feat: Initial bulk of type checking + setup by Gobot1234 · Pull Request #4761 · ansys/pyfluent

Gobot1234 · 2025-12-18T01:55:54Z

Context

Duplicate of #4500 but on the ansys remote. Looking at this code more, I realise I've ran pyupgrade on this so don't merge this till I get the rest of the ruff stuff set up. I'm also realising there are 3 issues tied up in this, apologies for that.
Fixes #4489 and fixes #4490
Helps with #4543

Change Summary

Added a bunch of initial types for the public facing library part of #4738

Gobot1234 · 2025-12-18T01:57:02Z

src/ansys/fluent/core/launcher/container_launcher.py

-        }
-        self.argvals, self.new_session = _get_argvals_and_session(argvals)
-        if self.argvals["start_timeout"] is None:
+        self.argvals, self.new_session = _get_argvals_and_session(


This was changed as sadly there's no way to type transform a typedict to have optional keys (python isn't typescript, sad)

src/ansys/fluent/core/launcher/launcher.py

src/ansys/fluent/core/session_pure_meshing.py

src/ansys/fluent/core/session_solver.py

Gobot1234 · 2025-12-18T02:03:02Z

I'll try and get to the conflicts soonish, probably more useful to get the ruff changes in first

…s/pyfluent into jhilton/typing-improvements

…ements

src/ansys/fluent/core/launcher/slurm_launcher.py

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

Copilot

Pull request overview

Copilot reviewed 93 out of 95 changed files in this pull request and generated 1 comment.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-02-18T10:38:14Z

src/ansys/fluent/core/launcher/launcher.py

    :class:`~ansys.fluent.core.session_solver.Solver`, \
-    :class:`~ansys.fluent.core.session_solver_icing.SolverIcing`, dict]
+    :class:`~ansys.fluent.core.session_solver_icing.SolverIcing`, tuple[str, str]]
        Session object or configuration dictionary if ``dry_run = True``.


The launch_fluent function has default values for dimension and precision but these should match the docstring and typical usage. The defaults are set to Dimension.THREE and Precision.DOUBLE, which matches issue #4490's goal of providing informative defaults instead of None. However, the function signature shows these defaults are provided, but the docstring in the Returns section still mentions "configuration dictionary if dry_run = True" when it should say "tuple[str, str] if dry_run = True" to match the new type annotation.

Suggested change

Session object or configuration dictionary if ``dry_run = True``.

Session object or tuple[str, str] if ``dry_run = True``.

Gobot1234 · 2026-02-23T16:06:09Z

@seanpearsonuk can this be merged?

Copilot

Pull request overview

Copilot reviewed 94 out of 96 changed files in this pull request and generated 3 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-02-23T16:08:38Z

src/ansys/fluent/core/launcher/container_launcher.py


-    def __call__(self):
-
+    def __call__(self) -> "Meshing | PureMeshing | Solver | SolverIcing | SolverAero":


DockerLauncher.__call__ is annotated to return only session types, but the dry_run branch returns config_dict (a dict). This makes the public typing incorrect (and inconsistent with launch_fluent() docs). Update the return annotation (and related overloads) to include the dict type, or change dry_run behavior to return a session/tuple consistently.

Suggested change

def __call__(self) -> "Meshing | PureMeshing | Solver | SolverIcing | SolverAero":

def __call__(self) -> "Meshing | PureMeshing | Solver | SolverIcing | SolverAero | dict[str, Any]":

Copilot · 2026-02-23T16:08:39Z

src/ansys/fluent/core/launcher/launcher.py

+    *,
+    dry_run: Literal[True],
+    **kwargs: Unpack[LaunchFluentArgs],
+) -> tuple[str, str]: ...


The launch_fluent(..., dry_run=True, ...) overload always claims a tuple[str, str] return, but container dry-runs return a dict (from DockerLauncher.__call__). Add overload(s) that distinguish standalone vs container dry-run returns, and include the container dict return type in the function return union.

Suggested change

) -> tuple[str, str]: ...

) -> tuple[str, str] | dict[str, Any]: ...

Copilot · 2026-02-23T16:08:39Z

src/ansys/fluent/core/launcher/launcher.py

+            assert_never(fluent_launch_mode)
+
+    return cast(
+        Meshing | PureMeshing | Solver | SolverIcing | SolverAero | tuple[str, str],


Gobot1234 · 2026-02-27T18:55:46Z

Note to self dimension should be removed from SolverAero, attempting to set anything other than 3 doesn't work

…ements

Copilot

Pull request overview

Copilot reviewed 94 out of 96 changed files in this pull request and generated no new comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot

Pull request overview

Copilot reviewed 94 out of 96 changed files in this pull request and generated 3 comments.

Comments suppressed due to low confidence (2)

src/ansys/fluent/core/launcher/container_launcher.py:285

DockerLauncher.__call__ is annotated to return only session types, but when dry_run is true it returns config_dict (a dict). This makes the method signature inaccurate and undermines the typing work in this PR. Update the return annotation to include the dry-run return type (and ideally tighten it to dict[str, Any]), or change the dry-run behavior to match the annotated return type.

    def __call__(self) -> "Meshing | PureMeshing | Solver | SolverIcing | SolverAero":
        if self.argvals["dry_run"]:
            config_dict, *_ = configure_container_dict(
                self._args,
                compose_config=self._compose_config,
                **self.argvals["container_dict"],
            )
            dict_str = dict_to_str(config_dict)
            print("\nDocker container run configuration:\n")
            print("config_dict = ")
            print(dict_str)
            return config_dict

src/ansys/fluent/core/launcher/launcher.py:125

The create_launcher() docstring says it raises ValueError for unknown launch modes, but the implementation raises DisallowedValuesError. Either update the docstring to reflect the actual exception type or change the code to raise ValueError as documented.

    Raises
    ------
    ValueError
        If an unknown Fluent launch mode is passed.
    """
    launchers = {
        LaunchMode.STANDALONE: StandaloneLauncher,
        LaunchMode.CONTAINER: DockerLauncher,
        LaunchMode.PIM: PIMLauncher,
        LaunchMode.SLURM: SlurmLauncher,
    }

    if fluent_launch_mode in launchers:
        return launchers[fluent_launch_mode](**kwargs)
    else:
        raise DisallowedValuesError(
            "launch mode",
            fluent_launch_mode,
            [f"LaunchMode.{m.name}" for m in LaunchMode],
        )

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

src/ansys/fluent/core/launcher/watchdog_exec

        ip, port, password = sys.argv[2:5]
-        allow_remote_host, certificates_folder, insecure_mode = sys.argv[6:9]
+        allow_remote_host, certificates_folder, insecure_mode, inside_container = (
+            sys.argv[6:10]
+        )


.github/workflows/ci.yml

+      - name: "Make venv"
+        run: uv venv --clear
+
+      - name: "Install extras"
+        run: uv pip install -e ".[ci_types]"
+


src/ansys/fluent/core/search.py

 @functools.cache
 def _get_api_tree_data():
    """Get API tree data."""
    api_tree_data_file_path = _get_api_tree_data_file_path()
    if api_tree_data_file_path.exists():
-        json_file = open(api_tree_data_file_path, "r")
+        json_file = open(api_tree_data_file_path)
        api_tree_data = json.load(json_file)
        return api_tree_data


Copilot

Pull request overview

Copilot reviewed 94 out of 96 changed files in this pull request and generated 4 comments.

Comments suppressed due to low confidence (1)

src/ansys/fluent/core/services/_protocols.py:38

ServiceProtocol currently defines an __init__(channel, metadata) signature (under TYPE_CHECKING), but many concrete service classes inheriting it require additional constructor parameters (for example fluent_error_state). Type checkers typically treat protocol inheritance like normal inheritance for signature compatibility, so this is likely to introduce a large number of incompatible-override errors (and StreamingService also doesn’t match channel/metadata). Consider removing __init__ from the protocol entirely, or widening it to accept *args, **kwargs (or adding overloads) so all service implementations can conform.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-03-23T14:13:03Z

src/ansys/fluent/core/field_data_interfaces.py

+    "VectorFieldDataRequest",
+)
+
+_to_field_name_str = naming_strategy().to_string


_to_field_name_str is assigned twice in a row; the first assignment is immediately overwritten. This is dead code and can confuse readers about which naming strategy instance is intended. Consider keeping only one assignment (either create a single _naming_strategy_instance and derive _to_field_name_str from it, or inline the call once).

Suggested change

_to_field_name_str = naming_strategy().to_string

Copilot · 2026-03-23T14:13:04Z

src/ansys/fluent/core/search.py

    api_tree_data_file_path = _get_api_tree_data_file_path()
    if api_tree_data_file_path.exists():
-        json_file = open(api_tree_data_file_path, "r")
+        json_file = open(api_tree_data_file_path)
        api_tree_data = json.load(json_file)
        return api_tree_data


_get_api_tree_data() opens the JSON file without a context manager, so the file handle may remain open until GC. Use with open(...) as f: (or explicitly close the file) to avoid leaking file descriptors, especially since this function is cached and could be called in long-lived processes.

Copilot · 2026-03-23T14:13:04Z

src/ansys/fluent/core/__init__.py

+from ansys.fluent.core.module_config import *

 # Logging has to be imported before importing other PyFluent modules
-from ansys.fluent.core.logger import set_console_logging_level  # noqa: F401
+from ansys.fluent.core.logger import *



Using from ... import * in the package ansys.fluent.core.__init__ makes the exported namespace depend entirely on each submodule’s __all__ and also stops submodules themselves (e.g., module_config, logger, services, etc.) from being exposed as attributes unless they’re explicitly imported as modules. This conflicts with the public API expectations in tests/test_public_api.py (it expects many submodules to be present in dir(ansys.fluent.core)) and is likely to cause API regressions. Consider switching back to explicit imports/aliases for public symbols and explicitly importing the intended public submodules (or defining a package-level __all__) instead of wildcard imports.

Copilot · 2026-03-23T14:13:04Z

tests/test_public_api.py

+        "data_model_cache",
+        "docker",
+        "examples",
+        "exceptions",
+        "field_data_interfaces",
+        "filereader",
+        "fluent_connection",
+        "generated",
+        "get_build_details",
+        "get_build_version",
+        "get_build_version_string",
+        "journaling",
+        "launch_fluent",
+        "launcher",
+        "logger",
+        "module_config",
+        "parametric",
+        "pyfluent_warnings",


This test assumes many submodules (e.g., docker, exceptions, module_config, logger, etc.) are present in dir(ansys.fluent.core), but the updated ansys.fluent.core.__init__ now mainly does wildcard imports of selected symbols and does not import most submodules as module attributes. As written, this will likely report a large set of “missing symbols” even if the submodules still exist as importable packages. Consider either (1) explicitly importing/exposing those submodules in ansys.fluent.core.__init__, or (2) changing this test to validate importability via importlib.import_module('ansys.fluent.core.<name>') rather than relying on dir().

Gobot1234 and others added 10 commits November 5, 2025 10:21

Add type level support for passing Path instances to download_file

4f2c990

Add typing_extensions

1ddf8e8

Fix inheritance for Meshing

7825ff4

Fix create_launcher allowing None

90d69a0

Update launchers typing to be more specific

654deb4

Fix ci

ec258c3

Fix ci again

00c3a85

remove accidental commits

b6d2e3e

run pyupgrade for moving typing to builtins and a couple of other things

0a48f90

fix some np.arrays that are meant to be ndarrays in types

7b36a1c

Gobot1234 commented Dec 18, 2025

View reviewed changes

mkundu1 mentioned this pull request Dec 30, 2025

100% type hints in PyFluent code #4738

Open

Gobot1234 added 5 commits January 5, 2026 16:50

pre-fix the conflict that will happen

c04b413

add types for services

08e5525

Merge branch 'jhilton/typing-improvements' of https://github.com/ansy…

7c2e9c3

…s/pyfluent into jhilton/typing-improvements

fix some low hanging fruit

0acf4ea

Merge remote-tracking branch 'origin/main' into jhilton/typing-improv…

e6a7bf9

…ements

github-actions bot added documentation Documentation related (improving, adding, etc) examples Publishing PyFluent examples maintenance General maintenance of the repo (libraries, cicd, etc) dependencies Related to dependencies labels Jan 15, 2026

Gobot1234 commented Jan 15, 2026

View reviewed changes

src/ansys/fluent/core/launcher/slurm_launcher.py Outdated Show resolved Hide resolved

Gobot1234 marked this pull request as ready for review January 15, 2026 03:19

Gobot1234 requested review from hpohekar, mkundu1, prmukherj and seanpearsonuk as code owners January 15, 2026 03:19

Copilot AI review requested due to automatic review settings January 15, 2026 03:19

Gobot1234 requested a review from mayankansys as a code owner January 15, 2026 03:19

Gobot1234 and others added 2 commits February 18, 2026 01:47

Update slightly silly test

2de9fc2

Update examples/00-fluent/DOE_ML.py

c8e8999

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

Copilot AI review requested due to automatic review settings February 18, 2026 10:35

Update examples/00-fluent/DOE_ML.py

9517350

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

Copilot started reviewing on behalf of Gobot1234 February 18, 2026 10:36 View session

Copilot AI reviewed Feb 18, 2026

View reviewed changes

Gobot1234 added 2 commits February 23, 2026 16:01

Merge branch 'main' into jhilton/typing-improvements

c751908

Update context_managers.py

5649990

Copilot AI review requested due to automatic review settings February 23, 2026 16:02

Copilot started reviewing on behalf of Gobot1234 February 23, 2026 16:03 View session

Update test_public_api.py

f73bcba

Copilot AI reviewed Feb 23, 2026

View reviewed changes

Merge branch 'main' into jhilton/typing-improvements

d464ad0

Merge remote-tracking branch 'origin/main' into jhilton/typing-improv…

74171c9

…ements

Copilot AI review requested due to automatic review settings March 12, 2026 16:56

Copilot started reviewing on behalf of Gobot1234 March 12, 2026 16:56 View session

Copilot AI reviewed Mar 12, 2026

View reviewed changes

Gobot1234 added 2 commits March 13, 2026 10:31

add tomli

4eb9f20

maybe the reordering fixes it

83c5f72

Copilot AI review requested due to automatic review settings March 17, 2026 13:28

Copilot started reviewing on behalf of Gobot1234 March 17, 2026 13:29 View session

Copilot AI reviewed Mar 17, 2026

View reviewed changes

maybe this fixes it?

317d529

seanpearsonuk approved these changes Mar 17, 2026

View reviewed changes

rebasline?

bfb4e3a

Copilot AI review requested due to automatic review settings March 23, 2026 14:05

Copilot started reviewing on behalf of Gobot1234 March 23, 2026 14:05 View session

Copilot AI reviewed Mar 23, 2026

View reviewed changes

	Session object or configuration dictionary if ``dry_run = True``.
	Session object or tuple[str, str] if ``dry_run = True``.


		def __call__(self):

		def __call__(self) -> "Meshing \| PureMeshing \| Solver \| SolverIcing \| SolverAero":

	) -> tuple[str, str]: ...
	) -> tuple[str, str] \| dict[str, Any]: ...

Conversation

Gobot1234 commented Dec 18, 2025

Context

Change Summary

Uh oh!

Gobot1234 Dec 18, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Gobot1234 commented Dec 18, 2025

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI Feb 18, 2026

Choose a reason for hiding this comment

Uh oh!

Gobot1234 commented Feb 23, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI Feb 23, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 23, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 23, 2026

Choose a reason for hiding this comment

Uh oh!

Gobot1234 commented Feb 27, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI Mar 23, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 23, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 23, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 23, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants