@cj-tech-master/excelts 9.5.4 → 9.5.5

package/dist/browser/modules/word/reader/comments-parser.d.ts

@@ -0,0 +1,26 @@
+/**
+ * DOCX reader — comments parsers.
+ *
+ * Split out from `docx-reader.ts` to keep that file's part-orchestration
+ * scope manageable. We accept `parseParagraph` as an injected dependency
+ * because the body paragraph parser still lives in `docx-reader.ts` (it
+ * has many cross-references that would not be cheap to disentangle); a
+ * direct import would form a cycle.
+ */
+import type { XmlElement } from "../../xml/types.js";
+import type { CommentDef, Paragraph } from "../types.js";
+import { type ReaderContext } from "./reader-context.js";
+/** Parse `word/comments.xml` into a list of CommentDef. */
+export declare function parseCommentsXml(xmlStr: string, ctx: ReaderContext, parseParagraph: (el: XmlElement, ctx: ReaderContext) => Paragraph): CommentDef[];
+/**
+ * Parse `word/commentsExtended.xml` into a paraId → metadata map.
+ *
+ * Modern Word stores comment "resolved" state and reply threading in a
+ * sidecar part keyed by the paragraph id of each comment's first
+ * paragraph. The reader merges this map into `CommentDef.done` /
+ * `CommentDef.parentId` once both parts have been parsed.
+ */
+export declare function parseCommentsExtendedXml(xmlStr: string): Map<string, {
+    done?: boolean;
+    parentId?: string;
+}>;

package/dist/browser/modules/word/reader/comments-parser.js

@@ -0,0 +1,92 @@
+/**
+ * DOCX reader — comments parsers.
+ *
+ * Split out from `docx-reader.ts` to keep that file's part-orchestration
+ * scope manageable. We accept `parseParagraph` as an injected dependency
+ * because the body paragraph parser still lives in `docx-reader.ts` (it
+ * has many cross-references that would not be cheap to disentangle); a
+ * direct import would form a cycle.
+ */
+import { parseXml } from "../../xml/dom.js";
+import { attrInt, attrVal, findChildrenNs } from "./parse-utils.js";
+import { createFieldState } from "./reader-context.js";
+/** Parse `word/comments.xml` into a list of CommentDef. */
+export function parseCommentsXml(xmlStr, ctx, parseParagraph) {
+    // Comments are a separate part — reset the field state so a body-level
+    // unterminated field does not bleed into comment parsing.
+    const savedField = ctx.field;
+    ctx.field = createFieldState();
+    try {
+        const doc = parseXml(xmlStr);
+        const root = doc.root;
+        const comments = [];
+        for (const commentEl of findChildrenNs(root, "comment")) {
+            const id = attrInt(commentEl, "id");
+            const author = attrVal(commentEl, "author");
+            if (id === undefined || !author) {
+                continue;
+            }
+            const content = [];
+            for (const child of commentEl.children) {
+                if (child.type === "element" && child.name.replace(/^w:/, "") === "p") {
+                    content.push(parseParagraph(child, ctx));
+                }
+            }
+            const comment = { id, author, content };
+            const date = attrVal(commentEl, "date");
+            if (date) {
+                comment.date = date;
+            }
+            const initials = attrVal(commentEl, "initials");
+            if (initials) {
+                comment.initials = initials;
+            }
+            comments.push(comment);
+        }
+        return comments;
+    }
+    finally {
+        ctx.field = savedField;
+    }
+}
+/**
+ * Parse `word/commentsExtended.xml` into a paraId → metadata map.
+ *
+ * Modern Word stores comment "resolved" state and reply threading in a
+ * sidecar part keyed by the paragraph id of each comment's first
+ * paragraph. The reader merges this map into `CommentDef.done` /
+ * `CommentDef.parentId` once both parts have been parsed.
+ */
+export function parseCommentsExtendedXml(xmlStr) {
+    const map = new Map();
+    const doc = parseXml(xmlStr);
+    const root = doc.root;
+    for (const child of root.children) {
+        if (child.type !== "element") {
+            continue;
+        }
+        // w15:commentEx
+        const name = child.name;
+        if (!name.endsWith("commentEx")) {
+            continue;
+        }
+        const paraId = child.attributes["w15:paraId"] ?? child.attributes["paraId"];
+        if (!paraId) {
+            continue;
+        }
+        const entry = {};
+        const done = child.attributes["w15:done"] ?? child.attributes["done"];
+        if (done === "1" || done === "true") {
+            entry.done = true;
+        }
+        else if (done === "0" || done === "false") {
+            entry.done = false;
+        }
+        const pid = child.attributes["w15:paraIdParent"] ?? child.attributes["paraIdParent"];
+        if (pid) {
+            entry.parentId = pid;
+        }
+        map.set(paraId, entry);
+    }
+    return map;
+}

package/dist/browser/modules/word/reader/doc-props-parsers.d.ts

@@ -0,0 +1,15 @@
+/**
+ * DOCX Reader - Document Properties Parsers
+ *
+ * Parses standard OPC document properties XML files:
+ * - docProps/core.xml (CoreProperties — title, author, dc:* fields)
+ * - docProps/app.xml (AppProperties — application info)
+ * - docProps/custom.xml (CustomProperty[] — user-defined name/value pairs)
+ * - word/fontTable.xml (FontDef[] — embedded font references)
+ */
+import type { AppProperties, CoreProperties, CustomProperty, FontDef } from "../types.js";
+declare function parseCoreProps(xmlStr: string): CoreProperties;
+declare function parseAppProps(xmlStr: string): AppProperties;
+declare function parseCustomPropsXml(xmlStr: string): CustomProperty[];
+declare function parseFontTableXml(xmlStr: string): FontDef[];
+export { parseCoreProps, parseAppProps, parseCustomPropsXml, parseFontTableXml };

package/dist/browser/modules/word/reader/doc-props-parsers.js

@@ -0,0 +1,190 @@
+/**
+ * DOCX Reader - Document Properties Parsers
+ *
+ * Parses standard OPC document properties XML files:
+ * - docProps/core.xml (CoreProperties — title, author, dc:* fields)
+ * - docProps/app.xml (AppProperties — application info)
+ * - docProps/custom.xml (CustomProperty[] — user-defined name/value pairs)
+ * - word/fontTable.xml (FontDef[] — embedded font references)
+ */
+import { findChild, parseXml, textContent } from "../../xml/dom.js";
+import {} from "../core/internal-utils.js";
+import { attrVal, findChildNs, findChildrenNs } from "./parse-utils.js";
+// =============================================================================
+// Core Properties Parser
+// =============================================================================
+function parseCoreProps(xmlStr) {
+    const doc = parseXml(xmlStr);
+    const root = doc.root;
+    const props = {};
+    const fields = [
+        ["dc:title", "title"],
+        ["dc:subject", "subject"],
+        ["dc:creator", "creator"],
+        ["dc:description", "description"],
+        ["cp:keywords", "keywords"],
+        ["cp:lastModifiedBy", "lastModifiedBy"],
+        ["cp:revision", "revision"],
+        ["cp:category", "category"]
+    ];
+    for (const [tag, prop] of fields) {
+        const el = findChild(root, tag);
+        if (el) {
+            const val = textContent(el);
+            if (val) {
+                props[prop] = val;
+            }
+        }
+    }
+    const createdEl = findChild(root, "dcterms:created");
+    if (createdEl) {
+        const val = textContent(createdEl);
+        if (val) {
+            props.created = new Date(val);
+        }
+    }
+    const modifiedEl = findChild(root, "dcterms:modified");
+    if (modifiedEl) {
+        const val = textContent(modifiedEl);
+        if (val) {
+            props.modified = new Date(val);
+        }
+    }
+    return props;
+}
+// =============================================================================
+// App Properties Parser
+// =============================================================================
+function parseAppProps(xmlStr) {
+    const doc = parseXml(xmlStr);
+    const root = doc.root;
+    const props = {};
+    const strFields = ["Application", "AppVersion", "Company", "Manager"];
+    const intFields = ["Pages", "Words", "Characters", "Lines", "Paragraphs"];
+    for (const field of strFields) {
+        const el = findChild(root, field);
+        if (el) {
+            const val = textContent(el);
+            if (val) {
+                props[field.charAt(0).toLowerCase() + field.slice(1)] = val;
+            }
+        }
+    }
+    for (const field of intFields) {
+        const el = findChild(root, field);
+        if (el) {
+            const val = textContent(el);
+            if (val) {
+                props[field.charAt(0).toLowerCase() + field.slice(1)] = parseInt(val, 10);
+            }
+        }
+    }
+    return props;
+}
+// =============================================================================
+// Custom Properties Parser
+// =============================================================================
+function parseCustomPropsXml(xmlStr) {
+    const doc = parseXml(xmlStr);
+    const root = doc.root;
+    const props = [];
+    for (const propEl of root.children) {
+        if (propEl.type !== "element" || propEl.name !== "property") {
+            continue;
+        }
+        const name = propEl.attributes["name"];
+        if (!name) {
+            continue;
+        }
+        let value;
+        for (const child of propEl.children) {
+            if (child.type !== "element") {
+                continue;
+            }
+            const tn = child.name;
+            const tv = textContent(child);
+            if (tn === "vt:lpwstr") {
+                value = { type: "string", value: tv };
+            }
+            else if (tn === "vt:i4") {
+                value = { type: "number", value: parseInt(tv, 10) };
+            }
+            else if (tn === "vt:r8") {
+                value = { type: "number", value: parseFloat(tv) };
+            }
+            else if (tn === "vt:bool") {
+                value = { type: "boolean", value: tv === "true" };
+            }
+            else if (tn === "vt:filetime") {
+                value = { type: "date", value: new Date(tv) };
+            }
+        }
+        if (value) {
+            props.push({ name, value });
+        }
+    }
+    return props;
+}
+// =============================================================================
+// Font Table Parser
+// =============================================================================
+function parseFontTableXml(xmlStr) {
+    const doc = parseXml(xmlStr);
+    const root = doc.root;
+    const fonts = [];
+    for (const fontEl of findChildrenNs(root, "font")) {
+        const f = { name: attrVal(fontEl, "name") ?? "" };
+        const p1 = findChildNs(fontEl, "panose1");
+        if (p1) {
+            f.panose1 = attrVal(p1, "val");
+        }
+        const cs = findChildNs(fontEl, "charset");
+        if (cs) {
+            f.charset = attrVal(cs, "val");
+        }
+        const fam = findChildNs(fontEl, "family");
+        if (fam) {
+            f.family = attrVal(fam, "val");
+        }
+        const pitch = findChildNs(fontEl, "pitch");
+        if (pitch) {
+            f.pitch = attrVal(pitch, "val");
+        }
+        // Signature
+        const sigEl = findChildNs(fontEl, "sig");
+        if (sigEl) {
+            const sig = {};
+            for (const key of ["usb0", "usb1", "usb2", "usb3", "csb0", "csb1"]) {
+                const v = attrVal(sigEl, key);
+                if (v !== undefined) {
+                    sig[key] = v;
+                }
+            }
+            if (Object.keys(sig).length > 0) {
+                f.sig = sig;
+            }
+        }
+        // Embedded fonts
+        for (const [tag, rIdKey, keyKey] of [
+            ["embedRegular", "embedRegular", "embedRegularKey"],
+            ["embedBold", "embedBold", "embedBoldKey"],
+            ["embedItalic", "embedItalic", "embedItalicKey"],
+            ["embedBoldItalic", "embedBoldItalic", "embedBoldItalicKey"]
+        ]) {
+            const el = findChildNs(fontEl, tag);
+            if (el) {
+                const rId = el.attributes["r:id"] ?? el.attributes["id"];
+                if (rId) {
+                    f[rIdKey] = rId;
+                    const fontKey = attrVal(el, "fontKey");
+                    if (fontKey) {
+                        f[keyKey] = fontKey;
+                    }
+                }
+            }
+        }
+        fonts.push(f);
+    }
+    return fonts;
+}
+export { parseCoreProps, parseAppProps, parseCustomPropsXml, parseFontTableXml };

package/dist/browser/modules/word/reader/docx-reader.d.ts

@@ -0,0 +1,27 @@
+/**
+ * DOCX Module - Reader / Parser
+ *
+ * Reads a DOCX ZIP file and parses it into a DocxDocument model.
+ * Uses the archive module for ZIP reading and XML module for parsing.
+ */
+import { type WordSecurityPolicy } from "../security/policy.js";
+import type { DocxDocument } from "../types.js";
+/** Options for reading a DOCX file. */
+export interface ReadDocxOptions {
+    /** Password for decrypting an encrypted DOCX file. */
+    readonly password?: string;
+    /**
+     * Optional security policy that controls hard limits enforced while reading
+     * the package (max total size, max single-part size, max part count). Falls
+     * back to {@link DEFAULT_SECURITY_POLICY} when omitted. Helpful as a
+     * defense-in-depth control against ZIP bombs / pathological inputs.
+     */
+    readonly securityPolicy?: WordSecurityPolicy;
+}
+/**
+ * Read a DOCX file from a Uint8Array buffer and parse it into a DocxDocument model.
+ *
+ * If the file is encrypted (CFB format), provide a password via the options parameter
+ * to decrypt it automatically.
+ */
+export declare function readDocx(buffer: Uint8Array, options?: ReadDocxOptions): Promise<DocxDocument>;