Python Forum

Hello,

I need to loop through HTML files, and make use the charset line is the very first item in the header so the title is correctly displayed.

Do you know of a simple way to do this?

Thank you.

#<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
look_for = re.compile("^Content-Type$", re.I)

soup = BeautifulSoup(open(INPUTFILE,'rb'), "lxml")
meta = soup.find("meta", {"http-equiv":look_for})
#if no meta, add one since BS doesn't
if not meta:
	print("No meta")
	metatag = soup.new_tag('meta')
	metatag.attrs['http-equiv'] = 'Content-Type'
	metatag.attrs['content'] = 'text/html; charset=utf-8'
	head.insert(0,metatag) #insert as first line in head
else:
	#check for dups, remove if any
	print("Found meta(s)")
	metas = soup.find_all("meta", {"http-equiv":look_for})
	if len(metas) > 1:
		print("Dups")
		for meta in metas[1:]:
			meta.decompose()
	#at this point, only one line left: Move it to top in head
	#TODO How to move utf-8 line at top in head?
	"""
	 <head>
	  <title>
	   Blah
	  </title>
	  <meta content="text/html; charset=utf-8" http-equiv="Content-Type"/>
	"""

---
Edit: A less elegant solution that does the job: Removing all the relevant meta lines, and inserting one at the top.

look_for = re.compile("^Content-Type$", re.I)

def insert_meta(soup):
	metatag = soup.new_tag('meta')
	metatag.attrs['http-equiv'] = 'Content-Type'
	metatag.attrs['content'] = 'text/html; charset=utf-8'
	soup.head.insert(0,metatag)

#<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
meta = soup.find("meta", {"http-equiv":look_for})
#if no meta, add one since BS doesn't
if not meta:
	print("No meta")
	insert_meta(soup)
else:
	print("Found meta(s)")
	#remove for dups, if any
	metas = soup.find_all("meta", {"http-equiv":look_for})
	for meta in metas:
		meta.decompose()
	insert_meta(soup)

Winfried