The tracer particle code has been updated in several ways that make it more flexible:
1) tracer particle velocities can now be turned on and off using the TracerParticleOutputVelocity parameter, rather than using compile-time parameters.
2) tracer particles can now be initialized in 1D and 2D simulations, as well as in 3D simulations (which was all that we could do before).
3) extensive comments have been added to make the source code more understandable for people who want to modify it in the future.
NOTE: this set of commits does not change the output format in any way, which maintains backwards compatibility. In particular, it does not go from streaming binary to HDF5. While my ultimate goal is to move to HDF5 (as per the enzo-dev mailing list discussion from a week or two ago), I made these changes to help a user and feel that it makes sense to commit them to the repo so they're not lost (and also since they ARE actually useful).
Is there any reason why this should not be merged?