CODEWITHSUNDEEP
shellawkgrepsh
mahi
mahi

Reputation: 37

match and print data into CSV file

I'm struggling to fetch data from a file having the same format as mentioned below into a new output file(CSV) using shell Script. 

[+++]          Added Names:          [+++]
     -> Add- abc.txt (732):
     -> Add- bcd.txt (490):
[+++]         Enabled Names:         [+++]
     -> Mod- cde.txt (105):
     -> Mod- efg.txt (105):
[+++]          Deleted Names:          [+++]
     -> Del- fgh.txt (4):
     -> Del- xyz.txt (45):
[+++]      Added non-names lines:     [+++]
     -> Added to test.txt (41):
     -> Added to bgh.txt (41):
        # This distribution may contain names under different licenses.

and the expected output should look like 

Add,abc.txt,732
Add,bcd.txt,490
Mod,cde.txt,105
Mod,efg.txt,105
Del,fgh.txt,4
Del,xyz.txt,45

Here Output renames the same as per last.The only thing which is not required is -> Added to test.txt (41): and -> Added to bgh.txt (41):.which is part of [+++] Added non-names lines: [+++]

Upvotes: 1

Views: 50

Answers (3)

anubhava
anubhava

Reputation: 785176

You can use this awk command with a custom input field separator:

awk -v OFS=, -F '[->:()[:blank:]]+' 'index($0, "[+++]") {
p=($0 ~ / (Added|Deleted|Enabled) Names:/); next} p && /-> /{print $2, $3, $4}' file

Add,abc.txt,732
Add,bcd.txt,490
Mod,cde.txt,105
Mod,efg.txt,105
Del,fgh.txt,4
Del,xyz.txt,45

Upvotes: 3

RomanPerekhrest
RomanPerekhrest

Reputation: 92854

Another GNU awk solution:

awk -v FPAT='[[:alnum:].]+' '/->/{ print $1,$2,$3 }' OFS=',' file

The output:

Add,abc.txt,732
Add,bcd.txt,490
Mod,cde.txt,105
Mod,efg.txt,105
Del,fgh.txt,4
Del,xyz.txt,45

Upvotes: 1

Yoda
Yoda

Reputation: 445

Another approach using gsub function:-

awk '/->/{gsub(/[->():]/,X);$1=$1;print}' OFS=, file
Add,abc.txt,732
Add,bcd.txt,490
Mod,cde.txt,105
Mod,efg.txt,105
Del,fgh.txt,4
Del,xyz.txt,45

Upvotes: 1

Related Questions