Get the substring

Shell Programming and Scripting
1

Hi All,

I have a ouput string likes 'u8wos' or 'u10acsd' or somthing else 'u{number}{any characters}'and I want to get the number behind the letter 'u' by bash shell.

Thanks
Damon

2

If there is only this constellation, ie. not more numbers at other positions and not other characters than alphanumerics, you can just delete all letters for example:

$> echo 'u8wos'| tr -d "[a-z]"
8
3

same behaviour as above

for S in u8wos u10acsd
do
   echo "S='${S//[a-z]/}'"
done

if something different comes before the 'u' and/or after the numbers

for S in u8wos u10acsd
do
   S=${S#*u}
   echo "S='${S%%[a-z]*}'"
done
4 
sed -n 's/u//;s/[0-9]\+/&u/;p'
5 
sed 's/u\([0-9]*\).*/\1/' infile
6

infile:

u123abecadlo
u234abecadl
u345abecad
u01mini

sed 's/u\([0-9]*\).*/\1/' infile

123
234
345
01

my

sed -n 's/u//;s/[0-9]\+/&u/;p' infile

result:

123uabecadlo
234abecadl
345uabecad
01umini
7

Yes, mine is right. Thank you to prove it.

Request is:

get the number behind the letter 'u' by bash shell.

Only numbers.

8

SORRY, YOU ARE WRONG.

other sed script:

sed 's/\(u\)\([0-9]+\)\(.*\)/\2\1\3/' infile

should be literally

sed 's/u\([0-9]+\).*/\1u/' infile
9

Nice.

---------- Post updated at 11:43 AM ---------- Previous update was at 11:41 AM ----------

Please post sample output.

10

I have bash.
I done:

for S in u8wos u10acsd
do
   echo "S='${S//[a-z]/}'"
done

result:

S='8'
S='10'

next I done:

for S in u8wos u10acsd
do
   S=${S#*u}
echo "S='${S%%[a-z]*}'"
done

result:

S='u8uuu'
S='u10uuuu'

Now I do:

for S in u8wos u10acsd
do
   echo "S='${S//[a-z]/}u'"
done

result:

S='8u'
S='10u'
11

John1212, what shell and version are you using on what OS?

$ for S in u8wos u10acsd; do S=${S#*u}; echo "S='${S%%[a-z]*}'"; done
S='8'
S='10'
12

VERY SORRY.
YOU'RE RIGHT.
I had to commit a mistake when typing. Sorry.