25871 2014-05-23 03:02:14 +0000 What's the "character that caused the state machine…" when it was a branch on EOF? What about the transition from markup declaration open state which does lookahead of multiple charaters, so what's "the" character? 2014-08-05 19:34:51 +0000 1 1 1 Unclassified WHATWG HTML unspecified Other other RESOLVED FIXED http://www.whatwg.org/specs/web-apps/current-work/#bogus-comment-state P3 normal Unsorted 1 contributor ian geoffers+w3cbugs ian mike contributor oldest_to_newest 106690 0 contributor 2014-05-23 03:02:14 +0000 Specification: http://www.whatwg.org/specs/web-apps/current-work/multipage/tokenization.html Multipage: http://www.whatwg.org/C#bogus-comment-state Complete: http://www.whatwg.org/c#bogus-comment-state Referrer: http://www.whatwg.org/specs/web-apps/current-work/multipage/ Comment: What's the "character that caused the state machine…" when it was a branch on EOF? What about the transition from markup declaration open state which does lookahead of multiple charaters, so what's "the" character? Posted from: 86.30.234.99 by geoffers@gmail.com User agent: Mozilla/5.0 (X11; Linux x86_64; rv:29.0) Gecko/20100101 Firefox/29.0 106734 1 ian 2014-05-23 19:52:30 +0000 EOF is a character. See the last paragraph of: http://www.whatwg.org/specs/web-apps/current-work/#preprocessing-the-input-stream What's the case where you enter the bogus comment state from a string? 106742 2 geoffers+w3cbugs 2014-05-23 21:54:09 +0000 Then how do we include an EOF character in a comment token? Surely every UA is non-conforming given they don't include the EOF character in that case? And the second case was me misreading the spec, so ignore that. 106818 3 ian 2014-05-27 06:10:50 +0000 Does the parenthetical at the end of the paragraph not sufficiently cover this? Maybe I should remove the parentheses? 106862 4 geoffers+w3cbugs 2014-05-27 22:27:25 +0000 That sounds to me like it's meant to be restating what has already been stated, and therefore contradicting the above, given the token should include the character that caused the transition (i.e., EOF) while also being empty. 106886 5 ian 2014-05-28 05:37:33 +0000 Well right now the other text says "the characters starting from and including the character that caused the state machine to switch into the bogus comment state, up to and including the character immediately before the last consumed character", which is a negative-length string in the case of "<!" followed by EOF. The parenthetical was trying to clean that up. But I can try to improve this. 109955 6 ian 2014-08-05 19:33:57 +0000 Please reopen if the fix isn't sufficient. 109956 7 contributor 2014-08-05 19:34:51 +0000 Checked in as WHATWG revision r8709. Check-in comment: Try to avoid negative-length strings and strings with EOF characters in them... http://html5.org/tools/web-apps-tracker?from=8708&to=8709