soup.find("tagName", { "id" : "articlebody" }) Why does this NOT return the <div> ... </div> tags and stuff in between? It returns nothing. And I know for a fact it exists because I'm staring right at it from
soup.prettify() soup.find("div", { "id" : "articlebody" }) also does not work.
(EDIT: I found that BeautifulSoup wasn't correctly parsing my page, which probably meant the page I was trying to parse isn't properly formatted in SGML or whatever)
113 Answers
You should post your example document, because the code works fine:
>>> import BeautifulSoup >>> soup = BeautifulSoup.BeautifulSoup('<html><body><div> ... </div></body></html') >>> soup.find("div", {"id": "articlebody"}) <div> ... </div> Finding <div>s inside <div>s works as well:
>>> soup = BeautifulSoup.BeautifulSoup('<html><body><div><div> ... </div></div></body></html') >>> soup.find("div", {"id": "articlebody"}) <div> ... </div> To find an element by its id:
div = soup.find(id="articlebody") Beautiful Soup 4 supports most CSS selectors with the .select() method, therefore you can use an id selector such as:
soup.select('#articlebody') If you need to specify the element's type, you can add a type selector before the id selector:
soup.select('div#articlebody') The .select() method will return a collection of elements, which means that it would return the same results as the following .find_all() method example:
soup.find_all('div',) # or soup.find_all(id="articlebody") If you only want to select a single element, then you could just use the .find() method:
soup.find('div',) # or soup.find(id="articlebody") I think there is a problem when the 'div' tags are too much nested. I am trying to parse some contacts from a facebook html file, and the Beautifulsoup is not able to find tags "div" with class "fcontent".
This happens with other classes as well. When I search for divs in general, it turns only those that are not so much nested.
The html source code can be any page from facebook of the friends list of a friend of you (not the one of your friends). If someone can test it and give some advice I would really appreciate it.
This is my code, where I just try to print the number of tags "div" with class "fcontent":
from BeautifulSoup import BeautifulSoup f = open('/Users/myUserName/Desktop/contacts.html') soup = BeautifulSoup(f) list = soup.findAll('div', attrs={'class':'fcontent'}) print len(list) Most probably because of the default beautifulsoup parser has problem. Change a different parser, like 'lxml' and try again.
1In the beautifulsoup source this line allows divs to be nested within divs; so your concern in lukas' comment wouldn't be valid.
NESTABLE_BLOCK_TAGS = ['blockquote', 'div', 'fieldset', 'ins', 'del'] What I think you need to do is to specify the attrs you want such as
source.find('div', attrs={'id':'articlebody'}) have you tried soup.findAll("div", {"id": "articlebody"})?
sounds crazy, but if you're scraping stuff from the wild, you can't rule out multiple divs...
I used:
soup.findAll('tag', attrs={'attrname':"attrvalue"}) As my syntax for find/findall; that said, unless there are other optional parameters between the tag and attribute list, this shouldn't be different.
Here is a code fragment
soup = BeautifulSoup(:"index.html") titleList = soup.findAll('title') divList = soup.findAll('div', attrs={ "class" : "article story"}) As you can see I find all tags and then I find all tags with inside
Happened to me also while trying to scrape Google.
I ended up using pyquery.
Install:
pip install pyquery Use:
from pyquery import PyQuery pq = PyQuery('<html><body><div> ... </div></body></html') tag = pq('div#articlebody') The Id property is always uniquely identified. That means you can use it directly without even specifying the element. Therefore, it is a plus point if your elements have it to parse through the content.
divEle = soup.find(id = "articlebody") from bs4 import BeautifulSoup from requests_html import HTMLSession url = 'your_url' session = HTMLSession() resp = session.get(url) # if element with id "articlebody" is dynamic, else need not to render resp.html.render() soup = bs(resp.html.html, "lxml") soup.find("div", {"id": "articlebody"}) soup.find("tagName",attrs={ "id" : "articlebody" }) 
