]>
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 |
c496ca96 | 15 | import socket |
d77c3dfd | 16 | import sys |
01951dda | 17 | import traceback |
d77c3dfd | 18 | import zlib |
d77c3dfd | 19 | |
01ba00ca | 20 | try: |
59ae15a5 | 21 | import urllib.request as compat_urllib_request |
01ba00ca | 22 | except ImportError: # Python 2 |
59ae15a5 | 23 | import urllib2 as compat_urllib_request |
01ba00ca PH |
24 | |
25 | try: | |
59ae15a5 | 26 | import urllib.error as compat_urllib_error |
01ba00ca | 27 | except ImportError: # Python 2 |
59ae15a5 | 28 | import urllib2 as compat_urllib_error |
01ba00ca PH |
29 | |
30 | try: | |
59ae15a5 | 31 | import urllib.parse as compat_urllib_parse |
01ba00ca | 32 | except ImportError: # Python 2 |
59ae15a5 | 33 | import urllib as compat_urllib_parse |
01ba00ca | 34 | |
799c0763 PH |
35 | try: |
36 | from urllib.parse import urlparse as compat_urllib_parse_urlparse | |
37 | except ImportError: # Python 2 | |
38 | from urlparse import urlparse as compat_urllib_parse_urlparse | |
39 | ||
6543f0dc JMF |
40 | try: |
41 | import urllib.parse as compat_urlparse | |
42 | except ImportError: # Python 2 | |
43 | import urlparse as compat_urlparse | |
44 | ||
01ba00ca | 45 | try: |
59ae15a5 | 46 | import http.cookiejar as compat_cookiejar |
01ba00ca | 47 | except ImportError: # Python 2 |
59ae15a5 | 48 | import cookielib as compat_cookiejar |
01ba00ca | 49 | |
3e669f36 | 50 | try: |
59ae15a5 | 51 | import html.entities as compat_html_entities |
9f37a959 | 52 | except ImportError: # Python 2 |
59ae15a5 | 53 | import htmlentitydefs as compat_html_entities |
3e669f36 | 54 | |
a8156c1d | 55 | try: |
59ae15a5 | 56 | import html.parser as compat_html_parser |
9f37a959 | 57 | except ImportError: # Python 2 |
59ae15a5 | 58 | import HTMLParser as compat_html_parser |
a8156c1d | 59 | |
348d0a7a | 60 | try: |
59ae15a5 | 61 | import http.client as compat_http_client |
9f37a959 | 62 | except ImportError: # Python 2 |
59ae15a5 | 63 | import httplib as compat_http_client |
348d0a7a | 64 | |
2eabb802 | 65 | try: |
0e283428 | 66 | from urllib.error import HTTPError as compat_HTTPError |
2eabb802 PH |
67 | except ImportError: # Python 2 |
68 | from urllib2 import HTTPError as compat_HTTPError | |
69 | ||
e0df6211 PH |
70 | try: |
71 | from urllib.request import urlretrieve as compat_urlretrieve | |
72 | except ImportError: # Python 2 | |
73 | from urllib import urlretrieve as compat_urlretrieve | |
74 | ||
75 | ||
5910e210 PH |
76 | try: |
77 | from subprocess import DEVNULL | |
78 | compat_subprocess_get_DEVNULL = lambda: DEVNULL | |
79 | except ImportError: | |
80 | compat_subprocess_get_DEVNULL = lambda: open(os.path.devnull, 'w') | |
81 | ||
9f37a959 | 82 | try: |
59ae15a5 | 83 | from urllib.parse import parse_qs as compat_parse_qs |
9f37a959 | 84 | except ImportError: # Python 2 |
59ae15a5 PH |
85 | # HACK: The following is the correct parse_qs implementation from cpython 3's stdlib. |
86 | # Python 2's version is apparently totally broken | |
87 | def _unquote(string, encoding='utf-8', errors='replace'): | |
88 | if string == '': | |
89 | return string | |
90 | res = string.split('%') | |
91 | if len(res) == 1: | |
92 | return string | |
93 | if encoding is None: | |
94 | encoding = 'utf-8' | |
95 | if errors is None: | |
96 | errors = 'replace' | |
97 | # pct_sequence: contiguous sequence of percent-encoded bytes, decoded | |
98 | pct_sequence = b'' | |
99 | string = res[0] | |
100 | for item in res[1:]: | |
101 | try: | |
102 | if not item: | |
103 | raise ValueError | |
104 | pct_sequence += item[:2].decode('hex') | |
105 | rest = item[2:] | |
106 | if not rest: | |
107 | # This segment was just a single percent-encoded character. | |
108 | # May be part of a sequence of code units, so delay decoding. | |
109 | # (Stored in pct_sequence). | |
110 | continue | |
111 | except ValueError: | |
112 | rest = '%' + item | |
113 | # Encountered non-percent-encoded characters. Flush the current | |
114 | # pct_sequence. | |
115 | string += pct_sequence.decode(encoding, errors) + rest | |
116 | pct_sequence = b'' | |
117 | if pct_sequence: | |
118 | # Flush the final pct_sequence | |
119 | string += pct_sequence.decode(encoding, errors) | |
120 | return string | |
121 | ||
122 | def _parse_qsl(qs, keep_blank_values=False, strict_parsing=False, | |
123 | encoding='utf-8', errors='replace'): | |
124 | qs, _coerce_result = qs, unicode | |
125 | pairs = [s2 for s1 in qs.split('&') for s2 in s1.split(';')] | |
126 | r = [] | |
127 | for name_value in pairs: | |
128 | if not name_value and not strict_parsing: | |
129 | continue | |
130 | nv = name_value.split('=', 1) | |
131 | if len(nv) != 2: | |
132 | if strict_parsing: | |
133 | raise ValueError("bad query field: %r" % (name_value,)) | |
134 | # Handle case of a control-name with no equal sign | |
135 | if keep_blank_values: | |
136 | nv.append('') | |
137 | else: | |
138 | continue | |
139 | if len(nv[1]) or keep_blank_values: | |
140 | name = nv[0].replace('+', ' ') | |
141 | name = _unquote(name, encoding=encoding, errors=errors) | |
142 | name = _coerce_result(name) | |
143 | value = nv[1].replace('+', ' ') | |
144 | value = _unquote(value, encoding=encoding, errors=errors) | |
145 | value = _coerce_result(value) | |
146 | r.append((name, value)) | |
147 | return r | |
148 | ||
149 | def compat_parse_qs(qs, keep_blank_values=False, strict_parsing=False, | |
150 | encoding='utf-8', errors='replace'): | |
151 | parsed_result = {} | |
152 | pairs = _parse_qsl(qs, keep_blank_values, strict_parsing, | |
153 | encoding=encoding, errors=errors) | |
154 | for name, value in pairs: | |
155 | if name in parsed_result: | |
156 | parsed_result[name].append(value) | |
157 | else: | |
158 | parsed_result[name] = [value] | |
159 | return parsed_result | |
348d0a7a | 160 | |
3e669f36 | 161 | try: |
59ae15a5 | 162 | compat_str = unicode # Python 2 |
3e669f36 | 163 | except NameError: |
59ae15a5 | 164 | compat_str = str |
3e669f36 PH |
165 | |
166 | try: | |
59ae15a5 | 167 | compat_chr = unichr # Python 2 |
3e669f36 | 168 | except NameError: |
59ae15a5 | 169 | compat_chr = chr |
3e669f36 | 170 | |
b31756c1 FV |
171 | def compat_ord(c): |
172 | if type(c) is int: return c | |
173 | else: return ord(c) | |
174 | ||
468e2e92 FV |
175 | # This is not clearly defined otherwise |
176 | compiled_regex_type = type(re.compile('')) | |
177 | ||
3e669f36 | 178 | std_headers = { |
96b31b65 | 179 | 'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20100101 Firefox/10.0 (Chrome) (iPhone)', |
59ae15a5 PH |
180 | 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7', |
181 | 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8', | |
182 | 'Accept-Encoding': 'gzip, deflate', | |
183 | 'Accept-Language': 'en-us,en;q=0.5', | |
3e669f36 | 184 | } |
f427df17 | 185 | |
d77c3dfd | 186 | def preferredencoding(): |
59ae15a5 | 187 | """Get preferred encoding. |
d77c3dfd | 188 | |
59ae15a5 PH |
189 | Returns the best encoding scheme for the system, based on |
190 | locale.getpreferredencoding() and some further tweaks. | |
191 | """ | |
192 | try: | |
193 | pref = locale.getpreferredencoding() | |
194 | u'TEST'.encode(pref) | |
195 | except: | |
196 | pref = 'UTF-8' | |
bae611f2 | 197 | |
59ae15a5 | 198 | return pref |
d77c3dfd | 199 | |
8cd10ac4 | 200 | if sys.version_info < (3,0): |
59ae15a5 PH |
201 | def compat_print(s): |
202 | print(s.encode(preferredencoding(), 'xmlcharrefreplace')) | |
8cd10ac4 | 203 | else: |
59ae15a5 PH |
204 | def compat_print(s): |
205 | assert type(s) == type(u'') | |
206 | print(s) | |
d77c3dfd | 207 | |
f4bfd65f PH |
208 | # In Python 2.x, json.dump expects a bytestream. |
209 | # In Python 3.x, it writes to a character stream | |
210 | if sys.version_info < (3,0): | |
211 | def write_json_file(obj, fn): | |
212 | with open(fn, 'wb') as f: | |
213 | json.dump(obj, f) | |
214 | else: | |
215 | def write_json_file(obj, fn): | |
216 | with open(fn, 'w', encoding='utf-8') as f: | |
217 | json.dump(obj, f) | |
218 | ||
59ae56fa PH |
219 | if sys.version_info >= (2,7): |
220 | def find_xpath_attr(node, xpath, key, val): | |
221 | """ Find the xpath xpath[@key=val] """ | |
5de3ece2 | 222 | assert re.match(r'^[a-zA-Z]+$', key) |
54543467 | 223 | assert re.match(r'^[a-zA-Z0-9@\s]*$', val) |
59ae56fa PH |
224 | expr = xpath + u"[@%s='%s']" % (key, val) |
225 | return node.find(expr) | |
226 | else: | |
227 | def find_xpath_attr(node, xpath, key, val): | |
228 | for f in node.findall(xpath): | |
229 | if f.attrib.get(key) == val: | |
230 | return f | |
231 | return None | |
232 | ||
d7e66d39 JMF |
233 | # On python2.6 the xml.etree.ElementTree.Element methods don't support |
234 | # the namespace parameter | |
235 | def xpath_with_ns(path, ns_map): | |
236 | components = [c.split(':') for c in path.split('/')] | |
237 | replaced = [] | |
238 | for c in components: | |
239 | if len(c) == 1: | |
240 | replaced.append(c[0]) | |
241 | else: | |
242 | ns, tag = c | |
243 | replaced.append('{%s}%s' % (ns_map[ns], tag)) | |
244 | return '/'.join(replaced) | |
245 | ||
d77c3dfd | 246 | def htmlentity_transform(matchobj): |
59ae15a5 PH |
247 | """Transforms an HTML entity to a character. |
248 | ||
249 | This function receives a match object and is intended to be used with | |
250 | the re.sub() function. | |
251 | """ | |
252 | entity = matchobj.group(1) | |
253 | ||
254 | # Known non-numeric HTML entity | |
255 | if entity in compat_html_entities.name2codepoint: | |
256 | return compat_chr(compat_html_entities.name2codepoint[entity]) | |
257 | ||
258 | mobj = re.match(u'(?u)#(x?\\d+)', entity) | |
259 | if mobj is not None: | |
260 | numstr = mobj.group(1) | |
261 | if numstr.startswith(u'x'): | |
262 | base = 16 | |
263 | numstr = u'0%s' % numstr | |
264 | else: | |
265 | base = 10 | |
266 | return compat_chr(int(numstr, base)) | |
267 | ||
268 | # Unknown entity in name, return its literal representation | |
269 | return (u'&%s;' % entity) | |
d77c3dfd | 270 | |
a8156c1d | 271 | 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 |
272 | class BaseHTMLParser(compat_html_parser.HTMLParser): |
273 | def __init(self): | |
274 | compat_html_parser.HTMLParser.__init__(self) | |
275 | self.html = None | |
276 | ||
277 | def loads(self, html): | |
278 | self.html = html | |
279 | self.feed(html) | |
280 | self.close() | |
281 | ||
282 | class AttrParser(BaseHTMLParser): | |
43e8fafd ND |
283 | """Modified HTMLParser that isolates a tag with the specified attribute""" |
284 | def __init__(self, attribute, value): | |
285 | self.attribute = attribute | |
286 | self.value = value | |
59ae15a5 PH |
287 | self.result = None |
288 | self.started = False | |
289 | self.depth = {} | |
59ae15a5 PH |
290 | self.watch_startpos = False |
291 | self.error_count = 0 | |
a921f407 | 292 | BaseHTMLParser.__init__(self) |
59ae15a5 PH |
293 | |
294 | def error(self, message): | |
295 | if self.error_count > 10 or self.started: | |
296 | raise compat_html_parser.HTMLParseError(message, self.getpos()) | |
297 | self.rawdata = '\n'.join(self.html.split('\n')[self.getpos()[0]:]) # skip one line | |
298 | self.error_count += 1 | |
299 | self.goahead(1) | |
300 | ||
59ae15a5 PH |
301 | def handle_starttag(self, tag, attrs): |
302 | attrs = dict(attrs) | |
303 | if self.started: | |
304 | self.find_startpos(None) | |
43e8fafd | 305 | if self.attribute in attrs and attrs[self.attribute] == self.value: |
59ae15a5 PH |
306 | self.result = [tag] |
307 | self.started = True | |
308 | self.watch_startpos = True | |
309 | if self.started: | |
310 | if not tag in self.depth: self.depth[tag] = 0 | |
311 | self.depth[tag] += 1 | |
312 | ||
313 | def handle_endtag(self, tag): | |
314 | if self.started: | |
315 | if tag in self.depth: self.depth[tag] -= 1 | |
316 | if self.depth[self.result[0]] == 0: | |
317 | self.started = False | |
318 | self.result.append(self.getpos()) | |
319 | ||
320 | def find_startpos(self, x): | |
321 | """Needed to put the start position of the result (self.result[1]) | |
322 | after the opening tag with the requested id""" | |
323 | if self.watch_startpos: | |
324 | self.watch_startpos = False | |
325 | self.result.append(self.getpos()) | |
326 | handle_entityref = handle_charref = handle_data = handle_comment = \ | |
327 | handle_decl = handle_pi = unknown_decl = find_startpos | |
328 | ||
329 | def get_result(self): | |
330 | if self.result is None: | |
331 | return None | |
332 | if len(self.result) != 3: | |
333 | return None | |
334 | lines = self.html.split('\n') | |
335 | lines = lines[self.result[1][0]-1:self.result[2][0]] | |
336 | lines[0] = lines[0][self.result[1][1]:] | |
337 | if len(lines) == 1: | |
338 | lines[-1] = lines[-1][:self.result[2][1]-self.result[1][1]] | |
339 | lines[-1] = lines[-1][:self.result[2][1]] | |
340 | return '\n'.join(lines).strip() | |
3b024e17 PH |
341 | # Hack for https://github.com/rg3/youtube-dl/issues/662 |
342 | if sys.version_info < (2, 7, 3): | |
343 | AttrParser.parse_endtag = (lambda self, i: | |
344 | i + len("</scr'+'ipt>") | |
345 | if self.rawdata[i:].startswith("</scr'+'ipt>") | |
346 | else compat_html_parser.HTMLParser.parse_endtag(self, i)) | |
9e6dd238 FV |
347 | |
348 | def get_element_by_id(id, html): | |
43e8fafd ND |
349 | """Return the content of the tag with the specified ID in the passed HTML document""" |
350 | return get_element_by_attribute("id", id, html) | |
351 | ||
352 | def get_element_by_attribute(attribute, value, html): | |
353 | """Return the content of the tag with the specified attribute in the passed HTML document""" | |
354 | parser = AttrParser(attribute, value) | |
59ae15a5 PH |
355 | try: |
356 | parser.loads(html) | |
357 | except compat_html_parser.HTMLParseError: | |
358 | pass | |
359 | return parser.get_result() | |
9e6dd238 | 360 | |
a921f407 JMF |
361 | class MetaParser(BaseHTMLParser): |
362 | """ | |
363 | Modified HTMLParser that isolates a meta tag with the specified name | |
364 | attribute. | |
365 | """ | |
366 | def __init__(self, name): | |
367 | BaseHTMLParser.__init__(self) | |
368 | self.name = name | |
369 | self.content = None | |
370 | self.result = None | |
371 | ||
372 | def handle_starttag(self, tag, attrs): | |
373 | if tag != 'meta': | |
374 | return | |
375 | attrs = dict(attrs) | |
376 | if attrs.get('name') == self.name: | |
377 | self.result = attrs.get('content') | |
378 | ||
379 | def get_result(self): | |
380 | return self.result | |
381 | ||
382 | def get_meta_content(name, html): | |
383 | """ | |
384 | Return the content attribute from the meta tag with the given name attribute. | |
385 | """ | |
386 | parser = MetaParser(name) | |
387 | try: | |
388 | parser.loads(html) | |
389 | except compat_html_parser.HTMLParseError: | |
390 | pass | |
391 | return parser.get_result() | |
392 | ||
9e6dd238 FV |
393 | |
394 | def clean_html(html): | |
59ae15a5 PH |
395 | """Clean an HTML snippet into a readable string""" |
396 | # Newline vs <br /> | |
397 | html = html.replace('\n', ' ') | |
6b3aef80 FV |
398 | html = re.sub(r'\s*<\s*br\s*/?\s*>\s*', '\n', html) |
399 | html = re.sub(r'<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html) | |
59ae15a5 PH |
400 | # Strip html tags |
401 | html = re.sub('<.*?>', '', html) | |
402 | # Replace html entities | |
403 | html = unescapeHTML(html) | |
7decf895 | 404 | return html.strip() |
9e6dd238 FV |
405 | |
406 | ||
d77c3dfd | 407 | def sanitize_open(filename, open_mode): |
59ae15a5 PH |
408 | """Try to open the given filename, and slightly tweak it if this fails. |
409 | ||
410 | Attempts to open the given filename. If this fails, it tries to change | |
411 | the filename slightly, step by step, until it's either able to open it | |
412 | or it fails and raises a final exception, like the standard open() | |
413 | function. | |
414 | ||
415 | It returns the tuple (stream, definitive_file_name). | |
416 | """ | |
417 | try: | |
418 | if filename == u'-': | |
419 | if sys.platform == 'win32': | |
420 | import msvcrt | |
421 | msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY) | |
898280a0 | 422 | return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename) |
59ae15a5 PH |
423 | stream = open(encodeFilename(filename), open_mode) |
424 | return (stream, filename) | |
425 | except (IOError, OSError) as err: | |
f45c185f PH |
426 | if err.errno in (errno.EACCES,): |
427 | raise | |
59ae15a5 | 428 | |
f45c185f PH |
429 | # In case of error, try to remove win32 forbidden chars |
430 | alt_filename = os.path.join( | |
431 | re.sub(u'[/<>:"\\|\\\\?\\*]', u'#', path_part) | |
432 | for path_part in os.path.split(filename) | |
433 | ) | |
434 | if alt_filename == filename: | |
435 | raise | |
436 | else: | |
437 | # An exception here should be caught in the caller | |
438 | stream = open(encodeFilename(filename), open_mode) | |
439 | return (stream, alt_filename) | |
d77c3dfd FV |
440 | |
441 | ||
442 | def timeconvert(timestr): | |
59ae15a5 PH |
443 | """Convert RFC 2822 defined time string into system timestamp""" |
444 | timestamp = None | |
445 | timetuple = email.utils.parsedate_tz(timestr) | |
446 | if timetuple is not None: | |
447 | timestamp = email.utils.mktime_tz(timetuple) | |
448 | return timestamp | |
1c469a94 | 449 | |
796173d0 | 450 | def sanitize_filename(s, restricted=False, is_id=False): |
59ae15a5 PH |
451 | """Sanitizes a string so it could be used as part of a filename. |
452 | If restricted is set, use a stricter subset of allowed characters. | |
796173d0 | 453 | Set is_id if this is not an arbitrary string, but an ID that should be kept if possible |
59ae15a5 PH |
454 | """ |
455 | def replace_insane(char): | |
456 | if char == '?' or ord(char) < 32 or ord(char) == 127: | |
457 | return '' | |
458 | elif char == '"': | |
459 | return '' if restricted else '\'' | |
460 | elif char == ':': | |
461 | return '_-' if restricted else ' -' | |
462 | elif char in '\\/|*<>': | |
463 | return '_' | |
627dcfff | 464 | if restricted and (char in '!&\'()[]{}$;`^,#' or char.isspace()): |
59ae15a5 PH |
465 | return '_' |
466 | if restricted and ord(char) > 127: | |
467 | return '_' | |
468 | return char | |
469 | ||
470 | result = u''.join(map(replace_insane, s)) | |
796173d0 PH |
471 | if not is_id: |
472 | while '__' in result: | |
473 | result = result.replace('__', '_') | |
474 | result = result.strip('_') | |
475 | # Common case of "Foreign band name - English song title" | |
476 | if restricted and result.startswith('-_'): | |
477 | result = result[2:] | |
478 | if not result: | |
479 | result = '_' | |
59ae15a5 | 480 | return result |
d77c3dfd FV |
481 | |
482 | def orderedSet(iterable): | |
59ae15a5 PH |
483 | """ Remove all duplicates from the input iterable """ |
484 | res = [] | |
485 | for el in iterable: | |
486 | if el not in res: | |
487 | res.append(el) | |
488 | return res | |
d77c3dfd FV |
489 | |
490 | def unescapeHTML(s): | |
59ae15a5 PH |
491 | """ |
492 | @param s a string | |
493 | """ | |
494 | assert type(s) == type(u'') | |
d77c3dfd | 495 | |
59ae15a5 PH |
496 | result = re.sub(u'(?u)&(.+?);', htmlentity_transform, s) |
497 | return result | |
d77c3dfd FV |
498 | |
499 | def encodeFilename(s): | |
59ae15a5 PH |
500 | """ |
501 | @param s The name of the file | |
502 | """ | |
d77c3dfd | 503 | |
59ae15a5 | 504 | assert type(s) == type(u'') |
d77c3dfd | 505 | |
59ae15a5 PH |
506 | # Python 3 has a Unicode API |
507 | if sys.version_info >= (3, 0): | |
508 | return s | |
0f00efed | 509 | |
59ae15a5 PH |
510 | if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5: |
511 | # Pass u'' directly to use Unicode APIs on Windows 2000 and up | |
512 | # (Detecting Windows NT 4 is tricky because 'major >= 4' would | |
513 | # match Windows 9x series as well. Besides, NT 4 is obsolete.) | |
514 | return s | |
515 | else: | |
6df40dcb PH |
516 | encoding = sys.getfilesystemencoding() |
517 | if encoding is None: | |
518 | encoding = 'utf-8' | |
519 | return s.encode(encoding, 'ignore') | |
d77c3dfd | 520 | |
8271226a PH |
521 | def decodeOption(optval): |
522 | if optval is None: | |
523 | return optval | |
524 | if isinstance(optval, bytes): | |
525 | optval = optval.decode(preferredencoding()) | |
526 | ||
527 | assert isinstance(optval, compat_str) | |
528 | return optval | |
1c256f70 | 529 | |
4539dd30 PH |
530 | def formatSeconds(secs): |
531 | if secs > 3600: | |
532 | return '%d:%02d:%02d' % (secs // 3600, (secs % 3600) // 60, secs % 60) | |
533 | elif secs > 60: | |
534 | return '%d:%02d' % (secs // 60, secs % 60) | |
535 | else: | |
536 | return '%d' % secs | |
537 | ||
ea6d901e PH |
538 | def make_HTTPS_handler(opts): |
539 | if sys.version_info < (3,2): | |
540 | # Python's 2.x handler is very simplistic | |
acebc9cd | 541 | return compat_urllib_request.HTTPSHandler() |
ea6d901e PH |
542 | else: |
543 | import ssl | |
544 | context = ssl.SSLContext(ssl.PROTOCOL_SSLv23) | |
545 | context.set_default_verify_paths() | |
546 | ||
547 | context.verify_mode = (ssl.CERT_NONE | |
548 | if opts.no_check_certificate | |
549 | else ssl.CERT_REQUIRED) | |
acebc9cd | 550 | return compat_urllib_request.HTTPSHandler(context=context) |
ea6d901e | 551 | |
1c256f70 PH |
552 | class ExtractorError(Exception): |
553 | """Error during info extraction.""" | |
2eabb802 | 554 | def __init__(self, msg, tb=None, expected=False, cause=None): |
9a82b238 PH |
555 | """ tb, if given, is the original traceback (so that it can be printed out). |
556 | If expected is set, this is a normal error message and most likely not a bug in youtube-dl. | |
557 | """ | |
558 | ||
559 | if sys.exc_info()[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError): | |
560 | expected = True | |
561 | if not expected: | |
298f833b | 562 | 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 | 563 | super(ExtractorError, self).__init__(msg) |
d5979c5d | 564 | |
1c256f70 | 565 | self.traceback = tb |
8cc83b8d | 566 | self.exc_info = sys.exc_info() # preserve original exception |
2eabb802 | 567 | self.cause = cause |
1c256f70 | 568 | |
01951dda PH |
569 | def format_traceback(self): |
570 | if self.traceback is None: | |
571 | return None | |
572 | return u''.join(traceback.format_tb(self.traceback)) | |
573 | ||
1c256f70 | 574 | |
55b3e45b JMF |
575 | class RegexNotFoundError(ExtractorError): |
576 | """Error when a regex didn't match""" | |
577 | pass | |
578 | ||
579 | ||
d77c3dfd | 580 | class DownloadError(Exception): |
59ae15a5 | 581 | """Download Error exception. |
d77c3dfd | 582 | |
59ae15a5 PH |
583 | This exception may be thrown by FileDownloader objects if they are not |
584 | configured to continue on errors. They will contain the appropriate | |
585 | error message. | |
586 | """ | |
8cc83b8d FV |
587 | def __init__(self, msg, exc_info=None): |
588 | """ exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """ | |
589 | super(DownloadError, self).__init__(msg) | |
590 | self.exc_info = exc_info | |
d77c3dfd FV |
591 | |
592 | ||
593 | class SameFileError(Exception): | |
59ae15a5 | 594 | """Same File exception. |
d77c3dfd | 595 | |
59ae15a5 PH |
596 | This exception will be thrown by FileDownloader objects if they detect |
597 | multiple files would have to be downloaded to the same file on disk. | |
598 | """ | |
599 | pass | |
d77c3dfd FV |
600 | |
601 | ||
602 | class PostProcessingError(Exception): | |
59ae15a5 | 603 | """Post Processing exception. |
d77c3dfd | 604 | |
59ae15a5 PH |
605 | This exception may be raised by PostProcessor's .run() method to |
606 | indicate an error in the postprocessing task. | |
607 | """ | |
7851b379 PH |
608 | def __init__(self, msg): |
609 | self.msg = msg | |
d77c3dfd FV |
610 | |
611 | class MaxDownloadsReached(Exception): | |
59ae15a5 PH |
612 | """ --max-downloads limit has been reached. """ |
613 | pass | |
d77c3dfd FV |
614 | |
615 | ||
616 | class UnavailableVideoError(Exception): | |
59ae15a5 | 617 | """Unavailable Format exception. |
d77c3dfd | 618 | |
59ae15a5 PH |
619 | This exception will be thrown when a video is requested |
620 | in a format that is not available for that video. | |
621 | """ | |
622 | pass | |
d77c3dfd FV |
623 | |
624 | ||
625 | class ContentTooShortError(Exception): | |
59ae15a5 | 626 | """Content Too Short exception. |
d77c3dfd | 627 | |
59ae15a5 PH |
628 | This exception may be raised by FileDownloader objects when a file they |
629 | download is too small for what the server announced first, indicating | |
630 | the connection was probably interrupted. | |
631 | """ | |
632 | # Both in bytes | |
633 | downloaded = None | |
634 | expected = None | |
d77c3dfd | 635 | |
59ae15a5 PH |
636 | def __init__(self, downloaded, expected): |
637 | self.downloaded = downloaded | |
638 | self.expected = expected | |
d77c3dfd | 639 | |
acebc9cd | 640 | class YoutubeDLHandler(compat_urllib_request.HTTPHandler): |
59ae15a5 PH |
641 | """Handler for HTTP requests and responses. |
642 | ||
643 | This class, when installed with an OpenerDirector, automatically adds | |
644 | the standard headers to every HTTP request and handles gzipped and | |
645 | deflated responses from web servers. If compression is to be avoided in | |
646 | a particular request, the original request in the program code only has | |
647 | to include the HTTP header "Youtubedl-No-Compression", which will be | |
648 | removed before making the real request. | |
649 | ||
650 | Part of this code was copied from: | |
651 | ||
652 | http://techknack.net/python-urllib2-handlers/ | |
653 | ||
654 | Andrew Rowls, the author of that code, agreed to release it to the | |
655 | public domain. | |
656 | """ | |
657 | ||
658 | @staticmethod | |
659 | def deflate(data): | |
660 | try: | |
661 | return zlib.decompress(data, -zlib.MAX_WBITS) | |
662 | except zlib.error: | |
663 | return zlib.decompress(data) | |
664 | ||
665 | @staticmethod | |
666 | def addinfourl_wrapper(stream, headers, url, code): | |
667 | if hasattr(compat_urllib_request.addinfourl, 'getcode'): | |
668 | return compat_urllib_request.addinfourl(stream, headers, url, code) | |
669 | ret = compat_urllib_request.addinfourl(stream, headers, url) | |
670 | ret.code = code | |
671 | return ret | |
672 | ||
acebc9cd PH |
673 | def http_request(self, req): |
674 | for h,v in std_headers.items(): | |
59ae15a5 PH |
675 | if h in req.headers: |
676 | del req.headers[h] | |
335959e7 | 677 | req.add_header(h, v) |
59ae15a5 PH |
678 | if 'Youtubedl-no-compression' in req.headers: |
679 | if 'Accept-encoding' in req.headers: | |
680 | del req.headers['Accept-encoding'] | |
681 | del req.headers['Youtubedl-no-compression'] | |
3446dfb7 | 682 | if 'Youtubedl-user-agent' in req.headers: |
335959e7 PH |
683 | if 'User-agent' in req.headers: |
684 | del req.headers['User-agent'] | |
685 | req.headers['User-agent'] = req.headers['Youtubedl-user-agent'] | |
3446dfb7 | 686 | del req.headers['Youtubedl-user-agent'] |
59ae15a5 PH |
687 | return req |
688 | ||
acebc9cd | 689 | def http_response(self, req, resp): |
59ae15a5 PH |
690 | old_resp = resp |
691 | # gzip | |
692 | if resp.headers.get('Content-encoding', '') == 'gzip': | |
aa3e9507 PH |
693 | content = resp.read() |
694 | gz = gzip.GzipFile(fileobj=io.BytesIO(content), mode='rb') | |
695 | try: | |
696 | uncompressed = io.BytesIO(gz.read()) | |
697 | except IOError as original_ioerror: | |
698 | # There may be junk add the end of the file | |
699 | # See http://stackoverflow.com/q/4928560/35070 for details | |
700 | for i in range(1, 1024): | |
701 | try: | |
702 | gz = gzip.GzipFile(fileobj=io.BytesIO(content[:-i]), mode='rb') | |
703 | uncompressed = io.BytesIO(gz.read()) | |
704 | except IOError: | |
705 | continue | |
706 | break | |
707 | else: | |
708 | raise original_ioerror | |
709 | resp = self.addinfourl_wrapper(uncompressed, old_resp.headers, old_resp.url, old_resp.code) | |
59ae15a5 PH |
710 | resp.msg = old_resp.msg |
711 | # deflate | |
712 | if resp.headers.get('Content-encoding', '') == 'deflate': | |
713 | gz = io.BytesIO(self.deflate(resp.read())) | |
714 | resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code) | |
715 | resp.msg = old_resp.msg | |
716 | return resp | |
0f8d03f8 | 717 | |
acebc9cd PH |
718 | https_request = http_request |
719 | https_response = http_response | |
bf50b038 JMF |
720 | |
721 | def unified_strdate(date_str): | |
722 | """Return a string with the date in the format YYYYMMDD""" | |
723 | upload_date = None | |
724 | #Replace commas | |
725 | date_str = date_str.replace(',',' ') | |
726 | # %z (UTC offset) is only supported in python>=3.2 | |
727 | date_str = re.sub(r' (\+|-)[\d]*$', '', date_str) | |
19e1d359 JMF |
728 | format_expressions = [ |
729 | '%d %B %Y', | |
730 | '%B %d %Y', | |
731 | '%b %d %Y', | |
732 | '%Y-%m-%d', | |
733 | '%d/%m/%Y', | |
734 | '%Y/%m/%d %H:%M:%S', | |
735 | '%d.%m.%Y %H:%M', | |
736 | '%Y-%m-%dT%H:%M:%SZ', | |
2e1fa03b | 737 | '%Y-%m-%dT%H:%M:%S', |
19e1d359 | 738 | ] |
bf50b038 JMF |
739 | for expression in format_expressions: |
740 | try: | |
741 | upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d') | |
742 | except: | |
743 | pass | |
744 | return upload_date | |
745 | ||
cbdbb766 | 746 | def determine_ext(url, default_ext=u'unknown_video'): |
73e79f2a PH |
747 | guess = url.partition(u'?')[0].rpartition(u'.')[2] |
748 | if re.match(r'^[A-Za-z0-9]+$', guess): | |
749 | return guess | |
750 | else: | |
cbdbb766 | 751 | return default_ext |
73e79f2a | 752 | |
d4051a8e JMF |
753 | def subtitles_filename(filename, sub_lang, sub_format): |
754 | return filename.rsplit('.', 1)[0] + u'.' + sub_lang + u'.' + sub_format | |
755 | ||
bd558525 | 756 | def date_from_str(date_str): |
37254abc JMF |
757 | """ |
758 | Return a datetime object from a string in the format YYYYMMDD or | |
759 | (now|today)[+-][0-9](day|week|month|year)(s)?""" | |
760 | today = datetime.date.today() | |
761 | if date_str == 'now'or date_str == 'today': | |
762 | return today | |
763 | match = re.match('(now|today)(?P<sign>[+-])(?P<time>\d+)(?P<unit>day|week|month|year)(s)?', date_str) | |
764 | if match is not None: | |
765 | sign = match.group('sign') | |
766 | time = int(match.group('time')) | |
767 | if sign == '-': | |
768 | time = -time | |
769 | unit = match.group('unit') | |
770 | #A bad aproximation? | |
771 | if unit == 'month': | |
772 | unit = 'day' | |
773 | time *= 30 | |
774 | elif unit == 'year': | |
775 | unit = 'day' | |
776 | time *= 365 | |
777 | unit += 's' | |
778 | delta = datetime.timedelta(**{unit: time}) | |
779 | return today + delta | |
bd558525 JMF |
780 | return datetime.datetime.strptime(date_str, "%Y%m%d").date() |
781 | ||
782 | class DateRange(object): | |
783 | """Represents a time interval between two dates""" | |
784 | def __init__(self, start=None, end=None): | |
785 | """start and end must be strings in the format accepted by date""" | |
786 | if start is not None: | |
787 | self.start = date_from_str(start) | |
788 | else: | |
789 | self.start = datetime.datetime.min.date() | |
790 | if end is not None: | |
791 | self.end = date_from_str(end) | |
792 | else: | |
793 | self.end = datetime.datetime.max.date() | |
37254abc | 794 | if self.start > self.end: |
bd558525 JMF |
795 | raise ValueError('Date range: "%s" , the start date must be before the end date' % self) |
796 | @classmethod | |
797 | def day(cls, day): | |
798 | """Returns a range that only contains the given day""" | |
799 | return cls(day,day) | |
800 | def __contains__(self, date): | |
801 | """Check if the date is in the range""" | |
37254abc JMF |
802 | if not isinstance(date, datetime.date): |
803 | date = date_from_str(date) | |
804 | return self.start <= date <= self.end | |
bd558525 JMF |
805 | def __str__(self): |
806 | return '%s - %s' % ( self.start.isoformat(), self.end.isoformat()) | |
c496ca96 PH |
807 | |
808 | ||
809 | def platform_name(): | |
810 | """ Returns the platform name as a compat_str """ | |
811 | res = platform.platform() | |
812 | if isinstance(res, bytes): | |
813 | res = res.decode(preferredencoding()) | |
814 | ||
815 | assert isinstance(res, compat_str) | |
816 | return res | |
c257baff PH |
817 | |
818 | ||
7459e3a2 PH |
819 | def write_string(s, out=None): |
820 | if out is None: | |
821 | out = sys.stderr | |
822 | assert type(s) == type(u'') | |
823 | ||
824 | if ('b' in getattr(out, 'mode', '') or | |
825 | sys.version_info[0] < 3): # Python 2 lies about mode of sys.stderr | |
826 | s = s.encode(preferredencoding(), 'ignore') | |
827 | out.write(s) | |
828 | out.flush() | |
829 | ||
830 | ||
48ea9cea PH |
831 | def bytes_to_intlist(bs): |
832 | if not bs: | |
833 | return [] | |
834 | if isinstance(bs[0], int): # Python 3 | |
835 | return list(bs) | |
836 | else: | |
837 | return [ord(c) for c in bs] | |
838 | ||
c257baff | 839 | |
cba892fa | 840 | def intlist_to_bytes(xs): |
841 | if not xs: | |
842 | return b'' | |
843 | if isinstance(chr(0), bytes): # Python 2 | |
844 | return ''.join([chr(x) for x in xs]) | |
845 | else: | |
846 | return bytes(xs) | |
c38b1e77 PH |
847 | |
848 | ||
849 | def get_cachedir(params={}): | |
850 | cache_root = os.environ.get('XDG_CACHE_HOME', | |
851 | os.path.expanduser('~/.cache')) | |
852 | return params.get('cachedir', os.path.join(cache_root, 'youtube-dl')) | |
c1c9a79c PH |
853 | |
854 | ||
855 | # Cross-platform file locking | |
856 | if sys.platform == 'win32': | |
857 | import ctypes.wintypes | |
858 | import msvcrt | |
859 | ||
860 | class OVERLAPPED(ctypes.Structure): | |
861 | _fields_ = [ | |
862 | ('Internal', ctypes.wintypes.LPVOID), | |
863 | ('InternalHigh', ctypes.wintypes.LPVOID), | |
864 | ('Offset', ctypes.wintypes.DWORD), | |
865 | ('OffsetHigh', ctypes.wintypes.DWORD), | |
866 | ('hEvent', ctypes.wintypes.HANDLE), | |
867 | ] | |
868 | ||
869 | kernel32 = ctypes.windll.kernel32 | |
870 | LockFileEx = kernel32.LockFileEx | |
871 | LockFileEx.argtypes = [ | |
872 | ctypes.wintypes.HANDLE, # hFile | |
873 | ctypes.wintypes.DWORD, # dwFlags | |
874 | ctypes.wintypes.DWORD, # dwReserved | |
875 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow | |
876 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh | |
877 | ctypes.POINTER(OVERLAPPED) # Overlapped | |
878 | ] | |
879 | LockFileEx.restype = ctypes.wintypes.BOOL | |
880 | UnlockFileEx = kernel32.UnlockFileEx | |
881 | UnlockFileEx.argtypes = [ | |
882 | ctypes.wintypes.HANDLE, # hFile | |
883 | ctypes.wintypes.DWORD, # dwReserved | |
884 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow | |
885 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh | |
886 | ctypes.POINTER(OVERLAPPED) # Overlapped | |
887 | ] | |
888 | UnlockFileEx.restype = ctypes.wintypes.BOOL | |
889 | whole_low = 0xffffffff | |
890 | whole_high = 0x7fffffff | |
891 | ||
892 | def _lock_file(f, exclusive): | |
893 | overlapped = OVERLAPPED() | |
894 | overlapped.Offset = 0 | |
895 | overlapped.OffsetHigh = 0 | |
896 | overlapped.hEvent = 0 | |
897 | f._lock_file_overlapped_p = ctypes.pointer(overlapped) | |
898 | handle = msvcrt.get_osfhandle(f.fileno()) | |
899 | if not LockFileEx(handle, 0x2 if exclusive else 0x0, 0, | |
900 | whole_low, whole_high, f._lock_file_overlapped_p): | |
901 | raise OSError('Locking file failed: %r' % ctypes.FormatError()) | |
902 | ||
903 | def _unlock_file(f): | |
904 | assert f._lock_file_overlapped_p | |
905 | handle = msvcrt.get_osfhandle(f.fileno()) | |
906 | if not UnlockFileEx(handle, 0, | |
907 | whole_low, whole_high, f._lock_file_overlapped_p): | |
908 | raise OSError('Unlocking file failed: %r' % ctypes.FormatError()) | |
909 | ||
910 | else: | |
911 | import fcntl | |
912 | ||
913 | def _lock_file(f, exclusive): | |
914 | fcntl.lockf(f, fcntl.LOCK_EX if exclusive else fcntl.LOCK_SH) | |
915 | ||
916 | def _unlock_file(f): | |
917 | fcntl.lockf(f, fcntl.LOCK_UN) | |
918 | ||
919 | ||
920 | class locked_file(object): | |
921 | def __init__(self, filename, mode, encoding=None): | |
922 | assert mode in ['r', 'a', 'w'] | |
923 | self.f = io.open(filename, mode, encoding=encoding) | |
924 | self.mode = mode | |
925 | ||
926 | def __enter__(self): | |
927 | exclusive = self.mode != 'r' | |
928 | try: | |
929 | _lock_file(self.f, exclusive) | |
930 | except IOError: | |
931 | self.f.close() | |
932 | raise | |
933 | return self | |
934 | ||
935 | def __exit__(self, etype, value, traceback): | |
936 | try: | |
937 | _unlock_file(self.f) | |
938 | finally: | |
939 | self.f.close() | |
940 | ||
941 | def __iter__(self): | |
942 | return iter(self.f) | |
943 | ||
944 | def write(self, *args): | |
945 | return self.f.write(*args) | |
946 | ||
947 | def read(self, *args): | |
948 | return self.f.read(*args) | |
4eb7f1d1 JMF |
949 | |
950 | ||
951 | def shell_quote(args): | |
952 | return ' '.join(map(pipes.quote, args)) | |
9d4660ca PH |
953 | |
954 | ||
f4d96df0 PH |
955 | def takewhile_inclusive(pred, seq): |
956 | """ Like itertools.takewhile, but include the latest evaluated element | |
957 | (the first element so that Not pred(e)) """ | |
958 | for e in seq: | |
959 | yield e | |
960 | if not pred(e): | |
961 | return | |
962 | ||
963 | ||
9d4660ca PH |
964 | def smuggle_url(url, data): |
965 | """ Pass additional data in a URL for internal use. """ | |
966 | ||
967 | sdata = compat_urllib_parse.urlencode( | |
968 | {u'__youtubedl_smuggle': json.dumps(data)}) | |
969 | return url + u'#' + sdata | |
970 | ||
971 | ||
972 | def unsmuggle_url(smug_url): | |
973 | if not '#__youtubedl_smuggle' in smug_url: | |
974 | return smug_url, None | |
975 | url, _, sdata = smug_url.rpartition(u'#') | |
976 | jsond = compat_parse_qs(sdata)[u'__youtubedl_smuggle'][0] | |
977 | data = json.loads(jsond) | |
978 | return url, data |