search for: a1b0

...e a series of cases each with categorical factors A, B. What is the best way to "glue" A and B together into a single factor? For example, given A0 B1 ... A1 B1 ... A0 B2 ... A1 B0 ... A0 B0 ... A1 B2 ... then I'd like to end up with a single factor with levels A0B0, A0B1, A0B2, A1B0, A1B1, A1B2 according to all the combinations which actually occur in the data (e.g. if (A1,B2) did not occur in the data, then A1B2 would not be a level in the resulting "glued" factor). I can see clumsy and tedious ways involving manipulating strings, paste(), etc.; but I'm sure i...