process.py

import json

# Open the file and read line by line
with open('/home/yiyangai/stephenqs/datasets/wikihow/WikihowText_data.json', 'r') as file:
    unique_first_level_categories = set()
    
    for line in file:
        try:
            # Attempt to parse each line as a JSON object
            item = json.loads(line.strip())
            if item['category_hierarchy'] != []:
                first_category = item['category_hierarchy'][0]
                unique_first_level_categories.add(first_category)
        except json.JSONDecodeError as e:
            print(f"Skipping line due to JSON error: {e}")

# Save the unique first-level categories to a text file
with open('unique_categories.txt', 'w') as output_file:
    for category in sorted(unique_first_level_categories):
        output_file.write(f"{category}\n")

print(f"Unique first-level categories have been saved to 'unique_categories.txt'")


#import json
#
## Open the file and read lines
#with open('/home/yiyangai/stephenqs/datasets/wikihow/WikihowText_data.json', 'r') as file:
#    data = []
#    for line in file:
#        data.append(json.loads(line))
#
## Now 'data' is a list of dictionaries
#unique_categories = set()
#for item in data:
#    hierarchy = tuple(item['category_hierarchy'])
#    unique_categories.add(hierarchy)
#    if len(unique_categories) >= 100:
#        break
#
#for hierarchy in unique_categories:
#    print(" > ".join(hierarchy))