Merge remote-tracking branch 'dstftw/escape-non-ascii-in-urls'

[youtube-dl] / youtube_dl / utils.py

diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index d920c65a4df4f006ad882a7dc885571182b3aa38..b644f4e920bf0353658ec9920abdb0541dbaf0e2 100644 (file)
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -280,6 +280,11 @@ if sys.version_info >= (2, 7):
         return node.find(expr)
 else:
     def find_xpath_attr(node, xpath, key, val):
+        # Here comes the crazy part: In 2.6, if the xpath is a unicode,
+        # .//node does not match if a node is a direct child of . !
+        if isinstance(xpath, unicode):
+            xpath = xpath.encode('ascii')
+
         for f in node.findall(xpath):
             if f.attrib.get(key) == val:
                 return f
@@ -299,6 +304,20 @@ def xpath_with_ns(path, ns_map):
     return '/'.join(replaced)
 
 
+def xpath_text(node, xpath, name=None, fatal=False):
+    if sys.version_info < (2, 7):  # Crazy 2.6
+        xpath = xpath.encode('ascii')
+
+    n = node.find(xpath)
+    if n is None:
+        if fatal:
+            name = xpath if name is None else name
+            raise ExtractorError('Could not find XML element %s' % name)
+        else:
+            return None
+    return n.text
+
+
 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
 class BaseHTMLParser(compat_html_parser.HTMLParser):
     def __init(self):
@@ -1418,6 +1437,24 @@ 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:
@@ -1552,3 +1589,13 @@ except AttributeError:
         if ret:
             raise subprocess.CalledProcessError(ret, p.args, output=output)
         return output
+
+
+def limit_length(s, length):
+    """ Add ellipses to overly long strings """
+    if s is None:
+        return None
+    ELLIPSES = '...'
+    if len(s) > length:
+        return s[:length - len(ELLIPSES)] + ELLIPSES
+    return s