Suppose I have a string "123456789".
I want to extract the 3rd, 6th, and 8th element. I guess I can use
cut -3, -6, -8
But if this gives
368
Suppose I want to separate them by a white space to get
3 6 8
What should I do?
Actually shell parameter expansion lets you do substring slicing directly, so you could just do:
x='123456789'
echo "${x:3:1}" "${x:6:1}" "${x:8:1}"
Update
To do this over an entire file, read the line in a loop:
while read x; do
echo "${x:3:1}" "${x:6:1}" "${x:8:1}"
done < file
(By the way, bash slicing is zero-indexed, so if you want the numbers '3', '6' and '8' you'd really want ${x:2:1} ${x:5:1} and {$x:7:1}.)
You can use the sed tool and issue this command in your teminal:
sed -r "s/^..(.)..(.).(.).*$/\1 \2 \3/"
Explained RegEx: http://regex101.com/r/fH7zW6
To "generalize" this on a file you can pipe it after a cat like so:
cat file.txt|sed -r "s/^..(.)..(.).(.).*$/\1 \2 \3/"
Perl one-liner.
perl -lne '#A = split //; print "$A[2] $A[5] $A[7]"' file
Using cut:
$ cat input
1234567890
2345678901
3456789012
4567890123
5678901234
$ cut -b3,6,8 --output-delimiter=" " input
3 6 8
4 7 9
5 8 0
6 9 1
7 0 2
The -b option selects only the specified bytes. The output delimiter can be specified using --output-delimiter.
Related
I have a string with multiple value outputs that looks like this:
SD performance read=1450kB/s write=872kB/s no error (0 0), ManufactorerID 27 Date 2014/2 CardType 2 Blocksize 512 Erase 0 MaxtransferRate 25000000 RWfactor 2 ReadSpeed 22222222Hz WriteSpeed 22222222Hz MaxReadCurrentVDDmin 3 MaxReadCurrentVDDmax 5 MaxWriteCurrentVDDmin 3 MaxWriteCurrentVDDmax 1
I would like to output only the read value (1450kB/s) using bash and sed.
I tried
sed 's/read=\(.*\)kB/\1/'
but that outputs read=1450kB but I only want the number.
Thanks for any help.
Sample input shortened for demo:
$ echo 'SD performance read=1450kB/s write=872kB/s no error' | sed 's/read=\(.*\)kB/\1/'
SD performance 1450kB/s write=872/s no error
$ echo 'SD performance read=1450kB/s write=872kB/s no error' | sed 's/.*read=\(.*\)kB.*/\1/'
1450kB/s write=872
$ echo 'SD performance read=1450kB/s write=872kB/s no error' | sed 's/.*read=\([0-9]*\)kB.*/\1/'
1450
Since entire line has to be replaced, add .* before and after search pattern
* is greedy, will try to match as much as possible, so in 2nd example it can be seen that it matched even the values of write
Since only numbers after read= is needed, use [0-9] instead of .
Running
sed 's/read=\(.*\)kB/\1/'
will replace read=[digits]kB with [digit]. If you want to replace the whole string, use
sed 's/.*read=\([0-9]*\)kB.*/\1/'
instead.
As Sundeep noticed, sed doesn't support non-greedy pattern, updated for [0-9]* instead
I have several file names, for ease I've put them in a file as follows:
01.action1.txt
04action2.txt
12.action6.txt
2.action3.txt
020.action9.txt
10action4.txt
15action7.txt
021action10.txt
11.action5.txt
18.action8.txt
As you can see the formats aren't consistent what I'm trying to do is extract the first numbers from these file names 1,4,12,2,20 etc
I have the following regex
(\.)?action\d{1,}.txt
Which is successfully matching .action[number].txt but I need to also match the leading 0 and apply it to my substitute with blank in sed so i'm only left with the leading numbers. I'm having trouble matching the leading 0 and applying the whole thing to sed.
Thanks
With GNU sed:
sed -r 's/0*([0-9]*).*/\1/' file
Output:
1
4
12
2
20
10
15
21
11
18
See: The Stack Overflow Regular Expressions FAQ
I don't know if the below awk is helpful but it works as well:
awk '{print $1 + 0}' file
1
4
12
2
20
10
15
21
11
18
Trying to replace a string of fixed length at certain position (a string of arbitrary numbers) with a specified string.
I have to :
for every line beginning with 1, in the 4-13 columns, replace existing value with 123456789 where column 4 is a space. 123456789
so a sample file looks like this in the first line:
110 000000000000000000000000000000000000000
and i want
110 123456789000000000000000000000000000000
So far I have:
sed -i "/^1/ s/(.{10})/ 123456789/4" $DEST/$FILE_NAME$DATE.txt
This doesn't do anything though...
With sed:
sed '/^1/s/\(.\{4\}\)\(.\{9\}\)/\1123456789/' "$DEST/$FILE_NAME$DATE.txt"
The preceding regex /^1/ makes the following substitute command apply only to lines starting with a 1.
The substitute command itself captures the first 4 chars 100<space> and the following 9 chars 000000000 into separate groups while keeping the first 4 chars and replacing the following nine chars by 123456789.
Btw, if you have GNU sed, you might simplify the command to:
sed -r '/^1/s/(.{4})(.{9})/\1123456789/'
... which looks simpler for understanding, but is not portable across all different sed versions.
Using awk, simple to understand solution
awk '/^1/ {print substr($0,1,4)"123456789"substr($0,14)}' file
110 123456789000000000000000000000000000000
If line starts with 1, print the 4 first characters + 123456789 + the rest of the line starting from 14 position.
My dummy text file (one continuous line) looks like this:
AAChvhkfiAFAjjfkqAPPMB
I want to:
Delete part of the text (specific range);
Copy-Paste (specific range of characters) within the file.
How I am doing this:
To cut part of the text at wanted positions (from 5 to 7 characters & from 10 to 14 characters) I use cut
echo 'AAChvhkfiAFAjjfkqAPPMB' | cut --complement -c 5-7,10-14
AAChfifkqAPPMB
But I really don't know how to copy-paste text. For example: to copy text from 15 to 18 characters and paste it after character 1 (also using previous cut command). To get the final result like this:
fkqAAAChfifkqAPPMB
So I do have to questions:
How to read text (from .. to) given range using perl, awk or sed & paste this text at specific position.
How to combine this text pasting with the previous cut command as after cutting text will move to the left side, hence wrong text will be copied.
Maybe something like this:
$ echo AAChvhkfiAFAjjfkqAPPMB | awk '{ print(substr($1, 0, 14) substr($1, 18) substr($1, 15, 3)) }'
AAChvhkfiAFAjjAPPMBfkq
In Perl I think substr would be a good candidate, try eg.
$a = '1234567890';
#from pos 2, replace 3 chars with nothing, return the 3 chars
$b=substr($a,2,3,'');
print "$a\t$b\n"; #1267890 345
#in posistion 0 (first), replace 0 characters (ie pure insert)
#with the content of $b
substr($a,0,0,$b);
print "$a\t$b\n"; #3451267890 345
See http://perldoc.perl.org/functions/substr.html for more details.
splice() may be a candidate as well.
In perl, you can use array slice, by splitting the string in a array :
my $string = "AAChvhkfiAFAjjfkqAPPMB1";
my #arr = split //, $string;
and slicing (print element 5 to 7 and 10 to 14):
print #array[5..7,10..14];
you can use splice() too to re-arrange the array.
perldoc said :
Removes the elements designated by OFFSET and LENGTH from an array, and replaces them with the elements of LIST, if any.
See http://perldoc.perl.org/perldata.html#Slices
quite straightforward with awk:
kent$ echo "AAChvhkfiAFAjjfkqAPPMB"|awk '
{for(i=5;i<=7;i++)$i="";
for(i=10;i<=14;i++)$i="";
for(i=15;i<=18;i++)t=sprintf("%s%s",t,$i);
$0=t""$0}1' OFS="" FS=""
fkqAAAChfifkqAPPMB
edit
to reverse the part of text, you just need to swap t and $i:
kent$ echo "AAChvhkfiAFAjjfkqAPPMB"|awk '
{for(i=5;i<=7;i++)$i="";
for(i=10;i<=14;i++)$i="";
for(i=15;i<=18;i++)t=sprintf("%s%s",$i,t);
$0=t""$0}1' OFS="" FS=""
AqkfAAChfifkqAPPMB
i have contents in a file
like
asdfb ... 1
adfsdf ... 2
sdfdf .. 3
I want to write a unix command that should be able to add 1 + 2 + 3 and give the result as 6
From what I am aware grep and awk would be handy, any pointers would help.
I believe the following is what you're looking for. It will sum up the last field in each record for the data that is read from stdin.
awk '{ sum += $NF } END { print sum }' < file.txt
Some things to note:
With awk you don't need to declare variables, they are willed into existence by assigning values to them.
The variable NF is the number of fields in the current record. By prepending it with a $ we are treating its value as a variable. At least this is how it appears to work anyway :)
The END { } block is only once all records have been processed by the other blocks.
An awk script is all you need for that, since it has grep facilities built in as part of the language.
Let's say your actual file consists of:
asdfb zz 1
adfsdf yyy 2
sdfdf xx 3
and you want to sum the third column. You can use:
echo 'asdfb zz 1
adfsdf yyy 2
sdfdf xx 3' | awk '
BEGIN {s=0;}
{s = s + $3;}
END {print s;}'
The BEGIN clause is run before processing any lines, the END clause after processing all lines.
The other clause happens for every line but you can add more clauses to change the behavior based on all sorts of things (grep-py things).
This might not exactly be what you're looking for, but I wrote a quick Ruby script to accomplish your goal:
#!/usr/bin/env ruby
total = 0
while gets
total += $1.to_i if $_ =~ /([0-9]+)$/
end
puts total
Here's one in Perl.
$ cat foo.txt
asdfb ... 1
adfsdf ... 2
sdfdf .. 3
$ perl -a -n -E '$total += $F[2]; END { say $total }' foo
6
Golfed version:
perl -anE'END{say$n}$n+=$F[2]' foo
6