@laisuk/opencc-fmmseg-wasm 0.3.9 → 0.4.1

package/README.md

@@ -12,25 +12,33 @@ This package provides high-quality Simplified Chinese ↔ Traditional Chinese co
 
 Features:
 
-* OpenCC-compatible conversion configs
-* Pure WebAssembly (no native binaries)
-* Browser-friendly
-* TypeScript-friendly APIs
-* Fast Rust backend
-* FMM-based phrase segmentation
-* Traditional Chinese regional variants
-* Japanese Shinjitai conversion support
-* Chinese script detection (`zho_check`)
-* Optional CJK Compatibility Ideograph normalization
-* In-memory Office / EPUB document conversion
-* Zero-dependency Node.js CLI
+- OpenCC-compatible conversion configs
+- Pure WebAssembly (no native binaries)
+- Browser-friendly
+- TypeScript-friendly APIs
+- Fast Rust backend
+- FMM-based phrase segmentation
+- Traditional Chinese regional variants
+- Japanese Shinjitai conversion support
+- Chinese script detection (`zho_check`)
+- Optional CJK Compatibility Ideograph normalization
+- In-memory Office / EPUB document conversion
+- Zero-dependency Node.js CLI
 
 Package profile:
 
-* 0 runtime dependencies
-* 1 WASM file
-* 18 conversion configs
-* 100% offline
+- 0 runtime dependencies
+- 1 WASM file
+- 20 conversion configs
+- 100% offline
+
+### 🌐 Live Demo
+
+Try `opencc-fmmseg-wasm` directly in your browser:
+
+**[Open the live WASM demo: CJK Conversion Tool](https://laisuk.github.io/opencc-fmmseg-wasm/)**
+
+No installation or server backend required.
 
 ---
 
@@ -40,79 +48,31 @@ Package profile:
 npm install @laisuk/opencc-fmmseg-wasm
 ```
 
----
-
-## Quick Start
-
-```javascript
-import init, {
-    OpenccWasm,
-    DetofuLevelWasm
-} from "@laisuk/opencc-fmmseg-wasm";
-
-await init();
-
-const cc = new OpenccWasm("s2t");
+Or installation from the mirror package:
 
-console.log(cc.convert("汉字", false));
-// 漢字
-
-console.log(cc.convertDetofu("儼驂騑於上路", false, DetofuLevelWasm.ExtB));
-// 俨骖騑于上路
+```bash
+npm install opencc-fmmseg-wasm
 ```
 
 ---
 
-## Using Config Enums
+## Quick Start
 
 ```javascript
 import init, {
-    OpenccWasm,
-    OpenccConfigWasm
+    OpenccWasm
 } from "@laisuk/opencc-fmmseg-wasm";
 
 await init();
 
-const cc = OpenccWasm.newWithEnum(
-    OpenccConfigWasm.S2hkp
-);
+const cc = new OpenccWasm("s2t");
 
-console.log(cc.convert("别随便录影侵犯个人隐私权", false));
-// 別隨便錄影侵犯個人私隱權
+console.log(cc.convert("汉字转换测试", false));
+// 漢字轉換測試
 ```
 
 ---
 
-## Supported Configs
-
-| Config  | Enum                     | Description                                           |
-|---------|--------------------------|-------------------------------------------------------|
-| `s2t`   | `OpenccConfigWasm.S2t`   | Simplified Chinese → Traditional Chinese              |
-| `s2tw`  | `OpenccConfigWasm.S2tw`  | Simplified Chinese → Taiwan Traditional               |
-| `s2twp` | `OpenccConfigWasm.S2twp` | Simplified Chinese → Taiwan Traditional (phrases)     |
-| `s2hk`  | `OpenccConfigWasm.S2hk`  | Simplified Chinese → Hong Kong Traditional            |
-| `s2hkp` | `OpenccConfigWasm.S2hkp` | Simplified Chinese → Hong Kong Traditional (phrases)  |
-| `t2s`   | `OpenccConfigWasm.T2s`   | Traditional Chinese → Simplified Chinese              |
-| `t2tw`  | `OpenccConfigWasm.T2tw`  | Traditional Chinese → Taiwan Traditional              |
-| `t2twp` | `OpenccConfigWasm.T2twp` | Traditional Chinese → Taiwan Traditional (phrases)    |
-| `t2hk`  | `OpenccConfigWasm.T2hk`  | Traditional Chinese → Hong Kong Traditional           |
-| `t2hkp` | `OpenccConfigWasm.T2hkp` | Traditional Chinese → Hong Kong Traditional (phrases) |
-| `tw2s`  | `OpenccConfigWasm.Tw2s`  | Taiwan Traditional → Simplified Chinese               |
-| `tw2sp` | `OpenccConfigWasm.Tw2sp` | Taiwan Traditional → Simplified Chinese (phrases)     |
-| `tw2t`  | `OpenccConfigWasm.Tw2t`  | Taiwan Traditional → Traditional Chinese              |
-| `tw2tp` | `OpenccConfigWasm.Tw2tp` | Taiwan Traditional → Traditional Chinese (phrases)    |
-| `hk2s`  | `OpenccConfigWasm.Hk2s`  | Hong Kong Traditional → Simplified Chinese            |
-| `hk2sp` | `OpenccConfigWasm.Hk2sp` | Hong Kong Traditional → Simplified Chinese (phrases)  |
-| `hk2t`  | `OpenccConfigWasm.Hk2t`  | Hong Kong Traditional → Traditional Chinese           |
-| `hk2tp` | `OpenccConfigWasm.Hk2tp` | Hong Kong Traditional → Traditional Chinese (phrases) |
-| `jp2t`  | `OpenccConfigWasm.Jp2t`  | Japanese Shinjitai → Traditional Chinese              |
-| `t2jp`  | `OpenccConfigWasm.T2jp`  | Traditional Chinese → Japanese Shinjitai              |
-
-The numeric enum values match the vendored Rust backend. Existing values are unchanged; `S2hkp = 17`, `Hk2sp = 18`,
-`T2hkp = 19`, and `Hk2tp = 20`.
-
----
-
 ## API
 
 ### Constructor
@@ -123,8 +83,8 @@ const cc = new OpenccWasm("s2t");
 
 Parameters:
 
-* `config` (optional): OpenCC config string
-* default: `"s2t"`
+- `config` (optional): OpenCC config string
+- default: `"s2t"`
 
 Example:
 
@@ -132,13 +92,22 @@ Example:
 const cc = new OpenccWasm("t2s");
 ```
 
+Taiwan phrase config example:
+
+```javascript
+const cc = new OpenccWasm("s2twp");
+
+cc.convert("预订‘奔驰’品牌出租车", true);
+// 預訂『賓士』品牌計程車
+```
+
 Hong Kong phrase config example:
 
 ```javascript
-const cc = new OpenccWasm("s2hkp");
+const cc = new OpenccWasm("hk2sp");
 
-cc.convert("别随便录影侵犯个人隐私权", false);
-// 別隨便錄影侵犯個人私隱權
+cc.convert("作業系統加密保護個人私隱權", false);
+// 操作系统加密保护个人隐私权
 ```
 
 ---
@@ -151,12 +120,12 @@ cc.convert(text, punctuation)
 
 Parameters:
 
-* `text`: input string
-* `punctuation`: whether to convert punctuation variants
+- `text`: input string
+- `punctuation`: whether to convert punctuation variants
 
 Returns:
 
-* converted string
+- converted string
 
 Example:
 
@@ -166,6 +135,85 @@ cc.convert("汉字", false);
 
 ---
 
+### setConfig
+
+```javascript
+cc.setConfig("t2s");
+```
+
+Returns:
+
+- `true` if valid
+- `false` if invalid
+
+---
+
+### getConfig
+
+```javascript
+cc.getConfig();
+```
+
+Returns current config string.
+
+---
+
+### isValidConfig
+
+```javascript
+OpenccWasm.isValidConfig("s2t");
+```
+
+---
+
+### getSupportedConfigs
+
+```javascript
+OpenccWasm.getSupportedConfigs();
+```
+
+Returns all supported config strings.
+
+Includes `s2hkp`, `hk2sp`, `t2hkp`, and `hk2tp`.
+
+---
+
+### getAvailableSlots
+
+```javascript
+const slots = OpenccWasm.getAvailableSlots();
+```
+
+Returns all canonical dictionary slot names accepted by `newWithCustomDicts` as a string array. The list is sourced from
+the core `DictSlot` definitions, so callers can use it to populate selectors or validate custom dictionary input without
+maintaining their own slot list.
+
+```javascript
+if (!OpenccWasm.getAvailableSlots().includes(slot)) {
+    throw new Error(`Unsupported dictionary slot: ${slot}`);
+}
+```
+
+---
+
+### zhoCheck
+
+Detect Chinese script type.
+
+```javascript
+cc.zhoCheck(text);
+```
+
+Returns:
+
+| Value | Meaning             |
+|-------|---------------------|
+| `0`   | Unknown / mixed     |
+| `1`   | Traditional Chinese |
+| `2`   | Simplified Chinese  |
+
+---
+
 ### normalizeCompat
 
 Normalize Unicode CJK Compatibility Ideographs before conversion.
@@ -176,11 +224,11 @@ cc.normalizeCompat(text)
 
 Parameters:
 
-* `text`: input string
+- `text`: input string
 
 Returns:
 
-* normalized string
+- normalized string
 
 Example:
 
@@ -197,9 +245,89 @@ console.log(cc.convert(normalized, false));
 // 天龙八部书里的乔峰是契丹人
 ```
 
-This is an optional pre-conversion pass for text that contains compatibility ideographs from Unicode compatibility
-ranges. Unmapped characters are preserved unchanged. Normal OpenCC conversion does not automatically run this pass, so
-call it explicitly when compatibility normalization is desired.
+This is an optional pre-conversion pass for text that contains CJK Compatibility Ideographs. Unmapped characters are
+preserved unchanged. Normal OpenCC conversion does not automatically run this pass, so call it explicitly when
+compatibility normalization is desired.
+
+### normalizeUnicodeCompat
+
+Normalize additional Unicode compatibility forms, CJK radicals, allographs, legacy glyphs, and selected
+compatibility-like punctuation before conversion.
+
+```javascript
+cc.normalizeUnicodeCompat(text)
+```
+
+Parameters:
+
+- `text`: input string
+
+Returns:
+
+- normalized string
+
+Example:
+
+```javascript
+const cc = new OpenccWasm("t2s");
+
+const input = "聼聼竒羙⽟䂖甁噐⾳";
+const normalized = cc.normalizeUnicodeCompat(input);
+
+console.log(normalized);
+// 聽聽奇美玉石瓶器音
+
+console.log(cc.convert(normalized, false));
+// 听听奇美玉石瓶器音
+```
+
+This pass uses the extended Unicode compatibility table and is separate from `normalizeCompat()`. It is useful for text
+containing radical forms, historical or allographic Han forms, and other compatibility-like characters that are not
+covered by the CJK Compatibility Ideograph ranges.
+
+Unmapped characters are preserved unchanged.
+
+### normalizeCompatExtended
+
+Apply complete compatibility normalization before conversion.
+
+```javascript
+cc.normalizeCompatExtended(text)
+```
+
+Parameters:
+
+- `text`: input string
+
+Returns:
+
+- normalized string
+
+Example:
+
+```javascript
+const cc = new OpenccWasm("t2s");
+
+const input = "天龍八部書裡的聼眾";
+const normalized = cc.normalizeCompatExtended(input);
+
+console.log(normalized);
+// 天龍八部書裡的聽眾
+
+console.log(cc.convert(normalized, false));
+// 天龙八部书里的听众
+```
+
+`normalizeCompatExtended()` combines the extended Unicode compatibility table with CJK Compatibility Ideograph
+normalization. Use this when input may contain characters handled by either normalization set.
+
+The normalization order is:
+
+1. extended Unicode compatibility normalization;
+2. CJK Compatibility Ideograph normalization.
+
+Normal OpenCC conversion does not automatically perform compatibility normalization. Call this method explicitly before
+`convert()` when complete compatibility normalization is desired.
 
 ---
 
@@ -213,12 +341,12 @@ cc.detofu(text, level)
 
 Parameters:
 
-* `text`: input string
-* `level`: `DetofuLevelWasm` threshold for the CJK extension ranges to replace
+- `text`: input string
+- `level`: `DetofuLevelWasm` threshold for the CJK extension ranges to replace
 
 Returns:
 
-* detofu-safe string
+- detofu-safe string
 
 Supported levels:
 
@@ -265,13 +393,13 @@ cc.convertDetofu(text, punctuation, level)
 
 Parameters:
 
-* `text`: input string
-* `punctuation`: whether to convert punctuation variants
-* `level`: `DetofuLevelWasm` threshold for the CJK extension ranges to replace
+- `text`: input string
+- `punctuation`: whether to convert punctuation variants
+- `level`: `DetofuLevelWasm` threshold for the CJK extension ranges to replace
 
 Returns:
 
-* converted detofu-safe string
+- converted detofu-safe string
 
 Example:
 
@@ -282,85 +410,6 @@ cc.convertDetofu("儼驂騑於上路", false, DetofuLevelWasm.ExtB);
 
 ---
 
-### setConfig
-
-```javascript
-cc.setConfig("t2s");
-```
-
-Returns:
-
-* `true` if valid
-* `false` if invalid
-
----
-
-### getConfig
-
-```javascript
-cc.getConfig();
-```
-
-Returns current config string.
-
----
-
-### isValidConfig
-
-```javascript
-OpenccWasm.isValidConfig("s2t");
-```
-
----
-
-### getSupportedConfigs
-
-```javascript
-OpenccWasm.getSupportedConfigs();
-```
-
-Returns all supported config strings.
-
-Includes `s2hkp`, `hk2sp`, `t2hkp`, and `hk2tp`.
-
----
-
-### getAvailableSlots
-
-```javascript
-const slots = OpenccWasm.getAvailableSlots();
-```
-
-Returns all canonical dictionary slot names accepted by `newWithCustomDicts` as a string array. The list is sourced from
-the core `DictSlot` definitions, so callers can use it to populate selectors or validate custom dictionary input without
-maintaining their own slot list.
-
-```javascript
-if (!OpenccWasm.getAvailableSlots().includes(slot)) {
-    throw new Error(`Unsupported dictionary slot: ${slot}`);
-}
-```
-
----
-
-### zhoCheck
-
-Detect Chinese script type.
-
-```javascript
-cc.zhoCheck(text);
-```
-
-Returns:
-
-| Value | Meaning             |
-|-------|---------------------|
-| `0`   | Unknown / mixed     |
-| `1`   | Traditional Chinese |
-| `2`   | Simplified Chinese  |
-
----
-
 ### newWithCustomDicts
 
 Construct a converter with in-memory custom dictionary pairs.
@@ -371,8 +420,8 @@ const cc = OpenccWasm.newWithCustomDicts(config, specs);
 
 Parameters:
 
-* `config`: OpenCC config string, such as `"s2t"`
-* `specs`: array of custom dictionary specs
+- `config`: OpenCC config string, such as `"s2t"`
+- `specs`: array of custom dictionary specs
 
 TypeScript-style spec shape:
 
@@ -455,14 +504,14 @@ Suffixes such as `.txt` are not accepted, even though case and surrounding white
 
 Merge contract:
 
-* Custom dictionaries are loaded from in-memory pairs only; no file I/O is involved.
-* The embedded compressed CBOR dictionary is loaded first.
-* Custom specs are applied to `DictionaryMaxlength` before `OpenCC::from_dictionary(...)`.
-* Conversion hot paths remain immutable after construction.
-* `Append` mode merges into the selected slot.
-* Duplicate or conflicting keys use last-wins semantics.
-* `Override` mode clears the selected slot first, then inserts the provided custom pairs.
-* Multiple specs are applied in array order.
+- Custom dictionaries are loaded from in-memory pairs only; no file I/O is involved.
+- The embedded compressed CBOR dictionary is loaded first.
+- Custom specs are applied to `DictionaryMaxlength` before `OpenCC::from_dictionary(...)`.
+- Conversion hot paths remain immutable after construction.
+- `Append` mode merges into the selected slot.
+- Duplicate or conflicting keys use last-wins semantics.
+- `Override` mode clears the selected slot first, then inserts the provided custom pairs.
+- Multiple specs are applied in array order.
 
 This API is useful for browser apps, user-defined terminology, database-loaded terms, generated dictionaries,
 `localStorage` or `IndexedDB` terms, testing, and embedded WASM environments. Customization happens at construction
@@ -470,6 +519,56 @@ time, not during conversion.
 
 ---
 
+## Supported Configs
+
+| Config  | Enum                     | Description                                           |
+|---------|--------------------------|-------------------------------------------------------|
+| `s2t`   | `OpenccConfigWasm.S2t`   | Simplified Chinese → Traditional Chinese              |
+| `s2tw`  | `OpenccConfigWasm.S2tw`  | Simplified Chinese → Taiwan Traditional               |
+| `s2twp` | `OpenccConfigWasm.S2twp` | Simplified Chinese → Taiwan Traditional (phrases)     |
+| `s2hk`  | `OpenccConfigWasm.S2hk`  | Simplified Chinese → Hong Kong Traditional            |
+| `s2hkp` | `OpenccConfigWasm.S2hkp` | Simplified Chinese → Hong Kong Traditional (phrases)  |
+| `t2s`   | `OpenccConfigWasm.T2s`   | Traditional Chinese → Simplified Chinese              |
+| `t2tw`  | `OpenccConfigWasm.T2tw`  | Traditional Chinese → Taiwan Traditional              |
+| `t2twp` | `OpenccConfigWasm.T2twp` | Traditional Chinese → Taiwan Traditional (phrases)    |
+| `t2hk`  | `OpenccConfigWasm.T2hk`  | Traditional Chinese → Hong Kong Traditional           |
+| `t2hkp` | `OpenccConfigWasm.T2hkp` | Traditional Chinese → Hong Kong Traditional (phrases) |
+| `tw2s`  | `OpenccConfigWasm.Tw2s`  | Taiwan Traditional → Simplified Chinese               |
+| `tw2sp` | `OpenccConfigWasm.Tw2sp` | Taiwan Traditional → Simplified Chinese (phrases)     |
+| `tw2t`  | `OpenccConfigWasm.Tw2t`  | Taiwan Traditional → Traditional Chinese              |
+| `tw2tp` | `OpenccConfigWasm.Tw2tp` | Taiwan Traditional → Traditional Chinese (phrases)    |
+| `hk2s`  | `OpenccConfigWasm.Hk2s`  | Hong Kong Traditional → Simplified Chinese            |
+| `hk2sp` | `OpenccConfigWasm.Hk2sp` | Hong Kong Traditional → Simplified Chinese (phrases)  |
+| `hk2t`  | `OpenccConfigWasm.Hk2t`  | Hong Kong Traditional → Traditional Chinese           |
+| `hk2tp` | `OpenccConfigWasm.Hk2tp` | Hong Kong Traditional → Traditional Chinese (phrases) |
+| `jp2t`  | `OpenccConfigWasm.Jp2t`  | Japanese Shinjitai → Traditional Chinese              |
+| `t2jp`  | `OpenccConfigWasm.T2jp`  | Traditional Chinese → Japanese Shinjitai              |
+
+The numeric enum values match the vendored Rust backend. Existing values are unchanged; `S2hkp = 17`, `Hk2sp = 18`,
+`T2hkp = 19`, and `Hk2tp = 20`.
+
+---
+
+## Using Config Enums
+
+```javascript
+import init, {
+    OpenccWasm,
+    OpenccConfigWasm
+} from "@laisuk/opencc-fmmseg-wasm";
+
+await init();
+
+const cc = OpenccWasm.newWithEnum(
+    OpenccConfigWasm.S2hkp
+);
+
+console.log(cc.convert("操作系统加密保护个人隐私权", false));
+// 作業系統加密保護個人私隱權
+```
+
+---
+
 ## Office / EPUB Conversion
 
 Office and EPUB conversion runs fully locally in the browser or Node.js. Files are passed in and returned as bytes;
@@ -484,8 +583,12 @@ docx, xlsx, pptx, odt, ods, odp, epub
 File size is limited by available browser or Node.js memory, but there is no upload or server-side limit. Font
 preservation is supported with the `keepFont` option.
 
-Use the instance method when possible. It reuses the converter configuration and any custom dictionaries already held by
-the `OpenccWasm` instance.
+Use the instance methods when possible. They reuse the converter configuration, custom dictionaries, and other converter
+state already held by the `OpenccWasm` instance.
+
+### Basic Office Conversion
+
+For normal OpenCC conversion, use:
 
 ```javascript
 cc.convertOfficeBytes(inputBytes, format, punctuation, keepFont)
@@ -493,21 +596,93 @@ cc.convertOfficeBytes(inputBytes, format, punctuation, keepFont)
 
 Parameters:
 
-* `inputBytes`: `Uint8Array` document bytes
-* `format`: `docx`, `xlsx`, `pptx`, `odt`, `ods`, `odp`, or `epub`
-* `punctuation`: whether to convert punctuation variants
-* `keepFont`: whether to preserve font declarations where supported
+- `inputBytes`: `Uint8Array` document bytes
+- `format`: `docx`, `xlsx`, `pptx`, `odt`, `ods`, `odp`, or `epub`
+- `punctuation`: whether to convert punctuation variants
+- `keepFont`: whether to preserve font declarations where supported
 
 Returns:
 
-* converted output bytes
+- converted output bytes
 
-The older free function remains available for compatibility:
+The existing free function remains available for compatibility:
 
 ```javascript
 convert_office_bytes(inputBytes, format, config, punctuation, keepFont)
 ```
 
+### Office Conversion Pipeline
+
+For compatibility normalization and optional DeTofu processing, use:
+
+```javascript
+cc.convertOfficeBytesPipeline(
+    inputBytes,
+    format,
+    punctuation,
+    keepFont,
+    normalizeMode,
+    detofuLevel
+)
+```
+
+The text-processing order is:
+
+```text
+Normalize -> OpenCC -> DeTofu
+```
+
+`normalizeMode` is a `NormalizeModeWasm` value:
+
+```javascript
+NormalizeModeWasm.None
+NormalizeModeWasm.Compat
+NormalizeModeWasm.UnicodeCompat
+NormalizeModeWasm.CompatExtended
+```
+
+The final `detofuLevel` argument is optional. When omitted or `undefined`, DeTofu is not applied.
+
+To enable DeTofu, pass a `DetofuLevelWasm` value:
+
+```javascript
+DetofuLevelWasm.ExtB
+DetofuLevelWasm.ExtC
+DetofuLevelWasm.ExtD
+DetofuLevelWasm.ExtE
+DetofuLevelWasm.ExtF
+DetofuLevelWasm.ExtG
+DetofuLevelWasm.ExtH
+DetofuLevelWasm.ExtI
+```
+
+`ExtB` applies all supported DeTofu mappings from Extension B onward.
+
+For example, extended compatibility normalization followed by OpenCC conversion, without DeTofu:
+
+```javascript
+const outputBytes = cc.convertOfficeBytesPipeline(
+    inputBytes,
+    "docx",
+    true,
+    true,
+    NormalizeModeWasm.CompatExtended
+);
+```
+
+To additionally apply DeTofu:
+
+```javascript
+const outputBytes = cc.convertOfficeBytesPipeline(
+    inputBytes,
+    "docx",
+    true,
+    true,
+    NormalizeModeWasm.CompatExtended,
+    DetofuLevelWasm.ExtB
+);
+```
+
 ### Browser Office Example
 
 ```javascript
@@ -537,6 +712,41 @@ a.click();
 URL.revokeObjectURL(a.href);
 ```
 
+### Browser Office Pipeline Example
+
+```javascript
+import init, {
+    OpenccWasm,
+    NormalizeModeWasm,
+    DetofuLevelWasm
+} from "@laisuk/opencc-fmmseg-wasm";
+
+await init();
+
+const cc = new OpenccWasm("t2s");
+const file = document.querySelector("input[type=file]").files[0];
+const inputBytes = new Uint8Array(await file.arrayBuffer());
+
+const outputBytes = cc.convertOfficeBytesPipeline(
+    inputBytes,
+    "docx",
+    true,
+    true,
+    NormalizeModeWasm.CompatExtended,
+    DetofuLevelWasm.ExtB
+);
+
+const blob = new Blob([outputBytes], {
+    type: "application/vnd.openxmlformats-officedocument.wordprocessingml.document"
+});
+
+const a = document.createElement("a");
+a.href = URL.createObjectURL(blob);
+a.download = "converted.docx";
+a.click();
+URL.revokeObjectURL(a.href);
+```
+
 ### Node.js Office Example
 
 ```javascript
@@ -558,6 +768,33 @@ const outputBytes = cc.convertOfficeBytes(
 fs.writeFileSync("output.docx", outputBytes);
 ```
 
+### Node.js Office Pipeline Example
+
+```javascript
+import fs from "fs";
+import init, {
+    OpenccWasm,
+    NormalizeModeWasm,
+    DetofuLevelWasm
+} from "@laisuk/opencc-fmmseg-wasm";
+
+await init();
+
+const cc = new OpenccWasm("t2s");
+const inputBytes = fs.readFileSync("input.epub");
+
+const outputBytes = cc.convertOfficeBytesPipeline(
+    inputBytes,
+    "epub",
+    true,
+    true,
+    NormalizeModeWasm.CompatExtended,
+    DetofuLevelWasm.ExtB
+);
+
+fs.writeFileSync("output.epub", outputBytes);
+```
+
 ---
 
 ## Browser Example
@@ -618,7 +855,7 @@ The package includes a zero-dependency Node.js CLI:
 ```bash
 opencc-fmmseg convert -i input.txt -o output.txt -c s2t -p
 opencc-fmmseg convert -i input.txt -o output.txt -c t2s -p --detofu all
-echo "别随便录影侵犯个人隐私权" | opencc-fmmseg convert -c s2hkp
+echo "操作系统加密保护个人隐私权" | opencc-fmmseg convert -c s2hkp
 echo "天龍八部書裡的喬峰是契丹人" | opencc-fmmseg convert -c t2s --norm-compat
 // 天龙八部书里的乔峰是契丹人
 echo "這個細路哥很靈活" | opencc-fmmseg convert -c hk2sp --custom-dict hkphrasesrev:append:my_hk_dict.txt  
@@ -649,6 +886,7 @@ opencc-fmmseg office -i input.docx -o output.docx -c s2t -p --keep-font
                               default when omitted value: all
 --keep-ids                  Preserve complete IDS expressions during conversion (default: false)
 -n, --norm-compat           Normalize CJK Compatibility Ideographs before conversion (default: false)
+-E, --norm-compat-extended  Normalize extended Unicode compatibility forms before conversion (default: false)
 -D, --custom-dict <slot:mode:file>
                             Load a custom dictionary.
                             May be specified multiple times.
@@ -677,12 +915,18 @@ tw2s, tw2sp, tw2t, tw2tp, hk2s, hk2sp, hk2t, hk2tp, jp2t, t2jp
 -F, --convert-filename      Convert generated output filename stem (default: false)
 --keep-font                 Preserve font-family information (default)
 --no-keep-font              Do not preserve font-family information
---custom-dict <slot:mode:file>
-                            Load a custom dictionary.
-                            May be specified multiple times.
-                            Examples:
-                              --custom-dict hkphrasesrev:append:my_hk_dict.txt
-                              --custom-dict stphrases:override:terms.txt
+--keep-ids                  Preserve complete IDS expressions during conversion (default: false) 
+-n, --norm-compat           Normalize CJK Compatibility Ideographs before conversion (default: false)
+-E, --norm-compat-extended  Normalize extended Unicode compatibility forms before conversion (default: false)
+--detofu [level]            Replace tofu-risk rare CJK extension chars after conversion
+                          level: all | ext-b | ext-c | ext-d | ext-e | ext-f | ext-g | ext-h | ext-i
+                          default when omitted value: all
+-D, --custom-dict <slot:mode:file>
+                          Load a custom dictionary.
+                          May be specified multiple times.
+                          Examples:
+                            --custom-dict hkphrasesrev:append:my_hk_dict.txt
+                            --custom-dict stphrases:override:terms.txt
 ```
 
 For `office`, the format is inferred from the input file extension when `--format` is omitted.
@@ -708,18 +952,19 @@ The WASM-facing enum is exported as `OpenccConfigWasm`, alongside `OpenccWasm`.
 
 ## Performance Notes
 
-* WebAssembly build disables Rayon parallelism by default.
-* Dictionaries are embedded into the WASM binary.
-* Browser caching significantly improves subsequent loads.
+- WebAssembly build disables Rayon parallelism by default.
+- Dictionaries are embedded into the WASM binary.
+- Browser caching significantly improves subsequent loads.
 
 ---
 
 ## Related Projects
 
-* Rust backend: https://github.com/laisuk/opencc-fmmseg
-* C API: https://github.com/laisuk/opencc-fmmseg/tree/master/capi/opencc-fmmseg-capi
-* .NET: https://github.com/laisuk/OpenccNet
-* Python: https://github.com/laisuk/opencc_purepy
+- Rust backend: https://github.com/laisuk/opencc-fmmseg
+- C API: https://github.com/laisuk/opencc-fmmseg/tree/master/capi/opencc-fmmseg-capi
+- .NET: https://github.com/laisuk/OpenccNet
+- Python: https://github.com/laisuk/opencc_purepy
+- Java: https://github.com/laisuk/OpenccJava
 
 ---