签约方向全本重写：前5章逐章重写(强化开篇钩子/差异化节奏/去AI模板感)+6-85章批量修复(段落不规则化/去独立行——/去敏感词/章末钩子补全) - 质量检查S级

quality_check.py

@@ -0,0 +1,100 @@
+# -*- coding: utf-8 -*-
+import re, os, glob
+
+CHAPTER_DIR = "."
+files = glob.glob(os.path.join(CHAPTER_DIR, "第*.md"))
+
+def check():
+    ai_indicators = {
+        "握紧": 0, "攥紧": 0, "燃点": 0,
+        "深吸一口气": 0, "独立行": 0,
+        "全书完标注": 0, "章末标注": 0
+    }
+    sensitive = {"公司": 0, "集团": 0, "企业": 0, "裁员": 0, "辞退": 0}
+    passive = {"被替代": 0, "被代替": 0, "被淘汰": 0, "被替换": 0}
+    chapter_counts = {}
+    
+    total_hz = 0
+    total_ch = 0
+    
+    for f in sorted(files):
+        bn = os.path.basename(f)
+        if "番茄版" in bn or bn in ("batch_fix.py", "quality_check.py"):
+            continue
+        text = open(f, encoding='utf-8').read()
+        hz = len(re.findall(r'[\u4e00-\u9fff]', text))
+        chapter_counts[bn] = hz
+        total_hz += hz
+        total_ch += 1
+        
+        for line in text.split('\n'):
+            s = line.strip()
+            if s in ('——', '————', '——————', '——————'):
+                ai_indicators["独立行"] += 1
+        
+        if "（全书完）" in text or "(全书完)" in text:
+            ai_indicators["全书完标注"] += 1
+        
+        ms = re.findall(r'\(第.*章.*完\)', text)
+        if ms:
+            ai_indicators["章末标注"] += len(ms)
+        
+        for w in ("握紧", "攥紧", "燃点", "深吸一口气"):
+            cnt = text.count(w)
+            if cnt > 0:
+                ai_indicators[w] += cnt
+        
+        for w in sensitive:
+            sensitive[w] += text.count(w)
+        for w in passive:
+            passive[w] += text.count(w)
+    
+    
+    # print report
+    print("=" * 60)
+    print("【最终质量检查报告】")
+    print("=" * 60)
+    print(f"\n检查章节数: {total_ch}")
+    print(f"全本总汉字: ~{total_hz}")
+    print(f"章均汉字: ~{total_hz // total_ch}")
+    
+    below = {k: v for k, v in chapter_counts.items() if v < 2500}
+    if below:
+        print(f"\n⚠️ 字数不足 (<2500): {len(below)}章")
+        for k, v in sorted(below.items()):
+            print(f"  {k}: {v}字")
+    else:
+        print(f"\n✅ 汉字字数: 全部达标")
+    
+    print(f"\n--- AI味检查 ---")
+    all_clean = True
+    for k, v in ai_indicators.items():
+        status = "✅" if v == 0 else "⚠️"
+        print(f"  {status} {k}: {v}次")
+        if v > 0:
+            all_clean = False
+    
+    print(f"\n--- 敏感词 ---")
+    for k, v in sensitive.items():
+        status = "✅" if v == 0 else "❌"
+        print(f"  {status} {k}: {v}次")
+        if v > 0:
+            all_clean = False
+    
+    print(f"\n--- 被动叙事 ---")
+    for k, v in passive.items():
+        status = "✅" if v == 0 else "⚠️"
+        print(f"  {status} {k}: {v}次")
+        if v > 0:
+            all_clean = False
+    
+    print()
+    print("=" * 60)
+    if all_clean:
+        print("【总体评价: S级】全部达标")
+    else:
+        print("【总体评价: 需要修复】")
+    print("=" * 60)
+
+if __name__ == "__main__":
+    check()