docrev 0.8.1 → 0.9.0

package/lib/{wordcomments.js → wordcomments.ts}

@@ -17,7 +17,45 @@ const MARKER_START_PREFIX = '⟦CMS:';
 const MARKER_END_PREFIX = '⟦CME:';
 const MARKER_SUFFIX = '⟧';
 
-function escapeXml(str) {
+interface ParsedComment {
+  author: string;
+  text: string;
+  anchor: string | null;
+  start: number;
+  end: number;
+  fullMatch: string;
+}
+
+interface PreparedComment extends ParsedComment {
+  isReply: boolean;
+  parentIdx: number | null;
+  commentIdx: number;
+  anchorFromReply?: boolean;
+  placesParentMarkers?: boolean;
+}
+
+interface PrepareResult {
+  markedMarkdown: string;
+  comments: PreparedComment[];
+}
+
+interface CommentWithIds extends PreparedComment {
+  id: string;
+  paraId: string;
+  paraId2: string;
+  durableId: string;
+  parentParaId?: string;
+}
+
+interface InjectionResult {
+  success: boolean;
+  commentCount: number;
+  replyCount?: number;
+  skippedComments: number;
+  error?: string;
+}
+
+function escapeXml(str: string): string {
   return str
     .replace(/&/g, '&')
     .replace(/</g, '&lt;')
@@ -26,7 +64,7 @@ function escapeXml(str) {
     .replace(/'/g, '&apos;');
 }
 
-function generateParaId(commentIdx, paraNum) {
+function generateParaId(commentIdx: number, paraNum: number): string {
   // Generate 8-character uppercase hex ID matching Word format
   // Word uses IDs like "3F25BC58", "0331C187"
   // Must be deterministic - same inputs always produce same output
@@ -41,14 +79,14 @@ function generateParaId(commentIdx, paraNum) {
  * - markedMarkdown: markdown with markers for parent comments only
  * - comments: array with author, text, isReply, parentIdx
  */
-export function prepareMarkdownWithMarkers(markdown) {
+export function prepareMarkdownWithMarkers(markdown: string): PrepareResult {
   // Match all comments with optional anchor
   const commentPattern = /\{>>(.+?)<<\}(?:\s*\[([^\]]+)\]\{\.mark\})?/g;
 
-  const rawMatches = [];
-  let match;
+  const rawMatches: ParsedComment[] = [];
+  let match: RegExpExecArray | null;
   while ((match = commentPattern.exec(markdown)) !== null) {
-    const content = match[1];
+    const content = match[1] ?? '';
     let author = 'Unknown';
     let text = content;
     const colonIdx = content.indexOf(':');
@@ -75,12 +113,13 @@ export function prepareMarkdownWithMarkers(markdown) {
   // First comment in a cluster = parent, all subsequent = replies to that parent
   // Comments are "adjacent" if there's minimal text between them (< 10 chars)
   const ADJACENT_THRESHOLD = 10;
-  const comments = [];
+  const comments: PreparedComment[] = [];
   let clusterParentIdx = -1;  // Index of first comment in current cluster
   let lastCommentEnd = -1;
 
   for (let i = 0; i < rawMatches.length; i++) {
     const m = rawMatches[i];
+    if (!m) continue;
 
     // Check if this comment is adjacent to the previous one
     const gap = lastCommentEnd >= 0 ? m.start - lastCommentEnd : Infinity;
@@ -94,7 +133,12 @@ export function prepareMarkdownWithMarkers(markdown) {
     if (clusterParentIdx === -1) {
       // First comment in cluster = parent (regardless of author)
       comments.push({
-        ...m,
+        author: m.author,
+        text: m.text,
+        anchor: m.anchor,
+        start: m.start,
+        end: m.end,
+        fullMatch: m.fullMatch,
         isReply: false,
         parentIdx: null,
         commentIdx: comments.length
@@ -103,7 +147,12 @@ export function prepareMarkdownWithMarkers(markdown) {
     } else {
       // Subsequent comment in cluster = reply to first comment
       comments.push({
-        ...m,
+        author: m.author,
+        text: m.text,
+        anchor: m.anchor,
+        start: m.start,
+        end: m.end,
+        fullMatch: m.fullMatch,
         isReply: true,
         parentIdx: clusterParentIdx,
         commentIdx: comments.length
@@ -119,7 +168,7 @@ export function prepareMarkdownWithMarkers(markdown) {
   for (const c of comments) {
     if (c.isReply && c.anchor && c.parentIdx !== null) {
       const parent = comments[c.parentIdx];
-      if (!parent.anchor) {
+      if (parent && !parent.anchor) {
         parent.anchor = c.anchor;
         parent.anchorFromReply = true;  // Parent's anchor came from a reply (markers placed by reply)
         c.placesParentMarkers = true;   // This reply should place the parent's markers
@@ -134,12 +183,14 @@ export function prepareMarkdownWithMarkers(markdown) {
 
   for (let i = comments.length - 1; i >= 0; i--) {
     const c = comments[i];
+    if (!c) continue;
 
     if (c.isReply) {
       // Reply: remove from document entirely (will be in comments.xml only)
       // Also consume leading whitespace to avoid double spaces
       let removeStart = c.start;
-      while (removeStart > 0 && /\s/.test(markedMarkdown[removeStart - 1])) {
+      const charBefore = markedMarkdown[removeStart - 1];
+      while (removeStart > 0 && charBefore && /\s/.test(charBefore)) {
         removeStart--;
       }
 
@@ -148,7 +199,7 @@ export function prepareMarkdownWithMarkers(markdown) {
         // Extract anchor text from the original match
         const anchorMatch = c.fullMatch.match(/\[([^\]]+)\]\{\.mark\}$/);
         if (anchorMatch) {
-          const anchorText = anchorMatch[1];
+          const anchorText = anchorMatch[1] ?? '';
           // Output markers with PARENT's index around the anchor text
           const parentIdx = c.parentIdx;
           const replacement = `${MARKER_START_PREFIX}${parentIdx}${MARKER_SUFFIX}${anchorText}${MARKER_END_PREFIX}${parentIdx}${MARKER_SUFFIX}`;
@@ -164,7 +215,8 @@ export function prepareMarkdownWithMarkers(markdown) {
       if (c.anchorFromReply) {
         // Anchor markers are placed by the reply, just remove this comment
         let removeStart = c.start;
-        while (removeStart > 0 && /\s/.test(markedMarkdown[removeStart - 1])) {
+        const charBefore = markedMarkdown[removeStart - 1];
+        while (removeStart > 0 && charBefore && /\s/.test(charBefore)) {
           removeStart--;
         }
         markedMarkdown = markedMarkdown.slice(0, removeStart) + markedMarkdown.slice(c.end);
@@ -180,7 +232,7 @@ export function prepareMarkdownWithMarkers(markdown) {
   return { markedMarkdown, comments };
 }
 
-function createCommentsXml(comments) {
+function createCommentsXml(comments: CommentWithIds[]): string {
   // Word expects date without milliseconds: 2025-12-30T08:33:00Z
   const now = new Date().toISOString().replace(/\.\d{3}Z$/, 'Z');
 
@@ -209,7 +261,7 @@ function createCommentsXml(comments) {
   return xml;
 }
 
-function createCommentsExtendedXml(comments) {
+function createCommentsExtendedXml(comments: CommentWithIds[]): string {
   let xml = '<?xml version="1.0" encoding="UTF-8" standalone="yes"?>\n';
   // Minimal namespaces matching golden file structure
   xml += '<w15:commentsEx xmlns:mc="http://schemas.openxmlformats.org/markup-compatibility/2006" xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main" xmlns:w14="http://schemas.microsoft.com/office/word/2010/wordml" xmlns:w15="http://schemas.microsoft.com/office/word/2012/wordml" mc:Ignorable="w14 w15">';
@@ -228,7 +280,7 @@ function createCommentsExtendedXml(comments) {
   return xml;
 }
 
-function generateDurableId(index) {
+function generateDurableId(index: number): string {
   // Generate unique 8-char hex ID for durableId
   // CRITICAL: Must stay within signed 32-bit range (< 0x7FFFFFFF = 2147483647)
   // Word interprets durableIds as signed 32-bit integers
@@ -237,7 +289,7 @@ function generateDurableId(index) {
   return id.toString(16).toUpperCase().padStart(8, '0');
 }
 
-function createCommentsIdsXml(comments) {
+function createCommentsIdsXml(comments: CommentWithIds[]): string {
   let xml = '<?xml version="1.0" encoding="UTF-8" standalone="yes"?>\n';
   // Minimal namespaces matching golden file structure
   xml += '<w16cid:commentsIds ';
@@ -257,7 +309,7 @@ function createCommentsIdsXml(comments) {
   return xml;
 }
 
-function createCommentsExtensibleXml(comments) {
+function createCommentsExtensibleXml(comments: CommentWithIds[]): string {
   const now = new Date().toISOString().replace(/\.\d{3}Z$/, 'Z');
 
   let xml = '<?xml version="1.0" encoding="UTF-8" standalone="yes"?>\n';
@@ -278,7 +330,7 @@ function createCommentsExtensibleXml(comments) {
 
 // Generate deterministic user IDs for authors (no hardcoded personal data)
 
-function createPeopleXml(comments) {
+function createPeopleXml(comments: CommentWithIds[]): string {
   // Extract unique authors
   const authors = [...new Set(comments.map(c => c.author))];
 
@@ -308,7 +360,7 @@ function createPeopleXml(comments) {
   return xml;
 }
 
-function generateUserId(author) {
+function generateUserId(author: string): string {
   // Generate a deterministic 16-char hex ID from author name
   let hash = 0;
   for (let i = 0; i < author.length; i++) {
@@ -321,7 +373,11 @@ function generateUserId(author) {
 /**
  * Inject comments at marker positions
  */
-export async function injectCommentsAtMarkers(docxPath, comments, outputPath) {
+export async function injectCommentsAtMarkers(
+  docxPath: string,
+  comments: PreparedComment[],
+  outputPath: string
+): Promise<InjectionResult> {
   try {
     if (!fs.existsSync(docxPath)) {
       return { success: false, commentCount: 0, skippedComments: 0, error: `File not found: ${docxPath}` };
@@ -341,7 +397,7 @@ export async function injectCommentsAtMarkers(docxPath, comments, outputPath) {
     let documentXml = zip.readAsText(documentEntry);
 
     // Assign IDs and paraIds (IDs start at 1, not 0 - Word convention)
-    const commentsWithIds = comments.map((c, idx) => ({
+    const commentsWithIds: CommentWithIds[] = comments.map((c, idx) => ({
       ...c,
       id: String(idx + 1),
       paraId: generateParaId(idx, 1),       // First paragraph (e.g., 10000001)
@@ -352,17 +408,21 @@ export async function injectCommentsAtMarkers(docxPath, comments, outputPath) {
     // Link replies to parent paraIds
     for (const c of commentsWithIds) {
       if (c.isReply && c.parentIdx !== null) {
-        c.parentParaId = commentsWithIds[c.parentIdx].paraId;
+        const parent = commentsWithIds[c.parentIdx];
+        if (parent) {
+          c.parentParaId = parent.paraId;
+        }
       }
     }
 
-    const injectedIds = new Set();
+    const injectedIds = new Set<string>();
 
     // Process only parent comments (non-replies) for document ranges
     const parentComments = commentsWithIds.filter(c => !c.isReply);
 
     for (let i = parentComments.length - 1; i >= 0; i--) {
       const comment = parentComments[i];
+      if (!comment) continue;
       const idx = comment.commentIdx;
 
       const startMarker = `${MARKER_START_PREFIX}${idx}${MARKER_SUFFIX}`;
@@ -392,8 +452,10 @@ export async function injectCommentsAtMarkers(docxPath, comments, outputPath) {
       const textMatch = runContent.match(/<w:t[^>]*>([\s\S]*?)<\/w:t>/);
       if (!textMatch) continue;
 
-      const fullText = textMatch[1];
-      const tElement = textMatch[0].match(/<w:t[^>]*>/)[0];
+      const fullText = textMatch[1] ?? '';
+      const tElementMatch = textMatch[0].match(/<w:t[^>]*>/);
+      if (!tElementMatch) continue;
+      const tElement = tElementMatch[0];
 
       const startInText = fullText.indexOf(startMarker);
       const endInText = fullText.indexOf(endMarker);
@@ -407,7 +469,7 @@ export async function injectCommentsAtMarkers(docxPath, comments, outputPath) {
       if (!anchorText && textAfter) {
         const wordMatch = textAfter.match(/^\s*(\S+)/);
         if (wordMatch) {
-          anchorText = wordMatch[1];
+          anchorText = wordMatch[1] ?? '';
           textAfter = textAfter.slice(wordMatch[0].length);
         }
       }
@@ -425,7 +487,7 @@ export async function injectCommentsAtMarkers(docxPath, comments, outputPath) {
       }
 
       // Find replies to this comment
-      const replies = commentsWithIds.filter(c => c.isReply && c.parentIdx === comment.commentIdx);
+      const replies = commentsWithIds.filter(c => c.isReply && c.parentIdx === comment?.commentIdx);
 
       // Start ranges for parent AND all replies (nested)
       replacement += `<w:commentRangeStart w:id="${comment.id}"/>`;
@@ -458,7 +520,7 @@ export async function injectCommentsAtMarkers(docxPath, comments, outputPath) {
     }
 
     // Add required namespaces to document.xml for comment threading
-    const requiredNs = {
+    const requiredNs: Record<string, string> = {
       'xmlns:w14': 'http://schemas.microsoft.com/office/word/2010/wordml',
       'xmlns:w15': 'http://schemas.microsoft.com/office/word/2012/wordml',
       'xmlns:w16cid': 'http://schemas.microsoft.com/office/word/2016/wordml/cid',
@@ -494,7 +556,8 @@ export async function injectCommentsAtMarkers(docxPath, comments, outputPath) {
         return injectedIds.has(c.id);
       } else {
         // Include reply if its parent was injected
-        return c.parentIdx !== null && injectedIds.has(commentsWithIds[c.parentIdx].id);
+        const parent = c.parentIdx !== null ? commentsWithIds[c.parentIdx] : undefined;
+        return parent && injectedIds.has(parent.id);
       }
     });
 
@@ -639,16 +702,7 @@ export async function injectCommentsAtMarkers(docxPath, comments, outputPath) {
       skippedComments: comments.length - includedComments.length,
     };
 
-  } catch (err) {
+  } catch (err: any) {
     return { success: false, commentCount: 0, skippedComments: 0, error: err.message };
   }
 }
-
-export async function injectComments(docxPath, markdown, outputPath) {
-  console.warn('Warning: Use prepareMarkdownWithMarkers + injectCommentsAtMarkers instead');
-  return { success: false, commentCount: 0, skippedComments: 0, error: 'Use marker-based flow' };
-}
-
-export async function buildWithComments(cleanDocxPath, comments, outputPath) {
-  return injectCommentsAtMarkers(cleanDocxPath, comments, outputPath);
-}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "docrev",
-  "version": "0.8.1",
+  "version": "0.9.0",
   "description": "Academic paper revision workflow: Word ↔ Markdown round-trips, DOI validation, reviewer comments",
   "type": "module",
   "types": "types/index.d.ts",
@@ -81,10 +81,15 @@
     "rev": "bin/rev.js"
   },
   "scripts": {
-    "build": "echo 'No build needed'",
-    "test": "node --test test/*.test.js",
+    "build": "tsc",
+    "build:watch": "tsc --watch",
+    "dev": "tsx bin/rev.ts",
+    "test": "tsx --test test/*.test.js",
+    "test:ts": "tsx --test test/*.test.ts",
     "test:watch": "node --test --watch test/*.test.js",
-    "test:coverage": "c8 --reporter=text --reporter=lcov node --test test/*.test.js"
+    "test:coverage": "c8 --reporter=text --reporter=lcov node --test test/*.test.js",
+    "typecheck": "tsc --noEmit",
+    "prepublishOnly": "npm run build"
   },
   "repository": {
     "type": "git",
@@ -119,10 +124,15 @@
     "nspell": "^2.1.5",
     "pdf-lib": "^1.17.1",
     "pdfjs-dist": "^5.4.530",
+    "tsx": "^4.21.0",
     "xml2js": "^0.6.2",
     "yaml": "^2.8.2"
   },
   "devDependencies": {
-    "c8": "^10.1.2"
+    "@types/adm-zip": "^0.5.7",
+    "@types/node": "^25.2.0",
+    "@types/xml2js": "^0.4.14",
+    "c8": "^10.1.2",
+    "typescript": "^5.9.3"
   }
 }

package/skill/REFERENCE.md

@@ -313,6 +313,73 @@ rev config user "Your Name"  # Set author name for replies
 rev config                   # Show current config
 ```
 
+## rev.yaml Settings
+
+### Tables
+
+Configure table formatting for PDF output:
+
+```yaml
+tables:
+  nowrap:
+    - Prior           # Column headers to keep on one line
+    - "$\\widehat{R}$"
+```
+
+In `nowrap` columns, distribution notation is auto-converted to LaTeX math:
+- `Normal(0, 0.5)` → `$\mathcal{N}(0, 0.5)$`
+- `Student-t(3, 0, 1)` → `$t_3(0, 1)$`
+- `Gamma(2, 0.5)` → `$\text{Gamma}(2, 0.5)$`
+- `Half-Normal(0, 1)` → `$\text{Half-Normal}(0, 1)$`
+- `Exponential(1)` → `$\text{Exp}(1)$`
+
+**Tip:** For complex tables, use simple tables (space-aligned with dashes) instead of pipe tables to avoid column width issues:
+
+```markdown
+  Parameter    Prior                       Description
+  ----------   -------------------------   ------------------
+  alpha        $\mathcal{N}(0, 0.5)$       Intercept prior
+  beta         $t_3(0, 2.5)$               Slope prior
+```
+
+### Postprocess Scripts
+
+Run custom scripts after output generation:
+
+```yaml
+postprocess:
+  pdf: ./scripts/fix-pdf.py      # Runs after PDF
+  docx: ./scripts/add-meta.js    # Runs after DOCX
+  tex: ./scripts/tweak-latex.sh  # Runs after TEX
+  pptx: ./scripts/fix-slides.ps1 # Runs after PPTX
+  all: ./scripts/notify.js       # Runs after ANY format
+```
+
+**Environment variables** available to scripts:
+
+| Variable | Description |
+|----------|-------------|
+| `OUTPUT_FILE` | Full path to generated file |
+| `OUTPUT_FORMAT` | Format: `pdf`, `docx`, `tex`, `pptx`, `beamer` |
+| `PROJECT_DIR` | Directory containing rev.yaml |
+| `CONFIG_PATH` | Full path to rev.yaml |
+
+**Supported script types:** `.js`, `.mjs` (Node), `.py` (Python), `.ps1` (PowerShell), `.sh` (Bash)
+
+**Example Python script** (add DOCX metadata):
+```python
+import os
+from docx import Document
+doc = Document(os.environ['OUTPUT_FILE'])
+doc.core_properties.author = "Research Team"
+doc.save(os.environ['OUTPUT_FILE'])
+```
+
+Use `--verbose` flag to see script output:
+```bash
+rev build pdf --verbose
+```
+
 ## Shell Completions
 
 ### rev completions

package/tsconfig.json

@@ -0,0 +1,26 @@
+{
+  "compilerOptions": {
+    "target": "ES2022",
+    "module": "NodeNext",
+    "moduleResolution": "NodeNext",
+    "lib": ["ES2022"],
+    "outDir": "./dist",
+    "rootDir": ".",
+    "declaration": true,
+    "declarationMap": true,
+    "sourceMap": true,
+    "strict": true,
+    "noImplicitAny": true,
+    "strictNullChecks": true,
+    "noImplicitReturns": true,
+    "noUncheckedIndexedAccess": false,
+    "esModuleInterop": true,
+    "skipLibCheck": true,
+    "forceConsistentCasingInFileNames": true,
+    "resolveJsonModule": true,
+    "allowJs": true,
+    "checkJs": false
+  },
+  "include": ["lib/**/*", "bin/**/*"],
+  "exclude": ["node_modules", "dist", "test"]
+}