定義された用語の意図しない使用方法を検出するスクリプトを追加 #1347

Author: faithandbrave

.github/workflows/defined_word_check.yml

@@ -0,0 +1,19 @@
+name: defined word check
+
+on: [push, pull_request, workflow_dispatch]
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    steps:
+    - name: Set up Python
+      uses: actions/setup-python@v5
+      with:
+        python-version: '3.x'
+    - name: Install dependencies
+      run: |
+        python -m pip install --upgrade pip
+        pip install requests
+    - uses: actions/checkout@v4
+    - name: check
+      run: python3 .github/workflows/script/defined_word_check.py

.github/workflows/script/defined_word_check.py

@@ -0,0 +1,68 @@
+# 定義された用語の意図しない使用を検出する
+
+import glob
+import os
+import sys
+import re
+
+DEFINED_WORD_LIST = [
+    # 先頭はターゲット用語、それ以外はその用語の許可された使用方法
+    ["未定義", "未定義動作", "未定義の動作", "動作は未定義", "挙動は未定義", "振る舞いは未定義"],
+]
+
+def check_defined_word(text: str, filename: str) -> bool:
+    found_error: bool = False
+
+    for words in DEFINED_WORD_LIST:
+        word = words[0]
+        allow_list = words[1:]
+
+        match_list = [m.start() for m in re.finditer(word, text)]
+        if len(match_list) == 0:
+            continue
+
+        for i in match_list:
+            ok_count = 0
+            for allow_word in allow_list:
+                index = allow_word.find(word)
+                n = len(allow_word)
+
+                if i - index < 0:
+                    continue
+
+                j = i - index
+                sliced = text[j:j+n]
+                if sliced == allow_word:
+                    ok_count += 1
+
+            if ok_count == 0:
+                start = text.rfind('\n', 0, i)
+                if start < 0:
+                    start = 0
+                else:
+                    start = start + 1
+
+                end = text.find('\n', i)
+                if end < 0:
+                    end = start + 20
+
+                around_word = text[start:end]
+                print("{}: the file includes unintended word use \"{}\". Around the word:\"{}\"".format(filename, word, around_word))
+                found_error = True
+
+    return not found_error
+
+if __name__ == '__main__':
+    found_error = False
+    for p in sorted(list(glob.glob("**/*.md", recursive=True))):
+        if p.startswith("start_editing/"):
+            continue
+
+        with open(p) as f:
+            text = f.read()
+
+        if not check_defined_word(text, p):
+            found_error = True
+
+    if found_error:
+        sys.exit(1)