SORT from start char pos to end char pos within record 1

[Calator]

Hi,

I need to sort a file from a start character position to an end character position within record, eg:

sort using a key specified as character positions 5-12 within the record

Can this be achieved with the Unix sort command? The entire record should be regarded as one field irrespective of embedded blanks (ie sort should not attempt to split my record into fields based on existing blanks!). I'm afraid that:

sort -k 1.5,1.12 infile

may not work ok

anyone already faced this issue and able to share a tested solution? Thanks

 

[Salem]

`-t SEPARATOR'
`--field-separator=SEPARATOR'
     Use character SEPARATOR as the field separator when finding the
     sort keys in each line.

Pick a character which isn't in your file, and use that as a separator. sort should then view each line as a single record.

 

[marsd]

You can always use gawk for a start.

function selectrangeSort() {
awk -v v1=$1 -v v2=$2 '{printf "%s\n", substr($0,v1,v2)}' $3 
}

More extravagantly:
function selectrangeSort() {
awk -v v1=$1 -v v2=$2 ' {array[i++] = substr($0,v1,v2)}
END { asort(array) ; for (m=1 ; m <= i ; m++) {print array[m]}}' $3
}

 

[LKBrwnDBA]

Try this:

sort +0.5 infile




----------------------------------------------------------------------------
The person who says it can't be done should not interrupt the person doing it. -- Chinese proverb