chunkr-ai 0.3.1__py3-none-any.whl → 0.3.3__py3-none-any.whl

chunkr_ai/__init__.py

@@ -1,3 +1,14 @@
 from .api.chunkr import Chunkr
+import tomllib
+from pathlib import Path
 
-__all__ = ["Chunkr"]
+# Read version from pyproject.toml
+try:
+    pyproject_path = Path(__file__).parent.parent.parent / "pyproject.toml"
+    with open(pyproject_path, "rb") as f:
+        pyproject_data = tomllib.load(f)
+    __version__ = pyproject_data["project"]["version"]
+except Exception:
+    __version__ = "unknown"
+
+__all__ = ["Chunkr", "__version__"]

chunkr_ai/api/auth.py

@@ -1,3 +1,28 @@
+import platform
+import sys
+import tomllib
+from pathlib import Path
+
+def _find_pyproject_toml(start_path: Path) -> Path | None:
+    """Search for pyproject.toml in current and parent directories."""
+    for parent in [start_path, *start_path.parents]:
+        candidate = parent / "pyproject.toml"
+        if candidate.is_file():
+            return candidate
+    return None
+
+# Read version from pyproject.toml
+try:
+    pyproject_path = _find_pyproject_toml(Path(__file__).resolve().parent)
+    if pyproject_path is not None:
+        with open(pyproject_path, "rb") as f:
+            pyproject_data = tomllib.load(f)
+        __version__ = pyproject_data["project"]["version"]
+    else:
+        __version__ = "unknown"
+except Exception:
+    __version__ = "unknown"
+
 class HeadersMixin:
     """Mixin class for handling authorization headers"""
     _api_key: str = ""
@@ -9,5 +34,9 @@ class HeadersMixin:
         return self._api_key
 
     def _headers(self) -> dict:
-        """Generate authorization headers"""
-        return {"Authorization": self.get_api_key()}
+        """Generate authorization headers and version information"""
+        user_agent = f"chunkr-ai/{__version__} (Python/{sys.version.split()[0]}; {platform.system()}/{platform.release()})"
+        return {
+            "Authorization": self.get_api_key(),
+            "User-Agent": user_agent
+        }

chunkr_ai/api/chunkr.py

@@ -42,7 +42,7 @@ class Chunkr(ChunkrBase):
         data = await prepare_upload_data(file, filename, config)
         assert self._client is not None
         r = await self._client.post(
-            f"{self.url}/api/v1/task/parse", json=data, headers=self._headers()
+            f"{self.url}/task/parse", json=data, headers=self._headers()
         )
         r.raise_for_status()
         return TaskResponse(**r.json()).with_client(cast(ChunkrClientProtocol, self), True, False)
@@ -55,7 +55,7 @@ class Chunkr(ChunkrBase):
         data = await prepare_upload_data(None, None, config)
         assert self._client is not None
         r = await self._client.patch(
-            f"{self.url}/api/v1/task/{task_id}/parse",
+            f"{self.url}/task/{task_id}/parse",
             json=data,
             headers=self._headers(),
         )
@@ -71,7 +71,7 @@ class Chunkr(ChunkrBase):
         }
         assert self._client is not None
         r = await self._client.get(
-            f"{self.url}/api/v1/task/{task_id}",
+            f"{self.url}/task/{task_id}",
             params=params,
             headers=self._headers()
         )
@@ -83,7 +83,7 @@ class Chunkr(ChunkrBase):
     async def delete_task(self, task_id: str) -> None:
         assert self._client is not None
         r = await self._client.delete(
-            f"{self.url}/api/v1/task/{task_id}", headers=self._headers()
+            f"{self.url}/task/{task_id}", headers=self._headers()
         )
         r.raise_for_status()
 
@@ -92,7 +92,7 @@ class Chunkr(ChunkrBase):
     async def cancel_task(self, task_id: str) -> None:
         assert self._client is not None
         r = await self._client.get(
-            f"{self.url}/api/v1/task/{task_id}/cancel", headers=self._headers()
+            f"{self.url}/task/{task_id}/cancel", headers=self._headers()
         )
         r.raise_for_status()
 

chunkr_ai/api/configuration.py

@@ -1,6 +1,6 @@
 from pydantic import BaseModel, Field, ConfigDict
 from enum import Enum
-from typing import Any, List, Optional, Union
+from typing import List, Optional, Union
 from pydantic import field_validator, field_serializer
 
 class CroppingStrategy(str, Enum):
@@ -20,15 +20,17 @@ class EmbedSource(str, Enum):
 class GenerationStrategy(str, Enum):
     LLM = "LLM"
     AUTO = "Auto"
+    IGNORE = "Ignore"
 
 class GenerationConfig(BaseModel):
     format: Optional[SegmentFormat] = None
     strategy: Optional[GenerationStrategy] = None
-    llm: Optional[str] = None
     crop_image: Optional[CroppingStrategy] = None
-    embed_sources: Optional[List[EmbedSource]] = None
     extended_context: Optional[bool] = None
+    description: Optional[bool] = None
     # Deprecated fields for backwards compatibility
+    llm: Optional[str] = None # Deprecated
+    embed_sources: Optional[List[EmbedSource]] = None # Deprecated
     html: Optional[GenerationStrategy] = None  # Deprecated: Use format=SegmentFormat.HTML and strategy instead
     markdown: Optional[GenerationStrategy] = None  # Deprecated: Use format=SegmentFormat.MARKDOWN and strategy instead
 
@@ -83,7 +85,7 @@ class TokenizerType(BaseModel):
         return {}
 
 class ChunkProcessing(BaseModel):
-    ignore_headers_and_footers: Optional[bool] = True
+    ignore_headers_and_footers: Optional[bool] = None # Deprecated
     target_length: Optional[int] = None
     tokenizer: Optional[Union[TokenizerType, Tokenizer, str]] = None
 
@@ -286,6 +288,7 @@ class Page(BaseModel):
     page_height: float
     page_width: float
     ss_sheet_name: Optional[str] = None
+    dpi: Optional[float] = None
 
 class Segment(BaseModel):
     bbox: BoundingBox
@@ -303,6 +306,8 @@ class Segment(BaseModel):
     confidence: Optional[float]
     text: str = ""
     segment_length: Optional[int] = None
+    embed: Optional[str] = None
+    description: Optional[str] = None
     # Spreadsheet-specific fields
     ss_cells: Optional[List[Cell]] = None
     ss_header_bbox: Optional[BoundingBox] = None
@@ -317,6 +322,7 @@ class Chunk(BaseModel):
     chunk_length: int
     segments: List[Segment]
     embed: Optional[str] = None
+    content: Optional[str] = None
 
 class OutputResponse(BaseModel):
     chunks: List[Chunk]
@@ -347,10 +353,6 @@ class Configuration(BaseModel):
     
 class OutputConfiguration(Configuration):
     input_file_url: Optional[str] = None
-    # Deprecated
-    json_schema: Optional[Any] = None
-    model: Optional[Model] = None
-    target_chunk_length: Optional[int] = None
     
 class Status(str, Enum):
     STARTING = "Starting"

chunkr_ai/api/decorators.py

@@ -2,10 +2,12 @@ import asyncio
 import functools
 import httpx
 import nest_asyncio
-from typing import Callable, Any, TypeVar, Awaitable, Union, overload
-try:
+from typing import Callable, Any, TypeVar, Awaitable, Union
+import sys
+
+if sys.version_info >= (3, 10):
     from typing import ParamSpec
-except ImportError:
+else:
     from typing_extensions import ParamSpec
 
 T = TypeVar('T')

chunkr_ai/api/misc.py

@@ -3,7 +3,7 @@ import base64
 import io
 from pathlib import Path
 from PIL import Image
-from typing import Union, Tuple, BinaryIO, Optional, Any
+from typing import Union, Tuple, BinaryIO, Optional
 
 async def prepare_file(file: Union[str, Path, BinaryIO, Image.Image, bytes, bytearray, memoryview]) -> Tuple[Optional[str], str]:
     """Convert various file types into a tuple of (filename, file content).
@@ -39,7 +39,7 @@ async def prepare_file(file: Union[str, Path, BinaryIO, Image.Image, bytes, byte
             base64.b64decode(potential_base64)
             # If we get here, it was a valid base64 string in bytes form
             return None, potential_base64
-        except:
+        except Exception:
             # Not a base64 string in bytes form, encode it as base64
             base64_str = base64.b64encode(file_bytes).decode()
             return None, base64_str
@@ -66,14 +66,14 @@ async def prepare_file(file: Union[str, Path, BinaryIO, Image.Image, bytes, byte
                     # Just test if it's valid base64, don't store the result
                     base64.b64decode(file)
                     return None, file
-                except:
+                except Exception:
                     raise ValueError(f"File not found: {file} and it's not a valid base64 string")
         except Exception as e:
             # If string can't be converted to Path or decoded as base64, it might still be a base64 string
             try:
-                base64.b64decode(file)
-                return None, file
-            except:
+                base64.b64decode(str(file))
+                return None, str(file)
+            except Exception:
                 raise ValueError(f"Unable to process file: {e}")
 
     # Handle file paths - convert to base64

chunkr_ai/api/task_response.py

@@ -1,5 +1,5 @@
 from datetime import datetime
-from typing import Optional, cast, Awaitable, Union
+from typing import Optional, cast, Union
 from pydantic import BaseModel, PrivateAttr
 import asyncio
 import json

{chunkr_ai-0.3.1.dist-info → chunkr_ai-0.3.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: chunkr-ai
-Version: 0.3.1
+Version: 0.3.3
 Summary: Python client for Chunkr: open source document intelligence
 Author-email: Ishaan Kapoor <ishaan@lumina.sh>
 License: MIT License

chunkr_ai-0.3.3.dist-info/RECORD

@@ -0,0 +1,16 @@
+chunkr_ai/__init__.py,sha256=xkXAzwvm1cFfrdJOOoZ2w9yBoudz_H6OF2HJimOma5I,409
+chunkr_ai/models.py,sha256=NvFJOpsgzEyYHhE-flp7Yr9tpTDvFmF4T87jttFRquU,1202
+chunkr_ai/api/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+chunkr_ai/api/auth.py,sha256=eb0XykbDvWgNWCkeI19XInDNnuU06s_Y-_KC89LhWH0,1408
+chunkr_ai/api/chunkr.py,sha256=3QAlZeq8zbiHp1HxgWpBBUAmvjabD9iBZxIkuGVsKJk,3822
+chunkr_ai/api/chunkr_base.py,sha256=8roSPoCADmaXM2r7zz2iHfZzIcY9NopOfa4j-dfk8RA,6310
+chunkr_ai/api/configuration.py,sha256=YDSN-hv5VyffKxDJBUaVE3u27BsDafxGNGfOBVRmRUk,11682
+chunkr_ai/api/decorators.py,sha256=B-neL5d4N-skq2rjnOfaCVSTz6HEye6udcykacbv7G4,4399
+chunkr_ai/api/misc.py,sha256=pNjbiD5reMdDSkjNTWHn0VgTVsGYn0fl751WuRtSkL8,5389
+chunkr_ai/api/protocol.py,sha256=LjPrYSq52m1afIlAo0yVGXlGZxPRh8J6g7S4PAit3Zo,388
+chunkr_ai/api/task_response.py,sha256=omnKkACjN3ijnbG6UncHLI_IDbsaJ1wHu1g4X9JbijU,8017
+chunkr_ai-0.3.3.dist-info/licenses/LICENSE,sha256=w3R12yNDyZpMiy2lxy_hvNbsldC75ww79sF0u11rkho,1069
+chunkr_ai-0.3.3.dist-info/METADATA,sha256=eQJqOqhXUch-11cqAC4dkEUc2ZkQxAe91vlYzbCEFkE,7086
+chunkr_ai-0.3.3.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+chunkr_ai-0.3.3.dist-info/top_level.txt,sha256=0IZY7PZIiS8bw5r4NUQRUQ-ATi-L_3vLQVq3ZLouOW8,10
+chunkr_ai-0.3.3.dist-info/RECORD,,

chunkr_ai-0.3.1.dist-info/RECORD

@@ -1,16 +0,0 @@
-chunkr_ai/__init__.py,sha256=6KpYv2lmD6S5z2kc9pqwuLP5VDHmOuu2qDZArUIhb1s,53
-chunkr_ai/models.py,sha256=NvFJOpsgzEyYHhE-flp7Yr9tpTDvFmF4T87jttFRquU,1202
-chunkr_ai/api/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-chunkr_ai/api/auth.py,sha256=0RSNFPvHt4Nrg8qtP2xvA2KbR0J_KUe1B_tKynbq9Fc,436
-chunkr_ai/api/chunkr.py,sha256=uSNYtB_mcs4-QRKsX7wZb8yv6ayXgRrJSDNZ-EbAyvc,3857
-chunkr_ai/api/chunkr_base.py,sha256=8roSPoCADmaXM2r7zz2iHfZzIcY9NopOfa4j-dfk8RA,6310
-chunkr_ai/api/configuration.py,sha256=y_jd3K5GB-P8N3uym4wqHDVq-Rq-VT_bhqJqgKs0PVg,11586
-chunkr_ai/api/decorators.py,sha256=w1l_ZEkl99C-BO3qRTbi74sYwHDFspB1Bjt1Arv9lPc,4384
-chunkr_ai/api/misc.py,sha256=AaGLxZlMzNgVPwErskDRKc2UVGkC0JwxLXU-enPwzA0,5354
-chunkr_ai/api/protocol.py,sha256=LjPrYSq52m1afIlAo0yVGXlGZxPRh8J6g7S4PAit3Zo,388
-chunkr_ai/api/task_response.py,sha256=VYa62E08VlZUyjn2YslnY4cohdK9e53HbEzsaYIXKXM,8028
-chunkr_ai-0.3.1.dist-info/licenses/LICENSE,sha256=w3R12yNDyZpMiy2lxy_hvNbsldC75ww79sF0u11rkho,1069
-chunkr_ai-0.3.1.dist-info/METADATA,sha256=_Lg59OcvE1hpsbc3zg20yQFGQ2bpAqOXSx_o6_1UlzY,7086
-chunkr_ai-0.3.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-chunkr_ai-0.3.1.dist-info/top_level.txt,sha256=0IZY7PZIiS8bw5r4NUQRUQ-ATi-L_3vLQVq3ZLouOW8,10
-chunkr_ai-0.3.1.dist-info/RECORD,,