Overview

Ccrawl is composed of several python modules and sub-packages:

• Module main defines all commands of the ccrawl command line tool. All commands are handled by the python click package which deals with options and arguments.

• Module core defines all internal classes used to represent collected objects, ie. typedefs, structs, unions, C++ classes, functions’ prototypes, macros, etc.

• Module parser implements ccrawl’s interface to the clang compiler by relying on the libclang python package. This interface allows to collect C/C++ definitions from input (header) files inside a database.

• Module db defines ccrawl’s interface with the underlying databases, allowing to query for various properties of stored C/C++ definitions and to output the full definition of a chosen structure (all needed types recusively) or to instanciate an “external” object associated with the chosen definition (for example a ctypes instance, an amoco struct or a ghidra data type instance.)

• Module conf provides the global configuration based on the traitlets package.

• Module graphs provides the Node, Link and CGraph classes that are used to encode the dependency graph of a type and locate cyclic dependencies in this graph.

• Module utils implements the pyparsing utilities for decomposing a C/C++ type into a ccrawl object.

• Module graphs implements the graph classes (Node, Link, CGraph) for computing the dependency-graph of a set of types.

• Sub-packages formatters deals with translating the queried definitions into a specific language (raw, C/C++, amoco, ctypes, etc) and sub-package ext deals with instanciating the queried definitions into specific python tools (amoco, ctypes, Ghidra bridge, etc.)