UniWorld – Global Unicode Showcase (TEST OUTPUT)

This document is a Unicode stress-test and demo for UniWorld.
It is intentionally long and visually dense. Treat it as both: - A presentation of what UniWorld enables. - A test fixture for rendering, fonts, and text handling across scripts.

1. What UniWorld Enables

• Correct text behavior for every script:
• Bidirectional layout for Arabic, Hebrew, mixed RTL/LTR.
• Reliable grapheme clusters for emoji, Indic conjuncts, and combining marks.
• Dictionary-based line breaking for Thai, Lao, Khmer, Myanmar (no-space scripts).
• Accurate display width for CJK full-width text and terminals.
• Single, testable core:
• Implemented in Rust with full conformance to:
  • UAX #9 – Bidirectional Algorithm.
  • UAX #14 – Line Breaking.
  • UAX #29 – Text Segmentation (grapheme, word, sentence).
  • UAX #15 – Normalization.
• Bindings:
• Python (uniworld), JavaScript/WASM, C, Go – same behavior, same tests.

Use this file to: - Inspect how your editor, renderer, or terminal handles complex Unicode. - Verify that UniWorld-based tools respect grapheme clusters, widths, and breaks. - Copy-paste samples into your own test harnesses.

2. Latin Scripts and Accents

Regular ASCII Latin:

• The quick brown fox jumps over the lazy dog.

Latin with diacritics and composed vs decomposed forms:

• Composed: é, ö, ñ, å, ç, ů, ẞ, Ǘ
• Decomposed (base + combining marks):
• e + COMBINING ACUTE: é
• o + COMBINING DIAERESIS: ö
• n + COMBINING TILDE: ñ

NFKC/NFKD-sensitive examples (ligatures, compatibility forms):

• Ligatures: ff, fi, fl, ffi, ffl (should normalize to ff, fi, fl, ffi, ffl).
• Fractions: ½ vs 1/2, ⅓ vs 1/3.

These samples exercise: - Normalization (NFC/NFD/NFKC/NFKD). - Grapheme clustering for combining marks.

3. Greek, Cyrillic, and Case Mapping

Greek:

• Lowercase: α β γ δ ε ζ η θ ι κ λ μ ν ξ ο π ρ σ τ υ φ χ ψ ω
• Uppercase: Α Β Γ Δ Ε Ζ Η Θ Ι Κ Λ Μ Ν Ξ Ο Π Ρ Σ Τ Υ Φ Χ Ψ Ω
• Final sigma test word: ΟΔΥΣΣΕΥΣ → lowercase should end with final sigma.

Cyrillic (Russian phrase):

• Привет, мир! Это тест Юникода.
  (Hello, world! This is a Unicode test.)

These samples exercise: - Case mapping (upper/lower/title). - Locale-neutral casing and Greek final sigma.

4. Hebrew and Arabic – Bidirectional Text

Hebrew:

• Plain: שלום עולם (shalom olam – “peace, world / hello, world”)
• With punctuation and numbers: שלום (שלום), 123, סוף.

Arabic:

• Plain: مرحبا بالعالم (marḥaban bil-ʿālam – “hello, world”)
• With Latin and digits:
• النسخة 2.0 من UniWorld
• تجربة UniWorld مع النص المختلط: English + العربية + 1234

Mixed BiDi paragraph:

• RTL: مرحبا بالعالم, then LTR: Hello, and back to RTL: שלום.

These lines should be rendered in correct visual order under UAX #9, with appropriate cursor mapping if UniWorld’s bidi and cursor logic is used.

5. South Asian Scripts – Indic and Beyond

5.1 Devanagari (Hindi)

• नमस्ते दुनिया (namaste duniya – “hello, world”)
• Conjuncts and matras: क् + ष → क्ष, श् + र → श्र
• Mixed word: प्रोग्रामिंग (programming)

5.2 Bengali

• হ্যালো বিশ্ব (hyalo biśśo – “hello, world”)
• Combined consonant clusters: ক্ষ, জ্ঞ, ম্প্র

5.3 Gurmukhi (Punjabi)

• ਸਤ ਸ੍ਰੀ ਅਕਾਲ (sat sri akal – traditional greeting)

5.4 Tamil

• வணக்கம் உலகம் (vaṇakkam ulagam – “hello, world”)

5.5 Sinhala

• හෙලෝ වර්ල්ඩ් (hello world, approximated)

These samples exercise: - Grapheme cluster boundaries for Indic consonant + virama + consonant sequences. - Line breaking around scripts with complex clusters.

6. Southeast Asian Scripts – Dictionary-based Line Breaking

These scripts traditionally do not use spaces between words, so line breaking depends on dictionary-based segmentation.

6.1 Thai

• Sentence (no spaces between words):
  สวัสดีชาวโลกนี่คือการทดสอบการตัดคำด้วย UniWorld

6.2 Lao

• ສະບາຍດີໂລກນີ້ແມ່ນການທົດສອບການຕັດຄໍາ

6.3 Khmer

• សួស្តីពិភពលោកនេះគឺជាការធ្វើតេស្តការកាត់ពាក្យ

6.4 Myanmar

• မင်္ဂလာပါကမ္ဘာလောကဤသည်မှာUniWorld၏စမ်းသပ်ခြင်းဖြစ်သည်

When viewed in a UniWorld-aware editor or test harness:

• Line breaks should respect dictionary segmentation (Thai/Lao/Khmer/Myanmar).
• Grapheme boundaries should never split inside stacked consonants or complex clusters.

7. CJK – Chinese, Japanese, Korean

7.1 Chinese

• Simplified: 你好，世界
• Traditional: 你好，世界 (same orthography here, different fonts/locale may differ)

7.2 Japanese

• Kanji + Hiragana + Katakana + Latin:
  こんにちは、UniWorld ライブラリへようこそ – Unicode テキスト処理のテストです。

7.3 Korean

• Hangul: 안녕하세요, 유니월드 라이브러리입니다.

Display width checks (should be full-width = 2 columns per CJK ideograph/Hangul syllable):

• Monospace grid (conceptual):

text ASCII: [H][e][l][l][o] -> width 5 CJK: [你][好][世][界] -> width 8 (4 × 2) Mixed: [H][i][!][你][好] -> width 2 + 1 + 4 = 7

These samples exercise: - East Asian Width handling in display_width. - Grapheme clustering for Hangul and kana with diacritics.

8. African and Other Scripts

8.1 Ethiopic (Amharic)

• ሰላም ዓለም (selam ālem – “hello, world”)

8.2 Tifinagh (Amazigh/Berber)

• ⵣⵉⵙⵉⵏ ⵏ ⵍⵎⵎⵉ (zisin n lmmi – sample text)

8.3 Cherokee

• ᎣᏏᏲ ᎢᎬᏱ (osiyo igvyi – “hello there, friend” approximation)

8.4 Canadian Aboriginal Syllabics (Cree, Inuktitut, Ojibwe)

Unified Canadian Aboriginal Syllabics (UCAS), used for Cree, Inuktitut, Ojibwe, and other Indigenous languages:

• Cree (Plains Cree syllabics): ᐊᐦᐊᐤ ᐊᐧᐊᐧᐱᐩ (sample greeting)
• Inuktitut: ᐃᓄᒃᑎᑐᑦ (Inuktitut – the language name)
• Word with hyphen: ᒥᓂᐦᐅᒋᐊᐧᐤ‑ᐃᐦᐃ (Canadian syllabics hyphen U+1400)

These test: - Line breaking: UCAS block (U+1400–U+167F) is AL (alphabetic) in UAX #14; hyphen U+1400 allows break after. - Grapheme clustering: one cluster per syllabic; no special GCB rules beyond defaults. - LTR; no dictionary-based segmentation (word boundaries follow spaces).

Section 8 (African and Other) tests: - Unicode range coverage. - Grapheme clustering in less-common scripts.

9. Emoji, ZWJ Sequences, and Flags

9.1 Basic Emoji

• Faces: 🙂 🙃 😄 😢 🤔 😴 😎 😇 🤖 👾
• Animals: 🐶 🐱 🐻 🐼 🐸 🐙 🐧 🐦 🦄
• Food: 🍎 🍕 🍣 🍜 🥐 🍩 ☕

Each should be a single grapheme cluster when treated by UniWorld.

9.2 Skin Tone Modifiers

• 👍 👍🏻 👍🏼 👍🏽 👍🏾 👍🏿
• 👋 👋🏻 👋🏼 👋🏽 👋🏾 👋🏿

Each base + modifier pair must be a single grapheme cluster (emoji base + Extend).

9.3 ZWJ Sequences (Emoji ZWJ)

ZWJ (Zero Width Joiner, U+200D) sequences combine multiple emoji into a single rendered glyph. UniWorld must treat each full ZWJ sequence as one grapheme cluster.

Examples (codepoint sequences):

• Family: U+1F468 ZWJ U+1F469 ZWJ U+1F467 ZWJ U+1F466 (man + woman + girl + boy)
• Profession: U+1F469 ZWJ U+1F4BB (woman technologist)
• Relationship: U+1F469 ZWJ U+2764 U+FE0F ZWJ U+1F468 (couple with heart)

These are the most complex grapheme clusters in Unicode: multiple ExtPict code points joined by ZWJ into a single user-perceived character. UniWorld handles them correctly; the rendered ZWJ emoji are omitted in this PDF only because headless browser print-to-PDF has layout issues with such sequences, not because of any Unicode limitation in the library.

9.4 Regional Indicator Flags

Regional Indicator pairs (U+1F1E6..U+1F1FF) form flag emoji:

• 🇺🇳 (UN)
• 🇺🇸 (US)
• 🇨🇦 (Canada)
• 🇧🇷 (Brazil)
• 🇯🇵 (Japan)

Regional indicators must pair: RI + RI = one cluster. UniWorld's grapheme segmentation must not split inside a flag pair.

10. Combining Marks and Edge Cases

Layered combining marks:

• à̀ (a + grave + grave)
• o͂̂ (o + tilde + circumflex)
• ḗ (e with acute and macron when normalized)

Visually similar but canonically distinct sequences:

• Å vs Å (precomposed vs decomposed).

These cases stress: - Canonical equivalence under NFC/NFD. - Cluster boundaries in the presence of multiple combining marks.

11. Simple Pictorials and Box Drawing

Unicode box-drawing table (small preview):

┌────────────────┐
│ UniWorld Box   │
├────────────────┤
│ [x] Graphemes  │
│ [x] Width      │
│ [x] Bidi       │
│ [x] Breaks     │
└────────────────┘

Monospace art with mixed characters:

Column ruler:
12345678901234567890
ASCII:   Hello, world!
CJK:     你好世界
Mixed:   Hi你好

These test: - Display width alignment in monospace contexts. - Handling of box drawing characters.

12. Mixed-Script Paragraph (Stress Test)

Hello, مرحبا, שלום, नमस्ते, สวัสดี, 你好, 한녕하세요 -- this single paragraph mixes Latin, Arabic, Hebrew, Devanagari, Thai, Han, and Hangul. 12345. UniWorld’s job is to ensure that segmentation (grapheme/word/sentence), bidi reordering, and line breaking all behave as the Unicode Standard intends, regardless of which binding (Rust, Python, JS/WASM, C, Go) you use.

You can use this paragraph to: - Test cursor movement (logical and visual) across mixed scripts. - Check line wrapping with different terminal widths. - Verify case folding (HELLO vs hello vs Straße vs STRASSE).

13. How to Use This File with UniWorld

Some suggested experiments:

• Grapheme boundaries:
• Run uniworld.segment.grapheme_boundaries() (Python) or grapheme_boundaries() (Rust/JS) on each section.

• Verify that:

  • Emoji ZWJ sequences and flags are single clusters.
  • Indic conjuncts with virama/nukta are not split.
  • Combining marks stay attached to their base.

• Word boundaries:

• Apply word_boundaries() and inspect how mixed scripts and punctuation are grouped.

• Confirm that dictionary-based segmentation improves Thai/Lao/Khmer/Myanmar breaks.

• Normalization:

• Compare NFC vs NFD on accented Latin samples and compatibility forms (ligatures, fractions).

• Ensure canonical-equivalent strings compare equal after normalization.

• Line breaking:

• Use line_break_opportunities_with_dictionary() on the SE Asian sections.

• Render with and without dictionary support to see the difference.

• Display width and truncation:

• Use display_width() and truncate_display_width() on CJK/emoji sections at various widths to confirm visual clamping.

14. Invitation

UniWorld exists so that every script gets first-class treatment in modern software:

• Editors and terminals that respect the people using them.
• Backends and services that handle global text correctly, not just ASCII.
• Applications that can be confidently localized into any writing system supported by Unicode.

If this document renders cleanly in your environment – with legible, correctly ordered text across scripts – then UniWorld is doing its job underneath. If you find edge cases, mis-renderings, or missing coverage, they become actionable test cases we can add to the suite.

This file is a test output, but it is also an invitation: build a world where every script works correctly by default.