Grep / awk

[GrouchyGaijin]

Hi Guys,

I have a decent command line script that will let me add, edit and view my Google tasks.
I've fed the output into Conky using kaivalagi's conkyText script.


This gives me:

GrouchyGaijin's list
   0 [ ] Test Task 
     Due Date: Fri, May 03, 2013
   1 [ ] Another Test Task
     Due Date: Fri, May 03, 2013

Is there a way I can automatically change the output of the script to go from the above example to

0; Test Task; Due Date: Fri, May 03, 2013
1; Another Test Task; Due Date: Fri, May 03, 2013

Or does that fact that the length and content of the output will change depending on the tasks make it impossible?

[papibe]

Hi GrouchyGaijin.

I think 'awk' would be more suitable for that job.

This works (except some trailing spaces):

$ awk '/\[ \]/ {n=$1; $1=$2=$3=""; printf("%s;%s; ",n,$0); getline; print $0}'  file.txt

0;   Test Task;      Due Date: Fri, May 03, 2013
1;   Another Test Task;      Due Date: Fri, May 03, 2013

Regards.

[GrouchyGaijin]

Thank you!
I truly do appreciate the help.

After I run my script I now get:

0;   Example Task;      Due Date: Fri, May 03, 2013
1;   Another example task;      Due Date: Fri, May 03, 2013
2;   Task3;      Due Date: Fri, May 03, 2013

Which is exactly what I asked for. You truly do rock.
I was wondering if you could tell me how to remove the 0 1 and 2 at the beginning but keep the semi-colon?

[papibe]

I highlighted the parts that take care of the number:

awk '/\[ \]/ {n=$1; $1=$2=$3=""; printf("%s;%s; ",n,$0); getline; print $0}'  file.txt

So if you get rid of that:

$ awk '/\[ \]/ {$1=$2=$3=""; printf(";%s; ",$0); getline; print $0}'  file.txt

;   Test Task;      Due Date: Fri, May 03, 2013
;   Another Test Task;      Due Date: Fri, May 03, 2013

Alternatively, you can also use this other version:

awk '/\[ \]/,gsub("^.*]","") {printf(";%s; ",$0); getline; print $0}'  file.txt

Let us know if you need more details to understand the code.
Regards.

[GrouchyGaijin]

Thank you.
Actually I have no understanding of what you did, but would really like to be able to do this without bothering people for help.
I'll check the awk man page, but often I simply don't understand the manual.

[schragge]

How about

awk -F'\n[ \t]*' -vRS='[0-9]+ *[[] []] *' -vOFS='; ' 'NR>1{print "",$1,$2}' file.txt

or even

awk -F\\n -vRS='[0-9]+ *[[] []]' 'NR>1,$0=";"$1";"$2' file.txt

[GrouchyGaijin]

@Schragge Thanks man, once again you come to my coding aid.