73 lines
2.4 KiB
Python
73 lines
2.4 KiB
Python
import asyncio
|
|
import tempfile
|
|
import unittest
|
|
|
|
from whoosh import index
|
|
from whoosh.fields import Schema, TEXT, ID
|
|
|
|
from app.services.search_service import SearchService, ChineseAnalyzer
|
|
|
|
|
|
class SearchServiceTest(unittest.TestCase):
|
|
def _build_service(self):
|
|
tmp = tempfile.mkdtemp()
|
|
schema = Schema(
|
|
project_id=ID(stored=True),
|
|
path=ID(unique=True, stored=True),
|
|
title=TEXT(stored=True, analyzer=ChineseAnalyzer()),
|
|
content=TEXT(stored=True, analyzer=ChineseAnalyzer()),
|
|
)
|
|
ix = index.create_in(tmp, schema)
|
|
writer = ix.writer()
|
|
writer.add_document(
|
|
project_id="1",
|
|
path="1:标签表.md",
|
|
title="标签表.md",
|
|
content="本文档介绍标签表的使用方法。",
|
|
)
|
|
writer.add_document(
|
|
project_id="1",
|
|
path="1:标签分类.md",
|
|
title="标签分类.md",
|
|
content="文档按标签进行分类，分类表存储记录。",
|
|
)
|
|
writer.add_document(
|
|
project_id="1",
|
|
path="1:标签说明.md",
|
|
title="标签说明.md",
|
|
content="标签用于内容分类。",
|
|
)
|
|
writer.add_document(
|
|
project_id="1",
|
|
path="1:配置表.md",
|
|
title="配置表.md",
|
|
content="该表用于系统配置。",
|
|
)
|
|
writer.commit()
|
|
|
|
service = SearchService()
|
|
service.ix = ix
|
|
return service
|
|
|
|
def test_phrase_keyword_matches_only_exact_content(self):
|
|
"""搜索「标签表」只应命中原文确实包含该短语的文件，
|
|
而不是「标签」+「表」分散出现在不同位置的文件。"""
|
|
service = self._build_service()
|
|
result = asyncio.run(service.search("标签表", "1"))
|
|
self.assertEqual([item["path"] for item in result], ["标签表.md"])
|
|
|
|
def test_single_token_search_returns_all_matches(self):
|
|
service = self._build_service()
|
|
result = asyncio.run(service.search("标签", "1"))
|
|
paths = {item["path"] for item in result}
|
|
self.assertEqual(paths, {"标签表.md", "标签分类.md", "标签说明.md"})
|
|
|
|
def test_no_match_returns_empty(self):
|
|
service = self._build_service()
|
|
result = asyncio.run(service.search("接口文档", "1"))
|
|
self.assertEqual(result, [])
|
|
|
|
|
|
if __name__ == "__main__":
|
|
unittest.main()
|