AWK script to generate seqence numbers

[kobewins]

I am working a script to add a kind of sequence number as the third column to the data file, starting with a given number.

The data file uses pipeline "|" as the field delimiter.

1|AUGMENTIN 250-62.5 TAB CHEW F| |
2|AUGMENTIN XR 1,000-62.5 TAB F| |
3|AXERT 12.5 MG TABLET F| |
4|AXERT 6.25 MG TABLET F| |
5|AZMACORT INHALER F| |
B|BACTROBAN NASAL 2% OINTMENT F| |
7|BECONASE AQ 0.042% SPRAY F| |
8|BENICAR 20 MG TABLET F| |
9|BENICAR 40 MG TABLET F| |
A|BENICAR 5 MG TABLET F| |

I'd like the output shown as the follwoing:

start=5
1|AUGMENTIN 250-62.5 TAB CHEW F|F0005|
2|AUGMENTIN XR 1,000-62.5 TAB F|F0006|
3|AXERT 12.5 MG TABLET F|F0007|
4|AXERT 6.25 MG TABLET F|F0008|
5|AZMACORT INHALER F|F0009|
B|BACTROBAN NASAL 2% OINTMENT F|F0010|
7|BECONASE AQ 0.042% SPRAY F|F0011|
8|BENICAR 20 MG TABLET F|F0011|
9|BENICAR 40 MG TABLET F|F0011|
A|BENICAR 5 MG TABLET F|F0011|

start=5
cat $data_file | awk '{print $1 "|" $2 "|"}' \
'{printf "F%04d|", ++start}' \
> $result_file

Of course, the above thought did't work out for me.

Any hints and help are greatly appreciated.

 

[feherke]

Hi

And this ?

awk -v start=5 -F '|' 'BEGIN{OFS="|"}{$3=sprintf("F%04d",start++i)}1' $data_file

Feherke.

http://rootshell.be/~feherke/

 

[PHV]

A starting point:
awk -F'|' '
BEGIN{start=5}
{printf "%s|%s|F%04d\n",$1,$2,start++}
' $data_file > $result_file


Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ181-2886

 

[feherke]

Hi

Hell. Typo. From where appears that "i" after the ++ ? Anyway, delete it. ( Although in [tt]gawk[/tt] works even with that. )

awk -v start=5 -F '|' 'BEGIN{OFS="|"}{$3=sprintf("F%04d",start++)}1' $data_file

Feherke.

http://rootshell.be/~feherke/

 

[kobewins]

Both scripts work perfectly. Thank you a lot for the help.

feherke, could you please explain what does that 1 stand for inside
awk -v start=5 -F '|' 'BEGIN{OFS="|"}{$3=sprintf("F%04d",start++)}1' .

 

[vgersh99]

'1' is a shorthand for '{print $0}'

vlad
+----------------------------+
| #include<disclaimer.h> |
+----------------------------+

 

[kobewins]

Now could anyone shed me some light on how to pass a unix shell variable into the awk script, as I want to something like the following?

-----------------
#!/bin/ksh

let st=5

awk -F'|' '
BEGIN{start=$st}
{printf "%s|%s|F%04d|\n",$1,$2,start++}
' $data_file > $result_file

-------------

The above code didn't not start from 5, instead it began with 0.

 

[feherke]

Hi

The value of $st could be passed to the [tt]awk[/tt] script, only if the shell ( the one who knows $st's value ) expands it. But shell expands variable names to their values only when are between double quotes ( " ). And your [tt]awk[/tt] commands are between single quotes ( ' ).

Feherke.

http://rootshell.be/~feherke/

 

[kobewins]

feherke,

I have tried to use both "$st" and $st, all gave me the same result - starting from 0. By the way, I am using AIX unix.

let st=5

awk -F'|' '
BEGIN{start="$st"}
{printf "%s|%s|F%04d|\n",$1,$2,start++}
' $data_file > $result_file

Thank you.

 

[PHV]

st=5
awk -F'|' '
BEGIN{start=[!]'[/!]$st[!]'[/!]}
{printf "%s|%s|F%04d|\n",$1,$2,start++}
' $data_file > $result_file

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ181-2886

 

[kobewins]

Thank you, PHV. It works with the 'forward tics' -- no command substitution.

I have been being confused by these:
"double quotes",
'forward tics', and
`back tics'

Thanks, again.