You're right, there is no easy way. I managed to do it using Teethless Mama's
recommendation but it also involved a lot of find, replace, and sort.

Thanks for the reply

"smartin" wrote:

Sei wrote:
I'm trying to separate the street number and street name but when with a
foreign address I'm having a hard time especially if the number is right in
the middle such as:

Av 47 1330 Col Hidalgo
Av 55 339 Co. Hidalgo
Av Aguas Calientas 1332 Col Puebl
Av Baja De Los Angeles 1283
Av Cesar Prieto 3139 Calocio
Av Colima 3033 Col Oscar G

Looking at your sample data I think the first question you you ask
yourself is "how do I know what a street number is?" The first two rows
have two numbers in them, which are probably a "street name" and "street
number". Given this, I suppose we/you could work something out that
ignores numbers following common street prefixes like "Av".

But what if you were confronted with data like

Av 47 1330 Col Hidalgo Ste 10
or
Av 47 1330A Col Hidalgo
or
Av 47 1330-1/2 Col Hidalgo

?

Just food for thought... Parsing data like this is seldom a
straightforward task, as there is seldom a typical solution.

As you take a long look at the data in front of you, and think about
what data may come, are there any patterns you can isolate and describe
to make identifying a "street number" a sure bet (or pretty close)?