Add Feature: XAS Plugin #580

PNOGillespie · 2023-12-13T12:33:52Z

Overview

This PR adds a plugin for calculating X-Ray Absorption (XAS) spectra using the XspectraCrystalWorkChain of AiiDA-QE. Currently this is implemented for K-edge calculations of neutral systems. Calculations of systems with hubbard parameters and magnetic configurations can be made possible, though would require a backend change to AiiDA-QE (see aiidateam/aiida-quantumespresso#969 for relevant PR)

Setting Panel

The Setting panel for the plugin allows users to simply select which element to calculate XAS for, as well as tuning the supercell size to be used in the calculation. This panel also provides the option to change the type of core-hole approximation between FCH (removing the excited electron with tot_charge = 1) and XCH (allowing the excited electron to occupy the conduction band. Since it is still an ongoing subject of debate in the literature, we allow the user to change this value - though recommended values for each element are provided, which will also occupy the default value in the dropdown box.

Note that only elements for which appropriate pseudopotentials are available will be presented in the panel. See later section for notes about pseudopotentials.

Results Panel

Overview

The final XAS spectrum is displayed in the Result panel. A dropdown box selects which element to display spectra for. The widget will display both the complete powder XAS for the input structure and the contributions from each symmetrically non-equivalent site. Note that all spectra are normalised with respect to area under the peak by default. Spectra from different sites are scaled in intensity according to the ratio of site multiplicity (number of symmetrically-equivalent sites) to total multiplicity (number of sites for the element).

Broadening Tools

Broadening effects present in experimental XAS data greatly affect the appearance of the final spectrum (see Bunau & Calandra (2013)) and so post-procesing of spectra to include these effects (which cannot be computed ab-initio with DFT) is often used in published works using DFT-derived spectra. We therefore provide a set of broadening tools for the final spectrum. The tools apply one of two Lorentzian broadening schemes (toggled by a checkbox widget):

Constant broadening applied to the entire spectrum.
Variable-energy broadening using a starting value ($\Gamma_{hole}$), maximum value ($\Gamma_{max}$, and a mid-point energy ($E_{Center}$).

The 1st scheme uses just $\Gamma_{hole}$ (which refers to the natural linewidth of the core level) to define the Lorentzian parameter. The 2nd scheme uses each parameter to define an s-curve function with an inflection point at $E_{Center}$. The broadening scheme is applied to all spectra on the plot simultaneously. Note that when the spectrum is downloaded, the plotted data on the Plotly widget (including broadening) is what is written to the CSV file - not the raw data. To view the raw data, the user can simply set $\Gamma_{hole}$ to 0.

Pseudopotentials

Calculations of XAS using QE require a pseudopotential to define the electron configuration of the final state of the excitation for the absorbing atom. In addition, a radial coordinate plot of the initial state (e.g. the 1s state of the absorbing atom in a K-edge calculation) is needed by XSpectra. To provide both of these, including a relevant pseudopotential for the ground-state of the absorbing element (and from which we generate the initial state wavefunction) a small pseudopotential library is provided with the plugin. This library is taken from another GitHub repository (https://github.com/PNOGillespie/Core_Level_Spectra_Pseudos) which is a small collection of pseudos which we have used already in XAS calculations. This repo has an archive file for each XC functional (currently only PBE) and a "Pseudo Table of Contents" (pseudo_toc.yaml) file to record the filenames and directories for each pseudo and associated core wavefunction file. The plugin uses this YAML file to import files, check for missing files, load the correct nodes for calculation, and set the default core-hole approximation. The only requirement for maintainers is just to copy the YAML file from Core_Level_Spectra_Pseudos into the plugin directory in order to update the local pseudo archive (which is done automatically on startup if the archive file is missing or there are pseudos missing which should be there).

Adds a working version of a plugin to calculate XANES spectra using the XspectraCrystalWorkChain of AiiDA-QE This commit requires a locally-situated copy of the core-hole pseudopotentials required for XSpectra to function, which is obtained from https://github.com/PNOGillespie/Core_Level_Spectra_Pseudos and downloaded and installed upon first activation of AiiDALab-QE

Refactors the system for importing core-hole pseudopotentials from using groups to using a `yaml` file to define which pseudos should be present in the ch pseudos archive. Greatly simplifies the logic for checking for the presence of pseudopotentials and populating lists of available pseudos. Yaml file `pseudo_toc` is obtained from the Core_Level_Spectra_Pseudos archive (https://github.com/PNOGillespie/Core_Level_Spectra_Pseudos) and must be updated on both the archive and the plugin when new pseudos are available on the CLS archive. Also adds a step in `workchain.py` to check for mismatches in which pseudos should be present and re-download the archive file if there is a mismatch. Added in order to check for inconsistencies and to automatically update the archive if `pseudo_toc.yaml` is updated with new pseudos.

Migrates the directory used for core-level spectra pseudos to `~/.local/lib/cls_pseudos` in order to allow integration tests to pass.

Adds a range of improvements and fixes to the plugin. Fixes: * Fixes an issue where the `Setting` panel of the plugin would cause a crash due to a variable being re-defined during runtime. * Adds a `.close()` statement to the end of the HTTP request when accessing the pseudo archive file for download. Additions: * The app now retrieves all contributions to the final spectrum for the selected element and plots each together. The spectra are aligned in terms of energy in the same way as the `XspectraCrystalWorkChain` uses to generate the final spectrum. Each contributing spectrum is scaled according to the ratio of site multiplicity to total multiplicity. The provided broadening tools will apply to all spectra on the widget simultaneously. * Adds a button to the `Result` panel to download a CSV file of the spectra presented on the Plotly widget. This will download a file titled `{element}_XAS_Spectra.csv` which contains all spectra on the widget (total spectrum and all site contributions). For sites with a multiplicity ratio different from 1, the weighted and unweighted spectra are printed in separate columns.

codecov · 2023-12-13T12:43:05Z

Codecov Report

Attention: 204 lines in your changes are missing coverage. Please review.

Comparison is base (8ad9d9c) 80.91% compared to head (b0a80ac) 77.55%.

Files	Patch %	Lines
src/aiidalab_qe/plugins/xas/result.py	10.65%	176 Missing ⚠️
src/aiidalab_qe/plugins/xas/workchain.py	34.28%	23 Missing ⚠️
src/aiidalab_qe/plugins/xas/setting.py	96.40%	5 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #580      +/-   ##
==========================================
- Coverage   80.91%   77.55%   -3.36%     
==========================================
  Files          49       54       +5     
  Lines        3426     3831     +405     
==========================================
+ Hits         2772     2971     +199     
- Misses        654      860     +206

Flag	Coverage Δ
python-3.10	`77.55% <49.25%> (?)`
python-3.8	`77.58% <49.25%> (-3.33%)`	⬇️
python-3.9	`77.58% <49.25%> (-3.33%)`	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

unkcpz

Really cool! Welcome XAS to join the QeApp family.
I give a broad look with two comments, @superstar54 is capable of giving it a detailed look.

unkcpz · 2023-12-15T09:45:55Z

src/aiidalab_qe/plugins/xas/pseudo_toc.yaml

+            C: C.pbe-n-kjgipaw_psl.1.0.0.UPF
+            Cu: Cu.pbe-n-van_gipaw.UPF
+            F: F.pbe-gipaw_kj_no_hole.UPF
+            Li: Li.pbe-s-rrkjus-gipaw.UPF
+            O: O.pbe-n-kjpaw_gipaw.UPF
+            Si: Si.pbe-van_gipaw.UPF


I don't want to slow down the progress, is that possible to integrate this to aiida-pseudo?

So, I will say for this that the solution being used here is not necessarily intended as a permanent solution to providing pseudopotentials for these types of calculations. In principle, integrating some sort of family in aiida-pseudo for core-hole pseudopotentials is the ideal long-term objective here - since then the procedure for assigning pseudos obviously follows that of other plugins.

The two issues I see with integrating the current library with aiida-pseudo are:

Some sort of new Class or other new feature will need to be made for aiida-pseudo in order to work with these types of pseudopotentials: for instance to ensure the correct core wavefunction data file is matched to the correct GIPAW pseudo, also that the core-hole and GIPAW pseudos are matched correctly.

The current library is a stub which covers only some elements, therefore potentially not enough to make it worth pursuing with just what's there at the moment.

I would be interested in your thoughts on this in any case @unkcpz.

unkcpz · 2023-12-15T09:47:27Z

src/aiidalab_qe/plugins/xas/result.py

+    """Download button with dynamic content
+    The content is generated using a callback when the button is clicked.
+    Modified from responses to https://stackoverflow.com/questions/61708701/how-to-download-a-file-using-ipywidget-button#62641240
+    """


We have the widget to use in https://github.com/aiidalab/aiidalab-widgets-base/blob/master/aiidalab_widgets_base/export.py

I can try it using the AWB version and see how that goes. In principle, it looks like it should be fine since the file for the output spectra is created outside of the widget, but I will let you know if there are any issues with using it here.

Sorry, @PNOGillespie I think what you did are correct, the export widget is for exporting the aiida archive, we don't yet have the generic widget for export the file.
Can you try to it to aiidalab-qe/src/aiidalab_qe/common/, @AndresOrtegaGuerrero also implement a similar widget.

Sure @unkcpz, I will work on using the export widget with the plugin, though I'm currently only a couple of days from vacation, so maybe in January at this rate 😆

I had intended it to show explicitly (for the user) that it would download a CSV file - also that the filename be related to the element selected to be displayed (e.g. O_K-edge_XAS.csv for spectra of oxygen). Does the widget allow for the filename or description text to be specified? @AndresOrtegaGuerrero

I don't mind working on changes to export in order to make this happen, if that's ok with everyone.

@PNOGillespie i think @unkcpz refers to the BandsWidget PR #581 , you can adapt the export (code) to your needs, like the filename name

Hi @unkcpz & @AndresOrtegaGuerrero, I've run some tests with the DownloadButton widget and compared it to the SpectrumDownloadButton widget I made for this plugin. I can get all the same functionality out of DownloadButton as I did for my own one, though the issue I notice is that the Plotly viewer runs noticeably slower when working with the DownloadButton.

The reason, I'm guessing, is that the DownloadButton needs the file's contents (payload) to be a bytes object, so each time the spectrum is re-drawn (e.g. if the spectrum broadening or element selection is changed) it needs to encode a new bytes object, which takes extra time (probably even more for spectra composed of multiple different absorbing atoms). The SpectrumDownloadButton doesn't have this issue because encoding is done during __on_click and so the input can just be a str object.

I do like the idea of a generalised DownloadButton widget (particularly as there's only really one way to do make this work anyway) but I think it needs some encoding methods implemented to enhance the performance.

superstar54

@PNOGillespie thanks for the work!

I did some test calculations. Here are some comments on the workchain part.

src/aiidalab_qe/plugins/xas/workchain.py

Enables the `supercell_min_parameter` setting for XAS calculations Also removes various extraneous lines of code which were commented out, had typos, or were requested for deletion.

for more information, see https://pre-commit.ci

superstar54

Hi @PNOGillespie, I see that you checked and downloaded the pseudo in the workchain.py part. I suggest moving them to the setting.py. Because the pseudo information is used in the setting panel. Of course, you can do this in another PR.

The PR will be good to me when you make the changes.

For the download button, let Andres finish his, and we can discuss which one to go with. You can update it in the future PR.

superstar54 · 2024-01-24T15:04:38Z

src/aiidalab_qe/plugins/xas/result.py

+        self.on_click(self.__on_click)
+
+    def __on_click(self, b):
+        if self.contents() is None:


You pass Non to contents when creating the download_data button

Suggested change

if self.contents() is None:

if self.contents is None:

superstar54 · 2024-01-24T15:23:02Z

src/aiidalab_qe/plugins/xas/setting.py

+        <h4>Element and Core-Hole Treatment Setting.</h4></div>"""
+    )
+
+    # TODO: The element selection should lock the "Confirm" button if no elements have been


Could you open a issue? so that we can work on this in the future.

I'll open one once this PR is concluded, no problem.

superstar54 · 2024-01-24T15:27:53Z

src/aiidalab_qe/plugins/xas/setting.py

+        }
+        return parameters
+
+    def set_panel_value(self, input_dict):


the element_and_ch_treatment_options is not set properly. I think you need to read the
core_hole_treatments, and elements_list from then input_dict, and set the widget values.

Ok, this did confuse me a little on review, but element_and_ch_treatment_options is indeed not set, however it also doesn't do anything. The core_hole_treatments and elements_list however are set correctly - I tested it just now - but that's defined by the code in _update_element_selection_panel() (lines 166+).

The short answer to this is that we can just delete all of the code in lines 141-159. I can discuss this in more detail separately if you would like.

PNOGillespie · 2024-01-25T14:11:04Z

src/aiidalab_qe/plugins/xas/setting.py

+        ch_pseudos = self.core_hole_pseudos
+        structure = self.input_structure
+        available_elements = [k for k in ch_pseudos]
+        elements_to_select = sorted(
+            [
+                kind.symbol
+                for kind in structure.kinds
+                if kind.symbol in available_elements
+            ]
+        )
+
+        element_and_ch_treatment_options = {}
+        for element in elements_to_select:
+            if element in xch_elements:
+                element_and_ch_treatment_options[element] = "xch_smear"
+            else:
+                element_and_ch_treatment_options[element] = "full"
+
+        self.element_and_ch_treatment_options = element_and_ch_treatment_options


Hi @superstar54. Does this seem like a reasonable way to address your previous comment?

I suggest this because the logic to define the initial values for the element_and_ch_treatment panel has already been defined in _update_element_select_panel(). This works fine with my own testing and sets the right behaviour for set_panel_value().

Suggested change

ch_pseudos = self.core_hole_pseudos

structure = self.input_structure

available_elements = [k for k in ch_pseudos]

elements_to_select = sorted(

[

kind.symbol

for kind in structure.kinds

if kind.symbol in available_elements

]

)

element_and_ch_treatment_options = {}

for element in elements_to_select:

if element in xch_elements:

element_and_ch_treatment_options[element] = "xch_smear"

else:

element_and_ch_treatment_options[element] = "full"

self.element_and_ch_treatment_options = element_and_ch_treatment_options

self._update_element_select_panel()

Changes requested changes for PR aiidalab#580: * Moves the loading of pseudos and core wavefunction data to `setting.py`. `setting.py` now collects labels for pseudos and core_wfc_data, while `workchain.py` loads the nodes from the labels. * Replaces `set_panel_value()` content with call to `self._update_element_select_panel()`. * Fixes typos and removes redundant code noted in latest review.

superstar54 · 2024-01-29T21:11:54Z

The codecov decreased because there is no test on the workchain and its result. We will come to this when we resolve #533.

unkcpz · 2024-01-29T21:22:10Z

The codecov decreased because there is no test on the workchain and its result. We will come to this when we resolve #533.

Thanks for mentioning this. I'd suggest we iron the test first.

We leave the pseudopotential setup and the download button to the feature PR.

unkcpz · 2024-01-30T10:51:07Z

We leave the pseudopotential setup and the download button to the feature PR.

As we talked over the phone, please open issues for these isuses that not addressed. And we should try to do it as soon as possible, otherwise they will never happened. @superstar54 @PNOGillespie

unkcpz · 2024-01-30T10:55:21Z

Also a mild reminder that try not do the hush merge next time. If you have DDL then we need to plan things ahead and make things happened under the plan. The integration test is failed so please try to fix it by this week @superstar54 @PNOGillespie

unkcpz · 2024-01-30T10:56:40Z

In fact there is a crucial break in the PR, the integration test shows:

---------------------------------------------------------------------------
FileNotFoundError                         Traceback (most recent call last)
/tmp/ipykernel_1713/2189467020.py in <cell line: 17>()
     15 from jinja2 import Environment
     16 
---> 17 from aiidalab_qe.app import App, static
     18 from aiidalab_qe.version import __version__

/opt/conda/lib/python3.9/site-packages/aiidalab_qe/app/__init__.py in <module>
      1 """Package for the AiiDAlab QE app."""
      2 
----> 3 from .main import App
      4 
      5 __all__ = [

/opt/conda/lib/python3.9/site-packages/aiidalab_qe/app/main.py in <module>
     12 from aiidalab_qe.app.result import ViewQeAppWorkChainStatusAndResultsStep
     13 from aiidalab_qe.app.structure import StructureSelectionStep
---> 14 from aiidalab_qe.app.submission import SubmitQeAppWorkChainStep
     15 from aiidalab_qe.common import QeAppWorkChainSelector
     16 

/opt/conda/lib/python3.9/site-packages/aiidalab_qe/app/submission/__init__.py in <module>
     20 from aiidalab_qe.common.setup_codes import QESetupWidget
     21 from aiidalab_qe.common.setup_pseudos import PseudosInstallWidget
---> 22 from aiidalab_qe.workflows import QeAppWorkChain
     23 
     24 from .resource import ParallelizationSettings, ResourceSelectionWidget

/opt/conda/lib/python3.9/site-packages/aiidalab_qe/workflows/__init__.py in <module>
     40 
     41 
---> 42 plugin_entries = get_entry_items("aiidalab_qe.properties", "workchain")
     43 
     44 

/opt/conda/lib/python3.9/site-packages/aiidalab_qe/workflows/__init__.py in get_entry_items(entry_point_name, item_name)
     32 # load entry point items
     33 def get_entry_items(entry_point_name, item_name="workchain"):
---> 34     entries = get_entries(entry_point_name)
     35     return {
     36         name: entry_point.get(item_name)

/opt/conda/lib/python3.9/site-packages/aiidalab_qe/workflows/__init__.py in get_entries(entry_point_name)
     25     entries = {}
     26     for entry_point in entry_points().get(entry_point_name, []):
---> 27         entries[entry_point.name] = entry_point.load()
     28 
     29     return entries

/opt/conda/lib/python3.9/site-packages/importlib_metadata/__init__.py in load(self)
    205         """
    206         match = self.pattern.match(self.value)
--> 207         module = import_module(match.group('module'))
    208         attrs = filter(None, (match.group('attr') or '').split('.'))
    209         return functools.reduce(getattr, attrs, module)

/opt/conda/lib/python3.9/importlib/__init__.py in import_module(name, package)
    125                 break
    126             level += 1
--> 127     return _bootstrap._gcd_import(name[level:], package, level)
    128 
    129 

/opt/conda/lib/python3.9/site-packages/aiidalab_qe/plugins/xas/__init__.py in <module>
      8 
      9 from .result import Result
---> 10 from .setting import Setting
     11 from .workchain import workchain_and_builder
     12 

/opt/conda/lib/python3.9/site-packages/aiidalab_qe/plugins/xas/setting.py in <module>
     17 from aiidalab_qe.plugins import xas as xas_folder
     18 
---> 19 PSEUDO_TOC = yaml.safe_load(resources.read_text(xas_folder, "pseudo_toc.yaml"))
     20 pseudo_data_dict = PSEUDO_TOC["pseudos"]
     21 xch_elements = PSEUDO_TOC["xas_xch_elements"]

/opt/conda/lib/python3.9/importlib/resources.py in read_text(package, resource, encoding, errors)
    137     bytes.decode().
    138     """
--> 139     with open_text(package, resource, encoding, errors) as fp:
    140         return fp.read()
    141 

/opt/conda/lib/python3.9/importlib/resources.py in open_text(package, resource, encoding, errors)
    119     """Return a file-like object opened for text reading of the resource."""
    120     return TextIOWrapper(
--> 121         open_binary(package, resource), encoding=encoding, errors=errors)
    122 
    123 

/opt/conda/lib/python3.9/importlib/resources.py in open_binary(package, resource)
     89     reader = _get_resource_reader(package)
     90     if reader is not None:
---> 91         return reader.open_resource(resource)
     92     absolute_package_path = os.path.abspath(
     93         package.__spec__.origin or 'non-existent file')

/opt/conda/lib/python3.9/importlib/_bootstrap_external.py in open_resource(self, resource)

FileNotFoundError: [Errno 2] No such file or directory: '/opt/conda/lib/python3.9/site-packages/aiidalab_qe/plugins/xas/pseudo_toc.yaml'

unkcpz · 2024-01-30T10:57:17Z

You can reproduce it by start the container docker run --rm -it -p 8888:8888 aiidalab/qe:edge and go to the app to see the error.

superstar54 · 2024-01-30T11:38:01Z

@unkcpz Sorry for the bug. It should be fixed in #609 .

unkcpz · 2024-01-30T11:40:12Z

No worries, I add the integration test to the required test. It is quite robust and can give useful information, so we need to put an eye on it next time.

PNOGillespie added 7 commits November 21, 2023 17:22

XAS: Migrate CLS Pseudos Directory

bc38545

Migrates the directory used for core-level spectra pseudos to `~/.local/lib/cls_pseudos` in order to allow integration tests to pass.

Merge remote-tracking branch 'origin/main' into xas_plugin

cc9785a

XAS: Fix for Pre-Commit Checks

ed59020

XAS: Fix Typo in variable_broad_select_help

3f6acbc

PNOGillespie changed the title ~~Xas plugin~~ Add Feature: XAS Plugin Dec 13, 2023

unkcpz previously requested changes Dec 15, 2023

View reviewed changes

superstar54 requested changes Jan 17, 2024

View reviewed changes

src/aiidalab_qe/plugins/xas/workchain.py Outdated Show resolved Hide resolved

src/aiidalab_qe/plugins/xas/workchain.py Show resolved Hide resolved

src/aiidalab_qe/plugins/xas/workchain.py Outdated Show resolved Hide resolved

PNOGillespie and others added 2 commits January 24, 2024 15:07

XAS: Enable supercell_min_parameter and Clean Code

9a9f162

Enables the `supercell_min_parameter` setting for XAS calculations Also removes various extraneous lines of code which were commented out, had typos, or were requested for deletion.

[pre-commit.ci] auto fixes from pre-commit.com hooks

1b24462

for more information, see https://pre-commit.ci

superstar54 requested changes Jan 24, 2024

View reviewed changes

PNOGillespie commented Jan 25, 2024

View reviewed changes

PNOGillespie and others added 4 commits January 26, 2024 16:17

Merge branch 'main' into xas_plugin

3ac8b6e

fix set_panel_value

cc963d4

add test for xas app's setting panel

b0a80ac

superstar54 enabled auto-merge (squash) January 29, 2024 21:16

superstar54 requested review from superstar54 and unkcpz January 29, 2024 21:18

superstar54 approved these changes Jan 29, 2024

View reviewed changes

superstar54 self-requested a review January 29, 2024 21:21

superstar54 approved these changes Jan 29, 2024

View reviewed changes

superstar54 merged commit 1950c34 into aiidalab:main Jan 29, 2024
16 of 18 checks passed

superstar54 mentioned this pull request Jan 30, 2024

Include xas yaml in package data #609

Merged

PNOGillespie deleted the xas_plugin branch February 18, 2024 18:29

superstar54 mentioned this pull request Mar 26, 2024

Implement X-ray absorption spectroscopy (XAS) (first working version) #350

Closed

5 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add Feature: XAS Plugin #580

Add Feature: XAS Plugin #580

PNOGillespie commented Dec 13, 2023

codecov bot commented Dec 13, 2023 •

edited

Loading

unkcpz left a comment

unkcpz Dec 15, 2023

PNOGillespie Dec 15, 2023

unkcpz Dec 15, 2023

PNOGillespie Dec 15, 2023

unkcpz Dec 18, 2023

PNOGillespie Dec 19, 2023 •

edited

Loading

AndresOrtegaGuerrero Dec 19, 2023

PNOGillespie Jan 22, 2024

superstar54 left a comment

superstar54 left a comment •

edited

Loading

superstar54 Jan 24, 2024

superstar54 Jan 24, 2024

PNOGillespie Jan 24, 2024

superstar54 Jan 24, 2024

PNOGillespie Jan 24, 2024

PNOGillespie Jan 25, 2024

superstar54 commented Jan 29, 2024

unkcpz commented Jan 29, 2024

unkcpz commented Jan 30, 2024 •

edited

Loading

unkcpz commented Jan 30, 2024

unkcpz commented Jan 30, 2024

unkcpz commented Jan 30, 2024

superstar54 commented Jan 30, 2024

unkcpz commented Jan 30, 2024

Add Feature: XAS Plugin #580

Add Feature: XAS Plugin #580

Conversation

PNOGillespie commented Dec 13, 2023

Overview

Setting Panel

Results Panel

Overview

Broadening Tools

Pseudopotentials

codecov bot commented Dec 13, 2023 • edited Loading

Codecov Report

unkcpz left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

PNOGillespie Dec 19, 2023 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

superstar54 left a comment

Choose a reason for hiding this comment

superstar54 left a comment • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

superstar54 commented Jan 29, 2024

unkcpz commented Jan 29, 2024

unkcpz commented Jan 30, 2024 • edited Loading

unkcpz commented Jan 30, 2024

unkcpz commented Jan 30, 2024

unkcpz commented Jan 30, 2024

superstar54 commented Jan 30, 2024

unkcpz commented Jan 30, 2024

codecov bot commented Dec 13, 2023 •

edited

Loading

PNOGillespie Dec 19, 2023 •

edited

Loading

superstar54 left a comment •

edited

Loading

unkcpz commented Jan 30, 2024 •

edited

Loading