Python code to consolidate CSS into HTML

Question

Looking for python code that can take an HTML page and insert any linked CSS style definitions used by that page into it - so any externally referenced css page(s) are not needed.

Needed to make single files to insert as email attachments from existing pages used on web site. Thanks for any help.

• hughdbrown: You can easily pull the CSS out, but what CSS path will you use to identify each piece of CSS? I'm not sure how easily you can do that programmatically.
• khachik: dup? stackoverflow.com/questions/781382/…
• Sven Marnach: No, not a dup. The aim here is to put everything into one file.
• Paulo Scardine: Note that Outlook HTML mail reader is Word, anything but inline style="..." in each tag may result very odd sometimes.

Answer 1

Sven's answer helped me, but it didn't work out of the box. The following did it for me:

import bs4 #BeautifulSoup 3 has been replaced
soup = bs4.BeautifulSoup(open("index.html").read())
stylesheets = soup.findAll("link", {"rel": "stylesheet"})
for s in stylesheets:
 t = soup.new_tag('style')
 c = bs4.element.NavigableString(open(s["href"]).read())
 t.insert(0,c)
 t['type'] = 'text/css'
 s.replaceWith(t)
open("output.html", "w").write(str(soup))

Answer 2

You will have to code this yourself, but BeautifulSoup will help you a long way. Assuming all your files are local, you can do something like:

from BeautifulSoup import BeautifulSoup
soup = BeautifulSoup(open("index.html").read())
stylesheets = soup.findAll("link", {"rel": "stylesheet"})
for s in stylesheets:
 s.replaceWith('<style type="text/css" media="screen">' +
 open(s["href"]).read()) +
 '</style>')
open("output.html", "w").write(str(soup))

If the files are not local, you can use Pythons urllib or urllib2 to retrieve them.

• Frédéric Hamidi: That, and hope he doesn't use any @import rules :)
• Sven Marnach: @Frédéric: Good point. I'm sure I missed quite a few cases, but that's just to get Scott going :)
• Scott Treseder: Thanks. This was a great help.

Answer 3

You can use pynliner. Example from their documentation:

html = "html string"
css = "css string"
p = Pynliner()
p.from_string(html).with_cssString(css)