@beyondwork/docx-react-component 1.0.27 → 1.0.29

package/src/io/source-package-provenance.ts

@@ -1,241 +0,0 @@
-import type { PersistedSourcePackage } from "../model/snapshot.ts";
-import { DOCX_MIME_TYPE } from "./opc/docx-package.ts";
-
-const SHA256_INITIAL_STATE = [
-  0x6a09e667,
-  0xbb67ae85,
-  0x3c6ef372,
-  0xa54ff53a,
-  0x510e527f,
-  0x9b05688c,
-  0x1f83d9ab,
-  0x5be0cd19,
-] as const;
-
-const SHA256_CONSTANTS = [
-  0x428a2f98, 0x71374491, 0xb5c0fbcf, 0xe9b5dba5,
-  0x3956c25b, 0x59f111f1, 0x923f82a4, 0xab1c5ed5,
-  0xd807aa98, 0x12835b01, 0x243185be, 0x550c7dc3,
-  0x72be5d74, 0x80deb1fe, 0x9bdc06a7, 0xc19bf174,
-  0xe49b69c1, 0xefbe4786, 0x0fc19dc6, 0x240ca1cc,
-  0x2de92c6f, 0x4a7484aa, 0x5cb0a9dc, 0x76f988da,
-  0x983e5152, 0xa831c66d, 0xb00327c8, 0xbf597fc7,
-  0xc6e00bf3, 0xd5a79147, 0x06ca6351, 0x14292967,
-  0x27b70a85, 0x2e1b2138, 0x4d2c6dfc, 0x53380d13,
-  0x650a7354, 0x766a0abb, 0x81c2c92e, 0x92722c85,
-  0xa2bfe8a1, 0xa81a664b, 0xc24b8b70, 0xc76c51a3,
-  0xd192e819, 0xd6990624, 0xf40e3585, 0x106aa070,
-  0x19a4c116, 0x1e376c08, 0x2748774c, 0x34b0bcb5,
-  0x391c0cb3, 0x4ed8aa4a, 0x5b9cca4f, 0x682e6ff3,
-  0x748f82ee, 0x78a5636f, 0x84c87814, 0x8cc70208,
-  0x90befffa, 0xa4506ceb, 0xbef9a3f7, 0xc67178f2,
-] as const;
-
-export function createPersistedSourcePackage(
-  bytes: Uint8Array | ArrayBuffer,
-  sourceLabel?: string,
-): PersistedSourcePackage {
-  const normalizedBytes = toUint8Array(bytes);
-
-  return {
-    format: "docx",
-    storage: "embedded-base64",
-    mimeType: DOCX_MIME_TYPE,
-    sourceLabel:
-      typeof sourceLabel === "string" && sourceLabel.trim().length > 0
-        ? sourceLabel
-        : undefined,
-    sha256Hex: sha256Hex(normalizedBytes),
-    bytesBase64: encodeBytesBase64(normalizedBytes),
-  };
-}
-
-export function decodePersistedSourcePackageBytes(
-  sourcePackage: PersistedSourcePackage,
-): Uint8Array {
-  return decodeBytesBase64(sourcePackage.bytesBase64);
-}
-
-export function hasValidPersistedSourcePackageDigest(
-  sourcePackage: PersistedSourcePackage,
-  bytes: Uint8Array,
-): boolean {
-  return sha256Hex(bytes) === sourcePackage.sha256Hex;
-}
-
-export function encodeBytesBase64(bytes: Uint8Array): string {
-  const bufferConstructor = getBufferConstructor();
-  if (bufferConstructor) {
-    return bufferConstructor.from(bytes).toString("base64");
-  }
-
-  let binary = "";
-  for (let offset = 0; offset < bytes.byteLength; offset += 0x8000) {
-    const chunk = bytes.subarray(offset, offset + 0x8000);
-    for (let index = 0; index < chunk.length; index += 1) {
-      binary += String.fromCharCode(chunk[index] ?? 0);
-    }
-  }
-
-  if (typeof btoa !== "function") {
-    throw new Error("Base64 encoding is unavailable in this runtime.");
-  }
-
-  return btoa(binary);
-}
-
-export function decodeBytesBase64(value: string): Uint8Array {
-  const bufferConstructor = getBufferConstructor();
-  if (bufferConstructor) {
-    return new Uint8Array(bufferConstructor.from(value, "base64"));
-  }
-
-  if (typeof atob !== "function") {
-    throw new Error("Base64 decoding is unavailable in this runtime.");
-  }
-
-  const binary = atob(value);
-  const bytes = new Uint8Array(binary.length);
-  for (let index = 0; index < binary.length; index += 1) {
-    bytes[index] = binary.charCodeAt(index);
-  }
-  return bytes;
-}
-
-export function sha256Hex(bytes: Uint8Array): string {
-  const padded = padSha256Message(bytes);
-  const state = Array.from(SHA256_INITIAL_STATE, (value) => value >>> 0);
-  const schedule = new Uint32Array(64);
-
-  for (let chunkOffset = 0; chunkOffset < padded.byteLength; chunkOffset += 64) {
-    for (let index = 0; index < 16; index += 1) {
-      const wordOffset = chunkOffset + index * 4;
-      schedule[index] =
-        (((padded[wordOffset] ?? 0) << 24) |
-          ((padded[wordOffset + 1] ?? 0) << 16) |
-          ((padded[wordOffset + 2] ?? 0) << 8) |
-          (padded[wordOffset + 3] ?? 0)) >>>
-        0;
-    }
-
-    for (let index = 16; index < 64; index += 1) {
-      schedule[index] = add32(
-        lowerSigma1(schedule[index - 2] ?? 0),
-        schedule[index - 7] ?? 0,
-        lowerSigma0(schedule[index - 15] ?? 0),
-        schedule[index - 16] ?? 0,
-      );
-    }
-
-    let [a, b, c, d, e, f, g, h] = state;
-
-    for (let index = 0; index < 64; index += 1) {
-      const t1 = add32(
-        h,
-        upperSigma1(e),
-        choose(e, f, g),
-        SHA256_CONSTANTS[index] ?? 0,
-        schedule[index] ?? 0,
-      );
-      const t2 = add32(upperSigma0(a), majority(a, b, c));
-
-      h = g;
-      g = f;
-      f = e;
-      e = add32(d, t1);
-      d = c;
-      c = b;
-      b = a;
-      a = add32(t1, t2);
-    }
-
-    state[0] = add32(state[0] ?? 0, a);
-    state[1] = add32(state[1] ?? 0, b);
-    state[2] = add32(state[2] ?? 0, c);
-    state[3] = add32(state[3] ?? 0, d);
-    state[4] = add32(state[4] ?? 0, e);
-    state[5] = add32(state[5] ?? 0, f);
-    state[6] = add32(state[6] ?? 0, g);
-    state[7] = add32(state[7] ?? 0, h);
-  }
-
-  return state.map((word) => word.toString(16).padStart(8, "0")).join("");
-}
-
-function padSha256Message(bytes: Uint8Array): Uint8Array {
-  const bitLength = bytes.byteLength * 8;
-  const paddedLength = Math.ceil((bytes.byteLength + 9) / 64) * 64;
-  const padded = new Uint8Array(paddedLength);
-  padded.set(bytes);
-  padded[bytes.byteLength] = 0x80;
-
-  const lengthOffset = paddedLength - 8;
-  for (let index = 0; index < 8; index += 1) {
-    const shift = (7 - index) * 8;
-    padded[lengthOffset + index] = Math.floor(bitLength / 2 ** shift) & 0xff;
-  }
-
-  return padded;
-}
-
-function rotateRight(value: number, shift: number): number {
-  return ((value >>> shift) | (value << (32 - shift))) >>> 0;
-}
-
-function add32(...values: number[]): number {
-  let total = 0;
-  for (const value of values) {
-    total = (total + value) >>> 0;
-  }
-  return total;
-}
-
-function choose(x: number, y: number, z: number): number {
-  return ((x & y) ^ (~x & z)) >>> 0;
-}
-
-function majority(x: number, y: number, z: number): number {
-  return ((x & y) ^ (x & z) ^ (y & z)) >>> 0;
-}
-
-function upperSigma0(value: number): number {
-  return (rotateRight(value, 2) ^ rotateRight(value, 13) ^ rotateRight(value, 22)) >>> 0;
-}
-
-function upperSigma1(value: number): number {
-  return (rotateRight(value, 6) ^ rotateRight(value, 11) ^ rotateRight(value, 25)) >>> 0;
-}
-
-function lowerSigma0(value: number): number {
-  return (rotateRight(value, 7) ^ rotateRight(value, 18) ^ (value >>> 3)) >>> 0;
-}
-
-function lowerSigma1(value: number): number {
-  return (rotateRight(value, 17) ^ rotateRight(value, 19) ^ (value >>> 10)) >>> 0;
-}
-
-function toUint8Array(bytes: Uint8Array | ArrayBuffer): Uint8Array {
-  return bytes instanceof Uint8Array ? new Uint8Array(bytes) : new Uint8Array(bytes);
-}
-
-function getBufferConstructor():
-  | {
-      from(
-        value: Uint8Array | string,
-        byteOffsetOrEncoding?: number | string,
-        length?: number,
-      ): Uint8Array & { toString(encoding: string): string };
-    }
-  | undefined {
-  const maybeBuffer = (globalThis as {
-    Buffer?: {
-      from(
-        value: Uint8Array | string,
-        byteOffsetOrEncoding?: number | string,
-        length?: number,
-      ): Uint8Array & { toString(encoding: string): string };
-    };
-  }).Buffer;
-
-  return maybeBuffer;
-}

package/src/legal/cross-references.ts

@@ -1,414 +0,0 @@
-import {
-  extractComplexFieldsFromBodyXml,
-  parseFieldsFromParagraphXml,
-} from "../io/ooxml/parse-fields.ts";
-import { parseMainDocumentXml } from "../io/ooxml/parse-main-document.ts";
-import type {
-  CanonicalDocument,
-  DocumentNode,
-  FieldNode,
-  FieldRegistry,
-  HyperlinkNode,
-  ParagraphNode,
-  TocEntry,
-} from "../model/canonical-document.ts";
-
-export interface CrossReferencePattern {
-  kind: "section" | "clause" | "article" | "schedule" | "exhibit" | "appendix";
-  label: string;
-  targetHint: string;
-}
-
-export interface CrossReference {
-  source: "hyperlink" | "field" | "w:ref" | "text-pattern";
-  kind: "bookmark" | "section" | "clause" | "article" | "schedule" | "exhibit" | "appendix";
-  target?: string;
-  instruction?: string;
-  label: string;
-  paragraphIndex?: number;
-  confidence: "high" | "medium";
-}
-
-interface FieldReference {
-  kind: "REF" | "PAGEREF" | "NOTEREF";
-  target: string;
-  instruction: string;
-}
-
-/**
- * Collect field references from a canonical document, leveraging the
- * field family classification when available.
- */
-export function collectFieldReferencesFromCanonicalDocument(
-  document: Pick<CanonicalDocument, "content"> | DocumentNode,
-): Array<{ family: string; target: string; instruction: string; paragraphIndex: number; displayText: string }> {
-  const root = "content" in document ? document.content : document;
-  const results: Array<{ family: string; target: string; instruction: string; paragraphIndex: number; displayText: string }> = [];
-  let paragraphIndex = -1;
-
-  walkDocument(root, (node) => {
-    if (node.type === "paragraph") {
-      paragraphIndex += 1;
-      for (const child of node.children) {
-        if (child.type === "field" && child.fieldFamily && child.fieldTarget) {
-          const family = child.fieldFamily;
-          if (family === "REF" || family === "PAGEREF" || family === "NOTEREF") {
-            results.push({
-              family,
-              target: child.fieldTarget,
-              instruction: child.instruction,
-              paragraphIndex,
-              displayText: flattenInlineText(child.children).trim(),
-            });
-          }
-        }
-      }
-    }
-  });
-
-  return results;
-}
-
-const CROSS_REFERENCE_PATTERN =
-  /\b(Section|Clause|Article|Schedule|Exhibit|Appendix)\s+(\d+(?:\.\d+)*|[A-Z]-\d+|[A-Z])(?=[^A-Za-z0-9]|$)/g;
-const W_REF_PATTERN = /<w:ref\b([^>]*)\/?>/g;
-const ATTRIBUTE_PATTERN = /(?:^|\s)(?:w:)?([A-Za-z][\w-]*)="([^"]*)"/g;
-const COMPLEX_FIELD_PATTERN =
-  /<w:instrText\b[^>]*>([\s\S]*?)<\/w:instrText>[\s\S]*?<w:fldChar\b[^>]*w:fldCharType="separate"[^>]*\/>[\s\S]*?<w:t\b[^>]*>([\s\S]*?)<\/w:t>[\s\S]*?<w:fldChar\b[^>]*w:fldCharType="end"[^>]*\/>/g;
-
-export function parseCrossReferencesFromDocumentXml(xml: string): CrossReference[] {
-  const parsed = parseMainDocumentXml(xml);
-  const results: CrossReference[] = [];
-
-  parsed.blocks.forEach((block, paragraphIndex) => {
-    if (block.type !== "paragraph") {
-      return;
-    }
-
-    for (const child of block.children) {
-      if (child.type === "hyperlink" && child.href.startsWith("#")) {
-        results.push({
-          source: "hyperlink",
-          kind: "bookmark",
-          target: child.href.slice(1),
-          label: flattenInlineText(child.children).trim() || child.href.slice(1),
-          paragraphIndex,
-          confidence: "high",
-        });
-      }
-    }
-
-    const { simpleFields } = parseFieldsFromParagraphXml(block.rawXml);
-    for (const field of simpleFields) {
-      const reference = parseFieldReferenceInstruction(field.instruction);
-      if (!reference) {
-        continue;
-      }
-
-      results.push({
-        source: "field",
-        kind: "bookmark",
-        target: reference.target,
-        instruction: reference.instruction,
-        label: stripXml(field.contentXml) || reference.target,
-        paragraphIndex,
-        confidence: "high",
-      });
-    }
-
-    for (const pattern of detectCrossReferencePatterns(flattenParagraphText(block as unknown as ParagraphNode))) {
-      results.push({
-        source: "text-pattern",
-        kind: pattern.kind,
-        target: pattern.targetHint,
-        label: pattern.label,
-        paragraphIndex,
-        confidence: "medium",
-      });
-    }
-  });
-
-  for (const field of extractComplexFieldsFromBodyXml(xml)) {
-    const reference = parseFieldReferenceInstruction(field.instruction);
-    if (!reference) {
-      continue;
-    }
-    results.push({
-      source: "field",
-      kind: "bookmark",
-      target: reference.target,
-      instruction: reference.instruction,
-      label: stripXml(field.contentXml) || reference.target,
-      confidence: "high",
-    });
-  }
-
-  for (const field of extractComplexFieldsWithRegex(xml)) {
-    const reference = parseFieldReferenceInstruction(field.instruction);
-    if (!reference) {
-      continue;
-    }
-
-    results.push({
-      source: "field",
-      kind: "bookmark",
-      target: reference.target,
-      instruction: reference.instruction,
-      label: field.label || reference.target,
-      confidence: "high",
-    });
-  }
-
-  for (const element of extractWRefElements(xml)) {
-    results.push({
-      source: "w:ref",
-      kind: "bookmark",
-      target: element.target,
-      label: element.label || element.target || "w:ref",
-      confidence: element.target ? "high" : "medium",
-    });
-  }
-
-  return dedupeCrossReferences(results);
-}
-
-export function collectCrossReferencesFromCanonicalDocument(
-  document: Pick<CanonicalDocument, "content"> | DocumentNode,
-): CrossReference[] {
-  const root = "content" in document ? document.content : document;
-  const results: CrossReference[] = [];
-  let paragraphIndex = -1;
-
-  walkDocument(root, (node) => {
-    if (node.type === "paragraph") {
-      paragraphIndex += 1;
-      for (const child of node.children) {
-        if (child.type === "hyperlink" && child.href.startsWith("#")) {
-          results.push({
-            source: "hyperlink",
-            kind: "bookmark",
-            target: child.href.slice(1),
-            label: flattenInlineText(child.children).trim() || child.href.slice(1),
-            paragraphIndex,
-            confidence: "high",
-          });
-          continue;
-        }
-
-        if (child.type === "field") {
-          const reference = parseFieldReferenceInstruction(child.instruction);
-          if (!reference) {
-            continue;
-          }
-          results.push({
-            source: "field",
-            kind: "bookmark",
-            target: reference.target,
-            instruction: reference.instruction,
-            label: flattenInlineText(child.children).trim() || reference.target,
-            paragraphIndex,
-            confidence: "high",
-          });
-        }
-      }
-
-      for (const pattern of detectCrossReferencePatterns(flattenParagraphText(node))) {
-        results.push({
-          source: "text-pattern",
-          kind: pattern.kind,
-          target: pattern.targetHint,
-          label: pattern.label,
-          paragraphIndex,
-          confidence: "medium",
-        });
-      }
-    }
-  });
-
-  return dedupeCrossReferences(results);
-}
-
-export function detectCrossReferencePatterns(text: string): CrossReferencePattern[] {
-  const matches: CrossReferencePattern[] = [];
-
-  for (const match of text.matchAll(CROSS_REFERENCE_PATTERN)) {
-    const referenceType = match[1]?.toLowerCase();
-    const target = match[2];
-    const label = `${match[1]} ${target}`.trim();
-
-    if (!referenceType || !target) {
-      continue;
-    }
-
-    matches.push({
-      kind: referenceType as CrossReferencePattern["kind"],
-      label,
-      targetHint: target,
-    });
-  }
-
-  return matches;
-}
-
-export function parseFieldReferenceInstruction(instruction: string): FieldReference | undefined {
-  const trimmedInstruction = instruction.trim().replace(/\s+/g, " ");
-  const match = /^(REF|PAGEREF|NOTEREF)\s+(?:"([^"]+)"|([^\s\\]+))/i.exec(trimmedInstruction);
-  if (!match) {
-    return undefined;
-  }
-
-  return {
-    kind: match[1].toUpperCase() as FieldReference["kind"],
-    target: (match[2] ?? match[3] ?? "").trim(),
-    instruction: trimmedInstruction,
-  };
-}
-
-function extractWRefElements(xml: string): Array<{ target?: string; label?: string }> {
-  const matches: Array<{ target?: string; label?: string }> = [];
-
-  for (const match of xml.matchAll(W_REF_PATTERN)) {
-    const attributes = readAttributes(match[1] ?? "");
-    matches.push({
-      target: attributes.anchor ?? attributes.name ?? attributes.id,
-      label: attributes.displayText ?? attributes.text,
-    });
-  }
-
-  return matches;
-}
-
-function extractComplexFieldsWithRegex(xml: string): Array<{ instruction: string; label: string }> {
-  const matches: Array<{ instruction: string; label: string }> = [];
-
-  for (const match of xml.matchAll(COMPLEX_FIELD_PATTERN)) {
-    const instruction = stripXml(match[1] ?? "");
-    const label = stripXml(match[2] ?? "");
-    if (!instruction) {
-      continue;
-    }
-    matches.push({ instruction, label });
-  }
-
-  return matches;
-}
-
-function readAttributes(source: string): Record<string, string> {
-  const attributes: Record<string, string> = {};
-
-  for (const match of source.matchAll(ATTRIBUTE_PATTERN)) {
-    const [, key, value] = match;
-    attributes[key] = value;
-  }
-
-  return attributes;
-}
-
-function stripXml(xml: string): string {
-  return xml.replace(/<[^>]+>/g, "").replace(/\s+/g, " ").trim();
-}
-
-function flattenParagraphText(paragraph: ParagraphNode): string {
-  return paragraph.children
-    .map((child) => {
-      switch (child.type) {
-        case "text":
-          return child.text;
-        case "hyperlink":
-          return flattenInlineText(child.children);
-        case "field":
-          return child.children ? flattenInlineText(child.children) : "";
-        case "tab":
-          return "\t";
-        case "hard_break":
-        case "column_break":
-          return "\n";
-        default:
-          return "";
-      }
-    })
-    .join("");
-}
-
-function flattenInlineText(children: HyperlinkNode["children"] | FieldNode["children"]): string {
-  return children
-    .map((child) => {
-      if (child.type === "text") {
-        return child.text;
-      }
-      if (child.type === "tab") {
-        return "\t";
-      }
-      if (child.type === "hard_break" || child.type === "column_break") {
-        return "\n";
-      }
-      return "";
-    })
-    .join("");
-}
-
-function dedupeCrossReferences(references: CrossReference[]): CrossReference[] {
-  const seen = new Set<string>();
-  const deduped: CrossReference[] = [];
-
-  for (const reference of references) {
-    const key = [
-      reference.source,
-      reference.kind,
-      reference.target ?? "",
-      reference.label,
-      String(reference.paragraphIndex ?? -1),
-    ].join("|");
-
-    if (seen.has(key)) {
-      continue;
-    }
-
-    seen.add(key);
-    deduped.push(reference);
-  }
-
-  return deduped;
-}
-
-/**
- * Collect TOC entries from a field registry's tocStructure.
- * Returns the entries in document order with their heading text and levels.
- * Returns an empty array if no TOC structure is available.
- */
-export function collectTocEntriesFromRegistry(
-  registry: FieldRegistry | undefined,
-): TocEntry[] {
-  return registry?.tocStructure?.entries ?? [];
-}
-
-/**
- * Assess whether the TOC structure in a field registry is current
- * with respect to the document's heading structure.
- */
-export function assessTocFreshness(
-  registry: FieldRegistry | undefined,
-): "current" | "stale" | "absent" {
-  if (!registry?.tocStructure) return "absent";
-  return registry.tocStructure.status;
-}
-
-function walkDocument(node: DocumentNode, visit: (node: DocumentNode) => void): void {
-  visit(node);
-
-  if ("children" in node && Array.isArray(node.children)) {
-    for (const child of node.children) {
-      walkDocument(child, visit);
-    }
-  }
-
-  if (node.type === "table") {
-    for (const row of node.rows) {
-      walkDocument(row, visit);
-    }
-  } else if (node.type === "table_row") {
-    for (const cell of node.cells) {
-      walkDocument(cell, visit);
-    }
-  }
-}