*****************************************************************

Date: Thu, 19 Aug 2004
From: Marco Verzocchi

awk '{print $2}' > input.txt | /home/mverzocc/binLinux/fastsum

Useful for adding a column of floats.
Goes through the input.txt file, reads the second column and
using a little C program performs the sum.

******************************************************************

Date: Fri, 13 Aug 2004 20:28:00 -0500 (CDT)
From: buehler <buehler@fnal.gov>

cat input.txt | awk '{if ($12>10) print $0}' > output.txt

It goes through input.txt line by line, checks if the number of events, 
$12 (12th column), is larger than 10 (for example), and if so prints
out the entire line ($0) into output.txt.

******************************************************************

Date: Fri, 28 Nov 2003 06:40:48 +0100 (CET)
From: Laurent Duflot <duflot@lal.in2p3.fr>

If you want to know how many lines there are in a file, you can do

wc -l myfile

******************************************************************

Date: Fri, 28 Nov 2003 13:48:34 -0600
From: Stilianos Kesisoglou <kesisogl@fnal.gov>

Look at the following small piece of code (syntax is for
sh/ksh/bash UNIX shells):
 
#!/bin/bash
 
#   Input file provided. Test if the file exists first.
inFile='myData';    
if [ -f ${inFile} ]; then echo "Input File Missing";
exit ; fi
 
#   Initialize index.
idx=0;
 
#   Loop over the contents of the inFile and make outFiles.

while [ ${idx} -lt $(wc -l < ${inFile}) ];

do awk 'BEGIN{X='$((2*${idx}))';Y='$((2*${idx}+3))'} {if (NR>=X &&
NR<=Y) print $0}' > outFile_${idx}
 
idx=$((${idx}+1))
   
done < ${inFile}
 
 Your main interest is inside the "while" loop. If you look at the awk
argument it is a concatenation of the following strings:
 
    1)    'BEGIN{X='
 
    2)    $((2*${idx}))
 
    3)    ';Y='
 
    4)    $((2*${idx}+3))
 
    5)    '} {if (NR>=X && NR<=Y) print $0}'
 
Because the shell resolves any variable references BEFORE it handles
the argument to awk, strings (2) and (4) are substituted by a number 
before the whole fixed string is made and passed to awk. So essentially 
awk takes as an input a fixed string BUT with numbers X and Y that change
for every loop execution.
 
    A general advise for shell script writing is to use the Bourne family
(sh, ksh or bash shells).  This family is by far superior in writing shell 
scripts than the C-Shell family.

******************************************************************

Date: Fri, 28 Nov 2003 14:33:27 -0600
From: Paul S. Russo <russo@fnal.gov>

Common text file manipulation patterns


To append file 2 to file1 giving file3:

$ cat file1 file2 > file3


To pipe all lines from file1 to the wc (word count) command:

$ cat file1 | wc


To get the first N lines of file1:

$ head -N file1

or equivalently:

$ head -n N file1


To get the rest of file1, skipping the first N lines:

$ tail +N file1


To get the last N lines of file1:

$ tail -N file1


To get only the lines from file1 that match a pattern:

$ egrep 'pattern' file1


To get lines from file1 which *do not* match a pattern:

$ egrep -v 'pattern' file1


To translate tabs to an equivalent number of spaces in file1:

$ cat file1 | col -x


To make it possible to search the output of the man command:

$ man topic | col -bx | less


To select columns 1, 5, 9, 10, and 11 from the output of the ls -l command:

$ ls | tr -s ' ' | cut -d ' ' -f 1,5,9,10,11

where tr -s ' ' squeezes multiple spaces to just one, and cut -d ' ' -f 1,5,9,10,11
divides each line into columns delimited by the ' ' character and selects
fields (columns) 1, 5, 9, 10, and 11 (attributes, size, name [including symlink]).


To see the contents of your PATH environment variable more clearly:

$ echo $PATH | tr ':' '\n'


To sort file1:

$ sort -o file1 file1


To sort the output of ls -l by size (column 5):

$ ls -l | sort -b -n -k 5,5


If you know both file1 and file2 are almost the
same but still different (like two directory listings
that are supposed to contain the same files but don't
quite):

To find which lines are in file1 but not in file2:

$ comm -23 file1 file2


To find which lines are in file2 but not in file1:

$ comm -13 file1 file2


To find which lines are in both file1 and file2:

$ comm -12 file1 file2


If the output of these commands is not making sense try:

$ sort -u -o file1 file1
$ sort -u -o file2 file2

then use the comm command.

******************************************************************

Date: Thu, 27 Nov 2003 21:21:31 -0600
From: Stilianos Kesisoglou <kesisogl@fnal.gov>

    The following command takes the lines from X=10 to Y=17 (included lines
X and Y) from the file "inFile" and places them on the file "outFile".

cat inFile | awk 'BEGIN{X=10;Y=17} {if (NR>=X && NR<=Y) print $0}' > outFile

******************************************************************

Date: Thu, 11 Sep 2003 12:43:22 -0500
From: Peter Tamburello <tamburel@fnal.gov>

"I have text file containing an integer number in each line.
Is there an easy way to get the sum of those numbers?"

awk '{sum+=$1; print sum}' file.txt | tail -1