Hacktoberfest 2026: the issues maintainers tagged for October, open and beginner-friendly. Browse Hacktoberfest issues

A simulation built from a situation keeps the old system's defaults after apply_reform

Open
#614 0 comments 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 2 days

Nobody has claimed this yet.

Assessment

Difficulty
4/5
Estimated time
3-5 days
Newbie friendliness
57/100
Issue type
Bug
Clarity
Clearly specified
Activity status
Active
Tech stack
python

Research direction

Start with SimulationBuilder, Simulation.apply_reform, and _invalidate_all_caches, then read the changes in #613 to understand _build_from_situation and skip_rejected_inputs. Reproduce the examples in the issue with the stated Python command, including a set_input call after construction. Done when applying a reform rebuilds situation inputs under the new system while preserving inputs added afterward, and the three salary calculations match the expected values.

Written by the indexing model from the issue text.

Description

Problem

A simulation built from a situation keeps, after its system changes, the inputs the situation builder made under the old system. SimulationBuilder gives the people a situation gives no value for a variable it gives others (or sets along axes) that variable's default and stores the whole array as an input, so those people keep the old system's default. #612 is the same leak into a reform simulation's baseline; this is the other direction, into the reformed simulation.

Reproduced on master b909d5a60 (and on #613's branch, which does not change it), from the repository root with uv run --frozen --extra dev python:

import sys; sys.path.insert(0, ".")
from policyengine_core.country_template import CountryTaxBenefitSystem, Simulation
from tests.core.test_reform_baseline_system import change_salary_default  # salary default 0 -> 1000

P = "2022-01"
sit = {"persons": {"a": {"salary": {P: 3000.0}}, "b": {}}, "households": {"h": {"parents": ["a", "b"]}}}
print(Simulation(situation=sit, reform=change_salary_default).calculate("salary", P))  # [3000. 1000.]

s = Simulation(situation=sit)
s.apply_reform(change_salary_default)
print(s.calculate("salary", P))  # [3000.    0.]: the reform's default is 1000

s2 = Simulation(tax_benefit_system=CountryTaxBenefitSystem(), situation=sit, reform=change_salary_default)
print(s2.calculate("salary", P))  # [3000.    0.]
  • apply_reform changes the system in place, and _invalidate_all_caches replays every recorded input, including the arrays the builder filled with the old default.
  • Simulation.__init__ with both tax_benefit_system and reform builds the situation under the system it was given, then applies the reform to that system (apply_reform_set).

policyengine-us applies its structural reforms with apply_reform right after Simulation.__init__, so a structural reform that changes an input variable's default (or neutralizes one the situation gives) reaches its household simulations this way.

Expected

After its system changes, a simulation built from a situation holds what the builder makes of that situation under the new system, plus any inputs set on it afterwards.

Notes

#613 adds Simulation._build_from_situation and SimulationBuilder.skip_rejected_inputs, which build a situation into a simulation under its own system. Rebuilding on apply_reform also needs to keep the inputs set after construction (set_input calls, which _invalidate_all_caches replays today) apart from the builder's.

Dominant language
Python
Stars
22
Forks
30
Avg merge
2d 17h
Merged PRs (30d)
26

Getting set up

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from PolicyEngine/policyengine-core

All issues in PolicyEngine/policyengine-core

Similar issues

More Python issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.