Depends on the sound, on the song, basically, it really depends on what it needs, which can vary from track to track.
I prefer compressing first, then add eq, though sometimes I'll filter, then compress, then eq. Other times I'll eq, then compresss. Threre's really no rule, you just experiment and tune in when it starts sounding like what you want.

A common thought would be to apply corrective EQ before compression - You don't want rumble, tubbiness or excessive sibilance triggering the compressor. Shaping EQ would be applied post-compressor, so the compressor isn't messing with your new shaping.

Hope that helps...

for me i EQ first then compress if needed and it work for me