UNIX for Dummies Questions & Answers

Hi,

I need a script to read the n° character from a text file.
eg: if the text file contains the line "123456" ,I nedd a command to display the number 4, as an example.

I tried with awk and printf but it seems only works with words separated with spaces, but in this case I have only one word "123456" and the characters are not separated.


Thanks in advance for your kindly help.

echo 12345 | sed 's/^.\{3\}\(.\).*/\1/'

Phew,

vgersh99 can you throw some light on that.
It bounced over my head.

Code:

^        - from the beginning of the line
.\{3\}  - any 3 characters
\(.\)    - capture one character [the 4-th chars from the beginning]
.*       - any number of characters

\1       - print the first 'capture'

echo 12345 | sed 's/^.\{3\}\(.\).*/\1/'

Then won't this work.

echo 12345 | sed 's/^\(...\)\(.\).*/\1\2/'

echo 12345 | sed 's/^.\{3\}\(.\).*/\1/'
4

echo 12345 | sed 's/^\(...\)\(.\).*/\1\2/'
1234

I don't know - depends on the the definition of 'works'

sed and awk wont work if the character you've got isn't on the first line of the file.

If you have GNU text utiliities installed you can do it using the versions tail and head in the distribution.

head -c 4 yourfile.txt | tail -c 1

If you wanted the 4th charcter for example. However this counts carriage returns and spaces as characters. If you want to ignore spaces and/or carraige returns you can do it but its a little more complicated.

It would be easy if you have access to perl:

cat youfile.txt | perl -e 'read(STDIN,$data,4); print substr($data,-1);'

again gives your the 4th character.

Sean