awk's -INTRO-

[phaza7]

If this was posted anywhere on this forum let me know cause I didn't see it anywhere when I search. So I will explore some of the uses of the powerful awk language. Yes, it's a language so may argue since it's can be used on the cmd-line within a Unix console. This tutorial assumes UR familiar with using Unix commands and editors.

Awk, is a pattern-scanning & processing language. Kewl Huh! It offers power user's a command-line tool to maniplate input. Correct me if I made a mistake, cause I'm no expert. Awk as I understood it is useful for doing interesting things with input. My first experience with Unix I didn't have anuff ram to run the gui. So I was forced to use the console until I built a new box. *"Awk first saw light of day with version 7 UNIX, around 1978."-O'Reilly "sed and awk" . Enough with the boring history for now let explore how we can use it's power.
type at console prompt:

$ echo -e "AO MEMBERS make a kewl community">AOer's
          $ echo -e "AO MEMBERS are very helpful">>AOer's
          $ echo -e "AO best site on the Web">>AOer's
          $ awk '{ print "AO ADDICTS & SENIOR'S RULE"} AOer's

output should look as follows:
AO ADDICTS & SENIOR'S RULE
AO ADDICTS & SENIOR'S RULE
AO ADDICTS & SENIOR'S RULE

The above example as I said before awk reads input. It prints the "AO ADDICTS' & SENIOR'S RULE" for each line of input read from the file "AOer's".

Now to better understand awk as a programming language. It is helpful to write awk scripts
this makes it second nature in using it since it is easy to learn Below is a description of awk's three procedures we can use which consists of the 3 routines: BEGIN, main loop, END.

1st routine-- BEGIN
#this is done before input is read, this can be useful for all sorts of things
# use your imagination.

2nd routine--main loop
# this is the action to perform on the input, kewl I like

3rd routine--END
#this is done after input is read, this is also useful

Always comment your scripts it make it easier to debug later. Comment with the # symbol they can be placed anywhere within your scripts. Do U understand the above, if not maybe this will help. Type at prompt

 $ echo a b c d | awk 'BEGIN {one=1, two =2}
                                                          > {print $(one + two)}'

c output. This help U understand the laws of using awk. Ok, Ok:-V !! Let's do something more interesting with awk. Let's make a file called "AOTOP10POSTERS"
I hope AO members aren't upset with me using them as working examples. If U don't know how to create a file from the console do a

$ cat>AOTOP10POSTERS
  Tedobl,2261
  Souleman,2031
  KorpDeath,1774
  Terr,1718
  ac1dap3ctrum,1466
  s0nic,1422
  Ennis,1272
  conf1rm3d k1ll,1168
  valhallen,1149
  victorkaum,1124

on new line hit ctrl-D or use your favorite editor.

Now we can work with awk to read the input of the new AO file at the prompt type:
awk -F, '{printf "%-16c\t %-04d\n", $1, $2}' then enter.
The '-F, ' tells awk the field separator is a comma. {printf "%-16c\t %-04d\n", $1, $2}' is the action to perform on each line of input read in. "printf" is a C language formatting function. "%" tells how to line up the fields literally. "-" is to left align. The numbers tells how many characters to display in each field. The trailings c's & d's represent c = any character and the d = the field is to be an integer. The \t is a tab & \n is a new line or carriage return.

Remember the 3 routines of awk's law we can use the BEGIN, main loop, END. Open your favorite editor and write the simple awk script. The BEGIN routine is done before input is read

BEGIN {
              { FS = "," } 
                  { print "\t\t AO's Top 10 Posters\n" }
              { "date" | getline d }
              { printf "\t %s\n", d }
              { print "Members Name\n" }
              { print "==========================================\n" }
              } #this ends the begin routine and makes a header for the input of file
#now for main loop
              { printf "%-16c\t %-04d\n", $1,$2 }
#this is how we want the input formatted
#beginning the end routine
            END { print " "       #makes a blank line
                       print NR, "records processed."    #shows numbers of input lines read
                       total = $2   #set variable total to second field
                       avg = total/3      #avgrage for top posters base on three posts per day
# and the number post for that user
                       print $1, avg}  #print the member name and the avg total for each user.

save as awkao then run within the console with the following statement:

 $ awk -f awkao AOTOP10POSTERS

output is:
AO'S Top 10 Posters
(Today's date )
Members name
================================
Tedol 2261
souleman 2031
Korpdeath 1744
Terr 1718
ac1dsp3ctrum 1466
s0nic 1422
Ennis 1272
conf1rm3d Kill 1168
valhallen 1149
victorkaum 1124

10 records processed.
tedobl 753
souleman 677
KorpDeath 591
Terr 572
ac1dsp3ctrum 488
s0nic 477
Ennis 424
conf1m3d k1ll 389
valhallen 383
victorkaum 374
Sweet :-V it works.
btw, This is the end of the short Intro to awk. If U found this helpful let me know then I'll make a 2nd tut going more in depth using advance awk features. thaxs for UR time 2 read this Hey! UNIX is user-friendly, it's just picky about it's friends.

[HTRegz]

For you windows users out there who are interested in playing with awk.... there is a version available for Win32 called gawk.

It is part of the GNU Win32 Project... you will also find many other gnu/open source *nix applications available here.


http://gnuwin32.sourceforge.net/

[t2k2]

Pardon my ignorance, as I am fairly new to programming and languages of this nature, but what advantages does this have over Perl or a tool like grep? Thanks for the tut phaza, and thanks for the link HTRegz. I just finished downloading it for Windows.


Cheers,

t2k2

[phaza7]

first, it has none over perl.
second, grep matches patterns in a file
third awk process's files; which later we can maniuplate with test to get a certain look in output file. grep is a cmd-line tool as well as awk. awk is a command that lets U do programming logic on input. if running a *nix flavor do a man awk. or try
awk To me since I love programming languages, awk is a handy tool for certain tasks to give a file a certain look. That I want. It can't do nearly what perl or python my favorite can do. So don't throw them out the window. Hope, this helps-thanxs for the feed back.

PS,
Awk is merely a editor.
Sed is a stream editor, handy for deleting headers of emails that I want to save in a folder. Awk is very powerful when combine with a shell script.

[t2k2]

Thanks for clarifying that for me. It looks like a pretty handy tool. I just wish I had more time to play with programming and scripting languages and what not...

[vescovono]

thanks for the tut phaza, amazing that someone did not have it here already. look forward to other updates on this topic.

awk has helped me out a bunch in the past, and is certainly a viable tool for any *nix user-owner-sa, etc, and it is very flexible in usage.

if you like spending $$$, i would also recommend the O'Reilly book on sed/awk. great reading.

just for grins, I took a quick look to add to this and found a few more sources:

http://www.vectorsite.net/tsawk.html

http://stud.wsi.edu.pl/~robert/awk/

thanks again phaza!