New atom selection mechanism #672

MBartkowiakSTFC · 2025-02-14T17:12:27Z

Description of work
The intention behind this PR is to implement a mechanism of selecting atoms which is:

Not attached to a specific trajectory file,
Transferable,
Easy to extend.

We can still discuss if the implementation in this PR meets the requirements.

This screenshot shows the carbon dioxide trajectory in which two out of three atoms have been selected in the molecules from the first half of the index range. This is not scientifically meaningful, but illustrates how the selection operations can be combined.

Fixes

New selection functions have been implemented to replace the classes we have been using so far.
The output of each selection function is a set of indices.
Selections are applied sequentially, and can be combined using set operations (union, intersection, difference).
ReusableSelection class stores a sequence of selection functions and uses JSON for saving and loading a sequence.
ReusableSelection.select_in_trajectory method takes a Trajectory instance as input, applies the selection and returns the resulting set of atom indices.
Analysis jobs save the JSON selection sequence as one of the job inputs, and an array of 0/1 values to mark the indices which got selected.

The JSON representation is saved as str:

The indices are saved as a mask array:

To test
All tests must pass.

…to maciej/simplify-atom-selection

ChiCheng45 · 2025-02-18T12:22:07Z

Looks good. Just one thing about the SMARTS pattern matching. In the RDKit molecule object we use the Chem.rdchem.BondType.UNSPECIFIED for all of the bondtypes. This means that most SMARTS patterns you'd think should work won't and that is why the SMARTS patterns we use are a bit unusual.

The correct thing would be to set all the correct bond types and atom properties but I think this might get difficult to get working for a general molecule e.g. setting single/double/aromatic bonds for a benzene or cyclooctatetraene or some big protein.

MDANSE/Src/MDANSE/Framework/AtomSelector/atom_selection.py

oerc0122 · 2025-02-18T14:15:32Z

MDANSE_GUI/Src/MDANSE_GUI/InputWidgets/AtomSelectionWidget.py

+VALID_SELECTION = "Valid selection"
+USELESS_SELECTION = "Selection did not change. This operation is not needed."
+MALFORMED_SELECTION = "This is not a valid JSON string."


Consider making these an Enum

MDANSE_GUI/Src/MDANSE_GUI/InputWidgets/AtomSelectionWidget.py

oerc0122 · 2025-02-18T14:20:05Z

MDANSE_GUI/Src/MDANSE_GUI/Widgets/SelectionWidgets.py

+        if self.mode_box.currentIndex() == 2:
+            return "difference"
+        elif self.mode_box.currentIndex() == 1:
+            return "intersection"
+        else:
+            return "union"


Might be better as an Enum

One complication here is that these values have to be JSON encoded and decoded (and remain human-readable when saved as a text string). I am sure that it can be done. Do you think it is worth the effort?

I was actually referring to the 1, 2...
If you had:

class Modes(Enum): DIFFERENCE = auto() INTERSECTION = auto() UNION = auto() ... if self.mode_box.currentIndex() == Modes.DIFFERENCE: return "difference" elif self.mode_box.currentIndex() == Modes.INTERSECTION: return "intersection" elif self.mode_box.currentIndex() == Modes.UNION: return "union"

It removes magic numbers and ensures internal consistency a bit better if we ever add other operations.

As an aside, Enums can act as strings

oerc0122 · 2025-02-18T14:23:21Z

MDANSE_GUI/Src/MDANSE_GUI/Widgets/SelectionWidgets.py

+            self.selection_field.setPlaceholderText("0,1,2")
+            self.selection_keyword = "index_list"
+            self.selection_separator = ","
+        if new_mode == "range":


Also probably Enum

…to maciej/simplify-atom-selection

Co-authored-by: Jacob Wilkins <46597752+oerc0122@users.noreply.github.com> Signed-off-by: Maciej Bartkowiak <108934199+MBartkowiakSTFC@users.noreply.github.com>

…SISNeutronMuon/MDANSE into maciej/simplify-atom-selection

…to maciej/simplify-atom-selection

oerc0122 · 2025-02-28T10:10:51Z

MDANSE/Src/MDANSE/Framework/AtomSelector/atom_selection.py

+
+
+def select_atoms(
+    trajectory: Trajectory, **function_parameters: Dict[str, Any]


It's currently recommended that for **kwargs you document the types of any variables that they might contain.

https://peps.python.org/pep-0484/#arbitrary-argument-lists-and-default-argument-values

oerc0122 · 2025-02-28T10:12:11Z

MDANSE/Src/MDANSE/Framework/AtomSelector/general_selection.py

+    Set[int]
+        Set of all the atom indices
+    """
+    return set(range(len(trajectory.chemical_system.atom_list)))


I'm wondering if we want a utility function for set(range(len(...))) or if there's a nicer way of writing it.

oerc0122 · 2025-02-28T10:13:20Z

MDANSE/Src/MDANSE/Framework/AtomSelector/general_selection.py

+
+
+def select_all(
+    trajectory: Trajectory, **function_parameters: Dict[str, Any]


For cases where the arguments are not used internally, it's recommended that they are prefixed with an _

oerc0122 · 2025-02-28T10:36:48Z

MDANSE/Src/MDANSE/Framework/AtomSelector/molecule_selection.py

+    trajectory : Trajectory
+        A trajectory instance to which the selection is applied
+    function_parameters : Dict[str, Any]
+        should include a list of str molecule names under key "atom_labels"


If there are sensible options, they should be in the function signature, i.e.

def select_molecules( trajectory: Trajectory, *, molecule_names: Sequence[str] | None = None, **_function_parameters: typing.Never ) -> Set[int]:

(The lone * is to enforce molecule_names should only be passed via kwargs).

oerc0122 · 2025-02-28T10:40:30Z

MDANSE/Src/MDANSE/Framework/AtomSelector/selector.py

+        self.operations = {}
+
+    def set_selection(
+        self, number: Union[int, None] = None, function_parameters: Dict[str, Any] = {}


Should function parameters be ** here?

MDANSE/Src/MDANSE/Framework/AtomSelector/spatial_selection.py

oerc0122 · 2025-02-28T11:39:09Z

MDANSE_GUI/Src/MDANSE_GUI/InputWidgets/AtomTransmutationWidget.py

@@ -137,9 +135,8 @@ def update_transmutation_textbox(self) -> None:
        map = self.transmuter.get_setting()


Not you, but map here is shadowing the python function.

MDANSE_GUI/Src/MDANSE_GUI/Widgets/SelectionWidgets.py

Co-authored-by: Jacob Wilkins <46597752+oerc0122@users.noreply.github.com> Signed-off-by: Maciej Bartkowiak <108934199+MBartkowiakSTFC@users.noreply.github.com>

…to maciej/simplify-atom-selection

…SISNeutronMuon/MDANSE into maciej/simplify-atom-selection

oerc0122 · 2025-03-04T16:05:58Z

MDANSE/Src/MDANSE/Framework/AtomSelector/atom_selection.py

+    index_list: Sequence[int] = None,
+    index_range: Sequence[int] = None,
+    index_slice: Sequence[int] = None,


Structly pep-484 doesn't allow implicit None.

C.f. the end of https://peps.python.org/pep-0484/#union-types

oerc0122 · 2025-03-04T16:08:10Z

MDANSE/Src/MDANSE/Framework/AtomSelector/group_selection.py

+    """
+    selection = set()
+    system = trajectory.chemical_system
+    if rdkit_pattern:


Does it make sense to have a default here?

oerc0122 · 2025-03-04T16:11:51Z

MDANSE/Src/MDANSE/Framework/AtomSelector/selector.py

-        set[int]
-            The atoms indices.
+        bool
+            True if the selection adds atoms, False otherwise


Suggested change

True if the selection adds atoms, False otherwise

True if the operation changes selection, False otherwise

oerc0122 · 2025-03-04T16:19:17Z

MDANSE/Src/MDANSE/Framework/AtomSelector/selector.py

+            if operation_type == "union":
+                selection = selection.union(temp_selection)
+            elif operation_type == "intersection":
+                selection = selection.intersection(temp_selection)
+            elif operation_type == "difference":
+                selection = selection.difference(temp_selection)


Could be either

Suggested change

if operation_type == "union":

selection = selection.union(temp_selection)

elif operation_type == "intersection":

selection = selection.intersection(temp_selection)

elif operation_type == "difference":

selection = selection.difference(temp_selection)

if operation_type == "union":

selection |= temp_selection

elif operation_type == "intersection":

selection &= temp_selection

elif operation_type == "difference":

selection -= temp_selection

Or the shorter, but slightly more esoteric

Suggested change

if operation_type == "union":

selection = selection.union(temp_selection)

elif operation_type == "intersection":

selection = selection.intersection(temp_selection)

elif operation_type == "difference":

selection = selection.difference(temp_selection)

if operation_type in ("union", "intersection", "difference"):

selection = getattr(selection, operation_type)(temp_selection)

oerc0122 · 2025-03-04T16:22:34Z

MDANSE/Src/MDANSE/Framework/AtomSelector/spatial_selection.py

+    Returns
+    -------
+    Set[int]
+        _description_


oerc0122 · 2025-03-04T16:23:38Z

MDANSE/Src/MDANSE/Framework/Configurators/AtomSelectionConfigurator.py

@@ -132,14 +134,11 @@ def get_information(self) -> str:

        return "\n".join(info) + "\n"

-    def get_selector(self) -> Selector:
+    def get_selector(self) -> "ReusableSelection":


Am I wrong in thinking this is imported?

Suggested change

def get_selector(self) -> "ReusableSelection":

def get_selector(self) -> ReusableSelection:

oerc0122 · 2025-03-04T16:25:12Z

MDANSE_GUI/Src/MDANSE_GUI/InputWidgets/AtomSelectionWidget.py

+class SelectionValidity(StrEnum):
+    VALID_SELECTION = "Valid selection"
+    USELESS_SELECTION = "Selection did not change. This operation is not needed."
+    MALFORMED_SELECTION = "This is not a valid JSON string."


Suggested change

MALFORMED_SELECTION = "This is not a valid JSON string."

MALFORMED_SELECTION = "This is not a valid selection string."

…to maciej/simplify-atom-selection

oerc0122

Looks good to me if it does what it's meant to, just a couple of possible stylistic suggestions. Feel free to ignore if you feel they remove clarity.

oerc0122 · 2025-03-06T13:26:47Z

MDANSE/Src/MDANSE/Framework/AtomSelector/selector.py

+        if number is None:
+            number = len(self.operations)
+        else:
+            number = int(number)


If it's not too long an option is:

Suggested change

if number is None:

number = len(self.operations)

else:

number = int(number)

number = int(number) if number is not None else len(self.operations)

oerc0122 · 2025-03-06T13:33:39Z

MDANSE/Src/MDANSE/Framework/AtomSelector/selector.py

+        return bool(
+            (
+                len(selection.difference(current_selection)) > 0
+                and operation_type == "union"
+            )
+            or (
+                len(current_selection.difference(selection)) > 0
+                and operation_type != "union"
+            )
+        )


Give if len(Collection) > 0 is equivalent to if Collection we can simplify this to:

Suggested change

return bool(

(

len(selection.difference(current_selection)) > 0

and operation_type == "union"

)

or (

len(current_selection.difference(selection)) > 0

and operation_type != "union"

)

)

return ((selection - current_selection) and operation_type == "union") or

(current_selection - selection) and operation_type != "union"))

oerc0122 · 2025-03-06T13:35:20Z

MDANSE/Src/MDANSE/Framework/AtomSelector/selector.py

+            if isinstance(v0, dict):
+                self.set_selection(number=k0, function_parameters=v0)
+            else:
+                raise TypeError(f"Selection {v0} is not a dictionary.")


Suggested change

if isinstance(v0, dict):

self.set_selection(number=k0, function_parameters=v0)

else:

raise TypeError(f"Selection {v0} is not a dictionary.")

if not isinstance(v0, dict):

raise TypeError(f"Selection {v0} is not a dictionary.")

self.set_selection(number=k0, function_parameters=v0)

oerc0122 · 2025-03-06T13:37:01Z

MDANSE/Src/MDANSE/Framework/AtomSelector/spatial_selection.py

+    if position_minimum is None:
+        lower_limits = np.array(3 * [-np.inf])
+    else:
+        lower_limits = np.array(position_minimum)
+    if position_maximum is None:
+        upper_limits = np.array(3 * [np.inf])
+    else:
+        upper_limits = np.array(position_maximum)


Suggested change

if position_minimum is None:

lower_limits = np.array(3 * [-np.inf])

else:

lower_limits = np.array(position_minimum)

if position_maximum is None:

upper_limits = np.array(3 * [np.inf])

else:

upper_limits = np.array(position_maximum)

lower_limits = (np.array(position_minimum)

if position_minimum is not None else

np.array([-np.inf] * 3))

upper_limits = (np.array(position_maximum)

if position_maximum is not None else

np.array([np.inf] * 3))

oerc0122 · 2025-03-06T13:39:02Z

MDANSE/Src/MDANSE/Framework/Configurators/AtomSelectionConfigurator.py

+        dict[str, list[int]]
+            For each atom type, a list of indices of selected atoms
+
+        """
        indicesPerElement = {}


Could be simplified to:

indicesPerElement = defaultdict(list) for i, v in enumerate(self["names"]): indicesPerElement[v].extend(self["indices"][i])

ChiCheng45 · 2025-03-06T16:03:28Z

Some issues I've found and comments.

The default selection string in the gui {"all": true}, this causes the job to crash.
Can the invert button be on its own line? It's slightly confusing as it's in the ALL atoms section.
Selecting or deselecting in the 3D view doesn't update the selection setting.
Odd behaviours with the atom transmutation and the reset.

Invalid SMARTS causes the job to fail.
The setting dragging causes the setting to disappear.

The setting {"0": {"function_name": "invert_selection"}} is ambiguous; at the moment, it's equivalent to selecting all atoms. I guess this is fine but maybe it should be an invalid setting similarly to something like {"0": {"function_name": "select_atoms", "atom_types": ["H"], "operation_type": "difference"}}.

MBartkowiakSTFC · 2025-03-06T16:59:02Z

Some issues I've found and comments.

The default selection string in the gui {"all": true}, this causes the job to crash.

Fixed.

Can the invert button be on its own line? It's slightly confusing as it's in the ALL atoms section.

I could argue that it is an operation on all atoms, since inverted selection = ALL atom indices - current selection
But this is mainly because I think it would take a lot of space in the layout, will be different to all the other widgets (since it is just one button) and it may still be hard to find after all this.

Selecting or deselecting in the 3D view doesn't update the selection setting.

Now that the selection is a sequence of operations, it is less clear how the clicks should be handled. Two options I can think of:

each click creates its own selection operation on a single atom
while you are clicking, indices are added to a single operation. If you do something else in the meantime, a new selection operation will be added next time you click.
Suggestions welcome.

Odd behaviours with the atom transmutation and the reset.

This is weird and I have not found the reason for it.

Invalid SMARTS causes the job to fail.

Fixed.

The setting dragging causes the setting to disappear.

Most likely I will need to subclass QListView to fix this. For now I will disable dragging.

The setting {"0": {"function_name": "invert_selection"}} is ambiguous; at the moment, it's equivalent to selecting all atoms. I guess this is fine but maybe it should be an invalid setting similarly to something like {"0": {"function_name": "select_atoms", "atom_types": ["H"], "operation_type": "difference"}}.

I could make sure that the selection widget always starts with an explicit "select all" instead of being empty. Because at the moment empty selection is assumed to mean "select all", but the first selection replaces it.

ChiCheng45 · 2025-03-06T17:14:30Z

Since we are letting the user enter SMARTS patterns for matching, I think we need to try to tackle #286 at some point. Currently, we don't have bonding information, so users can only use "C~C" for two carbon atoms bonded using any bond ("~") and things like "c:c" (aromatic carbons joined by an aromatic bond) and "c-c" (aromatic carbons joined by a single bond (e.g. biphenyl)) currently do not work.

MBartkowiakSTFC · 2025-03-06T17:25:05Z

Some issues I've found and comments.

The default selection string in the gui {"all": true}, this causes the job to crash.

Can the invert button be on its own line? It's slightly confusing as it's in the ALL atoms section.

Selecting or deselecting in the 3D view doesn't update the selection setting.

Odd behaviours with the atom transmutation and the reset.

Invalid SMARTS causes the job to fail.

The setting dragging causes the setting to disappear.

The setting {"0": {"function_name": "invert_selection"}} is ambiguous; at the moment, it's equivalent to selecting all atoms. I guess this is fine but maybe it should be an invalid setting similarly to something like {"0": {"function_name": "select_atoms", "atom_types": ["H"], "operation_type": "difference"}}.

OK, I think that I managed to improve matters regarding points 1, 4, 5, 6 and 7.
I am happy to ignore point 2 for now, just because it is not urgent. We can agree on a solution and fix it.
Point 3 will require more work. I think I will open a new issue.

MBartkowiakSTFC added 16 commits February 11, 2025 10:53

Merge branch 'protos' of https://github.com/ISISNeutronMuon/MDANSE in…

fdf460f

…to maciej/simplify-atom-selection

Add new selection functions

a32a0f9

Merge branch 'protos' of https://github.com/ISISNeutronMuon/MDANSE in…

ffe37cf

…to maciej/simplify-atom-selection

Merge branch 'protos' of https://github.com/ISISNeutronMuon/MDANSE in…

054657e

…to maciej/simplify-atom-selection

Add first unit tests for selection

881d4c6

Add tests for molecule and group selection

f21f07e

Add tests for selection saving

438a462

Update configurators to use the new selection mechanism

4f10e0f

Update configurators to use the new selection mechanism

f4042e0

Consistently rename load_from_json method

d2fa0cd

Make all tests pass

10263c8

Add a selection mechanism via widgets

2abad53

Add atom selection widgets

5371440

Add string parsing to index range selection widget

d8604f9

Add new selectors and change layout in AtomSelectionWidget

edaedc7

Save selection as array of bool values

795a7b6

MBartkowiakSTFC marked this pull request as draft February 14, 2025 17:12

oerc0122 reviewed Feb 18, 2025

View reviewed changes

MBartkowiakSTFC and others added 11 commits February 19, 2025 09:39

Merge branch 'protos' of https://github.com/ISISNeutronMuon/MDANSE in…

10581a8

…to maciej/simplify-atom-selection

Apply suggestions from code review

16c56f9

Co-authored-by: Jacob Wilkins <46597752+oerc0122@users.noreply.github.com> Signed-off-by: Maciej Bartkowiak <108934199+MBartkowiakSTFC@users.noreply.github.com>

Merge branch 'maciej/simplify-atom-selection' of https://github.com/I…

1013a97

…SISNeutronMuon/MDANSE into maciej/simplify-atom-selection

Adjust code after review suggestions

e1f971b

Correct the docstrings of selection functions

727fe53

Add inversion to the GUI

f517419

Add position selection widget

d20fa64

Add cube and sphere selection to the GUI

aaf0b68

Format code using black

47b2a8e

Convert list to set in select_atoms

e1ffb96

Merge branch 'protos' of https://github.com/ISISNeutronMuon/MDANSE in…

a5d2fd1

…to maciej/simplify-atom-selection

oerc0122 reviewed Feb 28, 2025

View reviewed changes

MBartkowiakSTFC and others added 6 commits February 28, 2025 18:08

Rewrite selection tests using parametrize

5253823

Apply suggestions from code review

d117d6e

Co-authored-by: Jacob Wilkins <46597752+oerc0122@users.noreply.github.com> Signed-off-by: Maciej Bartkowiak <108934199+MBartkowiakSTFC@users.noreply.github.com>

Merge branch 'protos' of https://github.com/ISISNeutronMuon/MDANSE in…

e463a9f

…to maciej/simplify-atom-selection

Merge branch 'maciej/simplify-atom-selection' of https://github.com/I…

c3fe813

…SISNeutronMuon/MDANSE into maciej/simplify-atom-selection

Respond to review, correct selection functions

baac644

Improve docstrings, add all_indices property

05e0bd8

oerc0122 reviewed Mar 4, 2025

View reviewed changes

MBartkowiakSTFC added 3 commits March 5, 2025 12:20

Merge branch 'protos' of https://github.com/ISISNeutronMuon/MDANSE in…

35db18d

…to maciej/simplify-atom-selection

Format the code according to linter

4eab203

Format the GUI code according to linter

3b61207

This was referenced Mar 5, 2025

[ENHANCEMENT] Atom selection saving #707

Open

[ENHANCEMENT] Quick trajectory operations for atom selection #708

Open

MBartkowiakSTFC added 2 commits March 6, 2025 10:20

Use keyword arguments in set_selection

77f6c4c

Implement review suggestions

970c49b

MBartkowiakSTFC mentioned this pull request Mar 6, 2025

Atom selection in scripts ISISNeutronMuon/MDANSE-Examples#15

Open

MBartkowiakSTFC requested a review from oerc0122 March 6, 2025 12:24

oerc0122 approved these changes Mar 6, 2025

View reviewed changes

MBartkowiakSTFC added 2 commits March 6, 2025 15:38

Add style corrections

c9d8acc

Validate SMARTS input

d3dbca4

Select all atoms explicitly, disable dragging

97b4a5a

Reset internal selection in the transmuter

b022132

MBartkowiakSTFC marked this pull request as ready for review March 6, 2025 17:25

MBartkowiakSTFC mentioned this pull request Mar 6, 2025

[BUG] Clicking in 3D view does not work with the new atom selection #709

Open

MBartkowiakSTFC merged commit 1dda714 into protos Mar 7, 2025
38 checks passed

MBartkowiakSTFC deleted the maciej/simplify-atom-selection branch March 7, 2025 09:24

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

New atom selection mechanism #672

New atom selection mechanism #672

MBartkowiakSTFC commented Feb 14, 2025

ChiCheng45 commented Feb 18, 2025 •

edited

Loading

oerc0122 Feb 18, 2025

oerc0122 Feb 18, 2025

MBartkowiakSTFC Feb 21, 2025

oerc0122 Feb 21, 2025

oerc0122 Feb 18, 2025

oerc0122 Feb 18, 2025

oerc0122 Feb 28, 2025

oerc0122 Feb 28, 2025

oerc0122 Feb 28, 2025

oerc0122 Feb 28, 2025

oerc0122 Feb 28, 2025

oerc0122 Feb 28, 2025

oerc0122 Mar 4, 2025

oerc0122 Mar 4, 2025

oerc0122 Mar 4, 2025

oerc0122 Mar 4, 2025

oerc0122 Mar 4, 2025

oerc0122 Mar 4, 2025

oerc0122 Mar 4, 2025

oerc0122 left a comment

oerc0122 Mar 6, 2025

oerc0122 Mar 6, 2025

oerc0122 Mar 6, 2025

oerc0122 Mar 6, 2025

oerc0122 Mar 6, 2025

ChiCheng45 commented Mar 6, 2025 •

edited

Loading

MBartkowiakSTFC commented Mar 6, 2025

ChiCheng45 commented Mar 6, 2025 •

edited

Loading

MBartkowiakSTFC commented Mar 6, 2025



		def select_atoms(
		trajectory: Trajectory, **function_parameters: Dict[str, Any]



		def select_all(
		trajectory: Trajectory, **function_parameters: Dict[str, Any]

		@@ -137,9 +135,8 @@ def update_transmutation_textbox(self) -> None:
		map = self.transmuter.get_setting()

	True if the selection adds atoms, False otherwise
	True if the operation changes selection, False otherwise

	def get_selector(self) -> "ReusableSelection":
	def get_selector(self) -> ReusableSelection:

	MALFORMED_SELECTION = "This is not a valid JSON string."
	MALFORMED_SELECTION = "This is not a valid selection string."

New atom selection mechanism #672

New atom selection mechanism #672

Conversation

MBartkowiakSTFC commented Feb 14, 2025

ChiCheng45 commented Feb 18, 2025 • 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

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

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

oerc0122 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

ChiCheng45 commented Mar 6, 2025 • edited Loading

MBartkowiakSTFC commented Mar 6, 2025

ChiCheng45 commented Mar 6, 2025 • edited Loading

MBartkowiakSTFC commented Mar 6, 2025

ChiCheng45 commented Feb 18, 2025 •

edited

Loading

ChiCheng45 commented Mar 6, 2025 •

edited

Loading

ChiCheng45 commented Mar 6, 2025 •

edited

Loading