1.8 KiB
Integration with Python
Export to Python
When the Erg script is compiled, a .pyc file is generated, which can simply be imported as a Python module. However, variables set to private on the Erg side cannot be accessed from Python.
# foo.er
.public = "this is a public variable"
private = "this is a private variable"
erg --compile foo.er
import foo
print(foo.public)
print(foo.private) # AttributeError:
Import from Python
All objects imported from Python are by default of type Object. Since no comparisons can be made at this point, it is necessary to refine the type.
Type Specification in the Standard Library
All APIs in the Python standard library are type specified by the Erg development team.
time = pyimport "time"
time.sleep! 1
Type Specification for User Scripts
Create a foo.d.er file that types the Python foo module.
Type hints on the Python side are ignored since they are not 100% guaranteed.
# foo.py
X = ...
def bar(x):
...
def baz():
...
...
# foo.d.er
foo = pyimport "foo"
.X = declare foo.'X', Int
.bar = declare foo.'bar', Int -> Int
.baz! = declare foo.'baz', () => Int
foo = pyimport "foo"
assert foo.bar(1) in Int
This ensures type safety by performing type checking at runtime. The declare function works roughly as follows.
declare|S: Subroutine| sub!: S, T =
# Actually, => can be cast to a function without block side effects
x =>
assert x in T.Input
y = sub!(x)
assert y in T.Output
y
Since this is a runtime overhead, a project is planned to statically type analyze Python scripts with Erg's type system.