app/vocabulary.py

@@ -0,0 +1,196 @@
+import re
+import pickle
+import os
+from typing import Dict, List, Tuple, Union
+from collections import Counter
+
+# 预编译正则表达式提高性能
+WORD_PATTERN = re.compile(r'\b[\w-]+\b')
+
+
+class VocabularyLevelEstimator:
+    """
+    词汇难度评估基类
+    使用预定义的单词-考试类型数据评估单词难度级别
+
+    类属性:
+    _test_raw: 原始单词-考试类型数据
+    _difficulty_dict: 转换后的单词-难度级别映射
+    """
+    _test_raw = None
+    _difficulty_dict = None
+    PICKLE_PATH = 'static/words_and_tests.p'  # 默认数据文件路径
+
+    @classmethod
+    def _load_data(cls):
+        """延迟加载数据，避免不必要的文件操作"""
+        if cls._test_raw is None:
+            cls._test_raw = cls.load_record(cls.PICKLE_PATH)
+            cls._difficulty_dict = cls.convert_test_type_to_difficulty_level(cls._test_raw)
+
+    @staticmethod
+    def load_record(pickle_fname: str) -> Dict[str, List[str]]:
+        """
+        加载pickle格式的单词-考试类型数据
+
+        参数:
+        pickle_fname: pickle文件名
+
+        返回:
+        字典格式的单词到考试类型列表的映射
+
+        异常:
+        FileNotFoundError: 当文件不存在时抛出
+        ValueError: 当pickle文件损坏时抛出
+        """
+        try:
+            # 文件校验
+            if not os.path.exists(pickle_fname):
+                raise FileNotFoundError(f"词汇数据文件 {pickle_fname} 未找到")
+            if not pickle_fname.endswith('.p'):
+                raise ValueError("仅支持.pickle格式文件")
+
+            with open(pickle_fname, 'rb') as f:
+                return pickle.load(f)
+        except pickle.PickleError as e:
+            raise ValueError(f"Pickle文件 {pickle_fname} 损坏: {str(e)}")
+
+    @staticmethod
+    def convert_test_type_to_difficulty_level(d: Dict[str, List[str]]) -> Dict[str, int]:
+        """
+        将考试类型映射为难度级别
+
+        难度级别定义:
+        0: 未知/未分类
+        4: CET4
+        5: OXFORD3000
+        6: CET6或GRADUATE
+        7: IELTS或OXFORD5000
+        8: BBC
+
+        参数:
+        d: 单词到考试类型列表的映射
+
+        返回:
+        单词到难度级别的映射
+        """
+        result = {}
+        for word, test_types in d.items():
+            word_lower = word.lower()  # 统一小写处理
+            if 'CET4' in test_types:
+                result[word_lower] = 4
+            elif 'OXFORD3000' in test_types:
+                result[word_lower] = 5
+            elif 'CET6' in test_types or 'GRADUATE' in test_types:
+                result[word_lower] = 6
+            elif 'IELTS' in test_types or 'OXFORD5000' in test_types:
+                result[word_lower] = 7
+            elif 'BBC' in test_types:
+                result[word_lower] = 8
+            else:
+                result[word_lower] = 0
+        return result
+
+    @classmethod
+    def get_word_level(cls, word: str) -> int:
+        """
+        获取单词难度级别
+
+        参数:
+        word: 要查询的单词
+
+        返回:
+        单词的难度级别(0-8)
+        """
+        cls._load_data()
+        return cls._difficulty_dict.get(word.lower(), 0)
+
+    @classmethod
+    def reload_data(cls, new_path=None):
+        """强制重新加载词汇数据"""
+        if new_path:
+            cls.PICKLE_PATH = new_path
+        cls._test_raw = None
+        cls._difficulty_dict = None
+        cls._load_data()
+
+
+class UserVocabularyLevel(VocabularyLevelEstimator):
+    """用户词汇水平评估"""
+
+    def __init__(self, user_data: Dict[str, List[int]]):
+        """
+        初始化用户词汇数据
+
+        参数:
+        user_data: 单词到时间戳列表的映射
+        """
+        # 获取每个单词的最新查询时间并排序
+        word_time = [(word, max(times)) for word, times in user_data.items() if times]
+        sorted_words = sorted(word_time, key=lambda x: x[1], reverse=True)
+        self.recent_words = [word for word, _ in sorted_words[:3]]  # 取最近3个单词
+
+    @property
+    def level(self) -> float:
+        """
+        计算用户词汇水平
+
+        返回:
+        最近查询的有效单词的平均难度级别
+        如果没有有效单词则返回0
+        """
+        levels = [self.get_word_level(word) for word in self.recent_words]
+        valid_levels = [lvl for lvl in levels if lvl > 0]
+        return sum(valid_levels) / len(valid_levels) if valid_levels else 0
+
+
+class ArticleVocabularyLevel(VocabularyLevelEstimator):
+    """文章词汇水平评估"""
+
+    def __init__(self, content: str):
+        """
+        初始化文章内容
+
+        参数:
+        content: 文章内容字符串
+        """
+        if not content or not isinstance(content, str):
+            self.top_levels = []
+            return
+
+        # 文本预处理：转换为小写并提取单词
+        words = WORD_PATTERN.findall(content.lower())
+
+        # 计算单词难度并筛选有效值
+        word_levels = [self.get_word_level(word) for word in words]
+        valid_levels = sorted([lvl for lvl in word_levels if lvl > 0], reverse=True)
+
+        # 取难度最高的5个单词
+        self.top_levels = valid_levels[:5] if valid_levels else []
+
+    @property
+    def level(self) -> float:
+        """
+        计算文章词汇水平
+
+        返回:
+        文章中最难5个单词的平均难度级别
+        如果没有有效单词则返回0
+        """
+        if not self.top_levels:
+            return 0
+        return sum(self.top_levels) / len(self.top_levels)
+
+    def word_frequency(self, top_n=10) -> Dict[str, int]:
+        """
+        获取文章词频统计
+
+        参数:
+        top_n: 返回的最高频单词数量
+
+        返回:
+        词频最高的top_n个单词及其频率
+        """
+        words = WORD_PATTERN.findall(self.content.lower())
+        word_freq = Counter(words)
+        return dict(word_freq.most_common(top_n))