Can you have an XML comment as the first line in an SVG file? For example:
<!-- Timestamp 1434061994 -->
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
<svg...
Is this against SVG spec? Or would it ever fail validation or would this cause any sort of problems I'm not seeing when implementing this in a website?
Yes, it is ok to have a comment be the first line in an SVG file, but only if there is no XML declaration (<?xml version="1.0" encoding="UTF-8"?>).
Nothing can appear before the XML declaration in an XML file. At most one XML declaration can appear in a file, and if an XML declaration is used, then it must be at the very top of the file. Anything before an XML declaration, including a comment, prevents the XML from being well-formed and should result in an error such as the following diagnostic by Xerces-J:
The processing instruction target matching "[xX][mM][lL]" is not
allowed.
If a comment appears before the XML declaration, then the XML is not well-formed, and if the XML is not well-formed, the SVG is not conforming.
Final note: An XML declaration is optional. Unless you want to specify a version other than 1.0 or an encoding other than UTF-8, you don't have to have an XML declaration in an XML (or SVG) file.
Related
Must SVG foreignObject contain only XML? That seems to be my reading of the spec, so is it required to put XHTML inside, if I wish to use foreignObject? Or can I just copy my regular HTML5 syntaxed code inside?
The contents of any document must reflect the mime type it's served as.
If you serve the document as image/svg+xml then the whole document must be valid XML including any foreignObject child elements which basically limits you to XHTML. XML fails on error so if you paste in HTML you'll likely just get a syntax error printed than have anything rendered.
If you serve the document as text/html then the whole document is HTML and any SVG in it would be parsed with an HTML parser, no namespaces required etc. Any foreignObject elements in such SVG will be parsed as if they were standard HTML. HTML attempts to render what it can of invalid documents.
I am new to coding. I got a file called index.html. The first 2 lines in it are given below.
<?xml version='1.0' encoding='utf-8'?>
<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en">
Is it an html or xml file?
HTML 5 defines an XML serialisation. Previously, XHTML 1 defined an implementation of HTML 4 in XML.
It is XML. It is also either HTML or XHTML.
Nice Question, First of all the first line is pure XML declarationXML Declaration, the second line is of html, and it contains meta data.
I want to replace all the with in my html file to support XML parser.
But I don't want to replace them directly, I'd like to add an entity in <!DOCTYPE > like below:
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"[<!ENTITY nbsp " ">]>
<html><head></head><body><div>Hello World!</div></body></html>
But when I view the file, there is an extra ]> on the top of the document:
Anyone know how to deal with it?
Thanks!
What you have is a valid way to include an entity declaration in an internal subset. The document is not otherwise valid, though, as you can check with the W3C Markup Validator: the required xmlns attribute on the html element is missing, and so is the required title attribute.
When served as text/html, the document is processed how browsers use to process HTML document, which means among other thing that internal subsets are not recognized; in fact, document type definitions are not read at all – instead, doctype declarations are just taken as magic strings so that some strings trigger “quirks mode”, some don’t. The doctype declaration is parsed in a simplistic manner, which makes the first “>” terminate it, so whatever comes after it is taken as character data.
The morale is that entity declarations just don’t work with “HTML”, internally or externally, when “HTML” means sending something to a browser and telling (in HTTP headers) it to be text/html – and that’s what servers normally tell when they send .html files.
Served as application/xhtml+xml and fixed to conform to XHTML syntax, your approach works on conforming browsers (online demo: http://www.cs.tut.fi/~jkorpela/test/nbsp.xhtml):
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"
[<!ENTITY nbsp " ">]>
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<title>Entity demo</title></head>
<body>
<div>Hello World!</div>
</body>
</html>
However, IE 8 and earlier don’t process HTML when served as application/xhtml+xml (the browser just launches a “Save As” dialog).
The conclusions depend on what you are doing and why (and in which sense) you need to “support XML parser”. It’s not really about parsing but about entity declarations. XHTML user agents are not required to understand predefined entities as in HTML (except for those defined in XML), but has this possibility realized somehow? And in general, it is better to convert to actual no-break space characters than to character references.
here
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"[<!ENTITY nbsp " ">
I have an XML file ("exchange.xml") and an XSL file ("exchange.xsl") for parsing it. exchange.xml has the following line:
<?xml-stylesheet type="text/xsl" href="exchange.xsl"?>
Indicating that exchange.xsl should be used to parse the file.
It parses as expected if I load exchange.xml in a browser. However, I want to have exchange.xml embedded in an HTML file which otherwise does not have any XML in it. How do I do this? As far as I understand it, when you embed XML in HTML with the XML tag's SRC ID, you then have to parse it yourself for it to be displayed. I want it to automatically be displayed using the XSL file I've already created for it.
It has also come to my attention that some browsers (e.g. Android default browser) won't parse the XML with the XSL file, and require a server-side transform. Is it possible to embed my XML into an HTML file but use a server-side transform?
As far as I am aware HTML 4 as specified by the W3C does not provide any way to embed XML markup in an HTML 4 document. And of the browsers I know only IE has an extension to HTML 4 that allows that, the so called XML data islands, http://msdn.microsoft.com/en-us/library/ms766512%28v=VS.85%29.aspx, where IE's HTML parser recognizes an xml element that can contain XML markup as its content or link to it with the src attribute. So unless you want to use something IE specific like an XML data islands the cross-browser way to load XML as data inside of an HTML document is to use XMLHttpRequest with Javascript.
You can use PHP: XSL class for transforming XML and include it via PHP in your page.
eg: showxml.php
<?php
function processXML($file,$styles){
... //example here
}
<html>
...
<div id="xml>
<?php echo processXML('foo.xml','bar.xsl'); ?>
</div>
...
</html>
I am building a menu and have it set up so that I use a standard <!--#include virtual = "myDoc.xml" --> SSI tag to include my xml document. The xml document includes the xsl document with <?xml-stylesheet type="text/xsl" href="myOtherDoc.xsl"?>. For some reason the xsl document is not working. The xml is being displayed as a blob.
Unfortunately your XSL transform won't work like that. You'd have to send the XML document alone to the browser where the built in xsl transformer would then reference the stylesheet and perform the transform.
What you've got is an HTML page already being rendered and you're including the XML as just a chunk of xml rendered into the output stream, but the browser won't know to transform it because it doesn't have the:
<?xml-stylesheet type="text/xsl" href="myOtherDoc.xsl"?>
...PI at the start of the page. Remember these are processed by the browser not the server.
You would need to transform the XML server side e.g.
<!-- #include virtual="doMenuXform.asp" -->