ERC20 Transaction Data Download

Jovay Network provides historical ERC20 transaction data in Apache Parquet format for developers and analysts who need to access on-chain token transfer records.

Data Overview

The ERC20 transaction data includes all token transfer events (ERC20 Transfer events) that occurred on the Jovay Network. This data is useful for:

• Building analytics dashboards
• Tracking token movements
• Auditing transaction history
• Creating custom reports

Download URL Format

The data is hosted on OSS and organized by date. The base URL is:

https://datasets.jovay.io/rwa/erc20/asset_operation/

Directory Structure

/rwa/erc20/asset_operation/
├── date=2025-12-03/asset_operation_full.parquet      # First full snapshot
├── date=2025-12-04/asset_operation_increment.parquet
├── date=2025-12-05/asset_operation_increment.parquet
└── ...

File Types

File Type	Description
asset_operation_full.parquet	Contains the complete historical data up to that date
asset_operation_increment.parquet	Contains only the new transactions for that specific date

Example Download URLs

Full data snapshot:

https://datasets.jovay.io/rwa/erc20/asset_operation/date=2025-12-03/asset_operation_full.parquet

Incremental data:

https://datasets.jovay.io/rwa/erc20/asset_operation/date=2025-12-08/asset_operation_increment.parquet

Data Schema

The parquet files contain the following columns:

Column	Type	Description
id	int64	Auto-incremented identifier (can be ignored for most use cases)
transaction_index	int64	Position of the transaction within the block
transaction_hash	string	The hash of the transaction containing this transfer
evt_index	int64	Index of the event within the transaction
block_number	int64	The block number where the transfer occurred
block_hash	string	The hash of the block
from_address	string	The sender address (0x format)
to_address	string	The recipient address (0x format)
token_address	string	The ERC20 token contract address (0x format)
timestamp	datetime64[ms]	The timestamp when the block was mined
trace_address	int64[]	The call-path to the EVM frame that emitted the log. Each element is the ordinal of the child call within its parent. [] = top-level tx call; [0] = first internal call; [0,2] = third child of the first child, etc.
amount	string	Human-readable token amount (with decimals applied)
raw_amount	string	Raw token amount (without decimals)
date	datetime64[ns]	The date of the transaction

Reading the Data

Prerequisites

Install the required Python packages:

pip install pandas fastparquet

Or using requirements.txt:

pandas
fastparquet

Python Example

import pandas as pd

pd.set_option('display.max_columns', None)  # Show all columns
pd.set_option('display.max_colwidth', None)  # Do not truncate column content
pd.set_option('display.width', None)  # Automatically detect terminal width

# Read parquet file (local)
df = pd.read_parquet('./asset_operation_full.parquet')

# Or read directly from URL
df = pd.read_parquet(
    'https://datasets.jovay.io/rwa/erc20/asset_operation/date=2025-12-03/asset_operation_full.parquet'
)

# View row count and shape
print(f"Row count: {len(df)}")
print(f"Shape: {df.shape}")

# View detailed column info
print(df.info())

# View first few rows
print(df.head())

# View statistics
print(df.describe())

Sample Output

Data columns (total 14 columns):
 #   Column             Non-Null Count  Dtype         
---  ------             --------------  -----         
 0   id                 6477 non-null   object        
 1   transaction_index  6477 non-null   int64         
 2   transaction_hash   6477 non-null   object        
 3   evt_index          6477 non-null   int64         
 4   block_number       6477 non-null   int64         
 5   block_hash         6477 non-null   object        
 6   from_address       6477 non-null   object        
 7   to_address         6477 non-null   object        
 8   token_address      6477 non-null   object        
 9   timestamp          6477 non-null   datetime64[ms]
 10  trace_address      0 non-null      object        
 11  amount             6477 non-null   object        
 12  raw_amount         6477 non-null   object        
 13  date               6477 non-null   datetime64[ns]

Sample data preview:

id	transaction_hash	from_address	to_address	amount	date
0	0x61c4fe...	0x0000...	0xabcd...	100000000000.0	2025-09-23
1	0xdf5012...	0x1234...	0x5678...	1000.0	2025-09-23

Query Examples

Filter by Token Address

# Filter transfers for a specific token
token_address = "0x13dd03e35e4bc8ba6172faee5d1e848b7e507f61"
token_transfers = df[df['token_address'] == token_address]
print(f"Transfers for token: {len(token_transfers)}")

Filter by Address

# Find all transfers from or to a specific address
address = "0xfde35125051c0f07448005dc85667d0459cc848b"
address_transfers = df[
    (df['from_address'] == address) | (df['to_address'] == address)
]
print(f"Transfers involving address: {len(address_transfers)}")

Filter by Date Range

# Filter by date range
df['date'] = pd.to_datetime(df['date'])
start_date = pd.to_datetime('2025-11-01')
end_date = pd.to_datetime('2025-11-30')
date_filtered = df[
    (df['date'] >= start_date) & (df['date'] <= end_date)
]
print(f"Transfers in November 2025: {len(date_filtered)}")

Aggregate by Token

# Count transfers per token
transfers_per_token = df.groupby('token_address').size().sort_values(ascending=False)
print(transfers_per_token)

Best Practices

1. Use incremental files for updates: If you already have historical data, download only the incremental files for new dates to save bandwidth.

2. Store locally: For repeated analysis, download the parquet files locally instead of reading from URL each time.

3. Handle large datasets: For very large datasets, consider using Dask or PyArrow for better memory management:

import pyarrow.parquet as pq

# Read specific columns only
table = pq.read_table('asset_operation_full.parquet', columns=['from_address', 'to_address', 'amount'])
df = table.to_pandas()

Data Update Schedule

• Full snapshots: Generated once (check for the first asset_operation_full.parquet)
• Incremental files: Generated daily with the day's transactions

Support

If you have questions about the data or encounter issues, please reach out through our official channels:

• Twitter: @JovayNetwork