21897 2013-05-02 01:18:44 +0000 Definition of 'valid URLs' points to The URL standard, which lacks requirement to escape spaces 2013-05-02 11:26:08 +0000 1 1 1 Unclassified HTML WG HTML5 spec unspecified PC Windows 3.1 CLOSED WORKSFORME http://www.w3.org/html/wg/drafts/html/master/infrastructure.html#valid-url P2 normal --- 1 xn--mlform-iua dave.null annevk mike public-html-admin public-html-wg-issue-tracking shadow2531 xn--mlform-iua public-html-bugzilla oldest_to_newest 87134 0 xn--mlform-iua 2013-05-02 01:18:44 +0000 ISSUE: 1) NU validator considers it an error if a string that is supposed to be a URL, contains an unescaped space character. (Percentage escaping of valid URLs is common knowledge.) 2) HTML5 says "A URL is a valid URL if it conforms to the authoring conformance requirements in the URL standard." http://www.w3.org/html/wg/drafts/html/master/infrastructure.html#valid-url 3) HOWEVER, the URL standard does has no requirement that spaces are written as percentage-encoded. NOTE: Probably relates to other characters that needs escape too. PROPOSAL: * If the URL standards editor plans to add this requirement, then clarify that this is currently not defined by the URL standard. * If the URL standard editor has no such plans, then define the requirement in HTML 5.1. (HTML5 CR does not have this issue I believe.) 87146 1 annevk 2013-05-02 09:24:23 +0000 U+0020 is not a URL unit (unless you write it %20). 87159 2 xn--mlform-iua 2013-05-02 11:26:08 +0000 (In reply to comment #1) > U+0020 is not a URL unit (unless you write it %20). My bad ... for not seeing that the 'URL code points' paragraph does not list the space character. Think the section ought to explicitly mention - perhaps in a note - that U+0009, U+000A, and U+000D are to be escaped. Currently, the way to grok it is to understand that space is not listed in the code points list. But that's a bug against the URL standard.