Extracting n characters from a test line

Question

How can I extract the 6 characters after Z in the following text line and put it into a file using grep, egrep, sed or awk?

B1_Site4_5aT4ZNHN691AQSB6B65_KYEC_SLT_2013-11-24-00-30_935985e7_100m_PASS1.tar

Stéphane Chazelas · Accepted Answer · 2014-01-10 21:49:02Z

file="B1_Site4_5aT4ZNHN691AQSB6B65_KYEC_SLT_2013-11-24-00-30_935985e7_100m_PASS1.tar"

with GNU grep when built with PCRE support (and with zsh or recent versions of ksh93 or bash for <<<):

grep -oP '(?<=Z).{6}' <<< "$file" > file

with ksh93, bash or recent versions of zsh:

tmp=${file#*Z}              # remove chars up to the first Z
echo "${tmp:0:6}" > file

Just for fun, awk

awk -F Z '{print substr($2, 1, 6)}' <<< "$file"

Stéphane Chazelas · Accepted Answer · 2014-01-11 08:22:13Z

expr B1_Site4_5aT4ZNHN691AQSB6B65_KYEC_SLT_2013-11-24-00-30_935985e7_100m_PASS1.tar : '.*Z\(.\{6\}\)' > file

Or just with the shell

string=B1_Site4_5aT4ZNHN691AQSB6B65_KYEC_SLT_2013-11-24-00-30_935985e7_100m_PASS1.tar
printf '%.6s\n' "${string#*Z}" > file

(the first one will consider the last Z followed by 6 characters, the second one the first; behaviour will vary if their not ZXXXXXX in there).

Graeme · Accepted Answer · 2014-01-10 22:04:29Z

2

>>echo \
B1_Site4_5aT4ZNHN691AQSB6B65_KYEC_SLT_2013-11-24-00-30_935985e7_100m_PASS1.tar |
sed 's/.*Z\(.\{6\}\).*/\1/'
>>NHN691

Adding a >file will put it in a file.

answered Jan 10 '14 at 21:40

Graeme

25.9k4 gold badges66 silver badges100 bronze badges

1

As an Aging Dinosaur, I believe sed is the right choice. However, put the '|' at the end of the echo line, and you don't have to backslash the newline to get the pipeline continued. – Bruce Ediger Jan 10 '14 at 22:03
1

@BruceEdiger, sed works on every line of its input, not the whole input, so somehow expr is more appropriate (echo has a few issues associated with it as well for arbitrary data). – Stéphane Chazelas Jan 10 '14 at 22:35

add a comment |

3 Answers 3