Getting defined substring with help of sed or egrep

Question

Everyone!! I want to get specific substring from stdout of command.
stdout:

{"response": {"id":"110200dev1","success":"true","token":"09ad7cc7da1db13334281b84f2a8fa54"},"success":"true"}

I need to get a hex string after token without quotation marks, the length of hex string is 32 letters.I suppose it can be done by sed or egrep. I don't want to use awk here. Because the stdout is being changed very often.

possible duplicate of [Parsing json with sed and awk](http://stackoverflow.com/questions/1955505/parsing-json-with-sed-and-awk) — tripleee, Jun 29 '14 at 17:19
@triplee It isn't duplicate because i try to get this substring without Perl expressions — Bilow Yuriy, Jun 29 '14 at 17:37
I do not understand the reason to not to use `awk`, please explain better. — Jotne, Jun 29 '14 at 17:43
@Jotne API is developing, because of it every week i need to change my script 'gawk '{ print substr($0,58,32)}'' — Bilow Yuriy, Jun 29 '14 at 17:56
@BilowYuriy : `awk` is not limited to just using `substr` ;) — Idriss Neumann, Jun 29 '14 at 17:58
@BilowYuriy `awk` supports regex, and `gnu awk` support lots of stuff. — Jotne, Jun 29 '14 at 18:03

hwnd · Answer 1 · 2014-06-29T18:05:46.823

4

grep's nature is extracting things:

grep -Po '"token":"\K[^"]+'

-P option interprets the pattern as a Perl regular expression.
-o option shows only the matching part that matches the pattern.
\K throws away everything that it has matched up to that point.

Or an option using sed...

sed 's/.*"token":"\([^"]*\)".*/\1/'

edited Jun 29 '14 at 18:05

answered Jun 29 '14 at 17:10

hwnd

67,942
4
86
123

I perform my script in MinGW, there aren't Perl patterns there :( – Bilow Yuriy Jun 29 '14 at 17:30
MinGW doesn't have **egrep** :( . – Bilow Yuriy Jun 29 '14 at 17:52
Well you should state your question and what is available more thoroughly next time. – hwnd Jun 29 '14 at 17:53
Ok. i will, thank you for your help, by the way your script with **sed** doesn't get me result, Cyrus's response was acceptable – Bilow Yuriy Jun 29 '14 at 18:03

score 4 · Accepted Answer · answered Jun 29 '14 at 17:22

4

This is an alternate gnu-awk solution when grep -P isn't available:

awk -F: '{gsub(/"/, "")} NF==2&&$1=="token"{print $2}' RS='[{},]' <<< "$string"
09ad7cc7da1db13334281b84f2a8fa54

answered Jun 29 '14 at 17:22

anubhava

713,503
59
514
593

1

Yep.Thank you, it works!! gawk -F: '{gsub(/"/, "")} NF==2&&$1=="token"{print $2}' RS='[{},]' – Bilow Yuriy Jun 29 '14 at 17:40

score 2 · Answer 3 · answered Jun 29 '14 at 17:32

2

With sed:

your-command | sed 's/.*"token":"\([^"]*\)".*/\1/'

answered Jun 29 '14 at 17:32

Cyrus

77,979
13
71
125

score 1 · Answer 4 · answered Jun 30 '14 at 07:12

1

YourStreamOrFile | sed -n 's/.*"token":"\([a-f0-9]\{32\}\)".*/\1/p'

doesn not return a full string if not corresponding

answered Jun 30 '14 at 07:12

NeronLeVelu

9,588
1
22
41

Getting defined substring with help of sed or egrep

4 Answers4