chunkr-ai 0.0.50__tar.gz → 0.3.0__tar.gz

{chunkr_ai-0.0.50/src/chunkr_ai.egg-info → chunkr_ai-0.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: chunkr-ai
-Version: 0.0.50
+Version: 0.3.0
 Summary: Python client for Chunkr: open source document intelligence
 Author-email: Ishaan Kapoor <ishaan@lumina.sh>
 License: MIT License
@@ -28,6 +28,7 @@ Project-URL: Homepage, https://chunkr.ai
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: httpx>=0.25.0
+Requires-Dist: matplotlib>=3.10.3
 Requires-Dist: nest-asyncio>=1.6.0
 Requires-Dist: pillow>=10.0.0
 Requires-Dist: pydantic>=2.0.0

{chunkr_ai-0.0.50 → chunkr_ai-0.3.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "chunkr-ai"
-version = "0.0.50"
+version = "0.3.0"
 authors = [{"name" = "Ishaan Kapoor", "email" = "ishaan@lumina.sh"}]
 description = "Python client for Chunkr: open source document intelligence"
 readme = "README.md"
@@ -12,6 +12,7 @@ license = {"file" = "LICENSE"}
 urls = {Homepage = "https://chunkr.ai"} 
 dependencies = [
     "httpx>=0.25.0",
+    "matplotlib>=3.10.3",
     "nest-asyncio>=1.6.0",
     "pillow>=10.0.0",
     "pydantic>=2.0.0",

{chunkr_ai-0.0.50 → chunkr_ai-0.3.0}/src/chunkr_ai/api/configuration.py

@@ -3,27 +3,34 @@ from enum import Enum
 from typing import Any, List, Optional, Union
 from pydantic import field_validator, field_serializer
 
-class GenerationStrategy(str, Enum):
-    LLM = "LLM"
-    AUTO = "Auto"
-
 class CroppingStrategy(str, Enum):
     ALL = "All"
     AUTO = "Auto"
 
-class EmbedSource(str, Enum):
-    HTML = "HTML"
+class SegmentFormat(str, Enum):
+    HTML = "Html"
     MARKDOWN = "Markdown"
-    LLM = "LLM"
+
+class EmbedSource(str, Enum):
     CONTENT = "Content"
+    HTML = "HTML"  # Deprecated
+    MARKDOWN = "Markdown"  # Deprecated
+    LLM = "LLM"
+
+class GenerationStrategy(str, Enum):
+    LLM = "LLM"
+    AUTO = "Auto"
 
 class GenerationConfig(BaseModel):
-    html: Optional[GenerationStrategy] = None
+    format: Optional[SegmentFormat] = None
+    strategy: Optional[GenerationStrategy] = None
     llm: Optional[str] = None
-    markdown: Optional[GenerationStrategy] = None
     crop_image: Optional[CroppingStrategy] = None
-    embed_sources: Optional[List[EmbedSource]] = Field(default_factory=lambda: [EmbedSource.MARKDOWN])
+    embed_sources: Optional[List[EmbedSource]] = None
     extended_context: Optional[bool] = None
+    # Deprecated fields for backwards compatibility
+    html: Optional[GenerationStrategy] = None  # Deprecated: Use format=SegmentFormat.HTML and strategy instead
+    markdown: Optional[GenerationStrategy] = None  # Deprecated: Use format=SegmentFormat.MARKDOWN and strategy instead
 
 class SegmentProcessing(BaseModel):
     model_config = ConfigDict(populate_by_name=True, alias_generator=str.title)
@@ -244,9 +251,45 @@ class SegmentType(str, Enum):
     TEXT = "Text"
     TITLE = "Title"
 
+class Alignment(str, Enum):
+    LEFT = "Left"
+    CENTER = "Center"
+    RIGHT = "Right"
+    JUSTIFY = "Justify"
+
+class VerticalAlignment(str, Enum):
+    TOP = "Top"
+    MIDDLE = "Middle"
+    BOTTOM = "Bottom"
+    BASELINE = "Baseline"
+
+class CellStyle(BaseModel):
+    bg_color: Optional[str] = None
+    text_color: Optional[str] = None
+    font_face: Optional[str] = None
+    is_bold: Optional[bool] = None
+    align: Optional[Alignment] = None
+    valign: Optional[VerticalAlignment] = None
+
+class Cell(BaseModel):
+    cell_id: str
+    text: str
+    range: str
+    formula: Optional[str] = None
+    value: Optional[str] = None
+    hyperlink: Optional[str] = None
+    style: Optional[CellStyle] = None
+
+class Page(BaseModel):
+    image: str
+    page_number: int
+    page_height: float
+    page_width: float
+    ss_sheet_name: Optional[str] = None
+
 class Segment(BaseModel):
     bbox: BoundingBox
-    content: str
+    content: str = ""
     page_height: float
     llm: Optional[str] = None
     html: Optional[str] = None
@@ -258,6 +301,16 @@ class Segment(BaseModel):
     segment_id: str
     segment_type: SegmentType
     confidence: Optional[float]
+    text: str = ""
+    segment_length: Optional[int] = None
+    # Spreadsheet-specific fields
+    ss_cells: Optional[List[Cell]] = None
+    ss_header_bbox: Optional[BoundingBox] = None
+    ss_header_ocr: Optional[List[OCRResult]] = None
+    ss_header_text: Optional[str] = None
+    ss_header_range: Optional[str] = None
+    ss_range: Optional[str] = None
+    ss_sheet_name: Optional[str] = None
 
 class Chunk(BaseModel):
     chunk_id: str
@@ -268,6 +321,8 @@ class Chunk(BaseModel):
 class OutputResponse(BaseModel):
     chunks: List[Chunk]
     file_name: Optional[str]
+    mime_type: Optional[str] = None
+    pages: Optional[List[Page]] = None
     page_count: Optional[int]
     pdf_url: Optional[str]
 

{chunkr_ai-0.0.50 → chunkr_ai-0.3.0}/src/chunkr_ai/models.py

@@ -1,5 +1,8 @@
 from .api.configuration import (
+    Alignment,
     BoundingBox,
+    Cell,
+    CellStyle,
     Chunk,
     ChunkProcessing,
     Configuration,
@@ -14,7 +17,9 @@ from .api.configuration import (
     OCRResult,
     OcrStrategy,
     OutputResponse,
+    Page,
     Segment,
+    SegmentFormat,
     SegmentProcessing,
     SegmentType,
     SegmentationStrategy,
@@ -22,11 +27,15 @@ from .api.configuration import (
     Pipeline,
     Tokenizer,
     TokenizerType,
+    VerticalAlignment,
 )
 from .api.task_response import TaskResponse
 
 __all__ = [
+    "Alignment",
     "BoundingBox",
+    "Cell",
+    "CellStyle",
     "Chunk",
     "ChunkProcessing",
     "Configuration",
@@ -41,7 +50,9 @@ __all__ = [
     "OCRResult",
     "OcrStrategy",
     "OutputResponse",
+    "Page",
     "Segment",
+    "SegmentFormat",
     "SegmentProcessing",
     "SegmentType",
     "SegmentationStrategy",
@@ -50,4 +61,5 @@ __all__ = [
     "Pipeline",
     "Tokenizer",
     "TokenizerType",
+    "VerticalAlignment",
 ]

{chunkr_ai-0.0.50 → chunkr_ai-0.3.0/src/chunkr_ai.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: chunkr-ai
-Version: 0.0.50
+Version: 0.3.0
 Summary: Python client for Chunkr: open source document intelligence
 Author-email: Ishaan Kapoor <ishaan@lumina.sh>
 License: MIT License
@@ -28,6 +28,7 @@ Project-URL: Homepage, https://chunkr.ai
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: httpx>=0.25.0
+Requires-Dist: matplotlib>=3.10.3
 Requires-Dist: nest-asyncio>=1.6.0
 Requires-Dist: pillow>=10.0.0
 Requires-Dist: pydantic>=2.0.0

{chunkr_ai-0.0.50 → chunkr_ai-0.3.0}/src/chunkr_ai.egg-info/SOURCES.txt

@@ -18,4 +18,6 @@ src/chunkr_ai/api/misc.py
 src/chunkr_ai/api/protocol.py
 src/chunkr_ai/api/task_response.py
 tests/test_chunkr.py
-tests/test_file_handling.py
+tests/test_excel.py
+tests/test_file_handling.py
+tests/test_pages.py

{chunkr_ai-0.0.50 → chunkr_ai-0.3.0}/src/chunkr_ai.egg-info/requires.txt

@@ -1,4 +1,5 @@
 httpx>=0.25.0
+matplotlib>=3.10.3
 nest-asyncio>=1.6.0
 pillow>=10.0.0
 pydantic>=2.0.0

{chunkr_ai-0.0.50 → chunkr_ai-0.3.0}/tests/test_chunkr.py

@@ -2,7 +2,6 @@ import pytest
 from pathlib import Path
 from PIL import Image
 import asyncio
-from typing import Awaitable
 
 from chunkr_ai import Chunkr
 from chunkr_ai.models import (
@@ -21,6 +20,7 @@ from chunkr_ai.models import (
     Status,
     TaskResponse,
     Tokenizer,
+    SegmentFormat,
 )
 
 @pytest.fixture
@@ -48,14 +48,21 @@ def client():
     client = Chunkr()
     yield client
 
+@pytest.fixture(params=[
+    pytest.param(None, id="none_pipeline"),
+    pytest.param(Pipeline.AZURE, id="azure_pipeline"),
+])
+def pipeline_type(request):
+    return request.param
+
 @pytest.fixture
 def markdown_embed_config():
     return Configuration(
         segment_processing=SegmentProcessing(
             Page=GenerationConfig(
-                html=GenerationStrategy.LLM,
-                markdown=GenerationStrategy.LLM,
-                embed_sources=[EmbedSource.MARKDOWN]
+                format=SegmentFormat.MARKDOWN,
+                strategy=GenerationStrategy.LLM,
+                embed_sources=[EmbedSource.CONTENT]
             )
         ),
     )
@@ -65,9 +72,9 @@ def html_embed_config():
     return Configuration(
         segment_processing=SegmentProcessing(
             Page=GenerationConfig(
-                html=GenerationStrategy.LLM,
-                markdown=GenerationStrategy.LLM,
-                embed_sources=[EmbedSource.HTML]
+                format=SegmentFormat.HTML,
+                strategy=GenerationStrategy.LLM,
+                embed_sources=[EmbedSource.HTML]  # Keep this for backwards compatibility testing
             )
         ),
     )
@@ -77,10 +84,10 @@ def multiple_embed_config():
     return Configuration(
         segment_processing=SegmentProcessing(
             Page=GenerationConfig(
-                html=GenerationStrategy.LLM,
-                markdown=GenerationStrategy.LLM,
+                format=SegmentFormat.MARKDOWN,
+                strategy=GenerationStrategy.LLM,
                 llm="Generate a summary of this content",
-                embed_sources=[EmbedSource.MARKDOWN, EmbedSource.LLM, EmbedSource.HTML]
+                embed_sources=[EmbedSource.CONTENT, EmbedSource.LLM, EmbedSource.HTML]
             )
         ),
     )
@@ -169,13 +176,15 @@ def model_fallback_config():
 def extended_context_config():
     return Configuration(
         segment_processing=SegmentProcessing(
-            picture=GenerationConfig(
+            Picture=GenerationConfig(
                 extended_context=True,
-                html=GenerationStrategy.LLM,
+                format=SegmentFormat.HTML,
+                strategy=GenerationStrategy.LLM,
             ),
-            table=GenerationConfig(
+            Table=GenerationConfig(
                 extended_context=True,
-                html=GenerationStrategy.LLM,
+                format=SegmentFormat.HTML,
+                strategy=GenerationStrategy.LLM,
             )
         ),
     )
@@ -471,12 +480,12 @@ async def test_tokenizer_custom_string(client, sample_path, custom_tokenizer_con
     assert response.status == "Succeeded"
     assert response.output is not None
 
-@pytest.mark.asyncio
-async def test_embed_sources_with_different_tokenizer(client, sample_path, xlm_roberta_with_html_content_config):
-    response = await client.upload(sample_path, xlm_roberta_with_html_content_config)
-    assert response.task_id is not None
-    assert response.status == "Succeeded"
-    assert response.output is not None
+# @pytest.mark.asyncio
+# async def test_embed_sources_with_different_tokenizer(client, sample_path, xlm_roberta_with_html_content_config):
+#     response = await client.upload(sample_path, xlm_roberta_with_html_content_config)
+#     assert response.task_id is not None
+#     assert response.status == "Succeeded"
+#     assert response.output is not None
 
 @pytest.mark.asyncio
 async def test_error_handling_continue(client, sample_path):
@@ -582,4 +591,129 @@ async def test_extended_context(client, sample_path, extended_context_config):
 
     except Exception as e:
         print(f"Error during extended context test: {e}")
-        raise # Re-raise the exception to fail the test explicitly
+        raise # Re-raise the exception to fail the test explicitly
+
+
+# Tests for new fields added in recent updates
+class TestNewFields:
+    """Test the newly added fields in the models"""
+
+    @pytest.mark.asyncio
+    async def test_output_has_mime_type_field(self, client, sample_path):
+        """Test that OutputResponse includes mime_type field"""
+        response = await client.upload(sample_path)
+        assert response.task_id is not None
+        assert response.status == "Succeeded"
+        assert response.output is not None
+        
+        # mime_type should be accessible (might be None for some file types)
+        assert hasattr(response.output, 'mime_type')
+        
+        # For PDF files, mime_type should be present
+        if response.output.mime_type:
+            assert "pdf" in response.output.mime_type.lower()
+
+    @pytest.mark.asyncio
+    async def test_output_has_pages_field(self, client, sample_path):
+        """Test that OutputResponse includes pages field"""
+        response = await client.upload(sample_path)
+        assert response.task_id is not None
+        assert response.status == "Succeeded"
+        assert response.output is not None
+        
+        # pages should be accessible (might be None for some configurations)
+        assert hasattr(response.output, 'pages')
+        
+        # If pages exist, validate structure
+        if response.output.pages:
+            assert len(response.output.pages) > 0
+            page = response.output.pages[0]
+            assert hasattr(page, 'image')
+            assert hasattr(page, 'page_number')
+            assert hasattr(page, 'page_height')
+            assert hasattr(page, 'page_width')
+            assert hasattr(page, 'ss_sheet_name')
+
+    @pytest.mark.asyncio
+    async def test_segments_have_spreadsheet_fields(self, client, sample_path):
+        """Test that Segment objects include new spreadsheet fields"""
+        response = await client.upload(sample_path)
+        assert response.task_id is not None
+        assert response.status == "Succeeded"
+        assert response.output is not None
+        assert len(response.output.chunks) > 0
+        
+        segment = response.output.chunks[0].segments[0]
+        
+        # All new spreadsheet fields should be accessible
+        assert hasattr(segment, 'segment_length')
+        assert hasattr(segment, 'ss_cells')
+        assert hasattr(segment, 'ss_header_bbox')
+        assert hasattr(segment, 'ss_header_ocr')
+        assert hasattr(segment, 'ss_header_text')
+        assert hasattr(segment, 'ss_header_range')
+        assert hasattr(segment, 'ss_range')
+        assert hasattr(segment, 'ss_sheet_name')
+        
+        # For PDF files, spreadsheet fields should be None
+        assert segment.ss_cells is None
+        assert segment.ss_range is None
+        assert segment.ss_sheet_name is None
+
+    @pytest.mark.asyncio
+    async def test_segment_length_field(self, client, sample_path):
+        """Test that segments can have length calculations"""
+        response = await client.upload(sample_path)
+        assert response.task_id is not None
+        assert response.status == "Succeeded"
+        assert response.output is not None
+        
+        # Check if any segments have length calculations
+        segments_with_length = []
+        for chunk in response.output.chunks:
+            for segment in chunk.segments:
+                if segment.segment_length is not None:
+                    segments_with_length.append(segment)
+        
+        # segment_length might be None depending on configuration
+        # but if present, should be positive
+        for segment in segments_with_length:
+            assert segment.segment_length > 0
+
+    @pytest.mark.asyncio
+    async def test_backwards_compatibility_preserved(self, client, sample_path):
+        """Test that all existing fields still work after adding new ones"""
+        response = await client.upload(sample_path)
+        assert response.task_id is not None
+        assert response.status == "Succeeded"
+        assert response.output is not None
+        
+        # All existing fields should still work
+        assert response.output.chunks is not None
+        assert response.output.file_name is not None
+        assert response.output.page_count is not None
+        assert response.output.pdf_url is not None
+        
+        # Chunk structure should be unchanged
+        chunk = response.output.chunks[0]
+        assert chunk.chunk_id is not None
+        assert chunk.chunk_length is not None
+        assert chunk.segments is not None
+        assert chunk.embed is not None or chunk.embed is None  # embed can be None
+        
+        # Segment structure should include all original fields
+        segment = chunk.segments[0]
+        assert segment.bbox is not None
+        assert segment.content is not None or segment.content == ""
+        assert segment.page_height is not None
+        assert segment.llm is not None or segment.llm is None
+        assert segment.html is not None or segment.html == ""
+        assert segment.image is not None or segment.image is None
+        assert segment.markdown is not None or segment.markdown == ""
+        assert segment.ocr is not None or segment.ocr == []
+        assert segment.page_number is not None
+        assert segment.page_width is not None
+        assert segment.segment_id is not None
+        assert segment.segment_type is not None
+        assert segment.confidence is not None or segment.confidence is None
+        assert segment.text is not None or segment.text == ""

chunkr_ai-0.3.0/tests/test_excel.py

@@ -0,0 +1,417 @@
+import pytest
+import json
+from pathlib import Path
+
+from chunkr_ai import Chunkr
+from chunkr_ai.models import (
+    Configuration,
+    ChunkProcessing,
+    SegmentProcessing,
+    GenerationConfig,
+    SegmentFormat,
+    GenerationStrategy,
+    EmbedSource,
+    Tokenizer,
+    OcrStrategy,
+    SegmentationStrategy,
+    Cell,
+    CellStyle,
+    Alignment,
+    VerticalAlignment,
+    Page,
+    Segment,
+    SegmentType,
+)
+
+
+@pytest.fixture
+def excel_sample_path():
+    """Path to the Excel test file"""
+    return Path("tests/files/excel/test.xlsx")
+
+
+@pytest.fixture
+def excel_expected_output():
+    """Expected output for Excel test file"""
+    with open("tests/files/excel/test.json", "r") as f:
+        return json.load(f)
+
+
+@pytest.fixture
+def client():
+    """Chunkr client instance"""
+    client = Chunkr()
+    yield client
+
+
+@pytest.fixture
+def excel_config():
+    """Configuration optimized for Excel processing"""
+    return Configuration(
+        high_resolution=True,
+        ocr_strategy=OcrStrategy.ALL,
+        segmentation_strategy=SegmentationStrategy.LAYOUT_ANALYSIS,
+        chunk_processing=ChunkProcessing(
+            target_length=512,
+            tokenizer=Tokenizer.WORD,
+        ),
+        segment_processing=SegmentProcessing(
+            Table=GenerationConfig(
+                format=SegmentFormat.MARKDOWN,
+                strategy=GenerationStrategy.AUTO,
+                embed_sources=[EmbedSource.MARKDOWN],
+            ),
+            Text=GenerationConfig(
+                format=SegmentFormat.MARKDOWN,
+                strategy=GenerationStrategy.AUTO,
+                embed_sources=[EmbedSource.MARKDOWN],
+            ),
+        ),
+    )
+
+
+class TestExcelBasicFunctionality:
+    """Test basic Excel file processing"""
+
+    @pytest.mark.asyncio
+    async def test_excel_upload_and_process(self, client, excel_sample_path, excel_config):
+        """Test that Excel file can be uploaded and processed successfully"""
+        response = await client.upload(excel_sample_path, excel_config)
+        
+        assert response.task_id is not None
+        assert response.status == "Succeeded"
+        assert response.output is not None
+        assert response.output.chunks is not None
+        assert len(response.output.chunks) > 0
+
+    @pytest.mark.asyncio
+    async def test_excel_mime_type(self, client, excel_sample_path, excel_config):
+        """Test that Excel files have correct MIME type"""
+        response = await client.upload(excel_sample_path, excel_config)
+        
+        assert response.output.mime_type == "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
+
+    @pytest.mark.asyncio
+    async def test_excel_pages_exist(self, client, excel_sample_path, excel_config):
+        """Test that Excel processing generates pages information"""
+        response = await client.upload(excel_sample_path, excel_config)
+        
+        assert response.output.pages is not None
+        assert len(response.output.pages) > 0
+        assert response.output.page_count is not None
+        assert response.output.page_count > 0
+
+    @pytest.mark.asyncio
+    async def test_excel_chunks_have_segments(self, client, excel_sample_path, excel_config):
+        """Test that Excel chunks contain segments with data"""
+        response = await client.upload(excel_sample_path, excel_config)
+        
+        assert len(response.output.chunks) > 0
+        chunk = response.output.chunks[0]
+        assert len(chunk.segments) > 0
+        assert chunk.chunk_length > 0
+
+
+class TestExcelSpreadsheetFields:
+    """Test Excel-specific spreadsheet fields"""
+
+    @pytest.mark.asyncio
+    async def test_segments_have_spreadsheet_fields(self, client, excel_sample_path, excel_config):
+        """Test that segments contain spreadsheet-specific fields"""
+        response = await client.upload(excel_sample_path, excel_config)
+        
+        # Find a segment with spreadsheet data
+        spreadsheet_segment = None
+        for chunk in response.output.chunks:
+            for segment in chunk.segments:
+                if segment.ss_cells and len(segment.ss_cells) > 0:
+                    spreadsheet_segment = segment
+                    break
+            if spreadsheet_segment:
+                break
+        
+        assert spreadsheet_segment is not None, "No segment with spreadsheet data found"
+        
+        # Test spreadsheet-specific fields
+        assert spreadsheet_segment.ss_cells is not None
+        assert len(spreadsheet_segment.ss_cells) > 0
+        assert spreadsheet_segment.ss_sheet_name is not None
+        assert spreadsheet_segment.ss_range is not None
+
+    @pytest.mark.asyncio
+    async def test_cells_have_required_fields(self, client, excel_sample_path, excel_config):
+        """Test that cells contain all required fields"""
+        response = await client.upload(excel_sample_path, excel_config)
+        
+        # Find a segment with cells
+        test_cell = None
+        for chunk in response.output.chunks:
+            for segment in chunk.segments:
+                if segment.ss_cells and len(segment.ss_cells) > 0:
+                    test_cell = segment.ss_cells[0]
+                    break
+            if test_cell:
+                break
+        
+        assert test_cell is not None, "No cell found in any segment"
+        
+        # Test required cell fields
+        assert test_cell.cell_id is not None
+        assert test_cell.text is not None
+        assert test_cell.range is not None
+        # Optional fields should be accessible
+        assert hasattr(test_cell, 'formula')
+        assert hasattr(test_cell, 'value')
+        assert hasattr(test_cell, 'hyperlink')
+        assert hasattr(test_cell, 'style')
+
+    @pytest.mark.asyncio
+    async def test_cell_styling_fields(self, client, excel_sample_path, excel_config):
+        """Test that cells with styling contain CellStyle information"""
+        response = await client.upload(excel_sample_path, excel_config)
+        
+        # Find a cell with styling
+        styled_cell = None
+        for chunk in response.output.chunks:
+            for segment in chunk.segments:
+                if segment.ss_cells:
+                    for cell in segment.ss_cells:
+                        if cell.style is not None:
+                            styled_cell = cell
+                            break
+                if styled_cell:
+                    break
+            if styled_cell:
+                break
+        
+        assert styled_cell is not None, "No styled cell found"
+        assert styled_cell.style is not None
+        
+        # Test CellStyle fields
+        style = styled_cell.style
+        assert hasattr(style, 'bg_color')
+        assert hasattr(style, 'text_color')
+        assert hasattr(style, 'font_face')
+        assert hasattr(style, 'is_bold')
+        assert hasattr(style, 'align')
+        assert hasattr(style, 'valign')
+
+    @pytest.mark.asyncio
+    async def test_excel_sheet_names(self, client, excel_sample_path, excel_config):
+        """Test that sheet names are properly captured"""
+        response = await client.upload(excel_sample_path, excel_config)
+        
+        # Check segments for sheet names
+        sheet_names = set()
+        for chunk in response.output.chunks:
+            for segment in chunk.segments:
+                if segment.ss_sheet_name:
+                    sheet_names.add(segment.ss_sheet_name)
+        
+        assert len(sheet_names) > 0, "No sheet names found in segments"
+        
+        # Check pages for sheet names
+        page_sheet_names = set()
+        if response.output.pages:
+            for page in response.output.pages:
+                if page.ss_sheet_name:
+                    page_sheet_names.add(page.ss_sheet_name)
+        
+        # At least one source should have sheet names
+        assert len(sheet_names) > 0 or len(page_sheet_names) > 0
+
+
+class TestExcelPages:
+    """Test Excel pages functionality"""
+
+    @pytest.mark.asyncio
+    async def test_pages_structure(self, client, excel_sample_path, excel_config):
+        """Test that pages have correct structure and fields"""
+        response = await client.upload(excel_sample_path, excel_config)
+        
+        assert response.output.pages is not None
+        assert len(response.output.pages) > 0
+        
+        page = response.output.pages[0]
+        assert page.image is not None
+        assert page.page_number is not None
+        assert page.page_height is not None
+        assert page.page_width is not None
+        # ss_sheet_name is optional for pages
+        assert hasattr(page, 'ss_sheet_name')
+
+    @pytest.mark.asyncio
+    async def test_page_count_consistency(self, client, excel_sample_path, excel_config):
+        """Test that page_count matches the actual number of pages"""
+        response = await client.upload(excel_sample_path, excel_config)
+        
+        assert response.output.page_count is not None
+        if response.output.pages:
+            assert response.output.page_count == len(response.output.pages)
+
+    @pytest.mark.asyncio
+    async def test_page_numbers_sequential(self, client, excel_sample_path, excel_config):
+        """Test that page numbers are sequential and start from 1"""
+        response = await client.upload(excel_sample_path, excel_config)
+        
+        if response.output.pages and len(response.output.pages) > 1:
+            page_numbers = [page.page_number for page in response.output.pages]
+            page_numbers.sort()
+            
+            # Should start from 1 and be sequential
+            for i, page_num in enumerate(page_numbers):
+                assert page_num == i + 1, f"Page numbers not sequential: {page_numbers}"
+
+
+class TestExcelSegmentTypes:
+    """Test Excel segment types and their properties"""
+
+    @pytest.mark.asyncio
+    async def test_segment_types_present(self, client, excel_sample_path, excel_config):
+        """Test that appropriate segment types are detected in Excel files"""
+        response = await client.upload(excel_sample_path, excel_config)
+        
+        segment_types = set()
+        for chunk in response.output.chunks:
+            for segment in chunk.segments:
+                segment_types.add(segment.segment_type)
+        
+        # Excel files should contain at least Table or Text segments
+        expected_types = {SegmentType.TABLE, SegmentType.TEXT}
+        assert len(segment_types.intersection(expected_types)) > 0, f"No expected segment types found. Got: {segment_types}"
+
+    @pytest.mark.asyncio
+    async def test_table_segments_have_cells(self, client, excel_sample_path, excel_config):
+        """Test that TABLE segments contain cell data"""
+        response = await client.upload(excel_sample_path, excel_config)
+        
+        table_segments = []
+        for chunk in response.output.chunks:
+            for segment in chunk.segments:
+                if segment.segment_type == SegmentType.TABLE:
+                    table_segments.append(segment)
+        
+        if table_segments:  # If we have table segments, they should have cells
+            found_cells = False
+            for segment in table_segments:
+                if segment.ss_cells and len(segment.ss_cells) > 0:
+                    found_cells = True
+                    break
+            assert found_cells, "TABLE segments should contain cell data"
+
+
+class TestExcelEmbedding:
+    """Test Excel embedding functionality"""
+
+    @pytest.mark.asyncio
+    async def test_chunks_have_embed_content(self, client, excel_sample_path, excel_config):
+        """Test that chunks generate embed content for Excel data"""
+        response = await client.upload(excel_sample_path, excel_config)
+        
+        # At least some chunks should have embed content
+        chunks_with_embed = [chunk for chunk in response.output.chunks if chunk.embed]
+        assert len(chunks_with_embed) > 0, "No chunks with embed content found"
+        
+        # Embed content should not be empty
+        for chunk in chunks_with_embed:
+            assert len(chunk.embed.strip()) > 0, "Empty embed content found"
+
+    @pytest.mark.asyncio
+    async def test_segment_length_calculation(self, client, excel_sample_path, excel_config):
+        """Test that segments have length calculations"""
+        response = await client.upload(excel_sample_path, excel_config)
+        
+        segments_with_length = []
+        for chunk in response.output.chunks:
+            for segment in chunk.segments:
+                if segment.segment_length is not None:
+                    segments_with_length.append(segment)
+        
+        # At least some segments should have length calculations
+        assert len(segments_with_length) > 0, "No segments with length calculations found"
+        
+        # Lengths should be positive
+        for segment in segments_with_length:
+            assert segment.segment_length > 0, f"Invalid segment length: {segment.segment_length}"
+
+
+class TestExcelEdgeCases:
+    """Test edge cases and error handling for Excel processing"""
+
+    @pytest.mark.asyncio
+    async def test_empty_cells_handling(self, client, excel_sample_path, excel_config):
+        """Test that empty cells are handled properly"""
+        response = await client.upload(excel_sample_path, excel_config)
+        
+        # Look for cells that might be empty
+        all_cells = []
+        for chunk in response.output.chunks:
+            for segment in chunk.segments:
+                if segment.ss_cells:
+                    all_cells.extend(segment.ss_cells)
+        
+        assert len(all_cells) > 0, "No cells found to test"
+        
+        # All cells should have a text field, even if empty
+        for cell in all_cells:
+            assert hasattr(cell, 'text'), "Cell missing text field"
+            assert cell.text is not None, "Cell text is None"
+
+    @pytest.mark.asyncio
+    async def test_range_format_validity(self, client, excel_sample_path, excel_config):
+        """Test that Excel ranges follow expected format"""
+        response = await client.upload(excel_sample_path, excel_config)
+        
+        ranges = []
+        for chunk in response.output.chunks:
+            for segment in chunk.segments:
+                if segment.ss_range:
+                    ranges.append(segment.ss_range)
+                if segment.ss_cells:
+                    for cell in segment.ss_cells:
+                        ranges.append(cell.range)
+        
+        assert len(ranges) > 0, "No ranges found to test"
+        
+        # Basic range format validation (e.g., "A1", "A1:B2")
+        import re
+        range_pattern = re.compile(r'^[A-Z]+\d+(:[A-Z]+\d+)?$')
+        valid_ranges = [r for r in ranges if range_pattern.match(r)]
+        
+        # Most ranges should follow the expected format
+        assert len(valid_ranges) > 0, f"No valid ranges found. Ranges: {ranges[:10]}..."
+
+
+# Integration test using the expected output fixture
+class TestExcelIntegration:
+    """Integration tests comparing against expected output"""
+
+    @pytest.mark.asyncio
+    async def test_compare_with_expected_structure(self, client, excel_sample_path, excel_config, excel_expected_output):
+        """Test that the output structure matches expected format"""
+        response = await client.upload(excel_sample_path, excel_config)
+        
+        expected = excel_expected_output["output"]
+        actual = response.output
+        
+        # Compare high-level structure
+        assert actual.mime_type == expected["mime_type"]
+        assert actual.page_count == expected["page_count"]
+        assert len(actual.chunks) > 0
+        assert len(actual.pages) > 0
+        
+        # Verify that we have similar data structure
+        expected_has_cells = any(
+            segment.get("ss_cells") 
+            for chunk in expected["chunks"] 
+            for segment in chunk["segments"]
+        )
+        actual_has_cells = any(
+            segment.ss_cells 
+            for chunk in actual.chunks 
+            for segment in chunk.segments 
+            if segment.ss_cells
+        )
+        
+        if expected_has_cells:
+            assert actual_has_cells, "Expected cells in output but none found" 

chunkr_ai-0.3.0/tests/test_pages.py

@@ -0,0 +1,261 @@
+import pytest
+from pathlib import Path
+
+from chunkr_ai import Chunkr
+from chunkr_ai.models import (
+    Configuration,
+    ChunkProcessing,
+    SegmentProcessing,
+    GenerationConfig,
+    SegmentFormat,
+    GenerationStrategy,
+    EmbedSource,
+    Tokenizer,
+    OcrStrategy,
+    SegmentationStrategy,
+    Page,
+)
+
+
+@pytest.fixture
+def client():
+    """Chunkr client instance"""
+    client = Chunkr()
+    yield client
+
+
+@pytest.fixture
+def sample_pdf_path():
+    """Path to the PDF test file"""
+    return Path("tests/files/test.pdf")
+
+
+@pytest.fixture
+def excel_sample_path():
+    """Path to the Excel test file"""
+    return Path("tests/files/excel/test.xlsx")
+
+
+@pytest.fixture
+def basic_config():
+    """Basic configuration for testing pages"""
+    return Configuration(
+        high_resolution=True,
+        ocr_strategy=OcrStrategy.ALL,
+        segmentation_strategy=SegmentationStrategy.LAYOUT_ANALYSIS,
+    )
+
+
+class TestPagesBasicFunctionality:
+    """Test basic pages functionality across different file types"""
+
+    @pytest.mark.asyncio
+    async def test_pdf_generates_pages(self, client, sample_pdf_path, basic_config):
+        """Test that PDF files generate pages information"""
+        response = await client.upload(sample_pdf_path, basic_config)
+        
+        assert response.task_id is not None
+        assert response.status == "Succeeded"
+        assert response.output is not None
+        
+        # Test pages structure
+        if response.output.pages:  # Pages might be optional for some file types
+            assert len(response.output.pages) > 0
+            page = response.output.pages[0]
+            assert page.image is not None
+            assert page.page_number is not None
+            assert page.page_height is not None
+            assert page.page_width is not None
+
+    @pytest.mark.asyncio
+    async def test_excel_generates_pages_with_sheet_info(self, client, excel_sample_path, basic_config):
+        """Test that Excel files generate pages with sheet information"""
+        response = await client.upload(excel_sample_path, basic_config)
+        
+        assert response.task_id is not None
+        assert response.status == "Succeeded"
+        assert response.output is not None
+        
+        # Excel should definitely have pages
+        assert response.output.pages is not None
+        assert len(response.output.pages) > 0
+        
+        page = response.output.pages[0]
+        assert page.image is not None
+        assert page.page_number is not None
+        assert page.page_height is not None
+        assert page.page_width is not None
+        # Excel pages should have sheet names
+        assert page.ss_sheet_name is not None
+
+    @pytest.mark.asyncio
+    async def test_page_count_consistency(self, client, sample_pdf_path, basic_config):
+        """Test that page_count matches the actual number of pages"""
+        response = await client.upload(sample_pdf_path, basic_config)
+        
+        assert response.output.page_count is not None
+        if response.output.pages:
+            assert response.output.page_count == len(response.output.pages)
+        else:
+            # If no pages array, page_count should still be meaningful
+            assert response.output.page_count > 0
+
+
+class TestPageStructure:
+    """Test the Page model structure and validation"""
+
+    @pytest.mark.asyncio
+    async def test_page_required_fields(self, client, sample_pdf_path, basic_config):
+        """Test that Page objects have all required fields"""
+        response = await client.upload(sample_pdf_path, basic_config)
+        
+        if response.output.pages and len(response.output.pages) > 0:
+            page = response.output.pages[0]
+            
+            # Required fields
+            assert page.image is not None
+            assert isinstance(page.page_number, int)
+            assert isinstance(page.page_height, (int, float))
+            assert isinstance(page.page_width, (int, float))
+            
+            # Optional fields should be accessible
+            assert hasattr(page, 'ss_sheet_name')
+
+    @pytest.mark.asyncio
+    async def test_page_numbers_start_from_one(self, client, sample_pdf_path, basic_config):
+        """Test that page numbers start from 1 and are sequential"""
+        response = await client.upload(sample_pdf_path, basic_config)
+        
+        if response.output.pages and len(response.output.pages) > 0:
+            page_numbers = [page.page_number for page in response.output.pages]
+            page_numbers.sort()
+            
+            # Should start from 1
+            assert page_numbers[0] == 1, f"Page numbers should start from 1, got: {page_numbers[0]}"
+            
+            # Should be sequential if multiple pages
+            if len(page_numbers) > 1:
+                for i in range(1, len(page_numbers)):
+                    assert page_numbers[i] == page_numbers[i-1] + 1, f"Page numbers not sequential: {page_numbers}"
+
+    @pytest.mark.asyncio
+    async def test_page_dimensions_positive(self, client, sample_pdf_path, basic_config):
+        """Test that page dimensions are positive values"""
+        response = await client.upload(sample_pdf_path, basic_config)
+        
+        if response.output.pages:
+            for page in response.output.pages:
+                assert page.page_height > 0, f"Invalid page height: {page.page_height}"
+                assert page.page_width > 0, f"Invalid page width: {page.page_width}"
+
+    @pytest.mark.asyncio
+    async def test_page_images_are_urls(self, client, sample_pdf_path, basic_config):
+        """Test that page images are valid URLs"""
+        response = await client.upload(sample_pdf_path, basic_config)
+        
+        if response.output.pages:
+            for page in response.output.pages:
+                assert page.image.startswith(('http://', 'https://')), f"Invalid page image URL: {page.image}"
+
+
+class TestMimeTypeHandling:
+    """Test MIME type handling for different file types"""
+
+    @pytest.mark.asyncio
+    async def test_pdf_mime_type(self, client, sample_pdf_path, basic_config):
+        """Test that PDF files have correct MIME type"""
+        response = await client.upload(sample_pdf_path, basic_config)
+        
+        assert response.output.mime_type is not None
+        # Should be PDF MIME type
+        assert response.output.mime_type in [
+            "application/pdf",
+            "application/x-pdf"
+        ], f"Unexpected PDF MIME type: {response.output.mime_type}"
+
+    @pytest.mark.asyncio
+    async def test_excel_mime_type(self, client, excel_sample_path, basic_config):
+        """Test that Excel files have correct MIME type"""
+        response = await client.upload(excel_sample_path, basic_config)
+        
+        assert response.output.mime_type is not None
+        # Should be Excel MIME type
+        expected_excel_types = [
+            "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
+            "application/vnd.ms-excel"
+        ]
+        assert response.output.mime_type in expected_excel_types, f"Unexpected Excel MIME type: {response.output.mime_type}"
+
+
+class TestBackwardsCompatibility:
+    """Test that new fields don't break existing functionality"""
+
+    @pytest.mark.asyncio
+    async def test_existing_fields_still_work(self, client, sample_pdf_path, basic_config):
+        """Test that all existing fields still work with new page functionality"""
+        response = await client.upload(sample_pdf_path, basic_config)
+        
+        # Test that all traditional fields still work
+        assert response.task_id is not None
+        assert response.status == "Succeeded"
+        assert response.output is not None
+        assert response.output.chunks is not None
+        assert len(response.output.chunks) > 0
+        assert response.output.file_name is not None
+        assert response.output.page_count is not None
+        assert response.output.pdf_url is not None
+        
+        # Test chunk structure
+        chunk = response.output.chunks[0]
+        assert chunk.chunk_id is not None
+        assert chunk.chunk_length is not None
+        assert chunk.segments is not None
+        assert len(chunk.segments) > 0
+        
+        # Test segment structure
+        segment = chunk.segments[0]
+        assert segment.segment_id is not None
+        assert segment.segment_type is not None
+        assert segment.bbox is not None
+
+    @pytest.mark.asyncio
+    async def test_optional_new_fields(self, client, sample_pdf_path, basic_config):
+        """Test that new optional fields are properly handled"""
+        response = await client.upload(sample_pdf_path, basic_config)
+        
+        # New fields should be accessible but might be None
+        assert hasattr(response.output, 'mime_type')
+        assert hasattr(response.output, 'pages')
+        
+        # For segments, spreadsheet fields should be accessible but None for PDFs
+        for chunk in response.output.chunks:
+            for segment in chunk.segments:
+                assert hasattr(segment, 'ss_cells')
+                assert hasattr(segment, 'ss_range')
+                assert hasattr(segment, 'ss_sheet_name')
+                assert hasattr(segment, 'segment_length')
+                
+                # For non-Excel files, these should be None
+                if response.output.mime_type != "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet":
+                    assert segment.ss_cells is None
+                    assert segment.ss_range is None
+                    assert segment.ss_sheet_name is None
+
+
+class TestErrorHandling:
+    """Test error handling for pages functionality"""
+
+    @pytest.mark.asyncio
+    async def test_missing_pages_handled_gracefully(self, client, sample_pdf_path, basic_config):
+        """Test that missing pages are handled gracefully"""
+        response = await client.upload(sample_pdf_path, basic_config)
+        
+        # Even if pages is None, the response should be valid
+        if response.output.pages is None:
+            # page_count should still be available
+            assert response.output.page_count is not None
+            assert response.output.page_count > 0
+        else:
+            # If pages exist, they should be valid
+            assert len(response.output.pages) > 0
+            assert response.output.page_count == len(response.output.pages)