How to extract a part of string?

Question

I have string contains a path

string="toto.titi.1.tata.2.abc.def"

I want to extract the substring which is situated after toto.titi.1.tata.2.. but 1 and 2 here are examples and could be other numbers.

In general: I want to extract the substring which situated after toto.titi.[i].tata.[j]..

[i] and [j] are a numbers

How to do it?

score 6 · Answer 1 · answered Jul 23 '13 at 13:55

6

Pure bash solution:

[[ $string =~ toto\.titi\.[0-9]+\.tata\.[0-9]+\.(.*$) ]] && result="${BASH_REMATCH[1]}"
echo "$result"

answered Jul 23 '13 at 13:55

dogbane

266,786
75
396
414

nice! I've searched for that before suggesting a `sed` solution – hek2mgl Jul 23 '13 at 13:56
Can also say `\d` instead of `[0-9]`. – devnull Jul 23 '13 at 14:03
No you can't. `bash` doesn't support `\d`. – dogbane Jul 23 '13 at 14:05

chepner · Answer 2 · 2013-07-26T13:55:05.750

5

An alternate bash solution that uses parameter expansion instead of a regular expression:

echo "${string#toto.titi.[0-9].tata.[0-9].}"

If the numbers can be multi-digit values (i.e., greater than 9), you would need to use an extended pattern:

shopt -s extglob
echo "${string#toto.titi.+([0-9]).tata.+([0-9]).}"

edited Jul 26 '13 at 13:55

answered Jul 23 '13 at 14:03

chepner

497,756
71
530
681

a good answer (+1). but you can avoid the use of `shopt -s extglob` by using `echo "${string#toto.titi.[0-9]*.tata.[0-9]*.}"` – MOHAMED Jul 26 '13 at 13:53
This is a pattern, not a regular expression. `[0-9]*` would match a single digit followed by *anything*, not a (possibly empty) series of digits. – chepner Jul 26 '13 at 13:54
For my case I m sure that my string contains only numbers. so for my case it's ok to use `[0-9]*` – MOHAMED Jul 26 '13 at 13:57
In that case, you wouldn't even need to specify the numbers; `${string#toto.titi.*.tata.*.}` would probably work (since the `#` operator will match the shortest string possible). – chepner Jul 26 '13 at 14:07

score 1 · Accepted Answer · answered Jul 23 '13 at 13:52

1

You can use cut

 echo $string | cut -f6- -d'.'

answered Jul 23 '13 at 13:52

jh314

27,144
16
62
82

score 0 · Answer 4 · answered Jul 23 '13 at 13:53

0

This does it:

echo ${string} | sed -re 's/^toto\.titi\.[[:digit:]]+\.tata\.[[:digit:]]+\.//'

answered Jul 23 '13 at 13:53

thb

13,796
3
40
68

score 0 · Answer 5 · answered Jul 23 '13 at 13:53

0

May be like this:

echo "$string" | cut -d '.' -f 6-

answered Jul 23 '13 at 13:53

phoxis

60,131
14
81
117

score 0 · Answer 6 · answered Jul 23 '13 at 13:54

0

You can use sed. Like this:

string="toto.titi.1.tata.2.abc.def"
string=$(sed 's/toto\.titi\.[0-9]\.tata\.[0-9]\.//' <<< "$string")
echo "$string"

Output:

abc.def

answered Jul 23 '13 at 13:54

hek2mgl

152,036
28
249
266

@dogbane I expect that under the hood it is almost the same. But however, I have updated the post – hek2mgl Jul 23 '13 at 13:59

score 0 · Answer 7 · answered Jul 23 '13 at 13:56

0

try this awk line:

awk -F'toto\\.titi\\.[0-9]+\\.tata\\.[0-9]+\\.' '{print $2}' file

with your example:

kent$  echo "toto.titi.1.tata.2.abc.def"|awk -F'toto\\.titi\\.[0-9]+\\.tata\\.[0-9]+\\.' '{print $2}'
abc.def

answered Jul 23 '13 at 13:56

Kent

189,393
32
233
301

How to extract a part of string?

7 Answers7