In Python, how do I remove the "root" tag in an HTML snippet?

Question

Suppose I have an HTML snippet like this:

<div>
  Hello <strong>There</strong>
  <div>I think <em>I am</em> feeing better!</div>
  <div>Don't you?</div>
  Yup!
</div>

What's the best/most robust way to remove the surrounding root element, so it looks like this:

Hello <strong>There</strong>
<div>I think <em>I am</em> feeing better!</div>
<div>Don't you?</div>
Yup!

I've tried using lxml.html like this:

lxml.html.fromstring(fragment_string).drop_tag()

But that only gives me "Hello", which I guess makes sense. Any better ideas?

Answer 1

This is a bit odd in lxml (or ElementTree). You'd have to do:

def inner_html(el):
    return (el.text or '') + ''.join(tostring(child) for child in el)

Note that lxml (and ElementTree) have no special way to represent a document except rooted with a single element, but .drop_tag() would work like you want if that <div> wasn't the root element.

Answer 2

For such a simple task you can use regexp like r'<(.*?)>(.*)</\1>' and get match #2 (\2 in perl terms) from it

You should also put flags like ms for correct multi-line working

Answer 3

You can use BeautifulSoup package. For this particular html I would go like this:

import BeautifulSoup

html = """<div>
  Hello <strong>There</strong>
  <div>I think <em>I am</em> feeing better!</div>
  <div>Don't you?</div>
  Yup!
</div>"""

bs = BeautifulSoup.BeautifulSoup(html)

no_root = '\n'.join(map(unicode, bs.div.contents))

BeautifulSoup has many nice features that will allow you to tweak this example for many other cases. Full documentation: http://www.crummy.com/software/BeautifulSoup/documentation.html.