Copy part of the content of the previous row into the following

Question

I have a file with the following structure:

GO:0000001      mitochondrion inheritance
GO:0000002      mitochondrial genome maintenance
GO:0000003      reproduction
alt_id: GO:0019952
alt_id: GO:0050876
GO:0000005      obsolete ribosomal chaperone activity
GO:0000006      high-affinity zinc uptake transmembrane transporter activity
GO:0000007      low-affinity zinc ion transmembrane transporter activity
GO:0000008      obsolete thioredoxin
alt_id: GO:0000013
GO:0000009      alpha-1,6-mannosyltransferase activity

Where it says alt_id it means that it is an alternative to the previous GO: code. I'd like to add to each alt_id the definition of the previous GO:, that is, I want an output like this:

GO:0000001      mitochondrion inheritance
GO:0000002      mitochondrial genome maintenance
GO:0000003      reproduction
alt_id: GO:0019952     reproduction
alt_id: GO:0050876     reproduction
GO:0000005      obsolete ribosomal chaperone activity
GO:0000006      high-affinity zinc uptake transmembrane transporter activity
GO:0000007      low-affinity zinc ion transmembrane transporter activity
GO:0000008      obsolete thioredoxin
alt_id: GO:0000013      obsolete thioredoxin
GO:0000009      alpha-1,6-mannosyltransferase activity

How can I copy the content of the previous row in the following? I work with Cygwin in a Windows-based environment.

Answer 1

With awk, not sure if it will work on Cygwin

$ awk '{ if(/^alt_id/){$0 = $0" "p} else{p = ""; for (i=2; i<=NF; i++) p = p" "$i} } 1' ip.txt
GO:0000001      mitochondrion inheritance
GO:0000002      mitochondrial genome maintenance
GO:0000003      reproduction
alt_id: GO:0019952  reproduction
alt_id: GO:0050876  reproduction
GO:0000005      obsolete ribosomal chaperone activity
GO:0000006      high-affinity zinc uptake transmembrane transporter activity
GO:0000007      low-affinity zinc ion transmembrane transporter activity
GO:0000008      obsolete thioredoxin
alt_id: GO:0000013  obsolete thioredoxin
GO:0000009      alpha-1,6-mannosyltransferase activity

• For every line not matching alt_id at start of line, use a variable (p) to save all columns from two onwards
• When line matches alt_id at start of line, append the contents of p variable to input line contained in $0 variable
• The final 1 is short cut to print the contents of $0

Answer 2

The task can be easy done by sed

sed '
    N  #append next line (operate with `line1\nline2`);
    /\nalt_id/s/\([^0-9]*\)\n.*/&\1/
       #if next line starts with `alt_id` the append end of present line
    P  #print present line (all before `\n`)
    D  #remove all before `\n`, starts from begin with remain part (line2)
    ' file

Other way is use hold-space

sed '
    /^alt_id:/G #if line starts by `alt_id:` append hold-space
    s/\n//      #remove `\n`ewline symbol
    t           #if removing success pass further commands (go to end)
    h           #if no (for other lines) copy it to hold-space
    s/\S*//     #remove all non-space symbols from start till first space
    x           #exchange hold-space and pattern-space ==
                #+put resedue into hold-space and return full line
    ' file