docrev 0.2.1 → 0.5.0

package/lib/grammar.js

@@ -0,0 +1,290 @@
+/**
+ * Grammar checker module with custom dictionary support
+ *
+ * Features:
+ * - Common grammar/style issues detection
+ * - Custom dictionary for project-specific terms
+ * - Learn mode to add words to dictionary
+ */
+
+import * as fs from 'fs';
+import * as path from 'path';
+
+// Default dictionary location
+const DEFAULT_DICT_NAME = '.rev-dictionary';
+
+/**
+ * Common grammar/style rules
+ */
+const GRAMMAR_RULES = [
+  {
+    id: 'passive-voice',
+    pattern: /\b(is|are|was|were|be|been|being)\s+(being\s+)?\w+ed\b/gi,
+    message: 'Possible passive voice',
+    severity: 'info',
+  },
+  {
+    id: 'weasel-words',
+    pattern: /\b(very|really|quite|extremely|fairly|rather|somewhat|just)\b/gi,
+    message: 'Weasel word - consider removing or being more specific',
+    severity: 'warning',
+  },
+  {
+    id: 'weak-start',
+    pattern: /^\s*(There (is|are|was|were)|It is)\b/gmi,
+    message: 'Weak sentence start - consider restructuring',
+    severity: 'info',
+  },
+  {
+    id: 'duplicate-words',
+    pattern: /\b(\w+)\s+\1\b/gi,
+    message: 'Duplicate word',
+    severity: 'error',
+  },
+  {
+    id: 'split-infinitive',
+    pattern: /\bto\s+(\w+ly)\s+\w+\b/gi,
+    message: 'Split infinitive',
+    severity: 'info',
+  },
+  {
+    id: 'sentence-length',
+    pattern: /[^.!?]*[.!?]/g,
+    check: (match) => {
+      const words = match.trim().split(/\s+/).length;
+      return words > 40;
+    },
+    message: 'Long sentence (>40 words) - consider breaking up',
+    severity: 'warning',
+  },
+  {
+    id: 'cliches',
+    pattern: /\b(at the end of the day|in terms of|it goes without saying|needless to say|as a matter of fact|first and foremost|last but not least)\b/gi,
+    message: 'Cliche - consider rephrasing',
+    severity: 'warning',
+  },
+  {
+    id: 'hedging',
+    pattern: /\b(seems to|appears to|tends to|might|may|could possibly|would seem)\b/gi,
+    message: 'Hedging language - be more direct if appropriate',
+    severity: 'info',
+  },
+  {
+    id: 'redundancy',
+    pattern: /\b(basic fundamentals|end result|free gift|future plans|past history|completely unique|absolutely essential|close proximity|each and every|first began|true fact|advance planning|final outcome)\b/gi,
+    message: 'Redundant phrase',
+    severity: 'warning',
+  },
+];
+
+/**
+ * Scientific writing specific rules
+ */
+const SCIENTIFIC_RULES = [
+  {
+    id: 'first-person',
+    pattern: /\b(I|we|my|our)\b/gi,
+    message: 'First person pronoun - check if appropriate for your journal',
+    severity: 'info',
+  },
+  {
+    id: 'significant',
+    pattern: /\bsignificant(ly)?\b(?!\s+(P|p|α|difference|effect|increase|decrease|correlation))/gi,
+    message: '"Significant" without statistical context - clarify or use different word',
+    severity: 'warning',
+  },
+  {
+    id: 'prove',
+    pattern: /\b(prove[ds]?|proof)\b/gi,
+    message: 'Avoid "prove" in science - use "demonstrate", "show", "suggest"',
+    severity: 'warning',
+  },
+  {
+    id: 'obviously',
+    pattern: /\b(obviously|clearly|of course)\b/gi,
+    message: 'If obvious, no need to say so; if not obvious, this doesn\'t help',
+    severity: 'warning',
+  },
+];
+
+/**
+ * Load custom dictionary from file
+ * @param {string} directory - Directory to search for dictionary
+ * @returns {Set<string>} Set of custom words
+ */
+export function loadDictionary(directory = '.') {
+  const dictPath = path.join(directory, DEFAULT_DICT_NAME);
+  const words = new Set();
+
+  if (fs.existsSync(dictPath)) {
+    const content = fs.readFileSync(dictPath, 'utf-8');
+    const lines = content.split('\n');
+
+    for (const line of lines) {
+      const word = line.trim().toLowerCase();
+      if (word && !word.startsWith('#')) {
+        words.add(word);
+      }
+    }
+  }
+
+  return words;
+}
+
+/**
+ * Save custom dictionary to file
+ * @param {Set<string>} words - Set of words
+ * @param {string} directory - Directory to save dictionary
+ */
+export function saveDictionary(words, directory = '.') {
+  const dictPath = path.join(directory, DEFAULT_DICT_NAME);
+
+  const header = `# Custom dictionary for docrev
+# Add one word per line
+# Lines starting with # are comments
+`;
+
+  const content = header + [...words].sort().join('\n') + '\n';
+  fs.writeFileSync(dictPath, content, 'utf-8');
+}
+
+/**
+ * Add word to custom dictionary
+ * @param {string} word - Word to add
+ * @param {string} directory - Directory containing dictionary
+ * @returns {boolean} True if word was added (not already present)
+ */
+export function addToDictionary(word, directory = '.') {
+  const words = loadDictionary(directory);
+  const normalizedWord = word.trim().toLowerCase();
+
+  if (words.has(normalizedWord)) {
+    return false;
+  }
+
+  words.add(normalizedWord);
+  saveDictionary(words, directory);
+  return true;
+}
+
+/**
+ * Remove word from custom dictionary
+ * @param {string} word - Word to remove
+ * @param {string} directory - Directory containing dictionary
+ * @returns {boolean} True if word was removed
+ */
+export function removeFromDictionary(word, directory = '.') {
+  const words = loadDictionary(directory);
+  const normalizedWord = word.trim().toLowerCase();
+
+  if (!words.has(normalizedWord)) {
+    return false;
+  }
+
+  words.delete(normalizedWord);
+  saveDictionary(words, directory);
+  return true;
+}
+
+/**
+ * Check text for grammar/style issues
+ * @param {string} text - Text to check
+ * @param {object} options - Options
+ * @param {boolean} options.scientific - Include scientific writing rules
+ * @param {string} options.directory - Directory for custom dictionary
+ * @returns {Array<{rule: string, message: string, severity: string, line: number, match: string}>}
+ */
+export function checkGrammar(text, options = {}) {
+  const { scientific = true, directory = '.' } = options;
+  const customDict = loadDictionary(directory);
+  const issues = [];
+
+  // Get all rules
+  const rules = scientific ? [...GRAMMAR_RULES, ...SCIENTIFIC_RULES] : GRAMMAR_RULES;
+
+  // Split into lines for line number tracking
+  const lines = text.split('\n');
+
+  for (let lineNum = 0; lineNum < lines.length; lineNum++) {
+    const line = lines[lineNum];
+
+    // Skip code blocks and YAML frontmatter
+    if (line.trim().startsWith('```') || line.trim().startsWith('---')) {
+      continue;
+    }
+
+    // Skip lines that are just markdown syntax
+    if (/^[#\-*>|]/.test(line.trim()) && line.trim().length < 5) {
+      continue;
+    }
+
+    for (const rule of rules) {
+      const pattern = new RegExp(rule.pattern.source, rule.pattern.flags);
+      let match;
+
+      while ((match = pattern.exec(line)) !== null) {
+        // Check if rule has additional check function
+        if (rule.check && !rule.check(match[0])) {
+          continue;
+        }
+
+        // Skip if word is in custom dictionary
+        const word = match[0].toLowerCase();
+        if (customDict.has(word)) {
+          continue;
+        }
+
+        issues.push({
+          rule: rule.id,
+          message: rule.message,
+          severity: rule.severity,
+          line: lineNum + 1,
+          column: match.index + 1,
+          match: match[0],
+          context: line.trim(),
+        });
+      }
+    }
+  }
+
+  return issues;
+}
+
+/**
+ * Get grammar check summary
+ * @param {Array} issues - List of issues from checkGrammar
+ * @returns {object} Summary stats
+ */
+export function getGrammarSummary(issues) {
+  const summary = {
+    total: issues.length,
+    errors: 0,
+    warnings: 0,
+    info: 0,
+    byRule: {},
+  };
+
+  for (const issue of issues) {
+    if (issue.severity === 'error') summary.errors++;
+    else if (issue.severity === 'warning') summary.warnings++;
+    else summary.info++;
+
+    summary.byRule[issue.rule] = (summary.byRule[issue.rule] || 0) + 1;
+  }
+
+  return summary;
+}
+
+/**
+ * List available grammar rules
+ * @param {boolean} scientific - Include scientific rules
+ * @returns {Array<{id: string, message: string, severity: string}>}
+ */
+export function listRules(scientific = true) {
+  const rules = scientific ? [...GRAMMAR_RULES, ...SCIENTIFIC_RULES] : GRAMMAR_RULES;
+  return rules.map(r => ({
+    id: r.id,
+    message: r.message,
+    severity: r.severity,
+  }));
+}