Shell Programming and Scripting

Dear friends,
I have one more query,
incase a file contains multiple tags of same name, then how to get the required string between the tags, in which the string begins with "O/M"

i.e., file1.txt contains following text(please note that all the following tags are in single line)
----------------------------
<aaa> some text1</aaa><bbb> some text2</bbb><ccc> some text3</ccc><kkk>W/X12345</kkk><ddd> some text4</ddd><eee> some text5</eee><fff> some text6</fff><kkk>3333</kkk><ggg> some text7</ggg><hhh> some text8</hhh>
----------------------------
please see that above file conatins "<kkk>" twice. i.e.,
<kkk>W/X12345</kkk>
and
<kkk>3333</kkk>

I have to extract the entire string that starts with "W/X" till the end tag("</kkk>")
i.e., I need to extract "W/X12345"

this paramenter,W/X12345, length may not be constant, some times it may be
W/X123
or
W/X87789989
or
W/X90798897989
ect...
please tell me how do I get this paramenter?


waiting for ur reply,
swamymns

lets say your string is str

echo "${str}"|sed 's#W/X.*[!<]//g'

This will give you the string str - contents in W/X....<
the pattern that you need to match is W/X.*[!<] meaning characters starting at W/X followed by any characters till '<'

Hi friend,
thanks for ur reply, but when I use the ur command,
echo "${str}"|sed 's#W/X.*[!<]//g'
where str contains the complete file content.

I ma getting following error...
sed: command garbled: s#W/X.*[!<]//g

please go through my query once again and kindly give me the solution..
waiting for ur reply,

swamymns

dear friends please help me..
incase a file contains multiple tags of same name, then how to get the required string between the tags, in which the string begins with "O/M"

i.e., file1.txt contains following text(please note that all the following tags are in single line)
----------------------------
<aaa> some text1</aaa><bbb> some text2</bbb><ccc> some text3</ccc><kkk>W/X12345</kkk><ddd> some text4</ddd><eee> some text5</eee><fff> some text6</fff><kkk>3333</kkk><ggg> some text7</ggg><hhh> some text8</hhh>
----------------------------
please see that above file conatins "<kkk>" twice. i.e.,
<kkk>W/X12345</kkk>
and
<kkk>3333</kkk>

I have to extract the entire string that starts with "W/X" till the end tag("</kkk>")
i.e., I need to extract "W/X12345"

this paramenter,W/X12345, length may not be constant, some times it may be
W/X123
or
W/X87789989
or
W/X90798897989
ect...
please tell me how do I get this paramenter?


waiting for ur reply,
swamymns

Hi,

This might not be very efficient, but hopefully will work in your case:
cat file_name|tr -d '\n'| awk -F"<[a-zA-Z/0-9]*>" '{for (i=0;i<=NF;i+=1)print $i}' |grep "^W/X*"


(file_name is where you put your filename)

Please let me know how this fares...

i m sorry the command should have been

echo "${str}"|sed 's#W/X.*[!<]##g'

I didnt notice that you said everything is on a single line. I assumed that its spread across all across a file, and a starting tag may/maynot have a matching closing tag. In case of single line:
cat file_name|awk -F"<[a-zA-Z/0-9]*>" '{for (i=0;i<=NF;i+=1)print $i}' |grep "^W/X*"