cjk-autospace/test/juzhen.test.mjs

// 聚珍切片 1 jsdom 測試（架構 §12）：驗證 DOM 變換正確性——包裝結構、
// bd-* class、邊界、冪等、revert 還原、copy-clean、lang 政策分支、作用域。

import assert from "node:assert/strict";
import { readFileSync } from "node:fs";
import { fileURLToPath } from "node:url";
import { test } from "node:test";

import { setupDom, countTag } from "./helpers.mjs";
import { createJuzhen, createCjkAutospace } from "../dist/juzhen.mjs";

// ----- 中西間隙（spacing）-----

test("spacing：CJK↔Latin 邊界以 jz-hws 包裹左側字（margin 留隙，不注入空白）", () =>
{
    const { document } = setupDom("<p>我用Neovim寫程式</p>");
    createJuzhen({ jiya: false, longWord: false, jinze: false }).render(document.body);
    const p = document.querySelector("p");
    assert.equal(countTag(p, "jz-hws"), 2, "用|Neovim 與 Neovim|寫 兩處邊界，各包左側字");
    assert.equal(p.textContent, "我用Neovim寫程式", "textContent 完全不變（無注入空白，純 copy-clean）");
    assert.equal(p.querySelector("jz-hws").textContent.length, 1, "jz-hws 內為單一邊界字");
});

test("spacing：textContent 完全不變（margin 留隙，不注入也不刪語義字元）", () =>
{
    const { document } = setupDom("<p>用Neovim，很爽。</p>");
    createJuzhen({ jiya: false, longWord: false, jinze: false }).render(document.body);
    assert.equal(document.querySelector("p").textContent, "用Neovim，很爽。");
});

test("spacing：冪等——重跑不重複插入 jz-hws", () =>
{
    const { document } = setupDom("<p>我用Neovim寫程式</p>");
    const jz = createJuzhen({ jiya: false, longWord: false, jinze: false });
    jz.render(document.body);
    const first = countTag(document.body, "jz-hws");
    jz.render(document.body);
    assert.equal(countTag(document.body, "jz-hws"), first, "第二次 render 不應增加");
});

test("spacing：revert 還原至原文（無 jz-hws，textContent 回到原樣）", () =>
{
    const { document } = setupDom("<p>我用Neovim寫程式</p>");
    const jz = createJuzhen({ jiya: false, longWord: false, jinze: false });
    jz.render(document.body);
    jz.revert(document.body);
    assert.equal(countTag(document.body, "jz-hws"), 0);
    assert.equal(document.querySelector("p").textContent, "我用Neovim寫程式");
});

test("spacing：全角省略號 …… 後接 CJK 不插中西間隙（問題 B：原 DOTS_CJK `|…` 誤匹配）", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">他竟然走了……留下一地</p>');
    createJuzhen({ lang: { style: "quanjiao" } }).render(document.body);
    const p = document.querySelector("p");
    // …… 屬 CJK 標點（已 charify 為 bd-liga），後接 CJK 不應有西文間隙。
    assert.equal(countTag(p, "jz-hws"), 0, "…… 與後接 CJK 之間不應有 jz-hws");
});

test("spacing：貨幣符號 € 視為西文——與 CJK 邊界補隙、手工空格正規化（問題 D）", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">售價 €1099 都行</p>');
    createJuzhen({ jiya: false, longWord: false, jinze: false }).render(document.body);
    const p = document.querySelector("p");
    assert.equal(/售價\s€/.test(p.textContent), false, "售價↔€ 之手工空格應正規化剝除");
    assert.ok(countTag(p, "jz-hws") >= 2, "價↔€ 與 9↔都 兩處中西邊界各補 jz-hws");
});

test("spacing：擴充西文集——拼音聲調 ā／希臘 α／西里爾 Я 皆視為西文補隙（問題 D 完整化）", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">中ā文α字Я尾</p>');
    createJuzhen({ jiya: false, longWord: false, jinze: false }).render(document.body);
    const p = document.querySelector("p");
    // ā(U+0101 拉丁擴充)、α(U+03B1 希臘，驗證希臘集補全)、Я(U+042F 西里爾) 皆為西文。
    assert.ok(countTag(p, "jz-hws") >= 5, "ā／α／Я 與相鄰 CJK 各成邊界、補 jz-hws");
});

test("spacing：CJK 標點介於 CJK 間一律不補中西隙（標點稽核：。！？「」（）《》……—— 等）", () =>
{
    const samples = [
        "前。後", "前！後", "前？後", "前，後", "前、後", "前；後", "前：後",
        "前「後", "前」後", "前（後", "前）後", "前《後", "前》後", "前【後", "前】後",
        "前〔後", "前〕後", "前……後", "前——後", "前·後", "前～後", "前“後", "前”後",
    ];
    for (const s of samples)
    {
        const { document } = setupDom('<p lang="zh-Hant">' + s + "</p>");
        createJuzhen({ lang: { style: "quanjiao" } }).render(document.body);
        assert.equal(countTag(document.querySelector("p"), "jz-hws"), 0, s + " 不應有 jz-hws");
    }
});

test("spacing：收尾彎引號 ” 後接拉丁／數字不補隙（標點稽核移除 QUOTE_AN ” 偏例，與 」）》 一致）", () =>
{
    // 原 pangu QUOTE_AN 令 ”+字母數字補隙，使 ” 成唯一會補隙之收尾標點（其他收尾標點皆不）。
    const { document } = setupDom('<p lang="zh-Hans">“中文”ABC 與 ”123</p>');
    createJuzhen({ jiya: false, longWord: false, jinze: false }).render(document.body);
    const p = document.querySelector("p");
    // “”ABC、”123 中之彎引號鄰拉丁／數字皆不補隙；惟「與」兩側作者空格為 CJK↔Latin
    // 邊界（與|ABC 之 ABC 結尾、123|與 之類）仍會處理，故只斷言不含「彎引號旁之隙」——
    // 以 textContent 不變 + 彎引號未被 jz-hws 包裹驗證。
    const hwsTexts = Array.from(p.querySelectorAll("jz-hws")).map((h) => h.textContent);
    assert.equal(hwsTexts.includes("”"), false, "收尾彎引號 ” 不應被包入 jz-hws（不補隙）");
});

test("spacing：跨樣式行內邊界（<strong>是</strong>Neovim）亦識別（I7）", () =>
{
    const { document } = setupDom("<p>這<strong>是</strong>Neovim</p>");
    createJuzhen({ jiya: false, longWord: false, jinze: false }).render(document.body);
    // 是|Neovim 邊界（跨 </strong>）應插入 jz-hws。
    assert.equal(countTag(document.querySelector("p"), "jz-hws"), 1);
});

test("spacing：avoid 子樹（code/pre）不處理", () =>
{
    const { document } = setupDom("<p>看<code>foo是bar</code>這</p>");
    createJuzhen({ jiya: false, longWord: false, jinze: false }).render(document.body);
    // code 內 foo是bar 不應被 charify/spacing；但 pill 邊界（看|code、code|這）會加間隙。
    assert.equal(document.querySelector("code").textContent, "foo是bar");
});

// ----- pill 邊界（spacing Pass B）-----

test("spacing：pill（code）兩側與 CJK 相鄰插入 jz-hws", () =>
{
    const { document } = setupDom("<p>用<code>vim</code>編輯</p>");
    createJuzhen({ jiya: false, longWord: false, jinze: false }).render(document.body);
    assert.equal(countTag(document.querySelector("p"), "jz-hws"), 2);
});

// ----- 標點擠壓（jiya）-----

test("jiya：標點 charify 為 jz-char>jz-inner，標 bd-* class，textContent 不變（copy-clean）", () =>
{
    const { document } = setupDom('<p lang="zh-Hans">他說：「好，走。」</p>');
    createJuzhen({ spacing: false, longWord: false, jinze: false }).render(document.body);
    const p = document.querySelector("p");
    assert.equal(p.textContent, "他說：「好，走。」", "標點被包裝但文字不變");
    assert.ok(countTag(p, "jz-char") >= 5);
    assert.ok(countTag(p, "jz-inner") >= 5);
    const pause = p.querySelector("jz-char.bd-pause");
    assert.ok(pause, "，／：應標 bd-pause");
    const stop = p.querySelector("jz-char.bd-stop");
    assert.ok(stop, "。應標 bd-stop");
});

test("jiya：開明式（zh-Hans）句內點號半形（jz-half），句末點號不收（機制 C）", () =>
{
    const { document } = setupDom('<p lang="zh-Hans">好，走。</p>');
    createJuzhen({ spacing: false, longWord: false, jinze: false }).render(document.body);
    const p = document.querySelector("p");
    const comma = p.querySelector("jz-char.bd-pause");   // ，
    const period = p.querySelector("jz-char.bd-stop");   // 。
    assert.ok(comma.classList.contains("jz-half"), "，開明式半形（halt）");
    assert.ok(!period.classList.contains("jz-half"), "。句末保持全形");
});

test("jiya：開明式括號引號半形（開類／閉類皆 jz-half）", () =>
{
    const { document } = setupDom('<p lang="zh-Hans">說「好」嗎</p>');
    createJuzhen({ spacing: false, longWord: false, jinze: false }).render(document.body);
    const p = document.querySelector("p");
    assert.ok(p.querySelector("jz-char.bd-open").classList.contains("jz-half"), "「半形");
    assert.ok(p.querySelector("jz-char.bd-close").classList.contains("jz-half"), "」半形");
});

test("jiya：全角式（zh-Hant）baseline 一律不收（孤立標點無 jz-half）", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">好，走。</p>');
    createJuzhen({ spacing: false, longWord: false, jinze: false }).render(document.body);
    assert.equal(countTag(document.querySelector("p"), "jz-char.jz-half"), 0, "全角式不擠（無相鄰標點）");
});

test("jiya 機制 A：句末標點後接閉引號，前者半形（。」→ 。jz-half，全角亦然）", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">好。」嗎</p>');
    createJuzhen({ spacing: false, longWord: false, jinze: false }).render(document.body);
    const period = document.querySelector("jz-char.bd-stop"); // 。
    assert.ok(period.classList.contains("jz-half"), "。鄰接閉引號 → 半形（行內連續擠壓）");
});

test("jiya 機制 A：？！→ 僅 ？半形，！保持全形 1em（修正居中字形單側收縮缺陷）", () =>
{
    const { document } = setupDom('<p lang="zh-Hans">怎么办？！真</p>');
    createJuzhen({ spacing: false, longWord: false, jinze: false }).render(document.body);
    const stops = document.querySelectorAll("jz-char.bd-stop"); // ？ ！
    assert.equal(stops.length, 2);
    assert.ok(stops[0].classList.contains("jz-half"), "？鄰接 ！ → 半形");
    assert.ok(!stops[1].classList.contains("jz-half"), "！保持全形（1em，不被單側收縮拖拽重疊）");
});

test("jiya 機制 A：連續開類，後者半形（「「→ 第二個 jz-half）", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">「「好</p>');
    createJuzhen({ spacing: false, longWord: false, jinze: false }).render(document.body);
    const opens = document.querySelectorAll("jz-char.bd-open");
    assert.equal(opens.length, 2);
    assert.ok(!opens[0].classList.contains("jz-half"), "首個開類左側對漢字／行首，不由鄰接收");
    assert.ok(opens[1].classList.contains("jz-half"), "第二個開類內側 → 半形");
});

test("jiya 機制 A：，「→ 兩側皆收半形（雙側獨立判斷，合占 1em）", () =>
{
    // ，右側有空白且右鄰標點「→ 收，；「左側有空白且左鄰標點，→ 收「。兩者皆半 = 1em。
    const { document } = setupDom('<p lang="zh-Hant">甲，「乙</p>');
    createJuzhen({ spacing: false, longWord: false, jinze: false }).render(document.body);
    const comma = document.querySelector("jz-char.bd-pause"); // ，
    const open = document.querySelector("jz-char.bd-open");    // 「
    assert.ok(comma.classList.contains("jz-half"), "，右空白鄰標點 → 半形");
    assert.ok(open.classList.contains("jz-half"), "「左空白鄰標點 → 半形（雙側收，非二選一）");
});

test("jiya 機制 A：。」→ 僅 。半形（」左為墨、無空白可挤，全角式合占 1.5em）", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">甲。」乙</p>');
    createJuzhen({ spacing: false, longWord: false, jinze: false }).render(document.body);
    const stop = document.querySelector("jz-char.bd-stop");   // 。
    const close = document.querySelector("jz-char.bd-close");  // 」
    assert.ok(stop.classList.contains("jz-half"), "。右空白鄰標點 → 半形");
    assert.ok(!close.classList.contains("jz-half"), "」左為墨、無空白 → 不收（全角式 1.5em）");
});

test("jiya 機制 A：跨透明 inline 邊界之連續標點亦正確擠（，「 分置 strong/em，雙側皆收）", () =>
{
    // 下游回饋：原「同 parent 直接兄弟」掃描漏掉跨 strong/em 邊界之連續標點。
    const { document } = setupDom('<p lang="zh-Hant">甲<strong>，</strong><em>「</em>乙</p>');
    createJuzhen({ spacing: false, longWord: false, jinze: false }).render(document.body);
    assert.ok(document.querySelector("jz-char.bd-pause").classList.contains("jz-half"), "跨邊界 ，半形");
    assert.ok(document.querySelector("jz-char.bd-open").classList.contains("jz-half"), "跨邊界 「半形（透明穿越）");
});

test("jiya 機制 A：多標點鏈 ，」。逐對判斷（，半／」半／。全）", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">甲，」。乙</p>');
    createJuzhen({ spacing: false, longWord: false, jinze: false }).render(document.body);
    const p = document.querySelector("p");
    const comma = p.querySelector("jz-char.bd-pause");  // ，右空白鄰」→ 半
    const close = p.querySelector("jz-char.bd-close");  // 」右空白鄰。→ 半
    const stop = p.querySelector("jz-char.bd-stop");    // 。左為墨、右鄰漢字 → 全
    assert.ok(comma.classList.contains("jz-half"), "，→ 半");
    assert.ok(close.classList.contains("jz-half"), "」右空白鄰標點。→ 半");
    assert.ok(!stop.classList.contains("jz-half"), "。左為墨、右側無標點鄰 → 全");
});

test("jiya：revert 還原標點（無 jz-char，textContent 不變）", () =>
{
    const { document } = setupDom('<p lang="zh-Hans">好，走。</p>');
    const jz = createJuzhen({ spacing: false, longWord: false, jinze: false });
    jz.render(document.body);
    jz.revert(document.body);
    assert.equal(countTag(document.body, "jz-char"), 0);
    assert.equal(countTag(document.body, "jz-inner"), 0);
    assert.equal(document.querySelector("p").textContent, "好，走。");
});

// ----- 全域擠壓政策（lang.style）＋逐 lang 覆寫（lang.policy）-----

const JIYA_ONLY = { spacing: false, longWord: false, jinze: false };

test("政策：全域 style=banjiao → 各 lang 句末點號亦半形（。得 jz-half）", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">好。</p>');
    createJuzhen({ ...JIYA_ONLY, lang: { style: "banjiao" } }).render(document.body);
    const period = document.querySelector("jz-char.bd-stop");
    assert.ok(period.classList.contains("jz-half"), "半角式：句末亦半形");
});

test("政策：全域 style=kaiming 套用於繁體（覆寫內建全角預設，句內 ，半形）", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">好，走</p>');
    createJuzhen({ ...JIYA_ONLY, lang: { style: "kaiming" } }).render(document.body);
    assert.ok(document.querySelector("jz-char.bd-pause").classList.contains("jz-half"), "全域開明 → 繁體句內亦擠");
});

test("政策：全域 style=quanjiao，但 policy 逐 lang 覆寫 zh-Hans→kaiming（最高優先）", () =>
{
    const { document } = setupDom(
        '<p lang="zh-Hans">简，</p><p lang="zh-Hant">繁，</p>',
    );
    createJuzhen({
        ...JIYA_ONLY,
        lang: { style: "quanjiao", policy: { "zh-Hans": "kaiming" } },
    }).render(document.body);
    const ps = document.querySelectorAll("p");
    assert.ok(ps[0].querySelector("jz-char.bd-pause").classList.contains("jz-half"), "簡體 policy 開明 → 擠");
    assert.ok(!ps[1].querySelector("jz-char.bd-pause").classList.contains("jz-half"), "繁體用全域全角 → 不擠");
});

test("政策：未設 style／policy 時回退內建預設（繁全角不擠、簡開明擠）", () =>
{
    const { document } = setupDom(
        '<p lang="zh-Hans">简，</p><p lang="zh-Hant">繁，</p>',
    );
    createJuzhen(JIYA_ONLY).render(document.body);
    const ps = document.querySelectorAll("p");
    assert.ok(ps[0].querySelector("jz-char.bd-pause").classList.contains("jz-half"), "簡體內建開明");
    assert.ok(!ps[1].querySelector("jz-char.bd-pause").classList.contains("jz-half"), "繁體內建全角");
});

// ----- 字形偏側（jz-ink-*）＋ margin 後備模式（.feedback）-----

test("偏側：jz-ink-* 恆於 charify 標記（halt 預設模式亦然，root 不設 margin 屬性）", () =>
{
    const { document } = setupDom('<p lang="zh-Hans">好，「走」。</p>');
    createJuzhen(JIYA_ONLY).render(document.body);
    // 簡體點號偏左；角括號「」墨色強偏一側 → 開 right、閉 left（圓方括號才 center）。
    assert.ok(document.querySelector("jz-char.bd-pause").classList.contains("jz-ink-left"), "簡體 ，偏左");
    assert.ok(document.querySelector("jz-char.bd-open").classList.contains("jz-ink-right"), "角括號「墨偏右 → right");
    assert.ok(document.querySelector("jz-char.bd-close").classList.contains("jz-ink-left"), "角括號」墨偏左 → left");
    assert.equal(document.body.getAttribute("data-jz-halfwidth"), null, "halt 預設模式不設 root 屬性");
});

test("偏側：繁體點號居中（jz-ink-center），！？恆居中", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">走，真？</p>');
    createJuzhen(JIYA_ONLY).render(document.body);
    assert.ok(document.querySelector("jz-char.bd-pause").classList.contains("jz-ink-center"), "繁體 ，居中");
    assert.ok(document.querySelector("jz-char.bd-stop").classList.contains("jz-ink-center"), "？居中");
});

test("margin 模式：render 於 root 設 data-jz-halfwidth=margin，revert 清除", () =>
{
    const { document } = setupDom('<p lang="zh-Hans">好，走。</p>');
    const jz = createJuzhen({ ...JIYA_ONLY, jiya: { halfWidth: "margin" } });
    jz.render(document.body);
    assert.equal(document.body.getAttribute("data-jz-halfwidth"), "margin", "margin 模式設 root 屬性");
    // 偏側 class 與 jz-half 並存，CSS 始能定向收縮。
    const comma = document.querySelector("jz-char.bd-pause");
    assert.ok(comma.classList.contains("jz-half") && comma.classList.contains("jz-ink-left"), "，半形＋偏左");
    jz.revert(document.body);
    assert.equal(document.body.getAttribute("data-jz-halfwidth"), null, "revert 清除 root 屬性");
});

test("margin 模式：消費端 bias 覆寫參考表（如把簡體 ，改判居中）", () =>
{
    const { document } = setupDom('<p lang="zh-Hans">好，走</p>');
    createJuzhen({
        ...JIYA_ONLY,
        jiya: { halfWidth: "margin", bias: { "zh-Hans": { "，": "center" } } },
    }).render(document.body);
    const comma = document.querySelector("jz-char.bd-pause");
    assert.ok(comma.classList.contains("jz-ink-center"), "覆寫後 ，判居中");
    assert.ok(!comma.classList.contains("jz-ink-left"), "不再用參考預設之偏左");
});

// ----- 長英文詞（longWord）-----

test("longWord：[A-Za-z]{6,} 包進 span[lang]，textContent 不變", () =>
{
    const { document } = setupDom("<p>用 TypeScript 寫</p>");
    createJuzhen({ spacing: false, jiya: false, jinze: false }).render(document.body);
    const p = document.querySelector("p");
    const span = p.querySelector('span[lang="en"]');
    assert.ok(span, "TypeScript 應被包裝");
    assert.equal(span.textContent, "TypeScript");
    assert.equal(p.textContent, "用 TypeScript 寫");
});

test("longWord：短詞（<6）不包裝", () =>
{
    const { document } = setupDom("<p>vim 好</p>");
    createJuzhen({ spacing: false, jiya: false, jinze: false }).render(document.body);
    assert.equal(countTag(document.body, 'span[lang="en"]'), 0);
});

// ----- 禁則（jinze）：行首行末避頭尾 -----

test("jinze 避頭：閉類／點號綁到前一字（。」不置行首），中段文字仍可斷", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">他說好。」嗎</p>');
    createJuzhen({ spacing: false, longWord: false }).render(document.body);
    const p = document.querySelector("p");
    const group = p.querySelector("jz-jinze");
    assert.ok(group, "應有綁定群組");
    assert.equal(group.textContent, "好。」", "前一字 好 ＋ 。」 綁為 nowrap 單元");
    assert.equal(countTag(group, "jz-char"), 2, "群組含 。與 」");
    assert.equal(p.textContent, "他說好。」嗎", "文字不變（他說 在外、嗎 在外）");
});

test("jinze 避尾：開類綁到後一字（「不置行末），長引文中段可斷", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">說「好」嗎</p>');
    createJuzhen({ spacing: false, longWord: false }).render(document.body);
    const group = document.querySelector("jz-jinze");
    assert.ok(group, "應有綁定群組");
    assert.equal(group.textContent, "「好」", "「綁後字、」綁前字 → 「好」一體");
});

test("jinze 避頭：孤立逗號亦綁前字（逗號不可置行首）", () =>
{
    const { document } = setupDom('<p lang="zh-Hans">中文，好</p>');
    createJuzhen({ spacing: false, longWord: false }).render(document.body);
    const group = document.querySelector("jz-jinze");
    assert.equal(group.textContent, "文，", "前字 文 ＋ ， 綁定（中 在外可斷）");
});

test("jinze：revert 還原（無 jz-jinze／jz-char，文字回原樣）", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">他說好。」嗎</p>');
    const jz = createJuzhen({ spacing: false, longWord: false });
    jz.render(document.body);
    jz.revert(document.body);
    assert.equal(countTag(document.body, "jz-jinze"), 0);
    assert.equal(countTag(document.body, "jz-char"), 0);
    assert.equal(document.querySelector("p").textContent, "他說好。」嗎");
});

// ----- 作用域與分塊（§3.6）-----

test("scope.include：未命中子樹不處理", () =>
{
    const { document } = setupDom('<p>我用Vim</p><div class="jz"><p>他用Emacs寫</p></div>');
    createJuzhen({
        jiya: false, jinze: false,
        scope: { include: ".jz" },
    }).render(document.body);
    const ps = document.querySelectorAll("p");
    assert.equal(countTag(ps[0], "jz-hws"), 0, "include 外不處理");
    assert.ok(countTag(ps[1], "jz-hws") >= 1, "include 內處理");
});

test("分塊：data-juzhen 白名單只啟用列出功能", () =>
{
    const { document } = setupDom('<p data-juzhen="spacing" lang="zh-Hans">用Vim，好</p>');
    createJuzhen({}).render(document.body);
    const p = document.querySelector("p");
    assert.ok(countTag(p, "jz-hws") >= 1, "spacing 啟用");
    assert.equal(countTag(p, "jz-char"), 0, "jiya 未在白名單，停用");
});

test("分塊：data-juzhen-off 黑名單停用列出功能，其餘繼承", () =>
{
    const { document } = setupDom('<p data-juzhen-off="jiya" lang="zh-Hans">用Vim，好</p>');
    createJuzhen({}).render(document.body);
    const p = document.querySelector("p");
    assert.ok(countTag(p, "jz-hws") >= 1, "spacing 繼承啟用");
    assert.equal(countTag(p, "jz-char"), 0, "jiya 被黑名單停用");
});

// ----- 下游回饋修復（問題一 1a/1b、問題二）-----

test("1a：pill 後接 CJK 標點時，pill 左緣仍補 jz-hws（穿越 jz-jinze）", () =>
{
    const { document } = setupDom("<p>代码<code>build</code>，后</p>");
    createJuzhen({ pillSelector: "code, kbd", lang: { default: "zh-Hans", style: "quanjiao" } })
        .render(document.body);
    // jinze 已把 code＋，納入 jz-jinze；spacing 仍須能找到 code 左鄰字「码」並補間隙。
    const hws = document.querySelector("jz-hws");
    assert.ok(hws && hws.textContent === "码", "pill 左側「码」包入 jz-hws（不因右側標點/jz-jinze 而遺失）");
});

test("1a：pill 後接 CJK 標點且有作者空格 → 空格被剝除、左緣補間隙", () =>
{
    const { document } = setupDom("<p>代码 <code>build</code>，后</p>");
    createJuzhen({ pillSelector: "code, kbd", lang: { default: "zh-Hans", style: "quanjiao" } })
        .render(document.body);
    assert.ok(document.querySelector("jz-hws"), "補了 jz-hws");
    assert.equal(document.querySelector("p").textContent, "代码build，后", "作者空格被剝除（與無空格輸入一致）");
});

test("1b：CJK↔CJK 贅餘作者空格被剝除（含跨樣式行內標籤）", () =>
{
    const { document } = setupDom('<p><span class="b">事件</span> 元件</p>');
    createJuzhen({ lang: { default: "zh-Hans" } }).render(document.body);
    assert.equal(document.querySelector("p").textContent, "事件元件", "跨 <span> 之 CJK↔CJK 空格剝除");
});

test("1b：CJK↔CJK 不插 jz-hws（僅剝空格，無留隙）", () =>
{
    const { document } = setupDom("<p>事件 元件</p>");
    createJuzhen({ spacing: true, jiya: false, jinze: false, longWord: false }).render(document.body);
    assert.equal(countTag(document.body, "jz-hws"), 0, "CJK↔CJK 不需間隙，僅刪空格");
    assert.equal(document.querySelector("p").textContent, "事件元件");
});

test("問題二：justifyAtoms:false → root 設 data-jz-atoms=inline，revert 清除", () =>
{
    const { document } = setupDom("<p lang='zh-Hans'>测试，（甲）。</p>");
    const jz = createJuzhen({ lang: { default: "zh-Hans" }, justifyAtoms: false });
    jz.render(document.body);
    assert.equal(document.body.getAttribute("data-jz-atoms"), "inline", "設 inline 原子屬性");
    jz.revert(document.body);
    assert.equal(document.body.getAttribute("data-jz-atoms"), null, "revert 清除");
});

test("問題二：justifyAtoms 預設 true → 不設屬性（維持 inline-block 原子）", () =>
{
    const { document } = setupDom("<p lang='zh-Hans'>测试，</p>");
    createJuzhen({ lang: { default: "zh-Hans" } }).render(document.body);
    assert.equal(document.body.getAttribute("data-jz-atoms"), null, "預設不設屬性");
});

// ----- v1 相容層（§3.5）-----

test("compat：createCjkAutospace().apply() 可跑，產生中西間隙（margin 模型，copy-clean）", () =>
{
    const { document } = setupDom("<p>我用Neovim</p>");
    createCjkAutospace().apply(document.body);
    const p = document.querySelector("p");
    assert.ok(p.querySelector("jz-hws"), "間隙仍產生（jz-hws 包左側字 + margin）");
    assert.equal(p.textContent, "我用Neovim", "textContent 不變（無注入空白）");
});

test("compat：punctuationSqueeze:false 關閉 jiya", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">好，走</p>');
    createCjkAutospace({ punctuationSqueeze: false }).apply(document.body);
    assert.equal(countTag(document.body, "jz-char"), 0);
});

test("compat：預設強制 kaiming，句內點號半形（jz-half）（不因預設 zh-Hant 而退化）", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">好，走</p>');
    createCjkAutospace().apply(document.body);
    const comma = document.querySelector("jz-char.bd-pause");
    assert.ok(comma && comma.classList.contains("jz-half"), "compat 不退化：仍擠句內");
});

// ----- text-indent 漏溢防護（下游回報）-----

test("CSS：jz-* 原子歸零「區塊容器首／末行級」繼承屬性，免 inline-block 重套（透明原則）", () =>
{
    // jsdom 無版面，無法量寬；改驗構建 CSS 確含原子歸零規則。
    // 機制見 src/css/_normalize.css：此類繼承屬性會被 inline-block 原子（jz-char／
    // jz-jinze）重新套用於其內部首／末行，故須於原子上歸零（段落本身不受影響）。
    // 整類僅兩項（headless Chrome 逐屬性實測：letter-spacing／word-spacing／text-align
    // 系皆不洩漏）。
    const cssPath = fileURLToPath(new URL("../dist/juzhen.css", import.meta.url));
    const css = readFileSync(cssPath, "utf8");
    assert.match(css, /text-indent\s*:\s*0/, "構建 CSS 應含 text-indent:0 重設");
    assert.match(css, /hanging-punctuation\s*:\s*none/, "構建 CSS 應含 hanging-punctuation:none 重設");
});

test("CSS：jz-* 原子 text-decoration:inherit，補裝飾盒跳過原子致下劃線斷缺（透明原則·欠套用面）", () =>
{
    // 另一機制：text-decoration 非繼承、由段落裝飾盒繪製並跳過原子級 inline-block，
    // 致下劃線於每個標點原子下斷缺。令原子 inherit 自繪、接續（headless Chrome 像素掃描
    // 實測 halt／margin coverage 皆回 1.0）。
    const cssPath = fileURLToPath(new URL("../dist/juzhen.css", import.meta.url));
    const css = readFileSync(cssPath, "utf8");
    assert.match(css, /text-decoration\s*:\s*inherit/, "構建 CSS 應含 text-decoration:inherit");
});

test("CSS：聚珍不內建 font-family（字型交消費端），半形顯式保留 locl（下游回饋）", () =>
{
    // 字型屬消費端領域，聚珍不內建任何字型堆疊（移除 _lang.css）；僅以 _normalize 之
    // `font: inherit` 使原子繼承消費端字型、不強加，並於半形 feature 顯式併入 locl，
    // 確保泛 CJK 字型之地域標點字形不因設 halt 而失。body locl 由消費端字型＋lang 決定。
    const cssPath = fileURLToPath(new URL("../dist/juzhen.css", import.meta.url));
    const css = readFileSync(cssPath, "utf8");
    assert.equal(/font-family/.test(css), false, "聚珍 CSS 不應內建任何 font-family");
    assert.match(css, /font-feature-settings:\s*"locl"\s+1,\s*"halt"\s+1/, "半形 feature 應顯式保留 locl");
});

// ----- charClass：自定義字符分類規則集（向下兼容＋擴充）-----

test("jiya：英文全角句號 ． 已內建為句末標點 bd-stop（本次納入；Noto CJK 有 halt）", () =>
{
    // ． U+FF0E（異於表意句號 。U+3002）已納入內置 BD_STOP，預設即 charify 為 bd-stop。
    const { document } = setupDom('<p lang="zh-Hans">中．文。</p>');
    createJuzhen({ lang: { style: "kaiming" } }).render(document.body);
    const p = document.querySelector("p");
    const dot = Array.from(p.querySelectorAll("jz-char")).find((c) => c.textContent === "．");
    assert.ok(dot && dot.classList.contains("bd-stop"), "． 預設 charify 為 bd-stop");
    // 開明式句末不挤（與 。一致）；banjiao 才挤——此處僅確認分類與政策一致。
    assert.equal(dot.classList.contains("jz-half"), false, "開明式句末標點不挤半（保持全角，同 。）");
});

test("charClass.biaodian.stop：把仍未分類之 ‥ 歸為全寬句末標點 → charify 為 bd-stop", () =>
{
    // ‥ U+2025 預設未分類；以 charClass 歸句末標點以示「追加」機制。
    const base = setupDom('<p lang="zh-Hans">中‥文</p>');
    createJuzhen({ lang: { style: "kaiming" } }).render(base.document.body);
    assert.equal(countTag(base.document.querySelector("p"), "jz-char"), 0, "‥ 預設未分類、不 charify");
    const { document } = setupDom('<p lang="zh-Hans">中‥文</p>');
    createJuzhen({ lang: { style: "kaiming" }, charClass: { biaodian: { stop: "‥" } } }).render(document.body);
    assert.ok(countTag(document.querySelector("p"), "jz-char.bd-stop") >= 1, "‥ 經 charClass 歸 bd-stop");
});

test("charClass.western：把未分類符號 ★ 歸為西文 → 與 CJK 邊界補 jz-hws", () =>
{
    const { document } = setupDom('<p lang="zh-Hans">中★文</p>');
    // 先確認預設不補隙
    createJuzhen({}).render(document.body);
    assert.equal(countTag(document.querySelector("p"), "jz-hws"), 0, "預設 ★ 非西文、不補隙");
    const dom2 = setupDom('<p lang="zh-Hans">中★文</p>');
    createJuzhen({ charClass: { western: "★" } }).render(dom2.document.body);
    assert.ok(countTag(dom2.document.querySelector("p"), "jz-hws") >= 1, "★ 歸西文後，中★／★文 邊界補隙");
});

test("charClass.cjk：把符號歸為中文 → 與西文邊界補 jz-hws", () =>
{
    const { document } = setupDom('<p lang="zh-Hans">〇A</p>');
    createJuzhen({ charClass: { cjk: "〇" } }).render(document.body);
    assert.ok(countTag(document.querySelector("p"), "jz-hws") >= 1, "〇 歸中文後，〇↔A 為中西邊界、補隙");
});

test("charClass：實例隔離——覆寫不污染內置默認（resolveRuleset 不 mutate DEFAULT_RULESET）", () =>
{
    // 先建帶覆寫之實例（★ 視為西文）。
    const a = setupDom('<p lang="zh-Hans">中★文</p>');
    createJuzhen({ charClass: { western: "★" } }).render(a.document.body);
    assert.ok(countTag(a.document.querySelector("p"), "jz-hws") >= 1, "覆寫實例：★ 補隙");
    // 再建無覆寫之實例：★ 應仍未分類、不補隙（前實例之覆寫未洩漏至默認）。
    const b = setupDom('<p lang="zh-Hans">中★文</p>');
    createJuzhen({}).render(b.document.body);
    assert.equal(countTag(b.document.querySelector("p"), "jz-hws"), 0,
        "無覆寫實例：★ 仍未分類、不補隙（默認規則集未被前實例污染）");
});

test("charClass.biaodian.pause：未分類符號歸句內點號 → 開明式挤半形（jz-half）", () =>
{
    const { document } = setupDom('<p lang="zh-Hans">中•文</p>');
    createJuzhen({ lang: { style: "kaiming" }, charClass: { biaodian: { pause: "•" } } }).render(document.body);
    const c = document.querySelector("jz-char.bd-pause");
    assert.ok(c, "• charify 為 bd-pause");
    assert.ok(c.classList.contains("jz-half"), "開明式句內點號挤半形（jz-half）");
});

test("charClass.biaodian.open：歸開括號 → 禁則避尾，與後字綁為 jz-jinze", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">甲•乙</p>');
    createJuzhen({ charClass: { biaodian: { open: "•" } } }).render(document.body);
    assert.ok(countTag(document.querySelector("p"), "jz-jinze") >= 1, "開類 • 與後字綁定（避尾）為 jz-jinze");
});

test("charClass.western：追加不破壞內置——€（內置）與 ★（追加）皆補隙", () =>
{
    const { document } = setupDom('<p lang="zh-Hans">價€元★好</p>');
    createJuzhen({ jiya: false, longWord: false, jinze: false, charClass: { western: "★" } }).render(document.body);
    const n = countTag(document.querySelector("p"), "jz-hws");
    // 價↔€、€↔元、元↔★、★↔好 四處中西邊界皆補隙（內置 € 不因追加而失）。
    assert.ok(n >= 4, "內置 € 與追加 ★ 之四邊界皆補隙，實得 " + n);
});

test("charClass.biaodian：多子類同傳，各自生效（stop／pause）", () =>
{
    const { document } = setupDom('<p lang="zh-Hans">甲‥乙•丙</p>');
    createJuzhen({ lang: { style: "kaiming" }, charClass: { biaodian: { stop: "‥", pause: "•" } } }).render(document.body);
    const p = document.querySelector("p");
    assert.ok(countTag(p, "jz-char.bd-stop") >= 1, "‥ → bd-stop");
    const pause = p.querySelector("jz-char.bd-pause");
    assert.ok(pause && pause.classList.contains("jz-half"), "• → bd-pause 且開明式挤半形");
});

// ----- 功能分級（§6.5.0a，text-level vs paragraph-level）-----

test("分級：data-jz-level=text 之子樹只跑文本級——jiya／spacing 生效、jinze 跳過", () =>
{
    const { document } = setupDom('<p data-jz-level="text" lang="zh-Hant">說「你好」用Neovim寫</p>');
    createJuzhen().render(document.body);
    const p = document.querySelector("p");
    // 文本級：標點 charify（jz-char）、中西間隙（jz-hws）照常。
    assert.ok(countTag(p, "jz-char") >= 1, "text-level 仍跑 jiya（jz-char）");
    assert.ok(countTag(p, "jz-hws") >= 1, "text-level 仍跑 spacing（jz-hws）");
    // 段落級：禁則跳過——無 jz-jinze（「」本會綁鄰字）。
    assert.equal(countTag(p, "jz-jinze"), 0, "text-level 段落級禁則跳過（無 jz-jinze）");
});

test("分級：data-jz-level=text 跳過段落級 longWord（charify 亦受閘）", () =>
{
    const { document } = setupDom('<p data-jz-level="text">用 configuration 設定</p>');
    createJuzhen().render(document.body);
    const p = document.querySelector("p");
    assert.equal(p.querySelectorAll('[data-jz="longWord"]').length, 0, "text-level 不跑 longWord（無 span[data-jz=longWord]）");
});

test("分級：選項 level.text 選擇器標記（h1 → text-level，禁則跳過）", () =>
{
    const { document } = setupDom('<h1 lang="zh-Hant">說「你好」結束</h1><p lang="zh-Hant">說「你好」結束</p>');
    createJuzhen({ level: { text: "h1" } }).render(document.body);
    const h1 = document.querySelector("h1");
    const p = document.querySelector("p");
    assert.equal(countTag(h1, "jz-jinze"), 0, "h1 命中 level.text → 段落級禁則跳過");
    assert.ok(countTag(h1, "jz-char") >= 1, "h1 文本級 jiya 仍生效");
    assert.ok(countTag(p, "jz-jinze") >= 1, "未命中之 p 仍跑禁則");
});

test("分級：data-jz-level=paragraph 巢狀於 text 子樹內，最近祖先勝（恢復全功能）", () =>
{
    const { document } = setupDom(
        '<div data-jz-level="text" lang="zh-Hant">'
        + '<p data-jz-level="paragraph">說「你好」結束</p></div>',
    );
    createJuzhen().render(document.body);
    const p = document.querySelector("p");
    assert.ok(countTag(p, "jz-jinze") >= 1, "最近 data-jz-level=paragraph 勝 → 禁則恢復");
});

test("分級：spacing 文本級支持單元素 root（render 於 <span>）", () =>
{
    const { document } = setupDom("<span>我用Neovim寫程式</span>");
    const span = document.querySelector("span");
    createJuzhen({ jiya: false, jinze: false, longWord: false }).render(span);
    assert.ok(countTag(span, "jz-hws") >= 1, "行內單元素 root 亦補中西間隙");
    assert.equal(span.textContent, "我用Neovim寫程式", "textContent 不變");
});

test("分級：預設（無標記）塊級元素為段落級——禁則照常", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">說「你好」結束</p>');
    createJuzhen().render(document.body);
    assert.ok(countTag(document.querySelector("p"), "jz-jinze") >= 1, "預設段落級，禁則綁定");
});

// ----- 垂懸字避免（§6.5.1，方案 C）-----

test("垂懸字：末 N 實義字＋尾隨標點包入 jz-orphan（預設 chars=2）", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">甲乙，丙丁。</p>');
    createJuzhen().render(document.body);
    const p = document.querySelector("p");
    const orphan = p.querySelector("jz-orphan");
    assert.ok(orphan, "應產生 jz-orphan");
    assert.equal(orphan.textContent, "丙丁。", "綁定末 2 實義字（丙丁）＋尾隨句號");
    assert.equal(p.textContent, "甲乙，丙丁。", "textContent 不變");
});

test("垂懸字：chars 選項可調（chars=3 綁末 3 實義字）", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">甲乙，丙丁。</p>');
    createJuzhen({ orphan: { chars: 3 } }).render(document.body);
    const orphan = document.querySelector("jz-orphan");
    assert.equal(orphan.textContent, "乙，丙丁。", "綁定末 3 實義字（乙丙丁）＋其間／尾標點");
});

test("垂懸字：標點／符號不計實義字數（逗號不計）", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">讀書，思考，寫作。</p>');
    createJuzhen().render(document.body);
    const orphan = document.querySelector("jz-orphan");
    // 實義字 讀書思考寫作（6），末 2＝寫作；逗號不計入。
    assert.equal(orphan.textContent, "寫作。", "末 2 實義字＝寫作，標點不計");
});

test("垂懸字：段落實義字 < N 不綁（單字段）", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">嗯。</p>');
    createJuzhen().render(document.body);
    assert.equal(countTag(document.querySelector("p"), "jz-orphan"), 0, "實義字 1 < 2，不綁");
});

test("垂懸字：邊界字落於 jz-jinze 內 → 綁定整個群組（跨 jinze 攀爬、不切割）", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">他說「好」</p>');
    createJuzhen({ orphan: { chars: 1 } }).render(document.body);
    const orphan = document.querySelector("jz-orphan");
    assert.equal(orphan.textContent, "「好」", "末 1 實義字 好 在 jz-jinze 內 → 綁整個 「好」");
    assert.equal(orphan.querySelectorAll("jz-jinze").length, 1, "jz-jinze 整體納入、未被切割");
});

test("垂懸字：orphan:false 關閉（無 jz-orphan）", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">甲乙，丙丁。</p>');
    createJuzhen({ orphan: false }).render(document.body);
    assert.equal(countTag(document.querySelector("p"), "jz-orphan"), 0, "關閉時不產生 jz-orphan");
});

test("垂懸字：冪等——重跑不重複包裹", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">甲乙，丙丁。</p>');
    const jz = createJuzhen();
    jz.render(document.body);
    const first = countTag(document.body, "jz-orphan");
    jz.render(document.body);
    assert.equal(countTag(document.body, "jz-orphan"), first, "第二次 render 不增加 jz-orphan");
});

test("垂懸字：revert 還原（無 jz-orphan，textContent 回原樣）", () =>
{
    const { document } = setupDom('<p lang="zh-Hant">甲乙，丙丁。</p>');
    const jz = createJuzhen();
    jz.render(document.body);
    jz.revert(document.body);
    assert.equal(countTag(document.body, "jz-orphan"), 0, "revert 清除 jz-orphan");
    assert.equal(document.querySelector("p").textContent, "甲乙，丙丁。", "文字回原樣");
});

test("垂懸字：段落級——data-jz-level=text 子樹跳過", () =>
{
    const { document } = setupDom('<p data-jz-level="text" lang="zh-Hant">甲乙，丙丁。</p>');
    createJuzhen().render(document.body);
    assert.equal(countTag(document.querySelector("p"), "jz-orphan"), 0, "text-level 不跑垂懸字");
});

// ----- 標點懸掛（§6.5.2，行端模式接線；版面凸出量於 headless Chrome 驗證）-----

test("懸掛：已全面停用——hanging 預設關，行端模式 squeeze", () =>
{
    const jz = createJuzhen();
    assert.equal(jz.options.lineEnd, "squeeze", "行端模式恒 squeeze");
});

test("懸掛：防逃逸——即使 hanging:true，行端模式仍鎖死 squeeze（不啟用 hang）", () =>
{
    const jz = createJuzhen({ hanging: true });
    // option 仍被接受（API 不破），但 lineEnd 鎖死 squeeze——hang 不可能激活（§6.5.6 停用）。
    assert.equal(jz.options.lineEnd, "squeeze", "hanging:true 亦不轉 hang（無逃逸）");
});

test("懸掛：防逃逸——hanging:true render 後不產生任何 jz-hang-* 標記", () =>
{
    const { document } = setupDom('<p lang="zh-Hans">春。夏。秋。冬。</p>');
    createJuzhen({ hanging: true, justifyAtoms: false }).render(document.body);
    assert.equal(countTag(document.body, "jz-char.jz-hang-r"), 0, "無 jz-hang-r");
    assert.equal(countTag(document.body, "jz-char.jz-hang-l"), 0, "無 jz-hang-l");
});

test("懸掛：防逃逸——下游 data-juzhen=\"hanging\" 亦不啟用（lineEnd 不受 per-node 影響）", () =>
{
    const { document } = setupDom('<p lang="zh-Hans" data-juzhen="spacing jiya jinze hanging">春。夏。秋。冬。</p>');
    const jz = createJuzhen({ hanging: true, justifyAtoms: false });
    jz.render(document.body);
    assert.equal(jz.options.lineEnd, "squeeze", "lineEnd 恒 squeeze");
    assert.equal(countTag(document.body, "jz-char.jz-hang-r"), 0, "data-juzhen 要求亦無 jz-hang-r");
    assert.equal(countTag(document.body, "jz-char.jz-hang-l"), 0, "data-juzhen 要求亦無 jz-hang-l");
});

test("懸掛：防逃逸——blocks features:[\"hanging\"] 亦不啟用", () =>
{
    const { document } = setupDom('<p class="q" lang="zh-Hans">春。夏。秋。冬。</p>');
    createJuzhen({
        hanging: true, justifyAtoms: false,
        blocks: [ { selector: ".q", features: [ "spacing", "jiya", "hanging" ] } ],
    }).render(document.body);
    assert.equal(countTag(document.body, "jz-char.jz-hang-r"), 0, "blocks 要求亦無 jz-hang-r");
    assert.equal(countTag(document.body, "jz-char.jz-hang-l"), 0, "blocks 要求亦無 jz-hang-l");
});