admin
2d42c7ecf2
從零重寫之第一垂直切片（依 ARCHITECTURE.md 契約），涵蓋並超越 v1 全部功能。 核心基座（src/core）：unicode 字元表＋標點子類＋字形側位；locale 語言與 全角／開明政策（全域 lang.style ＋逐 lang policy）；dom（jz-* 工廠＋可還原 標記）；finder 遍歷引擎（avoid／scope.include／邏輯 run／pill 鄰字／分塊閘）； pass 聚合管線。 功能 pass（src/typeset）： - spacing：中西間隙採 margin 模型——包裹邊界左側字、以 margin-right 留隙， 不注入空白字元（textContent 完全不變、複製緊湊、justify 不暴增、不撐框、 行首乾淨）。含跨樣式邊界、pill、原檔空格移除。 - gapTrim（lineedge）：layout pass，行末去隙，修正 justify 右緣內縮。 - longword：長英文詞 <span lang> + hyphens。 - jiya：標點 charify + 全角／開明／半角（全域 + 逐 lang）；半形以字型 OpenType halt 渲染（按字形正確定位、不重疊，居中字形 ！？ 與繁體 locl 句逗點號皆正確）； 行內連續擠壓。 - jinze：行首行末避頭尾（閉類／點號綁前字、開類綁後字，以詞元為邊界、長引文 中段可斷）；亦為 justify 綁定載體。 - lineedge：行端擠壓 layout pass（括號引號；ResizeObserver 重算，瀏覽器專用）。 §8 justify：jz-char inline-block 原子，句末標點單份間距。 v1 相容層（compat/v1）：createCjkAutospace().apply 映射，IIFE 全域別名。 構建：esbuild → ESM／IIFE／CSS；tsc → d.ts。測試：39 個 jsdom 案例。 demo：簡繁並列＋justify＋強制繁體開明＋標點寬度量測，載 Noto webfont 作參考 （擠壓依賴字型 halt）。間隔號 · 不觸發中西間隙；200% 等詞元不被禁則切斷。
319 lines
15 KiB
JavaScript
319 lines
15 KiB
JavaScript
// 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 { 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：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：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");
|
||
});
|
||
|
||
// ----- 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");
|
||
});
|
||
|
||
// ----- 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");
|
||
});
|