how to replace field for each record

Shell Programming and Scripting
1

Hello,

I have the following xml formatted file. I would like to get the newnumber field number and replace into customernumber for each record.
For example:

<XMLFORMAT>
<customernumberR11>9</customernumberR11>
<newnumberR11>30</newnumberR11>
<customerdetailR11>

</XMLFORMT>
<XMLFORMAT>
<customernumberR11>20</customernumberR11>
<newnumberR11>18</newnumberR11>
<customerdetailR11>

</XMLFORMT>
<XMLFORMAT>
<customernumberR11>11</customernumberR11>
<newnumberR11>32</newnumberR11>
<customerdetailR11>

</XMLFORMT>

expected output:
<XMLFORMAT>
<customernumberR11>30</customernumberR11>
<newnumberR11>30</newnumberR11>
<customerdetailR11>

</XMLFORMT>
<XMLFORMAT>
<customernumberR11>18</customernumberR11>
<newnumberR11>18</newnumberR11>
<customerdetailR11>

</XMLFORMT>
<XMLFORMAT>
<customernumberR11>32</customernumberR11>
<newnumberR11>32</newnumberR11>
<customerdetailR11>

</XMLFORMT>

Many thank...and please advise.

2

What have you done to attempt to solve this problem yourself?

Post your sample script, and we'll see how we can assist.

Cheers,
ZB

3

i have no idea to write the script to fix...

normally, I need to do it manually...:mad:

4

Try something like this

#!/usr/bin/perl

open( FH, "input.txt" ) || die "Couldn't open file...\n";

while ( <FH> ) {
   $data .= $_;
}

$data=~ s/(<customernumberR11>)[0-9]*(<\/customernumberR11>\n<newnumberR11>)([0-9]*)(<\/newnumberR11>)/$1$3$2$3$4/g;

print $data;

close( FH );

exit( 0 );

Cheers
ZB

5 
sed "/<customernumberR11>/{N;s/\(>[0-9]*<\)\(.*\)\(>[0-9]*<\)/\3\2\3/;}" filename
6

you have 186 posts, by now you should know how to start writing scripts or at least start something.

awk '/<customernumberR11>/{ line=$0 }
     /<newnumberR11>/{  
        current=$0
        gsub("<newnumberR11>|</newnumberR11>","",$0)
   	gsub(/>(.*)</ , ">"$0"<",line)
	print line
	print current	
	next
     }
     !/<customernumberR11>/ && !/<newnumberR11>/ {print}
' "file"
7

thank...but it look like not work

awk: syntax error near line 4
awk: illegal statement near line 4
awk: syntax error near line 5
awk: illegal statement near line 5
awk: syntax error near line 10
awk: bailing out near line 10

8

Try nawk instead of awk

Another way always with awk (or nawk)

awk -F '[<>]' '
   $2=="customernumberR11" { next }
   $2=="newnumberR11" { print "<customernumberR11>" $3 "</customernumberR11>"}
   1
' file
9

yes...I tried to you nawk or awk. but the script look like not work. It move the cusomernumber after the newnumber and duplicated the newnumber

<XMLFORMAT>

<newnumberR11>30</newnumberR11>
<customernumberR11>30</customernumberR11>
<newnumberR11>30</newnumberR11>
<customerdetailR11>

10

The two awk solutions (ghostdog74 and mine) work fine on my AIX box
Input file:

$ cat xml.txt
<XMLFORMAT>
<customernumberR11>9</customernumberR11>
<newnumberR11>30</newnumberR11>
<customerdetailR11>

</XMLFORMT>
<XMLFORMAT>
<customernumberR11>20</customernumberR11>
<newnumberR11>18</newnumberR11>
<customerdetailR11>

</XMLFORMT>
<XMLFORMAT>
<customernumberR11>11</customernumberR11>
<newnumberR11>32</newnumberR11>
<customerdetailR11>

</XMLFORMT>

$

The ghostdog74' solution :

$ cat xml2.sh
awk '/<customernumberR11>/{ line=$0 }
     /<newnumberR11>/{  
        current=$0
        gsub("<newnumberR11>|</newnumberR11>","",$0)
        gsub(/>(.*)</ , ">"$0"<",line)
        print line
        print current   
        next
     }
     !/<customernumberR11>/ && !/<newnumberR11>/ {print}
' xml.txt
$ sh xml2.sh
<XMLFORMAT>
<customernumberR11>30</customernumberR11>
<newnumberR11>30</newnumberR11>
<customerdetailR11>

</XMLFORMT>
<XMLFORMAT>
<customernumberR11>18</customernumberR11>
<newnumberR11>18</newnumberR11>
<customerdetailR11>

</XMLFORMT>
<XMLFORMAT>
<customernumberR11>32</customernumberR11>
<newnumberR11>32</newnumberR11>
<customerdetailR11>

</XMLFORMT>

$

My solution :

$ cat xml.sh
awk -F '[<>]' '
   $2=="customernumberR11" { next }
   $2=="newnumberR11" { print "<customernumberR11>" $3 "</customernumberR11>"}
   1
' xml.txt
$ sh xml.sh
<XMLFORMAT>
<customernumberR11>30</customernumberR11>
<newnumberR11>30</newnumberR11>
<customerdetailR11>

</XMLFORMT>
<XMLFORMAT>
<customernumberR11>18</customernumberR11>
<newnumberR11>18</newnumberR11>
<customerdetailR11>

</XMLFORMT>
<XMLFORMAT>
<customernumberR11>32</customernumberR11>
<newnumberR11>32</newnumberR11>
<customerdetailR11>

</XMLFORMT>

$
11

oh...very sorry..
forgot provide one more IMPORTANT information. The customernumber and newnumber field is not the same field number for each records. Sometime, the customernumber in row 1, sometime in row 2. Therefore, it difficult for me..

<XMLFORMAT>
....
...
<customernumberR11>9</customernumberR11>
<newnumberR11>30</newnumberR11>
<customerdetailR11>

</XMLFORMT>
<XMLFORMAT>
....
<customernumberR11>20</customernumberR11>
...
<newnumberR11>18</newnumberR11>
<customerdetailR11>

</XMLFORMT>
<XMLFORMAT>
....
...
...
<customernumberR11>11</customernumberR11>
....
..
<newnumberR11>32</newnumberR11>
<customerdetailR11>

</XMLFORMT>

12

while read line
do
str1=`grep 'customernumberR11' file1 | cut -c 20-21`
str2=`grep 'newnumberR11' file1 | cut -c 15-16`
if [ "$str1" -ne "$str2" ] ; then
sed -e "s/newnumberR11/ s/$str2/\$str1/g" $line
fi
done < file1

13

Try this :

awk -F '[<>]' '
   $2=="customernumberR11" { customernumber = 1; next}
   $2=="newnumberR11" { number = $3 ; newnumber = 1; next}
   customernumber && newnumber {
      printf("<customernumberR11>%d</customernumberR11>\n<newnumberR11>%d</newnumberR11>\n", number, number);
      customernumber = newnumber = 0;
   }
   1
' xml.txt