@laminapdf/wasm 0.2.0-alpha.3 → 0.2.0-alpha.5

package/dist/lamina_pdf_text.js

@@ -278,6 +278,25 @@ export class WasmDocument {
         }
         return takeFromExternrefTable0(ret[0]);
     }
+    /**
+     * An extraction stepped a page at a time, whose models the host runs: each step returns
+     * its progress, with the result it waits for under `request` (render that page at `dpi`,
+     * run the model, hand the result to `provide`), and once every page is done, the output in
+     * `options.format` (the extract and text modules).
+     *
+     * # Errors
+     * As `extract`, for the options and the pages asked for.
+     * @param {ExtractInit} options
+     * @returns {WasmExtractJob}
+     */
+    extractJob(options) {
+        let ret;
+        ret = wasm.wasmdocument_extractJob(this.__wbg_ptr, options);
+        if (ret[2]) {
+            throw takeFromExternrefTable0(ret[1]);
+        }
+        return WasmExtractJob.__wrap(ret[0]);
+    }
     /**
      * The document's text for retrieval and LLMs, as one string: Markdown, plain text, the
      * `laminapdf.extract/1` JSON, or the chunks as a JSON array (the extract and text modules).
@@ -331,6 +350,40 @@ export class WasmDocument {
         }
         return takeFromExternrefTable0(ret[0]);
     }
+    /**
+     * Whether the document says it is tagged (`/MarkInfo /Marked true`; the extract and text modules).
+     *
+     * # Errors
+     * Reading errors.
+     * @returns {boolean}
+     */
+    isTagged() {
+        let ret;
+        ret = wasm.wasmdocument_isTagged(this.__wbg_ptr);
+        if (ret[2]) {
+            throw takeFromExternrefTable0(ret[1]);
+        }
+        return ret[0] !== 0;
+    }
+    /**
+     * The named destination `name`, or `undefined` when the document has none by that name or
+     * it names no page here (the extract and text modules).
+     *
+     * # Errors
+     * `data-pending` and the reading errors.
+     * @param {string} name
+     * @returns {TargetJs | undefined}
+     */
+    namedDestination(name) {
+        const ptr0 = passStringToWasm0(name, wasm.__wbindgen_malloc, wasm.__wbindgen_realloc);
+        const len0 = WASM_VECTOR_LEN;
+        let ret;
+        ret = wasm.wasmdocument_namedDestination(this.__wbg_ptr, ptr0, len0);
+        if (ret[2]) {
+            throw takeFromExternrefTable0(ret[1]);
+        }
+        return takeFromExternrefTable0(ret[0]);
+    }
     /**
      * Opens a document from its bytes.
      *
@@ -371,6 +424,22 @@ export class WasmDocument {
         }
         return WasmDocument.__wrap(ret[0]);
     }
+    /**
+     * The outline's top-level items, each with its children; empty without an outline (the
+     * extract and text modules).
+     *
+     * # Errors
+     * `data-pending` and the reading errors.
+     * @returns {OutlineNodeJs[]}
+     */
+    outline() {
+        let ret;
+        ret = wasm.wasmdocument_outline(this.__wbg_ptr);
+        if (ret[2]) {
+            throw takeFromExternrefTable0(ret[1]);
+        }
+        return takeFromExternrefTable0(ret[0]);
+    }
     /**
      * A handle on the page with `id` (`p12.0`).
      *
@@ -435,6 +504,22 @@ export class WasmDocument {
         }
         return takeFromExternrefTable0(ret[0]);
     }
+    /**
+     * The visible links on page `index`, in annotation order (the extract and text modules).
+     *
+     * # Errors
+     * `page-index-out-of-range`, `data-pending` and the reading errors.
+     * @param {number} index
+     * @returns {PageLinkJs[]}
+     */
+    pageLinks(index) {
+        let ret;
+        ret = wasm.wasmdocument_pageLinks(this.__wbg_ptr, index);
+        if (ret[2]) {
+            throw takeFromExternrefTable0(ret[1]);
+        }
+        return takeFromExternrefTable0(ret[0]);
+    }
     /**
      * A handle on page `index` that follows the page by its ID (F2).
      *
@@ -451,6 +536,21 @@ export class WasmDocument {
         }
         return WasmPage.__wrap(ret[0]);
     }
+    /**
+     * The document's properties (the extract and text modules). Reads every object once.
+     *
+     * # Errors
+     * `data-pending` and the reading errors.
+     * @returns {PropertiesJs}
+     */
+    properties() {
+        let ret;
+        ret = wasm.wasmdocument_properties(this.__wbg_ptr);
+        if (ret[2]) {
+            throw takeFromExternrefTable0(ret[1]);
+        }
+        return takeFromExternrefTable0(ret[0]);
+    }
     /**
      * Rebases onto `bytes`, the output of a save whose rebase failed (`not-rebased`).
      *
@@ -514,6 +614,22 @@ export class WasmDocument {
         }
         return takeFromExternrefTable0(ret[0]);
     }
+    /**
+     * The structure tree's top-level elements in reading order, or `undefined` without one (the
+     * extract and text modules, beside `WasmPage.text`, whose runs it links by MCID).
+     *
+     * # Errors
+     * Reading errors.
+     * @returns {StructureNodeJs[] | undefined}
+     */
+    structureTree() {
+        let ret;
+        ret = wasm.wasmdocument_structureTree(this.__wbg_ptr);
+        if (ret[2]) {
+            throw takeFromExternrefTable0(ret[1]);
+        }
+        return takeFromExternrefTable0(ret[0]);
+    }
     /**
      * Undoes the newest transaction; `undefined` when there is nothing to undo.
      *
@@ -541,6 +657,73 @@ export class WasmDocument {
 }
 if (Symbol.dispose) WasmDocument.prototype[Symbol.dispose] = WasmDocument.prototype.free;
 
+/**
+ * An extraction in progress, from `WasmDocument.extractJob`; it holds its document weakly.
+ */
+export class WasmExtractJob {
+    static __wrap(ptr) {
+        const obj = Object.create(WasmExtractJob.prototype);
+        obj.__wbg_ptr = ptr;
+        Object.defineProperty(obj, '__wbg_inst', { value: __wbg_instance_id, writable: true });
+
+        WasmExtractJobFinalization.register(obj, { ptr, instance: __wbg_instance_id }, obj);
+        return obj;
+    }
+    __destroy_into_raw() {
+        const ptr = this.__wbg_ptr;
+        this.__wbg_ptr = 0;
+        WasmExtractJobFinalization.unregister(this);
+        return ptr;
+    }
+    free() {
+        const ptr = this.__destroy_into_raw();
+        wasm.__wbg_wasmextractjob_free(ptr, 0);
+
+    }
+    /**
+     * Stops the job: its next step fails with `cancelled`.
+     */
+    cancel() {
+        wasm.wasmextractjob_cancel(this.__wbg_ptr);
+    }
+    /**
+     * Hands the job the result of its open request: `{ kind: "layout", regions }` or
+     * `{ kind: "ocr", words }`, boxes as `[x, y, width, height]` in top-left points.
+     *
+     * # Errors
+     * `model-result-unexpected` when nothing was asked for or `result` is the other kind;
+     * `invalid-argument` / `unknown-option` for a malformed result.
+     * @param {ModelResultInit} result
+     */
+    provide(result) {
+        let ret;
+        ret = wasm.wasmextractjob_provide(this.__wbg_ptr, result);
+        if (ret[1]) {
+            throw takeFromExternrefTable0(ret[0]);
+        }
+    }
+    /**
+     * Extracts up to `budget` pages: `{ state: "progress", done, total, phase, request? }` or
+     * `{ state: "done", output }`, the output in the job's format. While `request` is set, a
+     * step does nothing until `provide` answers it.
+     *
+     * # Errors
+     * `cancelled`, `document-changed`, `job-finished`, `disposed` when the document is freed,
+     * and the reading errors.
+     * @param {number} budget
+     * @returns {({ state: "progress" } & ExtractProgressJs) | { state: "done"; output: string }}
+     */
+    step(budget) {
+        let ret;
+        ret = wasm.wasmextractjob_step(this.__wbg_ptr, budget);
+        if (ret[2]) {
+            throw takeFromExternrefTable0(ret[1]);
+        }
+        return takeFromExternrefTable0(ret[0]);
+    }
+}
+if (Symbol.dispose) WasmExtractJob.prototype[Symbol.dispose] = WasmExtractJob.prototype.free;
+
 /**
  * A font registered with a [`WasmBuilder`].
  */
@@ -828,7 +1011,8 @@ export class WasmPageBuilder {
     }
     /**
      * Opens an artifact until the matching `endTag`: `kind` is `layout`, `page`, `pagination`
-     * or `background`.
+     * or `background`, or a pagination artifact that says what it is: `header` (attached to the
+     * top edge), `footer` (the bottom) or `watermark`.
      *
      * # Errors
      * `invalid-argument` for another kind.
@@ -1648,6 +1832,11 @@ const WasmDocumentFinalization = (typeof FinalizationRegistry === 'undefined')
     : new FinalizationRegistry(({ ptr, instance }) => {
     if (instance === __wbg_instance_id) wasm.__wbg_wasmdocument_free(ptr, 1);
 });
+const WasmExtractJobFinalization = (typeof FinalizationRegistry === 'undefined')
+    ? { register: () => {}, unregister: () => {} }
+    : new FinalizationRegistry(({ ptr, instance }) => {
+    if (instance === __wbg_instance_id) wasm.__wbg_wasmextractjob_free(ptr, 1);
+});
 const WasmFontFinalization = (typeof FinalizationRegistry === 'undefined')
     ? { register: () => {}, unregister: () => {} }
     : new FinalizationRegistry(({ ptr, instance }) => {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@laminapdf/wasm",
-  "version": "0.2.0-alpha.3",
+  "version": "0.2.0-alpha.5",
   "description": "Lamina PDF's WebAssembly modules and their wasm-bindgen glue: lamina_pdf_base (open, read, edit, save, vector and image pages), lamina_pdf_text (all of base plus fonts and text) and lamina_pdf_extract (all of base plus text extraction); and lamina_pdf_render_hayro, laminapdf-render's hayro adapter, which @laminapdf/render-hayro wraps. Generated into dist/ by `cargo xtask wasm`; @laminapdf/sdk is the API to use.",
   "license": "MIT OR Apache-2.0",
   "homepage": "https://laminapdf.com",