1. pts_gen.py contains the x and z points generation code for the points in h5 file. This process will be automated within the xmf code in next iteration.
2. pts_remover.py takes in y_points file, removes the unnecessary points and rewrites the file.
3. JHTDB has a fixed set of y points with varying delta_y. x and z widths are fixed. Thus, a separate y_points file is needed for just the y values. y-dir is the direction normal to the boundary layer (normal to flow dir axis-x).
4. xmf_h5_save and xmf_h5_save_multi_time_steps.py are files which generate an h5 file along with an xmf file.
5. xmf file has all the header info regarding what data contains what information.
6. h5 file has the x,y,z,v1_set,v2_set ... ,vn_set dataset arranged in a nice format. The xmf file along with h5 file can be read by most post processors, including paraview,Visit,etc.
7. This file system can handle multiple time steps. A single xmf file and h5 file are needed.
8. The xmf-h5 system is much more robust and efficient as compared to plain text files. One can also set what precision is needed, reducing excessive data storage necessities.
9. Both xmf_h5_save and xmf_h5_save_multi_time_steps.py take in .mat files generated by the bulk_retrieve.m code. xmf_h5_save does this for a single time step....xmf_h5_save_multi_time_steps.py does it for multiple time steps.