Split header in three parts and detect variables and variable descriptions added to attribute dictionary

instruments/readers/nasa_ames_reader.py

@@ -23,6 +23,62 @@ import argparse
 
 import utils.g5505_utils as utils
 
+
+def split_header(header_lines):
+    header_lines_copy = []
+
+    for line in header_lines:
+        if isinstance(line, bytes):
+            decoded_line = line.decode('utf-8', errors='ignore').strip()
+            header_lines_copy.append(decoded_line)
+        else:
+            header_lines_copy.append(line.strip())
+
+    # Find the index where the variable descriptions start
+    var_desc_marker = "Days from the file reference point (start_time)"
+
+    try:
+        var_start_idx = header_lines_copy.index(var_desc_marker)
+    except ValueError:
+        raise Exception("Expected variable description marker not found.")
+
+    # Part 1: Everything before variable description
+    part1 = header_lines[:var_start_idx]
+
+    # Part 2: Variable descriptions — until the first key-value line (contains ':')
+    part2 = []
+    part3 = []
+    in_part3 = False
+
+    for line in header_lines[var_start_idx:]:
+        if not in_part3 and ':' in line.decode(encoding = "utf-8"):
+            in_part3 = True  # We assume this is where key-value pairs begin
+
+        if in_part3:
+            part3.append(line)
+        else:
+            part2.append(line)
+
+    return part1, part2, part3
+
+def extract_var_descriptions(part2):
+
+    nvars = int(part2[1].decode(encoding='utf-8').strip())
+    if not sum(float(i) for i in part2[2].decode(encoding='utf-8').strip().split()) == nvars:
+        line1 = part2[1].decode(encoding='utf-8')
+        line2 = part2[2].decode(encoding='utf-8')
+        raise RuntimeError(f'Inconsistent lines. Check lines {line1} and {line2}')
+
+    descriptions = []
+    for line_idx in range(4,4+nvars):
+        descriptions.append(part2[line_idx])
+
+    return descriptions
+
+
+
+
+
 def read_nasa_ames_as_dict(filename, instruments_dir: str = None, work_with_copy: bool = True):
 
     # If instruments_dir is not provided, use the default path relative to the module directory
@@ -43,23 +99,48 @@ def read_nasa_ames_as_dict(filename, instruments_dir: str = None, work_with_copy
     description_dict = config_dict.get('table_header',{})
 
     # Read all lines once
-    with open(filename, 'r') as file:
+    with open(filename, 'rb') as file:
         lines = file.readlines()
 
     # Extract header length from the first line
     header_length = int(lines[0].split()[0])
     file_header = lines[:header_length]
 
-    # Extract start date from line 7
+    # Split header in three parts, header preamble, var descriptions, and metadata pairs 
-    date_header = lines[6].split()
+    part1, part2, part3 = split_header(file_header)
-    start_date_str = f"{date_header[0]}-{date_header[1]}-{date_header[2]}"
+
-    start_date = datetime.strptime(start_date_str, "%Y-%m-%d")
+    var_descriptions = extract_var_descriptions(part2)
+
+    table_header = part3[len(part3)-1]
+    part3.remove(table_header)
+    
+    
+    for line in part3:
+        if 'Startdate:' in line.decode(encoding = "utf-8"):
+            line_parts = line.decode(encoding = "utf-8").split(':',1)
+            attribute_name = line_parts[0]
+            attribute_value = line_parts[1]
+            print(attribute_name,attribute_value)
+            #date_header = lines[6].split()
+            # Split the string by '.'
+            #filename_parts = attribute_value.split('.')
+
+            # Extract the datetime strings
+            start_str = attribute_value.strip()
+            #end_str = filename_parts[2]
+
+            # Parse into datetime objects
+            start_date = datetime.strptime(start_str, "%Y%m%d%H%M%S")
+            #end_date = datetime.strptime(end_str, "%Y%m%d%H%M%S")
+
+    #start_date_str = f"{date_header[0]}-{date_header[1]}-{date_header[2]}"
+    #start_date = datetime.strptime(start_date_str, "%Y-%m-%d")
 
     # Extract number of dependent variables from line 10
-    num_dep_vars = int(lines[9].split()[0])
+    #num_dep_vars = int(lines[9].split()[0])
 
     # Get variable names: start_time + vars from lines 13 to 13+num_dep_vars-1 (zero-indexed: 12 to 12+num_dep_vars)
-    vars_list = ["start_time"] + [lines[i].strip() for i in range(12, 12 + num_dep_vars)]
+    vars_list = table_header.decode(encoding="utf-8").strip().split() #["start_time"] + [lines[i].strip() for i in range(12, 12 + num_dep_vars)]
 
     # Get the last line of the header (data column names)
     dat_head_line = lines[header_length - 1]
@@ -79,10 +160,13 @@ def read_nasa_ames_as_dict(filename, instruments_dir: str = None, work_with_copy
         # Create header metadata dictionary
         header_metadata_dict = {
             'header_length': header_length,
-            'start_date': start_date_str,
+            'start_date': start_str,
-            'num_dep_vars': num_dep_vars,
+            #'num_dep_vars': num_dep_vars,
             'variable_names': vars_list,
-            'raw_header': file_header
+            'variable_descriptions' : var_descriptions,
+            'raw_header_part1': part1,
+            'raw_header_part2': part2,
+            'raw_header_part3': part3
         }