docrev 0.9.4 → 0.9.5

package/lib/commands/utilities.ts

@@ -26,6 +26,34 @@ import {
 // Use the actual BuildConfig from build.ts which allows string|Author[]
 type BuildConfig = ReturnType<typeof loadBuildConfig>;
 
+interface ZipLike {
+  addLocalFile(localPath: string, zipPath?: string): void;
+}
+
+/**
+ * Recursively add directory contents to a zip archive, filtering by predicate
+ */
+function addDirToZip(
+  zip: ZipLike,
+  dir: string,
+  shouldInclude: (name: string) => boolean,
+  zipPath = '',
+): void {
+  const entries = fs.readdirSync(dir, { withFileTypes: true });
+  for (const entry of entries) {
+    const fullPath = path.join(dir, entry.name);
+    const entryZipPath = path.join(zipPath, entry.name);
+
+    if (!shouldInclude(entry.name)) continue;
+
+    if (entry.isDirectory()) {
+      addDirToZip(zip, fullPath, shouldInclude, entryZipPath);
+    } else {
+      zip.addLocalFile(fullPath, zipPath || undefined);
+    }
+  }
+}
+
 // Type definitions for package.json
 interface PackageJson {
   version?: string;
@@ -523,33 +551,7 @@ export function register(program: Command, pkg?: PackageJson): void {
         return true;
       };
 
-      const addDir = (dir: string, zipPath = ''): void => {
-        const entries = fs.readdirSync(dir, { withFileTypes: true });
-        for (const entry of entries) {
-          const fullPath = path.join(dir, entry.name);
-          const entryZipPath = path.join(zipPath, entry.name);
-
-          if (!shouldInclude(entry.name)) continue;
-
-          if (entry.isDirectory()) {
-            addDir(fullPath, entryZipPath);
-          } else {
-            zip.addLocalFile(fullPath, zipPath || undefined);
-          }
-        }
-      };
-
-      // Add current directory
-      const entries = fs.readdirSync('.', { withFileTypes: true });
-      for (const entry of entries) {
-        if (!shouldInclude(entry.name)) continue;
-
-        if (entry.isDirectory()) {
-          addDir(entry.name, entry.name);
-        } else if (entry.isFile()) {
-          zip.addLocalFile(entry.name);
-        }
-      }
+      addDirToZip(zip, '.', shouldInclude);
 
       zip.writeZip(outputPath);
       console.log(fmt.status('success', `Backup created: ${outputPath}`));
@@ -741,32 +743,7 @@ export function register(program: Command, pkg?: PackageJson): void {
         return true;
       };
 
-      const addDir = (dir: string, zipPath = ''): void => {
-        const entries = fs.readdirSync(dir, { withFileTypes: true });
-        for (const entry of entries) {
-          const fullPath = path.join(dir, entry.name);
-          const entryZipPath = path.join(zipPath, entry.name);
-
-          if (!shouldInclude(entry.name)) continue;
-
-          if (entry.isDirectory()) {
-            addDir(fullPath, entryZipPath);
-          } else {
-            zip.addLocalFile(fullPath, zipPath || undefined);
-          }
-        }
-      };
-
-      const entries = fs.readdirSync('.', { withFileTypes: true });
-      for (const entry of entries) {
-        if (!shouldInclude(entry.name)) continue;
-
-        if (entry.isDirectory()) {
-          addDir(entry.name, entry.name);
-        } else if (entry.isFile()) {
-          zip.addLocalFile(entry.name);
-        }
-      }
+      addDirToZip(zip, '.', shouldInclude);
 
       zip.writeZip(outputPath);
       console.log(fmt.status('success', `Exported: ${outputPath}`));

package/lib/comment-realign.ts

@@ -213,14 +213,6 @@ function parseMdParagraphs(markdown: string): MdParagraph[] {
   return paragraphs;
 }
 
-/**
- * Strip existing comments from a specific author
- */
-function stripAuthorComments(text: string, author: string): string {
-  const pattern = new RegExp(`\\s*\\{>>${author}:[^<]*<<\\}`, 'g');
-  return text.replace(pattern, '');
-}
-
 /**
  * Normalize text for matching (remove citations, extra whitespace)
  */

package/lib/dependencies.ts

@@ -5,44 +5,36 @@
 import { execSync } from 'child_process';
 
 /**
- * Check if pandoc-crossref is available
+ * Check if a command is available by running it silently
  */
-export function hasPandocCrossref(): boolean {
+function commandExists(cmd: string): boolean {
   try {
-    execSync('pandoc-crossref --version', { stdio: 'ignore' });
+    execSync(cmd, { stdio: 'ignore' });
     return true;
   } catch {
     return false;
   }
 }
 
+/**
+ * Check if pandoc-crossref is available
+ */
+export function hasPandocCrossref(): boolean {
+  return commandExists('pandoc-crossref --version');
+}
+
 /**
  * Check if pandoc is available
  */
 export function hasPandoc(): boolean {
-  try {
-    execSync('pandoc --version', { stdio: 'ignore' });
-    return true;
-  } catch {
-    return false;
-  }
+  return commandExists('pandoc --version');
 }
 
 /**
  * Check if LaTeX is available (for PDF generation)
  */
 export function hasLatex(): boolean {
-  try {
-    execSync('pdflatex --version', { stdio: 'ignore' });
-    return true;
-  } catch {
-    try {
-      execSync('xelatex --version', { stdio: 'ignore' });
-      return true;
-    } catch {
-      return false;
-    }
-  }
+  return commandExists('pdflatex --version') || commandExists('xelatex --version');
 }
 
 /**

package/lib/git.ts

@@ -103,14 +103,18 @@ export function getChangedFiles(fromRef: string, toRef: string = 'HEAD'): Change
 }
 
 /**
- * Get commit history for a file
- * @param filePath - Path to file
- * @param limit - Maximum number of commits to return
+ * Run git log with a given format and optional file path, parse pipe-delimited output
  */
-export function getFileHistory(filePath: string, limit: number = 10): CommitInfo[] {
+function runGitLog(
+  format: string,
+  limit: number,
+  fields: (keyof CommitInfo)[],
+  filePath?: string,
+): CommitInfo[] {
   try {
+    const fileArg = filePath ? ` -- "${filePath}"` : '';
     const output = execSync(
-      `git log --format="%h|%ci|%s" -n ${limit} -- "${filePath}"`,
+      `git log --format="${format}" -n ${limit}${fileArg}`,
       { stdio: 'pipe' }
     ).toString().trim();
 
@@ -118,18 +122,26 @@ export function getFileHistory(filePath: string, limit: number = 10): CommitInfo
 
     return output.split('\n').map(line => {
       const parts = line.split('|');
-      return {
-        hash: parts[0] ?? '',
-        date: parts[1] ?? '',
-        message: parts[2] ?? '',
-        author: ''
-      };
+      const entry: CommitInfo = { hash: '', date: '', author: '', message: '' };
+      for (let i = 0; i < fields.length; i++) {
+        entry[fields[i]] = parts[i] ?? '';
+      }
+      return entry;
     });
   } catch {
     return [];
   }
 }
 
+/**
+ * Get commit history for a file
+ * @param filePath - Path to file
+ * @param limit - Maximum number of commits to return
+ */
+export function getFileHistory(filePath: string, limit: number = 10): CommitInfo[] {
+  return runGitLog('%h|%ci|%s', limit, ['hash', 'date', 'message'], filePath);
+}
+
 /**
  * Compare file content between two refs
  * @param filePath - Path to file
@@ -194,26 +206,7 @@ export function getWordCountDiff(
  * @param limit - Maximum number of commits to return
  */
 export function getRecentCommits(limit: number = 10): CommitInfo[] {
-  try {
-    const output = execSync(
-      `git log --format="%h|%ci|%an|%s" -n ${limit}`,
-      { stdio: 'pipe' }
-    ).toString().trim();
-
-    if (!output) return [];
-
-    return output.split('\n').map(line => {
-      const parts = line.split('|');
-      return {
-        hash: parts[0] ?? '',
-        date: parts[1] ?? '',
-        author: parts[2] ?? '',
-        message: parts[3] ?? ''
-      };
-    });
-  } catch {
-    return [];
-  }
+  return runGitLog('%h|%ci|%an|%s', limit, ['hash', 'date', 'author', 'message']);
 }
 
 /**

package/lib/import.ts

@@ -28,6 +28,7 @@ import {
   protectTables,
   restoreTables,
 } from './protect-restore.js';
+import { normalizeWhitespace } from './utils.js';
 
 const execAsync = promisify(exec);
 
@@ -1138,17 +1139,6 @@ export function insertCommentsIntoMarkdown(
   return result;
 }
 
-/**
- * Normalize text for comparison (handle whitespace differences)
- */
-function normalizeWhitespace(text: string): string {
-  return text
-    .replace(/\r\n/g, '\n') // Normalize line endings
-    .replace(/\t/g, '    ') // Tabs to spaces
-    .replace(/ +/g, ' ') // Collapse multiple spaces
-    .trim();
-}
-
 /**
  * Fix citation and math annotations by preserving original markdown syntax
  */

package/lib/merge.ts

@@ -178,18 +178,15 @@ export async function checkBaseMatch(basePath: string, reviewerPath: string): Pr
 }
 
 /**
- * Extract changes from a Word document compared to original
- * Uses sentence-level diffing for better conflict detection
- * @param originalText - Original text (from base document)
- * @param wordText - Text extracted from reviewer's Word doc
+ * Extract changes from diffs between original and modified text
+ * @param diffs - Array of diff changes
  * @param reviewer - Reviewer identifier
  */
-export function extractChanges(originalText: string, wordText: string, reviewer: string): ReviewerChange[] {
+function extractChangesFromDiffs(
+  diffs: Array<{ added?: boolean; removed?: boolean; value: string }>,
+  reviewer: string,
+): ReviewerChange[] {
   const changes: ReviewerChange[] = [];
-
-  // Use sentence-level diff for better granularity
-  const diffs = diffSentences(originalText, wordText);
-
   let originalPos = 0;
   let i = 0;
 
@@ -198,11 +195,9 @@ export function extractChanges(originalText: string, wordText: string, reviewer:
     if (!part) break;
 
     if (!part.added && !part.removed) {
-      // Unchanged
       originalPos += part.value.length;
       i++;
     } else if (part.removed && diffs[i + 1]?.added) {
-      // Replacement: removed followed by added
       const nextPart = diffs[i + 1];
       if (!nextPart) break;
       changes.push({
@@ -216,7 +211,6 @@ export function extractChanges(originalText: string, wordText: string, reviewer:
       originalPos += part.value.length;
       i += 2;
     } else if (part.removed) {
-      // Pure deletion
       changes.push({
         reviewer,
         type: 'delete',
@@ -228,7 +222,6 @@ export function extractChanges(originalText: string, wordText: string, reviewer:
       originalPos += part.value.length;
       i++;
     } else if (part.added) {
-      // Pure insertion
       changes.push({
         reviewer,
         type: 'insert',
@@ -244,61 +237,22 @@ export function extractChanges(originalText: string, wordText: string, reviewer:
   return changes;
 }
 
+/**
+ * Extract changes from a Word document compared to original
+ * Uses sentence-level diffing for better conflict detection
+ * @param originalText - Original text (from base document)
+ * @param wordText - Text extracted from reviewer's Word doc
+ * @param reviewer - Reviewer identifier
+ */
+export function extractChanges(originalText: string, wordText: string, reviewer: string): ReviewerChange[] {
+  return extractChangesFromDiffs(diffSentences(originalText, wordText), reviewer);
+}
+
 /**
  * Extract changes using word-level diff (more fine-grained)
  */
 export function extractChangesWordLevel(originalText: string, wordText: string, reviewer: string): ReviewerChange[] {
-  const changes: ReviewerChange[] = [];
-  const diffs = diffWords(originalText, wordText);
-
-  let originalPos = 0;
-  let i = 0;
-
-  while (i < diffs.length) {
-    const part = diffs[i];
-    if (!part) break;
-
-    if (!part.added && !part.removed) {
-      originalPos += part.value.length;
-      i++;
-    } else if (part.removed && diffs[i + 1]?.added) {
-      const nextPart = diffs[i + 1];
-      if (!nextPart) break;
-      changes.push({
-        reviewer,
-        type: 'replace',
-        start: originalPos,
-        end: originalPos + part.value.length,
-        oldText: part.value,
-        newText: nextPart.value,
-      });
-      originalPos += part.value.length;
-      i += 2;
-    } else if (part.removed) {
-      changes.push({
-        reviewer,
-        type: 'delete',
-        start: originalPos,
-        end: originalPos + part.value.length,
-        oldText: part.value,
-        newText: '',
-      });
-      originalPos += part.value.length;
-      i++;
-    } else if (part.added) {
-      changes.push({
-        reviewer,
-        type: 'insert',
-        start: originalPos,
-        end: originalPos,
-        oldText: '',
-        newText: part.value,
-      });
-      i++;
-    }
-  }
-
-  return changes;
+  return extractChangesFromDiffs(diffWords(originalText, wordText), reviewer);
 }
 
 /**
@@ -563,23 +517,15 @@ export function clearConflicts(projectDir: string): void {
 }
 
 /**
- * Merge multiple Word documents using three-way merge
+ * Core merge logic: extract changes from reviewer docs, detect conflicts, apply annotations
  */
-export async function mergeThreeWay(
-  basePath: string,
+async function mergeReviewerDocsCore(
+  baseText: string,
   reviewerDocs: ReviewerDoc[],
-  options: MergeOptions = {}
-): Promise<MergeResult & { baseText: string }> {
+  options: MergeOptions = {},
+): Promise<MergeResult> {
   const { diffLevel = 'sentence' } = options;
 
-  if (!fs.existsSync(basePath)) {
-    throw new Error(`Base document not found: ${basePath}`);
-  }
-
-  // Extract text from base document
-  const { text: baseText } = await extractFromWord(basePath);
-
-  // Extract changes from each reviewer relative to base
   const allChanges: ReviewerChange[][] = [];
   const allComments: ReviewerComment[] = [];
 
@@ -590,14 +536,12 @@ export async function mergeThreeWay(
 
     const { text: wordText } = await extractFromWord(doc.path);
 
-    // Choose diff level
     const changes = diffLevel === 'word'
       ? extractChangesWordLevel(baseText, wordText, doc.name)
       : extractChanges(baseText, wordText, doc.name);
 
     allChanges.push(changes);
 
-    // Also extract comments
     try {
       const comments = await extractWordComments(doc.path);
       allComments.push(...comments.map(c => ({ ...c, reviewer: doc.name })));
@@ -609,13 +553,10 @@ export async function mergeThreeWay(
     }
   }
 
-  // Detect conflicts
   const { conflicts, nonConflicting } = detectConflicts(allChanges);
 
-  // Apply non-conflicting changes as annotations
   let merged = applyChangesAsAnnotations(baseText, nonConflicting);
 
-  // Add comments with reviewer attribution
   for (const comment of allComments) {
     merged += `\n{>>${comment.reviewer}: ${comment.text}<<}`;
   }
@@ -628,7 +569,24 @@ export async function mergeThreeWay(
     comments: allComments.length,
   };
 
-  return { merged, conflicts, stats, baseText, originalText: baseText };
+  return { merged, conflicts, stats, originalText: baseText };
+}
+
+/**
+ * Merge multiple Word documents using three-way merge
+ */
+export async function mergeThreeWay(
+  basePath: string,
+  reviewerDocs: ReviewerDoc[],
+  options: MergeOptions = {}
+): Promise<MergeResult & { baseText: string }> {
+  if (!fs.existsSync(basePath)) {
+    throw new Error(`Base document not found: ${basePath}`);
+  }
+
+  const { text: baseText } = await extractFromWord(basePath);
+  const result = await mergeReviewerDocsCore(baseText, reviewerDocs, options);
+  return { ...result, baseText };
 }
 
 /**
@@ -640,60 +598,12 @@ export async function mergeReviewerDocs(
   reviewerDocs: ReviewerDoc[],
   options: MergeOptions = {}
 ): Promise<MergeResult> {
-  const { autoResolve = false } = options;
-
   if (!fs.existsSync(originalPath)) {
     throw new Error(`Original file not found: ${originalPath}`);
   }
 
   const originalText = fs.readFileSync(originalPath, 'utf-8');
-
-  // Extract changes from each reviewer
-  const allChanges: ReviewerChange[][] = [];
-  const allComments: ReviewerComment[] = [];
-
-  for (const doc of reviewerDocs) {
-    if (!fs.existsSync(doc.path)) {
-      throw new Error(`Reviewer file not found: ${doc.path}`);
-    }
-
-    const { text: wordText } = await extractFromWord(doc.path);
-    const changes = extractChanges(originalText, wordText, doc.name);
-    allChanges.push(changes);
-
-    // Also extract comments
-    try {
-      const comments = await extractWordComments(doc.path);
-      allComments.push(...comments.map(c => ({ ...c, reviewer: doc.name })));
-    } catch (e) {
-      if (process.env.DEBUG) {
-        const error = e as Error;
-        console.warn(`merge: Failed to extract comments:`, error.message);
-      }
-    }
-  }
-
-  // Detect conflicts
-  const { conflicts, nonConflicting } = detectConflicts(allChanges);
-
-  // Apply non-conflicting changes as annotations
-  let merged = applyChangesAsAnnotations(originalText, nonConflicting);
-
-  // Add comments
-  for (const comment of allComments) {
-    // Append comments at the end for now (position tracking is complex)
-    merged += `\n{>>${comment.reviewer}: ${comment.text}<<}`;
-  }
-
-  const stats = {
-    reviewers: reviewerDocs.length,
-    totalChanges: allChanges.flat().length,
-    nonConflicting: nonConflicting.length,
-    conflicts: conflicts.length,
-    comments: allComments.length,
-  };
-
-  return { merged, conflicts, stats, originalText };
+  return mergeReviewerDocsCore(originalText, reviewerDocs, options);
 }
 
 /**

package/lib/pdf-comments.ts

@@ -4,6 +4,8 @@
  * Converts CriticMarkup comments to LaTeX margin notes for PDF output
  */
 
+import { escapeLatex } from './utils.js';
+
 /**
  * LaTeX preamble for margin comments
  * Uses todonotes package with custom styling
@@ -117,20 +119,6 @@ export function convertCommentsToMarginNotes(
   };
 }
 
-/**
- * Escape LaTeX special characters
- * @param text - Text to escape
- * @returns Escaped text
- */
-function escapeLatex(text: string): string {
-  return text
-    .replace(/\\/g, '\\textbackslash{}')
-    .replace(/([#$%&_{}])/g, '\\$1')
-    .replace(/\^/g, '\\textasciicircum{}')
-    .replace(/~/g, '\\textasciitilde{}')
-    .replace(/\n/g, ' ');  // Replace newlines with spaces
-}
-
 /**
  * Result of track changes conversion
  */