繁體中文 | English

---

漢字 IDS 部件查詢

Glyphs 字型編輯器外掛，用於拆解漢字結構並查詢相關字符。

功能

• 字符拆解 — 輸入漢字，視覺化顯示其部件樹狀結構
• 多部件組合搜尋 — 輸入多個部件（如「氵木」）找出同時包含所有部件的字；採遞迴比對（部件藏在更深層也算，如「淋」含「木」），可再拆的中間部件也能當查詢詞（如「立里」找到「童」、「火林」找到「焚」），重複部件代表「至少 N 個」（如「木木」找含兩個以上木的字）
• 子部件同位查詢 — 對複合字（如「明」=⿰日月）右欄上半列每個子部件在原位的字：⿰1 暉暗暝 …（日在 ⿰1 同位）、⿰2 朋朗期 …（月在 ⿰2 同位）；嵌套則加 ·（如 ⿰1· 鴠 因鴠=⿰旦鳥、日嵌在旦內）
• 含本字組合查詢 — 右欄下半列含本字作為部件的字、按頂層 IDC + 位置細分（如搜「金」→ ⿰1 鐘鈴銀 …、⿰≡ 鍂、⿱≡· 鑫𨰻）；對稱位用 ≡、嵌套加 ·
• 多部件交集 — 搜兩個以上部件時、右欄交集按頂層 IDC 粗略分組（如搜「金童」→ ⿰ 鐘 …）
• 字集篩選 — 預設顯示字型檔內的字，亦可使用自訂字集
• 顏色篩選 — 依 Glyphs 顏色標籤過濾結果
• 筆畫數篩選 — 用底部滑桿選擇 ±0/±1/±2/±3/±5 筆畫差，快速從相關字中找到筆畫接近主字的造字參考（資料源自 CNS11643）
• 全字庫連結 — 一鍵查詢 CNS11643 全字庫資料

安裝

外掛程式管理員（推薦）

1. 視窗 > 外掛程式管理員
2. 搜尋「HanziIDSComponentExplorer」
3. 點擊 安裝
4. 重新啟動 Glyphs

手動安裝

下載 HanziIDSComponentExplorer.glyphsPlugin，雙擊安裝。

使用方式

1. 視窗 > 漢字 IDS 部件查詢 開啟視窗
2. 在搜尋欄輸入漢字（如「森」）或 Unicode（如「68EE」）
3. 查看拆解結果、同字根、衍生字
4. 輸入多個部件（如「氵木」）搜尋同時包含所有部件的字：中欄列出輸入部件、右欄列出符合的字；此時筆畫篩選以各部件筆畫加總為基準（如「氵木」= 7 畫）

參考字型資料夾（PUA 造字顯示）

造字（PUA）的顯示字型優先序：參考字型資料夾 → 與當前文件同名的已安裝字型 → 其他已安裝字型。開發中的字型不必安裝，放進參考資料夾即可：

1. 右下角篩選選單 > 開啟參考字型資料夾...（位於 ~/Library/Application Support/Glyphs 3/HanziIDSComponentExplorer/Fonts/）
2. 放入字型檔（.otf / .ttf / .ttc / .otc）
3. 切回工具視窗即生效；重新匯出覆蓋同檔名也會自動更新，不受 macOS 字型快取影響

系統需求

• Glyphs 3.0 或以上
• macOS 10.9 或以上

資料來源

• IDS 拆解資料 — CHISE IDS database 的 CNS 及 Unicode 資料，收錄超過 10 萬個字符
• 筆畫數資料 — CNS11643-OpenData 的 Tables/Properties/CNS_stroke.txt，覆蓋約 7.7 萬字（74.8%）。超出 CNS 範圍的 Ext-G/H 罕字筆畫為空，僅在「關閉筆畫篩選」時顯示

授權

程式碼以 Apache License 2.0 授權。

本外掛內含的 ids.pdata 為 CHISE IDS 衍生資料，受 GPL-2.0-or-later 約束。CNS11643 資料依政府資料開放授權條款使用。

詳見 NOTICE。

作者

殷慈遠 TzuYuan Yin — erikyin.net

致謝

• CHISE Project — IDS 資料庫
• 全字庫 — 字形資料參考
• 3type/EOD 拆字小組 — 資料格式啟發

---

Hanzi IDS Component Explorer

A Glyphs font editor plugin for decomposing Chinese characters and exploring related glyphs.

Features

• Character Decomposition — Visualize the component tree structure of any Chinese character
• Multi-Component Search — Enter multiple components (e.g. "氵木") to find characters containing all of them; uses recursive matching (a component nested deeper still counts, e.g. "淋" contains "木"), decomposable intermediate components also work as queries (e.g. "立里" finds "童", "火林" finds "焚"), and repeated components mean "at least N" (e.g. "木木" finds characters with two or more 木)
• Sub-component Co-position Lookup — For compound characters (e.g., 明=⿰日月), the upper right panel lists characters with each sub-component in the same position: ⿰1 暉暗暝 … (日 at ⿰1), ⿰2 朋朗期 … (月 at ⿰2); nesting adds · (e.g., ⿰1· 鴠 because 鴠=⿰旦鳥 has 日 nested inside 旦)
• Containing-self Lookup — The lower right panel lists characters that use the search character as a component, grouped by top-level IDC + position (e.g., search 金 → ⿰1 鐘鈴銀 …, ⿰≡ 鍂, ⿱≡· 鑫𨰻); symmetric positions use ≡, nesting adds ·
• Multi-Component Intersection — Searching 2+ components groups intersection by top-level IDC (coarse): e.g., 金童 → ⿰ 鐘 …
• Charset Filtering — Filter by current font glyphs or custom charset
• Color Filtering — Filter by Glyphs color labels
• Stroke Count Filtering — Discrete bottom slider (±0/±1/±2/±3/±5) to narrow related characters by stroke count difference from the current character (data from CNS11643)
• CNS Link — Quick lookup in CNS11643 database

Installation

Plugin Manager (Recommended)

1. Window > Plugin Manager
2. Search for "HanziIDSComponentExplorer"
3. Click Install
4. Restart Glyphs

Manual Installation

Download HanziIDSComponentExplorer.glyphsPlugin and double-click to install.

Usage

1. Open Window > HanziIDSComponentExplorer
2. Enter a Chinese character (e.g., "森") or Unicode (e.g., "68EE")
3. View decomposition, sister characters, and derived characters
4. Enter multiple components (e.g. "氵木") to find characters containing all of them: the middle column lists the input components, the right column lists the matches; stroke filtering then uses the sum of the components' stroke counts as the baseline (e.g. "氵木" = 7 strokes)

Reference Fonts Folder (PUA display)

PUA glyphs resolve their display font in this order: reference fonts folder → installed font matching the current document's family name → other installed fonts. Fonts under development don't need to be installed — drop them into the reference folder:

1. Bottom-right filter menu > Open Reference Fonts Folder... (located at ~/Library/Application Support/Glyphs 3/HanziIDSComponentExplorer/Fonts/)
2. Place font files (.otf / .ttf / .ttc / .otc) in the folder
3. Switch back to the tool window to apply; re-exporting over the same file name also updates automatically, unaffected by the macOS font cache

Requirements

• Glyphs 3.0+
• macOS 10.9+

Data Sources

• IDS decomposition — CHISE IDS database (CNS and Unicode), covering over 100,000 characters
• Stroke counts — Tables/Properties/CNS_stroke.txt from CNS11643-OpenData, covering ~77k characters (74.8%). Characters outside CNS11643 (Ext-G/H rare ideographs) have no stroke data and are only shown when stroke filtering is OFF

License

Source code is licensed under Apache License 2.0.

The bundled ids.pdata is derived from the CHISE IDS database and is subject to GPL-2.0-or-later. CNS11643 data is used under the Open Government Data License, Taiwan.

See NOTICE for details.

Author

TzuYuan Yin — erikyin.net

Acknowledgments

• CHISE Project — IDS database
• CNS11643 全字庫 — Glyph data reference
• 3type/EOD — Data format inspiration

---

Copyright 2026 TzuYuan Yin