unix script to extract values in xml file - unix

I have an XML file as below:
<xml>Workinstance name="suvi" permission="read" id="6543"</xml>
<xml>Projectinstance name="ram" permission="write" id="3534"</xml>
I want to display the workinstance id field from that XML file.


grep '<xml>Workinstance' file.xml | grep -o 'id="[^"]*' | cut -c5-


$ awk '/Workinstance/{ gsub(/.*id=\042|\042.*/,""); print } ' file
6543

Related

No way to move pound (£) symbol in UNIX?

Need get the delimiters at the starting of each line, below are sample input and output files for reference. In actual Delimiter used are( £{, ^$^)
Note - The file to be rearranged has huge data
Have tried the below but it does not work:
tr £{ \\n
sed 's/£{/\n/g'
awk '{ gsub("£{", "\n") } 1'
Input File:
£{firstlinecontinues£{secondstartsfromhereandit
keepsoncontinueingtillend£{herecomes3rdand£{fi
nallyfourthisalsohere
Output File:
£{firstlinecontinues
£{secondstartsfromhereanditkeepsoncontinueingtillend
£{herecomes3rdand
£{finallyfourthisalsohere

With GNU awk for multi-char RS and \s:
$ awk -v RS='£{' 'NR>1{gsub(/\s/,""); print RS $0}' file
£{firstlinecontinues
£{secondstartsfromhereanditkeepsoncontinueingtillend
£{herecomes3rdand
£{finallyfourthisalsohere

awk 'BEGIN{RS="(£{\|\^\$\^)"; OFS=ORS=""}{$1=$1;print $0 (FNR>1?"\n":"") RT}' file

Since the £ symbol is represented by two Octal codes, 302 and 243, I was able to produce the desired result with this perl command:
perl -pe 's/(\302\243)/\n$1/g' data.txt
NOTE: Here's what I see on my system:
echo "£" | od -c
0000000 302 243 \n
0000003

Find word count using wc and assign to a variable

i want the word count wc -w value be assigned to a variable
i've tried something like this, but i'm getting error, what is wrong?
winget="this is the first line"
wdCount=$winget | wc -w
echo $wdCount

You need to $(...) to assign the result:
wdCount=$(echo $winget | wc -w)
Or you could also avoid echo by using here-document:
wdCount=$(wc -w <<<$winget)

You can pass word count without the filename using the following:
num_of_lines=$(< "$file" wc -w)
See https://unix.stackexchange.com/a/126999/320461

You can use this to store the word count in variable:
word_count=$(wc -w filename.txt | awk -F ' ' '{print $1}'

To print GunZip FileName and selected rows -Continuation :

Would like to print first 2 rows from all the files located in the directory along with File Name.
All are *.gz extension files. Having around 100 files in that directory.
sample_jan.csv.gz
10,Jan,100
30,Jan,300
50,Jan,500
sample_feb.csv.gz
10,Feb,200
20,Feb,400
40,Feb,800
60,Feb,1200
Expected Output:
Filename:sample_jan.csv.gz
10,Jan,100
30,Jan,300
Filename:sample_feb.csv.gz
10,Feb,200
20,Feb,400
Tried below command where as Filename appears Blank
zcat sample_jan.csv.gz | awk 'FNR==1{print "Filename:" FILENAME} FNR<3' > Output.txt
Filename:-
10,Jan,100
30,Jan,300
Tried below command where as Filename appears Wrong
awk 'FNR==1{print "Filename:" FILENAME} FNR<3' <(gzip -dc sample_jan.csv.gz) > Output.txt
Filename:/dev/fd/63
10,Jan,100
30,Jan,300
Looking for your suggestions, dont have perl & python.

You can use this one-liner,
for file in *.gz; do echo "Filename: $file"; zcat "$file" | head -2 ; done

AWK print specific line based on search pattern

I have two sets of files test.csv data.xml.
I am trying to grep a specific field from test.csv and search the string in data.xml. If string is found then print the corresponding line in test.csv file.
Example
search string is field 3 server name
test.csv
111,xxx,serversugar,port90
222,yyy,servertorque,port190
333,aaa,serverastrix,port8080
422,yxy,servertorque,port290
data.xml
<group>
<hostname>servertorque</hostname>
<hostname>serverastrix</hostname></group>
Output expected
222,yyy,servertorque,port190
333,aaa,serverastrix,port8080
422,yxy,servertorque,port290

with GNU sed & grep in 2 steps
sed '/>\w\+</!d;s/.*>\(\w\+\).*/\1/' data.xml>pattern.txt
grep -wf pattern.txt test.csv
..output:
222,yyy,servertorque,port190
333,aaa,serverastrix,port8080
422,yxy,servertorque,port290

One way with awk
awk -v FS="[><,]" 'NR==FNR{a[$3]++;next}$3 in a' data.xml test.csv
Test:
$ cat data.xml
<group>
<hostname>servertorque</hostname>
<hostname>serverastrix</hostname></group>
$ cat test.csv
111,xxx,serversugar,port90
222,yyy,servertorque,port190
333,aaa,serverastrix,port8080
422,yxy,servertorque,port290
$ awk -v FS="[><,]" 'NR==FNR {a[$3]++;next} $3 in a' data.xml test.csv
222,yyy,servertorque,port190
333,aaa,serverastrix,port8080
422,yxy,servertorque,port290

I assume you need:
awk -F',' '$3==<string_you_need> { print $0 }' test.csv

to grep a pattern from file compare contents with another file and replace

I want to grep version number in one file and replace it in another file. I want to grep 4.3.0.5 in file 1 and replace it in File 2 at 4.3.0.2. I have the below command to get the number , but how can I cut/replace it in second file??
File1 :
App :4.3.0.5 (or) App: 4.3.0.5-SNAPSHOT
File2: Before editing
grid_application_distribution_url=nexus://com.abcd.efge.ce/App/4.3.0.2/tar.gz/config
File 2 : after editing (Desired Result:)
If $VERISON in File is WITHOUT the word SNAPSHOT then in file 2
grid_application_distribution_url=nexus://com.abcd.efge.ce/App/4.3.0.5/tar.gz/config
If $VERSION has SNAPSHOT then line in file 2 should be
grid_application_distribution_url=nexus-snapshot://com.abcd.efge.ce/App/4.3.0.5/tar.gz/config
VER=$(awk -F: '/^App/{sub(/ .*$/, "", $2); print $2}'/path/file1.txt)
echo $VER
if ($vER ~ /SNAPSHOT/)
/usr/bin/ssh -t -t server2.com "sub("=nexus:", ":=nexus-snapshot") /path/file2" && sub(/[^\/]+\/tar\.gz/, $VER"/tar.gz") /path/file2

Something like this is all you need:
awk -F': +' 'NR==FNR{v=$2;next} {sub(/[^/]+\/tar.gz/,v"/tar.gz")} 1' File1 File2 > tmp && mv tmp File2

This awk script can do the job (this is an enhancement of above answer from #EDMorton):
Splitting the command in 2 as per OP's request
VER=$(awk -F' *: *' '/^App/{print $2}' file1)
awk -v v="$VER" '{
split(v, arr, "-");
sub(/[^\/]+\/tar\.gz/, arr[1]"/tar.gz");
if (arr[2] ~ /SNAPSHOT/)
sub("=nexus:", ":=nexus-snapshot")
}1' file2 > tmpFile
mv tmpFile > file2

You can try with this:
VERSION=($(grep -r "App:" /path/File1| awk '{print ($2)}'))
sed -i "s/4.3.0.2/$VERSION/" File2
it will look for "4.3.0.2" and change by value in $VERSION. File2 will be updated with this change.
If you want the file to keep the same, delete the flag -i:
sed "s/4.3.0.2/$VERSION/" File2
You will get the result in stdout.
As indicated in comments, 4.3.0.2 is not like this every time. Adapted for format X.Y.Z.W:
sed "s/\/[0-9][0-9]*\.[0-9][0-9]*\.[0-9][0-9]*\.[0-9][0-9]\(\/tar.gz\)/\/$VERSION\1/" File2

Resources