chunkr-ai 0.0.3__tar.gz → 0.0.5__tar.gz

{chunkr_ai-0.0.3/src/chunkr_ai.egg-info → chunkr_ai-0.0.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: chunkr-ai
-Version: 0.0.3
+Version: 0.0.5
 Summary: Python client for Chunkr: open source document intelligence
 Author-email: Ishaan Kapoor <ishaan@lumina.sh>
 Project-URL: Homepage, https://chunkr.ai
@@ -17,7 +17,13 @@ Requires-Dist: pytest-xdist>=3.6.1; extra == "test"
 
 # Chunkr Python Client
 
-This is the Python client for the Chunkr API. It provides a simple interface to interact with Chunkr's services.
+This provides a simple interface to interact with the Chunkr API.
+
+## Getting Started
+
+You can get an API key from [Chunkr](https://chunkr.ai) or deploy your own Chunkr instance. For self-hosted deployment options, check out our [deployment guide](https://github.com/lumina-ai-inc/chunkr/tree/main?tab=readme-ov-file#self-hosted-deployment-options).
+
+For more information about the API and its capabilities, visit the [Chunkr API docs](https://docs.chunkr.ai).
 
 ## Installation
 
@@ -102,6 +108,80 @@ chunkr.upload(img)
 
 ### Configuration
 
+You can customize the processing behavior by passing a `Configuration` object:
+
+```python
+from chunkr_ai.models import Configuration, OcrStrategy, SegmentationStrategy, GenerationStrategy
+
+# Basic configuration
+config = Configuration(
+    ocr_strategy=OcrStrategy.AUTO,
+    segmentation_strategy=SegmentationStrategy.LAYOUT_ANALYSIS,
+    high_resolution=True,
+    expires_in=3600,  # seconds
+)
+
+# Upload with configuration
+task = chunkr.upload("document.pdf", config)
+```
+
+#### Available Configuration Examples
+
+- **Chunk Processing**
+  ```python
+  from chunkr_ai.models import ChunkProcessing
+  config = Configuration(
+      chunk_processing=ChunkProcessing(target_length=1024)
+  )
+  ```
+- **Expires In**
+  ```python
+  config = Configuration(expires_in=3600)
+  ```
+
+- **High Resolution**
+  ```python
+  config = Configuration(high_resolution=True)
+  ```
+
+- **JSON Schema**
+  ```python
+  config = Configuration(json_schema=JsonSchema(
+      title="Sales Data",
+      properties=[
+          Property(name="Person with highest sales", prop_type="string", description="The person with the highest sales"),
+          Property(name="Person with lowest sales", prop_type="string", description="The person with the lowest sales"),
+      ]
+  ))
+  ```
+
+- **OCR Strategy**
+  ```python
+  config = Configuration(ocr_strategy=OcrStrategy.AUTO)
+  ```
+
+- **Segment Processing**
+  ```python
+  from chunkr_ai.models import SegmentProcessing, GenerationConfig, GenerationStrategy
+  config = Configuration(
+      segment_processing=SegmentProcessing(
+          page=GenerationConfig(
+              html=GenerationStrategy.LLM,
+              markdown=GenerationStrategy.LLM
+          )
+      )
+  )
+  ```
+
+- **Segmentation Strategy**
+  ```python
+  config = Configuration(
+      segmentation_strategy=SegmentationStrategy.LAYOUT_ANALYSIS  # or SegmentationStrategy.PAGE
+  )
+  ```
+
+## Environment setup
+
 You can provide your API key and URL in several ways:
 1. Environment variables: `CHUNKR_API_KEY` and `CHUNKR_URL`
 2. `.env` file
@@ -112,13 +192,3 @@ chunkr = Chunkr(
     url="https://api.chunkr.ai"
 )
 ```
-
-## Run tests
-
-```python
-# Install dependencies
-uv pip install -e ".[test]"
-
-# Run tests
-uv run pytest
-```

chunkr_ai-0.0.5/README.md

@@ -0,0 +1,177 @@
+# Chunkr Python Client
+
+This provides a simple interface to interact with the Chunkr API.
+
+## Getting Started
+
+You can get an API key from [Chunkr](https://chunkr.ai) or deploy your own Chunkr instance. For self-hosted deployment options, check out our [deployment guide](https://github.com/lumina-ai-inc/chunkr/tree/main?tab=readme-ov-file#self-hosted-deployment-options).
+
+For more information about the API and its capabilities, visit the [Chunkr API docs](https://docs.chunkr.ai).
+
+## Installation
+
+```bash
+pip install chunkr-ai
+```
+
+## Usage
+
+We provide two clients: `Chunkr` for synchronous operations and `ChunkrAsync` for asynchronous operations.
+
+### Synchronous Usage
+
+```python
+from chunkr_ai import Chunkr
+
+# Initialize client
+chunkr = Chunkr()
+
+# Upload a file and wait for processing
+task = chunkr.upload("document.pdf")
+
+# Print the response
+print(task)
+
+# Get output from task
+output = task.output
+
+# If you want to upload without waiting for processing
+task = chunkr.start_upload("document.pdf")
+# ... do other things ...
+task.poll()  # Check status when needed
+```
+
+### Asynchronous Usage
+
+```python
+from chunkr_ai import ChunkrAsync
+
+async def process_document():
+    # Initialize client
+    chunkr = ChunkrAsync()
+
+    # Upload a file and wait for processing
+    task = await chunkr.upload("document.pdf")
+
+    # Print the response
+    print(task)
+
+    # Get output from task
+    output = task.output
+
+    # If you want to upload without waiting for processing
+    task = await chunkr.start_upload("document.pdf")
+    # ... do other things ...
+    await task.poll_async()  # Check status when needed
+```
+
+### Additional Features
+
+Both clients support various input types:
+
+```python
+# Upload from file path
+chunkr.upload("document.pdf")
+
+# Upload from opened file
+with open("document.pdf", "rb") as f:
+    chunkr.upload(f)
+
+# Upload from URL
+chunkr.upload("https://example.com/document.pdf")
+
+# Upload from base64 string
+chunkr.upload("data:application/pdf;base64,JVBERi0xLjcKCjEgMCBvYmo...")
+
+# Upload an image
+from PIL import Image
+img = Image.open("photo.jpg")
+chunkr.upload(img)
+```
+
+### Configuration
+
+You can customize the processing behavior by passing a `Configuration` object:
+
+```python
+from chunkr_ai.models import Configuration, OcrStrategy, SegmentationStrategy, GenerationStrategy
+
+# Basic configuration
+config = Configuration(
+    ocr_strategy=OcrStrategy.AUTO,
+    segmentation_strategy=SegmentationStrategy.LAYOUT_ANALYSIS,
+    high_resolution=True,
+    expires_in=3600,  # seconds
+)
+
+# Upload with configuration
+task = chunkr.upload("document.pdf", config)
+```
+
+#### Available Configuration Examples
+
+- **Chunk Processing**
+  ```python
+  from chunkr_ai.models import ChunkProcessing
+  config = Configuration(
+      chunk_processing=ChunkProcessing(target_length=1024)
+  )
+  ```
+- **Expires In**
+  ```python
+  config = Configuration(expires_in=3600)
+  ```
+
+- **High Resolution**
+  ```python
+  config = Configuration(high_resolution=True)
+  ```
+
+- **JSON Schema**
+  ```python
+  config = Configuration(json_schema=JsonSchema(
+      title="Sales Data",
+      properties=[
+          Property(name="Person with highest sales", prop_type="string", description="The person with the highest sales"),
+          Property(name="Person with lowest sales", prop_type="string", description="The person with the lowest sales"),
+      ]
+  ))
+  ```
+
+- **OCR Strategy**
+  ```python
+  config = Configuration(ocr_strategy=OcrStrategy.AUTO)
+  ```
+
+- **Segment Processing**
+  ```python
+  from chunkr_ai.models import SegmentProcessing, GenerationConfig, GenerationStrategy
+  config = Configuration(
+      segment_processing=SegmentProcessing(
+          page=GenerationConfig(
+              html=GenerationStrategy.LLM,
+              markdown=GenerationStrategy.LLM
+          )
+      )
+  )
+  ```
+
+- **Segmentation Strategy**
+  ```python
+  config = Configuration(
+      segmentation_strategy=SegmentationStrategy.LAYOUT_ANALYSIS  # or SegmentationStrategy.PAGE
+  )
+  ```
+
+## Environment setup
+
+You can provide your API key and URL in several ways:
+1. Environment variables: `CHUNKR_API_KEY` and `CHUNKR_URL`
+2. `.env` file
+3. Direct initialization:
+```python
+chunkr = Chunkr(
+    api_key="your-api-key",
+    url="https://api.chunkr.ai"
+)
+```

{chunkr_ai-0.0.3 → chunkr_ai-0.0.5}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "chunkr-ai"
-version = "0.0.3"
+version = "0.0.5"
 authors = [{"name" = "Ishaan Kapoor", "email" = "ishaan@lumina.sh"}]
 description = "Python client for Chunkr: open source document intelligence"
 readme = "README.md"

{chunkr_ai-0.0.3 → chunkr_ai-0.0.5}/src/chunkr_ai/api/config.py

@@ -1,4 +1,4 @@
-from pydantic import BaseModel, Field
+from pydantic import BaseModel, Field, model_validator
 from enum import Enum
 from typing import Optional, List, Dict
 
@@ -40,15 +40,14 @@ class ChunkProcessing(BaseModel):
 
 class Property(BaseModel):
     name: str
-    title: Optional[str]
+    title: Optional[str] = None
     prop_type: str
-    description: Optional[str]
-    default: Optional[str]
+    description: Optional[str] = None
+    default: Optional[str] = None
 
 class JsonSchema(BaseModel):
     title: str
     properties: List[Property]
-    schema_type: Optional[str]
 
 class OcrStrategy(str, Enum):
     ALL = "All"
@@ -121,10 +120,12 @@ class Configuration(BaseModel):
     ocr_strategy: Optional[OcrStrategy] = Field(default=None)
     segment_processing: Optional[SegmentProcessing] = Field(default=None)
     segmentation_strategy: Optional[SegmentationStrategy] = Field(default=None)
-    target_chunk_length: Optional[int] = Field(default=None)
 
-class Status(str, Enum):
-    STARTING = "Starting"
-    PROCESSING = "Processing"
-    SUCCEEDED = "Succeeded"
-    FAILED = "Failed"
+    @model_validator(mode='before')
+    def map_deprecated_fields(cls, values: Dict) -> Dict:
+        if isinstance(values, dict) and "target_chunk_length" in values:
+            target_length = values.pop("target_chunk_length")
+            if target_length is not None:
+                values["chunk_processing"] = values.get("chunk_processing", {}) or {}
+                values["chunk_processing"]["target_length"] = target_length
+        return values

{chunkr_ai-0.0.3 → chunkr_ai-0.0.5}/src/chunkr_ai/api/task.py

@@ -1,11 +1,18 @@
 from .protocol import ChunkrClientProtocol
-from .config import Configuration, Status, OutputResponse
+from .config import Configuration, OutputResponse
 import asyncio
 from datetime import datetime
+from enum import Enum
 from pydantic import BaseModel, PrivateAttr
 import time
 from typing import Optional, Union
 
+class Status(str, Enum):
+    STARTING = "Starting"
+    PROCESSING = "Processing"
+    SUCCEEDED = "Succeeded"
+    FAILED = "Failed"
+    
 class TaskResponse(BaseModel):
     configuration: Configuration
     created_at: datetime

{chunkr_ai-0.0.3 → chunkr_ai-0.0.5}/src/chunkr_ai/models.py

@@ -18,10 +18,9 @@ from .api.config import (
     SegmentProcessing,
     SegmentType,
     SegmentationStrategy,
-    Status
 )
 
-from .api.task import TaskResponse, TaskPayload
+from .api.task import TaskResponse, TaskPayload, Status
 
 __all__ = [
     'BoundingBox',

{chunkr_ai-0.0.3 → chunkr_ai-0.0.5/src/chunkr_ai.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: chunkr-ai
-Version: 0.0.3
+Version: 0.0.5
 Summary: Python client for Chunkr: open source document intelligence
 Author-email: Ishaan Kapoor <ishaan@lumina.sh>
 Project-URL: Homepage, https://chunkr.ai
@@ -17,7 +17,13 @@ Requires-Dist: pytest-xdist>=3.6.1; extra == "test"
 
 # Chunkr Python Client
 
-This is the Python client for the Chunkr API. It provides a simple interface to interact with Chunkr's services.
+This provides a simple interface to interact with the Chunkr API.
+
+## Getting Started
+
+You can get an API key from [Chunkr](https://chunkr.ai) or deploy your own Chunkr instance. For self-hosted deployment options, check out our [deployment guide](https://github.com/lumina-ai-inc/chunkr/tree/main?tab=readme-ov-file#self-hosted-deployment-options).
+
+For more information about the API and its capabilities, visit the [Chunkr API docs](https://docs.chunkr.ai).
 
 ## Installation
 
@@ -102,6 +108,80 @@ chunkr.upload(img)
 
 ### Configuration
 
+You can customize the processing behavior by passing a `Configuration` object:
+
+```python
+from chunkr_ai.models import Configuration, OcrStrategy, SegmentationStrategy, GenerationStrategy
+
+# Basic configuration
+config = Configuration(
+    ocr_strategy=OcrStrategy.AUTO,
+    segmentation_strategy=SegmentationStrategy.LAYOUT_ANALYSIS,
+    high_resolution=True,
+    expires_in=3600,  # seconds
+)
+
+# Upload with configuration
+task = chunkr.upload("document.pdf", config)
+```
+
+#### Available Configuration Examples
+
+- **Chunk Processing**
+  ```python
+  from chunkr_ai.models import ChunkProcessing
+  config = Configuration(
+      chunk_processing=ChunkProcessing(target_length=1024)
+  )
+  ```
+- **Expires In**
+  ```python
+  config = Configuration(expires_in=3600)
+  ```
+
+- **High Resolution**
+  ```python
+  config = Configuration(high_resolution=True)
+  ```
+
+- **JSON Schema**
+  ```python
+  config = Configuration(json_schema=JsonSchema(
+      title="Sales Data",
+      properties=[
+          Property(name="Person with highest sales", prop_type="string", description="The person with the highest sales"),
+          Property(name="Person with lowest sales", prop_type="string", description="The person with the lowest sales"),
+      ]
+  ))
+  ```
+
+- **OCR Strategy**
+  ```python
+  config = Configuration(ocr_strategy=OcrStrategy.AUTO)
+  ```
+
+- **Segment Processing**
+  ```python
+  from chunkr_ai.models import SegmentProcessing, GenerationConfig, GenerationStrategy
+  config = Configuration(
+      segment_processing=SegmentProcessing(
+          page=GenerationConfig(
+              html=GenerationStrategy.LLM,
+              markdown=GenerationStrategy.LLM
+          )
+      )
+  )
+  ```
+
+- **Segmentation Strategy**
+  ```python
+  config = Configuration(
+      segmentation_strategy=SegmentationStrategy.LAYOUT_ANALYSIS  # or SegmentationStrategy.PAGE
+  )
+  ```
+
+## Environment setup
+
 You can provide your API key and URL in several ways:
 1. Environment variables: `CHUNKR_API_KEY` and `CHUNKR_URL`
 2. `.env` file
@@ -112,13 +192,3 @@ chunkr = Chunkr(
     url="https://api.chunkr.ai"
 )
 ```
-
-## Run tests
-
-```python
-# Install dependencies
-uv pip install -e ".[test]"
-
-# Run tests
-uv run pytest
-```

{chunkr_ai-0.0.3 → chunkr_ai-0.0.5}/tests/test_chunkr.py

@@ -8,7 +8,9 @@ from chunkr_ai.models import (
     Configuration, 
     GenerationStrategy, 
     GenerationConfig,
+    JsonSchema,
     OcrStrategy, 
+    Property,
     SegmentationStrategy, 
     SegmentProcessing, 
     TaskResponse, 
@@ -129,7 +131,21 @@ def test_page_llm(chunkr, sample_path):
     assert response.task_id is not None
     assert response.status == "Succeeded"
     assert response.output is not None
-    
+  
+def test_json_schema(chunkr, sample_path):
+    response = chunkr.upload(sample_path, Configuration(
+        json_schema=JsonSchema(
+            title="Sales Data",
+            properties=[
+                Property(name="Person with highest sales", prop_type="string", description="The person with the highest sales"),
+                Property(name="Person with lowest sales", prop_type="string", description="The person with the lowest sales"),
+            ]
+        )
+    ))
+    assert isinstance(response, TaskResponse)
+    assert response.task_id is not None
+    assert response.status == "Succeeded"
+    assert response.output is not None
 
 async def test_async_send_file_path(async_chunkr, sample_path):
     response = await async_chunkr.upload(sample_path)
@@ -138,4 +154,5 @@ async def test_async_send_file_path(async_chunkr, sample_path):
     assert response.task_id is not None
     assert response.status == "Succeeded"
     assert response.output is not None
+    
     

chunkr_ai-0.0.3/README.md

@@ -1,107 +0,0 @@
-# Chunkr Python Client
-
-This is the Python client for the Chunkr API. It provides a simple interface to interact with Chunkr's services.
-
-## Installation
-
-```bash
-pip install chunkr-ai
-```
-
-## Usage
-
-We provide two clients: `Chunkr` for synchronous operations and `ChunkrAsync` for asynchronous operations.
-
-### Synchronous Usage
-
-```python
-from chunkr_ai import Chunkr
-
-# Initialize client
-chunkr = Chunkr()
-
-# Upload a file and wait for processing
-task = chunkr.upload("document.pdf")
-
-# Print the response
-print(task)
-
-# Get output from task
-output = task.output
-
-# If you want to upload without waiting for processing
-task = chunkr.start_upload("document.pdf")
-# ... do other things ...
-task.poll()  # Check status when needed
-```
-
-### Asynchronous Usage
-
-```python
-from chunkr_ai import ChunkrAsync
-
-async def process_document():
-    # Initialize client
-    chunkr = ChunkrAsync()
-
-    # Upload a file and wait for processing
-    task = await chunkr.upload("document.pdf")
-
-    # Print the response
-    print(task)
-
-    # Get output from task
-    output = task.output
-
-    # If you want to upload without waiting for processing
-    task = await chunkr.start_upload("document.pdf")
-    # ... do other things ...
-    await task.poll_async()  # Check status when needed
-```
-
-### Additional Features
-
-Both clients support various input types:
-
-```python
-# Upload from file path
-chunkr.upload("document.pdf")
-
-# Upload from opened file
-with open("document.pdf", "rb") as f:
-    chunkr.upload(f)
-
-# Upload from URL
-chunkr.upload("https://example.com/document.pdf")
-
-# Upload from base64 string
-chunkr.upload("data:application/pdf;base64,JVBERi0xLjcKCjEgMCBvYmo...")
-
-# Upload an image
-from PIL import Image
-img = Image.open("photo.jpg")
-chunkr.upload(img)
-```
-
-### Configuration
-
-You can provide your API key and URL in several ways:
-1. Environment variables: `CHUNKR_API_KEY` and `CHUNKR_URL`
-2. `.env` file
-3. Direct initialization:
-```python
-chunkr = Chunkr(
-    api_key="your-api-key",
-    url="https://api.chunkr.ai"
-)
-```
-
-## Run tests
-
-```python
-# Install dependencies
-uv pip install -e ".[test]"
-
-# Run tests
-uv run pytest
-```