sed command not working

[aliyesam]

i want sed to only output lines that do not have "01","03","05","07","10","11" on the 9th and 10th position
e.g look at the file below
I want only the three records from this file that have "33" ,"17" and "00" at 9th and 10th position.
433483433339
167710001710
167730600000

operating system is sun solaris and Linux .

oracle:$ sed '/^[0-9]\{8\}\(0[1357]\|1[01]\)/d' a.txt
000000001000
433483433339 <<< want this to be spitted out by sed 33 at 9th n 10th pos
121121211100
167710001710 <<< want this to be spitted out by sed 17 at 9th and 10th pos
167735250310
167735260510
167735280710
167735301010
167735431010
167735451010
167710101110
167730691110
167730600000 <<< want this to be spitted out by sed 00 at 9th and 10th pos

 

[feherke]

Hi

I think your [tt]sed[/tt] may use BRE ( Basic Regular Expression ), while the [tt]|[/tt] is ERE ( Extended Regular Expression ), at least according to The Open Group Base Specifications.

What I would try :
[ul]
[li]

sed -E '/^[0-9]\{8\}\(0[1357]\|1[01]\)/d' a.txt

[/li]
[li]

sed -E '/^[0-9]{8}(0[1357]|1[01])/d' a.txt

[/li]
[li]

sed -r '/^[0-9]{8}(0[1357]|1[01])/d' a.txt

[/li]
[li]

sed '/^[0-9]\{8\}0[1357]/d;/^[0-9]\{8\}1[01]/d' a.txt

[/li]
[/ul]

BTW, your code works fine as it is with GNU [tt]sed[/tt] 4.4.


Feherke.
feherke.github.io

 

[Beilstwh]

Why not the following. His requirements don't say that the file name is numeric, only that position 9-10 be one of 00,17, or 33

sed -E '/^.{8}(?:33|17|00)/d' a.txt

Bill
Lead Application Developer
New York State, USA