[BUG] dcc.Dropdown: virtualized list keeps the height measured at opening when options grow while the menu is open

Offen
#4,131 1 Kommentar 0 Reaktionen 1 zugewiesene Person Auf GitHub ansehen

Maintainer antworten meist innerhalb von 1 Tag

@KoolADE85 arbeitet bereits daran.

Seit 09.11.2026.

Bewertung

Dieses Issue wurde noch nicht bewertet.

Beschreibung

bug P3 size: 1

Describe your context

dash                 4.5.0

Reproduced in headless Chromium (Playwright); first seen by our users on dash 4.4.1.

Describe the bug

A multi dcc.Dropdown whose options are updated by a callback while its menu is open keeps the list height it measured when the menu opened. If the menu opens while the options are few (here 1) and the callback then brings back many (here 300), the virtualized list stays one row tall (34 px) and the user has to scroll inside a single row. Closing and opening the menu again shows it at the normal height.

This happens with cascading filters: an item is chosen, the callback narrows the options to it, the user clears it with the × and opens the menu again before the callback has returned all the options.

The cause seems to be in OptionsList (components/dash-core-components/src/utils/optionRendering.tsx): the container is measured once, in a useLayoutEffect with an empty dependency array, and listHeight = Math.min(measuredHeight, options.length * defaultOptionHeight). With one option the measured height is one row, and it is never measured again when options grows.

Minimal example

import time
from dash import Dash, Input, Output, dcc, html

ALL = [f"item-{i:03d}" for i in range(300)]

app = Dash(__name__)
app.layout = html.Div(
    [dcc.Dropdown(id="dd", options=ALL, multi=True, placeholder="Pick an item")],
    style={"width": 400, "padding": 40},
)


@app.callback(Output("dd", "options"), Input("dd", "value"))
def narrow(value):
    # like a cascading filter: once an item is chosen only it is offered;
    # cleared, all 300 come back (the server takes a moment)
    time.sleep(0.5)
    return value if value else ALL


if __name__ == "__main__":
    app.run(port=8077, debug=False)

Steps:

  1. open the menu, pick item-007, close the menu (the options become ["item-007"]);
  2. click the × (Clear selection) and open the menu at once, before the callback returns;
  3. wait for the 300 options to come back.

Measured with Playwright on .dash-dropdown-content .dash-options-list-virtualized:

State list height content height
menu opened with 300 options 124 px 10494 px
step 3: menu opened with 1 option, 300 options back 34 px 10500 px
menu closed and opened again 124 px 10494 px

Expected behavior

The list height follows the options: when options changes while the menu is open, the list is measured again (for example the layout effect depending on options.length, or measuredHeight reset when the options change), so 300 options show at the normal height without closing and reopening the menu.

We work around it client-side by closing and reopening a menu whose virtualized list is shorter than three rows while its content is taller, which is what users did by hand.

Image Image
Vorherrschende Sprache
Python
Sterne
24.5k
Forks
2.3k
Ø Merge
3 T. 7 Std.
Gemergte PRs (30 T.)
30

Entwicklungsumgebung

Erste Schritte

  1. Lesen Sie das ganze Issue und danach den Beitragsleitfaden des Projekts.
  2. Schreiben Sie ins Issue, dass Sie es übernehmen — das erspart doppelte Arbeit.
  3. Forken Sie das Repository und arbeiten Sie in einem Branch.
  4. Öffnen Sie einen Pull Request, der die Issue-Nummer nennt.

Mehr aus plotly/dash

Alle Issues in plotly/dash

Ähnliche Issues

Weitere Issues zu Python

Neue Issues direkt in Ihr Postfach

Eine kurze Übersicht über anfängerfreundliche GitHub-Issues.