Extension now interacts with AgentCommunication contract #2
Conversation
| # our component's API: we can pre-process its input args, post-process its | ||
| # output value, and add a docstring for users. | ||
| def my_component(recipient: str, amount_in_ether: str, data: str, key=None): | ||
| def wallet_component(recipient: str, amount_in_ether: str, data: str | None = None, key=None): |
There was a problem hiding this comment.
Adding to our discussion on Slack on having multiple components in the same package -> see https://discuss.streamlit.io/t/creating-multiple-custom-components-in-the-same-python-package/10906/7
tl;dr some work would be needed to maintain multiple components in parallel. Since this is still in the testing phase and we are validating this ideas internally, I decided to keep just one custom component for simplicity.
There was a problem hiding this comment.
But in that case, this repo and pypi package shouldn't be called python-web3-wallet-streamlit.
More something in terms of agent-communication-streamlit.
There was a problem hiding this comment.
Or you could duplicate this repository with the new name, and merge this PR there. That way we have 2 repositories:
- One that already works for sending classic xDai transactions.
- Second that is meant for sending messages to agent.
There was a problem hiding this comment.
The goal of this repository is to simply house a Streamlit wallet component allowing users to send messages to agents.
Previously, sending messages involved sending classic xDai transactions.. This logic changed after our internal discussion with Martin and now involves executing a transaction on a smart contract. Hence, I refactored the logic and now pressing the button triggers the aforementioned transaction. Per the above, I don't think it's necessary to create a new repo/create a new component, since the old logic has been deprecated in favor of the new one.
There was a problem hiding this comment.
I thought you got my reasoning over Slack -- on the one hand, we talk about how awesome it would be to have some tools/frameworks to drive people to Gnosis (AI), on the other, when we have a handy functional component that could be interesting for machine learning people who mostly work in Python and Streamlit, it's going to be deleted because keeping it involves a few boring clicks around Github and PyPi.
However, in the end, it's all your code in this repository, so if you want to get rid of it, I won't block it.
There was a problem hiding this comment.
on the one hand, we talk about how awesome it would be to have some tools/frameworks to drive people to Gnosis (AI)
This requires a bit more planning - it's not "build it and they will come", this involves DevRel, talking to the community to see what they want to build, we prepare something, give talks, etc etc - simply having 1...N components is not going to move the needle.
Happy to brainstorm on this better, but closing this as per above.
| # our component's API: we can pre-process its input args, post-process its | ||
| # output value, and add a docstring for users. | ||
| def my_component(recipient: str, amount_in_ether: str, data: str, key=None): | ||
| def wallet_component(recipient: str, amount_in_ether: str, data: str | None = None, key=None): |
There was a problem hiding this comment.
But in that case, this repo and pypi package shouldn't be called python-web3-wallet-streamlit.
More something in terms of agent-communication-streamlit.
| # our component's API: we can pre-process its input args, post-process its | ||
| # output value, and add a docstring for users. | ||
| def my_component(recipient: str, amount_in_ether: str, data: str, key=None): | ||
| def wallet_component(recipient: str, amount_in_ether: str, data: str | None = None, key=None): |
There was a problem hiding this comment.
Or you could duplicate this repository with the new name, and merge this PR there. That way we have 2 repositories:
- One that already works for sending classic xDai transactions.
- Second that is meant for sending messages to agent.
Co-authored-by: Peter Jung <peter@jung.ninja>
2 participants
Uh oh!
There was an error while loading. Please reload this page.