Remove specific lines after matching a pattern

Question

I have a file like below:

chs_de_le_q1gg
fd_tr_mn_0
fd_tr_mn_06
fd_tr_mn_070
fd_tr_mn_0716
fd_tr_mn_09013
fd_tr_mn_092433
fd_tr_mn_1020333
fd_tr_mn_12013332
dsdas_1
dsdas_12
dsdas_212
sdasd_4567
weqwe_32323
dasds_232322
2321321_rewrwe_3233
32_Ff
asdasd_1_fff

I need to remove the lines ending with digits ranging from 4-8 of length; also to remove lines starting from numbers.

the below works to match the lines which I want to delete, but then doing invert match on that match with adding -iv doesn't work.

cat test.txt | grep -oP '(?<![0-9])[0-9]{4,8}(?![0-9])'

Answer 1

To remove lines which:

• starts with a digit: ^[0-9] or ^\d

• or ends with 4~8 character length of digits: [0-9]{4,8}$ or \d{4,8}$.

  grep -vE '^[0-9]|[0-9]{4,8}$'  infile >output
  sed -E '/^[0-9]|[0-9]{4,8}$/d' infile >output
  grep -vP '^\d|\d{4,8}$' infile        >output
  

To remove lines which:

• starts with a digit: ^[0-9]

• or ends with exactly 4~8 character length of digits: [^0-9][0-9]{4,8}$ or \D\d{4,8}$.

  grep -vE '^[0-9]|[^0-9][0-9]{4,8}$'  infile >output
  sed -E '/^[0-9]|[^0-9][0-9]{4,8}$/d' infile >output
  grep -vP '^\d|\D\d{4,8}$' infile            >output
  

Answer 2

Using any POSIX awk:

$ awk '!( /^[0-9]/ || /(^|[^0-9])[0-9]{4,8}$/ )' file
chs_de_le_q1gg
fd_tr_mn_0
fd_tr_mn_06
fd_tr_mn_070
dsdas_1
dsdas_12
dsdas_212
asdasd_1_fff

Answer 3

Using Raku (formerly known as Perl_6)

Below deletes lines starting with a digit, or containing runs of 4-to-8 digits adjacent to the line ending (unless is synonymous with if not):

raku -ne '.put unless /^ \d | \d ** 4..8 $ /;'  

Sample Input:

chs_de_le_q1gg
fd_tr_mn_0
fd_tr_mn_06
fd_tr_mn_070
fd_tr_mn_0716
fd_tr_mn_09013
fd_tr_mn_092433
fd_tr_mn_1020333
fd_tr_mn_12013332
dsdas_1
dsdas_12
dsdas_212
sdasd_4567
weqwe_32323
dasds_232322
2321321_rewrwe_3233
32_Ff
asdasd_1_fff

Sample Output:

chs_de_le_q1gg
fd_tr_mn_0
fd_tr_mn_06
fd_tr_mn_070
dsdas_1
dsdas_12
dsdas_212
asdasd_1_fff

Note: Above would delete a line such as fd_tr_mn_123456789, since 8-digits are adjacent to the $ end-of-line. If the OP desires to retain runs of digits greater-than 8-digits adjacent to the line ending, the following regex works:

raku -ne '.put unless /^ \d | \D\d ** 4..8 $ /;' 

Above, a line such as fd_tr_mn_123456789 would be retained in the return.

https://raku.org