Trail

trail-py

The reference implementation of Trail - a small, total, declarative language for computing financial indicators, scores, and screening strategies over panels of securities. Trail expressions compile to vectorized Polars operations.

The language specification (grammar, reference, standard library) lives in trail-language/spec.

model quality on us_main period annual {
    operating_margin = income.operating_income / income.revenue
    score om_score weight 7 {
        2 if operating_margin > 0.12
        1 if operating_margin > 0.05
        else 0
    }
    export composite = weighted_score()
}

Install

pip install trail-lang        # or: uv add trail-lang

CLI

trail validate models/quality.trail                 # parse, kind-check, dry-run
trail run models/quality.trail --model quality      # evaluate a model
trail run models/quality.trail --model quality --config trail.yaml
trail catalog                                       # discover fields, functions, sources
trail catalog income                                # fields in a namespace
trail catalog cagr                                  # describe a function

The standard library is loaded implicitly; pass --no-stdlib to opt out.

Library

from trail.pipeline import prepare
from trail.compiler import compile_model
from trail.fixtures import load_panel  # a bundled demo panel

program = prepare("model m { export margin = income.operating_income / income.revenue }")
model = next(d for d in program.decls if type(d).__name__ == "ModelDecl")
result = compile_model(model, {}).run(load_panel())

Architecture

Parser (Lark LALR) → typed AST → macro expansion (def inlining) → kind-checked validation → Polars compiler. The engine carries only irreducible primitives; the large derived function library is written in Trail itself and shipped as trail/stdlib/*.trail (canonical copy in the spec).

Development

uv sync
uv run pytest -q
uv run ruff check .

License

MIT.

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

trail_lang-0.5.0.tar.gz (76.9 kB view details)

Uploaded Jul 16, 2026 Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

trail_lang-0.5.0-py3-none-any.whl (38.2 kB view details)

Uploaded Jul 16, 2026 Python 3

File details

Details for the file trail_lang-0.5.0.tar.gz.

File metadata

Download URL: trail_lang-0.5.0.tar.gz
Upload date: Jul 16, 2026
Size: 76.9 kB
Tags: Source
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for trail_lang-0.5.0.tar.gz
Algorithm	Hash digest
SHA256	`36762a3f17701d87045b346fbc04623ba581855cce6729798ded121129ab3dc7`
MD5	`6cc6de96887542ebe5d478ff45bdd561`
BLAKE2b-256	`dabe56be548ae9e04e756b7a766dfc180d082a556fdb76844e7a21a67853d50b`

See more details on using hashes here.

Provenance

The following attestation bundles were made for trail_lang-0.5.0.tar.gz:

Publisher: release.yml on trail-language/trail-py

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: trail_lang-0.5.0.tar.gz
- Subject digest: 36762a3f17701d87045b346fbc04623ba581855cce6729798ded121129ab3dc7
- Sigstore transparency entry: 2186184654
- Sigstore integration time: Jul 16, 2026
Source repository:
- Permalink: trail-language/trail-py@437b35752b869016640be71277e90a03d404b273
- Branch / Tag: refs/tags/v0.5.0
- Owner: https://github.com/trail-language
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: release.yml@437b35752b869016640be71277e90a03d404b273
- Trigger Event: push