text-corrector 0.1.0__py3-none-any.whl

text_corrector/__init__.py

@@ -0,0 +1,4 @@
+from .core import correct_text, tokenize, get_candidates
+
+__version__ = "0.1.0"
+__all__ = ["correct_text", "tokenize", "get_candidates"]

text_corrector/core.py

@@ -0,0 +1,105 @@
+import re
+from typing import List, Dict, Set, Tuple
+
+def tokenize(text: str) -> List[str]:
+    """Поддержка русских и английских букв"""
+    return re.findall(r'[a-zA-Zа-яА-ЯёЁ]+', text)
+
+
+def detect_suspects(words: List[str], dictionary_set: Set[str]) -> Tuple[List[int], List[str]]:
+    """Находит слова, которых нет в словаре"""
+    suspect_indices = []
+    suspect_words = []
+    for i, word in enumerate(words):
+        if word.lower() not in dictionary_set:
+            suspect_indices.append(i)
+            suspect_words.append(word)
+    return suspect_indices, suspect_words
+
+
+def levenshtein(a: str, b: str, max_dist: int = 2) -> int:
+    """Вычисляет расстояние Левенштейна с ранним выходом"""
+    a = a.lower()
+    b = b.lower()
+
+    len_a, len_b = len(a), len(b)
+
+    if abs(len_a - len_b) > max_dist:
+        return max_dist + 1
+
+    prev_row = list(range(len_b + 1))
+
+    for i in range(1, len_a + 1):
+        curr_row = [i]
+        min_in_row = i
+
+        for j in range(1, len_b + 1):
+            if a[i - 1] == b[j - 1]:
+                cost = 0
+            else:
+                cost = 1
+
+            insert = prev_row[j] + 1
+            delete = curr_row[j - 1] + 1
+            replace = prev_row[j - 1] + cost
+
+            curr_row.append(min(insert, delete, replace))
+            min_in_row = min(min_in_row, curr_row[-1])
+
+        if min_in_row > max_dist:
+            return max_dist + 1
+
+        prev_row = curr_row
+
+    return prev_row[len_b]
+
+
+def get_candidates(word: str, dictionary: Dict[str, float], max_dist: int = 2) -> List[Tuple[str, int, float]]:
+    """Ищет в словаре слова, похожие на заданное"""
+    word_lower = word.lower()
+    candidates = []
+
+    for dict_word, freq in dictionary.items():
+        if abs(len(word_lower) - len(dict_word)) > max_dist:
+            continue
+
+        dist = levenshtein(word_lower, dict_word, max_dist)
+        if dist <= max_dist:
+            candidates.append((dict_word, dist, freq))
+
+    return candidates
+
+
+def rank_candidates(candidates: List[Tuple[str, int, float]]) -> List[str]:
+    """Сортирует кандидатов: по расстоянию, затем по частоте"""
+    sorted_candidates = sorted(candidates, key=lambda x: (x[1], -x[2], x[0]))
+    return [c[0] for c in sorted_candidates]
+
+
+def correct_text(text: str, dictionary: Dict[str, float], max_dist: int = 2) -> str:
+    """Главная функция: возвращает исправленный текст"""
+    # Токенизация
+    words = tokenize(text)
+
+    # Детекция
+    dictionary_set = set(dictionary.keys())
+    indices, suspects = detect_suspects(words, dictionary_set)
+
+    # Поиск и замена
+    corrections = {}
+    for idx, suspect in zip(indices, suspects):
+        candidates = get_candidates(suspect, dictionary, max_dist)
+        ranked = rank_candidates(candidates)
+
+        if ranked:
+            best = ranked[0]
+            if suspect and suspect[0].isupper():
+                best = best.capitalize()
+            corrections[idx] = best
+
+    # Сборка результата
+    result_words = words.copy()
+    for idx, corrected in corrections.items():
+        result_words[idx] = corrected
+
+    return ' '.join(result_words)

text_corrector-0.1.0.dist-info/METADATA

@@ -0,0 +1,9 @@
+Metadata-Version: 2.4
+Name: text-corrector
+Version: 0.1.0
+Summary: Простая библиотека на основе расстояния Левенштейна для исправления опечаток
+License: MIT
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Requires-Python: >=3.8

text_corrector-0.1.0.dist-info/RECORD

@@ -0,0 +1,5 @@
+text_corrector/__init__.py,sha256=LweYlRH6XHOF4igHhZdwn41gRiCphmXgvIPgeMrmx9c,139
+text_corrector/core.py,sha256=aLBNXEbAN8VrAxnAHcBRgqySI_j97bmgzAgfRNzi8VM,3497
+text_corrector-0.1.0.dist-info/METADATA,sha256=-l4DrnPPrg6OUZxB7_SmWxFnupNenJWJgtkEMaAqJ90,394
+text_corrector-0.1.0.dist-info/WHEEL,sha256=mffPy8wBnZQn2VnJUU5jE99KsxaSfiyMHV9Yt0aLVxs,87
+text_corrector-0.1.0.dist-info/RECORD,,

text_corrector-0.1.0.dist-info/WHEEL

@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: hatchling 1.30.1
+Root-Is-Purelib: true
+Tag: py3-none-any