How to Start and Stop printing lines with Specific Text in Linux/Unix Terminal

Question

Print file on terminal Starting and Ending in a line with specific Text

For Example: file.txt

2019/10/17 23:01:02 -W- Blah Blah Blah.....
2019/10/17 23:01:02 -I- Blah Blah Blah.....
2019/10/17 23:01:02 -I- Blah Blah Blah.....
2019/10/17 23:01:02 -I- Start
2019/10/17 23:01:02 -I- ======================
2019/10/17 23:05:02 -I- Summary Details
2019/10/17 23:10:02 -I- End
2019/10/17 23:10:02 -I- Blah Blah Blah.....
2019/10/17 23:10:02 -I- Blah Blah Blah.....
2019/10/17 23:10:02 -I- Blah Blah Blah.....

I tried to display the lines with grep cmd, but I am unable to the required output.

Expected Result: Command to get the results based on system date

> some command in grep (or) awk (or) sed with system date for current date

2019/10/17 23:01:02 -I- Start
2019/10/17 23:01:02 -I- ======================
2019/10/17 23:05:02 -I- Summary Details
2019/10/17 23:10:02 -I- End

https://www.google.com/search?q=bash+how+to+filter+lines+between+patterns https://stackoverflow.com/questions/17988756/how-to-select-lines-between-two-marker-patterns-which-may-occur-multiple-times-w — KamilCuk, Oct 20 '19 at 09:33
Possible duplicate of [Bash, grep between two lines with specified string](https://stackoverflow.com/questions/22221277/bash-grep-between-two-lines-with-specified-string) — KamilCuk, Oct 20 '19 at 09:34

David C. Rankin · Accepted Answer · 2019-10-20T21:30:22.637

A simple way in awk is just keep a flag and set n = 1 when "Start" is encountered in field 4. Checking n == 1 gives you a test to print each line while n is 1. When "End" is reached, simply print that line and exit, e.g.

awk '$4 == "End" {print; exit} $4 == "Start" {n = 1} n == 1' file

(note: the operation associated with the rule n == 1 is simply the default operation of print. It is equivalent to writing n == 1 {print})

Example Use/Output

Using your data file in file, you would get:

$ awk '$4 == "End" {print; exit} $4 == "Start" {n = 1} n == 1' file
2019/10/17 23:01:02 -I- Start
2019/10/17 23:01:02 -I- ======================
2019/10/17 23:05:02 -I- Summary Details
2019/10/17 23:10:02 -I- End

It can also be written as:

awk '$4 == "Start" {n = 1} n == 1; $4 == "End" {exit}' file

Edit - Todays Date Only

If you want to match today's date only, you simply need to pass the date in the format you need as a variable to awk from the shell using the -v option, e.g.

awk -v date="$(date +%Y/%m/%d)" '
    $1 == date { 
        if($4 == "Start") n = 1
        if(n == 1) {print}
        if($4 == "End") exit
    }
' file

Example Input

$ cat file
2019/10/17 23:01:02 -W- Blah Blah Blah.....
2019/10/17 23:01:02 -I- Blah Blah Blah.....
2019/10/17 23:01:02 -I- Blah Blah Blah.....
2019/10/17 23:01:02 -I- Start
2019/10/17 23:01:02 -I- ======================
2019/10/17 23:05:02 -I- Summary Details
2019/10/17 23:10:02 -I- End
2019/10/17 23:10:02 -I- Blah Blah Blah.....
2019/10/17 23:10:02 -I- Blah Blah Blah.....
2019/10/17 23:10:02 -I- Blah Blah Blah.....
2019/10/20 23:01:02 -W- Blah Blah Blah.....
2019/10/20 23:01:02 -I- Blah Blah Blah.....
2019/10/20 23:01:02 -I- Blah Blah Blah.....
2019/10/20 23:01:02 -I- Start
2019/10/20 23:01:02 -I- ======================
2019/10/20 23:05:02 -I- Summary Details
2019/10/20 23:10:02 -I- End
2019/10/20 23:10:02 -I- Blah Blah Blah.....
2019/10/20 23:10:02 -I- Blah Blah Blah.....
2019/10/20 23:10:02 -I- Blah Blah Blah.....

Example Use/Output (for 2019/10/20)

$ awk -v date="$(date +%Y/%m/%d)" '
>     $1 == date {
>         if($4 == "Start") n = 1
>         if(n == 1) {print}
>         if($4 == "End") exit
>     }
> ' file
2019/10/20 23:01:02 -I- Start
2019/10/20 23:01:02 -I- ======================
2019/10/20 23:05:02 -I- Summary Details
2019/10/20 23:10:02 -I- End

Let me know if you have further questions.

Thanks for your reply. @David C.Rankin. How to edit awk to include current system date? I want my command to pick up current date from column 1 — Naga, Oct 20 '19 at 11:06
Isn't that what was done? you have the date `2019/10/17` in the output -- or are you saying forget the date in the input file and just replace it by the current system date? — David C. Rankin, Oct 20 '19 at 11:47
Yes everyday date will change, I need the command to pick up the current date automatically and show up the lines only for the current day. For example if I run the command today it should show up the start and end portion of lines for 2019/10/20, if I run it tomorrow it should show up the start and end lines for 2019/10/21 — Naga, Oct 20 '19 at 11:56

score 0 · Answer 2 · answered Oct 20 '19 at 14:52

Here is an awk script. With explanation: It will solve for printing partial terminating sections as well.

script.awk

BEGIN {FS=" -I- "}  # Field separator is " -I- ", therefore idicative field data is in fiedl $2
$2 ~ "^Start"{onSection = 1} # if indicative field start with "Start", mark section flag ON
onSection{print} # print output only if marked section flag ON
$2 ~"^End"{onSection = 0} # if indicative field ends with "End", mark section flag OFF

running

awk -f script.awk file.txt

or one liner:

awk 'BEGIN{FS=" -I- "}$2~"^Start"{f=1}f{print}$2~"^End"{f=0}' file.txt

output

2019/10/17 23:01:02 -I- Start
2019/10/17 23:01:02 -I- ======================
2019/10/17 23:05:02 -I- Summary Details
2019/10/17 23:10:02 -I- End

William Pursell · Answer 3 · 2019-10-20T21:46:35.677

0

sed -n '/Start$/,/End$/p' input

To filter on today's date:

sed -n "\@$(date +%Y/%m/%d)@"'{ /Start$/,/End$/p; }' input

edited Oct 20 '19 at 21:46

answered Oct 20 '19 at 21:40

William Pursell

204,365
48
270
300

How to Start and Stop printing lines with Specific Text in Linux/Unix Terminal

3 Answers3

script.awk

running

output