@xberg-io/xberg-wasm 1.2.2 → 1.2.3

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@xberg-io/xberg-wasm",
-  "version": "1.2.2",
+  "version": "1.2.3",
   "private": false,
   "description": "High-performance document intelligence library",
   "license": "MIT",

package/pkg/web/package.json

@@ -2,7 +2,7 @@
   "name": "xberg-wasm",
   "type": "module",
   "description": "High-performance document intelligence library",
-  "version": "1.2.2",
+  "version": "1.2.3",
   "license": "MIT",
   "repository": {
     "type": "git",

package/pkg/web/xberg_wasm.d.ts

@@ -4330,6 +4330,22 @@ export class WasmNodeContent {
     set title(value: string | null | undefined);
 }
 
+/**
+ * A registered OCR backend's declared name and language capabilities.
+ *
+ * Returned by `list_ocr_backend_capabilities`. See that function's documentation for the
+ * determinism guarantees and the important caveat about what an empty `supported_languages`
+ * means.
+ */
+export class WasmOcrBackendCapabilities {
+    free(): void;
+    [Symbol.dispose](): void;
+    static default(): WasmOcrBackendCapabilities;
+    constructor(name: string, supportedLanguages: string[]);
+    name: string;
+    supportedLanguages: string[];
+}
+
 /**
  * OCR backend types.
  */
@@ -7213,6 +7229,35 @@ export function listDocumentExtractors(): string[];
  */
 export function listEmbeddingBackends(): string[];
 
+/**
+ * List every registered OCR backend's name alongside its declared supported languages.
+ *
+ * This is the capability-enumeration counterpart to `list_ocr_backends`: where that function
+ * exposes only backend names, this exposes each backend's `supported_languages()` too, so a
+ * consumer (for example, a job-acceptance gate) does not need to hardcode a second list of
+ * backend languages.
+ *
+ * # Determinism
+ *
+ * The returned vector is sorted by `name`, regardless of registration order or the order
+ * reported by the underlying registry. `supported_languages` within each entry is **not**
+ * sorted — see `OcrBackendCapabilities.supported_languages` for why.
+ *
+ * # Cost
+ *
+ * Calling this is not free for every backend. In particular, `TesseractBackend`'s
+ * `supported_languages()` allocates a Tesseract API and initializes it (`init("", "eng")`) the
+ * first time it is called, to enumerate installed tessdata languages; subsequent calls are
+ * served from a cache.
+ *
+ * # Errors
+ *
+ * Returns an error only if the registry lock cannot be acquired in the current environment.
+ *
+ * # Example
+ */
+export function listOcrBackendCapabilities(): WasmOcrBackendCapabilities[];
+
 /**
  * List all registered OCR backends.
  *
@@ -7316,6 +7361,29 @@ export function listValidators(): string[];
  */
 export function mapUrl(uri: string, config?: WasmUrlExtractionConfig | null): Promise<WasmMapResult>;
 
+/**
+ * Check whether a specific registered OCR backend supports a language.
+ *
+ * Delegates to the named backend's own `OcrBackend.supports_language`, which is the correct
+ * per-language decision — do not infer support (or its absence) from whether
+ * `list_ocr_backend_capabilities` reports an empty `supported_languages` list for that
+ * backend, since an empty list can mean "does not enumerate" rather than "supports nothing"
+ * (see `OcrBackendCapabilities.supported_languages`).
+ *
+ * # Arguments
+ *
+ * * `backend` - Name of a registered OCR backend, as returned by `list_ocr_backends`. Lookup
+ *   is case-insensitive and resolves the same `paddleocr` alias as backend dispatch.
+ * * `language` - Language code to check (e.g. `"eng"`, `"deu"`).
+ *
+ * # Errors
+ *
+ * Returns an error if no backend with that name (or alias) is registered.
+ *
+ * # Example
+ */
+export function ocrBackendSupportsLanguage(backend: string, language: string): boolean;
+
 export function registerDocumentExtractor(backend: any): void;
 
 export function registerEmbeddingBackend(backend: any): void;
@@ -7517,6 +7585,7 @@ export interface InitOutput {
     readonly __wbg_wasmngramrange_free: (a: number, b: number) => void;
     readonly __wbg_wasmnochunkingreason_free: (a: number, b: number) => void;
     readonly __wbg_wasmnodecontent_free: (a: number, b: number) => void;
+    readonly __wbg_wasmocrbackendcapabilities_free: (a: number, b: number) => void;
     readonly __wbg_wasmocrboundinggeometry_free: (a: number, b: number) => void;
     readonly __wbg_wasmocrconfidence_free: (a: number, b: number) => void;
     readonly __wbg_wasmocrconfig_free: (a: number, b: number) => void;
@@ -7636,6 +7705,7 @@ export interface InitOutput {
     readonly findUnmarkedClaims: (a: number, b: number) => [number, number];
     readonly listDocumentExtractors: () => [number, number, number, number];
     readonly listEmbeddingBackends: () => [number, number, number, number];
+    readonly listOcrBackendCapabilities: () => [number, number, number, number];
     readonly listOcrBackends: () => [number, number, number, number];
     readonly listPostProcessors: () => [number, number, number, number];
     readonly listRenderers: () => [number, number, number, number];
@@ -7646,6 +7716,7 @@ export interface InitOutput {
     readonly mapUrl: (a: number, b: number, c: number) => any;
     readonly nermodel_detect: (a: number, b: number, c: number, d: any) => any;
     readonly nermodel_load: (a: any) => any;
+    readonly ocrBackendSupportsLanguage: (a: number, b: number, c: number, d: number) => [number, number, number];
     readonly registerDocumentExtractor: (a: any) => [number, number];
     readonly registerEmbeddingBackend: (a: any) => [number, number];
     readonly registerOcrBackend: (a: any) => [number, number];
@@ -9861,6 +9932,12 @@ export interface InitOutput {
     readonly wasmnodecontent_term: (a: number) => [number, number];
     readonly wasmnodecontent_text: (a: number) => [number, number];
     readonly wasmnodecontent_title: (a: number) => [number, number];
+    readonly wasmocrbackendcapabilities_default: () => number;
+    readonly wasmocrbackendcapabilities_name: (a: number) => [number, number];
+    readonly wasmocrbackendcapabilities_new: (a: number, b: number, c: number, d: number) => number;
+    readonly wasmocrbackendcapabilities_set_name: (a: number, b: number, c: number) => void;
+    readonly wasmocrbackendcapabilities_set_supportedLanguages: (a: number, b: number, c: number) => void;
+    readonly wasmocrbackendcapabilities_supportedLanguages: (a: number) => [number, number];
     readonly wasmocrboundinggeometry_default: () => number;
     readonly wasmocrboundinggeometry_height: (a: number) => number;
     readonly wasmocrboundinggeometry_left: (a: number) => number;
@@ -10811,7 +10888,6 @@ export interface InitOutput {
     readonly wasmsparseembeddingpreset_modelFile: (a: number) => [number, number];
     readonly wasmsparseembeddingpreset_modelRepo: (a: number) => [number, number];
     readonly wasmsparseembeddingpreset_name: (a: number) => [number, number];
-    readonly wasmsparseembeddingpreset_new: (a: number, b: number, c: number, d: number, e: number, f: number, g: number, h: number, i: number, j: number, k: number) => number;
     readonly qcms_profile_precache_output_transform: (a: number) => void;
     readonly lut_inverse_interp16: (a: number, b: number, c: number) => number;
     readonly lut_interp_linear16: (a: number, b: number, c: number) => number;
@@ -10824,9 +10900,10 @@ export interface InitOutput {
     readonly qcms_transform_data_bgra_out_lut_precache: (a: number, b: number, c: number, d: number) => void;
     readonly qcms_profile_is_bogus: (a: number) => number;
     readonly qcms_transform_release: (a: number) => void;
+    readonly wasmsparseembeddingpreset_new: (a: number, b: number, c: number, d: number, e: number, f: number, g: number, h: number, i: number, j: number, k: number) => number;
     readonly wasmsparseembeddingpreset_set_additionalFiles: (a: number, b: number, c: number) => void;
-    readonly wasmsparseembeddingpreset_set_description: (a: number, b: number, c: number) => void;
     readonly __cxa_atexit: (a: number, b: number, c: number) => number;
+    readonly wasmsparseembeddingpreset_set_description: (a: number, b: number, c: number) => void;
     readonly wasmsparseembeddingpreset_set_maxLength: (a: number, b: number) => void;
     readonly wasmsparseembeddingpreset_set_modelFile: (a: number, b: number, c: number) => void;
     readonly wasmsparseembeddingpreset_set_modelRepo: (a: number, b: number, c: number) => void;

package/pkg/web/xberg_wasm.js

@@ -25029,6 +25029,94 @@ export class WasmNodeContent {
 }
 if (Symbol.dispose) WasmNodeContent.prototype[Symbol.dispose] = WasmNodeContent.prototype.free;
 
+/**
+ * A registered OCR backend's declared name and language capabilities.
+ *
+ * Returned by `list_ocr_backend_capabilities`. See that function's documentation for the
+ * determinism guarantees and the important caveat about what an empty `supported_languages`
+ * means.
+ */
+export class WasmOcrBackendCapabilities {
+    static __wrap(ptr) {
+        const obj = Object.create(WasmOcrBackendCapabilities.prototype);
+        obj.__wbg_ptr = ptr;
+        WasmOcrBackendCapabilitiesFinalization.register(obj, obj.__wbg_ptr, obj);
+        return obj;
+    }
+    __destroy_into_raw() {
+        const ptr = this.__wbg_ptr;
+        this.__wbg_ptr = 0;
+        WasmOcrBackendCapabilitiesFinalization.unregister(this);
+        return ptr;
+    }
+    free() {
+        const ptr = this.__destroy_into_raw();
+        wasm.__wbg_wasmocrbackendcapabilities_free(ptr, 0);
+    }
+    /**
+     * @returns {WasmOcrBackendCapabilities}
+     */
+    static default() {
+        const ret = wasm.wasmocrbackendcapabilities_default();
+        return WasmOcrBackendCapabilities.__wrap(ret);
+    }
+    /**
+     * @returns {string}
+     */
+    get name() {
+        let deferred1_0;
+        let deferred1_1;
+        try {
+            const ret = wasm.wasmocrbackendcapabilities_name(this.__wbg_ptr);
+            deferred1_0 = ret[0];
+            deferred1_1 = ret[1];
+            return getStringFromWasm0(ret[0], ret[1]);
+        } finally {
+            wasm.__wbindgen_free_command_export(deferred1_0, deferred1_1, 1);
+        }
+    }
+    /**
+     * @param {string} name
+     * @param {string[]} supportedLanguages
+     */
+    constructor(name, supportedLanguages) {
+        const ptr0 = passStringToWasm0(name, wasm.__wbindgen_malloc_command_export, wasm.__wbindgen_realloc_command_export);
+        const len0 = WASM_VECTOR_LEN;
+        const ptr1 = passArrayJsValueToWasm0(supportedLanguages, wasm.__wbindgen_malloc_command_export);
+        const len1 = WASM_VECTOR_LEN;
+        const ret = wasm.wasmocrbackendcapabilities_new(ptr0, len0, ptr1, len1);
+        this.__wbg_ptr = ret;
+        WasmOcrBackendCapabilitiesFinalization.register(this, this.__wbg_ptr, this);
+        return this;
+    }
+    /**
+     * @param {string} value
+     */
+    set name(value) {
+        const ptr0 = passStringToWasm0(value, wasm.__wbindgen_malloc_command_export, wasm.__wbindgen_realloc_command_export);
+        const len0 = WASM_VECTOR_LEN;
+        wasm.wasmocrbackendcapabilities_set_name(this.__wbg_ptr, ptr0, len0);
+    }
+    /**
+     * @param {string[]} value
+     */
+    set supportedLanguages(value) {
+        const ptr0 = passArrayJsValueToWasm0(value, wasm.__wbindgen_malloc_command_export);
+        const len0 = WASM_VECTOR_LEN;
+        wasm.wasmocrbackendcapabilities_set_supportedLanguages(this.__wbg_ptr, ptr0, len0);
+    }
+    /**
+     * @returns {string[]}
+     */
+    get supportedLanguages() {
+        const ret = wasm.wasmocrbackendcapabilities_supportedLanguages(this.__wbg_ptr);
+        var v1 = getArrayJsValueFromWasm0(ret[0], ret[1]);
+        wasm.__wbindgen_free_command_export(ret[0], ret[1] * 4, 4);
+        return v1;
+    }
+}
+if (Symbol.dispose) WasmOcrBackendCapabilities.prototype[Symbol.dispose] = WasmOcrBackendCapabilities.prototype.free;
+
 /**
  * OCR backend types.
  * @enum {0 | 1 | 2 | 3}
@@ -40330,6 +40418,44 @@ export function listEmbeddingBackends() {
     return v1;
 }
 
+/**
+ * List every registered OCR backend's name alongside its declared supported languages.
+ *
+ * This is the capability-enumeration counterpart to `list_ocr_backends`: where that function
+ * exposes only backend names, this exposes each backend's `supported_languages()` too, so a
+ * consumer (for example, a job-acceptance gate) does not need to hardcode a second list of
+ * backend languages.
+ *
+ * # Determinism
+ *
+ * The returned vector is sorted by `name`, regardless of registration order or the order
+ * reported by the underlying registry. `supported_languages` within each entry is **not**
+ * sorted — see `OcrBackendCapabilities.supported_languages` for why.
+ *
+ * # Cost
+ *
+ * Calling this is not free for every backend. In particular, `TesseractBackend`'s
+ * `supported_languages()` allocates a Tesseract API and initializes it (`init("", "eng")`) the
+ * first time it is called, to enumerate installed tessdata languages; subsequent calls are
+ * served from a cache.
+ *
+ * # Errors
+ *
+ * Returns an error only if the registry lock cannot be acquired in the current environment.
+ *
+ * # Example
+ * @returns {WasmOcrBackendCapabilities[]}
+ */
+export function listOcrBackendCapabilities() {
+    const ret = wasm.listOcrBackendCapabilities();
+    if (ret[3]) {
+        throw takeFromExternrefTable0(ret[2]);
+    }
+    var v1 = getArrayJsValueFromWasm0(ret[0], ret[1]);
+    wasm.__wbindgen_free_command_export(ret[0], ret[1] * 4, 4);
+    return v1;
+}
+
 /**
  * List all registered OCR backends.
  *
@@ -40506,6 +40632,42 @@ export function mapUrl(uri, config) {
     return ret;
 }
 
+/**
+ * Check whether a specific registered OCR backend supports a language.
+ *
+ * Delegates to the named backend's own `OcrBackend.supports_language`, which is the correct
+ * per-language decision — do not infer support (or its absence) from whether
+ * `list_ocr_backend_capabilities` reports an empty `supported_languages` list for that
+ * backend, since an empty list can mean "does not enumerate" rather than "supports nothing"
+ * (see `OcrBackendCapabilities.supported_languages`).
+ *
+ * # Arguments
+ *
+ * * `backend` - Name of a registered OCR backend, as returned by `list_ocr_backends`. Lookup
+ *   is case-insensitive and resolves the same `paddleocr` alias as backend dispatch.
+ * * `language` - Language code to check (e.g. `"eng"`, `"deu"`).
+ *
+ * # Errors
+ *
+ * Returns an error if no backend with that name (or alias) is registered.
+ *
+ * # Example
+ * @param {string} backend
+ * @param {string} language
+ * @returns {boolean}
+ */
+export function ocrBackendSupportsLanguage(backend, language) {
+    const ptr0 = passStringToWasm0(backend, wasm.__wbindgen_malloc_command_export, wasm.__wbindgen_realloc_command_export);
+    const len0 = WASM_VECTOR_LEN;
+    const ptr1 = passStringToWasm0(language, wasm.__wbindgen_malloc_command_export, wasm.__wbindgen_realloc_command_export);
+    const len1 = WASM_VECTOR_LEN;
+    const ret = wasm.ocrBackendSupportsLanguage(ptr0, len0, ptr1, len1);
+    if (ret[2]) {
+        throw takeFromExternrefTable0(ret[1]);
+    }
+    return ret[0] !== 0;
+}
+
 /**
  * @param {any} backend
  */
@@ -41275,7 +41437,7 @@ function __wbg_get_imports() {
             const ret = arg0.value;
             return ret;
         },
-        __wbg_warn_ccfd5b4f6afa280c: function(arg0, arg1) {
+        __wbg_warn_3b898cbca4e507de: function(arg0, arg1) {
             console.warn(getStringFromWasm0(arg0, arg1));
         },
         __wbg_wasmarchiveentry_new: function(arg0) {
@@ -41626,6 +41788,10 @@ function __wbg_get_imports() {
             const ret = WasmMarkdownLink.__unwrap(arg0);
             return ret;
         },
+        __wbg_wasmocrbackendcapabilities_new: function(arg0) {
+            const ret = WasmOcrBackendCapabilities.__wrap(arg0);
+            return ret;
+        },
         __wbg_wasmocrelement_new: function(arg0) {
             const ret = WasmOcrElement.__wrap(arg0);
             return ret;
@@ -41835,12 +42001,12 @@ function __wbg_get_imports() {
             return ret;
         },
         __wbindgen_generic_0000000000000001: function(arg0, arg1) {
-            // Cast intrinsic for `Closure(Closure { owned: true, function: Function { arguments: [Externref], shim_idx: 669, ret: Result(Unit), inner_ret: Some(Result(Unit)) }, mutable: true }) -> Externref`.
+            // Cast intrinsic for `Closure(Closure { owned: true, function: Function { arguments: [Externref], shim_idx: 673, ret: Result(Unit), inner_ret: Some(Result(Unit)) }, mutable: true }) -> Externref`.
             const ret = makeMutClosure(arg0, arg1, wasm_bindgen__convert__closures_____invoke__h26802dab031c3123);
             return ret;
         },
         __wbindgen_generic_0000000000000002: function(arg0, arg1) {
-            // Cast intrinsic for `Closure(Closure { owned: true, function: Function { arguments: [], shim_idx: 1165, ret: Unit, inner_ret: Some(Unit) }, mutable: true }) -> Externref`.
+            // Cast intrinsic for `Closure(Closure { owned: true, function: Function { arguments: [], shim_idx: 1170, ret: Unit, inner_ret: Some(Unit) }, mutable: true }) -> Externref`.
             const ret = makeMutClosure(arg0, arg1, wasm_bindgen__convert__closures_____invoke__ha4d0d7342deba426);
             return ret;
         },
@@ -42370,6 +42536,9 @@ const WasmNoChunkingReasonFinalization = (typeof FinalizationRegistry === 'undef
 const WasmNodeContentFinalization = (typeof FinalizationRegistry === 'undefined')
     ? { register: () => {}, unregister: () => {} }
     : new FinalizationRegistry(ptr => wasm.__wbg_wasmnodecontent_free(ptr, 1));
+const WasmOcrBackendCapabilitiesFinalization = (typeof FinalizationRegistry === 'undefined')
+    ? { register: () => {}, unregister: () => {} }
+    : new FinalizationRegistry(ptr => wasm.__wbg_wasmocrbackendcapabilities_free(ptr, 1));
 const WasmOcrBoundingGeometryFinalization = (typeof FinalizationRegistry === 'undefined')
     ? { register: () => {}, unregister: () => {} }
     : new FinalizationRegistry(ptr => wasm.__wbg_wasmocrboundinggeometry_free(ptr, 1));

package/pkg/web/xberg_wasm_bg.wasm

@@ -1,4 +1,4 @@
 [diffend] Oversized file quarantined before diffing.
 name: package/pkg/web/xberg_wasm_bg.wasm
-size: 36416412 bytes
-sha256: 42be852e4482dfa6549302c597360a55107da23cf319f5882f0b59ab61eb05f1
+size: 36443440 bytes
+sha256: 7d6a595c701378198ac0c6d747c094faf0e4bc6825e0041c14eeb3a1906e39f3

package/pkg/web/xberg_wasm_bg.wasm.d.ts

@@ -146,6 +146,7 @@ export const __wbg_wasmnerconfig_free: (a: number, b: number) => void;
 export const __wbg_wasmngramrange_free: (a: number, b: number) => void;
 export const __wbg_wasmnochunkingreason_free: (a: number, b: number) => void;
 export const __wbg_wasmnodecontent_free: (a: number, b: number) => void;
+export const __wbg_wasmocrbackendcapabilities_free: (a: number, b: number) => void;
 export const __wbg_wasmocrboundinggeometry_free: (a: number, b: number) => void;
 export const __wbg_wasmocrconfidence_free: (a: number, b: number) => void;
 export const __wbg_wasmocrconfig_free: (a: number, b: number) => void;
@@ -265,6 +266,7 @@ export const extractBatch: (a: number, b: number, c: number) => any;
 export const findUnmarkedClaims: (a: number, b: number) => [number, number];
 export const listDocumentExtractors: () => [number, number, number, number];
 export const listEmbeddingBackends: () => [number, number, number, number];
+export const listOcrBackendCapabilities: () => [number, number, number, number];
 export const listOcrBackends: () => [number, number, number, number];
 export const listPostProcessors: () => [number, number, number, number];
 export const listRenderers: () => [number, number, number, number];
@@ -275,6 +277,7 @@ export const listValidators: () => [number, number, number, number];
 export const mapUrl: (a: number, b: number, c: number) => any;
 export const nermodel_detect: (a: number, b: number, c: number, d: any) => any;
 export const nermodel_load: (a: any) => any;
+export const ocrBackendSupportsLanguage: (a: number, b: number, c: number, d: number) => [number, number, number];
 export const registerDocumentExtractor: (a: any) => [number, number];
 export const registerEmbeddingBackend: (a: any) => [number, number];
 export const registerOcrBackend: (a: any) => [number, number];
@@ -2490,6 +2493,12 @@ export const wasmnodecontent_src: (a: number) => [number, number];
 export const wasmnodecontent_term: (a: number) => [number, number];
 export const wasmnodecontent_text: (a: number) => [number, number];
 export const wasmnodecontent_title: (a: number) => [number, number];
+export const wasmocrbackendcapabilities_default: () => number;
+export const wasmocrbackendcapabilities_name: (a: number) => [number, number];
+export const wasmocrbackendcapabilities_new: (a: number, b: number, c: number, d: number) => number;
+export const wasmocrbackendcapabilities_set_name: (a: number, b: number, c: number) => void;
+export const wasmocrbackendcapabilities_set_supportedLanguages: (a: number, b: number, c: number) => void;
+export const wasmocrbackendcapabilities_supportedLanguages: (a: number) => [number, number];
 export const wasmocrboundinggeometry_default: () => number;
 export const wasmocrboundinggeometry_height: (a: number) => number;
 export const wasmocrboundinggeometry_left: (a: number) => number;
@@ -3440,7 +3449,6 @@ export const wasmsparseembeddingpreset_maxLength: (a: number) => number;
 export const wasmsparseembeddingpreset_modelFile: (a: number) => [number, number];
 export const wasmsparseembeddingpreset_modelRepo: (a: number) => [number, number];
 export const wasmsparseembeddingpreset_name: (a: number) => [number, number];
-export const wasmsparseembeddingpreset_new: (a: number, b: number, c: number, d: number, e: number, f: number, g: number, h: number, i: number, j: number, k: number) => number;
 export const qcms_profile_precache_output_transform: (a: number) => void;
 export const lut_inverse_interp16: (a: number, b: number, c: number) => number;
 export const lut_interp_linear16: (a: number, b: number, c: number) => number;
@@ -3453,9 +3461,10 @@ export const qcms_transform_data_rgba_out_lut_precache: (a: number, b: number, c
 export const qcms_transform_data_bgra_out_lut_precache: (a: number, b: number, c: number, d: number) => void;
 export const qcms_profile_is_bogus: (a: number) => number;
 export const qcms_transform_release: (a: number) => void;
+export const wasmsparseembeddingpreset_new: (a: number, b: number, c: number, d: number, e: number, f: number, g: number, h: number, i: number, j: number, k: number) => number;
 export const wasmsparseembeddingpreset_set_additionalFiles: (a: number, b: number, c: number) => void;
-export const wasmsparseembeddingpreset_set_description: (a: number, b: number, c: number) => void;
 export const __cxa_atexit: (a: number, b: number, c: number) => number;
+export const wasmsparseembeddingpreset_set_description: (a: number, b: number, c: number) => void;
 export const wasmsparseembeddingpreset_set_maxLength: (a: number, b: number) => void;
 export const wasmsparseembeddingpreset_set_modelFile: (a: number, b: number, c: number) => void;
 export const wasmsparseembeddingpreset_set_modelRepo: (a: number, b: number, c: number) => void;