ChCh/CloudBot
Archived
11
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

Fixed title plugin

Browse source
This commit is contained in:
Luke Rogers 2012-10-12 22:34:48 +13:00
parent 00d80f0122
commit cf377e1800
2 changed files with 9 additions and 8 deletions
Download patch file Download diff file Expand all files Collapse all files

7
plugins/util/urlnorm.py
View file

@ -42,12 +42,12 @@ normalizers = ( Normalizer( re.compile(r'(?:https?://)?(?:[a-zA-Z0-9\-]+\.)?(?:a
lambda m: r'http://amazon.%s/dp/%s' % (m.group('tld'), m.group('ASIN'))),
Normalizer( re.compile(r'.*waffleimages\.com.*/([0-9a-fA-F]{40})'),
lambda m: r'http://img.waffleimages.com/%s' % m.group(1) ),
Normalizer( re.compile(r'(?:youtube.*?(?:v=|/v/)|youtu\.be/|yooouuutuuube.*?id=)([-_a-z0-9]+)'),
Normalizer( re.compile(r'(?:youtube.*?(?:v=|/v/)|youtu\.be/|yooouuutuuube.*?id=)([-_a-zA-Z0-9]+)'),
lambda m: r'http://youtube.com/watch?v=%s' % m.group(1) ),
)
def normalize(url):
def normalize(url, assume_scheme=False):
"""Normalize a URL."""
scheme, auth, path, query, fragment = urlparse.urlsplit(url.strip())
@ -69,6 +69,9 @@ def normalize(url):
scheme = "http"
path = path[4:]
if assume_scheme and not scheme:
scheme = assume_scheme.lower()
# Only perform percent-encoding where it is essential.
# Always use uppercase A-through-F characters when percent-encoding.
# All portions of the URI must be utf-8 encoded NFC from Unicode strings