Value.has_global_df_recursive()

Returns true when the value interacts with a global variable such as msg.sender, msg.data, etc. across multiple functions.

has_global_df_recursive() → bool

The function is the recursive/inter-procedural variant of has_global_df(). It traverses the data-flow graph across function boundaries and returns true if the Value interacts with a global variable at any point in its data flow, and false otherwise.

By "interacts," we mean the Value is used in an operation that reads or writes a global variable, or is combined with a value derived from a global. For example, if we analyze the lastDatasetId value, the following counts as an interaction because lastDatasetId is used to access the global mapping datasetOwners:

datasetOwners[lastDatasetId] = msg.sender;

The difference between has_global_df_recursive() and has_global_df() is scope: the non-recursive version analyzes only the current function, while has_global_df_recursive() follows the Value’s data flow recursively (across calls) until it either finds an interaction with a global variable or exhausts the flow.

Query Example

from glider import *

def query():
    instructions = (
        Functions()
        .with_signature("approve(address,uint256)")
        .exec(100)
        .filter(lambda function : function.get_contract().name == "ERC20")[0]
        .instructions()
        .exec()
    )

    for instruction in instructions:
        components = instruction.get_components()
        for component in components:
            if component.has_global_df_recursive():
                print("Value: " + component.expression)

    return instructions

Example Output