Search Results

The best way would probably be to:-

Convert this to seconds since the Epoch (197001010000 in the same format)
Add the required seconds to change the timezone
Convert back to your format....

Solaris 5_10 comes with Perl 5.8 by default which has the Time::Local module by default.

This allows conversion to epoch

perl -MTime::Local -ne '...

You could also try something like:
awk '
BEGIN { FS = OFS = ";"
}
FNR == NR {
a[$1,$2] = $0
next
}
($1,$2) in a {
$0 = a[$1,$2]
}
1' file1 file2
which produces the output you...

Hello,

Following may help you in same.

awk -F";" 'FNR==NR{X[$1 FS $2]=$3;next} ($1 FS $2){if(X[$1 FS $2]){print $1 FS $2 FS X[$1 FS $2]} else {print $1 FS $2}}' File1 File2


Output will...

You can call Perl within a bash script, just as you can call sed or awk, the example above, using the same quotes could be dropped into a script as a single command. (Though a few comments might be...

Tryawk 'NR==FNR {T[$12]=$28
next
}
$19 in T {sub (/tre/, "RSL", $29)
if ($29 != T[$19]) print...

Quoting Don Cragun:

So something like the following WARNING UNTESTED OFF THE TOP OF MY HEAD CODE FOLLOWS:

perl -ne 'BEGIN {open file1_fh, "<", "/path/to/file1";
while(<file1_fh>){@r=split/;/,$_;
push...

You could try something like:
awk '
BEGIN { FS = OFS = ";"
}
FNR == NR {
x[$1] = $2
next
}
$1 in x {
$2 = x[$1]
}
1' file[12]which (with your sample files) produces the output:
cellRef...

Don't use double quotes around the filenames (i.e. file[12]) as they need to be expanded by the shell. Assuming a typo in your samples' first lines, I'd suggest "replacing field 1 in file 2 with...

I read it as "n == 3". But you're right, it's not really clear.

If the intent is "n >= 3", then you either:
- Use the scrutinizer shorter solution designed for "n >= 3", or
- Change == 3 to >=...

That should work, what is your OS and version?


---
Alternative for 3 or more occurrences, not exactly 3 (not sure what you require):
awk '$2>=M[$1]{M[$1]=$2} ++A[$1]==3{print $1,M[$1]}' file
...

Here is a way to do it using awk.
$ awk '{cnts[$1]++} $2 > fld2[$1] {fld2[$1]=$2; fld0[$1]=$0} END {for (key in cnts) { if (cnts[key] == 3) print fld0[key] }}' input3
A 5
B 5
C 10

cat tone
two
three
four
five
six
seven
eight
nine
ten

tail -n 3 t
eight
nine
ten

awk '(s-NR)<3' s=$(cat t | wc -l) t
eight
nine
ten

man tail

Maybe some like this works
awk '(s-NR)<500' s=$(cat file | wc -l) file

This is how to delete
tail -n 500 oldfile > newfile
newfile then contains what you want.

Most efficient is to read the file once and store the lines in a circular buffer.
An attempt with awk

awk '
{s[i++]=$0}
{i=i%500}
(i in s){print s[i]}
' file

Perl with its compact arrays...

This removes the 500-1 last line. He like to save the last 500 lines, not delete.

Oh sorry, then it's simply
tail -500 fileFor academic interest, the circular buffer is still to be used but must be printed at the very end

awk '
{s[NR%n]=$0}
END {
for (i=NR+1;i<=NR+n;i++)...

Not just academic interest. With some implementations of tail the internal buffer is so small that it may not be able to handle 500 lines...


--
Alternative circular buffer (on Solaris use...

Another approach:
[ $( wc -l < file ) -gt 500 ] && tac file | head -500 | tac

For academic interest, I have bug-fixed my 2nd sample.
It also makes use of NR instead of an extra variable.

Scrutinizer presented a fix for Jotne's 1st sample
awk '(s-NR)<500' s="$(wc -l...

$ tac file | awk 'NR<=500' | tac

printf '%s\n' 1,-500d w q | ed -s file 2>/dev/null

Regards,
alister