802 | 574 | def __init__(self, msg, tb=None, expected=False, cause=None): |
9a82b238 PH |
575 | """ tb, if given, is the original traceback (so that it can be printed out). |
576 | If expected is set, this is a normal error message and most likely not a bug in youtube-dl. | |
577 | """ | |
578 | ||
579 | if sys.exc_info()[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError): | |
580 | expected = True | |
581 | if not expected: | |
298f833b | 582 | 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 | 583 | super(ExtractorError, self).__init__(msg) |
d5979c5d | 584 | |
1c256f70 | 585 | self.traceback = tb |
8cc83b8d | 586 | self.exc_info = sys.exc_info() # preserve original exception |
2eabb802 | 587 | self.cause = cause |
1c256f70 | 588 | |
01951dda PH |
589 | def format_traceback(self): |
590 | if self.traceback is None: | |
591 | return None | |
592 | return u''.join(traceback.format_tb(self.traceback)) | |
593 | ||
1c256f70 | 594 | |
55b3e45b JMF |
595 | class RegexNotFoundError(ExtractorError): |
596 | """Error when a regex didn't match""" | |
597 | pass | |
598 | ||
599 | ||
d77c3dfd | 600 | class DownloadError(Exception): |
59ae15a5 | 601 | """Download Error exception. |
d77c3dfd | 602 | |
59ae15a5 PH |
603 | This exception may be thrown by FileDownloader objects if they are not |
604 | configured to continue on errors. They will contain the appropriate | |
605 | error message. | |
606 | """ | |
8cc83b8d FV |
607 | def __init__(self, msg, exc_info=None): |
608 | """ exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """ | |
609 | super(DownloadError, self).__init__(msg) | |
610 | self.exc_info = exc_info | |
d77c3dfd FV |
611 | |
612 | ||
613 | class SameFileError(Exception): | |
59ae15a5 | 614 | """Same File exception. |
d77c3dfd | 615 | |
59ae15a5 PH |
616 | This exception will be thrown by FileDownloader objects if they detect |
617 | multiple files would have to be downloaded to the same file on disk. | |
618 | """ | |
619 | pass | |
d77c3dfd FV |
620 | |
621 | ||
622 | class PostProcessingError(Exception): | |
59ae15a5 | 623 | """Post Processing exception. |
d77c3dfd | 624 | |
59ae15a5 PH |
625 | This exception may be raised by PostProcessor's .run() method to |
626 | indicate an error in the postprocessing task. | |
627 | """ | |
7851b379 PH |
628 | def __init__(self, msg): |
629 | self.msg = msg | |
d77c3dfd FV |
630 | |
631 | class MaxDownloadsReached(Exception): | |
59ae15a5 PH |
632 | """ --max-downloads limit has been reached. """ |
633 | pass | |
d77c3dfd FV |
634 | |
635 | ||
636 | class UnavailableVideoError(Exception): | |
59ae15a5 | 637 | """Unavailable Format exception. |
d77c3dfd | 638 | |
59ae15a5 PH |
639 | This exception will be thrown when a video is requested |
640 | in a format that is not available for that video. | |
641 | """ | |
642 | pass | |
d77c3dfd FV |
643 | |
644 | ||
645 | class ContentTooShortError(Exception): | |
59ae15a5 | 646 | """Content Too Short exception. |
d77c3dfd | 647 | |
59ae15a5 PH |
648 | This exception may be raised by FileDownloader objects when a file they |
649 | download is too small for what the server announced first, indicating | |
650 | the connection was probably interrupted. | |
651 | """ | |
652 | # Both in bytes | |
653 | downloaded = None | |
654 | expected = None | |
d77c3dfd | 655 | |
59ae15a5 PH |
656 | def __init__(self, downloaded, expected): |
657 | self.downloaded = downloaded | |
658 | self.expected = expected | |
d77c3dfd | 659 | |
acebc9cd | 660 | class YoutubeDLHandler(compat_urllib_request.HTTPHandler): |
59ae15a5 PH |
661 | """Handler for HTTP requests and responses. |
662 | ||
663 | This class, when installed with an OpenerDirector, automatically adds | |
664 | the standard headers to every HTTP request and handles gzipped and | |
665 | deflated responses from web servers. If compression is to be avoided in | |
666 | a particular request, the original request in the program code only has | |
667 | to include the HTTP header "Youtubedl-No-Compression", which will be | |
668 | removed before making the real request. | |
669 | ||
670 | Part of this code was copied from: | |
671 | ||
672 | http://techknack.net/python-urllib2-handlers/ | |
673 | ||
674 | Andrew Rowls, the author of that code, agreed to release it to the | |
675 | public domain. | |
676 | """ | |
677 | ||
678 | @staticmethod | |
679 | def deflate(data): | |
680 | try: | |
681 | return zlib.decompress(data, -zlib.MAX_WBITS) | |
682 | except zlib.error: | |
683 | return zlib.decompress(data) | |
684 | ||
685 | @staticmethod | |
686 | def addinfourl_wrapper(stream, headers, url, code): | |
687 | if hasattr(compat_urllib_request.addinfourl, 'getcode'): | |
688 | return compat_urllib_request.addinfourl(stream, headers, url, code) | |
689 | ret = compat_urllib_request.addinfourl(stream, headers, url) | |
690 | ret.code = code | |
691 | return ret | |
692 | ||
acebc9cd PH |
693 | def http_request(self, req): |
694 | for h,v in std_headers.items(): | |
59ae15a5 PH |
695 | if h in req.headers: |
696 | del req.headers[h] | |
335959e7 | 697 | req.add_header(h, v) |
59ae15a5 PH |
698 | if 'Youtubedl-no-compression' in req.headers: |
699 | if 'Accept-encoding' in req.headers: | |
700 | del req.headers['Accept-encoding'] | |
701 | del req.headers['Youtubedl-no-compression'] | |
3446dfb7 | 702 | if 'Youtubedl-user-agent' in req.headers: |
335959e7 PH |
703 | if 'User-agent' in req.headers: |
704 | del req.headers['User-agent'] | |
705 | req.headers['User-agent'] = req.headers['Youtubedl-user-agent'] | |
3446dfb7 | 706 | del req.headers['Youtubedl-user-agent'] |
59ae15a5 PH |
707 | return req |
708 | ||
acebc9cd | 709 | def http_response(self, req, resp): |
59ae15a5 PH |
710 | old_resp = resp |
711 | # gzip | |
712 | if resp.headers.get('Content-encoding', '') == 'gzip': | |
aa3e9507 PH |
713 | content = resp.read() |
714 | gz = gzip.GzipFile(fileobj=io.BytesIO(content), mode='rb') | |
715 | try: | |
716 | uncompressed = io.BytesIO(gz.read()) | |
717 | except IOError as original_ioerror: | |
718 | # There may be junk add the end of the file | |
719 | # See http://stackoverflow.com/q/4928560/35070 for details | |
720 | for i in range(1, 1024): | |
721 | try: | |
722 | gz = gzip.GzipFile(fileobj=io.BytesIO(content[:-i]), mode='rb') | |
723 | uncompressed = io.BytesIO(gz.read()) | |
724 | except IOError: | |
725 | continue | |
726 | break | |
727 | else: | |
728 | raise original_ioerror | |
729 | resp = self.addinfourl_wrapper(uncompressed, old_resp.headers, old_resp.url, old_resp.code) | |
59ae15a5 PH |
730 | resp.msg = old_resp.msg |
731 | # deflate | |
732 | if resp.headers.get('Content-encoding', '') == 'deflate': | |
733 | gz = io.BytesIO(self.deflate(resp.read())) | |
734 | resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code) | |
735 | resp.msg = old_resp.msg | |
736 | return resp | |
0f8d03f8 | 737 | |
acebc9cd PH |
738 | https_request = http_request |
739 | https_response = http_response | |
bf50b038 JMF |
740 | |
741 | def unified_strdate(date_str): | |
742 | """Return a string with the date in the format YYYYMMDD""" | |
743 | upload_date = None | |
744 | #Replace commas | |
745 | date_str = date_str.replace(',',' ') | |
746 | # %z (UTC offset) is only supported in python>=3.2 | |
747 | date_str = re.sub(r' (\+|-)[\d]*$', '', date_str) | |
19e1d359 JMF |
748 | format_expressions = [ |
749 | '%d %B %Y', | |
750 | '%B %d %Y', | |
751 | '%b %d %Y', | |
752 | '%Y-%m-%d', | |
753 | '%d/%m/%Y', | |
754 | '%Y/%m/%d %H:%M:%S', | |
755 | '%d.%m.%Y %H:%M', | |
756 | '%Y-%m-%dT%H:%M:%SZ', | |
59040888 PH |
757 | '%Y-%m-%dT%H:%M:%S.%fZ', |
758 | '%Y-%m-%dT%H:%M:%S.%f0Z', | |
2e1fa03b | 759 | '%Y-%m-%dT%H:%M:%S', |
19e1d359 | 760 | ] |
bf50b038 JMF |
761 | for expression in format_expressions: |
762 | try: | |
763 | upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d') | |
764 | except: | |
765 | pass | |
766 | return upload_date | |
767 | ||
cbdbb766 | 768 | def determine_ext(url, default_ext=u'unknown_video'): |
73e79f2a PH |
769 | guess = url.partition(u'?')[0].rpartition(u'.')[2] |
770 | if re.match(r'^[A-Za-z0-9]+$', guess): | |
771 | return guess | |
772 | else: | |
cbdbb766 | 773 | return default_ext |
73e79f2a | 774 | |
d4051a8e JMF |
775 | def subtitles_filename(filename, sub_lang, sub_format): |
776 | return filename.rsplit('.', 1)[0] + u'.' + sub_lang + u'.' + sub_format | |
777 | ||
bd558525 | 778 | def date_from_str(date_str): |
37254abc JMF |
779 | """ |
780 | Return a datetime object from a string in the format YYYYMMDD or | |
781 | (now|today)[+-][0-9](day|week|month|year)(s)?""" | |
782 | today = datetime.date.today() | |
783 | if date_str == 'now'or date_str == 'today': | |
784 | return today | |
785 | match = re.match('(now|today)(?P<sign>[+-])(?P<time>\d+)(?P<unit>day|week|month|year)(s)?', date_str) | |
786 | if match is not None: | |
787 | sign = match.group('sign') | |
788 | time = int(match.group('time')) | |
789 | if sign == '-': | |
790 | time = -time | |
791 | unit = match.group('unit') | |
792 | #A bad aproximation? | |
793 | if unit == 'month': | |
794 | unit = 'day' | |
795 | time *= 30 | |
796 | elif unit == 'year': | |
797 | unit = 'day' | |
798 | time *= 365 | |
799 | unit += 's' | |
800 | delta = datetime.timedelta(**{unit: time}) | |
801 | return today + delta | |
bd558525 JMF |
802 | return datetime.datetime.strptime(date_str, "%Y%m%d").date() |
803 | ||
804 | class DateRange(object): | |
805 | """Represents a time interval between two dates""" | |
806 | def __init__(self, start=None, end=None): | |
807 | """start and end must be strings in the format accepted by date""" | |
808 | if start is not None: | |
809 | self.start = date_from_str(start) | |
810 | else: | |
811 | self.start = datetime.datetime.min.date() | |
812 | if end is not None: | |
813 | self.end = date_from_str(end) | |
814 | else: | |
815 | self.end = datetime.datetime.max.date() | |
37254abc | 816 | if self.start > self.end: |
bd558525 JMF |
817 | raise ValueError('Date range: "%s" , the start date must be before the end date' % self) |
818 | @classmethod | |
819 | def day(cls, day): | |
820 | """Returns a range that only contains the given day""" | |
821 | return cls(day,day) | |
822 | def __contains__(self, date): | |
823 | """Check if the date is in the range""" | |
37254abc JMF |
824 | if not isinstance(date, datetime.date): |
825 | date = date_from_str(date) | |
826 | return self.start <= date <= self.end | |
bd558525 JMF |
827 | def __str__(self): |
828 | return '%s - %s' % ( self.start.isoformat(), self.end.isoformat()) | |
c496ca96 PH |
829 | |
830 | ||
831 | def platform_name(): | |
832 | """ Returns the platform name as a compat_str """ | |
833 | res = platform.platform() | |
834 | if isinstance(res, bytes): | |
835 | res = res.decode(preferredencoding()) | |
836 | ||
837 | assert isinstance(res, compat_str) | |
838 | return res | |
c257baff PH |
839 | |
840 | ||
7459e3a2 PH |
841 | def write_string(s, out=None): |
842 | if out is None: | |
843 | out = sys.stderr | |
844 | assert type(s) == type(u'') | |
845 | ||
846 | if ('b' in getattr(out, 'mode', '') or | |
847 | sys.version_info[0] < 3): # Python 2 lies about mode of sys.stderr | |
848 | s = s.encode(preferredencoding(), 'ignore') | |
849 | out.write(s) | |
850 | out.flush() | |
851 | ||
852 | ||
48ea9cea PH |
853 | def bytes_to_intlist(bs): |
854 | if not bs: | |
855 | return [] | |
856 | if isinstance(bs[0], int): # Python 3 | |
857 | return list(bs) | |
858 | else: | |
859 | return [ord(c) for c in bs] | |
860 | ||
c257baff | 861 | |
cba892fa | 862 | def intlist_to_bytes(xs): |
863 | if not xs: | |
864 | return b'' | |
865 | if isinstance(chr(0), bytes): # Python 2 | |
866 | return ''.join([chr(x) for x in xs]) | |
867 | else: | |
868 | return bytes(xs) | |
c38b1e77 PH |
869 | |
870 | ||
871 | def get_cachedir(params={}): | |
872 | cache_root = os.environ.get('XDG_CACHE_HOME', | |
873 | os.path.expanduser('~/.cache')) | |
874 | return params.get('cachedir', os.path.join(cache_root, 'youtube-dl')) | |
c1c9a79c PH |
875 | |
876 | ||
877 | # Cross-platform file locking | |
878 | if sys.platform == 'win32': | |
879 | import ctypes.wintypes | |
880 | import msvcrt | |
881 | ||
882 | class OVERLAPPED(ctypes.Structure): | |
883 | _fields_ = [ | |
884 | ('Internal', ctypes.wintypes.LPVOID), | |
885 | ('InternalHigh', ctypes.wintypes.LPVOID), | |
886 | ('Offset', ctypes.wintypes.DWORD), | |
887 | ('OffsetHigh', ctypes.wintypes.DWORD), | |
888 | ('hEvent', ctypes.wintypes.HANDLE), | |
889 | ] | |
890 | ||
891 | kernel32 = ctypes.windll.kernel32 | |
892 | LockFileEx = kernel32.LockFileEx | |
893 | LockFileEx.argtypes = [ | |
894 | ctypes.wintypes.HANDLE, # hFile | |
895 | ctypes.wintypes.DWORD, # dwFlags | |
896 | ctypes.wintypes.DWORD, # dwReserved | |
897 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow | |
898 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh | |
899 | ctypes.POINTER(OVERLAPPED) # Overlapped | |
900 | ] | |
901 | LockFileEx.restype = ctypes.wintypes.BOOL | |
902 | UnlockFileEx = kernel32.UnlockFileEx | |
903 | UnlockFileEx.argtypes = [ | |
904 | ctypes.wintypes.HANDLE, # hFile | |
905 | ctypes.wintypes.DWORD, # dwReserved | |
906 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow | |
907 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh | |
908 | ctypes.POINTER(OVERLAPPED) # Overlapped | |
909 | ] | |
910 | UnlockFileEx.restype = ctypes.wintypes.BOOL | |
911 | whole_low = 0xffffffff | |
912 | whole_high = 0x7fffffff | |
913 | ||
914 | def _lock_file(f, exclusive): | |
915 | overlapped = OVERLAPPED() | |
916 | overlapped.Offset = 0 | |
917 | overlapped.OffsetHigh = 0 | |
918 | overlapped.hEvent = 0 | |
919 | f._lock_file_overlapped_p = ctypes.pointer(overlapped) | |
920 | handle = msvcrt.get_osfhandle(f.fileno()) | |
921 | if not LockFileEx(handle, 0x2 if exclusive else 0x0, 0, | |
922 | whole_low, whole_high, f._lock_file_overlapped_p): | |
923 | raise OSError('Locking file failed: %r' % ctypes.FormatError()) | |
924 | ||
925 | def _unlock_file(f): | |
926 | assert f._lock_file_overlapped_p | |
927 | handle = msvcrt.get_osfhandle(f.fileno()) | |
928 | if not UnlockFileEx(handle, 0, | |
929 | whole_low, whole_high, f._lock_file_overlapped_p): | |
930 | raise OSError('Unlocking file failed: %r' % ctypes.FormatError()) | |
931 | ||
932 | else: | |
933 | import fcntl | |
934 | ||
935 | def _lock_file(f, exclusive): | |
936 | fcntl.lockf(f, fcntl.LOCK_EX if exclusive else fcntl.LOCK_SH) | |
937 | ||
938 | def _unlock_file(f): | |
939 | fcntl.lockf(f, fcntl.LOCK_UN) | |
940 | ||
941 | ||
942 | class locked_file(object): | |
943 | def __init__(self, filename, mode, encoding=None): | |
944 | assert mode in ['r', 'a', 'w'] | |
945 | self.f = io.open(filename, mode, encoding=encoding) | |
946 | self.mode = mode | |
947 | ||
948 | def __enter__(self): | |
949 | exclusive = self.mode != 'r' | |
950 | try: | |
951 | _lock_file(self.f, exclusive) | |
952 | except IOError: | |
953 | self.f.close() | |
954 | raise | |
955 | return self | |
956 | ||
957 | def __exit__(self, etype, value, traceback): | |
958 | try: | |
959 | _unlock_file(self.f) | |
960 | finally: | |
961 | self.f.close() | |
962 | ||
963 | def __iter__(self): | |
964 | return iter(self.f) | |
965 | ||
966 | def write(self, *args): | |
967 | return self.f.write(*args) | |
968 | ||
969 | def read(self, *args): | |
970 | return self.f.read(*args) | |
4eb7f1d1 JMF |
971 | |
972 | ||
973 | def shell_quote(args): | |
a6a173c2 JMF |
974 | quoted_args = [] |
975 | encoding = sys.getfilesystemencoding() | |
976 | if encoding is None: | |
977 | encoding = 'utf-8' | |
978 | for a in args: | |
979 | if isinstance(a, bytes): | |
980 | # We may get a filename encoded with 'encodeFilename' | |
981 | a = a.decode(encoding) | |
982 | quoted_args.append(pipes.quote(a)) | |
983 | return u' '.join(quoted_args) | |
9d4660ca PH |
984 | |
985 | ||
f4d96df0 PH |
986 | def takewhile_inclusive(pred, seq): |
987 | """ Like itertools.takewhile, but include the latest evaluated element | |
988 | (the first element so that Not pred(e)) """ | |
989 | for e in seq: | |
990 | yield e | |
991 | if not pred(e): | |
992 | return | |
993 | ||
994 | ||
9d4660ca PH |
995 | def smuggle_url(url, data): |
996 | """ Pass additional data in a URL for internal use. """ | |
997 | ||
998 | sdata = compat_urllib_parse.urlencode( | |
999 | {u'__youtubedl_smuggle': json.dumps(data)}) | |
1000 | return url + u'#' + sdata | |
1001 | ||
1002 | ||
1003 | def unsmuggle_url(smug_url): | |
1004 | if not '#__youtubedl_smuggle' in smug_url: | |
1005 | return smug_url, None | |
1006 | url, _, sdata = smug_url.rpartition(u'#') | |
1007 | jsond = compat_parse_qs(sdata)[u'__youtubedl_smuggle'][0] | |
1008 | data = json.loads(jsond) | |
1009 | return url, data | |
02dbf93f PH |
1010 | |
1011 | ||
1012 | def parse_xml_doc(s): | |
1013 | assert isinstance(s, type(u'')) | |
1014 | return xml.etree.ElementTree.fromstring(s.encode('utf-8')) | |
1015 | ||
1016 | ||
1017 | def format_bytes(bytes): | |
1018 | if bytes is None: | |
1019 | return u'N/A' | |
1020 | if type(bytes) is str: | |
1021 | bytes = float(bytes) | |
1022 | if bytes == 0.0: | |
1023 | exponent = 0 | |
1024 | else: | |
1025 | exponent = int(math.log(bytes, 1024.0)) | |
1026 | suffix = [u'B', u'KiB', u'MiB', u'GiB', u'TiB', u'PiB', u'EiB', u'ZiB', u'YiB'][exponent] | |
1027 | converted = float(bytes) / float(1024 ** exponent) | |
1028 | return u'%.2f%s' % (converted, suffix) |