$(" ... html ...") strips leading and trailing text nodes in HTML string

[[email protected]…]

Hi,

I stumbled about this when trying to use $() with an HTML string that contains multiple top-level nodes and starts with a text node:

• $("1<br/>2<br/>3")
  • Expected result: ["1", <br>, "2", <br>, "3"]
  • Observed result: [<br>, "2", <br>]

Additional observations:

The nodes are created as expected if the HTML string starts with "<" and ends with ">", even if the characters are converted to text nodes.

• $("1<br/>2<br/>3 >") returns (incorrectly):
  • [<br>, "2", <br>, "3 &gt;"]

• $("< 1<br/>2<br/>3") returns (incorrectly):
  • ["&lt; 1", <br>, "2", <br>]

• $("< 1<br/>2<br/>3 >") returns (correctly):
  • ["&lt; 1", <br>, "2", <br>, "3 &gt;"]

• $("1<br/>2<br/>3<!-- -->") returns (incorrectly):
  • [<br>, "2", <br>, "3", (comment)]

• $("<!-- -->1<br/>2<br/>3") returns (incorrectly):
  • [(comment), "1", <br>, "2", <br>]

• $("<!-- -->1<br/>2<br/>3<!-- -->") returns (correctly):
  • [(comment), "1", <br>, "2", <br>, "3", (comment)]

(The last is what I'm using as a workaround right now.)

Tested on:

• Firefox 6.0
• Google Chrome 13.0.782.112 m

OS:

• Windows 7 x64 SP1

Regards, Philipp Serafin

[dmethvin]

jQuery isn't intended to process text nodes as if they were DOM elements. Most methods save a few like .content() remove or ignore them if they are in a jQuery collection.

I've marked the ticket for a documentation review, but the behavior is correct.

[addyosmani]

I've updated the docs for this here: ​http://api.jquery.com/jQuery/