Data storage

HDF5 data management: subscription, scheduled/manual saving and history read-back, with integrated plotting panels.

Overview

quatm/GUI/h5storage/ provides the HDF5 data storage GUI (a package split out of the former single file, with its public symbols re-exported at package level): three kinds of data manager — aggregate, single-stream and image — subscribe to data and save it automatically at a scheduled time, periodically, or on a Looper command; the current snapshot can also be written to an .h5 file in the background, and historical measurements read back from a file.

Components

Component	Module	Description
`H5StorageGui`	`main_window.py`	Main window: data tabs, save/scheduled-save controls and plot management
`H5StorageDataMgr`	`data_manager.py`	Host for the three data managers: subscription management, save/read-back, clearing
`H5SaveWorker`	`workers.py`	Worker thread that writes the `.h5` file, keeping disk I/O off the Qt event loop
`H5DataPlotter` / `EventPlotter`	`plotters.py`	Plotting tabs for measurement data / event data
`XAxisWidget` / `YAxisWidget`	`axes.py`	Data selection and filter-chain configuration for the x axis and each y axis
`BinningWidget` / `FilterWidget` / `FittingWidget`	`widgets.py` / `fitting.py`	binning, filtering and ODR fitting processing chain

Saving modes (H5StorageDataMgr)

Method	Description
`toggle_scheduled_save`	Toggles saving at a scheduled time (saves automatically and notifies when the time arrives)
`toggle_regular_save`	Toggles periodic automatic saving every N hours
`scheduled_save_by_looper`	Responds to save commands from the Looper (final / intermittent)
`savetoFile`	Writes the current data snapshot to an `.h5` file in the background
`loadFromFile`	Loads historical measurements from an existing `.h5` file and refreshes the interface
`selectDatadir`	Chooses the target h5 file manually (to avoid falling back to the daily default name)
`clear`	Clears the entire cache of the three data managers

Usage

from quatm.GUI.h5storage import H5StorageGui   # package-level re-export

win = H5StorageGui(parent=parent)
win.show()

Data saving is performed by the background H5SaveWorker: the GUI thread copies the data snapshot first and then hands it to the worker, so writing to HDF5 on disk never blocks the interface.