Parsing a XML using Shell Script?

Question

I am new on this and I would like some help. I have a XML file and I need to edite you content when some cretirias are met.

Currently I'm using the sed Unix commando, but this only exclude the entire line, but I need to exclude the entire block

What I have

<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<Package xmlns="http://soap.sforce.com/2006/04/metadata">
  <types>
    <name>ApexClass</name>
    <members>CC_CaseScenarioMatchingJobQueueable</members>
  </types>
  <types>
    <name>EmailTemplate</name>
    <members>CC_CaseScenarioMatchingJobQueueable</members>
  </types>
  <types>
    <name>CustomObject</name>
    <members>CC_CaseScenarioMatchingJobQueueable</members>
  </types>
</Package>

What I need to solve. When I for a key for example "ApexClass" I have to delete from the tag "" to "" related to this na "ApexClass"

<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<Package xmlns="http://soap.sforce.com/2006/04/metadata">
  <types>
    <name>EmailTemplate</name>
    <members>CC_CaseScenarioMatchingJobQueueable</members>
  </types>
  <types>
    <name>CustomObject</name>
    <members>CC_CaseScenarioMatchingJobQueueable</members>
  </types>
</Package>

This is my currently code.

job1:
  stage: test
  script:
    - cleanItems="$(cat config/destructiveClean.list | tr -d '\r' | tr '\n' ' ')"
    - >
      if [ ! -z "$cleanItems" ]; then
          echo "=== Additional cleaning ==="          
          for cleanItem in ${cleanItems}; do
              metadaType=${cleanItem%:*}
              echo -ne "Cleaning ${metadaType} "
              sed -i "/${metadaType}/d" destructiveChanges.xml
              echo -e "\rCleaning ${metadaType} "
          done
          isDeleted="$(grep -c "<types>" destructiveChanges.xml)"
          deleted=${isDeleted%:*}
          echo -e "isDeleted ${deleted} "        
          if [ "$deleted" == 1 ]; then
              echo "=== Deleting destructiveChanges.xml file ===" 
              rm - destructiveChanges.xml
          fi          
      else 
          echo -e "Additional cleaning should be added in the config\destructiveClean.list file"
      fi

Thank you very much for you help!!

Shawn · Accepted Answer · 2021-06-20 09:51:54Z

1

When dealing with structured data like XML, it's best to use a tool that actually understands the format, like xmlstarlet, which can be given an XPath expression that references the note you want to delete:

$ xmlstarlet ed -d '//_:types[_:name="ApexClass"]' input.xml
<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<Package xmlns="http://soap.sforce.com/2006/04/metadata">
  <types>
    <name>EmailTemplate</name>
    <members>CC_CaseScenarioMatchingJobQueueable</members>
  </types>
  <types>
    <name>CustomObject</name>
    <members>CC_CaseScenarioMatchingJobQueueable</members>
  </types>
</Package>

answered Jun 20, 2021 at 9:51

Shawn

53.9k3 gold badges29 silver badges74 bronze badges

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

3 Comments

Shawn Over a year ago

To edit the file itself, xmlstarlet ed --inplace -d ...

Erick Barbosa Over a year ago

Hi Shawn! Frstly, thank you so much for helping me with this. My job is running without errors, however, is not removing what I have for ApexClass variable. xmlstarlet ed --inplace -d '//_:types[_:name="${metadaType}"]' DestructiveChanges.xml The variable "metadaType" has the value "ApexClass". I added a Cat command after to ran this code and the DestructiveChanges.xml still the same. Could you help me again, please?

Erick Barbosa Over a year ago

Hi Shawn, could you help me with this, please? Thank you so much!

fpmurphy · Accepted Answer · 2021-06-21 04:24:25Z

xmlstarlet is typically not installed on a system by default, but xsltproc typically is and can be used to transform your input into the desired output using an appropriate stylesheet (XSL - XML Stylesheet Language).

$ cat demo.xsl
<?xml version="1.0" encoding="UTF-8"?>

<xsl:stylesheet version="1.0"
                xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
                xmlns:p="http://soap.sforce.com/2006/04/metadata" >
<xsl:output method='xml' encoding="UTF-8" />

<!-- the identity transform -->
<xsl:template match="@*|node()">
    <xsl:copy>
        <xsl:apply-templates select="@*|node()"/>
    </xsl:copy>
</xsl:template>

<!-- here is where the real work occurs -->
<!-- note the use of a namespace prefix (p:) -->
<xsl:template match='//p:types[p:name="ApexClass"]' />

</xsl:stylesheet>

$ xsltproc demo.xsl input.xml
<?xml version="1.0" encoding="UTF-8"?>
<Package xmlns="http://soap.sforce.com/2006/04/metadata">
  
  <types>
    <name>EmailTemplate</name>
    <members>CC_CaseScenarioMatchingJobQueueable</members>
  </types>
  <types>
    <name>CustomObject</name>
    <members>CC_CaseScenarioMatchingJobQueueable</members>
  </types>
</Package>

Collectives™ on Stack Overflow

Parsing a XML using Shell Script?

2 Answers 2

3 Comments

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

3 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related