Login or Register to Ask a Question and Join Our Community


UNIX for Dummies Questions & Answers

Combining awk tests

 
Thread Tools Search this Thread
Top Forums UNIX for Dummies Questions & Answers Combining awk tests
# 1  
Old 08-05-2009
Combining awk tests
It would be convenient to be able to combine awk tests. For example, suppose that I do this query:

Code:
awk '$1 != "Bob" || $1 != "Linda" {print $2}' datafile

Is there a reasonable way to combine the conditions into a single statement? For example, in egrep, I can do:

Code:
egrep -v "Bob|Linda" datafile

That will exclude any line with either Bob or Linda. Nice shorthand, but egrep lacks awk's field specificity. Any way to do this in awk? Sadly, it does not work to do:

Code:
awk '$1 != "Bob|Linda" {print $2}' datafile

# 2  
Old 08-05-2009
awk supports alternation

Code:
awk '$1 ~/Bob|Linda/ {print $2}' datafile

# 3  
Old 08-05-2009
Ah, so apparently it only works with the ~ "contains" operator, not == and !=, etc:

Code:
> cat datafile
Bob  here
Jane here
Ron  absent
Linda absent
Paul unknown

> awk '$1 ~/Bob|Linda/ {print $0}' datafile
Bob  here
Linda absent

> awk '$1 == /Bob|Linda/ {print $0}' datafile

> awk '$1 != /Bob|Linda/ {print $0}' datafile
Bob  here
Jane here
Ron  absent
Linda absent
Paul unknown

That's a shame, but it's more than I had before. Thanks for the reply.
# 4  
Old 08-06-2009
~ is the match operator for regular expressions.

Code:
awk '$1=="bob" || $1=="linda" ' filename

does what you want - exact match
# 5  
Old 08-06-2009
So it does, and apparently this works as well:

Code:
awk '$1 ~ "^bob$|^linda$"' filename

I want to test it a little, but the start-and-end full line match seems to work so far.

Thanks for the replies. It's been very helpful.
 
Login or Register to Ask a Question

Previous Thread | Next Thread

10 More Discussions You Might Find Interesting

1. Shell Programming and Scripting

awk Pattern Matching for PING tests

I have a script that logs into a server and pings several other servers in order to verify IP path between servers. The output can look like this, if good pings: Response from 1.1.1.4;_id=0, vlan_prio=0): seq=0 time=91.547 ms. Response from 1.1.1.4;_id=0, vlan_prio=0): seq=1 time=61.176 ms.... (7 Replies)
Discussion started by: he204035
7 Replies

2. Shell Programming and Scripting

What am I doing wrong in awk with multiple tests?

awk -v pat="$pattern" 'NR == 1 {print $0}; $0 ~ pat {print $0}' infile.csv > outfile.csv The first row of my file contains headers so I want them. Otherwise, I want only lines containing the BASH variable pattern which I am passing to awk with -v. This is giving me all the lines containing... (6 Replies)
Discussion started by: Michael Stora
6 Replies

3. Shell Programming and Scripting

Combining echo and awk

i have a script that has many lines similar to: echo $var | awk -F"--" '{print $2}' as you can see, two commands are being run here. echo and awk. id like to combine this into one awk statement. i tried: awk -F"--" "BEGIN{print $var; print $2}" but i get error messages. (10 Replies)
Discussion started by: SkySmart
10 Replies

4. Shell Programming and Scripting

awk problem - combining awk statements

i have a datafile that has several lines that look like this: 2,dataflow,Sun Mar 17 16:50:01 2013,1363539001,2990,excelsheet,660,mortar,660,4 using the following command: awk -F, '{$3=strftime("%a %b %d %T %Y,%s",$3)}1' OFS=, $DATAFILE | egrep -v "\-OLDISSUES," | ${AWK} "/${MONTH} ${DAY}... (7 Replies)
Discussion started by: SkySmart
7 Replies

5. Shell Programming and Scripting

How to combining awk commands?

I can achieve two tasks with 2 different awk commands: 1) awk -F";;WORD" '{print $2}' file | sed '/^$/d' #to find surface_word 2) awk -F"bw:|gloss:" '// {print $2}' file | sed '/\//!d; s:/*+*: + :g; s:^+::; s: *+ *$::;' #to find segmentation of surface_word Number 1) finds surface_word... (7 Replies)
Discussion started by: Viernes
7 Replies

6. Shell Programming and Scripting

combining awk and sed

Hi experts, I have a requirement, In which I need to display the first and last line of a zip file where the line starts with "L". I've writen the code like below using sed and awk. gunzip -c 20110203.1104.gz | awk '$1 ~ "^L" {print substr($0,178,15)}' | sed -n '1p;$p' Is it possible to do it... (8 Replies)
Discussion started by: senthil.ak
8 Replies

7. Shell Programming and Scripting

Combining awk statements

I have a pretty simple script below: #!/bin/sh for i in *.cfg do temp=`awk '/^InputDirectory=/' ${i}` input_dir=`echo ${temp} | awk '{ print substr( $0, 16) }'` echo ${input_dir} done As you can see its opening each cfg file and searching for the line that has "InputDirectory="... (3 Replies)
Discussion started by: ssbsts
3 Replies

8. Shell Programming and Scripting

Combining two awk scripts

I have a file like this consisting of blocks separated by > of two number X and T > 10 0 13 5.92346 16 10.3106 19 13.9672 22 16.9838 25 19.4407 28 21.4705 31 23.1547 34 24.6813 37 26.0695 40 27.3611 43 28.631 46 29.8366 49 30.9858 52 32.0934 55 33.1458 (6 Replies)
Discussion started by: kristinu
6 Replies

9. Shell Programming and Scripting

combining fields in awk

I am using: ps -A -o command,%cpu to get process and cpu usage figures. I want to use awk to split up the columns it returns. If I use: awk '{print "Process: "$1"\nCPU Usage: "$NF"\n"}' the $NF will get me the value in the last column, but if there is more than one word in the... (2 Replies)
Discussion started by: json4639
2 Replies

10. Shell Programming and Scripting

help combining lines in awk

I seem to have gotten myself in over my head on this one. I need help combining lines together. I have a text file containing 24,000 lines (exactly why I need awk) due to bad formatting it has separated the lines (ideally it should be 12,000 lines total). Example of file: ... (2 Replies)
Discussion started by: blueheed
2 Replies
Login or Register to Ask a Question