]>
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 FV |
10 | import locale |
11 | import os | |
4eb7f1d1 | 12 | import pipes |
c496ca96 | 13 | import platform |
d77c3dfd | 14 | import re |
13ebea79 | 15 | import ssl |
c496ca96 | 16 | import socket |
d77c3dfd | 17 | import sys |
01951dda | 18 | import traceback |
d77c3dfd | 19 | import zlib |
d77c3dfd | 20 | |
01ba00ca | 21 | try: |
59ae15a5 | 22 | import urllib.request as compat_urllib_request |
01ba00ca | 23 | except ImportError: # Python 2 |
59ae15a5 | 24 | import urllib2 as compat_urllib_request |
01ba00ca PH |
25 | |
26 | try: | |
59ae15a5 | 27 | import urllib.error as compat_urllib_error |
01ba00ca | 28 | except ImportError: # Python 2 |
59ae15a5 | 29 | import urllib2 as compat_urllib_error |
01ba00ca PH |
30 | |
31 | try: | |
59ae15a5 | 32 | import urllib.parse as compat_urllib_parse |
01ba00ca | 33 | except ImportError: # Python 2 |
59ae15a5 | 34 | import urllib as compat_urllib_parse |
01ba00ca | 35 | |
799c0763 PH |
36 | try: |
37 | from urllib.parse import urlparse as compat_urllib_parse_urlparse | |
38 | except ImportError: # Python 2 | |
39 | from urlparse import urlparse as compat_urllib_parse_urlparse | |
40 | ||
6543f0dc JMF |
41 | try: |
42 | import urllib.parse as compat_urlparse | |
43 | except ImportError: # Python 2 | |
44 | import urlparse as compat_urlparse | |
45 | ||
01ba00ca | 46 | try: |
59ae15a5 | 47 | import http.cookiejar as compat_cookiejar |
01ba00ca | 48 | except ImportError: # Python 2 |
59ae15a5 | 49 | import cookielib as compat_cookiejar |
01ba00ca | 50 | |
3e669f36 | 51 | try: |
59ae15a5 | 52 | import html.entities as compat_html_entities |
9f37a959 | 53 | except ImportError: # Python 2 |
59ae15a5 | 54 | import htmlentitydefs as compat_html_entities |
3e669f36 | 55 | |
a8156c1d | 56 | try: |
59ae15a5 | 57 | import html.parser as compat_html_parser |
9f37a959 | 58 | except ImportError: # Python 2 |
59ae15a5 | 59 | import HTMLParser as compat_html_parser |
a8156c1d | 60 | |
348d0a7a | 61 | try: |
59ae15a5 | 62 | import http.client as compat_http_client |
9f37a959 | 63 | except ImportError: # Python 2 |
59ae15a5 | 64 | import httplib as compat_http_client |
348d0a7a | 65 | |
2eabb802 | 66 | try: |
0e283428 | 67 | from urllib.error import HTTPError as compat_HTTPError |
2eabb802 PH |
68 | except ImportError: # Python 2 |
69 | from urllib2 import HTTPError as compat_HTTPError | |
70 | ||
e0df6211 PH |
71 | try: |
72 | from urllib.request import urlretrieve as compat_urlretrieve | |
73 | except ImportError: # Python 2 | |
74 | from urllib import urlretrieve as compat_urlretrieve | |
75 | ||
76 | ||
5910e210 PH |
77 | try: |
78 | from subprocess import DEVNULL | |
79 | compat_subprocess_get_DEVNULL = lambda: DEVNULL | |
80 | except ImportError: | |
81 | compat_subprocess_get_DEVNULL = lambda: open(os.path.devnull, 'w') | |
82 | ||
9f37a959 | 83 | try: |
59ae15a5 | 84 | from urllib.parse import parse_qs as compat_parse_qs |
9f37a959 | 85 | except ImportError: # Python 2 |
59ae15a5 PH |
86 | # HACK: The following is the correct parse_qs implementation from cpython 3's stdlib. |
87 | # Python 2's version is apparently totally broken | |
88 | def _unquote(string, encoding='utf-8', errors='replace'): | |
89 | if string == '': | |
90 | return string | |
91 | res = string.split('%') | |
92 | if len(res) == 1: | |
93 | return string | |
94 | if encoding is None: | |
95 | encoding = 'utf-8' | |
96 | if errors is None: | |
97 | errors = 'replace' | |
98 | # pct_sequence: contiguous sequence of percent-encoded bytes, decoded | |
99 | pct_sequence = b'' | |
100 | string = res[0] | |
101 | for item in res[1:]: | |
102 | try: | |
103 | if not item: | |
104 | raise ValueError | |
105 | pct_sequence += item[:2].decode('hex') | |
106 | rest = item[2:] | |
107 | if not rest: | |
108 | # This segment was just a single percent-encoded character. | |
109 | # May be part of a sequence of code units, so delay decoding. | |
110 | # (Stored in pct_sequence). | |
111 | continue | |
112 | except ValueError: | |
113 | rest = '%' + item | |
114 | # Encountered non-percent-encoded characters. Flush the current | |
115 | # pct_sequence. | |
116 | string += pct_sequence.decode(encoding, errors) + rest | |
117 | pct_sequence = b'' | |
118 | if pct_sequence: | |
119 | # Flush the final pct_sequence | |
120 | string += pct_sequence.decode(encoding, errors) | |
121 | return string | |
122 | ||
123 | def _parse_qsl(qs, keep_blank_values=False, strict_parsing=False, | |
124 | encoding='utf-8', errors='replace'): | |
125 | qs, _coerce_result = qs, unicode | |
126 | pairs = [s2 for s1 in qs.split('&') for s2 in s1.split(';')] | |
127 | r = [] | |
128 | for name_value in pairs: | |
129 | if not name_value and not strict_parsing: | |
130 | continue | |
131 | nv = name_value.split('=', 1) | |
132 | if len(nv) != 2: | |
133 | if strict_parsing: | |
134 | raise ValueError("bad query field: %r" % (name_value,)) | |
135 | # Handle case of a control-name with no equal sign | |
136 | if keep_blank_values: | |
137 | nv.append('') | |
138 | else: | |
139 | continue | |
140 | if len(nv[1]) or keep_blank_values: | |
141 | name = nv[0].replace('+', ' ') | |
142 | name = _unquote(name, encoding=encoding, errors=errors) | |
143 | name = _coerce_result(name) | |
144 | value = nv[1].replace('+', ' ') | |
145 | value = _unquote(value, encoding=encoding, errors=errors) | |
146 | value = _coerce_result(value) | |
147 | r.append((name, value)) | |
148 | return r | |
149 | ||
150 | def compat_parse_qs(qs, keep_blank_values=False, strict_parsing=False, | |
151 | encoding='utf-8', errors='replace'): | |
152 | parsed_result = {} | |
153 | pairs = _parse_qsl(qs, keep_blank_values, strict_parsing, | |
154 | encoding=encoding, errors=errors) | |
155 | for name, value in pairs: | |
156 | if name in parsed_result: | |
157 | parsed_result[name].append(value) | |
158 | else: | |
159 | parsed_result[name] = [value] | |
160 | return parsed_result | |
348d0a7a | 161 | |
3e669f36 | 162 | try: |
59ae15a5 | 163 | compat_str = unicode # Python 2 |
3e669f36 | 164 | except NameError: |
59ae15a5 | 165 | compat_str = str |
3e669f36 PH |
166 | |
167 | try: | |
59ae15a5 | 168 | compat_chr = unichr # Python 2 |
3e669f36 | 169 | except NameError: |
59ae15a5 | 170 | compat_chr = chr |
3e669f36 | 171 | |
b31756c1 FV |
172 | def compat_ord(c): |
173 | if type(c) is int: return c | |
174 | else: return ord(c) | |
175 | ||
468e2e92 FV |
176 | # This is not clearly defined otherwise |
177 | compiled_regex_type = type(re.compile('')) | |
178 | ||
3e669f36 | 179 | std_headers = { |
ae8f7871 | 180 | 'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20100101 Firefox/10.0 (Chrome)', |
59ae15a5 PH |
181 | 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7', |
182 | 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8', | |
183 | 'Accept-Encoding': 'gzip, deflate', | |
184 | 'Accept-Language': 'en-us,en;q=0.5', | |
3e669f36 | 185 | } |
f427df17 | 186 | |
d77c3dfd | 187 | def preferredencoding(): |
59ae15a5 | 188 | """Get preferred encoding. |
d77c3dfd | 189 | |
59ae15a5 PH |
190 | Returns the best encoding scheme for the system, based on |
191 | locale.getpreferredencoding() and some further tweaks. | |
192 | """ | |
193 | try: | |
194 | pref = locale.getpreferredencoding() | |
195 | u'TEST'.encode(pref) | |
196 | except: | |
197 | pref = 'UTF-8' | |
bae611f2 | 198 | |
59ae15a5 | 199 | return pref |
d77c3dfd | 200 | |
8cd10ac4 | 201 | if sys.version_info < (3,0): |
59ae15a5 PH |
202 | def compat_print(s): |
203 | print(s.encode(preferredencoding(), 'xmlcharrefreplace')) | |
8cd10ac4 | 204 | else: |
59ae15a5 PH |
205 | def compat_print(s): |
206 | assert type(s) == type(u'') | |
207 | print(s) | |
d77c3dfd | 208 | |
f4bfd65f PH |
209 | # In Python 2.x, json.dump expects a bytestream. |
210 | # In Python 3.x, it writes to a character stream | |
211 | if sys.version_info < (3,0): | |
212 | def write_json_file(obj, fn): | |
213 | with open(fn, 'wb') as f: | |
214 | json.dump(obj, f) | |
215 | else: | |
216 | def write_json_file(obj, fn): | |
217 | with open(fn, 'w', encoding='utf-8') as f: | |
218 | json.dump(obj, f) | |
219 | ||
59ae56fa PH |
220 | if sys.version_info >= (2,7): |
221 | def find_xpath_attr(node, xpath, key, val): | |
222 | """ Find the xpath xpath[@key=val] """ | |
5de3ece2 | 223 | assert re.match(r'^[a-zA-Z]+$', key) |
54543467 | 224 | assert re.match(r'^[a-zA-Z0-9@\s]*$', val) |
59ae56fa PH |
225 | expr = xpath + u"[@%s='%s']" % (key, val) |
226 | return node.find(expr) | |
227 | else: | |
228 | def find_xpath_attr(node, xpath, key, val): | |
229 | for f in node.findall(xpath): | |
230 | if f.attrib.get(key) == val: | |
231 | return f | |
232 | return None | |
233 | ||
d7e66d39 JMF |
234 | # On python2.6 the xml.etree.ElementTree.Element methods don't support |
235 | # the namespace parameter | |
236 | def xpath_with_ns(path, ns_map): | |
237 | components = [c.split(':') for c in path.split('/')] | |
238 | replaced = [] | |
239 | for c in components: | |
240 | if len(c) == 1: | |
241 | replaced.append(c[0]) | |
242 | else: | |
243 | ns, tag = c | |
244 | replaced.append('{%s}%s' % (ns_map[ns], tag)) | |
245 | return '/'.join(replaced) | |
246 | ||
d77c3dfd | 247 | def htmlentity_transform(matchobj): |
59ae15a5 PH |
248 | """Transforms an HTML entity to a character. |
249 | ||
250 | This function receives a match object and is intended to be used with | |
251 | the re.sub() function. | |
252 | """ | |
253 | entity = matchobj.group(1) | |
254 | ||
255 | # Known non-numeric HTML entity | |
256 | if entity in compat_html_entities.name2codepoint: | |
257 | return compat_chr(compat_html_entities.name2codepoint[entity]) | |
258 | ||
259 | mobj = re.match(u'(?u)#(x?\\d+)', entity) | |
260 | if mobj is not None: | |
261 | numstr = mobj.group(1) | |
262 | if numstr.startswith(u'x'): | |
263 | base = 16 | |
264 | numstr = u'0%s' % numstr | |
265 | else: | |
266 | base = 10 | |
267 | return compat_chr(int(numstr, base)) | |
268 | ||
269 | # Unknown entity in name, return its literal representation | |
270 | return (u'&%s;' % entity) | |
d77c3dfd | 271 | |
a8156c1d | 272 | 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 |
273 | class BaseHTMLParser(compat_html_parser.HTMLParser): |
274 | def __init(self): | |
275 | compat_html_parser.HTMLParser.__init__(self) | |
276 | self.html = None | |
277 | ||
278 | def loads(self, html): | |
279 | self.html = html | |
280 | self.feed(html) | |
281 | self.close() | |
282 | ||
283 | class AttrParser(BaseHTMLParser): | |
43e8fafd ND |
284 | """Modified HTMLParser that isolates a tag with the specified attribute""" |
285 | def __init__(self, attribute, value): | |
286 | self.attribute = attribute | |
287 | self.value = value | |
59ae15a5 PH |
288 | self.result = None |
289 | self.started = False | |
290 | self.depth = {} | |
59ae15a5 PH |
291 | self.watch_startpos = False |
292 | self.error_count = 0 | |
a921f407 | 293 | BaseHTMLParser.__init__(self) |
59ae15a5 PH |
294 | |
295 | def error(self, message): | |
296 | if self.error_count > 10 or self.started: | |
297 | raise compat_html_parser.HTMLParseError(message, self.getpos()) | |
298 | self.rawdata = '\n'.join(self.html.split('\n')[self.getpos()[0]:]) # skip one line | |
299 | self.error_count += 1 | |
300 | self.goahead(1) | |
301 | ||
59ae15a5 PH |
302 | def handle_starttag(self, tag, attrs): |
303 | attrs = dict(attrs) | |
304 | if self.started: | |
305 | self.find_startpos(None) | |
43e8fafd | 306 | if self.attribute in attrs and attrs[self.attribute] == self.value: |
59ae15a5 PH |
307 | self.result = [tag] |
308 | self.started = True | |
309 | self.watch_startpos = True | |
310 | if self.started: | |
311 | if not tag in self.depth: self.depth[tag] = 0 | |
312 | self.depth[tag] += 1 | |
313 | ||
314 | def handle_endtag(self, tag): | |
315 | if self.started: | |
316 | if tag in self.depth: self.depth[tag] -= 1 | |
317 | if self.depth[self.result[0]] == 0: | |
318 | self.started = False | |
319 | self.result.append(self.getpos()) | |
320 | ||
321 | def find_startpos(self, x): | |
322 | """Needed to put the start position of the result (self.result[1]) | |
323 | after the opening tag with the requested id""" | |
324 | if self.watch_startpos: | |
325 | self.watch_startpos = False | |
326 | self.result.append(self.getpos()) | |
327 | handle_entityref = handle_charref = handle_data = handle_comment = \ | |
328 | handle_decl = handle_pi = unknown_decl = find_startpos | |
329 | ||
330 | def get_result(self): | |
331 | if self.result is None: | |
332 | return None | |
333 | if len(self.result) != 3: | |
334 | return None | |
335 | lines = self.html.split('\n') | |
336 | lines = lines[self.result[1][0]-1:self.result[2][0]] | |
337 | lines[0] = lines[0][self.result[1][1]:] | |
338 | if len(lines) == 1: | |
339 | lines[-1] = lines[-1][:self.result[2][1]-self.result[1][1]] | |
340 | lines[-1] = lines[-1][:self.result[2][1]] | |
341 | return '\n'.join(lines).strip() | |
3b024e17 PH |
342 | # Hack for https://github.com/rg3/youtube-dl/issues/662 |
343 | if sys.version_info < (2, 7, 3): | |
344 | AttrParser.parse_endtag = (lambda self, i: | |
345 | i + len("</scr'+'ipt>") | |
346 | if self.rawdata[i:].startswith("</scr'+'ipt>") | |
347 | else compat_html_parser.HTMLParser.parse_endtag(self, i)) | |
9e6dd238 FV |
348 | |
349 | def get_element_by_id(id, html): | |
43e8fafd ND |
350 | """Return the content of the tag with the specified ID in the passed HTML document""" |
351 | return get_element_by_attribute("id", id, html) | |
352 | ||
353 | def get_element_by_attribute(attribute, value, html): | |
354 | """Return the content of the tag with the specified attribute in the passed HTML document""" | |
355 | parser = AttrParser(attribute, value) | |
59ae15a5 PH |
356 | try: |
357 | parser.loads(html) | |
358 | except compat_html_parser.HTMLParseError: | |
359 | pass | |
360 | return parser.get_result() | |
9e6dd238 | 361 | |
a921f407 JMF |
362 | class MetaParser(BaseHTMLParser): |
363 | """ | |
364 | Modified HTMLParser that isolates a meta tag with the specified name | |
365 | attribute. | |
366 | """ | |
367 | def __init__(self, name): | |
368 | BaseHTMLParser.__init__(self) | |
369 | self.name = name | |
370 | self.content = None | |
371 | self.result = None | |
372 | ||
373 | def handle_starttag(self, tag, attrs): | |
374 | if tag != 'meta': | |
375 | return | |
376 | attrs = dict(attrs) | |
377 | if attrs.get('name') == self.name: | |
378 | self.result = attrs.get('content') | |
379 | ||
380 | def get_result(self): | |
381 | return self.result | |
382 | ||
383 | def get_meta_content(name, html): | |
384 | """ | |
385 | Return the content attribute from the meta tag with the given name attribute. | |
386 | """ | |
387 | parser = MetaParser(name) | |
388 | try: | |
389 | parser.loads(html) | |
390 | except compat_html_parser.HTMLParseError: | |
391 | pass | |
392 | return parser.get_result() | |
393 | ||
9e6dd238 FV |
394 | |
395 | def clean_html(html): | |
59ae15a5 PH |
396 | """Clean an HTML snippet into a readable string""" |
397 | # Newline vs <br /> | |
398 | html = html.replace('\n', ' ') | |
6b3aef80 FV |
399 | html = re.sub(r'\s*<\s*br\s*/?\s*>\s*', '\n', html) |
400 | html = re.sub(r'<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html) | |
59ae15a5 PH |
401 | # Strip html tags |
402 | html = re.sub('<.*?>', '', html) | |
403 | # Replace html entities | |
404 | html = unescapeHTML(html) | |
7decf895 | 405 | return html.strip() |
9e6dd238 FV |
406 | |
407 | ||
d77c3dfd | 408 | def sanitize_open(filename, open_mode): |
59ae15a5 PH |
409 | """Try to open the given filename, and slightly tweak it if this fails. |
410 | ||
411 | Attempts to open the given filename. If this fails, it tries to change | |
412 | the filename slightly, step by step, until it's either able to open it | |
413 | or it fails and raises a final exception, like the standard open() | |
414 | function. | |
415 | ||
416 | It returns the tuple (stream, definitive_file_name). | |
417 | """ | |
418 | try: | |
419 | if filename == u'-': | |
420 | if sys.platform == 'win32': | |
421 | import msvcrt | |
422 | msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY) | |
898280a0 | 423 | return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename) |
59ae15a5 PH |
424 | stream = open(encodeFilename(filename), open_mode) |
425 | return (stream, filename) | |
426 | except (IOError, OSError) as err: | |
f45c185f PH |
427 | if err.errno in (errno.EACCES,): |
428 | raise | |
59ae15a5 | 429 | |
f45c185f PH |
430 | # In case of error, try to remove win32 forbidden chars |
431 | alt_filename = os.path.join( | |
432 | re.sub(u'[/<>:"\\|\\\\?\\*]', u'#', path_part) | |
433 | for path_part in os.path.split(filename) | |
434 | ) | |
435 | if alt_filename == filename: | |
436 | raise | |
437 | else: | |
438 | # An exception here should be caught in the caller | |
439 | stream = open(encodeFilename(filename), open_mode) | |
440 | return (stream, alt_filename) | |
d77c3dfd FV |
441 | |
442 | ||
443 | def timeconvert(timestr): | |
59ae15a5 PH |
444 | """Convert RFC 2822 defined time string into system timestamp""" |
445 | timestamp = None | |
446 | timetuple = email.utils.parsedate_tz(timestr) | |
447 | if timetuple is not None: | |
448 | timestamp = email.utils.mktime_tz(timetuple) | |
449 | return timestamp | |
1c469a94 | 450 | |
796173d0 | 451 | def sanitize_filename(s, restricted=False, is_id=False): |
59ae15a5 PH |
452 | """Sanitizes a string so it could be used as part of a filename. |
453 | If restricted is set, use a stricter subset of allowed characters. | |
796173d0 | 454 | Set is_id if this is not an arbitrary string, but an ID that should be kept if possible |
59ae15a5 PH |
455 | """ |
456 | def replace_insane(char): | |
457 | if char == '?' or ord(char) < 32 or ord(char) == 127: | |
458 | return '' | |
459 | elif char == '"': | |
460 | return '' if restricted else '\'' | |
461 | elif char == ':': | |
462 | return '_-' if restricted else ' -' | |
463 | elif char in '\\/|*<>': | |
464 | return '_' | |
627dcfff | 465 | if restricted and (char in '!&\'()[]{}$;`^,#' or char.isspace()): |
59ae15a5 PH |
466 | return '_' |
467 | if restricted and ord(char) > 127: | |
468 | return '_' | |
469 | return char | |
470 | ||
471 | result = u''.join(map(replace_insane, s)) | |
796173d0 PH |
472 | if not is_id: |
473 | while '__' in result: | |
474 | result = result.replace('__', '_') | |
475 | result = result.strip('_') | |
476 | # Common case of "Foreign band name - English song title" | |
477 | if restricted and result.startswith('-_'): | |
478 | result = result[2:] | |
479 | if not result: | |
480 | result = '_' | |
59ae15a5 | 481 | return result |
d77c3dfd FV |
482 | |
483 | def orderedSet(iterable): | |
59ae15a5 PH |
484 | """ Remove all duplicates from the input iterable """ |
485 | res = [] | |
486 | for el in iterable: | |
487 | if el not in res: | |
488 | res.append(el) | |
489 | return res | |
d77c3dfd FV |
490 | |
491 | def unescapeHTML(s): | |
59ae15a5 PH |
492 | """ |
493 | @param s a string | |
494 | """ | |
495 | assert type(s) == type(u'') | |
d77c3dfd | 496 | |
59ae15a5 PH |
497 | result = re.sub(u'(?u)&(.+?);', htmlentity_transform, s) |
498 | return result | |
d77c3dfd FV |
499 | |
500 | def encodeFilename(s): | |
59ae15a5 PH |
501 | """ |
502 | @param s The name of the file | |
503 | """ | |
d77c3dfd | 504 | |
59ae15a5 | 505 | assert type(s) == type(u'') |
d77c3dfd | 506 | |
59ae15a5 PH |
507 | # Python 3 has a Unicode API |
508 | if sys.version_info >= (3, 0): | |
509 | return s | |
0f00efed | 510 | |
59ae15a5 PH |
511 | if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5: |
512 | # Pass u'' directly to use Unicode APIs on Windows 2000 and up | |
513 | # (Detecting Windows NT 4 is tricky because 'major >= 4' would | |
514 | # match Windows 9x series as well. Besides, NT 4 is obsolete.) | |
515 | return s | |
516 | else: | |
6df40dcb PH |
517 | encoding = sys.getfilesystemencoding() |
518 | if encoding is None: | |
519 | encoding = 'utf-8' | |
520 | return s.encode(encoding, 'ignore') | |
d77c3dfd | 521 | |
8271226a PH |
522 | def decodeOption(optval): |
523 | if optval is None: | |
524 | return optval | |
525 | if isinstance(optval, bytes): | |
526 | optval = optval.decode(preferredencoding()) | |
527 | ||
528 | assert isinstance(optval, compat_str) | |
529 | return optval | |
1c256f70 | 530 | |
4539dd30 PH |
531 | def formatSeconds(secs): |
532 | if secs > 3600: | |
533 | return '%d:%02d:%02d' % (secs // 3600, (secs % 3600) // 60, secs % 60) | |
534 | elif secs > 60: | |
535 | return '%d:%02d' % (secs // 60, secs % 60) | |
536 | else: | |
537 | return '%d' % secs | |
538 | ||
13ebea79 | 539 | |
ea6d901e | 540 | def make_HTTPS_handler(opts): |
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) | |
555 | except ssl.SSLError as e: | |
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 | |
567 | if opts.no_check_certificate | |
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 |