Docx Generator

#!/usr/bin/env -S deno run --allow-read

/**
 * analyze-template.ts - Extract text content and structure from DOCX files
 *
 * Extracts paragraphs, tables, headers, footers, and placeholders from Word documents
 * for template analysis and content replacement planning.
 *
 * Usage:
 *   deno run --allow-read scripts/analyze-template.ts <input.docx> [options]
 *
 * Options:
 *   -h, --help       Show help
 *   -v, --verbose    Enable verbose output
 *   --pretty         Pretty-print JSON output
 *
 * Permissions:
 *   --allow-read: Read DOCX file
 */

import { parseArgs } from "jsr:@std/cli@1.0.9/parse-args";
import { basename } from "jsr:@std/path@1.0.8";
import JSZip from "npm:jszip@3.10.1";
import { DOMParser } from "npm:@xmldom/xmldom@0.9.6";

// === Types ===
export interface TextRun {
  text: string;
  bold?: boolean;
  italic?: boolean;
  underline?: boolean;
  fontSize?: number;
  fontFamily?: string;
  color?: string;
}

export interface ParagraphInfo {
  index: number;
  style?: string;
  alignment?: "left" | "center" | "right" | "justify";
  runs: TextRun[];
  fullText: string;
  isListItem: boolean;
  listLevel?: number;
}

export interface TableCellInfo {
  row: number;
  col: number;
  text: string;
  paragraphs: ParagraphInfo[];
}

export interface TableInfo {
  index: number;
  rows: number;
  cols: number;
  cells: TableCellInfo[];
}

export interface HeaderFooterInfo {
  type: "header" | "footer";
  section: "default" | "first" | "even";
  paragraphs: ParagraphInfo[];
}

export interface PlaceholderInfo {
  tag: string;
  location: string;
  paragraphIndex?: number;
  tableIndex?: number;
  cellLocation?: string;
}

export interface DocumentInventory {
  filename: string;
  paragraphCount: number;
  tableCount: number;
  imageCount: number;
  paragraphs: ParagraphInfo[];
  tables: TableInfo[];
  headersFooters: HeaderFooterInfo[];
  placeholders: PlaceholderInfo[];
  styles: string[];
}

interface ParsedArgs {
  help: boolean;
  verbose: boolean;
  pretty: boolean;
  _: (string | number)[];
}

// === Constants ===
const VERSION = "1.0.0";
const SCRIPT_NAME = "analyze-template";

const NS = {
  w: "http://schemas.openxmlformats.org/wordprocessingml/2006/main",
  r: "http://schemas.openxmlformats.org/officeDocument/2006/relationships",
};

// Placeholder pattern: {{PLACEHOLDER}} or ${placeholder}
const PLACEHOLDER_REGEX = /\{\{([^}]+)\}\}|\$\{([^}]+)\}/g;

// === Help Text ===
function printHelp(): void {
  console.log(`
${SCRIPT_NAME} v${VERSION} - Extract text content from DOCX templates

Usage:
  deno run --allow-read scripts/${SCRIPT_NAME}.ts <input.docx> [options]

Arguments:
  <input.docx>     Path to the Word document to analyze

Options:
  -h, --help       Show this help message
  -v, --verbose    Enable verbose output (to stderr)
  --pretty         Pretty-print JSON output (default: compact)

Examples:
  # Analyze document
  deno run --allow-read scripts/${SCRIPT_NAME}.ts template.docx > inventory.json

  # With verbose output
  deno run --allow-read scripts/${SCRIPT_NAME}.ts template.docx -v --pretty
`);
}

// === Utility Functions ===
// deno-lint-ignore no-explicit-any
function getElementsByTagNameNS(parent: any, ns: string, localName: string): any[] {
  const elements = parent.getElementsByTagNameNS(ns, localName);
  // deno-lint-ignore no-explicit-any
  return Array.from(elements) as any[];
}

// deno-lint-ignore no-explicit-any
function getTextContent(element: any): string {
  const textElements = getElementsByTagNameNS(element, NS.w, "t");
  return textElements.map((t) => t.textContent || "").join("");
}

// === Parsing Functions ===
// deno-lint-ignore no-explicit-any
function parseTextRun(rElement: any): TextRun {
  const run: TextRun = {
    text: getTextContent(rElement),
  };

  const rPr = getElementsByTagNameNS(rElement, NS.w, "rPr")[0];
  if (rPr) {
    // Bold
    if (getElementsByTagNameNS(rPr, NS.w, "b").length > 0) {
      run.bold = true;
    }
    // Italic
    if (getElementsByTagNameNS(rPr, NS.w, "i").length > 0) {
      run.italic = true;
    }
    // Underline
    if (getElementsByTagNameNS(rPr, NS.w, "u").length > 0) {
      run.underline = true;
    }
    // Font size
    const sz = getElementsByTagNameNS(rPr, NS.w, "sz")[0];
    if (sz) {
      const val = sz.getAttribute("w:val");
      if (val) run.fontSize = parseInt(val, 10) / 2; // Half-points to points
    }
    // Font family
    const rFonts = getElementsByTagNameNS(rPr, NS.w, "rFonts")[0];
    if (rFonts) {
      run.fontFamily = rFonts.getAttribute("w:ascii") || rFonts.getAttribute("w:hAnsi");
    }
    // Color
    const color = getElementsByTagNameNS(rPr, NS.w, "color")[0];
    if (color) {
      run.color = color.getAttribute("w:val");
    }
  }

  return run;
}

// deno-lint-ignore no-explicit-any
function parseParagraph(pElement: any, index: number): ParagraphInfo {
  const runs = getElementsByTagNameNS(pElement, NS.w, "r").map(parseTextRun);
  const fullText = runs.map((r) => r.text).join("");

  const para: ParagraphInfo = {
    index,
    runs,
    fullText,
    isListItem: false,
  };

  const pPr = getElementsByTagNameNS(pElement, NS.w, "pPr")[0];
  if (pPr) {
    // Style
    const pStyle = getElementsByTagNameNS(pPr, NS.w, "pStyle")[0];
    if (pStyle) {
      para.style = pStyle.getAttribute("w:val");
    }
    // Alignment
    const jc = getElementsByTagNameNS(pPr, NS.w, "jc")[0];
    if (jc) {
      const val = jc.getAttribute("w:val");
      if (val === "left" || val === "center" || val === "right" || val === "both") {
        para.alignment = val === "both" ? "justify" : val;
      }
    }
    // List item
    const numPr = getElementsByTagNameNS(pPr, NS.w, "numPr")[0];
    if (numPr) {
      para.isListItem = true;
      const ilvl = getElementsByTagNameNS(numPr, NS.w, "ilvl")[0];
      if (ilvl) {
        para.listLevel = parseInt(ilvl.getAttribute("w:val") || "0", 10);
      }
    }
  }

  return para;
}

// deno-lint-ignore no-explicit-any
function parseTable(tblElement: any, index: number): TableInfo {
  const rows = getElementsByTagNameNS(tblElement, NS.w, "tr");
  const cells: TableCellInfo[] = [];
  let maxCols = 0;

  rows.forEach((row, rowIndex) => {
    const tcs = getElementsByTagNameNS(row, NS.w, "tc");
    maxCols = Math.max(maxCols, tcs.length);

    tcs.forEach((tc, colIndex) => {
      const paragraphs = getElementsByTagNameNS(tc, NS.w, "p").map((p, i) =>
        parseParagraph(p, i)
      );
      cells.push({
        row: rowIndex,
        col: colIndex,
        text: paragraphs.map((p) => p.fullText).join("\n"),
        paragraphs,
      });
    });
  });

  return {
    index,
    rows: rows.length,
    cols: maxCols,
    cells,
  };
}

function findPlaceholders(
  text: string,
  location: string,
  paragraphIndex?: number,
  tableIndex?: number,
  cellLocation?: string
): PlaceholderInfo[] {
  const placeholders: PlaceholderInfo[] = [];
  let match;

  while ((match = PLACEHOLDER_REGEX.exec(text)) !== null) {
    placeholders.push({
      tag: match[0],
      location,
      paragraphIndex,
      tableIndex,
      cellLocation,
    });
  }

  return placeholders;
}

// deno-lint-ignore no-explicit-any
async function parseDocumentPart(
  zip: JSZip,
  partPath: string,
  type: "header" | "footer",
  section: "default" | "first" | "even"
): Promise<HeaderFooterInfo | null> {
  const file = zip.file(partPath);
  if (!file) return null;

  const xml = await file.async("string");
  const parser = new DOMParser();
  const doc = parser.parseFromString(xml, "text/xml");

  const paragraphs = getElementsByTagNameNS(doc, NS.w, "p").map((p, i) =>
    parseParagraph(p, i)
  );

  return {
    type,
    section,
    paragraphs,
  };
}

// === Core Logic ===
export async function analyzeDocument(
  docxPath: string,
  options: { verbose?: boolean } = {}
): Promise<DocumentInventory> {
  const { verbose = false } = options;

  // Read the DOCX file
  const data = await Deno.readFile(docxPath);
  const zip = await JSZip.loadAsync(data);

  const filename = basename(docxPath);

  // Parse main document
  const documentFile = zip.file("word/document.xml");
  if (!documentFile) {
    throw new Error("Invalid DOCX: missing word/document.xml");
  }

  const documentXml = await documentFile.async("string");
  const parser = new DOMParser();
  const doc = parser.parseFromString(documentXml, "text/xml");

  // Parse paragraphs
  const bodyParagraphs = getElementsByTagNameNS(doc, NS.w, "p");
  const paragraphs: ParagraphInfo[] = [];
  const allPlaceholders: PlaceholderInfo[] = [];

  let paragraphIndex = 0;
  for (const p of bodyParagraphs) {
    // Skip paragraphs inside tables (they're handled separately)
    // deno-lint-ignore no-explicit-any
    let parent = (p as any).parentNode;
    let inTable = false;
    while (parent) {
      if (parent.localName === "tc") {
        inTable = true;
        break;
      }
      parent = parent.parentNode;
    }

    if (!inTable) {
      const para = parseParagraph(p, paragraphIndex);
      paragraphs.push(para);

      // Find placeholders
      const placeholders = findPlaceholders(
        para.fullText,
        "paragraph",
        paragraphIndex
      );
      allPlaceholders.push(...placeholders);

      paragraphIndex++;
    }
  }

  if (verbose) {
    console.error(`Found ${paragraphs.length} paragraphs`);
  }

  // Parse tables
  const tableElements = getElementsByTagNameNS(doc, NS.w, "tbl");
  const tables = tableElements.map((tbl, i) => {
    const table = parseTable(tbl, i);

    // Find placeholders in table cells
    for (const cell of table.cells) {
      const placeholders = findPlaceholders(
        cell.text,
        "table",
        undefined,
        i,
        `R${cell.row + 1}C${cell.col + 1}`
      );
      allPlaceholders.push(...placeholders);
    }

    return table;
  });

  if (verbose) {
    console.error(`Found ${tables.length} tables`);
  }

  // Count images
  let imageCount = 0;
  zip.forEach((path) => {
    if (path.startsWith("word/media/")) {
      imageCount++;
    }
  });

  if (verbose) {
    console.error(`Found ${imageCount} images`);
  }

  // Parse headers and footers
  const headersFooters: HeaderFooterInfo[] = [];
  const headerFooterFiles = [
    { path: "word/header1.xml", type: "header" as const, section: "default" as const },
    { path: "word/header2.xml", type: "header" as const, section: "first" as const },
    { path: "word/header3.xml", type: "header" as const, section: "even" as const },
    { path: "word/footer1.xml", type: "footer" as const, section: "default" as const },
    { path: "word/footer2.xml", type: "footer" as const, section: "first" as const },
    { path: "word/footer3.xml", type: "footer" as const, section: "even" as const },
  ];

  for (const { path, type, section } of headerFooterFiles) {
    const hf = await parseDocumentPart(zip, path, type, section);
    if (hf && hf.paragraphs.length > 0) {
      headersFooters.push(hf);

      // Find placeholders in headers/footers
      for (const para of hf.paragraphs) {
        const placeholders = findPlaceholders(
          para.fullText,
          `${type}-${section}`,
          para.index
        );
        allPlaceholders.push(...placeholders);
      }
    }
  }

  // Collect unique styles
  const styles = new Set<string>();
  for (const para of paragraphs) {
    if (para.style) styles.add(para.style);
  }

  if (verbose) {
    console.error(`Found ${allPlaceholders.length} placeholders`);
  }

  return {
    filename,
    paragraphCount: paragraphs.length,
    tableCount: tables.length,
    imageCount,
    paragraphs,
    tables,
    headersFooters,
    placeholders: allPlaceholders,
    styles: Array.from(styles),
  };
}

// === Main CLI Handler ===
async function main(args: string[]): Promise<void> {
  const parsed = parseArgs(args, {
    boolean: ["help", "verbose", "pretty"],
    alias: { help: "h", verbose: "v" },
    default: { verbose: false, pretty: false },
  }) as ParsedArgs;

  if (parsed.help) {
    printHelp();
    Deno.exit(0);
  }

  const positionalArgs = parsed._.map(String);

  if (positionalArgs.length === 0) {
    console.error("Error: No input file provided\n");
    printHelp();
    Deno.exit(1);
  }

  const inputPath = positionalArgs[0];

  try {
    const inventory = await analyzeDocument(inputPath, {
      verbose: parsed.verbose,
    });

    const output = parsed.pretty
      ? JSON.stringify(inventory, null, 2)
      : JSON.stringify(inventory);
    console.log(output);
  } catch (error) {
    console.error(
      "Error:",
      error instanceof Error ? error.message : String(error)
    );
    Deno.exit(1);
  }
}

// === Entry Point ===
if (import.meta.main) {
  main(Deno.args);
}

#!/usr/bin/env -S deno run --allow-read --allow-write

/**
 * generate-from-template.ts - Generate DOCX from existing templates
 *
 * Modifies existing Word templates using placeholder replacement.
 * Finds and replaces tagged content (e.g., {{TITLE}}, ${author}) in
 * paragraphs, tables, headers, and footers.
 *
 * Usage:
 *   deno run --allow-read --allow-write scripts/generate-from-template.ts <template.docx> <spec.json> <output.docx>
 *
 * Options:
 *   -h, --help       Show help
 *   -v, --verbose    Enable verbose output
 *
 * Permissions:
 *   --allow-read: Read template and specification files
 *   --allow-write: Write output DOCX file
 */

import { parseArgs } from "jsr:@std/cli@1.0.9/parse-args";
import { basename, dirname, resolve } from "jsr:@std/path@1.0.8";
import JSZip from "npm:jszip@3.10.1";
import { DOMParser, XMLSerializer } from "npm:@xmldom/xmldom@0.9.6";

// === Types ===

export interface TextReplacement {
  /** The tag to find and replace (e.g., "{{TITLE}}" or "${author}") */
  tag: string;
  /** The replacement text */
  value: string;
}

export interface ImageReplacement {
  /** The placeholder tag for the image (e.g., "{{LOGO}}") */
  tag: string;
  /** Path to the image file (relative to spec file) */
  imagePath: string;
  /** Width in pixels */
  width: number;
  /** Height in pixels */
  height: number;
}

export interface TemplateSpec {
  /** Text replacements to apply */
  textReplacements?: TextReplacement[];
  /** Image replacements (replaces placeholder text with image) */
  imageReplacements?: ImageReplacement[];
  /** Whether to apply replacements in headers */
  includeHeaders?: boolean;
  /** Whether to apply replacements in footers */
  includeFooters?: boolean;
}

interface ParsedArgs {
  help: boolean;
  verbose: boolean;
  _: (string | number)[];
}

// === Constants ===
const VERSION = "1.0.0";
const SCRIPT_NAME = "generate-from-template";

// XML Namespaces
const NS = {
  w: "http://schemas.openxmlformats.org/wordprocessingml/2006/main",
  r: "http://schemas.openxmlformats.org/officeDocument/2006/relationships",
  rel: "http://schemas.openxmlformats.org/package/2006/relationships",
  wp: "http://schemas.openxmlformats.org/drawingml/2006/wordprocessingDrawing",
  a: "http://schemas.openxmlformats.org/drawingml/2006/main",
  pic: "http://schemas.openxmlformats.org/drawingml/2006/picture",
};

// Placeholder patterns: {{PLACEHOLDER}} or ${placeholder}
const PLACEHOLDER_PATTERNS = [
  /\{\{([^}]+)\}\}/g,  // {{TAG}}
  /\$\{([^}]+)\}/g,     // ${tag}
];

// === Help Text ===
function printHelp(): void {
  console.log(`
${SCRIPT_NAME} v${VERSION} - Generate DOCX from existing templates

Usage:
  deno run --allow-read --allow-write scripts/${SCRIPT_NAME}.ts <template.docx> <spec.json> <output.docx>

Arguments:
  <template.docx>  Path to the template Word document
  <spec.json>      Path to JSON specification for replacements
  <output.docx>    Path for output Word document

Options:
  -h, --help       Show this help message
  -v, --verbose    Enable verbose output

Specification Format:
  {
    "textReplacements": [
      { "tag": "{{TITLE}}", "value": "Quarterly Report" },
      { "tag": "{{DATE}}", "value": "December 2024" },
      { "tag": "\${author}", "value": "John Smith" }
    ],
    "includeHeaders": true,
    "includeFooters": true
  }

Examples:
  # Replace text in template
  deno run --allow-read --allow-write scripts/${SCRIPT_NAME}.ts \\
    template.docx replacements.json output.docx

  # With verbose output
  deno run --allow-read --allow-write scripts/${SCRIPT_NAME}.ts \\
    template.docx replacements.json output.docx -v
`);
}

// === Utility Functions ===

// deno-lint-ignore no-explicit-any
function getElementsByTagNameNS(parent: any, ns: string, localName: string): any[] {
  const elements = parent.getElementsByTagNameNS(ns, localName);
  // deno-lint-ignore no-explicit-any
  return Array.from(elements) as any[];
}

function escapeRegExp(string: string): string {
  return string.replace(/[.*+?^${}()|[\]\\]/g, "\\$&");
}

// === Text Replacement ===

interface ReplacementResult {
  modified: boolean;
  replacementCount: number;
}

function replaceTextInXml(
  xmlContent: string,
  replacements: TextReplacement[],
  verbose: boolean = false
): { content: string; result: ReplacementResult } {
  const parser = new DOMParser();
  const doc = parser.parseFromString(xmlContent, "text/xml");

  let totalReplacements = 0;
  let modified = false;

  // Find all text elements
  const textElements = getElementsByTagNameNS(doc, NS.w, "t");

  // deno-lint-ignore no-explicit-any
  for (const textEl of textElements as any[]) {
    let text = textEl.textContent || "";
    const originalText = text;

    for (const replacement of replacements) {
      // The tag should match exactly as provided
      const tag = replacement.tag;

      if (text.includes(tag)) {
        const regex = new RegExp(escapeRegExp(tag), "g");
        const matches = text.match(regex);
        if (matches) {
          totalReplacements += matches.length;
        }
        text = text.replace(regex, replacement.value);
      }
    }

    if (text !== originalText) {
      textEl.textContent = text;
      modified = true;
    }
  }

  const serializer = new XMLSerializer();
  return {
    content: serializer.serializeToString(doc),
    result: { modified, replacementCount: totalReplacements },
  };
}

/**
 * Handle split placeholders across multiple text runs.
 * Word often splits text like "{{TITLE}}" into multiple runs like "<w:t>{{</w:t><w:t>TITLE</w:t><w:t>}}</w:t>"
 * This function consolidates runs within a paragraph before replacement.
 */
function consolidateTextRuns(xmlContent: string): string {
  const parser = new DOMParser();
  const doc = parser.parseFromString(xmlContent, "text/xml");

  // Process each paragraph
  const paragraphs = getElementsByTagNameNS(doc, NS.w, "p");

  // deno-lint-ignore no-explicit-any
  for (const para of paragraphs as any[]) {
    const runs = getElementsByTagNameNS(para, NS.w, "r");

    // Collect all text content and check for split placeholders
    let fullText = "";
    // deno-lint-ignore no-explicit-any
    const textNodes: any[] = [];

    // deno-lint-ignore no-explicit-any
    for (const run of runs as any[]) {
      const tElements = getElementsByTagNameNS(run, NS.w, "t");
      for (const t of tElements) {
        fullText += t.textContent || "";
        textNodes.push(t);
      }
    }

    // Check if there are placeholders in the combined text that might be split
    const hasSplitPlaceholder = (
      (fullText.includes("{{") && fullText.includes("}}")) ||
      (fullText.includes("${") && fullText.includes("}"))
    );

    // Only consolidate if we have split placeholders and multiple text nodes
    if (hasSplitPlaceholder && textNodes.length > 1) {
      // Put all text in the first node and clear others
      if (textNodes.length > 0) {
        textNodes[0].textContent = fullText;
        for (let i = 1; i < textNodes.length; i++) {
          textNodes[i].textContent = "";
        }
      }
    }
  }

  const serializer = new XMLSerializer();
  return serializer.serializeToString(doc);
}

// === Core Logic ===

export async function generateFromTemplate(
  templatePath: string,
  spec: TemplateSpec,
  outputPath: string,
  options: { verbose?: boolean; specDir?: string } = {}
): Promise<void> {
  const { verbose = false } = options;

  // Read template
  const templateData = await Deno.readFile(templatePath);
  const zip = await JSZip.loadAsync(templateData);

  if (verbose) {
    console.error(`Loaded template: ${basename(templatePath)}`);
  }

  const replacements = spec.textReplacements || [];
  const includeHeaders = spec.includeHeaders !== false;
  const includeFooters = spec.includeFooters !== false;

  if (verbose) {
    console.error(`Processing ${replacements.length} text replacements`);
    console.error(`Include headers: ${includeHeaders}, Include footers: ${includeFooters}`);
  }

  let totalReplacements = 0;

  // Process main document
  const documentFile = zip.file("word/document.xml");
  if (documentFile) {
    let documentXml = await documentFile.async("string");

    // Consolidate split text runs
    documentXml = consolidateTextRuns(documentXml);

    // Apply replacements
    const { content, result } = replaceTextInXml(documentXml, replacements, verbose);
    zip.file("word/document.xml", content);
    totalReplacements += result.replacementCount;

    if (verbose && result.modified) {
      console.error(`Document: ${result.replacementCount} replacements`);
    }
  }

  // Process headers
  if (includeHeaders) {
    const headerFiles = ["word/header1.xml", "word/header2.xml", "word/header3.xml"];
    for (const headerPath of headerFiles) {
      const headerFile = zip.file(headerPath);
      if (headerFile) {
        let headerXml = await headerFile.async("string");
        headerXml = consolidateTextRuns(headerXml);
        const { content, result } = replaceTextInXml(headerXml, replacements, verbose);
        zip.file(headerPath, content);
        totalReplacements += result.replacementCount;

        if (verbose && result.modified) {
          console.error(`${headerPath}: ${result.replacementCount} replacements`);
        }
      }
    }
  }

  // Process footers
  if (includeFooters) {
    const footerFiles = ["word/footer1.xml", "word/footer2.xml", "word/footer3.xml"];
    for (const footerPath of footerFiles) {
      const footerFile = zip.file(footerPath);
      if (footerFile) {
        let footerXml = await footerFile.async("string");
        footerXml = consolidateTextRuns(footerXml);
        const { content, result } = replaceTextInXml(footerXml, replacements, verbose);
        zip.file(footerPath, content);
        totalReplacements += result.replacementCount;

        if (verbose && result.modified) {
          console.error(`${footerPath}: ${result.replacementCount} replacements`);
        }
      }
    }
  }

  if (verbose) {
    console.error(`Total replacements: ${totalReplacements}`);
  }

  // Write output file
  const outputData = await zip.generateAsync({
    type: "uint8array",
    compression: "DEFLATE",
    compressionOptions: { level: 6 },
  });

  await Deno.writeFile(outputPath, outputData);

  if (verbose) {
    console.error(`Wrote ${outputPath}`);
  }
}

// === Main CLI Handler ===
async function main(args: string[]): Promise<void> {
  const parsed = parseArgs(args, {
    boolean: ["help", "verbose"],
    alias: { help: "h", verbose: "v" },
    default: { verbose: false },
  }) as ParsedArgs;

  if (parsed.help) {
    printHelp();
    Deno.exit(0);
  }

  const positionalArgs = parsed._.map(String);

  if (positionalArgs.length < 3) {
    console.error(
      "Error: template.docx, spec.json, and output.docx are required\n"
    );
    printHelp();
    Deno.exit(1);
  }

  const templatePath = positionalArgs[0];
  const specPath = positionalArgs[1];
  const outputPath = positionalArgs[2];

  try {
    // Read specification
    const specText = await Deno.readTextFile(specPath);
    const spec = JSON.parse(specText) as TemplateSpec;

    const specDir = dirname(resolve(specPath));

    await generateFromTemplate(templatePath, spec, outputPath, {
      verbose: parsed.verbose,
      specDir,
    });

    console.log(`Created: ${outputPath}`);
  } catch (error) {
    console.error(
      "Error:",
      error instanceof Error ? error.message : String(error)
    );
    Deno.exit(1);
  }
}

// === Entry Point ===
if (import.meta.main) {
  main(Deno.args);
}

#!/usr/bin/env -S deno run --allow-read --allow-write

/**
 * generate-scratch.ts - Create DOCX from scratch using JSON specification
 *
 * Creates Word documents programmatically from a JSON specification
 * using the docx library. Supports paragraphs, tables, images, headers, and footers.
 *
 * Usage:
 *   deno run --allow-read --allow-write scripts/generate-scratch.ts <spec.json> <output.docx>
 *
 * Options:
 *   -h, --help       Show help
 *   -v, --verbose    Enable verbose output
 *
 * Permissions:
 *   --allow-read: Read specification file and image assets
 *   --allow-write: Write output DOCX file
 */

import { parseArgs } from "jsr:@std/cli@1.0.9/parse-args";
import { dirname, resolve } from "jsr:@std/path@1.0.8";
import {
  Document,
  Packer,
  Paragraph,
  TextRun,
  Table,
  TableRow,
  TableCell,
  WidthType,
  AlignmentType,
  HeadingLevel,
  BorderStyle,
  Header,
  Footer,
  PageBreak,
  ImageRun,
  ExternalHyperlink,
  // deno-lint-ignore no-explicit-any
} from "npm:docx@9.0.2" as any;

// === Types ===
export interface TextRunSpec {
  text: string;
  bold?: boolean;
  italic?: boolean;
  underline?: boolean;
  strike?: boolean;
  fontSize?: number; // points
  font?: string;
  color?: string; // hex without #
  highlight?: string;
  superScript?: boolean;
  subScript?: boolean;
}

export interface HyperlinkSpec {
  text: string;
  url: string;
  bold?: boolean;
  italic?: boolean;
}

export interface ImageSpec {
  path: string;
  width: number; // pixels
  height: number; // pixels
  altText?: string;
}

export interface ParagraphSpec {
  text?: string;
  runs?: (TextRunSpec | HyperlinkSpec | ImageSpec)[];
  heading?: 1 | 2 | 3 | 4 | 5 | 6;
  alignment?: "left" | "center" | "right" | "justify";
  bullet?: boolean;
  numbering?: boolean;
  spacing?: {
    before?: number;
    after?: number;
    line?: number;
  };
  indent?: {
    left?: number;
    right?: number;
    firstLine?: number;
  };
  pageBreakBefore?: boolean;
}

export interface TableCellSpec {
  content: ParagraphSpec[];
  width?: number; // percentage or points
  rowSpan?: number;
  colSpan?: number;
  shading?: string; // hex color
  verticalAlign?: "top" | "center" | "bottom";
}

export interface TableRowSpec {
  cells: TableCellSpec[];
  isHeader?: boolean;
}

export interface TableSpec {
  rows: TableRowSpec[];
  width?: number; // percentage
  borders?: boolean;
}

export interface HeaderFooterSpec {
  paragraphs: ParagraphSpec[];
}

export interface SectionSpec {
  content: (ParagraphSpec | TableSpec | { pageBreak: true })[];
  header?: HeaderFooterSpec;
  footer?: HeaderFooterSpec;
}

export interface DocumentSpec {
  title?: string;
  subject?: string;
  creator?: string;
  description?: string;
  styles?: {
    defaultFont?: string;
    defaultFontSize?: number;
  };
  sections: SectionSpec[];
}

interface ParsedArgs {
  help: boolean;
  verbose: boolean;
  _: (string | number)[];
}

// === Constants ===
const VERSION = "1.0.0";
const SCRIPT_NAME = "generate-scratch";

// === Help Text ===
function printHelp(): void {
  console.log(`
${SCRIPT_NAME} v${VERSION} - Create DOCX from scratch using JSON specification

Usage:
  deno run --allow-read --allow-write scripts/${SCRIPT_NAME}.ts <spec.json> <output.docx>

Arguments:
  <spec.json>      Path to JSON specification file
  <output.docx>    Path for output Word document

Options:
  -h, --help       Show this help message
  -v, --verbose    Enable verbose output

Specification Format:
  {
    "title": "Document Title",
    "creator": "Author Name",
    "sections": [
      {
        "content": [
          {
            "text": "Hello World",
            "heading": 1
          },
          {
            "runs": [
              { "text": "Bold text", "bold": true },
              { "text": " and ", "italic": true },
              { "text": "normal text" }
            ]
          }
        ]
      }
    ]
  }

Examples:
  # Generate document
  deno run --allow-read --allow-write scripts/${SCRIPT_NAME}.ts spec.json output.docx

  # With verbose output
  deno run --allow-read --allow-write scripts/${SCRIPT_NAME}.ts spec.json output.docx -v
`);
}

// === Conversion Helpers ===
function getAlignment(align?: string): typeof AlignmentType[keyof typeof AlignmentType] | undefined {
  switch (align) {
    case "left": return AlignmentType.LEFT;
    case "center": return AlignmentType.CENTER;
    case "right": return AlignmentType.RIGHT;
    case "justify": return AlignmentType.JUSTIFIED;
    default: return undefined;
  }
}

function getHeadingLevel(level?: number): typeof HeadingLevel[keyof typeof HeadingLevel] | undefined {
  switch (level) {
    case 1: return HeadingLevel.HEADING_1;
    case 2: return HeadingLevel.HEADING_2;
    case 3: return HeadingLevel.HEADING_3;
    case 4: return HeadingLevel.HEADING_4;
    case 5: return HeadingLevel.HEADING_5;
    case 6: return HeadingLevel.HEADING_6;
    default: return undefined;
  }
}

// === Element Builders ===
function buildTextRun(spec: TextRunSpec): typeof TextRun {
  return new TextRun({
    text: spec.text,
    bold: spec.bold,
    italics: spec.italic,
    underline: spec.underline ? {} : undefined,
    strike: spec.strike,
    size: spec.fontSize ? spec.fontSize * 2 : undefined, // points to half-points
    font: spec.font,
    color: spec.color,
    highlight: spec.highlight,
    superScript: spec.superScript,
    subScript: spec.subScript,
  });
}

async function buildImageRun(spec: ImageSpec, specDir: string): Promise<typeof ImageRun> {
  const imagePath = resolve(specDir, spec.path);
  const imageData = await Deno.readFile(imagePath);

  return new ImageRun({
    data: imageData,
    transformation: {
      width: spec.width,
      height: spec.height,
    },
    altText: spec.altText ? { title: spec.altText, description: spec.altText } : undefined,
  });
}

function buildHyperlink(spec: HyperlinkSpec): typeof ExternalHyperlink {
  return new ExternalHyperlink({
    link: spec.url,
    children: [
      new TextRun({
        text: spec.text,
        bold: spec.bold,
        italics: spec.italic,
        style: "Hyperlink",
      }),
    ],
  });
}

function isHyperlink(spec: TextRunSpec | HyperlinkSpec | ImageSpec): spec is HyperlinkSpec {
  return "url" in spec;
}

function isImage(spec: TextRunSpec | HyperlinkSpec | ImageSpec): spec is ImageSpec {
  return "path" in spec && "width" in spec;
}

// deno-lint-ignore no-explicit-any
async function buildParagraph(spec: ParagraphSpec, specDir: string): Promise<any> {
  // deno-lint-ignore no-explicit-any
  const children: any[] = [];

  if (spec.pageBreakBefore) {
    children.push(new PageBreak());
  }

  if (spec.text) {
    children.push(new TextRun(spec.text));
  }

  if (spec.runs) {
    for (const run of spec.runs) {
      if (isImage(run)) {
        children.push(await buildImageRun(run, specDir));
      } else if (isHyperlink(run)) {
        children.push(buildHyperlink(run));
      } else {
        children.push(buildTextRun(run));
      }
    }
  }

  return new Paragraph({
    children,
    heading: getHeadingLevel(spec.heading),
    alignment: getAlignment(spec.alignment),
    bullet: spec.bullet ? { level: 0 } : undefined,
    numbering: spec.numbering ? { reference: "default-numbering", level: 0 } : undefined,
    spacing: spec.spacing ? {
      before: spec.spacing.before,
      after: spec.spacing.after,
      line: spec.spacing.line,
    } : undefined,
    indent: spec.indent ? {
      left: spec.indent.left,
      right: spec.indent.right,
      firstLine: spec.indent.firstLine,
    } : undefined,
  });
}

// deno-lint-ignore no-explicit-any
async function buildTableCell(spec: TableCellSpec, specDir: string): Promise<any> {
  const paragraphs = await Promise.all(
    spec.content.map((p) => buildParagraph(p, specDir))
  );

  return new TableCell({
    children: paragraphs,
    width: spec.width ? { size: spec.width, type: WidthType.PERCENTAGE } : undefined,
    rowSpan: spec.rowSpan,
    columnSpan: spec.colSpan,
    shading: spec.shading ? { fill: spec.shading } : undefined,
    verticalAlign: spec.verticalAlign === "center" ? "center" :
                   spec.verticalAlign === "bottom" ? "bottom" : undefined,
  });
}

// deno-lint-ignore no-explicit-any
async function buildTable(spec: TableSpec, specDir: string): Promise<any> {
  const rows = await Promise.all(
    spec.rows.map(async (rowSpec) => {
      const cells = await Promise.all(
        rowSpec.cells.map((c) => buildTableCell(c, specDir))
      );
      return new TableRow({
        children: cells,
        tableHeader: rowSpec.isHeader,
      });
    })
  );

  return new Table({
    rows,
    width: spec.width ? { size: spec.width, type: WidthType.PERCENTAGE } : undefined,
    borders: spec.borders === false ? {
      top: { style: BorderStyle.NONE },
      bottom: { style: BorderStyle.NONE },
      left: { style: BorderStyle.NONE },
      right: { style: BorderStyle.NONE },
      insideHorizontal: { style: BorderStyle.NONE },
      insideVertical: { style: BorderStyle.NONE },
    } : undefined,
  });
}

function isTable(item: ParagraphSpec | TableSpec | { pageBreak: true }): item is TableSpec {
  return "rows" in item;
}

function isPageBreak(item: ParagraphSpec | TableSpec | { pageBreak: true }): item is { pageBreak: true } {
  return "pageBreak" in item;
}

// deno-lint-ignore no-explicit-any
async function buildHeaderFooter(spec: HeaderFooterSpec, specDir: string): Promise<any[]> {
  return Promise.all(spec.paragraphs.map((p) => buildParagraph(p, specDir)));
}

// === Core Logic ===
export async function generateFromSpec(
  spec: DocumentSpec,
  outputPath: string,
  options: { verbose?: boolean; specDir?: string } = {}
): Promise<void> {
  const { verbose = false, specDir = "." } = options;

  if (verbose) {
    console.error(`Creating document with ${spec.sections.length} section(s)`);
  }

  // Build sections
  // deno-lint-ignore no-explicit-any
  const sections: any[] = [];

  for (let i = 0; i < spec.sections.length; i++) {
    const sectionSpec = spec.sections[i];
    // deno-lint-ignore no-explicit-any
    const children: any[] = [];

    if (verbose) {
      console.error(`Processing section ${i + 1}: ${sectionSpec.content.length} items`);
    }

    for (const item of sectionSpec.content) {
      if (isPageBreak(item)) {
        children.push(new Paragraph({ children: [new PageBreak()] }));
      } else if (isTable(item)) {
        children.push(await buildTable(item, specDir));
      } else {
        children.push(await buildParagraph(item, specDir));
      }
    }

    // deno-lint-ignore no-explicit-any
    const section: any = { children };

    // Add header
    if (sectionSpec.header) {
      section.headers = {
        default: new Header({
          children: await buildHeaderFooter(sectionSpec.header, specDir),
        }),
      };
    }

    // Add footer
    if (sectionSpec.footer) {
      section.footers = {
        default: new Footer({
          children: await buildHeaderFooter(sectionSpec.footer, specDir),
        }),
      };
    }

    sections.push(section);
  }

  // Create document
  const doc = new Document({
    title: spec.title,
    subject: spec.subject,
    creator: spec.creator,
    description: spec.description,
    styles: spec.styles ? {
      default: {
        document: {
          run: {
            font: spec.styles.defaultFont,
            size: spec.styles.defaultFontSize ? spec.styles.defaultFontSize * 2 : undefined,
          },
        },
      },
    } : undefined,
    sections,
  });

  // Write output
  const buffer = await Packer.toBuffer(doc);
  await Deno.writeFile(outputPath, new Uint8Array(buffer));

  if (verbose) {
    console.error(`Wrote ${outputPath}`);
  }
}

// === Main CLI Handler ===
async function main(args: string[]): Promise<void> {
  const parsed = parseArgs(args, {
    boolean: ["help", "verbose"],
    alias: { help: "h", verbose: "v" },
    default: { verbose: false },
  }) as ParsedArgs;

  if (parsed.help) {
    printHelp();
    Deno.exit(0);
  }

  const positionalArgs = parsed._.map(String);

  if (positionalArgs.length < 2) {
    console.error("Error: Both spec.json and output.docx are required\n");
    printHelp();
    Deno.exit(1);
  }

  const specPath = positionalArgs[0];
  const outputPath = positionalArgs[1];

  try {
    const specText = await Deno.readTextFile(specPath);
    const spec = JSON.parse(specText) as DocumentSpec;

    const specDir = dirname(resolve(specPath));

    await generateFromSpec(spec, outputPath, {
      verbose: parsed.verbose,
      specDir,
    });

    console.log(`Created: ${outputPath}`);
  } catch (error) {
    console.error(
      "Error:",
      error instanceof Error ? error.message : String(error)
    );
    Deno.exit(1);
  }
}

// === Entry Point ===
if (import.meta.main) {
  main(Deno.args);
}