Options for Enum Generation
When generating with schema.enum, some options will be used.
Copy Candidates
If schema.enum contains a mutable object such as list, the question is whether to copy it or not.
options.enum_copy_style can be used to specify whether or not to copy.
If enum_copy_style is NO_COPY, the returned value is same instance to one of candidate.
For example:
>>> import ranjg
>>> from ranjg import options
>>> from ranjg.options import Options
>>> schema = {'enum': [[1, 2]]} # only one candidate: [1, 2]
>>> generated = ranjg.gen(schema, options=Options(enum_copy_style=options.NO_COPY))
>>> assert generated == [1, 2]
>>>
>>> # Rewrite schema indirectly using the fact that it is NO_COPY.
>>> generated.append(3)
>>> assert schema['enum'][0] == [1, 2, 3]
If enum_copy_style is SHALLOW_COPY or DEEP_COPY, the returned value is copy of one of candidate.
For example:
>>> import ranjg
>>> from ranjg import options
>>> from ranjg.options import Options
>>> schema = {'enum': [[1, 2]]} # only one candidate: [1, 2]
>>> generated = ranjg.gen(schema, options=Options(enum_copy_style=options.DEEP_COPY))
>>> assert generated == [1, 2]
>>>
>>> # Since generated is a copy, editing it will not affect the schema.
>>> generated.append(3)
>>> assert schema['enum'][0] == [1, 2]
- note
To copy, it uses package copy. The respective behavior of shallow copy and deep copy is also the same as
copy.copy()andcopy.deepcopy().