From 2adafdaadc057790cf06c16fa0840ee741f55a54 Mon Sep 17 00:00:00 2001 From: mailchuck Date: Sat, 12 Mar 2016 09:07:39 +0100 Subject: [PATCH] Improved URI detector --- src/bitmessageqt/safehtmlparser.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/bitmessageqt/safehtmlparser.py b/src/bitmessageqt/safehtmlparser.py index 78f89888..e56fd14c 100644 --- a/src/bitmessageqt/safehtmlparser.py +++ b/src/bitmessageqt/safehtmlparser.py @@ -21,7 +21,7 @@ class SafeHTMLParser(HTMLParser): 'th', 'thead', 'tr', 'tt', 'u', 'ul', 'var', 'video'] replaces = [["&", "&"], ["\"", """], ["<", "<"], [">", ">"], ["\n", "
"], ["\t", "    "], [" ", "  "], [" ", "  "], ["
", "
 "]] src_schemes = [ "data" ] - uriregex1 = re.compile(r'(\b(?:https?|telnet|gopher|file|wais|ftp):[\w/#~:.?+=&%@!\-.:;?\\-]+?(?=[.:?\-]*(?:[^\w/#~:;.?+=&%@!\-.:?\-]|$)))') + uriregex1 = re.compile(r'(?i)\b((?:(https?|ftp):(?:/{1,3}|[a-z0-9%])|www\d{0,3}[.]|[a-z0-9.\-]+[.][a-z]{2,4}/)(?:[^\s()<>]+|\(([^\s()<>]+|(\([^\s()<>]+\)))*\))+(?:\(([^\s()<>]+|(\([^\s()<>]+\)))*\)|[^\s`!()\[\]{};:\'".,<>?]))') uriregex2 = re.compile(r'