remove all lines lead with # signs (alone or with spaces/tabs before) 2

[w5000]

what would be universal solution to remove all lines like:

#sfsfs
# sfsfsf
# fsfsf
<TAB>#

 

[w5000]

could anyone pleae check and solution I created is ok?

$ echo "#\n\t #sfsfs\n#ssfsfsf       \n #sfdsfs\n#\n# sfdsdf\n     #dfdfds\n sfsfdfd sdfsfsf"
#
         #sfsfs
#ssfsfsf
 #sfdsfs
#
# sfdsdf
     #dfdfds
 sfsfdfd sdfsfsf
$ echo "#\n\t #sfsfs\n#ssfsfsf       \n #sfdsfs\n#\n# sfdsdf\n     #dfdfds\n sfsfdfd sdfsfsf"|awk '$1!~/^#/'
 sfsfdfd sdfsfsf
$

 

[feherke]

Hi

Yes, it is Ok.

Personally I usually avoid starting a script interpreter for such simple tasks. Generally [tt]grep[/tt] should be faster on such tasks. However this depends on the Awk/grep implementation too :

[blue]master #[/blue] time gawk '$1!~/^#/' w5000.txt > /dev/null
time 0m3.989s

[blue]master #[/blue] time mawk '$1!~/^#/' w5000.txt > /dev/null
time 0m2.427s

[blue]master #[/blue] time grep -v '^\s*#' w5000.txt > /dev/null
time 0m0.965s

Tested using [tt]gawk[/tt], [tt]mawk[/tt] and GNU [tt]grep[/tt] on your sample input*1?000?000.


Feherke.

http://free.rootshell.be/~feherke/

 

[PHV]

A portable sed way:

sed '/^[ <Tab>]*#/d'

Hope This Helps, PH.
FAQ219-2884
FAQ181-2886