def stream(last_synced_block_file, lag, provider_uri, output, start_block, entity_types,
period_seconds=10, batch_size=2, block_batch_size=10, max_workers=5, log_file=None, pid_file=None):
"""Streams all data types to console or Google Pub/Sub."""
configure_logging(log_file)
configure_signals()
entity_types = parse_entity_types(entity_types)
validate_entity_types(entity_types, output)
from polygonetl.streaming.item_exporter_creator import create_item_exporter
from polygonetl.streaming.eth_streamer_adapter import EthStreamerAdapter
from blockchainetl_common.streaming.streamer import Streamer
# TODO: Implement fallback mechanism for provider uris instead of picking randomly
provider_uri = pick_random_provider_uri(provider_uri)
logging.info('Using ' + provider_uri)
streamer_adapter = EthStreamerAdapter(
batch_web3_provider=ThreadLocalProxy(lambda: get_provider_from_uri(provider_uri, batch=True)),
item_exporter=create_item_exporters(output),
batch_size=batch_size,
max_workers=max_workers,
entity_types=entity_types
)
streamer = Streamer(
blockchain_streamer_adapter=streamer_adapter,
last_synced_block_file=last_synced_block_file,
lag=lag,
start_block=start_block,
period_seconds=period_seconds,
block_batch_size=block_batch_size,
pid_file=pid_file
)
streamer.stream()
def get_block_range_for_date(provider_uri, date, output, chain='polygon'):
"""Outputs start and end blocks for given date."""
provider = get_provider_from_uri(provider_uri)
web3 = Web3(provider)
eth_service = EthService(web3)
start_block, end_block = eth_service.get_block_range_for_date(date)
with smart_open(output, 'w') as output_file:
output_file.write('{},{}\n'.format(start_block, end_block))
def get_partitions(start, end, partition_batch_size, provider_uri):
"""Yield partitions based on input data type."""
if is_date_range(start, end) or is_unix_time_range(start, end):
if is_date_range(start, end):
start_date = datetime.strptime(start, '%Y-%m-%d').date()
end_date = datetime.strptime(end, '%Y-%m-%d').date()
elif is_unix_time_range(start, end):
if len(start) == 10 and len(end) == 10:
start_date = datetime.utcfromtimestamp(int(start)).date()
end_date = datetime.utcfromtimestamp(int(end)).date()
elif len(start) == 13 and len(end) == 13:
start_date = datetime.utcfromtimestamp(int(start) / 1e3).date()
end_date = datetime.utcfromtimestamp(int(end) / 1e3).date()
day = timedelta(days=1)
provider = get_provider_from_uri(provider_uri)
web3 = Web3(provider)
eth_service = EthService(web3)
while start_date <= end_date:
batch_start_block, batch_end_block = eth_service.get_block_range_for_date(
start_date)
partition_dir = '/date={start_date!s}/'.format(
start_date=start_date)
yield batch_start_block, batch_end_block, partition_dir
start_date += day
elif is_block_range(start, end):
start_block = int(start)
end_block = int(end)
for batch_start_block in range(start_block, end_block + 1,
partition_batch_size):
batch_end_block = batch_start_block + partition_batch_size - 1
if batch_end_block > end_block:
batch_end_block = end_block
padded_batch_start_block = str(batch_start_block).zfill(8)
padded_batch_end_block = str(batch_end_block).zfill(8)
partition_dir = '/start_block={padded_batch_start_block}/end_block={padded_batch_end_block}'.format(
padded_batch_start_block=padded_batch_start_block,
padded_batch_end_block=padded_batch_end_block,
)
yield batch_start_block, batch_end_block, partition_dir
else:
raise ValueError(
'start and end must be either block numbers or ISO dates or Unix times'
)
def export_token_transfers(start_block, end_block, batch_size, output,
max_workers, provider_uri, tokens):
"""Exports ERC20/ERC721 transfers."""
job = ExportTokenTransfersJob(
start_block=start_block,
end_block=end_block,
batch_size=batch_size,
web3=ThreadLocalProxy(
lambda: Web3(get_provider_from_uri(provider_uri))),
item_exporter=token_transfers_item_exporter(output),
max_workers=max_workers,
tokens=tokens)
job.run()
def export_geth_traces(start_block, end_block, batch_size, output, max_workers,
provider_uri):
"""Exports traces from geth node."""
job = ExportGethTracesJob(
start_block=start_block,
end_block=end_block,
batch_size=batch_size,
batch_web3_provider=ThreadLocalProxy(
lambda: get_provider_from_uri(provider_uri, batch=True)),
max_workers=max_workers,
item_exporter=geth_traces_item_exporter(output))
job.run()
def export_tokens(token_addresses,
output,
max_workers,
provider_uri,
chain='polygon'):
"""Exports ERC20/ERC721 tokens."""
with smart_open(token_addresses, 'r') as token_addresses_file:
job = ExportTokensJob(
token_addresses_iterable=(
token_address.strip()
for token_address in token_addresses_file),
web3=ThreadLocalProxy(
lambda: Web3(get_provider_from_uri(provider_uri))),
item_exporter=tokens_item_exporter(output),
max_workers=max_workers)
job.run()
def extract_tokens(contracts, provider_uri, output, max_workers):
"""Extracts tokens from contracts file."""
set_max_field_size_limit()
with smart_open(contracts, 'r') as contracts_file:
if contracts.endswith('.json'):
contracts_iterable = (json.loads(line) for line in contracts_file)
else:
contracts_iterable = csv.DictReader(contracts_file)
job = ExtractTokensJob(
contracts_iterable=contracts_iterable,
web3=ThreadLocalProxy(lambda: Web3(get_provider_from_uri(provider_uri))),
max_workers=max_workers,
item_exporter=tokens_item_exporter(output))
job.run()
def export_traces(start_block, end_block, batch_size, output, max_workers, provider_uri,
genesis_traces, daofork_traces, timeout=60, chain='polygon'):
"""Exports traces from parity node."""
if chain == 'classic' and daofork_traces == True:
raise ValueError(
'Classic chain does not include daofork traces. Disable daofork traces with --no-daofork-traces option.')
job = ExportTracesJob(
start_block=start_block,
end_block=end_block,
batch_size=batch_size,
web3=ThreadLocalProxy(lambda: Web3(get_provider_from_uri(provider_uri, timeout=timeout))),
item_exporter=traces_item_exporter(output),
max_workers=max_workers,
include_genesis_traces=genesis_traces,
include_daofork_traces=daofork_traces)
job.run()
def export_contracts(batch_size,
contract_addresses,
output,
max_workers,
provider_uri,
chain='polygon'):
"""Exports contracts bytecode and sighashes."""
with smart_open(contract_addresses, 'r') as contract_addresses_file:
contract_addresses = (contract_address.strip()
for contract_address in contract_addresses_file
if contract_address.strip())
job = ExportContractsJob(
contract_addresses_iterable=contract_addresses,
batch_size=batch_size,
batch_web3_provider=ThreadLocalProxy(
lambda: get_provider_from_uri(provider_uri, batch=True)),
item_exporter=contracts_item_exporter(output),
max_workers=max_workers)
job.run()
def export_receipts_and_logs(batch_size,
transaction_hashes,
provider_uri,
max_workers,
receipts_output,
logs_output,
chain='polygon'):
"""Exports receipts and logs."""
with smart_open(transaction_hashes, 'r') as transaction_hashes_file:
job = ExportReceiptsJob(
transaction_hashes_iterable=(
transaction_hash.strip()
for transaction_hash in transaction_hashes_file),
batch_size=batch_size,
batch_web3_provider=ThreadLocalProxy(
lambda: get_provider_from_uri(provider_uri, batch=True)),
max_workers=max_workers,
item_exporter=receipts_and_logs_item_exporter(
receipts_output, logs_output),
export_receipts=receipts_output is not None,
export_logs=logs_output is not None)
job.run()
def export_all_common(partitions, output_dir, provider_uri, max_workers,
batch_size):
for batch_start_block, batch_end_block, partition_dir in partitions:
# # # start # # #
start_time = time()
padded_batch_start_block = str(batch_start_block).zfill(8)
padded_batch_end_block = str(batch_end_block).zfill(8)
block_range = '{padded_batch_start_block}-{padded_batch_end_block}'.format(
padded_batch_start_block=padded_batch_start_block,
padded_batch_end_block=padded_batch_end_block,
)
file_name_suffix = '{padded_batch_start_block}_{padded_batch_end_block}'.format(
padded_batch_start_block=padded_batch_start_block,
padded_batch_end_block=padded_batch_end_block,
)
# # # blocks_and_transactions # # #
blocks_output_dir = '{output_dir}/blocks{partition_dir}'.format(
output_dir=output_dir,
partition_dir=partition_dir,
)
os.makedirs(os.path.dirname(blocks_output_dir), exist_ok=True)
transactions_output_dir = '{output_dir}/transactions{partition_dir}'.format(
output_dir=output_dir,
partition_dir=partition_dir,
)
os.makedirs(os.path.dirname(transactions_output_dir), exist_ok=True)
blocks_file = '{blocks_output_dir}/blocks_{file_name_suffix}.csv'.format(
blocks_output_dir=blocks_output_dir,
file_name_suffix=file_name_suffix,
)
transactions_file = '{transactions_output_dir}/transactions_{file_name_suffix}.csv'.format(
transactions_output_dir=transactions_output_dir,
file_name_suffix=file_name_suffix,
)
logger.info('Exporting blocks {block_range} to {blocks_file}'.format(
block_range=block_range,
blocks_file=blocks_file,
))
logger.info(
'Exporting transactions from blocks {block_range} to {transactions_file}'
.format(
block_range=block_range,
transactions_file=transactions_file,
))
job = ExportBlocksJob(
start_block=batch_start_block,
end_block=batch_end_block,
batch_size=batch_size,
batch_web3_provider=ThreadLocalProxy(
lambda: get_provider_from_uri(provider_uri, batch=True)),
max_workers=max_workers,
item_exporter=blocks_and_transactions_item_exporter(
blocks_file, transactions_file),
export_blocks=blocks_file is not None,
export_transactions=transactions_file is not None)
job.run()
# # # token_transfers # # #
token_transfers_file = None
if is_log_filter_supported(provider_uri):
token_transfers_output_dir = '{output_dir}/token_transfers{partition_dir}'.format(
output_dir=output_dir,
partition_dir=partition_dir,
)
os.makedirs(os.path.dirname(token_transfers_output_dir),
exist_ok=True)
token_transfers_file = '{token_transfers_output_dir}/token_transfers_{file_name_suffix}.csv'.format(
token_transfers_output_dir=token_transfers_output_dir,
file_name_suffix=file_name_suffix,
)
logger.info(
'Exporting ERC20 transfers from blocks {block_range} to {token_transfers_file}'
.format(
block_range=block_range,
token_transfers_file=token_transfers_file,
))
job = ExportTokenTransfersJob(
start_block=batch_start_block,
end_block=batch_end_block,
batch_size=batch_size,
web3=ThreadLocalProxy(
lambda: Web3(get_provider_from_uri(provider_uri))),
item_exporter=token_transfers_item_exporter(
token_transfers_file),
max_workers=max_workers)
job.run()
# # # receipts_and_logs # # #
cache_output_dir = '{output_dir}/.tmp{partition_dir}'.format(
output_dir=output_dir,
partition_dir=partition_dir,
)
os.makedirs(os.path.dirname(cache_output_dir), exist_ok=True)
transaction_hashes_file = '{cache_output_dir}/transaction_hashes_{file_name_suffix}.csv'.format(
cache_output_dir=cache_output_dir,
file_name_suffix=file_name_suffix,
)
logger.info(
'Extracting hash column from transaction file {transactions_file}'.
format(transactions_file=transactions_file, ))
extract_csv_column_unique(transactions_file, transaction_hashes_file,
'hash')
receipts_output_dir = '{output_dir}/receipts{partition_dir}'.format(
output_dir=output_dir,
partition_dir=partition_dir,
)
os.makedirs(os.path.dirname(receipts_output_dir), exist_ok=True)
logs_output_dir = '{output_dir}/logs{partition_dir}'.format(
output_dir=output_dir,
partition_dir=partition_dir,
)
os.makedirs(os.path.dirname(logs_output_dir), exist_ok=True)
receipts_file = '{receipts_output_dir}/receipts_{file_name_suffix}.csv'.format(
receipts_output_dir=receipts_output_dir,
file_name_suffix=file_name_suffix,
)
logs_file = '{logs_output_dir}/logs_{file_name_suffix}.csv'.format(
logs_output_dir=logs_output_dir,
file_name_suffix=file_name_suffix,
)
logger.info(
'Exporting receipts and logs from blocks {block_range} to {receipts_file} and {logs_file}'
.format(
block_range=block_range,
receipts_file=receipts_file,
logs_file=logs_file,
))
with smart_open(transaction_hashes_file, 'r') as transaction_hashes:
job = ExportReceiptsJob(
transaction_hashes_iterable=(
transaction_hash.strip()
for transaction_hash in transaction_hashes),
batch_size=batch_size,
batch_web3_provider=ThreadLocalProxy(
lambda: get_provider_from_uri(provider_uri, batch=True)),
max_workers=max_workers,
item_exporter=receipts_and_logs_item_exporter(
receipts_file, logs_file),
export_receipts=receipts_file is not None,
export_logs=logs_file is not None)
job.run()
# # # contracts # # #
contract_addresses_file = '{cache_output_dir}/contract_addresses_{file_name_suffix}.csv'.format(
cache_output_dir=cache_output_dir,
file_name_suffix=file_name_suffix,
)
logger.info(
'Extracting contract_address from receipt file {receipts_file}'.
format(receipts_file=receipts_file))
extract_csv_column_unique(receipts_file, contract_addresses_file,
'contract_address')
contracts_output_dir = '{output_dir}/contracts{partition_dir}'.format(
output_dir=output_dir,
partition_dir=partition_dir,
)
os.makedirs(os.path.dirname(contracts_output_dir), exist_ok=True)
contracts_file = '{contracts_output_dir}/contracts_{file_name_suffix}.csv'.format(
contracts_output_dir=contracts_output_dir,
file_name_suffix=file_name_suffix,
)
logger.info(
'Exporting contracts from blocks {block_range} to {contracts_file}'
.format(
block_range=block_range,
contracts_file=contracts_file,
))
with smart_open(contract_addresses_file,
'r') as contract_addresses_file:
contract_addresses = (
contract_address.strip()
for contract_address in contract_addresses_file
if contract_address.strip())
job = ExportContractsJob(
contract_addresses_iterable=contract_addresses,
batch_size=batch_size,
batch_web3_provider=ThreadLocalProxy(
lambda: get_provider_from_uri(provider_uri, batch=True)),
item_exporter=contracts_item_exporter(contracts_file),
max_workers=max_workers)
job.run()
# # # tokens # # #
if token_transfers_file is not None:
token_addresses_file = '{cache_output_dir}/token_addresses_{file_name_suffix}'.format(
cache_output_dir=cache_output_dir,
file_name_suffix=file_name_suffix,
)
logger.info(
'Extracting token_address from token_transfers file {token_transfers_file}'
.format(token_transfers_file=token_transfers_file, ))
extract_csv_column_unique(token_transfers_file,
token_addresses_file, 'token_address')
tokens_output_dir = '{output_dir}/tokens{partition_dir}'.format(
output_dir=output_dir,
partition_dir=partition_dir,
)
os.makedirs(os.path.dirname(tokens_output_dir), exist_ok=True)
tokens_file = '{tokens_output_dir}/tokens_{file_name_suffix}.csv'.format(
tokens_output_dir=tokens_output_dir,
file_name_suffix=file_name_suffix,
)
logger.info(
'Exporting tokens from blocks {block_range} to {tokens_file}'.
format(
block_range=block_range,
tokens_file=tokens_file,
))
with smart_open(token_addresses_file, 'r') as token_addresses:
job = ExportTokensJob(
token_addresses_iterable=(
token_address.strip()
for token_address in token_addresses),
web3=ThreadLocalProxy(
lambda: Web3(get_provider_from_uri(provider_uri))),
item_exporter=tokens_item_exporter(tokens_file),
max_workers=max_workers)
job.run()
# # # finish # # #
shutil.rmtree(os.path.dirname(cache_output_dir))
end_time = time()
time_diff = round(end_time - start_time, 5)
logger.info(
'Exporting blocks {block_range} took {time_diff} seconds'.format(
block_range=block_range,
time_diff=time_diff,
))