def _scalar_data_changed(self, data):
img_data = self.image_data
if data is None:
img_data.point_data.scalars = None
self.data_changed = True
return
dims = list(data.shape)
if len(dims) == 2:
dims.append(1)
# set the dimension indices
dim0, dim1, dim2 = self.dimensions_order
img_data.origin = tuple(self.origin)
img_data.dimensions = tuple(dims)
img_data.extent = 0, dims[dim0] - 1, 0, dims[dim1] - 1, 0, dims[
dim2] - 1
if VTK_MAJOR_VERSION <= 7:
if is_old_pipeline():
img_data.update_extent = 0, dims[dim0] - 1, 0, dims[
dim1] - 1, 0, dims[dim2] - 1
else:
update_extent = [
0, dims[dim0] - 1, 0, dims[dim1] - 1, 0, dims[dim2] - 1
]
self.change_information_filter.set_update_extent(update_extent)
if self.transpose_input_array:
img_data.point_data.scalars = numpy.ravel(numpy.transpose(data))
else:
img_data.point_data.scalars = numpy.ravel(data)
img_data.point_data.scalars.name = self.scalar_name
# This is very important and if not done can lead to a segfault!
typecode = data.dtype
if is_old_pipeline():
img_data.scalar_type = array_handler.get_vtk_array_type(typecode)
img_data.update() # This sets up the extents correctly.
else:
filter_out_info = self.change_information_filter.get_output_information(
0)
img_data.set_point_data_active_scalar_info(
filter_out_info, array_handler.get_vtk_array_type(typecode),
-1)
img_data.modified()
img_data.update_traits()
self.change_information_filter.update()
# Now flush the mayavi pipeline.
self.data_changed = True
def _scalar_data_changed(self, data):
img_data = self.image_data
if data is None:
img_data.point_data.scalars = None
self.data_changed = True
return
dims = list(data.shape)
if len(dims) == 2:
dims.append(1)
img_data.origin = tuple(self.origin)
img_data.dimensions = tuple(dims)
img_data.extent = 0, dims[0] - 1, 0, dims[1] - 1, 0, dims[2] - 1
img_data.update_extent = 0, dims[0] - 1, 0, dims[1] - 1, 0, dims[2] - 1
if self.transpose_input_array:
img_data.point_data.scalars = numpy.ravel(numpy.transpose(data))
else:
img_data.point_data.scalars = numpy.ravel(data)
img_data.point_data.scalars.name = self.scalar_name
# This is very important and if not done can lead to a segfault!
typecode = data.dtype
img_data.scalar_type = array_handler.get_vtk_array_type(typecode)
img_data.update() # This sets up the extents correctly.
img_data.update_traits()
self.change_information_filter.update()
# Now flush the mayavi pipeline.
self.data_changed = True
def _scalar_data_changed(self, data):
img_data = self.image_data
if data is None:
img_data.point_data.scalars = None
self.data_changed = True
return
dims = list(data.shape)
if len(dims) == 2:
dims.append(1)
img_data.origin = tuple(self.origin)
img_data.dimensions = tuple(dims)
img_data.extent = 0, dims[0]-1, 0, dims[1]-1, 0, dims[2]-1
img_data.update_extent = 0, dims[0]-1, 0, dims[1]-1, 0, dims[2]-1
if self.transpose_input_array:
img_data.point_data.scalars = numpy.ravel(numpy.transpose(data))
else:
img_data.point_data.scalars = numpy.ravel(data)
img_data.point_data.scalars.name = self.scalar_name
# This is very important and if not done can lead to a segfault!
typecode = data.dtype
img_data.scalar_type = array_handler.get_vtk_array_type(typecode)
img_data.update() # This sets up the extents correctly.
img_data.update_traits()
self.change_information_filter.update()
# Now flush the mayavi pipeline.
self.data_changed = True
def _scalar_data_changed(self, data):
img_data = self.image_data
if data is None:
img_data.point_data.scalars = None
self.data_changed = True
return
dims = list(data.shape)
if len(dims) == 2:
dims.append(1)
# set the dimension indices
dim0, dim1, dim2 = self.dimensions_order
img_data.origin = tuple(self.origin)
img_data.dimensions = tuple(dims)
img_data.extent = 0, dims[dim0]-1, 0, dims[dim1]-1, 0, dims[dim2]-1
if VTK_MAJOR_VERSION <= 7:
if is_old_pipeline():
img_data.update_extent = 0, dims[dim0]-1, 0, dims[dim1]-1, 0, dims[dim2]-1
else:
update_extent = [0, dims[dim0]-1, 0, dims[dim1]-1, 0, dims[dim2]-1]
self.change_information_filter.set_update_extent(update_extent)
if self.transpose_input_array:
img_data.point_data.scalars = np.ravel(np.transpose(data))
else:
img_data.point_data.scalars = np.ravel(data)
img_data.point_data.scalars.name = self.scalar_name
# This is very important and if not done can lead to a segfault!
typecode = data.dtype
if is_old_pipeline():
img_data.scalar_type = get_vtk_array_type(typecode)
img_data.update() # This sets up the extents correctly.
else:
filter_out_info = self.change_information_filter.get_output_information(0)
img_data.set_point_data_active_scalar_info(filter_out_info,
get_vtk_array_type(typecode), -1)
img_data.modified()
img_data.update_traits()
self.change_information_filter.update()
# Now flush the mayavi pipeline.
self.data_changed = True
def numpy2vtk( arr, fname ):
# Make the data.
# dims = array((128, 128, 128))
# vol = array((-5., 5, -5, 5, -5, 5))
# origin = vol[::2]
# spacing = (vol[1::2] - origin)/(dims -1)
# xmin, xmax, ymin, ymax, zmin, zmax = vol
# x, y, z = ogrid[xmin:xmax:dims[0]*1j,
# ymin:ymax:dims[1]*1j,
# zmin:zmax:dims[2]*1j]
# x, y, z = [t.astype('f') for t in (x, y, z)]
# scalars = sin(x*y*z)/(x*y*z)
dims = array( arr.shape )
vol = array((-5.,5, -5, 5, -5, 5))
origin = vol[::2]
spacing = (vol[1::2] - origin)/(dims -1)
xmin, xmax, ymin, ymax, zmin, zmax = vol
x, y, z = ogrid[xmin:xmax:dims[0]*1j,
ymin:ymax:dims[1]*1j,
zmin:zmax:dims[2]*1j]
x, y, z = [t.astype('f') for t in (x, y, z)]
scalars = arr
# Make the tvtk dataset.
spoints = tvtk.StructuredPoints(origin=origin, spacing=spacing,
dimensions=dims)
# The copy makes the data contiguous and the transpose makes it
# suitable for display via tvtk. Note that it is not necessary to
# make the data contiguous since in that case the array is copied
# internally.
s = scalars.transpose().copy()
spoints.point_data.scalars = ravel(s)
spoints.point_data.scalars.name = 'scalars'
# This is needed in slightly older versions of VTK (like the 5.0.2
# release) to prevent a segfault. VTK does not detect the correct
# data type.
spoints.scalar_type = get_vtk_array_type(s.dtype)
# Uncomment the next two lines to save the dataset to a VTK XML file.
w = tvtk.XMLImageDataWriter(input=spoints, file_name=fname+'.vti')
w.write()
def StructuredScalar(InputArray, OutFileName, lim1, lim2):
nGr = np.shape(InputArray)[0]
# Make the data.
nx1IC = ny1 = nz1 = int(lim1)
nx2IC = ny2 = nz2 = int(lim2)
im = jm = km = nGr * 1j
nf0 = nf1 = nf2 = nGr
Xg, Yg, Zg = np.mgrid[nx1IC:nx2IC:im, ny1:ny2:jm, nz1:nz2:km]
# Make the data.
dims = np.array((nf0, nf1, nf2))
vol = np.array((lim1, lim2, lim1, lim2, lim1, lim2))
origin = vol[::2]
spacing = (vol[1::2] - origin) / (dims - 1)
xmin, xmax, ymin, ymax, zmin, zmax = vol
x, y, z = np.ogrid[xmin:xmax:dims[0] * 1j, ymin:ymax:dims[1] * 1j,
zmin:zmax:dims[2] * 1j]
x, y, z = [t.astype('f') for t in (x, y, z)]
scalars = InputArray
# Make the tvtk dataset.
spoints = tvtk.StructuredPoints(origin=origin,
spacing=spacing,
dimensions=dims)
s = scalars.transpose().copy()
spoints.point_data.scalars = np.ravel(s)
spoints.point_data.scalars.name = 'scalars'
spoints.scalar_type = get_vtk_array_type(s.dtype)
fileOut = OutFileName + '.vti'
print fileOut
w = tvtk.XMLImageDataWriter(input=spoints, file_name=fileOut)
w.write()
# Make the tvtk dataset.
spoints = tvtk.StructuredPoints(origin=origin, spacing=spacing,
dimensions=dims)
# The copy makes the data contiguous and the transpose makes it
# suitable for display via tvtk. Note that it is not necessary to
# make the data contiguous since in that case the array is copied
# internally.
s = scalars.transpose().copy()
spoints.point_data.scalars = ravel(s)
spoints.point_data.scalars.name = 'scalars'
# This is needed in slightly older versions of VTK (like the 5.0.2
# release) to prevent a segfault. VTK does not detect the correct
# data type.
spoints.scalar_type = get_vtk_array_type(s.dtype)
# Uncomment the next two lines to save the dataset to a VTK XML file.
#w = tvtk.XMLImageDataWriter(input=spoints, file_name='spoints3d.vti')
#w.write()
# Now view the data.
@mayavi2.standalone
def view():
from mayavi.sources.vtk_data_source import VTKDataSource
from mayavi.modules.outline import Outline
from mayavi.modules.image_plane_widget import ImagePlaneWidget
mayavi.new_scene()
src = VTKDataSource(data = spoints)
mayavi.add_source(src)
spoints = tvtk.StructuredPoints(origin=origin,
spacing=spacing,
dimensions=dims)
# The copy makes the data contiguous and the transpose makes it
# suitable for display via tvtk. Note that it is not necessary to
# make the data contiguous since in that case the array is copied
# internally.
s = scalars.transpose().copy()
spoints.point_data.scalars = ravel(s)
spoints.point_data.scalars.name = 'scalars'
# This is needed in slightly older versions of VTK (like the 5.0.2
# release) to prevent a segfault. VTK does not detect the correct
# data type.
if is_old_pipeline():
spoints.scalar_type = get_vtk_array_type(s.dtype)
# Uncomment the next two lines to save the dataset to a VTK XML file.
#w = tvtk.XMLImageDataWriter(input=spoints, file_name='spoints3d.vti')
#w.write()
# Now view the data.
@mayavi2.standalone
def view():
from mayavi.sources.vtk_data_source import VTKDataSource
from mayavi.modules.outline import Outline
from mayavi.modules.image_plane_widget import ImagePlaneWidget
mayavi.new_scene()
src = VTKDataSource(data=spoints)
