synapse-sdk 1.0.0b4__tar.gz → 1.0.0b6__tar.gz

{synapse_sdk-1.0.0b4/synapse_sdk.egg-info → synapse_sdk-1.0.0b6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: synapse-sdk
-Version: 1.0.0b4
+Version: 1.0.0b6
 Summary: synapse sdk
 Author-email: datamaker <developer@datamaker.io>
 License: MIT

{synapse_sdk-1.0.0b4 → synapse_sdk-1.0.0b6}/synapse_sdk/plugins/categories/export/actions/export.py

@@ -1,6 +1,7 @@
 import json
 from abc import ABC, abstractmethod
 from datetime import datetime
+from itertools import tee
 from typing import Annotated, Any, Literal
 
 from pydantic import AfterValidator, BaseModel, field_validator
@@ -324,16 +325,18 @@ class ExportAction(Action):
         handler = TargetHandlerFactory.get_handler(target)
 
         self.params['results'], self.params['count'] = self.get_filtered_results(filters, handler)
-        export_items = handler.get_export_item(self.params['results'])
 
         # For the 'ground_truth' target, retrieve project information from the first result and add configuration
         if target == 'ground_truth':
             try:
-                first_result = next(iter(self.params['results']))
+                # Split generator into two using tee()
+                peek_iter, main_iter = tee(self.params['results'])
+                first_result = next(peek_iter)  # Peek first value only
                 project_pk = first_result['project']
                 project_info = self.client.get_project(project_pk)
                 self.params['configuration'] = project_info.get('configuration', {})
-            except StopIteration:
+                self.params['results'] = main_iter  # Keep original generator intact
+            except (StopIteration, KeyError):
                 self.params['configuration'] = {}
         # For the 'assignment' and 'task' targets, retrieve the project from the filter as before
         elif target in ['assignment', 'task'] and 'project' in self.params['filter']:
@@ -341,6 +344,7 @@ class ExportAction(Action):
             project_info = self.client.get_project(project_pk)
             self.params['configuration'] = project_info.get('configuration', {})
 
+        export_items = handler.get_export_item(self.params['results'])
         storage = self.client.get_storage(self.params['storage'])
         pathlib_cwd = get_pathlib(storage, self.params['path'])
         return self.entrypoint(self.run, export_items, pathlib_cwd, **self.params)

{synapse_sdk-1.0.0b4 → synapse_sdk-1.0.0b6}/synapse_sdk/plugins/categories/pre_annotation/actions/to_task.py

@@ -8,6 +8,7 @@ from pydantic import AfterValidator, BaseModel, field_validator
 from pydantic_core import PydanticCustomError
 
 from synapse_sdk.clients.backend import BackendClient
+from synapse_sdk.clients.backend.models import JobStatus
 from synapse_sdk.clients.exceptions import ClientError
 from synapse_sdk.plugins.categories.base import Action
 from synapse_sdk.plugins.categories.decorators import register_action
@@ -35,6 +36,14 @@ class CriticalError(Exception):
         super().__init__(self.message)
 
 
+class PreAnnotationToTaskFailed(Exception):
+    """Pre-annotation to task failed."""
+
+    def __init__(self, message: str = 'Pre-annotation to task failed'):
+        self.message = message
+        super().__init__(self.message)
+
+
 class ToTaskRun(Run):
     class AnnotateTaskEventLog(BaseModel):
         """Annotate task event log model."""
@@ -255,6 +264,25 @@ class ToTaskParams(BaseModel):
         return value
 
 
+class ToTaskResult(BaseModel):
+    """Result model for ToTaskAction.start method.
+
+    Args:
+        status (JobStatus): The job status from the action execution.
+        message (str): A descriptive message about the action result.
+    """
+
+    status: JobStatus
+    message: str
+
+    def model_dump(self, **kwargs):
+        """Override model_dump to return status as enum value."""
+        data = super().model_dump(**kwargs)
+        if 'status' in data and isinstance(data['status'], JobStatus):
+            data['status'] = data['status'].value
+        return data
+
+
 @register_action
 class ToTaskAction(Action):
     """ToTask action for pre-annotation data processing.
@@ -306,14 +334,20 @@ class ToTaskAction(Action):
         }
     }
 
-    def start(self):
+    def start(self) -> dict:
         """Start to_task action.
 
         * Generate tasks.
         * Annotate data to tasks.
+
+        Returns:
+            dict: Validated result with status and message.
         """
         if not self.run or not self.params:
-            raise ValueError('Run instance or parameters not properly initialized')
+            result = ToTaskResult(
+                status=JobStatus.FAILED, message='Run instance or parameters not properly initialized'
+            )
+            raise PreAnnotationToTaskFailed(result.message)
 
         # Type assertion to help the linter
         assert isinstance(self.run, ToTaskRun)
@@ -325,20 +359,23 @@ class ToTaskAction(Action):
         if isinstance(project_response, str):
             self.run.log_message_with_code('INVALID_PROJECT_RESPONSE')
             self.run.end_log()
-            return {}
+            result = ToTaskResult(status=JobStatus.FAILED, message='Invalid project response received')
+            raise PreAnnotationToTaskFailed(result.message)
         project: Dict[str, Any] = project_response
 
         data_collection_id = project.get('data_collection')
         if not data_collection_id:
             self.run.log_message_with_code('NO_DATA_COLLECTION')
             self.run.end_log()
-            return {}
+            result = ToTaskResult(status=JobStatus.FAILED, message='Project does not have a data collection')
+            raise PreAnnotationToTaskFailed(result.message)
 
         data_collection_response = client.get_data_collection(data_collection_id)
         if isinstance(data_collection_response, str):
             self.run.log_message_with_code('INVALID_DATA_COLLECTION_RESPONSE')
             self.run.end_log()
-            return {}
+            result = ToTaskResult(status=JobStatus.FAILED, message='Invalid data collection response received')
+            raise PreAnnotationToTaskFailed(result.message)
         data_collection: Dict[str, Any] = data_collection_response
 
         # Generate tasks if provided project is empty.
@@ -355,7 +392,8 @@ class ToTaskAction(Action):
         if not task_ids_count:
             self.run.log_message_with_code('NO_TASKS_FOUND')
             self.run.end_log()
-            return {}
+            result = ToTaskResult(status=JobStatus.FAILED, message='No tasks found to annotate')
+            raise PreAnnotationToTaskFailed(result.message)
 
         # Annotate data to tasks.
         method = self.params.get('method')
@@ -365,23 +403,39 @@ class ToTaskAction(Action):
             if not target_specification_name:
                 self.run.log_message_with_code('TARGET_SPEC_REQUIRED')
                 self.run.end_log()
-                return {}
+                result = ToTaskResult(
+                    status=JobStatus.FAILED, message='Target specification name is required for file annotation method'
+                )
+                raise PreAnnotationToTaskFailed(result.message)
 
             file_specifications = data_collection.get('file_specifications', [])
             target_spec_exists = any(spec.get('name') == target_specification_name for spec in file_specifications)
             if not target_spec_exists:
                 self.run.log_message_with_code('TARGET_SPEC_NOT_FOUND', target_specification_name)
                 self.run.end_log()
-                return {}
+                result = ToTaskResult(
+                    status=JobStatus.FAILED,
+                    message=f"Target specification name '{target_specification_name}' not found in file specifications",
+                )
+                raise PreAnnotationToTaskFailed(result.message)
             self._handle_annotate_data_from_files(task_ids, target_specification_name)
         elif method == AnnotationMethod.INFERENCE:
             self._handle_annotate_data_with_inference(task_ids)
         else:
             self.run.log_message_with_code('UNSUPPORTED_METHOD', method)
             self.run.end_log()
-            return {}
+            result = ToTaskResult(status=JobStatus.FAILED, message=f'Unsupported annotation method: {method}')
+            raise PreAnnotationToTaskFailed(result.message)
+
+        current_progress = self.run.logger.get_current_progress()
+        if current_progress['overall'] != 100:
+            result = ToTaskResult(
+                status=JobStatus.FAILED, message='Pre-annotation to task failed. Current progress is not 100%'
+            )
+            raise PreAnnotationToTaskFailed(result.message)
 
-        return {}
+        result = ToTaskResult(status=JobStatus.SUCCEEDED, message='Pre-annotation to task completed successfully')
+        return result.model_dump()
 
     def _handle_annotate_data_from_files(self, task_ids: List[int], target_specification_name: str):
         """Handle annotate data from files to tasks.
@@ -757,21 +811,21 @@ class ToTaskAction(Action):
         except Exception as e:
             return {'success': False, 'error': f'Failed to restart pre-processor: {str(e)}'}
 
-    def _extract_primary_file_url(self, task: Dict[str, Any]) -> Tuple[str, str]:
+    def _extract_primary_file_url(self, task: Dict[str, Any]) -> Tuple[Optional[str], Optional[str]]:
         """Extract the primary file URL from task data.
 
         Args:
             task (Dict[str, Any]): The task data.
 
         Returns:
-            Tuple[str, str]: The primary file URL and original name.
+            Tuple[Optional[str], Optional[str]]: The primary file URL and original name.
         """
         data_unit = task.get('data_unit', {})
         files = data_unit.get('files', {})
 
         for file_info in files.values():
             if isinstance(file_info, dict) and file_info.get('is_primary') and file_info.get('url'):
-                return file_info['url'], file_info['file_name_original']
+                return file_info['url'], file_info.get('file_name_original')
 
         return None, None
 

{synapse_sdk-1.0.0b4 → synapse_sdk-1.0.0b6}/synapse_sdk/plugins/categories/upload/actions/upload.py

@@ -18,6 +18,7 @@ from synapse_sdk.i18n import gettext as _
 from synapse_sdk.plugins.categories.base import Action
 from synapse_sdk.plugins.categories.decorators import register_action
 from synapse_sdk.plugins.enums import PluginCategory, RunMethod
+from synapse_sdk.plugins.exceptions import ActionError
 from synapse_sdk.plugins.models import Run
 from synapse_sdk.shared.enums import Context
 from synapse_sdk.utils.pydantic.validators import non_blank
@@ -684,30 +685,26 @@ class UploadAction(Action):
         # Validate the organized files
         if not self._validate_organized_files(organized_files, file_specification_template):
             self.run.log_message('Validation failed.', context=Context.ERROR.value)
-            self.run.log_message('Upload is aborted due to validation errors.', context=Context.ERROR.value)
-            return result
+            raise ActionError('Upload is aborted due to validation errors.')
 
         # Upload files to synapse-backend.
         if not organized_files:
             self.run.log_message('Files not found on the path.', context=Context.WARNING.value)
-            self.run.log_message('Upload is aborted due to missing files.', context=Context.ERROR.value)
-            return result
+            raise ActionError('Upload is aborted due to missing files.')
         uploaded_files = self._upload_files(organized_files)
         result['uploaded_files_count'] = len(uploaded_files)
 
         # Generate data units for the uploaded data.
         if not uploaded_files:
             self.run.log_message('No files were uploaded.', context=Context.WARNING.value)
-            self.run.log_message('Upload is aborted due to no uploaded files.', context=Context.ERROR.value)
-            return result
+            raise ActionError('Upload is aborted due to no uploaded files.')
         generated_data_units = self._generate_data_units(uploaded_files)
         result['generated_data_units_count'] = len(generated_data_units)
 
         # Setup task with uploaded synapse-backend data units.
         if not generated_data_units:
             self.run.log_message('No data units were generated.', context=Context.WARNING.value)
-            self.run.log_message('Upload is aborted due to no generated data units.', context=Context.ERROR.value)
-            return result
+            raise ActionError('Upload is aborted due to no generated data units.')
 
         self.run.log_message('Import completed.')
         return result

{synapse_sdk-1.0.0b4 → synapse_sdk-1.0.0b6/synapse_sdk.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: synapse-sdk
-Version: 1.0.0b4
+Version: 1.0.0b6
 Summary: synapse sdk
 Author-email: datamaker <developer@datamaker.io>
 License: MIT