The past few months have been quite eventful for the Ethereum javascripters. To begin with, an outstanding hackathon was held in collaboration with IPFS. You can read Dan Finlay’s wonderful summary here.
Additionally, during this period, Aaron Davis (Kumavis) made significant advancements towards a JS light client by leveraging IPFS’s libp2p to forge an in-browser mesh network and IPLD to establish the merklization foundation. This endeavor will be crucial for developing pure in-browser clients. Furthermore, Casey Detrio worked on a standard JSON RPC testing suite, which you can view the outcomes of here.
Following the Seattle Meetup, we (Axic and Wanderer) gathered for a week-long hackathon in Budapest to iron out details of ewasm. Meanwhile, in the JS realm, Fabian is conducting an extensive overhaul of Web3.js in preparation for the 1.0 release, while Nick Dodson has been occupied with ethjs. The remainder of this article will explore the diverse prospects that this technology could offer, along with delving into specific details regarding each unique project. All these projects are open source and welcome community engagement, so if you’re interested, feel free to check them out, introduce yourself, and submit a PR when you can!
EWASM
Ewasm’s objective is to investigate and replace the EVM with WebAssembly and, as a secondary aim, to create a client for the existing system that can be efficiently JITed (or transcompiled) to WebAssembly.
A crucial part of assessing WebAssembly for blockchain applications will be to establish a test network, and this year, the Ewasm team will concentrate on bringing that test network into existence. The testnet initiative will:
- allow practical engagement with ewasm for a broader audience
- facilitate related projects, such as experiments with Casper, by providing a versatile platform for exploration
The ewasm track is dedicated to ewasm exploration and progress, while the client integration track will focus on developing the network and bringing robust and light clients into being. There are numerous shared components in these two tracks. The Ewasm project is divided into two primary aspects: the Kernel Layer, which manages IPC and controls the state, and the core VM. This structure should allow us to use the same framework for distinct VM implementations.
To summarize, the key tasks for ewasm include:
- Establish an ewasm test network
- Create a reusable “kernel” module
- Revamp ethereumjs-vm
- Utilize ewasm-kernel for message passing
- Implement the latest EIPs
- Ewasm integration tools
- Solidity-ewasm integration (forthcoming effort for the solidity hackathon!)
We invite you to participate in the implementation initiative! We hold bi-weekly meetings on Tuesdays. Our communication platform is on Matrix at prima:matrix.org (or #ewasm on IRC or at gitter)
Networking
There are various justifications for developing an Ethereum networking implementation in JS. Firstly, it would enable us to create a complete and lightweight Ethereum JS node. These light clients would operate both in a Node.js environment and within a browser. A necessity for an in-browser light client involves “bridge” nodes. These nodes might also serve as signaling servers for the webrtc network which the browser light clients would utilize to relay messages from the RLPx network to the WebRTC framework. This initiative is being led by Metamask employing IPFS’s libp2p. Moreover, the RLPx implementation was recently updated by fanatid.
IPLD
Ethereum’s blockchain and on-chain statecan be interpreted as a diagram of hash-linked information. IPFS/IPLD is suggested as a universal framework to articulate and disseminate hash-linked information. Thus, we can view Ethereum as an application layer built on the hash-linked data availability platform. As a demonstration, Kumavis executed IPLD resolvers for the Ethereum data formats that indicate where hash-links are incorporated within the standard Ethereum formats (e.g., block and state trie node). This, combined with additional universal characteristics of libp2p (IPFS’s generalized p2p networking framework), allows for the development of lightweight Ethereum clients centered around the consensus protocol and the state transition mechanism. A key benefit of this strategy is that the networking layer is transport-neutral and can be utilized in environments lacking access to tcp/udp (such as web browsers) which standard Ethereum clients rely on. Currently, this initiative is still being researched. MetaMask aims to leverage this strategy to create a browser-compatible Ethereum light client through a secondary network, facilitated by hybrid nodes.
Web3.js 1.0 is on the way!
A new iteration of web3.js is under development. It represents the most significant refactor of the codebase since the establishment of the popular Ethereum library. It will introduce numerous convenient features like confirmation and receipt events on transactions, an appealing subscription API, and checksum validations on address entries.
The API remains in the early stages of completion, but if you’re keen to explore, you can view the documentation here.
The new iteration will also encompass several major changes, but such updates are essential to ensure the new API is correctly configured and to phase out certain deprecated methods, including synchronous calls. Version 1.0 will exclusively support promises and in specific scenarios “PromiseEvents” to better represent multiple events during a transaction’s execution. For those contemplating transitioning their applications to the new web3, a migration guide will be available upon release to facilitate the shift from 0.x.x smoothly.
In Mist, there will no longer be a default exposure of web3, as this promotes the undesirable habit of depending on the Mist-supplied web3, making breaking changes detrimental for dapps. Instead, there will be an “ethereumProvider”, that libraries like web3 can utilize to communicate with the underlying node. Web3.js will automatically identify any provided provider and make it accessible in its API for seamless instantiation.
For those keen to try it immediately, check out the 1.0 branch in the web3.js repository. Just be cautious, there may be unforeseen issues!
Ethjs
Ethjs is a newly optimized, lightweight JavaScript utility for Ethereum designed to work with the JSON RPC, much like web3.js but more streamlined, asynchronous only, and utilizing bn.js. Presently, ongoing efforts include:
- Integrating ABI methods for log decoding in ethjs-abi
- Corrected a minor decoding issue in ethjs-abi (addressing 0x addresses)
- Merged the new schema for personal recovery and signing ethjs-schema
- Seeking assistance to make ethjs-filter stateless (ready for Infura)
- Bug resolution in ethjs-contract
- Documentation enhancements throughout
- Upcoming release of ethjs version 0.2.7!
TestRPC
Currently working on the 4.0.0 release! This version will feature:
- Data persistence. You can now establish a test chain and maintain that data, akin to any other private chain!
- Refinement of data storage in memory, which should significantly alleviate memory concerns. Although there may be a minor performance impact, which would mostly be unnoticeable unless many transactions are executed, it will greatly enhance stability.
- Bundling for browser use (provider only).
- Simplified installations on Windows, and potentially other platforms.
We will transition the TestRPC to the Truffle GitHub organization since it is primarily managed by Truffle developers. There are substantial new additions to TestRPC on the horizon. Additionally, we are dedicating considerable effort to unify documentation and branding under the Truffle identity. Any thoughts regarding this transition would be valued. Finally, the TestRPC is in need of a new name that reflects its full capabilities. If you have suggestions, please share!
The Ethereum JS community is a thrilling and remarkable sphere to engage with. There are numerous impressive projects underway. If you’re interested in integrating, we have weekly Friday meetings at 3:00 EST / 10:00 PST / 18:00 UTC. Keep an eye on our Gitter channel for the chat link. Furthermore, we are organizing an upcoming hackathon. Do let us know if you are interested.
ADDENDUM [Mar. 22, 2017]: Please note that certain projects mentioned in this post are not directly endorsed by the Ethereum Foundation but have been included due to their relevance to the broader Ethereum JS ecosystem update by the author.
