@pdfluent/sdk-wasm 1.0.0-beta.8 → 1.1.0

package/package.json

@@ -1,15 +1,20 @@
 {
   "name": "@pdfluent/sdk-wasm",
   "type": "module",
-  "description": "PDFluent browser SDK — read, edit, annotate, redact, sign, and validate PDFs (including XFA) entirely client-side via WASM.",
-  "version": "1.0.0-beta.8",
-  "license": "SEE LICENSE IN LICENSE",
+  "description": "PDFluent browser SDK — find and replace PDF text programmatically while preserving fonts and layout, plus edit, annotate, redact, sign and validate PDFs (including XFA), entirely client-side via WASM.",
+  "version": "1.1.0",
+  "license": "SEE LICENSE IN LICENSE-OFFER",
   "files": [
     "xfa_wasm_bg.wasm",
     "xfa_wasm.js",
     "xfa_wasm.d.ts",
     "snippets",
-    "xfa_wasm_bg.wasm.d.ts"
+    "xfa_wasm_bg.wasm.d.ts",
+    "LICENSE",
+    "LICENSE-COMMERCIAL",
+    "LICENSE-OFFER",
+    "NOTICE",
+    "THIRD_PARTY_LICENSES.txt"
   ],
   "main": "xfa_wasm.js",
   "types": "xfa_wasm.d.ts",
@@ -22,11 +27,21 @@
     "pdfluent",
     "browser",
     "sdk",
+    "pdf-text-replace",
+    "find-and-replace",
+    "edit-pdf-text",
+    "pdf-editor",
     "xfa",
+    "acroform",
+    "pades",
     "forms",
     "annotate",
     "redact"
   ],
   "homepage": "https://pdfluent.com",
-  "author": "Innovation Trigger BV <team@pdfluent.com>"
+  "author": "Innovation Trigger BV <team@pdfluent.com>",
+  "repository": {
+    "type": "git",
+    "url": "https://github.com/pdfluent/pdfluent-sdk"
+  }
 }

package/xfa_wasm.d.ts

@@ -1,6 +1,85 @@
 /* tslint:disable */
 /* eslint-disable */
 
+/** A stamp with preset text, size, rotation, opacity and colour (red, 60pt, 45°, 25%). */
+export type WatermarkPreset = "draft" | "confidential" | "approved" | "final" | "copy" | "notForDistribution";
+
+/** Options for `PdfDoc.addWatermark`. Give `text`, `preset`, or both (then `text` wins). */
+export interface WatermarkOptions {
+    text?: string;
+    preset?: WatermarkPreset;
+    /** Points. Default 48 (60 with a preset). */
+    fontSize?: number;
+    /** Degrees counter-clockwise. Default 45. */
+    rotation?: number;
+    /** 0..1. Default 0.3 (0.25 with a preset). */
+    opacity?: number;
+    /** `"#rrggbb"`. Default mid grey (red with a preset). */
+    color?: string;
+    /** `"center"` (default), or a corner / exact point in PDF points. */
+    position?: "center" | { anchor: "topLeft" | "topRight" | "bottomLeft" | "bottomRight" | "exact"; x: number; y: number };
+    /** Default `"foreground"`. */
+    layer?: "foreground" | "background";
+    /** Repeat across the page. `true` = 200pt spacing. */
+    tile?: boolean | { hSpacing?: number; vSpacing?: number };
+    /** `"all"` (default), `"even"`, `"odd"`, 0-based page indices, or an inclusive 0-based range. */
+    pages?: "all" | "even" | "odd" | number[] | { from: number; to: number };
+}
+
+
+
+/** An image exported faithfully. `page` is 0-based: the first page it is used on. */
+export interface ExtractedImage {
+    page: number;
+    width: number;
+    height: number;
+    mime: "image/png" | "image/jpeg";
+    data: Uint8Array;
+}
+
+/** An image left out because it could not be exported without changing how it looks. */
+export interface SkippedImage {
+    page: number;
+    width: number;
+    height: number;
+    reason: "unsupported_filter" | "transparency" | "decode_array" | "cmyk_jpeg" | "color_space" | "bit_depth" | "image_mask" | "undecodable";
+    detail: string;
+}
+
+export interface ImageExtraction {
+    images: ExtractedImage[];
+    skipped: SkippedImage[];
+}
+
+
+
+/** Options for `signPdf`. The signature is a cryptographic (PAdES) signature, not a drawn one. */
+export interface SignPdfOptions {
+    reason?: string;
+    location?: string;
+    contactInfo?: string;
+    /** Default "Signature1". */
+    fieldName?: string;
+    /** Show the signature on a page (0-based) at a rectangle in PDF points. Invisible when absent. */
+    visibleRect?: { page: number; x: number; y: number; width: number; height: number };
+}
+
+
+
+/** What a user who opens the file with the user password may do. Every flag defaults to `true`. */
+export interface EncryptPermissions {
+    print?: boolean;
+    printHighQuality?: boolean;
+    modify?: boolean;
+    copy?: boolean;
+    annotate?: boolean;
+    fillForms?: boolean;
+    accessibility?: boolean;
+    assemble?: boolean;
+}
+
+
+
 /**
  * Chroma subsampling format
  */
@@ -23,27 +102,6 @@ export enum ChromaSampling {
     Cs400 = 3,
 }
 
-/**
- * Status of the currently-active license.
- */
-export class LicenseStatus {
-    private constructor();
-    free(): void;
-    [Symbol.dispose](): void;
-    /**
-     * Whether output is marked via `/Producer` metadata (Trial tier).
-     */
-    readonly outputIsMarked: boolean;
-    /**
-     * Source: `"Default" | "EnvVar" | "Explicit"`.
-     */
-    readonly source: string;
-    /**
-     * Tier name: `"Trial" | "Developer" | "Team" | "Business" | "Enterprise"`.
-     */
-    readonly tier: string;
-}
-
 /**
  * PDF document handle for analysis and text extraction.
  *
@@ -98,8 +156,16 @@ export class PdfDoc {
     /**
      * Apply a diagonal text watermark to every page.
      * `opacity` in 0.0..=1.0. Default rotation 45 deg, default colour gray.
+     *
+     * Kept for compatibility; `addWatermark` takes the full set of options.
      */
     addTextWatermark(text: string, opacity: number): Uint8Array;
+    /**
+     * Apply a text watermark with full control over text or preset, size,
+     * rotation, opacity, colour, position, layer, tiling and pages. Returns
+     * the new PDF bytes. Text only: there is no image watermark.
+     */
+    addWatermark(options: WatermarkOptions): Uint8Array;
     /**
      * Re-encode all content streams with deflate compression. Removes
      * no-op overhead. Returns the optimised PDF as a `Uint8Array`.
@@ -120,6 +186,58 @@ export class PdfDoc {
      * DSS (Document Security Store) info as JSON, or null if absent.
      */
     dssInfo(): string | undefined;
+    /**
+     * The document's blocks as JSON, in reading order (#573).
+     *
+     * Preview: output quality is still improving and the API may change in a minor release.
+     *
+     * `{"blocks": [...]}`, one entry per paragraph, heading, list item, table
+     * cell and running head, each with its kind, page, box, text, spans and
+     * place in the reading order. `docs/STRUCTURED_EXTRACTION.md` is the key
+     * to every field.
+     *
+     * A JSON string and not a structured value, which is what every other
+     * structured reader in this binding hands back: one shape crosses the
+     * boundary, `JSON.parse` is what the caller already does for
+     * `getTextPositions` and the annotations, and a `serde` mirror of the
+     * block model would be a second definition of it to keep in step.
+     *
+     * An XFA form is read from its flattened form, for the reason
+     * `extractMarkdown` above gives.
+     */
+    extractBlocks(first_page?: number | null, page_count?: number | null): string;
+    /**
+     * The document as a Docling `DoclingDocument`, as JSON (#629).
+     *
+     * Preview: output quality is still improving and the API may change in a minor release.
+     *
+     * The schema version the engine pins, which docling-core's
+     * `DoclingDocument.model_validate_json` loads as it is. `name` is
+     * `DoclingDocument.name`, which Docling fills with the file's stem;
+     * absent means `"document"`. An XFA form is read from its flattened form,
+     * for the reason `extractMarkdown` above gives.
+     */
+    extractDocling(first_page?: number | null, page_count?: number | null, name?: string | null): string;
+    /**
+     * The document as markdown, in reading order (#573).
+     *
+     * Preview: output quality is still improving and the API may change in a minor release.
+     *
+     * Headings become `#` to `###`, list items `-` and `1.`, a table a pipe
+     * table or an HTML one where markdown cannot hold its shape, and a change
+     * of page an HTML comment. The running heads a document repeats are not in
+     * it; they are in `extractBlocks`.
+     *
+     * `firstPage` and `pageCount` choose a window of pages; leave both out for
+     * the whole document. The document is read whole whatever the window is,
+     * because what a block IS is decided against the rest of the document.
+     *
+     * AN XFA FORM IS READ FROM ITS FLATTENED FORM, the way `text` and
+     * `getTextPositions` read one: a dynamic XFA document draws its fields
+     * from a template rather than from page content, so the blocks of the
+     * unflattened file are the blocks of an empty page.
+     */
+    extractMarkdown(first_page?: number | null, page_count?: number | null): string;
     /**
      * Extract the listed pages (0-based) into a new document.
      * Used to implement "split" workflows in the browser.
@@ -214,6 +332,22 @@ export class PdfDoc {
      * is permanently removed from every page.
      */
     redactSearch(query: string): Uint8Array;
+    /**
+     * Replace this document's content with new bytes in-place.
+     *
+     * Call this after [`PdfDocMut.save`] to update the render handle with
+     * the mutated PDF bytes so that subsequent [`renderPage`] calls reflect
+     * the changes, without needing to create a new `PdfDoc` object.
+     *
+     * ```js
+     * const handle = PdfDocMut.open(bytes);
+     * handle.replaceTextSpan(1, "old", "new");
+     * const newBytes = handle.save();
+     * doc.refresh(newBytes);          // engine now renders the edited content
+     * doc.renderPage(0, 1.5);         // ← shows "new" instead of "old"
+     * ```
+     */
+    refresh(data: Uint8Array): void;
     /**
      * Render a page to RGBA pixels.
      *
@@ -266,11 +400,18 @@ export class PdfDoc {
      */
     rotatePage(page_index: number, degrees: number): Uint8Array;
     /**
-     * Set a single AcroForm text-field value. Returns the new PDF bytes.
+     * Set a single AcroForm **text** field value. Returns the new PDF bytes.
+     *
+     * Updates `/V` (ASCII literal or UTF-16BE+BOM), per-widget `/AS`, and
+     * regenerates the `/AP` appearance stream so the fill is visible without
+     * `/NeedAppearances` processing. Read-only fields throw.
+     *
+     * **Text fields only.** For checkbox, radio, and choice fields use
+     * `PdfDocMut.setFormField`, which runs the full type-aware dispatch.
      */
     setFormField(path: string, value: string): Uint8Array;
     /**
-     * Bulk-set multiple AcroForm text fields from a JSON object
+     * Bulk-set multiple AcroForm **text** fields from a JSON object
      * `{"field.path": "value", ...}`. Returns the new PDF bytes.
      */
     setFormFields(fields_json: string): Uint8Array;
@@ -285,6 +426,25 @@ export class PdfDoc {
      * extraction fails.
      */
     text(page_index: number): string;
+    /**
+     * Convert this PDF to a Word `.docx` package.
+     *
+     * Returns the package as a `Uint8Array`.
+     *
+     * The browser SDK could not do this until 23-08-2026: the conversion
+     * crates were excluded from wasm32 by a comment that applied to the
+     * renderer, not to them.
+     */
+    toDocx(): Uint8Array;
+    /**
+     * Convert this PDF to a PowerPoint `.pptx` deck, one slide per page.
+     * See [`Self::to_docx`].
+     */
+    toPptx(): Uint8Array;
+    /**
+     * Convert this PDF to an Excel `.xlsx` workbook. See [`Self::to_docx`].
+     */
+    toXlsx(): Uint8Array;
     /**
      * Validate against a PDF/A level. Returns compliance report as JSON.
      */
@@ -449,26 +609,101 @@ export class PdfDocMut {
      */
     save(): Uint8Array;
     /**
-     * Set a single AcroForm text-field value.
+     * Set a single AcroForm field value (text, checkbox, radio, or choice).
      *
-     * Note: pdfluent's high-level `PdfFormMut` requires a
-     * `pdfluent::PdfDocument`. To stay in-place on this handle, we use
-     * the lower-level lopdf walk via the `pdf-forms` engine. For 1.0,
-     * this supports flat top-level field hierarchies (no `/Kids`
-     * recursion) — same constraint as the Rust core.
+     * Delegates to [`pdf_forms::apply_field_value`] — the single SDK
+     * writeback chain. It keeps `/V` (ASCII literal else UTF-16BE+BOM),
+     * per-widget `/AS`, and regenerated `/AP` streams consistent, resolves
+     * fully-qualified names through `/Kids`, and rejects read-only fields.
      */
     setFormField(path: string, value: string): void;
     /**
-     * Bulk-set multiple AcroForm text fields from a JSON object
+     * Bulk-set multiple AcroForm fields from a JSON object
      * `{"field.path": "value", ...}`.
      */
     setFormFields(fields_json: string): void;
+    /**
+     * Select multiple options on a multi-select list box.
+     *
+     * Delegates to [`pdf_forms::apply_choice_multi`]: writes `/V` as an array
+     * of text strings and rebuilds `/I` (the sorted selected-index cache) to
+     * match what Adobe Acrobat produces. `values` is a JS array of strings;
+     * pass an empty array to clear the selection. The field must be a
+     * multi-select list box (`/Ff` MultiSelect flag); for non-editable list
+     * boxes every value must be one of the field's `/Opt` options.
+     */
+    setMultiSelect(path: string, values: string[]): void;
     /**
      * Apply bold/italic style to a text run on `page_num` (1-based).
      */
     setTextRunStyle(page_num: number, runs_json: string, run_index: number, bold?: boolean | null, italic?: boolean | null): string;
 }
 
+/**
+ * Find-and-replace editor over a PDF document (SDK surface).
+ */
+export class TextEditor {
+    private constructor();
+    free(): void;
+    [Symbol.dispose](): void;
+    /**
+     * Find text occurrences. `query_json` is a JSON object:
+     *
+     * ```json
+     * {
+     *   "text": "Acme B.V.",          // required
+     *   "caseInsensitive": false,      // optional
+     *   "pages": [1, 5],               // optional inclusive 1-based range
+     *   "region": {                    // optional
+     *     "page": 1,
+     *     "rect": [x0, y0, x1, y1],
+     *     "relation": "intersects"     // or "contained"
+     *   },
+     *   "limit": 100                   // optional
+     * }
+     * ```
+     *
+     * Returns a JSON array of matches; each match carries an opaque `id`
+     * token valid until the next applied edit on this document.
+     */
+    findText(query_json: string): string;
+    /**
+     * Open a PDF from raw bytes.
+     */
+    static open(data: Uint8Array): TextEditor;
+    /**
+     * Number of pages.
+     */
+    pageCount(): number;
+    /**
+     * Apply per-match replacements located earlier with `findText`.
+     *
+     * `edits_json` is a JSON array of `{"id": "<match token>", "text":
+     * "<replacement>"}`. All edits commit in one atomic transaction under
+     * `options_json` (see [`TextEditor::replace_text`]); the default is
+     * all-or-nothing. Returns the JSON replacement report.
+     */
+    replaceMatches(edits_json: string, options_json: string): string;
+    /**
+     * Find and replace in one call. Returns the JSON replacement report.
+     *
+     * `options_json` may be empty/`"{}"` for defaults, or:
+     *
+     * ```json
+     * {
+     *   "fontFallback": "deny" | "injectStandard" | {"explicit": "F1"},
+     *   "signaturePolicy": "reject" | "allowPostSignatureChange",
+     *   "commitPolicy": "allOrNothing" | "bestEffort"
+     * }
+     * ```
+     */
+    replaceText(query_json: string, replacement: string, options_json: string): string;
+    /**
+     * Serialize the (possibly edited) document to PDF bytes.
+     */
+    save(): Uint8Array;
+}
+
 /**
  * The main XFA processing engine for WASM.
  *
@@ -566,39 +801,74 @@ export class XfaEngine {
 }
 
 /**
- * Activate the process-global license from a key string.
+ * Remove the password from a PDF whose password you know, and return the
+ * unprotected bytes; open them with `PdfDoc.open`.
+ *
+ * The owner password always works. The user password works only when the
+ * file has no permission restrictions. Throws `PdfluentError` with
+ * `E-SECURITY-DECRYPTION-FAILED` for a password that does not entitle the
+ * caller, and `E-WASM-INVALID-ARGUMENT` for a PDF without a password.
+ */
+export function decryptPdf(bytes: Uint8Array, password: string): Uint8Array;
+
+/**
+ * Encrypt a PDF with AES-256 and return the encrypted bytes.
  *
- * Throws on invalid key or if the process has already been activated to a
- * different tier this session.
+ * `userPassword` is needed to open the file (may be empty: then anyone can
+ * open it, and the permissions apply); `ownerPassword` lifts the permissions
+ * (empty: the user password serves as both). Refuses a PDF that is already
+ * encrypted. Throws `PdfluentError` with `E-WASM-INVALID-ARGUMENT` or
+ * `E-PARSE-INVALID-PDF`.
  */
-export function activateLicenseKey(key: string): void;
+export function encryptPdf(bytes: Uint8Array, user_password: string, owner_password: string, permissions?: EncryptPermissions | null): Uint8Array;
 
 /**
- * Return the current license status. Always succeeds; defaults to Trial.
+ * Extract the images of a PDF as PNG or JPEG files.
+ *
+ * Returns `{ images, skipped }`. An image that could not be exported exactly
+ * as it looks -- JPEG 2000, JBIG2 or fax compression, a transparency mask, a
+ * Decode array, CMYK, an ICC profile other than sRGB -- is listed in `skipped`
+ * with the reason, never emitted wrong. Each image is listed once, on the
+ * first page (0-based) that uses it.
+ */
+export function extractImages(bytes: Uint8Array): ImageExtraction;
+
+/**
+ * Sign a PDF with your own certificate (a .p12/.pfx file and its password)
+ * and return the signed bytes. The private key never leaves the page.
+ *
+ * This is a cryptographic PAdES signature (ETSI.CAdES.detached), appended as
+ * an incremental update so earlier signatures stay valid. Without a
+ * timestamp authority: no TSA or LTV in the browser. A self-signed
+ * certificate verifies, but readers such as Acrobat show its validity as
+ * unknown because they do not trust the issuer. Throws `PdfluentError` with
+ * `E-SECURITY-DECRYPTION-FAILED` for a wrong certificate password.
  */
-export function licenseStatus(): LicenseStatus;
+export function signPdf(bytes: Uint8Array, pkcs12: Uint8Array, password: string, options?: SignPdfOptions | null): Uint8Array;
 
 export type InitInput = RequestInfo | URL | Response | BufferSource | WebAssembly.Module;
 
 export interface InitOutput {
     readonly memory: WebAssembly.Memory;
-    readonly __wbg_licensestatus_free: (a: number, b: number) => void;
     readonly __wbg_pdfdoc_free: (a: number, b: number) => void;
     readonly __wbg_pdfdocmut_free: (a: number, b: number) => void;
+    readonly __wbg_texteditor_free: (a: number, b: number) => void;
     readonly __wbg_xfaengine_free: (a: number, b: number) => void;
-    readonly activateLicenseKey: (a: number, b: number, c: number) => void;
-    readonly licenseStatus: () => number;
-    readonly licensestatus_outputIsMarked: (a: number) => number;
-    readonly licensestatus_source: (a: number, b: number) => void;
-    readonly licensestatus_tier: (a: number, b: number) => void;
+    readonly decryptPdf: (a: number, b: number, c: number, d: number, e: number) => void;
+    readonly encryptPdf: (a: number, b: number, c: number, d: number, e: number, f: number, g: number, h: number) => void;
+    readonly extractImages: (a: number, b: number, c: number) => void;
     readonly pdfdoc_addFreeText: (a: number, b: number, c: number, d: number, e: number, f: number, g: number, h: number, i: number) => void;
     readonly pdfdoc_addHighlight: (a: number, b: number, c: number, d: number, e: number, f: number, g: number, h: number, i: number) => void;
     readonly pdfdoc_addStickyNote: (a: number, b: number, c: number, d: number, e: number, f: number, g: number) => void;
     readonly pdfdoc_addTextWatermark: (a: number, b: number, c: number, d: number, e: number) => void;
+    readonly pdfdoc_addWatermark: (a: number, b: number, c: number) => void;
     readonly pdfdoc_compress: (a: number, b: number) => void;
     readonly pdfdoc_convertToPdfa: (a: number, b: number, c: number, d: number) => void;
     readonly pdfdoc_deletePages: (a: number, b: number, c: number, d: number) => void;
     readonly pdfdoc_dssInfo: (a: number, b: number) => void;
+    readonly pdfdoc_extractBlocks: (a: number, b: number, c: number, d: number) => void;
+    readonly pdfdoc_extractDocling: (a: number, b: number, c: number, d: number, e: number, f: number) => void;
+    readonly pdfdoc_extractMarkdown: (a: number, b: number, c: number, d: number) => void;
     readonly pdfdoc_extractPages: (a: number, b: number, c: number, d: number) => void;
     readonly pdfdoc_flattenXfa: (a: number, b: number) => void;
     readonly pdfdoc_getAnnotations: (a: number, b: number, c: number) => void;
@@ -612,6 +882,7 @@ export interface InitOutput {
     readonly pdfdoc_pageWidth: (a: number, b: number) => number;
     readonly pdfdoc_redactRegion: (a: number, b: number, c: number, d: number, e: number, f: number, g: number) => void;
     readonly pdfdoc_redactSearch: (a: number, b: number, c: number, d: number) => void;
+    readonly pdfdoc_refresh: (a: number, b: number, c: number, d: number) => void;
     readonly pdfdoc_renderPage: (a: number, b: number, c: number, d: number) => void;
     readonly pdfdoc_renderPageToCanvas: (a: number, b: number, c: number, d: number, e: number) => void;
     readonly pdfdoc_renderPageToCanvasVector: (a: number, b: number, c: number, d: number, e: number) => void;
@@ -622,6 +893,9 @@ export interface InitOutput {
     readonly pdfdoc_setFormFields: (a: number, b: number, c: number, d: number) => void;
     readonly pdfdoc_signatures: (a: number, b: number) => void;
     readonly pdfdoc_text: (a: number, b: number, c: number) => void;
+    readonly pdfdoc_toDocx: (a: number, b: number) => void;
+    readonly pdfdoc_toPptx: (a: number, b: number) => void;
+    readonly pdfdoc_toXlsx: (a: number, b: number) => void;
     readonly pdfdoc_validatePdfA: (a: number, b: number, c: number, d: number) => void;
     readonly pdfdoc_verifySignatures: (a: number, b: number) => void;
     readonly pdfdocmut_addFreeText: (a: number, b: number, c: number, d: number, e: number, f: number, g: number, h: number, i: number) => void;
@@ -644,7 +918,15 @@ export interface InitOutput {
     readonly pdfdocmut_save: (a: number, b: number) => void;
     readonly pdfdocmut_setFormField: (a: number, b: number, c: number, d: number, e: number, f: number) => void;
     readonly pdfdocmut_setFormFields: (a: number, b: number, c: number, d: number) => void;
+    readonly pdfdocmut_setMultiSelect: (a: number, b: number, c: number, d: number, e: number, f: number) => void;
     readonly pdfdocmut_setTextRunStyle: (a: number, b: number, c: number, d: number, e: number, f: number, g: number, h: number) => void;
+    readonly signPdf: (a: number, b: number, c: number, d: number, e: number, f: number, g: number, h: number) => void;
+    readonly texteditor_findText: (a: number, b: number, c: number, d: number) => void;
+    readonly texteditor_open: (a: number, b: number, c: number) => void;
+    readonly texteditor_pageCount: (a: number) => number;
+    readonly texteditor_replaceMatches: (a: number, b: number, c: number, d: number, e: number, f: number) => void;
+    readonly texteditor_replaceText: (a: number, b: number, c: number, d: number, e: number, f: number, g: number, h: number) => void;
+    readonly texteditor_save: (a: number, b: number) => void;
     readonly xfaengine_exportJson: (a: number, b: number) => void;
     readonly xfaengine_exportSchema: (a: number, b: number) => void;
     readonly xfaengine_fromFields: (a: number, b: number, c: number) => void;