lemke_handClaude Sonnet 5 6acf6bd402 Add 'respawn' subcommand: CLI equivalent of the GUI's remote restart
restart/stop/start only ever manage a process on the local host
(pgrep/systemctl are inherently local) -- pointing $URL at a server
running elsewhere, as a different account, does nothing useful there:
it just finds no local process and starts a brand-new one right here.

respawn instead POSTs /admin/restart to $URL over HTTP, same as the
GUI's "Restart process" button (client.reinit(mode="restart") ->
client.restart()) -- the server re-execs *itself* (os.execv, same
PID/user/host it already runs as), so this correctly reaches the real
process regardless of where the CLI is invoked from. -w polls /health
afterward until it reports ready with a new instance_id, mirroring
StatusServerClient.restart(wait=True)'s own timing.

Verified against a local fake HTTP server (no-wait, --wait, and the
connection-refused path) -- not against the real one, deliberately.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
2026-09-16 08:14:51 +02:00
2026-09-11 18:25:50 +02:00
2026-08-18 21:56:17 +02:00
2026-08-28 18:20:00 +02:00
2026-09-16 07:50:53 +02:00
2026-09-09 17:14:38 +02:00
2026-08-18 21:56:17 +02:00
2018-07-06 12:38:44 +02:00
2026-08-18 21:56:17 +02:00
2026-08-18 21:56:17 +02:00
2026-08-18 21:56:17 +02:00
2026-09-10 19:31:54 +02:00
2021-01-13 13:11:22 +01:00
2026-08-18 21:56:17 +02:00
2026-08-21 15:10:25 +02:00
2025-06-24 15:33:02 +02:00

eco — Experiment Control

                       ___ _______
                      / -_) __/ _ \
 Experiment Control   \__/\__/\___/

eco is a Python-based control environment for experiments, developed and used at SwissFEL, PSI. It is used both as:

  • a library of experimental devices for higher-level Python applications or GUIs, and
  • an interactive command-line interface, e.g. from an IPython/Jupyter shell or notebook.

eco follows an object-oriented approach: every device is represented as a Python object with a small, predictable interface, so devices can be freely combined in generic control/acquisition routines and analysed with the scientific Python ecosystem. For a general introduction to object-oriented Python, see e.g. this short introduction.

Documentation

The full documentation — installation, core concepts, and worked examples (listening monitors, archiver data and strip charts, pipeline offload, motor configuration) — lives in docs/ and is built with Sphinx, configured to build on Read the Docs via .readthedocs.yaml.

Build it locally:

pip install -r docs/requirements.txt
sphinx-build -b html docs docs/_build/html

Installation

conda install -c paulscherrerinstitute eco

or, for development, in editable mode from a checkout:

git clone https://github.com/paulscherrerinstitute/eco.git
cd eco
pip install -e .

See Installation for beamline-specific setup (the eco launcher, .ecorc defaults) and the full dependency picture.

Creating a new device

New devices are implemented as a subclass of Assembly, which provides naming, aliasing, and shell representation:

from eco.elements.assembly import Assembly

class MyDevice(Assembly):
    def __init__(self, name=None):
        super().__init__(name=name)
        self._append(MySubObject, name="my_sub_object", is_setting=True, is_status=True)

is_setting=True marks the child as a setting of the assembly (shown by .settings() and captured when settings are saved); is_status=True marks it as contributing to the assembly's .status(). See Representing real devices — the Assembly in the full docs for the rest of the model (Adjustable, Detector, Namespace) and a from-scratch, runnable example of each.

S
Description
Experimental control package
Readme GPL-3.0
8.7 MiB
Languages
Python 86.7%
HTML 10.4%
Jupyter Notebook 2%
Shell 0.8%