[petercorke]

robot.plot(q) of a trajectory in a notebook with %matplotlib ipympl showed a blank figure, a long pause, and then the final frame. It now animates, whatever %matplotlib says, with no argument needed.

Cause

An ipympl figure updates only when the browser sends a draw message and the kernel handles it. While a cell is running (the animation loop in plot()) the kernel cannot, so only the first and the last frames were ever drawn. The render mode that was chosen for ipympl was notebook-widget for every plot. The code of the widget branch of step() has not changed since the render modes were written (2026-05-29); this is a limitation of the widget mode, not a recent regression.

Measured in JupyterLite 4.6.4 and ipympl 0.10 with real Chromium (Playwright), sampling the output every 0.5 s during a 100-frame trajectory: 1 distinct image while the cell ran with the old code, and the same with a synchronous canvas.draw(), with flush_events(), and with plt.pause(), and with ipykernel 6.29.5, 7.2.0 and 7.4.0. With the existing notebook-inline render mode: 13 of 13 samples different.

What changes

• plot() passes animation=True to the backend when q has more than one row (only to backends with a render mode, so not Swift).
• _resolve_render_mode(render_mode, animation) chooses notebook-inline for an animation under ipympl. A single pose is still notebook-widget, an interactive figure; the inline backend is unchanged; an explicit render_mode= is always used as given. The docstrings say so.
• PyPlot2 had its own copy of _resolve_render_mode and _isnotebook; it now imports the ones in PyPlot, so the 2D backend gets the same choice.

Checked in the browser with the fix and no render_mode: %matplotlib ipympl with a trajectory animates (13 of 13), %matplotlib inline with a trajectory animates (13 of 13), and %matplotlib ipympl with one pose is still a live widget canvas.

tests/test_render_mode.py (14 tests) covers the choice for each backend and animation value, an explicit mode, the 2D backend sharing the function, and what plot() passes for a trajectory, a single pose, and an explicit mode. Full suite: only the 20 test_servo_half_turn failures from the stale compiled extension in my worktree; mypy clean. tests/test_notebook_animation.py (4 tests) runs a real IPython kernel and checks, from the messages it sends, that a trajectory plotted under %matplotlib ipympl and under %matplotlib inline pushes a different frame for every step while the cell runs, that one pose under ipympl is still an interactive widget, and that an explicit render_mode is honoured. It fails without the fix (assert 0 >= 6 frames for ipympl) and takes about 6 s. It does not include a browser. ipympl is added to the dev extra for it; the test skips if ipykernel or ipympl is missing.

Not tried in VS Code's notebook UI, only in JupyterLab. Written with AI assistance (Claude Code).

🤖 Generated with Claude Code

[codacy-production]

Not up to standards ⛔

🔴 Issues 1 critical · 19 high · 1 medium

Alerts:
⚠ 21 issues (≤ 0 issues of at least minor severity)

Results:
21 new issues

Category	Results
Security	19 high 1 critical 1 medium

View in Codacy

🟢 Metrics 23 complexity · -6 duplication

Metric	Results
Complexity	23
Duplication	-6

View in Codacy

_{NEW Get contextual insights on your PRs based on Codacy's metrics, along with PR and Jira context, without leaving GitHub. Enable AI reviewer
TIP This summary will be updated as you push new changes.}

[codecov]

Codecov Report

❌ Patch coverage is 0% with 8 lines in your changes missing coverage. Please review.
✅ Project coverage is 0.00%. Comparing base (742b6ed) to head (8306713).
⚠️ Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
src/roboticstoolbox/backends/PyPlot/PyPlot.py	0.00%	3 Missing ⚠️
src/roboticstoolbox/robot/BaseRobot.py	0.00%	3 Missing ⚠️
src/roboticstoolbox/backends/PyPlot/PyPlot2.py	0.00%	2 Missing ⚠️

Additional details and impacted files

@@          Coverage Diff          @@
##            main    #860   +/-   ##
=====================================
  Coverage   0.00%   0.00%           
=====================================
  Files        146     146           
  Lines      16867   16846   -21     
=====================================
+ Misses     16867   16846   -21     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.