Using Idenso from Python

Symbolica 2.2.0 bundles Idenso and Spenso as native community modules. This example contracts one four-dimensional Minkowski metric with a vector and verifies the exact remaining expression.

Install and verify the modules

python -m venv .venv
. .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install "symbolica==2.2.0"
python -c "import symbolica.community.idenso; import symbolica.community.spenso"

There is no standalone idenso Python wheel. Import the community module before constructing expressions, and follow Symbolica’s installation and license terms.

Simplify one metric contraction

Save this as idenso_quickstart.py:

from symbolica.community.idenso import initialize, list_dangling, simplify_metrics
from symbolica.community.spenso import Representation, TensorName

initialize()
rep = Representation.mink(4)
mu, nu = rep("mu"), rep("nu")
g, q = TensorName.g(), TensorName("q")

expression = g(mu, nu) * q(mu)
reduced = simplify_metrics(expression)

assert reduced == q(nu).to_expression()
assert len(list_dangling(reduced)) == 1
print(reduced)

Run python idenso_quickstart.py. Success means the explicit metric disappears and q(nu) remains. The structural equality check is stronger than comparing printed text, whose formatting may change between Symbolica releases.

Continue with the controlled identity tutorial for a verification workflow, then use the algebra guide to compose metric, Dirac, color, and cooking stages deliberately.