product_emb / convert.py
aakarsh-yadav-tcgls
update more products
d54c8f8
Raw
History Blame Contribute Delete
915 Bytes
import json
# Function to convert $oid to a string and clean up the data
def process_data(input_file, output_file):
with open(input_file, "r") as infile:
data = json.load(infile) # Read the JSON file
# Transform the data
for record in data:
if isinstance(record.get("_id"), dict) and "$oid" in record["_id"]:
record["id"] = record["_id"]["$oid"] # Replace with string value
del record["_id"]
if "__v" in record: # Optionally remove the "__v" field
del record["__v"]
# Write the transformed data back to a new JSON file
with open(output_file, "w") as outfile:
json.dump(data, outfile, indent=4) # Pretty print with indent
# File paths
input_file = "./sampleDb.products.json"
output_file = "./products.json"
# Process the data
process_data(input_file, output_file)
print(f"Processed data written to {output_file}")