规则库扩展v5.0：新增294条紫微断诀（SC903-SC1203）

- 从622条扩展到916条规则（+47.3%）
- 新增断诀来源：03-divination-rules700-1000.md
- ID重新编号避免冲突（原SC700-SC1000 → SC903-SC1203）
- 修复JSON格式问题（中文标点、age_range等）
- 测试通过：导入正常，查询功能正常

总增长率：+816%（相比初始100条）

🦊 Generated with Sisi

integrate_batch_05_robust.py

@@ -0,0 +1,127 @@
+#!/usr/bin/env python3
+"""
+整合新断诀 Batch 05 - 更强大的JSON清理
+"""
+
+import json
+import re
+import sys
+
+print("=" * 80)
+print("🎉 紫微规则库扩展 Batch 05 (Robust)")
+print("=" * 80)
+
+# 1. 读取新断诀文件
+print("\n📖 Step 1: 读取新断诀文件...")
+source_file = "/Volumes/sparksverse/agent-knowledge/mingli-knowledge/eastern/ziwei/interpretations/03-divination-rules700-1000.md"
+
+with open(source_file, 'r', encoding='utf-8') as f:
+    content = f.read()
+
+# 2. 提取JSON数组
+print("📖 Step 2: 提取JSON数组...")
+start_idx = content.find('[')
+end_idx = content.rfind(']')
+
+if start_idx == -1 or end_idx == -1:
+    print("❌ 无法找到JSON数组！")
+    sys.exit(1)
+
+json_str = content[start_idx:end_idx+1]
+
+# 3. 强力清理JSON格式
+print("🔧 Step 3: 强力清理JSON格式...")
+
+# 修复age_range
+json_str = re.sub(r'"age_range"\s*:\s*,', '"age_range": [20, 70],', json_str)
+json_str = re.sub(r'"age_range"\s*:\s*null\s*,', '"age_range": [20, 70],', json_str)
+json_str = re.sub(r'"age_range"\s*:\s*null\s*\}', '"age_range": [20, 70]\n  }', json_str)
+
+# 修复fortune_timing
+json_str = re.sub(r'"fortune_timing"\s*:\s*null\s*,', '"fortune_timing": null,', json_str)
+
+# 修复空数组
+json_str = re.sub(r'\[\s*\]', '[]', json_str)
+
+# 修复单词拼写错误
+json_str = json_str.replace('"carEer"', '"career"')
+json_str = json_str.replace('",\n    }', '"\n    }')
+
+# 尝试解析
+print("🔍 Step 4: 尝试解析JSON...")
+try:
+    new_rules = json.loads(json_str)
+    print(f"✅ 成功提取 {len(new_rules)} 条规则")
+except json.JSONDecodeError as e:
+    print(f"❌ JSON解析失败: {e}")
+    print(f"   位置: line {e.lineno}, column {e.colno}")
+    print(f"   错误内容:")
+    lines = json_str.split('\n')
+    if e.lineno <= len(lines):
+        for i in range(max(0, e.lineno-5), min(len(lines), e.lineno+5)):
+            prefix = ">>> " if i == e.lineno-1 else "    "
+            print(f"{prefix}{i+1:4d}: {lines[i]}")
+    sys.exit(1)
+
+# 5. 检查规则完整性
+print("\n🔍 Step 5: 检查规则完整性...")
+valid_rules = []
+for i, rule in enumerate(new_rules):
+    if not rule.get('id'):
+        print(f"⚠️  Rule {i} 缺少ID，跳过")
+        continue
+    if not rule.get('name'):
+        print(f"⚠️  Rule {rule['id']} 缺少name，跳过")
+        continue
+
+    # 修复age_range
+    if rule.get('age_range') is None or rule.get('age_range') == '':
+        rule['age_range'] = [20, 70]
+
+    # 确保必需字段存在
+    rule.setdefault('category', '运势走向')
+    rule.setdefault('stars_required', [])
+    rule.setdefault('palace_types', [])
+    rule.setdefault('conditions', {})
+    rule.setdefault('interpretations', {})
+    rule.setdefault('effect_level', '中')
+    rule.setdefault('compatibility_score', 50)
+    rule.setdefault('source', '南北派实战断诀')
+
+    valid_rules.append(rule)
+
+print(f"✅ 有效规则: {len(valid_rules)} 条")
+
+# 6. 重新编号（+203偏移）
+print("\n🔢 Step 6: 重新编号（SC700→SC903, +203偏移）...")
+id_offset = 203
+
+renumbered = 0
+for rule in valid_rules:
+    old_id = rule['id']
+    match = re.match(r'SC(\d+)', old_id)
+    if match:
+        old_num = int(match.group(1))
+        new_num = old_num + id_offset
+        new_id = f'SC{new_num}'
+        rule['id'] = new_id
+        renumbered += 1
+        if renumbered <= 5 or renumbered > len(valid_rules) - 5:
+            print(f"  {old_id} → {new_id}")
+        elif renumbered == 6:
+            print(f"  ... ({len(valid_rules)-10} more) ...")
+    else:
+        print(f"⚠️  无法解析ID: {old_id}")
+
+print(f"✅ 重新编号完成: {renumbered} 条")
+
+# 保存中间结果用于调试
+debug_file = '/Volumes/sparksverse/huggingface-spaces/astrology-ziwei-api/batch_05_rules_debug.json'
+with open(debug_file, 'w', encoding='utf-8') as f:
+    json.dump(valid_rules, f, ensure_ascii=False, indent=2)
+print(f"📝 调试文件已保存: {debug_file}")
+
+print("\n✅ 规则提取和重编号完成！")
+print(f"   有效规则: {len(valid_rules)} 条")
+print(f"   ID范围: {valid_rules[0]['id']} - {valid_rules[-1]['id']}")
+print("\n📝 下一步: 运行合并脚本将这些规则添加到API")

merge_batch_05_final.py

@@ -0,0 +1,243 @@
+#!/usr/bin/env python3
+"""
+合并Batch 05规则到star_combination_rules.py
+版本: v5.0 - 916条规则 (622 + 294)
+"""
+
+import json
+import sys
+sys.path.insert(0, '.')
+
+print("=" * 80)
+print("🎉🎉🎉 规则库扩展5.0 - Batch 05合并")
+print("=" * 80)
+
+# 1. 读取现有规则库
+print("\n📖 读取现有规则库...")
+from star_combination_rules import STAR_COMBINATION_RULES, RuleCategory, EffectLevel
+
+original_rules = STAR_COMBINATION_RULES.copy()
+print(f"✅ 原有规则: {len(original_rules)}条")
+
+# 2. 读取Batch 05规则
+print("\n📖 读取Batch 05规则...")
+with open('batch_05_rules_clean.json', 'r', encoding='utf-8') as f:
+    batch_05_rules = json.load(f)
+
+print(f"✅ Batch 05: {len(batch_05_rules)}条 (SC903-SC1203)")
+
+# 3. Category和EffectLevel映射
+category_map = {
+    '性格特质': 'RuleCategory.PERSONALITY',
+    '事业财运': 'RuleCategory.CAREER',
+    '婚姻感情': 'RuleCategory.RELATIONSHIP',
+    '健康状况': 'RuleCategory.HEALTH',
+    '运势走向': 'RuleCategory.FORTUNE',
+    '才华能力': 'RuleCategory.TALENT',
+    '人际交往': 'RuleCategory.SOCIAL'
+}
+
+effect_map = {
+    '极强': 'EffectLevel.VERY_STRONG',
+    '强': 'EffectLevel.STRONG',
+    '中': 'EffectLevel.MEDIUM',
+    '弱': 'EffectLevel.WEAK',
+    '微弱': 'EffectLevel.VERY_WEAK'
+}
+
+def format_value(v):
+    """格式化值，处理None和字符串"""
+    if v is None:
+        return 'None'
+    elif isinstance(v, str):
+        # 转义字符串中的引号和反斜杠
+        v = v.replace('\\', '\\\\').replace("'", "\\'")
+        return f"'{v}'"
+    elif isinstance(v, list):
+        if not v:
+            return '[]'
+        # 格式化列表中的每个元素
+        formatted_items = [format_value(item) for item in v]
+        return '[' + ', '.join(formatted_items) + ']'
+    elif isinstance(v, dict):
+        if not v:
+            return '{}'
+        # 格式化字典
+        items = [f"'{k}': {format_value(v)}" for k, v in v.items()]
+        return '{' + ', '.join(items) + '}'
+    else:
+        return str(v)
+
+def format_rule_for_file(rule, is_json=False):
+    """格式化规则为Python文件格式"""
+    if is_json:
+        # 从JSON格式转换
+        category = category_map.get(rule.get('category', '运势走向'), 'RuleCategory.FORTUNE')
+        effect = effect_map.get(rule.get('effect_level', '中'), 'EffectLevel.MEDIUM')
+
+        conditions = rule.get('conditions', {})
+        interpretations = rule.get('interpretations', {})
+        age_range = rule.get('age_range', [20, 70])
+
+    else:
+        # 从现有规则格式转换
+        category_val = rule['category'].value if hasattr(rule['category'], 'value') else rule['category']
+        category = category_map.get(category_val, 'RuleCategory.FORTUNE')
+
+        effect_val = rule['effect_level'].value if hasattr(rule['effect_level'], 'value') else rule['effect_level']
+        effect = effect_map.get(effect_val, 'EffectLevel.MEDIUM')
+
+        conditions = rule.get('conditions', {})
+        interpretations = rule.get('interpretations', {})
+        age_range = rule.get('age_range', [20, 70])
+
+    # 构建规则字符串
+    rule_str = "    {\n"
+    rule_str += f"        \"id\": \"{rule['id']}\",\n"
+    rule_str += f"        \"name\": {format_value(rule.get('name', ''))},\n"
+    rule_str += f"        \"category\": {category},\n"
+    rule_str += f"        \"stars_required\": {format_value(rule.get('stars_required', []))},\n"
+    rule_str += f"        \"palace_types\": {format_value(rule.get('palace_types', []))},\n"
+
+    # Conditions
+    rule_str += f"        \"conditions\": {{\n"
+    rule_str += f"            'same_palace': {conditions.get('same_palace', False)},\n"
+    rule_str += f"            'transformations': {format_value(conditions.get('transformations'))},\n"
+    rule_str += f"            'brightness': {format_value(conditions.get('brightness'))},\n"
+    rule_str += f"            'palace_position': {format_value(conditions.get('palace_position'))}\n"
+    rule_str += f"        }},\n"
+
+    # Interpretations
+    rule_str += f"        \"interpretations\": {{\n"
+    rule_str += f"            'core_meaning': {format_value(interpretations.get('core_meaning', ''))},\n"
+    rule_str += f"            'personality': {format_value(interpretations.get('personality', []))},\n"
+    rule_str += f"            'career': {format_value(interpretations.get('career', []))},\n"
+    rule_str += f"            'relationship': {format_value(interpretations.get('relationship', []))},\n"
+    rule_str += f"            'fortune_timing': {format_value(interpretations.get('fortune_timing'))},\n"
+    rule_str += f"            'warnings': {format_value(interpretations.get('warnings', []))}\n"
+    rule_str += f"        }},\n"
+
+    rule_str += f"        \"effect_level\": {effect},\n"
+    rule_str += f"        \"compatibility_score\": {rule.get('compatibility_score', 50)},\n"
+    rule_str += f"        \"age_range\": {format_value(age_range)},\n"
+    rule_str += f"        \"source\": {format_value(rule.get('source', '南北派实战断诀'))}\n"
+    rule_str += "    }"
+
+    return rule_str
+
+# 4. 格式化所有规则
+print("\n📝 格式化规则...")
+formatted_rules = []
+
+# 原有规则
+for rule in original_rules:
+    formatted_rules.append(format_rule_for_file(rule, is_json=False))
+
+# 新规则
+for rule in batch_05_rules:
+    formatted_rules.append(format_rule_for_file(rule, is_json=True))
+
+rules_code = ',\n'.join(formatted_rules)
+
+# 5. 生成新文件
+print("\n📝 生成新的star_combination_rules.py...")
+
+file_content = f'''"""
+星曜组合断语库 - Star Combination Interpretation Rules
+从138页《紫微斗数南北派断诀》提取的916条星曜组合规则。
+
+Version: 5.0 (规则库扩展5.0：100条 → 916条！🎉🎉🎉🎉)
+Update History:
+  - v1.0 (2025-12-08): 初始100条规则（手动精选）
+  - v2.0 (2025-12-11): 新增120条规则（Google Notebook半自动提取）SC301-SC420
+  - v3.0 (2025-12-12): 新增100条规则（Batch 03）SC501-SC600
+  - v4.0 (2025-12-13): 新增292条规则（Batch 04）SC601-SC892
+  - v5.0 (2025-12-13): 新增294条规则（Batch 05）SC903-SC1203
+
+Last Updated: 2025-12-13
+"""
+
+from typing import Dict, List, Any, Optional
+from enum import Enum
+
+class RuleCategory(Enum):
+    """规则分类"""
+    PERSONALITY = "性格特质"
+    CAREER = "事业财运"
+    RELATIONSHIP = "婚姻感情"
+    HEALTH = "健康状况"
+    FORTUNE = "运势走向"
+    TALENT = "才华能力"
+    SOCIAL = "人际交往"
+
+class EffectLevel(Enum):
+    """影响程度"""
+    VERY_STRONG = "极强"
+    STRONG = "强"
+    MEDIUM = "中"
+    WEAK = "弱"
+    VERY_WEAK = "微弱"
+
+STAR_COMBINATION_RULES = [
+{rules_code}
+]
+
+def get_rule_by_id(rule_id: str) -> Optional[Dict[str, Any]]:
+    """根据ID获取规则"""
+    for rule in STAR_COMBINATION_RULES:
+        if rule['id'] == rule_id:
+            return rule
+    return None
+
+def get_rules_by_category(category: RuleCategory) -> List[Dict[str, Any]]:
+    """根据分类获取规则"""
+    return [rule for rule in STAR_COMBINATION_RULES if rule['category'] == category]
+
+def get_rules_by_stars(stars: List[str]) -> List[Dict[str, Any]]:
+    """根据星曜名称搜索规则"""
+    matched_rules = []
+    for rule in STAR_COMBINATION_RULES:
+        required_stars = set(rule['stars_required'])
+        input_stars = set(stars)
+        if required_stars.issubset(input_stars):
+            matched_rules.append(rule)
+    return matched_rules
+
+def get_all_rule_ids() -> List[str]:
+    """获取所有规则ID"""
+    return [rule['id'] for rule in STAR_COMBINATION_RULES]
+
+def get_rules_count() -> int:
+    """获取规则总数"""
+    return len(STAR_COMBINATION_RULES)
+
+def get_rules_by_palace(palace_type: str) -> List[Dict[str, Any]]:
+    """根据宫位类型获取规则"""
+    return [rule for rule in STAR_COMBINATION_RULES if palace_type in rule['palace_types']]
+'''
+
+# 6. 保存文件
+with open('star_combination_rules.py', 'w', encoding='utf-8') as f:
+    f.write(file_content)
+
+total_rules = len(original_rules) + len(batch_05_rules)
+print(f"✅ 生成完成！")
+print(f"   文件大小: {len(file_content) // 1024}KB")
+print(f"   规则总数: {total_rules}条")
+
+print("\n" + "=" * 80)
+print("🎉🎉🎉 规则库扩展5.0完成！")
+print("=" * 80)
+print(f"  v1.0 (2025-12-08): 100条规则")
+print(f"  v2.0 (2025-12-11): 220条规则 (+120)")
+print(f"  v3.0 (2025-12-12): 320条规则 (+100)")
+print(f"  v4.0 (2025-12-13): 622条规则 (+302)")
+print(f"  v5.0 (2025-12-13): 916条规则 (+294) ✨✨✨")
+print()
+print(f"  📈 总增长率: +816% (相比最初100条)")
+print(f"  📈 本次增长: +47.3% (相比v4.0的622条)")
+print(f"  🎯 新增ID范围: SC903-SC1203")
+print("=" * 80)
+
+print("\n✅ v5.0 规则库准备完成！")
+print("📝 下一步: 测试导入，然后部署到HuggingFace")