A simpler solution to my original post on regEx

linuxmaveric · 04-02-2008, 05:29 PM

Original quote "Can anyone help me figure out how to write a simple regular expression "using grep" that would search in files for a "city, state and zipcode" in the following order using commas to separate each field:

City, ST, 12345

It should search for any city and 2 capital letters that could represent a state ST and a zip code that can be any 5 numbers form [0-9]. No egrep or ruby just usning simple grep & regEx.

I created a simpler and easier solution for my problem that I originally posted. Using grep & regEx to find a address consisting of "City, State and zip" in a file. City could be anything, state always two capitals "ST", and a 5 digit zip.

Here is my basic solution:

grep ", [A-Z][A-Z] [0-9][0-9][0-9][0-9][0-9]" /filename

This would do the job nicely. But thanks everyone for the suggestions. They gave me the ideas to figure this out.

chrism01 · 04-02-2008, 06:30 PM

You do realise that'll match any 2 alpha 'state' code ?
If you only want ST, you should make that explicit.
Looks a bit like you are accepting negative zip codes: [-0-9] ?

osor · 04-02-2008, 08:08 PM

What’s wrong with:

Code:

grep ', [A-Z]\{2\} [0-9]\{5\}' filename

Note that the following lines will match (perhaps undesirably):

Code:

Washington, DC 2000112345
, AL 36210
ATTENTION, ON 20080403 THE SYSTEM WILL BE DOWN!

(granted, unless you have all valid postal abbreviations off-hand, you will not be able to overcome the problem chrism01 mentions).

linuxmaveric · 04-03-2008, 12:47 AM

"ST" is just an example I used. State could be anything. Thanks for the reply.
RR.

You are correct a typo witjh [-0-9] "fixed" [0-9.

revised version.
grep ", [A-Z][A-Z] [0-9][0-9][0-9][0-9][0-9]" /filename

Thanks for the critique.

BTW. There is nothing wrong with your code. I was just trying to make it very simple in structure. Using basic regex.
RR.

osor · 04-03-2008, 11:18 AM

Quote:

Originally Posted by linuxmaveric

You are correct I should tweak [0-9] to be more specific and only include numbers bewtween 1 thru 9.

I don’t remember anyone saying this. The digit 0 is a valid for zip codes in the US. What chrism01 said is that originally, you had [-0-9] whereas you should have used [0-9]. What I said is that trailing digits are matched, instead of exactly 5.

Quote:

Originally Posted by linuxmaveric

BTW. There is nothing wrong with your code. I was just trying to make it very simple in structure. Using basic regex.

Well, my code was in the form of a BRE (basic regular expression), which is different from an ERE (extended regular expression). Suit yourself.

chrism01 · 04-03-2008, 06:22 PM

Hopefully another part of your prog validates for real state codes, so you don't accept eg XX

linuxmaveric · 04-04-2008, 01:57 AM

Well, my code was in the form of a BRE (basic regular expression), which is different from an ERE (extended regular expression). Suit yourself.

Sorry if I offended you. This is all very new to me and I'm just learning like everyone else here in the forum. No need to get touchy. Your code is sound. Take care.