Initial commit

novels/round3/quality_check.py

@@ -0,0 +1,91 @@
+# -*- coding: utf-8 -*-
+import re, os
+
+D = '/app/working/workspaces/default/novels/round3/03-满级大佬回新手村'
+rr = re.compile(r'[\u4e00-\u9fff]')
+
+print('=== 1. 章节完整性 ===')
+files = sorted([f for f in os.listdir(D) if f.endswith('.md') and f.startswith('第')])
+print(f'  共 {len(files)} 章')
+for f in files:
+    print(f'    {f}')
+
+print('\n=== 2. 汉字字数 ===')
+for f in files:
+    with open(os.path.join(D, f), 'r', encoding='utf-8') as fp:
+        h = len(rr.findall(fp.read()))
+        flag = 'OK' if h >= 2000 else 'NEED'
+        print(f'  {flag} {f}: {h}')
+
+print('\n=== 3. 敏感词检查 ===')
+sensitive = [
+    ('公司负面', '公[^告]?司'),
+    ('集团', '集团'),
+    ('企业', '企业'),
+    ('裁员/辞退/被裁', '(被裁|裁员|辞退)'),
+    ('被替代/被代替', '(被替代|被代替)'),
+]
+found = 0
+for f in files:
+    with open(os.path.join(D, f), 'r', encoding='utf-8') as fp:
+        c = fp.read()
+    for name, pat in sensitive:
+        m = re.findall(pat, c)
+        if m:
+            found += 1
+            print(f'  ⚠️ {f}: {name} -> {m}')
+if found == 0:
+    print('  ✅ 无敏感词')
+
+print('\n=== 4. AI味检查 ===')
+issues = 0
+for f in files:
+    with open(os.path.join(D, f), 'r', encoding='utf-8') as fp:
+        c = fp.read()
+    
+    # 深吸
+    n1 = c.count('深吸')
+    if n1 > 2:
+        issues += 1
+        print(f'  ⚠️ {f}: 深吸{n1}次(>2)')
+    
+    # 握紧/攥紧
+    n2 = c.count('握紧') + c.count('攥紧')
+    if n2 > 4:
+        issues += 1
+        print(f'  ⚠️ {f}: 握紧/攥紧{n2}次(>4)')
+    
+    # 独立行——
+    lines = c.split('\n')
+    for i, line in enumerate(lines):
+        s = line.strip()
+        if s == '——':
+            issues += 1
+            print(f'  ⚠️ {f}: 独立行—— (行{i+1})')
+    
+    # 燃点
+    if '燃点' in c:
+        issues += 1
+        print(f'  ⚠️ {f}: 含燃点')
+    
+    # 章末标注
+    if '(第' in c and '(完)' in c:
+        issues += 1
+        print(f'  ⚠️ {f}: 含章末标注')
+
+if issues == 0:
+    print('  ✅ 无AI味问题')
+
+print('\n=== 5. 开篇核弹检查（第1章前500字） ===')
+ch1_file = files[0]
+with open(os.path.join(D, ch1_file), 'r', encoding='utf-8') as fp:
+    ch1 = fp.read()
+first_500 = ch1[:1000]
+rr_hz_500 = len(rr.findall(first_500))
+print(f'  前500字区域汉字数: {rr_hz_500}')
+print(f'  前200字: {ch1[:200]}')
+print(f'  前300-500字: {ch1[200:600]}')
+
+print('\n=== 总汉字数 ===')
+total = sum(len(rr.findall(open(os.path.join(D, f),'r',encoding='utf-8').read())) for f in files)
+print(f'  {total} 汉字，平均 {total//len(files)}/章')