]>
Commit | Line | Data |
---|---|---|
d77c3dfd FV |
1 | #!/usr/bin/env python |
2 | # -*- coding: utf-8 -*- | |
3 | ||
4 | import gzip | |
03f9daab | 5 | import io |
f4bfd65f | 6 | import json |
d77c3dfd FV |
7 | import locale |
8 | import os | |
9 | import re | |
10 | import sys | |
11 | import zlib | |
d77c3dfd | 12 | import email.utils |
921a1455 | 13 | import json |
d77c3dfd | 14 | |
01ba00ca | 15 | try: |
59ae15a5 | 16 | import urllib.request as compat_urllib_request |
01ba00ca | 17 | except ImportError: # Python 2 |
59ae15a5 | 18 | import urllib2 as compat_urllib_request |
01ba00ca PH |
19 | |
20 | try: | |
59ae15a5 | 21 | import urllib.error as compat_urllib_error |
01ba00ca | 22 | except ImportError: # Python 2 |
59ae15a5 | 23 | import urllib2 as compat_urllib_error |
01ba00ca PH |
24 | |
25 | try: | |
59ae15a5 | 26 | import urllib.parse as compat_urllib_parse |
01ba00ca | 27 | except ImportError: # Python 2 |
59ae15a5 | 28 | import urllib as compat_urllib_parse |
01ba00ca | 29 | |
799c0763 PH |
30 | try: |
31 | from urllib.parse import urlparse as compat_urllib_parse_urlparse | |
32 | except ImportError: # Python 2 | |
33 | from urlparse import urlparse as compat_urllib_parse_urlparse | |
34 | ||
01ba00ca | 35 | try: |
59ae15a5 | 36 | import http.cookiejar as compat_cookiejar |
01ba00ca | 37 | except ImportError: # Python 2 |
59ae15a5 | 38 | import cookielib as compat_cookiejar |
01ba00ca | 39 | |
3e669f36 | 40 | try: |
59ae15a5 | 41 | import html.entities as compat_html_entities |
9f37a959 | 42 | except ImportError: # Python 2 |
59ae15a5 | 43 | import htmlentitydefs as compat_html_entities |
3e669f36 | 44 | |
a8156c1d | 45 | try: |
59ae15a5 | 46 | import html.parser as compat_html_parser |
9f37a959 | 47 | except ImportError: # Python 2 |
59ae15a5 | 48 | import HTMLParser as compat_html_parser |
a8156c1d | 49 | |
348d0a7a | 50 | try: |
59ae15a5 | 51 | import http.client as compat_http_client |
9f37a959 | 52 | except ImportError: # Python 2 |
59ae15a5 | 53 | import httplib as compat_http_client |
348d0a7a | 54 | |
5910e210 PH |
55 | try: |
56 | from subprocess import DEVNULL | |
57 | compat_subprocess_get_DEVNULL = lambda: DEVNULL | |
58 | except ImportError: | |
59 | compat_subprocess_get_DEVNULL = lambda: open(os.path.devnull, 'w') | |
60 | ||
9f37a959 | 61 | try: |
59ae15a5 | 62 | from urllib.parse import parse_qs as compat_parse_qs |
9f37a959 | 63 | except ImportError: # Python 2 |
59ae15a5 PH |
64 | # HACK: The following is the correct parse_qs implementation from cpython 3's stdlib. |
65 | # Python 2's version is apparently totally broken | |
66 | def _unquote(string, encoding='utf-8', errors='replace'): | |
67 | if string == '': | |
68 | return string | |
69 | res = string.split('%') | |
70 | if len(res) == 1: | |
71 | return string | |
72 | if encoding is None: | |
73 | encoding = 'utf-8' | |
74 | if errors is None: | |
75 | errors = 'replace' | |
76 | # pct_sequence: contiguous sequence of percent-encoded bytes, decoded | |
77 | pct_sequence = b'' | |
78 | string = res[0] | |
79 | for item in res[1:]: | |
80 | try: | |
81 | if not item: | |
82 | raise ValueError | |
83 | pct_sequence += item[:2].decode('hex') | |
84 | rest = item[2:] | |
85 | if not rest: | |
86 | # This segment was just a single percent-encoded character. | |
87 | # May be part of a sequence of code units, so delay decoding. | |
88 | # (Stored in pct_sequence). | |
89 | continue | |
90 | except ValueError: | |
91 | rest = '%' + item | |
92 | # Encountered non-percent-encoded characters. Flush the current | |
93 | # pct_sequence. | |
94 | string += pct_sequence.decode(encoding, errors) + rest | |
95 | pct_sequence = b'' | |
96 | if pct_sequence: | |
97 | # Flush the final pct_sequence | |
98 | string += pct_sequence.decode(encoding, errors) | |
99 | return string | |
100 | ||
101 | def _parse_qsl(qs, keep_blank_values=False, strict_parsing=False, | |
102 | encoding='utf-8', errors='replace'): | |
103 | qs, _coerce_result = qs, unicode | |
104 | pairs = [s2 for s1 in qs.split('&') for s2 in s1.split(';')] | |
105 | r = [] | |
106 | for name_value in pairs: | |
107 | if not name_value and not strict_parsing: | |
108 | continue | |
109 | nv = name_value.split('=', 1) | |
110 | if len(nv) != 2: | |
111 | if strict_parsing: | |
112 | raise ValueError("bad query field: %r" % (name_value,)) | |
113 | # Handle case of a control-name with no equal sign | |
114 | if keep_blank_values: | |
115 | nv.append('') | |
116 | else: | |
117 | continue | |
118 | if len(nv[1]) or keep_blank_values: | |
119 | name = nv[0].replace('+', ' ') | |
120 | name = _unquote(name, encoding=encoding, errors=errors) | |
121 | name = _coerce_result(name) | |
122 | value = nv[1].replace('+', ' ') | |
123 | value = _unquote(value, encoding=encoding, errors=errors) | |
124 | value = _coerce_result(value) | |
125 | r.append((name, value)) | |
126 | return r | |
127 | ||
128 | def compat_parse_qs(qs, keep_blank_values=False, strict_parsing=False, | |
129 | encoding='utf-8', errors='replace'): | |
130 | parsed_result = {} | |
131 | pairs = _parse_qsl(qs, keep_blank_values, strict_parsing, | |
132 | encoding=encoding, errors=errors) | |
133 | for name, value in pairs: | |
134 | if name in parsed_result: | |
135 | parsed_result[name].append(value) | |
136 | else: | |
137 | parsed_result[name] = [value] | |
138 | return parsed_result | |
348d0a7a | 139 | |
3e669f36 | 140 | try: |
59ae15a5 | 141 | compat_str = unicode # Python 2 |
3e669f36 | 142 | except NameError: |
59ae15a5 | 143 | compat_str = str |
3e669f36 PH |
144 | |
145 | try: | |
59ae15a5 | 146 | compat_chr = unichr # Python 2 |
3e669f36 | 147 | except NameError: |
59ae15a5 | 148 | compat_chr = chr |
3e669f36 | 149 | |
3e669f36 | 150 | std_headers = { |
59ae15a5 PH |
151 | 'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20100101 Firefox/10.0', |
152 | 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7', | |
153 | 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8', | |
154 | 'Accept-Encoding': 'gzip, deflate', | |
155 | 'Accept-Language': 'en-us,en;q=0.5', | |
3e669f36 | 156 | } |
d77c3dfd | 157 | def preferredencoding(): |
59ae15a5 | 158 | """Get preferred encoding. |
d77c3dfd | 159 | |
59ae15a5 PH |
160 | Returns the best encoding scheme for the system, based on |
161 | locale.getpreferredencoding() and some further tweaks. | |
162 | """ | |
163 | try: | |
164 | pref = locale.getpreferredencoding() | |
165 | u'TEST'.encode(pref) | |
166 | except: | |
167 | pref = 'UTF-8' | |
bae611f2 | 168 | |
59ae15a5 | 169 | return pref |
d77c3dfd | 170 | |
8cd10ac4 | 171 | if sys.version_info < (3,0): |
59ae15a5 PH |
172 | def compat_print(s): |
173 | print(s.encode(preferredencoding(), 'xmlcharrefreplace')) | |
8cd10ac4 | 174 | else: |
59ae15a5 PH |
175 | def compat_print(s): |
176 | assert type(s) == type(u'') | |
177 | print(s) | |
d77c3dfd | 178 | |
f4bfd65f PH |
179 | # In Python 2.x, json.dump expects a bytestream. |
180 | # In Python 3.x, it writes to a character stream | |
181 | if sys.version_info < (3,0): | |
182 | def write_json_file(obj, fn): | |
183 | with open(fn, 'wb') as f: | |
184 | json.dump(obj, f) | |
185 | else: | |
186 | def write_json_file(obj, fn): | |
187 | with open(fn, 'w', encoding='utf-8') as f: | |
188 | json.dump(obj, f) | |
189 | ||
190 | ||
d77c3dfd | 191 | def htmlentity_transform(matchobj): |
59ae15a5 PH |
192 | """Transforms an HTML entity to a character. |
193 | ||
194 | This function receives a match object and is intended to be used with | |
195 | the re.sub() function. | |
196 | """ | |
197 | entity = matchobj.group(1) | |
198 | ||
199 | # Known non-numeric HTML entity | |
200 | if entity in compat_html_entities.name2codepoint: | |
201 | return compat_chr(compat_html_entities.name2codepoint[entity]) | |
202 | ||
203 | mobj = re.match(u'(?u)#(x?\\d+)', entity) | |
204 | if mobj is not None: | |
205 | numstr = mobj.group(1) | |
206 | if numstr.startswith(u'x'): | |
207 | base = 16 | |
208 | numstr = u'0%s' % numstr | |
209 | else: | |
210 | base = 10 | |
211 | return compat_chr(int(numstr, base)) | |
212 | ||
213 | # Unknown entity in name, return its literal representation | |
214 | return (u'&%s;' % entity) | |
d77c3dfd | 215 | |
a8156c1d PH |
216 | 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 |
217 | class IDParser(compat_html_parser.HTMLParser): | |
59ae15a5 PH |
218 | """Modified HTMLParser that isolates a tag with the specified id""" |
219 | def __init__(self, id): | |
220 | self.id = id | |
221 | self.result = None | |
222 | self.started = False | |
223 | self.depth = {} | |
224 | self.html = None | |
225 | self.watch_startpos = False | |
226 | self.error_count = 0 | |
227 | compat_html_parser.HTMLParser.__init__(self) | |
228 | ||
229 | def error(self, message): | |
230 | if self.error_count > 10 or self.started: | |
231 | raise compat_html_parser.HTMLParseError(message, self.getpos()) | |
232 | self.rawdata = '\n'.join(self.html.split('\n')[self.getpos()[0]:]) # skip one line | |
233 | self.error_count += 1 | |
234 | self.goahead(1) | |
235 | ||
236 | def loads(self, html): | |
237 | self.html = html | |
238 | self.feed(html) | |
239 | self.close() | |
240 | ||
241 | def handle_starttag(self, tag, attrs): | |
242 | attrs = dict(attrs) | |
243 | if self.started: | |
244 | self.find_startpos(None) | |
245 | if 'id' in attrs and attrs['id'] == self.id: | |
246 | self.result = [tag] | |
247 | self.started = True | |
248 | self.watch_startpos = True | |
249 | if self.started: | |
250 | if not tag in self.depth: self.depth[tag] = 0 | |
251 | self.depth[tag] += 1 | |
252 | ||
253 | def handle_endtag(self, tag): | |
254 | if self.started: | |
255 | if tag in self.depth: self.depth[tag] -= 1 | |
256 | if self.depth[self.result[0]] == 0: | |
257 | self.started = False | |
258 | self.result.append(self.getpos()) | |
259 | ||
260 | def find_startpos(self, x): | |
261 | """Needed to put the start position of the result (self.result[1]) | |
262 | after the opening tag with the requested id""" | |
263 | if self.watch_startpos: | |
264 | self.watch_startpos = False | |
265 | self.result.append(self.getpos()) | |
266 | handle_entityref = handle_charref = handle_data = handle_comment = \ | |
267 | handle_decl = handle_pi = unknown_decl = find_startpos | |
268 | ||
269 | def get_result(self): | |
270 | if self.result is None: | |
271 | return None | |
272 | if len(self.result) != 3: | |
273 | return None | |
274 | lines = self.html.split('\n') | |
275 | lines = lines[self.result[1][0]-1:self.result[2][0]] | |
276 | lines[0] = lines[0][self.result[1][1]:] | |
277 | if len(lines) == 1: | |
278 | lines[-1] = lines[-1][:self.result[2][1]-self.result[1][1]] | |
279 | lines[-1] = lines[-1][:self.result[2][1]] | |
280 | return '\n'.join(lines).strip() | |
9e6dd238 FV |
281 | |
282 | def get_element_by_id(id, html): | |
59ae15a5 PH |
283 | """Return the content of the tag with the specified id in the passed HTML document""" |
284 | parser = IDParser(id) | |
285 | try: | |
286 | parser.loads(html) | |
287 | except compat_html_parser.HTMLParseError: | |
288 | pass | |
289 | return parser.get_result() | |
9e6dd238 FV |
290 | |
291 | ||
292 | def clean_html(html): | |
59ae15a5 PH |
293 | """Clean an HTML snippet into a readable string""" |
294 | # Newline vs <br /> | |
295 | html = html.replace('\n', ' ') | |
296 | html = re.sub('\s*<\s*br\s*/?\s*>\s*', '\n', html) | |
297 | # Strip html tags | |
298 | html = re.sub('<.*?>', '', html) | |
299 | # Replace html entities | |
300 | html = unescapeHTML(html) | |
301 | return html | |
9e6dd238 FV |
302 | |
303 | ||
d77c3dfd | 304 | def sanitize_open(filename, open_mode): |
59ae15a5 PH |
305 | """Try to open the given filename, and slightly tweak it if this fails. |
306 | ||
307 | Attempts to open the given filename. If this fails, it tries to change | |
308 | the filename slightly, step by step, until it's either able to open it | |
309 | or it fails and raises a final exception, like the standard open() | |
310 | function. | |
311 | ||
312 | It returns the tuple (stream, definitive_file_name). | |
313 | """ | |
314 | try: | |
315 | if filename == u'-': | |
316 | if sys.platform == 'win32': | |
317 | import msvcrt | |
318 | msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY) | |
319 | return (sys.stdout, filename) | |
320 | stream = open(encodeFilename(filename), open_mode) | |
321 | return (stream, filename) | |
322 | except (IOError, OSError) as err: | |
323 | # In case of error, try to remove win32 forbidden chars | |
324 | filename = re.sub(u'[/<>:"\\|\\\\?\\*]', u'#', filename) | |
325 | ||
326 | # An exception here should be caught in the caller | |
327 | stream = open(encodeFilename(filename), open_mode) | |
328 | return (stream, filename) | |
d77c3dfd FV |
329 | |
330 | ||
331 | def timeconvert(timestr): | |
59ae15a5 PH |
332 | """Convert RFC 2822 defined time string into system timestamp""" |
333 | timestamp = None | |
334 | timetuple = email.utils.parsedate_tz(timestr) | |
335 | if timetuple is not None: | |
336 | timestamp = email.utils.mktime_tz(timetuple) | |
337 | return timestamp | |
1c469a94 | 338 | |
796173d0 | 339 | def sanitize_filename(s, restricted=False, is_id=False): |
59ae15a5 PH |
340 | """Sanitizes a string so it could be used as part of a filename. |
341 | If restricted is set, use a stricter subset of allowed characters. | |
796173d0 | 342 | Set is_id if this is not an arbitrary string, but an ID that should be kept if possible |
59ae15a5 PH |
343 | """ |
344 | def replace_insane(char): | |
345 | if char == '?' or ord(char) < 32 or ord(char) == 127: | |
346 | return '' | |
347 | elif char == '"': | |
348 | return '' if restricted else '\'' | |
349 | elif char == ':': | |
350 | return '_-' if restricted else ' -' | |
351 | elif char in '\\/|*<>': | |
352 | return '_' | |
627dcfff | 353 | if restricted and (char in '!&\'()[]{}$;`^,#' or char.isspace()): |
59ae15a5 PH |
354 | return '_' |
355 | if restricted and ord(char) > 127: | |
356 | return '_' | |
357 | return char | |
358 | ||
359 | result = u''.join(map(replace_insane, s)) | |
796173d0 PH |
360 | if not is_id: |
361 | while '__' in result: | |
362 | result = result.replace('__', '_') | |
363 | result = result.strip('_') | |
364 | # Common case of "Foreign band name - English song title" | |
365 | if restricted and result.startswith('-_'): | |
366 | result = result[2:] | |
367 | if not result: | |
368 | result = '_' | |
59ae15a5 | 369 | return result |
d77c3dfd FV |
370 | |
371 | def orderedSet(iterable): | |
59ae15a5 PH |
372 | """ Remove all duplicates from the input iterable """ |
373 | res = [] | |
374 | for el in iterable: | |
375 | if el not in res: | |
376 | res.append(el) | |
377 | return res | |
d77c3dfd FV |
378 | |
379 | def unescapeHTML(s): | |
59ae15a5 PH |
380 | """ |
381 | @param s a string | |
382 | """ | |
383 | assert type(s) == type(u'') | |
d77c3dfd | 384 | |
59ae15a5 PH |
385 | result = re.sub(u'(?u)&(.+?);', htmlentity_transform, s) |
386 | return result | |
d77c3dfd FV |
387 | |
388 | def encodeFilename(s): | |
59ae15a5 PH |
389 | """ |
390 | @param s The name of the file | |
391 | """ | |
d77c3dfd | 392 | |
59ae15a5 | 393 | assert type(s) == type(u'') |
d77c3dfd | 394 | |
59ae15a5 PH |
395 | # Python 3 has a Unicode API |
396 | if sys.version_info >= (3, 0): | |
397 | return s | |
0f00efed | 398 | |
59ae15a5 PH |
399 | if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5: |
400 | # Pass u'' directly to use Unicode APIs on Windows 2000 and up | |
401 | # (Detecting Windows NT 4 is tricky because 'major >= 4' would | |
402 | # match Windows 9x series as well. Besides, NT 4 is obsolete.) | |
403 | return s | |
404 | else: | |
405 | return s.encode(sys.getfilesystemencoding(), 'ignore') | |
d77c3dfd FV |
406 | |
407 | class DownloadError(Exception): | |
59ae15a5 | 408 | """Download Error exception. |
d77c3dfd | 409 | |
59ae15a5 PH |
410 | This exception may be thrown by FileDownloader objects if they are not |
411 | configured to continue on errors. They will contain the appropriate | |
412 | error message. | |
413 | """ | |
414 | pass | |
d77c3dfd FV |
415 | |
416 | ||
417 | class SameFileError(Exception): | |
59ae15a5 | 418 | """Same File exception. |
d77c3dfd | 419 | |
59ae15a5 PH |
420 | This exception will be thrown by FileDownloader objects if they detect |
421 | multiple files would have to be downloaded to the same file on disk. | |
422 | """ | |
423 | pass | |
d77c3dfd FV |
424 | |
425 | ||
426 | class PostProcessingError(Exception): | |
59ae15a5 | 427 | """Post Processing exception. |
d77c3dfd | 428 | |
59ae15a5 PH |
429 | This exception may be raised by PostProcessor's .run() method to |
430 | indicate an error in the postprocessing task. | |
431 | """ | |
432 | pass | |
d77c3dfd FV |
433 | |
434 | class MaxDownloadsReached(Exception): | |
59ae15a5 PH |
435 | """ --max-downloads limit has been reached. """ |
436 | pass | |
d77c3dfd FV |
437 | |
438 | ||
439 | class UnavailableVideoError(Exception): | |
59ae15a5 | 440 | """Unavailable Format exception. |
d77c3dfd | 441 | |
59ae15a5 PH |
442 | This exception will be thrown when a video is requested |
443 | in a format that is not available for that video. | |
444 | """ | |
445 | pass | |
d77c3dfd FV |
446 | |
447 | ||
448 | class ContentTooShortError(Exception): | |
59ae15a5 | 449 | """Content Too Short exception. |
d77c3dfd | 450 | |
59ae15a5 PH |
451 | This exception may be raised by FileDownloader objects when a file they |
452 | download is too small for what the server announced first, indicating | |
453 | the connection was probably interrupted. | |
454 | """ | |
455 | # Both in bytes | |
456 | downloaded = None | |
457 | expected = None | |
d77c3dfd | 458 | |
59ae15a5 PH |
459 | def __init__(self, downloaded, expected): |
460 | self.downloaded = downloaded | |
461 | self.expected = expected | |
d77c3dfd FV |
462 | |
463 | ||
0b8c922d | 464 | class Trouble(Exception): |
59ae15a5 | 465 | """Trouble helper exception |
dffe658b | 466 | |
59ae15a5 PH |
467 | This is an exception to be handled with |
468 | FileDownloader.trouble | |
469 | """ | |
0b8c922d | 470 | |
01ba00ca | 471 | class YoutubeDLHandler(compat_urllib_request.HTTPHandler): |
59ae15a5 PH |
472 | """Handler for HTTP requests and responses. |
473 | ||
474 | This class, when installed with an OpenerDirector, automatically adds | |
475 | the standard headers to every HTTP request and handles gzipped and | |
476 | deflated responses from web servers. If compression is to be avoided in | |
477 | a particular request, the original request in the program code only has | |
478 | to include the HTTP header "Youtubedl-No-Compression", which will be | |
479 | removed before making the real request. | |
480 | ||
481 | Part of this code was copied from: | |
482 | ||
483 | http://techknack.net/python-urllib2-handlers/ | |
484 | ||
485 | Andrew Rowls, the author of that code, agreed to release it to the | |
486 | public domain. | |
487 | """ | |
488 | ||
489 | @staticmethod | |
490 | def deflate(data): | |
491 | try: | |
492 | return zlib.decompress(data, -zlib.MAX_WBITS) | |
493 | except zlib.error: | |
494 | return zlib.decompress(data) | |
495 | ||
496 | @staticmethod | |
497 | def addinfourl_wrapper(stream, headers, url, code): | |
498 | if hasattr(compat_urllib_request.addinfourl, 'getcode'): | |
499 | return compat_urllib_request.addinfourl(stream, headers, url, code) | |
500 | ret = compat_urllib_request.addinfourl(stream, headers, url) | |
501 | ret.code = code | |
502 | return ret | |
503 | ||
504 | def http_request(self, req): | |
505 | for h in std_headers: | |
506 | if h in req.headers: | |
507 | del req.headers[h] | |
508 | req.add_header(h, std_headers[h]) | |
509 | if 'Youtubedl-no-compression' in req.headers: | |
510 | if 'Accept-encoding' in req.headers: | |
511 | del req.headers['Accept-encoding'] | |
512 | del req.headers['Youtubedl-no-compression'] | |
513 | return req | |
514 | ||
515 | def http_response(self, req, resp): | |
516 | old_resp = resp | |
517 | # gzip | |
518 | if resp.headers.get('Content-encoding', '') == 'gzip': | |
519 | gz = gzip.GzipFile(fileobj=io.BytesIO(resp.read()), mode='r') | |
520 | resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code) | |
521 | resp.msg = old_resp.msg | |
522 | # deflate | |
523 | if resp.headers.get('Content-encoding', '') == 'deflate': | |
524 | gz = io.BytesIO(self.deflate(resp.read())) | |
525 | resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code) | |
526 | resp.msg = old_resp.msg | |
527 | return resp | |
0f8d03f8 PH |
528 | |
529 | https_request = http_request | |
530 | https_response = http_response |