I have a file that looks like this:

1,var1
2,var2
3,var3
4,var1_val1
5,var2_val2
6,var1_val2
7,var3_val1
8,var2_val1
9,var3_val2

Output file should look like:

var1 1 4 6 
var2 2 8 5
var3 3 7 9

My code is quite complicated. It works, but it's very inefficient. Can this be done more efficiently:

def findv(var):
    with open(inputfile) as f:
        for line in f:
            elems=line.split(',')
            name=elems[0]
            if var!=name:
                continue
            field=elems[0]
        f.seek(0)
        for line in f:
            elems2=line.split(',')
            if elems2[1].endswith(var+'_val1'):
                first=elems2[0]
        f.seek(0)
        for line in f:
            elems3=line.split(',')
            if elems3[1].endswith(var+'_val3'):
                second=elems3[0]
    return var,field,first,second

main part of the code:

with open(inputfile) as f:
    with open(outputfile) as fout:
        for line in f:
            tmp=line.split(',')
        if current[1].endswith('val1') or current[1].endswith('val2'):
            continue
        v=tmp[1]
        result=findv(v)
        f2.write(result)

My function findv(var) is called each time a line in input file starts with varx and then searches through the file multiple times until it finds fields that correspond to varx_val1 and varx_val2.

I have a file that looks like this:

1,var1
2,var2
3,var3
4,var1_val1
5,var2_val2
6,var1_val2
7,var3_val1
8,var2_val1
9,var3_val2

Output file should look like:

var1 1 4 6 
var2 2 8 5
var3 3 7 9

My code is quite complicated. It works, but it's very inefficient. Can this be done more efficiently:

def findv(var):
    with open(inputfile) as f:
        for line in f:
            elems=line.split(',')
            name=elems[0]
            if var!=name:
                continue
            field=elems[0]
        f.seek(0)
        for line in f:
            elems2=line.split(',')
            if elems2[1].endswith(var+'_val1'):
                first=elems2[0]
        f.seek(0)
        for line in f:
            elems3=line.split(',')
            if elems3[1].endswith(var+'_val3'):
                second=elems3[0]
    return var,field,first,second

main part of the code:

with open(inputfile) as f:
    with open(outputfile) as fout:
        for line in f:
            tmp=line.split(',')
        if current[1].endswith('val1') or current[1].endswith('val2'):
            continue
        v=tmp[1]
        result=findv(v)
        f2.write(result)

My function findv(var) is called each time a line in input file starts with varx and then searches through the file multiple times until it finds fields that correspond to varx_val1 and varx_val2.

EDIT: I need to preserve the order of the input file, so var1 has to appear first in the output file, then var2, then var3 etc.