CODEWITHSUNDEEP
pythonxmlxpathlxmlelementtree
Hypothetical Ninja
Hypothetical Ninja

Reputation: 4077

Extract text with lxml

I have this text :

INTRODUCTION
This is a test document for xml.
I need to extract this sentence.

Conclusion
It should hopefully..

The line I need to extract this sentence. is in italics . The xml of the file looks like:

<?xml version="1.0" encoding="UTF-8" standalone="yes"?>\r\n
<w:document 
  xmlns:mc="http://schemas.openxmlformats.org/markup-compatibility/2006"
  xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main" 
  mc:Ignorable="w14 w15 wp14">
   <w:body>
      <w:p w:rsidR="00470EEF" w:rsidRDefault="00456755">
         <w:pPr>
            <w:rPr>
               <w:b/>
            </w:rPr>
         </w:pPr>
         <w:r w:rsidRPr="00456755">
            <w:rPr>
               <w:b/>
            </w:rPr>
            <w:t>INTRODUCTION</w:t>
         </w:r>
      </w:p>
      <w:p w:rsidR="00456755" w:rsidRPr="00B042E3" w:rsidRDefault="00456755">
         <w:pPr>
            <w:rPr>
               <w:color w:val="FFFF00"/>
            </w:rPr>
         </w:pPr>
         <w:r w:rsidRPr="00B042E3">
            <w:rPr>
               <w:color w:val="FFFF00"/>
            </w:rPr>
            <w:t>This is a test document for xml.</w:t>
         </w:r>
      </w:p>
      <w:p w:rsidR="00456755" w:rsidRDefault="00E971E1">
         <w:r>
            <w:rPr>
               <w:i/>
            </w:rPr>
            <w:t>I need to extract this sentence.</w:t>
         </w:r>
         <w:bookmarkStart w:id="0" w:name="_GoBack"/>
         <w:bookmarkEnd w:id="0"/>
      </w:p>
      <w:p w:rsidR="00456755" w:rsidRDefault="00456755"/>
      <w:p w:rsidR="00456755" w:rsidRDefault="00456755">
         <w:pPr>
            <w:rPr>
               <w:b/>
            </w:rPr>
         </w:pPr>
         <w:r w:rsidRPr="00456755">
            <w:rPr>
               <w:b/>
            </w:rPr>
            <w:t>Conclusion</w:t>
         </w:r>
      </w:p>
      <w:p w:rsidR="00456755" w:rsidRPr="00456755" w:rsidRDefault="00456755">
         <w:r w:rsidRPr="00456755">
            <w:t>It should hopefully</w:t>
         </w:r>
         <w:r>
            <w:t>..</w:t>
         </w:r>
      </w:p>
      <w:sectPr w:rsidR="00456755" w:rsidRPr="00456755">
         <w:pgSz w:w="11906" w:h="16838"/>
         <w:pgMar w:top="1440" w:right="1440" w:bottom="1440" w:left="1440" w:header="708" w:footer="708" w:gutter="0"/>
         <w:cols w:space="708"/>
         <w:docGrid w:linePitch="360"/>
      </w:sectPr>
   </w:body>
</w:document>

I tried :

tree = ET.parse(doc_xml)  
[b.tag for b in tree.iterfind(".//i")]

The above returns an empty list.

I've searched a lot but wasn't able to figure out how to do that as the text is contained within <w:i/>. I have seen this question where this was done easily using BeautifulSoup.

Edit: This isn't related exactly but this is an ElementTree approach to extract all text.

w = 'http://schemas.openxmlformats.org/wordprocessingml/2006/main' 
    for p in source.findall('.//{' + w + '}p'):
        print ''.join(t.text for t in p.findall('.//{' + w + '}t'))

Upvotes: 1

Views: 697

Answers (2)

falsetru
falsetru

Reputation: 369134

Slightly modifying the you will get what you want:

>>> w = 'http://schemas.openxmlformats.org/wordprocessingml/2006/main'    
>>> for t in tree.findall('.//{%(ns)s}i/../..//{%(ns)s}t' % {'ns': w}):
...     print t.text
... 
I need to extract this sentence.

BTW, if you use local-name(), you don't need to specify namespace (need to use xpath method, which is available in lxml):

>>> for t in tree.xpath('.//*[local-name()="i"]/../..//*[local-name()="t"]'):
...     print t.text
... 
I need to extract this sentence.

UPDATE

.. in the expression selects parent node of the current node. So, {...}i/../.. will select grand-parent node of i node.

Upvotes: 2

har07
har07

Reputation: 89295

Building my answer based on your code in Edit section :

w = 'http://schemas.openxmlformats.org/wordprocessingml/2006/main' 
for p in source.findall('.//{' + w + '}p[.//{' + w + '}i]'):
    print ''.join(t.text for t in p.findall('.//{' + w + '}t'))

Basically, the first XPath supposed to match all <w:p> elements having descendant node <w:i>, then as you know the next line extract all <w:t> nodes' text from matched <w:p> nodes.

Upvotes: 2

Related Questions