zero-padding and sorting - Linux Shell Scripting

Question

I am trying to do some text file manipulation in Linux.

I have a file, called names.txt that looks like this:

A1
X12
B4
Y5
C10
Z23
B8
C3
Z6

And I need it to look like this:

A01
B04
B08
C03
C10
X12
Y05
Z06
Z23

GOAL: I need to zero-pad the single digits and then alphabetize the results, and save to file sorted_names.txt

I'm thinking I need to count the number of characters per line first, and if the number of characters is less than 2, then add a zero. Lastly I would need to sort alphabetically.

For starters, I think I do this to count the number of characters per line:

cat names.txt | while read line
do

  count=$(echo $line | wc -c)
  echo $line $count

done

Then my thought was to loop through count:

for COUNT in $count
if [( $COUNT = "3" )];
then
    echo doZeroPadHere
fi

ruakh · Accepted Answer · 2012-11-23 02:18:42Z

2

Is it important to you to do it using only built-in Bash features? Because it seems easier to use sed and sort:

<names.txt sed 's/^\([A-Z]\)\([0-9]\)$/\10\2/' | sort >sorted_names.txt

answered Nov 23, 2012 at 2:18

ruakh

185k29 gold badges292 silver badges324 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

Sheila Over a year ago

Hi @ruak. Thanks for your help. Would you be willing to explain your code here so I can understand what you've done?

Klaas van Gend Over a year ago

<names.txt tells the shell to read the file and give the output to sed. Then we get a regex that will replace any line that fits the requirement and leaves all other lines intact. The regex uses ^ to indicate start of a line and $ to indicate end of a line. Essentially, it looks for lines with a single uppercase character $[A-Z]$ and a single digit $[0-9]$, which then are repeated \1 0 \2 (without the spaces). Then using sort to sort the output. voila.

Brian Campbell · Accepted Answer · 2012-11-23 03:36:27Z

1

Here's a solution using only Bash and sort:

while read line
do 
    printf "%s%02d\n" ${line:0:1} ${line:1}
done <names.txt | sort >sorted.txt

This reads lines from names.txt, and splits each one up into its first character (${line:0:1}) and the rest of the line after the first character (${line:1}). It uses printf (more details) to print the first character verbatim, and the rest of the line as a 0-padded number. It redirects its input from names.txt (avoiding a useless use of cat), pipes the output to sort, and redirects that into sorted.txt.

edited Nov 23, 2012 at 3:36

answered Nov 23, 2012 at 3:18

Brian Campbell

335k59 gold badges367 silver badges344 bronze badges

4 Comments

Sheila Over a year ago

Hi @Brian. When I try this, I get an error ":invalid number4" but the file sorted.txt is still generated correctly. Do you know what the error is and what it means?

Brian Campbell Over a year ago

@Sheila It happens because one of the lines contains something other than a number after the first character. ${line:1} means "all of the characters on the line after the first". The %02d part of the printf format string says "print out as a decimal integer 2 characters wide padded with 0". If the rest of the line isn't a number (for instance, your line says A0Z or D20x or something), you will get this error. If you want to find out which lines are the problem, add before the printf: [[ ${line:1} =~ [^0-9] ]] && echo $line >&2; this will print out the lines that cause problems.

Sheila Over a year ago

Thanks for your comments. The line with the problem was "P24". This doesn't seem to have any different characteristics than the others but it is not included in the sorted.txt file. Any thoughts?

Brian Campbell Over a year ago

@Sheila I'm not sure; could be a stray invisible character? Maybe a space after the end of the line? Can you post the exact file somewhere?

Steve · Accepted Answer · 2012-11-23 02:29:40Z

0

Here's one way using awk and sort:

awk '{ printf "%s%02d\n", substr($0,0,1), substr($0,2) | "sort" }' file

Results:

A01
B04
B08
C03
C10
X12
Y05
Z06
Z23

answered Nov 23, 2012 at 2:29

Steve

55.1k13 gold badges94 silver badges105 bronze badges

Comments

Andy Lester · Accepted Answer · 2012-11-23 03:34:37Z

0

Here's a Perl way to do it:

perl -lpe's/^([A-Z])(\d)$/${1}0$2/' names.txt

For each line, if it matches exactly one letter and one digit, change it to the letter, a zero, and the digit. Then print.

answered Nov 23, 2012 at 3:34

Andy Lester

94.2k16 gold badges106 silver badges162 bronze badges

Collectives™ on Stack Overflow

zero-padding and sorting - Linux Shell Scripting

4 Answers 4

2 Comments

4 Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

2 Comments

4 Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related