9280 2006-06-02 15:10:15 -0700 Quirksmode: Trailing ]> from internal DTD subset is rendered as body text 2012-08-15 19:20:15 -0700 1 1 1 Unclassified WebKit DOM 420+ PC OS X 10.4 RESOLVED WONTFIX P2 Normal --- 9610 1 adele webkit-unassigned abarth ap bugs-webkit chortos fhp.mail hyatt ian oldest_to_newest 44464 0 adele 2006-06-02 15:10:15 -0700 http://www.quirksmode.org/bugreports/archives/safari/index.html "When creating custom DTDs like the one below, all browsers except Opera see the end of the ATTLIST as the end of the DOCTYPE. The result is that they print "]>" on the screen. <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd" [ <!ATTLIST p behavior CDATA #IMPLIED> ]> Test page. http://www.quirksmode.org/oddsandends/dtd.html Workaround is not included Reported by ppk." 47036 1 joost 2006-06-24 14:17:58 -0700 Moving this into component XML, though i'm not quite sure it's the correct one.. 47046 2 joost 2006-06-24 14:27:42 -0700 Changing to correct component. 48845 3 9274 ap 2006-07-08 10:41:17 -0700 Created attachment 9274 html test case The original test case is served as text/html, but has a DTD based on XHTML 1.0 Transitional, which has caused some concerns about its correctness (see the comments at quirksmode.org). The attached test case has a DTD based on HTML 4.01 Strict; it exposes the same behavior - the validator accepts it (and parses as expected, which can be seen in its outline view), while major browsers don't. I am not sure whether this document really is valid HTML - its DTD is not HTML 4.01 Strict, since it has an internal subset. HTML 4.01 explicitly lists the three DTDs used in HTML documents, and this is not one of them. I have found discussions of this issue from as early as 1995 <http://1997.webhistory.org/www.lists/www-html.1995q2/0379.html>, but nothing authoritative enough to ignore the results of the W3C validator. Mozilla documentation seems to imply that they do handle internal subsets in HTML <http://developer.mozilla.org/en/docs/Mozilla's_DOCTYPE_sniffing>, but I'm not seeing this. 107895 4 chortos 2009-02-01 08:58:43 -0800 *** Bug 23673 has been marked as a duplicate of this bug. *** 107897 5 chortos 2009-02-01 09:06:21 -0800 This bug is still there. Both HTML and XHTML are affected. Besides, I think a better summary would be what I wrote in my bug report (before finding this one): Trailing ]> from internal DTD subset is rendered as body text 107904 6 ap 2009-02-01 10:00:37 -0800 Good suggestion, renaming. We need to check what HTML5 has to say about this issue. Note however that I cannot reproduce this with XHTML (and the document at URL given in bug 23673 is plain HTML, see <http://hixie.ch/advocacy/xhtml> for more detail). 107908 7 chortos 2009-02-01 10:32:10 -0800 My fault. Sorry about that, should be fixed now. Actually... I have to admit that the bug indeed does not appear when the page is XHTML. 281414 8 abarth 2010-09-20 00:35:49 -0700 We're parsing this correct w.r.t. HTML5. 696638 9 ap 2012-08-15 19:20:15 -0700 *** Bug 94149 has been marked as a duplicate of this bug. *** 9274 2006-07-08 10:41:17 -0700 2006-07-08 10:41:17 -0700 html test case entity.html text/html 298 ap PCFET0NUWVBFIEhUTUwgUFVCTElDICItLy9XM0MvL0RURCBIVE1MIDQuMDEvL0VOIgogICAgICAg ICAgICAiaHR0cDovL3d3dy53My5vcmcvVFIvaHRtbDQvc3RyaWN0LmR0ZCIKWwo8IUVOVElUWSBu dyAibm93aGVyZS5jb20iPgpdPgo8aHRtbD4KPGhlYWQ+CjxtZXRhIGh0dHAtZXF1aXY9IkNvbnRl bnQtVHlwZSIgY29udGVudD0idGV4dC9odG1sO2NoYXJzZXQ9aXNvLTg4NTktMSI+Cjx0aXRsZT5U ZXN0IG9mIGV4dHJhIHN0dWZmPC90aXRsZT4KPC9oZWFkPgo8Ym9keT4KPHA+Jm53OzwvcD4KPC9i b2R5Pgo8L2h0bWw+Cg==