Examples | Serializing and Deserializing of Types in the Database - arizvisa/ida-minsc GitHub Wiki

Exporting a structure from a database

need to save a structure? just pickle it.

import pickle
st = struc.search('_IO_STACK*')
with open(db.config.path('somefilerelativetodatabase'), 'wb') as outfile:
    pickle.dump(st, outfile)

Importing structure into a different database

import it into another database with:

import pickle
with open('/path/to/somefilerelativetodatabase', 'rb') as infile:
    st = pickle.load(infile)

Save all your structures

or do all of them.

a = [st for st in struc.iterate()]
pickle.dumps(a)

Local types

types are natively serializable so you can serialize them like that, but typically you export the local types as a C definition. However, if you don't want to generate a .h file you can just select them with

a = [(ti, name) for index, name, ti in db.types.iterate('_*')]
b = [(name, ti.serialize()) for ti in a]
with open('blah', 'wb') as outfile: pickle.dump(b, outfile)

then load them into another db.

with open('blah', 'rb') as infile: b = pickle.load(infile)
ordinal = db.types.count()
for name, serialized in b:
    ti = db.types.get(serialized)
    db.types.set(ordinal, name, ti)
    ordinal += 1

Function prototypes

how about all structures used in any of the parameters in a function?

f = yourfunc

results = []
for typeinfo in function.args(f):
    if struc.has(typeinfo):
        results.append(struc.by(typeinfo))
    continue
prototype = func.tag(f, '__typeinfo__')
serialized = pickle.dumps(results)

Now you can deserialize your data, and apply the type information that you extracted from your tag to it.

deserialized = pickle.loads(result)
func.tag(f, '__typeinfo__', prototype)

Import and Export the names of all undefined local types

# list undefined types
db.types.list(defined=False)

# save all of their names
exported = [name for ordinal, name, typeinfo in db.types.iterate(defined=False)]

# add empty types in a different database
[db.types.add(name) for name in exported]