Shell command- to extract a string from xml file

Question

There is a function Parse_xml as below

 Parse_XML() { TDIR=$1 _VERSION= _REVISION= _FILENAME= _COMPONENT= _DESCRIPT= _ISITOA=0 _NOLOG=0 _OAVERSION= local TMP=/tmp/tmpfile.txt-$$ local JUNK # find the cpq_package XML file and assign it to file local file= for xmlfile in *.xml do if [ -n "$(head ${xmlfile} | grep '<cpq_package')" ] ; then file="${xmlfile}" break fi done if [ -z "${file}" ] || [ ! -f "${file}" ] then _NOLOG=1 return fi ${echo} `grep \<version $file|awk -F = '{print $2}'|awk '{print $1}'|tr -d '"'` > $TMP read _VERSION JUNK < $TMP ${echo} `grep \<version $file|awk -F '=' '{print $3}'|awk '{print $1}'|tr -d '"'` > $TMP read _REVISION JUNK < $TMP _OAVERSION=${_VERSION} _VERSION=${_VERSION}${_REVISION} 

here the version and revisions fetched from xml file from this line

<version value="GPK5" revision="B" type_of_change="1"/> <version value="GPK5" revision="" type_of_change="1"/> 

here some of the revision are empty string and some are having 1 character so the command

 grep \<version CP057761.xml|awk -F = '{print $2}'|awk '{print $1}'|tr -d '"' 

is fetching all the version from xml and store in TMP file. And command

grep \<version CP057761.xml|awk -F '=' '{print $3}'|awk '{print $1}'|tr -d '"' 

is fetching revisions of all the version headers from xml with different versions.

so sometimes the revision of previous version if fetched and added to a version which has empty revision.

How I can modify this command

 ${echo} `grep \<version $file|awk -F = '{print $2}'|awk '{print $1}'|tr -d '"'` > $TMP read _VERSION JUNK < $TMP ${echo} `grep \<version $file|awk -F '=' '{print $3}'|awk '{print $1}'|tr -d '"'` > $TMP read _REVISION JUNK < $TMP _OAVERSION=${_VERSION} _VERSION=${_VERSION}${_REVISION} 

to search only the value in _VERSION variable in xml file and fetch it's particular version. so when it has revision, the _VERSION prints GPK5B and when its empty, the _VERSION prints GPK5.

I fixed the issue by searching the $_VERSION in grep of revision instead \<version. it fetched me only revisions with that particular version and read _REVISION JUNK $TMP fetched me the revision So basically I wanted only latest revision along with version. I regret, I wasn't clear with my question before.

Answer 1

Use an XML parser to parse XML data. xmlstarlet is one.

Given file.xml containing

<root> <version value="GPK5" revision="B" type_of_change="1"/> <version value="GPK5" revision="" type_of_change="1"/> </root> 

Then

xmlstarlet sel -t -m '//version' -v '@value' -v '@revision' -n file.xml 

Outputs

GPK5B GPK5 

Answer 2

Don't use sed nor regex to parse HTML/XML you cannot, must not parse any structured text like XML/HTML with tools designed to process raw text lines. If you need to process XML/HTML, use an XML/HTML parser. A great majority of languages have built-in support for parsing XML and there are dedicated tools like xidel, xmlstarlet or xmllint if you need a quick shot from a command line shell.. Never accept a job if you don't have access to proper tools.

xidel is the most advanced XML/HTML parser in command line out there.

His syntax is more intuitive than xmlstarlet and xmllint when you know xpath query language:

xidel -e '//version/(@value||""||@revision)' -s file.xml GPK5B GPK5