esrf-statusgui 0.3.1

Get the post-treatment progress of any ESRF-ID11 experiment. Option to launch the post-treatment from the GUI.

ESRF ID11 Status GUI

ESRF ID11 Status GUI is an ipywidgets-based control panel that monitors and orchestrates the post-treatment pipeline for ID11 diffraction and tomography experiments. It understands the ESRF data policy layout (default /data/visitor/<proposal>/<beamline>/<date>/RAW_DATA|PROCESSED_DATA) via esrf-pathlib and surfaces the progress of DCT, PCT, FF, and s3DXRD processing steps directly inside Jupyter notebooks.

Key Capabilities

• Navigate experiments by proposal, beamline, and acquisition date with live folder trees.
• Detect datasets automatically via status.json rules and inspect their processing state with color-coded accordions.
• Launch and track post-processing (DCT, FF, SFF) from the embedded toolbar, including notebook generation helpers.
• Preview HDF5/MAT content and diagnostic messages inline without leaving the notebook.
• Utilities to scaffold new experiment dates, mirror RAW/PROCESSED layouts, and manage ESRF-specific permissions.

Requirements

• Python 3.9 or newer.
• Jupyter Notebook/Lab (IPython front-end is mandatory for the widgets) with ipywidgets, matplotlib, and related GUI extras. Install with pip install -e ".[gui]" if needed.
• MySQL client libraries available (for the bundled mysqlclient) so StatusGUI can probe DCT databases; override host/user/password via STATUSGUI_SQL_*, standard MYSQL_* vars, or a ~/.my.cnf client block.
• Access to the ESRF storage tree (default root /data/visitor, configurable via ESRF_VISITOR_ROOT); the GUI relies on esrf-pathlib to resolve proposal, beamline, and session metadata.
• Optional domain tools: ImageD11 (FF/SFF notebooks). For DCT processing, start MATLAB via the facility module/container (e.g. module load dct), add the DCT code to the MATLAB path as usual, and share the session with matlab.engine.shareEngine so StatusGUI can attach to it.
• ESRF file loader: esrf-loadfile is pulled from PyPI; the package also ships a fallback loader for environments without it.
• DCT Python package: the optional dct dependency (used by the [extra] extra) is not published on PyPI yet. Installing pip install "esrf-statusgui[extra]" from PyPI will fail until it is available; install the ESRF dct toolbox separately via the facility module/Git checkout if needed.

Database checks default to the graindb.esrf.fr service (gtadmin user) used by the MATLAB tools and will transparently use mysqlclient or any available SQLAlchemy MySQL driver.

Installation

From PyPI (recommended once published):

pip install "esrf-statusgui[gui]"        # notebook-ready runtime
pip install "esrf-statusgui[extra]"      # science stack (ImageD11, DCT, Dans-Diffraction)
pip install "esrf-statusgui[full]"       # contributor setup without the science extras
pip install "esrf-statusgui[dev]"        # lint/format/test toolchain only

Mamba environment (conda)

module load mamba  # on ESRF systems
mamba create -n statusgui -c conda-forge python=3.12 pip setuptools
mamba activate statusgui
pip install "esrf-statusgui[gui]"

From source (editable) for development:

git clone https://gitlab.esrf.fr/graintracking/statusgui.git
cd statusgui
python -m venv .venv
source .venv/bin/activate  # Use .venv\Scripts\activate on Windows
pip install -e .

Editable extras mirror the PyPI variants: pip install -e ".[gui]", .[extra], .[full], and .[dev]. Both statusGUI and statusgui console scripts point to the same launcher entrypoint.

Quick Start (Jupyter)

1. Launch JupyterLab/Notebook on an ID11 machine with access to the ESRF storage.

2. In a new notebook cell:

   from esrf_statusgui.visualization.statusGUI import DatasetSelectionTab
   
   tab = DatasetSelectionTab()
   tab.display()
   

3. Pick an experiment (proposal) and beamline; the GUI builds the RAW/PROCESSED tree and populates four columns (Tomography, DCT, FF, s3DXRD) with datasets.

4. Expand a dataset to inspect processing steps, open generated notebooks, or trigger refreshes.

Working with alternative data roots

By default the browser targets /data/visitor. To use another data mirror, set the visitor root before instantiating the GUI:

from esrf_statusgui.file_utils.paths import set_visitor_root

set_visitor_root("/mnt/id11_archive")

from esrf_statusgui.visualization.statusGUI import DatasetSelectionTab

tab = DatasetSelectionTab()
tab.display()

You can also export ESRF_VISITOR_ROOT=/mnt/id11_archive before launching Jupyter. The helper esrf_statusgui.file_utils.paths.describe() returns structured metadata (proposal, beamline, session date) for any path resolved by the GUI. Once the widget is running you can also point it to a different mirror via the Data root field (Use path button) at the top of the dashboard; this updates the session root and repopulates the experiment/beamline lists. Mount-specific prefixes such as /gpfs/easy or /mnt/storage are normalised automatically via file_utils.paths.clean_dir_name, so visitor metadata resolves even when paths contain dots or dashes.

Running outside notebooks

python -m src.main instantiates the widget but expects an IPython kernel. When invoked from a plain terminal it prints a helpful message; prefer launching inside Jupyter or wrap it with Voila for browser delivery.

Automating DCT preprocessing

pySetupH5.setup_pre_processing accepts dicts (or a list of dicts) to seed RAW paths, group IDs, and destination names, surfacing missing datasets as inline errors instead of failing silently:

from esrf_statusgui.visualization.DCT.pySetupH5 import pySetupH5

setup = pySetupH5()
setup.setup_pre_processing(
    {
        "raw_path": "/gpfs/easy/data/visitor/ma1234/id11/20240101/RAW_DATA/dataset.h5",
        "dataset_name": "dataset_clean",
        "projections_group": 5,
        "flat_groups": [1, 3],
        "dark_group": 4,
    }
)

Flat/dark/projection groups default to 1-3/4/2 when omitted. The call returns the created dataset names and destination folders.

Understanding the Status Columns

• Tomography (PCT) – checks HDF5 descriptors, beamline metadata, and processed reconstructions.
• DCT – validates parameter/DB files, pre-processing, segmentation, indexing, forward simulation, and reconstruction steps.
• FF / s3DXRD – reports on far-field 3DXRD and s3DXRD pipelines, including availability of target files and optional visualisations.
• Post-processing toolbar – the Post_process widget adds dataset-level refresh, quick notebook launchers (createProcessingNotebook.py), and optional data copies (gtMoveData).

Color coding: red = not processed, orange = in progress/partial, green = complete. Tooltips and accordion bodies list the files checked for each component.

Utilities & Supporting Modules

• esrf_statusgui.brain.* – discovery logic, caching, dataset classification (DatasetSelectionLogic, SampleManager, Post_process).
• esrf_statusgui.exp_methods.* – per-technique status evaluators driven by status.json.
• esrf_statusgui.file_utils.* – helpers for building experiment directories, managing permissions, creating companion notebooks, and normalising paths (see file_utils.paths.clean_dir_name for GPFS/mnt sanitisation).
• src/main.py – lightweight entry point that instantiates the main widget.
• src/notebooks/StatusGUI.ipynb – demonstration notebook with a pre-wired dashboard.

Development Workflow

pip install -e ".[dev]"
pytest
pytest --maxfail=1 --disable-warnings -q  # focused run
ruff check src tests
black src tests

Tests exercise widget factories and CLI wiring. The GUI relies on Jupyter, so widget rendering tests use ipywidgets stubs. When adding new experiment methods, extend status.json, provide a status class in exp_methods, and register it in DatasetSelectionTab. See docs/development_checks.md for the full toolchain and pre-commit workflow.

Project Layout

• src/esrf_statusgui/visualization/ – ipywidgets components for dataset browsing and post-processing controls.
• src/esrf_statusgui/brain/ – core logic (filesystem traversal, dataset inference, status refresh).
• src/esrf_statusgui/exp_methods/ – domain-specific status objects for tomography, DCT, FF, and s3DXRD.
• src/esrf_statusgui/data_managment/ – loaders for HDF5/MAT files and parameter parsing.
• src/esrf_statusgui/file_utils/ – ESRF filesystem helpers, experiment scaffolding, and notebook generators.
• tests/ – pytest suite covering widget utilities and entry points.

Notebook snapshots & dependency versions

The notebooks committed under src/esrf_statusgui/notebooks/ (DCT tutorials as well as ImageD11 S3DXRD/TDXRD flows) are checked in as snapshots that match the graintracking/dct==0.2.0 pipeline used by this release and the ImageD11==2.1.2 stack (see the pinned dependency in pyproject.toml). When executing them, make sure your environment uses those versions. If you upgrade either dependency, regenerate the notebooks (for example via src/esrf_statusgui/file_utils/createProcessingNotebook.py) and update this note to reflect the new baseline.

Troubleshooting

• Empty experiment list – ensure the Jupyter server runs on a host that can reach the configured visitor root or override it via set_visitor_root.
• Status objects missing files – verify RAW/PROCESSED samples follow the ESRF naming convention; detection keys live in exp_methods/status.json.
• DCT helpers report missing MATLAB engine – install MATLAB R2023 engine bindings, start MATLAB via module load dct, add the DCT module to the MATLAB path, and call matlab.engine.shareEngine so the Python side can attach to the existing session.
• Widgets refuse to render – confirm ipywidgets extension is enabled (jupyter nbextension enable --py widgetsnbextension for classic notebook).

Known Limitations

• DCT parameter wizards still rely on manual tweaks for detector metadata, diffrz calibration, and session-date heuristics (see inline TODOs in tab_dct_parameters.py).
• Pair-matching status reports do not yet parse legacy pairmatch.mat / DifspotTable.mat outputs (see dct_status.py).

Contributing & Support

We welcome merge requests and issue reports. Please read CONTRIBUTING.md for workflow guidelines. For ESRF-internal support open a ticket in the grain tracking GitLab project or contact the ID11 software team.

License

Distributed under the terms of the MIT License. See LICENSE.md for details.