Fetch content based on XML content validation in bash

Question

I have an XML as follows

<demos>
 <demo>
  <a>00</a>
  <b>First</b>
  <c>ProjectA</c>
 </demo>
 <demo>
  <a>01</a>
  <b>Second</b>
  <c>ProjectB</c>
 </demo>
</demos>

Example 1: If within <a> tag has 00 and <c> tag has ProjectA, Output should be First

Example 2: If within <a> tag has 01 and <c> tag has ProjectB, Output should be Second

I want try with xml_grep or xpath as these are installed packages in linux by default. I tried different ways such as

xpath xtest.xml '//a[text()="01"]/text() | //b'

but this validation doesnt work.

Both If conditions in your question are true for the sample XML. What should be the output? — Tomalak
– Tomalak, Commented Mar 2, 2018 at 9:05
Are you asking about how to select a sibling? stackoverflow.com/questions/17040254/… — Micha Wiedenmann
– Micha Wiedenmann, Commented Mar 2, 2018 at 9:19
You should probably drop the [linux], [bash] and [shell] tags, imho this is an xpath question. — Micha Wiedenmann
– Micha Wiedenmann, Commented Mar 2, 2018 at 9:19
Which kind of command xpath do you have? On my system, I would have to run: xpath -e '//a[text()="01"]/text() | //b' xtest.xml instead. — Pierre François
– Pierre François, Commented Mar 2, 2018 at 9:55

Pierre François · Accepted Answer · 2018-03-02 09:47:20Z

1

For example 1, you will have to run:

xpath -q -e '//demo[(a/text()="00") and (c/text()="ProjectA")]/b/text()' xtest.xml

For example 2:

xpath -q -e '//demo[(a/text()="01") and (c/text()="ProjectB")]/b/text()' xtest.xml

Be aware that xpath is not installed by default on any Linux. On Ubuntu, you will first have to run sudo apt install libxml-xpath-perl.

answered Mar 2, 2018 at 9:47

Pierre François

6,1681 gold badge21 silver badges42 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

stackover flow · Accepted Answer · 2018-03-02 09:40:19Z

0

The following command worked with XML as file.xml

 xpath file.xml '//demo[a/text()="00" and c/text()="ProjectA"]/b/text()'

answered Mar 2, 2018 at 9:40

stackover flow

751 gold badge1 silver badge6 bronze badges

2 Comments

Pierre François Over a year ago

This doesn't work on Ubuntu with xpath installed along with libxml-xpath-perl. Probably, you have another flavour of xpath.

stackover flow Over a year ago

This is suse linux

RomanPerekhrest · Accepted Answer · 2018-03-02 09:42:38Z

0

Robust xmlstarlet solution:

xmlstarlet sel -t -m '//demo/a' --if '. = 00 and ./following-sibling::c="ProjectA"' \
-v './following-sibling::b' -n input.xml

The output:

First

xmlstarlet sel -t -m '//demo/a' --if '. = 01 and ./following-sibling::c="ProjectB"' \
-v './following-sibling::b' -n input.xml

The output:

Second

answered Mar 2, 2018 at 9:42

RomanPerekhrest

93.1k4 gold badges75 silver badges112 bronze badges

Collectives™ on Stack Overflow

Fetch content based on XML content validation in bash

3 Answers 3

Comments

2 Comments

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

Comments

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related