test: add basic test suite for fojin_bridge, skill_writer, verify_sources (P1)

31 tests across 3 modules; all HTTP calls mocked, no real API dependencies.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

tests/conftest.py

@@ -0,0 +1,9 @@
+"""Shared pytest fixtures for Buddha-skill tests."""
+
+import sys
+import os
+from pathlib import Path
+
+# Add tools/ to path so tests can import modules
+PROJECT_ROOT = Path(__file__).parent.parent
+sys.path.insert(0, str(PROJECT_ROOT / "tools"))

tests/test_fojin_bridge.py

@@ -0,0 +1,89 @@
+"""Tests for fojin_bridge.py — uses mocked HTTP, no real API calls."""
+
+import pytest
+from unittest.mock import MagicMock, patch
+from fojin_bridge import FojinBridge, FojinUnavailableError, create_bridge
+import requests
+
+
+@pytest.fixture
+def bridge():
+    return FojinBridge(mode="api", base_url="https://fojin.app")
+
+
+def test_bridge_init_defaults():
+    b = FojinBridge()
+    assert b.mode == "api"
+    assert b.base_url == "https://fojin.app"
+
+
+def test_bridge_init_strips_trailing_slash():
+    b = FojinBridge(base_url="https://fojin.app/")
+    assert b.base_url == "https://fojin.app"
+
+
+def test_search_texts_basic(bridge):
+    mock_response = MagicMock()
+    mock_response.json.return_value = {"total": 1, "results": [{"id": 1}]}
+    mock_response.raise_for_status = MagicMock()
+    with patch.object(bridge.session, "get", return_value=mock_response) as mock_get:
+        result = bridge.search_texts("般若")
+        assert result["total"] == 1
+        mock_get.assert_called_once()
+        call_args = mock_get.call_args
+        assert "q" in call_args.kwargs["params"]
+        assert call_args.kwargs["params"]["q"] == "般若"
+
+
+def test_search_texts_with_filters(bridge):
+    mock_response = MagicMock()
+    mock_response.json.return_value = {"total": 0, "results": []}
+    mock_response.raise_for_status = MagicMock()
+    with patch.object(bridge.session, "get", return_value=mock_response) as mock_get:
+        bridge.search_texts("禅", sources="cbeta", lang="lzh", page=2, size=50)
+        params = mock_get.call_args.kwargs["params"]
+        assert params["sources"] == "cbeta"
+        assert params["lang"] == "lzh"
+        assert params["page"] == 2
+        assert params["size"] == 50
+
+
+def test_get_text_content(bridge):
+    mock_response = MagicMock()
+    mock_response.json.return_value = {"content": "test content", "juan_num": 1}
+    mock_response.raise_for_status = MagicMock()
+    with patch.object(bridge.session, "get", return_value=mock_response):
+        result = bridge.get_text_content(123, 1)
+        assert result["content"] == "test content"
+
+
+def test_search_kg_entities(bridge):
+    mock_response = MagicMock()
+    mock_response.json.return_value = {"total": 1, "results": [{"id": 456, "name_zh": "玄奘"}]}
+    mock_response.raise_for_status = MagicMock()
+    with patch.object(bridge.session, "get", return_value=mock_response):
+        result = bridge.search_kg_entities("玄奘", entity_type="person")
+        assert result["results"][0]["name_zh"] == "玄奘"
+
+
+def test_fojin_unavailable_on_connection_error(bridge):
+    with patch.object(bridge.session, "get", side_effect=requests.ConnectionError("test")):
+        with pytest.raises(FojinUnavailableError):
+            bridge.search_texts("test")
+
+
+def test_fojin_unavailable_on_timeout(bridge):
+    with patch.object(bridge.session, "get", side_effect=requests.Timeout("test")):
+        with pytest.raises(FojinUnavailableError):
+            bridge.get_text(123)
+
+
+def test_test_connection_returns_false_on_failure(bridge):
+    with patch.object(bridge.session, "get", side_effect=requests.ConnectionError("test")):
+        assert bridge.test_connection() is False
+
+
+def test_create_bridge_from_env(monkeypatch):
+    monkeypatch.setenv("FOJIN_URL", "https://custom.fojin.test")
+    b = create_bridge()
+    assert b.base_url == "https://custom.fojin.test"

tests/test_skill_writer.py

@@ -0,0 +1,134 @@
+"""Tests for skill_writer.py — uses tmp_path fixture."""
+
+import json
+import os
+import pytest
+from skill_writer import slugify, create_teacher, list_teachers, update_teacher, DISCLAIMER
+
+
+def test_slugify_english():
+    # With pypinyin installed, English chars are passed through as-is (no lowercasing);
+    # without pypinyin the fallback lowercases. Either result must be alphanumeric+hyphen.
+    result = slugify("Hello World")
+    assert all(c.isalnum() or c == "-" for c in result)
+    assert len(result) > 0
+
+
+def test_slugify_chinese():
+    # Should use pypinyin if available, otherwise lowercase fallback
+    result = slugify("印光大师")
+    assert "-" in result or result.isalnum()
+    assert result.islower()
+
+
+def test_slugify_strips_punctuation():
+    # Punctuation is removed; case depends on pypinyin presence
+    result = slugify("Master!@#$")
+    assert all(c.isalnum() or c == "-" for c in result)
+    assert "master" in result.lower()
+
+
+def test_create_teacher_writes_files(tmp_path):
+    teacher_dir = create_teacher(
+        base_dir=str(tmp_path),
+        name="测试法师",
+        tradition="汉传",
+        school="测试宗",
+        era="1900-2000",
+        languages=["zh"],
+        teaching_content="# 教义\n测试教义内容",
+        voice_content="# 风格\n测试风格内容",
+    )
+    assert os.path.exists(os.path.join(teacher_dir, "teaching.md"))
+    assert os.path.exists(os.path.join(teacher_dir, "voice.md"))
+    assert os.path.exists(os.path.join(teacher_dir, "SKILL.md"))
+    assert os.path.exists(os.path.join(teacher_dir, "meta.json"))
+    assert os.path.exists(os.path.join(teacher_dir, "versions"))
+
+
+def test_create_teacher_meta_content(tmp_path):
+    teacher_dir = create_teacher(
+        base_dir=str(tmp_path),
+        name="测试法师",
+        tradition="汉传",
+        school="测试宗",
+        era="1900-2000",
+        languages=["zh"],
+        teaching_content="教义",
+        voice_content="风格",
+        sources=[{"type": "cbeta", "id": "T01n0001"}],
+    )
+    with open(os.path.join(teacher_dir, "meta.json"), encoding="utf-8") as f:
+        meta = json.load(f)
+    assert meta["name"] == "测试法师"
+    assert meta["tradition"] == "汉传"
+    assert meta["version"] == "1.0.0"
+    assert meta["disclaimer"] == DISCLAIMER
+    assert len(meta["sources"]) == 1
+
+
+def test_create_teacher_skill_md_includes_content(tmp_path):
+    teacher_dir = create_teacher(
+        base_dir=str(tmp_path),
+        name="测试法师",
+        tradition="汉传",
+        school="测试宗",
+        era="1900-2000",
+        languages=["zh"],
+        teaching_content="UNIQUE_TEACHING_MARKER",
+        voice_content="UNIQUE_VOICE_MARKER",
+    )
+    with open(os.path.join(teacher_dir, "SKILL.md"), encoding="utf-8") as f:
+        content = f.read()
+    assert "UNIQUE_TEACHING_MARKER" in content
+    assert "UNIQUE_VOICE_MARKER" in content
+    assert "master_" in content  # frontmatter
+
+
+def test_list_teachers_empty(tmp_path):
+    assert list_teachers(str(tmp_path)) == []
+
+
+def test_list_teachers_finds_created(tmp_path):
+    create_teacher(
+        base_dir=str(tmp_path),
+        name="法师一", tradition="汉传", school="宗A",
+        era="1900", languages=["zh"],
+        teaching_content="a", voice_content="b",
+    )
+    create_teacher(
+        base_dir=str(tmp_path),
+        name="法师二", tradition="汉传", school="宗B",
+        era="1950", languages=["zh"],
+        teaching_content="a", voice_content="b",
+    )
+    teachers = list_teachers(str(tmp_path))
+    assert len(teachers) == 2
+    names = {t["name"] for t in teachers}
+    assert names == {"法师一", "法师二"}
+
+
+def test_update_teacher_bumps_version(tmp_path):
+    teacher_dir = create_teacher(
+        base_dir=str(tmp_path),
+        name="测试", tradition="汉传", school="宗",
+        era="1900", languages=["zh"],
+        teaching_content="原教义", voice_content="原风格",
+    )
+    new_version = update_teacher(teacher_dir, teaching_patch="补充教义")
+    assert new_version == "1.1.0"
+    with open(os.path.join(teacher_dir, "teaching.md"), encoding="utf-8") as f:
+        content = f.read()
+    assert "原教义" in content
+    assert "补充教义" in content
+
+
+def test_update_teacher_archives_version(tmp_path):
+    teacher_dir = create_teacher(
+        base_dir=str(tmp_path),
+        name="测试", tradition="汉传", school="宗",
+        era="1900", languages=["zh"],
+        teaching_content="v1", voice_content="v1",
+    )
+    update_teacher(teacher_dir, teaching_patch="update")
+    assert os.path.exists(os.path.join(teacher_dir, "versions", "v1.0.0"))

tests/test_verify_sources.py

@@ -0,0 +1,66 @@
+"""Tests for verify_sources.py — pure logic, no API calls."""
+
+import pytest
+from verify_sources import full_to_short_cbeta, FULL_CBETA_RE, FOJIN_URL_RE
+
+
+def test_verify_sources_module_imports():
+    """Verify verify_sources.py can be imported without errors."""
+    import verify_sources
+    assert callable(getattr(verify_sources, "main", None))
+
+
+def test_full_to_short_cbeta_t_series():
+    assert full_to_short_cbeta("T08n0235") == "T0235"
+
+
+def test_full_to_short_cbeta_x_series():
+    assert full_to_short_cbeta("X62n1182") == "X1182"
+
+
+def test_full_to_short_cbeta_j_series():
+    assert full_to_short_cbeta("J36n0348") == "J0348"
+
+
+def test_full_to_short_cbeta_strips_volume_number():
+    # Volume number (middle digits) must be dropped
+    assert full_to_short_cbeta("T34n1718") == "T1718"
+    assert full_to_short_cbeta("T01n0001") == "T0001"
+
+
+def test_full_to_short_cbeta_invalid_returns_none():
+    assert full_to_short_cbeta("invalid") is None
+    assert full_to_short_cbeta("") is None
+    assert full_to_short_cbeta("123") is None
+
+
+def test_cbeta_id_format_recognition():
+    """CBETA IDs follow format like T48n2008, X62n1182, J36n0348."""
+    valid_ids = ["T48n2008", "X62n1182", "J36n0348", "T01n0001"]
+    for cbeta_id in valid_ids:
+        assert FULL_CBETA_RE.match(cbeta_id), f"{cbeta_id} should match FULL_CBETA_RE"
+
+
+def test_cbeta_id_rejects_invalid():
+    invalid_ids = ["T48", "n2008", "abc123", "t48n2008"]  # lowercase prefix is invalid
+    for cbeta_id in invalid_ids:
+        assert not FULL_CBETA_RE.match(cbeta_id), f"{cbeta_id} should not match FULL_CBETA_RE"
+
+
+def test_fojin_url_re_matches_cbeta_url():
+    line = "See https://fojin.app/texts/T08n0235 for reference"
+    m = FOJIN_URL_RE.search(line)
+    assert m is not None
+    assert m.group(2) == "T08n0235"
+
+
+def test_fojin_url_re_matches_numeric_id():
+    line = "Link: https://fojin.app/texts/12345"
+    m = FOJIN_URL_RE.search(line)
+    assert m is not None
+    assert m.group(2) == "12345"
+
+
+def test_fojin_url_re_no_match_on_unrelated_url():
+    line = "Visit https://example.com/texts/something"
+    assert FOJIN_URL_RE.search(line) is None