wxPython面板嵌入Matplotlib图表显示异常：无法调整大小与定位

Got it, let's work through this problem you're having—where resizing just crops your Matplotlib charts instead of scaling them, and that right-side chart is way too big. I've run into this exact issue before, so let's break down why it's happening and how to fix it.

The Root of the Problem #

Most of the time, this happens because:

• You're not using wx's sizers to manage layout properly (so your chart containers don't resize with the window)
• Your Matplotlib figure isn't listening to the container's size changes, so it just stays fixed while the container gets clipped
• The figure's default size/dpi isn't synced with wx's display settings

Step-by-Step Solution #

Here's a working example that fixes both the resizing and placement issues. I'll walk through the key parts:

import wx
import matplotlib.pyplot as plt
from matplotlib.backends.backend_wxagg import FigureCanvasWxAgg as FigureCanvas

class ChartPanel(wx.Panel):
    def __init__(self, parent, title):
        super().__init__(parent)
        self.title = title
        
        # Create Matplotlib figure and axis, sync with display DPI
        self.figure = plt.Figure(dpi=wx.GetDisplayPPI()[0])
        self.ax = self.figure.add_subplot(111)
        self.canvas = FigureCanvas(self, -1, self.figure)
        
        # Use sizer to make canvas fill the entire panel
        self.sizer = wx.BoxSizer(wx.VERTICAL)
        self.sizer.Add(self.canvas, 1, wx.EXPAND)
        self.SetSizer(self.sizer)
        
        # Bind resize event to update figure dynamically
        self.Bind(wx.EVT_SIZE, self.on_size)
        
        # Add sample data to test
        self.plot_sample_data()

    def plot_sample_data(self):
        self.ax.clear()
        self.ax.plot([1,2,3,4,5], [2,4,1,5,3])
        self.ax.set_title(self.title)
        self.canvas.draw()

    def on_size(self, event):
        # Skip tiny size changes to avoid errors
        size = self.GetClientSize()
        if size.width < 10 or size.height < 10:
            return
        
        # Convert wx pixel size to Matplotlib's inch-based figsize
        dpi = self.figure.dpi
        fig_width = size.width / dpi
        fig_height = size.height / dpi
        
        # Update figure size and redraw to scale properly
        self.figure.set_size_inches(fig_width, fig_height)
        self.canvas.draw()
        event.Skip()

class MainFrame(wx.Frame):
    def __init__(self):
        super().__init__(None, title="Resizable Matplotlib Charts", size=(800, 500))
        
        # Main panel with horizontal sizer for side-by-side charts
        main_panel = wx.Panel(self)
        main_sizer = wx.BoxSizer(wx.HORIZONTAL)
        
        # Add two chart panels with proportional sizing (1:1 ratio here)
        left_chart = ChartPanel(main_panel, "Left Chart")
        right_chart = ChartPanel(main_panel, "Right Chart")
        main_sizer.Add(left_chart, 1, wx.EXPAND | wx.ALL, 5)
        main_sizer.Add(right_chart, 1, wx.EXPAND | wx.ALL, 5)
        
        main_panel.SetSizer(main_sizer)
        self.Show()

if __name__ == "__main__":
    app = wx.App()
    frame = MainFrame()
    app.MainLoop()

Key Fixes Explained #

• Sizer Management: We use wx.BoxSizer for both the main panel and each chart's panel. The 1 in Add(..., 1, wx.EXPAND) means each chart takes equal space, and wx.EXPAND forces them to fill their allocated area. Want one chart bigger? Just change the number (e.g., 2 for the left chart makes it twice as wide).
• Resize Event Handling: The on_size method converts the wx panel's pixel size to Matplotlib's inch-based figsize (using the system's display DPI), updates the figure, and redraws it. This ensures the chart scales with the panel instead of getting clipped.
• DPI Sync: Using wx.GetDisplayPPI() matches the figure's DPI to your system's display, so scaling looks crisp and avoids weird sizing mismatches.

What This Does #

• When you resize the window, both charts scale smoothly (no more clipping)
• The charts stay proportional to each other (adjust the ratio in the sizer if needed)
• You can drag the window to any size, and the charts will adapt perfectly

内容的提问来源于stack exchange，提问作者John Doe