The serial terminal
===================

The pane along the bottom of the main window is the
serial terminal: everything the running script writes
to standard output -- every ``print()``, every warning,
every traceback -- streams into it live. It is the
script's voice during development, and reading it is
half of debugging. Open and close it from its button in
the bottom status bar; like the other panes, you can
drag it larger or collapse it entirely.

.. figure:: figures/serial-terminal.png
   :class: framed
   :alt: The serial terminal showing a script's FPS prints, the traceback that ended it, and the camera's banner

   The serial terminal: the script's prints, the
   traceback that ended the script, and the camera's
   banner after the stop.

The terminal keeps a deep scrollback (100,000 lines),
and its toolbar offers a filter box that narrows the
view to lines matching a search, a save button that
writes the whole buffer to a text file, and a wrap
toggle for long lines. The text zooms with
``Ctrl+scroll`` like the editor. Scroll up and the
auto-scroll pauses so you can read earlier output while
the script keeps printing; scroll back to the bottom
and it resumes.

The pane is output-only -- it displays what the camera
prints but does not accept typed input. For an
interactive REPL prompt on the camera, open a
:doc:`standalone terminal window <open-terminal>`
instead.

A debug-logging toggle in the toolbar interleaves the
camera debug protocol into the terminal, one
emoji-tagged line per message, at a few levels of
detail. It is a diagnostic aid for a connection that
misbehaves -- a camera that will not connect, or a
stream that stalls -- and off by default; leave it off
for normal work, where it only adds noise.

Tracebacks are wired into the editor. When a script
dies with an unhandled exception, the IDE parses the
traceback as it prints, jumps the editor to the
offending line, marks it in red with the error text
beside it, and -- when the failing file is a module on
the camera's drive rather than the open script -- opens
that file at the failing line. You go straight from
"it crashed" to the line that crashed.