-----------------------------------------------------------------
----------  MeanField Utilities  --------------------------------
-----------------------------------------------------------------

degeneracy_check.x:

  Determine which numbers of bands do not break a degenerate subspace, to avoid warnings or
  errors in Epsilon, Sigma, and BSE codes. Supply any number of binary WFN files as command-
  line arguments, and a list of numbers that are acceptable for all k-points in all files
  will be written to standard output.

mf_convert_wrapper.sh:

  Converts WFN/RHO/VXC files between binary and ASCII. This can be useful for moving files
  correctly between big- and little-endian machines, or for examining the contents of a
  file (though wfn_rho_vxc_info.x is more user-friendly). Using this wrapper script, the
  flavor (real/complex), type of file, and binary/ASCII is determined automatically.

wfn_rho_vxc_info.x:

  Prints the contents of the header of a (binary) WFN, RHO, or VXC file in a clearly labeled
  format, for human inspection.

wfnmerge.x:

  Merges many WFN files into one. It assumes that all input files have the 
  same number and same ordering of G-vectors for the charge density.
  The number and name of input files is read from "wfnmerge.inp", as well
  is the kgrid and the kshift.
 
  FAQ:
  Q: Why would someone want to use this?
  A: For example, maybe to do a converged calculation one needs to include a
      large number of unoccupied states. There may not be the resources (either
      CPUs or wallclock) to do this all in one shot, so it may be beneficial to
      split up a calculation of kpoints (e.g., 4x4x4 MP grid) into smaller
      pieces and then add up the final wavefunctions.
  Q: What is the deal with kpoint weights?
  A: Quantum Espresso renormalizes the kpoint weights that you use in a 
     calculation. If you are splitting up a MP grid into different groups of 
     kpoints, and each group is normalized, the relative weights between
     kpoints in different groups is no longer correct. BerkeleyGW does not 
     make use of the kpoint weights (except for in determining whether the 
     Fermi level is reasonable for metallic calculations). So for most uses
     the fact that these weights are incorrect does not matter. If it matters
     to you, you can simply modify wfnmerge to read in the kpoint weights of
     your choosing.

convert_old_to_new.x:

  Convert wavefunction, density, and exchange-correlation potential files from the old
  CWFE(q)/CD95/VXC formats (used in the pre-release version of the code prior to June 2011)
  to the current WFN/RHO/VXC formats. Some additional information is required which was not
  available in the old files, so an input file convert_old_to_new.inp is required. See the
  example in this directory. Input and output files are both binary.

