基于标准Python脚本实现弹出式蜡烛图：Jupyter代码改造方案咨询

Got it, let's fix this up. Your original code works great in Jupyter, but to get a pop-up version, we need to adjust how Plotly renders the chart, plus fix a small issue with the outdated data source (Morningstar no longer works with pandas_datareader). Here are two straightforward solutions:

Solution 1: Pop-Up Browser Window (Easiest, Full Interactivity) #

This keeps all of Plotly's awesome interactive features (zoom, hover data, etc.) and just opens the chart in a separate browser window instead of embedding it in Jupyter.

Modified Code #

import plotly.graph_objs as go
import plotly.offline as py_offline
import yfinance as yf  # Replaces broken pandas_datareader Morningstar
from datetime import datetime

# Grab AAPL stock data (2020-2024 as an example)
start = datetime(2020, 1, 1)
end = datetime(2024, 1, 1)
df = yf.download("AAPL", start=start, end=end)

# Build the candlestick trace
candlestick = go.Candlestick(
    x=df.index,
    open=df['Open'],
    high=df['High'],
    low=df['Low'],
    close=df['Close'],
    name='AAPL Candlesticks'
)

# Add some clean layout tweaks (optional but makes it look better)
layout = go.Layout(
    title='AAPL Stock Candlestick Chart',
    xaxis_title='Date',
    yaxis_title='Price (USD)',
    hovermode='x unified'  # Shows all data points for a given date on hover
)

fig = go.Figure(data=[candlestick], layout=layout)

# This is the key change: opens the chart in a new browser window
py_offline.show(fig, filename='aapl_candlestick_popup.html')

Key Changes: #

• Replaced the broken data source: Morningstar stopped working with pandas_datareader, so we're using yfinance instead (install it first with pip install yfinance).
• Swapped iplot() for py_offline.show(): iplot() is for Jupyter embedding, while show() generates an HTML file and pops it open in your default browser.
• Dropped init_notebook_mode(): You don't need this for standalone scripts—it's only for Jupyter.
• Added layout improvements: Makes the chart easier to read at a glance.

Solution 2: Native Desktop Pop-Up (No Browser Required) #

If you want a true desktop app-style pop-up (not just a browser window), we can use Dash (Plotly's web framework) paired with a tiny Tkinter trigger to launch it as a "windowed" app.

Code Implementation #

import dash
from dash import dcc, html
import plotly.graph_objs as go
import yfinance as yf
from datetime import datetime
import tkinter as tk
import webbrowser
from threading import Timer

# Get stock data same as before
start = datetime(2020, 1, 1)
end = datetime(2024, 1, 1)
df = yf.download("AAPL", start=start, end=end)

# Build the candlestick figure
candlestick = go.Candlestick(
    x=df.index,
    open=df['Open'],
    high=df['High'],
    low=df['Low'],
    close=df['Close'],
    name='AAPL Candlesticks'
)
fig = go.Figure(data=[candlestick])
fig.update_layout(title='AAPL Stock Candlestick Chart', xaxis_title='Date', yaxis_title='Price (USD)')

# Set up a minimal Dash app
app = dash.Dash(__name__)
app.layout = html.Div([dcc.Graph(figure=fig)])

# Function to trigger the pop-up
def launch_popup():
    # Tiny Tkinter window to trigger the browser (we'll close it automatically)
    root = tk.Tk()
    root.title("Candlestick Chart")
    root.geometry("100x50")
    
    # Wait 2 seconds then open the Dash local server in a browser
    Timer(2.0, lambda: webbrowser.open('http://127.0.0.1:8050')).start()
    
    # Close the tiny Tkinter window after 3 seconds
    root.after(3000, root.destroy)
    root.mainloop()

if __name__ == '__main__':
    # Start Dash server in the background
    Timer(1.0, app.run_server, kwargs={'debug': False, 'use_reloader': False}).start()
    # Launch the pop-up trigger
    launch_popup()

Notes: #

• Install dependencies first: pip install dash yfinance
• This starts a local Dash server (runs in the background) and opens it in your browser as a separate window. If you want a fully native desktop window (no browser), you'd need to use something like PyQt to embed the Plotly chart, but this is a great balance of simplicity and "native" feel.

Quick Troubleshooting #

• Data source issues: If yfinance isn't working, try using the Stooq data source with pandas_datareader:

  import pandas_datareader.data as web
  df = web.DataReader("AAPL", 'stooq')
  

• Pop-up blocked: Make sure your browser allows pop-ups from local files/localhost.

内容的提问来源于stack exchange，提问作者Jeremie