CODEWITHSUNDEEP
perlbash
user349418
user349418

Reputation: 2217

Text manipulation of a file

I'm rather noob on bash (or perl). Looking a way to modify some lines in a file, (basically removing the first 2 characters of each line, and add a "," at the end of each line)

###element
##.element
###test-element-00

becomes the following,

#element,
.element,
#test-element-00,

Thanks

Upvotes: 0

Views: 330

Answers (9)

Idelic
Idelic

Reputation: 15572

Bash can handle this just fine: line="##element"; echo "${line#??},"

Upvotes: 0

jon
jon

Reputation: 1

Here's yet another to way to do it using the ed command:

str='
###element
##.element
###test-element-00
'

# for in-place file editing use "ed -s file" and replace ",p" with "w"
# cf. http://wiki.bash-hackers.org/howto/edit-ed
cat <<-'EOF' | sed -e 's/^ *//' -e 's/ *$//' | ed -s <(echo "$str")
   H
   ,g/.\{2\}/s/^..\(.*\)$/\1,/
   ,p
   q
EOF

Upvotes: 0

ikegami
ikegami

Reputation: 385506

perl -ple's/..(.*)/$1,/' infile > outfile

perl -ple's/..//;s/$/,/' infile > outfile

Using in-place editing:

perl -i -ple's/..(.*)/$1,/' file

perl -i -ple's/..//;s/$/,/' file

Upvotes: 5

Dennis Williamson
Dennis Williamson

Reputation: 359855

OK, just for variety:

yes , | head -n $(wc -l < filename) | paste -d '' <(cut -c3- filename) -

or

yes , | head -n $(wc -l < filename) | paste -d '' <(grep -Po '..\K.*' filename) -

Requires a shell, such as Bash, that supports process substitution. For the grep version, support for Perl Compatible Regular Expressions is required.

Upvotes: 0

Shalini
Shalini

Reputation: 455

There are several ways of doing this, one way of doing this in perl would be :

$line =  '##element'
$line = substr ( $line, 2 ) . ',';

Upvotes: 0

kurumi
kurumi

Reputation: 25599

awk

$> awk '/^#/ { print substr($0,3)"," }' file
#element,
.element,
#test-element-00,

Ruby(1.9+)

$> ruby -ne 'print "#{$_.chomp[2..-1]},\n" if /^#/' file
#element,
.element,
#test-element-00,

Or just pure bash

$> while read -r line; do echo "${line:2},"; done < file
#element,
.element,
#test-element-00,

or sed if you prefer

$> sed 's/^..//;s/$/,/' file
#element,
.element,
#test-element-00,

Upvotes: 0

erickb
erickb

Reputation: 6309

Try something like this:

my @old_lines = qw(###element ##.element ###test-element-00);
my @new_lines;
for my $line (@old_lines) {
   $line =~ s/^##//;
   $line .= ',';
   push @new_lines, $line;
}

print "@new_lines";

Upvotes: 0

Finbar Crago
Finbar Crago

Reputation: 444

On bash the command you are looking for is sed, try:

cat filename | sed 's/^..\(.*\)$/\1,/'

Upvotes: 2

mu is too short
mu is too short

Reputation: 434585

A bit of simple sed(1) is all you need for that:

sed 's/^..//' < your_file | sed 's/$/,/'

Or you can use Perl if you really want to:

perl -pe 's/^..(.*)/$1,/' < your_file

Or you can even do it with cut(1), xargs(1), and printf(1):

cut -c2- < your_file | xargs printf '%s,\n'

We could probably be here all night coming with clever combinations of shell tools to perform the desired transformation, I'd probably just go with the perl one and move on to more productive things.

The sed approach can probably be done better but my sed is rusty.

Upvotes: 1

Related Questions