def from_dex_classes(classes):
"""Extract information from dex classes."""
doc = dict()
for class_, infos in classes.items():
pkg, name = package.split_class(class_)
_setitem(doc, "dex::package", pkg)
_setitem(doc, "dex::class", name)
pkg, name = package.split_class(infos.get("super"))
_setitem(doc, "dex::super-package", pkg)
_setitem(doc, "dex::super-class", name)
return doc
def test_can_split_class_identifier():
# ok
assert package.split_class("LProcess;") == (None, "Process")
assert package.split_class("Landroid/os/Process;") == [
"android/os", "Process"
]
assert package.split_class("[Landroid/os/Process;") == [
"android/os", "Process"
]
# errors
## no L at beginning
assert package.split_class("android/os/Process;") == (None, None)
## no ; at the end
assert package.split_class("Landroid/os/Process") == (None, None)
def from_dex_fields(fields):
"""Extract information from dex fields."""
doc = dict()
for infos in fields.values():
pkg, class_ = package.split_class(infos.get("class"))
_setitem(doc, "dex::package", pkg)
_setitem(doc, "dex::class", class_)
_setitem(doc, "dex::field", infos.get("name"))
return doc
def from_dex_methods(methods):
"""Extract information from dex methods."""
doc = dict()
for infos in methods.values():
pkg, class_ = package.split_class(infos.get("class"))
_setitem(doc, "dex::package", pkg)
_setitem(doc, "dex::class", class_)
_setitem(doc, "dex::method", infos.get("name"))
_setitem(doc, "dex::code", infos.get("sha256"), on_key=lower)
return doc
def from_dex(doc):
"""Index artifacts from dex."""
entries = defaultdict(set)
entries["source.debug"].add(doc.get("debug", False))
entries["source.format"].add(_lownone(doc.get("format")))
for x in doc.get("strings", []):
if x is not None and len(x) <= 1000:
entries["source.string"].add(x.lower())
for x in doc.get("invokes", []):
entries["source.invoke"].add(_lownone(x))
for ID, field in doc.get("fields", {}).items():
entries["source.field.path"].add(_lownone(ID))
entries["source.field.name"].add(_lownone(field.get("name")))
for ID, class_ in doc.get("classes", {}).items():
pack, name = package.split_class(ID)
entries["source.class.path"].add(_lownone(ID))
entries["source.class.name"].add(_lownone(name))
entries["source.class.package"].add(_lownone(pack))
for inter_ in class_.get("inter", []):
entries["source.class.inter"].add(_lownone(inter_))
for super_ in class_.get("super", []):
entries["source.class.super"].add(_lownone(super_))
for ID, method in doc.get("methods", {}).items():
entries["source.method.path"].add(_lownone(ID))
entries["source.method.name"].add(_lownone(method.get("name")))
entries["source.method.code"].add(_lownone(method.get("sha256")))
return dict(entries)