Comments on: Fixing Postcodes

By: Jonathan Snook

Jonathan Snook — Sun, 28 Dec 2008 11:12:40 +0000

Would the handling of the file be any different if the postal code data was wrapped in quotes, just like the text data? I’d presume that Excel would assume it text in that case. But you know what happens when you assume…

By: Garrett Fitzgerald

Garrett Fitzgerald — Thu, 11 Dec 2008 15:32:43 +0000

*answers own dumb question*

Ah, right, not all the world runs on Windows.

*headdesk*

By: Garrett Fitzgerald

Garrett Fitzgerald — Thu, 11 Dec 2008 15:31:42 +0000

...and if we're talking text editors, how come nobody's mentioned Notepad++ yet?

By: Garrett Fitzgerald

Garrett Fitzgerald — Thu, 11 Dec 2008 15:30:43 +0000

When I worked at the mailshop, we’d always import in Excel — but then we’d toss it over to Visual FoxPro, where we had all our manipulation routines. It became so second-nature to do the import without losing the leading zeros that I can’t even tell you at this point how we did it. :-)

By: Nate Walton

Nate Walton — Mon, 24 Nov 2008 15:08:11 +0000

@Chuck: I found that with this zip code data set I was doing a lot of work in my text editor as well. I use TextMate, and the regular expression search and replace was invaluable.

By: Erling Ormar Vignisson

Erling Ormar Vignisson — Mon, 24 Nov 2008 09:49:55 +0000

No sooner had I read this article, acknowledged the problem… and later that very day, was I tasked with helping to solve this very problem! Nicely done ;)

By: Chuck Keller

Chuck Keller — Sun, 23 Nov 2008 17:47:13 +0000

I found Excel’s ‘helpfulness’ to be maddening as well after having to clean up a client’s mailing list. If you need to do specific manipulation to data in columns or rows that the import wizard cannot handle, set the data to text, select the column or row and copy-paste into BBEdit. In BBEdit you can do grep search and replace to further clean up your data before pasting it back into Excel. Not an elegant solution but it does work.

By: Eric Meyer

Eric Meyer — Thu, 20 Nov 2008 12:42:57 +0000

That would be easier, Matt, if I weren't dealing with a planet's worth of postcodes. I can't just assume five digits when some countries use four, some use a mixture of numbers and letters, and some use words with numbers. Not to mention all the postcode data was hand-entered by survey-takers, so there might be just plain wrong entries, and I don't want to force-correct them into what might be completely inaccurate values. So I really did need to open the file without any correction of the data, either by the program or by me.

By: Matt Rushton

Matt Rushton — Thu, 20 Nov 2008 11:32:50 +0000

An easier option is to just open the file in Excel. Then select the column, choose Format Cells and then select the Custom Category.

Enter #0 and then as many additional 0s that you need to precede the numeric.

So where 1 is the cell value, #00 will show 01 and #000 will show 001.

By: Eric Meyer

Eric Meyer — Wed, 19 Nov 2008 18:00:51 +0000

And yet, there are tons of people who have to use Excel and use it to open data from other sources, so perhaps this will be of some use to them. Which is, as I said, part of why I posted it.

By: Juliana Padron

Juliana Padron — Wed, 19 Nov 2008 17:24:53 +0000

Excel?? OpenOffice Ruuuulezz… =)

By: Casey

Casey — Wed, 19 Nov 2008 15:10:23 +0000

I have to agree with Matthew. OpenOffice Calc is great for dealing with CSV and similar files. It’s the main reason I have it on my system. In fact, all CSV files are associated with OpenOffice Calc so all I have to do is double-click and I’m good to go!

By: Matthew Brundage

Matthew Brundage — Wed, 19 Nov 2008 14:45:51 +0000

In my experience, OpenOffice Calc is more intuitive when handling CSV, especially when exporting. If I have a spreadsheet that needs to become a database table, I always run it through OOCalc first.

By: Eric Meyer

Eric Meyer — Wed, 19 Nov 2008 14:29:12 +0000

It's "Argentina, Argentine Republic", which (like all the country names) comes the Wikipedia list of countries by continent, which is in turn based on UN and ISO data.

By: Lurker

Lurker — Wed, 19 Nov 2008 06:28:34 +0000

Why do you have .ar as the “Argentine Republic of Argentina” ?