"""UTF-8 test file for Python: non-ASCII strings, names and comments.

Free sample from 99helpers.com/file-types/py
Run: python3 sample-utf8.py
If any line below looks garbled (é, ’), something read this file in the
wrong encoding. Python itself always reads it as UTF-8.
"""

import sys
import unicodedata

# 1. No "# -*- coding: utf-8 -*-" line needed: UTF-8 is the default source
#    encoding in Python 3 (PEP 3120). Python 2 assumed ASCII.

# 2. Make print() write UTF-8 even where the console or a redirect expects
#    another encoding (older Python versions on Windows, for example).
sys.stdout.reconfigure(encoding="utf-8")

GREETINGS = {
    "Dutch": "Goedemorgen, café-eigenaar",
    "German": "Grüß Gott, Straße",
    "French": "Ça va, Zoë ? Très bien",
    "Greek": "Καλημέρα",
    "Russian": "Здравствуйте",
    "Arabic": "مرحبا",
    "Hebrew": "שלום",
    "Hindi": "नमस्ते",
    "Japanese": "こんにちは",
    "Chinese": "你好",
    "Korean": "안녕하세요",
    "Symbols": "€ £ ¥ ₹ ≤ ≥ ∞ ✓",
    "Emoji": "👋🌍🐍",
}

# 3. len() counts code points; encoding to UTF-8 shows the bytes on disk.
for language, text in GREETINGS.items():
    size = len(text.encode("utf-8"))
    print(f"{language:<9} {text}  ({len(text)} characters, {size} bytes)")

# 4. Names can be non-ASCII too (PEP 3131), though PEP 8 asks the standard
#    library to stick to ASCII, and most projects do the same.
café = "a variable called café"
print(café)

# 5. Normalisation: "é" can be one code point or "e" plus a combining accent.
#    They look identical but compare unequal until normalised.
composed, decomposed = "\u00e9", "e\u0301"
print(composed == decomposed, unicodedata.normalize("NFC", decomposed) == composed)

# 6. casefold() is the right way to compare text without case: "ß" folds to "ss".
print("Straße".lower(), "Straße".casefold())
print("STRASSE".casefold() == "Straße".casefold())