gloser 1.1.0

Simple i18n for Python

Gloser

The simplest way to add internationalization to your Python app

from gloser import Gloser

g = Gloser("translations.yaml")
print(g.t("welcome", name="Alice", locale="es"))  # → "¡Bienvenida, Alice!"

72 languages • Zero config • Simple yet powerful

---

Why Gloser?

• ✨ 5-minute setup — No complex configuration. Load a YAML file and start translating.
• 🌍 72 languages out of the box — Built-in locale defaults with pluralization rules, date/time formatting, and number formatting for 72 languages.
• 🎯 Developer-friendly — Natural Python API, clean YAML syntax, full type hints, and intuitive conventions.
• 🚀 Modern patterns — Built for web apps with thread-safe, stateless per-request locale handling.

---

Quick Start

1. Install

pip install gloser

2. Create your translations (translations.yaml)

welcome:
  en: "Welcome, {name}!"
  es: "¡Bienvenida, {name}!"
  fr: "Bienvenue, {name}!"

items:
  en:
    one: "You have {count} item"
    other: "You have {count} items"
  es:
    one: "Tienes {count} artículo"
    other: "Tienes {count} artículos"
  fr:
    one: "Vous avez {count} article"
    other: "Vous avez {count} articles"

3. Use in your code

from gloser import Gloser

g = Gloser("translations.yaml")

# Simple translation with t() shorthand
print(g.t("welcome", name="Alice", locale="en"))
# → "Welcome, Alice!"

# Automatic pluralization
print(g.t("items", count=1, locale="fr"))
# → "Vous avez 1 article"
print(g.t("items", count=5, locale="es"))
# → "Tienes 5 artículos"

That's it! You're internationalized.

---

Feature Showcase

Flexible YAML Structure

Gloser supports multiple YAML formats — use whatever works best for you:

Option 1: Single file with all translations (shown above)

welcome:
  en: "Welcome!"
  es: "¡Bienvenida!"

Option 2: One file per locale

# en.yaml
.locale: en
welcome: "Welcome!"
goodbye: "Goodbye!"

# es.yaml
.locale: es
welcome: "¡Bienvenida!"
goodbye: "¡Adiós!"

Option 3: Multi-document YAML

---
en:
  welcome: "Welcome!"
---
es:
  welcome: "¡Bienvenida!"

Mix and match: Load multiple files with different formats — Gloser merges them intelligently!

g = Gloser()
g.load_yaml_files("common.yaml", "en.yaml", "es.yaml")

🔢 Smart Pluralization

Gloser handles pluralization automatically using built-in rules for 72 languages, including complex ones like Polish with three forms:

from gloser import Gloser

g = Gloser("translations.yaml")  # Auto-loads locale defaults on demand

# English: simple (one/other)
g.t("files", count=1, locale="en")  # → "You have 1 file"
g.t("files", count=5, locale="en")  # → "You have 5 files"

# Polish: complex (one/few/many)
g.t("files", count=1, locale="pl")   # → "Masz 1 plik"
g.t("files", count=2, locale="pl")   # → "Masz 2 pliki"
g.t("files", count=5, locale="pl")   # → "Masz 5 plików"
g.t("files", count=22, locale="pl")  # → "Masz 22 pliki"

No configuration needed — rules for Arabic, Russian, Slovenian, and 69 other languages are built-in!

📅 Intelligent Date/Time Formatting

Locale-aware date formatting with natural, localized output.

from datetime import date
from gloser import Gloser

g = Gloser()  # Auto-loads date formats on demand

# Different locales, different formats
g.format_date(date(2025, 1, 9), "long", locale="en")
# → "January 9, 2025"

g.format_date(date(2025, 1, 9), "long", locale="de")
# → "9. Januar 2025"

g.format_date(date(2025, 1, 9), "long", locale="fr")
# → "9 janvier 2025"

g.format_date(date(2025, 1, 9), "long", locale="es")
# → "9 de enero de 2025"

You can also use date formatting within translations:

event_date:
  en: "Event on {date:long}"
  de: "Veranstaltung am {date:long}"

g.t("event_date", date=date(2025, 1, 9), locale="en")
# → "Event on January 9, 2025"

Built-in format styles: short, medium, long, full — all configured per locale.

Supported variables: {YYYY}, {MM}, {DD}, {monthname}, {mnm}, {dayname}, {dnm}, {hh}, {mm}, {ss}, {AMPM}, and more.

💰 Number Formatting

Automatic thousands and decimal separators for all locales.

from gloser import Gloser

g = Gloser()  # Uses built-in number formats for all locales

# Different locales use different separators
g.t("Price: {price:,.2f}", price=1234.56, locale="en")
# → "Price: 1,234.56"

g.t("Price: {price:,.2f}", price=1234.56, locale="de")
# → "Price: 1.234,56"  (Germany uses . for thousands, , for decimal)

g.t("Price: {price:,.2f}", price=1234.56, locale="fr")
# → "Price: 1 234,56"  (France uses space for thousands, , for decimal)

📋 Array-Style Lookups

Perfect for enums, weekdays, months, and fixed lists. In your YAML:

planet:
  en:
    1: Mercury
    2: Venus
    3: Earth
    4: Mars
  es:
    1: Mercurio
    2: Venus
    3: Tierra
    4: Marte

Then use it in code:

g = Gloser("translations.yaml")

g.t("planet", 3, locale="en")  # → "Earth"
g.t("planet", 3, locale="es")  # → "Tierra"

🌐 Web-Ready: Stateless Locale Handling

Perfect for Flask, FastAPI, Django, and other web frameworks.

# Initialize once at app startup
gloser = Gloser("translations.yaml")

# Get locale-specific translator per request
@app.route("/")
def index():
    locale = request.headers.get("Accept-Language", "en")[:2]
    t = gloser[locale]  # Thread-safe, stateless
    return t("welcome", name=current_user.name)

🥇 Ordinals

"1st, 2nd, 3rd" formatting with built-in rules for English, French, Spanish, and more.

from gloser import Gloser

g = Gloser("translations.yaml")  # Auto-loads ordinal rules on demand

g.t("place", position=1, locale="en")   # → "1st place"
g.t("place", position=2, locale="en")   # → "2nd place"
g.t("place", position=3, locale="en")   # → "3rd place"
g.t("place", position=11, locale="en")  # → "11th place"
g.t("place", position=22, locale="en")  # → "22nd place"

🛡️ Graceful Fallback

Missing translations? Gloser falls back to your default locale automatically.

# translations.yaml
hello:
  en: "Hello!"
  es: "¡Hola!"

welcome:
  en: "Welcome!"
  # Spanish translation missing

from gloser import Gloser

g = Gloser("translations.yaml", default_locale="en")

# Spanish translation exists
g.t("hello", locale="es")    # → "¡Hola!"

# Falls back to English when Spanish is missing
g.t("welcome", locale="es")  # → "Welcome!"

---

72 Languages Ready to Go

Gloser includes built-in locale defaults for 72 languages with pluralization rules, date/time formatting, number formatting, month names, and weekday names:

View all supported languages

af - Afrikaans	am - Amharic	ar - Arabic	az - Azerbaijani	be - Belarusian
bg - Bulgarian	bn - Bengali	bs - Bosnian	ca - Catalan	cs - Czech
da - Danish	de - German	dz - Dzongkha	el - Greek	en - English
es - Spanish	et - Estonian	fa - Persian	fi - Finnish	fr - French
ha - Hausa	he - Hebrew	hi - Hindi	hr - Croatian	hu - Hungarian
hy - Armenian	id - Indonesian	ig - Igbo	is - Icelandic	it - Italian
ja - Japanese	ka - Georgian	kk - Kazakh	km - Khmer	ko - Korean
ku - Kurdish	ky - Kyrgyz	lo - Lao	lt - Lithuanian	lv - Latvian
mn - Mongolian	ms - Malay	my - Burmese	nb - Norwegian Bokmål	ne - Nepali
nl - Dutch	nn - Norwegian Nynorsk	no - Norwegian	pl - Polish	pt - Portuguese
ro - Romanian	ru - Russian	si - Sinhala	sk - Slovak	sl - Slovenian
sn - Shona	so - Somali	sq - Albanian	sr - Serbian	sv - Swedish
sw - Swahili	ta - Tamil	th - Thai	tl - Tagalog	tr - Turkish
uk - Ukrainian	uz - Uzbek	vi - Vietnamese	xh - Xhosa	yo - Yoruba
zh - Chinese	zu - Zulu

Automatic deep merge: Your custom translations merge with built-in defaults, so you only need to specify what's different.

---

Examples

Check out the examples/ directory for complete working examples:

• example.py — Basic usage with multiple locales
• example_web_context.py — Web application pattern (Flask/FastAPI)
• example_plurals.py — Complex pluralization rules
• example_date_formatting.py — Date and time formatting
• example_number_formatting.py — Number formatting with locale separators
• example_ordinals.py — Ordinal numbers ("1st, 2nd, 3rd")
• example_arrays.py — Array-style lookups for enums and lists

---

Installation & Requirements

pip install gloser

Requirements:

• Python 3.8 or higher
• PyYAML 6.0 or higher (automatically installed)

Optional dependencies for development:

pip install gloser[dev]  # Includes pytest, pytest-cov

---

API Overview

Basic Usage

Note: Gloser provides t() as a convenient shorthand for translate(). Both work identically — use whichever you prefer!

from gloser import Gloser

# Initialize with default locale
g = Gloser(default_locale="en")

# Load translations from YAML
g.load_yaml("translations.yaml")

# Or load multiple files
g.load_yaml_files("common.yaml", "app.yaml")

# Or pass file(s) to constructor
g = Gloser("translations.yaml", default_locale="en")

# Translate with t() shorthand (recommended for brevity)
g.t("welcome", name="Alice", locale="en")

# Or use translate() if you prefer explicit naming
g.translate("welcome", name="Alice", locale="en")

# Or set current locale
g.set_locale("es")
g.translate("welcome", name="Alice")  # Uses Spanish
g.t("welcome", name="Alice")  # Shorthand works too

Web Application Pattern

# Initialize once at startup
gloser = Gloser("translations.yaml")

# Get locale-specific translator per request (stateless, thread-safe)
t = gloser["es"]  # or gloser.for_locale("es")
message = t("welcome", name="Alice")

Controlling Locale Defaults

By default, Gloser auto-loads built-in locale defaults as needed. You can control this behavior:

# Auto-load defaults on demand (default behavior)
g = Gloser("translations.yaml")

# Don't load any built-in defaults
g = Gloser("translations.yaml", load_defaults=False)

# Load specific locales immediately
g = Gloser("translations.yaml", load_defaults=["en", "es", "de"])

# Load all 72 locale defaults at startup
g = Gloser("translations.yaml")
g.load_defaults()  # Loads all available locale defaults

# Load additional defaults at runtime
g.load_defaults("fr", "it")  # Load French and Italian defaults

---

Testing & Quality

Gloser is thoroughly tested with 203 tests covering:

• ✅ All pluralization rules (Slavic, Semitic, Romance, Baltic, Germanic languages)
• ✅ Date and time formatting (72 locales)
• ✅ Number formatting with locale separators
• ✅ Ordinal numbers
• ✅ Array-style lookups
• ✅ Web application patterns
• ✅ Fallback behavior
• ✅ Edge cases and error handling

Run tests:

pytest
pytest --cov=gloser  # With coverage report

---

Contributing

Contributions are welcome! Whether it's:

• 🐛 Bug reports
• 💡 Feature requests
• 📝 Documentation improvements
• 🌍 New locale defaults
• ✨ Code contributions

Please open an issue or submit a pull request on GitHub.

---

License

MIT License - see LICENSE file for details.

---

Why "Gloser"?

"Gloser" is Norwegian for "glossary" — a fitting name for a translation library! 📚🇳🇴

---

Made with ❤️ by developers who believe i18n should be simple.