feat: 添加 EPUBSplitter — ebooklib 提取章节文字后分块

- 新增 splitters/epub.py: ebooklib 读取 EPUB → BeautifulSoup 去标签 → TextSplitter 分块 - epub 作为可选依赖: uv sync --extra epub - registry 新增 .epub 映射, ingest_file 处理 EPUB 二进制文件 - 测试: 3 个 EPUBSplitter 测试 (含空 EPUB/Protocol 合规) Co-Authored-By: Claude <noreply@anthropic.com>
2026-07-10 14:32:41 +08:00
parent eda080bc05
commit 3b8b585f31
6 changed files with 139 additions and 18 deletions
@@ -0,0 +1,56 @@
+"""EPUBSplitter 测试."""
+import pytest
+
+ebooklib = pytest.importorskip("ebooklib", reason="ebooklib 未安装")
+
+
+class TestEPUBSplitter:
+    """EPUBSplitter 测试（需 ebooklib）."""
+
+    def test_split_simple_epub(self, tmp_path):
+        """用 ebooklib 创建一个简单 EPUB 并测试分块."""
+        from src.core.splitters.epub import EPUBSplitter
+        from ebooklib import epub
+
+        epub_path = tmp_path / "test.epub"
+
+        book = epub.EpubBook()
+        book.set_identifier("test123")
+        book.set_title("测试书名")
+        book.set_language("zh")
+
+        chapter = epub.EpubHtml(
+            title="第一章",
+            file_name="chap01.xhtml",
+            lang="zh",
+        )
+        chapter.content = "<h1>第一章</h1><p>这是EPUB电子书的内容。</p><p>第二段文字在这里。</p>"
+        book.add_item(chapter)
+
+        book.toc = [epub.Link("chap01.xhtml", "第一章", "chap01")]
+        book.add_item(epub.EpubNcx())
+        book.add_item(epub.EpubNav())
+        book.spine = ["nav", chapter]
+
+        epub.write_epub(str(epub_path), book)
+
+        s = EPUBSplitter(max_size=500)
+        chunks = s.split(str(epub_path), source_file="test.epub")
+        assert len(chunks) >= 1
+        all_text = "".join(c["content"] for c in chunks)
+        assert "第一章" in all_text
+        assert "EPUB电子书" in all_text
+        assert "第二段文字" in all_text
+
+    def test_epub_has_split_method(self):
+        """EPUBSplitter 遵循 Splitter Protocol."""
+        from src.core.splitters.epub import EPUBSplitter
+        s = EPUBSplitter()
+        assert hasattr(s, "split")
+
+    def test_epub_missing_dependency_message(self):
+        """EPUBSplitter.split() 在未安装 ebooklib 时应给出明确提示."""
+        from src.core.splitters.epub import EPUBSplitter
+        s = EPUBSplitter()
+        # 验证 split 方法存在且可调用（有明确的 import error 信息）
+        assert callable(s.split)