Remove a comment # from a file

[Andrew_80]

Any one know How to remove a "#" from a file without using vi (using script).
I think I am half way the problem as I believe I should use ch_rc command, but I am not sure of the syntax.

Thanks

[Pete Randall]

Well, there's always sed:

sed 's|#AnyOtherTextNecessaryToIdentifyTheLine||g'`



Pete

[Sergejs Svitnevs]

sed 's/#//g' file > new_file

Regards,
Sergejs

[Andrew_80]

This is OK, But I need to edit a file like
"/etc/rc.config.d/savecrash" and remove the comments from three lines and the sed allow you to only create it on anew file. So, I am trying to save multiple steps.

Thanks

[Bruno Ganino]

...more
sed 's/#//n' infile > outfile
(n=1 to 512). Substitute for just the n'th occurrence.
Bruno

[Michael Steele_2]

grep -v \# file1 > file2

[Bruno Ganino]

see also command ed

[Paddy_1]

Use this perl script to display the file and redirect it as neccesary
-----------------------------

open I, $ARGV[0] or die "cannot open $ARGV[0]: $!";

while ($line=)
{
next if $line=~ /^#/;
print $line;
}
------------------------

[Andrew_80]

Remember I just want to remove the '#' from some specific lines not the whole file savecrash.

# SAVECRASH=1
# SAVECRASH_DIR=/var/adm/crash
# COMPRESS=2

I know how to change the values using

#ch_rc -a -p

and on the man pages it says you can remove characters also, but I am not sure How ?
I appreciate your responses.

Thanks

[Mark Grant · Accepted solution]

mv savecrash savecrash.prev
cat savecrash.prev | sed s/^#//g > savecrash

[Andrew_80]

But that will remove it from all the lines including text !!

[harry d brown jr]

You need to use "vi", "emacs" or "ed". Unless you are a pattern matching expert and know how to write code in perl, your request for another "answer", other than editors, is highly unlikely, especially since you want to have an answer that magically knows which "#" lines to remove or possibly "un-comment" lines and which not to.

If you don't like "vi", then copy the progam to your desktop and use "wordpad".

live free or die
harry

[Robert-Jan Goossens]

Why do'nt you want to use vi ??

[Steven Sim Kok Leong]

Hi,

I tend to agree with Harry.

In addition, if you just want to get rid of consecutive lines at the beginning or at the end of the file, you can use head and/or tail. For instance, you want the 4th line onwards from the file, then:

# tail +4 myfile > myfile2

Hope this helps. Regards.

Steven Sim Kok Leong

[Mark Grant]

My last post only removes the "#" not the text. It does do it for all lines though. There is no way that you can be sure you are uncommenting the right thing unless you give the script something to match against. Now that is going to start to be annoying when uncommenting large files.

If it is something you find you are doing a lot of just make a copy of the file with the comments in and another with the comments out and just cp the one you want when you want.

[Andrew_80]

Thanks a lot, The reason I don't want to use vi is that I have to do the same thing across 20 machines !

I will use Mark's solution
cat savecrash.prev | sed s/^#//g > savecrash
and specify each line.

Thanks again

[Jean-Luc Oudart]

If you know which line pattern you want to replace say # SAVECRASH
then you can use vi
this way
create "mycom" a command file :
:1,$s/^# SAVECRASH/SAVECRASH/
:wq

then run
vi < mycom >/dev/null

Rgds,
Jean-Luc

[Bruno Ganino]

Well, try
cat savecrash | sed s/# SAVECRASH//g > savecrash.tmp
cat savecrash.tmp | sed s/# COMPRESS//g > savecrash

P.S. option g=globale or n=occorrence
HTH
Bruno

[Bruno Ganino]

Before not right, This is OK! Sorry
Well, try
cat savecrash | sed s/# SAVECRASH/SAVECRASH/g > savecrash.tmp
cat savecrash.tmp | sed s/# COMPRESS/COMPRESS/g > savecrash
P.S. option g=globale or n=occorrence
HTH
Bruno