discomb_utilities.py

# get MF grid information from a DIS file
# authored by Mike Fienen, USGS - Wisconsin Water Science Center
# modified by Andrew Leaf 9/9/2013

import numpy as np

def read_nrow_ncol_vals(infile,NROW,NCOL,DTYPE,i):
    # read in NROW * NCOL values from a file read into indat
    # Returns TMP - a np array of type DTYPE
    # i is a counter that was already set as the file is parsed
    # i is returned as well to keep the party rockin'
    indat=open(infile,'r').readlines()
    
    ncells = NROW*NCOL
    contflag = True
    TMP = []    
    for i in np.arange(i,len(indat)):
        if contflag == False:
            break
        if len(TMP) < ncells:
            TMP.extend(indat[i].strip().split())
        else:
            contflag = False
    poo = len(TMP)
    TMP = np.array(TMP,dtype=DTYPE).reshape(NROW,NCOL)    
    return TMP, i

def read_meta_data(infile):
    indat=open(infile,'r').readlines()
    i=0
    DX = []
    DY = []
    uniform=False
    # remove the comment lines from the top of the file
    contflag = True
    while contflag:
        if '#' in indat[0]:
            junkus = indat.pop(0)
            i+=1
        else:
            contflag = False
            
    # get the control (uber) parameters
    tmp = indat.pop(0).strip().split()
    i+=1
    NLAY = int(tmp.pop(0))
    NROW = int(tmp.pop(0))
    NCOL = int(tmp.pop(0))

    # skip the CBD line
    junkus = indat.pop(0)
    i+=1    
    # check if grid is uniform (first number will be 0)
    if int(indat[0].split()[0])==0:
        uniform=True
        dx=float(indat[0].split()[1].split('(')[0])
        DX=(dx*np.arange(NCOL))
        junkus = indat.pop(0)
        i+=1
    if int(indat[0].split()[0])==0:
        dy=float(indat[0].split()[1].split('(')[0])
        DY=(dy*np.arange(NROW))
        junkus = indat.pop(0)
        i+=1
    # account for the header line in the next row
    i+=1
    if not uniform:
        junkus = indat.pop(0)
        i+=1
        # now read the DX values (make sure this is indented!)
        contflag = True
        for j in np.arange(len(indat)):
            i+=1
            if contflag == False:
                break
            if len(DX) < NCOL:
                DX.extend(indat[j].strip().split())
            else:
                contflag = False
        DX = np.array(DX,dtype=float)
                
        # now read the DY values
        contflag = True
        for j in np.arange(j,len(indat)):
            i+=1
            if contflag == False:
                break
            if len(DY) < NROW:
                DY.extend(indat[j].strip().split())
            else:
                contflag = False
        DY = np.array(DY,dtype=float)
                    
        # adjust DX and DY for pcolor plotting
        #DX = np.hstack([0.5*DX[0],DX])
        #DY = np.hstack([0.5*DY[0],DY])
        DX = np.cumsum(DX)
        DY = np.cumsum(DY)    
    return DX,DY,NLAY,NROW,NCOL,i