This commit is contained in:
Austin 2025-12-09 13:24:24 -05:00
parent 3697967c46
commit b0e51e2b4d
2 changed files with 363 additions and 0 deletions

View File

@ -293,5 +293,19 @@
sql: |
{{ fsc_utils.create_udf_stablecoin_data_parse() | indent(4) }}
- name: {{ schema }}.udf_encode_contract_call
signature:
- [function_abi, VARIANT]
- [input_values, ARRAY]
return_type: STRING
options: |
LANGUAGE PYTHON
RUNTIME_VERSION = '3.10'
PACKAGES = ('eth-abi')
HANDLER = 'encode_call'
COMMENT = '{{ fsc_utils.udf_encode_contract_call_comment() }}'
sql: |
{{ fsc_utils.create_udf_encode_contract_call() | indent(4) }}
{% endmacro %}

View File

@ -784,4 +784,353 @@ class udf_stablecoin_data_parse:
except Exception as error:
raise Exception(f'Error parsing peggedData content: {str(error)}')
{% endmacro %}
{% macro create_udf_encode_contract_call() %}
def encode_call(function_abi, input_values):
"""
Encodes EVM contract function calls into ABI-encoded calldata.
This function generates complete calldata (selector + encoded params) that can be
used directly in eth_call JSON-RPC requests to query contract state.
"""
import eth_abi
from eth_hash.auto import keccak
import json
def get_function_signature(abi):
"""
Generate function signature using the same logic as utils.udf_evm_text_signature.
Examples:
balanceOf(address)
transfer(address,uint256)
swap((address,address,uint256))
"""
def generate_signature(inputs):
signature_parts = []
for input_data in inputs:
if 'components' in input_data:
# Handle nested tuples
component_signature_parts = []
components = input_data['components']
component_signature_parts.extend(generate_signature(components))
component_signature_parts[-1] = component_signature_parts[-1].rstrip(",")
if input_data['type'].endswith('[]'):
signature_parts.append("(" + "".join(component_signature_parts) + ")[],")
else:
signature_parts.append("(" + "".join(component_signature_parts) + "),")
else:
# Clean up Solidity-specific modifiers
signature_parts.append(input_data['type'].replace('enum ', '').replace(' payable', '') + ",")
return signature_parts
signature_parts = [abi['name'] + "("]
signature_parts.extend(generate_signature(abi.get('inputs', [])))
if len(signature_parts) > 1:
signature_parts[-1] = signature_parts[-1].rstrip(",") + ")"
else:
signature_parts.append(")")
return "".join(signature_parts)
def function_selector(abi):
"""Calculate 4-byte function selector using Keccak256 hash."""
signature = get_function_signature(abi)
hash_bytes = keccak(signature.encode('utf-8'))
return hash_bytes[:4].hex(), signature
def get_canonical_type(input_spec):
"""
Convert ABI input spec to canonical type string for eth_abi encoding.
Handles tuple expansion: tuple -> (address,uint256,bytes)
"""
param_type = input_spec['type']
if param_type.startswith('tuple'):
components = input_spec.get('components', [])
component_types = ','.join([get_canonical_type(comp) for comp in components])
canonical = f"({component_types})"
# Preserve array suffixes: tuple[] -> (address,uint256)[]
if param_type.endswith('[]'):
array_suffix = param_type[5:] # Everything after 'tuple'
canonical += array_suffix
return canonical
return param_type
def prepare_value(value, param_type, components=None):
"""
Convert Snowflake values to Python types suitable for eth_abi encoding.
Handles type coercion and format normalization for all Solidity types.
"""
# Handle null/None values with sensible defaults
if value is None:
if param_type.startswith('uint') or param_type.startswith('int'):
return 0
elif param_type == 'address':
return '0x' + '0' * 40
elif param_type == 'bool':
return False
elif param_type.startswith('bytes'):
return b''
else:
return value
# CRITICAL: Check arrays FIRST (before base types)
# This prevents bytes[] from matching the bytes check
if param_type.endswith('[]'):
base_type = param_type[:-2]
if not isinstance(value, list):
return []
# Special handling for tuple arrays
if base_type == 'tuple' and components:
return [prepare_tuple(v, components) for v in value]
else:
return [prepare_value(v, base_type) for v in value]
# Base type conversions
if param_type == 'address':
addr = str(value).lower()
if not addr.startswith('0x'):
addr = '0x' + addr
return addr
if param_type.startswith('uint') or param_type.startswith('int'):
return int(value)
if param_type == 'bool':
if isinstance(value, str):
return value.lower() in ('true', '1', 'yes')
return bool(value)
if param_type.startswith('bytes'):
if isinstance(value, str):
if value.startswith('0x'):
value = value[2:]
return bytes.fromhex(value)
return value
if param_type == 'string':
return str(value)
return value
def prepare_tuple(value, components):
"""
Recursively prepare tuple values, handling nested structures.
Tuples can contain other tuples, arrays, or tuple arrays.
"""
if not isinstance(value, (list, tuple)):
# Support dict-style input (by component name)
if isinstance(value, dict):
value = [value.get(comp.get('name', f'field_{i}'))
for i, comp in enumerate(components)]
else:
return value
result = []
for i, comp in enumerate(components):
if i >= len(value):
result.append(None)
continue
comp_type = comp['type']
val = value[i]
# Handle tuple arrays within tuples
if comp_type.endswith('[]') and comp_type.startswith('tuple'):
sub_components = comp.get('components', [])
result.append(prepare_value(val, comp_type, sub_components))
elif comp_type.startswith('tuple'):
# Single tuple (not array)
sub_components = comp.get('components', [])
result.append(prepare_tuple(val, sub_components))
else:
result.append(prepare_value(val, comp_type))
return tuple(result)
try:
inputs = function_abi.get('inputs', [])
# Calculate selector using battle-tested signature generation
selector_hex, signature = function_selector(function_abi)
# Functions with no inputs only need the selector
if not inputs:
return '0x' + selector_hex
# Prepare values for encoding
prepared_values = []
for i, inp in enumerate(inputs):
if i >= len(input_values):
prepared_values.append(None)
continue
value = input_values[i]
param_type = inp['type']
# Handle tuple arrays at top level
if param_type.endswith('[]') and param_type.startswith('tuple'):
components = inp.get('components', [])
prepared_values.append(prepare_value(value, param_type, components))
elif param_type.startswith('tuple'):
# Single tuple (not array)
components = inp.get('components', [])
prepared_values.append(prepare_tuple(value, components))
else:
prepared_values.append(prepare_value(value, param_type))
# Get canonical type strings for eth_abi (expands tuples)
types = [get_canonical_type(inp) for inp in inputs]
# Encode parameters using eth_abi
encoded_params = eth_abi.encode(types, prepared_values).hex()
# Return complete calldata: selector + encoded params
return '0x' + selector_hex + encoded_params
except Exception as e:
# Return structured error for debugging
import traceback
return json.dumps({
'error': str(e),
'traceback': traceback.format_exc(),
'function': function_abi.get('name', 'unknown'),
'signature': signature if 'signature' in locals() else 'not computed',
'selector': '0x' + selector_hex if 'selector_hex' in locals() else 'not computed',
'types': types if 'types' in locals() else 'not computed'
})
{% endmacro %}
{% macro udf_encode_contract_call_comment() %}
Encodes EVM contract function calls into hex calldata format for eth_call RPC requests.
PURPOSE:
Converts human-readable function parameters into ABI-encoded calldata that can be sent
to Ethereum nodes via JSON-RPC. Handles all Solidity types including complex nested
structures like tuples and arrays.
PARAMETERS:
function_abi (VARIANT):
- JSON object containing the function ABI definition
- Must include: "name" (string) and "inputs" (array of input definitions)
- Each input needs: "name", "type", and optionally "components" for tuples
input_values (ARRAY):
- Array of values matching the function inputs in order
- Values should be provided as native Snowflake types:
* addresses: strings (with or without 0x prefix)
* uint/int: numbers
* bool: booleans
* bytes/bytes32: hex strings (with or without 0x prefix)
* arrays: Snowflake arrays
* tuples: Snowflake arrays in component order
RETURNS:
STRING: Complete calldata as hex string with 0x prefix
- Format: 0x{4-byte selector}{encoded parameters}
- Can be used directly in eth_call RPC requests
- Returns JSON error object if encoding fails
EXAMPLES:
-- Simple function with no inputs
SELECT crosschain_dev.utils.udf_encode_contract_call(
PARSE_JSON(''{"name": "totalSupply", "inputs": []}''),
ARRAY_CONSTRUCT()
);
-- Returns: 0x18160ddd
-- Function with single address parameter
SELECT crosschain_dev.utils.udf_encode_contract_call(
PARSE_JSON(''{
"name": "balanceOf",
"inputs": [{"name": "account", "type": "address"}]
}''),
ARRAY_CONSTRUCT(''0xA0b86991c6218b36c1d19D4a2e9Eb0cE3606eB48'')
);
-- Returns: 0x70a08231000000000000000000000000a0b86991c6218b36c1d19d4a2e9eb0ce3606eb48
-- Function with multiple parameters
SELECT crosschain_dev.utils.udf_encode_contract_call(
PARSE_JSON(''{
"name": "transfer",
"inputs": [
{"name": "to", "type": "address"},
{"name": "amount", "type": "uint256"}
]
}''),
ARRAY_CONSTRUCT(''0x1234567890123456789012345678901234567890'', 1000000)
);
-- Complex function with nested tuples
SELECT crosschain_dev.utils.udf_encode_contract_call(
PARSE_JSON(''{
"name": "swap",
"inputs": [{
"name": "params",
"type": "tuple",
"components": [
{"name": "tokenIn", "type": "address"},
{"name": "tokenOut", "type": "address"},
{"name": "amountIn", "type": "uint256"}
]
}]
}''),
ARRAY_CONSTRUCT(
ARRAY_CONSTRUCT(
''0xA0b86991c6218b36c1d19D4a2e9Eb0cE3606eB48'',
''0xC02aaA39b223FE8D0A0e5C4F27eAD9083C756Cc2'',
1000000
)
)
);
TYPICAL WORKFLOW:
1. Get function ABI from crosschain.evm.dim_contract_abis
2. Prepare input values as Snowflake arrays
3. Encode using this function
4. Execute via eth_call RPC (ai.live.udf_api)
5. Decode response using streamline_dev.utils.udf_evm_decode_trace
SUPPORTED TYPES:
- address: Ethereum addresses
- uint8, uint16, ..., uint256: Unsigned integers
- int8, int16, ..., int256: Signed integers
- bool: Boolean values
- bytes, bytes1, ..., bytes32: Fixed and dynamic byte arrays
- string: Dynamic strings
- Arrays: Any type followed by []
- Tuples: Nested structures with components
- Nested combinations: tuple[], tuple[][], etc.
NOTES:
- Function selector is automatically calculated using Keccak256
- Compatible with existing utils.udf_evm_text_signature and utils.udf_keccak256
- Handles gas-optimized function names (e.g., selector 0x00000000)
- Tuples must be provided as arrays in component order
- Empty arrays are valid for array-type parameters
ERROR HANDLING:
- Returns JSON error object on failure
- Check if result starts with "{" to detect errors
- Error object includes: error message, traceback, function name, types
RELATED FUNCTIONS:
- utils.udf_evm_text_signature: Generate function signature
- utils.udf_keccak256: Calculate function selector
- streamline_dev.utils.udf_evm_decode_trace: Decode call results
VERSION: 1.0
AUTHOR: Flipside Crypto Data Engineering
LAST UPDATED: 2024-12-09
{% endmacro %}