data: validate and repair Japanese dictionaries

tests/data/skk/expected.tsv

@@ -0,0 +1,3 @@
+かんじ	漢字 幹事 感じ
+えがお	笑顔
+きごう	斜線/入り 記号;付き 普通

tests/data/skk/source.skk.utf8

@@ -0,0 +1,6 @@
+;; converter fixture; the test transcodes this file to EUC-JP
+かんじ /漢字;common/幹事/
+えがお /笑顔;face/
+かんじ /感じ/漢字;duplicate/
+きごう /斜線\/入り;escaped slash/記号\;付き;escaped semicolon/普通/
+むこう /候補 with space/(concat "式/" "候補")/[無効/候補/]/#0/

tests/skk2ktrans_test.py

@@ -0,0 +1,34 @@
+#!/usr/bin/env python3
+
+import subprocess
+import tempfile
+import unittest
+from pathlib import Path
+
+
+ROOT = Path(__file__).resolve().parents[1]
+CONVERTER = ROOT / "map" / "skk2ktrans"
+FIXTURES = Path(__file__).parent / "data" / "skk"
+
+
+class Skk2KtransTest(unittest.TestCase):
+	def test_euc_jp_conversion_is_stable_and_literal(self):
+		source = (FIXTURES / "source.skk.utf8").read_text(encoding="utf-8")
+		expected = (FIXTURES / "expected.tsv").read_bytes()
+		with tempfile.TemporaryDirectory() as directory:
+			input_path = Path(directory) / "source.skk"
+			input_path.write_bytes(source.encode("euc_jp"))
+			result = subprocess.run(
+				[CONVERTER, input_path], capture_output=True, check=False)
+		self.assertEqual(result.returncode, 0, result.stderr.decode())
+		self.assertEqual(result.stdout, expected)
+
+	def test_invalid_euc_jp_is_rejected(self):
+		result = subprocess.run(
+			[CONVERTER], input=b"\xff\xff\n", capture_output=True, check=False)
+		self.assertNotEqual(result.returncode, 0)
+		self.assertIn(b"invalid EUC-JP", result.stderr)
+
+
+if __name__ == "__main__":
+	unittest.main()