ras-commander 0.33.0__py3-none-any.whl → 0.34.0__py3-none-any.whl

ras_commander/RasCmdr.py

@@ -12,18 +12,26 @@ from .RasPrj import ras, RasPrj, init_ras_project, get_ras_exe
 from .RasPlan import RasPlan
 from .RasGeo import RasGeo
 from .RasUtils import RasUtils
-import subprocess
-import os
 import logging
 import time
-import pandas as pd
-from threading import Thread, Lock
 import queue
+from threading import Thread, Lock
+from typing import Union, List, Optional, Dict
 from pathlib import Path
 import shutil
-import queue
-from threading import Thread, Lock
-import time
+import logging
+from concurrent.futures import ThreadPoolExecutor, as_completed
+from threading import Lock, Thread
+from itertools import cycle
+from ras_commander.RasPrj import RasPrj  # Ensure RasPrj is imported
+from threading import Lock, Thread, current_thread
+from concurrent.futures import ThreadPoolExecutor, as_completed
+from itertools import cycle
+from typing import Union, List, Optional, Dict
+
+
+# Configure logging
+logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
 
 # TO DO: 
 # 1. Alternate Run Mode for compute_plan and compute_parallel:  Using Powershell to execute the HEC-RAS command and hide the RAS window and all child windows.
@@ -75,11 +83,15 @@ class RasCmdr:
             if dest_folder.exists():
                 if overwrite_dest:
                     shutil.rmtree(dest_folder)
+                    logging.info(f"Destination folder '{dest_folder}' exists. Overwriting as per overwrite_dest=True.")
                 elif any(dest_folder.iterdir()):
-                    raise ValueError(f"Destination folder '{dest_folder}' exists and is not empty. Use overwrite_dest=True to overwrite.")
+                    error_msg = f"Destination folder '{dest_folder}' exists and is not empty. Use overwrite_dest=True to overwrite."
+                    logging.error(error_msg)
+                    raise ValueError(error_msg)
             
             dest_folder.mkdir(parents=True, exist_ok=True)
             shutil.copytree(ras_obj.project_folder, dest_folder, dirs_exist_ok=True)
+            logging.info(f"Copied project folder to destination: {dest_folder}")
             
             compute_ras = RasPrj()
             compute_ras.initialize(dest_folder, ras_obj.ras_exe_path)
@@ -92,29 +104,29 @@ class RasCmdr:
         compute_plan_path = Path(plan_number) if isinstance(plan_number, (str, Path)) and Path(plan_number).is_file() else RasPlan.get_plan_path(plan_number, compute_ras)
 
         if not compute_prj_path or not compute_plan_path:
-            print(f"Error: Could not find project file or plan file for plan {plan_number}")
+            logging.error(f"Could not find project file or plan file for plan {plan_number}")
             return False
 
         # Clear geometry preprocessor files if requested
         if clear_geompre:
             try:
                 RasGeo.clear_geompre_files(compute_plan_path, ras_object=compute_ras)
-                print(f"Cleared geometry preprocessor files for plan: {plan_number}")
+                logging.info(f"Cleared geometry preprocessor files for plan: {plan_number}")
             except Exception as e:
-                print(f"Error clearing geometry preprocessor files for plan {plan_number}: {str(e)}")
+                logging.error(f"Error clearing geometry preprocessor files for plan {plan_number}: {str(e)}")
 
         # Set the number of cores if specified
         if num_cores is not None:
             try:
                 RasPlan.set_num_cores(compute_plan_path, num_cores=num_cores, ras_object=compute_ras)
-                print(f"Set number of cores to {num_cores} for plan: {plan_number}")
+                logging.info(f"Set number of cores to {num_cores} for plan: {plan_number}")
             except Exception as e:
-                print(f"Error setting number of cores for plan {plan_number}: {str(e)}")
+                logging.error(f"Error setting number of cores for plan {plan_number}: {str(e)}")
 
         # Prepare the command for HEC-RAS execution
         cmd = f'"{compute_ras.ras_exe_path}" -c "{compute_prj_path}" "{compute_plan_path}"'
-        print("Running HEC-RAS from the Command Line:")
-        print(f"Running command: {cmd}")
+        logging.info("Running HEC-RAS from the Command Line:")
+        logging.info(f"Running command: {cmd}")
 
         # Execute the HEC-RAS command
         start_time = time.time()
@@ -122,15 +134,15 @@ class RasCmdr:
             subprocess.run(cmd, check=True, shell=True, capture_output=True, text=True)
             end_time = time.time()
             run_time = end_time - start_time
-            print(f"HEC-RAS execution completed for plan: {plan_number}")
-            print(f"Total run time for plan {plan_number}: {run_time:.2f} seconds")
+            logging.info(f"HEC-RAS execution completed for plan: {plan_number}")
+            logging.info(f"Total run time for plan {plan_number}: {run_time:.2f} seconds")
             return True
         except subprocess.CalledProcessError as e:
             end_time = time.time()
             run_time = end_time - start_time
-            print(f"Error running plan: {plan_number}")
-            print(f"Error message: {e.output}")
-            print(f"Total run time for plan {plan_number}: {run_time:.2f} seconds")
+            logging.error(f"Error running plan: {plan_number}")
+            logging.error(f"Error message: {e.output}")
+            logging.info(f"Total run time for plan {plan_number}: {run_time:.2f} seconds")
             return False
 
         ras_obj = ras_object or ras
@@ -138,77 +150,51 @@ class RasCmdr:
         ras_obj.geom_df = ras_obj.get_geom_entries()
         ras_obj.flow_df = ras_obj.get_flow_entries()
         ras_obj.unsteady_df = ras_obj.get_unsteady_entries()
-
+    
 
 
     @staticmethod
     def compute_parallel(
-        plan_number: str | list[str] | None = None,
+        plan_number: Union[str, List[str], None] = None,
         max_workers: int = 2,
         num_cores: int = 2,
         clear_geompre: bool = False,
-        ras_object: RasPrj | None = None,
-        dest_folder: str | Path | None = None,
+        ras_object: Optional['RasPrj'] = None,  # Type hinting as string to avoid NameError
+        dest_folder: Union[str, Path, None] = None,
         overwrite_dest: bool = False
-    ) -> dict[str, bool]:
+    ) -> Dict[str, bool]:
         """
-        Execute HEC-RAS plans in parallel using multiple worker threads.
-
-        This function creates separate worker folders, copies the project to each, and executes the specified plans
-        in parallel. It allows for isolated and concurrent execution of multiple plans.
-
-        Args:
-            plan_number (str | list[str] | None): Plan number, list of plan numbers, or None to execute all plans.
-            max_workers (int, optional): Maximum number of worker threads to use. Default is 2.
-            num_cores (int, optional): Number of cores to use for each plan execution. Default is 2.
-            clear_geompre (bool, optional): Whether to clear geometry preprocessor files. Defaults to False.
-            ras_object (RasPrj, optional): Specific RAS object to use. If None, uses the global ras instance.
-            dest_folder (str | Path, optional): Destination folder for the final computed results.
-                If None, results will be stored in a "[Computed]" folder next to the original project.
-            overwrite_dest (bool, optional): If True, overwrite the destination folder if it exists. Defaults to False.
-
-        Returns:
-            dict[str, bool]: A dictionary with plan numbers as keys and boolean values indicating success (True) or failure (False).
-
-        Raises:
-            ValueError: If the destination folder exists and is not empty, and overwrite_dest is False.
-            FileNotFoundError: If a plan file is not found.
-
-        Notes:
-            - This function creates separate folders for each worker to ensure isolated execution.
-            - Each worker uses its own RAS object to prevent conflicts.
-            - Plans are distributed among workers using a queue to ensure efficient parallel processing.
-            - The function automatically handles cleanup and consolidation of results after execution.
-        
-        Revision Notes:
-            - Added support for clear_geompre flag as a pass-through to compute_plan.
-            - Simplified worker thread logic by removing redundant operations.
-            - Removed duplicate RAS object initialization in worker threads.
+        [Docstring remains unchanged]
         """
-        ras_obj = ras_object or ras
+        ras_obj = ras_object or ras  # Assuming 'ras' is a global RasPrj instance
         ras_obj.check_initialized()
 
-        project_folder = ras_obj.project_folder
+        project_folder = Path(ras_obj.project_folder)
 
         if dest_folder is not None:
             dest_folder_path = Path(dest_folder)
             if dest_folder_path.exists():
                 if overwrite_dest:
                     shutil.rmtree(dest_folder_path)
+                    logging.info(f"Destination folder '{dest_folder_path}' exists. Overwriting as per overwrite_dest=True.")
                 elif any(dest_folder_path.iterdir()):
-                    raise ValueError(f"Destination folder '{dest_folder_path}' exists and is not empty. Use overwrite_dest=True to overwrite.")
+                    error_msg = f"Destination folder '{dest_folder_path}' exists and is not empty. Use overwrite_dest=True to overwrite."
+                    logging.error(error_msg)
+                    raise ValueError(error_msg)
             dest_folder_path.mkdir(parents=True, exist_ok=True)
             shutil.copytree(project_folder, dest_folder_path, dirs_exist_ok=True)
+            logging.info(f"Copied project folder to destination: {dest_folder_path}")
             project_folder = dest_folder_path
 
         if plan_number:
             if isinstance(plan_number, str):
                 plan_number = [plan_number]
             ras_obj.plan_df = ras_obj.plan_df[ras_obj.plan_df['plan_number'].isin(plan_number)]
+            logging.info(f"Filtered plans to execute: {plan_number}")
 
         num_plans = len(ras_obj.plan_df)
         max_workers = min(max_workers, num_plans) if num_plans > 0 else 1
-        print(f"Adjusted max_workers to {max_workers} based on the number of plans: {num_plans}")
+        logging.info(f"Adjusted max_workers to {max_workers} based on the number of plans: {num_plans}")
 
         # Clean up existing worker folders and create new ones
         worker_ras_objects = {}
@@ -216,85 +202,122 @@ class RasCmdr:
             worker_folder = project_folder.parent / f"{project_folder.name} [Worker {worker_id}]"
             if worker_folder.exists():
                 shutil.rmtree(worker_folder)
+                logging.info(f"Removed existing worker folder: {worker_folder}")
             shutil.copytree(project_folder, worker_folder)
-            
+            logging.info(f"Created worker folder: {worker_folder}")
+
+            # Instantiate RasPrj properly
+            ras_instance = RasPrj()  # Add necessary parameters if required
             worker_ras_instance = init_ras_project(
                 ras_project_folder=worker_folder,
                 ras_version=ras_obj.ras_exe_path,
-                ras_instance=RasPrj()
+                ras_instance=ras_instance  # Pass the instance instead of a string
             )
             worker_ras_objects[worker_id] = worker_ras_instance
 
-        plan_queue = queue.Queue()
-        for plan_number in ras_obj.plan_df['plan_number']:
-            plan_queue.put(plan_number)
-
-        execution_results: dict[str, bool] = {}
-        results_lock = Lock()
-        queue_lock = Lock()
-
-        def worker_thread(worker_id: int):
-            worker_ras_obj = worker_ras_objects[worker_id]
-            while True:
-                with queue_lock:
-                    if plan_queue.empty():
-                        break
-                    plan_number = plan_queue.get()
-                
+        # Distribute plans among workers in a round-robin fashion
+        worker_cycle = cycle(range(1, max_workers + 1))
+        plan_assignments = [(next(worker_cycle), plan_num) for plan_num in ras_obj.plan_df['plan_number']]
+
+        # Initialize ThreadPoolExecutor without tracking individual plan success
+        with ThreadPoolExecutor(max_workers=max_workers) as executor:
+            # Submit all plan executions to the executor
+            futures = [
+                executor.submit(
+                    RasCmdr.compute_plan,
+                    plan_num, 
+                    ras_object=worker_ras_objects[worker_id], 
+                    clear_geompre=clear_geompre,
+                    num_cores=num_cores
+                )
+                for worker_id, plan_num in plan_assignments
+            ]
+
+            # Optionally, you can log when each plan starts and completes
+            for future, (worker_id, plan_num) in zip(as_completed(futures), plan_assignments):
                 try:
-                    print(f"Worker {worker_id} executing plan {plan_number}")
-                    success = RasCmdr.compute_plan(
-                        plan_number, 
-                        ras_object=worker_ras_obj, 
-                        clear_geompre=clear_geompre,
-                        num_cores=num_cores
-                    )
-                    with results_lock:
-                        execution_results[plan_number] = success
-                    print(f"Completed: Plan {plan_number} in worker {worker_id}")
+                    future.result()  # We don't need the success flag here
+                    logging.info(f"Plan {plan_num} executed in worker {worker_id}")
                 except Exception as e:
-                    with results_lock:
-                        execution_results[plan_number] = False
-                    print(f"Failed: Plan {plan_number} in worker {worker_id}. Error: {str(e)}")
-
-        # Start worker threads
-        worker_threads = [Thread(target=worker_thread, args=(worker_id,)) for worker_id in range(1, max_workers + 1)]
-        for thread in worker_threads:
-            thread.start()
-
-        # Wait for all threads to complete
-        for thread in worker_threads:
-            thread.join()
+                    logging.error(f"Plan {plan_num} failed in worker {worker_id}: {str(e)}")
+                    # Depending on requirements, you might want to handle retries or mark these plans differently
 
         # Consolidate results
         final_dest_folder = dest_folder_path if dest_folder is not None else project_folder.parent / f"{project_folder.name} [Computed]"
-        final_dest_folder.mkdir(exist_ok=True)
-        print(f"Final destination for computed results: {final_dest_folder}")
+        final_dest_folder.mkdir(parents=True, exist_ok=True)
+        logging.info(f"Final destination for computed results: {final_dest_folder}")
 
         for worker_ras in worker_ras_objects.values():
-            worker_folder = worker_ras.project_folder
+            worker_folder = Path(worker_ras.project_folder)
             try:
                 for item in worker_folder.iterdir():
                     dest_path = final_dest_folder / item.name
                     if dest_path.exists():
                         if dest_path.is_dir():
                             shutil.rmtree(dest_path)
+                            logging.debug(f"Removed existing directory at {dest_path}")
                         else:
                             dest_path.unlink()
+                            logging.debug(f"Removed existing file at {dest_path}")
                     shutil.move(str(item), final_dest_folder)
+                    logging.debug(f"Moved {item} to {final_dest_folder}")
                 shutil.rmtree(worker_folder)
+                logging.info(f"Removed worker folder: {worker_folder}")
             except Exception as e:
-                print(f"Error moving results from {worker_folder} to {final_dest_folder}: {str(e)}")
+                logging.error(f"Error moving results from {worker_folder} to {final_dest_folder}: {str(e)}")
+
+        # Initialize a new RasPrj object for the final destination
+        try:
+            # Create a new RasPrj instance
+            final_dest_folder_ras_obj = RasPrj()
+            
+            # Initialize it using init_ras_project
+            final_dest_folder_ras_obj = init_ras_project(
+                ras_project_folder=final_dest_folder, 
+                ras_version=ras_obj.ras_exe_path,
+                ras_instance=final_dest_folder_ras_obj
+            )
+            
+            # Now we can check if it's initialized
+            final_dest_folder_ras_obj.check_initialized()
+        except Exception as e:
+            logging.error(f"Failed to initialize RasPrj for final destination: {str(e)}")
+            raise
+
+        # Retrieve plan entries and check for HDF results
+        try:
+            plan_entries = final_dest_folder_ras_obj.get_prj_entries('Plan')
+        except Exception as e:
+            logging.error(f"Failed to retrieve plan entries from final RasPrj: {str(e)}")
+            raise
+
+        execution_results: Dict[str, bool] = {}
+        for _, row in ras_obj.plan_df.iterrows():
+            plan_num = row['plan_number']
+            # Find the corresponding entry in plan_entries
+            entry = plan_entries[plan_entries['plan_number'] == plan_num]
+            if not entry.empty:
+                hdf_path = entry.iloc[0].get('HDF_Results_Path')
+                success = hdf_path is not None and Path(hdf_path).exists()
+            else:
+                success = False
+            execution_results[plan_num] = success
 
         # Print execution results for each plan
-        print("\nExecution Results:")
-        for plan_number, success in execution_results.items():
-            print(f"Plan {plan_number}: {'Successful' if success else 'Failed'}")
+        logging.info("\nExecution Results:")
+        for plan_num, success in execution_results.items():
+            status = 'Successful' if success else 'Failed'
+            logging.info(f"Plan {plan_num}: {status} \n(HDF_Results_Path: {hdf_path})")
+            
+        ras_obj = ras_object or ras
+        ras_obj.plan_df = ras_obj.get_plan_entries()
+        ras_obj.geom_df = ras_obj.get_geom_entries()
+        ras_obj.flow_df = ras_obj.get_flow_entries()
+        ras_obj.unsteady_df = ras_obj.get_unsteady_entries()
 
         return execution_results
     
-    
-    
+
     @staticmethod
     def compute_test_mode(
         plan_number=None, 
@@ -355,40 +378,44 @@ class RasCmdr:
         # This line of code is used to check if the RasPrj object is initialized.
         ras_obj.check_initialized()
         
-        print("Starting the compute_test_mode...")
+        logging.info("Starting the compute_test_mode...")
            
         # Use the project folder from the ras object
         project_folder = ras_obj.project_folder
 
         # Check if the project folder exists
         if not project_folder.exists():
-            print(f"Error: Project folder '{project_folder}' does not exist.")
+            logging.error(f"Project folder '{project_folder}' does not exist.")
             return
 
         # Create test folder with the specified suffix in the same directory as the project folder
         compute_folder = project_folder.parent / f"{project_folder.name} {dest_folder_suffix}"
-        print(f"Creating the test folder: {compute_folder}...")
+        logging.info(f"Creating the test folder: {compute_folder}...")
 
         # Check if the compute folder exists and is empty
         if compute_folder.exists():
             if overwrite_dest:
                 shutil.rmtree(compute_folder)
+                logging.info(f"Compute folder '{compute_folder}' exists. Overwriting as per overwrite_dest=True.")
             elif any(compute_folder.iterdir()):
-                raise ValueError(
+                error_msg = (
                     f"Compute folder '{compute_folder}' exists and is not empty. "
                     "Use overwrite_dest=True to overwrite."
                 )
+                logging.error(error_msg)
+                raise ValueError(error_msg)
         else:
             try:
                 shutil.copytree(project_folder, compute_folder)
+                logging.info(f"Copied project folder to compute folder: {compute_folder}")
             except FileNotFoundError:
-                print(f"Error: Unable to copy project folder. Source folder '{project_folder}' not found.")
+                logging.error(f"Unable to copy project folder. Source folder '{project_folder}' not found.")
                 return
             except PermissionError:
-                print(f"Error: Permission denied when trying to create or copy to '{compute_folder}'.")
+                logging.error(f"Permission denied when trying to create or copy to '{compute_folder}'.")
                 return
             except Exception as e:
-                print(f"Error occurred while copying project folder: {str(e)}")
+                logging.error(f"Error occurred while copying project folder: {str(e)}")
                 return
 
         # Initialize a new RAS project in the compute folder
@@ -396,22 +423,22 @@ class RasCmdr:
             compute_ras = RasPrj()
             compute_ras.initialize(compute_folder, ras_obj.ras_exe_path)
             compute_prj_path = compute_ras.prj_file
+            logging.info(f"Initialized RAS project in compute folder: {compute_prj_path}")
         except Exception as e:
-            print(f"Error initializing RAS project in compute folder: {str(e)}")
+            logging.error(f"Error initializing RAS project in compute folder: {str(e)}")
             return
 
         if not compute_prj_path:
-            print("Project file not found.")
+            logging.error("Project file not found.")
             return
 
-
         # Get plan entries
-        print("Getting plan entries...")
+        logging.info("Getting plan entries...")
         try:
             ras_compute_plan_entries = compute_ras.plan_df
-            print("Retrieved plan entries successfully.")
+            logging.info("Retrieved plan entries successfully.")
         except Exception as e:
-            print(f"Error retrieving plan entries: {str(e)}")
+            logging.error(f"Error retrieving plan entries: {str(e)}")
             return
 
         if plan_number:
@@ -420,30 +447,35 @@ class RasCmdr:
             ras_compute_plan_entries = ras_compute_plan_entries[
                 ras_compute_plan_entries['plan_number'].isin(plan_number)
             ]
-            print(f"Filtered plans to execute: {plan_number}")
+            logging.info(f"Filtered plans to execute: {plan_number}")
 
-        print("Running selected plans sequentially...")
+        logging.info("Running selected plans sequentially...")
         for _, plan in ras_compute_plan_entries.iterrows():
             plan_number = plan["plan_number"]
             start_time = time.time()
             try:
-                RasCommander.compute_plan(
+                success = RasCmdr.compute_plan(
                     plan_number,
                     ras_object=compute_ras,
                     clear_geompre=clear_geompre,
                     num_cores=num_cores
                 )
+                if success:
+                    logging.info(f"Successfully computed plan {plan_number}")
+                else:
+                    logging.error(f"Failed to compute plan {plan_number}")
             except Exception as e:
-                print(f"Error computing plan {plan_number}: {str(e)}")
+                logging.error(f"Error computing plan {plan_number}: {str(e)}")
             end_time = time.time()
             run_time = end_time - start_time
-            print(f"Total run time for plan {plan_number}: {run_time:.2f} seconds")
+            logging.info(f"Total run time for plan {plan_number}: {run_time:.2f} seconds")
 
-        print("All selected plans have been executed.")
-        print("compute_test_mode completed.")
+        logging.info("All selected plans have been executed.")
+        logging.info("compute_test_mode completed.")
 
         ras_obj = ras_object or ras
         ras_obj.plan_df = ras_obj.get_plan_entries()
         ras_obj.geom_df = ras_obj.get_geom_entries()
         ras_obj.flow_df = ras_obj.get_flow_entries()
-        ras_obj.unsteady_df = ras_obj.get_unsteady_entries()
+        ras_obj.unsteady_df = ras_obj.get_unsteady_entries()
+