🌍 Internationalization Examples

Applies to: FinLang v0.6+ Status: Reference Last verified: v0.7.9

FinLang supports locale-specific date and number formats for global datasets.

🔹 Common CLI Flags

🔹 Regional Examples

🇺🇸 United States

# (No flags needed; this is the default)

Why: US exports use MM/DD/YYYY and . decimal with , thousands — FinLang’s default assumptions. Day-first is not used.

🇬🇧 United Kingdom

finlang --dayfirst

Why: UK uses DD/MM/YYYY with the same punctuation as US. --dayfirst disambiguates dates like 01/12/2025 (1 Dec).

🇫🇷 France / 🇩🇪 Germany

finlang --decimal "," --thousands "." --dayfirst --encoding auto

Why: EU exports commonly use , decimal, . thousands, and DD.MM.YYYY or ISO YYYY-MM-DD. The parser often infers day‑first from dots, but adding --dayfirst is explicit and deterministic across files.

🇨🇭 Switzerland

finlang --decimal "." --thousands "'" --dayfirst

Why: Swiss formats frequently use apostrophe ' as thousands (e.g., 1'234.56) and DD.MM.YYYY for dates. Explicit --dayfirst ensures consistent behaviour across banks.

🇮🇹 Italy

finlang --decimal "," --thousands "." --dayfirst --encoding auto --map italian_bank.map.json

Why: Italian exports use the same EU number formatting as France/Germany (, decimal, . thousands, semicolon delimiter). Column headers are in Italian — use a custom map to resolve them.

Example map (italian_bank.map.json):

{
  "date": ["data"],
  "amount": { "aliases": ["importo"] },
  "counterparty": ["controparte"],
  "memo": ["descrizione"]
}

🇯🇵 Japan

finlang --encoding utf-8 --map japanese_bank.map.json

Why: Japanese exports typically use UTF-8 encoding, YYYY-MM-DD dates, . decimal (no special flags needed), and ¥ currency symbols. The ¥ symbol is stripped automatically by the parser. Column headers are in Japanese — use a custom map.

Example map (japanese_bank.map.json):

{
  "date": ["日付"],
  "amount": { "aliases": ["金額"] },
  "counterparty": ["取引先"],
  "memo": ["摘要"]
}

Note: Japanese wildcard matching works in .fin rules: counterparty ~ "*シェル*" matches counterparties containing シェル (Shell).

🇮🇳 India

finlang --dayfirst --encoding utf-8 --map indian_bank.map.json

Why: Indian exports typically use DD/MM/YYYY dates and ₹ currency symbols. The ₹ symbol is stripped automatically. Column headers may be in Hindi or English — use a custom map for Hindi headers.

Example map (indian_bank.map.json):

{
  "date": ["तारीख"],
  "amount": { "aliases": ["राशि"] },
  "counterparty": ["प्रतिपक्ष"],
  "memo": ["विवरण"]
}

Note: Hindi wildcard matching works in .fin rules: counterparty ~ "*बैंक*" matches counterparties containing बैंक (Bank).

⚠️ Known limitation: Indian lakhs notation (e.g., 1,25,000 instead of 125,000) is not currently supported. The non-standard thousands grouping causes parsing errors. Amounts without lakhs formatting work correctly. This is tracked for a future release.

🔹 Supported Currency Symbols

The parser automatically strips the following currency symbols from amount fields:

No CLI flag is needed — currency stripping is automatic. Non-breaking spaces (\u00A0, \u202F) commonly found in EU-formatted amounts are also stripped.

🧪 Quick Test Your Settings

Create a test.csv with the following content:

date,amount
01/12/2025,"1.234,56"

Then run:

finlang --input test.csv --output test_out.csv   --decimal "," --thousands "." --dayfirst   --rules empty.fin --headless

Expected: Amount parses to 1234.56, date is 1 December 2025.

⚠️ Mixed Formats in One File

Problem: File contains both US and EU formatted numbers.

Solution: Standardize before import, or split into separate runs:

# Process US format rows
finlang --decimal "." --thousands "," --input us_subset.csv ...

# Process EU format rows
finlang --decimal "," --thousands "." --dayfirst --input eu_subset.csv ...

FinLang processes with one locale per run to ensure determinism.

🔧 Troubleshooting

📖 Related Documentation

• CLI Reference – Complete flag listing
• FAQ – Common questions
• Benchmarks – Performance data
• Workflows – Integration patterns

© FinLang Ltd