How can I search for mixed tag names in an ATOM XML document?
I'm working with google APIs and they offer the option to return JSON or ATOM. ATOM looks like XML syntax and I want to use BeautifulSoup to parse it.
I have no problem turning this into a BeautifulSoup object, but I am having a hard time finding the element. Take a paragraph of the ATOM doc as an example:
from bs4 import BeautifulSoup
feed = """
<cse:DataObject type="cse_thumbnail">
<cse:Attribute name="width" value="160"/>
<cse:Attribute name="height" value="160"/>
<cse:Attribute name="src" value="https://encrypted-tbn1.gstatic.com/images?q=tbn:ANd9GcRAUAShHrU8LK9MLEMEcfg-rtYgLzaxUP-j30lNJJdP1P6FBdVIziH4LTY"/>
</cse:DataObject>
"""
soup = BeautifulSoup(feed)
print soup.find_all("cse:Attribute", {"value":"160"})
... it returns an empty list. What am I doing wrong?
source to share
Your code, as written, parses the XML as if it were HTML, and since HTML is not case sensitive, BeautifulSoup will convert all tag names to lowercase
Since HTML tags and attributes are not case sensitive , all three HTML parsers convert the tags and attribute names to lowercase. That is, the markup is
<TAG></TAG>
converted to<TAG></TAG>
. If you want to preserve mixed or uppercase tags and attributes, you need to parse the document as XML .
Finding the bottom of the tag works really well:
>>> soup.find_all("cse:attribute", {"value":"160"})
[<cse:attribute name="width" value="160"></cse:attribute>,
<cse:attribute name="height" value="160"></cse:attribute>]
As the quoted text mentions, an alternative is to use an XML parser which will preserve case. However BeautifulSoup with an lxml parser is missing namespaces from tag names ...
>>> soup = BeautifulSoup(feed, "xml")
>>> soup
<?xml version="1.0" encoding="utf-8"?>
<DataObject type="cse_thumbnail">
<Attribute name="width" value="160"/>
<Attribute name="height" value="160"/>
<Attribute name="src" value="https://encrypted-tbn1.gstatic.com/images?q=tbn:ANd9GcRAUAShHrU8LK9MLEMEcfg-rtYgLzaxUP-j30lNJJdP1P6FBdVIziH4LTY"/>
</DataObject>
>>> soup.find_all("cse:Attribute", {"value":"160"})
[]
>>> soup.find_all("cse:attribute", {"value":"160"})
[]
>>> soup.find_all("Attribute", {"value":"160"})
[<Attribute name="width" value="160"/>,
<Attribute name="height" value="160"/>]
... which may not be what you want.
source to share