Replace Tag of and xml file after pattern matching using shell

Question

I have an input xml file like below

INPUT FILE

<ReconSummary>
<entryName>Total Deep</entryName>
<Code>777</Code>
<License>L</License>
<Tran>H20</Tran>
<job>1234</job>
</ReconSummary>
<ReconSummary>
<entryName>Total Saurav</entryName>
<Code>666</Code>
<License>L</License>
<Tran>H20</Tran>
<job>1234</job>
</ReconSummary>
<ReconSummary>
<entryName>Total Ankur</entryName>
<Code>555</Code>
<License>L</License>
<Tran>H20</Tran>
<job>1234</job>
</ReconSummary>

I want to comment the tag if i find the pattern "Total Deep" so that the output file looks like below

OUTPUT

<!--<ReconSummary>
<entryName>Total Deep</entryName>
<Code>777</Code>
<License>L</License>
<Tran>H20</Tran>
<job>1234</job>
</ReconSummary>-->
<ReconSummary>
<entryName>Total Saurav</entryName>
<Code>666</Code>
<License>L</License>
<Tran>H20</Tran>
<job>1234</job>
</ReconSummary>
<ReconSummary>
<entryName>Total Ankur</entryName>
<Code>555</Code>
<License>L</License>
<Tran>H20</Tran>
<job>1234</job>
</ReconSummary>

Since i am new to shell scripting, can anyone help me out as to how i can apply this with the help of shell scripting?

• berndbausch: There are tools for XML processing. To start, see stackoverflow.com/questions/91791/….
• DEEP MUKHERJEE: i have to do with shell scripting
• berndbausch: Assuming that <ReconSummary> is not nested, use awk. When you detect this tag, collect lines in an array until you reach <entryName>. If it contains Total Deep, write "<!--", then the lines from the array. Write the rest until you hit the closing tag, then write "-->". If it contains something else, do the same but omit the comment strings.
• DEEP MUKHERJEE: Since i am new to shell scripting can you please help me out with the code for awk?
• berndbausch: It would take me too long to do that for free.

Answer 1

It's easy to use an XML parsing tool like xmlstarlet to delete bits of XML matching some particular XPATH pattern.

In this case, to delete all ReconSummary nodes whose entryName child node has the value Total Deep, you would do

xmlstarlet ed -d '//ReconSummary[entryName = "Total Deep"]' file.xml >newfile.xml

... assuming your file is a well-formed XML file (which your example isn't since it lack as single top-level node).

However, to comment out a bit of XML is a bit trickier, and xmlstarlet can't do it directly.

You can instead apply an XSL transformation which rewrites your XML.

Using an answer that I wrote to a similar question, the XSL transformation

<xsl:transform xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">
 <xsl:template match="/|node()|@*">
 <xsl:copy>
 <xsl:apply-templates select="node()|@*"/>
 </xsl:copy>
 </xsl:template>
 <xsl:template match="//ReconSummary[entryName = 'Total Deep']">
 <xsl:text disable-output-escaping="yes">&lt;!-- </xsl:text>
 <xsl:copy-of select="."/>
 <xsl:text disable-output-escaping="yes"> --&gt;</xsl:text>
 </xsl:template>
</xsl:transform>

would insert the XML comment markers <!-- and --> around every ReconSummary node whose entryName child node had the value Total Deep.

You would apply this XSL transformation to your XML file using either of

xmlstarlet tr transform.xsl file.xml >newfile.xml

or

xsltproc transform.xsl file.xml >newfile.xml

where you file is file.xml and the transformation is stored in transform.xsl.