. is the current node; it is short for self::node().
// is the descendant-or-self axis; it is short for /descendant-or-self::node()/.
Together, .// will select along the descendent-or-self axis starting from the current node. Contrast this with // which starts at the document root.
Example
Consider the following HTML:
<html>
<body>
<div id="id1">
<p>First paragraph</p>
<div>
<p>Second paragraph</p>
</div>
</div>
<p>Third paragraph</p>
</body>
</html>
//p will select all paragraphs:
<p>First paragraph</p>
<p>Second paragraph</p>
<p>Third paragraph</p>
On the other hand, if the current node is at the div element (with @id of "id1"), then .//p will select only the paragraphs under the current node:
<p>First paragraph</p>
<p>Second paragraph</p>
Notice that the third paragraph is not selected by .//p when the current node is the id1 div because the third paragraph is not under that div element.
与恶龙缠斗过久,自身亦成为恶龙；凝视深渊过久,深渊将回以凝视…
