deepanshs / mrsimulator-app Goto Github PK

Steps to Reproduce:

1. Create session with multiple spin_systems and/or methods
2. Go to features tab and select a different method and/or spin_system than the first
3. Click Simulate or Fit and table & feature select buttons default to 0.

Note: For spin_systems, this does not change the selected system under the spin_systems tab. It does, however, select the first method under the methods tab

Note: If on the second or greater page on feature select buttons (spin systems 8+), buttons do not update but the table & values does.

NOTE: This bug only affects mrsimulator-app when running locally. The current live dev version is not affected by this issue.

Steps to reproduce:

1. Load any example with experimental data
2. Go to methods tab
3. Change sigma to any valid value except 1
4. Press the Submit Method button

Expected behavior:
Values in method fields remain the same after Submit Method is clicked

Observed behavior:
Other value remain the same, but sigma always resets to 1 regardless of given value

I believe the issue originates in these lines of code:

Steps to reproduce.

1. Open the example, wollastonite.
2. Add a new method (isotope 29Si).
3. Goto the Features tab.
4. Run least-squares minimization. (This will raise an error as expected: LeastSquaresAnalysisError: Please attach measurement(s) for method(s) at index(es) [1] before performing the least-squares analysis.)
5. Next, Goto the Methods tab and click on the unselected method.

Expected behavior: The screen should show the method parameters of the selected method.

What is observed: The screen jumps to the Features tab. This should not happen.

Steps to reproduce:

1. Create a new document
2. Add BlochDecaySpectrum for 1H
3. Go to Methods tab and add Constant Baseline Offset of 1 under Signal Processing
4. Go to Spin Systems tab
5. Add new 1H spin system
6. Observe baseline on the spectrum
7. Repeat from step 5 for desired number of spin systems

Expected behavior:
The spectrum always shows a baseline at 1

Observed behavior:
The spectrum baseline is dependent on the number of spin systems. For n>1 spin systems and a defined baseline offset of b the observed offset is n*b

After running a fit, the fit_report tab should update with tables displaying information about the fit. This functionality uses JavaScript and element.innerHTML to populate the page.

When mrsimulator-app is opened on Firefox, no issues arise. However on Safari and Google Chrome, the tab does not populate after a fit. Looking at the server output suggests the fit was successful and the issue arises from browsers' differences in JavaScript.

This issue probably stems from element.innerHTML.

Steps to reproduce.

1. From examples page, load Wollastonite
2. Select the decompose spectrum button

Expected behavior: The spectrum decomposes into experiment, Q2 (3), Q2 (1), Q2 (2), and residual

Observed behavior: The spectrum decomposes into experiment, Q2 (3), Q2 (1), Q2 (2), and NMR Spectrum

NMR Spectrum is mislabeled on the graph legend

Steps to reproduce:

1. Create a new file in the mrsimulator-app
2. Add two spin_systems to the file
3. Add a method to the file
4. Under Signal Processing add two convolutions
5. Have the first convolution apply only to spin_system 0 and second convolution only apply to spin_system 1
6. Apply the processor
7. Download the simulation as a .mrsim file
8. Replacing PATHNAME with downloaded file pathname run the following code

import mrsimulator

sim, processors, _ = mrsimulator.load("PATHNAME")
# sim.config.decompose_spectrum = "spin_system"
sim.run()

for processor, method in zip(processors, sim.methods):
    processed_data = processor.apply_operations(method.simulation).real

Expected behavior:
Downloaded data is simulated and processed without error

Observed behavior:
The last line throws IndexError: list index out of range error

Note: If line 4 is uncommented then the code runs without error

Steps to reproduce.

1. Open the example, wollastonite.
2. Add a new method (isotope 29Si).
3. Go to Features tab (needed to load feature select buttons).
4. Go to Methods tab.
5. Select new method.
6. From top menu bar, select method > Remove selection
7. Go to Features tab.

Expected behavior: The screen changes to Features tab with 0th and only method selected.

Observed behavior: The app pauses unusually long and throws error undefined is not an object (evaluating 'list[value+1].click') and ends on Features tab.

A similar issue results when removing selected method through method > Remove selection on the fitting tab; the buttons are not updated and selecting invalid methods gives same error.