<p>Writing a web scraper with Go using the <code>net/html</code> package, on one unit test I noticed that it couldn't fail as I made it to behave, even if the source document use for the test don't have a <code><body></code> tag. The node looks like this:</p>
<pre><code>&{Parent:0x18813100 FirstChild:<nil> LastChild:<nil> PrevSibling:0x18813240 NextSibling:<nil> Type:3 DataAtom:body Data:body Namespace: Attr:[]}
</code></pre>
<p>and this is how the source document use for that test looks:</p>
<pre><code><!DOCTYPE html>
<html>
<head>
<meta charset='utf-8'>
</head>
</html>
</code></pre>
<p>That's how the net/html was made? I would like to know! :)</p>
<hr/>**评论：**<br/><br/>HectorJ: <pre><p>It seems it does: <a href="https://github.com/golang/net/blob/master/html/parse.go#L678" rel="nofollow">https://github.com/golang/net/blob/master/html/parse.go#L678</a></p>
<pre><code>p.parseImpliedToken(StartTagToken, a.Body, a.Body.String())
</code></pre>
<p><a href="https://github.com/golang/net/blob/master/html/parse.go#L1956" rel="nofollow">https://github.com/golang/net/blob/master/html/parse.go#L1956</a></p>
<pre><code>// parseImpliedToken parses a token as though it had appeared in the parser's
// input.
</code></pre></pre>HadronHubbub: <pre><p>In HTML, <a href="https://html.spec.whatwg.org/multipage/semantics.html#the-body-element" rel="nofollow">the body tags are optional.</a> Any HTML parser that conforms to the specification will do the same thing and give you a body element even if there are no body tags. (Unfortunately you'll find that many "HTML" parsing libraries get this wrong and basically act like they're parsing XML without draconian error-handling).</p></pre>
