datamint 1.7.3__py3-none-any.whl → 1.7.4__py3-none-any.whl

datamint/apihandler/base_api_handler.py

@@ -138,7 +138,7 @@ class BaseAPIHandler:
             if isinstance(data, aiohttp.FormData):  # Check if it's aiohttp.FormData
                 # Handle FormData by extracting fields
                 form_parts = []
-                for options,headers,value in data._fields:
+                for options, headers, value in data._fields:
                     # get the name from options
                     name = options.get('name', 'file')
                     if hasattr(value, 'read'):  # File-like object
@@ -161,7 +161,7 @@ class BaseAPIHandler:
         if session is None:
             async with aiohttp.ClientSession() as s:
                 return await self._run_request_async(request_args, s, data_to_get)
-            
+
         async with self.semaphore:
             try:
                 _LOGGER.debug(f"Running request to {request_args['url']}")
@@ -222,7 +222,7 @@ class BaseAPIHandler:
 
     def _run_request(self,
                      request_args: dict,
-                     session: Session = None):
+                     session: Session | None = None):
         if session is None:
             with Session() as s:
                 return self._run_request(request_args, s)
@@ -281,24 +281,25 @@ class BaseAPIHandler:
     @staticmethod
     def convert_format(bytes_array: bytes,
                        mimetype: str,
-                       file_path: str = None
+                       file_path: str | None = None
                        ) -> pydicom.dataset.Dataset | Image.Image | cv2.VideoCapture | bytes | nib_FileBasedImage:
+        """ Convert the bytes array to the appropriate format based on the mimetype."""
         content_io = BytesIO(bytes_array)
-        if mimetype == 'application/dicom':
+        if mimetype.endswith('/dicom'):
             return pydicom.dcmread(content_io)
-        elif mimetype in ('image/jpeg', 'image/png', 'image/tiff'):
+        elif mimetype.startswith('image/'):
             return Image.open(content_io)
-        elif mimetype == 'video/mp4':
+        elif mimetype.startswith('video/'):
             if file_path is None:
-                raise NotImplementedError("file_path=None is not implemented yet for video/mp4.")
+                raise NotImplementedError("file_path=None is not implemented yet for video/* mimetypes.")
             return cv2.VideoCapture(file_path)
         elif mimetype == 'application/json':
             return json.loads(bytes_array)
         elif mimetype == 'application/octet-stream':
             return bytes_array
-        elif mimetype == 'application/nifti':
+        elif mimetype.endswith('nifti'):
             if file_path is None:
-                raise NotImplementedError("file_path=None is not implemented yet for application/nifti.")
+                raise NotImplementedError(f"file_path=None is not implemented yet for {mimetype}.")
             return nib.load(file_path)
 
         raise ValueError(f"Unsupported mimetype: {mimetype}")

datamint/apihandler/root_api_handler.py

@@ -8,10 +8,11 @@ import asyncio
 import aiohttp
 from medimgkit.dicom_utils import anonymize_dicom, to_bytesio, is_dicom
 from medimgkit import dicom_utils
+from medimgkit.io_utils import is_io_object
+from medimgkit.format_detection import guess_type, guess_extension
 import pydicom
 from pathlib import Path
 from datetime import date
-import mimetypes
 from PIL import Image
 import cv2
 from nibabel.filebasedimages import FileBasedImage as nib_FileBasedImage
@@ -26,13 +27,6 @@ _LOGGER = logging.getLogger(__name__)
 _USER_LOGGER = logging.getLogger('user_logger')
 
 
-def _is_io_object(obj):
-    """
-    Check if an object is a file-like object.
-    """
-    return callable(getattr(obj, "read", None))
-
-
 def _infinite_gen(x):
     while True:
         yield x
@@ -65,7 +59,7 @@ class RootAPIHandler(BaseAPIHandler):
                                             publish: bool = False,
                                             metadata_file: Optional[str | dict] = None,
                                             ) -> str:
-        if _is_io_object(file_path):
+        if is_io_object(file_path):
             name = file_path.name
         else:
             name = file_path
@@ -91,15 +85,12 @@ class RootAPIHandler(BaseAPIHandler):
             name = new_file_path
             _LOGGER.debug(f"New file path: {name}")
 
-        if mimetype is None:
-            mimetype = mimetypes.guess_type(name)[0]
         is_a_dicom_file = None
         if mimetype is None:
-            is_a_dicom_file = is_dicom(name) or is_dicom(file_path)
-            if is_a_dicom_file:
-                mimetype = 'application/dicom'
-            elif name.endswith('.nii') or name.endswith('.nii.gz'):
-                mimetype = 'application/x-nifti'
+            mimetype, _ = guess_type(file_path, use_magic=True)
+            if mimetype == 'application/gzip' and name.lower().endswith('nii.gz'):
+                # Special case for gzipped NIfTI files
+                mimetype = 'image/x.nifti'
 
         filename = os.path.basename(name)
         _LOGGER.debug(f"File name '{filename}' mimetype: {mimetype}")
@@ -451,7 +442,7 @@ class RootAPIHandler(BaseAPIHandler):
 
         files_path, is_multiple_resources = RootAPIHandler.__process_files_parameter(files_path)
 
-        ### Discard DICOM reports
+        # Discard DICOM reports
         if discard_dicom_reports:
             files_path = [f for f in files_path if not RootAPIHandler._is_dicom_report(f)]
             old_size = len(files_path)
@@ -678,7 +669,7 @@ class RootAPIHandler(BaseAPIHandler):
                 is_list = False
                 new_file_path = [file_path]
         # Check if is an IO object
-        elif _is_io_object(file_path):
+        elif is_io_object(file_path):
             is_list = False
             new_file_path = [file_path]
         elif not hasattr(file_path, '__len__'):
@@ -728,8 +719,8 @@ class RootAPIHandler(BaseAPIHandler):
 
     def get_resources(self,
                       status: Optional[ResourceStatus] = None,
-                      from_date: Optional[date] = None,
-                      to_date: Optional[date] = None,
+                      from_date: date | str | None = None,
+                      to_date: date | str | None = None,
                       tags: Optional[Sequence[str]] = None,
                       modality: Optional[str] = None,
                       mimetype: Optional[str] = None,
@@ -747,8 +738,8 @@ class RootAPIHandler(BaseAPIHandler):
 
         Args:
             status (ResourceStatus): The resource status. Possible values: 'inbox', 'published', 'archived' or None. If None, it will return all resources.
-            from_date (Optional[date]): The start date.
-            to_date (Optional[date]): The end date.
+            from_date (date | str | None): The start date.
+            to_date (date | str | None): The end date.
             tags (Optional[list[str]]): The tags to filter the resources.
             modality (Optional[str]): The modality of the resources.
             mimetype (Optional[str]): The mimetype of the resources.
@@ -767,9 +758,15 @@ class RootAPIHandler(BaseAPIHandler):
         """
         # Convert datetime objects to ISO format
         if from_date:
-            from_date = from_date.isoformat()
+            if isinstance(from_date, str):
+                date.fromisoformat(from_date)
+            else:
+                from_date = from_date.isoformat()
         if to_date:
-            to_date = to_date.isoformat()
+            if isinstance(to_date, str):
+                date.fromisoformat(to_date)
+            else:
+                to_date = to_date.isoformat()
 
         # Prepare the payload
         payload = {
@@ -953,6 +950,9 @@ class RootAPIHandler(BaseAPIHandler):
             >>> api_handler.download_resource_file('resource_id', save_path='path/to/dicomfile.dcm')
                 saves the file in the specified path.
         """
+        if save_path is None and add_extension:
+            raise ValueError("If add_extension is True, save_path must be provided.")
+
         url = f"{self._get_endpoint_url(RootAPIHandler.ENDPOINT_RESOURCES)}/{resource_id}/file"
         request_params = {'method': 'GET',
                           'headers': {'accept': 'application/octet-stream'},
@@ -964,7 +964,8 @@ class RootAPIHandler(BaseAPIHandler):
             mimetype = None
             if auto_convert or add_extension:
                 resource_info = self.get_resources_by_ids(resource_id)
-                mimetype = resource_info['mimetype']
+                mimetype = resource_info.get('mimetype', guess_type(response.content)[0])
+                
 
             if auto_convert:
                 try:
@@ -985,15 +986,15 @@ class RootAPIHandler(BaseAPIHandler):
             raise e
 
         if save_path is not None:
-            if add_extension:
-                ext = mimetypes.guess_extension(mimetype)
+            if add_extension and mimetype is not None:
+                ext = guess_extension(mimetype)
                 if ext is not None and not save_path.endswith(ext):
                     save_path += ext
             with open(save_path, 'wb') as f:
                 f.write(response.content)
 
-        if add_extension:
-            return resource_file, save_path
+            if add_extension:
+                return resource_file, save_path
         return resource_file
 
     def download_resource_frame(self,

datamint/dataset/base_dataset.py

@@ -283,10 +283,12 @@ class DatamintBaseDataset:
         """Post-process data after loading metadata."""
         self._check_integrity()
         self._calculate_dataset_length()
-        self._precompute_frame_data()
+        if self.return_frame_by_frame:
+            self._precompute_frame_data()
+        self.subset_indices = list(range(self.dataset_length))
         self._setup_labels()
 
-        if self.discard_without_annotations and self.return_frame_by_frame:
+        if self.discard_without_annotations:
             self._filter_unannotated()
 
     def _calculate_dataset_length(self) -> None:
@@ -301,9 +303,8 @@ class DatamintBaseDataset:
 
     def _precompute_frame_data(self) -> None:
         """Precompute frame-related data for efficient indexing."""
-        self.num_frames_per_resource = self.__compute_num_frames_per_resource()
-        self._cumulative_frames = np.cumsum([0] + self.num_frames_per_resource)
-        self.subset_indices = list(range(self.dataset_length))
+        num_frames_per_resource = self.__compute_num_frames_per_resource()
+        self._cumulative_frames = np.cumsum([0] + num_frames_per_resource)
 
     def _setup_labels(self) -> None:
         """Setup label sets and mappings."""
@@ -989,9 +990,11 @@ class DatamintBaseDataset:
             return Path(resource['file'])
         else:
             ext = guess_extension(resource['mimetype'], strict=False)
+            _LOGGER.debug(f"Guessed extension for resource {resource['id']}|{resource['mimetype']}: {ext}")
             if ext is None:
                 _LOGGER.warning(f"Could not guess extension for resource {resource['id']}.")
                 ext = ''
+                raise Exception
             return Path('images', f"{resource['id']}{ext}")
 
     def _get_annotation_file_path(self, annotation: dict | Annotation) -> Path | None:

{datamint-1.7.3.dist-info → datamint-1.7.4.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: datamint
-Version: 1.7.3
+Version: 1.7.4
 Summary: A library for interacting with the Datamint API, designed for efficient data management, processing and Deep Learning workflows.
 Requires-Python: >=3.10
 Classifier: Programming Language :: Python :: 3
@@ -19,7 +19,7 @@ Requires-Dist: humanize (>=4.0.0,<5.0.0)
 Requires-Dist: lazy-loader (>=0.3.0)
 Requires-Dist: lightning
 Requires-Dist: matplotlib
-Requires-Dist: medimgkit
+Requires-Dist: medimgkit (>=0.2.0)
 Requires-Dist: nest-asyncio (>=1.0.0,<2.0.0)
 Requires-Dist: nibabel (>=4.0.0)
 Requires-Dist: numpy

{datamint-1.7.3.dist-info → datamint-1.7.4.dist-info}/RECORD

@@ -1,17 +1,17 @@
 datamint/__init__.py,sha256=7rKCCsaa4RBRTIfuHB708rai1xwDHLtkFNFJGKYG5D4,757
 datamint/apihandler/annotation_api_handler.py,sha256=HnWiG2ebq08mdaazTXVbkuwvh6fmKIKt8uqAOf3Y1jU,57013
 datamint/apihandler/api_handler.py,sha256=cdVSddrFCKlF_BJ81LO1aJ0OP49rssjpNEFzJ6Q7YyY,384
-datamint/apihandler/base_api_handler.py,sha256=dWLiowyuP1XiGfNc-_D5oQSmJdfR0Quy0ToeJlkRA8s,11911
+datamint/apihandler/base_api_handler.py,sha256=t2no7gTIdPFfR_TXlZmh3rsncaL9p1G8eIS2m9Q7ALE,11978
 datamint/apihandler/dto/annotation_dto.py,sha256=qId1RK1VO7dXrvGJ7dqJ31jBQB7Z8yy5x0tLSiMxTB4,7105
 datamint/apihandler/exp_api_handler.py,sha256=hFUgUgBc5rL7odK7gTW3MnrvMY1pVfJUpUdzRNobMQE,6226
-datamint/apihandler/root_api_handler.py,sha256=aPJSPYEedm3oBKyEMJAz5J2l3eh1iw_nDVloWgQoeR4,56857
+datamint/apihandler/root_api_handler.py,sha256=IHpr2Bki1WPTKzM133U68c7p_50Fq5t-JEkqGFlSl8k,57152
 datamint/client_cmd_tools/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 datamint/client_cmd_tools/datamint_config.py,sha256=md7dnWrbl10lPtXKbmD9yo6onLJsajeG8Vz0ZWH1v4M,8181
 datamint/client_cmd_tools/datamint_upload.py,sha256=890USkrtaH23mNjTRYVHWce2q9jSmkUNHIP_e8fnJRM,29502
 datamint/configs.py,sha256=Bdp6NydYwyCJ2dk19_gf_o3M2ZyQOmMHpLi8wEWNHUk,1426
 datamint/dataset/__init__.py,sha256=4PlUKSvVhdfQvvuq8jQXrkdqnot-iTTizM3aM1vgSwg,47
 datamint/dataset/annotation.py,sha256=qN1IMjdfLD2ceQ6va3l76jOXA8Vb_c-eBk1oWQu6hW0,7994
-datamint/dataset/base_dataset.py,sha256=9aMzfrBBw8_Atro4j325OgBPXSQtcbRwXrdIpS7bM3Q,49251
+datamint/dataset/base_dataset.py,sha256=vYM1Q7uVdtF6DfkbJjTjdleF-kG612rSBa2c2ynohX4,49393
 datamint/dataset/dataset.py,sha256=fdcgxB9NKvPEdr9S6TOeAIqFW38PdhmCiYsit6u5Wxc,27314
 datamint/examples/__init__.py,sha256=zcYnd5nLVme9GCTPYH-1JpGo8xXK2WEYvhzcy_2alZc,39
 datamint/examples/example_projects.py,sha256=7Nb_EaIdzJTQa9zopqc-WhTBQWQJSoQZ_KjRS4PB4FI,2931
@@ -22,7 +22,7 @@ datamint/logging.yaml,sha256=a5dsATpul7QHeUHB2TjABFjWaPXBMbO--dgn8GlRqwk,483
 datamint/utils/logging_utils.py,sha256=DvoA35ATYG3JTwfXEXYawDyKRfHeCrH0a9czfkmz8kM,1851
 datamint/utils/torchmetrics.py,sha256=lwU0nOtsSWfebyp7dvjlAggaqXtj5ohSEUXOg3L0hJE,2837
 datamint/utils/visualization.py,sha256=yaUVAOHar59VrGUjpAWv5eVvQSfztFG0eP9p5Vt3l-M,4470
-datamint-1.7.3.dist-info/METADATA,sha256=YTALWrY1wwiyjk2a-4FOx_LBxTKxd__QrchCCukTcDA,4090
-datamint-1.7.3.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
-datamint-1.7.3.dist-info/entry_points.txt,sha256=mn5H6jPjO-rY0W0CAZ6Z_KKWhMLvyVaSpoqk77jlTI4,145
-datamint-1.7.3.dist-info/RECORD,,
+datamint-1.7.4.dist-info/METADATA,sha256=Fej5PDwzs3NQPuPf9ijJt0x57sY5f4Y9JGx_IEAwkX4,4100
+datamint-1.7.4.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
+datamint-1.7.4.dist-info/entry_points.txt,sha256=mn5H6jPjO-rY0W0CAZ6Z_KKWhMLvyVaSpoqk77jlTI4,145
+datamint-1.7.4.dist-info/RECORD,,