script help

[sheevm]

Hi All,

I have the following lines as input:


10:27:01 0 0 0 0 100
1 0 1 0 99
2 3 0 0 97
3 6 1 0 92



I want to format the ouput as:

Ex:
10:27:01-0-1-2-3-0-0-3-6-0-1-0-1-0-0-0-0-100-99-97-92


Thanks!!
Rajim

[James R. Ferguson]

Hi Rajim:

# perl -nale 'push @a,join "-", @F;END{print join "-",@a}' file

Regards!

...JRF...

[sheevm]

Hi James,

Thanks. I just executed your command it kind of worked. but not exactly. Pardon me, I do not know perl.

10:27:01 0 0 0 0 100
1 0 1 0 99
2 3 0 0 97
3 6 1 0 92

I want (2nd row through 4th row coulmn data)

10:27:01-0-1-2-3-0-0-3-6-0-1-0-1-0-0-0-01-100-99-97-92

Your command gives

10:27:01-0-0-0-0-100-1-0-1-0-99-2-3-0-0-97-3-6-1-0-92

Thanks.

Can you suggest me a good book on perl for dummies

[Sandman! · Accepted solution]

Try the awk script below...


awk '{
if(NR==1) {
tmp=$1
for (i=2;i<=NF;++i) {
if (x[i])
x[i]=x[i]"-"$i
else
x[i]=$i
}
}
else
for (i=1;i<=NF;++i)
x[i+1]=x[i+1]"-"$i
}END{
for (i=1;i<=NF+1;++i) {
if (z)
z=z"-"x[i]
else
z=tmp
}
print z
}' infile

[Hein van den Heuvel]

Rajim,

That was a horrible problem specification.
>> I want (2nd row through 4th row coulmn data)

Your data suggests you wnat 1st thru 5th unless there is a timestamp, in which case you want 2 - 6

That appears to be a horrible output format.
What if you ever get an other line of data?

Anyway, here is a chunk of perl with can read multiple blocks of your data pattern:

---- input ---
10:27:01 0 0 0 0 100
1 0 1 0 99
2 3 0 0 97
3 6 1 0 92
10:27:23 0 a 0 0 80
1 b 1 0 81
2 c 0 0 82
3 d 1 0 83

--------------- output --------

10:27:01-0-1-2-3-0-0-3-6-0-1-0-1-0-0-0-0-100-99-97-92
10:27:23-0-1-2-3-a-b-c-d-0-1-0-1-0-0-0-0-80-81-82-83

----------- perl ------
while (<>) {
@words = split;
if (/^\d*:\d+:/) {
&myprint;
undef @data;
$line = 0;
$time = shift @words;
}
$col = 0;
while (defined ($x = shift @words)) {
$data[$line][$col++] = $x;
}
$line++;
}
&myprint;

sub myprint {
if ($time) {
print "$time";
for $c (0..4) {
for $l (0..$line-1) {
print '-',$data[$l][$c];
}
}
print "\n";
}
}

[Peter Nikitka]

Hi,

this seems to me a simple awk solution:

awk -v col=5 'function out () {
printf dtm;for (j=1;jNF == (col+1) {if(found) out();dtm=$1; for (i=2;i<=NF;i++) a[i-1]=$i;found++}
NF == col {for (i=1;i<=NF;i++) a[i]=a[i]"-"$i}
END {if(found) out()}' output-file

For customization, set
awk -v col=5 ...
to the appropriate value. Multiple blocks (starting with an additional date field) are processed and reported.
Lines not containing col or (col+1) columns are ignored.

mfG Peter

[sheevm]

Thanks for all of your responses.

I will be submitting points soon.