SED替换匹配行，但如果被注释掉则不行

Question

My text contains a line like this:

DHCPD_INTERFACE="eth4"

There are other lines containing #DHCPD_INTERFACE=

I want to replace the whole line DHCPD_INTERFACE="eth4" matching on DHCPD_INTERFACE= but not if the line contains a #

I've tried: sed -i.bak '/^#/!s/DHCPD_INTERFACE=/DHCPD_INTERFACE="eth2"/g' dhcpd

But this doesn't replace the whole line.

This DHCPD_INTERFACE="eth4" should be updated to DHCPD_INTERFACE="eth2" but no line containing a # and DHCPD_INTERFACE= should be changed.

How can I do this.. ? Thanks

Answer 1

您可以使用anchor（ ^ ）来匹配行的开头：

sed 's/^\(DHCPD_INTERFACE="eth\)4"/\12"/' file

Answer 2

You can first replace the lines with a '#', then replace the lines with "eth4", and finally get back the ones with the '#'.

    sed -i.bak s/#DHCPD_INTERFACE=/foo/g dhcpd
    sed -i s/DHCPD_INTERFACE="eth4"/DHCPD_INTERFACE="eth2"/g dhcpd
    sed -i s/foo/#DHCPD_INTERFACE=/g dhcpd

Not the best solution, but still a working one. Hope I helped !

Answer 3

sed solution:

The exemplary test file contents:

DHCPD_INTERFACE="eth4"
#DHCPD_INTERFACE="eth4"

---

sed 's/\([^#]\|^\)DHCPD_INTERFACE="eth4"/\1DHCPD_INTERFACE="eth2"/g' test

The output:

DHCPD_INTERFACE="eth2"
#DHCPD_INTERFACE="eth4"