Expand range of numbers in file - bash

Question

I have a file with delimited integers which I've extracted from elsewhere and dumped into a file. Some lines contain a range, as per the below:
Files 1,2,3,4,5,6,7,8,9,10 are OK
Users 1,2,3-9,10 have problems
Cars 1-5,5-10 are in the depot
Trains 1-10 are on time
Any way to expand the ranges on the text file so that it returns each individual number, with the , delimiter preserved? The text either side of the integers could be anything, and I need it preserved.
Files 1,2,3,4,5,6,7,8,9,10 are OK
Uses 1,2,3,4,5,6,7,8,9,10 have problems
Cars 1,2,3,4,5,6,7,8,9,10 are in the depot
Trains 1,2,3,4,5,6,7,8,9,10 are on time
I guess this can be done relatively easily with awk, let alone any other scripting language. Any help very much appreciated

Answer 1

You haven't tagged with perl but I'd recommend it in this case:
perl -pe 's/(\d+)-(\d+)/join(",", $1..$2)/ge' file
This substitutes all occurrences of one or more digits, followed by a hyphen, followed by one or more digits. It uses the numbers it has captured to create a list from the first number to the second and joins the list on a comma.
The e modifier is needed here so that an expression can be evaluated in the replacement part of the substitution.
To avoid repeated values and to sort the list, things get a little more complicated. At this point, I'd recommend using a script, rather than a one-liner:
use strict;
use warnings;
use List::MoreUtils qw(uniq);
while (<>) {
s/(\d+)-(\d+)/join(",", $1..$2)/ge;
if (/(.*\s)((\d+,)+\d+)(.*)/) {
my #list = sort { $a <=> $b } uniq split(",", $2);
$_ = $1 . join(",", #list) . $4 . "\n";
}
} continue {
print;
}
After expanding the ranges (like in the one-liner), I've re-parsed the line to extract the list of values. I've used uniq from List::MoreUtils (a core module) to remove any duplicates and sorted the values.
Call the script like perl script.pl file.

Answer 2

A solution using awk:
{
result = "";
count = split($0, fields, /[ ,-]+/, seps);
for (i = 1; i <= count; i++) {
if (fields[i] ~ /[0-9]+/) {
if (seps[i] == ",") {
numbers[fields[i]] = fields[i];
} else if (seps[i] == "-") {
for (j = fields[i] + 1; j <= fields[i+1]; j++) {
numbers[j] = j;
}
} else if (seps[i] == " ") {
numbers[fields[i]] = fields[i];
c = asort(numbers);
for (r = 1; r < c; r++) {
result = result numbers[r] ",";
}
result = result numbers[c] " ";
}
} else {
result = result fields[i] seps[i];
}
}
print result;
}

Answer 3

$ cat tst.awk
match($0,/[0-9,-]+/) {
split(substr($0,RSTART,RLENGTH),numsIn,/,/)
numsOut = ""
delete seen
for (i=1;i in numsIn;i++) {
n = split(numsIn[i],range,/-/)
for (j=range[1]; j<=range[n]; j++) {
if ( !seen[j]++ ) {
numsOut = (numsOut=="" ? "" : numsOut ",") j
}
}
}
print substr($0,1,RSTART-1) numsOut substr($0,RSTART+RLENGTH)
}
$ awk -f tst.awk file
Files 1,2,3,4,5,6,7,8,9,10 are OK
Users 1,2,3,4,5,6,7,8,9,10 have problems
Cars 1,2,3,4,5,6,7,8,9,10 are in the depot
Trains 1,2,3,4,5,6,7,8,9,10 are on time

Answer 4

another awk
$ awk '{while(match($0, /[0-9]+-[0-9]+/))
{k=substr($0, RSTART, RLENGTH);
split(k,a,"-");
f=a[1];
for(j=a[1]+1; j<=a[2]; j++) f=f","j;
sub(k,f)}}1' file
Files 1,2,3,4,5,6,7,8,9,10 are OK
Users 1,2,3,4,5,6,7,8,9,10 have problems
Cars 1,2,3,4,5,5,6,7,8,9,10 are in the depot
Trains 1,2,3,4,5,6,7,8,9,10 are on time
note that the Cars 1-5,5-10 will end up two 5 values when expanded due to overlapping ranges.