0 
Source file example:
<HDR></HDR><b></b><c></c>

(XML file created in a single line)

OR 

Source file example:
<HDR>
</HDR>
<b>
</b>
<c>
</c>

I need to remove all the content of the file before <b> in both of the source format. I tried using the below method 

sed 's/^.*b/b/'

But this is not replacing it. Please let me know if there is an alternative way.

Improve this question
4
  • Please be clearer as to what you want to accomplish and why. A little context might be useful in understanding your needs and helping you find a solution. Commented Sep 3, 2018 at 7:01
  • 2
    The XML is not well formed as there is no root tag around the whole document. Commented Sep 3, 2018 at 7:02
  • I have a header in the XML which needs to be replaced with different header Commented Sep 3, 2018 at 7:14
  • It seems then your problem is different from what you describe. Please add complete specifications of what you're trying to accomplish (replacing header tag?) in order to avoid the XY problem. Commented Sep 3, 2018 at 7:42

3 Answers 3

Reset to default
5

Assuming your XML document is well formed, like

<document>
<HDR>
</HDR>
<b>
</b>
<c>
</c>
</document>

Then you may use XMLStarlet to remove all HDR tags like so:

xmlstarlet ed -d '//HDR' file.xml >newfile.xml

To only remove the HDR tags that are immediately followed by a b tag:

xmlstarlet ed -d '//HDR[following-sibling::*[1][name() = "b"]]' file.xml >newfile.xml

XMLStarlet may also be used to modify the contents of tags:

$ xmlstarlet ed -u '//HDR[following-sibling::*[1][name() = "b"]]' -v 'New header value' file.xml
<?xml version="1.0"?>
<document>
  <HDR>New header value</HDR>
  <b/>
  <c/>
</document>

$ xmlstarlet ed -i '//HDR[following-sibling::*[1][name() = "b"]]' -t attr -n 'new_attribute' -v 'hello' file.xml
<?xml version="1.0"?>
<document>
  <HDR new_attribute="hello"/>
  <b/>
  <c/>
</document>
Improve this answer
2
  • 3
    It has not been stressed enough that XML files should be treated as such and handled by dedicated tools, as opposed to be regarded as text files and messed with with the wrong tools. Commented Sep 3, 2018 at 7:47
  • This command is not added to our version, we can't ask the admins to install it. Thanks !! Commented Sep 3, 2018 at 12:20
2

Question:

remove all contents of the file before <b>

Answer:

perl -0777 -lape 's/^.*<b>/<b>/s'

Test run:

==> in1.txt <==
<HDR></HDR><b></b><c></c>

==> in2.txt <==
<HDR>
</HDR>
<b>
</b>
<c>
</c>

$ perl -i -0777 -lape 's/^.*<b>/<b>/s' in{1,2}.txt

==> in1.txt <==
<b></b><c></c>

==> in2.txt <==
<b>
</b>
<c>
</c>
Improve this answer
2
  • I tried this , It removed the every content after the <b> tag as well, now there <b> tag alone in the new file Commented Sep 3, 2018 at 7:25
  • @user7952074 that means there's a <b> tag at the end of the file :-) Please follow @Kusalananda's advice and stick to well-formed XML and proper XML manipulation tools. Commented Sep 3, 2018 at 7:39
1

Type 1:

 echo "<HDR></HDR><b></b><c></c>" | sed 's/^.*<b>/<b>/' 
 <b></b><c></c>
  • will replace everything up to <b> with <b>

Type 2:

sed  -n '/<b>/,$p' file
<b>
</b>
<c>
</c>
  • will print the first occurrence of <b> to end of the file ($).
Improve this answer

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.