Simple GUI to run the msmapper code on Beamline I16
The Miller Space Mapper (msmapper) program converts x-ray diffraction scans with area detectors into reciprocal space units. msmapper is developed by Peter Chang & SciSoft Group, Diamond Light Source Ltd. Links:
By Dan Porter, Diamond Light Source Ltd. 2024-2025
$ module load i16_msmapper
$ i16_msmapper
requirements: tkinter, numpy, matplotlib, h5py, hdfmap, msmapper
available from: https://github.com/DiamondLightSource/i16_msmapper
Latest version from github:
python -m pip install --upgrade git+https://github.com/DiamondLightSource/i16_msmapper.git
The following commands can be used on a beamline or DLS linux workstation (including NXuser)
$ module load msmapper/1.9
$ python -m pip install --upgrade git+https://github.com/DiamondLightSource/i16_msmapper.git
$ python -m i16_msmapperMSMapper can be run outside Diamond by downloading the executable file. The following options are for Windows but files for other operating systems are available and the process is similar.
- Install i16_msmapper as above
- Access the MSMapper files for different operating systems here: https://alfred.diamond.ac.uk/MSMapper/master/builds-snapshot/
- Download "MSMapper-1.7.0.v20240513-1606-win32.x86_64.zip" or equivalent
- Unzip the file to your choosen location
- Open the unzipped folder and copy the path of the executable (shift-right-click on
msmapperrc.exeand select copy as path) - Run i16_msmapper using
python -m i16_msmapper - From the main screen, select Tools>set shell command, replace the executable in the command with the copied one (ctrl+v)
- Click OK
- Use i16_msmapper as normal
MSMapper (Miller-Space-Mapper) is a java application that re-maps detector image pixels into a 3D voxel-grid with coordinates defined by the reciprocal lattice and motor coordinates stored in the NeXus file.
MSMapper is available on Diamond workstations via the module load system
$ module load msmapper
$ msmapper -bean /location/of/bean.json
# -OR-
$ rs_map -s 0.002 --monitor rc -o /dls/i16/data/2022/mm12345-1/processing/12345_remap.nxs /dls/i16/data/2022/mm12345-1/12345.nxsThe bean describes the options given to the MSMapper software
bean = {
inputs
[list] List of filenames of .nxs scan files
output
[str] Output filename - must be in processing directory, or somewhere you can write to
outputMode
[str] Type of output generated, see below for options
toCrystalFrame
[bool] if False, Q-mappings will be in the basis of the lab frame, rather than the Crystal frame
splitterName
[str] one of the following strings "nearest", "gaussian", "negexp", "inverse"
splitterParameter
[float] splitter's parameter is distance to half-height of the weight function.
If you use None or "" then it is treated as "nearest"
scaleFactor
[float] the oversampling factor for each image; to ensure that are no gaps in between pixels in mapping
monitorName
[str] sets the monitor to normalise by "rc", "ic1monitor"
thirdAxis
[list] direction of volume's third axis
aziPlaneNormal
[list] normal of azimuthal reference plane
correctPolarization
[bool] Set true to correct for polarization factor caused by transformation from laboratory frame to scattering plane
region
[sx, ex, sy, ey] Set rectangular region that defines the area with its bounds that contribute to output
step
[float, list] a single value or list if 3 values and determines the lengths of each side of the voxels in the volume
start
[list] location in HKL space of the bottom corner of the array.
shape
[list] size of the array to create for reciprocal space volume
reduceToNonZero
[bool] if True, attempts to reduce the volume output
}
Q-coordinates are in the basis of the sample frame (z along c, as Busing & Levy B matrix), as defined in the input file, unless toCrystalFrame=False, in which case the coordinates are given in the lab frame (version 1.9+).
| Output Mode | Description |
|---|---|
| Area_HK | Area in Miller-space (H,K) |
| Area_KL | Area in Miller-space (K,L) |
| Area_LH | Area in Miller-space (L,H) |
| Area_QPP | Area in q-space (parallel, perpendicular to sample surface) |
| Area_QXY | Area in q-space (X,Y) |
| Area_QYZ | Area in q-space (Y,Z) |
| Area_QZX | Area in q-space (Z,X) |
| Coords_HKL | Coordinates in Miller space |
| Coords_Q | Coordinates in q-space (momentum transfer) |
| Line_2Theta | Line in q-space (2 x theta is scattering angle, also in degrees) |
| Line_H | Line in Miller space (H) |
| Line_K | Line in Miller space (K) |
| Line_L | Line in Miller space (L) |
| Line_QX | Line in q-space (X) |
| Line_QY | Line in q-space (Y) |
| Line_QZ | Line in q-space (Z) |
| Line_Theta | Line in q-space (2 x theta is scattering angle) |
| Volume_HKL | Volume in Miller space |
| Volume_Q | Volume in q-space (crystal frame) |
| Volume_QCP | Volume in cylindrical polar q-space (crystal frame) |
| Volume_QES | Volume in equatorial stereographic q-space (crystal frame) |
bean = {
"inputs": ['i16/2024/mm12345-1/123456.nxs', 'i16/2024/mm12345-1/123457.nxs'], # Filename of scan file
"output": 'i16/2024/mm12345-1/processing/123456_rsmap.nxs',
# Output filename - must be in processing directory, or somewhere you can write to
"splitterName": "gaussian", # one of the following strings "nearest", "gaussian", "negexp", "inverse"
"splitterParameter": 2.0,
# splitter's parameter is distance to half-height of the weight function.
# If you use None or "" then it is treated as "nearest"
"scaleFactor": 2.0,
# the oversampling factor for each image; to ensure that are no gaps in between pixels in mapping
"step": [0.002, 0.002, 0.002],
# a single value or list if 3 values and determines the lengths of each side of the voxels in the volume
"start": None, # location in HKL space of the bottom corner of the array.
"shape": None, # size of the array to create for reciprocal space volume
"reduceToNonZero": False # True/False, if True, attempts to reduce the volume output
}