CSS - line-height:0 ignored in XHTML strict in Firefox - firefox

I'm using :first-line { line-height:0px; } to hide the empty first line in a block statement
curiously if I define doctype in XHTML strict Firefox ignores this statement and prints the empty line (seeable through the border)
I validated my quellcode and used Firefox' webdeveloper to check, if there is anything wrong, but they told no error
the (stripped) example quellcode is:
<?xml version="1.0" ?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml"><head><title>T</title>
<style type="text/css"><!--
.v { white-space:pre; display:inline-block; border:1px solid black; }
.v:first-line { line-height:0px; }
--></style>
</head><body>
<p class="v">
Foo
Bar
</p>
</body></html>
Is there anything wrong with my code and other browsers (I've tested Opera,Gnome,Safari) just ignore the strict, or is Firefox buggy abiding strict?

I'm not sure if it's really a bug …
I guess the internals of the :first-line selector fail to recognize an empty line with just a newline character in it, but I'm not sure if this is wanted or not. There is nothing inside the W3C Recommendations to prove this behaviour as wrong or right:
CSS 2.1 Specifications:
5.12.1 The :first-line pseudo-element
The :first-line pseudo-element applies special styles to the contents
of the first formatted line of a paragraph.
--
16.6 White space: the 'white-space' property
'white-space'
...
pre This value prevents user agents from collapsing sequences of white space. Lines are only broken at preserved newline characters.
If you're generating your HTML code dynamically just try to eliminate the empty line before outputting it. That's the only advice I can give at the moment.

Try using two colons instead of one. From Mozilla:
The :first-line CSS pseudo-element applies styles to the first line of
a element. However, the selector :first-line does not match any real
HTML element.

Related

WebVIew problems HTML5 doctype

Imagine a fairly simple HTML document
<!DOCTYPE HTML >
<html>
<body>
<table> <tr> <td> This is a test </td> </tr> </table>
</body>
</html>
where we are applying this css
body {background-color: ffffff;
font-size:100px;
font-style: normal;
font-family: MankSans,Arial,Helvetica, sans-serif;
}
In every single modern browser out there the font properties will apply. in my WebView however, they wont.
Looking around I found out that the problem is that the WebView goes into quirksmode, where the td tag lacks inheritance so the body css doesnt apply.
Im aware that there are ways around this, like using a more explicit 4.01 doctype, or adding this to the css
table, thead, tbody, tr, td, th {
font-size: inherit;
font-family: inherit;
}
these however dont fix the root of the problem, which is that WebView decides against all logic to suddenly "missunderstand" the HTML5 doctype and switch to quirksmode, when all other WebKit-based browsers (Safari, Chrome, etc) behave correctly.
Is there any way to programmatically fix this?
Ok, my stupid mistake
it turns out that the doctype is getting screwed because before sending the HTML document to the webview im treating it, and adding an internal ID to all tags (this internal ID is required for features inside my app)
but in case someone screws up in the same way I did and wonder whats going on, I'll offer a lenghty explanation.
DOMDocumentType has several fields, among them you have name, publicId and systemId.
for example, consider the typical HTML 4.01 transitional doctype
it can be divided in
name = html
publicId = -//W3C//DTD XHTML 1.0 Transitional//EN
systemID = http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd
and you can add a Custom ID to it as long as the format is preserved, like this
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd" CustomID="1">
the problem with the HTML 5 DOCTYPE is that while it looks like a)
<!DOCTYPE html>
for the DOM, it actually is b)
<!DOCTYPE html PUBLIC "" "">
long story short, adding my CustomID="1" at the end of a) caused the DOM to not recognize it as a valid DOCTYPE, hence switching my WebView into quirksmode. If you want to have an HTML5 DOCTYPE and add a custom id to it you should do
<!DOCTYPE html PUBLIC "" "" CustomId="1">
w3schools.com tells us that the id attribute is not valid in: base, head, html, meta, , script, style, and title.
they should warn us that, poorly placed, it can also screw your DOCTYPE =)

#font-face unicode-range attribute

In some html documents I'm using webfonts for only a couple of words. Performance-wise loading a complete font-file seems wasteful. This is where the unicode-range parameter of the #font-face declaration comes in:
http://www.w3.org/TR/css3-fonts/#unicode-range-desc
With it I supposedly can define what characters of the font-file to load, thus improving performance greatly.
But I just can't get it to work. And the odd thing is that it diplays all the characters in firefox, and it fails to load the font in safari just if I include the unicode-range parameter in my declaration.
Any help would be appreciated, below is the html I was testing it with:
<!doctype html>
<html lang="en">
<head>
<style text="text/css">
#font-face {
font-family: 'dream';
src: url(Fonts/Digital-dream/DIGITALDREAM.ttf) format("truetype");
unicode-range: U+FF21;
}
*{
font-family:dream;
font-weight:normal;
}
</style>
</head>
<body>
<p>ASDWEWQDSCF</p>
</body>
</html>
You are misunderstanding the purpose of that value. From that page:
This descriptor defines the range of Unicode characters supported by a given font
So this isn't the glyphs (or characters) to download, this is actually telling the browser what characters are in the font so that the browser can determine if its even worth downloading the font in the first place. If the text being styled isn't in the given unicode-range then the font won't be downloaded.
Example XIII on the page you reference shows a great example. Imagine 3 #font-face rules that share the same font-family attribute. The first rule, however, points to a giant 4.5MB TTF file that has every possible glyph needed. The second rule lists a much smaller 1.2MB TTF but says to only use it only if all of the characters fit into the Japanese glyph range. The third rule lists a very small 190KB file that can be download if the text being styling is only roman.

Microdata itemprop causes W3C validator error

After adding Microdata to my pages, I got many errors from W3C validator complaining the itemprop:
there is no attribute "itemprop"
From code like this:
<p itemprop="description">...</p>
This is my DOCTYPE and html tag
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml"
xmlns:fb="http://www.facebook.com/2008/fbml">
How to fix the validator errors?
p.s. Previously I have the validator error for itemscope as well. But after I changed it to itemscope="itemscope" then the error is fixed.
Documents with HTML 5 plus Microdata used to validate successfully but because of Bug #14020 the validator has become more strict on declaring documents as "valid". Your document is valid HTML 5 + Microdata, but is not strictly an HTML 5-only document.
You can use http://Validator.nu/ to validate HTML 5 + Microdata.
The site linked to in Fabian's answer is not the W3C site he mentions, so I wouldn't trust that as much for HTML 5 as I might have for earlier (pre-2000) versions of HTML.
The reason you had to change itemscope to itemscope="itemscope" is that previous browsers and specifications have defined incompatible interpretations (sometimes true, sometimes false) of code such as itemscope (no value), itemscope="" (an empty string is interpreted as false by XPath) and itemscope="false" (any non-empty string sometimes interpreted as true). Thus the statement in the spec that "The values 'true' and 'false' are not allowed on boolean attributes." However, "true" and "false" can appear in certain attribute values because they are allowed on enumerated attributes such as draggable. See bullet #4 regarding coding boolean values.
The workarounds (elsewhere) to insert invalid code with scripting may hide that code from the validator, but it won't create a document that is any more valid than using static HTML code because the HTML 5 specification is defined in terms of the internal document model, not the external representation. See HTML 5 Specifications focus on the DOM.
OK, here is what I did to make this work with the Validator:
Referring to this page: http://www.w3.org/TR/2011/WD-microdata-20110525/
I enclosed the main in my page (the "wrapper" if you will) with the following:
<div id="layout" itemscope>
If you have itemscope in the div tag for your page or for the div containing microdata, then the W3C Validtor will like it just fine.
The DOCTYPE needs to be HTML5 for microdata to validate.
<!DOCTYPE html>
<html xmlns:fb="http://www.facebook.com/2008/fbml">
.....
It will work with paragraph tag:
http://www.w3.org/TR/2011/WD-microdata-20110525/
if you just want to remove this tag you can do so by removing tag from file which is located in main root folder
"wp-includes/general-template.php" at 891 line
you can just remove extra tag.

Smarty strip tags not working

Is there a particular reason why {strip} tags in smarty don't work? I have two templates. In one of them, it works. In the other it doesn't. But I put them in the exact same place
{strip}
<!DOCTYPE html PUBLIC"-// W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html>
......
</html>
{/strip}
I would just like to know what are the possible causes to this.
If you have improperly nested or improperly closed Smarty tags elsewhere on the page, it could be breaking the strip pair, since they're the outermost pair.
Leaving something like {literal} unclosed or having a syntax error that's causing the page to stop rendering before the end of the page (view source; do you have the whole source, up to and including the final </html> tag?) would cause the final {/strip} to not get included/evaluated.

contentEditable on nodes in a XML/compound document?

I have an XML document that I'm displaying in a web browser, with a stylesheet attached:
<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<?xml-stylesheet type="text/css" href="abc.css"?>
<myxml xmlns:xhtml="http://www.w3.org/1999/xhtml">
<para>I wish i was editable</para>
<xhtml:script type="text/javascript" src="abc.js"/>
</myxml>
With the xhtml namespace declaration, and xhtml:script tag, I can execute javascript.
What I'd like to do is to make arbitrary non-XHTML elements in this document content editable. (Actually, they'll be in another namespace)
Even if I explicitly add #contentEditable="true" (ie without resorting to Javascript), the content is not actually editable (in Firefox 3.0.4).
Is it possible to edit it in any of the current browsers? (I had no problems with <div contentEditable="true">Edit me</div> in an XHTML 1.0 Transitional doc)
I can't even edit an xhtml:div in this document (in Firefox); if I could do that, that may offer a way forward.
In Firefox 3, #content-editable="true" only makes the relevant element editable if the
content type is text/html (which also happens if a local filename ends with .html)
It doesn't work for content types app/xhtml+xml or text/xml (local filenames ending with .xhtml or .xml)
I've logged an enhancement for this: https://bugzilla.mozilla.org/show_bug.cgi?id=486931
contentEditable works (tested in Firefox and Chrome) on elements which are foreign to html/xhtml if I use this doctype:
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
"http://www.w3.org/TR/html4/loose.dtd">
and a .html file extension (instead of .xml).
I don't have to include any html elements at all (eg head, body, div, p).
css isn't applied though (if my xml is in a namespace, which i guess makes sense, given the doctype!).
Not an elegant solution.
Firefox is one of the few browsers that strictly enforces the XHTML spec. So, to make an element editable, you must specify the contenteditable attribute as true. Note that the whole attribute name is lower case. In your example the first "E" in editable was capitalized.
Another quirk that should be mentioned is that IE(6,7,8) act exactly the opposite. To make an element editable in IE, you MUST add contentEditable="true" exactly. For what ever reason, contenteditable="true" (as well as any other variation in capitalization) does not work.

Resources