From d05cfe06006c4a44032e95dde047d5e12be8674c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Sat, 13 Sep 2014 20:59:16 +0700 Subject: [YoutubeDL/utils] Clarify rationale for URL escaping in comment, move escape routines to utils and add some tests --- youtube_dl/utils.py | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) (limited to 'youtube_dl/utils.py') diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index d920c65a4..9124c3621 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -1418,6 +1418,24 @@ def uppercase_escape(s): lambda m: unicode_escape(m.group(0))[0], s) + +def escape_rfc3986(s): + """Escape non-ASCII characters as suggested by RFC 3986""" + if sys.version_info < (3, 0) and isinstance(s, unicode): + s = s.encode('utf-8') + return compat_urllib_parse.quote(s, "%/;:@&=+$,!~*'()?#[]") #"%/;:@&=+$,!~*'()?#[]+" #?#[]+ + + +def escape_url(url): + """Escape URL as suggested by RFC 3986""" + url_parsed = compat_urllib_parse_urlparse(url) + return url_parsed._replace( + path=escape_rfc3986(url_parsed.path), + params=escape_rfc3986(url_parsed.params), + query=escape_rfc3986(url_parsed.query), + fragment=escape_rfc3986(url_parsed.fragment) + ).geturl() + try: struct.pack(u'!I', 0) except TypeError: -- cgit v1.2.3