Build whole models from mathspec (whole-model-only) by FabianHofmann · Pull Request #958 · PyPSA/linopy

FabianHofmann · 2026-09-18T06:57:17Z

Note

The following content was generated by AI.

Changes proposed in this Pull Request

A mathspec spec (a YAML declaration of a linear model) builds a whole linopy model from linopy/spec/. mathspec (>=0.4,<0.5, from PyPI, Python >= 3.12) is the optional extra linopy[spec] and imported lazily, so import linopy never pulls it in.

This branch was re-ported onto master (after #1041, #1042, #1051–#1059), migrated to mathspec 0.4 and leaned down. The older 67-commit history is kept under the tag spec-whole-model-v1. The simplification from #1068 is merged into this branch.

API

One constructor. Model.from_spec(spec, sources, **model_kwargs) builds variables, SOS, constraints, the objective and named expressions with a variable term. spec is a Path, YAML text, dict or mathspec.Spec. A one-line str ending in .yaml/.yml raises a ValueError that asks for a Path.
Data lives in model.parameters. Every parameter (unfilled, NaN at holes) and every relation is stored under its declared name. assign_coords, copy and the netcdf round trip therefore carry the spec data. There is no model.spec accessor and no stamp on variables or constraints.
Results are functions of (spec, model). linopy.spec.evaluate(spec, model, *names) folds named expressions over the solution and the parameters. linopy.spec.parameter(spec, model, name) returns filled values with the declared dtype. linopy.spec.__all__ is SpecDataError, evaluate, parameter.
Typesetting is mathspec's own: mathspec.typeset(spec, fmt) and typeset_line.
Rebuild from a saved model. Model.from_spec(path, read_netcdf(f).parameters) rebuilds without the original sources.
linopy.testing.synthetic_sources takes every spec form from_spec takes.

Sources

NaN in a DataArray, Dataset or wide DataFrame is a missing row under the missing: policy. Long tables still refuse NaN; a long table may carry its values in a column named after the parameter.
Sets, extension arrays, Categorical and masked indexes work as dimension sources. Each source key is read once, so a lazy Mapping runs one query per parameter.
Labels that differ only by type get a "differ in type, not spelling" error with a datetime hint.

mathspec 0.4 semantics

Relations are joins: sum(x, over=d, by=rel[c]), at(x, by=rel[c]), sums over several axes. Grouped sums on terms use master's Relation group-by.
missing: per parameter and relation (refused default, neutral, a literal, absent). Absence travels as a mask, never as NaN.
missing: refused only raises for holes a live row demands. A hole is excused under a where: mask, a masked variable, an absent row of another parameter, an unfilled shift edge or a failed lookup. Build and evaluate share one check that names the parameter, the declaration and the missing labels.
ordered:, conditions: and the dual(c) sign convention are supported.

Changes for all models

Object-dtype parameters are stored in netcdf as codes plus categories, so NaN holes round-trip instead of coming back as ''.
Mixed-type object parameters (for example ['a', 1]) raise on write instead of being written as strings.
read_netcdf keeps parameter coordinates that no data variable uses.

Limits

Relations: one key column into one value column only. Bare, multi-column and multi-key relations, and quadratic constraints, are refused with SpecDataError. Follow-up: spec: full mathspec relation joins and quadratic constraints (follow-up to #958) #1063 (needed by mathspec's pypsa.yaml).
Rows that absence leaves empty are refused; linopy is stricter than mathspec's "not built and reported" (spec: rows emptied by absence are dropped silently; revisit strict refusal #993).

Docs, CI, benchmarks

New page doc/spec-data.rst (source contract, SQLite example, holes, rebuild, variants via mathspec). Notebook examples/building-models-from-specs.ipynb, doc/api.rst, user guide and release notes.
CI clones the mathspec examples at the installed version and sweeps operators/, ports/ and the top level. On Python 3.11, where mathspec is not installable, the spec tests skip.
Benchmark benchmarks/models/spec_commitment.py (commitment.yaml plus a sum_by-shaped relation case). codspeed and benchmark-smoke install no spec extra, so these benchmarks skip in CI.

API review findings deferred to issues

Updating data and re-solving (editing model.parameters changes evaluate but not the model); applying a spec to an existing model and composing programs; bare relations and sets derived from data; sparse data densified; data-only named expressions not in model.expressions; lazy per-parameter reads.

Local verification

uv run ruff check . && uv run ruff format --check .     # pass
uv run mypy linopy test                                 # pass
uv run pytest -q --doctest-modules test -k spec         # 405 passed, 15 skipped
# Python 3.11 without mathspec:
uv run --python 3.11 --extra dev pytest --collect-only --doctest-modules linopy test   # 5159 collected

Checklist

AI-generated content is marked (see AGENTS.md).
Code changes are sufficiently documented; i.e. new functions contain docstrings and further explanations may be given in doc.
Unit tests for new features were added (if applicable).
A note for the release notes doc/release_notes.rst of the upcoming release is included.
I consent to the release of this PR's code under the MIT license.

codspeed · 2026-09-18T07:00:16Z

Merging this PR will not alter performance

✅ 181 untouched benchmarks
⏩ 181 skipped benchmarks¹

_{Comparing spec-whole-model (de6530f) with master (9ad435d)}

181 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

github-actions · 2026-09-18T07:00:35Z

Build cost — v1 vs legacy

v1 build peak & time relative to legacy, on this commit — not a comparison against master (that is CodSpeed).

peak — v1 / legacy	time — v1 / legacy

Full table (time + peak, mean)

benchmarks/drivers/test_build.py::test_build[basic-n=10]
                  time (s)         peak (KiB) 
 name                 mean   │           mean 
──────────────────────────────────────────────
 (legacy)   0.09022 (1.10)   │   15.03 (1.00) 
 (v1)        0.08229 (1.0)   │    15.00 (1.0) 

benchmarks/drivers/test_build.py::test_build[basic-n=250]
                  time (s)         peak (MiB) 
 name                 mean   │           mean 
──────────────────────────────────────────────
 (legacy)   0.09851 (1.12)   │   12.04 (1.00) 
 (v1)        0.08763 (1.0)   │    12.04 (1.0) 

benchmarks/drivers/test_build.py::test_build[cumsum-severity=0]
                  time (s)        peak (KiB) 
 name                 mean   │          mean 
─────────────────────────────────────────────
 (legacy)   0.03844 (1.07)   │   15.20 (1.0) 
 (v1)        0.03598 (1.0)   │   15.20 (1.0) 

benchmarks/drivers/test_build.py::test_build[cumsum-severity=100]
                  time (s)        peak (MiB) 
 name                 mean   │          mean 
─────────────────────────────────────────────
 (legacy)   0.05552 (1.06)   │   44.93 (1.0) 
 (v1)        0.05256 (1.0)   │   44.93 (1.0) 

benchmarks/drivers/test_build.py::test_build[cumsum-severity=50]
                  time (s)        peak (MiB) 
 name                 mean   │          mean 
─────────────────────────────────────────────
 (legacy)   0.04195 (1.04)   │   11.51 (1.0) 
 (v1)        0.04034 (1.0)   │   11.51 (1.0) 

benchmarks/drivers/test_build.py::test_build[expression_arithmetic-n=10]
                 time (s)         peak (KiB) 
 name                mean   │           mean 
─────────────────────────────────────────────
 (legacy)   0.1001 (1.06)   │   24.34 (1.06) 
 (v1)       0.09443 (1.0)   │    23.04 (1.0) 

benchmarks/drivers/test_build.py::test_build[expression_arithmetic-n=250]
                 time (s)         peak (MiB) 
 name                mean   │           mean 
─────────────────────────────────────────────
 (legacy)   0.1211 (1.17)   │   16.12 (1.00) 
 (v1)        0.1033 (1.0)   │    16.12 (1.0) 

benchmarks/drivers/test_build.py::test_build[knapsack-n=10000]
                  time (s)          peak (KiB) 
 name                 mean   │            mean 
───────────────────────────────────────────────
 (legacy)   0.02413 (1.07)   │   752.18 (1.10) 
 (v1)         0.0225 (1.0)   │    685.15 (1.0) 

benchmarks/drivers/test_build.py::test_build[knapsack-n=100]
                  time (s)        peak (KiB) 
 name                 mean   │          mean 
─────────────────────────────────────────────
 (legacy)   0.02414 (1.08)   │   3.12 (1.33) 
 (v1)        0.02234 (1.0)   │    2.34 (1.0) 

benchmarks/drivers/test_build.py::test_build[kvl_cycles-severity=0]
                  time (s)          peak (MiB) 
 name                 mean   │            mean 
───────────────────────────────────────────────
 (legacy)   0.06437 (1.21)   │   126.16 (1.44) 
 (v1)        0.05337 (1.0)   │     87.71 (1.0) 

benchmarks/drivers/test_build.py::test_build[kvl_cycles-severity=100]
                  time (s)          peak (MiB) 
 name                 mean   │            mean 
───────────────────────────────────────────────
 (legacy)   0.06332 (1.21)   │   126.16 (1.44) 
 (v1)        0.05235 (1.0)   │     87.71 (1.0) 

benchmarks/drivers/test_build.py::test_build[kvl_cycles-severity=50]
                  time (s)          peak (MiB) 
 name                 mean   │            mean 
───────────────────────────────────────────────
 (legacy)   0.06326 (1.20)   │   126.16 (1.44) 
 (v1)        0.05269 (1.0)   │     87.71 (1.0) 

benchmarks/drivers/test_build.py::test_build[masked-n=100]
                  time (s)          peak (KiB) 
 name                 mean   │            mean 
───────────────────────────────────────────────
 (legacy)   0.05572 (1.02)   │    715.12 (1.0) 
 (v1)        0.05456 (1.0)   │   787.73 (1.10) 

benchmarks/drivers/test_build.py::test_build[masked-n=10]
                  time (s)        peak (KiB) 
 name                 mean   │          mean 
─────────────────────────────────────────────
 (legacy)   0.05359 (1.08)   │   4.54 (1.27) 
 (v1)        0.04942 (1.0)   │    3.57 (1.0) 

benchmarks/drivers/test_build.py::test_build[merge_balance-severity=0]
                 time (s)          peak (KiB) 
 name                mean   │            mean 
──────────────────────────────────────────────
 (legacy)   0.3762 (1.05)   │   704.12 (1.09) 
 (v1)        0.3572 (1.0)   │    643.85 (1.0) 

benchmarks/drivers/test_build.py::test_build[merge_balance-severity=100]
                 time (s)        peak (MiB) 
 name                mean   │          mean 
────────────────────────────────────────────
 (legacy)   0.3941 (1.04)   │   18.34 (1.0) 
 (v1)        0.3798 (1.0)   │   18.34 (1.0) 

benchmarks/drivers/test_build.py::test_build[merge_balance-severity=50]
               time (s)       peak (MiB) 
 name              mean   │         mean 
─────────────────────────────────────────
 (legacy)   0.39 (1.04)   │   9.54 (1.0) 
 (v1)       0.374 (1.0)   │   9.54 (1.0) 

benchmarks/drivers/test_build.py::test_build[milp-n=10]
                  time (s)        peak (KiB) 
 name                 mean   │          mean 
─────────────────────────────────────────────
 (legacy)   0.07643 (1.07)   │   3.77 (1.12) 
 (v1)        0.07111 (1.0)   │    3.37 (1.0) 

benchmarks/drivers/test_build.py::test_build[milp-n=50]
                  time (s)          peak (KiB) 
 name                 mean   │            mean 
───────────────────────────────────────────────
 (legacy)   0.07651 (1.09)   │   216.59 (1.10) 
 (v1)        0.07003 (1.0)   │    196.23 (1.0) 

benchmarks/drivers/test_build.py::test_build[nodal_balance-severity=0]
                  time (s)         peak (KiB) 
 name                 mean   │           mean 
──────────────────────────────────────────────
 (legacy)   0.03872 (1.07)   │   938.49 (1.0) 
 (v1)        0.03603 (1.0)   │   938.49 (1.0) 

benchmarks/drivers/test_build.py::test_build[nodal_balance-severity=100]
                  time (s)       peak (MiB) 
 name                 mean   │         mean 
────────────────────────────────────────────
 (legacy)   0.03975 (1.09)   │   9.66 (1.0) 
 (v1)        0.03659 (1.0)   │   9.66 (1.0) 

benchmarks/drivers/test_build.py::test_build[nodal_balance-severity=50]
                  time (s)       peak (MiB) 
 name                 mean   │         mean 
────────────────────────────────────────────
 (legacy)   0.03942 (1.09)   │   5.32 (1.0) 
 (v1)         0.0362 (1.0)   │   5.32 (1.0) 

benchmarks/drivers/test_build.py::test_build[nodal_balance_sparse-severity=0]
                  time (s)       peak (MiB) 
 name                 mean   │         mean 
────────────────────────────────────────────
 (legacy)   0.01952 (1.00)   │   1.28 (1.0) 
 (v1)        0.01944 (1.0)   │   1.28 (1.0) 

benchmarks/drivers/test_build.py::test_build[nodal_balance_sparse-severity=100]
                  time (s)       peak (MiB) 
 name                 mean   │         mean 
────────────────────────────────────────────
 (legacy)   0.01947 (1.01)   │   1.28 (1.0) 
 (v1)         0.0193 (1.0)   │   1.28 (1.0) 

benchmarks/drivers/test_build.py::test_build[nodal_balance_sparse-severity=50]
                  time (s)       peak (MiB) 
 name                 mean   │         mean 
────────────────────────────────────────────
 (legacy)   0.01948 (1.02)   │   1.28 (1.0) 
 (v1)        0.01919 (1.0)   │   1.28 (1.0) 

benchmarks/drivers/test_build.py::test_build[piecewise-n=1000]
                 time (s)          peak (KiB) 
 name                mean   │            mean 
──────────────────────────────────────────────
 (legacy)   0.1884 (1.02)   │   949.04 (1.06) 
 (v1)        0.1851 (1.0)   │    893.73 (1.0) 

benchmarks/drivers/test_build.py::test_build[piecewise-n=10]
                 time (s)         peak (KiB) 
 name                mean   │           mean 
─────────────────────────────────────────────
 (legacy)   0.1863 (1.05)   │   10.03 (1.07) 
 (v1)        0.1782 (1.0)   │     9.38 (1.0) 

benchmarks/drivers/test_build.py::test_build[qp-n=1000]
                  time (s)          peak (KiB) 
 name                 mean   │            mean 
───────────────────────────────────────────────
 (legacy)   0.04827 (1.08)   │   147.70 (1.06) 
 (v1)        0.04489 (1.0)   │    139.87 (1.0) 

benchmarks/drivers/test_build.py::test_build[qp-n=10]
                  time (s)        peak (KiB) 
 name                 mean   │          mean 
─────────────────────────────────────────────
 (legacy)   0.04751 (1.07)   │   2.60 (1.09) 
 (v1)        0.04431 (1.0)   │    2.38 (1.0) 

benchmarks/drivers/test_build.py::test_build[rolling-severity=0]
                  time (s)          peak (KiB) 
 name                 mean   │            mean 
───────────────────────────────────────────────
 (legacy)   0.03976 (1.09)   │   696.75 (1.03) 
 (v1)        0.03637 (1.0)   │    673.70 (1.0) 

benchmarks/drivers/test_build.py::test_build[rolling-severity=100]
                  time (s)         peak (MiB) 
 name                 mean   │           mean 
──────────────────────────────────────────────
 (legacy)    0.08343 (1.0)   │   137.97 (1.0) 
 (v1)       0.08367 (1.00)   │   137.97 (1.0) 

benchmarks/drivers/test_build.py::test_build[rolling-severity=50]
                  time (s)        peak (MiB) 
 name                 mean   │          mean 
─────────────────────────────────────────────
 (legacy)   0.06422 (1.12)   │   69.22 (1.0) 
 (v1)        0.05726 (1.0)   │   69.22 (1.0) 

benchmarks/drivers/test_build.py::test_build[sos-n=1000]
                  time (s)          peak (KiB) 
 name                 mean   │            mean 
───────────────────────────────────────────────
 (legacy)   0.04559 (1.10)   │   402.33 (1.00) 
 (v1)        0.04161 (1.0)   │    402.30 (1.0) 

benchmarks/drivers/test_build.py::test_build[sos-n=10]
                  time (s)        peak (KiB) 
 name                 mean   │          mean 
─────────────────────────────────────────────
 (legacy)   0.04495 (1.10)   │   3.19 (1.19) 
 (v1)        0.04083 (1.0)   │    2.69 (1.0) 

benchmarks/drivers/test_build.py::test_build[sparse_network-n=10]
                  time (s)         peak (KiB) 
 name                 mean   │           mean 
──────────────────────────────────────────────
 (legacy)   0.05029 (1.07)   │   29.00 (1.54) 
 (v1)        0.04702 (1.0)   │    18.84 (1.0) 

benchmarks/drivers/test_build.py::test_build[sparse_network-n=250]
                  time (s)         peak (MiB) 
 name                 mean   │           mean 
──────────────────────────────────────────────
 (legacy)   0.06051 (1.10)   │   37.95 (1.43) 
 (v1)        0.05504 (1.0)   │    26.51 (1.0) 

benchmarks/drivers/test_build.py::test_build[storage-n=10]
                  time (s)          peak (KiB) 
 name                 mean   │            mean 
───────────────────────────────────────────────
 (legacy)    0.09397 (1.0)   │    410.93 (1.0) 
 (v1)       0.09719 (1.03)   │   427.84 (1.04) 

benchmarks/drivers/test_build.py::test_build[storage-n=250]
                 time (s)         peak (MiB) 
 name                mean   │           mean 
─────────────────────────────────────────────
 (legacy)    0.1014 (1.0)   │     9.94 (1.0) 
 (v1)       0.1038 (1.02)   │   10.22 (1.03)

📊 Interactive plots + CSV: download the semantics-report-v1-vs-legacy artifact from this run.

_{Report-only · not a gate · refreshed on every push · obsolete once legacy is dropped.}

…pec 0.4 Core hooks (spec stamps, add_spec/from_spec, netcdf spec block, expression stamps) in master's shape; linopy.spec rewritten for mathspec 0.4 nodes, joins, missing:, ordered:, conditions:. Phase 4 fixes 28 spec tests.

…trict; fold assumptions and groups)

…enchmark updates

…on replaced objective or undeclared SOS

Fold duplicated helpers (prefixing, netcdf filters, lookups, zeros, gathers, as_expression), drop dead checks and single-use wrappers; align only dense input. Test string parameters given as rows.

…y handles, KeyError for unknown names Fold the binary-operator branches of evaluate into one table, drop the redundant source-key cache and guards in attach, and build the row presence mask with Index.isin.

Keep every parameter unfilled in ModelSpec.parameters and fill on read; drop the .spec stamp, presence dataset, add_spec, lower_text and the file-name sniffing of str specs. Public surface is ModelSpec, NamedExpression, SpecDataError, Declaration and attach. Notebook updated.

from_spec writes raw parameters and relations into model.parameters; linopy.spec.evaluate and linopy.spec.parameter replace ModelSpec. Object parameters round-trip netcdf with NaN holes; grid sources accept NaN as a missing row; sets, extension arrays and named value columns are accepted; str file names are refused with a Path hint. Docs: spec-data page.

A hole in a refused parameter is excused where where: masks, a masked variable, an absent row, an unfilled shift edge or a failed lookup make the term absent anyway. build and evaluate share one check that names the parameter, the declaration and the missing labels.

…dth, unshadow SHIFTED

…True Model.from_spec gains store=False; evaluate takes sources= and otherwise reads model.parameters through the same attach path. Drop linopy.spec.parameter.

…th mathspec join semantics

… term

…ot hold

…tion once per build

FabianHofmann added enhancement New feature or request math-spec YAML/text math spec: schema, from_spec/add_spec, operators, expr container labels Sep 18, 2026

FabianHofmann marked this pull request as ready for review September 18, 2026 07:45

This was referenced Sep 19, 2026

Compose spec fragments into a whole model via given #959

Closed

Build and change models via math-spec #922

Closed

Feat/lazy expression container #900

Closed

FabianHofmann mentioned this pull request Sep 28, 2026

spec: rows emptied by absence are dropped silently; revisit strict refusal #993

Open

FabianHofmann changed the title ~~Build whole models from math-spec (whole-model-only)~~ Build whole models from mathspec (whole-model-only) Sep 28, 2026

brynpickering mentioned this pull request Oct 26, 2026

Link blocks / datarecords with mathspec manta-plan/manta#214

Open

FabianHofmann added 3 commits October 26, 2026 15:41

refactor(spec): lean down linopy.spec (drop drift, retain, Sources, s…

ff7e60e

…trict; fold assumptions and groups)

test(spec): rewrite spec tests for mathspec 0.4; docs, notebook and b…

f168f8f

…enchmark updates

FabianHofmann mentioned this pull request Oct 26, 2026

spec: full mathspec relation joins and quadratic constraints (follow-up to #958) #1063

Open

FabianHofmann force-pushed the spec-whole-model branch from fd3c8d5 to f168f8f Compare October 26, 2026 15:38

FabianHofmann added 5 commits October 26, 2026 20:48

fix(spec): AttributeError for unsolved .solution, export lower, warn …

5bce1c4

…on replaced objective or undeclared SOS

docs: note the Objective.expression copy in the breaking changes

72fe22c

refactor(spec): compactify linopy.spec

e0ca698

Fold duplicated helpers (prefixing, netcdf filters, lookups, zeros, gathers, as_expression), drop dead checks and single-use wrappers; align only dense input. Test string parameters given as rows.

refactor(spec): dataclass ModelSpec, Declaration base for typeset-onl…

cee4093

…y handles, KeyError for unknown names Fold the binary-operator branches of evaluate into one table, drop the redundant source-key cache and guards in attach, and build the row presence mask with Index.isin.

FabianHofmann mentioned this pull request Oct 27, 2026

Simplify linopy.spec: one constructor, data in model.parameters, no accessor #1068

Merged

5 tasks done

FabianHofmann added 2 commits October 27, 2026 11:15

FabianHofmann added 5 commits October 27, 2026 14:08

fix(ci): skip spec_helpers in doctest collection, type-safe roster wi…

8aa04dd

…dth, unshadow SHIFTED

refactor(spec): build from sources, store parameters only with store=…

1b43ea4

…True Model.from_spec gains store=False; evaluate takes sources= and otherwise reads model.parameters through the same attach path. Drop linopy.spec.parameter.

feat(spec): relations from tables: bare, multi-key and multi-value wi…

75c1457

…th mathspec join semantics

feat(spec): build a quadratic constraint whose data leaves no product…

b88a5c3

… term

fix(spec): evaluate points to sources= for a relation store=True cann…

eb7bf6e

…ot hold

FabianHofmann mentioned this pull request Oct 27, 2026

spec: check pypsa.yaml builds against mathspec's PyPSA reference objectives #1075

Open

FabianHofmann added 2 commits October 27, 2026 16:58

test(spec): solve pypsa.yaml on synthetic data its assumptions accept

332633d

perf(spec): fold each named expression, variable, parameter and condi…

de6530f

…tion once per build

FabianHofmann merged commit d2592b5 into master Oct 28, 2026
22 checks passed

FabianHofmann deleted the spec-whole-model branch October 28, 2026 07:31

FabianHofmann mentioned this pull request Oct 28, 2026

perf: dense fast paths in the core and a faster spec build #1078

Merged

4 of 5 tasks

Navigation Menu

Repository navigation

Build whole models from mathspec (whole-model-only) - #958

Build whole models from mathspec (whole-model-only)#958
FabianHofmann merged 17 commits into
masterfrom
spec-whole-model

FabianHofmann commented Sep 18, 2026 •

edited

Loading

Uh oh!

codspeed Bot commented Sep 18, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Sep 18, 2026 •

edited

Loading

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

FabianHofmann commented Sep 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Changes proposed in this Pull Request

API

Sources

mathspec 0.4 semantics

Changes for all models

Limits

Docs, CI, benchmarks

Checklist

Uh oh!

codspeed Bot commented Sep 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Merging this PR will not alter performance

Footnotes

Uh oh!

github-actions Bot commented Sep 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Build cost — v1 vs legacy

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

FabianHofmann commented Sep 18, 2026 •

edited

Loading

codspeed Bot commented Sep 18, 2026 •

edited

Loading

github-actions Bot commented Sep 18, 2026 •

edited

Loading