@gukhanmun/wasm 0.1.0-dev.15 → 0.1.0-dev.18

package/dist/index.d.ts

@@ -123,6 +123,19 @@ type NumeralStrategy = "hangul-phonetic" | "positional-arabic" | "additive-arabi
  *   accuracy matters more than latency.
  */
 type ContextWindow = "off" | "per-block" | "per-section" | "per-document";
+/**
+ * Selects how the homophone marker decides that a reading needs its hanja shown
+ * in `rendering: "hangul-only"`.  Corresponds to Rust `HomophoneDetection`.
+ *
+ * - `"context-local"` — Gloss a reading only when a different-meaning homophone
+ *   actually appears within the {@link ContextWindow}.  This keeps hangul-only
+ *   output clean and is the default.
+ * - `"dictionary-wide"` — Also gloss readings shared by other hanja forms
+ *   anywhere in the dictionary, even when no homophone appears in the text.
+ *   With a large reference dictionary this glosses most Sino-Korean words; words
+ *   that should always be glossed are better expressed with `requireHanja`.
+ */
+type HomophoneDetection = "context-local" | "dictionary-wide";
 /**
  * Controls how the pipeline handles reader errors encountered during HTML
  * scanning.  Corresponds to Rust `Recovery`.
@@ -324,13 +337,22 @@ interface GukhanmunOptions {
   readonly initialSoundLaw?: boolean;
   /**
    * Context window for homophone disambiguation.  The `HomophoneMarker`
-   * middleware sets `homophone = true` on annotations whose hangul reading is
-   * shared by another hanja form within this window.  Defaults to
-   * `"per-block"`.
+   * middleware sets `homophone = true` on annotations whose hangul reading
+   * collides within this window.  Defaults to `"per-block"`.
    *
    * @see {@link ContextWindow}
    */
   readonly homophoneWindow?: ContextWindow;
+  /**
+   * Strategy that decides which readings count as homophones needing a hanja
+   * gloss.  Defaults to `"context-local"`, which glosses a reading only when a
+   * different-meaning homophone appears within {@link homophoneWindow}.  Use
+   * `"dictionary-wide"` to also gloss readings shared by other dictionary
+   * entries.
+   *
+   * @see {@link HomophoneDetection}
+   */
+  readonly homophoneDetection?: HomophoneDetection;
   /**
    * Context window for first-occurrence filtering.  The
    * `FirstOccurrenceFilter` middleware clears `requireHanja` /
@@ -572,4 +594,4 @@ declare class GukhanmunError extends Error {
  */
 declare function load(options?: GukhanmunOptions): Promise<Gukhanmun>;
 //#endregion
-export { type ContextWindow, type DictionaryEntry, type DictionarySource, type Directives, type ErrorCode, type FileDictionarySource, type Format, type Gukhanmun, GukhanmunError, type GukhanmunFactory, type GukhanmunOptions, type HtmlOptions, type NumeralStrategy, type OriginalGloss, type Preset, type Recovery, type RenderMode, type Segmentation, load };
+export { type ContextWindow, type DictionaryEntry, type DictionarySource, type Directives, type ErrorCode, type FileDictionarySource, type Format, type Gukhanmun, GukhanmunError, type GukhanmunFactory, type GukhanmunOptions, type HomophoneDetection, type HtmlOptions, type NumeralStrategy, type OriginalGloss, type Preset, type Recovery, type RenderMode, type Segmentation, load };

package/dist/index.js

@@ -42,22 +42,36 @@ let wasmInit;
 * Loads and caches the WASM module.  The module URL is resolved relative to
 * this source file so it works with Deno's module graph and `import.meta.url`.
 *
-* Node.js `fetch()` does not support `file://` URLs, so on Node.js the WASM
-* binary is read via `node:fs/promises` and passed directly as a `Uint8Array`
-* to the web-target init function, bypassing the streaming-fetch path.
-* Deno and Bun support `fetch()` for `file://` URLs and use the default path.
+* Node.js `fetch()` does not support `file://` URLs, so when the binary is a
+* `file:` URL on Node.js it is read via `node:fs/promises` and passed directly
+* as a `Uint8Array` to the web-target init function, bypassing the
+* streaming-fetch path.  Every other case uses `fetch`: Deno and Bun support
+* `fetch()` for `file://` URLs, and a non-`file:` URL (such as the `https:`
+* URL of a JSR install run on Deno, where `process.versions.node` is also
+* defined) must be fetched since `node:fs` would reject it.
 */
 function ensureWasm() {
 	if (!wasmInit) {
 		const glueHref = new URL("./wasm/web/gukhanmun_wasm.js", import.meta.url).href;
 		wasmInit = (async () => {
-			const mod = await import(glueHref);
-			if (isNodeLike()) {
-				const wasmUrl = new URL("./wasm/web/gukhanmun_wasm_bg.wasm", import.meta.url);
-				const buf = await (await import("node:fs/promises")).readFile(wasmUrl);
+			const mod = await import(
+				/* webpackIgnore: true */
+				glueHref
+);
+			const wasmUrl = new URL("./wasm/web/gukhanmun_wasm_bg.wasm", import.meta.url);
+			if (isNodeLike() && wasmUrl.protocol === "file:") {
+				const buf = await (await import(
+					/* webpackIgnore: true */
+					"node:fs/promises"
+)).readFile(wasmUrl);
 				const bytes = new Uint8Array(buf.buffer, buf.byteOffset, buf.byteLength);
 				await mod.default({ module_or_path: bytes });
-			} else await mod.default();
+			} else {
+				const response = await fetch(wasmUrl);
+				if (!response.ok) throw new Error(`Failed to fetch WASM binary: HTTP ${response.status}`);
+				const bytes = new Uint8Array(await response.arrayBuffer());
+				await mod.default({ module_or_path: bytes });
+			}
 			return mod;
 		})();
 	}
@@ -80,11 +94,17 @@ async function resolveDictionary(source) {
 	else {
 		const str = String(source.data);
 		if (str.includes("://")) url = new URL(str);
-		else if (isNodeLike()) url = (await import("node:url")).pathToFileURL(str);
+		else if (isNodeLike()) url = (await import(
+			/* webpackIgnore: true */
+			"node:url"
+)).pathToFileURL(str);
 		else throw new GukhanmunError("invalid-input", "File path strings require a Node.js environment; use URL or ArrayBuffer in browsers");
 	}
 	if (isNodeLike() && url.protocol === "file:") {
-		const fs = await import("node:fs/promises");
+		const fs = await import(
+			/* webpackIgnore: true */
+			"node:fs/promises"
+);
 		let buf;
 		try {
 			buf = await fs.readFile(url);
@@ -186,6 +206,7 @@ function resolveOptions(opts = {}) {
 		numerals: opts.numerals ?? "hangul-phonetic",
 		initialSoundLaw: opts.initialSoundLaw ?? (koKp ? false : true),
 		homophoneWindow: opts.homophoneWindow ?? (koKp ? "off" : "per-block"),
+		homophoneDetection: opts.homophoneDetection ?? "context-local",
 		firstOccurrenceWindow: opts.firstOccurrenceWindow ?? "off",
 		recovery: opts.recovery ?? "strict"
 	};
@@ -229,6 +250,7 @@ function buildRawOptions(opts, resolved) {
 		numerals: resolved.numerals,
 		initialSoundLaw: resolved.initialSoundLaw,
 		homophoneWindow: resolved.homophoneWindow,
+		homophoneDetection: resolved.homophoneDetection,
 		firstOccurrenceWindow: resolved.firstOccurrenceWindow,
 		recovery: resolved.recovery
 	};

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@gukhanmun/wasm",
-  "version": "0.1.0-dev.15",
+  "version": "0.1.0-dev.18",
   "description": "WebAssembly implementation of the Gukhanmun hanja-to-hangul converter.",
   "type": "module",
   "exports": {
@@ -45,7 +45,7 @@
     "@gukhanmun/types": "*"
   },
   "devDependencies": {
-    "@gukhanmun/types": "0.1.0-dev.15+3fd33cbb18850ceb854950ab7837d01d307ba573"
+    "@gukhanmun/types": "0.1.0-dev.18+25611433dc656544ad355f3b96f45650c351cab9"
   },
   "scripts": {
     "build": "tsdown && node -e \"const fs=require('node:fs'),p=require('node:path');fs.cpSync('wasm','dist/wasm',{recursive:true,force:true,filter:(s)=>p.basename(s)!=='.gitignore'})\""