[jQuery] Re: Body Browser Classes

2007-04-06 Thread d . wachss



On Apr 4, 10:48 pm, Glen Lipka [EMAIL PROTECTED] wrote:
 Feedback desired:

 !-- Old Internet Explorer -- Lately, I have been developing CSS
and HTML for a deep Web 2.0 complex app.
 Usually, I avoid CSS hacks like the plague.  But recently, I have had to
 resort to the Holly Hack or the StarHTML Hack.  But then it occured to me
 that jQuery provides a better way.

 A simple plugin could be written (has this already been written?) that tags
 the BODY (or other node) with a browser class resulting in:
 body class=FF or body class=IE6 or body class=Saf or whatever.
 Then your CSS would be:

 body.ie6 div.troublesome {height: 100%} rather than
 * html div.troublesome etc

 This makes your CSS avoid bizarre invalid hacks and use normal conditional
 classes that are self-documenting.  Everyone knows that body.IE6 means you
 are adjusting for browser differences.  And jQuery is much better at
 detection than crazy hacks.

There is an easier  way that is not jQuery-or-Javascript-dependent,
since mostly we just want to know is the browser IE  or not. Use
conditional comments around the BODY tag. See 
http://www.puidokas.com/updating-my-css-for-ie7/
. I use :
  !-- Old Internet Explorer --
  !--[if lt IE 7]
body class=IE IElt7
  ![endif]--
  !-- New Internet Explorer --
  !--[if gte IE 7]
body class=IE
  ![endif]--
  !-- Any other browser --
  !--[if !IE]--
body class=notIE 
  !--![endif]--

Danny Wachsstock



[jQuery] Re: Body Browser Classes

2007-04-06 Thread Glen Lipka



!--[if lt IE 7]
link rel=stylesheet href=ie.css type=text/css
![endif]--


I think the alternative is for people who want to keep their classes

together even if some of them are browser specific.
For example, in the app I am working on the CSS is enormous.  (dont ask)
And the whole thing is changing pretty rapidly.
And I am going to want to change certain things section-by-section.  In this
scenerio, I really do not want to seperate out an ie6.css.
I just want to comment in my css /* override for blah blah because IE6 sucks
*/.

The benefit of his method is that it doesn't use JS, which I thought was an
interesting implementation and avoids the depends on JS issue.

Glen


[jQuery] Re: Body Browser Classes

2007-04-05 Thread Nathan Young -X \(natyoung - Artizen at Cisco\)

Hi.

I love this idea, it's inventive and the need is there.

The only feedback I have is that CSS that depends on JS will absolutely
not fly where I work and play.

Nathan

 Feedback desired:
  
 Lately, I have been developing CSS and HTML for a deep Web 
 2.0 complex app.  Usually, I avoid CSS hacks like the plague. 
  But recently, I have had to resort to the Holly Hack or the 
 StarHTML Hack.  But then it occured to me that jQuery 
 provides a better way. 
  
 A simple plugin could be written (has this already been 
 written?) that tags the BODY (or other node) with a browser 
 class resulting in:
 body class=FF or body class=IE6 or body class=Saf 
 or whatever.  Then your CSS would be:
  
 body.ie6 div.troublesome {height: 100%} rather than
 * html div.troublesome etc
  
 This makes your CSS avoid bizarre invalid hacks and use 
 normal conditional classes that are self-documenting.  
 Everyone knows that body.IE6 means you are adjusting for 
 browser differences.  And jQuery is much better at detection 
 than crazy hacks. 
  
 I wish all my CSS could do it right and find the common 
 ground that all the browsers love.  But this seems like a better way.
  
 What is your opinion?
  
 Anyone want to write a plugin that allows for $(body).browserTag()?
 Personally, I think this would be a cool thing in the 
 basecode, but I wont push it.
  
 Glen
 


[jQuery] Re: Body Browser Classes

2007-04-05 Thread Klaus Hartl


Glen Lipka schrieb:
I hear everyone's points.  I think the easy to remove it in 2012 
argument doesn't resonate for me, since anything created today will be 
changed within 1-3 years (in my experience) and IE6 isn't going away 
that soon.  The screen-reader/mobile stuff seems like they would have a 
different CSS base than the main browser audience anyway, so wouldn't 
applicable.  The conditional comments are good, but need seperate CSS 
and limit the option NOT to do that.  JS being required is a good 
point, but I guess it depends on how much tweaking versus wholesale 
changes are used.  With our without hacks.


I think, you overlooked the delivering smaller style sheets to modern 
browsers argument, which is to me more valid than the easy to remove one.


Although I wonder, you were saying that hacks feel dirty, doesn't a 
style sheet messed up with hacks feel much more dirty than two separate 
style sheets, of which one is free from hacks?


To avoid overlooking properties that are overruled in IE's bugsheet is 
just a matter of proper commenting your style sheets.



Last not least I recommend to all using hacks to read

1. Molly about Long term hack strategies:
http://www.informit.com/articles/article.asp?p=170511rl=1

2. Tantek - inventor of the CSS hack! - about hacks:
http://tantek.com/log/2005/11.html


-- Klaus


[jQuery] Re: Body Browser Classes

2007-04-05 Thread Glen Lipka

Good Articles.  I found it interesting to think about that hacks should be
ugly.

I think your point about browser specific sheets is well taken.  Firebug
makes this a good thing when you see what styles are applied and can see
which file they are in.  I wish IE had similar.

Another article I found interesting was how OpenLazlo deals with this:
http://weblog.openlaszlo.org/archives/category/documentation/

Wouldn't it be cool if you could invlude a jQuery plugin that made all the
browsers act the same for css?  I know it would be dependant on JS, but it
would make development of alot of projects much faster.  Lots of projects I
see don't let you in without JS.

Glen


On 4/5/07, Klaus Hartl [EMAIL PROTECTED] wrote:



Last not least I recommend to all using hacks to read

1. Molly about Long term hack strategies:
http://www.informit.com/articles/article.asp?p=170511rl=1

2. Tantek - inventor of the CSS hack! - about hacks:
http://tantek.com/log/2005/11.html


-- Klaus