Nancy Johnson wrote:> I have a question related to the same issue.> > I am
converting a very poorly constructed table based site to CSS.> > It is a tight
fixed width site. Since it is a child of a larger site,> I don't feel I can
convert to a more fluid site or change the look and> feel.> > I have used> /***
IE Fix ***/> * html> > to acommodate both IE6 and IE7 issues which seem to
work.> > My more recent readings indicate that this method isn't isn't>
recommended as a css fix and doesn't fix IE7, although it seemed to in> most
instances> > I also had one instance where IE6 needed fixing and not IE7.
*html fixed both.> > Is @import?> > or> > <!--[if IE.xx]>> <link
rel="stylesheet" type="text/css" media="screen"> href="/css/IEstylesheet.css"
/> <![endif]-->> > a better way?> > Thanks in advance,> > NancyHi Nancy. This
is a hotly disputed topic with a myriad of threads in the list archives and a
special page in this list's wiki "To hack or not to hack" [1] or there is also
much debate whether to use CSS hacks or IE Conditional Comments in the html. If
you are working for a client and delivering a product which you will not be
around to maintain, then the safest method is to use IE Conditional Comments.
This could be done also by.<!--[if IE]><style type="text/css">@import
"ie-style.css";</style><![endif]-->and separating the particular versions of IE
in the one stylesheet. This above import filter will be hidden from IE 4. A
full list of import hacks [2] show various methods.If the pages are always
maintained by yourself. This method may work better for you.* html #content
{...}This will target IE 5~6 and IE 7 in quirks mode [3] and for IE 7 in
standards mode [3] this works.*+html /* target IE 7 without the xml prolog
*/*:first-child+html /* target IE 7 with the xml prolog */Note that with the
xml prolog, Opera will also be targeted by similar hacks meant only for IE
[3].Hacks like.*html #content {...}or #content {*background:red;}cause parsing
errors so whatever style declarations are included may not be able to be
validated.All the import hacks that only target IE are invalid so use with
caution. When using any hacks it is wise to know the full consequence of the
method used. Some of the hacks that target IE/win 5~6 may also target IE/mac.
There is one well know method to hide these rules from IE/mac or only show
rules for IE/mac. When using any hacks it is good practice to comment what
browsers are targeted and why the have been used. A good is resource of many
methods with the pro and cons are found here [4].There is also the common
inappropriate use of hacks. Some authors will go straight for a hack if a part
of a page layout has a few pixels in difference. Browsers default settings are
always slightly difference. There could also be a scenario like.<div
id="sidebar">.....</div><div id="content">.....</div>#sidebar
{float:left;}#content {width:600px;margin-left:200px;}Instead of hacking, one
solution is to avoid the effects of hasLayout [4] in IE by removing the
haslayout trigger of width.#content {margin-left:200px;}I personally use * html
and *:first-child+html since I prefer to keep the hacks together with the rules
they relate to. I use the xml prolog to send IE 6 into quirks mode so I do not
have to hack for IE 5 separately. I generally find that I do "not" have to hack
for IE7, but this also means that I am also coding to keep IE happy. This is
something I'm slowly moving away from though.I finding early 2008 to be very
hacky time, but with the coming release of IE8 early this year. This could
become very important. Good luck with your project.1.
<http://css-discuss.incutio.com/?page=ToHackOrNotToHack>2.
<http://imfo.ru/csstest/css_hacks/import.php>3.
<http://css-class.com/test/css/selectors/ie7hacktargetingopera.htm>4.
<http://www.webdevout.net/css-hacks>Alanhttp://css-class.com/
______________________________________________________________________
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/
