how to remove all comments from a xml file least the ones containing the "HELP" word?-CodePudding

I need to know the xml matching code to let me clean a file of all comments least the ones containing the HELP word.

I am currently using perl to clean multiline comments (like failed tests) from a xml, least the ones I hand picked and are helpful to the end user:
perl -i -w -0777pe 's///sg' somefile.xml

But, If there is a way to obtain the same result using a xml mathing string, I would prefer as there may have some exception that regex may not handle, but for now this is what I have to use.

Obs.: I will use it with xmlstarlet on linux, so it would be better if the solution work with it too.

CodePudding user response：

Here is XSLT based solution.

The XSLT is very simple. One single line template removes not needed comments. The rest is just a boilerplate code for a so called Identity Transform pattern.

$ xml tr path/to/transformation.xslt path/to/source.xml

Input XML

<?xml version="1.0"?>
<root>
    <!--The Sunshine State-->
    <state>FL</state>
    <!--HELP is needed-->
    <city>Miami</city>
</root>

XSLT

<?xml version="1.0"?>
<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
    <xsl:output method="xml" encoding="utf-8" indent="yes" omit-xml-declaration="no"/>
    <xsl:strip-space elements="*"/>

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

    <!--keeps comment if it contains HELP-->
    <xsl:template match="comment()[not(contains(., 'HELP'))]" />
</xsl:stylesheet>

xmlstarlet
xmlstarlet ed -d '//comment()[not(contains(.,"HELP"))]' path/to/source.xml

Output

<?xml version='1.0' encoding='utf-8' ?>
<root>
  <state>FL</state>
  <!--HELP is needed-->
  <city>Miami</city>
</root>