Beginner awk Arrays will encounter several common problems

• Original topic:

process The following file contents , take out the domain name and sort the order according to the domain name :( Baidu and Sohu face test )

For a detailed explanation of the answers, please refer to:

http://lidao.blog.51cto.com/3388056/1912219

File contents:

http://www.etiantian.org/index.htmlhttp://www.etiantian.org/1.htmlhttp://post.etiantian.org/index.htmlhttp:// Mp3.etiantian.org/index.htmlhttp://www.etiantian.org/3.html

Required results:

mp3.etiantian.org 1post.etiantian.org 2www.etiantian.org 3

Answer:

1	awk-F "/+" ‘{hotel[$2]++}END{for(pol in hotel) print pol,hotel[pol]}‘url.txt|sort -rnk2

• There are a few questions about some of the little partners who are beginners to awk arrays.

• 
  

  • Why to use the awk array.

  • Why use the $ as room number.

Let's take a look at the problem with this question.

First of all, what are our goals and what are our results?

Here we simplify the problem, the goal is simplified to statistical Www,mp3.post repeated several times, the reader can adjust themselves, and statistics www.etiantian.org,mp3.etiantian.org, post.etiantian.org these domain names.

• www has appeared several times

• MP3 appeared several times.

• The post appeared several times

• First Milestone - simple and easy to understand approach:

if a www is present, the number of repetitions of www is recorded with a , and if mp3 appears then B is used to record the number of repetitions of MP3 , If post is present, the repeat quantity of post is recorded by C .

The answers are as follows:

[[email protected]] #awk-F "[/]+" ' $2~/www/{a++}$2~/mp3/{b++}$2~/post/{c++}end{print "www:" A, "MP3:" B, "Post:" C} ' Url.txt

Demo Process:

[Email protected]]# awk-f "[/]+" ' $2~/www/{a++}$2~/mp3/{b++}$2~/post/{c++}end{print "www:" A, "MP3:" B, "Post:" C} ' Url.txtwww:3 mp3:1 Post:2

Although this is easy to understand, because there are only three kinds of domain name Www,post,mp3 , the general work will have a lot of domain names, you will be crazy.

So we want a method, a thing (a name) can be loaded in these three cases, and do not affect each other.

This is what we call an array (hotel), and these three cases are the elements of the array we're talking about (hotel rooms).

• second Milestone-a Simple and practical awk Array

Or just the idea, divided into three kinds of situations, the different circumstances, this time is put into the hotel's three rooms.

[[email protected] awkfile] #awk-F "[/]+" ' $2~/www/{h["www"]++}$2~/mp3/{h["MP3"]++}$2~/post/{h["POST"]++}end{printh ["www"],h["MP3"],h["POST"]} ' URL.TXT3 1 2

Tip: If you specify a room name (the array element name) for the awk array, be sure to add double quotes. Otherwise , awk will think of this as a variable.

[[email protected] awkfile] #awk ' Begin{h[www]=1;h[bbs]=2;print H[www],h[bbs]} ' 2 2[[email protected] awkfile] #awk ' begin{h["www"]=1;h["bbs"]=2;printh["www"],h["BBS"]} ' 1 2

The example above is that if you don't add double quotes,awk will consider it a variable.

• A third milestone - Seat

Still a lot of trouble how to solve it?

Is there any way to get into the same room when you encounter the same content?

If you encounter www automatically enter www room

If you encounter MP3 automatically enters MP3 room

If you encounter post automatically enter the post room

yes, think about how we get www or mp3 or post through awk ?

They are the second column! awk can be used as a notation.

so , this array can be written as.

h[$2]++ , auto-seat, after the self-added operation (Auto plus 1).

[[email protected] awkfile] #awk-F "[/.] + "' {h[$2]++;p rint h[" www "]} ' Url.txt122233[[email protected]show Awkfile] #awk-F" [/.] + "' {h[$2]++;p rint" www: "h[" www "]," MP3: "h[" MP3 "]," Post: "h[" POST "]} ' url.txtwww:1 mp3:post:www:2 mp3:post:www:2 mp3: Post:1www:2 mp3:1 post:1www:3 mp3:1 post:1www:3 mp3:1 post:2 So although we can see the results we want, but also show the implementation process, we just want results!

• Fourth Milestone - Output Results

as long as the final result can be passed END mode to resolve.

[[email protected] awkfile] #awk-F "[/.] + "' {h[$2]++}end{print" www: "h[" www "]," MP3: "h[" MP3 "]," Post: "h[" POST "]} ' Url.txtwww:3 mp3:1 post:2

This is a room to show the contents of a room, or that sentence when the room a lot of time, you are crazy.

awk naturally has a way of solving this problem, dedicated loops.

[[email protected] awkfile] #awk-F "[/.] + "' {h[$2]++}end{for (Pol in h) print Pol,h[pol]} ' url.txtwww 3mp3 1post 2

• Here we re-understood the awk array again . Review the start of our question.

• 
  

  • Why to use the awk array.

He can do more complicated calculations, and of course we can do the most commonly used statistics (de-weight).

• 
  

  • Why use the $ as room number.

It would be a nightmare if we had to use the room number manually to specify the room.

The Www,mp3,post can be a good indication of the content. Of course, it involves the prior knowledge, choose the right knife, cut out what you want.

• Take a look at how hotel[$2]++ is performing, see.

650) this.width=650; "src=" Https://s3.51cto.com/wyfs02/M00/92/3F/wKiom1j9piqyAjPYAAKAbD6GgNs208.png "title=" Old boy It education-awk array calculation process-Take www as an example. png "alt=" Wkiom1j9piqyajpyaakabd6ggns208.png "/>

• Related Blog content:

awk Array Statistics count (de-weight)

http://lidao.blog.51cto.com/3388056/1912219

awk Array to accumulate (calculate sum)

http://lidao.blog.51cto.com/3388056/1914563

This article is from the "Lee blog" blog, make sure to keep this source http://lidao.blog.51cto.com/3388056/1918859

Beginner awk Arrays will encounter several common problems