Thread: Algorithm for all possible combinations?

Hi i'm doing a simple translation program built from a a glossary and some statistic from a corpus.
Now i have realized that to get my program to work i need a algorithm that can get all the possible combinations from a N*M array.

EX:
[1(A1,A2,A3,......,An),2(B1,B2,B3,......,Bn),3(C1,C 2,C3,......,Cn),........N(N1,N2,N3,......,Nn)]

I haven't been able to solve this and would really need some help.

Sry if i wasn't clear enough. The first array have a unknown amount of arrays in it and everyone of those arrays have an unknown amount of string in them.

I think i explained badly again i need the algorithm to work for any amount of values but i will know how many there are for every iteration. For an example
if i need to translate how are you then every word have a couple of different translations that i get from a glossary so the array will be [(all translations of how),(all translation of are),(all translation of you)] .

ok with just two words the sentence in swedish "jag är" translates correctly in english to "I am" but if you look it up in a glossery every word will have more then one translation so "jag" can be translated to "I" and "ego" and "är" can be translated to "am", "are" and "is" so the array will look like this:
[("i","ego"),("am","are","is")]
the possible combinations are:
I am
I are
I is
ego am
ego are
ego is

Then the program will use some statistic to compare the different sentences and hopefully it will get the right one most of the time.

In this case the word did't have that many translation and it was a short sentence but the problem is of course that this can vary alot.

I hope the example is enough and i really appreciate that you are trying to help me.

thx it worked perfectly!!!