1
0
mirror of https://github.com/bec-project/bec_widgets.git synced 2026-05-07 15:24:20 +02:00
Files
bec_widgets/bec_widgets/utils/crosshair.py
T

651 lines
26 KiB
Python
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
from __future__ import annotations
from collections import defaultdict
from typing import Any
import numpy as np
import pyqtgraph as pg
from qtpy.QtCore import QObject, QPointF, Qt, Signal
from qtpy.QtGui import QCursor, QTransform
from qtpy.QtWidgets import QApplication
from bec_widgets.utils.error_popups import SafeSlot
from bec_widgets.widgets.plots.image.image_item import ImageItem
class CrosshairScatterItem(pg.ScatterPlotItem):
def setDownsampling(self, ds=None, auto=None, method=None):
pass
def setClipToView(self, state):
pass
def setAlpha(self, *args, **kwargs):
pass
class Crosshair(QObject):
# QT Position of mouse cursor
positionChanged = Signal(tuple)
positionClicked = Signal(tuple)
# Plain crosshair position signals mapped to real coordinates
crosshairChanged = Signal(tuple)
crosshairClicked = Signal(tuple)
# Signal for 1D plot
coordinatesChanged1D = Signal(tuple)
coordinatesClicked1D = Signal(tuple)
# Signal for 2D plot
coordinatesChanged2D = Signal(tuple)
coordinatesClicked2D = Signal(tuple)
def __init__(
self,
plot_item: pg.PlotItem,
precision: int | None = None,
*,
min_precision: int = 2,
parent=None,
):
"""
Crosshair for 1D and 2D plots.
Args:
plot_item (pyqtgraph.PlotItem): The plot item to which the crosshair will be attached.
precision (int | None, optional): Fixed number of decimal places to display. If *None*, precision is chosen dynamically from the current view range.
min_precision (int, optional): The lower bound (in decimal places) used when dynamic precision is enabled. Defaults to 2.
parent (QObject, optional): Parent object for the QObject. Defaults to None.
"""
super().__init__(parent)
self.is_log_y = None
self.is_log_x = None
self.is_derivative = None
self.plot_item = plot_item
self._precision = precision
self._min_precision = max(0, int(min_precision)) # ensure nonnegative
self.v_line = pg.InfiniteLine(angle=90, movable=False)
self.v_line.skip_auto_range = True
self.h_line = pg.InfiniteLine(angle=0, movable=False)
self.h_line.skip_auto_range = True
# Add custom attribute to identify crosshair lines
self.v_line.is_crosshair = True
self.h_line.is_crosshair = True
self.plot_item.addItem(self.v_line, ignoreBounds=True)
self.plot_item.addItem(self.h_line, ignoreBounds=True)
# Initialize highlighted curve in a case of multiple curves
self.highlighted_curve_index = None
# Add TextItem to display coordinates
self.coord_label = pg.TextItem("", anchor=(1, 1), fill=(0, 0, 0, 100))
self.coord_label.setVisible(False) # Hide initially
self.coord_label.skip_auto_range = True
self.plot_item.addItem(self.coord_label)
# Signals to connect
self.proxy = pg.SignalProxy(
self.plot_item.scene().sigMouseMoved, rateLimit=60, slot=self.mouse_moved
)
self.positionChanged.connect(self.update_coord_label)
self.plot_item.scene().sigMouseClicked.connect(self.mouse_clicked)
# Connect signals from pyqtgraph right click menu
self.plot_item.ctrl.derivativeCheck.checkStateChanged.connect(self.check_derivatives)
self.plot_item.ctrl.logXCheck.checkStateChanged.connect(self.check_log)
self.plot_item.ctrl.logYCheck.checkStateChanged.connect(self.check_log)
self.plot_item.ctrl.downsampleSpin.valueChanged.connect(self.clear_markers)
# Initialize markers
self.items = []
self.marker_moved_1d = {}
self.marker_clicked_1d = {}
self.marker_2d_row = None
self.marker_2d_col = None
self.update_markers()
self.check_log()
self.check_derivatives()
self._connect_to_theme_change()
@property
def precision(self) -> int | None:
"""Fixed number of decimals; ``None`` enables dynamic mode."""
return self._precision
@precision.setter
def precision(self, value: int | None):
"""
Set the fixed number of decimals to display.
Args:
value(int | None): The number of decimals to display. If `None`, dynamic precision is used based on the view range.
"""
self._precision = value
@property
def min_precision(self) -> int:
"""Lower bound on decimals when dynamic precision is used."""
return self._min_precision
@min_precision.setter
def min_precision(self, value: int):
"""
Set the lower bound on decimals when dynamic precision is used.
Args:
value(int): The minimum number of decimals to display. Must be non-negative.
"""
self._min_precision = max(0, int(value))
def _current_precision(self) -> int:
"""
Get the current precision based on the view range or fixed precision.
"""
if self._precision is not None:
return self._precision
# Dynamically choose precision from the smaller visible span
view_range = self.plot_item.vb.viewRange()
x_span = abs(view_range[0][1] - view_range[0][0])
y_span = abs(view_range[1][1] - view_range[1][0])
# Ignore zero spans that can appear during initialisation
spans = [s for s in (x_span, y_span) if s > 0]
span = min(spans) if spans else 1.0
exponent = np.floor(np.log10(span)) # order of magnitude
decimals = max(0, int(-exponent) + 1)
return max(self._min_precision, decimals)
def _connect_to_theme_change(self):
"""Connect to the theme change signal."""
qapp = QApplication.instance()
if hasattr(qapp, "theme_signal"):
qapp.theme_signal.theme_updated.connect(self._update_theme)
self._update_theme()
@SafeSlot(str)
def _update_theme(self, theme: str | None = None):
"""Update the theme."""
if theme is None:
qapp = QApplication.instance()
if hasattr(qapp, "theme"):
theme = qapp.theme.theme
else:
theme = "dark"
self.apply_theme(theme)
def apply_theme(self, theme: str):
"""Apply the theme to the plot."""
if theme == "dark":
text_color = "w"
label_bg_color = (50, 50, 50, 150)
elif theme == "light":
text_color = "k"
label_bg_color = (240, 240, 240, 150)
else:
text_color = "w"
label_bg_color = (50, 50, 50, 150)
self.coord_label.setColor(text_color)
self.coord_label.fill = pg.mkBrush(label_bg_color)
self.coord_label.border = pg.mkPen(None)
@SafeSlot(int)
def update_highlighted_curve(self, curve_index: int):
"""
Update the highlighted curve in the case of multiple curves in a plot item.
Args:
curve_index(int): The index of curve to highlight
"""
self.highlighted_curve_index = curve_index
self.clear_markers()
self.update_markers()
def update_markers(self):
"""Update the markers for the crosshair, creating new ones if necessary."""
if self.highlighted_curve_index is not None and hasattr(self.plot_item, "visible_curves"):
# Focus on the highlighted curve only
self.items = [self.plot_item.visible_curves[self.highlighted_curve_index]]
elif hasattr(self.plot_item, "visible_items"): # PlotBase general case
# Handle visible items in the plot item
self.items = self.plot_item.visible_items()
else: # Non PlotBase case
# Handle all items
self.items = self.plot_item.items
# Create or update markers
for item in self.items:
if isinstance(item, pg.PlotDataItem): # 1D plot
pen = item.opts["pen"]
color = pen.color() if hasattr(pen, "color") else pg.mkColor(pen)
name = item.name() or str(id(item))
if name in self.marker_moved_1d:
# Update existing markers
marker_moved = self.marker_moved_1d[name]
marker_moved.setPen(pg.mkPen(color))
# Update clicked markers' brushes
for marker_clicked in self.marker_clicked_1d[name]:
alpha = marker_clicked.opts["brush"].color().alpha()
marker_clicked.setBrush(
pg.mkBrush(color.red(), color.green(), color.blue(), alpha)
)
# Update z-values
marker_moved.setZValue(item.zValue() + 1)
for marker_clicked in self.marker_clicked_1d[name]:
marker_clicked.setZValue(item.zValue() + 1)
else:
# Create new markers
marker_moved = CrosshairScatterItem(
size=10, pen=pg.mkPen(color), brush=pg.mkBrush(None)
)
marker_moved.skip_auto_range = True
marker_moved.is_crosshair = True
self.marker_moved_1d[name] = marker_moved
self.plot_item.addItem(marker_moved)
# Set marker z-value higher than the curve
marker_moved.setZValue(item.zValue() + 1)
# Create glowing effect markers for clicked events
marker_clicked_list = []
for size, alpha in [(18, 64), (14, 128), (10, 255)]:
marker_clicked = CrosshairScatterItem(
size=size,
pen=pg.mkPen(None),
brush=pg.mkBrush(color.red(), color.green(), color.blue(), alpha),
)
marker_clicked.skip_auto_range = True
marker_clicked.is_crosshair = True
self.plot_item.addItem(marker_clicked)
marker_clicked.setZValue(item.zValue() + 1)
marker_clicked_list.append(marker_clicked)
self.marker_clicked_1d[name] = marker_clicked_list
elif isinstance(item, pg.ImageItem): # 2D plot
if self.marker_2d_row is not None and self.marker_2d_col is not None:
continue
# Create horizontal ROI for row highlighting
if item.image is None:
continue
self.marker_2d_row = pg.ROI(
[0, 0], size=[item.image.shape[0], 1], pen=pg.mkPen("r", width=2), movable=False
)
self.marker_2d_row.skip_auto_range = True
if item.image_transform is not None:
self.marker_2d_row.setTransform(item.image_transform)
self.plot_item.addItem(self.marker_2d_row)
# Create vertical ROI for column highlighting
self.marker_2d_col = pg.ROI(
[0, 0], size=[1, item.image.shape[1]], pen=pg.mkPen("r", width=2), movable=False
)
if item.image_transform is not None:
self.marker_2d_col.setTransform(item.image_transform)
self.marker_2d_col.skip_auto_range = True
self.plot_item.addItem(self.marker_2d_col)
@SafeSlot()
def update_markers_on_image_change(self):
"""
Update markers when the image changes, e.g. when the
image shape or transformation changes.
"""
for item in self.items:
if not isinstance(item, pg.ImageItem):
continue
if self.marker_2d_row is not None:
self.marker_2d_row.setSize([item.image.shape[0], 1])
self.marker_2d_row.setTransform(item.image_transform)
if self.marker_2d_col is not None:
self.marker_2d_col.setSize([1, item.image.shape[1]])
self.marker_2d_col.setTransform(item.image_transform)
# Get the current mouse position
views = self.plot_item.vb.scene().views()
if not views:
return
view = views[0]
global_pos = QCursor.pos()
view_pos = view.mapFromGlobal(global_pos)
scene_pos = view.mapToScene(view_pos)
if self.plot_item.vb.sceneBoundingRect().contains(scene_pos):
plot_pt = self.plot_item.vb.mapSceneToView(scene_pos)
self.mouse_moved(manual_pos=(plot_pt.x(), plot_pt.y()))
def snap_to_data(
self, x: float, y: float
) -> tuple[None, None] | tuple[defaultdict[Any, list], defaultdict[Any, list]]:
"""
Finds the nearest data points to the given x and y coordinates.
Args:
x(float): The x-coordinate of the mouse cursor
y(float): The y-coordinate of the mouse cursor
Returns:
tuple: x and y values snapped to the nearest data
"""
y_values = defaultdict(list)
x_values = defaultdict(list)
# Iterate through items in the plot
for item in self.items:
if isinstance(item, pg.PlotDataItem): # 1D plot
name = item.name() or str(id(item))
plot_data = item._getDisplayDataset()
if plot_data is None:
continue
x_data, y_data = plot_data.x, plot_data.y
if x_data is not None and y_data is not None:
if self.is_log_x:
min_x_data = np.min(x_data[x_data > 0])
else:
min_x_data = np.min(x_data)
max_x_data = np.max(x_data)
if x < min_x_data or x > max_x_data:
y_values[name] = None
x_values[name] = None
continue
closest_x, closest_y = self.closest_x_y_value(x, x_data, y_data)
y_values[name] = closest_y
x_values[name] = closest_x
elif isinstance(item, pg.ImageItem): # 2D plot
name = item.objectName() or str(id(item))
image_2d = item.image
if image_2d is None:
continue
# Map scene coordinates (plot units) back to image pixel coordinates
if item.image_transform is not None:
inv_transform, _ = item.image_transform.inverted()
xy_trans = inv_transform.map(QPointF(x, y))
else:
xy_trans = QPointF(x, y)
# Define valid pixel coordinate bounds
min_x_px, min_y_px = 0, 0
max_x_px = image_2d.shape[0] - 1 # columns
max_y_px = image_2d.shape[1] - 1 # rows
# Clip the mapped coordinates to the image bounds
px = int(np.clip(xy_trans.x(), min_x_px, max_x_px))
py = int(np.clip(xy_trans.y(), min_y_px, max_y_px))
# Store snapped pixel positions
x_values[name] = px
y_values[name] = py
if x_values and y_values:
if all(v is None for v in x_values.values()) or all(
v is None for v in y_values.values()
):
return None, None
return x_values, y_values
return None, None
def closest_x_y_value(self, input_x: float, list_x: list, list_y: list) -> tuple:
"""
Find the closest x and y value to the input value.
Args:
input_x (float): Input value
list_x (list): List of x values
list_y (list): List of y values
Returns:
tuple: Closest x and y value
"""
# Convert lists to NumPy arrays
arr_x = np.asarray(list_x)
# Get the indices where x is not NaN
valid_indices = ~np.isnan(arr_x)
# Filter x array to exclude NaN values
filtered_x = arr_x[valid_indices]
# Find the index of the closest value in the filtered x array
closest_index = np.abs(filtered_x - input_x).argmin()
# Map back to the original index in the list_x and list_y arrays
original_index = np.where(valid_indices)[0][closest_index]
return list_x[original_index], list_y[original_index]
@SafeSlot(object, tuple)
def mouse_moved(self, event=None, manual_pos=None):
"""
Handles the mouse moved event, updating the crosshair position and emitting signals.
Args:
event(object): The mouse moved event, which contains the scene position.
manual_pos(tuple, optional): A tuple containing the (x, y) coordinates to manually set the crosshair position.
"""
# Determine target (x, y) in *plot* coordinates
if manual_pos is not None:
x, y = manual_pos
else:
if event is None:
return # nothing to do
scene_pos = event[0] # SignalProxy bundle
if not self.plot_item.vb.sceneBoundingRect().contains(scene_pos):
return
view_pos = self.plot_item.vb.mapSceneToView(scene_pos)
x, y = view_pos.x(), view_pos.y()
# Update crosshair visuals
self.v_line.setPos(x)
self.h_line.setPos(y)
self.update_markers()
scaled_x, scaled_y = self.scale_emitted_coordinates(x, y)
self.crosshairChanged.emit((scaled_x, scaled_y))
self.positionChanged.emit((x, y))
snap_x_vals, snap_y_vals = self.snap_to_data(x, y)
if snap_x_vals is None or snap_y_vals is None:
return
if all(v is None for v in snap_x_vals.values()) or all(
v is None for v in snap_y_vals.values()
):
return
precision = self._current_precision()
for item in self.items:
if isinstance(item, pg.PlotDataItem):
name = item.name() or str(id(item))
sx, sy = snap_x_vals[name], snap_y_vals[name]
if sx is None or sy is None:
continue
self.marker_moved_1d[name].setData([sx], [sy])
sx_s, sy_s = self.scale_emitted_coordinates(sx, sy)
self.coordinatesChanged1D.emit(
(name, round(sx_s, precision), round(sy_s, precision))
)
elif isinstance(item, pg.ImageItem):
name = item.objectName() or str(id(item))
px, py = snap_x_vals[name], snap_y_vals[name]
if px is None or py is None:
continue
# Respect image transforms
if isinstance(item, ImageItem) and item.image_transform is not None:
row, col = self._get_transformed_position(px, py, item.image_transform)
self.marker_2d_row.setPos(row)
self.marker_2d_col.setPos(col)
else:
self.marker_2d_row.setPos([0, py])
self.marker_2d_col.setPos([px, 0])
self.coordinatesChanged2D.emit((name, px, py))
def mouse_clicked(self, event):
"""Handles the mouse clicked event, updating the crosshair position and emitting signals.
Args:
event: The mouse clicked event
"""
# we only accept left mouse clicks
if event.button() != Qt.MouseButton.LeftButton:
return
self.update_markers()
if self.plot_item.vb.sceneBoundingRect().contains(event._scenePos):
mouse_point = self.plot_item.vb.mapSceneToView(event._scenePos)
x, y = mouse_point.x(), mouse_point.y()
scaled_x, scaled_y = self.scale_emitted_coordinates(mouse_point.x(), mouse_point.y())
self.crosshairClicked.emit((scaled_x, scaled_y))
self.positionClicked.emit((x, y))
x_snap_values, y_snap_values = self.snap_to_data(x, y)
if x_snap_values is None or y_snap_values is None:
return
if all(v is None for v in x_snap_values.values()) or all(
v is None for v in y_snap_values.values()
):
# not sure how we got here, but just to be safe...
return
precision = self._current_precision()
for item in self.items:
if isinstance(item, pg.PlotDataItem):
name = item.name() or str(id(item))
x, y = x_snap_values[name], y_snap_values[name]
if x is None or y is None:
continue
for marker_clicked in self.marker_clicked_1d[name]:
marker_clicked.setData([x], [y])
x_snapped_scaled, y_snapped_scaled = self.scale_emitted_coordinates(x, y)
coordinate_to_emit = (
name,
round(x_snapped_scaled, precision),
round(y_snapped_scaled, precision),
)
self.coordinatesClicked1D.emit(coordinate_to_emit)
elif isinstance(item, pg.ImageItem):
name = item.objectName() or str(id(item))
x, y = x_snap_values[name], y_snap_values[name]
if x is None or y is None:
continue
if isinstance(item, ImageItem) and item.image_transform is not None:
row, col = self._get_transformed_position(x, y, item.image_transform)
self.marker_2d_row.setPos(row)
self.marker_2d_col.setPos(col)
else:
self.marker_2d_row.setPos([0, y])
self.marker_2d_col.setPos([x, 0])
coordinate_to_emit = (name, x, y)
self.coordinatesClicked2D.emit(coordinate_to_emit)
else:
continue
def _get_transformed_position(
self, x: float, y: float, transform: QTransform
) -> tuple[QPointF, QPointF]:
"""
Maps the given x and y coordinates to the transformed position using the provided transform.
Args:
x (float): The x-coordinate to transform.
y (float): The y-coordinate to transform.
transform (QTransform): The transformation to apply.
"""
origin = transform.map(QPointF(0, 0))
row = transform.map(QPointF(0, y)) - origin
col = transform.map(QPointF(x, 0)) - origin
return row, col
def clear_markers(self):
"""Clears the markers from the plot."""
for marker in self.marker_moved_1d.values():
self.plot_item.removeItem(marker)
for markers in self.marker_clicked_1d.values():
for marker in markers:
self.plot_item.removeItem(marker)
self.marker_moved_1d.clear()
self.marker_clicked_1d.clear()
def scale_emitted_coordinates(self, x, y):
"""Scales the emitted coordinates if the axes are in log scale.
Args:
x (float): The x-coordinate
y (float): The y-coordinate
Returns:
tuple: The scaled x and y coordinates
"""
if self.is_log_x:
x = 10**x
if self.is_log_y:
y = 10**y
return x, y
def update_coord_label(self, pos: tuple):
"""Updates the coordinate label based on the crosshair position and axis scales.
Args:
pos (tuple): The (x, y) position of the crosshair.
"""
x, y = pos
x_scaled, y_scaled = self.scale_emitted_coordinates(x, y)
precision = self._current_precision()
text = f"({x_scaled:.{precision}f}, {y_scaled:.{precision}f})"
for item in self.items:
if isinstance(item, pg.ImageItem):
image = item.image
if image is None:
continue
if item.image_transform is not None:
inv_transform, _ = item.image_transform.inverted()
pt = inv_transform.map(QPointF(x, y))
px, py = pt.x(), pt.y()
else:
px, py = x, y
# Clip to valid pixel indices
ix = int(np.clip(px, 0, image.shape[0] - 1)) # column
iy = int(np.clip(py, 0, image.shape[1] - 1)) # row
intensity = image[ix, iy]
text += f"\nIntensity: {intensity:.{precision}f}"
break
# Update coordinate label
self.coord_label.setText(text)
self.coord_label.setPos(x, y)
self.coord_label.setVisible(True)
def check_log(self):
"""Checks if the x or y axis is in log scale and updates the internal state accordingly."""
self.is_log_x = self.plot_item.axes["bottom"]["item"].logMode
self.is_log_y = self.plot_item.axes["left"]["item"].logMode
self.clear_markers()
def check_derivatives(self):
"""Checks if the derivatives are enabled and updates the internal state accordingly."""
self.is_derivative = self.plot_item.ctrl.derivativeCheck.isChecked()
self.clear_markers()
@SafeSlot()
def reset(self):
"""Resets the crosshair to its initial state."""
if self.marker_2d_row is not None:
self.plot_item.removeItem(self.marker_2d_row)
self.marker_2d_row = None
if self.marker_2d_col is not None:
self.plot_item.removeItem(self.marker_2d_col)
self.marker_2d_col = None
self.clear_markers()
def cleanup(self):
self.reset()
self.plot_item.removeItem(self.v_line)
self.plot_item.removeItem(self.h_line)
self.plot_item.removeItem(self.coord_label)