Makrony i optymalizacje

checktrans

@@ -1,8 +1,109 @@
-#!/bin/bash
-#Transcription checking script:
-# Transcribes checktrans-cases.ssml, compares result with checktrans-expected-results.txt, prints found differences if any.
-pushd $(dirname $0) >/dev/null
-../../../local/bin/RHVoice-transcribe-sentences checktrans-cases.ssml /dev/stdout |
-diff -b -u0 checktrans-expected-results.txt /dev/stdin &&
-echo "All $(wc -l <checktrans-expected-results.txt) tests passed." &&
-popd >/dev/null
+#!/usr/bin/env python3
+
+import subprocess
+import sys
+from pathlib import Path
+from xml.sax.saxutils import escape
+
+
+def normalize_transcription(text: str) -> str:
+    text = text.strip()
+    if text.startswith("pau "):
+        text = text[4:]
+    if text.endswith(" pau"):
+        text = text[:-4]
+    return " ".join(text.split())
+
+
+def split_test_line(line: str) -> tuple[str, str]:
+    quote = None
+    for index, char in enumerate(line):
+        if quote is None:
+            if char in ("'", '"'):
+                quote = char
+            elif char == "=":
+                return line[:index], line[index + 1 :]
+        elif char == quote:
+            quote = None
+    if quote is not None:
+        raise ValueError("unterminated quoted string")
+    raise ValueError("missing '=' separator")
+
+
+def parse_test_field(field: str) -> str:
+    stripped = field.strip()
+    if not stripped:
+        return ""
+    if stripped[0] in ("'", '"'):
+        quote = stripped[0]
+        if len(stripped) < 2 or stripped[-1] != quote:
+            raise ValueError("unterminated quoted string")
+        inner = stripped[1:-1]
+        if quote in inner:
+            raise ValueError(f"unexpected {quote} inside quoted string")
+        return inner
+    return stripped
+
+
+def load_tests(path: Path) -> dict[str, str]:
+    tests: dict[str, str] = {}
+    for line_number, raw_line in enumerate(path.read_text(encoding="utf-8").splitlines(), start=1):
+        stripped = raw_line.strip()
+        if not stripped or stripped.startswith(("#", ";")):
+            continue
+        try:
+            raw_case, raw_expected = split_test_line(raw_line)
+            case = parse_test_field(raw_case)
+            expected = parse_test_field(raw_expected)
+        except ValueError as exc:
+            raise ValueError(f"{path}:{line_number}: {exc}") from exc
+        tests[case] = expected
+    return tests
+
+
+def main() -> int:
+    script_dir = Path(__file__).resolve().parent
+    tests_path = script_dir / "checktrans-tests.ini"
+    transcriber = script_dir / "../../../local/bin/RHVoice-transcribe-sentences"
+
+    try:
+        tests = load_tests(tests_path)
+    except ValueError as exc:
+        print(exc, file=sys.stderr)
+        return 1
+
+    ssml_lines = ['<speak xml:lang="pl">']
+    ssml_lines.extend(f"<s>{escape(case)}</s>" for case in tests)
+    ssml_lines.append("</speak>")
+    ssml_input = "\n".join(ssml_lines) + "\n"
+
+    result = subprocess.run(
+        [str(transcriber), "/dev/stdin", "/dev/stdout"],
+        cwd=script_dir,
+        input=ssml_input,
+        text=True,
+        capture_output=True,
+    )
+    if result.returncode != 0:
+        if result.stderr:
+            sys.stderr.write(result.stderr)
+        return result.returncode
+
+    expected = [normalize_transcription(tests[case]) for case in tests]
+    actual = [normalize_transcription(line) for line in result.stdout.splitlines()]
+
+    if actual == expected:
+        print(f"All {len(expected)} tests passed.")
+        return 0
+
+    cases = list(tests)
+    for case, expected_text, actual_text in zip(cases, expected, actual):
+        if expected_text != actual_text:
+            print(f"Failed: {case}")
+            print(f'Expected: "{expected_text}"')
+            print(f'Got: "{actual_text}"')
+    return 1
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())