w3c验证服务是否损坏，或者我迷失了方向？

Question

I am trying to validate the format of an RSS feed I'm making. The XML generated looks like this:

<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
<channel>
    <atom:link href="https://foo.bar.net/rss.php" rel="self" type="application/rss+xml" />
    <title>foo.bar Latest News</title>
    <link>https://foo.bar.net/</link>
    <description>Latest news and information of interest at foo.bar</description>
    <pubDate>Thu, 08 Dec 2011 00:08:09 -0500</pubDate>
    <item>
        <title>some title</title>
        <link>https://foo.bar.net/</link>
        <description>some description.</description>
        <pubDate>Wed, 07 Dec 2011 23:04:42 -0500</pubDate>
        <guid>https://foo.bar.net/rss.php?itemId=1</guid>
    </item>
</channel>
</rss>

If I take the text from this page and copy it from here, into the w3c validator at http://validator.w3.org/feed/check.cgi , this validates. If I take the same text, copied from the browser source, I get the error:

Sorry

This feed does not validate.
line 2, column 0: XML parsing error: <unknown>:2:0: XML or text declaration not at start of entity [help]

<?xml version="1.0" encoding="UTF-8"?>

If I hit the validate button again (no cutting or pasting this time, just what was already in the text window), I get:

Congratulations!

This is a valid RSS feed.

Say what? My RSS reader won't recognize this as a valid feed either, and there seems to be absolutely difference. What the heck is going on here?

...and yes, I did click on the "help" link. It says my feed is not well formed according to the XML specification. Having worked in xml for quite a while, I'm not seeing anything jumping out at me. Could this be some odd encoding issue, and if so, how do I get around it?

Answer 1

It looks to me as if, when you pasted the XML fragment into the validator, you left some whitespace in front of the XML declaration - it's complaining that the XML declaration is on line 2.

(Perhaps a nicer design would have been for the validation service to strip any leading whitespace before starting the parse...)