Gunlaug Sørtun wrote: > Alan Gresley wrote: >> Opera 9.5 seems not to recover from this import until the second ruleset. >> >> @import --url("chaos-060801-test6.css"); >> >> Here's a demo. >> >> http://css-class.com/test/css/imports/import-invalid-opera950.htm > > So I've noticed. Now, what does the CSS specs and/or the latest > discussions on [EMAIL PROTECTED] say about the correctness of this new > error-recovery behavior? > > I do not want to call it a bug if it isn't one, and I don't have time to > do in-depth research on what I would call a marginal case. > I think browsers have much freedom in how they recover from non-valid > rules, but my test shows that Opera won't recover even after a valid > @import - it has to be a regular valid style-rule. > > I seem to remember that Opera 6 and earlier needed similar dummy-rules > after certain hacks - the original "be nice to Opera" solutions was one > such case. Haven't seen much need for such dummy-rules since Opera 7.2 > was released though. > > Georg
I think it's a bug (don't put total trust in me) since the CSS parser should stop error recovery at the semicolon *;*. This is indicated in the specs [1]. "User agents must ignore an invalid at-keyword together with everything following it, up to and including the next semicolon (;) or block ({...}), whichever comes first." Both your test and my test shows that Opera 9.5 eats the next at rule or ruleset before recovering. [1] <http://www.w3.org/TR/CSS21/syndata.html#parsing-errors> -- Alan http://css-class.com/ Nearly all men can stand adversity, but if you want to test a man's character, give him power - Abraham Lincoln ______________________________________________________________________ css-discuss [EMAIL PROTECTED] http://www.css-discuss.org/mailman/listinfo/css-d List wiki/FAQ -- http://css-discuss.incutio.com/ List policies -- http://css-discuss.org/policies.html Supported by evolt.org -- http://www.evolt.org/help_support_evolt/