SmartPy

31 March 2026 11:48

but this solution is far better

SmartPy

31 March 2026 11:46

If your goal is to compile from Python code instead of invoking python3 contract.py, you can do it through a test scenario: instantiate the contract, add it to the scenario, and then read the compiled code and initial storage from origination_result. Like that:

import smartpy as sp

@sp.module
def main():
    class C(sp.Contract):
        def __init__(self, x: sp.int):
            self.data.x = x

        @sp.entrypoint
        def ep(self):
            pass

def compile_contract(contract_ref, *args):
    scenario = sp.test_scenario()
    contract = contract_ref(*args)
    scenario += contract
    return {
        "code_json": contract.origination_result["code_json"],
        "storage_json": contract.origination_result["storage_json"],
        "code_tz": contract.origination_result["code_tz"],
        "storage_tz": contract.origination_result["storage_tz"],
    }

result = compile_contract(main.C, 0)

SmartPy

21 March 2026 14:10

So the michelson runtime/tezos runtime will return a list of operations to the chain to be run after the entrypoint logic is done, so this will mitigate against common reentrancy attacks, however there are still things to be aware of, e.g. see:
https://octez.tezos.com/docs/alpha/michelson_anti_patterns.html

smartpy tracks the effects of functions and entrypoints that perform operations - and should not let you run effectful stuff from non-effectful code - this should in fact be a compilation error.

See with_operations: https://smartpy.tezos.com/manual/data-types/effects.html#effects

SmartPy

20 March 2026 10:14

And is possible to read an address balance?

SmartPy

17 March 2026 11:40

Is the inability to read the balance of another smart contract a limitation of SmartPy, or is Michelson designed that way?

SmartPy

16 March 2026 16:51

Yes, that's the only way currently.
Which alternative are you thinking of? It's not exactly that is truly necessary, simply that currently it's the way.

SmartPy

14 March 2026 18:41

No...
That's a good request. Maybe a design could be x = my_list.pip(error=...) or x=my_list.pop(default=...).
I'm a bit responsible for the fact there is no such thing.

What you can do it matching over head and tail the way I showed in my zip example.

SmartPy

14 March 2026 18:20

It's okay, no need to mention me.

SmartPy

14 March 2026 18:17

It's more a design choice because of how Tezos internal way of representing list is a chained list so accessing the index n costs o(n). A request could be to allow it in constructor as constructor cost isn't a problem.

SmartPy

14 March 2026 18:09

have you github:tag? I want to mention you in commit for this version, even if I had to change it in the future

SmartPy

14 March 2026 18:07

yes I noticed that problem about nat and negative form subs

SmartPy

14 March 2026 18:06

I think it's a wrong idea to do it in a 1:1 at least in this case. The version I'm proposing is far easier to read and corresponds to the requirements.
My version is close to yours (AI is probably the reason). Mine fixes a bug: there will be an error if the tez subtraction is lower than zero so your error will never be raised. As the requirements aren't saying we should fail I just used a if. You can also keep your version but invert the assert and the subtraction.

SmartPy

14 March 2026 18:04

I'll make some edit because I need to have same name for var and fun, also for the views

SmartPy

14 March 2026 17:59

here the updated version I'm working on lately

SmartPy

14 March 2026 17:58

Oh yes, it was different from the solidity one

SmartPy

31 March 2026 11:47

Oh thanks, because i'm working on a toolchain for automatic execution and testing and through python code I was calling the bash for compiling code

SmartPy

31 March 2026 11:24

It's possible to compile a smartpy contract via library function instead of calling it with 'python3 contract.py' ?

SmartPy

20 March 2026 12:28

And is smartpy/michelson safe to reentrancy pattern?

SmartPy

17 March 2026 11:40

It's a Michelson limitation

SmartPy

16 March 2026 16:55

Just for the record, actually, because I was keeping the raw code and the scenario separate to keep the contracts cleaner, but then I realized it wasn’t compiling—or rather, it wasn’t generating the output files correctly for deployment. So what’s the bare minimum required to get them?

@sp.add_test()
def test():
sc = sp.test_scenario("SmartContract")
c1 = main.SmartContract()
sc += c1

SmartPy

16 March 2026 16:45

The only way to compile a contract locally is to use the scenario? If so, why is it necessary?

SmartPy

14 March 2026 18:35

there's a way to pop the first element of a list? or the last

SmartPy

14 March 2026 18:19

Now this could also be debatable in the context of the Tezlink layer 2

SmartPy

14 March 2026 18:12

If you will be able in the next smartpy versions to add index access to sp.list as normal python it would be really usefull😄

SmartPy

14 March 2026 18:07

In fact you can remove it when you know what you're doing, nobody should do that basically for now.

SmartPy

14 March 2026 18:07

It's a good idea to put it back, sorry.

SmartPy

14 March 2026 18:05

I'll ask to coordinator for parameter, if it's mandatory to have payee and shares splitted or map is good too

SmartPy

14 March 2026 18:02

a question, why did you not use @sp.add_test?

SmartPy

14 March 2026 17:58

Now i'm trying to make a 1:1 replica

SmartPy

14 March 2026 17:57

Yes, the payment splitter.
Here is my version. Please check the validity before pushing it. Also, I would use black over it and column split for comments.