delete a line in a file

sparky90ful · 09-27-2013, 12:19 PM

How can i delete a line please?
For example i have this file.txt

Code:

5=e
4=a
3=c
2=b
1=k

and i wat with another script to delete the line where is the 5 but not only in the result i want to delet it forever
thanks

Habitual · 09-27-2013, 12:34 PM

If 5=e is the first line always then,

Code:

sed -i '1d' /home/path/to/files.txt

Else, more info.

sparky90ful · 09-27-2013, 12:42 PM

Quote:

Originally Posted by Habitual

If 5=e is the first line always then,

Code:

sed -i '1d' /home/path/to/files.txt

Else, more info.

no is not the fhirs line
i want to delet the line where is finding the string which i give it no mather cuz is fhirs second third fourt or anywhere else i want to delete the entire line where is finding the given string

Firerat · 09-27-2013, 12:58 PM

http://sed.sourceforge.net/sed1line.txt

don't forget you need the "" with the -i on macOSX

Code:

sed -i "" -e '<see link for this bit>' /path/to/sh.db

Habitual · 09-27-2013, 02:41 PM

Quote:

Originally Posted by Firerat

don't forget you need the "" with the -i on macOSX

Note to self: sparky90ful is a MacOS user.

WarTurkey · 09-27-2013, 03:10 PM

Could do:

Code:

if grep -V '5=e' ./input_file.txt > ./output_file.txt ; then mv ./output_file.txt ./input_file.txt ; fi

EDIT: Nevermind, I am not sure that this would work on OSX :\

Firerat · 09-27-2013, 03:26 PM

seems a lot of work when

Code:

Var="5=e"
sed -i "" -e "/${Var}/d" /path/to/Input

will do it
, yes I know, no real need for the ${Var}
... unless you are writing a script, and 5=e wasn't always 5=e

sparky90ful · 09-28-2013, 03:46 AM

i have did it with

Code:

grep -w "^$key" sh.sh

but now i am stuck in another thing i want to dysplay the value where exist my result of grep for example i have this:

Code:

1=a
2=b
3=c
4=e 
5=f
6=g
7=h
8=i
9=j
10=k
11=l
12=m
13=n
14=o
15=p
16=q
17=r
18=s
19=t
20=u
a=1
b=2
c=3
d=4
e=5
f=6
g=7
h=8
i=9
j=10
k=11

i want to when i grep <1> to print what is after the " = " but i cant do it because my grep print even if he find <1> after the " ="
so i want to search the lines befor the " = " and print what is after " = "
i dont know if you understand but i realy need help

Firerat · 09-28-2013, 04:08 AM

Quote:

Originally Posted by sparky90ful

i have did it with

Code:

grep -w "^$key" sh.sh

the problem with using grep is you need to write a temporary file and then move it to your target file
sometimes it is convenient to do it this way, for instance when working on the command line, trying a few things out.
But for scripts, it is not the correct tool.

grep = global regular expresion print

sed = stream editor

which should you use to edit a file

sparky90ful · 09-28-2013, 04:15 AM

the thing is for example i want to serch 1 so i want cuz he find me the lines where is 1 but before " = " so i dont wnt cuz he find <a=1> or <ddd=321> i wat to find <1=ghfh> or <211=op14> or <421=kj5417> and when he find this print what is after " = " so the result will be " ghf " " op14 " " kj5417 " and that is

i hope now is god explained

Firerat · 09-28-2013, 04:24 AM

Quote:

Originally Posted by sparky90ful

but now i am stuck in another thing i want to dysplay the value where exist my result of grep for example i have this:

Code:

g=7
h=8
i=9
j=10
k=11

i want to when i grep <1> to print what is after the " = " but i cant do it because my grep print even if he find <1> after the " ="
so i want to search the lines befor the " = " and print what is after " = "
i dont know if you understand but i realy need help

I'm going to introduce you to awk

like sed you can have 'awk one liners' , do a web search

awk is very powerful

but here we shall use the basics

Code:

awk -F= '$1 == "g" {print $2}' sh.db

can make it easier to read/understand

Code:

awk -F= '{ if ( $1 == "g" ) {print $2} }' sh.db

-F< field separator >

$1 $2 .... are the fields, don't get them confused with shell script variables

sparky90ful · 09-28-2013, 04:35 AM

Quote:

Originally Posted by Firerat

I'm going to introduce you to awk

like sed you can have 'awk one liners' , do a web search

awk is very powerful

but here we shall use the basics

Code:

awk -F= '$1 == "g" {print $2}' sh.db

can make it easier to read/understand

Code:

awk -F= '{ if ( $1 == "g" ) {print $2} }' sh.db

-F< field separator >

$1 $2 .... are the fields, don't get them confused with shell script variables

i can't understand this.......soo awk=search -F=field separator so i tell him to search a field separator and do wat "g" i replace g with what i search befor field separator??? for my example yo do how with awk?

Code:

the thing is for example i want to serch 1 so i want cuz he find me the lines where is 1 but before " = " so 
i dont wnt cuz he find <a=1> or <ddd=321> 
i wat to find <1=ghfh> or <211=op14> or <421=kj5417> and when he find this print what is after " = " 
so the result will be " ghf " " op14 " " kj5417 " and that is  i hope now is god explained

Firerat · 09-28-2013, 04:51 AM

this Input

Code:

foo=bar
barbar=foo
foobar=barfoo

Code:

awk -F= '$1 == "foo"{print $2}' Input

gives you

bar

long hand
fields are separated by =
if field 1 is equivalent to "foo" then print field 2

I shall expand the example

Code:

awk -v KEY=foo -F= '$1 == KEY {print $2}'

sparky90ful · 09-28-2013, 04:54 AM

i think i get it i test it and i come back with a replay
now i see but the thing is i want to give me not only "bar" but even "barfoo" so foo must be the right word searched or foobar or coolfooten or any who contain tis foo i want what is after =

Firerat · 09-28-2013, 04:55 AM

simpler grep and cut

Code:

KEY=foo
grep "^${KEY}=" sh.db | cut -d= -f2