Shell Programming and Scripting

INPUT FILE:

Code:

9780743565219    "GODS OF NEWPORT"    "JAKES, JOHN"  2006

OUTPUT FILE I NEED to CREATE FROM INPUT FILE:

Code:

cd /data/audiobooks/9780743565219
~/Desktop/mp3-to-m4b 9780743565219-GODS OF NEWPORT "GODS OF NEWPORT" "JAKES, JOHN" 2006 n

---------- Post updated at 04:19 PM ---------- Previous update was at 04:17 PM ----------

There will be multiple rows on the input file, I only provided one line for the example..

For the output, I need the newline after:
"cd /data/audiobooks/9780743565219"

I NEED all quotes to remain like the example shows, I color coded the variable data to make it easier to see what changes.

Code:

nawk -F'"' '{gsub(" ","",$1);print "cd /data/audiobooks/" $1; print "~/Desktop/mp3-to-m4b" OFS $1 "-" $2 OFS FS $2 FS OFS FS $4 FS OFS $5 OFS "n"}' myFile

VERY CLOSE!! VGERSH!! only one problem, there is too much of a space between the 97834343434 number and the hyphen, and the word after the hyphen.. they should all be touching..

Here is output from your code

cd /data/audiobooks/9780743566629
~/Desktop/mp3-to-m4b 9780743566629 -ME TO WE "ME TO WE" "KIELBURGER, CRAIG"

I need
cd /data/audiobooks/9780743566629
~/Desktop/mp3-to-m4b 9780743566629-ME TO WE "ME TO WE" "KIELBURGER, CRAIG"

given:

Code:

9780743565219    "GODS OF NEWPORT"    "JAKES, JOHN"  2006

using:

Code:

nawk -F'"' '{gsub(" ","",$1);print "cd /data/audiobooks/" $1; print "~/Desktop/mp3-to-m4b" OFS $1 "-" $2 OFS FS $2 FS OFS FS $4 FS OFS $5 OFS "n"}' myFile

I get:

Code:

cd /data/audiobooks/9780743565219
~/Desktop/mp3-to-m4b 9780743565219-GODS OF NEWPORT "GODS OF NEWPORT" "JAKES, JOHN"   2006 n

Is something wrong?
Provide a sample data which is 'broken' - please use code tags, when posting data/code samples.

I just tested vgersh99's code and it works as expected for me. Check to ensure that when the code was cut and pasted, or retyped, that "-" (no spaces inside of the quotes) was coded rather than " -" (has a leading space before the dash). The space might not be obvious in the code and would cause the problem you illustrated.

If that seems not to be the problem, then post the command you are running.

I apologize, my text editor was displaying incorrectly, I am not sure why. The code is good, but can you please explain it a little?

Glad to hear that you've figured it out. Here is a bit of explaination:

First some comments inline with the code.

Code:

nawk -F '"' '
{
    gsub( " ", "", $1); # remove all blanks from first token
                        # literally for each blank (" ") substitute "" (nothing)

    print "cd /data/audiobooks/" $1;    # print the cd command (print automatically adds a new line)

    # print the mp3-to-m4b command with the necessary parameters
    print "~/Desktop/mp3-to-m4b" OFS $1 "-" $2 OFS FS $2 FS OFS FS $4 FS OFS $5 OFS "n"
}'

The -F '"' parameter tells awk to use the double quote (") as the field seperator rather than the default (whitespace). This allows the tokens within double quotes to be seen as a single token, but does have some side effects. The obvious side effect is the trailing spaces on $1 that are removed with the gsub() call.

There are other ways to pick up tokens within double quotes; this is probably the easiest and given the input data, the side effects aren't impossible to work with.

In awk, the contents of two variables, and/or strings, can be concatenated simply by placing them next to each other. Hence, print a "-" b will print the values contained in a and b with a dash character between them; no spaces.

The OFS variable is set by awk to the output field seperator which is a space by default. The FS is set as the input field seperator (set to a double quote on the command line). The way FS is used on the print statement serves to cause output to be enclosed in double qoutes.

I prefer to use commas on a print statement, which has the same effect of separating the parameters with the OFS value. In this case I'd also use escaped quote marks as I believe it makes the code easier to read -- both are a matter of personal preference, or may be specified in coding standards depending on where you are working:

Code:

print "~/Desktop/mp3-to-m4b", $1 "-" $2, "\""$2"\"", "\""$4"\"", $5, "n"

Some will argue that the escaped quotes is confusing, but to me it's more obvious than using FS. An alternative is to use printf() which is my ultimate preference because of the formatting control:

Code:

printf( "~/Desktop/mp3-to-m4b %s-%s \"%s\" \"%s\" %s n\n", $1, $2, $2, $4, $5 );

A HTML based primer for awk is available here:
Awk - A Tutorial and Introduction - by Bruce Barnett

I think it's a great reference.