Ctrl+K

swissCheese

swissCheese#

Overview

Swiss cheese is a class that contains multiple methods to support the creation, use and management of Swiss Cheese masks. It also acts as a container for another class defined in maskUtils called eye. The nomenclature comes from the cheese analogy as the individual holes in a swiss cheese are referred to as “eyes”.

An eye has 5 attributes:

Attribute

Summary

xUnits

the units in the x-axis that define the bin limits of the eye. These must match mantid standard units such as Wavelength, dSpacing or TOF

xMin

The minimium value, in units of xUnits above which events will be masked

xMax

The maximum value, in units of xUnits above which events will be masked

inputWorkspaceIndexSet

this is a string specifying which pixels the eye affect. It follows mantid format: a comma separated list that may contain ranges specified using hyphens or colons e.g. “3452,5678,12000:12100”

isLite

as the pixel ranges in inputWorkspaceIndexSet will have different meaning depending on whether data are Lite or Native, the isLite Boolean must indicate which applies

When swissCheese is instantiated, it contains a single attribute, which is an empty list called eyeList which acts as a container for eye instances. Below, the available methods to generate and organize the list of eyes. In addition save and load methods exist to for the swissCheese object.

A swiss cheese mask is instantiated with

import snapwrap.maskUtils as mut

cheese = swissCheese()

At this point, the cheese has no eyes, these can be added using various methods as described below.

notchFromList#

A method which is especially beneficial for diamond-anvil datasets, this allows the user to specify “notches”. Notches are specific types of eye that extend across all detector pixels and, thus can “notch-out” entire sheets of events across a specified range of x-units. Required arguments are:

  • xUnits (typically ‘dSpacing’ or ‘Wavelength’),

  • notchList a list of lists where the sublists have length of 2 and contain [xMin,xMax]

  • isLite the usual specifier

The corresponding “eyes” are created and added to the swiss cheese

notchFromUB#

An alternate method to create notches, if you have ISAW format (commonly used at the SNS to store UB matrices), this utility will load the UB of a diamond anvil and calculate what wavelengths reflections are expected to satisfy the Bragg condition. The width of the notch corresponding to each reflection is specified by a polynomial of arbitrary degree.

i.e. the width of notch n, at wavelength lam_n is calculated as width a0+a1lam_n+a2lam_n_… with user specified coefficients. Required arguments are:

Note: currently, this only works for diamond anvils.

  • wsName the name of a mantid workspace that will hold the loaded UB matrix

  • UBPath the full path to the ISAW format UB matrix file

  • widthCoef the list of [a0,a1,a2…] coefficients used to calculate the width.

  • isLite the normal compression setting

  • lamMin defaulting to 0.5 Å this is the minimum lambda that eyes will be created for.

extractFromWorkspaceHistory#

It is possible to manually create a swiss cheese mask in the mantidworkbench show instrument view, using various provided drawing tools. After creating the mask and applying it to the workspace, this can be extracted into swiss cheese form using this method. The only argument is the workspace name.

makeMaskBinsTables#

This method converts all eyes into a mantid TableWorkspace that mantid can then apply using its algorithm MaskBinsFromTable. It will process all present eyes, treating those with different units separately and creating a separate TableWorkspace for these.

save and load#

These methods allow writing created swiss cheese masks to disk and then reloading them. For saving the required arguments are filePath and filePrefix. The corresponding mask will then be saved to a file at: {filePath}{filePrefix}_{unit}.json (where {unit} is the unit of the corresponding eyes).

For loading, the only required parameters is the full filepath.

Contents