DAMASK_EICMD/processing/pre/geom_fromEuclideanDistance.py

288 lines
12 KiB
Python
Raw Normal View History

#!/usr/bin/env python
# -*- coding: UTF-8 no BOM -*-
import os,re,sys,math,numpy,string,damask
from scipy import ndimage
from optparse import OptionParser, OptionGroup, Option, SUPPRESS_HELP
scriptID = '$Id$'
scriptName = scriptID.split()[1]
#--------------------------------------------------------------------------------------------------
class extendableOption(Option):
#--------------------------------------------------------------------------------------------------
# used for definition of new option parser action 'extend', which enables to take multiple option arguments
# taken from online tutorial http://docs.python.org/library/optparse.html
ACTIONS = Option.ACTIONS + ("extend",)
STORE_ACTIONS = Option.STORE_ACTIONS + ("extend",)
TYPED_ACTIONS = Option.TYPED_ACTIONS + ("extend",)
ALWAYS_TYPED_ACTIONS = Option.ALWAYS_TYPED_ACTIONS + ("extend",)
def take_action(self, action, dest, opt, value, values, parser):
if action == "extend":
lvalue = value.split(",")
values.ensure_value(dest, []).extend(lvalue)
else:
Option.take_action(self, action, dest, opt, value, values, parser)
def periodic_3Dpad(array, rimdim=(1,1,1)):
rimdim = numpy.array(rimdim,'i')
size = numpy.array(array.shape,'i')
padded = numpy.empty(size+2*rimdim,array.dtype)
padded[rimdim[0]:rimdim[0]+size[0],
rimdim[1]:rimdim[1]+size[1],
rimdim[2]:rimdim[2]+size[2]] = array
p = numpy.zeros(3,'i')
for side in xrange(3):
for p[(side+2)%3] in xrange(padded.shape[(side+2)%3]):
for p[(side+1)%3] in xrange(padded.shape[(side+1)%3]):
for p[side%3] in xrange(rimdim[side%3]):
spot = (p-rimdim)%size
padded[p[0],p[1],p[2]] = array[spot[0],spot[1],spot[2]]
for p[side%3] in xrange(rimdim[side%3]+size[side%3],size[side%3]+2*rimdim[side%3]):
spot = (p-rimdim)%size
padded[p[0],p[1],p[2]] = array[spot[0],spot[1],spot[2]]
return padded
#--------------------------------------------------------------------------------------------------
# MAIN
#--------------------------------------------------------------------------------------------------
synonyms = {
'grid': ['resolution'],
'size': ['dimension'],
}
identifiers = {
'grid': ['a','b','c'],
'size': ['x','y','z'],
'origin': ['x','y','z'],
}
mappings = {
'grid': lambda x: int(x),
'size': lambda x: float(x),
'origin': lambda x: float(x),
'homogenization': lambda x: int(x),
'microstructures': lambda x: int(x),
}
features = [
{'aliens': 1, 'names': ['boundary','biplane'],},
{'aliens': 2, 'names': ['tripleline',],},
{'aliens': 3, 'names': ['quadruplepoint',],}
]
neighborhoods = {
'neumann':numpy.array([
[-1, 0, 0],
[ 1, 0, 0],
[ 0,-1, 0],
[ 0, 1, 0],
[ 0, 0,-1],
[ 0, 0, 1],
]),
'moore':numpy.array([
[-1,-1,-1],
[ 0,-1,-1],
[ 1,-1,-1],
[-1, 0,-1],
[ 0, 0,-1],
[ 1, 0,-1],
[-1, 1,-1],
[ 0, 1,-1],
[ 1, 1,-1],
#
[-1,-1, 0],
[ 0,-1, 0],
[ 1,-1, 0],
[-1, 0, 0],
#
[ 1, 0, 0],
[-1, 1, 0],
[ 0, 1, 0],
[ 1, 1, 0],
#
[-1,-1, 1],
[ 0,-1, 1],
[ 1,-1, 1],
[-1, 0, 1],
[ 0, 0, 1],
[ 1, 0, 1],
[-1, 1, 1],
[ 0, 1, 1],
[ 1, 1, 1],
])
}
parser = OptionParser(option_class=extendableOption, usage='%prog options [file[s]]', description = """
Produce geom files containing Euclidean distance to grain structural features:
boundaries, triple lines, and quadruple points.
""" + string.replace(scriptID,'\n','\\n')
)
parser.add_option('-t','--type', dest = 'type', action = 'extend', type = 'string', metavar = '<string LIST>',
help = 'feature type (%s) '%(', '.join(map(lambda x:'|'.join(x['names']),features))) )
parser.add_option('-n','--neighborhood', dest='neighborhood', choices = neighborhoods.keys(), metavar = 'string',
help = 'type of neighborhood (%s) [neumann]'%(', '.join(neighborhoods.keys())))
parser.add_option('-s', '--scale', dest = 'scale', type = 'float',
help = 'voxel size [%default]')
parser.set_defaults(type = [])
parser.set_defaults(neighborhood = 'neumann')
parser.set_defaults(scale = 1.0)
(options,filenames) = parser.parse_args()
feature_list = []
for i,feature in enumerate(features):
for name in feature['names']:
for myType in options.type:
if name.startswith(myType):
feature_list.append(i) # remember valid features
break
#--- setup file handles ---------------------------------------------------------------------------
files = []
if filenames == []:
files.append({'name':'STDIN',
'input':sys.stdin,
'output':sys.stdout,
'croak':sys.stderr,
})
else:
for name in filenames:
if os.path.exists(name):
files.append({'name':name,
'input':open(name),
'output':[open(features[feature]['names'][0]+'_'+name,'w') for feature in feature_list],
'croak':sys.stdout,
})
#--- loop over input files ------------------------------------------------------------------------
for file in files:
if file['name'] != 'STDIN': file['croak'].write('\033[1m'+scriptName+'\033[0m: '+file['name']+'\n')
else: file['croak'].write('\033[1m'+scriptName+'\033[0m\n')
theTable = damask.ASCIItable(file['input'],file['output'][0],labels = False)
theTable.head_read()
#--- interpret header ----------------------------------------------------------------------------
info = {
'grid': numpy.zeros(3,'i'),
'size': numpy.zeros(3,'d'),
'origin': numpy.zeros(3,'d'),
'homogenization': 0,
'microstructures': 0,
}
newInfo = {
'grid': numpy.zeros(3,'i'),
'origin': numpy.zeros(3,'d'),
'microstructures': 0,
}
extra_header = []
for header in theTable.info:
headitems = map(str.lower,header.split())
if len(headitems) == 0: continue # skip blank lines
for synonym,alternatives in synonyms.iteritems():
if headitems[0] in alternatives: headitems[0] = synonym
if headitems[0] in mappings.keys():
if headitems[0] in identifiers.keys():
for i in xrange(len(identifiers[headitems[0]])):
info[headitems[0]][i] = \
mappings[headitems[0]](headitems[headitems.index(identifiers[headitems[0]][i])+1])
else:
info[headitems[0]] = mappings[headitems[0]](headitems[1])
else:
extra_header.append(header)
file['croak'].write('grid a b c: %s\n'%(' x '.join(map(str,info['grid']))) + \
'size x y z: %s\n'%(' x '.join(map(str,info['size']))) + \
'origin x y z: %s\n'%(' : '.join(map(str,info['origin']))) + \
'homogenization: %i\n'%info['homogenization'] + \
'microstructures: %i\n'%info['microstructures'])
if numpy.any(info['grid'] < 1):
file['croak'].write('invalid grid a b c.\n')
continue
if numpy.any(info['size'] <= 0.0):
file['croak'].write('invalid size x y z.\n')
continue
#--- read data ------------------------------------------------------------------------------------
microstructure = numpy.zeros(info['grid'].prod(),'i') # initialize as flat array
i = 0
while theTable.data_read():
items = theTable.data
if len(items) > 2:
if items[1].lower() == 'of': items = [int(items[2])]*int(items[0])
elif items[1].lower() == 'to': items = xrange(int(items[0]),1+int(items[2]))
else: items = map(int,items)
else: items = map(int,items)
s = len(items)
microstructure[i:i+s] = items
i += s
neighborhood = neighborhoods[options.neighborhood]
convoluted = numpy.empty([len(neighborhood)]+list(info['grid']+2),'i')
structure = periodic_3Dpad(microstructure.reshape(info['grid'],order='F'))
for i,p in enumerate(neighborhood):
stencil = numpy.zeros((3,3,3),'i')
stencil[1,1,1] = -1
stencil[p[0]+1,
p[1]+1,
p[2]+1] = 1
convoluted[i,:,:,:] = ndimage.convolve(structure,stencil)
distance = numpy.ones((len(feature_list),info['grid'][0],info['grid'][1],info['grid'][2]),'d')
convoluted = numpy.sort(convoluted,axis = 0)
uniques = numpy.where(convoluted[0,1:-1,1:-1,1:-1] != 0, 1,0) # initialize unique value counter (exclude myself [= 0])
for i in xrange(1,len(neighborhood)): # check remaining points in neighborhood
uniques += numpy.where(numpy.logical_and(
convoluted[i,1:-1,1:-1,1:-1] != convoluted[i-1,1:-1,1:-1,1:-1], # flip of ID difference detected?
convoluted[i,1:-1,1:-1,1:-1] != 0), # not myself?
1,0) # count flip
for i,feature_id in enumerate(feature_list):
distance[i,:,:,:] = numpy.where(uniques >= features[feature_id]['aliens'],0.0,1.0) # seed with 0.0 when enough unique neighbor IDs are present
for i in xrange(len(feature_list)):
distance[i,:,:,:] = ndimage.morphology.distance_transform_edt(distance[i,:,:,:])*[options.scale]*3
for i,feature in enumerate(feature_list):
newInfo['microstructures'] = int(math.ceil(distance[i,:,:,:].max()))
#--- write header ---------------------------------------------------------------------------------
theTable = damask.ASCIItable(file['input'],file['output'][i],labels = False)
theTable.labels_clear()
theTable.info_clear()
theTable.info_append(extra_header+[
scriptID + ' ' + ' '.join(sys.argv[1:]),
"grid\ta %i\tb %i\tc %i"%(info['grid'][0],info['grid'][1],info['grid'][2],),
"size\tx %f\ty %f\tz %f"%(info['size'][0],info['size'][1],info['size'][2],),
"origin\tx %f\ty %f\tz %f"%(info['origin'][0],info['origin'][1],info['origin'][2],),
"homogenization\t%i"%info['homogenization'],
"microstructures\t%i"%(newInfo['microstructures']),
])
theTable.head_write()
theTable.output_flush()
# --- write microstructure information ------------------------------------------------------------
formatwidth = int(math.floor(math.log10(distance[i,:,:,:].max())+1))
theTable.data = distance[i,:,:,:].reshape((info['grid'][0],info['grid'][1]*info['grid'][2]),order='F').transpose()
theTable.data_writeArray('%%%ii'%(formatwidth),delimiter=' ')
file['output'][i].close()
#--- output finalization --------------------------------------------------------------------------
if file['name'] != 'STDIN':
file['input'].close()