On 3/10/2008 3:46 PM, happytoday wrote:
> On Mar 10, 2:32 pm, Ed Morton <mor...@[EMAIL PROTECTED]
> wrote:
>
>>On 3/10/2008 1:13 AM, happytoday wrote:
>>
>>
>>>I am piping sed command to an awk program but I got that error
>>>message :
>>># sed -f namestate list | bystate
>>>ksh: bystate: not found
>>>#
>>
>>That has nothing to do with awk so you should post the question to
>>comp.unix.shell or similair. Having said that, below is a suggestion to
improve
>>your awk script...
>>
>>
>>
>>
>>
>>
>>>Though bystate is a chmod 777 and contain that commands :
>>
>>># cat bystate
>>>#!/bin/sh
>>>awk -F, '{
>>> print $4 ", " $0
>>> }' $* |
>>>sort |
>>>awk -F, '
>>>$1 == LastState {
>>> print "\t" $2
>>>}
>>>$1 != LastState {
>>> LastState = $1
>>> print $1
>>> print "\t" $2
>>>}'
>>
>>... |
>>awk -F, '
>>$1 != LastState { LastState = $1; print $1 }
>>{ print "\t" $2 }
>>'
>>
>> Ed.- Hide quoted text -
>>
>>- Show quoted text -
>
>
> Can I know How can I implement that command with Dos versions:
>
> # sed -f namestate list | bystate
>
> How can I convert that script to be run under awk95.exe
>
>
> # cat bystate
> #!/bin/sh
> awk -F, '{
> print $4 ", " $0
> }' $* |
> sort |
> awk -F, '
> $1 == LastState {
> print "\t" $2
>
>
> }
>
>
> $1 != LastState {
> LastState = $1
> print $1
> print "\t" $2
You need to ask in a DOS group as you're asking how to call awk in a pipe
with
multiple OS-specific programs and so that has nothing to do with the awk
language as discussed in this NG.
Alternatively, drop the OS-specific part (mainly the call to UNIX sort)
and
rewrite it in GNU awk which has built in sort functions and will, I
believe, run
on both DOS and UNIX. If you post some small sample input and expected
output
someone could probably help you write a GNU awk script.
Ed.
|
