radis.test.tools.test_slit module¶
Test slit methods
Examples
Run all tests:
pytest (in command line, in project folder)
Run only fast tests (i.e: tests that have a ‘fast’ label):
pytest -m fast
- linear_dispersion(w, f=750, phi=-6, m=1, gr=300)[source]¶
dlambda / dx Default values correspond to Acton 750i
- Parameters:
f (focal length (mm)) – default 750 (SpectraPro 2750i)
phi (angle in degrees (°)) – default 9
m (order of dispersion) – default 1
gr (grooves spacing (gr/mm)) – default 300
- test_against_specair_convolution(plot=True, close_plots=True, verbose=True, debug=False, *args, **kwargs)[source]¶
- test_all_slit_shapes(FWHM=0.4, verbose=True, plot=True, close_plots=True, *args, **kwargs)[source]¶
Test all slit generation functions and make sure we get the expected FWHM
- test_auto_correct_dispersion(verbose=True, plot=True, close_plots=True, *args, **kwargs)[source]¶
A test case to show the effect of wavelength dispersion (cf spectrometer reciprocal function) on the slit function
Test succeeds if there is a difference between convoluted spectra (with and without slit dispersion)
- test_convoluted_quantities_units(*args, **kwargs)[source]¶
Test that units are correctly convoluted after convolution
- test_cut_slices(verbose=True, plot=True, close_plots=True, *args, **kwargs)[source]¶
A test case to verify that _cut_slices does cut the spectrum into slices
Test fails if a
SlitDispersionWarning
is triggered and succeeds if the spectrum is sliced into 8 segments.
- test_linear_dispersion_effect(verbose=True, plot=True, close_plots=True, *args, **kwargs)[source]¶
A test case to show the effect of wavelength dispersion (cf spectrometer reciprocal function) on the slit function
Test succeeds if a
SlitDispersionWarning
is correctly triggered and if the FWHM is properly dilated.
- test_normalisation_mode(plot=True, close_plots=True, verbose=True, *args, **kwargs)[source]¶
Test norm_by = ‘area’ vs norm_by = ‘max’
- test_resampling(rtol=0.01, verbose=True, plot=True, warnings=True, *args, **kwargs)[source]¶
Test what happens when a spectrum in nm or cm-1, is convolved with a slit function in nm. In particular, slit function is generated in the spectrum unit, and spectrum is resampled if not evenly spaced
- test_slit_energy_conservation(verbose=True, plot=True, close_plots=True, *args, **kwargs)[source]¶
Convoluted and non convoluted quantities should have the same area (difference arises from side effects if the initial spectrum is not 0 on the sides