but, my philosophy is to correct before you enhance. For example, I like to use the compressor first because I like to correct excessive dynamics on the track. Then I'll use EQ to tweak the sound how I want it.
If there is some weird sound, however, like a ringing noise...or low rumble, I'll EQ to correct it before I compress it, then EQ again to shape the sound.
You also have to ask yourself either:
Do I want to boost part of the frequency spectrum after I've just smashed it down with the compressor?.....or
Do I want to smash down the frequencies I just boosted with EQ?
EQ-> Compress: will exagerate the EQ. For example, if you boost 2k then compress everything, the 2k will tend to trigger the compressor more. EVERYTHING will then be turned down - making the 2k bump more noticeable. If you primarily cut with the EQ this is much less an issue. Rule of thumb is corrective EQ before compression. Why have that boomy sound you don't want anyway triggering the compressor?
Compress->EQ: This allows the EQ to be a bit more predictable- I prefer this when I have serious sculpting to do with the EQ. So use shaping EQ after compression.
Or do whatever sounds best. With compression and EQ you *really* have to use your ears.