Multi-File Apps¶

Create apps with multiple Python files using <panel-file> child elements, and declare dependencies with <panel-requirements>.

Defining multiple files¶

Use <panel-file> children to declare separate Python files. Mark one as entrypoint (or the first file is used by default):

<panel-live mode="editor">
  <panel-file name="app.py" entrypoint>
import panel as pn
from utils import greet
pn.panel(greet("World")).servable()
  </panel-file>
  <panel-file name="utils.py">
def greet(name):
    return f"Hello, {name}!"
  </panel-file>
</panel-live>

At runtime, non-entrypoint files are written to Pyodide's virtual filesystem before the entrypoint executes.

`<panel-file>` attributes¶

Attribute	Default	Description
`name`	`"app.py"`	Filename in the virtual filesystem
`entrypoint`	—	Boolean attribute; marks the file to execute
`src`	—	URL to fetch file content from

Loading files from URLs¶

Use the src attribute to load file content from external URLs:

<panel-live mode="editor">
  <panel-file name="app.py" src="https://example.com/app.py" entrypoint></panel-file>
  <panel-file name="utils.py" src="https://example.com/utils.py"></panel-file>
</panel-live>

Declaring requirements¶

Use <panel-requirements> to declare pip packages. One package per line or space-separated. Pip specifier format. Comments (#) and blank lines are stripped.

<panel-live mode="editor">
  <panel-file name="app.py" entrypoint>
import panel as pn
import pandas as pd
df = pd.DataFrame({"x": [1, 2, 3], "y": [4, 5, 6]})
pn.panel(df).servable()
  </panel-file>
  <panel-requirements>
pandas>=2.0
# hvplot for interactive plots
hvplot
  </panel-requirements>
</panel-live>

Requirements are installed via micropip before code execution, with progress shown in the status bar.

JavaScript API equivalent¶

const ctrl = await PanelLive.mount({
  mode: 'editor',
  files: {
    'app.py': 'import panel as pn\nfrom utils import greet\npn.panel(greet("World")).servable()',
    'utils.py': 'def greet(name): return f"Hello, {name}!"',
  },
  entrypoint: 'app.py',
  requirements: ['pandas', 'hvplot'],
}, '#container');