Question

I have got a problem i am trying to solve using ONLY awk.

I have a csv file in the structure:

Easting  Northing    Latitude    Longitude   Locality Name

Easting  "Northing"  "Latitude"  "Longitude"     "LocalityName"
364208  176288           51.48441   -2.51685     "Fishponds"
358596  172813           51.45278   -2.59726     "Bristol City Centre"
358886  177828           51.49789   -2.59367     "Southmead"
358839  177839           51.49798   -2.59435     "Southmead"
358980  177882           51.49838   -2.59232     "Southmead"
359009  177863           51.49821   -2.5919          "Southmead"
358839  177529           51.4952        -2.59431     "Southmead"
359475  168262           51.41192   -2.58409     "Hengrove Park"
358945  173526           51.45921   -2.59232     "Bristol"
358943  173525           51.4592    -2.59235     "Bristol"
358941  173524           51.45919   -2.59238     "Bristol"
358940  173523           51.45919   -2.59239     "Bristol"
358945  173528           51.45923   -2.59232     "Bristol"
358936  173520           51.45916   -2.59245     "Bristol"
358936  173521           51.45917   -2.59245     "Bristol"
358932  173516           51.45912   -2.5925          "Bristol"

etc… I am trying to write an awk script which will count each instance of Locality name and the print print this so an out put would be:

Fishponds 1
Bristol City Centre 1
Southmead 5
Hengrove park 1
Bristol 8

So far i have got this:

BEGIN { i = 0; state = 0; names[NR]; FS=","; }

{
#for each element in names array, check if already exists.
    for(j=0;j<=i;j++)
    {
        if(names[j] == $5)
        {
        state = 1;
        break;
        }
    }
# If the name doesnt already exist add to names array
    if(state == 0)
    {
        names[i] = $5;
        i++;
    }
    state = 0;
}

END { 
    for(x=0;x<=i;x++)
    {
    print names[x];
    }
}

Which hopefully sorts the locations and removes the duplicates, but i still cannot think of a good way to count the instances of each location then list them back.

Answer 1

Simpler soulution:

awk -F '"' 'NR>3 {locname[$2]++}
            END { for (n in locname) {print n, locname[n] } }' INPUTFILE

First the input file separator is set to ", so the second field will be the location name. Skipping the first line (header). Utilizing an array (key is the second field) to count the occurrences. After the last line print the array’s keys, and values.