Nomad changes

tools/benchmark-harness/fixtures/epub/epub2_cover.json

@@ -0,0 +1,16 @@
+{
+  "document": "../../../../test_documents/epub/epub2_cover.epub",
+  "file_type": "epub",
+  "file_size": 11794,
+  "expected_frameworks": ["kreuzberg"],
+  "metadata": {
+    "description": "epub test: epub2_cover",
+    "source": "pandoc-generated",
+    "size_category": "small"
+  },
+  "ground_truth": {
+    "text_file": "../../../../test_documents/ground_truth/epub/epub2_cover.txt",
+    "markdown_file": "../../../../test_documents/ground_truth/epub/epub2_cover.md",
+    "source": "pandoc"
+  }
+}

tools/benchmark-harness/fixtures/epub/epub2_no_cover.json

@@ -0,0 +1,16 @@
+{
+  "document": "../../../../test_documents/epub/epub2_no_cover.epub",
+  "file_type": "epub",
+  "file_size": 3584,
+  "expected_frameworks": ["kreuzberg"],
+  "metadata": {
+    "description": "epub test: epub2_no_cover",
+    "source": "pandoc-generated",
+    "size_category": "small"
+  },
+  "ground_truth": {
+    "text_file": "../../../../test_documents/ground_truth/epub/epub2_no_cover.txt",
+    "markdown_file": "../../../../test_documents/ground_truth/epub/epub2_no_cover.md",
+    "source": "pandoc"
+  }
+}

tools/benchmark-harness/fixtures/epub/features.json

@@ -0,0 +1,16 @@
+{
+  "document": "../../../../test_documents/epub/features.epub",
+  "file_type": "epub",
+  "file_size": 8970,
+  "expected_frameworks": ["kreuzberg"],
+  "metadata": {
+    "description": "EPUB test document: features",
+    "source": "pandoc-generated",
+    "size_category": "small"
+  },
+  "ground_truth": {
+    "markdown_file": "../../../../test_documents/ground_truth/epub/features.md",
+    "source": "pandoc",
+    "text_file": "../../../../test_documents/ground_truth/epub/features.txt"
+  }
+}

tools/benchmark-harness/fixtures/epub/img.json

@@ -0,0 +1,16 @@
+{
+  "document": "../../../../test_documents/epub/img.epub",
+  "file_type": "epub",
+  "file_size": 20478,
+  "expected_frameworks": ["kreuzberg"],
+  "metadata": {
+    "description": "epub test: img",
+    "source": "pandoc-generated",
+    "size_category": "small"
+  },
+  "ground_truth": {
+    "text_file": "../../../../test_documents/ground_truth/epub/img.txt",
+    "markdown_file": "../../../../test_documents/ground_truth/epub/img.md",
+    "source": "pandoc"
+  }
+}

tools/benchmark-harness/fixtures/epub/simple.json

@@ -0,0 +1,16 @@
+{
+  "document": "../../../../test_documents/epub/simple.epub",
+  "file_type": "epub",
+  "file_size": 31662481,
+  "expected_frameworks": ["kreuzberg"],
+  "metadata": {
+    "description": "epub test: simple",
+    "source": "pandoc-generated",
+    "size_category": "small"
+  },
+  "ground_truth": {
+    "text_file": "../../../../test_documents/ground_truth/epub/simple.txt",
+    "markdown_file": "../../../../test_documents/ground_truth/epub/simple.md",
+    "source": "pandoc"
+  }
+}

tools/benchmark-harness/fixtures/epub/test.json

@@ -0,0 +1,23 @@
+{
+  "document": "../../../../test_documents/vendored/markitdown/epub/test.epub",
+  "file_type": "epub",
+  "file_size": 2677,
+  "expected_frameworks": [
+    "kreuzberg",
+    "markitdown",
+    "pandoc",
+    "pymupdf4llm",
+    "tika",
+    "unstructured"
+  ],
+  "metadata": {
+    "description": "Document from markitdown test suite",
+    "source": "markitdown",
+    "size_category": "small"
+  },
+  "ground_truth": {
+    "text_file": "../../../../test_documents/ground_truth/epub/test.txt",
+    "source": "pandoc",
+    "markdown_file": "../../../../test_documents/ground_truth/epub/test.md"
+  }
+}

tools/benchmark-harness/fixtures/epub/wasteland.json

@@ -0,0 +1,16 @@
+{
+  "document": "../../../../test_documents/epub/wasteland.epub",
+  "file_type": "epub",
+  "file_size": 25840,
+  "expected_frameworks": ["kreuzberg"],
+  "metadata": {
+    "description": "EPUB test document: wasteland",
+    "source": "pandoc-generated",
+    "size_category": "small"
+  },
+  "ground_truth": {
+    "markdown_file": "../../../../test_documents/ground_truth/epub/wasteland.md",
+    "source": "pandoc",
+    "text_file": "../../../../test_documents/ground_truth/epub/wasteland.txt"
+  }
+}

tools/benchmark-harness/fixtures/epub/winter_sports.json

@@ -0,0 +1,16 @@
+{
+  "document": "../../../../test_documents/epub/winter_sports.epub",
+  "file_type": "epub",
+  "file_size": 210376,
+  "expected_frameworks": ["kreuzberg"],
+  "metadata": {
+    "description": "epub test: winter_sports",
+    "source": "pandoc-generated",
+    "size_category": "small"
+  },
+  "ground_truth": {
+    "text_file": "../../../../test_documents/ground_truth/epub/winter_sports.txt",
+    "markdown_file": "../../../../test_documents/ground_truth/epub/winter_sports.md",
+    "source": "pandoc"
+  }
+}