chore: import upstream snapshot with attribution

2026-07-13 13:33:56 +08:00
commit 461bf6fd40
1313 changed files with 1079898 additions and 0 deletions
@@ -0,0 +1,146 @@
+[
+    {
+        "element_id": "eda37931eb954fcc8dec8804c7e8fa4c",
+        "metadata": {
+            "category_depth": 0,
+            "file_directory": "test_unstructured/documents/html_files",
+            "filename": "example.html",
+            "filetype": "text/html",
+            "languages": [
+                "eng"
+            ],
+            "last_modified": "2025-06-12T11:12:20",
+            "page_number": 1,
+            "parent_id": "037b418b76eb4ac1bd40326ff67e67b0",
+            "text_as_html": "<div class=\"Page\" data-page-number=\"1\" />"
+        },
+        "text": "",
+        "type": "UncategorizedText"
+    },
+    {
+        "element_id": "97eb491421584ad892074d039779fbfa",
+        "metadata": {
+            "category_depth": 1,
+            "file_directory": "test_unstructured/documents/html_files",
+            "filename": "example.html",
+            "filetype": "text/html",
+            "languages": [
+                "eng"
+            ],
+            "last_modified": "2025-06-12T11:12:20",
+            "page_number": 1,
+            "parent_id": "eda37931eb954fcc8dec8804c7e8fa4c",
+            "text_as_html": "<header class=\"Header\"><h1 class=\"Title\">Header</h1><time class=\"CalendarDate\">Date: October 30, 2023</time></header>"
+        },
+        "text": "Header Date: October 30, 2023",
+        "type": "Header"
+    },
+    {
+        "element_id": "4afb6e4a90e14835b958dadb77cd8331",
+        "metadata": {
+            "category_depth": 1,
+            "file_directory": "test_unstructured/documents/html_files",
+            "filename": "example.html",
+            "filetype": "text/html",
+            "languages": [
+                "eng"
+            ],
+            "last_modified": "2025-06-12T11:12:20",
+            "page_number": 1,
+            "parent_id": "eda37931eb954fcc8dec8804c7e8fa4c",
+            "text_as_html": "<form class=\"Form\"><label class=\"FormField\" for=\"company-name\">From field name</label><input class=\"FormFieldValue\" value=\"Example value\" /></form>"
+        },
+        "text": "From field name Example value",
+        "type": "UncategorizedText"
+    },
+    {
+        "element_id": "d8f996f2bc9a49f4979aac58a2a9ee93",
+        "metadata": {
+            "category_depth": 1,
+            "file_directory": "test_unstructured/documents/html_files",
+            "filename": "example.html",
+            "filetype": "text/html",
+            "languages": [
+                "eng"
+            ],
+            "last_modified": "2025-06-12T11:12:20",
+            "page_number": 1,
+            "parent_id": "eda37931eb954fcc8dec8804c7e8fa4c",
+            "text_as_html": "<section class=\"Section\" />"
+        },
+        "text": "",
+        "type": "UncategorizedText"
+    },
+    {
+        "element_id": "d2c12f995ab248808900f66aec479e9d",
+        "metadata": {
+            "category_depth": 2,
+            "file_directory": "test_unstructured/documents/html_files",
+            "filename": "example.html",
+            "filetype": "text/html",
+            "languages": [
+                "eng"
+            ],
+            "last_modified": "2025-06-12T11:12:20",
+            "page_number": 1,
+            "parent_id": "d8f996f2bc9a49f4979aac58a2a9ee93",
+            "text_as_html": "<table class=\"Table\"><thead><tr><th>Description</th><th>Row header</th></tr></thead><tbody><tr><td>Value description</td><td><span>50 $</span><span>(1.32 %)</span></td></tr></tbody></table>"
+        },
+        "text": "Description Row header Value description 50 $ (1.32 %)",
+        "type": "Table"
+    },
+    {
+        "element_id": "8e3f0d85329343008593f43afcad3327",
+        "metadata": {
+            "category_depth": 1,
+            "file_directory": "test_unstructured/documents/html_files",
+            "filename": "example.html",
+            "filetype": "text/html",
+            "languages": [
+                "eng"
+            ],
+            "last_modified": "2025-06-12T11:12:20",
+            "page_number": 1,
+            "parent_id": "eda37931eb954fcc8dec8804c7e8fa4c",
+            "text_as_html": "<section class=\"Section\" />"
+        },
+        "text": "",
+        "type": "UncategorizedText"
+    },
+    {
+        "element_id": "5deaad75854741ccb69767881ef399db",
+        "metadata": {
+            "category_depth": 2,
+            "file_directory": "test_unstructured/documents/html_files",
+            "filename": "example.html",
+            "filetype": "text/html",
+            "languages": [
+                "eng"
+            ],
+            "last_modified": "2025-06-12T11:12:20",
+            "page_number": 1,
+            "parent_id": "8e3f0d85329343008593f43afcad3327",
+            "text_as_html": "<h2 class=\"Subtitle\">2. Subtitle</h2>"
+        },
+        "text": "2. Subtitle",
+        "type": "Title"
+    },
+    {
+        "element_id": "9e61f29755bc4b6dbb41ea575d41edb6",
+        "metadata": {
+            "category_depth": 2,
+            "file_directory": "test_unstructured/documents/html_files",
+            "filename": "example.html",
+            "filetype": "text/html",
+            "languages": [
+                "eng"
+            ],
+            "last_modified": "2025-06-12T11:12:20",
+            "page_number": 1,
+            "parent_id": "8e3f0d85329343008593f43afcad3327",
+            "text_as_html": "<p class=\"NarrativeText\">Paragraph text</p>"
+        },
+        "text": "Paragraph text",
+        "type": "NarrativeText"
+    }
+]
@@ -0,0 +1,38 @@
+[
+    {
+        "element_id": "8157292897eb45e68229b5816da6e46c",
+        "metadata": {
+            "category_depth": 0,
+            "file_directory": "test_unstructured/documents/html_files",
+            "filename": "example_with_alternative_text.html",
+            "filetype": "text/html",
+            "languages": [
+                "eng"
+            ],
+            "last_modified": "2025-06-12T11:12:20",
+            "page_number": 1,
+            "parent_id": "6f5f66a6a23642a7958aeff927e343cd",
+            "text_as_html": "<div class=\"Page\" data-page-number=\"1\" />"
+        },
+        "text": "",
+        "type": "UncategorizedText"
+    },
+    {
+        "element_id": "70c89a734fe1497293a2e01b2f35b887",
+        "metadata": {
+            "category_depth": 1,
+            "file_directory": "test_unstructured/documents/html_files",
+            "filename": "example_with_alternative_text.html",
+            "filetype": "text/html",
+            "languages": [
+                "eng"
+            ],
+            "last_modified": "2025-06-12T11:12:20",
+            "page_number": 1,
+            "parent_id": "8157292897eb45e68229b5816da6e46c",
+            "text_as_html": "<header class=\"Header\"><img class=\"Logo\" alt=\"New York logo\" /><img class=\"Image\" alt=\"A line graph showing the comparison of 5 year cumulative total return for stocks\" /></header>"
+        },
+        "text": "New York logo A line graph showing the comparison of 5 year cumulative total return for stocks",
+        "type": "Header"
+    }
+]
@@ -0,0 +1,38 @@
+[
+    {
+        "element_id": "7aeac07bad7f44359c3b2403d2ca3d2a",
+        "metadata": {
+            "category_depth": 0,
+            "file_directory": "test_unstructured/documents/html_files",
+            "filename": "example_with_inline_fields.html",
+            "filetype": "text/html",
+            "languages": [
+                "eng"
+            ],
+            "last_modified": "2025-06-12T11:12:20",
+            "page_number": 1,
+            "parent_id": "8bf6d9316275493499a373771a6e46d0",
+            "text_as_html": "<div class=\"Page\" data-page-number=\"1\" />"
+        },
+        "text": "",
+        "type": "UncategorizedText"
+    },
+    {
+        "element_id": "f8f71f7ce10748f08bee0f9922a04406",
+        "metadata": {
+            "category_depth": 1,
+            "file_directory": "test_unstructured/documents/html_files",
+            "filename": "example_with_inline_fields.html",
+            "filetype": "text/html",
+            "languages": [
+                "eng"
+            ],
+            "last_modified": "2025-06-12T11:12:20",
+            "page_number": 1,
+            "parent_id": "7aeac07bad7f44359c3b2403d2ca3d2a",
+            "text_as_html": "<header class=\"Header\"><p class=\"NarrativeText\">Table of Contents</p><address class=\"Address\">68 Prince Street Palmdale, CA 93550</address><a class=\"Hyperlink\">www.google.com</a><span class=\"UncategorizedText\">More text</span></header>"
+        },
+        "text": "Table of Contents 68 Prince Street Palmdale, CA 93550 www.google.com More text",
+        "type": "Header"
+    }
+]
@@ -0,0 +1,44 @@
+[
+    {
+        "element_id": "2428404551304d4db5925f6afee11ed5",
+        "metadata": {
+            "category_depth": 0,
+            "filetype": "text/html",
+            "languages": [
+                "eng"
+            ],
+            "parent_id": "517f8559ba594270bdd67e1b02bf19a2",
+            "text_as_html": "<table class=\"Table\"><tr><th>Header 1</th><th>Header 2</th></tr><tr><td>Row 1, Cell 1</td><td>Row 1, Cell 2</td></tr><tr><td>Row 2, Cell 1</td><td>Row 2, Cell 2</td></tr></table>"
+        },
+        "text": "Header 1 Header 2 Row 1, Cell 1 Row 1, Cell 2 Row 2, Cell 1 Row 2, Cell 2",
+        "type": "Table"
+    },
+    {
+        "element_id": "9f91cae321c74b31bb1c83ac86cd7afb",
+        "metadata": {
+            "category_depth": 0,
+            "filetype": "text/html",
+            "languages": [
+                "eng"
+            ],
+            "parent_id": "517f8559ba594270bdd67e1b02bf19a2",
+            "text_as_html": "<table class=\"Table\"><tr><th colspan=\"3\">Big Table Header</th></tr><tr><td rowspan=\"2\">Merged Cell 1</td><td>Cell 2</td><td>Cell 3</td></tr><tr><td colspan=\"2\">Merged Cell 4 and 5</td></tr><tr><td>Cell 6</td><td>Cell 7</td><td>Cell 8</td></tr><tr><td>Cell 9</td><td colspan=\"2\">A cell with a lot of text. Lorem ipsum dolor sit amet, consectetur adipiscing elit. Sed do eiusmod tempor incididunt ut labore et dolore magna aliqua.</td></tr><tr><td>Cell 10</td><td>Cell 11</td><td>Cell 12</td></tr></table>"
+        },
+        "text": "Big Table Header Merged Cell 1 Cell 2 Cell 3 Merged Cell 4 and 5 Cell 6 Cell 7 Cell 8 Cell 9 A cell with a lot of text. Lorem ipsum dolor sit amet, consectetur adipiscing elit. Sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Cell 10 Cell 11 Cell 12",
+        "type": "Table"
+    },
+    {
+        "element_id": "da6c34391e544b3480e45d68f40870fa",
+        "metadata": {
+            "category_depth": 0,
+            "filetype": "text/html",
+            "languages": [
+                "eng"
+            ],
+            "parent_id": "517f8559ba594270bdd67e1b02bf19a2",
+            "text_as_html": "<table class=\"TableOfContents\"><tr><th>Chapter</th><th>Title</th><th>Page</th></tr><tr><td>1</td><td>Introduction</td><td>1</td></tr><tr><td>2</td><td>Getting Started</td><td>5</td></tr><tr><td>3</td><td>Basic Concepts</td><td>12</td></tr><tr><td>4</td><td>Advanced Topics</td><td>25</td></tr><tr><td>5</td><td>Conclusion</td><td>40</td></tr></table>"
+        },
+        "text": "Chapter Title Page 1 Introduction 1 2 Getting Started 5 3 Basic Concepts 12 4 Advanced Topics 25 5 Conclusion 40",
+        "type": "Table"
+    }
+]