GX-TXT: a workbench for measurement data

Version
0.9.8.10-dev-111
Status
In development
Updated on
October 28, 2026
Licence
GPL-3.0-or-later
Platforms
Linux (Windows coming)
Requirements
Tcl/Tk (wish), Python 3, GNU Octave; NumPy, SciPy, Matplotlib depending on modules

DownloadSource code

GX-TXT is the workbench I use to keep together everything that happens to a measurement after I’ve taken it: the instrument capture, the processing, the plots, the notes. It started as a cleanup of the scripts from the IR remote control project, so I would stop losing data and stop rewriting plotting code every time, and it grew into a framework. It is still in beta: I’m listing it now because the source is already on GitHub and I’m using it in my articles.

The presentation video

What it is for

The point is to manage the data of an experiment in a single environment. Oscilloscope CSV captures, spectra, tables, audio and images go into a Session as Datasets with stable identifiers, units, metadata and provenance. From there you compare them, analyse them (statistics, spectra, curve fitting), hand them to Python or GNU Octave scripts and bring them onto 2D and 3D plots. Whatever you derive stays linked to the data it came from, and the original data is never modified.

Specialised work is done by dedicated viewers: curves (including the Curve Digitizer and the Curve Workbench), signal and statistical analysis, spectroscopy, documents, 3D, circuits and electromagnetics. You can build your own, with the Interface Builder for the interface and an integrated C++ environment, with its own debugger, to program them. Heavier calculations can run on remote compute nodes, and the results come back into the Session with their provenance.

Development status

  • It is a development snapshot of the 0.9.8.10-dev series, not a finished release: some parts are complete, others (the 3D viewer, for instance) are still at an early stage.
  • It is distributed as source, with no installer: the dependencies (Tcl/Tk, Python 3, GNU Octave and the libraries of the modules you enable) must be installed separately.
  • The native components are built for Linux; Windows will come later.
  • Before using it on real data, try it on a test Session.

The current status, with the notes for each version, is on GitHub.

Trying it

On a system with Tcl/Tk and the scientific runtimes installed, from the source folder:

wish gxtxt_pipeline.tcl

Create or open a Session, import data from Acquisition, inspect it under Datasets and carry on with modules, scripts and viewers.

Where to find it

Documentation and guides

The manual, the guides and the articles on GX-TXT are split by area. Each area has its own page, with the manual guides and the articles that cover a single aspect; all the articles are also in the GX-TXT category. For a first look there is the introductory article GX-TXT: Your Experiment and Your Project in One Integrated Environment.

Main window

Session, Acquisition, Datasets and the other parts of the main interface.

Analysis modules

Statistics, spectra, curve fitting and the other modules that work on Datasets.

Viewers

One section per viewer, with its manual and its articles. The viewers documented so far:

Tutorials and applications for each viewer are on the Viewers page.

Scripting with GNU Octave and Python

The GX-TXT Octave Script Workspace – Complete Guide and the GX-TXT Python Script Workspace – Complete Guide to the editors, scripts and debuggers, the GNU Octave course (16 lessons, first edition completed), the Scientific Python in GX-TXT course (60 lessons, run in GX-TXT), the Script Debugger for Octave and Python, and the articles on the APIs, such as the Temporal Set API.

Viewer development

Interface Builder and the integrated C++ environment, with its debugger, for building your own viewers.

Topic guides

Subjects that run across interface, modules and viewers, covered in a single guide.