import numpy as np
from functools import cached_property
from glue.core import BaseData
from glue.core.subset_group import GroupedSubset
from bqplot import Lines
from glue_jupyter.bqplot.image import BqplotImageView
from jdaviz.core.registries import viewer_registry
from jdaviz.core.marks import SliceIndicatorMarks, ShadowSpatialSpectral
from jdaviz.configs.cubeviz.helper import layer_is_cube_image_data
from jdaviz.configs.default.plugins.viewers import JdavizViewerMixin
from jdaviz.configs.specviz.plugins.viewers import SpecvizProfileView
from jdaviz.core.events import AddDataMessage, RemoveDataMessage
from jdaviz.core.freezable_state import FreezableBqplotImageViewerState
from jdaviz.utils import get_subset_type
__all__ = ['CubevizImageView', 'CubevizProfileView',
'WithSliceIndicator', 'WithSliceSelection']
[docs]
class WithSliceIndicator:
@property
def slice_component_label(self):
return str(self.state.x_att)
@cached_property
def slice_indicator(self):
# SliceIndicatorMarks does not yet exist
slice_indicator = SliceIndicatorMarks(self)
self.figure.marks = self.figure.marks + slice_indicator.marks
return slice_indicator
@property
def slice_values(self):
def _get_component(layer):
try:
return layer.layer.get_component(self.slice_component_label).data
except (AttributeError, KeyError):
# layer either does not have get_component (because its a subset)
# or slice_component_label is not a component in this layer
# either way, return an empty array and skip this layer
return np.array([])
try:
return np.asarray(np.unique(np.concatenate([_get_component(layer) for layer in self.layers])), # noqa
dtype=float)
except ValueError:
return np.array([])
def _set_slice_indicator_value(self, value):
# this is a separate method so that viewers can override and map value if necessary
# NOTE: on first call, this will initialize the indicator itself
self.slice_indicator.value = value
[docs]
class WithSliceSelection:
@property
def slice_index(self):
# index in state.slices corresponding to the slice axis
return 2
@property
def slice_component_label(self):
slice_plg = self.jdaviz_helper.plugins.get('Slice', None)
if slice_plg is None: # pragma: no cover
raise ValueError("slice plugin must be activated to access slice_component_label")
return slice_plg._obj.slice_indicator_viewers[0].slice_component_label
@property
def slice_values(self):
# TODO: make a cached property and invalidate cache on add/remove data
# TODO: add support for multiple cubes (but then slice selection needs to be more complex)
# if slice_index is 0, then we want the equivalent of [:, 0, 0]
# if slice_index is 1, then we want the equivalent of [0, :, 0]
# if slice_index is 2, then we want the equivalent of [0, 0, :]
take_inds = [2, 1, 0]
take_inds.remove(self.slice_index)
for layer in self.layers:
try:
data_obj = layer.layer.data.get_component(self.slice_component_label).data
except (AttributeError, KeyError):
continue
else:
break
else:
return np.array([])
return np.asarray(data_obj.take(0, take_inds[0]).take(0, take_inds[1]), dtype=float)
@property
def slice(self):
return self.state.slices[self.slice_index]
@slice.setter
def slice(self, slice):
# NOTE: not intended for user-access - this should be controlled through the slice plugin
# in order to sync with all other viewers/slice indicators
slices = [0, 0, 0]
slices[self.slice_index] = slice
self.state.slices = tuple(slices)
@property
def slice_value(self):
return self.slice_values[self.slice]
@slice_value.setter
def slice_value(self, slice_value):
# NOTE: not intended for user-access - this should be controlled through the slice plugin
# in order to sync with all other viewers/slice indicators
# find the slice nearest slice_value
slice_values = self.slice_values
if not len(slice_values):
return
self.slice = np.argmin(abs(slice_values - slice_value))
[docs]
@viewer_registry("cubeviz-image-viewer", label="Image 2D (Cubeviz)")
class CubevizImageView(JdavizViewerMixin, WithSliceSelection, BqplotImageView):
# categories: zoom resets, (zoom, pan), subset, select tools, shortcuts
# NOTE: zoom and pan are merged here for space consideration and to avoid
# overflow to second row when opening the tray
tools_nested = [
['jdaviz:homezoom', 'jdaviz:prevzoom'],
['jdaviz:pixelboxzoommatch', 'jdaviz:boxzoom',
'jdaviz:pixelpanzoommatch', 'jdaviz:panzoom'],
['bqplot:truecircle', 'bqplot:rectangle', 'bqplot:ellipse',
'bqplot:circannulus'],
['jdaviz:spectrumperspaxel'],
['jdaviz:sidebar_plot', 'jdaviz:sidebar_export']
]
default_class = None
_state_cls = FreezableBqplotImageViewerState
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
# provide reference from state back to viewer to use for zoom syncing
self.state._viewer = self
self._subscribe_to_layers_update()
self.state.add_callback('reference_data', self._initial_x_axis)
# Hide axes by default
self.state.show_axes = False
@property
def _default_spectrum_viewer_reference_name(self):
return self.jdaviz_helper._default_spectrum_viewer_reference_name
@property
def _default_flux_viewer_reference_name(self):
return self.jdaviz_helper._default_flux_viewer_reference_name
@property
def _default_uncert_viewer_reference_name(self):
return self.jdaviz_helper._default_uncert_viewer_reference_name
@property
def active_image_layer(self):
"""Active image layer in the viewer, if available."""
# Find visible layers
visible_layers = [layer for layer in self.state.layers
if (layer.visible and layer_is_cube_image_data(layer.layer))]
if len(visible_layers) == 0:
return None
return visible_layers[-1]
def _initial_x_axis(self, *args):
# Make sure that the x_att is correct on data load
ref_data = self.state.reference_data
if ref_data and ref_data.ndim == 3:
for att_name in ["Right Ascension", "RA", "Galactic Longitude"]:
if att_name in ref_data.component_ids():
x_att = att_name
self.state.x_att_world = ref_data.id[x_att]
break
else:
x_att = "Pixel Axis 0 [z]"
self.state.x_att = ref_data.id[x_att]
[docs]
def set_plot_axes(self):
self.figure.axes[1].tick_format = None
self.figure.axes[0].tick_format = None
self.figure.axes[1].label = "y: pixels"
self.figure.axes[0].label = "x: pixels"
# Make it so y axis label is not covering tick numbers.
self.figure.axes[1].label_offset = "-50"
[docs]
def data(self, cls=None):
return [layer_state.layer # .get_object(cls=cls or self.default_class)
for layer_state in self.state.layers
if hasattr(layer_state, 'layer') and
isinstance(layer_state.layer, BaseData)]
[docs]
@viewer_registry("cubeviz-profile-viewer", label="Profile 1D (Cubeviz)")
class CubevizProfileView(SpecvizProfileView, WithSliceIndicator):
# categories: zoom resets, zoom, pan, subset, select tools, shortcuts
tools_nested = [
['jdaviz:homezoom', 'jdaviz:prevzoom'],
['jdaviz:boxzoom', 'jdaviz:xrangezoom', 'jdaviz:yrangezoom'],
['jdaviz:panzoom', 'jdaviz:panzoom_x', 'jdaviz:panzoom_y'],
['bqplot:xrange'],
['jdaviz:selectslice', 'jdaviz:selectline'],
['jdaviz:sidebar_plot', 'jdaviz:sidebar_export']
]
def __init__(self, *args, **kwargs):
kwargs.setdefault('default_tool_priority', ['jdaviz:selectslice'])
super().__init__(*args, **kwargs)
self.hub.subscribe(self, RemoveDataMessage,
handler=self._check_if_data_removed)
# TODO: Find out why this is not working
# self.hub.subscribe(self, DataCollectionDeleteMessage,
# handler=self._check_if_data_removed)
self.hub.subscribe(self, AddDataMessage,
handler=self._check_if_data_added)
@property
def _default_flux_viewer_reference_name(self):
return self.jdaviz_helper._default_flux_viewer_reference_name
def _check_if_data_removed(self, msg):
# isinstance and the data uuid check will be true for the data
# that is being removed
self.figure.marks = [m for m in self.figure.marks
if not (isinstance(m, ShadowSpatialSpectral)
and m.data_uuid == msg.data.uuid)]
def _check_if_data_added(self, msg=None):
# When data is added, make sure that all spatial subset layers
# that correspond with that data are checked for intersections
# with spectral subset layers
for layer in self.state.layers:
if layer.layer.data.label == msg.data.label:
if (isinstance(layer.layer, GroupedSubset) and
get_subset_type(layer.layer.subset_state) == 'spatial'):
self._expected_subset_layer_default(layer)
def _is_spatial_subset(self, layer):
subset_state = getattr(layer.layer, 'subset_state', None)
return get_subset_type(subset_state) == 'spatial'
def _get_spatial_subset_layers(self, data_label=None):
if data_label:
return [ls for ls in self.state.layers if (ls.layer.data.label == data_label and
self._is_spatial_subset(ls))]
return [ls for ls in self.state.layers if self._is_spatial_subset(ls)]
def _is_spectral_subset(self, layer):
subset_state = getattr(layer.layer, 'subset_state', None)
return get_subset_type(subset_state) == 'spectral'
def _get_spectral_subset_layers(self, data_label=None):
if data_label:
return [ls for ls in self.state.layers if (ls.layer.data.label == data_label and
self._is_spectral_subset(ls))]
return [ls for ls in self.state.layers if self._is_spectral_subset(ls)]
def _get_marks_for_layers(self, layers):
layers_list = list(self.state.layers)
# here we'll assume that all custom marks are subclasses of Lines/GL but don't directly
# use Lines/LinesGL (so an isinstance check won't be sufficient here)
layer_marks = self.native_marks
# and now we'll assume that the marks are in the same order as the layers, this should
# be the case as long as the order isn't manually resorted. If for any reason the layer
# is added but the mark has not yet been created, this will ignore that entry rather than
# raising an IndexError.
inds = [layers_list.index(layer) for layer in layers]
return [layer_marks[ind] for ind in inds if ind < len(layer_marks)]
def _on_subset_delete(self, msg):
# delete any ShadowSpatialSpectral mark for which either of the spectral or spatial marks
# no longer exists by matching the uuid of the msg subset to the uuid of the subsets
# in ShadowSpatialSpectral
super()._on_subset_delete(msg)
self.figure.marks = [m for m in self.figure.marks
if not (isinstance(m, ShadowSpatialSpectral)
and msg.subset.uuid in [m.spatial_uuid, m.spectral_uuid])]
def _expected_subset_layer_default(self, layer_state):
"""
This gets called whenever the layer of an expected new subset is added, we want to set the
default for the linewidth depending on whether it is spatial or spectral, and handle
creating any necessary marks for spatial-spectral subset intersections.
"""
def _marks_are_same(m, other):
# Check if ShadowSpatialSpectral mark already exists for particular
# data, spatial subset, and spectral subset combo
if (m.data_uuid == other.data_uuid
and m.spatial_uuid == other.spatial_uuid
and m.spectral_uuid == other.spectral_uuid):
return True
return False
def _is_unique(m):
unique = True
for m in existing_shadows_for_data:
if _marks_are_same(m, new_shadow):
unique = False
break
return unique
super()._expected_subset_layer_default(layer_state)
subset_type = get_subset_type(layer_state.layer)
if subset_type is None:
return
this_mark = self._get_marks_for_layers([layer_state])[0]
# new ShadowSpatialSpectral marks to be added
new_marks = []
# ShadowSpatialSpectral marks that already exists in the viewer
existing_shadows_for_data = [m for m in self.figure.marks
if isinstance(m, ShadowSpatialSpectral)]
if subset_type == 'spatial':
layer_state.linewidth = 1
# need to add marks for every intersection between THIS spatial subset and ALL spectral
# subset marks corresponding to this data
spectral_layers = [sub_layer for sub_layer in
self._get_spectral_subset_layers(layer_state.layer.data.label)]
spectral_marks = self._get_marks_for_layers(spectral_layers)
for index, spectral_mark in enumerate(spectral_marks):
new_shadow = ShadowSpatialSpectral(spatial_spectrum_mark=this_mark,
spectral_subset_mark=spectral_mark,
spatial_uuid=layer_state.layer.uuid,
spectral_uuid=spectral_layers[index].layer.uuid,
data_uuid=layer_state.layer.data.uuid)
if _is_unique(new_shadow):
new_marks.append(new_shadow)
# change opacity for live-collapsed spectra from spatial subsets in Cubeviz:
spatial_layers = [sub_layer for sub_layer in
self._get_spatial_subset_layers(layer_state.layer.data.label)]
spatial_marks = self._get_marks_for_layers(spatial_layers)
for layer, mark in zip(spatial_layers, spatial_marks):
# update profile opacities for spatial subset:
if isinstance(mark, Lines):
mark.set_trait(
'opacities',
# set the alpha for the spectrum in the profile viewer
# to be 50% more opaque than the alpha for the spatial subset
# in the flux-viewer
[min(1.5 * layer.alpha, 1)]
)
elif subset_type == 'spectral':
# need to add marks for every intersection between THIS spectral subset and ALL spatial
# subset marks corresponding to this data
spatial_layers = [sub_layer for sub_layer in
self._get_spatial_subset_layers(layer_state.layer.data.label)]
spatial_marks = self._get_marks_for_layers(spatial_layers)
for index, spatial_mark in enumerate(spatial_marks):
new_shadow = ShadowSpatialSpectral(spatial_spectrum_mark=spatial_mark,
spectral_subset_mark=this_mark,
spatial_uuid=spatial_layers[index].layer.uuid,
spectral_uuid=layer_state.layer.uuid,
data_uuid=layer_state.layer.data.uuid)
if _is_unique(new_shadow):
new_marks.append(new_shadow)
else:
return
# NOTE: += or append won't pick up on change
self.figure.marks = self.figure.marks + new_marks