http://bugzilla.spamassassin.org/show_bug.cgi?id=4132
------- Additional Comments From [EMAIL PROTECTED] 2005-02-11 10:46 ------- Even for HTML, encoding is compulsory if the ampersand is followed by letters or #, which could be the start of an entity. Try running your HTML through a validator, and you'll see. People remain unaware of it because you can usually get away with it because of browser error correction. The problem is that eventually you'll run into a URL where the letters following the ampersand happen to be an entity name, and then things will break. Worse yet, you could have URLs that work okay now but then start failing years from now when new browsers come out that understand a new entity. You avoid all this if you write valid HTML in the first place, which means encoding ampersands followed by such characters, and it's far easier just to encode all ampersands, even if there are a few (like ones followed by spaces) that you can technically avoid encoding. Also, browser error correction differs. If you have something like "¬ify=1" in a URL, some browsers will interpret the "¬" as an entity even though there's no "word boundary" after it. Others won't, even if they recognize the "not" entity. ------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee.
