Re: [css-d] parse error on *html
On 29/07/2011 6:36 PM, Mark Henderson wrote: Philippe wrote: In a decent browser, it won't select anything and nothing though... That hack relies on the fact that the simple little mind of IE 6&7 ‘thinks’ there is an element that wraps around the root element (html). Of course there is none. And Alan responded: This is not correct. For IE6 and earlier, there is an element outside . [..] This means that IE6- sees this. ... Perhaps this comes down to interpretation. Putting things in a different light, let's assume that Philippe was talking with regards to the specifications (that's how I took it) and they say that html _IS_ the root. http://www.w3.org/TR/html4/struct/global.html#h-7.3 http://en.wikipedia.org/wiki/HTML_element#Document_structure_elements So it doesn't matter that the IE developers forgot they had implemented some hidden node outside of, to my mind what Philippe said /was/ still correct. Just an opinion. I should have said that it is not completely correct about there being no node wrapping around in IE6-. The rest of what Philippe said is correct. -- Alan Gresley http://css-3d.org/ http://css-class.com/ __ css-discuss [css-d@lists.css-discuss.org] 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/
Re: [css-d] parse error on *html
That is what I meant. "if you know it is correct" as in 'I know it's there and why'. I guess I could have been clearer. Sent from my iPhone On Jul 29, 2011, at 4:17 AM, "Philip TAYLOR (Webmaster, Ret'd)" wrote: > > > > Tom Livingston wrote: > > > Though I'm rusty on *HTML hacks- havent used one in years - > > the validator is a tool, not law. If you know that is correct > > and need it for a fix, then it's fine. > > With respect, the validator is more likely to know if something > is "correct" than a mere human; I would be inclined to rephrase > your observation along the lines of : > > > the validator is a tool, not law. If you need it for a fix, > > then go ahead and use it, but be aware that the validator > > told you it was an error for a very good reason. > > Philip Taylor > __ css-discuss [css-d@lists.css-discuss.org] 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/
Re: [css-d] parse error on *html
Philippe wrote: > In a decent browser, it won't select anything and nothing though... > That hack relies on the fact that the simple little mind of IE 6& 7 > ‘thinks’ there is an element that wraps around the root element > (html). Of course there is none. And Alan responded: > This is not correct. For IE6 and earlier, there is an element outside > . [..] > This means that IE6- sees this. > > > ... > > Perhaps this comes down to interpretation. Putting things in a different light, let's assume that Philippe was talking with regards to the specifications (that's how I took it) and they say that html _IS_ the root. http://www.w3.org/TR/html4/struct/global.html#h-7.3 http://en.wikipedia.org/wiki/HTML_element#Document_structure_elements So it doesn't matter that the IE developers forgot they had implemented some hidden node outside of , to my mind what Philippe said /was/ still correct. Just an opinion. -- /*Mark __ css-discuss [css-d@lists.css-discuss.org] 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/
Re: [css-d] parse error on *html
Tom Livingston wrote: > Though I'm rusty on *HTML hacks- havent used one in years - > the validator is a tool, not law. If you know that is correct > and need it for a fix, then it's fine. With respect, the validator is more likely to know if something is "correct" than a mere human; I would be inclined to rephrase your observation along the lines of : > the validator is a tool, not law. If you need it for a fix, > then go ahead and use it, but be aware that the validator > told you it was an error for a very good reason. Philip Taylor __ css-discuss [css-d@lists.css-discuss.org] 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/
Re: [css-d] parse error on *html
On 29/07/2011 12:27 PM, Philippe Wittenbergh wrote: On Jul 29, 2011, at 9:45 AM, David Laakso wrote: There should be a space between the star and html: * html .whatever {...} Indeed. This will not make it valid. Nope. * html {} is perfectly valid (CSS1, 2, 3, 4, and beyond). In a decent browser, it won't select anything and nothing though... That hack relies on the fact that the simple little mind of IE 6& 7 ‘thinks’ there is an element that wraps around the root element (html). Of course there is none. Philippe -- Philippe Wittenbergh http://l-c-n.com/ This is not correct. For IE6 and earlier, there is an element outside . I received this message offlist (not stating the list) in 2008. | I mean the fact that | * html {...} | Ever applied to anything was a bug. I wrote the | original selector-matching code that had this bug | in it. We had a special hidden "root container" | node in our element tree for some internal reasons, | and I forgot to explicitly skip it in the selector | node matching code. We didn't discover this for | years. :) This means that IE6- sees this. ... This bug was partly corrected in IE7 but the universal selector '*' was now selecting the as if it was an element. This is why a sibling selecting was needed in the IE7 only hack. *+html { /* IE7 style */ } + For anyone following this thread, the below article gives some history of IE hacks (with links). The article is worded to suit the times of IE8 beta release. http://css-class.com/test/bugs/ie/ie-hacks.htm -- Alan Gresley http://css-3d.org/ http://css-class.com/ __ css-discuss [css-d@lists.css-discuss.org] 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/
Re: [css-d] parse error on *html
On Jul 29, 2011, at 9:45 AM, David Laakso wrote: > There should be a space between the star and html: > > * html .whatever {...} Indeed. > This will not make it valid. Nope. * html {} is perfectly valid (CSS1, 2, 3, 4, and beyond). In a decent browser, it won't select anything and nothing though... That hack relies on the fact that the simple little mind of IE 6 & 7 ‘thinks’ there is an element that wraps around the root element (html). Of course there is none. Philippe -- Philippe Wittenbergh http://l-c-n.com/ __ css-discuss [css-d@lists.css-discuss.org] 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/
Re: [css-d] parse error on *html
Mr. Modo is correct, none of these suggestions will validate but I was just suggesting an alternate method to achieving the same end. Some people do sleep better having the offending code outside their main sheet so it does pass the validator. Merely my opinion and another possible course to take. HTH Sent from my iPhone On Jul 28, 2011, at 8:45 PM, David Laakso wrote: > On 7/28/11 8:21 PM, John wrote: >> Just ran the latest version of my page through CSS validator, and it came up >> with 1 error: >> >> Parse Error *html .livebar{height:100%; /*For IE6 since overflow:auto does >> not trigger hasLayout*/} >> >> Do I need to be concerned about this? This is a bit of code which makes IE6 >> behave itself, correct, and as such, the validator might be choking on it, >> tho it is not actually a problem? >> >> the page in question is: >> >> http://coffeeonmars.com/test/16NewIndex.html >> >> >> thank you, >> >> John > > > > > John, > > There should be a space between the star and html: > > * html .whatever {...} > > This will not make it valid. > Wrapping it in a conditional comment won't make it valid, either [but the > validator will not see it]. > Putting it a separate style sheet and wrapping it in a conditional comment > won't make it valid, either [but the validator will not see it]. > > My personal opinion is simply to leave as the last item in your style sheet > and not lose any sleep over it not being valid... > > Best, > ~quasimodo > > > > -- > http://chelseacreekstudio.com/ > http://chelseacreekstudio.com/fa/ > > __ > css-discuss [css-d@lists.css-discuss.org] > 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/ __ css-discuss [css-d@lists.css-discuss.org] 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/
Re: [css-d] parse error on *html
On 29 July 2011 12:21, John wrote: > Just ran the latest version of my page through CSS validator, and it came up > with 1 error: > > Parse Error *html .livebar{height:100%; /*For IE6 since overflow:auto does > not trigger hasLayout*/} I'm pretty sure it should have a space there, as in * html, not *html. Nothing wrong with * html hacks as long as you understand the implications, and they are inside a @media block at the END of the style sheet, although I'm personally with Tom and prefer conditional comments. HTH -- /*Mark __ css-discuss [css-d@lists.css-discuss.org] 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/
Re: [css-d] parse error on *html
On 7/28/11 8:21 PM, John wrote: Just ran the latest version of my page through CSS validator, and it came up with 1 error: Parse Error *html .livebar{height:100%; /*For IE6 since overflow:auto does not trigger hasLayout*/} Do I need to be concerned about this? This is a bit of code which makes IE6 behave itself, correct, and as such, the validator might be choking on it, tho it is not actually a problem? the page in question is: http://coffeeonmars.com/test/16NewIndex.html thank you, John John, There should be a space between the star and html: * html .whatever {...} This will not make it valid. Wrapping it in a conditional comment won't make it valid, either [but the validator will not see it]. Putting it a separate style sheet and wrapping it in a conditional comment won't make it valid, either [but the validator will not see it]. My personal opinion is simply to leave as the last item in your style sheet and not lose any sleep over it not being valid... Best, ~quasimodo -- http://chelseacreekstudio.com/ http://chelseacreekstudio.com/fa/ __ css-discuss [css-d@lists.css-discuss.org] 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/
Re: [css-d] parse error on *html
Though I'm rusty on *HTML hacks- havent used one in years - the validator is a tool, not law. If you know that is correct and need it for a fix, then it's fine. However, I'd recommend another method for teaching IE 6 a lesson, and wrap it in conditional comments. Either embedded in the head, or an external IE sheet. HTH Sent from my iPhone On Jul 28, 2011, at 8:21 PM, John wrote: > Just ran the latest version of my page through CSS validator, and it came up > with 1 error: > > Parse Error *html .livebar{height:100%; /*For IE6 since overflow:auto does > not trigger hasLayout*/} > > Do I need to be concerned about this? This is a bit of code which makes IE6 > behave itself, correct, and as such, the validator might be choking on it, > tho it is not actually a problem? > > the page in question is: > > http://coffeeonmars.com/test/16NewIndex.html > > > thank you, > > John > __ > css-discuss [css-d@lists.css-discuss.org] > 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/ __ css-discuss [css-d@lists.css-discuss.org] 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/
[css-d] parse error on *html
Just ran the latest version of my page through CSS validator, and it came up with 1 error: Parse Error *html .livebar{height:100%; /*For IE6 since overflow:auto does not trigger hasLayout*/} Do I need to be concerned about this? This is a bit of code which makes IE6 behave itself, correct, and as such, the validator might be choking on it, tho it is not actually a problem? the page in question is: http://coffeeonmars.com/test/16NewIndex.html thank you, John __ css-discuss [css-d@lists.css-discuss.org] 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/