def create_comparative_analysis_files(versions, count_p2sh, non_std_only):
# The following analysis reads/writes from/to large data files. Some of the steps can be ignored if those files have
# already been created (if more updated data is not requited). Otherwise lot of time will be put in re-parsing large
# files.
# Set version and chainstate dir name
str_version = "-".join([str(v) for v in versions])
fins = dict()
for v in versions:
# When using snapshots of the chainstate, we store it as 'chainstate/version
chainstate = 'chainstate/' + str(v)
# When not using a snapshot, we directly use the chainstate under btc_core_dir
# chainstate = 'chainstate'
# Check if the directory for data exists, create it otherwise.
if not path.isdir(CFG.data_path + str(v)):
mkdir(CFG.data_path + str(v))
# Set the name of the output data files
f_utxos, f_parsed_txs, f_parsed_utxos, f_dust = set_out_names(
v, count_p2sh, non_std_only)
fins[str(v)] = [f_parsed_txs, f_parsed_utxos]
# Parse all the data in the chainstate.
parse_ldb(f_utxos, fin_name=chainstate, version=v)
# Parses transactions and utxos from the dumped data.
transaction_dump(f_utxos, f_parsed_txs, version=v)
utxo_dump(f_utxos, f_parsed_utxos, version=v)
# Check if the directory for figures exist, create it otherwise.
if not path.isdir(CFG.figs_path + str_version):
mkdir(CFG.figs_path + str_version)
def run_experiment(coin, chainstate, count_p2sh, non_std_only):
"""
Runs the whole experiment. You may comment the parts of it you are not interested in to save time.
:param coin: Coin to be used in the experiment (bitcoin, litecoin, bitcoin cash, ...)
:type coin: str
:param chainstate: Chainstate path.
:type chainstate: str
:param count_p2sh: Whether P2SH outputs are included in the experiment or not.
:type count_p2sh: bool
:param non_std_only: Whether the experiment is performed only counting non standard outputs.
:type non_std_only:bool
:return:
"""
# The following analysis reads/writes from/to large data files. Some of the steps can be ignored if those files have
# already been created (if more updated data is not requited). Otherwise lot of time will be put in re-parsing large
# files.
# Set the name of the output data files
f_utxos, f_parsed_txs, f_parsed_utxos, f_dust = set_out_names(
count_p2sh, non_std_only)
# Parse all the data in the chainstate.
print "Parsing the chainstate."
parse_ldb(f_utxos, fin_name=chainstate)
# Parses transactions and utxos from the dumped data.
print "Adding meta-data for transactions and UTXOs."
transaction_dump(f_utxos, f_parsed_txs)
utxo_dump(f_utxos,
f_parsed_utxos,
coin,
count_p2sh=count_p2sh,
non_std_only=non_std_only)
# Print basic stats from data
print "Running overview analysis."
overview_from_file(f_parsed_txs, f_parsed_utxos)
# Generate plots from tx data (from f_parsed_txs)
print "Running transaction based analysis."
tx_based_analysis(f_parsed_txs)
# Generate plots from utxo data (from f_parsed_utxos)
print "Running UTXO based analysis."
utxo_based_analysis(f_parsed_utxos)
# # Aggregates dust and generates plots.
print "Running dust analysis."
dust_analysis(f_parsed_utxos, f_dust)
dust_analysis_all_fees(f_parsed_utxos)
# Generate plots with filters
print "Running analysis with filters."
utxo_based_analysis_with_filters(f_parsed_utxos)
tx_based_analysis_with_filters(f_parsed_txs)
# When using snapshots of the chainstate, we store it as 'chainstate/version
chainstate = 'chainstate/' + str(version)
# Check if the directory for data exists, create it otherwise.
if not path.isdir(CFG.data_path + str(version)):
mkdir(CFG.data_path + str(version))
# Set the name of the output data files
f_utxos = str(version) + "/utxos.json"
f_parsed_utxos = str(version) + "/parsed_utxos.json"
f_parsed_txs = str(version) + "/parsed_txs.json"
f_dust = str(version) + "/dust.json"
non_std_utxos = str(version) + "/parsed_non_std_utxos.json"
# Parse all the data in the chainstate.
parse_ldb(f_utxos, fin_name=chainstate, version=version)
# Parses transactions and utxos from the dumped data.
transaction_dump(f_utxos, f_parsed_txs, version=version)
utxo_dump(f_utxos, f_parsed_utxos, version=version)
# Check if the directory for figures exist, create it otherwise.
if not path.isdir(CFG.figs_path + str(version)):
mkdir(CFG.figs_path + str(version))
# Generate plots from tx and utxo data
plots_from_file(["total_len"] * 2,
y=["tx"] * 2,
xlabel="Total length (bytes)",
log_axis="x",
def run_experiment(version, chainstate, count_p2sh, non_std_only):
"""
Runs the whole experiment. You may comment the parts of it you are not interested in to save time.
:param version: Bitcoin core version, used to decide the folder in which to store the data.
:type version: float
:param chainstate: Chainstate path.
:type chainstate: str
:param count_p2sh: Whether P2SH outputs are included in the experiment or not.
:type count_p2sh: bool
:param non_std_only: Whether the experiment is performed only counting non standard outputs.
:type non_std_only:bool
:return:
"""
# The following analysis reads/writes from/to large data files. Some of the steps can be ignored if those files have
# already been created (if more updated data is not requited). Otherwise lot of time will be put in re-parsing large
# files.
# Check if the directories for both data and figures exist, create them otherwise.
if not path.isdir(CFG.data_path + str(version)) and not path.isdir(CFG.figs_path + str(version)):
mkdir(CFG.data_path + str(version))
mkdir(CFG.figs_path + str(version))
# Set the name of the output data files
f_utxos, f_parsed_txs, f_parsed_utxos, f_dust = set_out_names(version, count_p2sh, non_std_only)
# Parse all the data in the chainstate.
print "Parsing the chainstate."
parse_ldb(f_utxos, fin_name=chainstate, version=version)
# Parses transactions and utxos from the dumped data.
print "Adding meta-data for transactions and UTXOs."
transaction_dump(f_utxos, f_parsed_txs, version=version)
utxo_dump(f_utxos, f_parsed_utxos, count_p2sh=count_p2sh, non_std_only=non_std_only, version=version)
# Print basic stats from data
print "Running overview analysis."
overview_from_file(f_parsed_txs, f_parsed_utxos, version)
# Generate plots from tx data (from f_parsed_txs)
print "Running transaction based analysis."
tx_based_analysis(f_parsed_txs, version)
# Generate plots from utxo data (from f_parsed_utxos)
print "Running UTXO based analysis."
utxo_based_analysis(f_parsed_utxos, version)
# Aggregates dust and generates plots it.
print "Running dust analysis."
dust_analysis(f_parsed_utxos, f_dust, version)
compare_dust(version)
# Comparative data analysis (transactions and UTXOs)
print "Running comparative data analysis."
comparative_data_analysis(f_parsed_txs, f_parsed_utxos, version)
# Generate plots with filters
print "Running analysis with filters."
utxo_based_analysis_with_filters(f_parsed_utxos, version)
tx_based_analysis_with_filters(f_parsed_txs, version)
def printMostUpdateUTXOfile():
f_utxos = "utxos"
parse_ldb(f_utxos)
fileStat = os.stat("data/utxos")
return fileStat.st_size / 68