using 'sed'...

Hailey's_Comet · 03-23-2007, 12:18 AM

I am writing a script to allow a user to properly set up a "saa7130/4" tv tuner card, and the only thing I can't figure out is this...

I need to 'cat' a file that has this:

Code:

and put it in the variable $choices for:

Code:

choice=$(kdialog --title "$title" --radiolist "$message" $choices)

but i need it in the form:

Code:

"3" "15" "23" "45" "73" "92"

I wish that I could find an adequate explanation for regexp, but can't. Please explain with response if possible...

Jerre Cope · 03-23-2007, 12:40 AM

Maybe awk:

cat file | awk '{ printf( "\"%s\"\t",$1)}' | awk '{ print $0}'

I used to pipe things around in sed and awk, now I just use perl. perldocs.org has all you need to get started. All the things you learned about regular expressions in awk and sed will be useful in perl as well.

Hailey's_Comet · 03-23-2007, 02:49 AM

could you either explain that regexp, or point me to a good page/site that can?

Hailey's_Comet · 03-23-2007, 03:22 AM

Actually, this is exactly what I'm looking for:

cat text | awk '{ printf("\"%s\" ",$1)}' | awk '{ print $0}'

to get:

"3" "23" "33" "73"

and not:

"3" (tab) "23" (tab) "33" (tab) "73"

thank you!

pixellany · 03-23-2007, 06:31 AM

At http://tldp.org:
Bash guide for beginners
Advanced Bash scripting guide

Sed and awk:
http://www.grymoire.com/Unix/

Hailey's_Comet · 03-27-2007, 02:34 PM

thanks for the link pixellany, I am going to read it this weekend. Until I can afford some C/C++ books, I am sticking to shell programming. (Not that I mind much, after all I am a DOS wizard, why not bash too?)