Menu
I'm trying to finally solve some encoding issues that pop up from trying to scrape HTML with lxml. Here are three sample HTML documents that I've encountered:
1.
<!DOCTYPE html>
<html lang='en'>
<head>
<title>Unicode Chars: 은 —’</title>
<meta charset='utf-8'>
</head>
<body></body>
</html>
2.
<!DOCTYPE html>
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="ko-KR" lang="ko-KR">
<head>
<title>Unicode Chars: 은 —’</title>
<meta http-equiv="content-type" content="text/html; charset=utf-8" />
</head>
<body></body>
</html>
3.
<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE html
PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<title>Unicode Chars: 은 —’</title>
</head>
<body></body>
</html>
My basic script:
from lxml.html import fromstring
...
doc = fromstring(raw_html)
title = doc.xpath('//title/text()')[0]
print title
The results are:
Unicode Chars: ì ââ
Unicode Chars: 은 —’
Unicode Chars: 은 —’
So, obviously an issue with sample 1 and the missing <meta http-equiv="Content-Type" content="text/html; charset=utf-8" /> tag. The solution from here will correctly recognize sample 1 as utf-8 and so it is functionally equivalent to my original code.
The lxml docs appear conflicted:
From here the example seems to suggest we should use UnicodeDammit to encode the markup as unicode.
from BeautifulSoup import UnicodeDammit
def decode_html(html_string):
converted = UnicodeDammit(html_string, isHTML=True)
if not converted.unicode:
raise UnicodeDecodeError(
"Failed to detect encoding, tried [%s]",
', '.join(converted.triedEncodings))
# print converted.originalEncoding
return converted.unicode
root = lxml.html.fromstring(decode_html(tag_soup))
However here it says:
[Y]ou will get errors when you try [to parse] HTML data in a unicode string that specifies a charset in a meta tag of the header. You should generally avoid converting XML/HTML data to unicode before passing it into the parsers. It is both slower and error prone.
If I try to follow the the first suggestion in the lxml docs, my code is now:
from lxml.html import fromstring
from bs4 import UnicodeDammit
...
dammit = UnicodeDammit(raw_html)
doc = fromstring(dammit.unicode_markup)
title = doc.xpath('//title/text()')[0]
print title
I now get the following results:
Unicode Chars: 은 —’
Unicode Chars: 은 —’
ValueError: Unicode strings with encoding declaration are not supported.
Sample 1 now works correctly but sample 3 results in an error due to the <?xml version="1.0" encoding="utf-8"?> tag.
Is there a correct way to handle all of these cases? Is there a better solution than the following?
dammit = UnicodeDammit(raw_html)
try:
doc = fromstring(dammit.unicode_markup)
except ValueError:
doc = fromstring(raw_html)
780
lxml provides a very simple and powerful API for parsing XML and HTML. It supports one-step parsing as well as step-by-step parsing using an event-driven API (currently only for XML).
lxml is also a similar parser but driven by XML features than HTML. It has dependency on external C libraries. It is faster as compared to html5lib. Lets observe the difference in behavior of these two parsers by taking a sample tag example and see the output.
lxml is faster than html. parser or html5lib parser. This is because lxml parser that you will invoke in beautiful soup is natively written in C ( uses the libxml2 C library ) , hwere as the html. parser is written in python.
lxml is a tool for working with HTML and XML documents, represented as an element tree. It evaluates XPath and CSS selectors to find matching elements. It facilitates navigating from one element to another. It facilitates extracting the text, attribute values or HTML for a particular element.
lxml has several issues related to handling Unicode. It might be best to use bytes (for now) while specifying the character encoding explicitly:
#!/usr/bin/env python
import glob
from lxml import html
from bs4 import UnicodeDammit
for filename in glob.glob('*.html'):
with open(filename, 'rb') as file:
content = file.read()
doc = UnicodeDammit(content, is_html=True)
parser = html.HTMLParser(encoding=doc.original_encoding)
root = html.document_fromstring(content, parser=parser)
title = root.find('.//title').text_content()
print(title)
Unicode Chars: 은 —’
Unicode Chars: 은 —’
Unicode Chars: 은 —’
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With
