finlang 0.7.5

FinLang: a deterministic, auditable DSL for financial rules

FinLang — The Financial Rules Engine

Deterministic. Auditable. Global.
Designed for explainable processing in regulated environments.

---

🌐 Overview

FinLang is a domain-specific language (DSL) and high-performance CLI engine for financial transaction processing.
It replaces opaque machine-learning categorization with transparent, deterministic rules — delivering explainability, auditability, and global compatibility.

Built for audit-friendly logic and deterministic processing.
A deterministic alternative where explainability and reproducibility matter.

---

📝 The FinLang DSL

FinLang rules are human-readable, Git-friendly, and designed for precision.
The engine processes rules top-to-bottom; the last matching rule sets the category, while flags accumulate.

# Example: Basic categorization and flagging
rule "GROCERIES: Tesco" {
  match:
    - counterparty ~ "*TESCO*"
  set:
    - category = "Groceries"
    - flags += "Supermarket"
}

# Example: Numeric range and exact match
rule "TRAVEL: High Value Flight" {
  match:
    - counterparty == "BRITISH AIRWAYS"
    - amount in -5000.00 .. -500.00
  set:
    - category = "Travel"
    - flags += "HighValue"
}

---

⚙️ Key Features (v0.7.5)

Feature	Description
Deterministic DSL	Human-readable .fin rules language — explainable logic, Git-friendly.
High-Performance Engine	Vectorized core (Pandas + NumPy + PyArrow) — 27K+ rows/sec validated throughput.
Dual Backend	Standard (Engine: c) or FastIO (Engine: pyarrow) with automatic fallback.
Growth Loop	Automated Discover → Suggest → Categorize workflow — 97.8% success on addressable patterns.
Global I18n Support	US/UK/EU/Commonwealth formats, £ € $ ¥ ₹ stripping, localized decimals/dates/delimiters.
Audit Trail System	Every decision logged (before/after state diffs); stateless for reproducibility.
Exclude Marker	Boolean exclude column — rule-driven, auditable, supports blacklist/whitelist exception patterns.
CR/DR Semantics	Case-insensitive CR/DR, accounting negatives (123.45), trailing minus 123.45-.
Amount Synthesis	Auto-computes amount = abs(credit) – abs(debit) across 9 edge cases.
Strict Parsing	Locale-aware normalization with configurable thresholds (--strict-parse).
Flag Integrity	Append-only (flags +=) with deterministic deduplication.

---

📦 Installation

Requirements: Python 3.10—3.14

From PyPI (Recommended):

pip install finlang

With Fast I/O (PyArrow):

pip install "finlang[fastio]"

(Enables --fastio for accelerated CSV I/O.)

From Source (Development):

git clone https://github.com/FinLang-Ltd/finlang.git
cd finlang
pip install -e .[fastio]

---

🚀 Quick Start — The 5-Step Growth Loop

1️⃣ Initial Categorization

finlang --input transactions.csv --output baseline.csv \
  --rules my_rules.fin --include-pack retail,transport

2️⃣ Discover Gaps

finlang-discover --input baseline.csv \
  --candidates candidates.csv --all-candidates all_candidates.csv \
  --min-count 5

3️⃣ Suggest Rules (Exact Mode Recommended)

finlang-suggest --input candidates.csv --output suggested_rules.fin \
  --rules my_rules.fin --emit-match exact

4️⃣ Merge and Re-run

cat my_rules.fin suggested_rules.fin > merged.fin
finlang --input transactions.csv --output improved.csv \
  --rules merged.fin --include-pack retail,transport

✅ Expected Result: 5–10% coverage improvement; zero duplicates in exact mode.

---

📊 Performance Benchmarks

Measured with --audit-mode none (max throughput).

Dataset	Test	Rules	Time (s)	Rows/sec	Notes
100 K (UK Synthetic)	Growth Loop	121	2.54	39,370 ✅	Baseline
100 K (after Growth Loop)	Growth Loop	764	4.96	20,161 ✅	+6.3× rules → ≈ 2× slower
5M × 50 cols	Benchmark Harness	—	187.90	26,600 ✅	High volume validation

v0.7.4 improvement: Cache invalidation fix delivered 3–5% faster runtimes across most data shapes; ~5% integrity test improvement. Headline enterprise throughput: ~27K rows/sec (peak observed: 28.4K).
Cumulative: 10% faster than v0.6.4 (208s → 188s), +12% throughput.
Audit Overhead: Enabling --audit-mode lite/full reduces throughput by ≈38% due to diff calculation; provides full decision provenance.

Note: These figures are validated benchmark results from controlled tests (5M × 50 columns). Actual performance varies depending on dataset, ruleset, and audit mode.
See docs/benchmarks.md for details.

---

🔐 Cryptographic Integrity Verification (Benchmark)

SHA-256 fingerprint verification benchmarked on large datasets:

Rows	Full Validation	Engine (FastIO)	Result
5M	~5 min	133K rows/s	✅ All fingerprints match
10M	~10 min	156K rows/s	✅ All fingerprints match
20M	~18 min	167K rows/s	✅ All fingerprints match

What this benchmark validated: Every row's immutable fields (date, amount, counterparty) were verified via SHA-256 hash before and after engine processing. Zero cross-row contamination detected. Zero data corruption detected.

Note: This benchmark was performed in the test suite. SHA-256 verification is not currently part of the standard runtime CLI — it is included for validation purposes and will be available as a CLI flag in a future release.

---

🌍 Internationalization Matrix

Region	Example Number	Date Order	CLI Flags
🇺🇸 US / 🇨🇦 Canada	1,234.56	MM/DD	(defaults)
🇬🇧 UK / 🇦🇺 Commonwealth	1,234.56	DD/MM	--dayfirst
🇪🇺 Continental Europe	1.234,56	DD/MM	--decimal "," --thousands "." --dayfirst
🇨🇭 Switzerland	1'234.56	DD/MM	--thousands "'" --dayfirst

Auto-Detection and Normalization: BOM-safe UTF-8 encodings, , ; | \t delimiters, and automatic currency symbol stripping.

---

🧠 The Growth Loop Explained

Discover → Suggest → Categorize → Repeat

FinLang's Growth Loop accelerates rule creation through data-driven discovery.

• Discover uncategorized counterparties
• Suggest new rules in seconds (1:1 mapping in exact mode)
• Merge + Re-run for incremental coverage gains
• Validated Result: 97.8% success on addressable patterns
• ROI: 8.8 transactions categorized per new rule

📄 See: docs/growth_loop_best_practices.md

---

🧾 Known Limitations (v0.7.x)

• ⚠️ --emit-match fuzzy (default) uses naive tokenization and may produce broad patterns (e.g. *PLC*).
  → Use --emit-match exact for production workflows.
• ⚠️ Hyphenated/apostrophe names may affect fuzzy matching (< 1% impact).
• ⚠️ No support for non-Gregorian calendars or non-Western numerals.

---

📘 Documentation

• docs/release_notes/v0_7_5.md
• docs/runtime_contract.md
• docs/cli_reference.md
• docs/rulepacks.md
• docs/benchmarks.md
• docs/growth_loop_best_practices.md
• docs/amount_synthesis.md
• docs/i18n_examples.md
• docs/stateless_processing.md

Command-line help:

finlang --help
finlang-discover --help
finlang-suggest --help

---

🧩 Example CLI Usage

finlang --input bank.csv --output categorized.csv \
  --rules examples/rules.demo.fin \
  --include-pack retail,transport,subs \
  --fastio --audit audit_log.json --audit-mode lite

---

📜 License & Commercial Use

FinLang is open source under the GNU Affero General Public License (AGPL-3.0).
Commercial licenses and enterprise support are available via FinLang Ltd.

📧 [e-mail]
🌐 https://finlang.io

---

Contributing

Contributions are welcome! Before submitting a PR, please review and accept our Contributor Licence Agreement (CLA).

---

📌 Version Summary

Component	Version	Status
Core Engine	v0.7.5	✅ Production-Ready
CLI Suite	v0.7.5	✅ Validated
Discover/Suggest	v0.7.5	✅ 97.8% accuracy
Integrity Test	v0.7.5	✅ 20M rows verified
Docs	v0.7.5	✅ Complete
Python Support	3.10—3.14	✅ Tested