Developer Quick Start

If you've never used PyRS before, you can get started quickly by doing the following.

1. Install basic dependencies: Conda, Python, and PyQt
2. Create a new Conda environment with additional dependencies:

$ conda create -n pyrs -c mantid -c mantid/label/nightly mantid-workbench -c conda-forge

3. Activate the conda environment

$ conda activate pyrs

4. From the PyRS directory, run the setup script in developer mode

$ python setup.py develop

5. From the PyRS directory, start the user interface

$ python scripts/pyrsplot

Data reduction workflow

The workflow is as follows. The word "script" is used to denote a distict step in processing the data rather than actual script.

1. Start with a nexus file. As in, /HFIR/HB2B/IPTS-22731/nexus/HB2B_439.nxs.h5
2. This is read in by something that will split into sub-scans and create a project file [script 1 - currently convert_nexus_to_hidra.py]
3. This is read in by something that will create powder patterns and add them to the project file [script 2 - currently reduce_HB2B.py]
4. This is read in by the gui to do the peak fitting and add the results to the project file [script 3 - currently somewhere in the gui]
5. This is read in by the gui and used to create the summary .csv file [script 4 - currently being written]

Some other things to note * scripts 1-2 will be refactored/combined to create a new reduce_HB2B.py that can be run by autoreduction * scripts 2-4 (more like functionality 2-4) will all live in the gui in some way * going back any number of steps will delete the following steps from the project file. This is to prevent users from having a project file with powder patterns that are not associated with the peak fitting from the project file.

Running and developing PyRS

To start main window from analysis machine

$ PYTHONPATH=$PWD:$PYTHONPATH QT_API=pyqt python scripts/pyrsplot

To develop

$ pyrsdev.sh

To run the tests

$ pyrstest.sh