Curve-boa

This is a fork of the Titanoboa interpreter, specifically for the needs of Curve Finance. In this fork we experiment with adding new features before they are merged into the main repository.

A Vyper interpreter with pretty tracebacks, forking, debugging features and more! Titanoboa's goal is to provide a modern, advanced and integrated development experience for vyper users.

Architecture

Titanoboa achieves feature parity with the vyper compiler while providing an interpreted experience. How does it do this? Internally, titanoboa uses vyper as a library to compile source code to bytecode, and then runs the bytecode using py-evm, adding instrumenting hooks to provide introspection. The use of py-evm means that the entire experience is highly configurable, down to the ability to patch opcodes and precompiles at the EVM level.

Documentation

Usage and quickstart are below. For more detailed documentation, please see the documentation.

Installation

pip install curve-boa

For latest dev version:

pip install git+https://github.com/curvefi/titanoboa

If you are installing titanoboa from git alongside brownie, you may have to manually install titanoboa after installing brownie

pip install brownie
pip install git+https://github.com/curvefi/titanoboa

Sometimes, using pypy can result in a substantial performance improvement for computation heavy contracts. Pypy can usually be used as a drop-in replacement for CPython.

To get a performance boost for mainnet forking, install with the forking-recommended extra (pip install "git+https://github.com/curvefi/titanoboa#egg=titanoboa[forking-recommended]", or pip install curve-boa[forking-recommended]). This installs plyvel to cache RPC results between sessions, and ujson which improves json performance.

If you are running titanoboa on a local Vyper project folder, you might need to run python setup.py install on your Vyper project if you encounter errors such as ModuleNotFoundError: No module named 'vyper.version'

Background

Titanoboa (/ˌtaɪtənəˈboʊə/; lit. 'titanic boa') is an extinct genus of giant boid (the family that includes all boas and anacondas) snake that lived during the middle and late Paleocene. Titanoboa was first discovered in the early 2000s by the Smithsonian Tropical Research Institute who, along with students from the University of Florida, recovered 186 fossils of Titanoboa from La Guajira in northeastern Colombia. It was named and described in 2009 as Titanoboa cerrejonensis, the largest snake ever found at that time. It was originally known only from thoracic vertebrae and ribs, but later expeditions collected parts of the skull and teeth. Titanoboa is in the subfamily Boinae, being most closely related to other extant boines from Madagascar and the Pacific.

Titanoboa could grow up to 12.8 m (42 ft) long, perhaps even up to 14.3 m (47 ft) long, and weigh around 730–1,135 kg (1,610–2,500 lb). The discovery of Titanoboa cerrejonensis supplanted the previous record holder, Gigantophis garstini, which is known from the Eocene of Egypt. Titanoboa evolved following the extinction of all non-avian dinosaurs, being one of the largest reptiles to evolve after the Cretaceous–Paleogene extinction event. Its vertebrae are very robust and wide, with a pentagonal shape in anterior view, as in other members of Boinae. Although originally thought to be an apex predator, the discovery of skull bones revealed that it was more than likely specialized in preying on fish.

Usage / Quick Start

Hello, world

import boa
boa.eval("empty(uint256)")

Basic

# simple.vy
@external
def foo() -> uint256:
    x: uint256 = 1
    return x + 7

>>> import boa

>>> simple = boa.load("examples/simple.vy")
>>> simple.foo()
    8
>>> simple.foo()._vyper_type
    uint256

Passing __init__

>>> import boa

>>> erc20 = boa.load("examples/ERC20.vy", 'titanoboa', 'boa', 18, 1)
>>> erc20.name()
    titanoboa
>>> erc20.symbol()
    boa
>>> erc20.balanceOf(erc20.address)
    0
>>> erc20.totalSupply()
    1000000000000000000

As a blueprint

>>> import boa
>>> s = boa.load_partial("examples/ERC20.vy")
>>> blueprint = s.deploy_as_blueprint()
>>> deployer = boa.load("examples/deployer.vy", blueprint)
>>> token = s.at(deployer.create_new_erc20("token", "TKN", 18, 10**18))
>>> token.totalSupply()
>>> 1000000000000000000000000000000000000

Expecting BoaErrors / handling reverts

>>> import boa
>>> erc20 = boa.load("examples/ERC20.vy", "titanoboa", "boa", 18, 0)
>>> with boa.env.prank(boa.env.generate_address()):
...     with boa.reverts():
...         erc20.mint(boa.env.eoa, 100)  # non-minter cannot mint
...
>>> with boa.env.prank(boa.env.generate_address()):
...     # you can be more specific about the failure reason
...     with boa.reverts(rekt="non-minter tried to mint"):
...         erc20.mint(boa.env.eoa, 100)

From within IPython

In [1]: %load_ext boa.ipython
        import boa
        boa.interpret.set_cache_dir()  # cache source compilations across sessions

In [2]: %vyper msg.sender  # evaluate a vyper expression directly
Out[2]: '0x0000000000000000000000000000000000000065'

In [3]: %%vyper
   ...:
   ...: MY_IMMUTABLE: immutable(uint256)
   ...:
   ...: @external
   ...: def __init__(some_number: uint256):
   ...:     MY_IMMUTABLE = some_number * 2
   ...:
   ...: @external
   ...: def foo() -> uint256:
   ...:     return MY_IMMUTABLE
   ...:
Out[3]: <boa.vyper.contract.VyperDeployer at 0x7f3496187190>

In [4]: d = _

In [4]: c = d.deploy(5)

In [5]: c.foo()
Out[5]: 10

Evaluating arbitrary code

>>> erc20 = boa.load("examples/ERC20.vy", 'titanoboa', 'boa', 18, 1)
>>> erc20.balanceOf(erc20.address)
    0
>>> erc20.totalSupply()
    1000000000000000000
>>> erc20.eval("self.totalSupply += 10")  # manually mess with total supply
>>> erc20.totalSupply()
1000000000000000010
>>> erc20.eval("self.totalSupply")  # same result when eval'ed
1000000000000000010
>>> erc20.eval("self.balanceOf[msg.sender] += 101")  # manually mess with balance
>>> erc20.balanceOf(boa.env.eoa)
1000000000000000101
>>> erc20.eval("self.balanceOf[msg.sender]")  # same result when eval'ed
1000000000000000101

Note that in eval() mode, titanoboa uses slightly different optimization settings, so gas usage may not be the same as using the external interface.

Forking

Create a fork of mainnet given rpc.

In [1]: import boa; boa.env.fork(url="<rpc server address>")

In [2]: %load_ext boa.ipython

In [3]: %%vyper Test
   ...: interface HasName:
   ...:     def name() -> String[32]: view
   ...:
   ...: @external
   ...: def get_name_of(addr: HasName) -> String[32]:
   ...:     return addr.name()
Out[3]: <boa.vyper.contract.VyperDeployer at 0x7f3496187190>

In [4]: c = Test.deploy()

In [5]: c.get_name_of("0xD533a949740bb3306d119CC777fa900bA034cd52")
Out[5]: 'Curve DAO Token'

Cast current deployed addresses to vyper contract

>>> import boa; boa.env.fork(url="<rpc server address>")
>>> c = boa.load_partial("examples/ERC20.vy").at("0xD533a949740bb3306d119CC777fa900bA034cd52")
>>> c.name()
    'Curve DAO Token'

Network Mode

>>> import boa
>>> boa.env.set_network_env("<rpc server address>")
>>> from eth_account import Account
>>> # in a real codebase, always load private keys safely from an encrypted store!
>>> boa.env.add_account(Account(<a private key>))
>>> c = boa.load("examples/ERC20.vy", "My Token", "TKN", 10**18, 10)
>>> c.name()
    'My Token'

Jupyter Integration

You can use Jupyter to execute titanoboa code in network mode from your browser using any wallet. We provide a BrowserSigner as a drop-in replacement for eth_account.Account. The BrowserRPC may be used to interact with the RPC server from the browser.

For a full example, please see this example Jupyter notebook

JupyterLab

Before being able to use the plugin, you need to install it. You can do this by running the following command in the terminal:

pip install curve-boa
jupyter lab extension enable boa

To activate our IPython extension, you need to run the following command in the notebook:

%load_ext boa.ipython

For ease of use, add the following to ipython_config.py:

c.InteractiveShellApp.extensions = ["boa.ipython"]
c.InteractiveShellApp.exec_lines = ['import boa']

We provide a multi-user setup with JupyterLab in try.vyperlang.org, where the extension is installed and activated. The source code for this website is available in the GitHub repository.

Colab

It is also possible to run our plugin in Google Colab. To do this, you need to install the plugin by running the following commands:

!pip install curve-boa
%load_ext boa.ipython

IPython extensions

This activates the %%vyper, %%contract and %%eval magics.

Basic tests

$ python -m tests.integration.sim_veYFI