Renamed to_yaml() as serialize_metadata() and introduce input parameter output_format, which allows yaml or json.

instruments/readers/g5505_text_reader.py

@@ -178,7 +178,7 @@ def read_txt_files_as_dict(filename: str, instruments_dir: str = None, work_with
         #if numerical_variables:
         dataset = {}
         dataset['name'] = 'data_table'#_numerical_variables'
-        dataset['data'] = utils.dataframe_to_np_structured_array(df) #df_numerical_attrs.to_numpy()
+        dataset['data'] = utils.convert_dataframe_to_np_structured_array(df) #df_numerical_attrs.to_numpy()
         dataset['shape'] = dataset['data'].shape
         dataset['dtype'] = type(dataset['data'])       
         #dataset['data_units'] = file_obj['wave']['data_units']   
@@ -191,7 +191,7 @@ def read_txt_files_as_dict(filename: str, instruments_dir: str = None, work_with
             for column_name in df.columns: 
                 column_attr_dict = description_dict['table_header'].get(column_name,
                                                                         {'note':'there was no description available. Review instrument files.'})   
-                dataset['attributes'].update({column_name: utils.parse_attribute(column_attr_dict)})     
+                dataset['attributes'].update({column_name: utils.convert_attrdict_to_np_structured_array(column_attr_dict)})     
 
         #try:
         #    dataset['attributes'] = description_dict['table_header'].copy()

src/hdf5_ops.py

@@ -10,7 +10,12 @@ import numpy as np
 import utils.g5505_utils as utils
 import logging
 import datetime
+
+import os
+import h5py
+
 import yaml
+import json
 
 class HDF5DataOpsManager():
     def __init__(self, file_path, mode = 'r+') -> None:
@@ -254,7 +259,7 @@ def construct_attributes_dict(attrs_obj):
 
     return attr_dict
 
-def print_metadata(name, obj, folder_depth, yaml_dict): 
+def __print_metadata__(name, obj, folder_depth, yaml_dict): 
 
     # TODO: should we enable deeper folders ?
     if len(obj.name.split('/')) <= folder_depth:
@@ -298,29 +303,60 @@ def print_metadata(name, obj, folder_depth, yaml_dict):
         #elif len(obj.name.split('/')) == 3:
         #    print(yaml.dump())
 
-def to_yaml(input_filename_path,folder_depth: int = 4):
-
+def serialize_metadata(input_filename_path, folder_depth: int = 4, output_format: str = 'yaml') -> str:
+    """
+    Serialize metadata from an HDF5 file into YAML or JSON format.
+    
+    Parameters
+    ----------
+    input_filename_path : str
+        The path to the input HDF5 file.
+    folder_depth : int, optional
+        The folder depth to control how much of the HDF5 file hierarchy is traversed (default is 4).
+    output_format : str, optional
+        The format to serialize the output, either 'yaml' or 'json' (default is 'yaml').
+    
+    Returns
+    -------
+    str
+        The output file path where the serialized metadata is stored (either .yaml or .json).
+    
+    """
+    
+    # Choose the appropriate output format (YAML or JSON)
+    if output_format not in ['yaml', 'json']:
+        raise ValueError("Unsupported format. Please choose either 'yaml' or 'json'.")
+    
+    # Initialize dictionary to store YAML/JSON data
     yaml_dict = {}
-
+    
+    # Split input file path to get the output file's base name
     output_filename_tail, ext = os.path.splitext(input_filename_path)
-
-    with h5py.File(input_filename_path,'r') as f:
-
+    
+    # Open the HDF5 file and extract metadata
+    with h5py.File(input_filename_path, 'r') as f:
+        # Construct attributes dictionary and top-level structure
         attrs_dict = construct_attributes_dict(f.attrs)
-        yaml_dict[f.name] =  {"name": f.name, "attributes": attrs_dict, "datasets":{}}
-        f.visititems(lambda name, obj: print_metadata(name,obj,folder_depth,yaml_dict))
-
-    #with open(output_filename_tail+".json","w") as yaml_file:
-    #    json_obj = json.dumps(yaml_dict,indent=4,sort_keys=False,)
-    #    yaml_file.write(json_obj)
-
-    with open(output_filename_tail+".yaml","w") as yaml_file:
-        yaml_output = yaml.dump(yaml_dict,sort_keys=False)
-        #for key in yaml_dict:
-        #    yaml_output = yaml.dump(yaml_dict[key],sort_keys=False)
-        yaml_file.write(yaml_output )
-
-    return output_filename_tail+".yaml"
+        yaml_dict[f.name] = {
+            "name": f.name,
+            "attributes": attrs_dict,
+            "datasets": {}
+        }
+        # Traverse HDF5 file hierarchy and add datasets
+        f.visititems(lambda name, obj: __print_metadata__(name, obj, folder_depth, yaml_dict))
+    
+    
+    # Serialize and write the data
+    output_file_path = output_filename_tail + '.' + output_format
+    with open(output_file_path, 'w') as output_file:
+        if output_format == 'json':
+            json_output = json.dumps(yaml_dict, indent=4, sort_keys=False)
+            output_file.write(json_output)
+        elif output_format == 'yaml':
+            yaml_output = yaml.dump(yaml_dict, sort_keys=False)
+            output_file.write(yaml_output)
+    
+    return output_file_path
 
 
 def get_groups_at_a_level(file: h5py.File, level: str):