chunkr-ai 0.0.11__tar.gz → 0.0.14__tar.gz

Sign up to get free protection for your applications and to get access to all the features.
Files changed (32) hide show
  1. {chunkr_ai-0.0.11/src/chunkr_ai.egg-info → chunkr_ai-0.0.14}/PKG-INFO +2 -2
  2. {chunkr_ai-0.0.11 → chunkr_ai-0.0.14}/README.md +1 -1
  3. {chunkr_ai-0.0.11 → chunkr_ai-0.0.14}/pyproject.toml +1 -1
  4. chunkr_ai-0.0.14/src/chunkr_ai/api/api.py +0 -0
  5. chunkr_ai-0.0.14/src/chunkr_ai/api/base.py +173 -0
  6. chunkr_ai-0.0.14/src/chunkr_ai/api/chunkr.py +77 -0
  7. chunkr_ai-0.0.14/src/chunkr_ai/api/chunkr_async.py +114 -0
  8. chunkr_ai-0.0.11/src/chunkr_ai/api/chunkr.py → chunkr_ai-0.0.14/src/chunkr_ai/api/chunkr_base.py +63 -75
  9. {chunkr_ai-0.0.11 → chunkr_ai-0.0.14}/src/chunkr_ai/api/config.py +1 -1
  10. {chunkr_ai-0.0.11 → chunkr_ai-0.0.14}/src/chunkr_ai/api/protocol.py +4 -4
  11. chunkr_ai-0.0.14/src/chunkr_ai/api/task.py +69 -0
  12. chunkr_ai-0.0.14/src/chunkr_ai/api/task_async.py +60 -0
  13. chunkr_ai-0.0.14/src/chunkr_ai/api/task_base.py +82 -0
  14. {chunkr_ai-0.0.11 → chunkr_ai-0.0.14/src/chunkr_ai.egg-info}/PKG-INFO +2 -2
  15. {chunkr_ai-0.0.11 → chunkr_ai-0.0.14}/src/chunkr_ai.egg-info/SOURCES.txt +2 -0
  16. {chunkr_ai-0.0.11 → chunkr_ai-0.0.14}/tests/test_chunkr.py +15 -25
  17. chunkr_ai-0.0.11/src/chunkr_ai/api/chunkr_async.py +0 -144
  18. chunkr_ai-0.0.11/src/chunkr_ai/api/chunkr_base.py +0 -85
  19. chunkr_ai-0.0.11/src/chunkr_ai/api/task.py +0 -168
  20. chunkr_ai-0.0.11/src/chunkr_ai/api/task_async.py +0 -103
  21. chunkr_ai-0.0.11/src/chunkr_ai/api/task_base.py +0 -31
  22. {chunkr_ai-0.0.11 → chunkr_ai-0.0.14}/LICENSE +0 -0
  23. {chunkr_ai-0.0.11 → chunkr_ai-0.0.14}/setup.cfg +0 -0
  24. {chunkr_ai-0.0.11 → chunkr_ai-0.0.14}/src/chunkr_ai/__init__.py +0 -0
  25. {chunkr_ai-0.0.11 → chunkr_ai-0.0.14}/src/chunkr_ai/api/__init__.py +0 -0
  26. {chunkr_ai-0.0.11 → chunkr_ai-0.0.14}/src/chunkr_ai/api/auth.py +0 -0
  27. {chunkr_ai-0.0.11 → chunkr_ai-0.0.14}/src/chunkr_ai/api/misc.py +0 -0
  28. {chunkr_ai-0.0.11 → chunkr_ai-0.0.14}/src/chunkr_ai/api/schema.py +0 -0
  29. {chunkr_ai-0.0.11 → chunkr_ai-0.0.14}/src/chunkr_ai/models.py +0 -0
  30. {chunkr_ai-0.0.11 → chunkr_ai-0.0.14}/src/chunkr_ai.egg-info/dependency_links.txt +0 -0
  31. {chunkr_ai-0.0.11 → chunkr_ai-0.0.14}/src/chunkr_ai.egg-info/requires.txt +0 -0
  32. {chunkr_ai-0.0.11 → chunkr_ai-0.0.14}/src/chunkr_ai.egg-info/top_level.txt +0 -0
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.2
2
2
  Name: chunkr-ai
3
- Version: 0.0.11
3
+ Version: 0.0.14
4
4
  Summary: Python client for Chunkr: open source document intelligence
5
5
  Author-email: Ishaan Kapoor <ishaan@lumina.sh>
6
6
  Project-URL: Homepage, https://chunkr.ai
@@ -80,7 +80,7 @@ async def process_document():
80
80
  # If you want to upload without waiting for processing
81
81
  task = await chunkr.start_upload("document.pdf")
82
82
  # ... do other things ...
83
- await task.poll_async() # Check status when needed
83
+ await task.poll() # Check status when needed
84
84
  ```
85
85
 
86
86
  ### Additional Features
@@ -62,7 +62,7 @@ async def process_document():
62
62
  # If you want to upload without waiting for processing
63
63
  task = await chunkr.start_upload("document.pdf")
64
64
  # ... do other things ...
65
- await task.poll_async() # Check status when needed
65
+ await task.poll() # Check status when needed
66
66
  ```
67
67
 
68
68
  ### Additional Features
@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
4
4
 
5
5
  [project]
6
6
  name = "chunkr-ai"
7
- version = "0.0.11"
7
+ version = "0.0.14"
8
8
  authors = [{"name" = "Ishaan Kapoor", "email" = "ishaan@lumina.sh"}]
9
9
  description = "Python client for Chunkr: open source document intelligence"
10
10
  readme = "README.md"
File without changes
@@ -0,0 +1,173 @@
1
+ from .config import Configuration
2
+ from .task import TaskResponse
3
+ from .auth import HeadersMixin
4
+ from abc import abstractmethod
5
+ from dotenv import load_dotenv
6
+ import io
7
+ import json
8
+ import os
9
+ from pathlib import Path
10
+ from PIL import Image
11
+ import requests
12
+ from typing import BinaryIO, Tuple, Union
13
+
14
+ class ChunkrBase(HeadersMixin):
15
+ """Base class with shared functionality for Chunkr API clients."""
16
+
17
+ def __init__(self, url: str = None, api_key: str = None):
18
+ load_dotenv()
19
+ self.url = (
20
+ url or
21
+ os.getenv('CHUNKR_URL') or
22
+ 'https://api.chunkr.ai'
23
+ )
24
+ self._api_key = (
25
+ api_key or
26
+ os.getenv('CHUNKR_API_KEY')
27
+ )
28
+ if not self._api_key:
29
+ raise ValueError("API key must be provided either directly, in .env file, or as CHUNKR_API_KEY environment variable. You can get an api key at: https://www.chunkr.ai")
30
+
31
+ self.url = self.url.rstrip("/")
32
+
33
+ def _prepare_file(
34
+ self,
35
+ file: Union[str, Path, BinaryIO, Image.Image]
36
+ ) -> Tuple[str, BinaryIO]:
37
+ """Convert various file types into a tuple of (filename, file-like object).
38
+
39
+ Args:
40
+ file: Input file, can be:
41
+ - String or Path to a file
42
+ - URL string starting with http:// or https://
43
+ - Base64 string
44
+ - Opened binary file (mode='rb')
45
+ - PIL/Pillow Image object
46
+
47
+ Returns:
48
+ Tuple[str, BinaryIO]: (filename, file-like object) ready for upload
49
+
50
+ Raises:
51
+ FileNotFoundError: If the file path doesn't exist
52
+ TypeError: If the file type is not supported
53
+ ValueError: If the URL is invalid or unreachable
54
+ ValueError: If the MIME type is unsupported
55
+ """
56
+ # Handle URLs
57
+ if isinstance(file, str) and (file.startswith('http://') or file.startswith('https://')):
58
+ response = requests.get(file)
59
+ response.raise_for_status()
60
+ file_obj = io.BytesIO(response.content)
61
+ filename = Path(file.split('/')[-1]).name or 'downloaded_file'
62
+ return filename, file_obj
63
+
64
+ # Handle base64 strings
65
+ if isinstance(file, str) and ',' in file and ';base64,' in file:
66
+ try:
67
+ # Split header and data
68
+ header, base64_data = file.split(',', 1)
69
+ import base64
70
+ file_bytes = base64.b64decode(base64_data)
71
+ file_obj = io.BytesIO(file_bytes)
72
+
73
+ # Try to determine format from header
74
+ format = 'bin'
75
+ mime_type = header.split(':')[-1].split(';')[0].lower()
76
+
77
+ # Map MIME types to file extensions
78
+ mime_to_ext = {
79
+ 'application/pdf': 'pdf',
80
+ 'application/vnd.openxmlformats-officedocument.wordprocessingml.document': 'docx',
81
+ 'application/msword': 'doc',
82
+ 'application/vnd.openxmlformats-officedocument.presentationml.presentation': 'pptx',
83
+ 'application/vnd.ms-powerpoint': 'ppt',
84
+ 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet': 'xlsx',
85
+ 'application/vnd.ms-excel': 'xls',
86
+ 'image/jpeg': 'jpg',
87
+ 'image/png': 'png',
88
+ 'image/jpg': 'jpg'
89
+ }
90
+
91
+ if mime_type in mime_to_ext:
92
+ format = mime_to_ext[mime_type]
93
+ else:
94
+ raise ValueError(f"Unsupported MIME type: {mime_type}")
95
+
96
+ return f"file.{format}", file_obj
97
+ except Exception as e:
98
+ raise ValueError(f"Invalid base64 string: {str(e)}")
99
+
100
+ # Handle file paths
101
+ if isinstance(file, (str, Path)):
102
+ path = Path(file).resolve()
103
+ if not path.exists():
104
+ raise FileNotFoundError(f"File not found: {file}")
105
+ return path.name, open(path, 'rb')
106
+
107
+ # Handle PIL Images
108
+ if isinstance(file, Image.Image):
109
+ img_byte_arr = io.BytesIO()
110
+ format = file.format or 'PNG'
111
+ file.save(img_byte_arr, format=format)
112
+ img_byte_arr.seek(0)
113
+ return f"image.{format.lower()}", img_byte_arr
114
+
115
+ # Handle file-like objects
116
+ if hasattr(file, 'read') and hasattr(file, 'seek'):
117
+ # Try to get the filename from the file object if possible
118
+ name = getattr(file, 'name', 'document') if hasattr(file, 'name') else 'document'
119
+ return Path(name).name, file
120
+
121
+ raise TypeError(f"Unsupported file type: {type(file)}")
122
+
123
+ def _prepare_upload_data(
124
+ self,
125
+ file: Union[str, Path, BinaryIO, Image.Image],
126
+ config: Configuration = None
127
+ ) -> Tuple[dict, dict]:
128
+ """Prepare files and data dictionaries for upload.
129
+
130
+ Args:
131
+ file: The file to upload
132
+ config: Optional configuration settings
133
+
134
+ Returns:
135
+ Tuple[dict, dict]: (files dict, data dict) ready for upload
136
+ """
137
+ filename, file_obj = self._prepare_file(file)
138
+ files = {"file": (filename, file_obj)}
139
+ data = {}
140
+
141
+ if config:
142
+ config_dict = config.model_dump(mode="json", exclude_none=True)
143
+ for key, value in config_dict.items():
144
+ if isinstance(value, dict):
145
+ files[key] = (None, json.dumps(value), 'application/json')
146
+ else:
147
+ data[key] = value
148
+
149
+ return files, data
150
+
151
+ @abstractmethod
152
+ def upload(self, file: Union[str, Path, BinaryIO, Image.Image], config: Configuration = None) -> TaskResponse:
153
+ """Upload a file and wait for processing to complete.
154
+
155
+ Must be implemented by subclasses.
156
+ """
157
+ pass
158
+
159
+ @abstractmethod
160
+ def start_upload(self, file: Union[str, Path, BinaryIO, Image.Image], config: Configuration = None) -> TaskResponse:
161
+ """Upload a file for processing and immediately return the task response.
162
+
163
+ Must be implemented by subclasses.
164
+ """
165
+ pass
166
+
167
+ @abstractmethod
168
+ def get_task(self, task_id: str) -> TaskResponse:
169
+ """Get a task response by its ID.
170
+
171
+ Must be implemented by subclasses.
172
+ """
173
+ pass
@@ -0,0 +1,77 @@
1
+ from .chunkr_base import ChunkrBase
2
+ from .config import Configuration
3
+ from .task import TaskResponse
4
+ from pathlib import Path
5
+ from PIL import Image
6
+ import requests
7
+ from typing import Union, BinaryIO
8
+ from .misc import prepare_upload_data
9
+
10
+ class Chunkr(ChunkrBase):
11
+ """Chunkr API client"""
12
+
13
+ def __init__(self, url: str = None, api_key: str = None):
14
+ super().__init__(url, api_key)
15
+ self._session = requests.Session()
16
+
17
+ def upload(self, file: Union[str, Path, BinaryIO, Image.Image], config: Configuration = None) -> TaskResponse:
18
+ task = self.create_task(file, config)
19
+ return task.poll()
20
+
21
+ def update(self, task_id: str, config: Configuration) -> TaskResponse:
22
+ task = self.update_task(task_id, config)
23
+ return task.poll()
24
+
25
+ def create_task(self, file: Union[str, Path, BinaryIO, Image.Image], config: Configuration = None) -> TaskResponse:
26
+ files= prepare_upload_data(file, config)
27
+ if not self._session:
28
+ raise ValueError("Session not found")
29
+ r = self._session.post(
30
+ f"{self.url}/api/v1/task",
31
+ files=files,
32
+ headers=self._headers()
33
+ )
34
+ r.raise_for_status()
35
+ return TaskResponse(**r.json()).with_client(self)
36
+
37
+ def update_task(self, task_id: str, config: Configuration) -> TaskResponse:
38
+ files = prepare_upload_data(None, config)
39
+ if not self._session:
40
+ raise ValueError("Session not found")
41
+ r = self._session.patch(
42
+ f"{self.url}/api/v1/task/{task_id}",
43
+ files=files,
44
+ headers=self._headers()
45
+ )
46
+
47
+ r.raise_for_status()
48
+ return TaskResponse(**r.json()).with_client(self)
49
+
50
+ def get_task(self, task_id: str) -> TaskResponse:
51
+ if not self._session:
52
+ raise ValueError("Session not found")
53
+ r = self._session.get(
54
+ f"{self.url}/api/v1/task/{task_id}",
55
+ headers=self._headers()
56
+ )
57
+ r.raise_for_status()
58
+ return TaskResponse(**r.json()).with_client(self)
59
+
60
+
61
+ def delete_task(self, task_id: str) -> None:
62
+ if not self._session:
63
+ raise ValueError("Session not found")
64
+ r = self._session.delete(
65
+ f"{self.url}/api/v1/task/{task_id}",
66
+ headers=self._headers()
67
+ )
68
+ r.raise_for_status()
69
+
70
+ def cancel_task(self, task_id: str) -> None:
71
+ if not self._session:
72
+ raise ValueError("Session not found")
73
+ r = self._session.get(
74
+ f"{self.url}/api/v1/task/{task_id}/cancel",
75
+ headers=self._headers()
76
+ )
77
+ r.raise_for_status()
@@ -0,0 +1,114 @@
1
+ from .chunkr_base import ChunkrBase
2
+ from .config import Configuration
3
+ from .misc import prepare_upload_data
4
+ from .task_async import TaskResponseAsync
5
+ import httpx
6
+ from pathlib import Path
7
+ from PIL import Image
8
+ from typing import Union, BinaryIO
9
+
10
+ class ChunkrAsync(ChunkrBase):
11
+ """Asynchronous Chunkr API client"""
12
+
13
+ def __init__(self, url: str = None, api_key: str = None):
14
+ super().__init__(url, api_key)
15
+ self._client = httpx.AsyncClient()
16
+
17
+ async def upload(self, file: Union[str, Path, BinaryIO, Image.Image], config: Configuration = None) -> TaskResponseAsync:
18
+ if not self._client or self._client.is_closed:
19
+ self._client = httpx.AsyncClient()
20
+ try:
21
+ task = await self.create_task(file, config)
22
+ return await task.poll()
23
+ except Exception as e:
24
+ await self._client.aclose()
25
+ raise e
26
+
27
+ async def update(self, task_id: str, config: Configuration) -> TaskResponseAsync:
28
+ if not self._client or self._client.is_closed:
29
+ self._client = httpx.AsyncClient()
30
+ try:
31
+ task = await self.update_task(task_id, config)
32
+ return await task.poll()
33
+ except Exception as e:
34
+ await self._client.aclose()
35
+ raise e
36
+
37
+ async def create_task(self, file: Union[str, Path, BinaryIO, Image.Image], config: Configuration = None) -> TaskResponseAsync:
38
+ if not self._client or self._client.is_closed:
39
+ self._client = httpx.AsyncClient()
40
+ try:
41
+ files = prepare_upload_data(file, config)
42
+ r = await self._client.post(
43
+ f"{self.url}/api/v1/task",
44
+ files=files,
45
+ headers=self._headers()
46
+ )
47
+ r.raise_for_status()
48
+ return TaskResponseAsync(**r.json()).with_client(self)
49
+ except Exception as e:
50
+ await self._client.aclose()
51
+ raise e
52
+
53
+ async def update_task(self, task_id: str, config: Configuration) -> TaskResponseAsync:
54
+ if not self._client or self._client.is_closed:
55
+ self._client = httpx.AsyncClient()
56
+ try:
57
+ files = prepare_upload_data(None, config)
58
+ r = await self._client.patch(
59
+ f"{self.url}/api/v1/task/{task_id}",
60
+ files=files,
61
+ headers=self._headers()
62
+ )
63
+
64
+ r.raise_for_status()
65
+ return TaskResponseAsync(**r.json()).with_client(self)
66
+ except Exception as e:
67
+ await self._client.aclose()
68
+ raise e
69
+
70
+ async def get_task(self, task_id: str) -> TaskResponseAsync:
71
+ if not self._client or self._client.is_closed:
72
+ self._client = httpx.AsyncClient()
73
+ try:
74
+ r = await self._client.get(
75
+ f"{self.url}/api/v1/task/{task_id}",
76
+ headers=self._headers()
77
+ )
78
+ r.raise_for_status()
79
+ return TaskResponseAsync(**r.json()).with_client(self)
80
+ except Exception as e:
81
+ await self._client.aclose()
82
+ raise e
83
+
84
+ async def delete_task(self, task_id: str) -> None:
85
+ if not self._client or self._client.is_closed:
86
+ self._client = httpx.AsyncClient()
87
+ try:
88
+ r = await self._client.delete(
89
+ f"{self.url}/api/v1/task/{task_id}",
90
+ headers=self._headers()
91
+ )
92
+ r.raise_for_status()
93
+ except Exception as e:
94
+ await self._client.aclose()
95
+ raise e
96
+
97
+ async def cancel_task(self, task_id: str) -> None:
98
+ if not self._client or self._client.is_closed:
99
+ self._client = httpx.AsyncClient()
100
+ try:
101
+ r = await self._client.get(
102
+ f"{self.url}/api/v1/task/{task_id}/cancel",
103
+ headers=self._headers()
104
+ )
105
+ r.raise_for_status()
106
+ except Exception as e:
107
+ await self._client.aclose()
108
+ raise e
109
+
110
+ async def __aenter__(self):
111
+ return self
112
+
113
+ async def __aexit__(self, exc_type, exc_val, exc_tb):
114
+ await self._client.aclose()
@@ -1,20 +1,35 @@
1
- from .chunkr_base import ChunkrBase
2
1
  from .config import Configuration
3
2
  from .task import TaskResponse
3
+ from .task_async import TaskResponseAsync
4
+ from .auth import HeadersMixin
5
+ from abc import abstractmethod
6
+ from dotenv import load_dotenv
7
+ import os
4
8
  from pathlib import Path
5
9
  from PIL import Image
6
- import requests
7
- from typing import Union, BinaryIO
8
- from .misc import prepare_upload_data
10
+ from typing import BinaryIO, Union
9
11
 
10
- class Chunkr(ChunkrBase):
11
- """Chunkr API client"""
12
+ class ChunkrBase(HeadersMixin):
13
+ """Base class with shared functionality for Chunkr API clients."""
12
14
 
13
15
  def __init__(self, url: str = None, api_key: str = None):
14
- super().__init__(url, api_key)
15
- self._session = requests.Session()
16
+ load_dotenv()
17
+ self.url = (
18
+ url or
19
+ os.getenv('CHUNKR_URL') or
20
+ 'https://api.chunkr.ai'
21
+ )
22
+ self._api_key = (
23
+ api_key or
24
+ os.getenv('CHUNKR_API_KEY')
25
+ )
26
+ if not self._api_key:
27
+ raise ValueError("API key must be provided either directly, in .env file, or as CHUNKR_API_KEY environment variable. You can get an api key at: https://www.chunkr.ai")
28
+
29
+ self.url = self.url.rstrip("/")
16
30
 
17
- def upload(self, file: Union[str, Path, BinaryIO, Image.Image], config: Configuration = None) -> TaskResponse:
31
+ @abstractmethod
32
+ def upload(self, file: Union[str, Path, BinaryIO, Image.Image], config: Configuration = None) -> Union[TaskResponse, TaskResponseAsync]:
18
33
  """Upload a file and wait for processing to complete.
19
34
 
20
35
  Args:
@@ -22,32 +37,32 @@ class Chunkr(ChunkrBase):
22
37
  config: Configuration options for processing. Optional.
23
38
 
24
39
  Examples:
25
- ```
40
+ ```python
26
41
  # Upload from file path
27
- chunkr.upload("document.pdf")
42
+ await chunkr.upload("document.pdf")
28
43
 
44
+ # Upload from opened file
45
+ with open("document.pdf", "rb") as f:
46
+ await chunkr.upload(f)
47
+
29
48
  # Upload from URL
30
- chunkr.upload("https://example.com/document.pdf")
49
+ await chunkr.upload("https://example.com/document.pdf")
31
50
 
32
51
  # Upload from base64 string (must include MIME type header)
33
- chunkr.upload("data:application/pdf;base64,JVBERi0xLjcKCjEgMCBvYmo...")
34
-
35
- # Upload from opened file
36
- with open("document.pdf", "rb") as f:
37
- chunkr.upload(f)
52
+ await chunkr.upload("data:application/pdf;base64,JVBERi0...")
38
53
 
39
54
  # Upload an image
40
55
  from PIL import Image
41
56
  img = Image.open("photo.jpg")
42
- chunkr.upload(img)
57
+ await chunkr.upload(img)
43
58
  ```
44
59
  Returns:
45
60
  TaskResponse: The completed task response
46
61
  """
47
- task = self.create_task(file, config)
48
- return task.poll()
62
+ pass
49
63
 
50
- def update(self, task_id: str, config: Configuration) -> TaskResponse:
64
+ @abstractmethod
65
+ def update(self, task_id: str, config: Configuration) -> Union[TaskResponse, TaskResponseAsync]:
51
66
  """Update a task by its ID and wait for processing to complete.
52
67
 
53
68
  Args:
@@ -57,11 +72,11 @@ class Chunkr(ChunkrBase):
57
72
  Returns:
58
73
  TaskResponse: The updated task response
59
74
  """
60
- task = self.update_task(task_id, config)
61
- return task.poll()
75
+ pass
62
76
 
63
- def create_task(self, file: Union[str, Path, BinaryIO, Image.Image], config: Configuration = None) -> TaskResponse:
64
- """Upload a file for processing and immediately return the task response. It will not wait for processing to complete. To wait for the full processing to complete, use `task.poll()`
77
+ @abstractmethod
78
+ def create_task(self, file: Union[str, Path, BinaryIO, Image.Image], config: Configuration = None) -> Union[TaskResponse, TaskResponseAsync]:
79
+ """Upload a file for processing and immediately return the task response. It will not wait for processing to complete. To wait for the full processing to complete, use `task.poll()`.
65
80
 
66
81
  Args:
67
82
  file: The file to upload.
@@ -70,60 +85,44 @@ class Chunkr(ChunkrBase):
70
85
  Examples:
71
86
  ```
72
87
  # Upload from file path
73
- task = chunkr.start_upload("document.pdf")
88
+ task = await chunkr.create_task("document.pdf")
74
89
 
75
90
  # Upload from opened file
76
91
  with open("document.pdf", "rb") as f:
77
- task = chunkr.start_upload(f)
78
-
92
+ task = await chunkr.create_task(f)
93
+
79
94
  # Upload from URL
80
- task = chunkr.start_upload("https://example.com/document.pdf")
95
+ task = await chunkr.create_task("https://example.com/document.pdf")
81
96
 
82
97
  # Upload from base64 string (must include MIME type header)
83
- task = chunkr.start_upload("data:application/pdf;base64,JVBERi0xLjcKCjEgMCBvYmo...")
98
+ task = await chunkr.create_task("data:application/pdf;base64,JVBERi0xLjcKCjEgMCBvYmo...")
84
99
 
85
100
  # Upload an image
86
101
  from PIL import Image
87
102
  img = Image.open("photo.jpg")
88
- task = chunkr.start_upload(img)
103
+ task = await chunkr.create_task(img)
89
104
 
90
105
  # Wait for the task to complete - this can be done when needed
91
- task.poll()
106
+ await task.poll()
92
107
  ```
93
-
94
- Returns:
95
- TaskResponse: The initial task response
96
108
  """
97
- files= prepare_upload_data(file, config)
98
- r = self._session.post(
99
- f"{self.url}/api/v1/task",
100
- files=files,
101
- headers=self._headers()
102
- )
103
- r.raise_for_status()
104
- return TaskResponse(**r.json()).with_client(self)
105
-
106
- def update_task(self, task_id: str, config: Configuration) -> TaskResponse:
107
- """Update a task by its ID.
109
+ pass
110
+
111
+ @abstractmethod
112
+ def update_task(self, task_id: str, config: Configuration) -> Union[TaskResponse, TaskResponseAsync]:
113
+ """Update a task by its ID and immediately return the task response. It will not wait for processing to complete. To wait for the full processing to complete, use `task.poll()`.
108
114
 
109
115
  Args:
110
116
  task_id: The ID of the task to update
111
- config: The new configuration to use
117
+ config: Configuration options for processing. Optional.
112
118
 
113
119
  Returns:
114
120
  TaskResponse: The updated task response
115
121
  """
116
- files = prepare_upload_data(None, config)
117
- r = self._session.patch(
118
- f"{self.url}/api/v1/task/{task_id}",
119
- files=files,
120
- headers=self._headers()
121
- )
122
-
123
- r.raise_for_status()
124
- return TaskResponse(**r.json()).with_client(self)
122
+ pass
125
123
 
126
- def get_task(self, task_id: str) -> TaskResponse:
124
+ @abstractmethod
125
+ def get_task(self, task_id: str) -> Union[TaskResponse, TaskResponseAsync]:
127
126
  """Get a task response by its ID.
128
127
 
129
128
  Args:
@@ -132,34 +131,23 @@ class Chunkr(ChunkrBase):
132
131
  Returns:
133
132
  TaskResponse: The task response
134
133
  """
135
- r = self._session.get(
136
- f"{self.url}/api/v1/task/{task_id}",
137
- headers=self._headers()
138
- )
139
- r.raise_for_status()
140
- return TaskResponse(**r.json()).with_client(self)
141
-
134
+ pass
142
135
 
136
+ @abstractmethod
143
137
  def delete_task(self, task_id: str) -> None:
144
138
  """Delete a task by its ID.
145
139
 
146
140
  Args:
147
141
  task_id: The ID of the task to delete
148
142
  """
149
- r = self._session.delete(
150
- f"{self.url}/api/v1/task/{task_id}",
151
- headers=self._headers()
152
- )
153
- r.raise_for_status()
154
-
143
+ pass
144
+
145
+ @abstractmethod
155
146
  def cancel_task(self, task_id: str) -> None:
156
147
  """Cancel a task by its ID.
157
148
 
158
149
  Args:
159
150
  task_id: The ID of the task to cancel
160
151
  """
161
- r = self._session.get(
162
- f"{self.url}/api/v1/task/{task_id}/cancel",
163
- headers=self._headers()
164
- )
165
- r.raise_for_status()
152
+ pass
153
+
@@ -142,4 +142,4 @@ class Status(str, Enum):
142
142
  PROCESSING = "Processing"
143
143
  SUCCEEDED = "Succeeded"
144
144
  FAILED = "Failed"
145
- CANCELLED = "Cancelled"
145
+ CANCELLED = "Cancelled"
@@ -1,14 +1,14 @@
1
- from typing import runtime_checkable, Protocol
1
+ from typing import Optional, runtime_checkable, Protocol
2
2
  from requests import Session
3
- from httpx import AsyncClient
3
+ from aiohttp import ClientSession
4
4
 
5
5
  @runtime_checkable
6
6
  class ChunkrClientProtocol(Protocol):
7
7
  """Protocol defining the interface for Chunkr clients"""
8
8
  url: str
9
9
  _api_key: str
10
- _session: Session
11
- _client: AsyncClient
10
+ _session: Optional[Session] = None
11
+ _client: Optional[ClientSession] = None
12
12
 
13
13
  def get_api_key(self) -> str:
14
14
  """Get the API key"""