Shell Programming and Scripting

Create the file not_found (or other name of your choice) containing the awk program.
Make the file executable : chmod +rx not_found
Execute :

Code:

not_found /home/padlist $(find . -type f)

This command display all the strings from /home/padlist that have not been found in any of the files of curent directory and sub-directories.

Jean-Pierre.

Hi.
What is the purpose of d2? And, do you intend to search for the entire string, CL1ACC01,"-13181,-32907", for example, including quotes?... cheers, drl

No, I just plan to search for the 1st set of characters up to the 1st coma. So really I only need the CLxxxxxx info. I've tried Jean-Pierre's method and for some reason it returns nothing, even with a search string I know it won't find. Otherwise I get no errors when I run it.

John

Hi, John.

My perl program, and I think aigles' awk program will read entire lines from padlist as the patterns.

So either we will need to adjust our programs or provide a preliminary step to extract the text up to (and not including) the first comma.

This is an important lesson in communicating requirements. For the future, it is best to provide a sample input and expected output. Even if the data is fake or trivial, we would been able to see more clearly what you wanted. This is also what ghostdog74 was getting at with his question.

You didn't answer my question about "d2" in your script ... cheers, drl

DRL,
Sorry I did not answer your question about d2. I had intended to use comas as a delimiter for a read and just prior to your last post did I realize the awk program would use the entire line for the search string. But I still cannot get it to work for me. So I'll just keep at it.

John

My awk program assumes space as Input Field Separator for the first file.
To use comma separator modify the program :

Code:

#!/usr/bin/awk  -f

#
# Initialize input Field Separator
#

BEGIN {
   FS = ","
}

#
# Strings file.
#

An example:
Strings file: strings.txt

Code:

lulu,xxx
bob,xxx
albert,xxx
gerald,xxx

Input files : Test/*.txt

Code:

Test/bob.txt: bob
Test/john.txt: john
Test/lulu.txt: lulu
Test/toto.txt: toto

Execution:

Code:

$ not_found strings.txt Test/*.txt
albert
gerald
$

If nawk is available on your system, replace awk by nawk in the first line of the program.


Jean-Pierre.

Jean-Pierre,
It's still not working for me. Here is what I have.

String file: D43E6309_padlist

CL123456 xxx
CL1ACC01 xxx
CL1ACF01 xxx

Input files: Too many and too big to show here.

awk program: compare_die2

#!/usr/bin/awk -f
# Filename: compare_die2
#
# String file
#
NR==FNR {
string_found[$1] = 0; # 0 = No, >0 = Yes
next;
}
#
# New data file
# Build array with strings not yet found
#
FNR==1 {
for (str in string_found) {
if (string_found[str] == 0) {
strings[str]++;
strings_count++;
}
}
if (strings_count == 0) exit;
}
#
# Input data
# Search data for strings not yet found
#
{
for (str in strings) {
if ($0 ~ str) {
string_found[str]++;
delete strings[str];
if (--strings_count == 0) exit;
}
}
}
#
# No more files or allstrings have been found
# Print strings not found
#
END {
for (str in strings_found)
if (string_found[str] == 0) print str;
}


I did a chmod 777 compare2_die. Here is the line I executed.

/home/rihnj/compare_die2 /home/rihnj/D43E6309_padlist $(find . -type f) > strings_not_found

Do you see what I've coded wrong or missed? strings_not_found file always comes up empty. The CL123456 search string is not to be found anywhere in directories where the search is being done.

Thanks, John