Techniques for traversing representations of source code

A system comprising:

• a processor;
• and
• a memory storing instructions which when executed by the processor causes the processor to:
• obtain source code;
• generate a representation of the source code, the representation including a plurality of nodes;
• select an entry point in the representation of the source code, wherein the entry point is a node of the plurality of nodes;
• push metadata associated with the entry point to a stack, wherein the entry point includes a first method body and the metadata is a reference to a class declaration of the first method body;
• begin to traverse the representation of the source code at the entry point, on a node-by-node basis, to identify one or more potential vulnerabilities;
• store a history of the traverse including metadata associated with visited concrete implementations of method invocations;
• continue to traverse the representation of the source code based at least in part on the history of the traverse, where when concrete implementations are stored in the metadata, select a subset of the concrete implementations to traverse;
• in response to traversing the representation of the source code, identify the one or more potential vulnerabilities;
• and
• output the one or more potential vulnerabilities.

A non-transitory computer readable medium storing instructions which when executed by a processor causes the processor to:

• obtain source code;
• generate a representation of the source code, the representation including a plurality of nodes;
• select an entry point in the representation of the source code, wherein the entry point is a node of the plurality of nodes;
• push metadata associated with the entry point to a stack, wherein the entry point includes a first method body and the metadata is a reference to a class declaration of the first method body;
• begin to traverse the representation of the source code at the entry point, on a node-by-node basis, to identify one or more potential vulnerabilities;
• store a history of the traverse including metadata associated with visited concrete implementations of method invocations;
• continue to traverse the representation of the source code based at least in part on the history of the traverse, where when concrete implementations are stored in the metadata, select a subset of the concrete implementations to traverse;
• in response to traversing the representation of the source code, identify the one or more potential vulnerabilities;
• and
• output the one or more potential vulnerabilities.