Bash-remove-n-match: Difference between revisions
Jump to navigation
Jump to search
(Created page with "Really unique problem and resolution using awk found in reddit. [https://www.reddit.com/r/bash/comments/qva5bm/delete_n_occurrence_of_a_pattern/?%24deep_link=true&correlation_...") |
mNo edit summary |
||
| Line 4: | Line 4: | ||
<pre> | <pre> | ||
input text1.txt: | input text1.txt has: | ||
hi 1 | hi 1 | ||
hi 2 | hi 2 | ||
| Line 22: | Line 22: | ||
<pre> | <pre> | ||
awk '!/hi/||++n>4' | awk '!/hi/||++n>4' | ||
</pre> | |||
Possible alternate that would still require some work: | |||
<pre> | |||
count=4 awk -e ' | |||
/hi [[:digit:]]+/ && s++ < ENVIRON["count"] { next }; | |||
{print}; | |||
' << EOT | |||
hi 1 | |||
hi 2 | |||
hi 3 | |||
hi 4 | |||
hi 5 | |||
hello | |||
EOT | |||
</pre> | </pre> | ||
Revision as of 10:13, 17 November 2021
Really unique problem and resolution using awk found in reddit. [1]
input text1.txt has: hi 1 hi 2 hi 3 hi 4 hi 5 hello I want to delete 4 lines containing 'hi' wanted output: hi 5 hello
Unique answer:
awk '!/hi/||++n>4'
Possible alternate that would still require some work:
count=4 awk -e '
/hi [[:digit:]]+/ && s++ < ENVIRON["count"] { next };
{print};
' << EOT
hi 1
hi 2
hi 3
hi 4
hi 5
hello
EOT