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