9983 2010-06-22 21:32:50 +0000 [parser] Insertion point not saved for external scripts 2010-10-09 15:35:22 +0000 1 1 1 Unclassified HTML WG pre-LC1 HTML5 spec (editor: Ian Hickson) unspecified All All VERIFIED INVALID P1 critical --- 1 w3c ian hsivonen ian mike Ms2ger public-html-admin public-html-wg-issue-tracking public-html-bugzilla oldest_to_newest 36356 0 w3c 2010-06-22 21:32:50 +0000 In step 8 of <http://www.whatwg.org/specs/web-apps/current-work/#running-a-script>, the spec adds external scripts to the <http://www.whatwg.org/specs/web-apps/current-work/#set-of-scripts-that-will-execute-as-soon-as-possible>. When the fetching algorithm is complete, the spec invokes <http://www.whatwg.org/specs/web-apps/current-work/#executing-a-script-block>. However, that sequence of steps does not save the current insertion point (among other things), which seems very unlikely to be compatible with the web. In implementing the above requirement in WebKit, we assumed the spec meant to invoke <http://www.whatwg.org/specs/web-apps/current-work/#scriptTagParserResumes>, which saves the current insertion point (and does other things that would make sense at that point). Should we change the spec to say this explicitly? 36363 1 ian 2010-06-23 06:57:06 +0000 The spec is unambiguous and intentional in its current requirements. The insertion point is only saved when the script is inserted by the parser. Do you have a specific example of something incompatible with the exact requirements in the spec? 36364 2 w3c 2010-06-23 07:02:22 +0000 Ah, I see. I misunderstood how "ready to be parser-executed" worked. Thanks.