Add `ifail` `enum` by chris-ashe · Pull Request #4275 · ukaea/PROCESS

chris-ashe · 2026-05-27T15:19:59Z

Description

Checklist

I confirm that I have completed the following checks:

My changes follow the PROCESS style guide
I have justified any large differences in the regression tests caused by this pull request in the comments.
I have added new tests where appropriate for the changes I have made.
If I have had to change any existing unit or integration tests, I have justified this change in the pull request comments.
If I have made documentation changes, I have checked they render correctly.
I have added documentation for my change, if appropriate.

codecov-commenter · 2026-05-27T15:30:07Z

Codecov Report

❌ Patch coverage is 41.66667% with 21 lines in your changes missing coverage. Please review.
✅ Project coverage is 48.65%. Comparing base (616f749) to head (0f3cc80).
⚠️ Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
process/core/scan.py	9.09%	10 Missing ⚠️
process/core/io/vary_run/tools.py	25.00%	3 Missing ⚠️
process/core/solver/solver_handler.py	25.00%	3 Missing ⚠️
process/core/io/plot/scans.py	33.33%	2 Missing ⚠️
process/core/io/vary_run/config.py	33.33%	2 Missing ⚠️
process/core/final.py	50.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #4275      +/-   ##
==========================================
+ Coverage   48.63%   48.65%   +0.02%     
==========================================
  Files         151      151              
  Lines       29673    29686      +13     
==========================================
+ Hits        14432    14445      +13     
  Misses      15241    15241

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

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

Copilot

Pull request overview

This PR introduces a typed IntEnum for solver ifail exit conditions (SolverOutputCondition) and updates several runtime and test call sites to use the enum instead of hard-coded numeric values, improving readability and consistency around solver status handling.

Changes:

Added SolverOutputCondition enum to centralize solver output (ifail) codes.
Replaced ifail == 1 / ifail != 1 checks with SolverOutputCondition.CONVERGED in core, tracking, and vary-run utilities.
Updated regression/integration tests to assert against the enum values.

Reviewed changes

Copilot reviewed 9 out of 9 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
tracking/tracking_data.py	Uses `SolverOutputCondition.CONVERGED` for strict convergence checks when reading MFILEs.
tests/regression/test_process_input_files.py	Updates regression assertion to use `SolverOutputCondition.CONVERGED`.
tests/integration/test_vmcon.py	Replaces numeric `ifail` expectations with enum values.
process/data_structure/numerics.py	Adds the new `SolverOutputCondition` `IntEnum` definition.
process/core/solver/solver_handler.py	Uses enum values in VMCON retry logic and imports the enum.
process/core/scan.py	Uses enum values in optimisation/solver status reporting and VMCON error handling.
process/core/io/vary_run/tools.py	Uses enum values when determining feasibility/convergence from MFILE contents.
process/core/io/vary_run/config.py	Uses enum values to report convergence status and generate README status messages.
process/core/final.py	Uses enum value for final feasible/unfeasible output header selection.

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

@@ -73,7 +74,7 @@ def run(self):
                # to next attempt


+class SolverOutputCondition(IntEnum):
+    """Enum for the possible conditions that can be returned by the solvers.
+    This is for the `ifail` condition
+    """


+    IMPROPER_INPUT = 0
+    """Solver failed due to improper input (e.g. invalid parameters, or failure to
+    satisfy solver preconditions)"""
+    CONVERGED = 1
+    """Solver converged successfully"""
+
+    MAX_ITERATIONS = 2
+    """Solver failed to converge within the maximum number of iterations"""
+
+    MAX_LINE_SEARCHES = 3
+    """Line search required 10 function calls without finding a better solution"""
+
+    UPHILL_SEARCH = 4
+    """Uphill search direction was calculated"""
+
+    NO_SOLUTION = 5
+    """No feasible solution or bad approximation of Hessian"""
+
+    SINGLE_MATRIX_OR_BOUNDS = 6
+    """Singular matrix in quadratic subproblem or restriction by artificial bounds"""


clmould

Some docs points:

Can the SolverOutputCondition enum be included in the vmcon page in the docs, eg something like

a brief sentence to say that the ifail values are now contained in this enum
an extra column in the ifail table to link the ifail values to the enum

Also the troubleshooting docs page mentions ifail = 3 - think it'd be worth adding a reference to the the enum for this value and adding a brief description of what this means here

clmould · 2026-06-19T11:21:22Z

            )
            process_output.oblnkl(constants.IOTTY)

            logger.critical("Solver returns with ifail /= 1. %s", ifail)


Can the SolverOutputCondition also be incorporated into this error message as it would make the message more descriptive for a user to see

timothy-nunn · 2026-06-24T10:45:21Z

+    MAX_LINE_SEARCHES = 3
+    """Line search required 10 function calls without finding a better solution"""
+
+    UPHILL_SEARCH = 4
+    """Uphill search direction was calculated"""


Might be worth noting that these are solver specific and won't always be raised (e.g. using fsolve or some custom solver)

…clarity and maintainability

Co-authored-by: clmould <86794332+clmould@users.noreply.github.com>

…rity and consistency

timothy-nunn

This method should be typed to return the enum

PROCESS/process/core/solver/solver.py

Line 131 in b211d2c

def solve(self) -> int:
The same with the VMCON solve method. I would then also update the setting of self.info in this method to use the enum:

PROCESS/process/core/solver/solver.py

Line 173 in b211d2c

def solve(self) -> int:
You can probably remove all references to UPHILL_SEARCH because it is not used anymore as far as I can tell

timothy-nunn · 2026-07-14T07:56:17Z

+    """
+
+    USER_TERMINATED = -1
+    # Solver was terminated by the user (e.g. via a keyboard interrupt)


Can you make these comments into docstrings so that they show up in the docs/IDE

timothy-nunn · 2026-07-14T07:57:19Z

+    # Uphill search direction was calculated
+
+    NO_SOLUTION = 5
+    # No feasible solution or bad approximation of Hessian


Suggested change

# No feasible solution or bad approximation of Hessian

# No feasible solution or bad approximation of Hessian (in VMCON only).

Solvers that don't approximate the Hessian will only raise a 5 for no solution

timothy-nunn · 2026-07-14T08:00:05Z

+    MAX_LINE_SEARCHES = 3
+    # Line search required 10 function calls without finding a better solution


This one is solver specific, most solvers do not have an integrated line search

clmould · 2026-07-13T17:24:47Z

    c1(x1,x2) = x1 + x2 - 3 = 0
    c2(x1,x2) = -x1**2/4 - x2**2 + 1 >= 0

    Note that this test is supposed to fail with ifail=5


Suggested change

Note that this test is supposed to fail with ifail=SolverOutputCondition.NO_SOLUTION (ifail = 5)

clmould · 2026-07-14T08:56:33Z


 <a name="table-1"></a>

 |  `ifail`  |  Description  |  Meaning  | Recommendation  |


Can this have a separate column for the SolverOutputCondition please? Otherwise the words are rendering across multiple lines so a new column would be easier to read:

clmould · 2026-07-14T08:58:19Z

            )
            process_output.oblnkl(constants.IOTTY)

            logger.critical("Solver returns with ifail /= 1. %s", ifail)


Can this message also incorporate the SolverOutputCondition please

chris-ashe added Convergence Solver or convergence-related problems Refactor labels May 27, 2026

chris-ashe marked this pull request as ready for review June 2, 2026 13:45

chris-ashe requested a review from a team as a code owner June 2, 2026 13:45

Copilot AI review requested due to automatic review settings June 2, 2026 13:45

Copilot started reviewing on behalf of chris-ashe June 2, 2026 13:45 View session

Copilot AI reviewed Jun 2, 2026

View reviewed changes

clmould self-assigned this Jun 5, 2026

chris-ashe force-pushed the add_ifail_enum branch from 6764908 to 6396b58 Compare June 19, 2026 11:00

clmould requested changes Jun 19, 2026

View reviewed changes

timothy-nunn reviewed Jun 24, 2026

View reviewed changes

chris-ashe requested a review from clmould July 9, 2026 14:18

chris-ashe and others added 6 commits July 13, 2026 11:31

Add ifail enum to define solver output conditions

c9c9b9b

Refactor ifail checks to use SolverOutputCondition enum for improved …

ceb1aeb

…clarity and maintainability

Post rebase fixes

766b1f4

Update process/data_structure/numerics.py

30a49e3

Co-authored-by: clmould <86794332+clmould@users.noreply.github.com>

Refactor ifail handling to use SolverOutputCondition enum for cla…

02fa7b5

…rity and consistency

Refactor ifail handling to use SolverOutputCondition enum for cla…

9e8ad48

…rity and consistency

chris-ashe force-pushed the add_ifail_enum branch from ace13ef to 9e8ad48 Compare July 13, 2026 10:32

Post rebase pre commit fix

0f3cc80

timothy-nunn requested changes Jul 14, 2026

View reviewed changes

clmould requested changes Jul 14, 2026

View reviewed changes

	# No feasible solution or bad approximation of Hessian
	# No feasible solution or bad approximation of Hessian (in VMCON only).

		MAX_LINE_SEARCHES = 3
		# Line search required 10 function calls without finding a better solution


	Note that this test is supposed to fail with ifail=SolverOutputCondition.NO_SOLUTION (ifail = 5)


		<a name="table-1"></a>

		\| `ifail` \| Description \| Meaning \| Recommendation \|

Uh oh!

Conversation

chris-ashe commented May 27, 2026

Description

Checklist

Uh oh!

codecov-commenter commented May 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

clmould left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

timothy-nunn left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

codecov-commenter commented May 27, 2026 •

edited

Loading