modify scripts

tools/puzzle.py

@@ -3,6 +3,8 @@ import json
 import random
 import time
 import logging
+import argparse
+from collections import defaultdict
 from pathlib import Path
 from openai import AzureOpenAI
 
@@ -158,6 +160,91 @@ def process_folder(folder):
 
         time.sleep(2)  # 每批之间暂停
 
+# redo逻辑
+def redo_results():
+    files = sorted(Path(RESULT_DIR).glob('*.json'))
+    for f in files:
+        if 'matched' in f.name:
+            continue
 
-if __name__ == "__main__":
-    process_folder(WORDS_DIR)
+        logging.info(f"Redoing {f}")
+        try:
+            with open(f, 'r', encoding='utf-8') as fp:
+                data = json.load(fp)
+            words = data.get("req")
+            if not words:
+                logging.warning(f"No req in {f}")
+                continue
+
+            prompt = build_prompt(words)
+            resp_text = call_openai_with_retry(prompt)
+            if resp_text is None:
+                logging.warning(f"Failed to get response: {f}")
+                continue
+
+            try:
+                resp_json = json.loads(resp_text)
+                if len(words) == len(resp_json):
+                    logging.info(f"get correct response. rewrite file. {f}")
+                    f.unlink()
+                    save_result(int(f.name[:5]), words, resp_json, True)
+                else:
+                    logging.warning(f"response not complete: {f}, req len: {len(words)}, rsp len: {len(resp_json)}")
+            except json.JSONDecodeError:
+                logging.warning(f"response is not valid JSON: {f}")
+
+            time.sleep(2)  # 每批之间暂停
+
+        except Exception as e:
+            logging.error(f"Error processing {f}: {e}")
+
+
+# 检测是否无重复字母
+def has_no_repeated_letters(word):
+    return len(set(word)) == len(word)
+
+def generate_wordlist():
+    """
+    从 RESULT_DIR 下的 matched 文件中提取无重复字母的单词，并按 f 分类写入 words_{f}.txt
+    """
+    word_map = defaultdict(list)
+
+    for file in sorted(os.scandir(RESULT_DIR), key=lambda f: int(f.name[:5])):
+        if 'matched' not in file.name:
+            continue
+
+        with open(file.path, 'r', encoding='utf-8') as f:
+            data = json.load(f)
+
+        rsp = data.get('rsp', [])
+        for item in rsp:
+            word = item.get('w')
+            freq = item.get('f')
+            if word and freq and has_no_repeated_letters(word):
+                word_map[freq].append(word)
+
+    # 写入文件
+    for freq, words in word_map.items():
+        filename = os.path.join(RESULT_DIR, f'words_{freq}.txt')
+        with open(filename, 'w', encoding='utf-8') as f:
+            for word in words:
+                f.write(word + '\n')
+        logging.info(f'✅ 写入完成: {filename} ({len(words)} 个单词)')
+
+
+def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument('cmd', help='执行的命令: init / redo / gen')
+    args = parser.parse_args()
+
+    if args.cmd == 'init':
+        process_folder(WORDS_DIR)
+    elif args.cmd == 'redo':
+        redo_results()
+    elif args.cmd == 'gen':
+        generate_wordlist()
+    else:
+        print("❌ 未知命令，请使用: all / redo / gen")
+
+if __name__ == '__main__':
+    main()