diff options
| author | Sergey M․ <dstftw@gmail.com> | 2017-03-22 01:59:27 +0700 |
|---|---|---|
| committer | Sergey M․ <dstftw@gmail.com> | 2017-03-22 01:59:27 +0700 |
| commit | e1e35d1ac66ab99202e8265ac811906de2aa87dd (patch) | |
| tree | 41b05849fdc68262e645e558158724ead08c0c8e | |
| parent | 21fbf0f955f584ad2d02608850a69a2fd74b65a6 (diff) | |
[pornhub] Improve extraction and style (closes #12515)
| -rw-r--r-- | youtube_dl/extractor/pornhub.py | 35 |
1 files changed, 15 insertions, 20 deletions
diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py index eb316ad14..b25f1f193 100644 --- a/youtube_dl/extractor/pornhub.py +++ b/youtube_dl/extractor/pornhub.py @@ -20,6 +20,7 @@ from ..utils import ( js_to_json, orderedSet, # sanitized_Request, + remove_quotes, str_to_int, ) # from ..aes import ( @@ -131,38 +132,32 @@ class PornHubIE(InfoExtractor): tv_webpage = dl_webpage('tv') - encoded_url = self._search_regex(r'(var.*mediastring.*)</script>', - tv_webpage, 'encoded url') - assignments = encoded_url.split(";") + assignments = self._search_regex( + r'(var.+?mediastring.+?)</script>', tv_webpage, + 'encoded url').split(';') + js_vars = {} def parse_js_value(inp): - inp = re.sub(r'/\*[^*]*\*/', "", inp) - - if "+" in inp: - inps = inp.split("+") - return functools.reduce(operator.concat, map(parse_js_value, inps)) - + inp = re.sub(r'/\*(?:(?!\*/).)*?\*/', '', inp) + if '+' in inp: + inps = inp.split('+') + return functools.reduce( + operator.concat, map(parse_js_value, inps)) inp = inp.strip() if inp in js_vars: return js_vars[inp] - - # Hope it's a string! - assert inp.startswith('"') and inp.endswith('"') - return inp[1:-1] + return remove_quotes(inp) for assn in assignments: assn = assn.strip() - if len(assn) == 0: + if not assn: continue - - assert assn.startswith("var ") - assn = assn[4:] - vname, value = assn.split("=", 1) - + assn = re.sub(r'var\s+', '', assn) + vname, value = assn.split('=', 1) js_vars[vname] = parse_js_value(value) - video_url = js_vars["mediastring"] + video_url = js_vars['mediastring'] title = self._search_regex( r'<h1>([^>]+)</h1>', tv_webpage, 'title', default=None) |