Search Results

Hi,
It's wrong
elif [ $bad_record_count <= $error_limit ];
expected
elif [ $bad_record_count -le $error_limit ];
and else
if [[ $chksanret = 0 ]]
expected
if [ $chksanret = 0 ]
or
if [...

modified code and the order of the files matters.
awk '
{dep=substr($0,361,9)}
FNR==NR {
f1[$1]=dep
next
}
NF && (!($1 in f1) || f1[$1]==dep)' prod_last.dat prod.dat

there you go:

awk '
{dep=substr($0,361,9)}
FNR==NR {
f1[$1]=dep
next
}
NF && $1 in f1 && f1[$1]==dep' prod.dat prod_last.dat

put in a file and attach it to the thread.
The current implementation is based on you "mock-up" and might not be relevant at all to the issue at hand.

a bit of a longhand - making some assumptions based on the file samples.....

awk -f mia.awk file1.txt file2.txt
where mia.awk is:

FNR==NR {
if (split($1,a, "_") ==3)
f1[$1]
next
}...

In AIX the ps command itself has such a provision: ps -T

On the other hand:
process_count=`ps -ef|grep $current_user|grep -v $current_process|grep -v $parent_process|grep -v grep|grep -v ps|grep...

s/SDV/STR/

There should be no space after the equal sign:
HEADRREC="0012PVGRSCDVSDV 005 00000000000000000000 2014 0.00"
Try variable expansion to get the first field:
CUTVAR=${HEADRREC%% *}

CUTVAR=$(echo "$HEADERREC"|awk '{print $1}')

Make it CUTVAR=$(echo "$HEADERREC"|awk '{print $1}') (shell's command substitution)

It should work then:
$ cat script
#!/bin/ksh

DIV=$1

if [[ "$DIV" = @(ter|bom|hre|nte|mol) ]]
then
echo "something"
else
echo "Please enter a correct division"
fi

Output:
$ ./script...

>> test.log 2>&1 to also redirect stderr.

You can use group command (http://www.gnu.org/software/bash/manual/html_node/Command-Grouping.html#Command-Grouping) { ...; } and append output to file:
{ command 1; command 2; command 3; } >>...

You might try sed book "Definitive Guide to sed": sed book - Definitive Guide to sed (http://www.sed-book.com/)

This would break if there are file names with spaces and/or special characters. To avoid that you would need to use properly quoting:
print "$FILE"
tail -1 "$FILE"

Also, wc -c only counts bytes,...

Why don't you use a loop:-
for file in sent/*
do
tail -1 $file | wc -c
done
OR
for file in $( find sent/ -type f )
do
tail -1 $file | wc -c
done
OR you can write a script which will...

Hi MIA651,

One way using perl. It prints the file name, a colon and the number of characters in last line:
$ perl -lne 'if ( eof ) { printf qq|%s: %d\n|, $ARGV, length; }' *
3.txt: 12
AAAB.txt:...

Using awk in the find command:


find sent/ -type f -exec awk 'END{printf("%d\t%s\n",length($0),FILENAME)}' {} \;


Prints out number of characters and then the filename.

Try:
find sent/ -type f -exec tail -1 {} \; | awk '{print length}'
To also count the linefeed at the end of the line, like wc -c or wc -m use awk '{print length+1}'



--
Use wc -c counts...

That can do the job if your AWK implementation treats a multicharacter RS as a regular expression and if you set RS to a regular expression which mimicks the default value of FS. In effect, what is...

Try awk '{for (i=1;i<=NF;i++) if (length($i)>max) max=length($i)} END{print max}' file

You can use awk with substr funtion.

Syntax:-

awk ' { print substr(string, start, length) } '

When setting a shell variable you can't have spaces around the "=". You can in awk; but not in shell.

awk -v OFS=, 'NR>4{exit}{gsub("-"," -");print $1,$2,$3+0,$4+0}' OFMT='%.2f' input

try this... (with some assumption)

awk -v OFS=, 'NR>5{exit}{gsub("-"," -");print $1,$2,$3,$4}' xtr3.rpl