Print processing file name in FOR loop inside the result file - BASH

Question

I am running a for loop inside a bash file which will check some files (.ts) for a specific string and print the matching lines in a result file.

Here is the code:

#! /bin/bash

for file in *.ts;
do awk -f test_function.awk $file > result.txt;
done

And this is the test_function.awk file:

match($0, /<name>(.*)<\/name>/,n){ nm=n[1] }
match($0, /<source>(.*)<\/source>/,s){ src=s[1] }
/unfinished/{ print "name: " nm, "source: " src }

And this is one of the input files that contains "unfinished" and needs to be included in the output:

<context>
    <name>AccuCapacityApp</name>
    <message>
        <source>Capacity</source>
        <translation type="unfinished">Kapazität</translation>
    </message>
    <message>
        <source>Charge Level</source>
        <translation type="unfinished"></translation>
    </message>
    <message>
        <source>Sel (Yes)</source>
        <translation type="unfinished">Sel (Ja)</translation>
    </message>
    <message>
        <source>Esc (No)</source>
        <translation type="unfinished">Esc (Nein)</translation>
    </message>
</context>

It gives output like this:

name: AccuCapacityApp source: Capacity
name: AccuCapacityApp source: Charge Level
name: AccuCapacityApp source: Sel (Yes)

And this is one of the input files that doesn't contain "unfinished" and needs to be excluded from the output:

<context>
    <name>ATM FSM state</name>
    <message>
        <source>Hunting</source>
        <translation>Sync-Suche</translation>
    </message>
    <message>
        <source>Pre-Sync</source>
        <translation>Pre-Sync</translation>
    </message>
    <message>
        <source>Sync</source>
        <translation>Sync</translation>
    </message>
</context>

What I want to do is to print the processing file name in the beginning of each paragrapgh of matching lines in the result file, ONLY when the matching strings are found, like following:

Processign file: alpha.txt
name: AccuCapacityApp source: Capacity
name: AccuCapacityApp source: Charge Level
name: AccuCapacityApp source: Sel (Yes)

Processing file: gamma.txt
name: AccuCapacityApp source: Capacity
name: AccuCapacityApp source: Charge Level
name: AccuCapacityApp source: Sel (Yes)

How can I achieve this?

I know the file name can be appended and then the matching lines can be appended to the result file. But I want to have a blank result file each time I run the bash file and only write the filename and content when the matching string is found. So I think appending the file name will not work. I have tried printing the file name with echo ${file##*/}, echo $file and {print FILENAME};{print "\t" $0} but unable to print as desired.

I don't know where you found for file in $(ls *.ts); but it's a complete anti-pattern. Use for file in *.ts; and quote your variables inside the loop. — Tom Fenech
– Tom Fenech, Commented Sep 8, 2017 at 8:34

Tom Fenech · Accepted Answer · 2017-09-11 10:13:05Z

1

Based on your update, I think this does what you want:

match($0, /<name>(.*)<\/name>/,m){ nm = m[1] }
match($0, /<source>(.*)<\/source>/,m){ src = m[1] }
/unfinished/ { list[++n] = src }
ENDFILE {
    for (i = 1; i <= n; ++i) {
        print "name:", nm, "source:", list[i]
    }
    n = 0
}

Only save elements when unfinished is found, the loop through the list at the end of each file. n keeps a count of the number of matches in the current file.

Use the script like this (no need for a shell loop):

awk -f test_function.awk *.ts > result.txt

Note that ENDFILE is a GNU awk extension, but then so is the third argument to match that you were already using, so I guess that's OK for you.

edited Sep 11, 2017 at 10:13

answered Sep 8, 2017 at 8:38

Tom Fenech

75.1k13 gold badges119 silver badges154 bronze badges

Sign up to request clarification or add additional context in comments.

9 Comments

John Over a year ago

Thank you. The "unfinished" is the string my code is looking a match for.

John Over a year ago

Your first solution works but prints the file name even the matching string is not found. Your second solution doesnt check for "unfinished" string (as you already mentioned it's purpose was unclear). I will try to modify your code and add it myself.

Tom Fenech Over a year ago

I guess you can probably add /unfinished/ { f = 1 } then only print if f is set. Remember that you need to set it back to 0 between files.

John Over a year ago

Umm, I tried this:

match($0, /<name>(.*)<\/name>/,n){ nm=n[1] } match($0, /<source>(.*)<\/source>/,s){ src=s[1] }/unfinished/ { f = 1 } ENDFILE {      if (f==1) {         print "Processing file:", FILENAME         print "name: " nm, "source: " src         print ""         f=0 # for next file     } }

But it didn't work. No matter where I reset the f, it doesn't work. Can you please have a look?

Tom Fenech Over a year ago

The problem is that you haven't shown us an example of your input that corresponds to the output that you want, and "it didn't work" doesn't really give me much to go on. At first glance, it looks like you need to check all three things if (nm != "" && src != "" && f) (then reset all 3 vars).

|

Collectives™ on Stack Overflow

Print processing file name in FOR loop inside the result file - BASH

1 Answer 1

9 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

9 Comments

Your Answer

Sign up or log in

Post as a guest

Related