diff --git a/tests/unit/test_pagination.py b/tests/unit/test_pagination.py index 1d7c55e30..06adbe903 100755 --- a/tests/unit/test_pagination.py +++ b/tests/unit/test_pagination.py @@ -1,9 +1,6 @@ -"""通用分页器单元测试.""" - -from __future__ import annotations +"""pagination 单元测试.""" import pytest -from pydantic import ValidationError from packages.application.common.pagination import ( PaginatedResponse, @@ -12,377 +9,171 @@ from packages.application.common.pagination import ( paginate, ) +# ── PaginationParams ──────────────────────────────────────────────────────── + class TestPaginationParams: - """PaginationParams 测试""" - def test_default_values(self): - """默认值正确""" params = PaginationParams() assert params.page == 1 assert params.page_size == 20 - def test_offset_first_page(self): - """第一页 offset 为 0""" + def test_custom_values(self): + params = PaginationParams(page=3, page_size=50) + assert params.page == 3 + assert params.page_size == 50 + + def test_offset_calculation(self): params = PaginationParams(page=1, page_size=20) assert params.offset == 0 - def test_offset_second_page(self): - """第二页 offset 计算正确""" - params = PaginationParams(page=2, page_size=20) - assert params.offset == 20 + params = PaginationParams(page=3, page_size=20) + assert params.offset == 40 - def test_offset_custom_page_size(self): - """自定义 page_size 的 offset""" - params = PaginationParams(page=3, page_size=10) - assert params.offset == 20 + params = PaginationParams(page=10, page_size=50) + assert params.offset == 450 def test_limit_equals_page_size(self): - """limit 等于 page_size""" - params = PaginationParams(page_size=50) - assert params.limit == 50 + params = PaginationParams(page_size=30) + assert params.limit == 30 def test_page_must_be_at_least_1(self): - """page 不能小于 1""" - with pytest.raises(ValidationError): + with pytest.raises(ValueError): PaginationParams(page=0) - def test_page_negative_raises(self): - """page 不能为负数""" - with pytest.raises(ValidationError): - PaginationParams(page=-1) - def test_page_size_must_be_at_least_1(self): - """page_size 不能小于 1""" - with pytest.raises(ValidationError): + with pytest.raises(ValueError): PaginationParams(page_size=0) def test_page_size_max_100(self): - """page_size 最大 100""" - with pytest.raises(ValidationError): + with pytest.raises(ValueError): PaginationParams(page_size=101) - def test_page_size_100_is_valid(self): - """page_size=100 是合法的""" - params = PaginationParams(page_size=100) - assert params.page_size == 100 + +# ── PaginationMeta ────────────────────────────────────────────────────────── class TestPaginationMeta: - """PaginationMeta 测试""" - def test_from_params_first_page(self): - """第一页元数据""" params = PaginationParams(page=1, page_size=10) meta = PaginationMeta.from_params(params, total=25) - assert meta.page == 1 assert meta.page_size == 10 assert meta.total == 25 - assert meta.total_pages == 3 + assert meta.total_pages == 3 # ceil(25/10) assert meta.has_next is True assert meta.has_prev is False def test_from_params_last_page(self): - """最后一页元数据""" params = PaginationParams(page=3, page_size=10) meta = PaginationMeta.from_params(params, total=25) - assert meta.page == 3 assert meta.total_pages == 3 assert meta.has_next is False assert meta.has_prev is True def test_from_params_middle_page(self): - """中间页元数据""" params = PaginationParams(page=2, page_size=10) - meta = PaginationMeta.from_params(params, total=50) - - assert meta.page == 2 - assert meta.total_pages == 5 + meta = PaginationMeta.from_params(params, total=25) assert meta.has_next is True assert meta.has_prev is True + def test_from_params_single_page(self): + params = PaginationParams(page=1, page_size=20) + meta = PaginationMeta.from_params(params, total=5) + assert meta.total_pages == 1 + assert meta.has_next is False + assert meta.has_prev is False + def test_from_params_zero_total(self): - """总数为 0 时""" params = PaginationParams(page=1, page_size=20) meta = PaginationMeta.from_params(params, total=0) - assert meta.total == 0 assert meta.total_pages == 0 assert meta.has_next is False assert meta.has_prev is False - def test_from_params_exact_multiple(self): - """总数刚好是 page_size 的整数倍""" + def test_from_params_exact_page_size(self): params = PaginationParams(page=1, page_size=10) - meta = PaginationMeta.from_params(params, total=30) - - assert meta.total_pages == 3 - - def test_from_params_single_page(self): - """单页即可放下所有数据""" - params = PaginationParams(page=1, page_size=100) - meta = PaginationMeta.from_params(params, total=50) - + meta = PaginationMeta.from_params(params, total=10) assert meta.total_pages == 1 - assert meta.has_next is False - assert meta.has_prev is False + + def test_from_params_one_extra(self): + params = PaginationParams(page=1, page_size=10) + meta = PaginationMeta.from_params(params, total=11) + assert meta.total_pages == 2 + + +# ── PaginatedResponse ─────────────────────────────────────────────────────── class TestPaginatedResponse: - """PaginatedResponse 测试""" - - def test_create_success(self): - """创建分页响应""" - params = PaginationParams(page=1, page_size=10) - data = [1, 2, 3] - - response = PaginatedResponse.create(data, params, total=25) - - assert response.data == [1, 2, 3] + def test_create_response(self): + params = PaginationParams(page=1, page_size=5) + data = [1, 2, 3, 4, 5] + response = PaginatedResponse.create(data, params, total=15) + assert response.data == data assert response.pagination.page == 1 - assert response.pagination.total == 25 + assert response.pagination.total == 15 assert response.pagination.total_pages == 3 - def test_create_empty_data(self): - """空数据分页响应""" - params = PaginationParams(page=1, page_size=20) - response = PaginatedResponse.create([], params, total=0) - assert response.data == [] - assert response.pagination.total == 0 - assert response.pagination.total_pages == 0 +# ── paginate function ─────────────────────────────────────────────────────── -class TestPaginateFunction: - """paginate 函数测试（内存分页）""" - +class TestPaginate: def test_first_page(self): - """第一页分页""" items = list(range(30)) params = PaginationParams(page=1, page_size=10) - result = paginate(items, params) - assert result.data == list(range(10)) assert result.pagination.total == 30 assert result.pagination.total_pages == 3 assert result.pagination.has_next is True assert result.pagination.has_prev is False - def test_second_page(self): - """第二页分页""" - items = list(range(30)) - params = PaginationParams(page=2, page_size=10) - - result = paginate(items, params) - - assert result.data == list(range(10, 20)) - assert result.pagination.page == 2 - def test_last_page(self): - """最后一页分页""" items = list(range(25)) params = PaginationParams(page=3, page_size=10) - result = paginate(items, params) - assert result.data == list(range(20, 25)) assert len(result.data) == 5 assert result.pagination.has_next is False + assert result.pagination.has_prev is True - def test_empty_list(self): - """空列表分页""" - params = PaginationParams(page=1, page_size=20) - result = paginate([], params) - - assert result.data == [] - assert result.pagination.total == 0 - assert result.pagination.total_pages == 0 - - def test_page_beyond_total(self): - """页码超出总数""" + def test_single_page(self): items = list(range(5)) - params = PaginationParams(page=10, page_size=10) - - result = paginate(items, params) - - assert result.data == [] - assert result.pagination.total == 5 - assert result.pagination.total_pages == 1 - - def test_custom_page_size(self): - """自定义每页数量""" - items = list(range(100)) - params = PaginationParams(page=1, page_size=50) - - result = paginate(items, params) - - assert len(result.data) == 50 - assert result.pagination.total_pages == 2 - - def test_single_item(self): - """单条数据""" - items = ["only_one"] params = PaginationParams(page=1, page_size=10) - result = paginate(items, params) - - assert result.data == ["only_one"] - assert result.pagination.total == 1 - assert result.pagination.total_pages == 1 - - def test_generic_type_preserved(self): - """泛型类型数据正确""" - items = [{"id": 1, "name": "a"}, {"id": 2, "name": "b"}] - params = PaginationParams(page=1, page_size=10) - - result = paginate(items, params) - - assert len(result.data) == 2 - assert result.data[0]["id"] == 1 - - -# ── PaginationParams 补充边界 ─────────────────────────────────────────────── - - -class TestPaginationParamsEdgeCases: - """PaginationParams 补充边界场景.""" - - def test_page_size_1_minimum(self): - """page_size=1 是允许的最小值.""" - params = PaginationParams(page_size=1) - assert params.page_size == 1 - assert params.limit == 1 - - def test_page_size_100_maximum(self): - """page_size=100 是允许的最大值.""" - params = PaginationParams(page_size=100) - assert params.page_size == 100 - - def test_offset_page_1_size_100(self): - """第1页每页100条 offset=0.""" - params = PaginationParams(page=1, page_size=100) - assert params.offset == 0 - - def test_offset_page_100_size_100(self): - """第100页每页100条 offset=9900.""" - params = PaginationParams(page=100, page_size=100) - assert params.offset == 9900 - - def test_large_page_number_accepted(self): - """极大页码（超过实际页数）允许.""" - params = PaginationParams(page=999999, page_size=20) - assert params.page == 999999 - assert params.offset == (999999 - 1) * 20 - - -# ── PaginationMeta 补充边界 ───────────────────────────────────────────────── - - -class TestPaginationMetaEdgeCases: - """PaginationMeta 补充边界场景.""" - - def test_total_0_page_1(self): - """total=0, page=1 时 total_pages=0, 无上下页.""" - params = PaginationParams(page=1, page_size=20) - meta = PaginationMeta.from_params(params, total=0) - assert meta.total_pages == 0 - assert meta.has_next is False - assert meta.has_prev is False - - def test_total_0_page_beyond(self): - """total=0, page>1 时 has_prev=True（因为page>1）.""" - params = PaginationParams(page=3, page_size=20) - meta = PaginationMeta.from_params(params, total=0) - assert meta.total_pages == 0 - assert meta.has_next is False - assert meta.has_prev is True - - def test_exact_last_page(self): - """刚好是最后一页时 has_next=False.""" - params = PaginationParams(page=5, page_size=10) - meta = PaginationMeta.from_params(params, total=50) - assert meta.total_pages == 5 - assert meta.has_next is False - assert meta.has_prev is True - - def test_one_more_than_exact(self): - """比整数页多1条时总页数+1.""" - params = PaginationParams(page=1, page_size=10) - meta = PaginationMeta.from_params(params, total=51) - assert meta.total_pages == 6 - - def test_page_exactly_total_pages(self): - """page == total_pages 时 has_next=False.""" - params = PaginationParams(page=3, page_size=10) - meta = PaginationMeta.from_params(params, total=30) - assert meta.has_next is False - - def test_total_1_page_1_size_1(self): - """1条数据1页.""" - params = PaginationParams(page=1, page_size=1) - meta = PaginationMeta.from_params(params, total=1) - assert meta.total_pages == 1 - assert meta.has_next is False - assert meta.has_prev is False - - -# ── paginate 补充边界 ────────────────────────────────────────────────────── - - -class TestPaginateEdgeCases: - """paginate 补充边界场景.""" - - def test_single_item_list(self): - """单元素列表.""" - result = paginate([42], PaginationParams(page=1, page_size=10)) - assert result.data == [42] - assert result.pagination.total == 1 - assert result.pagination.total_pages == 1 - - def test_page_exactly_last(self): - """刚好在最后一页.""" - items = list(range(25)) - result = paginate(items, PaginationParams(page=3, page_size=10)) - assert result.data == list(range(20, 25)) - assert result.pagination.has_next is False - - def test_page_past_end_returns_empty(self): - """页码超过总数返回空.""" - items = list(range(5)) - result = paginate(items, PaginationParams(page=10, page_size=10)) - assert result.data == [] - assert result.pagination.total == 5 - - def test_empty_list_page_1(self): - """空列表第1页.""" - result = paginate([], PaginationParams(page=1, page_size=10)) - assert result.data == [] - assert result.pagination.total == 0 - assert result.pagination.total_pages == 0 - - def test_page_size_1_iterates_all(self): - """page_size=1 时每页1条.""" - items = ["a", "b", "c"] - r1 = paginate(items, PaginationParams(page=1, page_size=1)) - r2 = paginate(items, PaginationParams(page=2, page_size=1)) - r3 = paginate(items, PaginationParams(page=3, page_size=1)) - assert r1.data == ["a"] - assert r2.data == ["b"] - assert r3.data == ["c"] - - def test_does_not_mutate_input(self): - """不修改输入列表.""" - items = [1, 2, 3, 4, 5] - original = items[:] - paginate(items, PaginationParams(page=1, page_size=2)) - assert items == original - - def test_page_size_greater_than_total(self): - """每页条数大于总数.""" - items = list(range(5)) - result = paginate(items, PaginationParams(page=1, page_size=100)) assert result.data == items assert result.pagination.total_pages == 1 + + def test_empty_list(self): + items = [] + params = PaginationParams(page=1, page_size=10) + result = paginate(items, params) + assert result.data == [] + assert result.pagination.total == 0 + assert result.pagination.total_pages == 0 + + def test_page_beyond_end(self): + items = list(range(5)) + params = PaginationParams(page=10, page_size=10) + result = paginate(items, params) + assert result.data == [] + assert result.pagination.total == 5 + + def test_page_size_larger_than_items(self): + items = list(range(5)) + params = PaginationParams(page=1, page_size=100) + result = paginate(items, params) + assert result.data == items + assert result.pagination.total_pages == 1 + + def test_middle_page(self): + items = list(range(100)) + params = PaginationParams(page=5, page_size=10) + result = paginate(items, params) + assert result.data == list(range(40, 50)) + assert result.pagination.has_next is True + assert result.pagination.has_prev is True diff --git a/tests/unit/test_text_splitter.py b/tests/unit/test_text_splitter.py index 3e221601e..831253df2 100755 --- a/tests/unit/test_text_splitter.py +++ b/tests/unit/test_text_splitter.py @@ -1,412 +1,100 @@ -"""文本分段工具单元测试.""" - -from __future__ import annotations - -import pytest +"""text_splitter 单元测试.""" from packages.application.tts_job.text_splitter import split_text class TestSplitText: - """split_text 函数测试""" - - def test_empty_string_returns_empty_list(self): - """空字符串返回空列表""" + def test_empty_text_returns_empty(self): assert split_text("") == [] - def test_whitespace_only_returns_empty_list(self): - """纯空白字符返回空列表""" - assert split_text(" \n \t ") == [] - - def test_short_text_returns_single_segment(self): - """短文本直接返回单段""" - text = "这是一段短文本。" - result = split_text(text, max_chars=500) - assert result == [text] - - def test_text_length_equals_max_chars(self): - """文本长度恰好等于 max_chars 时返回单段""" - text = "a" * 100 - result = split_text(text, max_chars=100) - assert len(result) == 1 - assert len(result[0]) == 100 - - def test_splits_on_sentence_boundary(self): - """在句子边界处分段""" - # 构造长文本，确保超过 max_chars - sentences = ["今天天气真好。我们一起去公园散步吧。", "公园里有很多花。还有很多小朋友在玩耍。"] * 10 - text = "".join(sentences) - - result = split_text(text, max_chars=200) - - assert len(result) >= 2 - # 每段都不超过 max_chars - for seg in result: - assert len(seg) <= 200 - - def test_all_segments_within_max_chars(self): - """所有分段都不超过 max_chars""" - text = "这是第一句话。这是第二句话。这是第三句话。这是第四句话。这是第五句话。" * 10 - - result = split_text(text, max_chars=100) - - for seg in result: - assert len(seg) <= 100 - - def test_long_single_sentence_hard_cut(self): - """超长单句会被硬切""" - text = "a" * 1000 # 没有标点 - - result = split_text(text, max_chars=200) - - assert len(result) > 1 - for seg in result: - assert len(seg) <= 200 - - def test_newline_is_sentence_end(self): - """换行符作为句子结束符""" - text = "第一行内容\n第二行内容\n第三行内容" * 10 - - result = split_text(text, max_chars=50) - - assert len(result) > 1 - for seg in result: - assert len(seg) <= 50 - - def test_chinese_punctuation(self): - """中文标点（。！？；）作为句子结束符""" - text = "你好！今天吃什么？我吃米饭；你呢？我也吃米饭。" * 10 - - result = split_text(text, max_chars=80) - - for seg in result: - assert len(seg) <= 80 - - def test_english_punctuation(self): - """英文标点（.!?;）作为句子结束符""" - text = "Hello! How are you? I'm fine; thank you. Good bye." * 10 - - result = split_text(text, max_chars=80) - - for seg in result: - assert len(seg) <= 80 - - def test_merged_short_segments(self): - """过短的段落会被合并""" - # 构造很多短句 - text = "你好。再见。谢谢。抱歉。好的。不行。可以。去吧。" * 5 # 每句3-4字 - - result = split_text(text, max_chars=100) - - # 合并后段数应该比单纯按句切的少 - assert len(result) < len(text) // 3 # 粗略估计 - for seg in result: - assert len(seg) <= 100 - - def test_preserves_content(self): - """分段后内容总和与原文基本一致（忽略strip的空白）""" - text = "这是测试文本。包含多个句子。用来验证分段正确性。" * 5 - - result = split_text(text, max_chars=50) - - # 合并所有分段，去掉空白后应该与原文去掉空白后基本一致 - combined = "".join(result).replace(" ", "") - original = text.strip().replace(" ", "") - assert combined == original - - def test_custom_max_chars(self): - """支持自定义 max_chars""" - text = "测试" * 100 # 200字 - - result_50 = split_text(text, max_chars=50) - result_100 = split_text(text, max_chars=100) - - # max_chars 越小，段数应该越多 - assert len(result_50) >= len(result_100) - - def test_single_char_text(self): - """单字符文本""" - assert split_text("好", max_chars=10) == ["好"] - - def test_text_with_only_punctuation(self): - """纯标点文本""" - text = "。。。。。。。。。。" # 10个句号 - result = split_text(text, max_chars=5) - - assert len(result) >= 1 - for seg in result: - assert len(seg) <= 5 - - def test_mixed_content(self): - """中英文混合内容""" - text = "今天的天气是 sunny and warm。我们去了 park 玩。真的很开心！" * 5 - - result = split_text(text, max_chars=80) - - for seg in result: - assert len(seg) <= 80 - - -# ── 短文本与空文本补充 ────────────────────────────────────────────────────── - - -class TestSplitTextEmptyAndShort: - """空文本与短文本补充场景.""" - - def test_whitespace_only_returns_empty(self): - """纯空白文本返回空列表.""" + def test_whitespace_only(self): assert split_text(" \n\t ") == [] - def test_single_char(self): - """单字符文本.""" - assert split_text("好", max_chars=10) == ["好"] - - def test_exactly_max_chars_no_split(self): - """刚好等于 max_chars 不分割.""" - text = "a" * 100 - result = split_text(text, max_chars=100) + def test_short_text_single_segment(self): + text = "你好世界。" + result = split_text(text, max_chars=500) assert len(result) == 1 assert result[0] == text - def test_one_over_max_chars_splits(self): - """超过 max_chars 1 个字符就会分割.""" - text = "a" * 101 + def test_exact_max_chars(self): + text = "a" * 500 + result = split_text(text, max_chars=500) + assert len(result) == 1 + assert len(result[0]) == 500 + + def test_splits_on_sentence_boundary(self): + # 两个长句子，各300字左右，超过50字阈值 + sent1 = "你" * 300 + "。" + sent2 = "我" * 300 + "。" + text = sent1 + sent2 + result = split_text(text, max_chars=500) + assert len(result) == 2 + assert result[0] == sent1 + assert result[1] == sent2 + + def test_long_sentence_hard_cut(self): + # 一个超长句子，没有句末标点，会被硬切 + text = "长" * 800 + result = split_text(text, max_chars=500) + assert len(result) >= 2 + assert all(len(seg) <= 500 for seg in result) + # 合起来应该等于原文本 + assert "".join(result) == text + + def test_short_segments_merged(self): + # 多个短句应该被合并 + sentences = [f"第{i}句。" for i in range(10)] + text = "".join(sentences) + result = split_text(text, max_chars=200) + # 每句5字左右，10句才50字，应该合并成1段 + assert len(result) < 10 + assert len(result[0]) <= 200 + + def test_preserves_content(self): + text = "今天天气真好。我们去公园玩吧！你觉得怎么样？好的，走吧。" + result = split_text(text, max_chars=20) + # 合并后内容应一致 + assert "".join(result) == text + + def test_multiple_punctuation_types(self): + # 构造足够长的文本触发分段 + text = "第一" * 30 + "。" + "第二" * 30 + "！" + "第三" * 30 + "？" + "第四" * 30 + "；" result = split_text(text, max_chars=100) assert len(result) >= 2 + assert "".join(result) == text - def test_none_raises(self): - """None 输入抛 AttributeError（strip 失败）.""" - with pytest.raises(AttributeError): - split_text(None) + def test_custom_max_chars(self): + text = "a" * 100 + "。" + "b" * 100 + "。" + result = split_text(text, max_chars=150) + assert len(result) == 2 + assert "a" in result[0] + assert "b" in result[1] - -# ── 句子边界分段补充 ────────────────────────────────────────────────────── - - -class TestSplitTextSentenceBoundaries: - """句子边界分段补充场景.""" - - def test_split_on_fullwidth_period(self): - """全角句号分段.""" - text = "第一句很长的内容。" * 20 - result = split_text(text, max_chars=60) - assert len(result) > 1 - for seg in result: - assert len(seg) <= 60 - - def test_split_on_fullwidth_question(self): - """全角问号分段.""" - text = "你知道这是为什么吗？" + "是的。" * 20 - result = split_text(text, max_chars=60) - assert len(result) > 1 - - def test_split_on_fullwidth_exclamation(self): - """全角感叹号分段.""" - text = "真是太棒了！" + "内容。" * 20 - result = split_text(text, max_chars=60) - assert len(result) > 1 - - def test_split_on_newline(self): - """换行符分段.""" - lines = ["这是第一行很长的一段文字内容" * 3 for _ in range(5)] - text = "\n".join(lines) - result = split_text(text, max_chars=80) - assert len(result) > 1 - - def test_split_on_semicolon(self): - """全角分号分段.""" - text = "第一项内容；" + "其他内容。" * 20 - result = split_text(text, max_chars=60) - assert len(result) > 1 - - def test_english_period_splits(self): - """英文句号分段.""" - text = "Hello world. " * 30 - result = split_text(text, max_chars=80) - assert len(result) > 1 - - def test_short_sentences_stay_merged(self): - """短句（都 < 50字的句子不会单独成段，会累积到一起.""" - text = "你好。我好。大家好。" - result = split_text(text, max_chars=200) - assert len(result) == 1 - - -# ── 长句强制切段补充 ────────────────────────────────────────────────────── - - -class TestSplitTextLongSentenceForce: - """超长单句强制切段补充.""" - - def test_no_punctuation_forced_split(self): - """完全没有标点的超长文本硬切.""" - text = "字" * 300 - result = split_text(text, max_chars=100) - assert len(result) == 3 - for seg in result: - assert len(seg) == 100 - - def test_force_split_preserves_content(self): - """硬切不丢字符.""" - text = "a" * 250 - result = split_text(text, max_chars=100) - assert sum(len(s) for s in result) == 250 - - def test_mixed_long_and_short(self): - """长句短句混合.""" - long_part = "非常长的句子没有标点符号" * 15 - text = long_part + "。结尾。" - result = split_text(text, max_chars=100) - assert len(result) > 1 - for seg in result: - assert len(seg) <= 100 - - -# ── 短段合并补充 ───────────────────────────────────────────────────────── - - -class TestSplitTextShortSegmentMerge: - """短段合并补充场景.""" - - def test_multiple_short_sentences_merged(self): - """多个短句合并成一段.""" - sentences = ["你好。", "我好。", "大家好。", "天气好。", "心情好。"] - text = "".join(sentences) - result = split_text(text, max_chars=200) - assert len(result) == 1 - - def test_short_tail_merged(self): - """尾部短段被合并到前一段.""" - # 前面一段接近 max_chars，尾部很短 - long_part = "一二三四五六七八九十" * 9 + "。" # ~90字 - tail = "完。" # 2字 - text = long_part + tail - result = split_text(text, max_chars=100) - # 尾部短的应该被合并 - assert len(result) <= 2 - - -# ── 边界情况补充 ───────────────────────────────────────────────────────── - - -class TestSplitTextEdgeCases: - """边界情况补充.""" - - def test_only_punctuation(self): - """纯标点符号.""" - text = "。。。。。" - result = split_text(text, max_chars=10) - assert len(result) == 1 - - def test_mixed_chinese_english(self): - """中英文混合.""" - text = "你好Hello。World!" * 20 - result = split_text(text, max_chars=100) - assert len(result) > 1 - for seg in result: - assert len(seg) <= 100 - - def test_strip_whitespace(self): - """首尾空白被去除.""" - text = " 你好世界。 " - result = split_text(text, max_chars=100) - assert result == ["你好世界。"] - - def test_total_length_preserved(self): - """分段后总长度等于原文 strip 后长度.""" - text = "这是一段用于测试的文本内容。" * 20 - result = split_text(text, max_chars=100) + def test_newline_as_sentence_end(self): + text = "第一段\n第二段\n第三段" + result = split_text(text, max_chars=50) + assert len(result) >= 1 assert "".join(result) == text.strip() - def test_custom_small_max_chars(self): - """很小的 max_chars.""" - text = "一二三四五六七八九十。" * 5 + def test_minimum_segment_length(self): + # 句子太短（<50字）不会立即分段 + text = "短句一。短句二。短句三。" + result = split_text(text, max_chars=200) + assert len(result) == 1 + + def test_trailing_content_added(self): + # 最后一段不完整的句子也要加上 + text = "完整的句子。剩余内容" + result = split_text(text, max_chars=50) + assert "".join(result) == text + + def test_no_empty_segments(self): + text = "。。。。。" # 全是标点 + result = split_text(text, max_chars=2) + assert all(len(seg) > 0 for seg in result) + + def test_chinese_and_english_mixed(self): + text = "Hello世界。这是测试Test文本。Mixed混合。" result = split_text(text, max_chars=20) - assert len(result) > 1 - for seg in result: - assert len(seg) <= 20 - - -# ── 更多边界场景补充 ───────────────────────────────────────────────────────── - - -class TestSplitTextMoreEdgeCases: - """更多边界场景补充""" - - def test_max_chars_one(self): - """max_chars=1 每个字符一段""" - text = "一二三四五" - result = split_text(text, max_chars=1) - assert len(result) == 5 - for seg in result: - assert len(seg) == 1 - - def test_consecutive_newlines(self): - """连续多个换行符""" - text = "第一段\n\n\n第二段\n\n第三段" - result = split_text(text, max_chars=100) - # 合并后应该是一段（内容不长且合并逻辑会被合并） - assert len(result) >= 1 - assert "第一段" in result[0] - for seg in result: - assert len(seg) <= 100 - - def test_only_newlines_only(self): - """只有换行符（纯空白被strip掉返回空""" - assert split_text("\n\n\n\n") == [] - - def test_leading_trailing_whitespace(self): - """首尾空白被去除""" - text = " 你好世界。 " - result = split_text(text, max_chars=100) - assert result == ["你好世界。"] - - def test_very_long_single_sentence_many_segments(self): - """超长单句被切成很多段""" - text = "字" * 1000 - result = split_text(text, max_chars=100) - assert len(result) == 10 - for seg in result: - assert len(seg) == 100 - - def test_mixed_punctuation_types(self): - """全角半角标点混合""" - text = "你好!再见。谢谢?抱歉；好的" - result = split_text(text, max_chars=200) - assert len(result) == 1 - - def test_last_segment_short_merged_to_previous(self): - """尾部极短段被合并到前一段""" - # 构造第一段接近max_chars，结尾有个短句尾巴 - long_part = "一二三四五六七八九十" * 9 + "。" # ~90字 - tail = "完" # 1字 - text = long_part + tail - result = split_text(text, max_chars=100) - # 尾巴应该被合并 - combined = "".join(result) - assert combined == text.strip() - assert len(result) <= 2 - - def test_all_short_sentences_merged_into_one(self): - """大量短句全部合并成一段""" - sentences = ["你好。", "我好。", "他好。", "大家好。", "才是真的好。"] - text = "".join(sentences) - result = split_text(text, max_chars=200) - assert len(result) == 1 - - def test_punctuation_only_long(self): - """很长的纯标点文本""" - text = "。" * 200 - result = split_text(text, max_chars=50) - assert len(result) >= 4 - for seg in result: - assert len(seg) <= 50 - - def test_tab_not_sentence_end(self): - """制表符不是句子结束符""" - text = "这是一段\t包含制表符的文本内容" + "字" * 100 - result = split_text(text, max_chars=50) - # 制表符不在句子结束符集合中，不会触发分段 - # 制表符会保留在分段内容中 - has_tab = any("\t" in seg for seg in result) - assert has_tab + assert len(result) >= 2 + assert "".join(result) == text