CODEWITHSUNDEEP
textawkgrepstring-matching
Antoine
Antoine

Reputation: 637

save multiple matches in a list (grep or awk)

I have a file that looks something like this:

# a mess of text
Hello. Student Joe Deere has
id number 1. Over.
# some more messy text
Hello. Student Steve Michael Smith has
id number 2. Over.
# etc.

I want to record the pairs (Joe Deere, 1), (Steve Michael Smith, 2), etc. into a list (or two separate lists with the same order). Namely, I will need to loop over those pairs and do something with the names and ids.

(names and ids are on distinct lines, but come in the order: name1, id1, name2, id2, etc. in the text). I am able to extract the lines of interest with

VAR=$(awk '/Student/,/Over/' filename.txt)

I think I know how to extract the names and ids with grep, but it will give me the result as one big block like

`Joe Deere 1 Steve Michael Smith 2 ...`

(and maybe even with a separator between names and ids). I am not sure at this point how to go forward with this, and in any case it doesn't feel like the right approach.

I am sure that there is a one-liner in awk that will do what I need. The possibilities are infinite and the documentation monumental.

Any suggestion?

Upvotes: 1

Views: 105

Answers (3)

glenn jackman
glenn jackman

Reputation: 246942

grep -oP 'Hello. Student \K.+(?= has)|id number \K\d+' file | paste - -

Upvotes: 0

RavinderSingh13
RavinderSingh13

Reputation: 133545

Could you please try following too.

awk '
/id number/{
  sub(/\./,"",$3)
  print val", "$3
  val=""
  next
}
{
  gsub(/Hello\. Student | has.*/,"")
  val=$0
}
'  Input_file

Upvotes: 1

Ed Morton
Ed Morton

Reputation: 203712

$ cat tst.awk
/^id number/ {
    gsub(/^([^ ]+ ){2}| [^ ]+$/,"",prev)
    printf "(%s, %d)\n", prev, $3
}
{ prev = $0 }

$ awk -f tst.awk file
(Joe Deere, 1)
(Steve Michael Smith, 2)

Upvotes: 2

Related Questions