Question

i need to “parse” a cvs file with sth like 400 000 lines.
The cvs looks like :

ADDRESS ZIPCODE TOWN      NAME       DESCRIPTION

I need to separate the 1st colum like this :

ADDRESS     ZIPCODE     TOWN      NAME       DESCRIPTION

it looks easy : add tabulation (before and after) ZIPCODE everytime i see 5 digits (all address are french Zip code).

But i don’t know how ?
I try by a sed commande line but i can’t find the good syntax. or maybe a VBA should be more suitable.

Answer 1

sed -r 's/( [0-9]{5} )/\t\1\t/' yourFile.csv

Note

• column command is good solution to your problem, if your csv is really well formatted. e.g. there is no “,” in your description text.

• the sed line above will check each line of your file, find the first occurrence of 5 numbers/digits and wrap it with TAB. I assume that the zipcodes in your file would be [space]number{5}[space]

• the sed line would fail if there was same format in your “ADDRESS” column.